OXFORD

INTERNATIONAL
CRIMINAL LAW
Second Edition

Antonio Cassese
O ^ 'P u 0 0 ^ 2 ^

OXFORD
U N IV E R S IT Y PRESS
Great Clarendon Street, Oxford OX2 6DP
Oxford University Press is a department of the University of Oxford.
It furthers the University’s objective of excellence in research, scholarship,
and education by publishing worldwide in
Oxford New York
Auckland Cape Town Dar es Salaam Hong Kong Karachi
Kuala Lumpur Madrid Melbourne Mexico City Nairobi
New Delhi Shanghai Taipei Toronto
With offices in
Argentina Austria Brazil Chile Czech Republic France Greece
Guatemala Hungary Italy Japan Poland Portugal Singapore
South Korea Switzerland Thailand Turkey Ukraine Vietnam
Oxford is a registered trade mark of Oxford University Press
in the UK and in certain other countries
Published in the United States
by Oxford University Press Inc., New York
© Antonio Cassese, 2008
The moral rights of the author have been asserted
Crown copyright material is reproduced under Class Licence
Number C01P0000148 with the permission of OPSI
and the Queen’s Printer for Scotland
Database right Oxford University Press (maker)
First edition published 2003
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 in writing of Oxford University Press,
or as expressly permitted by law, or under terms agreed with the appropriate
reprographics rights organization. Enquiries concerning reproduction
outside the scope of the above should be sent to the Rights Department,
Oxford University Press, at the address above
You must not circulate this book in any other binding or cover
and you must impose the same condition on any acquirer
British Library Cataloguing in Publication Data
Data available
Library of Congress Cataloging in Publication Data
Data applied for
Typeset by Newgen Imaging Systems (P) Ltd, Chennai, India
Printed in Great Britain
on acid-free paper by
Ashford Colour Press Ltd, Gosport, Hampshire

ISBN 978-0-19-920310-9
1 3 5 7 9 10 8 6 4 2
 CONTENTS

Preface xiii
Abbreviations xv
Table o f Cases xix
Table o f Treaties and Conventions xl
Table o f National Legislation xliv
Table o f International Instruments xlvii
Table o f Statutes of International Tribunals xlviii

PART I IN T R O D U C T IO N

1 FUNDAMENTALS OF INTERN A TIO N A L CR IM IN A L LAW 3

1.1 The notion of international criminal law (ICL) 3

1.2 General features of ICL 4
1.3 The notion of international crimes 11
1.4 Sources of ICL 13
1.5 The historical evolution of international crimes 27

2 GENERAL PRINCIPLES OF INTERN A TIO N A L CR IM IN A L LAW 32

2.1 Preliminary remarks 32
2.2 The principle of individual criminal responsibility 33
2.3 The principle of legality of crimes 36
2.4 Articulations of the principle of legality 41
2.5 The principle of legality of penalties 51

3 TH E ELEMENTS OF INTERN A TIO N A L CRIMES 53

3.1 The objective and subjective elements of crimes 53
3.2 The structural elements of international crimes 53
3.3 General features of the subjective element 56
3.4 Substantive rules setting out the mental element required for crimes 57
VI CONTENTS

3.5 General notions of mens rea common to most

legal systems of the world 58
3.6 General categories of mens rea: intent 60
3.7 Recklessness 66
3.8 Culpable or gross negligence 70
3.9 The ICC Statute 73
3.10 Judicial determination of the mental element 74

PA R T II S U B ST A N T IV E C R IM IN A L LAW

SECTION I INTERNATIONAL CRIMES

4 WAR CRIM ES 81

4.1 The notion 81

4.2 The need for a link between the oifence and an armed conflict 82
4.3 Establishing whether a serious violation of IHL has been criminalized 84
4.4 The objective elements 87
4.5 The subjective elements 92
4.6 The definition of war crimes in the ICC Statute 94

5 CRIM ES AGAINST H U M A N IT Y 98

5.1 The notion 98

5.2 The origin of the notion 101
5.3 The objective elements 109
5.4 The subjective elements 114
5.5 The possible authors 116
5.6 The possible victims 117
5.7 Customary international law and Article 7 of the ICC Statute 123

6 GENOCIDE 127

6.1 The notion 127

6.2 The 1948 Convention on Genocide 127
6.3 Developments in the case law on genocide 131
 CONTENTS vii

6.4 The objective elements 133

6.5 The subjective elements 137
6.6 Problematical aspects of genocide 138
6.7 Genocide and crimes against hum anity 144
6.8 Article 6 of the ICC Statute and customary international law 146

7 TORTURE AS A DISCRETE CRIM E, AND AGGRESSION 148

7.1 Introduction 148

7.2 Torture as a discrete crime 149
7.3 The crime of aggression 152

8 TERRORISM AS AN INTERN A TIO N A L CRIM E 162

8.1 A current misconception: the alleged lack of a generally

agreed definition of terrorism 162
8.2 Factors pointing to a generally agreed definition
of terrorism in time of peace 164
8.3 The ingredients of international terrorism as a
discrete international crime in time of peace 166
8.4 Specific sub-categories of international terrorism
as a discrete international crime 169
8.5 International terrorism in armed conflicts:
a sub-category of war crimes 171
8.6 International terrorism as a sub-category of
crimes against hum anity 175
8.7 Summing up 177
8.8 The general question of multiplicity of offences 178

SECTION II MODES OF CRIM INAL LIABILITY

9 P ER PETRA TION AND JOINT CR IM IN A L EN TER PR ISE 187

9.1 General 187

9.2 Perpetration 188
9.3 Co-perpetration 189
9.4 Participation in a joint criminal enterprise to
commit international crimes 189
viii CONTENTS

10 O T H E R MODES OF LIABILITY 214

10.1 Aiding and abetting 214

10.2 Incitement as a form of participation in international crimes 218
10.3 Inchoate crimes: general 219
10.4 Attempt 220
10.5 Planning 225
10.6 Conspiracy 227
10.7 Incitement to genocide 229
10.8 Ordering 230

11 CRIM INAL LIABILITY FOR OMISSIONS 233

11.1 General 233

11.2 Rules imposing a positive obligation to act 234
11.3 Culpable omission of an act mandated by an international criminal rule 235
11.4 The responsibility of superiors 236

SECTION III CIRCUMSTANCES EXCLUDING

CRIM INAL LIABILITY

12 JUSTIFICATIONS AND EXCUSES 255

12.1 The distinction between justifications and excuses 255

12.2 ICL: general 258
12.3 Self-defence 259
12.4 Excuses: two m ain categories 262
12.5 Excuses based on lack of individual autonomy 263

O T H E R EXCUSES: SUPERIOR ORDER, NECESSITY,

DURESS, AND MISTAKE 268

13.1 Excuses where lack of mens rea derives from

external circumstances 268
13.2 The question of superior orders: may they be pleaded as a defence? 268
13.3 Necessity and duress 280
13.4 Mistake of fact 290
\ 13.5 Mistake oflaw 294
 CONTENTS ix

4 IM M U N IT IES 302
14.1 General: various classes of immunities 302
14.2 Functional and personal immunities provided for in international
customary law 303
14.3 The customary rule lifting functional immunities with respect to
international crimes 305
14.4 International personal immunities 309
14.5 National personal immunities 314

PART I I I P R O S E C U T IO N A N D P U N IS H M E N T BY
IN T E R N A T IO N A L C O U R TS

15 THE ESTABLISHMENT OF INTERN A TIO N A L 317

CRIM INAL TRIBUNALS

15.1 Abortive early attempts (1919-1945) 317

15.2 The Nuremberg and Tokyo Tribunals (1945-1947) 319
15.3 The work of the ILC (1950-1954) 323
15.4 The Post-Cold War ‘new world order’ and the establishment of ad hoc
Tribunals (1993-1994) 324
15.5 The drafting and adoption of the Statute of the ICC (1994-1998) 328
15.6 Other criminal tribunals 330

16 INTERNATIONAL VERSUS NATIONAL JU RISDICTION 336

16.1 Primacy and complementarity 336

16.2 The primacy of international ad hoc Tribunals 339
16.3 The complementarity of the ICC 342
16.4 The Nuremberg scheme versus the ICC scheme 344
16.5 The need for State cooperation 346
16.6 Models of cooperation 346
16.7 Cooperation of States under the ICTY and ICTR scheme 347
16.8 Cooperation of States under the ICC scheme 349
16.9 The question of surrender of nationals 351
X CONTENTS

17 THE A D O P T IO N OF TH E ESSENTIAL FEATURES OF

THE ADVERSARIAL SYSTEM 353

17.1 The adversarial versus the inquisitorial system: general 353

17.2 How the two models work: a comparison 355
17.3 The transposition of the adversarial model on to
the international legal level 366
17.4 The principal elements of the inquisitorial model
incorporated into international procedure 371
17.5 Towards a mixed’ procedural model 376

18 GENERAL PRINCIPLES GO VERNIN G INTERNATIONAL

CRIM INAL TRIALS 378

18.1 The nature and role of principles 378

18.2 The principle that judges must be independent and impartial 379
18.3 The presumption of innocence 380
18.4 The principle of fair and expeditious trial 383
18.5 The principle that the accused should be present at his trial 389

c
19 STAGES OF INTERN A TIO N A L PROCEEDINGS IN O U TLINE
A P R E -T R IA L AND TRIAL

(A) p r o s e c u t o r ’s i n v e s t i g a t i o n s a n d

P R E - T R I A L P R O C E E D IN G S
395

395

19.1 General 395

19.2 The setting in motion of international criminal investigations 395
19.3 Conditions the prosecutor must fulfil before
initiating an investigation 397
19.4 Conduct of investigations by the prosecutor 400
19.5 Pre-trial proceedings 407

(b ) T R I A L P R O C E E D IN G S 409

19.6 Case presentation 409

19.7 Rules of evidence 413
19.8 Control of proceedings 417
19.9 Deliberations 418
19.10 Sentencing 420
19.11 Reparation or compensation to victims 422
 CONTENTS Xi

20 APPEALS AND EN FORCEM ENT 424

20.1 General # 424

(a ) APPEALS 425

20.2 Appeals against interlocutory decisions 425

20.3 Appeals against judgment or sentence 427

(b ) r e v ie w 429
20.4 Review of judgment or sentence 429
20.5 Review of other final decisions 430
(c) E N F O R C E M E N T OF S E N T E N C E S 431

20.6 Place of imprisonment 431

20.7 Conditions of detention 431
20.8 Reduction or commutation of sentence and pardon 432
20.9 Supervision of imprisonment 434

21 THE SPECIFICITY OF INTERN A TIO N A L TRIALS 435

21.1 The need for international trials 435

21.2 Merits of international criminal justice 438
21.3 The main traits of international criminal proceedings 440
21.4 Main problems besetting international criminal proceedings 442

Index 445
 x lv i TABLE OF N A T IO N A L LEGISLATIO N

Slovenia s. 1(1)... 44
1991 1994
Constitution of Slovenia. . . 351 Criminal Justice and Public Order Act
2000 . . . 45
Revised Constitution of Slovenia 2000
Art. 47 . . . 351 Terrorism Act
s. 1 . . . 165
Spain 2001
1985 International Criminal Court Act
M ilitary Criminal Code (C. 1 7 )... 349
Art. 2 1 ... 271 s. 2 (1 )... 349
s. 5 (5 )... 349
Law on Judicial Power
s. 5 (8 )... 349
Art. 21(2)... 310
Art. 23 . . . 310
United States of America
Sweden Foreign Sovereign Immunities Act
. . . 308
1999
Model Penal C ode. . . 62
Crim inal Code (c. 24)
Art. 2(2)(c). . . 58, 67
s. 8 . . . 271
Art. 2(2)(d). . . 59
1863
Switzerland
Lieber C o d e. . . 28
1927 Art. 4 4 ... 18
Penal Military Code 1948 (amended 1970)
Art. 109 . . . 84
Federal Criminal Code . . . 338
2001
1956
Federal Law on Cooperation with the
Uniform Code of M ilitary Justice
ICC . .. 349
Rule 850(a)(b). . . 419
2002
1965
Crim inal C ode. . . 88
New York Penal Code
Rule 15.5(3)... 67
United Kingdom Rule 1505(4)... 70
1215 1996
Magna Carta Antiterrorism and Effective Death Penalty
Art. 39 . . . 37, 357 A c t... 165
1957 War Crimes, US Code, Title 18, Chapter
Homicide A c t. . . 265 1 8 ... 436
1964
Crim inal Justice and Public Order A c t. . . 360 Yugoslavia
1967 1992
Crim inal Justice Act Constitution of the Federal Republic of
s. 9 . . . 358 Yugoslavia
1976 Art. 1 7 -3 ... 351
Sexual Offences (Amendment) Act
 TABLE OF IN T E R N A T IO N A L
IN ST R U M E N T S
1948 UN General Assembly Resolution 51/210
Universal Declaration of Human Rights . . . 11, . .. 165
99
Art. 5 . . . 18 1997
UN Doc. A/C6/234 . . . 134 UN General Assembly Resolution 52/165
. .. 165
1970
UN Declaration on Friendly Relations . . . 11, 1998
156-7 UN General Assembly Resolution 53/183
. .. 165
1975 UN Security Council Resolution 1165 . . . 326
Declaration of the United Nations General UN Security Council Resolution 1166 . . . 326
A ssem bly... I l l
1999
1989 UN General Assembly Resolution 54/110
UN General Assembly Resolution 44/ . .. 165
39 . . . 328
2000
1993 UN General Assembly Resolution 55/158 .
UN Security Council Resolution 808 Doc .. 165
S/25704 . . . 326, 398
para. 2 . . . 132, 307 2001
UN Security Council Resolution 827 ... UN General Assembly Resolution 56/88 . . . 165
326, 398
para. 2 . . . 326
2002
1994 UN General Assembly Resolution 57/27 . . . 165
UN General Assembly Resolution 50/53 . . . 165
UN Security Council Resolution 935 . . . 327 2003
UN General Assembly Resolution 58/81 . . . 165
1995
UN General Assembly Resolution 49/10 . . . 165 2004
UN General Assembly Resolution 69/46... 165
1996
UN General Assembly Resolution 51/110 .
.. 165
 TABLE OF STA TU TES OF
IN T E R N A T IO N A L TR IB U N A L S
1945 ICTY RPE (Rules of Procedure and
C harter of the IMT at Nuremberg . . . 31,41, Evidence). . . 369, 385,433
104,107,127, 161, 196, 306, 321, 322, 323, 378, Rule 9(ii) and (iii). . . 340
387, 441 Rule 11 bis . . . 340, 345
Art. 6 . . . 107,113,196,228 Rule 15(a). . . 380
Art. 6 (a )... 152 Rule 15(c). . . 370
Art. 6(c). . . 105, 117,118, 119,121, 127 Rule 28(A) • • • 398, 406
Art. 8 . . . 271 Rule 40(A). . . 400
Art. 9 . . . 34 Rule 40(B). . . 400
Art. 9-11 . . . 196 Rule 40 bis . . . 400
Art. 1 0 ... 34 R ule4 7 ... 398
Art. 12 . . . 367, 391 R ule47(B )... 404,406,415
Art. 16 . . . 384 Rule 50 . . . 404
Art. 19 . . . 414 Rule 51(A). . . 404
Art. 24(j). . . 409 Rule 54(D-1). . , 348
Rule 62 b is. . . 407
Rule 65 te r. . . 370,371
1946
Rule 65 ter (G ). . . 411
Charter of the Tokyo International Tribunal Rule 65 ter (L)(i)... 408
1 9 4 6 ... 109,322,368,387 Rule 66(A)(i). . . 407
Art. 9 . . . 384 Rule 66(A)(ii) • • • 386
Statute of the International Military Tribunal for Rule 66(B). . . 386
the Far East (IMTFE) Tokyo T ribunal. . . 41, Rule 67(A)(II)(b). . . 266
148, 161, 323, 378 Rule 67(C) • • • 386
Rule 68 . . . 440
1993 Rule 70 . . . 416
Rule 72(B). . . 426
ICTY Statute . . . 15, 19, 22, 26, 41, 99, 108, 132, Rule 73 (B ). . . 426
181, 326, 327, 348, 369, 379, 387, 389, 402,405 Rule 73 b is. . . 370
Arts 1 - 8 ... 41
Rule 73 bis (C ). . . 408
Art. 2 . . . 16,182
Rule 73 te r. . . 370,411
Art. 3 . . . 86,180,182
Rule 75(B). . . 417
Art. 4 . . . 132, 135, 136
Rule 77(A). . . 417
Art. 4(3)(a). . . 206
Rule 78 bis . . . 371
Art. 5 . . . 19,124
Rule 7 9 ... 417
Art. 7 (1 )... 25, 69, 226, 231, 232, 235, 248 Rule 80 . . . 417
Art. 7 (2 )... 306, 310
Rule 84 . . . 409
Art. 7 (3 )... 47, 243
Rule 84 bis. . . 409
Art. 9 . . . 339
Rule 85(B )... 411
Art. 18(1)... 395
R ule86(C )... 412
Art. 18(2)... 308
Rule 87(A). . . 414
A rt.19(1)... 406
Rule 89(D) • • • 414
Art. 21 . . . 384
Rule 89F . . . 372
Art. 21(2)... 384, 387
Rule 90(F). . . 411
Art. 21(3)... 380 Rule 90(H)(i). . . 410
Art. 21(4)(d). . . 392
Rule 90(H)(ii). . . 410
Art. 2 2 ... 387
Rule 90(H)(iii). . . 411
Art. 24(1)... 51,420
Rule 92 . . . 382
Art. 2 4 (3 )... 422
Rule 92 te r ... 372
Art. 2 6 ... 429
Rule 92 bis . . . 372, 415
Art. 2 7 ... 431, 434
Rule 94(B). . . 372
Art. 28 . . . 433
Rule 95 . . . 415
Art. 29 . . . 347, 394
 TABLE OF STATUT ES OF IN T E R N A T IO N A L TR IB U N A L S x lix

Rule 9 6 .. . 415 Art. 1 . . . 343, 399

Rule 9 7 .. . 416 Art. 5 . . . 113, 120, 154
Rule 105. .. 422 Art. 5 (1 )... 399
Rule 106. .. 422 Art. 5 (2 )... 154,155,439
Rule 115. .. 428 Art. 6 . . . 56, 146
Rule 116 b is ... 426 Arts 6 - 8 . . . 73
Rule 119. .. 429 Art. 1 . . . 56,109, 123, 124, 125, 126
Rule 120- 1 ... 429 Art. 7(1)... 63,124,125
Rule 121- 6 . . . 407 Art. 7(l)(a)-(j). . . 124
Rule 123. .. 433 Art. 7(l)(c)-(k). . . 125
Rule 125. .. 433 Art. 7(l)(e). . . 124
Art. 7 (l)(f). . . 150
Art. 7(l)(g). . . 124, 126
1994
Art. 7(l)(h). . . 125, 126
ICTR S tatu te... 15,22,26,41 Art. 7(l)(i). . . 126
172, 327, 379, 387, 397, 422 Art. 7(l)(j) ■• ■ 13>126
Art. 1 ... 327 Art. 7(l)(k). . . 41,49,57,114,124
Art. 2 ... 132 Art. 7(2)(a). . . 125
Art. 2(2) (a). . . 50 Art. 7(2)(b). . . 109, 123
Art. 4 . . . 16,90 Art. 7(2)(c). . . 110,123
Art. 4 (d ). .. 173 Art. 7(2) (d ). . . 110, 124
Art. 6 . . . 124 Art. 7(2)(e). . . 111,124
Art. 6(1). .. 225 Art. 7(2)(f). . . 113,126
Art. 6(2). .. 310 Art. 7(2)(g). . . 113
Art. 6(3). .. 243 Art. 7(2)(h). . . 13, 126
Art. 8 ... 339 Art. 7(2)(i). . . 113, 126, 175
Art. 17(1) ... 395 Art. 7 (3 )... 126
Art. 2 0 .. 384 Art. 8 . . . 56,94,96, 172
Art. 20(2) ... 384,387 Art. 8 (1 )... 279, 399
Art. 20(3) ... 380 Art. 8(2)(b). . . 94
Art. 20(4)(d)... 392 Art. 8(2)(b)(iv). . . 64,96
Art. 21.. 387 Art. 8(2)(b)(viii). . . 95
Art. 2 3 .. 52 Art. 8(2)(b)(xii). . . 95
Art. 23(1) . . . 420 Art. 8(2)(b)(xvii). . . 95
Art. 23(3) ... 422 Art. 8(2)(b)(xxvi). . . 97
Art. 2 4 .. 430 Art. 8(2)(c). . . 96
Art. 2 5 .. 429,430 Art. 8(2)(c)-(f). . . 81, 88
Art. 2 6 .. 431,434 Art. 8 (2) (e). . . 94
Art. 2 7 .. 433 Art. 8(2)(e)(vii). . . 97, 267
Art. 2 8 .. 347 Art. 8(2)(f). . . 96
ICTR RPE. . 385,428,433 Art. 9 . . . 56
Rule 6 8 .. . 440 Art. 10 . . . 172
Rule 72.. . 426 Art. 13(a). . . 344
Rule 7 2(D )... 426 Art. 13(b). . . 344, 397
Rule 117. .. 426 Art. 13(c). . . 344
Rule 120. .. 429, 430 Art. 14 . . . 344, 397
Rule 121-2... 429 Art. 15 . . . 343, 344, 396
Rule 1 2 4 -2 6 ... 433 Art. 15(1)... 396
Art. 15(3)—(5 )... 399
1998 Art. 17 . . . 343, 403
Art. 17(l)(d). . . 345
ICC Statute (Statute of the International
Art. 17(2)... 343
Criminal C o u rt). . . 14, 15, 17, 19, 22, 26, 41,
Art. 17(3)... 343
48, 51, 73, 95, 97, 99, 108, 110, 132, 146, 154,
Art. 18 . . . 343
212, 213, 224, 262, 279, 280, 298, 314, 317, 349,
Art. 18(1)... 344, 399
351, 370, 374, 376, 378, 379, 387, 399,403, 404,
Art. 1 9 ... 343, 400
419, 428, 439,440
Art. 21(1)... 15
Preamble, para. 10 . . . 343
Art. 21(l)(c). . . 22
 TABLE OF STATUT ES OF IN T E R N A T IO N A L TRIBU N A L S

Art. 2 2(1)... 41 Art. 67(1) (i). . . 419

Art. 2 2 (2 )... 48, 50 Art. 68 . . . 387
Art. 2 3 ... 52 Art. 69 . . . 417
Art. 2 5(1)... 212 Art. 6 9 (7 )... 415
Art. 2 5 (3 )... 213 Art. 70 . . . 418
Art. 25(3)(c). . . 146, 215 Art. 7 2 ... 350,351,417
Art. 25(3)(d). . . 146, 212, 213 Art. 75 . . . 423
Art. 25(3)(e). . . 146 Art. 75(1)... 423
Art. 25(3)(f). . . 146, 224 Art. 7 6 ... 413
Art. 26 . . . 97, 267 Art. 77 . . . 52, 422
Art. 2 7 (2 )... 311,312 Art. 77(2)... 422
Art. 2 8 ... 146,243 Art. 78 . . . 422
Art. 28(a)(i). . . 74 Art. 78(3)... 183
Art. 28(l)(a). . . 147 Art. 79 . . . 422, 423
Art. 28(2)(a). . . 147 Art. 81 . . . 428
Art. 30 . . . 56, 60, 73, 74, 212 Art. 82 . . . 426
Art. 3 0(1)... 73 Art. 82(2)... 427
Art. 3 0 (2 )... 62, 73 Art. 83 . . . 428
Art. 3 0 (3 )... 73 Art. 8 4 ... 430
Art. 3 1 ... 261 Art. 86 . . . 349
Art. 31(l)(a). . . 263 Arts 8 6 -1 0 2 ... 403
Art. 31(l)(b). . . 266 Art. 8 7(7)... 351,403
Art. 31(l)(c). . . 259, 261, 262 Art. 9 0 (6 )... 350
Art. 31(l)(d). . . 14,289 Art. 9 0 (7 )... 350
Art. 32(1)... 290, 294 Art. 93(4)... 350
Art. 33 . . . 279,280 Art. 101(1)... 350
Art. 33(1)... 279 Art. 103(l)(a). . . 431
Art. 3 3(2)... 279 Art. 104 . . . 433
Art. 36(9) (a). . . 379 Art. 106(2)... 432
Art. 4 0 (1 )... 379 Art. 1 10... 433
Art. 46 . . . 430 Art. 110(1)... 432
Art. 51(1)... 26 Art. 110(2)... 432
Art. 51(2)... 26 Art. 121 . . . 95
Art. 51(3)... 26 Art. 123 . . . 95
Art. 52(c). . . 344 Art. 124 . . . 97
Art. 53(1)... 399,405 IC C R P E ... 378,418
Art. 5 3(2)... 405 Rule 34 . . . 380
Art. 53(3)... 405 Rule 61 . . . 120
Art. 53(3)(b). . . 405 Rule 73(4)—(6) . . . 417
Art. 54(l)(a). . . 373,401 Rules 7 6 -8 4 ... 408
Art. 55(1)... 404 Rule 9 4 ... 423
Art. 5 5(2)... 404 Rule 9 4 - 9 ... 423
Art. 56 . . . 401 Rule 95 . . . 423
Art. 5 6 (3 )... 426 Rule 97(2)... 423
Art. 57(3)(d). . . 403, 427 Rule 121(10)... 408
Art. 5 8 ... 374,401,407 Rule 1 2 4 ... 391
Art. 60 . . . 374 Rule 131 . . . 408
Art. 6 0 (1 )... 408 Rule 140 . . . 413
Art. 61 . . . 404 Rule 140(2)(c). . . 413
Art. 6 1(7)... 407 Rule 143 . . . 413
Art. 6 3(1)... 391 R ule2 1 1 ... 433,434
Art. 6 3 (2 )... 391 Rule 211(l)(c). . . 434
Art. 64 . . . 384,417
Art. 64(8)(b). . . 413
2000
Art. 66 . . . 380, 384
Art. 6 6 (3 )... 415 Draft Statute of the Special Court of Sierra
Art. 6 7 ... 384 Leone (SCSL)
Art. 67(1)... 387 Art. 3 (d )... 173
 TABLE OF STATUT ES OF IN T E R N A T IO N A L TRIBU N A L S li

2002 2005
SCSL Statute ... 26, 41, 172, 387, 397,429 Statute of the Iraqi High T ribunal. . . 153
Art. I . . . 398 Art. 14, para. 3 . . . 153
Art. 9 (2 )... 342
Art. 15(1)... 395
2007
Art. 17 . . . 384
Art. 19(1)... 420 Statute of the Special Tribunal for
Art. 19(3)... 422 Lebanon . . . 149, 374, 375- 6, 376, 441, 442
Art. 20(l)(a)... 427 Art. 11(1)... 395
Art. 23 ... 433 Art. 18(1)... 406
Art. 2 0 (2 )... 412
SCSL RPE
Art. 2 2 ... 391
Rule 72(B)... 426
Art. 2 4(1)... 420
Rules 72(D). . . 426
Art. 25 . . . 423

2004
Statute of the Extraordinary Chambers in the
Court of Cambodia (ECCC). . . 374-5
Rules of the ECCC . . . 374-5
 PREFA CE

In this book I have tried succinctly to expound the fundamentals of both substantive and
procedural international criminal law. In so doing, I have made an effort to conceptual­
ize as much as possible; that is, give what I hope is a coherent theoretical framework to
the patchwork of disparate rules, principles, concepts, and legal constructs that at present
make up international criminal law.
I would be content if this book could serve as a general introduction, for both students
and practitioners, to this fascinating branch of law and as a stimulus to other scholars or
practitioners to delve deeper into the basic notions of international criminal law.
All the national or international cases that seemed relevant to a particular matter under
discussion have been cited. The purpose of my mentioning cases (mostly in footnotes, in
order to make the text smoother) is not only to support a specific proposition by refer­
ence to the jurisprudence relating thereto, or to show how courts have applied a rule of
law, or what interpretation they have placed on it. My aim is also to point to the historical
and human dimension of cases. For this purpose, I have as far as possible recounted the
facts behind the courts’ legal findings. For one should never forget that this body of law,
more than any other, results from a myriad of smaller or greater tragedies. Each crime is
a tragedy, for the victims and their relatives, the witnesses, the community to which they
belong, and even the perpetrator who, when brought to trial, will endure the ordeal of
criminal proceedings and, if found guilty, may suffer greatly, in the form of deprivation
of life, at worst, or of personal liberty, at best. Law, it is well known, filters and rarefies the
halo of horror and suffering surrounding crimes. As a consequence, when one reads a law
book or a judgment, one may tend to forget the violent origin of criminal law prescriptions.
That origin, however, remains the underpinning of those prescriptions. To recall it may
serve as a reminder of the true historical source of criminal law. This branch of law is about
human folly, wickedness, and aggressiveness. It deals with the darkest side of our nature. It
also deals with how society confronts vicious violence and seeks to stem it as far as possible
so as ‘to make gentle the life on this world’.
To provide the English-speaking reader with details of cases in other languages, I
have relied extensively upon relevant judgments in Dutch, French, German, Italian, and
Spanish, besides the most significant cases in English. Translations are mine, unless indi­
cated to the contrary.
The reader interested in consulting the treaties and other documents cited in this book,
as well as the relevant legal literature in English, may use the Oxford University Press com­
panion web site: www.oup.com/uk/best.textbooks/law/cassese_ internationalcriminallaw
I am grateful to Laura Magi for skilfully helping me revise, update, and enrich this text.
In this second edition I have restructured the book, revised and updated all the chapters,
and expunged some sections that have now appeared to me to be less relevant.
I am much beholden to Paola Gaeta for kindly reading and making insightful comments
on some chapters. Of course, the responsibility for any misapprehension that may remain
rests solely with me.
 1
FUNDAMENTALS OF
INTERNATIONAL
CRIMINAL LAW

1.1 THE NOTION OF INTERNATIONAL

CRIMINAL LAW (i Cl)

International criminal law (henceforth: ICL) is a body of international rules designed

both to proscribe certain categories of conduct (war crimes, crimes against humanity,
genocide, torture, aggression, terrorism) and to make those persons who engage in
such conduct criminally liable. They consequently either authorize states, or impose
upon them the obligation, to prosecute and punish such criminal conducts. ICL also
regulates international proceedings before international courts and tribunals, for
prosecuting and trying persons accused of such crimes.
The first limb of this body makes up substantive law. This is the set of rules indicat­
ing what acts are prohibited, with the consequence that their authors are criminally
accountable for their commission; they also set out the subjective elements required
for such acts to be regarded as criminalized, the possible circumstances under which
persons accused of such crimes may nevertheless not be held criminally liable, and
also the conditions on which states may or must, under international rules, pros­
ecute or bring to trial persons accused of one of those crimes. This whole corpus of
rules is premised on the general notion that international legal prescriptions are cap­
able of imposing obligations directly on individuals, without the intermediary of the
state wielding authority over such individuals. As the IMT at Nuremberg stated in
1946, ‘the essence of the [Tribunal’s] Charter [i.e. the international treaty establishing
the Tribunal and regulating its powers] is that individuals have international duties
which transcend the national obligations of obedience imposed by the individual state
(Goring and others, at 223). An Indonesian court echoed this dictum in 2002 by saying
that ‘each individual must comply with international obligations beyond the laws of
its nation (Abilio Soares, at 87).
The set of rules regulating international proceedings before international criminal
courts and tribunals, that is procedural criminal law, governs the action by prosecut­
ing authorities and the various stages of international trials.
4 IN T E R N A T IO N A L C R IM IN A L LAW

1.2 GENERAL FEATURES OF ICL

ICL is a branch of public international law. The rules making up this body of law
emanate from sources of international law (treaties, customary law, etc.).1Hence, they
are subject, among other things, to the principles of interpretation proper to that law.
However, one should not be unm indful of some unique features of ICL.

1. First, ICL is a relatively new branch of international law. The list of international
crimes, that is of the acts for whose accomplishment international law makes the
authors criminally responsible, has come into being by gradual accretion. Initially, in
the late nineteenth century, and for a long time after, only war crimes were punishable.
(Piracy, traditionally considered an international crime, is not discussed in this
book for, in addition to having become obsolete, it does not meet the requirements of
international crimes proper; see infra, 1.3). It is only since the Second World War that
new categories of crime have developed, while that of war crimes has been restated: in
1945 and 1946, the Statutes of the International Military Tribunal at Nuremberg (IMT)
and the International Military Tribunal for the Far East (IMTFE), respectively, were
adopted, laying down new classes of international criminality. Thus, in 1945 crimes
against humanity and crimes against peace (chiefly wars of aggression) were added, fol­
lowed in 1948 by genocide as a special subcategory of crimes against humanity (soon to
become an autonomous class of crimes), and then in the 1980s, by torture as a discrete
crime. Recently, international terrorism has been criminalized, subject to certain con­
ditions. As for rules on international criminal proceedings, they were first laid down in
the Statutes of the IMT and the IMTFE, then in those of the ICTY and the ICTR, and
more recently in the Rome Statute of the ICC, in the Statute of the Special Court for
Sierra Leone (SCSL), and of the Special Tribunal for Lebanon (STL), as well as the Rules
of the Extraordinary Courts for Cambodia (ECCC). Nonetheless they are still scant
and, what is even more important, they only pertain to the specific criminal court for
which they have been adopted; that is, they have no general scope. A fully fledged cor­
pus of generally applicable international procedural rules is only gradually evolving.
2. Furthermore, ICL is still a very rudimentary branch of law. The gradual broadening
of substantive criminal law has been a complex process. Among other things, when a
new class of crime has emerged, its constituent elements (the objective and subjective
conditions of the crime, or, in other words, actus reus and mens rea) have not been
immediately clear, nor has any scale of penalties been laid down in international rules.
This process can be easily explained. Three main features of the formation of ICL
stand out.
The first is that, for a long time, either treaties or (more seldom) customary rules
have confined themselves to prohibiting certain acts (for instance, killing prisoners of

1 For a succinct survey of these sources, I take the liberty of referring the reader to my book, International
Law, 2nd edn, (Oxford: Oxford University Press, 2005), 153-237.
 F U N D A M E N T A L S OF I N T E R N A T IO N A L C R IM IN A L LAW 5

war or bombing civilians). These prohibitions were, however, addressed to states, not
directly to individuals: belligerent Powers were legally obliged to prevent their offi­
cials (or, more generally, their nationals) from committing the prohibited acts. It fol­
lowed that, if any such act was performed, the state to which the individual belonged
was responsible under international law vis-ä-vis the state of which the victims were
nationals. Gradually, by bringing to trial before their courts enemy servicemen who
had breached international rules of warfare, states made individuals directly and per­
sonally accountable: gradually, state responsibility was either accompanied or replaced
by individual criminal liability. W hen this occurred, the inference became warranted
that international customary or treaty rules addressed themselves not only to states but
also to individuals, by criminalizing their deviant behaviour in time of war. However,
this criminalization was insufficient and inadequate: international rules did not pro­
vide for either the objective and subjective requirements of the crimes or for the crim ­
inal consequences of the prohibited conduct; in other words, they did not lay down the
conditions for its criminal repression and punishment.
It follows that international law left to national courts the task of prosecuting and
punishing the alleged perpetrators of those acts. As a consequence, municipal courts
of each state applied their procedural rules (legal provisions on jurisdiction and on
the conduct of criminal proceedings) and rules on ‘the general p art’ of substantive
criminal law; that is, on the definition and character of the objective and subjective
elements of crimes, on defences, etc. Among other things, very often national courts,
faced with the indeterminacy of most international criminal rules, found it necessary
to flesh them out and give them legal precision by drawing upon their own criminal
law. They thus refined notions initially left rather loose and woolly by treaty or cus­
tomary law.2
Finally (and this is the third of the features referred to above), when international
criminal courts were set up (first in 1945-7, then in 1993-4 and more recently in 1998
and 2002-7), they did indeed lay down in their Statutes the various classes of crime to
be punished; however, these classes were couched merely as offences over which each
court had jurisdiction. In other words, the crimes were not enumerated as in a crim ­
inal code, but simply as a specification of the jurisdictional authority of the relevant
court. The value and scope of those enumerations was therefore only germane to the
court’s jurisdiction and did not purport to have a general reach.
Given these characteristics of the evolution of ICL, it should not be surprising that
even the recent addition of the sets of written rules referred to above has not proved suf­
ficient to build a coherent legal system, as is shown by the heavy reliance by the newly
created international courts upon customary rules or unwritten general principles.

2 Still very recently a national court, the Hague Court of Appeal in the van Anraat case, faced with the
problem of determining the mental element of aiding and abetting (or complicity), discussed whether to
apply Dutch criminal law rather than ICL, in view of the unclear status of ICL on the matter (see §7). The
Court in the event applied Dutch law (§§ 11.9—11.19 and 12.4). The Court, however, concluded that ‘From an
international criminal law perspective, these requirements [set out in Dutch criminal law] for the contribu­
tion of the so-called “aider or abettor” are not essentially more severe’ Ch §12.4).
6 I N T E R N A T IO N A L C R IM IN A L LAW

As for procedural law, it was scantily delineated in the Statutes of the IMT and the
Tokyo Tribunal. Only recently has it been fortified, when various international crim ­
inal courts and tribunals have been set up, as noted above. Nonetheless, even pro­
cedural law remains at a rather underdeveloped stage and in any case has no general
purport (in that each international tribunal has its own rules of procedure).

3. ICL also presents the unique characteristic that, more than any other segment
of international law, it simultaneously derives its origin from and continuously draws
upon both international humanitarian law and human rights law, as well as national
criminal law.
International hum anitarian law (IHL) embraces principles and rules designed to
regulate warfare both by restraining states in the conduct of armed hostilities and by
protecting those persons who do not take part, or no longer take part (having fallen
into the hands of the enemy), in combat. As ICL, at its origin, was chiefly concerned
with offences committed during armed hostilities in time of war (war crimes), it was
only natural for it to build heavily upon international hum anitarian rules: violations
of these rules, which normally only generated state responsibility, gradually came to be
considered as breaches of law also entailing individual criminal liability. For instance,
previously the indiscriminate bombing of civilians was only considered a wrongful
act attributable to the relevant belligerent state and entailed the international respon­
sibility of that state vis-a-vis the enemy belligerent. Gradually the same act also came
to be regarded as a war crime for which those ordering and executing the indiscrim in­
ate attack had to bear individual criminal liability.
The description of the prohibited conduct that thus came to be criminalized was
to be found in rules of IHL; consequently those applying ICL had perforce to refer to
that body of law to establish which particular conduct IHL rules enjoined states to
refrain from, hence which conduct, if taken, amounted to a crime of the individuals
concerned.
Hum an rights law essentially consists of international treaties and conventions
granting fundamental rights to individuals by simultaneously restricting the author­
ity yielded by states over such individuals. It also includes the copious case law of
international bodies such as the European Court of Human Rights (ECHR), the Inter-
American Court of Hum an Rights (IACHR) and the UN Hum an Rights Committee
(HRC). This corpus of legal provisions and decisions has contributed to the develop­
ment of criminal law in many respects. It has expanded or strengthened, or created
greater sensitivity to, the values (hum an dignity, the need to safeguard life and limb as
far as possible, etc.) to be protected through the prohibition of attacks on such values.
Furthermore, hum an rights law lays down the fundamental rights of suspects and
accused persons, of victims and witnesses; it also sets out the basic safeguards of fair
trial. In short, this increasingly im portant segment of law has impregnated the whole
area of ICL.
In addition, most customary rules of ICL have primarily evolved from municipal
case law relating to international crimes (chiefly war crimes).
 F U N D A M E N T A L S OF I N T E R N A T IO N A L C R I M I N A L LAW 7

This element, as well as the paucity of treaty rules on the matter, explains why ICL
to a great extent results from the gradual transposition on to the international level
of rules and legal constructs proper to national criminal law or to national trial pro­
ceedings. The grafting of municipal law notions and rules on to international law has
not, however, been a smooth process. National legal orders do not contain a uniform
regulation of criminal law. On the contrary, they are split into many different sys­
tems, from among which two principal ones emerge: that prevailing in common law
countries (the UK, the USA, Australia, Canada, many African and Asian countries),
and that obtaining in civil law countries, chiefly based on a legal system of Romano-
Germanic origin (they include states of continental Europe, such as France, Germany,
Italy, Belgium, the countries of N orthern Europe such as Norway, Sweden, Denmark,
as well as Latin American countries, many Arab countries, as well as Asian states
including, for instance, China). The heterogeneous and composite origin of many
international rules of both substantive and procedural crim inal law, a real patchwork
o f normative standards, complicates matters, as we shall see.3
It follows that ICL is an essentially hybrid branch of law: it is public international
law impregnated with notions, principles, and legal constructs derived from national
criminal law, IHL as well as hum an rights law. However, the recent establishment of
international criminal tribunals, and in particular of the ICC, has given a stupendous
impulse to the evolution of a corpus of international criminal rules proper. It can
therefore be safely maintained that we are now heading for the formation of a fully
fledged body of law in this area.

4. A further major feature of ICL, in particular of substantive criminal law, closely

bound up with the feature to which I have just drawn attention, ought to be emphasized.
This law has a twofold relationship with the general body of public international law.
The first relationship is one of mutual subsidiarity or support. Strikingly, most of the
offences that ICL proscribes and for the perpetration of which it endeavours to punish
the individuals that allegedly committed them, are also regarded by international law
as wrongful acts by states to the extent that they are large-scale and systematic: they
are international delinquencies entailing the aggravated responsibility’ of the state on
whose behalf the perpetrators may have acted.4 This holds true not only for genocide

3 As is already noted above, this applies in particular to the so-called general part of criminal law’; that
is the set of rules regulating the subjective elements of crimes, the various forms or categories of criminal
liability (for instance, joint responsibility for common criminal purpose, aiding and abetting, and so on),
conditions excluding criminal liability, etc. It was only natural for each national court pronouncing on war
crimes or crimes against humanity to apply the general notions of criminal law prevailing in that coun­
try. As a result, one is confronted with hundreds of national cases where judges have relied upon different
conceptions of, or approaches to, the ‘general part’, or have even resorted to the national definition of some
subjective or objective elements of the relevant international crime. For instance, in Fröhlich, a British Court
of Appeal (established in Germany under Control Council Law no. 10), to satisfy itself that the offence of the
accused (a German charged with, and convicted by a Court of first instance of, killing four Russian prisoners
of war) amounted to a war crime consisting of murder, applied the German notion o f‘murder’ (280-2).
4 On the notion of‘aggravated State responsibility’, see Cassese, op. cit., at 262-75.
8 IN T E R N A T IO N A L C R IM IN A L LAW

and crimes against humanity, but also for systematic torture, large-scale terrorism, and
massive war crimes. Thus, when one of these crimes is committed by an individual not
acting in a private capacity, a dual responsibility may follow: criminal liability of the
individual, falling under ICL, and state responsibility, regulated by international rules
on this matter.5Admittedly, there is at present a tendency in the international commu­
nity to give pride of place to the former category of responsibility whilst playing down the
latter. Political motivations underpin this trend, chiefly the inclination of states to avoid
invoking the aggravated responsibility of other states except when they are prompted to
do so out of self-interest or on strong political grounds. It is nevertheless a fact that, the­
oretically, both legal avenues remain open and may be utilized, as is shown by the pro­
ceedings for genocide recently instituted by some states before the International Court
of Justice6 while at the same time genocide trials are taking place before the ICTY.7
The second relationship between public international law and ICL is more com­
plex. Two somewhat conflicting philosophies underlie each area of law. ICL primarily
addresses the conduct of individuals and aims at protecting society against the most
harmful transgressions of legal standards perpetrated by them (whether they be state
agents or persons acting in a private capacity). It therefore aims to punish the authors
of those transgressions, while however safeguarding the rights of suspects or accused
persons from any arbitrary prosecution and punishment. It follows among other
things that one of the mainstays of ICL is the exigency that its prohibitions be as clear,
detailed, and specific as possible. This is required by a basic demand of modern legal
civilization: anybody, before engaging in a particular conduct, is entitled to be aware
of whether such conduct is criminally prohibited or instead allowed. Another, closely
linked, fundamental requirement is that no one should be punished for conduct that
was not considered as criminal at the time when it was taken. In short, any person
suspected or accused of a crime is entitled to a set of significant rights protecting him
from possible abuse by the prosecuting authorities.
Public international law, on the other hand, primarily regulates the behaviour
of states. It pursues, in essence, the purpose of reconciling as much as possible the
conflicting interests and concerns of sovereign entities (although in modern times
somehow it also takes into account the interests and exigencies of individuals and
non-state entities). True, part of general international law is concerned with both the
violations by states of the most fundamental legal standards and the ensuing state

5 It is notable that the four Geneva Conventions of 1949, while they institute a special legal regime for the
criminal repression of grave breaches’ of the Conventions, at the same time provide for the ‘state responsi­
bility’ of contracting Parties for the case of commission of such ‘grave breaches’. See, for instance, Articles
129-30 of the Third Convention (on Prisoners of War), concerning the penal sanctions for ‘grave breaches’
and Article 131 on state responsibility. (Under the latter provision, ‘No High Contracting Party shall be
allowed to absolve itself or any other High Contracting Party of any liability incurred by itself or by another
High Contracting Party in respect of breaches referred to in the preceding Article.’)
6 See the cases brought by Bosnia and Herzegovina and by Croatia against the Federal Republic of
Yugoslavia (Application o f the Convention on the Prevention and Punishment o f Genocide, Bosnia and
Herzegovina v. Serbia and Montenegro, and Croatia v. Yugoslavia).
7 See, for instance, the judgment in Krstic (TJ) as well as the indictments against Milosevic (of 8 October
2001 and 22 November 2001), and the revised indictment against Karadzic of 28 April 2000.
 F U N D A M E N T A L S OF I N T E R N A T IO N A L C R IM IN A L LAW 9

responsibility. This area of international law is, however, relatively less conspicuous
than the corresponding segment of ICL. In fact, the thrust of general international
law is legally to regulate and facilitate a m inim um of peaceful intercourse between
states, much more than calling to account states for their breaches of law. To put it
differently, the normative role of law is more im portant and effective than its repressive
function. W hat is even more im portant from our present viewpoint is that, in order to
take account of the conflicting interests and preoccupations of states, the law-making
process is often actuated by dint of gradual evolution of sweeping and often loose rules
through custom or even so-called ‘soft law’ (that is, standards and guidelines devoid
of legally binding force). Often even treaties lay down ambiguous, or at any rate not
well-determined provisions; this happens whenever the need to reconcile conflicting
state interests makes it necessary to agree upon vague formulas. In short, the need for
detailed, clear, and unambiguous legal regulation is less strong in the general area of
public international law than in the specific area of international criminal law, where
this need becomes of crucial relevance, given that the fundamental rights of suspects
or accused persons are at stake.
The inherent requirements underlying ICL (not less than any national body of
criminal law) may therefore collide with the traditional characteristics of pub­
lic international law. The tension between the different philosophy and approach
underlying each of these two bodies of law (public international law and crim inal law)
explains the unease with which national criminal lawyers look upon ICL. In particular,
those criminal lawyers that are conversant with the Romano-Germanic tradition and
live in civil law countries take issue with the loose character of many provisions of
ICL. Notably, they assail the fact that ICL relies to a large extent upon custom.
Be that as it may, what counts on the practical side is that, as a result of the contrast
between the relative indeterminacy and ‘malleability’ of international criminal rules
deriving from their largely customary nature, and the imperative requirement that
criminal rules be clear and specific, the role o f national or international courts is con­
spicuously crucial. It falls to courts, both national and international, to try to cast light
on, and give legal precision to, rules of customary nature, whenever their content and
purport is still surrounded by uncertainty, as well as to spell out and elaborate upon
the frequently terse content of treaty provisions. In particular, courts play an indis­
pensable role in (i) the ascertainment of the existence and contents of customary rules;
(ii) the interpretation and clarification of treaty provisions; and (iii) the elaboration,
based on general principles and rules, of legal constructs indispensable for the appli­
cation of international criminal rules. It is mainly due to judicial decisions that ICL is
progressing so rapidly.8

8 These characteristic features of this body of law have in some respects a negative connotation, while
other features may prove advantageous. The drawback is that the rights of the accused risk being jeopard­
ized by the normative flux that still characterizes this branch of international law. It is chiefly for courts to
endeavour as far as possible to safeguard the rights of the accused from any unwarranted deviation from the
fundamental principles of criminal law and human rights law. The advantage of the unique nature of ICL
 10 I N T E R N A T IO N A L C R IM IN A L LAW

5. Closely bound up with the characteristic just underlined is another major trait
of current ICL. More than other branches of public international law, but like those
legal areas where rapid changes in technology impose speedy normative updating (for
instance, the law of the environment or the law of international trade) ICL is changing
very quickly. This is because unfortunately, in the world community there is a stagger­
ing increase in atrocities, whether or not linked to armed conflict. There is, therefore,
a widely felt need to respond to them by, among other things, criminal repression.
However, what is even more striking in this branch of law is that legal change (i) goes
hand in hand with increasing sophistication of the legal system (we are now moving
from a rudim entary jumble of rules and principles to a fairly consistent body of law);
and (ii) is accompanied by a gradual shift in its philosophical underpinning: in par­
ticular, a shift from the doctrine of substantive justice (whereby the need to protect
society requires the punishm ent of harm ful actions even if such actions had not been
previously criminalized) to that of strict legality (whereby the need to protect indi­
viduals’ hum an rights, in particular to safeguard individuals from arbitrary action of
the executive or judicial powers, requires that no one may be punished for any action
not considered criminal when performed). On this matter see infra, 2.3.
6. Finally, let me stress a significant characteristic of ICL, which, however, is not
unique to it. Like most national legal systems, international rules criminalize not only
conduct causing harm (for example, murder, rape, torture, shelling of innocent civil­
ians) but also conduct creating an unacceptable risk of harm (for.example, conspiracy
to commit genocide, not followed by genocidal acts). The rationale behind this legal
regulation is that—as in this area crim inal conduct is normally of great magnitude
and seriously offends against fundamental values—international hum anitarian and
criminal rules also aim at criminalizing any actions that may carry a serious risk of
causing grave harm. In other words, those rules also play &preventative role. This fea­
ture of ICL manifests itself in three major ways: (i) by criminalizing the early stage or
the preparation of crimes that are then committed; (ii) by the prohibition of so-called
inchoate crimes (or prelim inary criminal offences); and (iii) by the prohibition of spe­
cific conduct likely to cause serious risk.9

is that change and adaptation to evolving historical circumstances occur more easily and smoothly than in
legal systems based on codes and other forms of written law. In this respect, courts may become instrumen­
tal in reconciling the demands for change with the requirement of respect for the rights of the accused.
9 As for the first aspect, suffice it to stress that ICL among other things prohibits planning. As for incho­
ate crimes, it may be sufficient to recall that international rules criminalize attempt and (in the case of at
least the most serious crime, genocide), conspiracy and incitement (see infra, 10. 3 - 8). All these inchoate
offences that constitute the preparatory stage of other offences may be punished even if the crime they are
intended to bring about does not in fact occur. Criminalization of these offences is a way of preventing
them from occurring, to the extent of preventing the perpetration of the crime to which they intend to lead.
By criminalizing such conduct, international rules endeavour to forestall the danger that the execution
of those offences may cause major harm. They also serve to stigmatize attempting, inciting, or conspir­
ing as criminal in itself. Thus the message is conveyed not only that people should not commit crimes but
also not incite, conspire, or attempt such crimes; if they do so, they will be labelled as criminals and pun­
ished accordingly, (for this reason, some have criticized these crimes, especially conspiracy, as ‘thought
 F U N D A M E N T A L S OF IN T E R N A T IO N A L C R I M I N A L LAW 11

1.3 THE N O TION OF IN TER NATIONA L CRIMES

International crimes are breaches of international rules entailing the personal crim ­
inal liability of the individuals concerned (as opposed to the responsibility of the state
of which the individuals may act as organs).
Before considering the various categories of such crimes, it should be specified that
international crimes result from the cumulative presence of the following elements:

1. They consist of violations of international customary rules (as well as treaty pro­
visions, where such provisions exist and either codify or spell out customary law or
have contributed to its formation).
2. Such rules are intended to protect values considered im portant by the whole
international community and consequently binding all states and individuals. The
values at issue are not propounded by scholars or thought up by starry-eyed philoso­
phers. Rather, they are laid down in a string of international instruments, which, how­
ever, do not necessarily spell them out in so many words.10
3. There exists a universal interest in repressing these crimes. Subject to certain con­
ditions, under international law their alleged authors may in principle be prosecuted

crimes’, but this is inaccurate as each offence requires some overt conduct in addition to the mens rea
requirement.)
With regard to the third of the elements referred to, it may be pointed out that criminalization of risk
occurs any time a criminal rule envisages, among the possible subjective elements of criminal conduct, reck­
lessness or dolus eventualis (see infra, 3.7). Such criminalization may also specifically derive from the spe­
cific content of individual provisions. For instance, Article 7 of the Geneva Convention of 1929 on Prisoners
of War provided, among other things, that ‘As soon as possible after their capture, prisoners of war shall be
evacuated to depots sufficiently removed from the fighting zone for them to be out of danger [...] Prisoners
shall not be unnecessarily exposed to danger while awaiting evacuation from a fighting zone.’ (At present a
rule corresponding to that provision is Article 23 of the Third Geneva Convention of 1949.)
In 1947 a Dutch Court Martial in Indonesia applied this provision in Koshiro. The accused, an officer in
the Japanese Navy in charge of Japanese forces at Makassar in the Netherlands East Indies, was charged,
among other things, with unnecessarily exposing a large number of Allied prisoners of war to danger, in
that in 1944 a large ammunition depot had been built by the prisoners of war at a distance of about 50 yards
from the prisoner of war camp, and stocked with ammunition (the air-raid shelters constructed in the camp
were inadequate). The Court Martial found the accused guilty. The district in which the camp and the depot
were situated was the immediate target for Allied planes several times, and as a result ‘the ammunition depot
might have been hit, with disastrous consequences for the prisoners’ (211).
10 They include the 1945 UN Charter, the 1948 Universal Declaration of Human Rights, the 1950 Euro­
pean Convention on Human Rights, the two 1966 UN Covenants on Human rights, the 1969 American
Convention on Human Rights, the UN Declaration on Friendly Relations of 1970, and the 1981 African
Charter on Human and Peoples’ Rights.
Other treaties also enshrine those values, although from another perspective: they do not proclaim the
values directly, but prohibit conduct that infringes them: for instance, the 1948 Convention on Genocide,
the 1949 Conventions on the protection of victims of armed conflict, and the two Additional Protocols
of 1977, the 1984 Convention against Torture, and the various treaties providing for the prosecution and
repression of specific forms of terrorism.
 12 IN T E R N A T IO N A L C R IM IN A L LAW

and punished by any state, regardless of any territorial or nationality link with the
perpetrator or the victim.
4. Finally, if the perpetrator has acted in an official capacity, i.e. as a de jure or de
facto state official, the state on whose behalf he has performed the prohibited act is
barred from claiming enjoyment of the im munity from the civil or criminal jurisdic­
tion of foreign states accruing under customary law to state officials acting in the exer­
cise of their functions (although, if the state official belongs to one of three categories,
namely head of state, foreign minister, or diplomatic agent, and is still serving, then he
enjoys complete personal im m unity as long as he or she is in office: see Pinochet,u Fidel
Castro (Legal Grounds 1-4), and the Congo v. Belgium case, §§57-61).

Under this definition international crimes include war crimes, crimes against hum an­
ity, genocide, torture (as distinct from torture as one of the categories of war crimes
or crimes against humanity), aggression, and some extreme forms of international
terrorism. By contrast, the notion at issue does not embrace other classes.
First of all, it does not encompass piracy (a phenomenon that was im portant and
conspicuous during the seventeenth to the nineteenth centuries). Indeed, as I have
tried to show elsewhere,112piracy was (and is) not punished for the sake of protecting a
community value: all states were (and still are) authorized to capture on the high seas
and bring to trial pirates in order to safeguard their joint interest to fight a common
danger and a consequent (real or potential) damage. This is to some extent supported
by the fact that when piracy was committed on behalf o f a state (and was then called
‘privateering’), there was no universal jurisdiction over it. This shows that the objective
conduct amounting to piracy—identical to the conduct amounting to ‘privateering’—
was not considered so abhorrent as to amount to an international crime. After all,
piracy could be just a simple m atter of theft on the high seas, although it more usually
involved more nasty conduct, such as making sailors walk the plank, murdering or
raping passengers of the ship attacked, or mutilating members of the crew. Probably
it was simply because piracy by definition occurred outside any state’s territorial jur­
isdiction that a useful repressive mechanism evolved allowing all or any state to bring
pirates to justice.
Secondly, the notion of international crimes does not include (a) illicit traf­
fic in narcotic drugs and psychotropic substances; (b) unlawful arms trade; (c) the
smuggling of nuclear and other potentially deadly materials; (d) money laundering;
(e) slave trade; or (f) traffic in women. For one thing, this broad range of crimes is
only provided for in international treaties or resolutions of international organiza­
tions, not in customary law. For another, these offences are normally perpetrated by
private individuals or crim inal organizations; states usually fight against them, often

11 Pinochet (House of Lords, judgment of 24 March 1999), speeches of Lord Browne-Wilkinson (at
112-15), Lord Hope of Craighead (at 145-52), Lord Saville of Newdigate (at 169-70), Lord Millet (171-91),
and of Lord Phillips of Worth Matravers (at 181-90).
12 See Cassese, op. cit., at 15,143-4, as well as my paper on ‘When may Senior State Officials be Tried for
International Crimes? Some comments on the Congo v. Belgium Case’, 13 EJIL (2002), at 857-8.
 F U N D A M E N T A L S OF IN T E R N A T IO N A L C R IM IN A L LAW 13

by joint official action. In other words, as a rule these offences are committed against
states. They do not involve states as such or, if they involve state agents, these agents
typically act for private gain, perpetrating what national legislation normally regards
as ordinary crimes.
Nor does the list of international crimes include apartheid, provided for in a
Convention of 1973 (which entered into force in 1976). It would seem that this offence
has not yet reached the status of a customary law crime, probably because it was held
to be limited in time and space. Moreover, the 101 states party to the Convention do
not include any western country: only two major segments of the international com­
m unity (developing and eastern European countries) have agreed to label apartheid as
an international crime, whereas another grouping, that of western states, has refused
to do so. There is therefore a case for m aintaining that under customary international
law apartheid, although probably prohibited as a state delinquency, is not, however,
regarded as a crime entailing the criminal liability of individuals. Nevertheless, the
fact that Article 7(l)(j) of the Statute of the ICC grants the Court jurisdiction over
apartheid and Article 7(2)(h) provides a definition of this crime, might gradually
facilitate the formation of a customary rule. This development could occur if and when
cases concerning ‘inhum ane acts’ committed in the context of an institutionalized
regime of systematic oppression and domination by one racial group over any other
racial group or groups and committed with the intention of m aintaining that regime’
are ever brought before the Court.

1.4 SOURCES OF ICL

W hat are the law-making processes from which one can draw the rules m aking up
ICL to be applied by international criminal courts? An attempt will be made here to
answer this question. However, two prelim inary issues need to be clarified.
First, there is the problem of the extent to which the same sources may be used
by national courts and within what constraints must be left open. In many respects
each national legal system provides for its own mechanism for the implementation of
international rules. In particular, each system lays down the conditions under which
international rules of criminal law may be applied. For instance, in many states, in
order for courts to be authorized to pronounce on international crimes, it is necessary
for the legislature to have passed the appropriate legislation (a) defining the crimes;
(b) providing for the relevant penalties; and (c) granting courts jurisdiction over those
crimes. In consequence, the system of sources utilized by national courts for the pur­
pose of trying persons accused of international crimes is to a large extent bound up
with the general m anner in which the national system puts international rules into
effect at the domestic level. That is not to say that sources of ICL vary from state to
state; it is simply to say that the way national courts apply this body of law may vary.
For instance, courts of all states may and do apply both treaties and international
H IN T E R N A T IO N A L C R IM IN A L LAW

customary law as well as general principles of international law. Nonetheless, depend­

ing on the rank of each category of international rules within the national legal system
and their status vis-ä-vis national legislation, treaty rules may prevail over, or be pre­
vailed over by, national laws.
Secondly, many criminal lawyers, particularly in countries of Romano-Germanic
tradition, being used to interpreting and applying criminal rules laid down in written
codes of criminal law, tend to believe that the major source of ICL can be found in the
Statute of the ICC or at least that that Statute is a sort of code of international crim ­
inal law’. This is a wrong assumption, partly based on the fact that admittedly that
Statute is the only international written instrum ent laying down international rules
on both the ‘general part’ of ICL and the definition of crimes. The truth of the matter
is, however, that that Statute makes up a set of rules only applicable by the ICC: the
Statute does not apply to other international criminal tribunals (the ICTY, the ICTR,
the SCSL, the STL, and so on), each of which is regulated by, and must apply, its own
Statute or Charter. It is the ICC that must comply with the provisions defining the
various crimes under the C ourt’s jurisdiction and also apply the Statute’s provisions
on mens rea, defences, etc. In other words, the Statute, far from constituting an ‘inter­
national criminal code’, only lays down the rules that must guide the Court when it
exercises its jurisdiction over the crimes it is called upon to adjudicate. This conclusion
does not of course detract from the importance of the ICC Statute as a set of rules that
clarify many points in ICL and which, in this respect, may also prove useful to con­
sider by other courts. Thus, some provisions of the Statute may be held to codify cus­
tom ary international law (this, for instance, applies to some provisions on war crimes,
or on crimes against humanity; see infra, 4.6 and 5.7); others may be deemed to lay
down a rule that clearly chooses between two conflicting interpretations previously
offered in international case law (this, for instance, applies to the provision on duress,
namely Article 31(l)(d)). Others instead go beyond what is prescribed by customary
international law (see 4.6 or 5.7.2-3 or 12.3.2). Furthermore, it should not be ruled out
that, particularly after the ICC begins its judicial activity proper, some of the Statute’s
provisions may gradually turn into customary international law as a result of other
international courts and tribunals broadly accepting and applying these provisions as
encapsulating the world community’s opinio juris on the matter.
Let us now turn to the question at issue. Since ICL is but a branch of public inter­
national law, the sources of law from which one may derive the relevant rules (i) are
those proper to international law, and (ii) must be resorted to in the hierarchical order
dictated by international law.
Hence, one may draw upon primary sources (treaties, customary law), secondary
sources (law-making processes envisaged by customary rules or treaty provisions,
such as SC resolutions, when an international criminal tribunal has been established
by a SC resolution), general principles of ICL or general principles of law, or in the
final analysis such subsidiary sources as general principles of law recognized by the
community of states.
 F U N D A M E N T A L S OF I N T E R N A T IO N A L C R IM IN A L LAW 15

The order in which one may use such sources may be derived from the structure and
hierarchy of the sources of international law. Such order (which at present is codified
to a large extent in Article 21(1) of the ICC Statute),13 is as follows.
One should first of all look for treaty rules or for rules laid down in such inter­
national instruments as binding resolutions of the UN SC (as is the case with the ICTY
and the ICTR), when these treaty rules or resolutions contain the provisions conferring
jurisdiction on the court and setting out the procedure. W hen such rules are lacking
or contain gaps, one should resort to customary law or to treaties implicitly or expli­
citly referred to in the aforementioned rules. W hen even this set of general or treaty
rules is of no avail, one should apply general principles of ICL, such as for example the
principle of non-retroactivity of criminal law (see infra, 2.4.2) or the principle of com­
m and responsibility (see infra, 11.4). These principles can be inferred, by a process of
induction and generalization, from treaty provisions or customary rules. W hen even
these principles do not prove helpful, one could rely, as a fallback, on general principles
of law (such as the principle of respect for hum an rights). If one still does not find the
applicable rule or, more often, if the rule contains a gap or is at any rate insufficient,
one may have resort to general principles of criminal law common to the nations of
the world (such as the ban on denial of justice, the doctrine of res judicata, i.e. of the
binding force of a judicial decision; see infra, 1.4.5).
Let us now consider these various sources in some detail.

1.4.1 THE STATUTES OF COURTS AND TRIBUNALS

Chief among these texts are the London Agreement of 8 August 1945, setting out the
substantive and procedural law of the IMT of Nuremberg, and the 1998 Statute of the
ICC, a long and elaborate instrum ent that lays down both a list of crimes subject to the
jurisdiction of the Court and some general principles of ICL, and in addition sets forth
the main rules on the proceedings before the Court. Of considerable importance are
also the Statutes of the SCSL, laid down in an Annex to the Agreement between the UN
and Sierra Leone of 16 January 2002, and of the STL, enshrined in an Agreement between
the UN and Lebanon of 10 June 2007 and endorsed in SC resolution 1757 (2007).
Other international instrum ents endowed with legally binding force and regulat­
ing international tribunals are the resolutions passed in 1993 and 1994, respectively
by the UN SC to adopt the Statutes of the ICTY and the ICTR. These resolutions,
taken on the strength of Chapter VII of the UN Charter, are legally binding on all UN
member states pursuant to Article 25 of the UN Charter. They constitute ‘secondary’

13 T. The Court shall apply:

(a) In the first place, this Statute, Elements of Crimes and its Rules of Procedure and Evidence;
(b) In the second place, where appropriate, applicable treaties and the principles and rules of inter­
national law, including the established principles of the international law of armed conflict;
(c) Failing that, general principles of law derived by the Court from national laws of legal systems of
the world including, where appropriate, the national laws of States that would normally exercise
jurisdiction over the crime, provided that those principles are not inconsistent with this Statute
and with international law and internationally recognized norms and standards.’
 16 INTERNATIONAL CRIMINAL LAW

international legislation (in that they were adopted by virtue of provisions contained
in a treaty, the UN Charter).
For the interpretation of these instrum ents one must rely upon the rules of
interpretation laid down in the 1969 Vienna Convention on the Law of Treaties. Indeed,
in many respects these resolutions, and their annexed Statutes, may be equated with
international treaties. The ICTY AC upheld this view in a num ber of decisions.14

1.4.2 O T H E R T R E A TIES

Often provisions of the Statutes of courts and tribunals refer, if only implicitly, to
international treaties. For instance, Article 2 of the ICTY Statute, conferring on the
Tribunal jurisdiction over grave breaches of the Geneva Conventions of 1949, expli­
citly refers to these Geneva Conventions with regard to the notion of protected per­
sons’ and ‘protected property’. Article 4 of the ICTR Statute, granting jurisdiction over
violations of Article 3 (common to the four 1949 Geneva Conventions) and the Second
Additional Protocol, admittedly incorporates in terms only the main provisions of com­
mon Article 3 and the Additional Protocol; nevertheless, for its interpretation the Tribu­
nal may need to look at all the others provisions of the Conventions or the Protocol.
International treaties may come into play from another viewpoint. By definition
treaties are only binding upon the contracting states and any international body they
may establish. Nonetheless, they may also be taken into account, whenever this is
legally admissible, as evidence of the crystallization of customary rules.
O f course, given the overriding importance of the nullum crimen principle (see infra,
2.3), an international court is not allowed to apply treaties other than those conferring
on it jurisdiction over certain categories of crimes, if such treaties provide for other cat­
egories of crime. For instance, if the Statute of a court or tribunal grants jurisdiction over
crimes against humanity and genocide only, the court or tribunal may not have recourse
to a treaty prohibiting war crimes, and try an accused for such class of crimes.
Treaties relevant to our subject matter are those laying down substantive rules of
International Hum anitarian Law (IHL) (for instance, the Regulations annexed to
the Fourth Hague Convention of 1907, the four Geneva Conventions of 1949, the two
Geneva Additional Protocols of 1977, various recent treaties prohibiting the use of
certain specific weapons,15 and so on); that is, rules the serious violation of which
may amount to war crimes (or, in the case of the Geneva Conventions and the First
Additional Protocol, to ‘grave breaches’ of these Conventions or Protocol). Other

14 See, for instance, Tadic (Interlocutory Appeal) (§§71-93), as well as Tadic (Appeal) (§§282-6 and
287-305). An ICTY Trial Chamber rightly held in Slobodan Milosevic (decision on preliminary motions) that
‘the Statute of the International Tribunal is interpreted as a treaty’ (§47).
15 See, for instance, the 1925 Geneva Protocol for the Prohibition of the Use in War of Asphyxiating,
Poisonous or Other Gases, and of Bacteriological Methods of Warfare, or the 1980 UN Convention on
Prohibitions on Restrictions on the Use of Certain Conventional Weapons which may be deemed to be
excessively injurious or to have Indiscriminate Effects, or the 1997 Ottawa Convention on the Prohibition of
the Use, Stockpiling, Production and Transfer of Anti-personnel Mines and their Destruction.
 FUNDAMENTALS OF INTERNATIONAL CRIMINAL LAW 17

treaties refer to other international crimes: for instance, the 1948 Convention on
Genocide (the fundamental provisions of which have subsequently turned into cus­
tomary law); the 1984 Convention against Torture, various international treaties on
terrorism, etc.
The rules for interpreting treaties are those laid down in Articles 31-3 of the 1969
Vienna Convention on the Law of Treaties, which is declaratory of customary inter­
national law on the construction of both treaties and, arguably, other written rules
as well.

1.4.3 C USTOM ARY LAW

As pointed out above, written rules on our subject matter (belonging either to treaties
or to other international instrum ents endowed with normative force, such as binding
resolutions of the UN SC) are not numerous. In addition, when they exist, as is the case
with the ICC Statute, formally speaking they are only applicable to the court to which
they grant jurisdiction; for instance, ICC Statute provisions could not be binding on
the ICTY, unless they codify customary rules or contribute to the formation of such
rules, in other words, unless they acquire the value of universal norms.
Hence, one has frequently to rely upon customary rules or general principles either
to clarify the content of treaty provisions or to fill gaps in these provisions. Resort
to customary law may also prove necessary for the purpose of pinpointing general
principles of criminal law, whenever the application of such principles becomes
necessary (see below).
On this score ICL bears a strong resemblance to the crim inal law of such common
law countries as England, where next to statutory offences there exist many common
law offences, developed through judicial precedents. However, the deficiency deriving
from the unwritten nature of customary law is less conspicuous than in ICL. The exist­
ence of a huge wealth of judicial precedents built up over centuries, the hierarchical
structure of the judiciary coupled with the doctrine of ‘judicial precedent’ (whereby
each court is bound by the decisions of higher courts), as well as the extrapolation by legal
scholars of general principles from that copious case law, tend to meet the exigencies of
legal certainty and foreseeability proper to any system of criminal law. In contrast, ICL
is still in its infancy, or at least adolescence: consequently, many of its rules still suffer
from their loose content, contrary to the principle of specificity proper to criminal law
(see infra, 2.4.1). As noted above (1.2), the role of international, as well as national, courts
thus becomes crucial for the building of a less rudimentary corpus of legal rules.
W hat has been said above, with regard to treaties and the nullum crimen principle,
also holds true for customary law. A court or tribunal may not apply a customary rule
criminalizing conduct that does not fall within one of the categories of crimes over
which it has jurisdiction under its Statute.
As noted above, both customary rules and principles may normally be drawn or
inferred from judicial decisions, which to a very large have been handed down, chiefly
in the past, by national courts (whereas by now there exists a conspicuous num ber of
 i8 INTERNATIONAL CRIMINAL LAW

judgments delivered by international courts and tribunals). As each state court tends
to apply the general notions of national criminal law even when adjudicating inter­
national crimes, it often proves arduous to find views and concepts that are so uniform
and consistent as to evidence the formation of an international customary rule. The
same holds true for principles.
In addition, differences originating from varying legal approaches may influence the
appraisal by an international judge of the significance of case law. Judges trained in
common law systems naturally tend to attach great importance to cases as precedents’
and are inclined to apply such ‘precedents’ without asking themselves whether they
evince the formation of, or crystallize, an international customary rule, or instead
testify to the proper interpretation of a treaty or customary rule offered by another
court. In contrast, judges from civil law countries, where judicial precedents have less
weight and criminal codes enjoy a decisive legal status, tend to play down judicial
decisions, or at least to first ask themselves, before relying upon such decisions, what
legal status should be attached to them in international proceedings. This difference in
cultural background and legal training of international judges often leads to different
legal decisions.
Many examples may be cited of instances where national or international courts
have taken into consideration case law (plus, if need be, treaties and other international
instruments) to establish whether a customary rule had evolved on a specific matter.
For instance, in Furundzija an ICTY TC held that a rule on the definition of rape had
come into being at the customary law level.16In a case decided in 1950 a Brussels Court
M artial had already ruled that torture in time of armed conflict was prohibited by a
customary international law rule.17

16 After noting that rape was prohibited in treaty law, it pronounced as follows: ‘The prohibition of rape
and serious sexual assault in armed conflict has also evolved in customary international law. It has gradually
crystallised out of the express prohibition of rape in Article 44 of the [1863] Lieber Code and the general
provisions contained in Article 46 of the Regulations annexed to Hague Convention IV, read in conjunction
with the “Martens clause” laid down in the preamble to the Convention. While rape and sexual assaults
were not specifically prosecuted by the Nuremberg Tribunal, rape was expressly classified as a crime against
humanity under Article II(l)(c) of Control Council Law no. 10. The Tokyo International Military Tribunal
convicted Generals Toyoda and Matsui of command responsibility for violations of the laws or customs
of war committed by their soldiers in Nanking, which included widespread rapes and sexual assaults. The
former Foreign Minister of Japan, Hirota, was also convicted for these atrocities. This decision and that of
the United States Military Commission in Yamashita, along with the ripening of the fundamental prohib­
ition of “outrages upon personal dignity” laid down in common Article 3 into customary international law,
has contributed to the evolution of universally accepted norms of international law prohibiting rape as well
as serious sexual assault. These norms are applicable in any armed conflict. It is indisputable that rape and
other serious sexual assaults in armed conflict entail the criminal liability of the perpetrator. (§§168-9).
17 In K.W. German officers had been accused of ill-treating civilians in occupied Belgium. After noting
that Article 46 of the Hague Regulations imposed upon the Occupant to respect the life of individuals but
did not expressly forbid acts of violence or cruelty, the Court Martial held that a customary rule had evolved
on the matter. To this effect it relied upon the celebrated Martens Clause as well as Article 5 of the Universal
Declaration of Human Rights, concluding that ‘hanging a human being by his hands tied behind his back
from a pulley specially rigged for the purpose’ was torture, whereas ‘blows to the face, delivered so repeat­
edly and violently that they caused it to swell up and, in several cases, broke some teeth’ amounted to cruel
treatment (at 566). See also Auditeurv. K. (at 654).
 FUNDAMENTALS OF INTERNATIONAL CRIMINAL LAW 19

In many cases courts have resorted to customary law to determine the content and
scope of an international rule that made a crime punishable without, however, prop­
erly defining the prohibited conduct. For instance, in Kupreskic and others an ICTY
TC had carefully to consider treaties and case law to establish what was meant by
persecution as a crime against hum anity (§§567-626).18 In Tadic (AJ, 1999) the ICTY
AC had to find whether the doctrine of acting in pursuance of a common criminal
purpose (or a joint criminal enterprise) covered the case where one of the perpetrators
committed an act that, while outside the common design, was nevertheless a foresee­
able consequence of pursuing that common purpose or design. After considering vari­
ous national cases and two international treaties, as well as the legislation of a number
of civil law and common law countries, the C ourt gave an affirmative answer. It noted
that since there was no uniformity in the national legislation of the major legal systems
of the world (§§204-25), the Chamber could not consider that a general rule had been
generated by the general principles of criminal law recognized by the nations of the
world (§225). Rather, the law on the matter was customary in nature
the consistency and cogency of the case law and the treaties referred to above, as well as their
consonance with the general principles on criminal responsibility laid down in the [ICTY]
Statute and general international criminal law and in national legislation, warrant the con­
clusion that the case law reflects customary rules of international criminal law. (§226).

In the same case the Chamber upheld the Prosecutor’s submissions that the ICTY
Statute did not provide that crimes against hum anity could not be committed for
purely personal motives. The Court undertook a careful examination of case law as
evidence of customary international law’ (§§248-69) and concluded that:
the relevant case law and the spirit of international rules concerning crimes against human­
ity make it clear that under customary law, ‘purely personal motives’ do not acquire any
relevance for establishing whether or not a crime against humanity has been perpetrated
(§270).19

18 The Court found that under customary law persecution must contain the following elements: (i) the
elements required for all crimes against humanity under the ICTY Statute (namely, to be part of a wide­
spread or systematic attack on the civilian population, etc.); (ii) tobeagrossor blatant denial of a fundamen­
tal right reaching the same level of gravity as the other acts prohibited under Article 5 of the ICTY Statute (on
crimes against humanity); and (iii) to be based on discriminatory grounds (§627). Similarly, in Kunarac and
others an ICTY TC held that ‘at the time relevant to the indictment’, enslavement as a crime against human­
ity was prohibited by customary international law ‘as the exercise of any or all of the powers attaching to the
right of ownership over a person’ (§539). It reached this conclusion after along survey of treaties and national
and international cases (§§518-38).
19 In other words, the AC held for the Prosecution on the Prosecutor’s Cross-Appeal that a crime against
humanity could he committed for purely personal motives, since whether the crime is committed for purely
personal reasons or not is irrelevant.
In Slobodan Milosevic (Decision on Preliminary Motions) an ICTY TC concluded that the provision of
the ICTY Statute whereby the ‘official position’ of an accused does not relieve him of criminal responsibility
reflected customary international law, as evidenced by numerous treaty provisions on the matter, the adop­
tion by a very large majority of the ICC Statute at the Rome Diplomatic Conference, the adoption by the ILC
of the Draft Code of Crimes against the Peace and Security of Mankind, as well as by case law (§§26-33).
 20 INTERNATIONAL CRIMINAL LAW

In some cases courts reached the conclusion that, contrary to the submissions of
one of the parties, a specific matter was not governed by customary international
rules. Thus, for instance, in Tadic (AJ 1999) the ICTY AC held that
customary international law, as it results from the gradual development of international
instruments and national case law into general rules, does not presuppose a discriminatory
or persecutory intent for all crimes against humanity. (§§288-92).

Conversely, as pointed out above, in some cases international or national courts,

following an approach akin to that of common law courts, did not take into consid­
eration case law for the purpose of determining whether it had brought about the
crystallization of an international customary rule. Rather, they viewed and used case
law as a set of precedents that could be of assistance in establishing the applicable law.
(One should note, however, that on a typical common law approach, precedents are
binding, not merely of assistance. Obiter dicta are of assistance, but by definition they
are not precedents.)20

1.4.4 G E N E R A L P R I N C I P L E S OF IC L A N D G E N E R A L
P R I N C I P L E S O F IL

General principles of ICL include principles specific to criminal law, such as the prin­
ciples of legality21 (see infra, 2.3), of specificity (see infra, 2.4.1), the presumption of
innocence (see infra, 18.3), the principle of equality of arms (see infra, 18.4.1), the
principle of command responsibility (see 11.4), a corollary in ICL of the principle of
responsible command existing in IHL,22 etc.23 The application of these principles at
the international level normally results from their gradual transposition over time

For instance, in Kvocka and others an ICTY TC, when discussing the issue of how to distinguish
co-perpetrators from aiders and abettors in the case of participation by a number of persons in a joint crim­
inal enterprise, merely relied upon case law as such (‘A number of cases assist the Trial Chamber in its
assessment of the level of participation required to incur criminal responsibility as either a co-perpetrator
or an aider and abettor in a criminal endeavour in which several participants are involved’: §290; and see §§
291-312). Perhaps, the Chamber was trying to discover the content of international customary law but did
not say in so many words that that was what it was doing.
On this principle see among othr things ICTY AC, Hadzihasanovic, Alagic and Kubura, Decision on
Interlocutory Appeal Challenging Jurisdiction in Relation to Command Responsibility, §§ 32-6.
On these principles see ICTY AC, Hadzihasanovic, Alagic and Kubura, Decision on Interlocutory
Appeal Challenging Jurisdiction in Relation to Command Responsibility, §§14-18.
TC II of the ICTY, in Delalic et al. in 1998 mentioned the nullum crimen sine lege and the nulla poena
sine lege principles, noting that they are well recognised in the world’s major criminal justice systems as
being fundamental principles of criminality’ (§402). The Chamber also referred to another ‘fundamen­
tal principle’, namely ‘the prohibition against ex post facto criminal laws with its derivative rule of non­
retroactive application of criminal laws and criminal sanctions’, as well as ‘the requirement of specificity
and the prohibition of ambiguity in criminal legislation’ (ibid.). The Chamber then pointed out that: ‘the
above pi inciples of legality exist and are recognised in all the world’s major criminal justice systems’ (§403).
However, the Chamber warned, ‘[i]t is not certain to what extent they have been admitted as part of inter­
national legal practice, separate and apart from the existence of the national legal systems. This is essentially
because of the ditferent methods of criminalisation of conduct in national and international criminal justice
systems’ (§403).
 FUNDAMENTALS OF INTERNATIONAL CRIMINAL LAW 21

from national legal systems on to the international order. They are now firmly embed­
ded in ICL.
General principles of international law (IL) consist of principles inherent in the inter­
national legal system. Hence, their identification does not require an in-depth compara­
tive survey of all the major legal systems of the world, but can be carried out by way of
generalization and induction from the main features of the international legal order.
As noted above, resort to general principles of ICL may be had when treaty or cus­
tomary rules (or else legal provisions produced through a secondary source, such as
the ICTY or the ICTR Statutes) are unclear or incomplete. If even these principles
prove of no avail, one can then draw upon general principles of IL, if any.
From the point of view of legal logic it is only after looking for the existence of a
principle belonging to one of the two categories just mentioned, that a court may then
turn to general principles of criminal law recognized by the com m unity of nations (on
which see infra, 1.4.5). This is because this last category is a subsidiary source of law, as
was pointed out above (1.4), whereas general principles of ICL, as well as general prin
ciples of international law, derive from two primary sources of law, namely custom and
treaty. In practice, however, international courts, once they find that no general rule
exists on a specific issue, turn immediately to the subsidiary source and try to ascer­
tain whether a general principle of criminal law common to all the countries of the
world has evolved. The reasons behind this approach are clear: the two categories oi
principles we are discussing in this section are very general; they may therefore prove
of scant assistance in the search for a legal regulation of a specific issue. In contrast,
the exploration of the principal criminal systems of the world is more likely to provide
a normative standard applicable to the case at issue.
By way of illustration of the various modalities of resort to principles, one can m en­
tion Furundzija. In that case, the TC first surveyed international treaties and case law
to establish whether there existed any rule of customary international law defining
rape; having reached a negative conclusion, the Chamber embarked upon an examin­
ation of national legislation in order to identify a possible common definition ofthat
offence. It concluded that such a common definition did exist, except for one point
(whether or not the forced sexual penetration of the mouth by the male sexual organ
amounted to rape), on which a major discrepancy in the various legal systems could
be discerned. The Tribunal held that at this stage it was appropriate to look for general
principles of ICL or, if such principles are of no avail, to the general principles of inter­
national law’ (§182). It then applied the general principle of respect for hum an dignity’
both as a principle underpinning IHL and hum an rights law, and as a principle per­
meating the whole body of international law (§183). It also applied the general prin ­
ciple nullum crimen sine lege (§184), probably as a general principle of criminal law.
Arguably a more compelling approach was taken in Kupreskic and others.2A24

24 In that case TC II held that: ‘[A]ny time the Statute [of the ICTY] does not regulate a specific matter,
and the Report of the Secretary-General [submitted to the SC and endorsed by it as a document accompany­
ing the resolution establishing the Tribunal] does not prove to be of any assistance in the interpretation of
the Statute, it falls to the International Tribunal to draw upon (i) rules of customary international law or
 22 INTERNATIONAL CRIMINAL LAW

1.4.5 G E N E R A L P R I N C IP L E S OF C R I M I N A L L A W R E C O G N I Z E D BY
T H E C O M M U N IT Y OF N A T IO N S

While the general principles just mentioned may be inferred from the whole system
of ICL or of international law, the principles we will now discuss may be drawn from
a comparative survey of the principal legal systems of the world. Their articulation is
therefore grounded not merely on interpretation and generalization, but rather on a
comparative law approach.
As stressed above, this source is subsidiary in nature; hence, recourse to it can only
be made if reliance upon the other sources (treaties, custom, general principles of ICL,
general principle of international law, rules produced through a secondary source
such as the provisions of the ICTY or ICTR Statutes) has turned out to be of no avail. It
is at this stage that the search for general principles shared by the major legal systems
of the community of nations may be initiated. This is precisely the approach taken in
Article 21(l)(c) of the ICC Statute.25 Pursuant to this provision, resort to the general
principles under discussion is the extrema ratio for the ICC.
Clearly, a principle of criminal law may belong to this class only if a court finds that
it is shared by common law and civil law systems as well as other legal systems such
as those of the Islamic world, some Asian countries such as China and Japan, and the

(ii) general principles of ICL; or, lacking such principles, (iii) general principles of criminal law common to
the major legal systems of the world; or, lacking such principles, (iv) general principles of law consonant with
the basic requirements of international justice’ (§591).
The Tribunal applied general criteria, when dealing with the question of determining how a double con­
viction for a single criminal action should be reflected in sentencing. After finding that no general prin­
ciple could be garnered from the various legal systems, the Tribunal stated the following: ‘Faced with this
discrepancy in municipal legal systems, the TC considers that a fair solution can be derived both from the
object and purpose of the provisions of the Statute as well as the general concepts underlying the Statute and
from the general principles of justice applied by jurists and practised by military courts” referred to by the
International Military Tribunal at Nuremberg’ (§717).
The TC came back to the same problem when it dealt with the issue of how a TC should act in the case of an
erroneous legal classification of facts by the Prosecutor. It carefully examined various legal systems for the
purpose of establishing whether principles of criminal law common to the major legal systems of the world
exist on the matter (§§728-37). The Chamber concluded that no such principle could be found and added:
It therefore falls to the Trial Chamber to endeavour to look for a general principle of law consonant with the
fundamental features and the basic requirements of international criminal justice’ (§738).
It then set out two basic, potentially conflicting, requirements (that ‘the rights of the accused be fully safe­
guarded and that the Prosecutor and, more generally, the International Tribunal be in a position to exercise
all the powers expressly or implicitly deriving from the Statute or inherent in their functions, that are neces­
sary for them to fulfil their mission efficiently and in the interests of justice’: §§738-9). The TC concluded
that a careful balancing of these two requirements, as delineated by it, enabled a satisfactory legal solution
to be attained (§§742-8). One could note that, in actual practice, rather than applying a general principle or
conception of law, the TC outlined—others could say crafted—a principle based on such general concepts
as fair trial and equality of arms.
25 Failing that [i.e. an applicable rule of the ICC Statute, of the Elements of Crimes and the ICC Rules of
Procedure and Evidence, as well as of applicable treaties and the ‘principles and rules of international [cus­
tomary] law’] [the Court shall apply] general principles of law derived by the Court from national laws of
legal systems of the world including, as appropriate, the national laws of States that would normally exercise
jurisdiction over the crime, provide that those principles are not inconsistent with this Statute and with
international law and internationally recognized norms and standards.
 FUNDAMENTALS OF INTERNATIONAL CRIMINAL LAW 23

African continent. (It is more and more frequently pointed out in the legal literature
that limiting comparative legal analysis to civil law and common law systems alone is
too restrictive).26
International courts have sounded a note of warning about resorting to general
principles. They have emphasized that one ought not to transpose legal constructs
typical of national legal systems into international law, whenever these constructs do
not harmonize with the specific features of the international legal system. The ICTY
has taken this approach. In 1998 a TC in Furundzija set out an articulate delineation of
the limitations inherent in resorting to general principles. After mentioning the need
to look for ‘principles of criminal law common to the major legal systems of the world’
(§177), the TC went on to specify the following:
Whenever international criminal rules do not define a notion of criminal law, reliance upon
national legislation is justified, subject to the following conditions: (i) unless indicated by an
international rule, reference should not be made to one national legal system only, say that
of common-law or that of civil-law States. Rather, international courts must draw upon the
general concepts and legal institutions common to all the major legal systems of the world.
This presupposes a process of identification of the common denominators in these legal
systems so as to pinpoint the basic notions they share; (ii) since ‘international trials exhibit
a number of features that differentiate them from national criminal proceedings’ [reference
is made here to Judge Cassese’s Separate and Dissenting Opinion in Erdemovic, 7 October
1997], account must be taken of the specificity of international criminal proceedings when
utilising national law notions. In this way a mechanical importation or transposition from
national law into international criminal proceedings is avoided, as well as the attendant
distortions of the unique traits of such proceedings (§178).27

26 This distinction (still to a large extent upheld in such standard works as R. David and C. Juaffret Spinosi,
Les grands systemes de d roil cont emp ora ins, 10th edn (Paris, 1992)); as is well known, David divided the legal
world into four families: common law, civil law, socialist law, other conceptions of law), is held to be on the
wane by such writers as, for instance, J. Gordley, ‘Common Law and Civil Law: eine überholte Unterscheidung,
3 Zeitschrift fü r Europäisches Privatrecht (1993), 498 ff.; H.P. Glenn, La civilization de la common law, 45
Revue internationale de droit compare (1993), 599 if.; B.S. Markesinis (ed.), The Gradual Convergence: Foreign
Ideas, Foreign Influences, and English Law on the Eve of the 21st Century (Oxford: Clarendon Press, 1994). See
also H.P. Glenn, Legal Traditions of the World (Oxford: Oxford University Press, 2000).
Recently a distinguished author (U. Mattei, ‘Three Patterns of Law: Taxonomy and Change in the World’s
Legal Systems’, 45 American J. of Comparative Law (1997), 5-44) has suggested a tripartite scheme: in his
view there exist three patterns of law, according to the relative prevalence o f ‘the rule of professional law’,
‘the rule of political law’ and ‘the rule of traditional law’. The ‘rule of professional law’, which predominates
in the Western world (North America, western Europe, South Africa, and Oceania) can be subdivided, in
his opinion, into three subsystems: common law, civil law, and mixed systems (such as Scotland, Louisiana,
Quebec, South Africa) including the Scandinavian countries (ibid., 41-2).
27 The same TC conclusively set out this notion in Kupreskic et al. (§677 and see also §539). It held that ‘[I]t
is now clear that to fill possible gaps in international customary and treaty law, international and national
criminal courts may draw upon general principles of criminal law as they derive from the convergence of the
principal penal systems of the world. Where necessary, the TC shall use such principles to fill any lacunae
in the Statute of the International Tribunal and in customary law’ (§677; see also §539). It would seem that
the ICTR AC mechanically transposed onto ICL the notion o f‘abuse of process doctrine’ upheld in common
law countries but unknown to most countries of Romano-Germanic tradition, in Barayagwisa (Decision,
§§73-101).
 24 INTERNATIONAL CRIMINAL LAW

International courts have often relied upon these principles. For instance, the ICTY
has had the opportunity to discuss this subsidiary source of law fairly often. In some
cases the ICTY found that there existed general principles common to the major legal
systems of the world, and accordingly applied them.
Thus, in Erdemovic (sentencing judgment), the TC, in discussing the defences of
duress, state of necessity, and superior order, held that ‘a rigorous and restrictive
approach to this matter should be taken, adding that such approach was in line with
the general principles of law as expressed in numerous national laws and case law’
(§19). However, it actually relied only on French law and case law (see ibid., n. 13).
In the same case the TC set about looking for the scale of penalties applicable for
crimes against humanity. It found that among the various elements to be taken into
account were the penalties associated with [crimes against humanity] under inter­
national law and national laws, which are expressions of general principles of law rec­
ognised by all nations’ (§26). After a brief survey of international practice, it pointed
out that ‘[a]s in international law, the States which included crimes against hum anity
in their national laws provided that the commission of such crimes would entail the
imposition of the most severe penalties permitted in their respective systems’ (§30).
However, the TC did not give any specific indication of these laws. It then concluded
as follows:

The Trial Chamber thus notes that there is a general principle of law common to all nations
whereby the severest penalties apply for crimes against humanity in national legal systems.
It thus concludes that there exists in international law a standard according to which a crime
against humanity is one of extreme gravity demanding the most severe penalties when no
mitigating circumstances are present (§31).28

It may be respectfully noted that the Court not only failed to indicate on what
national laws it had relied but also omitted to specify whether it had taken into account,
in addition to general penal legislation, national laws on war crimes as well as those on
genocide, to establish whether these last laws provide for penalties as serious as those
attaching to crimes against humanity. It would therefore seem that the legal propos­
ition set out by the Court does not carry the weight it could have, had it been supported
by convincing legal reasoning.
In Furundzija, the TC had to find a definition of one of the categories of war crimes
and crimes against humanity, namely rape. After going through international treaties
and having considered the relevant case law for the purpose of establishing whether it
evinced the formation of a customary rule on the matter, the Chamber stated that no
elements other than the few resulting from such examination could be
drawn from international treaty or customary law, nor is resort to general principles of
ICL or to general principles of international law of any avail. The Trial Chamber therefore

Subsequently, after surveying the general practice regarding prison sentences in the case law of the
former Yugoslavia, the Court found that reference to this practice was ‘in fact a reflection of the general
principle of law internationally recognised by the community of nations whereby the most severe penalties
may be imposed for crimes against humanity’ (§40).
 FUNDAMENTALS OF INTERNATIONAL CRIMINAL LAW 25

considers that, to arrive at an accurate definition of rape based on the criminal law principle
of specificity [...] it is necessary to look for principles of criminal law common to the major
legal systems of the world (§177).
After undertaking this examination, the Court reached the conclusion that in spite
of inevitable discrepancies, most legal systems in the common and civil law worlds
consider rape to be the forcible sexual penetration of the hum an body by the penis or
the forcible insertion of any other object into either the vagina or the anus (§181).29
Far more numerous are, however, the cases where the ICTY has ruled out the exist­
ence of a general principle of law recognized by all nations.30

29 However, as noted above, on one point, namely whether forced oral penetration could be defined rape
or sexual assault, the Court found that there was no uniformity in national legislation.
In Kupreskic and others, TCII took into consideration the question of general principles on a number of
occasions. Thus it considered whether there were ‘principles of criminal law common to the major systems
of the world’outlining the criteria for deciding whether there has been a violation of one or more provisions
when the same conduct can be regarded as breaching more than one provision of criminal law (the question
of cumulation of offences), and concluded that such criteria did exist (§§680-95).
In Blaskic, the TC held that the principle on the various forms of individual criminal responsibility laid
down in Article 7(1) of the ICTY Statute was consonant ‘with the general principles of criminal law’ as well
as international customary law (§264). Subsequently, in appraising the various elements to be considered for
the determination of the appropriate penalty, the Chamber held that the ‘principle of proportionality’ [of the
penalty to the gravity of the crime] is a ‘general principle of criminal law’ (§796).
30 Thus, in Tadic, the TC rightly excluded a principle whereby unus testis nullus testis (one witness is no
witness), i.e. a principle requiring corroboration of evidence. It found that this principle was not even uni­
versally upheld in civil law systems (§§256, 535-9). In Erdemovic (AJ, 1997), Judges McDonald and Vohrah
in their Joint Separate Opinion, as well as Judge Li in his Separate and Dissenting Opinion, held that there
was no general principle on the question of whether duress can serve as a defence to the killing of innocent
civilians (§§46-58 and 4, respectively). Judge Cassese, in his Dissenting Opinion, contended, on the basis of
the international case law, that no special rule excluding duress as a defence in a case of murder had evolved
in ICL and that, in the absence of such a special rule, the Tribunal had to apply the general rule, which was to
recognize duress as a defence without specifying to which crimes it applied and to which crimes it did not.
Consequently, and subject to the strict requirements enumerated in his dissent, duress could be admitted as
a complete defence even to the crime of killing innocent persons: see §§11-49).
Similarly, in Tadic (AJ, 1999) the AC held that the criminal doctrine of acting in pursuance of
a common purpose, although rooted in the national law of many states, did not amount to a gen­
eral principle common to the major legal systems of the world (§§224-5). In Kupreskic and others, the
TC looked for general principles common to the major systems of the world on the question of how a double
conviction for a single action must be reflected in sentencing, and concluded that no such principles could be
discerned (§§713-16). It reached the same negative conclusion in another area: the specific question of how a
Trial Chamber should proceed when certain legal ingredients of a charge [made by the Prosecutor] have not
been proved but the evidence shows that, if the facts were differently characterised, an international crime
under the jurisdiction of the Tribunal would nevertheless have been perpetrated’ (§§728-38). The Court
therefore held that, lacking a general principle common to the major legal systems of the world, it fell to it
‘to endeavour to look for a general principle of law consonant with the fundamental features and the basic
requirements of international criminal justice’ (§738).
It is also notable that in Aleksovski, the AC pointed out that the principle of stare decisis, or binding pre­
cedent, tended to underpin the general trend of both common and civil law. However, the AC rightly held
that in the event the issue was to be settled in light not of a general principle common to the systems of the
world, but of international law (§98).
 26 INTERNATIONAL CRIMINAL LAW

1.4.6 R E G U L A T I O N S A N D O T H E R R U L E S O F IL

International proceedings are normally governed by ‘Rules of procedure and evi­

dence (RPE) that may be adopted by the international court itself, by virtue of a pro­
vision contained in the court’s Statute (this is the case with the ICTY and the ICTR).
The adoption of such Rules is thus provided for in an international instrum ent (the
court s Statute) adopted on the strength and by virtue of an international treaty (the
UN Charter). It follows that the passing of such rules of procedure amounts to ‘tertiary
legislation’.
In the case of the ICC, under Article 51(1) and (2) it is the Assembly of States Parties
that adopts the Rules of Procedure and Evidence by a two-thirds majority. However,
under Article 51(3), ‘in urgent cases where the Rules [of Procedure and Evidence] do
not provide for a specific situation before the Court, the judges may, by a two-thirds
majority, draw up provisional Rules to be applied until adopted, amended, or rejected
at the next ordinary or special session of the Assembly of the States Parties’. Clearly,
in this case, the law-making process leading to the adoption of the Rules constitutes a
‘secondary’ source of law, for it is envisaged in a treaty (the ICC Statute).
This set of rules must not conflict either with the prim ary (or ‘secondary’) legis­
lation governing the same matter (the Statute of the ICTY, the ICTR, the ICC, the
SCSL, or the STL) or with rules and principles laid down in customary law. In case of
inconsistency, a court should refrain from applying the relevant regulation or rule of
procedure, or else it must construe and apply them in such a m anner that they prove
consonant with the overriding rules.31
As for the principles of interpretation, once again they should be those upheld in
international law and codified in the Vienna Convention on the Law of Treaties: see to
this effect the judgment of the ICTY AC in Jelisic (AJ), where the Court rightly relied
upon the Vienna Convention to construe a Rule (98 bis (B)) of the RPE (§35).

1.4.7 T H E ROLE OF JU D IC IA L D E C ISIO N S A N D T H E O P IN IO N

OF SC H O L A R S

As stated above, judicial decisions—even of the same court—do not constitute per se
a source of ICL. Formally speaking they may only amount to a ‘subsidiary means for
the determination of international rules of law (see Article 38(l)(d) of the ICJ Statute,
which reflects customary international law).
Nevertheless, given the characteristics ofICL (see supra, 1.2) one should set great store
by national or international judicial decisions. They may prove of crucial importance,

In Blaskic (subpoena) the ICTY AC asked itself whether the term ‘subpoena’ used in Rule 54 of the RPE
should be understood ‘to mean an injunction accompanied by a threat of penalty in case of non-compliance’,
or instead should be taken to designate a binding order not necessarily implying the assertion of a power to
imprison or fine. The Court held that, since under customary international law tribunals were not empowered
to issue to states subpoenas capable of being enforced by a penalty, the term was to be given a narrow inter­
pretation: it was to be construed as indicating compulsory orders, which, only when addressed to individuals
acting in their private capacity, could imply the possible imposition of a penalty (§§21,24-5 and 38).
 FUNDAMENTALS OF INTERNATIONAL CRIMINAL LAW 27

not only for ascertaining whether a customary rule has evolved, but also as a means to
establish the most appropriate interpretation to be placed on a treaty rule.
In Aleksovski (AJ), the ICTY AC held that it could depart from a previous decision
by the same AC if it had cogent reasons for so doing (at §§92-111). One may wonder
whether the Chamber purported to establish a form of precedent at the Tribunal. The
objection is possible that this would be trying to pull oneself up by one’s own boot­
straps: one cannot establish a doctrine of precedent by precedent, for it would be tauto­
logical. In any event, it could be contended that this decision was not really precedent.
Besides, according to the traditional and strict doctrine of precedent, one court has to
follow another court’s decision, if the prior decision dealt with the same issue, whether
or not it has cogent reasons for departing from it.
It would therefore seem that the Aleksovski approach should be construed to the
effect that one AC’s decision may only be persuasive authority for another AC.
However, a decision by an AC in the very same case (e.g. the AC directing a TC to
do x or y) is binding on the TC. That, however, is not really a matter of precedent but
rather of the hierarchy of power between the appellate and trial levels: the AC has the
power to order’ the TC to act in a certain way as a matter of the division of labour
between them and their respective powers.
Legal literature, although it carries less weight than case law, may significantly con­
tribute to the elucidation of international rules.

1.5 THE HISTORICAL EVOLUTION OF

INTERNATIONAL CRIMES

Traditionally, individuals have been subject to the (executive and judicial) jurisdic­
tion of the state on whose territory they live. Hence, their possible violations of inter­
national rules (for example, ill-treatment of foreigners, attacks on foreign diplomats,
wrongful expulsion of foreigners by state officials, etc.) were prosecuted and punished
by the competent authorities of the state where these acts had been performed (under
the doctrine of territorial jurisdiction). Clearly, such prosecution and punishment
only occurred if the territorial state authorities were entitled to do so under their
national legislation, and provided they were willing so to proceed. If they did not,
the state of which the victim had the nationality was authorized internationally to
claim from the delinquent state that it either punish the perpetrators or pay compen­
sation. As what was involved was the responsibility of the state (for failure to bring to
trial and punish the offenders), the individuals who had materially breached inter­
national rules could not be called to account by the foreign state, unless they were their
nationals (think of the case of a Russian killing a Russian diplomat in Berlin: Russian
judges were entitled to try the former, provided he had returned to Russia or the Ger­
m an authorities had extradited him). In particular, if the international wrongful act
had been performed by one or more state officials (for instance, prosecutors or judges
 28 INTERNATIONAL CRIMINAL LAW

had wrongfully refrained from instituting criminal proceedings against the material
offender, or political leaders had wilfully instigated him to commit the offence), they
were entitled to im m unity abroad in that they had acted in an official capacity. Hence,
if they travelled to the territory of the aggrieved state and were arrested and brought
to trial, they were entitled to claim im m unity from jurisdiction as well as from sub­
stantive law (if they had the status of head of state, foreign minister, or diplomat, they
could also invoke personal immunities and inviolability; in consequence, they could
not even be arrested, let alone put in the dock).
A few exceptions to these general rules on territorial jurisdiction existed. One of them
was piracy, a practice that was widespread in the seventeenth and eighteenth centuries,
and has recently regained some importance, albeit limited to one area of the world—
East Asia.32All states of the world were empowered to search for and arrest pirates on the
high seas; they were also empowered to bring them to trial, regardless of the nationality
of the victims and of whether the proceeding state had been directly damaged by piracy.
The pirates were regarded as enemies of humanity (hostes humani generis) in that they
hampered the freedom of the high seas and threatened private property.
Another exception was constituted by war crimes. This category of international
crimes gradually emerged in the second half of the nineteenth century. Together with
piracy (which, however, is a much older category), it constituted the first exception to
the concept of collective responsibility’ prevailing in the international community
(under this notion only states as such could be held responsible for acts performed by
individuals—state officials; hence the whole state community ‘paid’ for the wrongful
act, if the state was then to grant reparation to the aggrieved state).
Two factors gave great impulse to the emergence of the class of war crimes. The first
was the codification of the customary law of warfare, as it was then called, at both a pri­
vate or semi-private level and at state level. At the private level, there emerged the famous
Lieber Code, in 186333 (which, issued by Army order no. 100 of President Lincoln, as
‘Instructions for the Government of the United States in the Field’, was applied during
the American Civil War, 1861-5). Also notable was the adoption by the Institut de Droit
International of the important Oxford Manual, in 1880.34 At the state level, a remarkable

32 An authoritative definition of piracy can now be found in Article 101 of the 1982 Convention on the
Law of the Sea:
Piracy consists of any of the following acts:
(a) any illegal acts of violence or detention, or any act of depredation, committed for private ends by the
crew or the passengers of a private ship, or a private aircraft, and directed
(i) on the high seas, against another ship or aircraft, or against persons or property on board such
ship or aircraft;
(ii) against a ship, aircraft, persons or property in a place outside the jurisdiction of any state;
(b) any act or voluntary participation in the operation of a ship or of an aircraft with knowledge of facts
making it a pirate ship or aircraft,
(c) any act of inciting or of intentionally facilitating an act described in subparagraph (a) and (b).
33 Text in Friedman, 1,158-86.
34 See Les Lois de la Guerre sur Terre, Manuel publie par Vlnstitut de Droit International (Brussels and
Leipzig: C. Muquardt, 1880).
 FUNDAMENTALS OF INTERNATIONAL CRIMINAL LAW 29

impulse was given by the Hague codification (1899-1907). Secondly, some important tri­
als were held at the end of the American Civil War (which in fact amounted to an inter­
national armed conflict, from the viewpoint of law), notably Henry Wirz (a case of serious
ill-treatment of prisoners of war), heard by a US Military Commission (1865). Later on,
many cases were brought in 1902 before US Courts Martial at the end of the US armed
conflict (1899-1901) against insurgents in the Philippines (which Spain had ceded by
treaty to the USA in 1898).35
Traditionally such crimes were defined as violations of the laws of warfare com­
mitted by combatants in international armed conflicts. War crimes entailed two
things. First, individuals acting as state officials (chiefly low-ranking servicemen)
could be brought to trial and punished for alleged violations of the laws of warfare by
the enemy belligerent. The exceptional character of war (a pathological occurrence in
international dealings, leading to utterly inhum an behaviour) warranted this devi­
ation from traditional law (which, as already pointed out above, granted to any state
official acting in an official capacity immunity from prosecution by foreign states).36
Secondly, individuals could be punished, not only by the enemy state but also by their
own state.
In actual fact for many years war crimes were chiefly prosecuted and punished by
the culprits’ own national authorities after the end of the hostilities.37 This holds true

35 One should mention in particular General Jacob H. Smith (about a superior order to deny quarter), the
case of Major Edwin F. Glenn (concerning an order to torture a detained enemy), that of Lieutenant Preston
Brown (about the killing of an unarmed prisoner of war), and Augustine de La Pena (again, a case of torture
of an enemy detained person).
US courts held many other trials in relation to crimes committed in armed conflict. See the numerous
cases cited in W. Winthrop, Military Law and Precedents, 2nd edn (Buffalo, NY: William S. Helm & Co.,
1920), 839-62.
36 The contrary view of A. Verdross, Völkerrecht (Berlin: Springer Verlag, 1937) at 298 was (and is) wrong.
(According to the distinguished Austrian international lawyer, ‘punishment [of authors of war crimes] must
be ruled out when the action was not performed on one’s own impulse, but must be exclusively attributed to
the state of which the person is a national’). H. Kelsen (Peace through Law (Chapel Hill: University of North
Carolina Press, 1944, at 97) shared Verdross’ view.
Characteristically, the 1912 British Manual on Land Warfare stipulated that ‘war crimes is the technical
expression for such an act of enemy soldiers and enemy civilians as may be visited by punishment on capture
of the offenders’(§ 441).
37 According to the authoritative History of the United Nations War Crimes Commission and the Develop­
ment of the Laws of War, compiled by the ‘United Nations War Crimes Commission’ (London: His Majesty’s
Stationery Office, 1948, at 29) “The right of the belligerent to punish as war criminals persons who violate the
laws or customs of war is a well-recognized principle of international law. It is the right of which a belligerent
may effectively avail himself during the war in cases when such offenders fall into his hands, or after he has
occupied all or part of enemy territory and is thus in the position to seize war criminals who happen to be
there. [...] And although the Treaty of Peace brings to an end the right to prosecute war criminals, no rule of
international law prevents the victorious belligerent from imposing upon the defeated state the obligation,
as one of the provisions of the armistice or of the Peace Treaty, to surrender for trial persons accused of war
crimes.’
This view, also shared by H. Kelsen (Peace through Law, cit., at 108-10) does not seem, however, to reflect
the status of traditional international law. As was conclusively demonstrated by A. Merignhac ( De la sanc­
tion des infractions au droit des gens commises, au cours de la guerre europeenne, par les empires du cen­
tre’, 24 RGDIP (1917), 28-56) and L. Renault (‘De l’application du droit penal aux faits de la guerre’, ibid.,
 30 INTERNATIONAL CRIMINAL LAW

for the numerous war crimes trials held in 1902 by US Courts M artial for offences
committed by Americans in the armed conflict in the Philippines, as well as for the tri­
als conducted in 1921 by the German Supreme Court against alleged German perpe­
trators (although in this case Germany had been compelled by the Allies to hold such
trials). In some instances, however, it was a belligerent that brought to trial an enemy
serviceman during the armed conflict: see two cases brought before Austrian Military
Courts during the First World War (the case of the Russian prisoner of war J.K. (1915)
(at 17-20), and Stanislaus Bednarek (1916) (at 1-2)). In contrast, the Henry Wirz case,
referred to above, was decided by enemy courts after the end of the war.
Interestingly, after the First World War, in 1919, some Extraordinary Courts Martial
were established in the Ottom an Empire to try persons responsible for the massacres
and atrocities committed in 1915-16 against Armenians. The authors of those crimes,
which would at present be classified as crimes against hum anity or, depending upon
the facts of each case, genocide, were instead tried under the Ottom an Criminal Code
for massacres’ ‘deportation, or ‘looting’, and sometimes for ‘massacres with the goal
of annihilating’ Armenians (see infra, 5.2).
After the Second World War the prosecution was normally effected by the victor
state, as well as by one of its allies, on the basis either of the principle of territoriality
(the crime had been committed on its territory), or of passive nationality (it was suf­
ficient for the victim to have the nationality of an allied country). Although various
national legislations also made provision tor punishment on the basis of the principle
of active nationality’ (the law-breaker had the nationality of the prosecuting state), in
practice scant use was made of this principle, for obvious reasons. (One notable excep­
tion is the trials conducted by German courts against Germans during 1946-51, under
a set of provisions jointly passed by the four Allies, the Control Council Law no. 10.)
The creation of the IMT and the subsequent trial at Nuremberg of the major German
criminals (followed in 1946 by the Tokyo Trial), marked a crucial turning point. First,
two new categories of crime were envisaged: crimes against peace and crimes against
humanity. Secondly, until 1945 (with the exception of the provisions of the 1919 Treaty
of Versailles relating to the German Emperor, which, however, remained a dead letter),
senior state officials had never been held personally responsible for their wrongdoings.
Until that time states alone could be called to account by other states, plus servicemen
(normally low-ranking people) accused of misconduct during international wars. In
1945, for the first time in history, the principle was laid down—and carried through,
in contrast to what had happened in 1919—that other state agents (high-ranking offic­
ers, politicians, prom inent administrators or financiers, as well as men in charge of
official state propaganda) could also be made answerable for gross misconduct in
time of armed conflict. Those men were no longer protected by state sovereignty;
they could be brought to trial before organs—representative if not of the whole inter­
national community, at least of the large group of the allied victors—and punished by

25 (1918), 5-29), state practice shows that belligerents are entitled to prosecute and punish their servicemen
as well as enemy military both during the armed conflict and after the end of hostilities.
 fundamentals of international crim inal law 31

foreign states.38 For the first time the basic principle was proclaimed that, faced with
the alternative of complying with either national legal commands or international
standards, state officials and individuals must opt for the latter. As the IMT forcefully
stated, ‘the very essence of the Charter [instituting the IMT] is that individuals have
international duties which transcend the national obligations of obedience imposed
by the individual State’ (at 223).
After the adoption in 1948 of the Convention on Genocide (which laid down
genocide as a discrete crime), the 1949 Geneva Conventions marked a great advance
as regards the extension both of substantive law (new categories of war crimes were
added: they were termed ‘grave breaches of the Geneva C onventions) and of the law
for the enforcement of substantive prohibitions. W ith regard to this last issue, the
relevant provisions represented a momentous departure from customary law, for the
Conventions laid down the principle of universality of jurisdiction (a contracting
state could bring to trial a person held in its custody and accused of a grave breach ,
regardless of his nationality, of the nationality of the victim, and of the place where the
alleged offence had been committed). It is probable that the exceedingly bold charac­
ter of this regulation contributed to its remaining ineffective for many years.
The Geneva Conventions were followed by the two Additional Protocols in 1977, the
Convention against Torture in 1984 (which significantly contributed to the emergence
of torture as a distinct crime), and a string of treaties against terrorism since the 1970s
(which contributed to the evolution of an international crime of terrorism).
Later on, as the ICTY AC authoritatively held in Tadic (IA) (§§94-137), the notion
of war crimes was gradually extended to serious violations of international hum ani­
tarian rules governing internal armed conflict.

38 The idea propounded by such distinguished international lawyers as the American Hyde (C. C.
Hyde, ‘Punishment of War Criminals’, Proceedings of the ASIL (1943), at 43-4) and the Austrian Kelsen
(H. Kelsen, Peace through Law, cit., at 111-16) that the international Court should consist of neutral nationals
was not upheld, clearly for political reasons; that is, because the victors wished to be and remain in control of
the trials.
 2
GENERAL PRINCIPLES OF
INTERNATIONAL
CRIMINAL LAW

2.1 PRELIMINARY REMARKS

In every legal order general principles are needed, which set out the overall orientation
of the system, provide sweeping guidelines for the proper interpretation of the law
whenever specific rules on legal construction prove insufficient or unhelpful, and also
enable courts to fill the gaps of written or unwritten norms. ICL, being a branch of
public international law, shares of course with any other sector of this corpus of rules
the general principles proper to it. However, given the unique features and the over­
arching purpose of this body of law (see supra, 1.2), on many occasions those general
principles may turn out to be of scant assistance. More useful and relevant appear to be
the general principles proper to ICL, for they are more attuned to its specificities.1

An international court recently questioned reliance on such principles. In Delalic and others an ICTY
TC, after noting that these principles exist and are recognised in all the world’s major criminal justice
systems’ stated that ‘[i]t is not certain to what extent they have been admitted as part of international legal
practice, separate and apart from the existence of the national legal systems. This is essentially because of
the different methods of criminalisation of conduct in national and international criminal justice systems’
(§403). The Chamber then explained the difference between the two levels (national and international) as
follows Whereas the criminalisation process in a national criminal justice system depends upon legislation
which dictates the time when conduct is prohibited and the content of such prohibition, the international
criminal justice system attains the same objective through treaties and conventions, or after a customary
practice of the unilateral enforcement of a prohibition by State’ (§404).
With respect, this explanation is not compelling. It would rather seem that the difference between national
criminal laws and international criminal rules lies in the still rudimentary character of the latter. This body
of law has not yet attained the degree of sophistication proper to national legal systems. It follows that the
principles in question are not yet applicable at the international level in all their implications and ramifica­
tions. Whether or not this legal justification is more cogent that the one advanced by the TC, one can, how­
ever, share at least the substance of the conclusions reached by the Chamber, which were as follows: ‘It could
be postulated, therefore, that the principles of legality in ICL are different from their related national legal
systems with respect to their application and standards. They appear to be distinctive, in the obvious objective
of maintaining a balance between the preservation ofjustice and fairness towards the accused and taking into
account the preservation of world order. To this end, the affected State or States must take into account the fol­
lowing factors, inter alia-, the nature of international law; the absence of international legislative policies and
 GENERAL PRINCIPLES OF INTERNATIONAL CRIMINAL LAW 33

In ICL there exist principles that are not specific to this body of law, for the same
principles can also be found in most state legal systems of the world. Nonetheless, as
we shall see, often the unique features of the international legal order and the way law
takes shape therein, condition the content and scope of some of those principles. One
may therefore conclude that some of those principles ultimately bear scant resem­
blance to those of municipal systems, for they are uniquely shaped to suit the charac­
teristic features of the world legal order.

2.2 THE PRINCIPLE OF INDIVIDUAL

CRIMINAL RESPONSIBILITY

In ICL the general principle applies that no one may be held accountable for an act he
has not performed or in the commission of which he has not in some way participated,
or for an omission that cannot be attributed to him.
The ICTY AC set this fundamental principle out most clearly in Tadic (AJ).2 The
principle in fact lays down two notions. First, nobody may be held accountable for
criminal offences perpetrated by other persons. The rationale behind this propos­
ition is that in modern criminal law the notion of collective responsibility is no longer
acceptable. In other words, a national, ethnic, racial, or religious group to which a per­
son may belong is not accountable for acts performed by a member of the group in his
individual capacity. By the same token, a member of any such group is not criminally
liable for acts contrary to law performed by leaders or other members of the group and
to which he is extraneous. The principle of individual autonomy whereby the individ­
ual is normally endowed with free will and the independent capacity to choose his
conduct is firmly rooted in modern criminal law, including ICL.
Secondly, a person may only be held criminally liable if he is somehow culpable for
any breach of criminal rules. In other words, he may only be deemed accountable if
he is somehow involved in the commission of a crime and in addition entertains a

standards; the ad hoc processes of technical drafting; and the basic assumption that international criminal
law norms will be embodied into the national criminal law of the various States’ (§405).
2 Before ascertaining whether the Appellant could be found guilty under the notion of participation in
a common criminal purpose, it stated that nobody may be held criminally responsible for acts or transac­
tions in which he has not personally engaged or in some other way participated’. ‘The basic assumption
must be that in international law as much as in national systems, the foundation of criminal responsibility is
the principle of personal culpability; nobody may be held criminally responsible for acts or transactions in
which he has not personally engaged or in some other way participated (nullapoena sine culpa). In national
legal systems this principle is laid down in Constitutions, in laws, or in judicial decisions. In international
criminal law the principle is laid down, inter alia, in Article 7(1) of the Statute of the International Tribunal
which states that; A person who planned, instigated, ordered, committed... [a crime]... shall be individually
responsible for the crime’ (emphasis added) (§186).
An ICTY TC recently restated in Kordic and Cerkez that this is a general principle applicable at the inter­
national level (§364).
34 INTERNATIONAL CRIMINAL LAW

frame of mind that expresses or implies his mental participation in the offence, or his
culpably negligent (or deliberate) omission to prevent or punish the commission of
crimes by his subordinates. As a consequence, objective criminal liability is ruled out.
It follows from the first notion that, among other things, no one may be held
answerable for acts or omissions of organizations to which he belongs, unless he bears
personal responsibility for a particular act, conduct, or omission.
An exception was, however, provided for in Articles 9 and 10 of the Statute of the
IMT at Nuremberg. Article 9 (1) stipulated that
At the trial of any individual member of any group or organization the Tribu nal may declare
(in connection with any act of which the individual may be convicted) that the group or
organization of which the individual was a member was a criminal organization.

Under Article 10
In cases where a group or organization is declared criminal by the Tribunal, the competent
national authority of any Signatory shall have the right to bring individuals to trial for mem­
bership therein before national, military, or occupation courts. In any such case the crim­
inal nature of the group or organization is considered proved and shall not be questioned.

Thus, mere membership in a criminal organization was regarded as criminal, whether or

not participation in that organization was voluntary. The idea behind the whole scheme
for post-war trials for war crimes, first propounded by Colonel Murray C. Bernays in the
US Pentagon in 1944, and eventually upheld by the Secretary of War, Henry L. Stimson,
was that ‘It will never be possible to catch and convict every Axis war criminal, or even
any great number of them, under the old concepts and procedures.’3 Given also that
Anglo-American law to some extent upholds the notion of corporate criminal liability, it
was suggested that it was for an international court to adjudicate and punish the crimes
of the leaders and of the criminal organizations. Thereafter, every member of the Nazi
Government or of those organizations would be subject to arrest, trial, and punishment
in the national courts of each state concerned. ‘Proof of membership, without more,
would establish guilt of participation in the mentioned conspiracy, and the individ­
ual would be punished in the discretion of the court.’4 This scheme was confirmed by
Control Council Law no. 10, of 20 December 1945, which provided in Article II(l)(d) that
acts ‘recognized as a crime’ included ‘membership in categories of a criminal group or
organization declared criminal by the International Military Tribunal’.
In its judgment in Goring and others the IMT eventually labelled some organiza­
tions as criminal: the Leadership Corps of the Nazi Party; the Gestapo and the SD; and
the SS. However, the Tribunal discarded the doctrine o f ‘objective’ or ‘group responsi­
bility’ and brought back the provisions of the Statute to traditional concepts of crim ­
inal law. It made the following qualifying points (at 255-79).

3 See Memo by Colonel Bernays of 15 September 1944, in B. F. Smith, The American Road to Nuremberg—
The Documentary Record—1944-45 (Stanford, Cal.: Hoover Institution Press, 1982), at 35.
4 Ibid., at 36.
 GENERAL PRINCIPLES OF INTERNATIONAL CRIMINAL LAW 35

First, it held that the labelling of a group or organization as crim inal should not be
based on arbitrary action’ but on ‘well-settled legal principles’, chiefly the principle
that ‘criminal guilt is personal’ and ‘that mass punishments should be avoided’. In
addition, ‘the Tribunal should make such declaration of criminality so far as possible
in a m anner to insure that innocent persons will not be punished’.
Secondly, the Tribunal reduced the notion of ‘criminal organization’ to that of
‘criminal conspiracy’:
A criminal organization is analogous to a criminal conspiracy in that the essence of both is
cooperation for criminal purposes. There must be a group bound together and organized for
a common purpose. The group must be formed or used in connection with the commission
of crimes denounced by the Charter.
It followed that one ‘should exclude persons who had no knowledge of the criminal
purposes or acts of the organization and those who were drafted by the state for m em­
bership, unless they were personally implicated in the commission of acts declared
criminal by Article 6 of the Charter as members of the organization’.
Thirdly, the Tribunal issued three ‘recommendations’ to other courts with regard to
penalties to be inflicted on members of criminal organizations.567
Fourthly, the Tribunal, each time it termed an organization criminal, added a simi­
lar caveat: one could hold criminally liable only those members of the organization
who had had ‘knowledge that it was being used for the commission’ of international
crimes, or were ‘personally implicated’ in the commission of such crimes, and in add­
ition had not ceased to belong to the organization prior to 1 September 1939 (the start
of the war of aggression by Germany).
It would appear that subsequent courts complied. Consequently, members of
German organizations termed criminal by the IMT were not punished for the mere
fact of belonging to one of them. Furthermore, other Tribunals upheld the principle of
personal responsibility laid down by the IMT in is judgment.

5 They were as follows: T. That so far as possible throughout the four zones of occupation in Germany the
classifications, sanctions and penalties be standardized. Uniformity of treatment so far as practical should
be a basic principle [...] 2. [Control Council] Law no. 10 [...] leaves punishment entirely in the discretion
of the trial court even to the extent of inflicting the death penalty. The De-Nazification Law of 5 March
1946, however, passed for Bavaria, Greater-Hesse, and Württemberg-Baden, provides definite sentences for
punishment in each type of offense. The Tribunal recommends that in no case should imprisonment imposed
under Law no. 10 upon any members of an organization or group declared by the Tribunal to be crimma
exceed the punishment fixed by the De-Nazification Law. No person should be punished under both laws.
3. The Tribunal recommends to the Control Council that Law no. 10 be amended to prescribe limitations
on the punishment which may be imposed for membership in a criminal group or organization so that such
punishment shall not exceed the punishment prescribed by the De-Nazification Law' (at 267-8).
6 Emphasis added. See ibid., at 262,268,273.
7 Thus, in Krupp and others, where the 12 accused were officials of the Krupp industrial enterprises who
occupied high positions in the political, financial, industrial, and economic life of Germany, a US Tribunal
sitting at Nuremberg held that the defendants could be held criminally liable only if it could be proved
that they had actually and personally’ committed the offences charged. ‘The mere fact without more that
a defendant was a member of the Krupp Directorate or an official of the firm is not sufficient [for criminal
liability to arise].' It then cited a rule of the American Corpus Juris Secundum on corporate liability, whereby
 36 INTERNATIONAL CRIMINAL LAW

2.3 THE PRINCIPLE OF LEGALITY OF CRIMES

To grasp fully the significance of this principle a few words of introduction are
necessary.
National legal systems tend to embrace, and ground their criminal law on either
the doctrine of substantive justice or that of strict legality. Under the former doctrine
the legal order must prim arily aim at prohibiting and punishing any conduct that is
socially harmful or causes danger to society, whether or not that conduct has already
been legally criminalized at the moment it is taken. The param ount interest is defend­
ing society against any deviant behaviour likely to cause damage or jeopardize the
social and legal system. Hence this doctrine favours society over the individual (favor
societatis). Extreme and reprehensible applications of this doctrine can be found in
the Soviet legal system (1918-58) or in the Nazi criminal law (1933-45). However,
one can also find some variations of this doctrine in modern democratic Germany,
where the principles of objective justice’ (materielle Gerechtigkeit) have been upheld
as a reaction to oppressive governments trampling upon fundamental hum an rights,
and courts have had recourse to the celebrated ‘Radbruch’s formula’. Radbruch, the
distinguished German professor of jurisprudence, created this ‘formula’ in 1946. In
terms subsequently taken up in some German cases,8 he propounded the notion that

officers of a corporation, normally not criminally responsible for corporate acts performed by other officers
or agents, are nevertheless liable if they actually and personally do the acts constituting the offence, or such
acts are done by their direction or permission, so that an officer is liable ‘where his scienter or authority is
established, or where he is the actual present and efficient actor’. The Tribunal added that the same principles
must apply in the case of war crimes (at 627-8).
Another US Tribunal sitting at Nuremberg took a similar stand in Flick and others (at 1189), and then
in Krauch and others (I G. Farhen trial, at 1108). In this latter case the 23 accused were all officials of
I. G. Farben industrial enterprises, charged among other things with war crimes. The Tribunal took pains
to emphasize that they did not bear collective responsibility but could only be found guilty of individual
criminal liability. It noted the following: ‘It is appropriate here to mention that the corporate defendant,
Farben, is not before the bar of this Tribunal and cannot be subjected to criminal penalties in these proceed­
ings. We have used the term Farben as descriptive of the instrumentality of cohesion in the name of which
the enumerated acts of spoliation were committed. But corporations act through individuals and, under
the conception of personal individual guilt to which previous reference has been made, the Prosecution, to
discharge the burden imposed upon it in this case, must establish by competent proof beyond a reasonable
doubt that an individual defendant was either a participant in the illegal act or that, being aware thereof,
he authorized or approved it. Responsibility does not automatically attach to an act proved to be criminal
merely by virtue of a defendant’s membership in the Vorstand [administration board]. Conversely, one may
not utilize the corporate structure to achieve an immunity from criminal responsibility for illegal acts which
he directs, counsels, aids, orders, or abets. But the evidence must establish action of the character we have
indicated, with knowledge of the essential elements of the crime’ (at 1153).
8 The German Federal Constitutional Court referred to that ‘formula’ in its judgment of 24 October 1996
in Streletz and Kessler. The question at issue was whether the accused, former senior officials of the former
German Democratic Republic (GDR) charged with incitement to commit intentional homicide for their
responsibility in ordering the shooting and killing by border guards of persons trying to flee from the GDR,
could invoke as a ground of justification the fact that their actions were legal under the law applicable in the
GDR at the material time, which did not make them liable to criminal prosecution. The defendants sub­
mitted that holding them criminally liable would run contrary to the ban on the retroactive application of
criminal law and Article 103(2) of the German Constitution laying down the nullum crimen principle. The
 GENERAL PRINCIPLES OF INTERNATIONAL CRIMINAL LAW 37

positive law must be regarded as contrary to justice and not applied where the incon­
sistency between statute law and justice is so intolerable that the former must give way
to the latter. This ‘formula has been widely accepted in the legal literature.9
In contrast, the doctrine of strict legality postulates that a person may only be held
criminally liable and punished if at the moment when he performed a certain act the
act was regarded as a criminal offence by the relevant legal order or, in other words,
under the applicable law. Historically, this doctrine stems from the opposition of the
baronial and knightly class to the arbitrary power of monarchs, and found expression
in Article 39 of Magna Charta libertatum of 1215 (so-called ‘Magna C arta).10 One
must, however, wait for the principal thinkers of the Enlightenment to find its proper
philosophical and political underpinning. Montesquieu and then the great American
proclamations of 1774 and of the French revolution (1789) conceived of the doctrine as
a way of restraining the power of the rulers and safeguarding the prerogatives of the
legislature and the judiciary. As the distinguished German criminal lawyer Franz von
Liszt wrote in 1893, the nullum crimen sine lege and nulla poena sine lege principles ‘are
the bulwark of the citizen against the state’s omnipotence; they protect the individual
against the ruthless power of the majority, against the Leviathan. However paradox­
ical it may sound, the Criminal Code is the criminal s magna charta. It guarantees his
right to be punished only in accordance with the requirements set out by the law and
only within the limits laid down in the law.’11
At present, most democratic civil law countries tend to uphold the doctrine of strict
legality as an overarching principle. In these countries the doctrine is normally held
to articulate four basic notions: (i) criminal offences may only be provided for in writ­
ten law, namely legislation enacted by Parliament, and not in customary rules (less
certain and definite than statutes) or in secondary legislation (which emanates from
the government and not from the parliamentary body expressing popular will), this
principle is referred to by the maxim nullum crimen sine lege scripta (criminal offences
must be provided for in written legislation); (ii) criminal legislation must abide by the
principle of specificity, whereby rules criminalizing hum an conduct must be as spe­
cific and clear as possible, so as to guide the behaviour of citizens; this is expressed
by the Latin tag nullum crimen sine lege stricta (criminal offences must be provided

Court dismissed the defendants’ submissions. Among other things, it noted that the prohibition on retro­
active law derived its justification from the special trust reposed in criminal statutes enacted by a democratic
legislature respecting fundamental rights.
9 Of course, the notion propounded by Radbruch could simply be termed the Natural Justice view that an
unjust law is no law and must be disregarded. As such, it might be susceptible to the criticism of positivists
that it makes the law subjective, since the sense of justice varies from person to person.
10 ‘It is only through the legal judgment by his peers and on the strength of the law of the land that a
freeman may be apprehended or imprisoned or disseised or outlawed or exiled or in any other manner
destroyed, nor may we go upon him or send upon him.’
11 F. von Liszt, ‘Die deterministischen Gegner der Zweckstrafe, 13 Zeitschrift fü r die gesamte
Strafrechtswissenschaft (1893), 325-70, at 357 (an English translation of some excerpts from this paper
has been published in 5 JICJ (2007) 1009-13). The Latin tag nullum crimen had been coined by another
German criminal lawyer, P. J. A. Feuerbach, Lehrbuch des gemeinen in Deutschland gültigen peinlichen
Rechts, llth edn (Geissen: Heyer, 1832) at 12-19 (English trans. In 5 JICJ(2007), at 1005-8).
 38 INTERNATIONAL CRIMINAL LAW

for through specific legislation); (iii) criminal rules may not be retroactive; that is, a
person may only be punished for behaviour that was considered criminal at the time
the conduct was undertaken; therefore he may not be punished on the strength of a law
passed subsequently; the maxim referred to in this case is nullum crimen sineproevia
lege (criminal offences must be provided for in a prior law);12 (iv) resort to analogy in
applying criminal rules is prohibited.
Plainly, as stated above, the purpose of these principles is to safeguard citizens as
far as possible against both the arbitrary power of government and possibly excessive
judicial discretion. In short, the basic underpinning of this doctrine lies in the postu­
late offavor rei (in favour of the accused) (as opposed to favor societatis, or in favour
of society).
In contrast, in common law countries, where judge-made law prevails or is at least
firmly embedded in the legal system, there is a tendency to adopt a qualified approach
to these principles. For one thing, common law offences (as opposed to statutory
offences) result from judge-made law and therefore may lack those requirements of
rigidity, foreseeability, and certainty proper to written legislation. For another, com­
mon law offences are not strictly subject to the principle of non-retroactivity, as is
shown by recent English cases contemplating new offences, or at any rate the extin­
guishing of traditional defences (see, for instance, R. v. R. (1992), which held that the
fact of marriage was no longer a common law defence to a husband’s rape of his wife).13
It is notable that the European Court of Human Rights did not regard such cases as
questionable, or at any rate contrary to the fundamental provisions of the European
Convention (see SW and CR v. United Kingdom, 1995).
Thus, the condition is not the same in every legal system. Let us now see which of the
two aforementioned doctrines is applied in international law.
One could state that international law, being based on customary processes, is more
akin to English law than to French, German, Argentinean, or Chinese law. However,
this would not be sufficient. The main problem is that for a long period, and until
recently, international law has applied the doctrine of substantive justice, and it is only

The German Federal Constitutional Court set out the principle in admirable terms in its aforemen­
tioned decision of 24 October 1996 in Streletz and Kessler. In illustrating the scope of Article 103(2) of the
German Constitution, laying down the principle at issue, it stated the following: ‘(l.a) Article 103 §2 of the
Basic Law protects against retroactive modification of the assessment of the wrongfulness of an act to the
offender’s detriment [...] Accordingly, it also requires that a statutory ground of justification which could
be relied on at the time when an act was committed should continue to be applied even where, by the time
criminal proceedings begin, it has been abolished. However, where justifications are concerned, in contrast
to the definition of offences and penalties, the strict reservation of Parliament’s law-making prerogative does
not apply. In the sphere of the criminal law grounds of justification may also be derived from customary law
or case-law.’
It would seem that the English law used to be that a man could not rape his wife because, by agreeing to
marry, she had implicitly consented to sexual intercourse for all time. This was obviously a somewhat medi­
aeval approach. The defence existed only as a matter of common law—it was not in any statute. The judge
in R. v. R. rightly held that societal attitudes had changed and that it was no longer acceptable to hold that
a husband could in law never be held guilty of raping his wife; hence he did not allow the old common law
defence. In fairness, it was not the introduction of a new offence—rape had always been an offence. It was a
question of disallowing a (retrograde) common law defence.
 GENERAL PRINCIPLES OF INTERNATIONAL CRIMINAL LAW 39

in recent years that it is gradually replacing it with the doctrine of strict legality, albeit
with some im portant qualifications.
That international law has long applied the former doctrine is not to be attributed
to a totalitarian or authoritarian streak in the international community. Rather, the
rationale for that attitude was that states were not prepared to enter into treaties laying
down criminal rules, nor had customary rules evolved covering this area. In practice,
there only existed customary rules prohibiting and punishing war crimes, although in
a rather rudim entary or unsophisticated m anner (see supra, 1.2 and 2.1,2.4.1). Hence
the need for the international community to rely upon the doctrine of substantive
justice when new and extremely serious forms of criminality (crimes against peace,
crimes against humanity) suddenly appeared on the international scene.
The IMT clearly enunciated this doctrine in Goring and others. From the outset
the Tribunal had to face the powerful objections of German defence counsel that the
Tribunal was not allowed to apply ex postfacto law. These objections were grounded in
the general principles of criminal law embedded in civil law countries, and also upheld
in German law before and after the Nazi period. The French Judge H. Donnedieu de
Vabres, coming from a country where the nullum crimen principle is deeply implanted,
also showed himself to be extremely sensitive to the principle. As a consequence,
when dealing with the crimes against peace of which the defendants stood accused,
the Tribunal, before stating that in fact such crimes were already prohibited when
they were perpetrated (at 219-23)-—a finding that still seems highly questionable
noted that in any case it was not contrary to justice to punish those crimes even if the
relevant conduct was not criminalized at the time of their commission
In the first place, it is to be observed that the maxim nullum crimen sine lege is not a limita­
tion of sovereignty, but is in general a principle of justice. To assert that it is unjust to punish
those who in defiance of treaties and assurances have attacked neighbouring states without
warning is obviously untrue, for in such circumstances the attacker must know that he is
doing wrong, and so far from it being unjust to punish him, it would be unjust if his wrong
were allowed to go unpunished (219; emphasis added).
In other words, substantive justice punishes acts that harm society deeply and are
regarded as abhorrent by all members of society, even if these acts were not prohibited
as criminal when they were performed.14

14 In his Dissenting Opinion in the Tokyo trial (Araki and others), Judge B. V. A. Roling spelled out
the same principle, again with regard to crimes against peace. He noted that in national legal systems the
nullum crimen principle ‘is not a principle of justice but a rule of policy; this rule was valid only if expressly
adopted, so as to protect citizens against arbitrariness of courts [...] as well as arbitrariness of legislators [...]
the prohibition of ex postfacto law is an expression of political wisdom, not necessarily applicable in present
international relations. This maxim of liberty may, if circumstances necessitate it, be disregarded even by
powers victorious in a war fought for freedom’ (at 1059). Judge Roling then delineated two classes of criminal
offence: ‘Crime in international law is applied to concepts with different meanings. Apart from those indi­
cated above [war crimes], it can also indicate acts comparable to political crimes in domestic law, where the
decisive element is the danger rather than the guilt, where the criminal is considered an enemy rather than a
villain and where the punishment emphasizes the political measure rather than the judicial retribution’ (at
1060). Judge Roling applied these concepts to crimes against peace and concluded that such crimes were to
 40 INTERNATIONAL CRIMINAL LAW

As stated above, after the Second World War the doctrine of substantive just­
ice (upheld in a num ber of cases, among which one may cite Peleus and later on
Eichmann)15was gradually replaced by that of strict legality. Twofactors brought about
this change.
First, states agreed upon and ratified a number of im portant hum an rights treaties
which laid down the nullum crimen principle, to be strictly complied with by national
courts.16 The same principle was also set out in such im portant treaties as the Third
and Fourth Geneva Conventions of 1949, respectively, on Prisoners of War and on
Civilians.17 The expansive force and striking influence of these treaties could not but
impact on international criminal proceedings, leading to the acceptance of the notion
that also in such proceedings the nullum crimen principle must be respected as a funda­
mental part of a set of basic hum an rights of individuals. In other words, the principle
came to be seen from the viewpoint of the hum an rights of the accused, and no longer
as essentially encapsulating policy guidelines dictating the penal strategy of states at
the international level.
The second factor is that gradually the network of ICL greatly expanded both
through a number of international treaties criminalizing conduct of individuals
(think of the 1948 Convention on Genocide, the 1949 Geneva Conventions, the 1984
Convention on Torture, and the various treaties on terrorism) and by dint of the accu­
mulation of case law. In particular, case law contributed to either the crystallization of
customary international rules of crim inal law (for instance, on the mental element of
crimes against humanity) or to clarifying or specifying elements of crimes, defences,

be punished because of the dangerous character of the individuals who committed them, hence on security
considerations. In his view, however, given the novel nature of these crimes, it followed that persons found
guilty of them could not be punished by a death sentence (ibid.).
15 In Peleus, in his summing up the Judge Advocate stated: ‘You have heard a suggestion made that this
Court has no right to adjudicate upon this case because it is said you cannot create an olfence by a law which
operates retrospectively so as to expose someone to punishment for acts which at the time he did them were
not punishable as crimes. That is the substance of the Latin maxim [nullum crimen sine lege, nulla poena
sine lege] that has been used so much in this Court. My advice to you is that the maxim and the principle
[of legality] that it expresses has nothing whatever to do with this case. It has reference only to municipal
oi domestic law of a particular State, and you need not be embarrassed by it in your consideration of the
problems that you have to deal with here’ (at 132). It should be noted that the defendants had been accused
of killing survivors of a sunken merchant vessel, the Greek steamship Peleus; they had raised the pleas of
‘operational necessity’ and superior orders.
The British Judge Advocate in Burgholz (No. 2) took a clearer stand. After noting that the Allies had set up
tribunals in Germany and Japan ‘with the object of bringing to justice certain persons who have outraged
the basic principles of decency and humanity’, he pointed out: ‘It may well be that no particular concrete law
can be pointed to as having been broken, and you remember what Defence Counsel Dr. Meyer-Labastille
said yesterday on the principle of “no punishment without pre-existing law”. That principle I agree with but
to this extent, that I do not regard it as limiting punishment of persons who have outraged human decency
in their conduct’ (at 79).
As for Eichmann, see the judgment of the Supreme Court of Israel, at 281.
16 See, for instance, Article 15 of the UN Covenant on Civil and Political Rights, Article 7 of the European
Convention on Human Rights, or Article 9 of the American Convention on Human Rights.
17 See Article 99(1) of the Third Convention and Article 67 of the Fourth Convention. See also Article
75(4)(c) of the First Additional Protocol of 1977.
 GENERAL PRINCIPLES OF INTERNATIONAL CRIMINAL LAW 41

and other im portant segments of ICL. As a consequence, the principle of strict legal­
ity has been laid down first, albeit implicitly, in the two ad hoc Tribunals (ICTY and
ICTR),18 and then, explicitly, in the Statute of the ICC, Article 22(1) of which provides
that ‘A person shall not be criminally responsible under this Statute unless the con­
duct in question constitutes, at the time it takes place, a crime within the jurisdiction
of the Court.’
The conclusion is therefore warranted that nowadays this principle must be com­
plied with also at the international level, albeit subject to a num ber of significant quali­
fications, which we shall presently consider.19

2.4 ARTICULATIONS OF THE PRINCIPLE

OF LEGALITY

2.4.1 T H E P R IN C IP L E OF SP E C IF IC IT Y

Under the principle of specificity, criminal rules must be as detailed as possible, so as

to clearly indicate to their addressees the conduct prohibited, namely, both the object­
ive elements of the crime and the requisite mens rea. The principle is aimed at ensuring
that all those who may fall under the prohibitions of the law know in advance which
specific behaviour is allowed or proscribed. They may thus foresee the consequences of
their action and freely choose either to comply with, or instead breach, legal standards
of behaviour. Clearly, the more accurate and specific the crim inal rule, the greater is
the protection accorded to the agent from arbitrary action of either enforcement offi­
cials or courts of law.
The principle is still far from being fully applicable in international law, which still
includes many rules that are loose in their scope and purport. In this regard, suffice it
to mention, as an extreme or most conspicuous instance, the provision first enshrined
in the London Charter of 1945 and then restated in many international instrum ents
(Control Council Law no. 10, the Statutes of the Tokyo Tribunal, the ICTY, the ICTR
and the SCSL), whereby crimes against hum anity encompass ‘other inhum ane acts’.20

18 See for instance Articles 1-8 of the ICTY Statute, as well as §29 of the UN Secretary-General’s Report
to the Security Council for the establishment of the Tribunal (S/25704) (‘It should be pointed out that, in
assigning to the International Tribunal the task of prosecuting persons responsible for serious violations
of international humanitarian law, the Security Council would not be creating or purporting to legislate
that law. Rather, the International Tribunal would have the task of applying existing international
humanitarian law’).
19 On this principle see, among other decisions by the ICTY, AC, Tadic Interlocutory Appeal 1995, §92;
Delalic and others (TJ, 1998), §§402-7; Jelisic, TJ, §61; Hadzihasanovic, Alagic and Kubura, Decision on Inter­
locutory Appeal Challenging Jurisdiction in Relation to Command Responsibility, AJ, §§32-6.
20 The ICC Statute fleshes the notion out, by providing that crimes against humanity may include 'other
inhumane acts of a similar character [to the other, specifically enumerated, classes of such crimes] inten­
tionally causing great suffering, or serious injury to body or to mental or physical health (Art. 7(l)(k)).
 42 INTERNATIONAL CRIMINAL LAW

Similarly, the provisions of the four 1949 Geneva Conventions on grave breaches
among other things enumerate, as grave breaches’, ‘torture or inhuman treatment’.
In addition, many rules contain notions that are not defined at the ‘legislative’ level,
such as ‘rape’, ‘torture’, ‘persecution’, ‘enslavement’, etc. Furthermore, most inter­
national rules proscribing conduct as criminal do not specify the subjective element of
the crime. Nor are customary rules on defences crystal clear: they do not indicate the
relevant excuses or justifications in unquestionable terms.
Given this indeterminacy and the consequent legal uncertainty for the possible address­
ees of international criminal rules, the contribution of courts to giving precision to law,
not infrequent even in civil law systems, and quite normal in common law countries,
becomes of crucial importance at the international level, as has already been pointed out
above (1.4.7). Both national and international courts play an immensely important role
in gradually clarifying notions, or spelling out the objective and subjective ingredients of
crimes, or better outlining such general legal concepts as excuses, justifications, etc.
Thus, for instance, the District Court of Tel Aviv, in Ternek spelled out, by way of
construction, the notion of other inhum ane acts’ in a m anner that seems acceptable
(at 540, and §7).21 Similarly, in defining the concept of ‘rape’ a TC of the ICTY in
Furundzija had recourse to general principles of ICL as well as general principles com­
mon to the major legal systems of the world, and general principles of law.22
One should not underestimate, however, another drawback of ICL: the lack of a cen­
tral criminal court endowed with the authority to clarify for the whole international
community the numerous hazy or unclear criminal rules. To put it differently: the
contribution of courts to the gradual specification and precision of legal rules, empha­
sized above, suffers from the major shortcoming that such judicial refinement is

- 1 the Court stated that: 'The defence counsel argue, secondly, that the words “other inhumane acts”
which appear in the definition of “crimes against humanity” should be interpreted subject to the principle of
ejusdem generis. That is, that an "other inhumane act” should be of the type of the specific action mentioned
e ore it, in the same definition, which are “murder, extermination, enslavement, starvation and deport­
ation [ .] We believe that there is truth in the defence counsels second claim. The punishment determined
in Article 1 of the [Israeli] Law [of 1950 on the Doing of Justice to Nazis and their Collaborators] for “crimes
against humanity” is death (subject to extenuating circumstances pursuant to Article 11(b) of the Law),
and it can be assumed that the legislator intended to inflict the most extreme punishment known to the
penal code only for those inhumane actions which resemble in their type and severity “murder, extermin­
ation, enslavement, starvation and deportation of a civilian population”. If we measure by this yardstick the
actions proven against the defendant [beating with bare hand other detainees and making detainees kneel
m the Concentration camp of Auschwitz-Birkenau, where the defendant herself was an inmate, with the role
of custodian of Block 7] we shall find that even if some of these actions could be considered inhumane from
known aspects, they do not, under the circumstances, reach the severity of the actions which the legislator
intended to include in the definition of “crimes against humanity” in Article 1 of the Law’ (§7).
- It is worth citing the relevant passage, for that TC proved alert to the principle of specificity. It stated
the following: ‘This TC notes that no elements [for defining rape] other than those emphasised may be drawn
from international treaty or customary law, nor is resort to general principles of international criminal law
or to general principles of international law of any avail. The TC therefore considers that, to arrive at an
accurate definition of rape based on the criminal law principle of specificity (Bestimmtheitgrundsatz, also
referred to by the maxim "nullum crimen sine lege stricta"), it is necessary to look for principles of criminal
aw common to the major legal systems of the world. These principles may be derived, with all due caution,
im m n a t i n r m l ln .T o ’ /<C1r7'7'\
 GENERAL PRINCIPLES OF INTERNATIONAL CRIMINAL LAW 43

‘decentralized’ and fragmentary. In addition, when such process is effected by national

courts, it suffers from the another flaw: each court tends to apply the general notions
of criminal law proper to the legal system within which such court operates. Hence,
the possibility frequently arises of a contradictory and cacophonic interpretation or
application of international criminal rules.
Fortunately, the draftsmen of the ICC Statute made a significant contribution, when
they endeavoured to define as precisely as possible the various categories of crimes.
(However, as the Statute is not intended to codify international customary law, one ought
always to take it with a pinch of salt, for in some cases it may go beyond existing law,
whereas in other instances it is narrower in scope than current rules of customary inter­
national law. Furthermore, formally speaking that Statute is only binding on the ICC.)
For the time being, internation al criminal rules still make up a body of law in need
of legal precision and some major refinement at the level of definitions and general
principles. To take account of these features and at the same time safeguard the right of
the accused, currently some notions play a role that is far greater than in most national
systems: the defence of mistake o f law (see infra, 13.5), the principle of strict interpret­
ation (barring extensive or broad constructions of criminal rules; see infra, 2.4.3), the
principle of favor rei (imposing that in case of doubt a rule should be interpreted in
such a m anner as to favour the accused; see infra, 2.4.4). These notions act as counter­
vailingfactors, aimed at compensating for the present flaws and lacunae of ICL.

2.4 .2 T H E P R IN C IP L E OF N O N -R E T R O A C T IV IT Y

A. G eneral
As stated above, a logical and necessary corollary of the doctrine of strict legality is that
criminal rules (that is, rules criminalizing certain classes of hum an conduct) may not
cover acts performed prior to their enactment, unless such rules are more favourable to
the accused. Otherwise the executive power, the judiciary, or even the legislature could
arbitrarily punish persons for actions that were allowed when they were carried out.
In contrast, the ineluctable corollary of the doctrine of substantive justice is that,
for the purpose of defending society against new and unexpected forms of crim inal­
ity, one may go so far as to prosecute and punish conduct that was legal when taken.
These two approaches lead to contrary conclusions. The question is: which approach
has been adopted in international law?
It seems indisputable that the London Agreement of 1945 provided for two categor­
ies of crime that were new: crimes against peace and crimes against humanity. The
IMT did act upon the Charter provisions dealing with both categories. In so doing, it
applied ex post facto law; in other words, it applied international law retroactively, as
the defence counsel at Nuremberg rightly stressed.23

23 Seethe Motion adopted by all defence counsel on 19 November 1945, in Trial of the Major War Criminals
Before the International Military Tribunal, Nuremberg 14 November 1945-1 October 1946 (Nuremberg,
1947), vol. I, at 168-9.
 44 INTERNATIONAL CRIMINAL LAW

Many tribunals sitting in judgment over Germans in the aftermath of the Second
World War,24 as well as the German Supreme Court in the British Occupied Zone,25
endorsed the legal approach taken by the IMT, for all its deficiencies. This stand, while
having scant persuasive force with regard to the past, nonetheless contributed to the
slow consolidation of the principle of non-retroactivity in ICL.
Subsequently, as a logical consequence of the emergence of the nullum crimen
sine lege principle a general rule prohibiting the retroactive application of criminal
law gradually evolved in the international community. Thus, the principle of non-
retroactivity of criminal rules is now solidly embedded in ICL. It follows that courts
may only apply substantive criminal rules that existed at the time of commission of
the alleged crime. This, of course, does not entail that courts are barred from refin­
ing and elaborating upon, by way of legal construction, existing rules. The ICTY AC
clearly set out this notion in Aleksovski (AJ).26

B. Expansive adaptation of some legal ingredients of crimes laid down in

international rules to newsocial conditions
One should duly take account of the nature ofICL, to a large extent made up of customary
rules that are often identified, clarified or spelled out, or given legal determinacy by
courts. In short, that body of law to a large extent consists of judge-made law (with
no doctrine of precedent). Consequently, one should reconcile the principle of non­
retroactivity with these inherent characteristics ofICL. In this respect some im port­
ant rulings of the European Court of Hum an Rights may prove of great assistance. In
particular, in CR v. United Kingdom27 the Court held that the European Convention

24 See in particular the Justice case (at 974-85), Einsatzgruppen (at 458-9), Flick and others (at 1189),
Krauch and others (I. G. Farben case) (at 1097-8, 1125), Krupp (at 1331), High Command (at 487), Hostages
(at 1238-42).
25 See the Bl. case (at 5), the B. and A. case (at 297), the H. case (at 232-3), the N. case (at 335), and Angek-
lagterH. (at 135).
After commenting on the significance and legal purport of the nullum crimen principle, the AC added
that this principle does not prevent a court, either at the national or international level, from determining an
issue through a process of interpretation and clarification as to the elements of a particular crime; nor does
it prevent a court from relying on previous decisions which reflect an interpretation as to the meaning to be
ascribed to particular ingredients of a crime’ (§127).
In 1989 a British national went back to see his estranged wife, who had been living for some time with
her parents, and attempted to have sexual intercourse with her against her will; he also assaulted her, squeez­
ing her neck with both hands. He was charged with attempted rape and assault occasioning actual bodily
harm, and convicted. Before the European Court he repeated the claim already advanced before British
courts, that at the time when the facts occurred, marital rape was not prohibited in the UK. Indeed, at that
time a British Statute only prohibited as rape sexual intercourse with a woman who did not consent to it if
such intercourse was ‘unlawful’ (see section 1(1) of the Sexual Offences (Amendment) Act 1976); hence the
question turned on determining whether forced marital intercourse was unlawful’. Various English courts
had ruled, until 1990, that a husband could not be convicted of raping his wife, for the status of marriage
involved that the woman had given her consent to her husband having intercourse with her during the sub­
sistence of the marriage and could not unilaterally withdraw such consent. In contrast, Scottish courts had
first held that that view did not apply where the parties to a marriage were no longer cohabiting, and then
ruled, in 1989, that the wife’s implied consent was a legal fiction, the real question being whether as a matter
of fact the wife consented to the acts complained of. The word unlawful’ in the Act referred to above was
 GENERAL PRINCIPLES OF INTERNATIONAL CRIMINAL LAW 45

could not be read ‘as outlawing the gradual clarification of the rules of criminal liabil­
ity through judicial interpretation from case to case, provided that the resulting devel­
opment is consistent with the essence of the offence and could reasonably be foreseen
(§34).28 In a subsequent case, Cantoni v. France, the Court insisted on the notion that,
in order for criminal law (that is, a statutory provision or a judge-made rule) to be
in keeping with the nullum crimen principle, it is necessary for the law to meet the
requirements of accessibility and foreseeability. It added two im portant points. First,
a criminal rule may be couched in vague terms. W hen this happens, there may exist
‘grey areas at the fringe of the definition’:
This penumbra of doubt in relation to borderline facts does not in itself make a provi­
sion incompatible with Article 7 [of the European Convention on Human Rights, laying
down the nullum crimen principle], provided that it proves to be sufficiently clear in the
large majority of cases. The role of adjudication vested in the courts is precisely to dissipate
such interpretational doubts as remain, taking into account the changes in everyday
practice. (§33.)
The second point related to the notion of foreseeability. The C ourt noted that the
scope of this notion:
depends to a considerable degree on the content of the text in issue, the field it is designed to
cover, and the number and status of those to whom it is addressed [....] A law may still sat­
isfy the requirement of foreseeability even if the person concerned has to take appropriate
legal advice to assess, to a degree that is reasonable in the circumstances, the consequences
which a given action may entail [__] This is particularly true in relation to persons carrying
on a professional activity, who are used to having to proceed with a high degree of caution
when pursuing their occupation. They can on this account be expected to take special caie
in assessing the risk that such activity entails (§35).29
It would seem that the following legal propositions could be inferred from the
C ourt’s reasoning. First, while interpretation and clarification of existing rules is
always admissible, adaptation is only compatible with legal principles subject to strin­
gent requirements. Secondly, such requirements are that the evolutive adaptation, by
courts of law, of criminal prohibitions, namely the extension of such legal ingredients

deleted in 1994 by the Criminal Justice and Public Order Act. This being the legal situation in the UK, before
the European Court the applicant argued that the British courts had gone beyond a reasonable interpret­
ation of the existing law and indeed extended the definition of rape in such a way as to include facts that until
then had not constituted a criminal offence.
Both the European Commission and the European Court held instead that the British courts had not
breached Article 7(1) of the European Convention on Human Rights (‘No one shall be held guilty of any
criminal offence on account of any act or omission which did not constitute a criminal offence under
national or international law at the time when it was committed ).
28 See also S.W. v. United Kingdom, §§37-47.
29 In the case at issue the applicant was the owner of a supermarket, convicted of unlawfully selling
pharmaceutical products in breach of the Public Health Code. In his application he had contended that the
definition of medicinal product contained in the relevant provision of that Code was very imprecise and left
a wide discretion to the courts.
 46 INTERNATIONAL CRIMINAL LAW

of an offence as actus reus in order to cover conduct previously not clearly considered
as criminal must

(i) be in keeping with the rules of criminal liability relating to the subject matter,
more specifically with the rules defining ‘the essence of the offence’;
(ii) conform with, and indeed implement and actualize, fundamental principles
of ICL or at least general principles of law; and
(iii) be reasonably foreseeable by the addressees; in other words the extension,
although formally speaking it turns out to be to the detriment of the accused,
could have been reasonably anticipated by him, as consonant with general
principles of criminal law.30

To put it differently, courts may not create a new criminal offence, with new legal
ingredients (a new actus reus or a new mens rea). They can only adapt provisions
envisaging criminal offences to changing social conditions—as long as this adjustment
(resulting in the broadening of actus reus or, possibly, in lowering the threshold of the
subjective element, for instance, from intent to recklessness, or from recklessness to
culpable negligence) is consonant with, or even required by, general principles.
This process, particularly if it proves to be to the detriment of the accused (which is
normally the case) must presuppose the existence of a broad criminal prohibition (for
instance, the proscription of rape) and no clear-cut and explicit enumeration, in law,
of the acts embraced by this definition. It is in the penumbra left by law around this
definition that the adaptation may be carried out. Admittedly, the frontier between
such adaptation process and the analogical process, which is instead banned (see
below), is rather thin and porous. It falls to courts to proceed with great caution and
determine on a case-by-case basis whether the ‘adaptation’ under discussion is legally
warranted and consonant with general principles, and in addition does not unduly
prejudice the rights of the accused.
An instance ofthis process of adaptation of existing law can be seen in the judgment
delivered by the ICTY AC in Tadic (IA), where the AC unanimously held that some
customary 1 ules of international law criminalized certain categories of conduct in
internal armed conflict (see §§94-137).31 It is well known that until that decision many

30 The notions of foreseeability and accessibility were taken up by the ICTY AC in Hadzihasanovic and
0tk§34 t'DeCiSi0n ° n Inlerloculory appeal Challenging Jurisdiction in Relation to Command Responsibility).

31 Before pointing to practice and opiniojuris supporting the view that some customary rules had evolved
in the international community criminalizing conduct in internal armed conflict, the AC emphasized the
rationale behind this evolution, as follows: ‘A State-sovereignty-oriented approach has been gradually sup­
planted by a human-being-oriented approach. Gradually the maxim of Roman law hominum causa omne
jus constitutum est (all law is created for the benefit of human beings) has gained a firm foothold in the inter­
national community as well. It follows that in the area of armed conflict the distinction between interstate
wars and civil wars is losing its value as far as human beings are concerned. Why protect civilians from bel­
ligerent violence, or ban rape, torture or the wanton destruction of hospitals, churches, museums or private
property, as well as proscribe weapons causing unnecessary suffering when two sovereign States are engaged
in war, and yet refrain from enacting the same bans or providing the same protection when armed violence
 GENERAL PRINCIPLES OF INTERNATIONAL CRIMINAL LAW 47

commentators, states as well as the ICRC, had held the view that violations of the
hum anitarian law of internal armed conflict did not amount to war crimes proper, for
such crimes could only be perpetrated within the context of an international armed
conflict. The ICTY AC authoritatively held that the contrary was true and clearly iden­
tified a set of international customary rules prohibiting as criminal certain classes of
conduct. Since then this view has been generally accepted.
Similarly, contrary to the submission made by defence counsel in Hadzihasanovic
and others,32 an adaptation’ of existing rules (corroborated by a logical construction)
warrants the contention that persons may be held accountable under the notion of
command responsibility even in internal armed conflicts. Two arguments support
this proposition. First, generally speaking the notion is widely accepted in inter­
national hum anitarian law that each army or military unit engaging in fighting either
in an international or in an internal armed conflict must have a commander charged
with holding discipline, ensuring compliance with the law, and executing the orders
from above (with the consequence that whenever the commander culpably fails to
ensure such compliance, he may be called to account). The notion at issue is crucial
to the existence and enforcement of the whole body of international hum anitarian
law, because without a chain of command and a person in control of each military
unit, anarchy and chaos would ensue and no one could ensure compliance with law
and order. Secondly, and with specific regard to the Statute of the ICTY, Article 7(3)
of this Statute is couched in sweeping terms and clearly refers to the commission by
subordinates of any crime falling under the jurisdiction of the Tribunal: any time such
a crime has been perpetrated involving the responsibility of a superior, this superior
may be held accountable for criminal omission (of course, if he is proved to have the
requisite mens rea: see infra, 11.4.4). If this is so, it is sufficient to show that ciimes
perpetrated in internal armed conflicts fall under the Tribunal s jurisdiction, as held
in 1995 in Tadic (IA), for inferring that as a consequence the Tribunal has jurisdiction
over a commander who failed to prevent or punish such crimes.3"

2.4.3 T H E B A N O N A N A L O G Y

National courts (particularly in civil law countries) as well as international courts nor­
mally refrain from applying ICL by analogy; that is, they do not extend the scope and
purport of a criminal rule to a matter that is unregulated by law (analogia legis). In
national law the prohibition on the application of criminal rules by analogy (which

has erupted “only” within the territory of a sovereign State? If international law. while of course safeguard­
ing the legitimate interests of States, must gradually turn to the protection of human beings, it is only natural
that the aforementioned dichotomy should gradually lose its weight (§97).
32 See the ICTY TC Decision on joint Challenge to Jurisdiction, on 7 December 2001, §§15-39.
33 The notions set out in the text are to a large extent coincident with the rulings in Hadzihasanovic and
others made in 2002 by the ICTY TC (Decision on Joint Challenge to Jurisdiction), at §§150-79, and later, in
2003, by the AC (Decision on Interlocutory Appeal Challenging Jurisdiction in Relation to Command Respon­
sibility), at §§10-36.
 48 INTERNATIONAL CRIMINAL LAW

was not provided for in the German Nazi state or in the Soviet Union, and was banned
in China only in 1997, when a new Criminal Code was enacted) is rooted in the need
to safeguard citizens and in particular to prevent their being punished for actions that
were not considered illegal when they were performed. By the same token, the prohib­
ition is intended to narrow down arbitrary judicial decisions.
The same principle applies in international law. Its rationale is the need to protect
individuals from arbitrary behaviour of states or courts (which is another side, or a
direct consequence, of the exigency that no one be accused of an act that at the time
of its commission was not a criminal offence). In other words, the prim ary rationale
is to safeguard the rights of the accused as much as possible. To satisfy this require­
ment, analogy is prohibited with regard to both treaty and customary rules. Such rules
(for instance, norms proscribing certain specific crimes against humanity) may not be
applied by analogy to classes of acts that are unregulated by law.
Article 22(2) of the ICC. Statute thus codifies existing customary law where it
provides that ‘The definition of a crime shall be strictly construed and shall not be
extended by analogy. In case of ambiguity, the definition shall be interpreted in favour
of the persons being investigated, prosecuted or convicted.’ For example, one is not
allowed to apply by analogy the rule prohibiting a specific weapon (such as blinding
weapons) to a new weapon or, at any rate, to another weapon not prohibited. Nor
may one apply by analogy a rule prohibiting a particular use of a specific weapon (for
instance, the use of napalm and other incendiary weapons contrary to Protocol III to
the 1980 UN Convention on Conventional Weapons) to another use of that weapon.
Consequently, one is not allowed to criminalize the use of those weapons when their
use was permitted.
As the aforementioned provision of the ICC Statute makes clear, a prohibition
closely bound up with that of analogy is the ban on broad or extensive interpretation
ol international criminal rules, and the consequent duty for states, courts, and other
relevant officials and individuals to resort to strict interpretation. This principle entails
that one is not allowed to broaden surreptitiously, by way of interpretation, the scope
ot i ules criminalizing conduct, so as to make them applicable to instances not specif­
ically envisaged by those rules.
An example of strict construction can be found in some post-Second World War
cases relating to the notion of crimes against humanity. In Altstotter and others a US
M ilitary Tribunal sitting at Nuremberg held that that notion, as laid down in Control
Council Law no. 10,

must be strictly construed to exclude isolated cases of atrocity or persecution whether

committed by private individuals or by governmental authorities. As we construe it, that
section [of the aforementioned Law] provides for the punishment of crimes committed
against German nationals only where there is proof of conscious participation in systematic
governmentally organized or approved procedures, amounting to atrocities and offences
of that kind specified in the act and committed against populations or amounting to
persecution on political, racial or religious grounds (at 284-5).
 GENERAL PRINCIPLES OF INTERNATIONAL CRIMINAL LAW 49

The finding was cited with approval in Flick and others, handed down by another
US Military Tribunal sitting at Nuremberg (at 1216), where the Tribunal also held
that under a strict interpretation of the same notion, crimes against hum anity do not
encompass offences against property, but only those against persons (at 1215).34*
Three qualifications must, however, be set out restricting the ban on analogy.
First, international law only prohibits the so-called analogia legis (that is, the
extension of a rule so as to cover a matter that is formally unregulated by law). It
does not bar the regulation of a matter not covered by a specific provision or rule, by
resorting to general principles ofICL, or to general principles of criminal justice, or to
principles common to the major legal systems of the world (so-called analogia juris).
National and international courts or tribunals have repeatedly affirmed that it is
permissible to rely upon such principles for establishing whether an international rule
covers a specific matter in dispute. To be sure, the question has always been framed as
one of interpretation, rather than analogical application. Nevertheless, whatever the
terminology employed, the fact remains that gaps or lacunae have been filled by resort
to those principles. It should, however, be clear that drawing upon general principles
should never be used to criminalize conduct that was previously not prohibited by a
criminal rule. It may only serve to spell out and clarify, or give a clear legal contour to,
prohibitions that have already been laid down in either customary law or treaties. In
other words, this approach may only be resorted to for the interpretation of existing
rules, not for the creation of new classes of criminal conduct. To hold the contrary
would mean to admit serious departures from the nullum crimen principle, contrary
to the whole thrust of current ICL.
Secondly, in quite a few cases international rules themselves invite or request
analogy, through the ejusdem generis canon of statutory construction (whereby when
in a legal rule general words follow the enumeration of a particular class of persons or
things, the general words must be construed as applying to persons or things of the
same kind or class as those enumerated). For instance, the customary and treaty rules
prohibiting and penalizing as crimes against humanity other inhumane acts, as well
as the provisions of the 1949 Geneva Conventions criminalizing as grave breaches of
the Convention ‘inhum an acts’ in addition to torture, impose upon the interpreter the
need to look at acts and conduct analogies in gravity to those prohibited. This indeed
was the reasoning of the Tel Aviv District Court in Ternek.36 The draftsmen of the ICC
Statute took the same logical approach when they criminalized in Article 7(l)(k) ‘other
inhumane acts o f a similar character intentionally causing great suffering, or serious
injury to body or to mental or physical health’ (emphasis added).

34 Subsequently the Dutch Special Court of Cassation took up in Albrecht (at 397-8) and in Bellmer
(at 543), as well as in Haase (at 432), the same strict interpretation advanced in Altstötter and others.
33 Resort to general principles of law recognized by civilized nations is termed by Anzilotti (op. cit.,
106-7) analogia juris. It should be noted, however, that according to the celebrated international lawyer
those general principles did not constitute an autonomous source of international law.
36 See supra, n. 21.
 50 INTERNATIONAL CRIMINAL LAW

Thirdly, in some cases international law allows a logical approach that at first glance
runs foul of the ban on analogy, but which is in fact permissible because it applies
to general principles. An example will clarify this proposition. In the case of a new
weapon that does not fall under any specific prohibition precisely because of its novel
features, analogical extension of an existing treaty ban is not allowed, as pointed
out above. Nevertheless, one is authorized to enquire whether the new weapon is at
variance with the general principle proscribing weapons that are inherently indiscrim­
inate or cause unnecessary suffering. For this purpose, one may justifiably look at
those weapons that have been prohibited by treaty because they are either indiscrim­
inate or cause superfluous sufferings. The object of this enquiry will not be the appli­
cation of these treaty prohibitions by analogy, but rather to better ascertain whether
the characteristics of the new weapon are such as to make them contrary to the general
principle. It would seem that the District Court of Tokyo in Shimoda and others took
precisely this approach (although, of course, it had been requested to pronounce on a
question of civil liability, not of criminal law).37

2.4.4 T H E P R IN C IP L E O F F A V O U R IN G T H E A C C U S E D

Another principle is closely intertwined with the ban on analogy, and is designed to
invigorate it. This is the principle requiring, when faced with conflicting interpreta­
tions of a rule, the construction that favours the accused: see also ICC Statute, Article
22(2). An ICTR TC upheld this principle in Akayesu with regard to the interpretation
of the word ‘killing’ in the Genocide Convention and the Statute of the ICTR.38 An
ICTY TC reaffirmed the principle in Krstic. The question was how to interpret the
notion of extermination’ as a crime against humanity. The Chamber pointed out that

After noting that the use of an atomic bomb was ‘believed to be contrary to the principle of inter­
national law prohibiting means of injuring the enemy which cause unnecessary suffering or are inhuman,
the District Court of Tokyo noted that the bomb was a new weapon. It then pointed out that the employment
of asphyxiating, poisonous, and other gases and bacteriological methods of warfare was prohibited, noting
that it could safely be concluded that besides poisons, poisonous gases and bacteria, the use of means of
injuring the enemy which cause injury at least as great as or greater than these prohibited materials is pro­
hibited by international law’. The Court concluded that ‘it is not too much to say that the sufferings brought
about by the atomic bomb are greater than those caused by poisons and poisonous gases; indeed the act of
dropping this bomb may be regarded as contrary to the fundamental principle of the law of war which pro­
hibits the causing of unnecessary suffering’ (at 1694-5).
38 With regard to the word 'meurtre (in French) and ‘killing’ in English, contained in the phrase ‘killing
members of the group’ (as a category of genocide), the TC noted the following: ‘The TC is of the opinion that
the term “killing” used in the English version is too general, since it could very well include both intentional
and unintentional homicides, whereas the term “meurtre”, used in the French version, is more precise. It is
accepted that there is murder when death has been caused with the intention to do so, as provided for, inci­
dentally, in the Penal Code of Rwanda, which stipulates in its Article 311 that “Homicide committed with
intent to cause death shall be treated as murder”. Given the presumption of innocence of the accused, and
pursuant to the general principles of criminal law, the Chamber holds that the version more favourable to
the accused should be upheld and finds that Article 2(2)(a) of the Statute must be interpreted in accordance
with the definition of murder given in the Penal Code of Rwanda, according to which “meurtre” (killing) is
homicide committed with the intent to cause death’ (§§500-1).
 GENERAL PRINCIPLES OF INTERNATIONAL CRIMINAL LAW 51

the ICC Statute provides that extermination may embrace acts calculated to bring
about the destruction of part of the population, namely only a limited num ber of
victims; it stressed that under customary law extermination generally involves a large
number of victims. It went on to hold as follows:
this definition [that is, that contained in the ICC Statute] was adopted after the time the
offences in this case were committed. In accordance with the principle that where there is
a plausible difference of interpretation or application, the position which most favours the
accused should be adopted, the Chamber determines that, for the purpose of this case, the
definition should be read as meaning the destruction of a numerically significant part of
the population concerned (§502).
It should be noted that the principle of construction in favour of the accused (favor
ret) has also been conceived of as a standard governing the appraisal of evidence: in
this case the principle is known as in dubio pro reo (in case of doubt, one should hold
for the accused). For instance, in Flick and others, a US Military Tribunal sitting at
Nuremberg held that it must be guided among other things by the standard whereby
‘If from credible evidence two reasonable inferences may be drawn, one of guilt and
the other of innocence, the latter must be taken (at 1189).39 The notion was also upheld
in Stakic.40

2.5 THE PR IN C IPL E OF LEGALITY OF PENALTIES

It is common knowledge that in many states, particularly in those of Romano-

Germanic tradition, it is considered necessary to lay down in law a tariff relating to
sentences for each crime, so as: (i) to ensure the uniform application of criminal law by
all courts of the state; and (ii) to make the addressees cognizant of the possible punish­
ment that may be meted out if they transgress a particular criminal provision.
This principle is not applicable at the international level, where these tariffs do
not exist. Indeed, states have not yet agreed upon a scale of penalties, due to widely
differing views about the gravity of the various crimes, the seriousness of guilt for
each criminal offence, and the consequent harshness of punishment. It follows that
courts enjoy much greater judicial discretion in punishing persons found guilty of
international crimes. However, some statutes of international tribunals set forth
limitations on the absolute discretion of judges. Thus, for instance, Article 24(1) of
the ICTY Statute provides, first, that penalties will be limited to imprisonment (thus
ruling out the death sentence), and, secondly, that ‘In determining the terms of

39 Another US Military Tribunal sitting at Nuremberg upheld the principle in Krauch and others
(.I. G. Farben case) (at 1108).
40 ‘The TC explicitly distances itself from the Defence submission that the principle in dubio pro reo
should apply as a principle for the interpretation of the substantive criminal law of the Statute. As this prin­
ciple is applicable to findings of fact and not of law, the TC has not taken it into account in its interpretation
of the law’ (TJ, §416).
 52 INTERNATIONAL CRIMINAL LAW

imprisonment, the Trial Chambers shall have recourse to the general practice regard-
ing prison sentences in the courts of the former Yugoslavia.’ This last provision was
applied in various cases,41 although it was generally not held mandatory. Article 23 of
the ICTR Statute is identical, but it refers of course to the general practice regarding
prison sentences in the courts of Rwanda.
As for the Statute of the ICC, Article 23 provides that A person convicted by the
C ourt may be punished only in accordance with this Statute’ and Article 77 confines
itself to envisaging imprisonment for a maximum of 30 years, while at the same time
adm itting life imprisonment ‘when justified by the extreme gravity of the crime and
the individual circumstances of the convicted person’. It thus implicitly rules out
the death penalty, but does not establish a scale of sentences, nor does it suggest that
the Court should take into account the scale of penalties of the relevant territorial or
national state. The Court is thus left with a very broad margin of appreciation.

41 See, for instance, in Erdemovic and Tadic‘ (Sentencing J. 1997) (§§7-10), Tadic (Sentencing J. 1999)
(§§10 13), Delaliif and others (§§1193-212), and Kupreskic and others (§§839-47).
 3
THE ELEMENTS OF
INTERNATIONAL CRIMES

3.1 THE OBJECTIVE AND SUBJECTIVE

ELEMENTS OF CRIMES

In any legal system, crimes consist of two elements: (i) conduct (an act or omission,
contrary to a rule imposing a specific behaviour; this is called actus reus, that is a culp­
able act); and (ii) a state of mind, a psychological element required by the legal order
for the conduct to be blameworthy and consequently punishable (also called culpable
frame of m ind or mens rea).
In international law also, there exist rules prescribing that individuals (whether
acting as state officials or as private persons) take a certain conduct (for instance, they
must refrain from killing civilians or from injuring prisoners of war in an armed con­
flict, or from engaging in large-scale torture of persons held in detention, or from
murdering a multitude of persons belonging to a certain ethnic, national, religious,
or racial group). As in national legal systems, also in international law, conduct con­
trary to a substantive rule of this corpus of law is not sufficient for individual criminal
responsibility to arise. A mental element is also required, in some way directed to or
linked with the commission of the criminal act.

3.2 THE STRUCTURAL ELEMENTS OF

INTERNATIONAL CRIMES

We will consider the specific acts, actions, or omissions falling under the notion of
actus reus when we move on to examine the various classes of international crime. It
is necessary now to dwell on some general notions relating to the essential structure
of such crimes.
Two main features characterize international crimes proper.1 First, they consist of
conduct taken or acts performed by either (i) state officials (for instance, servicemen

1 See on this matter P. Gaeta, ‘International Criminalization of Prohibited Conduct’, in A. Cassese (ed.),
Oxford Companion to International Criminal Justice (forthcoming).
 54 INTERNATIONAL CRIMINAL LAW

engaged in war, or political leaders planning or ordering genocide, etc.); or (ii) private
individuals.
What is notable is that this conduct is either (a) linked to an international or internal
armed conflict or, absent of such a conflict, (b) has a political or ideological dimension,
or is somehow linked or otherwise connected to (instigated, influenced, tolerated, or
acquiesced in) the behaviour of state authorities or organized non-state groups or
entities.
Thus, it is characteristic of such crimes that, when perpetrated by private individu­
als, they are somehow connected with a state policy or at any rate with ‘system
crim inality’.2 On this score international crimes are thus different from criminal
offences committed for personal purposes (private gain, satisfaction of personal greed,
desire for revenge, etc.) as is the case with ordinary criminal offences such as theft,
robbery, assault, kidnapping for extorting a ransom, etc., or such other crimes that
have a transnational dimension but pursue private goals, such as piracy, slave trade,
trade in women and children, counterfeiting currency, drug dealing, etc.
ITie fundamental hallmark of international crimes, which I have just highlighted, is
also called ‘the international element’ or ‘a context of organized violence’ of such crimes.3
The second notable feature of international crimes, inextricably intertwined with
the one I have just emphasized, is that they normally possess a twofold dimension
or are double-layered. They constitute criminal offences in domestic legal systems:
serious bodily harm, murder, rape, sexual assault, torture, persecution, etc., in that
they infringe municipal rules of criminal law. In addition, they have an international
dimension, in that they breach values recognized as universal in the world commu­
nity and enshrined in international customary rules and treaties. It follows that nor­
mally these crimes consist of an ‘underlying offence’ (for example, m urder or torture)
with the requisite objective and subjective elements of such offence, plus an objective
and mental element required by the international rules that contemplate the crime at
issue. For instance, we will see {infra 5.3-4) that murder as a crime against humanity
requires (i) the objective element of m urder (causing the death of another person) as
well as a mental element (intent to bring about by one’s action the death of another
person); plus (ii) a broader objective context (the existence of a widespread or system­
atic attack on the civilian population, whether in time of armed conflict or in time of
peace) and a mental element: awareness of the existence of such broader context.
These features relate to the vast majority of cases. There are, however, also crimes
that do not possess this double dimension, in that they do not encompass an under-
lying criminal offence. For instance, the use of prohibited weapons in time of war or
the indiscriminate attack of civilians in an internal armed conflict is per se an inter­
national crime, without necessarily having a ‘domestic’ underpinning. It follows that

The notion of system criminality as opposed to individual criminality was set out by the great Dutch
scholar and judge B.V.A. Röling, ‘The Law of War and the National Jurisdiction Since 1945’ 100 Hague
Recueil, 1960-11,335ff; see also ‘The Significance of the Laws of War’, in A. Cassese (ed.), Current Problems of
International Law (Milano: Giuffre, 1975), 137-9.
3 G. Werle, Principles of International Criminal Law (The Hague: T. M. C. Asser Press, 2005), 94-5.
 THE ELEMENTS OF INTERNATIONAL CRIMES 55

what is required for the crime to be perpetrated is a conduct defined in international

rules (for example, using a weapon that is proscribed by international law) as well as a
mental element (the intent to use the weapon). The same holds true for a sub-category
of crimes against humanity, namely persecution (see infra, 5.1 and 5.3, sub. 8).
Furthermore, as is normally the case in domestic legal systems with all criminal
offences, international crimes also can be split into conduct, consequences, and cir­
cumstances, from the point of view of their objective structure.1
The conduct is described by the international rule that imposes a certain behaviour
(for instance, respect civilians in a civil war, or protect prisoners of war in an inter­
national armed conflict) and therefore criminalizes any act or omission contrary to
such a rule. Consequences are the effects caused by one’s conduct. Between conduct
and consequences there is, of course, a causation nexus: for instance, I fire a missile
at a hospital and thus bring about the destruction of the building and the death of
dozens of civilians and wounded persons. From this point of view crimes may be held
to belong to two different categories: crimes of conduct and crimes of result. The for­
mer category comprises offences consisting in the breach of an international rule that
imposes a specific behaviour; there, it is irrelevant whether or not this breach brings
about any harm or injury to prospective victims. (Think, for instance, of the rule that
obliges belligerents to refrain from declaring that no quarter will be given; that is,
that in combat operations enemies will not be captured but will be killed, even if they
surrender; the same holds true for the rule prohibiting the use of a certain means of
warfare, for instance dum dum bullets or chemical or bacteriological weapons; the use
of these weapons constitutes a breach of IHL and a war crime, even if in a specific case
no damage to the adversary is in fact caused by such use.) Crimes of result embrace
violations of rules that confine themselves to imposing the achievement of a certain
end, regardless of the modalities for the realization of that end; for instance, causing
disproportionate casualties among civilians when attacking a military objective, oi
starving prisoners of war.
Consequences of a crime are the effects of criminal conduct. Most international
crim inal rules focus on the harm caused by hum an behaviour and proscribe conduct
that is such as to bring about such harm: for instance, they criminalize the killing of
civilians, the wounding of prisoners of war, the rape of women. Hie rationale behind
this emphasis is that the prim ary goal of international criminal law is to prevent and
punish behaviour that injures protected persons. On this score consequences are pai -
ticularly relevant with regard to crimes of result’, as defined above.
Circumstances are ‘any objective or subjective facts, qualities or motives with regard
to the subject of the crime (such as the perpetrator and any accomplices), the object of
the crime (such as the victim or other impaired interests) or any other modalities of
the crime (such as means or time and place of commission) .45 Thus, for instance, in the

4 See A. Eser,’ Mental Element—Mistake of Fact and Mistake of Law’ in Cassese, Gaeta, Jones, ICC Com­
mentary, vol. 1, 911-20.
5 A. Eser, op. cit, 919.
 56 INTERNATIONAL CRIMINAL LAW

rule imposing on m ilitary commanders and, more generally, on superior authorities

to prevent and repress crimes by their subordinates, one of the circumstances of the
crime is that a person is a military commander or a civilian superior. Similarly, in the
rule banning as a crime against hum anity the ‘forcible transfer of population’, the use
ot force in bringing about the movement of a multitude of civilians from one place to
another is a circumstance’ required by the rule.

3.3 GENERAL FEATURES OF THE

SUBJECTIVE ELEMENT
It IS not easy to identify the various forms and shades of the mental element in ICL
Two problems arise.
First, substantive rules concerning crimes often do not specify the subjective
element required for each specific offence. An exception may be found in the various
substantive provisions of the ICC Statute: Articles 6 (on genocide), 7 (on crimes against
humanity), and 8 (on war crimes), and the accompanying ‘Elements of Crime’ elabo­
rated pursuant to Article 9. Most of the time these provisions set out the subjective
dem ent required for each class of crime. However, in this respect the provisions of
he Statute are hedged about with two major limitations. They are only designed to
set out the categories of crime over which the ICC may exercise jurisdiction; in other
words they are not couched as provisions of a criminal code. Furthermore, they are
not intended to codify, or restate, or contribute to the development of customary
international law. Their legal value is therefore limited (although, of course, they may
gradually have a bearing on, and bring about a change in, existing law).
Secondly, there is no customary rule setting out a general definition of the various
categories of mens rea (such as intent, recklessness, or negligence). In this respect the
only exception is Article 30 of the ICC Statute, on ‘mental element’. However, it is
doubtful that it reflects customary international law. In addition, as we shall see, even
at the level of treaty law, it is not certain that it encompasses all the various possible
subjective elements of international crimes.
This difficult condition is compounded by the failure of national case lawto castlight
on the matter It is state courts that have handed down the bulk of judicial decisions
dealing with this matter, and each court has applied the rules of criminal law proper to
its own domestic system. Depending on the legal tradition to which it belonged, each
court has placed its own interpretation on the notion of intent, fault, or negligence
Consequently, to tackle the first of the two problems outlined above, one should first
identify all the international substantive provisions which themselves lay down, if only
implicitly, the subjective element required for their violation to am ount to an inter­
national crime. One ought also to draw upon the case law of international tribunals to
the extent that they have pronounced on the matter. To come to grips with the second
 THE ELEMENTS OF INTERNATIONAL CRIMES 57

nroblem, one must start from the assumption that here, as in other fields of ICL, what
matters is to identify the possible existence of general rules of international law or in
the absence of such rules, principles common to the major legal systems of the wor
To pinpoint such rules, one may chiefly rely on: (i) the case law of courts, with specia
attention being paid to the judicial decisions of international tribunals, in particular
the ICTY and the ICTR (these decisions have in fact proved to be of crucial importance
in the gradual elaboration of the various mental elements of each category of inter­
national crime); and (ii) the existence of some basic notions common to all major lega
systems of the world, as evidence of a convergence of these systems and confirmation
that parallel principles have also taken shape at the international level.
I shall briefly mention some instances of how the first of the two problems is some­
times solved. I shall then concentrate on the second problem; that is, the general defi­
nitions of the various categories of subjective element that one may deduce from a
perusal of international rules and the relevant case law.

3.4 SUBSTANTIVE RULES SETTING OUT THE

MENTAL ELEMENT REQUIRED FOR CRIMES

With regard to such substantive rules, one may recall, as major illustrations, a set
of im portant treaty provisions. The first is Article 2 of the 1948 Genocide Conven­
tion (now turned into an international customary rule), whereby genocide as an
international crime requires that there be The intent to destroy, in whole or in part, a
national, ethnical, racial or religious group, as such’.
Similarly Article 1 of the 1984 Convention on Torture prohibits torture when it
is, among other things, ‘intentionally inflicted’. Although the Convention belongs to
a different body of law, namely hum an rights law, some elements of its definition of
torture have been incorporated into international hum anitarian law, as IC 1Y ICs
held in Delalic and others (§§452-60), Furundzija (§§143-59), and Kunarac and others
(§§465-97). By the same token, Article 7 (l)(k) of the ICC Statute defines as crimes
against humanity ‘other inhumane acts’ of a character similar to that of the crimes
against hum anity the same provision enumerates before, it they ‘intentionally cause
great suffering or serious injury to body or to mental or physical health. Plainly, al
these provisions require intentional conduct, thereby automatically excluding any
other subjective frame of m ind such as recklessness, negligence, etc.
Furthermore, Article 85(3) and (4) of the First Additional Protocol of 1977 makes
punishable a host of violations of the Protocol so long as they are committed w i-
fully’. Under the interpretation of this adverb authoritatively suggested in the ICKC
Commentary, the word ‘wilful’ implies that
t h e a c c u s e d m u s t h a v e a c t e d c o n s c i o u s l y a n d w i t h i n t e n t , i.e . w i t h h i s m i n d o n t h e a c t
a n d it s c o n s e q u e n c e s , a n d w i l l i n g t h e m (‘c r i m i n a l i n t e n t ’ o r ‘m a l i c e a f o r e t h o u g h t ); [th is ]
 58 INTERNATIONAL CRIMINAL LAW

encompasses the concepts o f‘wrongful intent’ or ‘recklessness’, viz., the attitude of an agent
who, without being certain of a particular result, accepts the possibility of it happening
(§3474, p. 995).

In other words, under this interpretation those violations of the Protocol may entail
the criminal liability of the perpetrator if they are committed either intentionally or
with dolus eventualis: on this notion, see infra 3.7).

3.5 GENERAL NOTIONS OF MENS REA COMMON TO

MOST LEGAL SYSTEMS OF THE WORLD
Byway of introduction, it may prove fitting to undertake a brief comparative survey of
the attitude taken towards the definition of the major facets of mens rea by the major
legal systems of the world. It is apparent that, in spite of broad differences in term in­
ology, most legal systems tend to take the same basic approach to the specific regula­
tion of each aspect of mens rea, and its implications. They tend to require one of the
following frames of mind, for conduct to be considered criminally punishable (these
are listed in decreasing order of culpability):

1. Intention, namely awareness that a certain conduct will bring about a certain
result in the ordinary course of events, and will to attain that objective: for example, I
use a gun to shoot at a person because I want to cause his death and anticipate that as a
consequence of my shooting he will die. This class of mens rea is normally called intent
(dol direct, Vorsatz, dolus directus).
2. Awareness that undertaking a course of conduct carries with it an unreasonable
or unjustifiable risk of producing harm ful consequences, and the decision nevertheless
to go on to take that risk. For instance, I perceive the risk that using a certain weapon
may entail killing dozens or even hundreds of innocent civilians, and nevertheless
willingly ignore this risk. This class is normally called recklessness (or dol eventuel,
Eventualvorsatz (or Eventualdolus, or bedingter Vorsatz), dolus eventualis).6
3. Failure to pay sufficient attention to or to comply with certain generally accepted
standards of conduct thereby causing harm to another person when the actor believes
that the harmful consequences of his action will not come about, thanks to the meas­
ures he has taken or is about to take. For instance, an attendant at a mental hospital

Under Art. 2(2)(c) of the US Model Penal Code, ‘A person acts recklessly with respect to a material element
of an offense when he consciously disregards a substantial and unjustifiable risk that the material element exists
or will result from his conduct. The risk must be of such a nature and degree that, considering the nature and
purpose of the actor’s conduct and the circumstances known to him, its disregard involves a gross deviation from
the standard of conduct that a law-abiding person would observe in the actor s situation5(emphasis added).
On dolus eventualis, see in particular G. Fletcher, Rethinking Criminal Law, 1978 reprint (Oxford- Oxford
University Press, 2 0 0 0 ), 445-9.
 THE ELEMENTS OF INTERNATIONAL CRIMES 59

causes the death of a patient by releasing a flow of boiling water into the bath; one
of two persons playing with a loaded gun points it at the other and pulls the trigger
believing that it will not fire because neither bullet is opposite the barrel; however, as
the gun is a revolver, it does fire, killing the other person.7
This class is normally referred to as advertent or culpable negligence (negligence
consciente, bewusste Fahrlässigkeit) where the agent s conduct seriously or blatantly
fails to meet the standards of the reasonable man test.8
4. Failure to respect generally accepted standards of conduct without, however,
being aware of or anticipating the risk that such failure may bring about harm ful effects.
To prevent road accidents, some countries envisage this state of mind for drivers who
act negligently (for instance, cause the death of a pedestrian by not stopping at the stop
sign, or by driving at excessive speed or in a state of intoxication).
This class is normally termed inadvertent negligence (negligence inconsciente,
unbewusste Fahrlässigkeit).
These are, of course, only general trends of national criminal law. The courts of some
states often do not draw such a fine distinction between the aforementioned shades
on the scale of criminal culpability.9 Similarly, national laws or military manuals

7 See A. Ashworth, Principles of Criminal Law, 5th edn (Oxford: Oxford University Press, 2006), 191-5.
See also A. P. Simester and G. R. Sullivan, Criminal Law—Theory and Doctrine (Oxford: Hart, 2002), 139-40.
According to D. L. Hart (‘Negligence, Mens Rea and Criminal Responsibility', in Punishment and Respon­
sibility, Oxford: Oxford University Press, 1968, at 149), ‘Negligence is gross if the precautions to be taken
against harm are very simple, such as persons who are but poorly endowed with physical and mental capaci-
ties can easily take.’ A. P. Simester and G. R. Sullivan (at 140) provide a telling example: ‘It may be negligent
to drive around a particular bend at 50 mph; if so, it is grossly negligent to do so at 80 mph. It will also be
gross negligence if the risk created by the defendant is very obvious.
8 Under Art. 2(2) (d) of the US Model Penal Code, ‘A person acts negligently with respect to a material element
of an offense when he should be aware ofa substantial and unjustifiable risk that the material element exists or will
result from his conduct. The risk must be of such a nature and degree that the actor’s failure to perceive it, consid­
ering the nature and purpose of his conduct and the circumstances known to him, involves a gross deviation from
the standard of care that a reasonable person would observe in the actor’s situation (emphasis added).
9 For instance, in 1975 in Robert Strong the Court of Appeals of New York held that, from the point of view
of the mental state of the defendant at the time the crime was committed, the essential distinction between
the crime of ‘manslaughter in the second degree’ (that is, recklessly causing the death of a person, or inten­
tionally causing or aiding a person to commit suicide, or committing upon a female an abortion causing her
death), and ‘criminally negligent homicide’ (that is, causing the death of a person with criminal negligence)
is that in the former class of crime ‘the actor perceives the risk but consciously disregards i t , whereas in the
latter the actor ‘negligently fails to perceive the risk’. In the case at issue the accused, a leader of a Muslim
sect with a sizeable following, purportedly exercising his powers of ‘mind over matter’ used to perform
ceremonies such as walking though fire, performing surgical operations without anaesthesia, or stopping a
follower’s heartbeat and breathing while he plunged knives into his chest without any injury to the person.
Although he had performed this last-mentioned ceremony countless times without once causing an injury,
in the case brought before the court the follower had died as a result of the wounds. The jury found that the
defendant was guilty of manslaughter in the second degree, as charged, without considering whether he
could have been guilty of the lesser crime of criminally negligent homicide. The Court of Appeals held that
in this case the jury could have found that the defendant ‘failed to perceive the risk inherent in his actions
[...] The defendant’s conduct and claimed lack of perception, together with the belief of the victims and the
defendant’s followers, if accepted by the jury, would justify a verdict of criminally negligent homicide’ rather
than manslaughter in the second degree (568-9).
 6o INTERNATIONAL CRIMINAL LAW

may set out notions that do not necessarily fit in the above enumeration of forms of
mens rea.10
Depending on the category of crime and the degree of responsibility, international
customary rules (resulting from opinio juris seu necessitatis, i.e. the conviction that a
certain behaviour is necessary or is dictated by a legal rule, and international practice,
as evidenced by case law, treaty provisions if any, the views of state officials, and the
convergence of the major legal systems of the world) envisage various modalities of
the mental element. As mentioned above, the ICC Statute includes a provision, Article
30, that specifically deals with this matter (see infra, 3.9). However, this provision has
a limited purport, for it only applies to the crimes falling under ICC jurisdiction and
m addition does not reflect or codify customary rules. It therefore may not apply to
othei international courts or tribunals, which are bound either by their own Statute
or, if such Statutes do not regulate the matter (which is indeed the case), by customary
international law. ’

3.6 GENERAL CATEGORIES OF MENS REA: INTENT

By intent or intention (dolus directus) is meant awareness that by engaging in a certain
action or by omitting to act I shall bring about a certain result (such as, for example, the
death of a civilian) coupled with the will to cause such result. For instance: I want to
kill a civilian. So I shoot him and he dies as a result of my act. I must therefore answer
for this crime. Or else, I think he is dead but in fact he has not died; he only dies later
of exposure because he is left in the cold. It does not matter that my conduct did not
kill him I am guilty of m urder because: (i) I intended him to die (mens rea); and
(u) he died as a result of my acts (because he would never have been lying exposed were
it not for my acts). As a rule, my intent only has to be linked to a certain result (the
death of the victim).
International rules require intent for most international crimes, although, as we
shall see, under certain circumstances other states of m ind are admissible.
As an illustration of intent, Enigster may be mentioned. The accused, a Jewish
internee in a Nazi concentration camp having the rank of Schieber or group leader, had
been charged with crimes against humanity, in particular, grievous injuries, against
his fellow inmates. In examining the alleged grievous attack on another inmate
named Schweizer, the District C ourt of Tel Aviv had to establish whether all the neces­
sary elements were present; it therefore asked itself, among other things, whether the
requisite intent also existed. It noted that in this respect no special testimony had been

Thus, for example, according to the Australian Defence Force Discipline Manual, ‘A person can be
said to have acted recklessly when he is aware that certain harmful consequences are likely to flow from a
paihcular act but he performs the act despite the risk. A person acts negligently when he performs an act
without consideration of the probably harmful consequences which will flow from it but where those harm-
tul consequences would be foreseeable by a reasonable man’ (§533).
 THE ELEMENTS OF INTERNATIONAL CRIMES 6l

brought to the Court; it nonetheless had to determine whether the accused had that
intent. The Court noted the following:
As to ‘intent’, it is a well known rule that any person in his right mind is held to intend
the natural consequences of his actions. As it appears from the severe results of the blows
struck by the defendant, the blows were landed with some significant force, and for this
reason, and barring any proof that the defendant landed the blows other than from his own
free will, it must be concluded of his mind, that he intended to cause Schweizer grievous
damage.11
Premeditation, which is normally not required for international crim inal respon­
sibility, occurs when the intent to engage in conduct contrary to an international
substantive rule is formed before the conduct is actually embarked upon. As the
Turin M ilitary Tribunal pointed out in Sävecke in 1999 (at 14) and repeated in Engel
in 2000 (at 13), premeditation necessarily requires two elements: one of a temporal
nature, namely that some time must pass between the formation of the crim inal
intent and its being carried out; the other of a psychological nature, namely that the
crim inal intent must persist from the moment of its formation until the perpetra­
tion of the crime.11213
In some instances premeditation may coincide with, or overlap, planning the crim ­
inal action. However, while planning, as we shall see, has an autonomous scope and
legal significance, premeditation has not. In ICL premeditation may only be material
to sentencing, for it may amount to an aggravating circumstance.

3.6.1 T H E R O L E O F K N O W L E D G E W I T H IN (A N D W IT H O U T ) IN T E N T

‘K now ledge’ is n o t a n o tio n fam iliar to civil law co u n tries, w here it is n o t reg a rd e d as an
au to n o m o u s category o f m ens rea, bein g ab sorbed eith er by in te n t o r by recklessness.

11 The Court went on to say that ‘In regard to this it must be remembered that the defendant denied the
entire action and did not give any explanation that could have shown another intent or arouse doubts as to
his evil intent. In addition, it is clear from the testimony that no Germans were present while the blows were
being landed, and it was not proven, as mentioned above, that the defendant was bound by the orders of the
Germans, to do the deed he did in general, and in the way he did it, in particular’ (§14). See also Gotzfnd, at
22-3,62. , . ,
On the notion of ‘deliberate (attack on a civilian population) in crimes against humanity, see some Indo­
nesian cases: Herman Sedyono and others (at 69); Asep Kuswani (at 47-8); and Yayat Sudrajat (at 8).
12 In 1971 a US military judge took a similar stand in Calley, although less accurately spelled out, when
he issued instructions to the Court-Martial. He pointed out that premeditated murder (which under US law
is a distinct category from, and not an aggravating circumstance for, unpremeditated murder) is a murder
where the actor had ‘a premeditated design to kill’; this expression means ‘formation of a specific intent to
kill and consideration of the act [...] or the acts intended to bring about death [.. .1 prior to doing them. It is
not necessary that the “premeditated design to kill” shall have been entertained for any particular or consid -
erable length of time, but it must precede the killing.’ In contrast, in the case of unpremeditated murder, only
‘intent to kill’ is required (whereas in the case of voluntary manslaughter the person entertains ‘an intent to
kill but kills in the heat of sudden passion caused by adequate provocation’) (at 1708-10). See also Manuel
Goncalves Leto Bere (at 10). ,
13 In the two cases quoted above, the Turin Military Tribunal held that premeditation had been proved
and consequently considered it an aggravating circumstance: see Sdvecke, at 14-15, and Engel at 13.
 62
INTERNATIONAL CRIMINAL LAW

In contrast, the notion as a distinct class of mental attitude in criminal behaviour is

widespread in some common law countries, particularly the United States, where one
may find a clear-cut definition in the Model Penal Code. There it is stated at section
2.02 that

A person acts knowingly with respect to a material element of an offence when:

(1) if the element involves the nature of his conduct or the attendant circumstances, he is
aware that his conduct is of that nature or that such circumstances exist; and
(ii) if the element involves a result of his conduct, he is aware that it is practically certain
that his conduct will cause such a result.14

In such countries as the UK, some distinguished commentators consider knowledge

as having the same value and intensity as intent, with the difference that intent relates
to the consequences specified in the definition of the crime’ (for instance, death as a
result of killing, in the case of voluntary murder), whereas knowledge ‘relates to circum­
stances forming part of the definition of the crime’15 (for instance, the circumstance
that property belongs to another person, in the case of criminal damage to property).
In short, it would seem that in some common law countries, knowledge denotes two
different forms of mental attitude, depending on the contents of the substantive criminal
rule at stake: (i) if the substantive penal rule prescribes the existence of a particularfact
or circumstance for the crime to materialize, knowledge means awareness of the exist­
ence of this fact or circumstance; (ii) if instead the substantive criminal rule focuses on
the result of one’s conduct, then knowledge means (a) awareness that one’s action is most
likely to bring about that harmful result, and nevertheless (b) taking the high risk of
causing that result. Plainly, in category (i) knowledge is part of intent (which involves not
only the will to accomplish a certain action and thereby attain a certain result, but also
awareness of the factual circumstances implicated in the action). Instead, in category (ii)
knowledge substantially coincides with recklessness, as defined below (see infra, 3.7).
International rules, probably under the influence of US negotiators, uphold
the notion under discussion, in both versions. Also Article 30(2) of the ICC Statute
incorporates both versions, in that it stipulates that knowledge ‘means awareness that
a circumstance exists or a consequence will occur in the ordinary course of events’.
In addition, some international rules also rely upon or require a third notion of
knowledge, i.e. as the mere fact of being apprised of a certain fact. Here, knowledge is
disconnected from intent or recklessness; it is not part of, nor is it closely connected
to intent or recklessness (as instead in the murder of a civilian, where there is intent
to cause the death of a human being and awareness of his status as a civilian; or, as in

14 See Model Penal Code and Commentaries (Official Draft and Revised Comments), Part I, vol. 1
(Philadelphia, Pa.: The American Law Institute, 1985), 225-6.
The Model Penal Code then specifies that ‘when knowledge of the existence of a particular fact is an elem­
ent of an offence, such knowledge is established if a person is aware of a high probability of its existence,
unless he actually believes that it does not exist’ (at 227).
15 See Ashworth, Principles, 191-7. In contrast, the notion is discussed only in passing by Smith and
Hogan (see at 103 and 117).
 THE ELEMENTS OF INTERNATIONAL CRIMES 63

the bombing of a military objective situated in a densely populated area, where there is
intent to bring about the destruction of the military objective and the deliberate taking
o f the risk o f killing civilians in the knowledge that those living around that objective
have the status of civilians). In the third category under discussion, knowledge consti­
tutes an element per se of mens rea, an element that is normally required in addition
to another, distinct, mental element. Such is, for example, the case with crimes against
humanity; there, in addition to the intent required for the underlying offence (such as
murder, rape, torture, or extermination) the substantive criminal rules also require
that the agent have knowledge of a factual circumstance, namely that those offences
were part o f a widespread or systematic attack directed against a civilian population
(see, e.g., Article 7(1) of the ICC Statute).
Let us see instances of the three notions of knowledge.
1 Knowledge as part of intent can be found, for instance, in Article 85(3)(e) of
the First Additional Protocol of 1977. It enumerates among the grave breaches of the
Protocol (which must be committed wilfully’ and cause ‘death or serious injury to
body or health’) the fact of ‘making a person the object of attack in the knowledge
that he is hors de combat’. Here, knowledge means awareness of the requisite circum­
stances, namely that the person is hors de combat.
As another example of knowledge as awareness of facts, hence as part of intent,
one can mention that, to be held responsible for complicity in planning or waging an
aggressive war, it must be proved either that an accused participated in the prepar­
ation or execution of these plans (and in this case the criminal intent may be inferred
from such participation), or that the accused was apprised of the plans, in addition
to taking some sort of action furthering their implementation. In Goring and others,
in considering the charges of crimes against peace made against Schacht (President
of the Reichsbank and Minister without Portfolio until 1943), the IMT noted that he
was responsible for rearmament of Germany, but this as such was not a crime; for it to
become a crime it must be shown that he carried out rearmament as part of the Nazi
plans to wage aggressive wars. However, the Tribunal found that while organizing
rearmament, Schacht did not know of the Nazi aggressive plans; hence it acquitted
him (at 307-10). A US Military Tribunal at Nuremberg took the same position and
came to the same conclusion in Krauch and others (I. G. Farben case), where it also
held that the defendants’ lack of knowledge of Hitler’s aggressive plans proved that
they lacked the requisite criminal intent (at 1115-17).
Another im portant instance where knowledge is required by international criminal
rules is aiding and abetting an international crime (for example, a war crime such as
killing a prisoner of war or an enemy civilian). Here criminal responsibility arises if
the aider and abettor knows that his action will assist the commission of a specific
crime by the principal. Various courts have taken this position.16 As the ICTY TC put

16 For instance, a US Military Tribunal sitting at Nuremberg, in Einsatzgruppen (at 568-73), two British
courts respectively in Schonfeld (at 64) and Zyklon B (at 93), the German Supreme Court in the British
Occupied Zone in the Synagogue case (at 229), and the AC in Tadic (§229).
 64 INTERNATIONAL CRIMINAL LAW

it in Furundzija, the accomplice need not share the mens rea of the principal: ‘mere
knowledge that his actions assist the perpetrator in the commission of the crime is suf­
ficient to constitute mens rea in aiding and abetting the crime’ (§236).17
As will be shown below (11.4.4), knowledge is also required in most cases oi com­
mand responsibility,18 Thus, international rules on command responsibility require
knowledge of circumstances, in the case of a commander who knows that his sub­
ordinates have committed crimes, and yet fails to take any action to repress those
crimes. He is criminally liable if, in addition to knowledge (or rather, in spite of that
knowledge), he culpably fails to take any action for the prosecution and punishment
of the culprits (intentional omission to take the prescribed action). Here, awareness of
the fact that troops under the control or authority of the commander have committed
international crimes is a mental element constituting the prelim inary sine qua non
condition of intent, and is part and parcel of intent.
2. Secondly, some international rules focus on result, and hence substantially con­
sider knowledge as am ounting or equivalent to recklessness. Thus, Article 85(3)(b) of
the First Additional Protocol considers as a grave breach ‘launching an indiscriminate
attack affecting the civilian population or civilian objects in the knowledge that such
attack will cause excessive loss of life, injury to civilians or damage to civilian objects’.
A fairly similar definition is laid down in Article 8(2)(b)(iv) of the ICC Statute.
3. In the cases considered above, knowledge is not an autonomous criminal state of
mind, but only as a means o f entertaining criminal intent or recklessness.
In contrast (and we thus move on to the third category), in some instances
knowledge cannot be reduced to either of those classes of mental state, and it remains
indispensable as a subjective element on its own. One example has already been given
above. It refers to crimes against humanity: the accused must know of a widespread

In Veit Harlan the Court of Assizes of Hamburg held in 1950 that in the case at issue there existed the
requisite subjective element of the offence of complicity in a crime against humanity (persecution of Jews),
in that the accused, a film director who had produced a strong anti-Semitic film at the behest of Goebbels’
knew the intention of Goebbels, namely to justify through the film, beyond the usual propaganda, the per­
secutory measures against Jews that had been taken and planned’ (at 66), and in addition ‘had taken into
account the possible materializing of the [adverse] consequences of the film, such consequences having been
described [in general terms) by the Supreme Court [in the British Occupied Zone]’ (at 66).
17 In this case the accused interrogated the victim while she was being subjected to rape and serious
sexual assaults by another person; the TC found that the accused’s presence and continued interrogation
o the victim while she was being subjected to violence amounted to aiding and abetting the crime, for the
accused provided assistance, encouragement, or moral support to the sexual offender, and knew that these
acts assisted the commission of the rape and sexual assault.
18 The issue was well put by the US Judge Advocate in his instructions to a US Court Martial in Medina
[A] commander is [...] responsible if he has actual knowledge that troops or other persons subject to his
control are in the process of committing or are about to commit a war crime and he wrongfully fails to take
the necessary and reasonable steps to insure compliance with the law of war. You will observe that these legal
requirements placed upon a commander require actual knowledge plus a wrongful failure to act. Thus mere
presence at the scene without knowledge will not suffice. That is, the commander-subordinate relationship
alone will not allow an inference of knowledge. While it is not necessary that a commander actually see an
atrocity being committed, it is essential that he know that his subordinates are in the process of committing
atrocities or are about to commit atrocities’ (at 1732).
 THE ELEMENTS OF INTERNATIONAL CRIMES 65

or systematic attack against a civilian population. It is not that he intends the civilian
population to be subject to the attack, nor that he knows that there is a risk of them
being subjected to an attack—both of which are beside the point. W hat one wants, is
simply to be sure that he knew of the attack. In these instances knowledge is irredu­
cible to other mental elements and exists per se (see ICTR TC, Kayishema, at §§133-4
and ICTY TC Kupreskic and others, at §556).
Finally, let it be emphasized that in ICL knowledge as awareness of circumstances
does not mean awareness of the legal appraisal of those circumstances. It only denotes
cognizance of the factual circumstances envisaged in a particular international rule.
International law, like most national systems, does not require awareness of the
illegality of an act for the act to be regarded as an international crime. As we shall see
(13.5.1) it starts from the assumption that everybody must know the law; it therefore
makes culpable even acts that were performed without the author being fully aware
of their unlawfulness (as long as the required intent, recklessness, knowledge, etc.
are there).19 International law only takes into account knowledge, or lack of know­
ledge, of the law when the defence of mistake of law can be regarded as admissible,
for the law on a particular matter is uncertain or unclear (see infra, 13.5.2). In other
words, international rules do not attach importance to the subjective mental attitude
of the perpetrator with regard to law, unless this subjective attitude coincides with the
objective condition of the law, namely its uncertainty.

3.6.2 S P E C IA L IN T E N T ( D O L U S S P E C I A L I S )

International rules may require a special intent (dolus specialis, dol aggrave) for particular
classes of crime. Such rules, in addition to providing for the intent to bring about a cer­
tain result by undertaking certain conduct (for example, death by killing), may also
require that the agent pursue a specific goal that goes beyond the result of his conduct.
International rules require a special intent for genocide: the agent must possess the
intent to destroy, in whole or in part, a national, ethnical, racial or religious group.
Thus, it is not sufficient for the person to intend to kill, or cause serious mental or bod­
ily harm, or deliberately inflict on a group seriously adverse and discriminatory condi­
tions of life, or forcibly transfer children from one group to another, etc. It also must
be proved that he did all this with the (further and dominant) intention of destroying
a group. For, as the German Federal Court of Justice (Bundesgerichtshojj stated in
Jorgic on 30 April 1999, in the crime of genocide a single person is the object of an

19 In Burgholz {No. 2), the British Judge Advocate, in delineating to the Military Court the scope of mens
rea in international crimes, stated: ‘[Y]ou might think it difficult to say that any man could have a guilty
mind in respect of his conduct if he is not aware that his conduct is in breach of any law, or if there is no
formalized law to fit his participatory conduct and to involve the breach thereof. But Mens Rea goes a little
further than that. If a man ought to have known that he was doing wrong, then the law presumes a guilty
mind, and the requirements of the doctrine of Mens Rea are fulfilled if you find the accused either knew that
they were doing wrong or ought to have known: the fact that they may have had no conscious thought of
wrongdoing will not protect them from conviction if a breach of law has been committed (84-5).
 66 INTERNATIONAL CRIMINAL LAW

attack not as an individual but rather in his capacity as a member of a group whose
social existence the perpetrator intends to destroy [...] the particular inhum anity that
characterizes genocide as distinct from m urder lies in that the perpetrator or perpe­
trators do not see the victim as a hum an being but only as a member of a persecuted
group’ (at 401).20
Similarly, a special intent is required in some categories of crimes against humanity,
namely persecution. Here, in addition to the intent necessary for the commission of
the underlying offence (murder, rape, serious bodily assault, expulsion from a village,
an area or a country, etc.) a discriminatory intent is called for, namely the will to dis­
criminate against members of a particular national, ethnic, religious, racial, or other
group. As an ICTY TC put it in Kupreskic and others (§634), and another TC restated
it in Kordic and Cerkez (§§214 and 220), the acts of the accused must have been aimed
at singling out and attacking certain individuals on discriminatory grounds’, for the
purpose of ‘removal of those persons from the society in which they live alongside
the perpetrator, or eventually from hum anity itself’. In Blaskic, another TC worded
that intent as follows: the specific intent to cause injury to a hum an being because he
belongs to a particular community or group’ (§235).
The rules on crimes of international terrorism require a special intent: that of spread­
ing terror in the population by killing, hijacking, blowing up buildings, etc. (see infra,
8.3.2). Also the rules criminalizing aggression require special intent (see infra, 7.3.3(b)).
In all these cases pursuance of a special goal is essential, while its full attainment
is not necessary for the crime to be consummated. Clearly, the m urder of dozens of
Muslims, Kurds, or Jews may be termed genocide if the required special intent is pre­
sent, regardless of whether the general purpose of destroying the group as such is
achieved; the same holds true for terrorist attacks, which may amount to international
crimes even if in fact a specific attack does not achieve the purpose of terrorizing the
population; similarly, the forcible expulsion of a number of Muslims from their homes
amounts to persecution even if not all Muslims are in fact driven out of the area.

3.7 RECKLESSNESS
Recklessness or dolus eventualis is a state of mind where a person foresees that his or her
action is likely to produce its prohibited consequences, and nevertheless willingly takes the
risk o f so acting. In this case the degree of culpability is less than in intent. There, the actor
anticipates and pursues a certain result and in addition knows that he will achieve it by
his action; here instead he only envisages that result as possible or likely and deliberately

20 That a specific or special intent is required for genocide has also been stressed
in Akayesu (TJ §498),
Musema (TJ §§ 164-7), Jelisic (AJ §§45-6); Krstic (TJ, §§569-99; A), §§24-38).
 THE ELEMENTS OF INTERNATIONAL CRIMES 67

takes the risk; however, he does not necessarily will or desire the result. Recklessness,
thus, is made up of foresight and a volitional act (deliberately taking the risk).
Instances of recklessness are clearly envisaged in some international rules. Thus, for
instance, the rule on superiors’ responsibility provides that the superior is criminally
liable for the crimes of his subordinates i f ‘he consciously disregarded information
which clearly indicated’ that his subordinates were about to commit, or were com­
mitting, international crimes (see infra, 11.4.4). In this case the superior is liable to
punishment for consciously having taken the risk, knowing that his subordinates were
likely to commit or were committing crimes.
Furthermore, in the case of responsibility for crimes perpetrated by a multitude of
persons pursuant to a common design, or joint criminal enterprise (see infra, 9.4.4),
as the ICTY AC held in Tadic (AJ), what is required is that, under the circumstances
of the case, (i) it was foreseeable that a non-concerted crime might be perpetrated by
one or other members of a group or collectivity jointly pursuing a crim inal intent; and
(ii) the accused consciously and deliberately took that risk (§§227-8).
The notion of recklessness was also applied in many cases brought before German
courts after the Second World War. These courts, which administered criminal justice
under Control Council Law no. 10, were seized with crimes against hum anity commit­
ted by Germans against other Germans. Most cases concerned denunciations to the
Gestapo, with all the ensuing inhum an consequences. In many cases those courts held
that, for the denunciation to amount to a crime against humanity, it was not necessary
for the author of the denunciation to foresee and will all the nefarious consequences of
his act; it was sufficient that he be aware of the authoritarian and arbitrary system of
Nazi violence then prevailing in Germany and of the consequent risk that the victim
would be subjected to persecution and great suffering. In this connection the German
Supreme Court in the British Occupied Zone employed the German equivalent of the
notion of recklessness, namely Eventualvorsatz (or bedingter Vorsatz).21

21 According to an ICTY TC in Stakic, ‘The technical definition of dolus eventuahs is the following: if the
actor engages in life-endangering behaviour, his killing becomes intentional if he “reconciles himself” or
“makes peace” with the likelihood of death. Thus, if the killing is committed with “manifest indifference to
the value of human life”, even conduct of minimal risk can qualify as intentional homicide. Large scale 1
ings that would be classified as reckless murder in the United States would meet the continental criteria о
dolus eventualis. The Trial Chamber emphasises that the concept of dolus eventualis does not include a stan -
ard of negligence or gross negligence’ (§587). In Blaskic and ICTY TC defined recklessness as the situation
where ‘the outcome is foreseen by the perpetrator as only a probable or possible consequence of his conduct;
according to the TC the agent takes‘a deliberate risk in the hope that the risk does not cause injury (§ )■
A good definition of this notio n -as set out in the criminal law of the State of New Y ork-can be found
in Rule 15.5(3) of the New York Penal Code, whereby ‘A person acts recklessly with respect to a result or to
a circumstance described by a statute defining an offence when he is aware of and consciously disregards a
substantial and unjustifiable risk that such result will occur or that such circumstance exists. The risk must
be of such nature and degree that disregard thereof constitutes a gross deviation from the standard of con­
duct that a reasonable person would observe in the situation.’ See also Art. 2(2)(c) of the US Model Penal
Code cited above, at n. 6.
22 For instance, one can mention K. and M„ decided by the Offenburg Tribunal (Landgericht) on 4 Tune
1946. In January 1944, K., the principal accused, a member of the Nazi party, over a dinner with friends an
 68 INTERNATIONAL CRIMINAL LAW

The Supreme C ourt in the British Occupied Zone also required recklessness in
other cases not dealing with denunciations. For instance, in L. and others (the so-
called Pig-cart parade case) the events had occurred on 5 May 1933. In a parade by
SA (assault troopers) through the main streets of a small German town a prominent
socialist senator and a Jewish inhabitant were publicly humiliated and subjected to

acquaintances had a discussion with Könninger, a soldier who was on home leave. Already tipsy, Könninger
inveighed against the German leadership, noting among other things that the war was about to be lost A
tew weeks later K. reported Könninger's tirade to various persons including some dignitaries attending a
P r / " 16611" 8 at a restaurant- As a result, the Gestapo arrested Könninger and brought him to trial. In July
1944 he was sentenced to death for defeatism and executed. Before the Offenburg court K. submitted that
he had not intended to have the victim prosecuted and punished for his utterances. The court held, how­
ever, that when he reported his statements to the party meeting, ‘he must expect that his words would have
adverse consequences for Könninger. The accused caused proceedings against Könninger to be instituted,
witnesses to be heard, and the victim eventually to be sentenced. It is entirely credible that the accused K.
did not intend all that. However, he was to expect that this would be the result of his talk at the restaurant
He must foresee this result. He tacitly approved it. There was therefore recklessness on his part’ (67). The
court found K. guilty of a crime against humanity (persecution on political grounds) under Article II(l)(c)
of Control Council Law no. 10.
at A very similar case is W„ brought before the Tribunal of Waldshut (judgment of 16 February 1949,

K„ decided on 27 July 1948 by the German Supreme Court in the British Occupied Zone, is also interest-
mgHn February 1942 the accused, a member of the Waffen SS working at the headquarters of the Gestapo
m D., had denounced at his headquarters a Jewish businessman (M.) because the latter had gone to the
apartment of a non-Jew. The denunciation led to the Jew being taken into preventive custody for three
weeks^ The accused was found guilty of a crime against humanity. On appeal the Supreme Court confirmed
the judgment It held that under the relevant rules the accused had engaged in ‘offensive conduct that was
conscious and deliberate; he must be aware that he was ‘handing over the victim directly or indirectly
to forces which [would], on account of the facts in the denunciation, treat him solely according to their
purposes and ideas without being bound by considerations of justice or legal certainty’. According to the
court the accused ‘was aware that the denunciation could have entailed the most grave consequences for M„
as the accused knew of the criminal and arbitrary manner in which the Gestapo abused its power at the time’.’
is mental element was sufficient; it was not required for the perpetrator to have acted with ‘an inhumane
cast of m ind, nor was ‘approval of the result’ required (50-1).
R. was heard by the Supreme Court in the British Occupied Zone, and decided on 27 July 1948 In March
1944 m Hamburg the accused, a member of the Nazi party, had an argument with a soldier in uniform
who had insulted the political leaders while drunk. Later on he reported him to the police, and as a result
the victim was arrested on the Eastern front, brought back to Germany in September 1944, charged with
undermining military morale and brought to trial. He was sentenced first to five years' imprisonment and
hen a death sentence was sought, but not imposed due to the Russian occupation. The Court held that
tor the denunciation to be a crime against humanity, it was necessary that ‘the offensive behaviour of the
perpetrator be conscious and intentional (or at least the perpetrator took the risk), that it actually occurred
and the perpetrator, through his act, willed that the victim be handed over to powers that did not obey the
rule of law, or at least, that he took this possibility into account’. The Court insisted that the mental element
of the crime was met if the perpetrator had intended ‘to deliver the victim to the uncontrollable power
machinery of the power and the State or at the very least he had taken the risk that he would be treated
arbitrarily. And the Court added that ‘negligence’ (Fahrlässigkett) was not sufficient (at 47).
te Supreme Court took the same position in O. (judgment of 19 October 1948) (at 106-7) and in Vi
(judgment on the same date), where it restated that, for the accused’s denunciation of another person to the
police to be characterized as a crime against humanity, it was necessary that a mental element be present
namely that she at least was cognizant of and took into account the possibility that the victim, as a conse-
quence of her denunciation, would be treated in an arbitrary manner’ (at 115-16). The same judgment was
restated in other cases of denunciation, /. and R. (at 170), S. (at 260-1), and F. (at 367).
 THE ELEMENTS OF INTERNATIONAL CRIMES 69

inhuman treatment (they were led along in a pig cart, with demeaning inscriptions
hung around their necks and were vilified in various ways). The defendants took part
in the parade. The Court held that, as far as the involvement of three accused went, ‘it
was inconceivable’ that they, who were old officials of the Nazi party, did not at least
think it possible and consider that in the case at issue, through their participation, per­
sons were being assaulted by a system of violence and injustice; more is not required
for the mental element’ (at 232). In contrast, in the case of another defendant, who had
simply followed the procession among the onlookers and in civilian clothes, the Court
held that he was not guilty because he ‘had not participated in causing the offence
nor had he at least entertained dolus eventualis in taking part in the causation of the
offence (at 234).23 It would thus seem that, according to the Court at least, some of the
defendants took an unjustified risk of the victims being assaulted.
As for the case law of international tribunals, it bears mentioning Blaskic (where
the AC held that to establish liability under Article 7(1) of the ICTY Statute for order­
ing the commission of a crime, it is required that a person orders an act or omission
with the awareness of a substantial likelihood that a crime will be committed in the
execution of that order’, because ordering with such awareness has to be regarded as
accepting the crime’, at §42) and Stakic (where the TC held that recklessness or dolus
eventualis could suffice for the crime of murder as a war crime and for extermination
as a crime against humanity, at §§587 and 642).24

23 Another significant case is P. and others. On the night after Germany’s partial capitulation (5 May
1945) four young German marines had tried to escape from Denmark back to Germany. The next day they
were caught by Danes and delivered to the German troops, who court-martialled and sentenced three of
them to death for desertion; on the very day of the general capitulation of Germany, i.e. 10 May 1945, the
three were executed. The German Supreme Court found that some of the participants in the trial before the
Court-Martial were guilty of complicity in a crime against humanity. According to the Supreme Court,
the glaring discrepancy between the offence and the punishment proved that the execution of the three
marines had constituted a clear manifestation of the Nazis’ brutal and intimidatory justice. The acts per­
formed by the defendants involved a crime against humanity. As for the mental element of the crime, t e
Court-held that intent (indisputably present in the case of the judges who had sentenced the marines to
death and of the military commander who had confirmed the sentence and ordered the execution) was not
necessarily required; recklessness, for instance in the case of the prosecutor, was sufficient: it is sufficient for
the defendant concerned to have taken into account the possibility and have consented to the fact that his
conduct would contribute to cause the resulting killing’ (224).
In Eschner, the accused, an SS officer who had held an important position in the concentration camp of
Gross-Rosenbetween 1941 and 1945, was accused, amongother things, o f h a v i n g requested KapoV a crim­
inal by profession, to ‘get rid of’ a camp inmate who had tried to escape; the inmate had probably died. The
Wurzburg Tribunal held that the accused knew the violent behaviour of Kapo and approvingly too into
account that the inmate might suffer death as a result of the intended ill-treatment. Thus he willed recklessly
the death of a man contrary to law.’ However, in view of the fact that the inmate’s death was not certain, t e
court found the accused guilty o f‘attempted murder by recklessness (253).
24 As for the ICTR, see for instance Musema, TJ, at §215 and Kayishema and Ruzindana, TJ> at §146.
 70 INTERNATIONAL CRIMINAL LAW

3.8 CULPABLE OR GROSS NEGLIGENCE

Generally speaking, negligence entails that the person (i) is expected or required to
abide by certain standards of conduct or take certain specific precautions with which
any reasonable person should comply; (ii) acts in disregard of these standards or
precautions; and (iii) either (a) does not advert at all to the risk of harm to another
person involved in his conduct, which falls short of the standards or precautions
(simple negligence), or (b) is aware of that risk but believes that it will not occur, and
in addition takes a conduct that is blatantly at odds with the prescribed standards
(gross negligence). Mere negligence is the least degree of culpability. Normally it is not
sufficient for individual criminal liability to arise.
It would seem that, given the intrinsic nature of international crimes (which always
am ount to serious attacks on fundamental values) in ICL negligence operates as a
standard of liability only when it reaches the threshold of gross or culpable negligence
(culpa gravis). Given the nature of international crimes, the mental element under dis­
cussion only becomes relevant when there exist some specific conditions relating to
the objective elements of the crime; that is, the values attacked are fundamental and
the harm caused is serious.25
That national legal systems may penalize a mental state that is less grave than
the one criminalized at the international level should not be surprising. Given the
consequences following from, and the stigma inherent in, international crimes, it is
only natural that international criminal rules should be more exacting, with regard to
subjective requirements of the offence, than some national criminal legislation.26

25 This definition of culpable negligence is in some respects at variance with that upheld in some com­
mon law and civil law countries. For instance, under the New York Penal Code, Rule 1505(4): ‘A person
acts with criminal negligence with respect to a result or to a circumstance described by a statute defining
an offence when he fails to perceive a substantial and unjustifiable risk that such result will occur or that
such circumstance exists. The risk must be of such nature and degree that the failure to perceive it consti­
tutes a gross deviation from the standard of care that a reasonable person would observe in the situation ’
Clearly, this definition corresponds to what we termed above 'inadvertent negligence', or culpa levis (see
supra, 3.5).
26 Case law bears out the above international notion. John G. Schultz, a case brought before a US Court
of Military Appeals in 1952, deserves mention. Schultz, driving a car, had struck and killed two Japanese
pedestrians m 1950 in Japan (although Japan was still under US military occupation, this of course was
not a war crime). The US Court stated the following: ‘A careful perusal of the penal codes of most civilized
nations leads us to the conclusion that homicide involving less than culpable negligence is not universally
recognized as an offense. Even in those American jurisdictions-still relatively few in num ber-w hich have
given statutory recognition to either negligent homicide or vehicular homicide, the degree of negligence
required is often held to be culpable’ or gross’- t h e same as that required for involuntary manslaughter.
mposing criminal liability for less than culpable negligence is a relatively new concept in criminal law
and has not, as yet, been given universal acceptance by civilized nations’, 4 CMR (1952), 104, 115-16 (CMA
Lexis 661). On this case see also infra, 4.3, n. 8.
A definition of negligence as a possible subjective element in international crimes can be found in the
instructions given by the Judge Advocate to a Canadian Court Martial in Major A. G. Seward. The defendant
in f; a^ ° ng °ther *hing’ been Charg£d With ne^ i8entfi Performing his military duty while in Somalia in
19 U. The particulars of his negligence were stated to be that he ‘by issuing an instruction to his subordinates
t at prisoners could be abused, [he] failed to properly exercise command over his subordinates, as it was his
 THE ELEMENTS OF INTERNATIONAL CRIMES 71

Gross negligence is clearly required by the customary rules on superiors responsi­

bility (see infra 10.4) whereby a superior is responsible for the crimes of his subordi­
nates if he did not know but ‘should have known that they were about to commit, or
were committing, or had committed crimes. In this case, the superior was required to
become cognizant of, and verify, all the information necessary to monitor the activ­
ities and the conduct of his subordinates. If he disregards these standards of conduct,
he acts with gross negligence and is consequently liable for dereliction of duty, if all the
other conditions are fulfilled.27
Culpable negligence has also been considered sufficient in other circumstances. A
case where a court held negligence to be the mental element of a war crime is Stenger
and Crusius, decided in 1921 by the Leipzig Supreme Court.28
Another court also took into account negligence, this time with regard to crimes
against humanity: Hinselmann and others, decided by the British C ourt of Appeal in
the British Zone of Control in Germany, in 1947. A Trial Court had convicted a group

duty to do’. As a result of his instructions, some of his subordinates had beaten up and killed a Somali civil­
ian. In instructing the Court Martial on the notion of negligence, the Judge Advocate stated: ‘[A]s a matter
of law the alleged negligence must go beyond mere error in judgment. Mere error in judgment does not con­
stitute negligence. The alleged negligence must be either accompanied by a lack of zeal in the performance of
the military duty imposed, or it must amount to a measure of indifference or a want of care by Major Seward
in the matter at hand, or to an intentional failure on his part to take appropriate precautionary measures
(at 1081). The Court Martial found the defendant guilty on this count. In commenting on this finding y
the Court Martial, the Court Martial Appeal Court of Canada stated that the Court Martial ‘must be taken
to have concluded that the respondent did issue an “abuse” order and that his doing so was no mere error
in judgment. He himself confirmed that he was taking a “calculated risk” m doing so and that nothing in
his training or in Canadian doctrine would permit the use of that word during the giving of orders (ibid.).
Arguably, recklessness more than negligence was at issue in this case.
27 Among the cases that may be cited to support the applicability of gross negligence in cases of superior
responsibility, Schmitt stands out. This case, concerning the commander of a concentration camp in
Breendonck, was brought before the Brussels Military Tribunal, which held in 1950 that although it is true
that generally speaking jurisprudence does not consider that, in case of murder, simple lack of action or
negligence are punishable, this however does no longer apply when a person’s failure to act amounts to the
non-fulfilment of a duty [...] in this case failure to take action amounts to material conduct sufficient for the
realisation of criminal intent’ (at 936-7).
28 In the battle near Saarburg in Loraine between the French and the German Army, on 21 August 1914 the
accused, Crusius, a captain of the German army, thought that Major-General Stenger had verbally ordered
the killing of all French wounded. Acting under this erroneous assumption, he passed on this alleged order
to his company. The Court concluded that Crusius was guilty of causing ‘death through culpable negligence
(fahrlässige Tötung) and sentenced him to two years’ imprisonment. The Court held that: ‘the act of will
which in the further course of events caused the objectively illegal outcome [...] included an act of careless­
ness which ran contrary to his duty, and neglect of the consideration required in the situation at hand which
was perfectly reasonable to expect from the accused. Had he applied the care required of him, he would not
have failed to notice what many of his men realized immediately, namely that the indiscriminate killing o
all wounded represented an outrageous and by no means justifiable war manoeuvre [...] Captain Crusius
was certainly familiar with the provisions of the field operating procedures which require a written order as
the basis for troop command by the higher troop leaders, as well as the drill manual which makes the writ­
ten order a rule, especially concerning orders for brigades and higher. This circumstance is also not entire y
without significance, particularly in view of the personality of the accused who was described as a diligent,
zealous and benevolent officer. In view of the accused’s background and personality, he should have antici­
pated the illegal outcome which was easily demonstrated even if his mental and emotional states at the time
were to be fully taken into consideration (at 2567-8).
 72 INTERNATIONAL CRIMINAL LAW

of German doctors and police officers of crimes against humanity, under Control
Council Law no. 10 (Article 1(c)). It had found that they were concerned with carry­
ing out, in 1944-45, sterilization operations on a number of persons of gypsy blood,
to prevent the increase of the race’ (three doctors had performed the operations and
two police officers had induced persons to sign consent to the operations by threats).
Counsel for one of the doctors, Günther (a gynaecological specialist), argued that
there was no evidence that he knew that the gypsies were being sterilized on account of
their race. In counsel s view, the case against Günther could only be one of negligence;
however, negligence was not sufficient to constitute an offence under Control Council
Law no. 10, which required extremely gross negligence. Hence, Günther, if he were to
be convicted at all, could only be convicted under section 230 of the German Criminal
Code.29 Ihe Prosecutor countered that Günther must have known the correct proced­
ure in the case of sterilization, but made no enquiries, and saw no legal documents. 30
The Court of Appeals found that the appellant’s frame of m ind amounted to
negligence: a German law of 1933, as amended in 1935, made it clear that sterilization
operations were illegal unless: (i) they were performed to avert a serious threat to the
life and health of the person operated upon, and with the consent of that person; or
(ii) they were carried out in pursuance of an order of the Eugenics Court. The Court
of Appeals noted that in the case at issue neither of these conditions was fulfilled.31
Ihe crucial point was, however, whether negligence (Fahrlässigkeit) could suffice for
the requisite mens rea in the case of a crime against humanity. The Court of Appeals
held that in the case at issue there was no suggestion that the operations were cruelly
performed, and the evidence was inadequate to establish a degree of negligence which
could have amounted in any event to a Crime against Hum anity’. It consequently
reduced the sentence of two years’ imprisonment to six months.32

29 Under this provision, ‘Whoever through negligence causes bodily harm to another is punished by a pecu­
niary penalty or imprisonment up to three years’ (see A. Schönke, Strafgesetzbuch fü r das Deutsche Reich-
Kommentar, 2nd edn (Munich and Berlin: Beck, 1944), at 484; and see 172-3 for the notion of negligence).
In addition, in his view there was no difference ‘in the degree of negligence required to constitute an
offence under Section 230 and that required to constitute an offence under [Control Council] Law 10’.
The operations were of so special a nature, and the limits within which they could be legally performed
so narrow, that Günther was put upon his enquiry before he operated. His failure to make the necessary
enquiry was negligence. Although ‘negligence’ as used by British lawyers [in English law there is negligence
when the conduct of a person fails to measure up to an objective standard and the person ought to have fore­
seen the risk involved in his conduct; see, for instance, Smith and Hogan, 90-6.] and ‘Fahrlässigkeit’ as used
by German lawyers are not co-extensive terms [in German law there is negligence when a person, acting in
breach of a duty of precaution brings about a certain result he has not willed, and this result occurs either
because the person is not cognizant of the breach of duty, or else is aware that the breach may occur, but trusts
that the result will not materialize; see, for instance, Jescheck, Lehrbuch, at 563] there was undoubtedly Fahr­
lässigkeit on Günther s part; and the sterilization of the persons operated upon was a bodily injury.’ (68-60).
32 As mentioned above, counsel for the appellant had argued that negligence, if any, on the part of
Günther was not serious enough to constitute an offence under Control Council Law no. 10; German law was
therefore applicable. However, under this law, unless the rule under which a person was charged expressly
stated that negligence was sufficient, the person could not be convicted of a criminal offence if the act consti­
tuting it was merely negligent and not intentional. The Court dismissed this argument. The Court of Appeal
stated as follows: ‘We do not accept the proposition that this is necessarily so [namely that negligence may
not amount to the requisite subjective element unless this is explicitly provided for in the relevant law]
 THE ELEMENTS OF INTERNATIONAL CRIMES 73

It maybe clearly inferred from this finding that, for the Court of Appeals, crimes against
humanity may result from negligence, provided, however, that negligence is gross.
Finally, it should be pointed out that there are also cases where culpable negligence
has been so conceived of as to border on recklessness.33

3.9 THE ICC STATUTE

As stated above, the ICC Statute contains the only international provision setting out
a general definition of the subjective element of international crimes: Article 30. This
provision envisages intent and knowledge as the only mental elements of those crimes
(as set forth in Articles 6-8 of the ICC Statute). Article 30(1) provides that ‘Unless
otherwise provided, a person shall be criminally responsible and liable for punishment
for a crime within the jurisdiction of the Court only if the material elements are com­
mitted with intent and knowledge.’ Paragraph 2 then defines those two notions.34
Article 30 raises two problems. First, it does not refer expressly to recklessness
or culpable negligence, although recklessness (dolus eventualis) may be held to be
encompassed by the definition of intent laid down in paragraph 2. Secondly, it always
requires both intent and knowledge, whereas there may be cases where only intent, as

where a charge under [Control Council] Law 10 is tried in a Control Commission Court; but, in the present
case, there is no suggestion that the operations were cruelly performed, and the evidence was inadequate to
establish a degree of negligence which could have amounted in any event to a Crime against Humanity. The
Court consequently set aside Gunther’s conviction under Control Council Law no. 10 and substituted it with
a finding that he was guilty of an offence under section 230 of the German Criminal Code (at 60).
33 Thus in Medina, in 1971 a US military judge issued to the Court-Martial instructions with regard to
command responsibility arising in a case where the commander allegedly had actual knowledge that troops or
other persons subject to his control were in the process of committing war crimes (killing of innocent civilians
in the Vietnamese village of My Lai), and wrongfully failed to take the necessary and reasonable steps to ensure
compliance with the laws of war. The military judge pointed out that the legal requirements of international
law ‘placed upon a commander require actual knowledge plus a wrongful failure to act. He then stated that
the omission to exercise control must constitute culpable negligence and then pointed out that culpable neg­
ligence is a degree of carelessness greater than simple negligence. For purposes of making the distinction
between the two, you are advised that simple negligence is the absence of due care, that is an omission by a
person who is under a duty to exercise due care, which exhibits a lack of that degree of care for the safety of oth­
ers which a reasonable, prudent commander would have exercised under the same or similar circumstances.
Culpable negligence, on the other hand, is a higher degree of negligent omission, one that is accompanied by
a gross, reckless, deliberate, or wanton disregard for the foreseeable consequences to others of that omission;
it is an omission showing a disregard of human safety. It is higher in magnitude than simple inadvertence, but
falls short of intentional wrong. The essence of wanton or reckless conduct is intentional conduct by way of
omission where there is a duty to act, which conduct involves a high degree of likelihood that substantial harm
will result to others’ (at 1732-4). See also above, the Major A. G. Seward case (cited in nt. 26).
34 Para. 2 provides that; ‘For the purposes of this article, a person has intent where (a) in relation to con­
duct, that person means to engage in the conduct; (b) in relation to a consequence, that person means to
cause that consequence or is aware that it will occur in the ordinary course of events.
Para. 3 provides that; ‘For the purposes of this article, “knowledge” means awareness that a circumstance
exists or a consequence will occur in the ordinary course of events. “Know” and “knowingly” shall be con­
strued accordingly.’
 74 INTERNATIONAL CRIMINAL LAW

defined in the provision, is sufficient, and other cases where instead only knowledge
(which, according to the definition given in the provision, may be regarded as equiva­
lent to recklessness) would be sufficient.
To solve the first problem one may focus on the initial proviso of the rule (‘Unless
otherwise provided’): whenever a provision of the Statute or a rule of international
customary law requires a different mental element, this will be considered sufficient
by the Court. For instance, Article 28(a)(i) provides for the responsibility of superiors
where the ‘military commander or person [...] owing to the circumstances at the time,
should have known that the forces [under his effective command, control or author­
ity] were committing or about to commit [...] crimes’. Plainly, this provision envisages
culpable negligence (see supra, 3.8 as well as 11.4.4). This case would be covered by the
proviso just referred to.
Nonetheless, when a specific substantive provision of the Statute does not specify
the mental element required, one may deduce from Article 30 that one must take that
substantive provision to require intent and knowledge. In this manner the Statute may
eventually require a mental element higher than that set down in customary law. Indeed,
differences may arise between customary international law and treaty law whenever a
customary rule concerning a specific crime considers as a sufficient requirement for
that crime a subjective element other than intent (for instance, culpable negligence).
As for the second problem (the use of the conjunctive ‘and’), one ought to note that
in international law the standard of construction applies that a purely grammatical
consti uction must yield to a logical interpretation whenever this is dictated by the
principle of effectiveness (ut res magis valeat quam pereat) and is consonant with the
object and purpose of the rule. It is therefore admissible to construe the word ‘and’ as
also including the word or when this is logically required.35

3.10 JU D IC IA L D E TERM IN A TIO N OF THE

M ENTAL ELEM ENT

As in national law, in ICL a culpable state of m ind is normally proved in court by

circumstantial evidence. In other words, one may infer from the facts of the case
whether or not the accused, when acting in a certain way, willed, or was aware, that
his conduct would bring about a certain result. To put it differently, one may normally
deduce from factual circumstances whether the action contrary to ICL was accom­
panied by a mental attitude denoting some degree of fault.
This is the position taken by national and international courts. For instance, one
can refer to the statement made by the Judge Advocate addressing a Canadian Military
Court in Johann Neitz. The question at issue was whether the accused, who had shot at

35 An application of this rule of construction was made by an ICTY TC in Tadic, decision of 7 May 1997
§§712-13. ’
 THE ELEMENTS OF INTERNATIONAL CRIMES 75

a member of the Royal Canadian Air Force taken prisoner by Germans, wounding the
prisoner without killing him, had intended to cause his death. The Judge Advocate put
the issue to the Military Court as follows:
Intention is not capable of positive proof, and, accordingly, it is inferred from the overt
acts. Evidence of concrete acts is frequently much better evidence than the evidence of an
individual for, after all, an individual alone honestly knows what he is thinking. The Court
cannot look into the mind to see what is going on there. The individual may protest vehe­
mently what his intentions were, but such evidence is subject to human frailty and human
perfidy. Accordingly, intention is presumed from the overt act. It is a simple application of
the principle that actions speak louder than words, and, I add, often more truthfully. It is
also a well-established maxim of law that a man is presumed to have intended the natural
consequences of his acts. If one man deliberately strikes another over the head with an axe,
the law presumes he intended to kill the other. Similarly so, if one man deliberately shoots a
gun at another, an intent to kill will be presumed [...] If a man points a gun at another and
deliberately fires, it is presumed that he intends to kill the other. However, this is a presump­
tion of fact, but it may be rebutted’ (at 209). (The Court found the accused had committed a
war crime with intent to kill and sentenced him to life imprisonment.)

Interestingly, in Jelisic an ICTY TC, in order to establish whether the accused had
entertained the special intent required for genocide, examined various statements he
had made to the effect that he wished to exterminate Muslims, for he hated them and
wanted to kill them all (§§102-4). The Court concluded, however, that these utterances
revealed a disturbed personality and consequently, for lack of the requisite special
intent, the acts of the accused were not ‘the physical expression of an affirmed resolve
to destroy in whole or in part a group as such’ (§107). The AC, while holding that the
TC had erred in acquitting the defendant of genocide (Jelisic, AJ, §§53-72), surpris­
ingly did not uphold the Appellant’s request that the case be remitted to a TC for retrial
(§§73-7)- It held that such remittal was not in the interests of justice’ (§77).

36 A court of Bosnia and Herzegovina took the same approach in Tepez with regard to intent. In setting
out the mental element of the crimes of torture and murder of civilians, the Sarajevo Cantonal Court stated
that ‘The accused perpetrated the crime deliberately; he was aware that together with others from Rajko
Kuj’s group he was taking part in torture, beatings and killing of prisoners. Since the accused repeated these
actions many times, he definitely wished to do that and was aware that repeated beatings of prisoners with
hard objects, fists and boots in vital parts of their bodies can certainly result in their death. By repeating
these actions it is evident that the accused wanted these people killed’ (at 7).
With regard to the subjective element of command responsibility, an ICTY TC pointed out in Delalic
and others, that it could be established 'by way of circumstantial evidence. The TC pointed out that m the
absence of direct evidence of the superior’s knowledge of the offences committed by his subordinates, such
knowledge cannot be presumed, but must be established by way of circumstantial evidence’ (§386).
Again, with regard to ‘knowledge’ that the subordinates were committing or had committed crimes in the
case of command responsibility, an ICTY TC stated in Kordic and Cerkez that, ‘Depending on the position
of authority held by a superior, whether military or civilian, de jure or de facto, and his level of responsibil­
ity in the chain of command, the evidence required to demonstrate actual knowledge may be different. For
instance, the actual knowledge of a military commander may be easier to prove considering the fact that he
will presumably be part of an organized structure with established reporting and monitoring systems. In
the case of de facto commanders of more informal military structure, or of civilian leaders holding de facto
positions of authority, the standard of proof will be higher’ (§428).
 PART II

SUBSTANTIVE
CRIMINAL LAW
 SECTION I

INTERNATIONAL
CRIMES
 4
WAR CRIMES

4.1 TH E NOTION

War crimes are serious violations of customary or treaty rules belonging to the corpus of
the international humanitarian law of armed conflict (IHL). As the AC of the ICTY stated
in Tadic (IA), (i) war crimes must consist o f‘a serious infringement’ of an international
rule, that is to say must constitute a breach of a rule protecting important values, and the
breach must involve grave consequences for the victim’; (ii) the rule violated must either
belong to the corpus of customary law or be part of an applicable treaty; and (iii) ‘the vio­
lation must entail, under customary or conventional law, the individual criminal respon­
sibility of the person breaching the rule’ (§94); in other words, the conduct constituting a
serious breach of international law, in addition to being an interstate violation involving
the responsibility of the state to which the serviceman belongs, must be criminalized.
In the same decision the AC gave the following example of a non-serious violation:
‘the fact of a combatant simply appropriating a loaf of bread in an occupied village’
would not amount to such a breach, ‘although it may be regarded as falling foul of
the basic principle laid down in Art. 46(1) of the [1907] Hague Regulations [on Land
Warfare] (and the corresponding rule of customary international law) whereby “private
property must be respected” by any army occupying an enemy territory’ (§94).
War crimes may be perpetrated in the course of either international or internal
armed conflicts; that is, civil wars or large-scale and protracted armed clashes
breaking out within a sovereign state. Traditionally, war crimes were held to embrace
only violations of international rules regulating war proper; that is international
armed conflicts and not civil wars. After the ICTY AC decision in Tadic (IA) of 1995
(see infra, 4.3), it is now widely accepted that serious infringements of international
hum anitarian law on internal armed conflicts may also be regarded as amounting to
war crimes proper, if the relevant conduct has been criminalized. As evidence of this
new trend, suffice it to mention Article 8(2)(c-f) of the ICC Statute.
IHL is a vast body of substantive rules comprising what are traditionally called
‘the law of the Hague’ and ‘the law of Geneva’. The former set of rules includes some
Hague Conventions of 1899 or 1907 on international warfare. These rules, in add­
ition to providing for the various categories of lawful combatants, primarily regulate
combat actions (means and methods of warfare) and the treatment of persons who no
longer take part in armed hostilities (prisoners of war). The so-called ‘law of Geneva’
comprises the various Geneva Conventions (at present the four Conventions of 1949
 82 INTERNATIONAL CRIMINAL LAW

plus the two Additional Protocols of 1977), and is essentially designed to regulate the
treatment of persons who do not, or no longer, take part in armed conflict (civilians,
the wounded, the sick and shipwrecked, as well as prisoners of war). Furthermore,
Article 3, common to the four Geneva Conventions and the Second Additional
Protocol, regulate, internal armed conflict. The Third Geneva Convention of 1949 also
regulates the various classes of lawful combatants, thereby updating the Hague rules.
In addition, the First Additional Protocol of 1977 to some extent updates those rules of
the Hague law which deal with means and methods of combat, for the sake of sparing
civilians as far as possible from armed hostilities. It is thus clear that the traditional
distinction between the two sets of rules is fading away; even assuming it has not
become obsolete, its purpose now is largely descriptive.
War crimes may be perpetrated by military personnel against enemy servicemen or
civilians, or by civilians against either members o f the enemy armed forces or enemy
civilians (for instance, in occupied territory). Conversely, crimes committed by ser­
vicemen against their own military (whatever their nationality) do not constitute war
crimes.1 Such offences may nonetheless fall within the ambit of the military law of the
relevant belligerent.

4.2 TH E NEED FOR A LINK BETW EEN THE

OFFENCE AN D AN ARM ED CONFLICT

Criminal offences, to amount to war crimes, must also have a link with an inter­
national or internal armed conflict. Many courts, chiefly the ICTY12 and the ICTR,3
have restated this proposition, which can easily be deduced from the whole body of

1 This point was clarified in Pilz by the Dutch Special Court of Cassation. A young Dutchman in the
occupied Netherlands had enlisted in the German army and while attempting to escape from his unit had
been fired upon and wounded. Pilz, a German doctor serving in the German army with the rank of Haupt-
sturmfuhrer, prevented medical and other aid or assistance being given by a doctor and hospital orderly to
the wounded Dutchman, and in addition, ‘in abuse of his authority as a superior’, ‘ordered or instructed a
subordinate to kill the wounded [man] by means of a firearm’ (at 1210), as a result of which the Dutchman
died. The Court held that the offence was not a war crime, for ‘the wounded person was part of the occupying
army and the nationality of this person is therefore irrelevant, given that, by entering the military service of
the occupying forces, he removed himself from the protection of international law and placed himself under
the laws of the occupying power’ (at 1210): consequently, the offence constituted a crime ‘within the province
of the internal law of Germany’ (at 1211).
See also the decision in Motosuke delivered by a Temporary Court Martial of the Netherlands East Indies,
at Amboina. Motosuke, a Japanese officer, had been accused, among other things, of having ordered the exe­
cution by shooting of a Dutch national named Barends, who, during the occupation of Ceram by Japanese
armed forces, had joined the Gunkes, a corps of volunteer combatants composed mainly of Indonesian
natives serving with the Japanese army. The Court held that by joining the Japanese forces, Barends had lost
his nationality. His killing by Japanese forces was not considered a war crime (at 682-4).
2 See Tadic (TJ), at §573; Delalic and others (TJ, §193).
3 See the following decisions by TCs: Akayesu (§§630-4, 638—44), Kayishema and Ruzindana (§§185-9,
590-624), Musema (§§259-62, 275, and 974). In all these cases the Court eventually found that the link
required was lacking.
 WAR CRIMES 83

international hum anitarian law of armed conflict. This applies in particular to offences
committed by civilians, although courts have also required the link or nexus with an
armed conflict in the case of crimes perpetrated by members of the military.4
Special attention should be paid to crimes committed by civilians against other civil­
ians. They may constitute war crimes, provided there is a link or connection between the
offence and the armed conflict. In the absence of such a link, the breach simply consti­
tutes an ordinary’ criminal offence under the law applicable in the relevant territory.5

4 In this respect a case worth mentioning is Lehnigk and Schuster, decided by the Italian Court of Assize
of S. Maria Capua Vetere in 1994. In October 1943, after Italy had declared war against Germany and while
the German troops were pulling out as a result of the military advance of the Allied forces in Southern Italy,
a German unit including the two accused killed 22 Italian civilians who had taken shelter in a farm, to avoid
being caught in the adverse consequences of the armed conflict under way. A case was brought against the
two Germans in absentia (in Germany one of the two accused had been acquitted because the crime was
covered by a statute of limitation, while the legal condition of the other was unclear, although criminal pro­
ceedings had been instituted against him). The Italian Court first asked itself whether the crime with which
the two accused were charged should be regarded as ordinary murder or ‘murder against the laws and cus­
toms of war’, or in other words a war crime (at 8). In this respect the Court stated that a murder may amount
to war crime only if it was proved that there exists ‘an objective link [of the offence] with the demands of war’
or, in other words, if the offence had ‘a war-like nature’; namely it had a link with war and did not ‘prove to
be generically linked to war’ (at 9). The Court then dwelled at length on the facts and concluded that what
some witnesses had stated (namely that the German unit had killed the civilians in the farm, in the dark,
because they had seen light signals from the farm and feared that there could be partisans or enemy troops)
was not correct; the killing was not carried out as a response to, or out of fear of, enemy action, and did not
serve any military purpose; indeed the Germans had killed the civilians only out of intolerance and hatred
for the Italian people’ (at 26-30); hence, the murder was not linked to war and could not be classified as a war
crime (at 30). That these conclusions totally lack legal merit is patent: the Court undisputedly misinterpreted
the laws of war. Clearly, even assuming that the killing only resulted from hatred, it was still a war crime:
subjective motives do not have legal relevance in this context.
In Kunarac and others the ICTY AC clarified the link between a criminal offence and an armed conflict
necessary for the offence to constitute a war crime: ‘What ultimately distinguishes a war crime from a purely
domestic offence is that a war crime is shaped by or dependent upon the environment—the armed con­
flict—in which it is committed. It need not have been planned or supported by some form of policy. The
armed conflict need not have been causal to the commission of the crime, but the existence of an armed
conflict must, at a minimum, have played a substantial part in the perpetrator’s ability to commit it, his
decision to commit it, the manner in which it was committed or the purpose for which it was committed.
Hence, if it can be established, as in the present case, that the perpetrator acted in furtherance of or under the
guise of the armed conflict, it would be sufficient to conclude that his acts were closely related to the armed
conflict. The Trial Chamber’s finding on that point is unimpeachable’ (AJ §58). The AC added that ‘In deter­
mining whether or not the act in question is sufficiently related to the armed conflict, the Trial Chamber may
take into account, inter alia, the following factors: the fact that the perpetrator is a combatant; the fact that
the victim is a non-combatant; the fact that the victim is a member of the opposing party; the fact that the
act may be said to serve the ultimate goal of a military campaign; and the fact that the crime is committed as
part of or in the context of the perpetrator’s official duties’ (§59).
In Brdanin the TC had found that members of the Bosnian Serb police and the army had committed rapes
in Teslic during weapons’ searches (TJ, §§140, 154, 523). On appeal the defendant argued that these were
‘individual domestic crimes’ with no link with the armed conflict. The AC dismissed the contention, stating
that the rapes had been perpetrated in the context of an armed conflict: ‘crimes committed by combatants
and by members of forces accompanying them while searching for weapons during an armed conflict, and
taking advantage of their position, clearly fall into the category of crimes committed “in the context of the
armed conflict” ’ (AJ §256).
5 The Swiss Appellate Military Tribunal aptly confirmed this proposition in Niyonteze. The accused was a
Rwandan arrested in Switzerland and accused of having instigated, and in some cases ordered, the murder of
civilians in Rwanda in 1994 in his capacity as mayor of a local ‘community’ (commune). The Tribunal could
 84 INTERNATIONAL CRIMINAL LAW

4.3 ESTABLISHING WHETHER A SERIOUS

VIOLATION OF IHL HAS BEEN
CRIMINALIZED
As pointed out above, in order for a serious violation of IHL to become a war crime, it
is necessary that the violation be criminalized. The question then becomes one of how
to determine whether this is the case.
The point of departure is the observation that the failure of the relevant rules of IHL
to provide for any courts or criminal proceedings in the event of the rule being breached
is not determinative of the issue. W hat matters is that criminal or m ilitary courts have
in fact adjudicated breaches of IHL. Various courts rightly held this view.6
A second, general and preliminary, remark concerns the need to avoid the follow­
ing simplistic proposition: to determine whether a particular act may be termed a war
crime, one need only establish that the act breaches IHL, since all violations of the laws
of war are war crimes under national law and military manuals. The Judge Advocate at
a Canadian Military Court pronouncing in 1946 on a war crime in Johann Neitz took
this view. After noting that, under Canadian law, a war crime was any ‘violation of the
laws and usages of war committed during any war in which Canada had been or may
be engaged at any tim e’, the Judge Advocate added

The test of criminal responsibility is therefore not properly applicable, and the issue upon
any charge is not ‘did the accused commit a crime?’as we understand the word ‘crime’under
our criminal law, but ‘did he violate the laws and usages of war’? (195-6).

not apply the Genocide Convention since Switzerland had not yet ratified it. The Tribunal held, therefore,
that it would apply the laws of warfare and the provisions of the Geneva Conventions applicable to internal
armed conflicts as well as the Second Protocol of 1977. Faced with the question whether a civilian could be
held responsible for war crimes where he had instigated or ordered the murder of other civilians, the Tribu­
nal held that ‘Anyone, whether military or civilian, who attacks a civilian protected by the Geneva Conven­
tions [...] breaches these Conventions and consequently falls under Article 109 of the Swiss Penal Military
Code [providing for the punishment of war crimes]. This Appellate Tribunal thus differs from the judgments
of the ICTR, which require a close link between the breach and an armed conflict and confine the applica­
tion of the Geneva Conventions to persons discharging functions within the armed forces or the civilian
government (Musema §§259[-62] and Akayesu §§642-3). Nevertheless this Tribunal considers that in any
case there must exist a link between the breach and an armed conflict. If, within the framework of a civil war,
where civilians of the two sides are both protected by the Geneva Conventions, a protected person commits
a breach against another protected person, it is necessary to establish a link between this act and the armed
conflict. If such link is lacking, the breach does not constitute a war crime but an ordinary offence (infraction
de droit commun)’ (39-40). In the case at bar, the Tribunal found this link in the fact that the accused was the
mayor of the commune, and exercised dejure and defacto authority over the local citizens; it was thus in his
capacity as a ‘public official’ or civil servant that he committed the crimes (40-1).
The Tribunal Militatre de Cassation upheld the ruling in its decision of 27 April 2001 on the same
case (§9).
6 See, for instance, the IMT in Goring and others (at 220-1), a US Military Tribunal sitting at Nuremberg
in List and others (the so-called Hostages case) (at 635), and in Ohlendorf and others (the so-called
Einsatzgruppen case) (at 658), as well as the US Supreme Court in Ex parte Quirin (at 465).
 WAR CRIMES 85

This approach is not convincing, as not all violations of international hum anitarian
law amount to war crimes, as pointed out in Tadic (IA) (§94), although they may give
rise to state responsibility.
These points having been established, several situations need to be distinguished.
First, it may be that a violation has been consistently considered a war crime by national
or international courts (this is, for example, true for the most blatant violations, such as
unlawfully killing prisoners of war or innocent civilians, shelling hospitals, refusing
quarter, killing shipwrecked or wounded persons, and so on). The existence of war
crimes cases on a particular matter may sometimes be considered sufficient for holding
the breach to be a war crime. However, strictly speaking, the existence of a few (possibly
isolated) war crimes decisions may not be enough. It would be better if it were possible
to show that the breach is considered a war crime under customary international law,
in which case there would have to be widespread evidence that states customai iiy pros­
ecute such breaches as war crimes and that they do so because they believe themselves
to be acting under a binding rule of international law {opinio juris).
A second possible instance is that a breach is termed a war crime by the Statute of an
international tribunal. In this case, even if the breach has never been brought before a
national or international tribunal, it may justifiably be regarded as a war crime—or, at
least, as a war crime falling under the jurisdiction of that international tribunal.
A third, and more difficult, category is when the case law and statutes of international
tribunals are absent or silent on the m atter/ In such a case, how is one to determine
whether violating a prohibition of international hum anitarian law amounts to a war
crime? In light of the case law (see List and others (Hostages case), John G. Schultz,
Tadic (IA), and Blaskic, to which I will presently return) and the general principles
of ICL, one is entitled, in seeking an answer to the question, to examine: (1) military
manuals; (ii) the national legislation of states belonging to the major legal systems of
the world; or, if these elements are lacking, (iii) the general principles of criminal just­
ice common to nations of the world, as set out in international instruments, acts, reso­
lutions and the like; and (iv) the legislation and judicial practice of the state to which
the accused belongs or on whose territory the crime has allegedly been committed.
Let us now take a look at how courts have gone about this matter.
In List and others (Hostages case) the defendants were high-ranking officers in
the German armed forces charged with war crimes and crimes against humanity.
They were accused of offences committed by troops under their command during
the occupation of Greece, Yugoslavia, Albania, and Norway, these offences mainly
being reprisal killings, purportedly carried out in an attempt to m aintain order in the
occupied territories in the face of guerrilla opposition, or wanton destruction of prop­
erty not justified by military necessity. They claimed that Control Council Law no. 10,
on the basis of which they stood accused, was an ex post facto act and retroactive in
nature. The Tribunal rejected the contention, holding that the crimes defined in that 7

7 An example is the prohibition on the use of weapons that are inherently indiscriminate or cause
unnecessary suffering.
 86 INTERNATIONAL CRIMINAL LAW

Law were crimes under pre-existing rules of international law, ‘some by conventional
law and some by customary law’. It went on to state that the war crimes at issue were
such under the Hague Regulations of 1907 and then added
In any event, the practices and usages of war which gradually ripened into recognized cus­
toms with which belligerents were bound to comply, recognized the crimes specified herein
as crimes subject to punishment. It is not essential that a crime be specifically defined and
charged in accordance with a particular ordinance, statute or treaty if it is made a crime by
international convention, recognized customs and usages or war, or the general principles
of criminal justice common to civilized nations generally (634-5).

The Tribunal then noted that the acts at issue were traditionally punished, adding
that, although no courts had been established nor penalties provided for the commis­
sion of these crimes, ‘this is not fatal to their validity. The acts prohibited are without
deterrent effect unless they are punishable as crimes’ (635).
It was the AC of the ICTY that best addressed the issue under discussion, in
Tadic (IA). The question in dispute was whether the accused could be held crim in­
ally liable for breaches of international hum anitarian law allegedly committed in an
internal armed conflict; in other words, whether he could be held responsible for war
crimes perpetrated in a civil war. The AC first considered whether there were cus­
tom ary rules of international hum anitarian law governing internal armed conflicts,
and answered in the affirmative (§§96-127). It then asked itself whether violations
of those rules could entail individual crim inal responsibility. For this purpose, the
C ourt examined national cases, m ilitary manuals, national legislation, and resolu­
tions of the UN Security Council. It concluded in the affirmative (§§128-34) and then
added that in the case at issue this conclusion was fully warranted ‘from the point of
view of substantive justice and equity’, because violations of IHL in internal armed
conflicts were punished as crim inal offences in the countries concerned, that is both
the old Socialist Federal Republic of Yugoslavia and in Bosnia and Herzegovina. As
the C ourt noted, ‘Nationals of the former Yugoslavia as well as, at present, those of
Bosnia-Herzegovina were therefore aware, or should have been aware, that they were
amenable to the jurisdiction of their national criminal courts in cases of violation of
international hum anitarian law’ (§135; see also §136).
An ICTY TC returned to the question in Blaskic. The defence contended that viola­
tions of common Article 3 of the four 1949 Geneva Conventions (on internal armed
conflict) did not entail criminal liability. The TC dismissed this contention by noting,
first, that those violations were envisaged in Article 3 of the ICTY Statute, conferring
jurisdiction on the Tribunal, and secondly, that the criminal code of Yugoslavia, taken
over in 1992 as the criminal code of Bosnia and Herzegovina (the place where the
alleged offences had been committed), provided that war crimes perpetrated either
in international or in internal armed conflicts involved the criminal liability of the
perpetrator (§176).8

8 The question was also dealt with, albeit in less compelling terms, by a US Court of Military Appeals
m John G. Schultz. The accused, a former captain of the US Air Force who had returned to civilian life had
 WAR CRIMES 87

4.4 THE OBJECTIVE ELEMENTS

4.4.1 GENERAL

In order to identify the main legal features of the prohibited conduct, it is necessary
to consider in each case the content of the substantive rule that has been allegedly
breached. This should not be surprising. No authoritative and legally binding list of
war crimes exists in customary law. An enumeration can only be found in Article 8
of the ICC Statute, which is not, however, intended to codify customary law. It should
also be noted, more generally, that the principle of legality or nullum crimen sine lege
(traditionally laid down in national legal systems, particularly those of civil law coun­
tries) is upheld in ICL only in a limited way (see supra, 2.3). Hence in each case the
objective element of the crime can essentially be inferred from the substantive rule of
international hum anitarian law allegedly violated.
For a subcategory of war crimes, namely those acts that are provided for in terms
and defined by the 1949 Geneva Conventions and Additional Protocol I of 1977 as
‘grave breaches’, a further requirement is provided for: such acts must be committed
within the context of an international armed conflict. The ICTY AC held in Tadic (IA)
that a customary rule was in statu nascendi, that is in the process of forming, whereby
‘grave breaches’ could also be perpetrated in internal armed conflicts; instead, accord­
ing to Judge Abi-Saab’s Separate Opinion in that case, such a rule had already evolved.

killed two Japanese pedestrians in Japan in 1950. He was tried by a US General Court Martial on charges of
involuntary manslaughter and drunken driving, in violation of Articles of War (respectively, 93 and 96). The
Judge Advocate General of the Air Force appealed the case on, among other grounds, the issue of whether
the Court Martial had jurisdiction over the accused and the offences charged. The Court of Appeals, having
found that the accused was neither a ‘retainer to the camp’ nor a ‘person accompanying or serving with the
US Armies’, hence not amenable to a US Court Martial’s jurisdiction on these grounds, asked itself whether
he fell under the category o f‘any other person who by the law of war is subject to trial by military tribunals.
To answer this question it noted, among other things, that US jurisdiction extended to two types of offences:
first, crimes committed against the civilian population made ‘punishable by the penal codes of all civilized
nations’, namely war crimes; secondly, ‘crimes condemned by local statute which the military occupying
power must take cognizance of inasmuch as the civil authority is superseded by the military’. The Court
first looked into the first category, to establish whether the offence at issue fell within such category. Hav­
ing reached a negative conclusion, it turned to the second category, and concluded that the offence came
within its purview. Let us now briefly see how the Court discussed the class of war crimes in a lengthy obiter
dictum.
The Court noted that this category ‘finds its basis in the customs and usages of civilized nations’. It then
went on to say that, ‘It is [...] no obstacle to finding a particular offence to be a violation of the law of war that
it has not yet been precisely labelled as such. On the other hand, of course, we are not free to add offences
at will. In deciding whether an offence comes within the common law of war, we must consider the inter­
national attitude towards that offence. The power to define such offences is derived from Articles of War 12
and 15 [...] and it is no objection that Congress has not codified that branch of international law or defined
the acts which that law condemns [.. .j We shall assume that a crime may become a violation of the law of
war if universally recognized as an offence even though it contains no element of specific criminal intent. A
careful perusal of the penal codes of most civilized nations leads us to the conclusion that homicide involv­
ing less than culpable negligence is not universally recognized as an offence’ (114-16).
 88 INTERNATIONAL CRIMINAL LAW

At present, in light of the recent trends in the legislation or practice of states,9 the con­
tention is perhaps warranted that a customary rule has indeed emerged. However, it
would seem plausible to interpret this rule to the effect that it only confers on states
the power to search for and bring to trial or extradite alleged authors of grave breaches
committed in internal armed conflicts; the rule does not go so far as to also impose
upon states an obligation to seek out and try or extradite those alleged authors (as is
instead the case for grave breaches perpetrated in international armed conflicts).

4.4.2 C LA SSES OF W A R C R IM E

War crimes can be classified under different headings. The following classification
is based on some objective criteria, and may prove useful, although of course it only
serves descriptive purposes: (i) war crimes committed in international armed conflicts
(that is, between two or more states, or between a state and a national liberation move­
ment, pursuant to Article 1(4) of the First Additional Protocol of 1977); and (ii) war
crimes perpetrated in internal armed conflicts (that is, large-scale armed hostilities,
other than internal disturbances and tensions, or riots or isolated or sporadic acts
of armed violence, between state authorities and rebels, or between two or more
organized armed groups within a state). Traditionally, states and courts have held that
war crimes may only be committed during wars proper. Violations of international
law committed in the course of internal armed conflicts were not criminalized. Thus,
a glaring and preposterous disparity existed. As stated above, in 1995, a seminal judg­
ment of the ICTY AC in Tadic (IA) (§§97-137) signalled a significant advance: the AC
held that war crimes could be committed not only in international armed conflicts but
also in internal armed conflicts. Since then the view has been generally upheld and the
ICC Statute definitively consecrates it in Article 8(2)(c)-(f).
Both classes include the following:

1. Crimes committed against persons not taking part, or no longer taking part, in
armed hostilities. In practice by far the most numerous crimes are committed against
civilians,101or armed resistance movements in occupied territory,11 and include sexual

9 For instance, Article 8 of the Netherlands Criminal Law in Time of War Act (1952) provides that
national courts have jurisdiction over all violations of the laws and customs of war. The law has been
interpreted to apply to internal conflicts as well (Article 1 (3) states that the term ‘war’ includes civil war).
Article 3 of that law provides that courts may exercise universal jurisdiction over violations of the laws and
customs of war. In Switzerland an amendment to the Criminal Code of 13 December 2002 provides for crim­
inal jurisdiction over violations of IHL in internal armed conflict as well. In Germany Section 1 of the Code
of Crimes Against International Law applies the universality principle to all international crimes such as
genocide, war crimes, and crimes against humanity, whether or not committed in Germany. This provision
is strengthened by Section 153ff of the Code of Criminal Procedure, which also lays down a duty of investi­
gation and prosecution for international crimes committed abroad.
10 See, for instance, von Falkenhausen and others (at 867-93), Bellmer (at 541-4), Lages (at 2-3), Wagener
and others (at 148), Sch. O. (at 305-7), Sergeant W. (decision of 18 May 1966, at 1-3; decision of 14 July 1966,
at 2). For fairly recent cases see for instance Major Malinky Shmuel and others (at 10-137), Calley (at 1164-84),
Tzofan and others (Yehuda Meir case) at 724-46, Sablic and others (at 37-135).
11 See, for instance, the SIPO Brussels case (at 11518-26), Aliens and others (at 225-47).
 WAR CRIMES 89

violence against women.12 In particular, they are perpetrated against persons detained
in internm ent or concentration camps.13 They are also committed against prisoners
of war.14
In the case of international armed conflicts, serious war crimes against one of the
‘protected persons’ (wounded, shipwrecked persons, prisoners of war, civilians on
the territory of the Detaining Power or subject to the belligerent occupation of an
Occupying Power) or protected objects’ provided for in the 1949 Geneva Conventions,
as well as the First Additional Protocol are termed ‘grave breaches’. Grave breaches are
defined in the following provisions: Articles 50, 51, 130, and 147 of the First, Second,
Third, and Fourth Geneva Conventions, respectively, as well as in Article 85 of the
First Additional Protocol. They include wilful killing, torture or inhum an treatment,
including biological experiments, wilfully causing great suffering or serious injury to
body or health, extensive destruction and appropriation of property, not justified by
military necessity and carried out unlawfully and wantonly. The essential feature of
‘grave breaches’ is that they are subject to universal jurisdiction’ of all states parties to
the Convention and the Protocol: any contracting state is authorized as well as obliged
to search for and bring to trial—or, alternatively, extradite to a requesting state—any
person suspected or accused of a grave breach (whatever his or her nationality and the
territory where the grave breach has allegedly been perpetrated) who happens to be
on its territory.

12 In this respect it is worth mentioning two cases brought after the Second World War before the Dutch
Temporary Court Martial in Batavia (Indonesia). The first is Washio Awochi. The accused, a Japanese civil­
ian who managed a club for Japanese civilians in Indonesia, had procured or arranged the procurement of
girls and women for the club’s visitors, forcing them into prostitution; they were not free to leave the part
of the club where they had been confined. The Court held that the defendant was guilty of the war crime
o f‘forcing into prostitution and sentenced him to 10 years’ imprisonment (at 1-15). In Takeuchi Hiroe the
accused, a Japanese national, had used violence or threats of violence against a young Indonesian woman,
and had forced her to have sexual intercourse with him. The Court found him guilty of the war crime of rape
and sentenced him to five years’ imprisonment (at 1-5).
See also some cases of rape brought before the ICTY: Furundzija (§§165-89) and Kunarac and others
(§§436-64 and 630-87,717-45, 785-98, 806-22).
13 Among the numerous cases on this matter, one may recall various ones concerning the ill-treatment
of persons detained in the concentration camps instituted in Poland, such as Auschwitz (see Mulka and
others), in Germany, at Dachau (see Martin Gottfried and others), by the German occupying troops in
Majdanek (see Götzfrid, at 2-70), in camps in Belgium (see, for instance, Köpperlmann as well as K.W. (at
565-7) and K. (at 653-5), in Amersfoort (Netherlands) (see for instance Kotälla), or in Bolzano (Italy) (see,
for instance, Mittermair, at 2-5, Mitterstieler, at 2-7, Lanz, at 2-4, Cologna, at 2-9, Koppelstatter and oth­
ers, at 3-7) or in the Italian camp of Fossoli (see Gutweniger, at 2-4), or in internment camps in the former
Yugoslavia (see, for instance, Saric, 2-6). Such crimes may even be perpetrated by internees against other
internees (see, for instance, Ternek, at 3-11, and Enigster, at 5-26).
14 See, for instance, some cases brought after the First World War before the Leipzig Supreme Court.
Heynen (at 2543-7), Müller (at 2549-52) and Neumann (at 2553-6). See also other cases, relating to the Sec­
ond World War: Mälzer (at 53-5), Feurstein and others (at 1-26), Krauch and others (at 668-80), Weiss and
Mundo (at 149), Gozawa Sadaichi and others (at 195-228), General Seeger and others (Vosges case), at 17-22;
St Die case, at 58-61; La Grande Fosse case, at 23-7; Essen lynching case, at 88-92.
 90 INTERNATIONAL CRIMINAL LAW

In the case of internal armed conflict,15 the same violations are prohibited and may
amount to a war crime if they are serious. In this connection reference should be made
to Article 3 common to the four 1949 Geneva Conventions, Additional Protocol II
(especially Article 4 thereof),16 as well as Article 4 of the ICTR Statute.17 As noted
above, there is no treaty provision characterizing violations of these rules as ‘grave
breaches and consequently attaching to such classification all the ensuing conse­
quences at the procedural level (power and duty to exercise universal jurisdiction over
the alleged offender). Nor, it would seem, has a customary rule evolved imposing upon
states (and the rebellious group engaged in a civil war) the obligation to search for and
bring to trial (or extradite) persons suspected or accused of a grave breach perpetrated
in an internal armed conflict.
2. Crimes against enemy combatants or civilians, committed by resorting to pro­
hibited methods of warfare. Examples include intentionally directing attacks against
the civilian population in the combat area or individual civilians in the combat area
not taking part in hostilities; committing acts or threats of violence the prim ary pur­
pose of which is to spread terror among the civilian population; intentionally launch­
ing an indiscriminate attack affecting the civilian population or civilian objects in
the knowledge that such attack will cause excessive loss of life, injury to civilians, or
damage to civilian objects; intentionally making non-defended localities or demilita­
rized zones the object of attack; intentionally making a person the object of attack in
the knowledge that he is hors de combat; intentionally attacking medical buildings,
material, medical units and transport, and personnel; intentionally using starvation
of civilians, as a method of warfare by depriving civilians of objects indispensable
to their survival, including wilfully impeding relief supplies; intentionally launching
an attack in the knowledge that such attack will cause widespread, long-term, and
severe damage to the natural environment; utilizing the presence of civilians or other
protected persons with a view to rendering certain points, areas, or military forces
immune from military operations; declaring that no quarter will be given, that is, that
enemy combatants will be killed and not taken prisoner.
It should be noted that, the substantive rules of IHL on this m atter being pur­
posely loose, so far very few cases have been brought before national or international
courts concerning alleged violations of rules on the conduct of hostilities entailing
the criminal liability of the perpetrators.18 Strikingly, more cases involving the alleged

For a case where a court has endeavoured to define the notion o f‘internal armed conflict’ see Ministere
public and Centre pour Vegalite des chances et la lutte contre le racism v. C. and B. (at 5-7). Other cases where
courts had to pronounce on whether or not the conflict was internal, include: Osvaldo Romo Mena (decision
o t e Supreme Court of Chile of 26 October 1995, at 3, and decision of 9 September 1998, at 2-5) Chilean
state o f emergency case (at 1-3), and G. (Swiss Military Tribunal, at 7).
„ 16 * “ “ where a court has held that Additional Protocol II was applicable, see Applicability o f the
econd Additional Protocol to the Conflict in Chechnya (Chechnya case) (at 2-3). See also Constitutional
Conformity of Protocol II (§25).
17 For a case of war crimes in civil war, see Nwaoga (at 494-5).
18 For instance, see the General Jacob H. Smith case, decided by a US Court Martial on 3 May 1902
concerning the order that no quarter should be given (at 799-813). Before the Nuremberg IMT Admirals
 WAR CRIMES 91

breach of rules of IHL on the conduct of hostilities have been brought before interstate
courts, pronouncing on state responsibility.19
3. Crimes against enemy combatants and civilians, involving the use of prohibited
means o f warfare. Examples include employing weapons, projectiles, and materials
which are of a nature to cause superfluous injury or unnecessary suffering; employ­
ing poison or poisoned weapons, or asphyxiating, poisonous, or other gases, and all
analogous liquids, materials, or devices; using chemical or bacteriological weapons;
employing expanding bullets or weapons, the prim ary effect of which is to injure by
fragments not detectable by X-rays, or blinding laser weapons;20 employing booby-
traps or land mines indiscriminately, that is, in such a way as to hit both combatants and
civilians alike, or anti-personnel mines which are not detectable; employing napalm and
other incendiary weapons in a manner prohibited by the 1980 Protocol III to the afore­
mentioned Convention (for instance, by making a military objective ‘located within a
concentration of civilians the object of attack by air-delivered incendiary weapons’).
W hat I have pointed out above, with regard to breaches of international rules on
methods of war, a fortiori applies to violations of rules on means of warfare, the latter
category of rules being even more difficult to apply than the legal standards on the
conduct of hostilities.
4. Crimes against specially protected persons and objects (such as medical per­
sonnel units or transport, personnel participating in relief actions, hum anitarian

Donitz and Reeder were charged with, but acquitted of, waging unrestricted submarine warfare (see Trial of
the Major War Criminals, vol. I, 311-12 and 316-17). See also the William L. Calley case, revolving around
the killing of Vietnamese civilians in the village of My Lai (see the Instructions of the Military Judge to the
Court Martial, March 1971, at 1703-27, as well as the decision of 16 February 1973 of the US Army Court o
Military Review, 1131). The van Anraat case, brought before the Hague District Court and subsequently t e
Hague Appeal Court, relates to complicity in a war crime (the accused sold chemicals for the manufacture
of prohibited chemical weapons to Iraqi authorities).
As for international courts, see ICTY, TC Blaskic (unlawful attacks against civilians and civilian prop­
erty; destruction of institutions dedicated to religion); Galic (on sniping and shelling at civilians in Sara­
jevo); Strugar (on the shelling of Dubrovnik); Martic (on the shelling of Zagreb).
On this issue, see the important remarks by P. Gaeta, ‘A neglected category of war crimes; violations of the
law on the conduct of hostilities’ 19 EJIL (2008), forthcoming.
19 For instance, see some cases brought before the Eritrea-Ethiopia Claims Commission, Partial Award,
Central front, Ethiopia’s Claim 2, judgment of 28 April 2004, as well as the judgment of 19 December 2005
(Partial Award, Western and Eastern Fronts, Ethiopia’s Claims l and 2). See also a few cases brought before the
European Court on Human Rights; Isayeva, Ysupova and Bazaieva v. Russia, Isayevav. Russia and Khatsiyeva
and others. See also some cases brought before the Inter-American Court of Human Rights: Plande Sanchez
Massacre v. Guatemala; case Las Palmerasv. Colombia; case of the massacre ofMapiripan v. Colombia. See also
the decision of the Inter-American Commission of Human Rights on Juan Carlos Abella Argentina.
Seldom have national courts pronounced upon the conduct of hostilities, and always within the context o
civil, not criminal, action. See for instance Shimoda and others v. The State ofJapan, judgment of the District
Court of Tokyo (on the atomic bombing of Hiroshima and Nagasaki), at 1688-1702.
20 According to the definition of the 1995 Protocol IV to the Convention on Prohibitions or Restrictions
on the Use of Certain Conventional Weapons Which Maybe Deemed to be Excessively Injurious or to Have
Indiscriminate Effects, adopted at Geneva on 10 October 1980, the latter are ‘laser weapons specifically
designed, as their sole combat function or as one of their combat functions, to cause permanent blindness to
un-enhanced vision, that is to the naked eye or to the eye with corrective eyesight devices.’
 92 INTERNATIONAL CRIMINAL LAW

organizations such as the Red Cross, or Red Crescent, or Red Lion and Sun units, UN
personnel belonging to peace-keeping missions, etc.).
5. Crimes consisting of improperly using protected signs and emblems (such as a flag
of truce; the distinctive emblems of the Red Cross, or Red Crescent, or Red Lion and
Sun, plus the emblem provided for in the Third Additional Protocol of 8 December
2005 (the emblem composed of a red frame in the shape of a square on edge on a white
ground ); perfidious use of a national flag or of military uniform and insignia, etc.).
6. Conscripting or enlisting children under the age o f fifteen years or using them to
participate actively in hostilities (in either international or internal armed conflicts).
According to the AC of the SCSL (Norman, Decision on Preliminary Motion Based on
Lack o f Jurisdiction, §53) child recruitment was criminalised before it was explicitly
set out as a criminal prohibition in treaty law and certainly by November 1996, the
starting point of the time frame relevant to the indictment’(against the defendants
in that case). This proposition was restated by a TC of the SCSL in Brima and others
(§§727-8), where the elements of the crime were set out (§729).

4.5 THE SUBJECTIVE ELEMENTS

The subjective or mental element (mens rea) of the crime is sometimes specified by
the international rule prohibiting a certain conduct.
Thus, for instance, Article 130 ofthe Third Geneva Convention of 1949 (on prisoners
of war) enumerates among the grave breaches’ of the Convention the ‘wilful killing [of
prisoners of war], torture or inhum an treatment, including biological experiments’ as
well as wilfully causing great suffering or serious injury to body or health’ of a prison­
ers of war, or 'wilfully depriving a prisoner of war of the rights of fair and regular trial
prescribed in [the] Convention. The word ‘wilful’ obviously denotes criminal intent,
namely the intention to bring about the consequences of the act prohibited by the
international rule (for instance, in the case of ‘wilful killing’ proof must be produced
ofthe intention to cause the death of the victim; in the case o f ‘wilfully causing great
suffering it must be proved that the perpetrator had the intention to cause great suf­
fering, etc.). The same holds true for other similar provisions, such as Article 147 of the
Fourth Geneva Convention (on civilians), as well as provisions of other treaties, such as
Article 15 ofthe 1999 Second Hague Protocol for the Protection of Cultural Property
in the Event of Armed Conflict. (This provision, in enumerating the serious violations
ofthe Protocol entailing individual criminal liability, makes such liability contingent
upon the fact that the author of the ‘offence’ has perpetrated it ‘intentionally’.)
One can also mention Article 85(3) of the First Additional Protocol of 1977. This
provision subordinates the criminalization of such acts as attacking civilians or
undefended localities, or demilitarized zones, or perfidiously using the distinctive
emblem of the Red Cross, Red Crescent or Red Lion and Sun, to three conditions:
(i) the acts must be committed ‘wilfully’; (ii) they must be carried out in violation ofthe
 WAR CRIMES 93

relevant provisions of the Protocol; and (iii) they must cause death or serious injury to
body or health. Thus, the provisions clearly require intent or at least recklessness (so-
called dolus eventualis), which exists whenever somebody, although aware of the likely
pernicious consequences of his conduct, knowingly takes the risk of bringing about
such consequences (see supra, 3.7).
For other acts, the same provision also requires ‘knowledge as a condition of crim ­
inal liability. This, for instance, applies to ‘launching an indiscriminate attack affect­
ing the civilian population or civilian objects in the knowledge that such attack will
cause excessive loss of life, injury to civilians or damage to civilian objects’ (Article
85(3)(b)); or to ‘launching an attack against works or installations containing danger­
ous forces in the knowledge that such attack will cause excessive loss of life, injury to
civilians or damage to civilian objects’ (Article 85(3)(c)). As we have seen (3.6.1), in
criminal law ‘knowledge’ is normally part of ‘intent’ (dolus) and refers to awareness
of the circumstances forming part of the definition of the crime. However, in the con­
text of the provision at issue, ‘knowledge’ must be interpreted to mean ‘predictability
of the likely consequences of the action’ (recklessness or dolus eventualis). Therefore,
for an act such as that just mentioned to be regarded as a war crime, evidence must
be produced not only of the intention to launch an attack, for instance an attack on
a military objective normally used by civilians (e.g. a bridge, a road, etc.), but also of
the foreseeability that the attack was likely to cause excessive loss of life or injury to
civilians or civilian objects. In other instances, international rules require knowledge
in the sense of awareness of a circumstance of fact, as part of crim inal intent (dolus).
Thus, Article 85(3)(e) of the same Protocol makes it a crime to wilfully attack a person
‘in the knowledge that he is hors de com bat.
W hen international rules do not provide, not even implicitly, for a subjective element,
it would seem appropriate to hold that what is required is the intent or, depending upon
the circumstances, recklessness as prescribed in most legal systems of the world for the
underlying offence (murder, rape, torture, destruction of private property, pillage, etc.).
Often, international courts and tribunals have gradually identified the requisite mental
element based on the nature of the underlying offence. Thus, for instance, in the case
of murder as a war crime, the jurisprudence of the ICTR and the ICTY has consist­
ently held that what is required is that ‘the death of the victim must result from an act
or omission of the accused committed with the intent either to kill or to cause serious
bodily harm in the reasonable knowledge that it would likely result in death’ (Krstic, TJ,
§483; Blaskic, TJ, §217; Kvocka and others, TJ, §132; Stakic, TJ, §§584-6). In other words,
either intent or at least dolus eventualis or recklessness (see supra, 3.7) are required.21
Generally speaking, it appears admissible to contend that, for at least some limited
categories of war crimes, gross or culpable negligence (culpa gravis) may be sufficient;
that is, the author of the crime, although aware of the risk involved in his conduct,

21 As an ICTY TC held in Stakic, ‘both a dolus directus and a dolus eventualis are sufficient to establish
the crime of murder under Article 3. In French and German law, the standard form of criminal homicide
(:meurtre, Totschlag) is defined simply as intentionally killing another human being. German law takes dolus
eventualis as sufficient to constitute intentional killing’ (§587).

i
 94 INTERNATIONAL CRIMINAL LAW

is nevertheless convinced that the prohibited consequence will not occur (whereas
in the case of ‘recklessness’ or dolus eventualis the author knowingly takes the risk).
supra, 3.8. Indeed, the consequent broadening of the range of acts amenable to inter­
national prosecution is in keeping with the general object and purpose of international
hum anitarian law. This modality of mens rea may, for instance, apply to cases of com­
m and responsibility (see infra, 11.4.4), where the commander should have known that
war crimes were being committed by his subordinates. Also, it could be contended
that it may apply to such cases as wanton destruction of private property. In contrast,
it may seem difficult to consider culpable negligence a sufficient subjective element of
the crime in cases involving the taking of hum an life.

4.6 THE DEFINITION OF WAR CRIMES IN

THE ICC STATUTE

Generally speaking, the Rome Statute appears to be praiseworthy in many respects

as far as substantive criminal law is concerned. Many crimes have been defined with
the required degree of specificity, and the general principles of criminal liability have
been set out in detail.
As far as war crimes more specifically are concerned, it is no doubt commendable
that they have been regulated in such a detailed manner. Furthermore, the notion of
war crimes has rightly been extended to offences committed in time of internal armed
conflict. However, in some areas the relevant provision of the Rome Statute, Article 8,
marks a retrograde step with respect to existing international law.
First of all, there is a perplexing phrase, ‘within the established framework of
international law , that appears in Article 8(2)(b) and (e), dealing with crimes likely to
be perpetrated while in combat (that is, crimes involving the wrongful use of means
or methods of combat), respectively in international armed conflicts and in non­
international armed conflicts. These two provisions are worded as follows:
[For the purpose of this Statute ‘war crimes’ means] Other serious violations of the laws and
customs applicable in international armed conflict [in armed conflicts not of an international
character: litt (e)], within the establishedframework of international law, namely any of the
following acts.

It is notable that in the other provisions of Article 8 no mention is made of ‘the

established framework of international law’. Hence one could argue that there is only
one possible explanation of this odd phrase: the offences listed in the two aforemen­
tioned provisions are to be considered as war crimes for the purpose of the Statute only
if they are regarded as such by customary international law. In other words, whilst for
the other classes of war crimes the Statute confines itself to setting out the content of
the prohibited conduct, and the relevant provision can thus be directly and immedi­
ately applied by the Court, in the case of the two provisions under consideration things
 WAR CRIMES 95

are different. The Court may consider that the conduct envisaged in these provisions
amounts to a war crime only if and to the extent that general international law already
regards the offence as a war crime. It would follow, for example, that ‘declaring that no
quarter will be given’ (Article 8(2)(b)(xii)) will no doubt be taken to amount to a war
crime, because indisputably denial of quarter is prohibited by customary international
law and, if effected, amounts to a war crime. By contrast, offences such as ‘The transfer,
directly or indirectly, by the Occupying Power of parts of its own civilian population
into the territory it occupies, or the deportation or transfer of all or parts of the popu­
lation of the occupied territory within or outside this territory’ (Article 8(2)(b)(viii))
cannot ipso facto be regarded as war crimes. The Court will first have to establish
whether: (i) under general international law they are considered as breaches of the
international hum anitarian law of armed conflict; and, in addition, (ii) whether under
customary international law their commission amounts to a war crime.
Were the above explanation regarded as sound, it would follow that for two broad
categories of war crime the Statute does not set out a self-contained legal regime, but
presupposes a mandatory examination, by the Court, on a case-by-case basis, of the
current status of general international law. This method, while commendable in some
respects, may, however, entail that the Statute’s provisions eventually constitute only
a tentative and interim regulation of the matter, for the final say rests with the C ourt s
determination. W hether or not such a regulation is considered satisfactory, it seems
indisputable that it leaves greater freedom to sovereign states or, to put it differently,
makes the net of international prohibitions less tight and stringent.
Secondly, the legal regulation of means of warfare seems to be narrower than that
laid down in customary international law.
The use in international armed conflict of modern weapons which (a) cause
superfluous injury or unnecessary suffering; or (b) are inherently indiscriminate, is
not banned per se and therefore does not amount to a crime under the ICC Statute
whereas arguably such use constitutes a war crime under customary international
law, at least in those instances where the weapon at issue or the way it is used indis­
putably infringes those two principles or one of them .22 Thus, in the event the two
principles are deprived of their overarching legal value, at least with regard to indi­
viduals (the principles still act as standards applicable to states, with the consequence
that those states that breach them incur international responsibility). This seems all
the more questionable because even bacteriological weapons, which undoubtedly
are already prohibited by general international law, might be used without entailing
the commission of a crime falling under the jurisdiction of the Court (it would seem
that the use of this category of weapons is not covered by the ban on ‘asphyxiating,
poisonous or other gases and all analogous liquids, materials or devices, contained in
Article 8(2)(b)(xviii) and clearly relating to chemical weapons only).

22 th e ban will only take effect, and its possible breach amount to a crime, if an amendment to this end is
made to the Statute pursuant to Articles 121 and 123. In practice, as it is extremely unlikely that such amend­
ment will ever be agreed upon, those weapons may eventually be regarded as lawful.
 96 INTERNATIONAL CRIMINAL LAW

A similar criticism may be made of the sub-article on damage to the environment.

Under Article 8(2)(b)(iv)) ‘Intentionally launching an attack in the knowledge that
such attack will cause [...] widespread, long-term and severe damage to the natural
environment which would be clearly excessive in relation to the concrete and direct
overall military advantage anticipated’ constitutes a war crime. It should be noted that
Article 55(1) of Additional Protocol I—to which any provision on environmental war
crimes must accord a sort of precedential value—provides

Care shall be taken in warfare to protect the natural environment against widespread,
long-term and severe damage. This protection includes a prohibition on the use of methods
or means of warfare which are intended or may be expected to cause such damage to the
natural environment and thereby to prejudice the health or survival of the population.

Article 55 makes no mention of the ‘excessive’ or disproportionate character of the

attack nor o f ‘anticipated military advantage’ (let alone of the ‘direct overall military
advantage anticipated, a phrase that gives belligerents a very great latitude and renders
judicial scrutiny almost impossible). Moreover, in paragraph 2 it prohibits reprisals by
way ot attack against the natural environment. Article 8 of the ICC Statute there­
fore takes a huge leap backwards by allowing the defence that ‘widespread, long-term
and severe damage to the natural environment’ caused by the perpetrator—not just
damage, but widespread, long-term and severe damage, intentionally caused—was not
clearly excessive (perhaps it was excessive, but not ‘clearly excessive’) in relation to the
concrete and direct overall m ilitary advantage anticipated. This seems indefensible.
Thirdly, one may entertain some misgivings concerning the distinction, upheld
in Article 8, between the regulation of international armed conflict, on the one side,
and internal conflicts on the other. In so far as Article 8 separates the law applicable
to the former category of armed conflict from that applicable to the latter category,
it is somewhat retrograde, as the current trend has been to abolish the distinction
and to have simply one corpus of law applicable to all conflicts. It can be confusing—
and unjust to have one law for international armed conflict and another for internal
armed conflict.
More specific flaws may be discerned. For instance, when it comes to crimes in
internal armed conflicts perpetrated against adversaries hors de combat (combatants
who have laid down their weapons), the wounded, the sick, as well as civilians, the rele­
vant provision (Article 8(2)(c)) admits that such crimes may be committed in broad
categories of armed conflict (any armed conflict not of an international character’,
excluding situations of internal disturbances and tensions, such as riots, isolated and
sporadic acts of violence or other acts of a similar nature’). In contrast, the threshold
required by the provision for crimes committed in combat is higher: Article 8(2)(f))
stipulates that the relevant provisions only apply ‘to armed conflicts that take place in
the territory of a state when there is protracted armed conflict between governmental
authorities and organized groups or between such groups’ (emphasis added). It follows
that for a crime belonging to the second class to be perpetrated, an added requirement
is envisaged, namely that the internal armed clash be ‘protracted’. Allegedly the main
 WAR CRIMES 97

reason for this distinction is that in the first class, there already existed a set of provi­
sions laid down in Article 3 common to the four Geneva Conventions and that fur­
thermore these provisions are held to have turned into customary international law.
On the contrary, no previous treaty or customary rule existed regulating methods of
combat in internal armed conflict. While making progress in this area, the majority of
states gathered at the Rome Conference preferred to tread gingerly, so the explanation
goes, so as to take due account of states’ concerns. Assuming that this explanation is
correct, the fact remains that a dichotomy was created, which appears contrary to the
fundamental object and purpose of international hum anitarian law.
Furthermore, the prohibited use of weapons in internal armed conflicts is not
regarded as a war crime. This regulation does not reflect the current status of general
international law.23
The above ICC Statute restrictions on modern regulation of armed conflict are
compounded by two more factors: (i) allowance has been made for superior orders to
relieve subordinates of their responsibility for the execution of orders involving the
commission of war crimes (whereas under the ICC Statute for crimes against hum an­
ity or genocide superior orders a priori may not be pleaded); (ii) Article 124 allows
states to declare, upon becoming parties to the Statute, that the Court s jurisdiction
shall not become operative for a period of seven years with regard to war crimes (com­
m itted by their nationals or on their territory), whereas no similar allowance is made
for other categories of international crime.24
One is therefore left with the impression that the framers of the ICC Statute were
eager to shield their servicemen as much as possible from being brought to trial for
war crimes.
To summarize, a tentative appraisal of the provisions on war crimes of the ICC
Statute cannot but be chequered: in many respects the Statute marks a great advance
in ICL, in others it proves instead faulty; in particular, it is m arred by being too obse­
quious to state sovereignty.

23 As the AC of the ICTY stressed in Tadic (IA), in modern warfare it no longer makes sense to distin­
guish between international and internal armed conflicts: ‘Why protect civilians from belligerent violence,
or ban rape, torture or the wanton destruction of hospitals, churches, museums or private property, as well
as proscribe weapons causing unnecessary suffering when two sovereign States are engaged in war, and yet
refrain from enacting the same bans or providing the same protection when armed violence has erupted
“only” within the territory of a sovereign State?’ (§97, emphasis added).
The AC rightly answered this question by finding that the prohibition of weapons causing unnecessary suf­
fering, as well as the specific ban on chemical weapons, also applies to internal armed conflicts (§§119-24).
24 One should also note an odd provision, which applies to all the crimes envisaged in the Rome Statute.
While children may be conscripted or enlisted as from the age of 15 (Article 8(2)(b)(xxvi), and (e)(vu)), the
Court has no jurisdiction over persons under the age of 18 at the commission of the crime (Art. 26). Thus a
person between 15 and 17 is regarded as a lawful combatant and may commit a crime without being brought
to court and punished. A commander could therefore recruit minors into his army expressly for the purpose
of forming terrorist units whose members would be immune from prosecution. Moreover, in modern war­
fare, particularly in developing countries, young persons are more and more involved in armed hostilities
and thus increasingly in a position to commit war crimes and crimes against humanity.
 5
CRIMES AGAINST HUMANITY

5.1 THE NOTION

Under general international law the category of crimes against hum anity is sweeping
but sufficiently well defined. It covers actions that share a set of common features:

1 . They are particularly odious offences in that they constitute a serious attack on
hum an dignity or a grave humiliation or degradation of one or more persons.
2 . They are not isolated or sporadic events, but are part of a widespread or sys­
tematic practice of atrocities that either form part of a governmental policy or are
tolerated, condoned, or acquiesced in by a government or a de facto authority. Clearly,
it is required that a single crime be an instance of a repetition of similar crimes or be
part of a string of such crimes (widespread practice), or that it be the manifestation
of a policy or a plan of violence worked out, or inspired by, state authorities or by the
leading officials of a de facto state-like organization, or of an organized political group
(systematic practice). However, this contextual element does not necessarily mean that
the individual act amounting to crime against hum anity (murder, torture, rape, per­
secution, etc.) be repeated in time and space or, in other words that the same offence
be committed on a large scale. It may also be sufficient for the offence at issue (murder,
torture, persecution, etc.) to be part of a massive attack on the civilian population
(see, however, infra, 5.6), whatever the form taken by such large-scale violence. This
conclusion is warranted by the very rationale behind the prohibition and criminaliza­
tion of this category of heinous conduct (international rules intend to proscribe and
make punishable any offence against humanity, whatever its features, which is part of
massive despicable violence against hum an beings, for they consider that such attacks,
in whatever form, offend against humanity). It is also borne out by case law.11

1 See infra the German cases in denunciations, etc., mentioned in notes 3, 37 and 38.
The same conclusion is also indirectly corroborated by more recent case law relating to the elements from
which one can infer the existence of a policy. For instance, in Blaskic an ICTY TC, in addressing the issue of
the ‘systematic’ character of the crimes at issue, held that ‘The systematic character refers to four elements
which for the purposes of this case may be expressed as follows:—the existence of a political objective, a
plan pursuant to which the attack is perpetrated or an ideology, in the broad sense of the word, that is, to
destroy, persecute or weaken a community—the perpetration of a criminal act on a very large scale against a
group of civilians or the repeated and continuous commission of inhumane acts linked to one another—the
preparation and use of significant public or private resources, whether military or other—the implication
 CRIMES AGAINST HUMANITY 99

3. They are prohibited and may consequently be punished regardless of whether

they are perpetrated in time o f war or peace. While in 1945 a link with an armed con­
flict was required, at present customary law no longer attaches any importance to such
nexus. Thus, while in 1945 the ‘contextual element’ of the crime was the existence of an
armed conflict, at present such element resides in a ‘widespread or systematic’ attack
on the population.
4. The victims of the crime may be civilians or, where crimes are committed during
armed conflict, persons who do not take part (or no longer take part) in armed hos­
tilities, as well as, under customary international law (but not under the Statute of the
ICTY, ICTR, and the ICC), enemy combatants.
Before embarking upon an exposition of the history of the notion and the vari­
ous classes of crimes, it may be fitting to note that to a large extent many concepts
underlying this category of crimes derive from, or overlap with, those of hum an rights
law (the rights to life, not to be tortured, to liberty and security of the person, etc.),
laid down in provisions of international hum an rights instrum ents (e.g. the Universal
Declaration of Hum an Rights, the UN Covenant on Civil and Political Rights). Indeed,
while ICL concerning war crimes largely derives from, or is closely linked with, IHL,
ICL concerning crimes against hum anity is to a great extent predicated upon inter­
national hum an rights law. IHL (which traditionally regulates warfare between or
within states), and international human rights law (which regulates what states may
do to their own citizens and, more generally, to individuals under their control), are
in essence two distinct bodies of law, each arising from separate concerns and con­
siderations. The former is largely rooted in notions of reciprocity— one need not be a
great hum anist to be in favour of laws of war for international conflicts, as it is simple
self-interest for a state to ensure that its soldiers are treated well in exchange for treat­
ing enemy soldiers well and that its civilians are spared the horrors of war. The latter
is more geared to community concerns, as it intends to protect hum an beings per se
regardless of their national or other allegiance.
Let us now return to the large-scale or massive nature of the crimes. That this fea­
ture is a necessary ingredient of the crimes may be inferred from the first provisions
setting out a list of such offences. They clearly, if implicitly, required that the offence,

of high-level political and/or military authorities in the definition and establishment of the methodical
plan. This plan, however, need not necessarily be declared expressly or even stated clearly and precisely.
It may be surmised from the occurrence of a series of events, inter alia—the general historical circum­
stances and the overall political background against which the criminal acts are set—the establishment
and implementation of autonomous political structures at any level of authority in a given territory the
general content of a political programme, as it appears in the writings and speeches of its authors media
propaganda—the establishment and implementation of autonomous military structures—the mobilisa­
tion of armed forces—temporally and geographically repeated and co-ordinated military offensives links
between the military hierarchy and the political structure and its political programme—alterations to the
“ethnic” composition of populations—discriminatory measures, whether administrative or other (banking
restrictions, laissez-passer [...]—the scale of the acts of violence perpetrated in particular, murders and
other physical acts of violence, rape, arbitrary imprisonment, deportations and expulsions or the destruc­
tion of non-military property, in particular, sacral sites’ (§§203-4).
 100 INTERNATIONAL CRIMINAL LAW

to constitute an attack on humanity, be of extreme gravity and not be a sporadic event

but part of a pattern of misconduct. Subsequent case law has consistently borne out
that this is a major feature of the crimes.2
The link or connection with a systematic policy of a government or a de facto author­
ity was emphasized by the German Supreme Court in the British zone of occupation,
in the numerous and significant decisions on crimes against hum anity it delivered in
the years 1948-52. By way of illustration, one can mention /. and R. In i 950 the Court
of Assizes of Hamburg summed up the case law in Veit Harlan.3
However, when the atrocities are part of a government policy, the perpetrators need
not identify themselves with this policy, as the District Court of Tel-Aviv held in 1951
in Enigster (a case concerning a Jew imprisoned in a Nazi concentration camp, who
persecuted his fellow Jewish inmates). The Tel-Aviv Court rightly stated that
a person who was himself persecuted and confined in the same camp as his victims can,
from the legal point of view, be guilty of a crime against humanity if he performs inhumane
acts against his fellow prisoners. In contrast to a war criminal, the perpetrator of a crime
against humanity does not have to be a man who identified himself with the persecuting
regime or its evil intention (542).

In 1949, in Albrecht, the Dutch Special Court of Cassation delivered one of the first decisions on crimes
against humanity, after the Nuremberg Judgment of the International Military Tribunal. The defendant,
a German Sturmscharführer (commander of a storm company) of the Waffen SS (German state Security
Police), had been accused of killing a Dutch national and ill-treating five others. The Court was called upon
to decide if the offences perpetrated by Albrecht were to be regarded as war crimes or as crimes against
humanity. It opted for the first category, adding that they could not also be classified as crimes against
humanity. Addressing this last class of crimes the Court stated that: ‘[C]rimes of this category are charac­
terised either by their seriousness and their savagery, or by their magnitude, or by the circumstance that
they were part of a system designed to spread terror, or that they were a link in a deliberately pursued policy
against certain groups of the population’ (at 750). A judgment of the Dutch Court of Cassation in 1981 sub­
stantially supported this view (see Menten at 362-3).
In /. and R., a trial court had sentenced for crimes against humanity a German who had denounced
to the police two other Germans for listening to a foreign radio, which amounted under German law to
national treason; as a consequence the two persons had been arrested and sentenced to imprisonment; they
had died as a result of harsh prison conditions. The Supreme Court overruled the acquittal pronounced by
the trial court and the Appeals Court. It pointed out, among other things, that the aggressive behaviour of
the agent and the inhuman injury to the victim had to be objectively connected with the Nazi system of vio­
lence and tyranny. This connection does not need [...] to lie in support for the tyranny, but may, for example,
also consist of the use of the system of violence and tyranny. [Furthermore], the agent need not act systemat­
ically; it is sufficient that his single action be connected with the system and thereby lose the character of an
isolated occurrence. The Court went on to explain that the denunciation by the accused was closely linked
with the arbitrary and violent Nazi system, that there existed no freedom, and the state suppressed any devi­
ant behaviour by violence and harsh punishment. The denunciation at issue had been intended to achieve
the handing over of two persons to an arbitrary police system based on terror: hence ‘he who caused such a
consequence through his denunciation, objectively committed a crime against humanity’ (167-71).
Vett Harlan dealt with a charge of complicity in a crime against humanity. (The accused, a film director,
had contributed to the persecution of Jews by his film Jud Süss, produced in 1940.) The Court of Assizes,’
basing itself on numerous judicial precedents on the matter, gave the following definition of crimes against
humanity: ‘One must regard as a crime against humanity any conscious and willed attack that, in connec­
tion with the Nazi system of violence and arbitrariness, harmfully interferes with the life and existence of a
person or his relationships with his social sphere, or interferes with his assets and values, thereby offending
against his human dignity as well as humanity as such’ (52).
 CRIMES AGAINST HUMANITY 101

In summary, murder, extermination, torture, rape, political, racial, or religious per­

secution and other inhumane acts reach the threshold of crimes against hum anity
only if they are part of a practice.4Isolated inhumane acts of this nature may constitute
grave infringements of hum an rights or, depending on the circumstances, war crimes,
but fall short of the stigma attaching to crimes against humanity. On the other hand,
an individual may be guilty of crimes against humanity even if he perpetrates one or
two of the offences mentioned above, or engages in one such offence against only a few
civilians, provided those offences are part of a consistent pattern of misbehaviour by a
number of persons linked to that offender (for example, because they engage in armed
action on the same side, or because they are parties to a common plan, or for any othei
similar reason).

5.2 THE ORIGIN OF THE NOTION

The notion of crimes against hum anity was propounded for the first time in 1915, on
the occasion of mass killings of Armenians in the Ottom an Empire. On 28 May 1915
the French, British, and Russian Governments decided to react strongly. They there­
fore jointly issued a declaration stating that
In view of these new crimes of Turkey against humanity and civilisation, the Allied govern­
ments announce publicly to the Sublime Porte that they will hold personally responsible [for]
these crimes all members of the Ottoman Government and those of their agents who are
implicated in such massacres.5
The expression ‘crimes against hum anity was not in the original proposal em an­
ating from the Russian Foreign Minister, Sazonov. He had suggested instead a protest

4 In Limaj and others an ICTY TC held that, as a rule, the widespread or systematic attack required for
crimes against humanity occurs at the behest of a state: ‘Due to structural factors and organisational and mili­
tary capabilities, an “attack directed against a civilian population” will most often be found to have occurred
at the behest of a State. Being the locus of organised authority within a given territory, able to mobilise and
direct military and civilian power, a sovereign State by its very nature possesses the attributes that permit it to
organise and deliver an attack against a civilian population; it is States which can most easily and efficiently
marshal the resources to launch an attack against a civilian population on a “widespread” scale, or upon a
“systematic” basis. In contrast, the factual situation before the Chamber involves the allegation of an attack
against a civilian population perpetrated by a non-state actor with extremely limited resources, personnel and
organisation’(§191). This statement is acceptable to the extent that it is intended merely to reflect what happens
in practice, not as the formulation of a legal requirement (plainly, a widespread or systematic attack on the
population can be carried out by non-state groups or paramilitary units with the acquiescence of state author­
ities or in circumstances where such authorities lack the effective power to put an end to such attacks).
For insistence on the notion that the context of a ‘widespread or systematic’ attack is a fundamental
requirement for crimes against humanity, see a string of recent Indonesian cases concerning East Timor:
Abilio Soares (at 98-9), Herman Sedyono and others (at 66-8), Endar Priyanto (at 32.3), Eurico Guterres (at
27-8), Asep Kuswani ( at 45), Letkol Inf. Soedjarwo (at 22-3), Yayat Sudrajat (at 6-7).
5 Emphasis added. For the full text of the note, see the dispatch of the US Ambassador in France, Sharp,
to the US Secretary of State, Bryan, of 28 May 1915, in Papers Relating to the Foreign Relations of the United
States, 1915, Supplement (Washington: US Government Printing Office, 1928), at 981.
 102 INTERNATIONAL CRIMINAL LAW

against crimes against Christianity and civilisation’. However, the French Foreign
Minister Delcasse took issue with the reference to crimes against Christianity. He
feared that the Muslim populations under French and British colonial domination
might take umbrage at that expression, because it excluded them; consequently, they
might feel discriminated against. Hence, he proposed, instead of ‘crimes against
C hristianity, crimes against hum anity’. This proposal was accepted by the Russian
and British Foreign Ministers, and passed into the joint Declaration.6 It would seem
that the three states were neither aware of, nor interested in, the general philosophical
implications of the phrase they had used. Indeed, they did not ask themselves, nor did
they try to establish in practice, whether by ‘hum anity’ they meant ‘all hum an beings’
or rather the feelings of hum anity shared by men and women of modern nations’ or
even the concept of hum anity propounded by ancient and modern philosophy’. It is
probable that, although they used strong language criminalizing the perpetrators of
the massacre, in fact they were only intent on solving a short-term political problem,
as is shown by the lack of any practical follow-up to their joint protest.7
In any event, various initiatives to act diplomatically on behalf of hum anity subse­
quently failed.8
Similarly, the special Commission set up after the First World War proposed in its
report to the Versailles Conference that an international criminal tribunal be created
and that its jurisdiction extend to ‘offences against the laws of hum anity’.9However, the

6 See the Russian dispatch of II May 1915, published in A. Beylerian, Les Grandes Puissances, VEmpire
Ottoman et les Armeniens dans les archives franfaises (1914-1918)—Recueil de documents (Paris, 1983), at
23 (doc. no. 29). The Russian draft referred to ‘crimes against Christianity and civilisation’ (‘crimes de la
Turquie contre la chretiente et la civilisation). The French Foreign Minister, Delcasse, changed the expres­
sion to crimes against humanity (‘crimes contre Vhumanite), in addition to making another, minor change
(ibid., at 23, footnotes with an asterisk).
The political reasons for this change, in particular for dropping any reference to Christianity, were set out
by the French Ministry in a Note of 20 May 1915 to the British Embassy (ibid., at 26, doc. 34: ‘L’interet qu’il
y a a menager le sentiment des populations musulmanes qui vivent sous la souverainete de la France et de
1Angleterre fera sans doute estimer au gouvernement britannique comme au gouvernement fran^ais qu’il
convient de s’abstenir de specifier que 1’interet des deux puissances parait ne se porter que du cote des ele­
ments Chretiens’.). The two French suggestions were eventually accepted by Great Britain and Russia and the
text of the note was changed accordingly.
7 On 11 August 1915, during the massacre of Armenians, the American Ambassador to Turkey,
Morgenthau, had proposed to the US Secretary of State, Robert Lansing, among other things, that ‘The
United States Government on behalf of humanity urgently request the Turkish Government to cease at
once the present campaign and to permit the survivors to return to their homes if not in the war zones, or
else to receive proper treatment.’ However, the Secretary of State did not adopt this suggestion, contenting
himself merely with asking whether the protest of the German Ambassador to the Turkish Government had
improved conditions. See Papers Relating to the Foreign Relations of the United States, op. cit., at 986.
8 The Peace Treaty of Sevres of 10 August 1920 provided in Article 230 that the ‘Ottoman Government’
undertook to hand over to the Allies the persons requested by these Powers as responsible for the massacres
perpetrated, during the war, on territories which constituted part of the Ottoman Empire; the Allies reserved
the right to ‘designate’ the tribunal which would try those persons. However, the Treaty was never ratified,
and its replacement, the Peace Treaty of Lausanne, of 24 July 1923, provided in an annexed Declaration for
an amnesty for crimes committed between 1914 and 1922.
9 See Report presented to the preliminary Peace Conference by the Commission on the Responsibility
of the Authors of the War and on the Enforcement of Penalties’, in Carnegie Endowment for International
 CRIMES AGAINST HUMANITY 103

‘Memorandum of Reservations’ submitted by the two distinguished representatives

of the United States, Robert Lansing and James Brown Scott, paralysed any action by
the Conference. They emphasized that while war crimes should be punished because
‘the laws and customs of war are a standard certain (at 64), the laws and principles
of hum anity are not certain, varying with time, place and circumstance, and accord­
ing, it may be, to the conscience of the individual judge. There is no fixed and univer­
sal standard of hum anity’ (at 73). This, the US delegates said, ‘if for no other reason,
should exclude them from consideration in a court of justice, especially one charged
with the administration of criminal law’ (at 64). As a result of the American oppos­
ition, no provision was made for crimes against humanity.
It is notable that in 1919 a few Extraordinary Courts Martial were established in
the Ottom an Empire to try the presumed authors of the 1915-16 deportation, m as­
sacres and looting of Armenians. According to a distinguished author, at least 28
such Court Martial trials were held. Judging from the verdicts that are available,
those courts tried in 1919-20 officials of the Ottom an Empire under the Ottom an
Criminal Code,11 and found many of them responsible for massacres, deportation,
and looting,12 or massacres ‘for the purpose of destroying and annihilating (ifna ve
im ha’si emrinde) Armenians.13
During the Second World War, the Allies became aware that some of the most hein­
ous acts of barbarity perpetrated by the Germans were not prohibited by traditional
international law. The laws of warfare only proscribed violations involving the adver­
sary or the enemy populations, whereas the Germans had also performed inhum an
acts for political or racial reasons against their own citizens (Jews, trade union mem ­
bers, social democrats, communists, gypsies, members of the church), as well as other
persons not covered by the laws of warfare.14 In addition, in 1945 persecution for pol­
itical or racial purposes was not prohibited, even if perpetrated against civilians of
occupied territories.

Peace, Division of International Law, Pamphlet No. 32, Violations of the Laws and Customs of War, Report of
Majority and Dissenting Reports of American and Japanese Members of the Commission of Responsibilities,
Conference of Paris 1919 (Oxford: Clarendon Press, 1919), at 25-6.
10 T. Akcam, Armenien und der Völkermord: Die Istanbuler Prozesse und die Türkische Nationalbewegung
(Hamburg: Hamburger Edition, 1996), at 162-5.
11 They applied in particular Articles 102 (negligence in the execution of ones duties or failure to carry
out a superior order), 130 (undue interference with civilian or military officials), 170 (murder) and 171
(premeditated coercion to destroy or rob supplies or goods) or 172 (abuse of an official position), together
with Articles 45 (aiding and abetting), and 55 (co-perpetration).
12 See in particular Tal&t Pa?a and others, at 106-16; Kemäl Bey and others, at 155-8 (or 171-5); Kerim
Bey and others, at 166-8.
13 See, in particular, Ahmed Mithad Bey and others, at 147-53; Mehmed ‘A ll Bey and others (at 159-65 or
17 7 - 84 ); Bahäeddin §äkir and others, at 169-73. See also Talät Pa$a and others (at 106-16).
14 For instance, citizens of the Allies (e.g. French Jews under the Vichy regime (1940-4)); nationals
of states not formally under German occupation and, therefore, not protected by the international rules
safeguarding the civilian population of occupied territories: this applied to Austria, annexed by Germany
in 1938, and Czechoslovakia (following the Munich Treaty in 1938, the Sudeten territory was annexed by
Germany, and the rest of the country became the so-called Protectorate of Bohemia and Moravia, in 1939).
The Germans also harassed and murdered stateless Jews and gypsies.
 104 INTERNATIONAL CRIMINAL LAW

In 1945, at the strong insistence of the USA, the Allies thus decided that a better
course of action than simply to execute all the major war criminals would be to bring
them to trial. The London Agreement embodying the Charter of the International
M ilitary Tribunal (IMT) included a provision under which the Tribunal was to try
and punish persons guilty, among other things, o f‘crimes against hum anity’ (the use
of this specific term was suggested by a leading scholar, Hersch Lauterpacht, to Robert
Jackson, the US delegate to the London Conference, who was subsequently appointed
chief US prosecutor at Nuremberg).15 These crimes were defined as
murder, extermination, enslavement, deportation, and other inhumane acts committed
against any civilian population, before or during the war, or persecutions on political, racial,
or religious grounds in execution of or connexion with any crimes within the jurisdiction of
the Tribunal [i.e. either crimes against peace’ or ‘war crimes’], whether or not in violation of
the domestic law of the country where perpetrated.

One major shortcoming of this definition is that it closely linked crimes against
hum anity to the other two categories of offences. Article 6(c) indeed required, for
crimes against hum anity to come under the jurisdiction of the IMT, that they be per­
petrated in execution of or in connection w ith’ war crimes or crimes against peace.
This link was not spelled out, but it was clear that it was only within the context of
a war or of the unleashing of aggression that these crimes could be prosecuted and
punished. As rightly pointed out by Schwelb,16 this association meant that only those
criminal activities were punished which ‘directly affected the interests of other States’
(either because these activities were connected with a war of aggression or a conspir­
acy to wage such a war, or because they were bound up with war crimes, that is crimes
against enemy combatants or enemy civilians). Plainly, in 1945 the Allies did not feel
that they should ‘legislate’ in such a way as to prohibit inhum an acts regardless of
their consequences or implications fo r third states. At that stage, what happened within
a national system, even if contrary to fundamental values of humanity, was still of
exclusive concern to that state if it had no spill-over effects on other states: it fell within
its own ‘domestic jurisdiction’.
Despite this limitation, the creation of the new category marked a great advance.
First, it indicated that the international community was widening the category of acts
considered o f ‘meta-national’ concern. This category came to include all actions ru n ­
ning contrary to those basic values that are, or should be, considered inherent in any
hum an being (in the notion, hum anity did not mean ‘m ankind’ or ‘hum an race’ but
‘the quality’ of being humane).
Secondly, inasmuch as crimes against hum anity were made punishable even if per­
petrated in accordance with domestic laws, the 1945 Charter showed that in some
special circumstances there were limits to the ‘omnipotence of the State’ (to quote
the British Chief Prosecutor, Sir Hartley Shawcross) and that ‘the individual h u m an

15 See on this point M. Koskenniemi, ‘Hersch Lauterpacht and the Development of International
Criminal Law’, 2 JICJ (2004), at 811.
16 E. Schwelb, ‘Crimes Against Humanity’, 23 BYIL (1946), at 207.
 CRIMES AGAINST HUMANITY 10 5

being, the ultimate unit of all law, is not disentitled to the protection of m ankind
when the State tramples upon his rights in a m anner which outrages the conscience
of m ankind’.17
A num ber of courts have explicitly or implicitly held by that Article 6(c) of the
London Agreement simply crystallized or codified a nascent rule of general inter­
national law prohibiting crimes against humanity. It seems more correct to contend
that that provision constituted new law. This explains both the limitations to which the
new notion was subjected (and to which reference has already been made above) and
the extreme caution and indeed reticence of the IMT in applying the notion.
The reticence and what could be viewed as the embarrassment of the IMT on the
matter are striking. Six points, in particular, should be stressed.
First, the IMT tackled the issue of ex post facto law only with regard to crimes against
peace (in particular, aggression), whereas it did not pronounce at all upon the no less
delicate question of whether or not crimes against hum anity constituted a new cat­
egory of offence. (However, the reason for this omission may also be found in the fact
that the German defence counsel, in the joint motion of 19 November 1945 by which
they complained about the retroactive application of criminal law by the IMT,18 only
referred to crimes against peace; this probably occurred because they felt that such
offences as murder, extermination, or persecution constituted breaches of the law in
most countries of the world and in any case had been committed by Nazi authorities
on a very large scale.)
Secondly, when dealing with ex post facto law, the IMT was rather reticent and
indeed vague, as is apparent from, inter alia, the glaring discrepancy between the
English and the French text of the judgment,19 both authoritative.
Thirdly, the IMT held that no evidence had been produced to the effect that crimes
against hum anity had been committed before the war, in execution of or in connection
with German aggression.20 The IMT thus markedly narrowed the scope, in casu, of
the category of crimes against humanity, although it asserted that it did so on grounds
linked to the evidence produced.

17 Sir Hartley Shawcross, in Speeches of the ChiefProsecutors at the Close of the Case Against the Individual
Defendants (London: HM Stationery Office, Cmd. 6964,1946), at 63.
18 See Trial of the Major War Criminals before the International Military Tribunal, vol. I, at 168-70.
19 In the English text, the IMT stated that ‘the maxim nullum crimen sine lege is not a limitation of sov­
ereignty, but is in general a principle of justice’ (at 219; emphasis added), while in the French text it is stated
that ‘Nullum crimen sine lege ne limite pas la souverainete des Etats; elle ne formule qu’une regie generale-
ment suivie (at 231; emphasis added). Furthermore, the phrase in the English text, ‘On this view of the case
alone, it would appear that the maxim has no application to the present facts’ (at 219) does not appear in the
French text.
20 The tribunal stated that: ‘To constitute crimes against humanity, the acts relied on before the out-
break of war must have been in execution of, or in connection with, any crime within the jurisdiction of
the Tribunal. The Tribunal is of the opinion that revolting and horrible as many of these crimes were, it has
not been satisfactorily proved that they were done in execution of, or in connection with, any such crime.
The Tribunal therefore cannot make a general declaration that the acts before 1939 were crimes against
humanity within the meaning of the Charter’ (at 254).
 10 6 INTERNATIONAL CRIMINAL LAW

Fourthly, probably aware of the novelty of that class of crimes and hence of the
possible objection that the nullum crimen principle (see above, 2 .3) was being breached
by applying criminal law retroactively, the IMT tended to find that some defendants
accused of various classes of crime were guilty both of war crimes and of crimes
against hum anity (this was the case with 14 defendants): in other words, the Tribunal
avoided clearly identifying the distinction between the two classes, preferring instead
to find that in many cases the defendant was answerable for both.
Fifthly, in the only two cases where the IMT found a defendant guilty exclusively
of crimes against hum anity (Streicher and von Schirach), the Tribunal did not specify
the nature, content, and scope of the link between crimes against hum anity and war
crimes (in the case of Streicher) or crimes against humanity and aggression (in the
case of von Schirach); rather, the Tribunal confined itself to a generic reference to the
connection between the classes of crimes, without any further elaboration.
Finally, it is striking that in the part of the judgment referring to Streicher, the
English text is markedly different from the French.21
In summary, in all probability the IMT applied new law, or substantially new law,
when it found some defendants guilty of crimes against hum anity alone or of these
crimes in conjunction with others. However, this was not in breach of a general norm
strictly prohibiting retroactive criminal law. As noted above (2.3), immediately after
the Second World War, the nullum crimen sine lege principle could be regarded as a
moral maxim destined to yield to superior exigencies whenever it would have been
contrary to justice not to hold persons accountable for appalling atrocities. The strict

In the English text it is stated that Streicher s incitement to murder and extermination at the time
when Jews in the East were being killed under the most horrible conditions clearly constitutes persecution
on political and racial grounds in connection with War Crimes, as defined in the Charter, and constitutes
a crime against humanity’ (at 304). By contrast, in the French text it is stated that Streicher’s persecution of
Jews was itself a war crime as well as a crime against humanity (‘Le fait que Streicher poussait au meurtre
et a 1’extermination, a l’epoque meme oil, dans l’Est, les Juifs etaient massacres dans les conditions les plus
horribles, realise la persecution pour des motifs politiques et raciaux” prevue parmi les crimes de guerre
definis par le Statut, et constitue egalement un crime contre l’Humanite’ (at 324). Clearly, this wording
reflects the position of the French Chief Prosecutor, Francois de Menthon (see his opening statement, of
17 January 1946, in IMT, vol. 5, at 371. The French Prosecutor stated that ‘This horrible accumulation and
maze of Crimes against Humanity both include and go beyond the two more precise juridical notions of
Crimes against Peace and War Crimes. But I think—and I will revert later separately to Crimes against Peace
and War Crimes—that this body of Crimes against Humanity constitutes, in the last analysis, nothing less
than the perpetration for political ends and in a systematic manner, of common law crimes such as theft,
looting, ill treatment, enslavement, murders, and assassinations, crimes that are provided for and punish­
able under the penal laws of all civilized states. No general objection of a juridical nature, therefore, appears
to hamper your task ofjustice. Moreover, the Nazis accused would have no ground to argue on alleged lack of
written texts to justify the penal qualification that you will apply to their crimes.’). The wording at issue also
reflected the reservations and misgivings of the French Judge, H. Donnedieu de Vabres, who in 1947 set forth
his views in scholarly papers in which he argued that crimes against humanity simultaneously constituted
war climes and hence, the Tribunal did not breach the nullum crimen, nulla poena sine lege principle (see
H. Donnedieu de Vabres, ‘Le Jugement de Nuremberg et le principe de legalite des debts et des peines’, in 27
Revue de droit penal et de criminologie (1946-47), 826-7; see also his Hague Academy lectures: ‘Le proces de
Nuremberg devant les principes modernes du droit penal international’, HR (1947-1), 525-7 (in particular
n. 1 at 526)).
 CRIMES AGAINST HUMANITY 10 7

legal prohibition of ex post facto law had not yet found expression in international law;
nor did it constitute a general principle of law universally accepted by all states. The
IMT set out the view that ‘the maxim nullum crimen sine lege [...] is in general a prin­
ciple of justice’ allowing the punishm ent of actions not proscribed by law at the time
of their commission, when it would be unjust’ for such wrongs to be allowed to go
unpunished’ (at 219).22
In the wake of the major war trials, momentous changes in international law took
place. On 11 December 1946 the UN GA unanimously adopted a resolution affirm­
ing’ the principles of the Charter of the Nuremberg International Tribunal and its
judgment. On 13 February 1946 it passed resolution 3(1) recommending the extradi­
tion and punishment of persons accused of the crimes provided for in the Nuremberg
Charter. These resolutions show that the category of crimes against hum anity was in
the process of becoming part of customary international law.23

22 However, as pointed out above, the IMT expressed this view only with regard to aggressive war; in
addition it hastened to add (at 219-23) that in any event, under international law, such wars were already
regarded as criminal before the outbreak of the Second World War.
Interestingly, the first of the two propositions referred to in the text was repeatedly set forth, with specific
regard to crimes against humanity, by the German Supreme Court in the British Occupied Zone. According
to this Court, ‘[retroactive punishment is unjust when the action, at the time of its commission, falls foul
not only of a positive rule of criminal law, but also of the moral law. This is not the case for crimes against
humanity. In the view of any morally-oriented person, serious injustice was perpetrated, the punishment
of which would have been a legal obligation of the state. The subsequent cure of such dereliction of a duty
through retroactive punishment is in keeping with justice. This also does not entail any violation of legal
security but rather the re-establishment of its basis and presuppositions’ (case against Bl, at 5).
See also the following judgments: B. and A. case, at 297; H. case (18 October 1949), at 232-3; N. case, at 335,
H. case (11 September 1950), at III, 135.
Other judgments include elaborate reasoning concerning the distinction to be drawn between law
enacted by the Occupying Powers and German law: see, for example, G. case, at 362-4; M. et al. case, at
378-81 (this judgment sets out important reasons in support of the view that crimes against humanity could
be punished retroactively: see 380-1).
23 Strikingly, the French Court of Cassation, in Sobanski Wladyslav (also called the Boudarel case), in
1993 placed a patently flawed interpretation on the second resolution and the Charter of the IMT, to which
the resolution referred. It held that the resolution and Article 6 of the Tribunal’s Statute only related to
offences perpetrated on behalf of the Axis European States’, hence it could not apply to atrocities commit­
ted elsewhere. The specific question brought to the Court revolved around the scope of the French law of
26 December 1964. (Under this law, crimes against humanity by their nature are not covered by any statute
of limitation; the law stated that such crimes were those referred to in the UN resolution of 13 February 1946
which in turn adverted to the definition set out in the Statute of the Tribunal.) In the case at bar the ques­
tion was whether such law applied to the accused Boudarel, a French serviceman who, after deserting the
French army, had sided with the Viet Minh and allegedly committed atrocities against French prisoners of
war in 1952-4. By interpreting the GA resolution and the IMT Statute as recalled above, and consequently
by also restrictively construing the French law of 1964, the Court concluded that the law did not apply to the
accused, who consequently could not be tried. According to the Court, his alleged crimes were covered by a
law of 1966 granting amnesty for all crimes committed in Indochina before 1 October 1957 (at 354-5).
To refute the legal grounds set forth in the judgment, it may suffice to quote the sort o f‘authentic inter­
pretation’ of Article 6 of the IMT Statute, propounded by Robert H. Jackson, the protagonist of the London
Conference that led to the adoption, on 8 August 1945, of that Statute. After the Conference he wrote that
‘The most serious disagreement [at the Conference], and one on which the United States declined to recede
from its position even if it meant the failure of the Conference, concerned the definition of crimes. The Soviet
Delegation proposed and until the last meeting pressed a definition which, in our view, had the effect of
 io8 INTERNATIONAL CRIMINAL LAW

In addition to the Charter of the Tokyo International Tribunal, a number of inter­

national instruments were then drawn up embodying the prohibition of crimes against
humanity, some of which improved and expanded the provisions of the London
Agreement, for instance, the Peace Treaties with Italy, Romania, Hungary, Bulgaria, and
Finland, each of which included terms providing for the punishment of these crimes.24
In particular, after 1945 the link between crimes against hum anity and war was grad­
ually dropped. This is evidenced by Article II(l)(c) of such ‘multinational’ legislation as
Control Council Law no. 10 passed by the four victorious Powers four months after the
London Agreement; that is on 20 December 1945, by national legislation (such as the
Canadian25 and the French26criminal codes), case law,27 as well as international treaties
such as the 1948 Genocide Convention, the 1968 Convention on the Non-Applicability
of Statutory Limitations to War Crimes and Crimes against Humanity, and the 1973
Convention on Apartheid. This evolution gradually led to the abandonment of the
nexus between crimes against humanity and war: at present, as stated above, custom­
ary international law bans crimes against humanity whether they are committed in
time of war or peace.28 The same holds true for the ICC Statute, which confirms the
rupture of the link between these crimes and armed conflict.
On the other hand, some treaties and other binding international instruments
enshrining the Statutes of international courts and tribunals restrict the scope of cus­
tomary rules. To be more accurate (because strictly speaking those Statutes do not lay
down substantive rules of criminal law but only provide for the definition of those crimes
over which each relevant court or tribunal is endowed with jurisdiction), such treaties
and other instruments may indirectly contribute to the restriction of the customary rules.
Thus, the Statutes of the ICTY (1993), the ICTR (1994), and the ICC (1998) provide that

declaring certain acts crimes only when committed by the Nazis. The United States contended that the crim­
inal character of such acts could not depend on who committed them and that international crimes could
only be defined in broad terms applicable to statesmen of any nation guilty of the proscribed conduct. At
the final meeting the Soviet qualifications were dropped and agreement was reached on a generic definition
acceptable to all’ (International Conference on Military Trials, at vii-viii).
24 See, for instance, Article 45 of the Peace Treaty with Italy, Article 6 of the Treaty with Romania, and
Article 5 of that with Bulgaria.
Para. 7 (3.76) of the Canadian Criminal Code provides that:c“[CJrimes against humanity” means mur­
der, extermination, enslavement, deportation, persecution or any other inhumane act or omission that is
committed against any civilian population or any identifiable group of persons, whether or not it constitutes
a contravention of the law in force at the time and in the place of its commission, and that, at that time and in
that place, constitutes a contravention of customary international law or conventional international law or is
criminal according to the general principles of law recognized by the community of nations.’
26 Article 212-1, para. 1 of the French Criminal Code (enacted by Law no. 92-1336 of 16 December 1992,
modified by Law no. 93-913 of 19 July 1993), which entered into force on 1 March 1994, provides that: ‘La
deportation, la reduction en esclavage ou la pratique massive et systematique d ’executions sommaires,
d ’enlevements de personnes suivis de leur disparition, de la torture ou d ’actes inhumains, inspires par
des motifs politiques, philosophiques, raciaux ou religieux et organises en execution d u n plan concerte a
1encontre d un groupe de population civile sont punies de la reclusion criminelle ä perpetuite.’
27 See e.g. the Einsatzgruppen case, at 49; Altstötter and others (Justice case), at 974. See, however, the Flick
case, at 1213 and the Weizaecker case, at 112.
28 See on this point the dictum of the ICTY AC in its Decision of 2 October 1995 in Tadic (IA), §141.
 CRIMES AGAINST HUMANITY 10 9

the crimes at issue can only be committed against civilians, whereas in some respects
customary law upholds a broader notion of victims of such crimes (see infra, 5.6).

5.3 THE OBJECTIVE ELEMENTS

The conduct prohibited was loosely described in the London Agreement of 1945, and
similarly in Control Council Law no. 10 and the Charter of the Tokyo International
Tribunal, as well the ICTY and the ICTR. Gradually case law has contributed to defin­
ing the legal contours of the actus reus. In the event, the various categories have been
largely spelled out in the ICC Statute, Article 7, of which may be held to a large extent
either to crystallize nascent notions or to codify the bulk of existing customary law
(see infra, 5.7).
At present, ICL always requires for the crimes under discussion a general context
of criminal conduct, consisting of a widespread or systematic practice of unlawful
attacks against the population (see supra, 5.1, at 2; see, however, also the qualifications
set out infra in 5.6).
If such context does exist, the following classes of offence constitute crimes against
humanity:
1 . Murder. As a rule, the mental element of this conduct is the intent to bring about
the death of another person; intentional killing may or may not be premeditated, that
is planned and willed in advance of the act of killing (with the mental status persisting
over time between the first moment when the intention took shape and the later phys­
ical act of killing). However, for murder as a crime against hum anity a lesser mental
element is required by case law: it is sufficient for the perpetrator ‘to cause the victim
serious injury with reckless disregard for hum an life .2:
2. Extermination; that is mass or large-scale killing, as well as ‘the intentional
infliction of conditions of life, inter alia the deprivation of access to food and medi­
cine, calculated to bring about the destruction of part of a population’ (Article 7(2)(b)
of the ICC Statute).
The ICTR has defined the notion of extermination in a few cases.29
30 A Chamber of
the ICTY offered a better definition in Krstic. It held that:
for the crime of extermination to be established, in addition to the general requirements for
a crime against humanity, there must be evidence that a particular population was targeted

29 AkayesuM,§§589-90; Rutaganda, TJ, §80; Kupreskic, TJ, §561; Musema, TJ, §215.
30 Akayesu (§§591-2), Kambanda (§§141-7), Kayishema and Ruzindana (§§141-7), Rutaganda (§§82-4),
Musema (§§217-19). The ICTR has held that the requisite elements of the offence are as follows: (i) the accused
or his subordinate participated in the killing of certain named or described persons; (ii) the act or omission
was unlawful and intentional; (iii) the unlawful act or omission must be part of a widespread or systematic
attack; and (iv) the attack must be against the civilian population. This definition does not seem to be satis­
factory, for it is loose and does not indicate the unique objective features of the crime.
 110 INTERNATIONAL CRIMINAL LAW

and th at its members were killed or otherwise subjected to conditions of life calculated to
bring about the destruction o f a num erically significant p a rt of the population (§503).
The TC also specified that ‘In accordance with the Tadic (AJ),’ [...] it is unnecessary
that the victims were discriminated against for political, social or religious grounds’
(§499).31
It is submitted that one ought not to exclude from this class of crimes extermination
carried out by groups of terrorists for the purpose o f spreading terror. (Of course, the
necessary condition that the terrorist attack exterminating a group of persons be part
of a widespread or systematic attack, must be fulfilled.) See also infra, 8.6.
3. Enslavement. This notion was gradually elaborated upon by case law, notably by
two US Military Tribunals sitting at Nuremberg, in Milch (at 773-91) and in Pohl and
others (at 970), and then refined by a TC of the ICTY in Kunarac and others (§§515-43).
According to the ICC Statute, which crystallizes a nascent notion, enslavement ‘means
the exercise of any or all of the powers attaching to the right of ownership over a per­
son and includes the exercise of such power in the course of trafficking in persons, in
particular women and children’ (Article 7(2)(c)). The ICTY TC in Kunarac and others
convincingly propounded a set of elements that clarify this definition (§§542-3). In
addition, the TC set out clearly the reasons for which it found two of the defendants
guilty of enslavement (§§728-82).
4. Deportation orforcible transfer of population; that is, the ‘forced displacement of the
persons concerned by expulsion or other coercive acts from the area in which they are
lawfully present, without grounds permitted under international law’ (Article 7(2) (d)).
An ICTY TC emphasized in Krstic that:
Both deportation and forcible transfer relate to involuntary and unlaw ful evacuation of
individuals from the territory in which they reside. Yet the two are not synonym ous in
custom ary international law. D eportation presum es transfer beyond State borders, whereas
forcible transfer relates to displacem ent w ithin a State (§521).

In that case the TC found that, on 12-13 July 1995, about 25,000 Bosnian Muslim
civilians were forcibly bussed outside the enclave of Srebrenica to the territory under
Bosnian Muslim control, always within the same state (Bosnia and Herzegovina). The
transfer was compulsory and was carried out ‘in furtherance of a well organised pol­
icy whose purpose was to expel the Bosnian Muslim population from the enclave’. The
Chamber concluded that the civilians transported from Srebrenica were not subjected
to deportation but to forcible transfer, a crime against hum anity (§§527-32).
5. Imprisonment or other severe deprivation of physical liberty in violation of
fundamental rules of international law.
An ICTY TC, in Kordic and Cerkez, was the first international court to offer a
definition of this offence. It held that imprisonment as a crime against hum anity must
‘be understood as arbitrary imprisonment, that is to say, the deprivation of liberty of

31 In the same case the TC found that the accused was guilty of extermination (§§504-5).
 CRIMES AGAINST HUMANITY 111

the individual without due process of law, as part of a widespread or systematic attack
directed against a civilian population’ (§§302-3).
6. Torture; that is ‘the intentional infliction of severe pain or suffering, whether
physical or mental, upon a person in the custody or under the control of the accused’,
except when pain or suffering is inherent in or incidental to lawful sanctions (Article
7(2)(e) of the ICC Statute).
In Delalic and others an ICTY TC noted that the definition of torture contained
in the 1984 Torture Convention was broader than, and included, that laid down in
the 1975 Declaration of the United Nations General Assembly and in the 1985 Inter-
American Convention, and considered it to reflect a consensus which the TC regarded
as ‘representative of customary international law’ (§459). Another TC of the ICTY,
ruling in Furundzija, shared that conclusion, although on different legal grounds. It
held that, as shown by the broad convergence of international instrum ents and inter­
national jurisprudence, there was general acceptance of the main elements contained
in the definition set out in Article 1 of the Torture Convention. It considered, however,
that some specific elements pertained to torture as considered from the specific view­
point of ICL relating to armed conflicts. It held that torture as a crime committed in
an armed conflict must contain the following elements:
It (i) consists of the infliction, by act or omission, of severe pain or suffering, whether phys­
ical or mental; in addition, (ii) this act or omission must be intentional; (iii) it must aim at
obtaining information or a confession, or at punishing, intimidating, humiliating or coer­
cing the victim or a third person, or at discriminating, on any ground, against the victim
or a third person; (iv) it must be linked to an armed conflict; (v) at least one of the persons
involved in the torture process must be a public official or must at any rate act in a non­
private capacity, e.g. as a defacto organ of a State or any other authority-wielding entity.
The TC went on to note that:
As is apparent from this enumeration of criteria, the Trial Chamber considers that among
the possible purposes of torture one must also include that of humiliating the victim. This
proposition is warranted by the general spirit of international humanitarian law: the pri­
mary purpose of this body of law is to safeguard human dignity. The proposition is also
supported by some general provisions of such important international treaties as the Geneva
Conventions and Additional Protocols, which consistently aim at protecting persons not
taking part, or no longer taking part, in the hostilities from outrages upon personal dig­
nity’. The notion of humiliation is, in any event close to the notion of intimidation, which is
explicitly referred to in the Torture Convention’s definition of torture (§162).
Subsequently, in Kunarac and others, another TC of the ICTY broadened that defin­
ition. Starting from the correct assumption that one ought to distinguish between the
definition of torture under international hum an rights law and that applicable under
ICL, the TC held, among other things, that ‘the presence of a State official or of any
other authority-wielding person in the torture process is not necessary for the offence
to be regarded as torture under IHL’ (§496). Another TC shared this view in Kvocka
and others (§§137-41).
 112 I N T E R N A T IO N A L C R IM IN A L LAW

In Brdanin the ICTY AC made an interesting contribution to the delineation of the

notion. The appellant had submitted that the TC had erred in law in its determination
of what acts constitute torture; in his view torture, to amount to an international crime,
must involve physical pain equivalent in intensity to the pain accompanying serious
physical injury, such as organ failure, impairment of bodily function or even death. To
support his contention, the appellant had stressed that this notion of torture was that
recently propounded by the US Department of Justice in a legal memorandum. The
AC dismissed the submission. After noting that no matter how powerful or influential
a country is, its practice does not automatically become customary international law’
(§247), the Chamber held that ‘acts inflicting physical pain may am ount to torture even
when they do not cause pain of the type accompanying serious injury’ (251).
Finally, in Naletilic and Martinovic the ICTY AC added an im portant specifica­
tion, given the general purport of the definition of torture eventually set out in inter­
national case law. It clarified that the concrete and specific determination of whether
an act causing severe mental or physical pain amounts to torture must be made on a
case-by-case basis.32
7. Sexual violence. This class of offence includes: (i) rape, a category of crime that
was not defined in international law until a TC of the ICTR set out a rather terse def­
inition in Akayesu (rape is ‘a physical invasion of a sexual nature, committed under
circumstances which are coercive’, §597), taken up by a TC of the ICTY in Delalic
and others (§479). Subsequently, two ICTY TCs delivered im portant judgments, in
Furundzija and Kunarac and others;33 (ii) sexual slavery; (iii) enforced prostitution;

32 As stated in the Kunarac et al. Appeal Judgement, torture “is constituted by an act or an omis­
sion giving rise to ‘severe pain or suffering, whether physical or mental’, but there are no more specific
requirements which allow an exhaustive classification and enumeration of acts which may constitute
torture. Existing case-law has not determined the absolute degree of pain required for an act to amount
to torture. Thus, while the suffering inflicted by some acts may be so obvious that the acts amount per se
to torture, in general allegations of torture must be considered on a case-by-case basis so as to determine
whether, in light of the acts committed and their context, severe physical or mental pain or suffering was
inflicted. Similar case-by-case analysis is necessary regarding the crime of wilfully causing great suffer­
ing’ (§299).
In Furundzija, the TC held that neither international customary or treaty law, nor general princi­
ples of ICL, nor general principles of international law offered any possible definition of rape. It therefore
resorted to the principles of criminal law common to the major legal systems of the world, deriving them,
with caution, from national laws. It concluded that the objective elements of rape are as follows: ‘(i) the sex­
ual penetration, however slight: (a) of the vagina or anus of the victim by the penis of the perpetrator or any
other object used by the perpetrator; or (b) of the mouth of the victim by the penis of the perpetrator; (ii) by
coercion or force or threat of force against the victim or a third person’ (§185).
Subsequently, in Kunarac and others, another TC of the same ICTY placed a different interpretation on
one of the elements of the definition set out in Furundzija; that is the element o f coercion, or force, or threat
of force. According to this TC that element must be taken to mean that there is rape whenever sexual auton­
omy is violated, or in other terms the person subjected to the act has not freely agreed to it or is otherwise not
a voluntary participant. Therefore, that element may be set out as follows: ‘sexual penetration occurs without
the consent of the victim. Consent for this purpose must be consent given voluntarily, as a result of the vic­
tim’s free will, assessed in the context of the surrounding circumstances’ (§460, and see §§438-60).
It would appear that the two definitions are in substance equivalent, for ‘coercion, or force, or threat of
force’ in essence imply or mean Tack of consent’.
 C R IM E S A G A IN S T H U M A N I T Y 113

(iv) forced pregnancy, namely ‘the unlawful confinement of a woman forcibly made
pregnant, with the intent of affecting the ethnic composition of any population or
carrying out other grave violations of international law’ (Article 7(2)(f) of the Rome
Statute for an ICC) (perhaps this sub-category is not yet contemplated by customary
international law: see infra, 5.7.3); (v) enforced sterilization; and (vi) any other form of
sexual violence o f comparable gravity.
8. Persecution against any identifiable group or collectivity on political, racial,
national, ethnic, cultural, religious, gender, or other grounds, that are universally
recognized as impermissible grounds of discrimination under international law;
persecution means the intentional and severe deprivation of fundamental rights
contrary to international law by reason of the identity of the group or collectivity
(Article 7(2)(g) of the Rome Statute for an ICC).
An ICTY TC propounded an elaborate definition of this crime in Kupreskic and
others (§§616-27). It found that the defendants were guilty of persecution, for:
the “deliberate and systematic killing of Bosnian Muslim civilians” as well as their “organ­
ised detention and expulsion from Ahmici [the village where the crimes were committed]’
can constitute persecution. This is because these acts qualify as murder, imprisonment, and
deportation, which are explicitly mentioned in the Statute under Article 5 (§629).
The TC also found that the comprehensive destruction of Bosnian Muslim homes
and property constituted ‘a gross or blatant denial of fundamental human rights, and,
being committed on discriminatory grounds, amounted to persecution (§§630-1).34
9. Enforced disappearance of persons, namely ‘the arrest, detention or abduction of
persons by, or with the authorization, support or acquiescence of, a State or a political
organization, followed by a refusal to acknowledge that deprivation of freedom or to
give information on the fate or whereabouts of those persons, with the intention of
removing them from the protection of the law for a prolonged period of time (Article
7(2)(i) of the Rome Statute for an ICC). It may be noted that with respect to this crime
the ICC Statute has not codified existing customary law but contributed to the crys­
tallization of a nascent rule, evolved prim arily out of treaty law (that is, the numerous
treaties on hum an rights prohibiting various acts falling under this heading), as well
as the case law of the Inter-American Commission and Court of Hum an Rights, in
addition to a number of UN General Assembly resolutions. These various strands have
been instrum ental in the gradual formation of a customary rule prohibiting enforced
disappearance of persons. The ICC Statute has upheld and laid down in a written
provision of the criminalization of this conduct.
10. Other inhumane acts of a similar character and gravity, intentionally causing
great suffering, or serious injury to body or to mental or physical health. This notion
harks back to Article 6(c) of the Nuremberg Statute, which simply criminalized ‘other

34 In Brdanin the ICTY AC rejected the Appellant’s submission that the dismissal of Bosnian Muslims
and Bosnian Croats by the Bosnian Serb authorities had been justified by the security reasons provided for
in Art. 27 of the IVth Geneva Convention of 1949. It held that, as such dismissals were based on the ethnicity
of the individuals concerned, they amounted to persecution as a crime against humanity (A], §§166-7).
 114 IN T E R N A T IO N A L C R IM IN A L LAW

inhumane acts’, by a provision lacking any precision and therefore at odds with the
principle of specificity proper to criminal law (see above, 2.4.1).
The provision was subsequently interpreted in such cases as Ternek35 on the strength
of the ejusdem generis principle, thereby acquiring some degree of precision, as well as in
Kupreskic and others, where an ICTY TC dwelt at length on the interpretation of the clause
(§§563-6). The rule was recently restated in Article 7(l)(k) of the ICC Statute, which to a
large extent codifies and in some respects develops customary international law.
In spite of its relatively loose character (which, however, has been rightly narrowed
down by the case law, as just noted), the rule is im portant for it may function as a
residual clause’ covering and criminalizing instances of inhum an behaviour that do
not neatly fall under any of the other existing categories of crimes against humanity
(for instance, it can cover acts of terrorism not falling under the sub-category of m ur­
der, torture, etc; see infra, 8.6). O f course, the clause may serve this purpose only sub­
ject to strict conditions concerning the gravity of the inhum an conduct.

5.4 THE SUBJECTIVE ELEMENTS

The relevant rules of international law require two mental elements for the crimes
under discussion: (i) the mens rea proper to the underlying offence (murder, rape,
torture, deportation, etc.); and (ii) awareness of the existence of a widespread or sys­
tematic practice.
In most cases the first mental element is intent; that is the intention to bring about
a certain result. However, as noted above (5.3) in the case of murder, case law has
considered that what is required for such conduct to amount to a crime against hum an­
ity is inter alia either the intent to kill, proper to murder, or a different intentional elem­
ent, namely ‘the intent to inflict serious injury in reckless disregard of human life’.
More generally, where an accused, acting as an ‘agent of a system’, does not directly
and immediately cause the inhumane acts, it is not necessary that he anticipate all the
specific consequences of his misconduct; it is sufficient for him to be aware o f the risk
that his action might bring about serious consequences for the victim, on account of
the violence and arbitrariness of the system to which he delivers the victim.36 Thus,
recklessness (or dolus eventualis) m aybe sufficient (see supra 3.7).
The second requirement is that the agent be cognisant o f the link between his mis­
conduct and a widespread or systematic practice (the ‘contextual’ practice may refer

35 The District Court of Tel-Aviv held in a decision of 14 December 1951 that the definition o f ‘other
inhumane acts’, was to apply only to such other inhumane acts as resembled in their nature and their gravity
those specified in the definition (at §7 or p. 538).
36 This point was particularly stressed by the German Supreme Court in the British Occupied Zone,
with particular reference to cases of denunciation of Jews or political opponents to the police or Gestapo,
for instance in T. and K., in which the accused had been charged with burning down a synagogue in 1938
(at 198-202). See also Finta, decisions by the Ontario Court of Appeal (at 1-153) and the Supreme Court of
Canada (at 701-877).
 CR IM E S AGAINST H U M A N I T Y H5

either to offences of the same category or to other large-scale attacks on the civilian
population directed to offend the dignity and humanity of the population, as long as a
link exists between the crime against hum anity at issue and the practice). As the ICTY
AC held in Tadic (AJ, 1999), the perpetrator needs to know that there is an attack on
the civilian population and that his acts comprise part of the attack (§248); a TC held
in Blaskic that the perpetrator needs at least to be aware of the risk that his act is part
of the attack, and then takes that risk (TJ §§247, 251). This does not, however, entail
that he needs to know the details of the attack (.Kunarac and others, TJ §434). The
rationale behind this requirement is clear: ICL intends to punish persons who, being
aware of the fact that the crimes they are perpetrating (or plan to perpetrate) are part
of a general framework of criminality, are thereby encouraged to misbehave and also
hope subsequently to enjoy im punity (if this requirement is lacking, depending on the
circumstances misconduct will amount to either a war crime or an ordinary criminal
offence under domestic law).
W hen crimes against humanity take the form of persecution, another mental elem­
ent also is required: a persecutory or discriminatory animus. The intent must be to
subject a person or group to discrimination, ill-treatment, or harassment, so as to
bring about great suffering or injury to that person or group on religious, political, or
other such grounds. This added element for persecution amounts to a special criminal
intent (dol special).
Finally, courts have not required, as part of the mens rea, that the perpetrator should
have a specifically racist or inhum an frame of m ind.37
To sum up, the requisite subjective element or mens rea in crimes against hum anity
is not simply limited to the criminal intent (or recklessness) requiredfor the underlying
offence (murder, extermination, deportation, rape, torture, persecution, etc.).38 The
viciousness of these crimes goes far beyond the underlying offence, however wicked

37 On this point, a number of cases brought before the German Supreme Court in the British Occupied
Zone are also relevant. Most of these cases concern denunciations by Germans to the police or military
authorities of Jews or political opponents, with the consequence that the denounced persons were arrested
and imprisoned or severely ill-treated; some such cases concern the burning of synagogues in 1938.
In the Sch. case, in 1943 a person had denounced his landlord to the Gestapo for his statements against
Hitler; as a result the man had been arrested and sentenced to death. It is notable that the German Supreme
Court held that the existence of a link or nexus between an offence against humanity and a general policy
or a systematic practice of abuses did not necessarily imply that the author of the crime against humanity
intended by his action to further or promote the violent and brutal practice of the regime within which the
crime had been committed. Nor was it required that the agent should approve the final result of his action. In
other words, the Court simply required an objective link between that act and the policy or practice, as well
as the awareness of the policy or practice, not necessarily the intention to commit the crime for the purpose
of pursuing that policy or practice, ora state of mind which approved the outcome of the crime (at 124). In K.
(at 50) the German Supreme Court in the British Occupied Zone held that for the mens rea in a crime against
humanity to exist, it is not necessary for the agent to have acted ‘out of inhumane convictions’.
The Barbie (at 137-41 and 331-7) and Touvier (337) cases, brought before the French Court of Cassation,
confirm this approach.
38 In some cases courts have held that the subjective element may be culpable negligence (see supra, 3.8):
see Hinselmann and others (at 58-60) (see supra, 3.8). In some German cases it was held that instead mere
negligence or Fahrlässigkeit was not sufficient (see, for instance, R., at 45-9).
 116 IN T E R N A T IO N A L C R IM IN A L LAW

or despicable it may be. This additional element—which helps to distinguish crimes

against hum anity from war crimes—consists of awareness of the broader context into
which this crime fits.

5.5 THE POSSIBLE AUTHORS

Normally it is state organs, i.e. individuals acting in an official capacity such as m ili­
tary commanders, servicemen, etc. who perpetrate crimes against humanity. Is this a
necessary element of the crimes; that is, must the offence be perpetrated by organs or
agents of a state or a governmental authority or on behalf of such bodies, or may such
crimes be committed by individuals not acting in an official capacity? In the latter
case, must the offence be approved or at least condoned or countenanced by a govern­
mental body for it to am ount to a crime against humanity?
The case law seems to indicate that the crimes we are discussing may be committed
by individuals acting in their private capacity, provided they behave in unison, as it
were, with a general state policy and find support for their misdeeds in such policy.
This is clearly shown by the numerous cases brought after 1945 before the German
Supreme Court in the British Occupied Zone and concerning denunciations to the
German authorities of Jews or political opponents by private German individuals.39
An interesting problem that may arise is whether crimes against hum anity may be
committed by state officials acting in a private capacity. It would seem that in such
cases some sort of explicit or implicit approval or endorsement by state or govern­
mental authorities is required, or else that it is necessary for the offence to be clearly
encouraged by a general governmental policy, or at least to fit clearly within such a
policy. This is best illustrated by the Weller case. This case, which seems to have been
unknown until it was cited by the ICTY in Kupreskic (§555), gave rise to six different
judgments by German courts after the Second World War.40

39 See, for instance, the judgments in B„ (at 6-10), in P, (at 11-18), in V, (at 20-5), in R„ (at 46-9), in K„
(at 49-52), in M„ (at 91-5), in H„ (at 385-91), in P, decision of 10 May 1949 (at 17-19), in Ehel M (at 67-9)
in A., (at 144-7), in S„ (at 56-7).
Given its significance (and its historical value as well), it may be useful to dwell on it at some length.
The facts, as set out m almost all the six judgments, are as follows. In early 1940, in the small German
town of Mönchengladbach (near Düsseldorf), various Jewish families were obliged to move together into
one house; eventually 16 persons lived there. One night, in May 1940, three (probably drunken) persons
broke into the house. One of them was the accused Weller, a member of the SS, who was in civilian clothing;
another wore the SA uniform, and the third wore the blue uniform of the German Navy. They obliged all 16
inhabitants to assemble in their night clothes in the basement, then went to the kitchen, where they sum­
moned the 16 persons, one by one. There, 11 (or 10, according to some ofthejudgments) ofthe 16 inhabitants
of the house were beaten with a ‘heavy leather whip’ and verbally abused. The next day the injured parties
reported to the Jewish community (Jüdische Gemeinde), which turned to the local Gestapo. The head ofthe
Gestapo informed the wronged Jews that ‘Weller’s and the other persons’ actions were an isolated event
which would in no way be approved’ (judgment ofthe 16 June 1948, at 3). Thereafter Weller was summoned by
the Gestapo and strongly taken to task by the district leader of the NSDAP (the national socialist party). It is
not clear (nor was it established by the various German courts dealing with the case after 1945) whether in
 CR IM E S A GAINST H U M A N I T Y 117

5.6 THE POSSIBLE VICTIMS

Article 6(c) of the London Agreement establishing the IMT clearly prohibited two
distinct categories of crimes: (i) inhumane acts such as murder, extermination, enslave­
ment, and deportation of any civilian population, i.e. any group of civilians what­
ever their nationality; and (ii) persecution on political, racial, or religious grounds.

1940 Weller had in fact been fined 20RM for bodily harm, as alleged, instead of imprisonment for not less
than two months (this being the penalty which was usually imposed by German law for bodily harm). After
the war, the case was brought before the District Court (Landgericht) of Mönchengladbach. The court found
Weller guilty of grievous bodily harm and sentenced him to 18 months’ imprisonment. While admitting
that he had acted out of racist motives, the court ruled that his action could nevertheless not be regarded as
a crime against humanity. In this connection the court held that three requirements were to be met for such
a crime to exist: (i) a significant breach of human dignity (this the court held to have been established in the
case at issue, and lay in the ill treatment of Jews); (ii) the racial motivation of the offence (this could also be
found in this case); and (iii) the action must be perpetrated ‘by abusing the authority of the state or of t e
police’ (at 7-12). The court found that this third element was lacking. It held that a crime against humanity
must be 'either systematically organized by the government or carried out with its approval’ (at 10). In the
case at issue, one was faced with the ‘occasional persecution of various persons by one person’, not with
abuses perpetrated by the ‘holder of political power or at least by a person acting under the protection of
or with the approval of [those holding] political power’ (ibid.). In short, the necessary ‘link between crimes
against humanity and State authority’ was lacking. _
On appeal, the case was passed on, ‘to ensure uniform jurisprudence’ (at 5), by the Court of Appeal in
Düsseldorf to the Supreme Court (Oberster Gerichtshof) for the British Occupied Zone. This court over­
turned the decision of the District Court and held that the offence did indeed constitute a crime against
humanity. According to the Supreme Court, it was sufficient for the attack on human dignity to be connected
to the national socialist system of power and hegemony (at 7-9). The same Supreme Court, when again seized
with the case (the Prosecutor contending that the sentence newly passed by the Court of Assize was too
light), emphasized that the offence amounted to a crime against humanity, although it had been committed
by Weller ‘on his own initiative and out of racial hatred’ (decision of 10 October 1949, at 2, or 150). The court
also pointed out that the ‘punishment’ (fine of 20RM) allegedly inflicted in 1940 and on which the accused
so much insisted, was a measure that, assuming it had been taken, ‘would not serve justice, but only scorn
the victims’ (at 5, or 153). . ...
The Supreme Court pointed out the following: ‘The national-socialist leadership often, and quite readily,
utilized for its criminal goals and plans actions which appeared to have, or actually had, originated from
quite personal decisions. This was true even of actions that were outwardly disapproved of, perhaps because
it was felt that some sort of consideration should be shown and it was inappropriate openly to admit such
actions [...] The link, in this sense, with the national-socialist system of power and tyranny does in the case
at issue manifestly exist. The state and the party had long before the action at issue made Jews out to be sub­
humans, not worthy to be respected as human beings [...] Also the action of the accused fitted into the numer­
ous persecutory measures which then affected the Jews in Germany, or could at any time affect them. As the
trial court established, the accused, influenced by official propaganda, acted from racial hatred. In the deci­
sion [of the Düsseldorf Court of Appeal] [...] it is rightly pointed out that the link with the national-socialist
system of power and tyranny exists not only in the case of those actions which are ordered and approved by
the holders of hegemony. That link also exists when those actions can only be explained by the atmosphere
and condition created by the authorities in power. The trial court was wrong when it attached decisive value to
the fact that after his action the accused was “rebuked” and that even the Gestapo disapproved of the excess as
an isolated infringement. This action nevertheless fitted into the persecution of Jews carried out by the state
and the party. This is proved by the fact that the accused, assuming he was the subject of an order for summary
punishment (Straßefehl) or a criminal measure (Strafverfügung) for the payment of 20RM—a matter that m
any case has not been clarified-was in any event not held criminally accountable in a manner commensurate
to the gravity of his guilt [...] Given the gravity of the abuse, the harm caused to the victims brought about
consequences extending beyond the single individuals and affecting the whole of humanity (at 206 7).
 ii8 IN T E R N A T IO N A L C R IM IN A L LAW

Since the customary international law of crimes against hum anity that has emerged
is largely based on Article 6(c), it is fitting to look into the fundamental elements of
that provision.
It is apparent from the wording of Article 6(c) that the actus reus is different for
these two classes of crimes. Murder, extermination, and other ‘inhumane acts’ (of
similar gravity) largely constitute offences already covered by all national legal sys­
tems, and also are committed against civilians. ‘Persecutions’, instead, embrace
actions that at the time of their commission may not be prohibited by national legal
systems, for persecution may take the form of acts other than murder, extermination,
enslavement, or deportation. Furthermore, since no mention is made of the pos­
sible victims of persecutions, or rather, as it is not specified that such persecutions
should target ‘any civilian population’, the inference is warranted that not only any
civilian group but also members of the armed forces may be the victims of this class
of crime.
For the purposes of our discussion, it is useful to deal separately with each of the two
classes of crime against humanity.

5.6.1 ‘m u r d e r -t y p e ’ C R IM E S A G A IN S T H U M A N IT Y

‘Murder-type’ crimes against hum anity embrace offences that are perpetrated ‘against
any civilian population’. The words ‘any’ and ‘civilian’ need careful interpretation. As
for ‘any’, it is apparent, both from the text of the provision and from the legislative
history of Article 6(c), that it was intended to cover civilians other than those associ­
ated with the enemy, who were already protected by the traditional rules of the law of
warfare. In other words, by using ‘any’, the draftsmen intended to protect the civilian
population of the state committing crimes against humanity, as well as civilians of its
allied countries or of countries under its control, although formally under no military
occupation.
As for the word civilian’, it is apparent that it was intended to refer to persons other
than lawful combatants, whether or not such persons were civilians fighting alongside
enemy military forces. In other words, this phrase does not cover belligerents.41 The
rationale for the relatively limited scope of this part of Article 6(c) is that enemy com­
batants were already protected by the traditional laws of warfare, while it was deemed
unlikely that a belligerent might commit atrocities against its own servicemen or those
of allied countries. In any event, such atrocities, if any, would come under the jurisdic­
tion of the Courts Martial of the country concerned; in other words, they would fall
within the scope of national legislation.
Nonetheless, after the Second World War courts gradually inclined towards pla­
cing a liberal interpretation on the term ‘civilians’. For instance, the Supreme Court

41 Seethe categories of belligerents envisaged in the Regulations annexed to the Fourth Hague Convention
of 1899/1907 (subsequently supplemented by Article 4 of the Third Geneva Convention of 1949 and Articles
43-4 of the First Additional Protocol of 1977).
 119
C R IM E S AGAINST H U M A N I T Y

of Germany in the British Occupied Zone propounded a broad construction of

Article 6(c). This court held in at least three cases that military persons could be the
victims of crimes against humanity even in situations where the crime did not take
the form of persecution. In other words, the court held that the crime at issue could
be perpetrated against military personnel even if the offence was not one of those
envisaged in the second part of Article 6(c) or in the corresponding second part of
Article II(l)Cc) of Control Council Law no. 10. As a consequence, the court substan­
tially broadened the notion o f‘any civilian population’ included in the first part o ft at
provision. These three cases will be briefly summarized.
In a decision of 27 July 1948 in id, the court pronounced upon the guilt of a
member of the Nazi Party and Nazi commandos (NSKK), who in 1944 had denounced
a non-commissioned officer in uniform and member of the Nazi Party and the SA
(assault units), for insulting the leadership of the Party. As a result of this denunci­
ation, the victim had been brought to trial three times and eventually sentenced to
death (the sentence had not been carried out because in the interim the Russians had
occupied Germany). The court held that the denunciation could constitute a crime
against humanity if it could be proved that the agent had intended to hand over the
victim to the ‘uncontrollable power structure of the [Nazi] party and State, knowing
that as a consequence of his denunciation, the victim was likely to be caught up m an
arbitrary and violent system (at 47).
In 1948, in P. and others the same court applied the notion of crimes against hum an­
ity to members of the military. In the night following Germany’s partial capitulation
(5 May 1945), four German marines had tried to escape from Denmark back to
Germany. The next day they were caught by Danes and delivered to the German troops,
who court-martialled and sentenced three of them to death for desertion; on the very
day of the general capitulation of Germany (10 May 1945), the three were executed.
The German Supreme Court found that the five members of the Court Martial were
guilty of complicity in a crime against humanity. According to the Supreme Court,
the glaring discrepancy between the offence and the punishment proved that the exe­
cution of the three marines had constituted a clear manifestation of the Nazis brutal
and intimidatory justice, which denied the very essence of hum anity in blind defer­
ence to the superior exigencies of the Nazi State. In this case as well, there had taken
place ‘an intolerable degradation of the victims to mere means for the pursuit of a goal,
hence the depersonalisation and reification of hum an beings’ (at 220); consequent y,
by sentencing to death those marines, the members of the Court M artial had also
injured humanity as a whole. With regard to the wording of the relevant provision
on crimes against humanity (namely, Article II(l)(c) of Control Council Law no. 10,
which referred only to offences ‘against civilian populations’), the court observed the
following;
Whoever notes the expressly emphasized illustrative character of the instances and classes
of instance mentioned there, cannot come to the conclusion that action between soldiers
may not constitute crimes against humanity. [Admittedly], a single and isolated excess
 120 IN T E R N A T IO N A L C R IM IN A L LAW

would not constitute a crime against humanity pursuant to the legal notion of such crimes
[However], it has already been shown [in the judgment] that the action at issue can belong
to the criminal system and criminal tendency of the Nazi era. For the offence to be a crime
against humanity, it is not necessary that the action should support or sustain Nazi tyranny,
or that the accused should intend so to act (at 228).

Finally, in its decision of 18 October 1949 in H., the court dealt with a case in which a
German judge had presided over two trials by a Naval Court Martial against two officers
of the German Navy: one against a commander of submarines who had been accused
of criticizing Hitler in 1944, the other against a lieutenant-commander of the German
naval forces, charged with procuring two foreign identity cards for himself and his wife
m 1944. The judge had initially sentenced both officers to death (the first had been exe­
cuted, while the sentence against the second had been commuted by Hitler to ten years’
imprisonment). The Supreme Court held that the judge could be held guilty of crimes
against humanity to the extent that his action was undertaken deliberately in connec­
tion with the Nazi system of violence and terror (at 233-4,238,241-4),
After the Second World War other courts, with the notable exception of the French
Court of Cassation in Barbie,42 tended instead to place a strict interpretation on the
term civilians and consequently to rule out from the notion of victims of crimes
against hum anity persons who belonged, or had belonged, to the military. Indicative
in this respect is Neddermeier, brought before a British Court of Appeal established
under Control Council Law no. 10.43
Ihe trend towards loosening the strict requirement that the victims be civilians also
continued, however, in more recent times. It is significant that the ICTY has placed a
liberal inteipretation on the narrow notion ofvictim s of crimes against hum anity set
out in Article 5 of its Statute (according to which those crimes can only be committed
against any civilian population’). In its decision in Mrksic and others (rendered under
Rule 61 of the Rules of Procedure and Evidence), the court held that crimes against
hum anity could be committed even where the victims at one time bore arms.44

42 In 1985 the French Court of Cassation in Barbie held that the victims of crimes against humanity
could include the opponents of [a] policy [of ideological supremacy, manifesting itself in inhumane acts
139 ^ SeCUti0n committed in a systematic manner], whatever the form of their “opposition” ’ (at 137 and

43 The accused had been convicted by the High Court of Brunswick of crimes against humanity, pursu­
ant to Article II(l)(c) of Control Council Law no. 10. The court had found that he had caused a number of
Polish workers to be beaten (the Poles, originally brought to Germany as prisoners of war, had subsequently
been compelled to sign agreements to surrender such status and be treated as civilians). Before the Appeal
Court the Defence claimed among other things that the offence did not amount to a crime against humanity
because there was no element of cruelty’. The Prosecution admitted that, if the victims of ill-treatment were
to be considered as prisoners of war, a conviction under the label of war crimes ‘could be substituted’ for the
conviction for crimes against humanity. The court held that the victims had the status of prisoners of war
and not civilians’. It consequently set aside the conviction for crimes against humanity and substituted for
it that for war crimes (at 58-60).
44 In Kupreskic et al, a TC held that ‘the presence of those actively involved in the conflict should not
prevent the characterization of a population as civilian and those actively involved in a resistance move­
ment can qualify as victims of crimes against humanity’. In Kunarac and others, an ICTY TC held that ‘as a
minimum, the perpetrator must have known or considered the possibility that the victim of his crime was a
 CR IM E S A G A IN S T H U M A N I T Y 121

A different issue that arose in cases brought before the United States Military
Tribunals sitting at Nuremberg is whether victims of extermination through euthan­
asia as a crime against hum anity may be nationals of the state concerned, 01 whethei
such victims must perforce be foreigners. In these cases some defendants had been
accused of participating in euthanasia programmes for the chronically disabled or ter­
minally ill. The Tribunals wrongly held that euthanasia amounted to a crime against
hum anity only if carried out against foreigners, i.e. non-nationals of the state practis­
ing euthanasia.45

5.6.2 ‘P E R S E C U T I O N - T Y P E ’ C R I M E S A G A I N S T H U M A N I T Y

As stated above, it is apparent from Article 6(c) that in the case of persecution, the
victims of crimes against hum anity need not necessarily be civilians; they may also
include military personnel. There is an obvious rationale for this regulation, trad­
itional laws of warfare, while they protected servicemen against such illegal actions
by the enemy as treachery and use of prohibited means or methods of warfare, did not
safeguard them against persecution either by the enemy, or by the Allies or by the very
authorities to which military personnel belonged.
The textual and logical construction of Article 6(c) was confirmed implicitly in
Pilz by the Dutch Special Court of Cassation and explicitly by French courts in Barbie
and Touvier.46

civilian [...] in case of doubt as to whether a person is a civilian, that person shall be considered to be a civil­
ian. The Prosecution must show that the perpetrator could not reasonably have believed that the victim was
a member of the armed forces’ (§435).
45 In Karl Brandt, the Tribunal found that the defendant had participated in a programme for the
extermination of disabled persons, and that this programme had quickly been extended to Jews and then
to concentration camp inmates (those inmates deemed to be unfit for labour were ruthlessly weeded out
and sent to extermination camps in great numbers). The Tribunal stressed that it was difficult to believe
Brandt’s assertion that he was not implicated in the extermination of Jews or of concentration camp inmates;
however, even if it were true, ‘the evidence [was] conclusive that almost at the outset of the programme non-
German nationals were selected for euthanasia and extermination’ (at 197-8).
The same Tribunal also took this restrictive (and undisputedly fallacious) view in Greifelt and others
(at 654-5).
46 As recalled above, Pilz was a German medical doctor serving with the German army occupying
the Netherlands. He had prevented a young Dutchman, who had enlisted in the German army and been
wounded while attempting to escape from his unit, from being treated and had then ordered a subordinate
to kill the Dutchman. The Dutch Special Court of Cassation held that the offence did not amount to a war
crime, because the victim, even if still a Dutch national, belonged to the German army. It then asked itself
whether it could amount to a crime against humanity, and answered in the negative, noting that the vic­
tim ‘was not part of the civilian population of occupied territory, nor [could] the acts with which he [was]
charged be seen as forming part of a system of persecution on political, racial or religious grounds (at 1211).
Clearly it can be deduced from this reasoning that had the victim, a member of the military, been the object
of persecution on one of those grounds, the offence might have amounted to a crime against humanity.
In Barbie, in a decision rendered on 20 December 1985 the French Court of Cassation held that crimes
against humanity in the form of persecution had been perpetrated against members of the French Resist­
ance movements (at 136). Subsequently, the Paris Court of Appeal took the same view in a judgment of
9 July 1986, again in Barbie, followed by the Chambre d ’accusation of the same Court of Appeal in a judg­
ment of 13 April 1992 in Touvier (at 352). In this last decision the Chambre d ’accusation held that: Jews and
 122 I N T E R N A T IO N A L C R IM IN A L LAW

5.6.3 T H E G R A D U A L B R O A D E N IN G OF T H E C A T E G O R Y O F V I C T I M S

As a result of the gradual disappearance in customary international law of the nexus

between crimes against humanity and armed conflict, so too has the emphasis on civil­
ians as the exclusive class of victims of such crimes dwindled, if not disappeared. For if
crimes against humanity may be committed in time o f peace as well, it no longer makes
sense to require that such crimes be perpetrated against civilians alone. Why should
members of military forces be excluded, since they in any case would not be protected
by IHL in the absence of any armed conflict? Plainly, in times of peace military person­
nel too may become the object of crimes against humanity at the hands of their own
authorities. By the same token, in time of armed hostilities, there is no longer any reason
for excluding servicemen, whether or not hors de combat (wounded, sick, or prisoners
of war), from protection against crimes against humanity (chiefly persecution), whether
committed by their own authorities, by allied forces, or by the enemy.
The broadening of the category of persons safeguarded by the relevant rules of cus­
tomary international law is consonant with the overall trend in IHL toward expanding
the scope of protection of the basic values of hum an dignity, regardless of the legal sta­
tus of those entitled to such protection. This trend has manifested itself in, inter alia,
the adoption of international treaties protecting hum an rights and treaties prohibit­
ing crimes such as genocide, apartheid, or torture, in the passing of some significant
resolutions by the United Nations General Assembly, and in certain pronouncements
of the International C ourt of Justice. Nowadays, international hum an rights stand­
ards also clearly protect individuals against abuses and misdeeds of their own gov­
ernmental authorities. It follows that there no longer exists any substantial reason
for refusing to apply the notion of crimes against hum anity to vicious and inhum ane
actions undertaken on a large scale by governments against the hum an dignity of
their own military or the m ilitary personnel of allies or other non-enemy countries (or
even of the enemy). It is worth noting that, had this expansion of the notion of crimes
against hum anity not occurred, a strict interpretation of the notion of civilians would
lead in times of armed conflict to a questionable result. Some categories of combatants
who, in modern armed conflicts (particularly in internal conflicts) often find them ­
selves in a twilight area, would remain unprotected—or scantily protected—against
serious atrocities. Consider, for example, members of param ilitary forces or members
of police forces who occasionally or sporadically take part in hostilities. These are per­
sons whose legal status may be uncertain, as one may not be sure whether they are to
be regarded as combatants or civilians.47 It could therefore follow that, under a strict
and traditional interpretation of the crimes at issue, and assuming that these persons

members of the Resistance persecuted in a systematic manner in the name of a state practising a policy of
ideological supremacy, the former by reason of their membership of a racial or religious community, the latter
by reason of their opposition to that policy, can equally be the victims of crimes against humanity’ (at 352).
47 Under Article 43(3) of the First Additional Protocol of 1977 ‘Whenever a Party to a conflict incorp-
porates a paramilitary or armed law enforcement agency into its armed forces it shall so notify the other
Parties to the conflict.’ If such notification has not been made, the status of the paramilitary or police force
maybe uncertain.
 C R IM E S AGAINST H U M A N I T Y 123

were at the same time regarded as combatants, they would ultimately be unprotected
by the prohibition against such crimes.
By way of conclusion on this point, the proposition is warranted that the scope of the
customary rules on crimes against hum anity is much broader than normally admit­
ted. Private individuals may also perpetrate those crimes (provided the governmental
authorities approve of or condone, or at any rate fail to repress their action, or their
action fits into a widespread or systematic practice of official misconduct). Furthermore,
the victims of the crimes belonging to the subclass of persecutory offences, as well as—it
is here contended—those of the other subclass, may embrace both civilians and com­
batants. In addition, such victims need not have the nationality of an enemy country
but may belong to the country whose authorities order, approve, fail to punish, or con­
done the pattern of misbehaviour amounting to crimes against humanity.

5.7 CUSTOMARY IN TER N A TIO N A L LAW AND

ARTICLE 7 OF TH E ICC STATUTE

Let us now ask ourselves whether Article 7 of the ICC Statute, contemplating crimes
against hum anity as one of the categories of criminal conduct over which the Court
has jurisdiction, departs from or instead restates customary international law.
A comparison between customary international law and the ICC Statute shows
that by and large the latter is based on the former. However, many differences may be
discerned. In some respects, Article 7 elaborates upon and clarifies, in other respects
it is narrower than, customary international law; in others, it instead broadens cus­
tomary rules.

5.7.1 A R E A S W H E R E A R T I C L E 7 SETS F O R T H E L E M E N T S OF
C U S T O M A R Y I N T E R N A T I O N A L LAW

Article 7 specifies and elaborates upon customary international law in many respects.
First, it specifies that a crime against hum anity must be committed ‘with knowledge of
the attack’. The provision thus makes it clear that the requisite mens rea must include
the awareness that the individual criminal act is part of a widespread or systematic
attack on a civilian population.
Secondly, Article 7 clarifies the objective elements of some of the underlying offences,
by making explicit notions that, until set out in this Article, were only implicit and could
therefore be determined only by way of interpretation. These notions are further elabor­
ated upon in the ‘Elements of Crimes’ adopted by the Preparatory Commission.48

48 This applies to the following notions: (i) ‘Extermination’, which, pursuant to Article 7(2)(b), ‘includes
the intentional infliction of conditions of life, inter alia, the deprivation of access to food and medi­
cine, calculated to bring about the destruction of part of a population’; (ii) ‘Enslavement’, which under
Article 7(2) (c) refers to ‘the exercise of any or all the powers attaching to the right of ownership over a person
 124 I N T E R N A T IO N A L C R IM IN A L LAW

Finally, one should emphasize that the ‘Elements of Crime’ have clarified an im port­
ant aspect of mens rea. In commenting on the need for the offender to have knowledge
of a widespread or systematic attack on a civilian population, it is stated there that:
However, the last element should not be interpreted as requiring proof that the perpetrator
had knowledge of all characteristics of the attack or the precise details of the plan or policy
of the state or organization. In the case of an emerging widespread or systematic attack on a
civilian population, the intent clause of the last element indicates that this mental element is
satisfied if the perpetrator intended to further such attack.

5.7.2 A R E A S W H E R E A R T I C L E 7 IS N A R R O W E R T H A N C U S T O M A R Y
I N T E R N A T I O N A L LAW

On some points, Article 7 departs from customary law by setting out notions at odds
with that body of law.
First, Article 7(1) defines the victim or target of crimes against hum anity as ‘any
civilian population. This provision, which thus adopts a position similar to that taken
in the statutes of the ICTY (Article 5) and the ICTR (Article 6), excludes non-civilians

and includes the exercise of such power in the course of trafficking in persons, in particular women and
children’. This notion is made more specific in the ‘Elements of Crime’, where it is stated that the conduct at
issue takes place when ‘the perpetrator exercised any or all of the powers attaching to the right of ownership
over one or more persons, such as purchasing, selling, lending or bartering such a person or persons, or by
imposing on them a similar deprivation of liberty’, and it is added (in a footnote) that deprivation ofliberty
may include exacting forced labour or otherwise reducing a person to a servile status’; (iii) ‘Deportation or
forcible transfer of population’, which under Article 7(2)(d) is defined as ‘forced displacement of the persons
concerned by expulsion or other coercive acts from the area in which they are lawfully present, without
grounds permitted under international law’. In the 'Elements of Crime’ the important specification is added
that the persons deported or forcibly transferred ‘were lawfully present in the area from which they were so
deported or transferred’ and that ‘the perpetrator was aware of the factual circumstances that established
the lawfulness of such presence’; (iv) ‘Torture’; Article 7(2) (e) sets out a definition of torture that, rightly, is
broader than that laid down in customary international law with regard to torture as an international crime
per se as established by an ICTY TC in Kunarac and others. In general international law, for the torture as a
discrete crime to have occurred, it is necessary, amongst other things, that a public official be involved, either
as the perpetrator or as one of the participants or accomplices (see infra, 7.2). By contrast, under Article 7,
torture may amount to a crime against humanity even if committed by civilians against other civilians with­
out any involvement of public officials or military personnel. Indeed, Article 7(2)(e) defines torture as ‘the
intentional infliction of severe pain or suffering, whether physical or mental, upon a person in the custody
01 under the control of the accused’. Consequently, as long as the single act of torture is part of a widespread
or systematic practice, even torture inflicted without any participation of a public official is punishable as a
crime against humanity. The only involvement of public authorities is required by the ‘Elements of Crime’:
it is necessary for the widespread or systematic practice constituting the general context of the crime to take
place pursuant to or in furtherance of a state or organizational policy’ of torture; (v) ‘Imprisonment’, which
under Article 7(l)(e) embraces other severe deprivation of physical liberty in violation of fundamental rules
of international law’; (vi) ‘Rape’, which under Article 7(l)(g) is not the sole form of sexual violence punish­
able under international law; as spelled out by an ICTY TC in Furundzija, in addition to the violent physical
penetration of the victim’s body, other forms of serious sexual violence are criminalized by international
law: ‘sexual slavery, enforced prostitution [...] enforced sterilization, or any other form of sexual violence of
comparable gravity’; (vii) ‘Other inhumane acts’ are defined in Article 7(l)(k) as acts ‘of a similar character
[to those listed in Article 7(1), from (a) to (j)] intentionally causing great suffering, or serious injury to body
or to mental or physical health.’ '
 CR IM E S A G A IN S T H U M A N I T Y 12 5

(i.e. the military) from the victims of the crimes under discussion. Thus, any of the
acts enumerated in Article 7(l)(c) to (k), if perpetrated against an enemy combatant,
would only amount to a war crime or a grave breach of the 1949 Geneva Conventions.
The question arises whether the term ‘civilian population’ includes belligerents hors
de combat who have laid down their weapons, either because they are wounded or
because they have been captured. As we have seen above, the case law of the ICTY
has answered this question in the affirmative. It would seem to be consonant with the
hum anitarian object and purpose of Article 7 to suggest the same solution with regard
to this provision.
Secondly, Article 7, in defining attack directed against any civilian population’
narrows the scope of the notion of ‘widespread or systematic practice requii ed as a
context of a specific offence, for the offence to amount to a crime against humanity.
Indeed, in paragraph 2(a) that provision stipulates that attack means a course of con­
duct involving the multiple commission of acts referred to in paragraph 1 against any
civilian population, pursuant to or in furtherance of a State or organizational policy to
commit such attack’. It would seem that the Statute requires that the offender, in com­
mitting a crime against humanity, pursue or promote such a practice. It would follow
that any practice simply tolerated or condoned by a state or an organization would not
constitute an attack on the civilian population or a widespread or systematic practice.
For instance, in the case of murder, or rape, or forced pregnancy, why should it be
required that the general practice constitute a policy pursued by a state or an organ­
ization? Would it not be sufficient for the practice to be accepted, or tolerated, or acqui­
esced in by the state or the organization, for those offences to constitute crimes against
humanity? Clearly, this requirement goes beyond what is required under international
customary law and unduly restricts the notion under discussion. The Elements of
Crime’ make this restriction even broader and more explicit. There it is stated that ‘the
policy to commit such attack’ ‘requires that the State or organization actively promote
or encourage such an attack against a civilian population (emphasis added).
Thirdly, Article 7 is less liberal than customary international law with regard to
one element of the definition of persecution. Under Article 7(l)(h), persecution, in
order to fall under the jurisdiction of the ICC, must be perpetrated ‘in connection
with any act referred to in this paragraph or any crime within the jurisdiction of the
C ourt’. Instead, under customary international law no such link is required. In other
words, it is not necessary for persecution to consist of (a) conduct defined as a war
crime or a crime against humanity or linked to any such crime; plus (b) a discrim ­
inatory intent. Under general international law, persecution may also consist of acts
not punishable as war crimes or crimes against humanity, as long as such acts (a) result
in egregious violations of fundamental hum an rights; (b) are part of a widespread or
systematic practice; and (c) are committed with a discriminatory intent. Article 7(1)
(h) imposes a further burden on the Prosecution: it must be proved that, in addition to
discriminatory acts based on one of the grounds described in this provision, the actus
reus consists of one of the acts prohibited in Article 7(1) or of a war crime or genocide
(or aggression, if this crime is eventually accepted as falling under the jurisdiction of
 126 I N T E R N A T IO N A L C R IM IN A L LAW

the Court), or must be ‘connected’ with such acts or crimes. Besides adding a require­
ment not provided for in general international law, Article 7 uses the phrase ‘in con­
nection w ith’, which is unclear and susceptible to many interpretations.

5.7.3 A R E A S W H E R E A R T I C L E 7 IS B R O A D E R T H A N C U S T O M A R Y
IN T E R N A T IO N A L LAW

Article 7 expands general international law in at least two respects.

First, it broadens the classes of conduct amounting to crimes against humanity.
Thus, it includes within this category ‘forced pregnancy’ (Article 7(l)(g) and (2)(f));
(see supra, 5.3, sub 7); ‘enforced disappearance of persons’ (Article 7(l)(i) and (2)(i));
and the crime of apartheid (Article 7(l)(j) and (2)(h)) (as noted supra at 5.3 sub 9, the
ICC Statute has, however, contributed to the recent formation of a customary rule on
the matter).
Secondly, in dealing with the crime of persecution, it greatly expands the category
of discriminatory grounds. While under customary international law these grounds
may be political, racial, ethnic, or religious, Article 7(l)(h) adds ‘cultural’ grounds,
gender as defined in paragraph 3 [of the same provision]’, as well as other grounds
that are universally recognized as impermissible under international law’.
 6
GENOCIDE

6.1 THE NOTION

Genocide is the intentional destruction, through one offive well-specified categories of con­
duct, of one of some groups as such (national, ethnical, racial, or religious) or of members of
one of these groups as such.
Article 6(c) of the Charter of the IMT did not envisage genocide as a crime falling
under the Tribunal’s jurisdiction. However, in referring to crimes against humanity it
used a wording (‘murder, extermination, enslavement, deportation, and other inhumane
acts committed against any civilian population’ and ‘persecutions on political, racial or
religious grounds’) that encompassed large-scale massacres of ethnic, racial or religious
groups. In dealing with the extermination of Jews and other ethnic or religious groups, the
IMT referred in its judgment to the crime of persecution (Goring and others, at 247-55).
The extermination of Jews as a crime against hum anity was discussed in a few other
cases: Hoess, decided by a Polish court in 1947 (at 12-18), and Greifelt and others, heard
in 1948 by a US Military Tribunal (at 2-36). In the latter judgment (and in Altstotter
and others, at 1128,1156), the word ‘genocide’ was used to describe the criminal con­
duct, without however elevating genocide to a distinct category of criminality. In other
cases (for instance, Kramer and others (the Belsen trial), at 4, 117-21; and see 106) the
killing of Jews in concentration camps was dealt with as a war crime.
Thus, at this stage prosecution and punishment of massacres of ethnic or religious
groups did not require evidence of the ‘special intent’ typical of genocide (see infra,
6.5), but simply proof of the subjective and objective elements of either crimes against
hum anity or war crimes.

6.2 THE 1948 CONVENTION ON GENOCIDE

Genocide acquired autonomous significance as a specific crime in 1948, when the UN
GA adopted the Genocide Convention.

6.2.1 M A I N F E A T U R E S OF T H E C O N V E N T I O N

A careful look at the Convention shows that it pursued two goals: (i) to oblige Contracting
Parties to criminalize genocide and punish their authors within the legal system of each
 128 I N T E R N A T IO N A L C R IM IN A L LAW

Party, and accordingly (ii) to provide for the judicial cooperation of those contracting
states for the suppression of the crime. This is already made clear by the preamble, where
the draughtsmen, after declaring that genocide is a crime under international law, set out
their conviction that ‘in order to liberate mankind from such an odious scourge, inter­
national co-operation is required’.1The various provisions of the Convention bear out that
this is its main purpose. In Article I it is stipulated that the Contracting Parties undertake
to prevent and punish’ genocide. Article III imposes upon Contracting Parties the obli­
gation to punish not only the perpetration of genocide but also conduct somehow linked
to the crime, which the provision defines by using criminal law categories: conspiracy,
incitement, attempt, and complicity. By Article IV states assume the obligation to punish
persons committing genocide or related conduct even if they are ‘constitutionally respon­
sible rulers or public officials. Article V provides for the enactment of the necessary
criminal legislation, with particular regard to penalties. Article VI deals with criminal
jurisdiction over the offence, and Article VII addresses the issue of extradition.
It thus seems clear, both from the text of the Convention and the preparatory
works,123that the Genocide Convention is very much like some previous international
treaties such as the 1926 Convention on Slavery (followed by the Protocol of 1953),
the 1929 International Convention for the Suppression of Counterfeiting Currency, or
the more recent UN Convention Against Torture of 1984, which (i) provide for a set
ol international obligations that contracting states are required to implement within
their own domestic legal systems, and in addition (ii) arrange for judicial cooperation
in the matter regulated by the treaty.
It was perhaps the naive assumption of the Convention’s draughtsmen that, after
the horrendous genocide of European Jews in the Second World War and the stiff
punishm ent of many of its planners and perpetrators at the hands of criminal courts,
contracting states themselves would not dare to engage in genocide. Plausibly it is this
assumption that to some extent accounts for the odd (or, rather, ingenuous) provision
in Article VI stipulating that persons accused of genocide must be prosecuted and
tried by the judicial authorities of the territory in which ‘the act was com m itted’ (plus
a future international criminal court that in 1948 looked like a radiant daydream).2

1 Emphasis added.
For the preparatory work, see for instance N. Robinson, The Genocide Convention—A Commentary (New
York: Institute of Jewish Affairs, 1960). It is crystal dear, for instance with regard to Article III, that the
authors of the Convention only had in mind action to be taken by each contracting state at the domestic level.
This is also apparent from the statement of the Swedish delegate: ‘The discussion at the beginning of this meet­
ing seems to me to have shown that the significance of the terms corresponding to the French and English
expressions here in question [used in Article III]—incitement, conspiracy, attempt, complicity, etc.—is sub­
ject to certain variations in many systems of criminal law represented here. When these expressions have to be
translated in order to introduce the text of the Convention into our different criminal codes in other languages, it
will no doubt be necessary to resign ourselves to the fact that certain differences in meaning are inevitable.”It
would therefore be advisable to indicate in the Committee’s report that Article IV of the Convention does not
bind signatory States to punish the various types of acts to a greater extent than the corresponding acts aimed
at the most serious crimes, as, for example, murder and high treason, already recognized under national law.’
(A/760, at 4 and A/C.6 SR.84, at 7, reported in Robinson, op.cit., at 70; emphasis added).
3 That the 1948 Convention was conceived of as a treaty having the scope I have just described, can also
be inferred from another circumstance: both in 1947-8 and subsequently, states have consistently shied away
 G E N O C ID E 129

6.2.2 T H E D U A L R E G I M E O F R E S P O N S I B I L I T Y F O R G E N O C I D E ,
A C C O R D I N G T O T H E ICJ

In the judgement delivered on 26 February 2007 in the Bosnia v. Serbia case, the
International Court of Justice (ICJ) chose to place an expansive interpretation on the
Convention. It preferred to look upon it as a treaty that also imposes on contracting states
as such, that is as international subjects, specific obligations relating to their own behav­
iour towards groups protected under Article II (1) (national, ethnical, racial, or religious
groups). This led the Court to propound the notion that the Convention upholds a dual­
ity of responsibility’ for genocide: according to the Court the same acts may give rise
both to individual criminal liability and state responsibility (§§163 and 173).
The Court first of all construed Article I as imposing not only a duty to prevent
and punish genocide, but also an obligation for contracting states to refrain from
engaging in genocide (§§162—6). This interpretation, as the C ourt rightly noted, is fully
warranted having regard to the object and purpose of the Convention. It broadens the
scope of Article I and also makes the set of obligations it is designed to impose more
consistent: it would be paradoxical’ for states to be obliged to prevent and punish
genocide, while being free themselves to engage in genocide.4 The interpretation ‘is
also supported by the purely hum anitarian and civilizing purpose of the Convention’.1
I would add that this obligation, as set out by the Court, does not remain unchecked:
it is the ICJ that can ensure the judicial safeguard of compliance with such obligation,
pursuant to Article IX of the Convention. However, the Court did not stop here. It
interpreted Article III as implying that contracting states are also under the obliga­
tion to refrain from engaging in any of the sets of conduct envisaged in that provision:
conspiracy, direct and public incitement, attempt to commit genocide, or complicity
in genocide.6
Thus the Court ended up contemplating the same prohibited conduct both with
regard to individuals and with respect to states. Both individuals and states may
incur, respectively, criminal liability and state responsibility for the same unlawful
behaviour (acts of genocide, conspiracy, incitement, attempt, or complicity). This view
has been criticized by a num ber of commentators.7 According to a more convincing

from the notion that they -as such-m ight be held criminally accountable for genocide. In their view, states
as international subjects may not commit crimes proper: they can only incur state responsibility foi inter­
nationally wrongful acts. Hence, it would be inappropriate to apply criminal law categories to their conduct.
4 Ibid., §166. 5 Ibid., §162.
6 In the view of the Court, although the concepts used in paragraphs (b) to (e) of Article III, and particu­
larly that of “complicity”, refer to well known categories of criminal law and, as such, appear particularly
well adapted to the exercise of penal sanctions against individuals [...] it would however not be in keeping
with the object and purpose of the Convention to deny that the international responsibility of a State—even
though quite different in nature from criminal responsibility—can be engaged through one of the acts, other
than genocide itself, enumerated in Article III’. (Ibid., §167.)
7 See P. Gaeta, ‘Genocide d ’Etat et responsabilite penale individuelle’, in 111 RG DIP (2007) 272-84;
‘On What Conditions Can a State Be Held Responsible For Genocide?’, in 18 EJIL (2007), forthcoming;
A. Cassese, ‘On the Use of Criminal Law Nations in Determining State Responsibility for Genocide’, 5 JICJ
(2007), 875-87.
 130 IN T E R N A T IO N A L C R IM IN A L LAW

view the Convention (and the customary rules evolved as a result of its broad accept­
ance by states and the passing of national legislation along the same lines) chiefly
provides for criminal liability o f individuals for any of the acts of genocide enumerated
in Article III of the Convention (and in addition imposes on contracting states only
the obligation to prevent and repress genocide by individuals, be they state officials
or private individuals). As for state responsibility for genocide, it arises in the event
of a breach of the customary rule of international law obliging states to refrain from
engaging in genocide as a conduct involving a genocidal policy pursued or tolerated
by the state.8 Thus, as has been rightly noted,9 the subjective and objective conditions
on which the arising of, respectively, state and individual responsibility for genocide is
contingent, may and indeed do differ.

6.2.3 M A I N M E R I T S OF T H E C O N V E N T I O N

The Convention has numerous merits. Among other things, (i) it sets out a careful def­
inition of the crime; (ii) it punishes other acts connected with genocide (conspiracy,
complicity, etc.); (iii) it prohibits genocide regardless of whether it is perpetrated in time
of war or peace; (iv) thanks to the Convention and its very broad acceptance by states,
at the level of state responsibility it is now widely recognized that customary rules on
genocide impose erga omnes obligations; that is, lay down obligations towards all other
member states of the international community, and at the same time confer on any state
the right to require that acts of genocide be discontinued. Furthermore, those rules now
form part ofjus cogens or the body of peremptory norms; that is, they may not be dero­
gated from by international agreement (nor a fortiori by national legislation).
One should, however, be mindful of the flaws or omissions of the Convention. These
are the most blatant ones;

1. The definition of genocide does not embrace cultural genocide (that is, the destruc­
tion of the language and culture of a group).10Probably it was felt that cultural genocide
is a rather nebulous concept. Similarly, genocide does not encompass the extermin­
ation of a group on political grounds. This was a deliberate omission. One may wonder
whether the elimination of political groups fits with the notion of genocide. Killing all
the communists in a country is extermination, but is it genocide? Many would think
not. The Convention confined itself to the physical destruction of relatively stable
groups to which persons in most instances belong ‘involuntarily’ and, often, by birth
(clearly, in the case of religious groups, membership may be voluntary).

8 This is the approach substantially underpinning the section on genocide of the Report of the UN
International Commission of Inquiry on Darfur, UN doc. S/2005/60, §§439-522.
9 P. Gaeta, op. cit., Genocide d ’Etat, cit, supra, at n. 7.
10 See, for instance, the decision of the High Court of Australia in Kruger v. Commonwealth (at 32). It
should be noted that some countries, in passing legislation on genocide, have broadened the category of
protected groups. For instance, in Ethiopia Art. 281 of the 1957 Penal Code also uses genocide with regard to
‘political groups’. Other countries include ‘social groups’ within the definition of genocide: Peru (Art. 129,
Criminal Code); Paraguay (Art. 308, Criminal Code); Lithuania (Art. 71, Criminal Code).
 G E N O C ID E 131

2. The four classes of protected groups (national, ethnical, racial, and religious) are
not defined, nor are criteria for their definition provided.
3. The enforcement mechanism envisaged in the Convention is ineffective (in
Article IV the Convention contemplates trials before the courts of the state on the
territory of which genocide has occurred, or before a future ‘international penal tribu­
nal’. This is a flaw because it is the territorial state authorities (or persons supported by
such authorities) that normally tend to commit acts of genocide; so national prosecu­
tors will be reluctant to bring prosecutions; furthermore, no international penal tri­
bunal existed at the time, nor for 50 years afterwards. Moreover, Article VIII provides
that any contracting party ‘may call upon the competent organs of the United Nations
to take such action’ under the Charter as they consider appropriate’ for the prevention
or suppression of genocide, whereas Article IX confers on the ICJ jurisdiction over
disputes between states concerning the interpretation, application, or fulfilment of
the Convention.
Indeed, at the enforcement level the Convention has long proved a failure. Only
once did a United Nations body pronounce on a specific instance of massacres, that it
defined as genocide: this occurred in the case of Sabra and Shatila, when the UN GA
characterized the mass killing of Palestinians perpetrated there by Christian falangist
troops as ‘an act of genocide’ in its resolution 37/123 D of 16 December 1982. (However,
the GA did not set out the legal reasons for this ‘finding’, nor did it draw any legal
consequences from it.) Subsequently in 1993, for the first time a state brought a case of
genocide before the ICJ: Bosnia v. Serbia. In 1999 Croatia also instituted before the ICJ
proceedings against Serbia for violations of the Genocide Convention.

6.3 DEVELO PM EN TS IN THE CASE

LAW ON GENOCIDE

If we leave aside a few decisions handed down by the Extraordinary Courts M artial of
the Ottoman Empire in 1920 and dealing with ‘the massacres of Armenians carried
out with the goal of annihilating them’11 (at that time the notion of genocide had not
yet been fully developed), it is striking that, until the 1990s, only a few cases of geno­
cide were brought before national courts. Chief among them is Eichmann (decided
in 1961 by the District Court of Jerusalem and subsequently, in 1962, by the Israeli
Supreme Court). Eichmann was tried for ‘crimes against the Jewish people’, an offence
under Israeli law which incorporated all the elements of the definition of genocide
(and the Supreme Court of Israel held that ‘the crimes against the Jewish People’ cor­
responded to genocide, Eichmann, SC, at 287).

11 For instance, see in particular Ahmed Mithad Bey and others, at 147-53; Mehmed All Bey and others, at
159-65; Bahaeddin §akir and others, at 169-73.
 13 2 IN T E R N A T IO N A L C R IM IN A L LAW

By contrast, much headway has been made both at the level of prosecution and p un­
ishment of genocide by international criminal tribunals (which have prodded national
courts also to deal with this crime) and at the normative level.
Genocide as a crime of individuals began to be punished following the establish­
ment of the ICTY and the ICTR. Genocide having been provided for in the Statutes
of both Tribunals as well as the ICC (followed by provisions relating to the Special
Panels for East Timor and the Extraordinary Chambers for Cambodia),12the first two
courts have had the opportunity to try quite a few persons accused of this crime. They
have delivered im portant judgments on the matter: the ICTR, particularly in Akayesu
(§§204-28) and Kayishema and Ruzindana (§§41-9); the ICTY in Jelisic (§§78-83) and
Krstic (§§539-69).
After the establishment of the ICTY and the ICTR, some national courts began to
institute criminal proceedings against persons accused of serious crimes in the former
Yugoslavia. German courts have thus pronounced on some cases of genocide.13Trials on
genocide have also been conducted in other countries (for instance, in Ethiopia, where
the High Court tried former President Mengistu in absentia; see Mengistu and others.
At the norm-setting level, some major advances stand out. The major substantive
provisions of the Convention gradually turned into customary international law. In its
Advisory Opinion on Reservations to the Convention on Genocide, the ICJ held that ‘the
principles underlying the Convention are principles which are recognized by civilized
nations as binding on states, even without any conventional obligation’ (at 24). This view
was confirmed by the Court in Bosnia v. Serbia (§§161). It is notable that the UN SG took
the same view of the customary status of the Genocide Convention (or, more accurately,
of the substantive principles it lays down), a view that was endorsed implicitly by the UN
SC,14and explicitly by the ICTR in Akayesu (§495) and by the ICTY in Krstic (§541).

See ICTY Statute, Art. 4; ICTR Statute, Art. 2; Art. 4 of Regulation 2000/15, s. 4 (as amended by regu­
lation 2001/30) of the SPET; as well as Art. 4 of the law establishing the ECC.
13 See Jorgic, decided in 1997 by the Higher State Court (Oberlandsgericht) of Düsseldorf. The Court
found the defendant guilty of genocide and sentenced him to life imprisonment. The most significant part of
the judgment is that relating to mens rea. The Court held that the intent to destroy a group means destroy­
ing the group as a social unit in its specificity, uniqueness and feeling of belonging: the biological-physical
destruction of the group is not required’ (section III, para. 1). The Court’s findings about the factual and
psychological elements from which one can infer the existence of ‘intent’ are extremely interesting. The
judgment was upheld by the Federal High Court (Bundesgerichtshof) in 1999, followed by the Constitutional
Court in 2000. See also Sokolovic and Kusljic in 2001. On these cases see K. Ambos and S. Wirth, ‘Genocide
and War Crimes in the Former Yugoslavia Before German Criminal Courts’ in H. Fischer, C. Kress and
S. R. Lüder (eds), International and National Prosecution of Crimes Under International Law (Berlin: Arno
Spitz, 2001), 783-97, R. Rissing-van Saan, The German Federal Supreme Court and the Prosecution of
International Crimes Committed in the Former Yugoslavia, 2 JTCT (2005), 381-99.
14 See Report of the Secretary-General Pursuant to Para. 2 of Security Council Resolution 808 (1993)
UN Doc. S/25704, §45.
 G E N O C ID E 133

6.4 THE OBJECTIVE ELEM ENTS

Article II of the Genocide Convention, and the corresponding rule of customary law,
clearly defines the conduct that may am ount to genocide:
(i) killing members (hence more than one member) of what we could term a
‘protected group’, namely a national or ethnical, racial, or religious group;
(ii) causing serious bodily or mental harm to members of a ‘protected group’;
(iii) deliberately inflicting on the group conditions of life calculated to bring about
its physical destruction in whole or in part;
(iv) imposing measures intended to prevent birth within the group; or
(v) forcibly transferring children of the group to another group.

While the definition of the four classes of group is an intricate problem that requires
serious interpretative efforts (see infra, 6.6.1), the various classes of action falling
under genocide seem to be relatively clear. They were to a large extent spelled out in
Akayesu (TJ), as well as other judgments of the ICTR:
(i) as for killing members of the group, ‘killing’ must be interpreted as ‘m urder’, i.e.
voluntary or intentional killing;15
(ii) as for causing serious bodily or mental harm, these term s ‘do not necessarily
mean that the harm is permanent and irremediable’: Akayesu §§502-4; Gacumbitsi,
TJ, §291. As an ICTY TC put it in Krstic:
In line with the Akayesu Judgement [§502], the Trial Chamber states that serious harm need
not cause permanent and irremediable harm, but it must involve harm that goes beyond
temporary unhappiness, embarrassment or humiliation. It must be harm that results in a
grave and long-term disadvantage to a person’s ability to lead a normal and constructive life.
In subscribing to the above case-law, the Chamber holds that inhuman treatment, torture,
rape, sexual abuse and deportation are among the acts which may cause serious bodily or
mental injury (§513).
See also ICTY, Blagojevic and Jokic, TJ, §645. The harm may include acts of bodily or
mental torture, sexual violence, and persecution (.Rutaganda, TJ, §51).
(iii) with regard to deliberately inflicting on the group conditions of life
calculated to bring about its physical destruction, in Akayesu the TC held that
this expression includes among other things, ‘subjecting a group of people to a
subsistence diet, systematic expulsion from homes and the reduction of essential
medical services below m inim um requirem ent[s]’: (§§505-6), or the ‘deliberate
deprivation of resources indispensable for survival, such as food or medical serv­
ices’ (Kayishema and Ruzindana, §115); according to an ICTY TC in Brdanin,
‘also included is the creation of circum stances that would lead to a slow death,

15 Akayesu (§§500-1). See also Semanza (TJ), at §319) and Kayishema and Ruzindana (AJ), §151.
 134 IN T E R N A T IO N A L C R IM IN A L LAW

such as lack of proper housing, clothing and hygiene or excessive work or physical
exertion’ (§691).
(iv) as for ‘imposing measures intended to prevent births within the group’, in
Akayesu it was held that these measures could consist o f ‘sexual mutilation, the prac­
tice of sterilization, forced birth control [and the] separation of the sexes and prohib­
ition of marriages’ (§507); in addition, the measures at issue may be not only physical
but also mental (§508); they may include rape as an act directed to prevent births when
the woman raped refuses subsequently to procreate (§508); see also Rutaganda, TJ, §53
and Musema, TJ, §158.
(v) forcibly transferring children of the group to another group may embrace
threats or intimidation leading to the forcible transfer of children to another group
(Akayesu, §509).
Another interesting problem relating to actus reus is whether genocide may also
include the killing, with the required intent, of only one single member of a protected
group. In Akayesu the Trial Chamber, when dealing with the constituent elements of
genocide, held the view that there may be genocide even if one of the acts prohibited by
the relevant rules on this matter is committed ‘against one’ member of a group (§521).
Arguably, this broad interpretation is not consistent with the text of the norms on
genocide, which speak instead o f‘members of a group’ (see above).
It would seem that Article II does not cover the conduct currently termed in non­
technical language ‘ethnic cleansing’; that is the forcible expulsion of civilians belong­
ing to a particular group from an area, a village, or a town. (In the course of the drafting
of the Genocide Convention, Syria proposed an amendment designed to add a sixth
class of acts of genocide: ‘Imposing measures intended to oblige members of a group
to abandon their homes in order to escape the threat of subsequent ill-treatment’.
However, the draughtsmen rejected this proposal.)16
Some courts have indeed excluded the forced expulsion of persons belonging to a
particular ethnic, racial, or religious group from the notion of genocide.17 However,

16 For the Syrian proposal see UN Doc. A/C6/234.

17 See, for instance, Jelisic, judgment of 14 December 1999 (§§107-8). The Prosecution had alleged that
Jelisic had contributed to the campaign of ethnic cleansing in Brcko in eastern Bosnia and had, for a period,
acted as the principal executioner at the Luka camp ‘with the intent to destroy, in whole or in part, a racial,
ethnic or religious group’ (Jelisic, TC, oral ruling, 19 October 1999). The Prosecution asserted that the
accused demonstrated considerable authority, that he had received instructions to kill as many Muslims
as possible and that his genocidal intent could be shown by the accused’s own words, as was reported to the
judges by the witnesses. In this regard, they characterized Jelisic as ‘an effective and enthusiastic participant
in the genocidal campaign’ and noted, in addition, that the group targeted by Jelisic was significant, ‘not only
because it included all the dignitaries of the Bosnian Muslim community in the region but also because of
its size’. The Trial Chamber ruled, however, that Jelisic could not be found guilty of the crime of genocide.
Although he had pleaded guilty to both war crimes and crimes against humanity (§§24-58), with respect
to the crime of genocide the Trial Chamber issued the following pronouncement: ‘In conclusion, the acts of
Goran Jelisic are not the physical expression of an affirmed resolve to destroy in whole or in part a group as
such. All things considered, the Prosecutor has not established beyond all reasonable doubt that genocide
was committed in Brcko during the period covered by the indictment. Furthermore, the behaviour of the
accused appears to indicate that, although he obviously singled out Muslims, he killed arbitrarily rather
 G E N O C ID E 135

in other cases courts have asserted that that expulsion, under certain circumstances,
could be held to amount to genocide.18 Probably the better view is that upheld by the
German Constitutional Court in Jorgic, namely that ‘systematic expulsion can be a
m ethod of destruction and therefore an indication, though not the sole substantiation,
of an intention to destroy’ (at §24). (A similar view was propounded by an ICTY TC in
Krstic (at §§589-98).)
In Krstic an ICTY TC clarified the actus reus by defining the notion of the destruc­
tion of a group ‘in part’. The Prosecution had accused the defendant of genocide for
having planned and participated in the massacre in a limited locality (the area of
Srebrenica), of between 7,000 and 8,000 Bosnian Muslims, all of them men of military
age. The question arose of whether the ‘protected group’was constituted by the ‘Bosnian
Muslims of Srebrenica’ or instead by ‘Bosnian Muslims’. The Chamber answered the
query by noting that the group was that of Bosnian Muslims, and the Bosnian Muslims
of Srebrenica constituted ‘a part of the protected group’ under Article 4 of the ICTY
Statute (§560) (which was based on Article II of the Genocide Convention and was
held by the Chamber to be declaratory of customary international law: §§541-80). The
Chamber added that ‘the intent to eradicate a group within a limited geographical
area such as the region of a country or even a municipality’ could be characterized as
genocide (§589).19 As for the fact that the persons systematically killed at Srebrenica

than with the clear intention to destroy a group. The Trial Chamber therefore concludes that it has not been
proved beyond all reasonable doubt that the accused was motivated by the dolus specialis of the crime of
genocide. The benefit of the doubt must always go to the accused and, consequently, Goran Jelisic must be
found not guilty on this count5(§§107-8).
On ethnic cleansing it is also worth mentioning the decision delivered on 31 August 2001 by the Supreme
Court of Kosovo in Vuckovic: ‘Indeed, the essential characteristic of the criminal act of genocide is the
intended destruction of a national, ethnical, racial or religious group. However, the appealed verdict only
considered that the accused, forcefully expelling population from their houses in unbearable living condi­
tions, was ready to accept the consequence that the part or entire group of Albanian population of these
villages will be exterminated. Such motivation does not characterize the intent to destroy an ethnic group in
whole or in part. More generally, according to the Supreme Court, the exactions committed by the Milosevic
regime in 1999 cannot be qualified as criminal acts of genocide, since their purpose was not the destruction
of the Albanian ethnic group in whole or in part, but its forceful departure from Kosovo as a result of a sys­
tematic campaign of terror including murders, rapes, arsons and severe mistreatments5(at 2-3).
See also Kusljic (decision of the German Bundesgerichtshof of 21 February 2001), at 7-10.
18 In the confirmation of the second indictment of 16 November 1995 (pertaining to the fall of the UN
safe area of Srebrenica) against Radovan Karadzic and Ratko Mladic, for instance, Judge Riad expressly
characterized ethnic cleansing5as a form of genocide: Karadzic and Mladic, confirmation of indictment of
16 November 1995.
An ICTY Trial Chamber observed in the Karadzic and Mladic Rule 61 Decision that the character of the
acts in question may permit the inference of genocidal intent: Karadzic and Mladic, Rule 61 Decision of
11 July 1996, §94. See also Nikolic, Rule 61 Decision, ICTY Trial Chamber, §34.
However, a subsequent judgment of the Trial Chamber suggests a retreat from the Trial Chamber’s above-
mentioned and relatively expansive stance (see Jelisic, supra, n. 17).
19 It then pointed out the following: ‘the intent to destroy a group, even if only in part, means seeking to
destroy a distinct part of the group as opposed to an accumulation of isolated individuals within it. Although
the perpetrators of genocide need not seek to destroy the entire group protected by the Convention, they
must view the part of the group they wish to destroy as a distinct entity which must be eliminated as such.
A campaign resulting in the killings, in different places spread over a broad geographical area, of a finite
 13 6 I N T E R N A T IO N A L C R IM IN A L LAW

were ‘only men of m ilitary age’, the TC emphasized that, while these men were being
massacred, at the same time the rest of the Bosnian Muslim population was being for­
cibly transferred out of the area.20
The Chamber concluded that the killing of all the Bosnian Muslim men of military
age in Srebrenica accompanied by the intent to destroy in part the Bosnian Muslim
group within the meaning of Article 4 of the ICTY Statute must qualify as genocide.
Before m aking this ruling, the TC had also discussed the question of the extent to
which, while appraising whether or not genocide had been perpetrated in the case
at issue, it could take into account evidence or facts relating to the cultural or social
destruction of a group, as opposed to its physical or biological destruction.21

number of members of a protected group might not thus qualify as genocide, despite the high total number
of casualties, because it would not show an intent by the perpetrators to target the very existence of the group
as such. Conversely, the killing of all members of the part of a group located within a small geographical
area, although resulting in a lesser number of victims, would qualify as genocide if carried out with the
intent to destroy the part of the group as such located in this small geographical area. Indeed, the physical
destruction may target only a part of the geographically limited part of the larger group because the per­
petrators of the genocide regard the intended destruction as sufficient to annihilate the group as a distinct
entity in the geographic area at issue. In this regard, it is important to bear in mind the total context in which
the physical destruction is carried out (§590).
20 In this respect it stressed that ‘The Bosnian Serb forces could not have failed to know, by the time they
decided to kill all the men, that this selective destruction of the group would have a lasting impact upon the
entire group. Their death precluded any effective attempt by the Bosnian Muslims to recapture the terri­
tory. Furthermore, the Bosnian Serb forces had to be aware of the catastrophic impact that the disappear­
ance of two or three generations of men would have on the survival of a traditionally patriarchal society,
an impact the Chamber has previously described in detail. The Bosnian Serb forces knew, by the time
they decided to kill all of the military aged men, that the combination of those killings with the forcible
transfer of the women, children and elderly would inevitably result in the physical disappearance of the
Bosnian Muslim population at Srebrenica. Intent by the Bosnian Serb forces to target the Bosnian Muslims
of Srebrenica as a group is further evidenced by their destroying homes of Bosnian Muslims in Srebrenica
and Potocari and the principal mosque in Srebrenica soon after the attack. Finally, there is a strong indica­
tion of the intent to destroy the group as such in the concealment of the bodies in mass graves, which were
later dug up, the bodies mutilated and reburied in other mass graves located in even more remote areas,
thereby preventing any decent burial in accord with religious and ethnic customs and causing terrible dis­
tress to the mourning survivors, many of whom have been unable to come to a closure until the death of
their men is finally verified. The strategic location of the enclave, situated between two Serb territories, may
explain why the Bosnian Serb forces did not limit themselves to expelling the Bosnian Muslim population.
By killing all the military aged men, the Bosnian Serb forces effectively destroyed the community of the
Bosnian Muslims in Srebrenica as such and eliminated all likelihood that it could ever re-establish itself
on that territory’ (§§595-7).
21 On this point it set out the following interesting remarks (which it then applied in the ruling just
cited): ‘The Trial Chamber is aware that it must interpret the Convention with due regard for the principle of
nullum crimen sine lege. It therefore recognises that, despite recent developments, customary international
law limits the definition of genocide to those acts seeking the physical or biological destruction of all or part
of the group. Hence, an enterprise attacking only the cultural or sociological characteristics of a human
group in order to annihilate these elements which give to that group its own identity distinct from the rest of
the community would not fall under the definition of genocide. The Trial Chamber however points out that
where there is physical or biological destruction there are often simultaneous attacks on the cultural and
religious property and symbols of the targeted group as well, attacks which may legitimately be considered
as evidence of an intent to physically destroy the group. In this case, the Trial Chamber will thus take into
account as evidence of intent to destroy the group the deliberate destruction of mosques and houses belong­
ing to members of the group’ (§580).
 G E N O C ID E 137

6.5 THE SUBJECTIVE ELEMENTS

The mental requirement for genocide as a crime involving international criminal

liability is provided for in Article 11(1) of the Convention on Genocide (and in the
corresponding customary rule): the ‘intent to destroy, in whole or in part, a national,
ethnical, racial or religious group’. Genocide is a typical crime based on the ‘deperson­
alization of the victim’; that is a crime where the victim is not targeted on account of
his or her individual qualities or characteristics, but only because he or she is a m em­
ber of a group. As the German Federal Court of Justice rightly held in Jorgic in 1999,
the perpetrators of genocide do not target a person ‘in his capacity as an individual’;
they ‘do not see the victim as a hum an being but only as a member of the persecuted
group’ (at 401).22
This intent amounts to dolus specialis; that is, to an aggravated criminal intention,
required in addition to the criminal intent accompanying the underlying offence (kill­
ing; causing serious bodily or mental harm; inflicting conditions of life calculated to
physically destroy the group; imposing measures designed to prevent births within
the group; forcibly transferring children). It logically follows that other categories of
mental element are excluded: recklessness (or dolus eventualis) and gross negligence.
The ICTR TCs have contributed greatly to elucidating the subjective element of
genocide. In Akayesu, an ICTR TC held that the commission of genocide required ‘a
special intent or dolus specialis’. ‘Special intent’ is defined by the ICTR as ‘the specific
intention, required as a constitutive element of the crime, which demands that the
perpetrator clearly seeks to produce the act charged’ (§498). The TC added that intent
‘is a mental factor which is difficult, even impossible to determine. This is the reason
why, in the absence of a confession from the accused, his intent can be inferred from
a certain num ber of presumptions of fact’ (§523). It added that one can in particular
infer the special intent ‘from all acts or utterances of the accused, or from the general
context in which other culpable acts were perpetrated systematically against the same
group, regardless of whether such other acts were committed by the same perpetrator
or even by other perpetrators’.23

22 In the same case the German Constitutional Court held the following view: ‘The Higher State Court
and Federal Constitutional Court take the view that para. 220(a) of the StGB [the German Criminal Code]
protects the group. They have unanimously interpreted the intention of StGB para. 220a as meaning that
the destruction of the group as a social entity in its specificity and particularity and sense of togetherness,
or even geographically limited part of the group, need not extend to its physical and biological extermin­
ation [...] It is enough if the culprit takes upon himself the intent of the central controlling structure that
inevitably must be in place for the elements of the crime to be met, even if toward a part of the group [...] the
statutory definition of genocide defends a supra-individual object of legal protection, i.e. the social existence
of the group [...] The text of the law does not therefore compel the interpretation that the culprit’s intent must
be to exterminate physically at least a substantial number of the members of the group [...] the intent can
be deduced as a rule from the circumstances of an attack carried out under a structurally organized central
control on the group, of which the culprit is aware, and which he wills’ (§§19-22).
23 The interpretation given in Akayesu has to a very large extent been followed by the Trial Chambers
of the ICTR: in Kayishema and Ruzindana (§§87-118) as well as in Rutaganda (§§44-63) and in Musema,
where the Tribunal in particular considered the issues of complicity in and conspiracy to commit genocide
 138 IN T E R N A T IO N A L C R IM IN A L LAW

6 .6 PROBLEM ATICAL ASPECTS OF GENOCIDE

There are three issues concerning genocide that are at the same time intricate and con­
troversial, and which therefore deserve our attention: (i) how to identify the various
protected groups; (ii) whether acts of genocide always require an underlying geno-
cidal policy by a state or organized authority; (iii) how to discern genocidal intent.

6 .6 .1 H O W T O I D E N T I F Y T H E ‘P R O T E C T E D ’ G R O U P S

The major problems concerning the objective element of genocide relate to the notion
o f the group victim of the crime as well as the identification o f the four groups enu­
merated in the rule (national, ethnical, racial, religious). The former problem may be
framed as follows: what do the Convention and the corresponding customary rule
mean by group’? In other words, when can one state with certainty that one is faced
with a group protected by the Convention? The latter question, which is obviously
closely related to the former, is ‘By what standards or criteria can one identify each of
the four groups?’ Can one rely upon an objective test for each group? If so, where does
one find such a test?
Normally the various classes of groups are defined objectively, on account of some
alleged objective features each group exhibits. By national group is meant a multi­
tude of persons distinguished by their nationality or national origin (for instance,
the French citizens living abroad in a particular country, the US nationals of Irish
descent). Race is a notion whose scientific validity has been debunked by anthropolo­
gists; it must nevertheless be perforce interpreted and applied when used in a legal
provision. In the Genocide Convention race seems to embrace groups that share some
hereditary physical traits or features, such as the colour of skin. Ethnicity refers to
groups that share a language and cultural traditions. Religion is probably the least
controversial standard; it refers to groups sharing the same religion or set of spiritual
beliefs and faith, as well as modes of worship.
The case law of the ICTR and ICTY has contributed considerably to clarifying the
notion of group, moving from an objective to a subjective evaluation. The im port­
ance of Akayesu in particular needs to be stressed. In this case, an ICTR TC not only
emphasized that genocide is the most grave international crime or, as it put it, ‘the
crime of crimes’ (§16), but also, and more importantly, set out a definition o f‘group’. In
its view, this word, in the provisions on genocide, refers only to ‘stable groups’,
constituted in a permanent fashion, and membership of which is determined by birth, with
the exclusion of the more mobile’ groups which one joins through individual voluntary
commitment, such as political and economic groups. Therefore, a common criterion in the
four types of groups protected by the Genocide Convention is that membership in such

(§§884-941). (In three other cases concerning genocide the accused had pleaded guilty and therefore the
TC only dealt with sentencing: see Kambanda (sentence of 4 September 1998), Serushago (sentence of 5
February 1999), and Ruggiu (sentence of 1 June 2000).)
 G E N O C ID E 139

groups would seem to be normally not challengeable by its members, who belong to it
automatically, by birth, in a continuous and often irremediable manner (§511).

According to the TC, the groups protected against genocide should not be limited
to the four groups envisaged in the relevant rules, but—in order to respect the inten­
tion of the draughtsmen of the Genocide Convention, who clearly intended to protect
any identifiable group—should include any stable and permanent group’ (§516). This
proposition without further elaboration appears unconvincing, given that the fram ­
ers of the Convention, as clearly expressed in the text of that instrum ent, evinced an
intention to protect only the four groups explicitly indicated there. The Chamber then
propounded a definition of each of the four groups envisaged in the relevant rules. It
defined ‘national groups’ as a collection of people who are perceived to share a legal
bond of common citizenship, coupled with reciprocity of rights and duties’ (§512), an
‘ethnic group’ as ‘a group whose members share a common language or culture’ (§513),
a ‘racial group’ as a group ‘based on the hereditary physical traits often identified with
a geographical region, irrespective of linguistic, cultural, national or religious factors’
(§514), and a ‘religious group’ as a group ‘whose members share the same religion,
denomination or mode of worship’ (§515).
It should be noted that in the particular case of the genocide of Tutsis by Hutus in
Rwanda, the question of how to identify a protected group played a major role. Indeed,
these two groups shared language, religion, and culture, lived in the same areas, and
in addition there was a high rate of mixed marriages. The ICTR stressed that the two
terms of Tutsi and Hutus before colonization by the Germans (1885-1916) and then
by the Belgians (1916-1962) referred to individuals and not to groups, the distinction
being based on lineage rather than ethnicity (§81). (Furthermore, Tutsis were origin­
ally shepherds, whereas Hutus were farmer.) However, in 1931 Belgians introduced
a permanent distinction by dividing the population into three ethnic groups (Hutu,
Tutsi, and Twa), making it mandatory for each Rwandan to carry an identity card that
mentioned his or her ethnicity (§83). The TC concluded that thus in fact the members
of the various groups ended up considering themselves as distinct from members of
the other groups.24
It would thus seem that for the TC in Akayesu the question of whether or not a
multitude of persons made up a group protected by the rules against genocide was
primarily a question of fact: the court had to establish whether (i) those persons were
in fact treated as belonging to one of those protected groups; and in addition (ii) they
considered themselves as belonging to one of such groups.
One may find the same admixture of objective and subjective criteria in Kayishema
and Ruzindana. There an ICTR TC stated that

24 The TC noted that ‘in Rwanda, in 1994, the Tutsi constituted a group referred to as ‘ethnic’ in offi­
cial classifications. Thus, the identity cards at the time included a reference to ‘ubwoko’ in Kinyarwnda or
‘ethnie (ethnic group) in French which, depending on the case, referred to the designation Hutu or Tutsi,
for example [...] [In addition] all the Rwandan witnesses who appeared before it [the Trial Chamber] invari­
ably answered spontaneously and without hesitation the questions of the Prosecutor regarding their ethnic
identity’ (§702).
 140 I N T E R N A T IO N A L C R IM IN A L LAW

An ethnic group is one whose members share a common language and culture: or a group
which distinguishes itself, as such (self-identification); or a group identified as such by
others, including perpetrators of the crimes (identification by others) (§98).

In Rutaganda the ICTR pushed the subjective standard even further. It noted
that:

The concepts of national, ethnical, racial and religious groups have been researched exten­
sively and [...] at present, there are no generally and internationally accepted precise defini­
tions thereof. Each of these concepts must be assessed in the light of a particular political,
social and cultural context. Moreover, the Chamber notes that for the purposes of applying
the Genocide Convention, membership of a group is, in essence, a subjective rather than an
objective concept. The victim is perceived by the perpetrator of genocide as belonging to a
group slated for destruction. In some instances, the victim may perceive himself/herself as
belonging to the said group (§56).

Also, two ICTY TCs, as well as the UN International Commission of Inquiry on

Darfur, shared this subjective approach.25

6.6.2 W H E T H E R G E N O C I D E ALWAYS R E Q U I R E S A
G E N O C ID A L P O L I C Y O R P L A N

Various commentators have noted that acts of genocide by individuals or groups

always presuppose a policy or at least a collective activity of a state, an entity or a
group, policy or collective activity in which the individual perpetrators of genocide
participate by their conduct.26 Contrary to this view, the ICTY AC held in Jelisic that
‘the existence of a plan or policy is not a legal ingredient of the crime’, although ‘it may
facilitate proof of the crime’ (AJ, §48).27
I submit that both views do not construe the existing law correctly. Arguably a con­
textual element is not required by the customary and treaty rules on genocide for some
instances of genocide, whilst it is needed for other categories. At least with regard to
two categories of acts of genocide ((i) killing members of a protected group; (ii) causing
serious bodily or mental harm to members of a protected group), one or more indi­
viduals may engage in the crime of genocide without any general policy or collective
action being required for their being prosecuted and punished for that crime. One
or more individuals may, for example, kill a num ber of members of a religious group
with genocidal intent, even if no state authorities or collectivity persecute and intend
to destroy that group. Similarly, one or more persons may engage in rape or torture of

25 See Jelisic (§§70-1) and Krstic (§§556-7 and 559-60). In 2005 the UN International Commission of
Inquiry on Darfur shared this approach when discussing whether the so-called African tribes (essentially
consisting of sedentary farmers) in Darfur made up an ethnic group distinct from the so-called Arab tribes
(essentially consisting of nomadic shepherds), in spite of their sharing the same language (Arabic) and
religion (Muslim) and not distinguishing themselves from one another as far the colour of their skin was
concerned (§§498-501 and 508-12).
26 for instance, see C. Kress, ‘The Darfur Report and Genocidal Intent’, in 3 JICJ (2005), 562-78.
27 The ICTR AC had already set out this proposition in its AJ in Kayishema and Ruzindana (§138).
 G E N O C ID E 14 1

members of an ethnic or racial group with the intent of thereby destroying the group
in whole or in part. In other words, international rules do not require the existence of
either a widespread or systematic practice or a plan as a legal ingredient of the crime
of genocide.28 This conclusion is material at the procedural level, for it implies that the
Prosecutor in a national or international trial need not lead evidence on that practice
or contextual element. In reality, however, even genocidal acts belonging to one of the
two categories at issue are hardly conceivable as isolated or sporadic events. Normally
they are in fact part of a pattern of conduct tolerated, approved, or condoned by gov­
ernmental authorities. These circumstances remain nevertheless factual events, not
provided for or required by the relevant treaty and customary rules.
Instead, the other three categories of genocide perforce not only presuppose, but
necessarily take the shape of, some sort of collective or even organized action (I am
referring to (i) deliberately inflicting on a protected group or members thereof con­
ditions of life calculated to bring about its physical destruction in whole or in part;
(ii) imposing measures intended to prevent births w ithin a protected group; (iii) for­
cibly transferring children of a protected group to another group). Plainly, actions
such as deliberate deprivation of resources indispensable for the survival of m em ­
bers of a protected group, e.g. food or medical supplies, or such action as systematic
expulsion from home with a view to bringing about conditions of life leading to the
destruction of the group, constitute actions that are necessarily carried out on a large
scale and by a m ultitude of individuals in pursuance of a common plan, possibly with
the support or at least the acquiescence of the authorities. Similarly, such measures
designed to prevent births as prohibition of marriages, separation of the sexes, forced
birth control, sterilization, large-scale sexual mutilation, are all activities that only
state organs or other official authorities may undertake, or authorize to undertake, or
at least approve or condone.

6.6.3 H O W T O I D E N T I F Y G E N O C I D A L I N T E N T

The ICTR TCs have contributed greatly to elucidating the subjective element of geno­
cide. As noted, in Akayesu an ICTR TC held that intent ‘is a mental factor which is
difficult, even impossible to determine’ (§523).29
Indeed, normally to prove the existence of genocidal intent one has to infer such
intent from factual circumstances. Only seldom can one find documents or state­
ments by which one or more persons explicitly declare that they intend to destroy

28 The UN International Commission of Inquiry on Darfur held in its report that mass killing large-
scale rape as well as massive forcible expulsion of civilians from their homes, committed by or on behalf of
the Sudanese authorities in Darfur did not amount to genocide, for lack of genocidal intent attributable to
the Sudanese Government. Nevertheless, in its view single individuals participating in such crimes might
be found guilty of genocide by a court of law if such court were satisfied that the defendants had pursued a
genocidal intent (§§520-1). However, the Commission did not draw any distinction between mass murder
and rape, on the one side, and forcible expulsion, on the other.
29 The approach taken in Akayesu has to a very large extent been followed by the ICTR TCs: in Kayishema
andRuzindana (§§87-118) as well as in Rutaganda (§§44-63) and in Musema (§§884-941).
 142 I N T E R N A T IO N A L C R IM IN A L LAW

a whole group. An instance of such statements can be found in the minutes (drafted
by Eichmann) of the discussion held at Wannsee (Berlin) on 20 January 1942 to plan
the extermination of the European Jews,30 as well as in the speech Heinrich Himmler
(head of the SS) made on 4 October 1943 in Poznan to SS officers31 to the same effect.
In other instances utterances against a particular group expressing the intent to
destroy (or to contribute to destroy) the group, were not taken to express genocidal
intent proper. A case in point is Jelisic. An ICTY TC held that his repeated statements
against Muslims and the consequent criminal offences perpetrated by him against
many Muslims did not manifest genocidal intent but were expression of ‘a disturbed
personality’ (§§102-7). The AC took a different (and a more correct) view, ruling that
the accused had instead entertained genocidal intent (§§55-72), although it then oddly
declined to reverse the acquittal for genocide entered by the TC and remit the case for
further proceedings.
In Krstic an ICTY TC made a considerable contribution, in various respects, to
the definition of mens rea of genocide. The Prosecution, as noted above, accused the
defendant of genocide for having planned and participated in the massacre in a lim ­
ited locality (the area of Srebrenica), of between 7,000 and 8,000 Bosnian Muslims, all
of them men of military age. The following question then arose: was this intent present
in this case where only men of military age were systematically killed? The Chamber
answered the query in the affirmative. It emphasized that the rest of the Bosnian
Muslim population had been forcibly transferred out of the area, with the inevitable
result of the physical disappearance of the whole Muslim population of Srebrenica.32
The Chamber concluded that the intent to kill all the Bosnian Muslim men of military
age in Srebrenica evinced the intention to destroy in part the Bosnian Muslim group
and therefore must qualify as genocidal intent.
As pointed out above, the special intent under discussion is normally deduced from
the factual circumstances. Hence, in those cases where the actus reus is murder or
bodily or mental harm the question whether those acts were part of a plan or policy
or of widespread or systematic practice may eventually acquire importance from an
evidentiary viewpoint (although, as noted above, not as a legal ingredient of the
crime), as an element capable of proving (or confirming) that there was indeed geno­
cidal intent.
This is dear from what an ICTR TC held in some cases, for instance in Akayesu
and in Kayishema and Ruzindanda. In the former case the TC inferred the special
intent from the speeches by which the accused called, ‘more or less explicitly’, for the
commission of genocide (§729). It also deduced intent from the very high number
of deliberate and systematic atrocities committed against the Tutsis (§730) and the
numerous and systematic acts of rape and sexual violence against Tutsi women
(§§731-3). Also in Kayishema and Ruzindanda the TC inferred genocidal intent from

30 See the English translation online: www.h-net.org/-german/gtext/nazi/wanneng2.

31 He insisted in that speech on the ‘extermination [Ausrottung] of the Jewish people’ (German text and
English translation online: www.holocaust-history.org/himmler-poznan/speech-text).
32 See §§593-7, cited supra, in n. 20.
 G E N O C ID E 143

the high number of Tutsis killed (§531 and 533), the fact that they had been massacred
regardless of gender or age (§532), as well as the fact that the attacks had been carried
out in a consistent and methodical way (§§534-6 and 543). The utterances of the two
defendants were also taken into account (for instances, Tutsis had been called ‘cock­
roaches’, had been referred to as ‘d irt’ or ‘filth’ (§538); in particular, Ruzindana had
stated that babies whose mothers had been killed must not be spared ‘because those
attacking the country initially left as children’ (at §542).
Similarly, in Musema an ICTR TC inferred special intent to destroy Tutsis from
the numerous atrocities committed against them (§928), form large-scale attacks
launched against Tutsi civilians (§930)'and, more generally, from the widespread
and systematic perpetration of other criminal acts against members of the Tutsi
group’(§931) in which the defendant participated. These acts were accompanied by
humiliating utterances.33
W hen the objectively genocidal act is part of a whole pattern of conduct taking place
in the same state (or region or geographical area), or, a fortiori, of a policy planned or
pursued by the governmental authorities (or by the leading officials of an organized
political or military group), then it may become easier to deduce not only the intent34but
also lack of intent from the facts of the case. Thus, the UN Commission of Inquiry on
Darfur held that a range of acts or conducts by the Sudanese governmental authorities

33 According to the TC such humiliating utterances clearly indicated that the intention underlying each
specific act was to destroy the Tutsi group as a whole. The Chamber notes, for example, that during the rape
of Nyiramusugi Musema declared: “The pride of the Tutsis will end today.” In this context, the acts of rape
and sexual violence were an integral part of the plan conceived to destroy the Tutsi group. Such acts targeted
Tutsi women, in particular, and specifically contributed to their destruction and therefore that of the Tutsi
group as such. Witness N testified before the Chamber that Nyiramusugi, who was left for dead by those who
raped her, had indeed been killed in a way. Indeed, the Witness specified that “what they did to her is worse
than death”’(§933).
34 As the Hague Court of Appeal held in the van Anraat case with regard to the Iraqi genocide of Kurds
in 1987-8. The Court stated that ‘From a number of documents, including the aforementioned reports and
statements in the case file, it appears that the offences put forward in the charges refer to the air attacks that
were carried out partly during the so-called Anfal Campaign by or under the command of the perpetrators.
Moreover, they show that those attacks, however horrifying and shocking they were, formed part of a con­
siderably larger complex of many years of actions against the Kurds in the Northern Iraqi territory, which is
mainly inhabited by the Kurdish population. Apparently these actions involved the systematic destruction
of hundreds of Kurdish villages. Hundreds of thousands of Kurdish civilians were chased from their home
towns and deported to other places and tens of thousands of Kurds were killed. In one of his reports, Van
der Stoel described the policy that constituted the basis for the so-called Anfal Campaign, as a policy that
without a doubt had the characteristics of a genocidal design. In view of the said facts and circumstances,
the Court believes that the actions taken by the perpetrators, in any case even the ones that have not been
included in the charges, as outlined in the above, as to their nature at least produce strong indications that
the leaders of the Iraqi regime, also regarding the actions that have been put down in the charges, let them­
selves be guided by a genocidal intention with regards to at least a substantial part of the Kurdish population
group in (Northern) Iraq’ (§7). The Court however held that ‘Nevertheless, [...] a final judicial judgment
regarding the important as well as internationally significant question whether certain actions by certain
persons as mentioned in the charges should be designated as genocide, deserves a better motivated judgment
(which should be based on conclusive evidence) than the one on which the Court was able to establish its
observation’ (ibid.).
 144 INTERNATIONAL CRIMINAL LAW

committed in breach of international rules evinced that the intent to destroy an ethnic
group in whole or in part was lacking.35
If instead no policy or plan or widespread practice may be discerned, it may turn out
to be extremely difficult to prove the required intent. The Commission of Inquiry on
Darfur stated that the fact that no genocidal intent could be imputed to the Sudanese
authorities did not exclude that such special intent might be entertained by single
individual Sudanese servicemen or militias fighting on behalf of or together with the
Sudanese armed forces. To establish the existence of such intent in specific cases was,
according to the Commission, a task falling to a competent court of law (§§520-1).

6.7 GENOCIDE AND CRIMES

AGAINST HUMANITY

As emphasised above, large-scale massacres of ethnic or religious groups were first

criminalized as a subclass of the category of crimes against humanity. However, after
the adoption of the Genocide Convention of 1948 and the gradual transformation of
its main substantive provisions into customary international law, genocide became a
category of crimes per se, with its own specific actus reus and mens rea.
True, both categories share at least three elements: (i) they encompass very serious
offences that shock our sense of hum anity in that they constitute attacks on the most
fundamental aspects of hum an dignity; (ii) they do not constitute isolated events but
are instead always part of a larger context, either because they are large-scale and m as­
sive infringements of hum an dignity or because they are linked to a broader practice
of misconduct; and (iii) although they need not be perpetrated by state officials or by
officials of entities such as insurgents, they are usually carried out with the complicity,
connivance, or at least the toleration or acquiescence of the authorities.

jS The Commission scrutinized various elements and concluded that the attacks by Arab militias (under
governmental control) on villages inhabited by so-called African tribes did not disclose genocidal intent. As
it put it: the intent of the attackers was not to destroy an ethnic group as such, or part of the group. Instead,
the intention was to murder all those men they considered as rebels, as well as forcibly expel the whole popu­
lation so as to vacate the villages and prevent rebels from hiding among, or getting support from, the local
population (§514).The Commission went on to note that ‘Another element that tends to show the Sudanese
Government’s lack of genocidal intent can be seen in the fact that persons forcibly dislodged from their vil­
lages are collected in IDP camps. In other words, the populations surviving attacks on villages are not killed
outright, so as to eradicate the group; they are rather forced to abandon their homes and live together in areas
selected by the Government. While this attitude of the Sudanese Government may be held to be in breach of
international legal standards on human rights and international criminal law rules, it is not indicative of any
intent to annihilate the group. This is all the more true because the living conditions in those camps, although
open to strong criticism on many grounds, do not seem to be calculated to bring about the extinction of the
ethnic group to which the IDPs [Internally Displaced Persons] belong. Suffice it to note that the Government
of Sudan generally allows humanitarian organizations to help the population in camps by providing food,
clean water, medicines and logistical assistance (construction of hospitals, cooking facilities, latrines,
etc.)’ (§515).
 G E N O C ID E 145

However, the objective and subjective elements of the two crimes differ in many
respects (see also supra, 6.5). As for the objective element, the two crimes may
undoubtedly overlap to some extent: for instance, killing members of an ethnic or
religious group may as such fall under both categories; the same holds true for causing
serious bodily or mental harm to members of a racial or religious group, or even for
the other classes of protected group. However, crimes against hum anity have a broader
scope, for they may encompass acts that, as such, do not come within the purview
of genocide (for instance, imprisonment and torture)—unless they am ount to acts
inflicting on members of a group conditions of life calculated to bring about the phys­
ical destruction of the group. By the same token, there may be acts of genocide that are
not normally held (at least under the Statutes of the ICTY, ICTR, and the ICC) to fall
within the other category of crime (for instance, killing detained military personnel
belonging to a particular religious or racial group, by reason of their membership of
that group). Thus, from the viewpoint of their objective elements, the two categories
are normally ‘reciprocally special’, in that they form overlapping circles which never­
theless intersect only tangentially.
By contrast, from the perspective of the mens rea, the two categories do not overlap at
all. In the case of crimes against humanity, international law requires the intent to com­
mit the underlying offence plus knowledge of the widespread or systematic practice con­
stituting the general context of the offence. For genocide, what is required is instead the
special intent to destroy, in whole or in part, a particular group, in addition to the intent
to commit the underlying offence. From this viewpoint, the two categories are therefore
‘mutually exclusive’. They form two circles that do not intersect. The only exception is the
case where the underlying actus reus is the same, for instance, murder; in this case, the
intent to kill is required in both categories; nevertheless genocide remains an autono­
mous category, for it is only genocide that also requires the intent to destroy a group.
Similarly, it is only for crimes against humanity that knowledge of the widespread or
systematic practice is required. As for persecution, the intent of seriously discriminat­
ing against members of a particular group is shared by both crimes against humanity
and genocide. For persecution-type crimes against humanity, however, it is sufficient to
prove that the perpetrator intentionally carried out large-scale and severe deprivations
of the fundamental rights of a particular group, whereas for genocide it is necessary to
prove the intent to destroy a group, in whole or in part.36
I should add that, depending on the group targeted and the accompanying intent,
the same objective conduct may give rise to a combination of both genocide and crimes
against humanity. For instance, the Hutus’ massacres of Tutsis in Rwanda in 1994
amounted to genocide, whereas their simultaneous or concomitant killing of moderate
Hutus constituted a crime against humanity.

36 It should be noted that in Kayishema and Ruzindana the majority of the ICTR TC dismissed the charge
of crime against humanity by wrongly holding that it was already covered and indeed ‘completely absorbed’
by genocide (§§577-9); Judge Khan dissented.
 146 INTERNATIONAL CRIMINAL LAW

6.8 ARTICLE 6 OF THE ICC STATUTE AND

CUSTOMARY INTERNATIONAL LAW
Article 6 of the ICC Statute reproduces word for word Article II of the Genocide
Convention and the corresponding customary rule. In contrast, Article III of the
Convention (and the corresponding customary rule) on responsibility for forms of
participation in the crime other than perpetration, namely conspiracy, incitement,
attempt, and complicity, have not been taken up in the provision on genocide, either
because the notion has not been accepted by the Rome Diplomatic Conference (as
was the case with conspiracy, a concept that has not found the support of all the civil
law countries present at Rome), or because the relevant notion is laid down in general
terms (i.e. in terms applicable to other crimes as well) in other provisions of the ICC
Statute: this applies to incitement (at present envisaged in Article 25(3)(e)), attempt
(which is provided for in Article 25(3)(f)), and complicity (which is contemplated in
Article 25(3)(c) and (d)).
It follows that in at least one respect there is an inconsistency between customary
international law and the Rome Statute. The former prohibits and makes punishable
conspiracy to commit genocide’; that is, an inchoate crime consisting of the planning
and organizing of genocide not necessarily followed by the perpetration of the crime,
whereas Article 6 does not contain a similar prohibition.
It should be noted that in the process of drafting Article 6, in February 1997 it was
suggested in the Working Group of the Preparatory Committee that ‘the reference to
“intent to destroy in whole or in part [...] a group as such” was understood to refer
to the specific intention to destroy more than a small num ber of individuals who are
members of a group’.37 This suggestion was aptly assailed by two commentators, who
noted that nothing in the Genocide Convention could justify such a restrictive inter­
pretation and that, in addition, international practice belied this interpretation, for
‘successful counts or prosecutions of crimes against humanity, of which genocide is
a species, have involved relatively small numbers of victims’.38 It would seem that the
customary international rule, as codified in Article 6, does not require that the victims
of genocide be numerous. The only thing that can be clearly inferred from the rule is
that genocide cannot be held to occur when there is only one victim (see above 6.4).
However, as long as the other requisite elements are present, the killing or commis­
sion of the other enumerated offences against more than one person may amount to
genocide.
Finally, one should note a further view put forth with regard to the mens rea element
of genocide. According to the proponent of this view, the ICC Statute ‘appears to allow’
that ‘genocide may be committed with a lower level of mens rea’ than the very high
intent requirement mentioned above, for it ‘contemplates [in Article 28, on command

37 UN Doc. A/AAC.249/1997/L.5 Annex I, p. 3, n. 1.

38 L. Sadat Wexler and J. Paust, in 13(3) Nouvelles Etudes Penales (1998), at 5.
 GENOCIDE 147

responsibility] liability of commanders for genocide committed by their subordinates

even if they have no real knowledge of the crime’.39 It may be objected that this could
be true only with regard to the case where the superior knows that genocide is about
to be perpetrated, or is being committed, and deliberately refrains from forestalling
the crime or stopping it. Indeed in this case, according to a widespread opinion, the
superior may be equated to a co-perpetrator, or at least an aider and abettor (see infra,
11.4.2-4). Instead, one could not accuse a superior of genocide (as a co-perpetrator or
an accomplice) when the superior fails to punish the subordinates who have engaged
in genocide; or when, although he has information that should enable him to conclude
that genocide is being committed or may be committed, fails to act, in breach of his
supervisory obligations (see Article 28(l)(a) and (2)(a)). In these cases the superior
would be guilty of a different offence: intentional, reckless, or grossly negligent breach
of his supervisory duties. It follows that, with regard to such cases, it would not be
correct to assert that he should be held responsible for genocide, although with a sub­
jective element lower than specific intent.

39 W. A. Schabas, in O. Triffterer (ed.), Commentary, at margin 4.

 7
TORTURE AS A DISCRETE
CRIME, AND AGGRESSION

7.1 INTRODUCTION

In this and the following chapter we will discuss three classes of international crimes
that share two main features. First, they are normally not regarded as being included
in the so-called core crimes’ (a category comprising the most heinous offences: war
crimes, crimes against humanity, and genocide). Secondly, at least at present normally
they do not fall under the jurisdiction of international criminal tribunals or courts
(whereas at least aggression was provided for in the Nuremberg and Tokyo Statutes
and currently is prohibited at national level in some instances: see infra, 7.3.1, nn 5 and
6). It may prove useful briefly to dwell on this second distinguishing trait.
The reasons for the current exclusion of those three classes from international
jurisdiction differ for each class. In the case of aggression, the reason for this exclu­
sion is that the offence is too politically charged to be defined in sufficiently clear and
exhaustive criminal provisions and consequently entrusted to international inde­
pendent judicial bodies for adjudication.
As for torture as a discrete international crime (see infra, 7.2), the fact that to date
no international court or tribunal has been authorized to exercise its jurisdiction
over such crime may probably be explained by noting that torture as a crime con­
nected with armed conflicts (a war crime) or as large-scale or widespread crim inal
conduct (a crime against humanity) has been considered more in need of attention.
In contrast, in the opinion of states torture (i) practised by state officials or with their
connivance or complicity, and (ii) disconnected from a wider context (armed con­
flict, or widespread or systematic practice), is a m atter pertaining to their domestic
domain, where international intrusions are not welcome; hence it in principle falls
under their own crim inal jurisdiction. (It is common knowledge that despite the
major merits of the 1984 Convention against Torture, state prosecutors and courts
are still somewhat loath to prosecute and punish torturers allegedly com m itting
offences abroad against foreigners.)
Finally, many states still feel that on practical grounds terrorism is better investi­
gated and prosecuted at the state level by individual or joint enforcement and judicial
action. This view is strengthened by the feeling that such offence is still controversial
 TORTURE AS A DISCRETE CRIME, AND AGGRESSION 149

at the international level, for there is no agreement yet on the acceptance of what some
states deem to be a necessary exception to the crime (see infra, 8.1-2). As a conse­
quence, even in the only Statute of international tribunal envisaging such crime (that
of the Special Tribunal for Lebanon, STL) no international definition of the offence is
laid down, and instead a reference to Lebanese law is made.
Whatever the reasons for the present legal condition, the failure to extend inter­
national adjudication to these three classes of crimes is a matter of regret. Indeed,
entrusting an international judicial body with the task of pronouncing upon tor­
ture as a crime per se, aggression, or international terrorism would offer at least two
major advantages. First, it would significantly contribute, at a judicial level, to rein in
im punity for these odious crimes. Secondly, it would ensure—more and better than
any national court can do—full respect both for the principle of im partiality of courts
and for the fundamental rights of the accused.

7.2 TORTURE AS A DISCRETE CRIME

7.2.1 T O R T U R E AS A W A R C R I M E , A C R I M E A G A I N S T H U M A N I T Y ,
A N D A D IS C R E T E C R I M E

Torture is not only prohibited when it is part of a widespread or systematic practice

thus amounting to a crime against humanity. Torture is also proscribed when it is
done as a single act, outside any large-scale practice. In this case, if torture is perpe­
trated in time of armed conflict, and is connected to the conflict, it is a war crime. It
may also be a discrete crime under customary international law, whether committed
in time of peace or in time of armed conflict. There is an im portant difference between
these various categories.
In time of war or internal armed conflict a serviceman may incur criminal liability
for a war crime if he engages in torture against an enemy military or an enemy civil­
ian. Also a private individual not acting in an official capacity may perpetrate torture
in time of war; in this case, to qualify as a war crime, it must be committed against
(i) a member of the enemy belligerent army (or other lawful combatants) or (ii) a pro­
tected person who either has the nationality of the enemy or (particularly in the case
of internal armed conflict) is under the control of the adversary. In these two classes
of criminal conduct, to qualify as a war crime torture must be linked to the armed
conflict. Thus for instance, acts of torture performed by a civilian against another
civilian fall under the category of ‘ordinary crimes’ if committed outside the context
of, and without any nexus with, the armed conflict (for example, torture practised by
a civilian on a neighbour out of sadism, or on another civilian to take vengeance for
a previous personal wrong). In such war crimes it is not therefore necessary that a
150 INTERNATIONAL CRIMINAL LAW

state official be involved in the torture process, as was instead incorrectly held by the
German Federal Court of Justice in Sokolovic.1
Torture in time of internal or international armed conflict or in time of peace, to
amount to a crime against humanity, needs, among other things, to be part of a wide­
spread or systematic practice or attack on the population, as this is a general require­
ment of all crimes against humanity. Moreover, the accused must know that his acts
of torture form part of a widespread or systematic pattern of violence against civilians
(or, under customary international law, servicemen). Private persons may commit tor­
ture; again, there is no need for the participation of a state official in the specific act
of torture. It is, however, implicit in the very definition of this class of crimes that, in
addition to the specific case of torture being prosecuted, numerous acts of torture are
being or have been perpetrated without being punished by the authorities, or, in any
case, that acts of torture are part of a widespread or systematic pattern of violence. In
other words, there must be implicit approval or condonation by the authorities, or at
least they must have failed to take appropriate action to bring the culprits to book.
To put it differently, there must be at least some sort of ‘passive involvement’ of the
authorities. However, it is not required that a state official be involved in the torture
process, as was instead incorrectly maintained in Furundzija (§162).
Things are different with regard to torture as a discrete crime, i.e. not a crime against
humanity nor a war crime. Torture as a discrete crime may be perpetrated either in
time of peace or in time of armed conflict, as was rightly held in 2001 by the ICTY in
Kunarac and others (§§488-97) with a slight departure from the previous judgments of
the ICTR in Akayesu (§593) and the ICTY in Furundzija (§162). Under Article 1.1 of the
UN Torture Convention of 1984, the ‘pain or suffering’ that is a necessary ingredient of
torture must be inflicted ‘by or at the instigation of or with the consent or acquiescence
of a public official or other person acting in an official capacity. The need for this sort
of participation of a de jure or de facto state official stems from: (i) the fact that in this
case torture is punishable under international rules even when it constitutes a single or
sporadic episode; and (ii) the consequent necessity to distinguish between torture as a
common or ‘ordinary’ crime (for example, torture of a woman by her husband, or of a
young man by a sadist) and torture as an international crime covered by international
rules on human rights.
It would seem that, although they differ in many respects, the three categories of
torture (as a war crime, as a crime against humanity, as a discrete crime) share one
fundamental element: it is not exclusively required that the purpose of torture be the
extraction of a confession or admission of guilt from the victim. Instead, the aim of
torture as an international crime may be: (i) to obtain information or a confession; or
(ii) to punish, intimidate, or humiliate a person; or (iii) to coerce the victim or a third
person to do or omit something; or (iv) to discriminate, on any ground, against the
victim or a third person (see, among other cases, Furundzija (§162), which, however,

1 At 16-19. The Court required, for a war crime to exist, that among other things torture be practised ‘by
a state organ or with state approval’ (at 16). The Court therefore expressed misgivings about the notion of
torture laid down in Article 7(l)(f) of the ICC Statute (at 17-18).
 TORTURE AS A DISCRETE CRIME, AND AGGRESSION 151

referred specifically to torture as a crime against humanity). It should be noted that

the ICC ‘Element of Crimes’ do not require any specific purpose for torture as a crime
against hum anity (see Art. 7(l)(f)).

7.2.2 T H E E M E R G E N C E O F A C U S T O M A R Y R U L E O N T O R T U R E
AS A C R I M E P E R SE

The ban on torture perpetrated in the above circumstances has had a long evolution.
Significant contributions to this process, at the norm-setting level, were made by an
im portant Declaration passed by the UN GA (res. 3452(XXX) of 9 December 1975), by
the increasing importance of the 1984 UN Convention on Torture, by general treaties
on hum an rights and the judicial practice of the bodies responsible for their enforce­
ment, by national case law (in particular cases such as Pinochet), and by the judgments
of the ICTY in Furundzija (§146) and the European Court of Hum an Rights in Aksoy
(§62) and Selmouni (§§96-105). Suffice it to mention that in Filartiga a US court held
that ‘the torturer has become, like the pirate or the slave trader before him, hostis
humani generis, an enemy of all m ankind’ (at 980). And in 1998 in Furundzjia the
ICTY, after mentioning the hum an rights treaties and the resolutions of international
organizations prohibiting torture, stated that:
the existence of this corpus of general and treaty rules proscribing torture shows that the
international community, aware of the importance of outlawing this heinous phenomenon,
has decided to suppress any manifestation of torture by operating both at the interstate level
and at the level of individuals. No legal loopholes have been left (§146).

By now a general rule has evolved in the international community, (i) prohibiting
individuals from perpetrating torture, regardless of whether it is committed on a large
scale; and (ii) authorizing all states to prosecute and punish the alleged author of such
acts, irrespective of where the acts were perpetrated and the nationality of the perpet­
rator or the victim.

7.2.3 O B J E C T I V E A N D S U B J E C T I V E E L E M E N T S O F T O R T U R E A S A
C R I M E P E R SE

As for the conduct prohibited, one may safely rely upon the definition of torture laid
down in Article 1(1) of the 1984 UN Convention. As held by the ICTY,2 ‘there is now
general acceptance [in the world community] of the main elements contained’ in that
definition. The objective elements of torture may therefore be held to consist of: (i) ‘any
act by which severe pain or suffering, whether physical or mental, is [...] inflicted on
a person’; (ii) ‘such pain or suffering is inflicted by or at the instigation of or with the
consent or acquiescence of a public official or other person acting in an official cap­
acity’; and (iii) such pain or suffering does not arise ‘only from’ nor is it ‘inherent in or
incidental to lawful sanctions’.

2 See Delalic and others (§§455-74), in Furundzija (§§257), and Kunarac and others (§§483-97).
 15 2 INTERNATIONAL CRIMINAL LAW

The requirements for mens rea may be deduced from the very nature of torture, as
set out in the definition just referred to. It should be noted that Article 1 of the 1984
Convention, which has to a large extent become part of customary law, provides that
the infliction of pain or suffering must be ‘intentional’. It appears, therefore, that crim ­
inal intent 0dolus) is always required for torture to be an international crime. Other
less stringent subjective criteria (recklessness, culpable negligence) are not sufficient
(except where superior responsibility is at stake: see infra, 11.4).

7.3 TH E CRIM E OF AGGRESSION

7.3.1 T H E S U D D E N E M E R G E N C E O F T H E N O T I O N A N D ITS
I M M E D I A T E F A L L IN G I N T O L E T H A R G Y

Aggression was first considered as an international crime of individuals in 1945, when

the London Agreement was adopted.3 It was punished as such in 1946-7 by a num ber
o f international criminal tribunals.4 On 11 December 1946 the UN GA unanimously
adopted resolution 95(1), by which it ‘affirmed’ the ‘principles of international law rec­
ognized by the Charter of the Nuremberg Tribunal and the judgment of the Tribunal’.
Thus, all the states that at that stage were members of the UN eventually approved of
both the definition of crimes against peace and its application by the IMTs. At that
stage the crime fell into oblivion. In 1974, when the UN General Assembly adopted a
resolution containing the famous Definition of Aggression, the existence and punish­
ability of aggression as a crime was substantially glossed over.5

3 The London Agreement of 8 August 1945 establishing the IMT. Article 6(a) of the IMT Charter, annexed
to the Agreement, provided as follows: ‘The following acts, or any of them, are crimes coming within the
jurisdiction of the Tribunal for which there shall be individual responsibility: (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.’
4 The IMT dwelt at some length in its judgment on this category of crimes to prove that: (i) it had already
been established before 1945; and (ii) consequently punishing the Nuremberg defendants for having com­
mitted these crimes did not fall foul of the nullum crimen sine lege principle. The IMT went so far as to define
aggression as the ‘supreme international crime’ (at 186). Twelve defendants were found guilty on this count
and sentenced either to death or to long terms of imprisonment. Control Council Law no. 10 (of 20 December
1945) also provided for aggression in Art. II (l)(a). Subsequently the Tokyo International Military Tribunal
found 25 defendants guilty of aggression. Some of the US Military Tribunals established at Nuremberg also
pronounced on aggression (see Krauch and others (so-called IG Farben case), at 1081ff; Krupp and others,
1327ff; von Weizsäcker and others (so-called Ministries trial), 308fF; Wilhelm von Leeb and others (so-called
High Command trial), 462ff), as well as the French Tribunals that adjudicated the Röcklingen and others
case (at 1-7 and 404-12).
5 See res. 3314 (XXIX) of 14 December 1974. It was deliberately incomplete, for Article 4 provided that the
definition was not exhaustive and left to the SC a broad area of discretion, by stating that it was free to char­
acterize other acts as aggression under the Charter. Furthermore, the resolution did not specify that aggres­
sion could entail both state responsibility and individual criminal liability: in Article 5(2) of the Definition
 TORTURE AS A DISCRETE CRIME, AND AGGRESSION 153

Not surprisingly, since 1947 there have been no international trials for alleged
crimes of aggression, although undisputedly in many instances states have engaged in
acts of aggression in breach of Art. 2(4) of the UN Charter, and in a few cases the SC
has determined that such acts were committed by states.6 Only recently have alleged
cases of aggression been brought before some national courts,7 or have national
Prosecutors been requested to open investigations into alleged instances of aggres­
sion (such requests, however, have not been granted).8 It is a fact that, although many
national criminal codes provide for the crime of aggression,9 no criminal action at
the judicial level is being initiated. Similarly, although the Statute of the Iraqi High
Tribunal (IHT) grants jurisdiction over the crime of aggression against other Arab
countries,10so far nobody has been tried for such crime. All this is compounded by the

it simply provided that war of aggression is a crime against international law, adding that it ‘gives rise to
international responsibility’.
The definition propounded in the Draft Code of Crimes Against Peace and Security of Mankind, adopted
by the ILC in 1996, although it specifically dealt with criminal liability for aggression, was rather circular
and in fact did not provide any definition. Article 16 of the Draft Code provided that ‘An individual, who,
as leader or organizer, actively participates in or orders the planning, preparation, initiation or waging of
aggression committed by a state, shall be responsible for a crime of aggression’ (UN Doc. A/51/332)).
6 TheSC defined as ‘acts of aggression’ certain actions or raids by South Africa and Israel; see, for example,
resolution 573 of 4 October 1985, on Israeli attacks on PLO targets, and resolution 577 of 6 December 1985,
on South Africa’s attacks on Angola.
7 See R. v. Jones et ah, decided by the House of Lords on 29 March 2006. The appellants, who in 2003 had
unlawfully entered British or NATO military bases in the UK to prevent what they considered to be prepara­
tions for a war of aggression against Iraq, had been charged with or convicted of causing criminal damage
or aggravated trespass in British military bases. The House of Lords held that aggression is criminalized in
international law; however, absent any statutory enactment in the UK incorporating the international cus­
tomary law criminalizing aggression, the appellants were not entitled to rely upon that criminalization as a
defence for the illegality of their action. On this decision, see C. Villarino Villa, in 4 JICJ (2006), 866-77.
8 Pursuant to Article 80 of the German Criminal Code (which criminalizes ‘whoever prepares a war
of aggression’ in which Germany ‘is supposed to participate’) Germany’s Chief Federal Prosecutor
was requested in 1999 to initiate prosecution into the alleged aggression against the Federal Republic of
Yugoslavia (Serbia and Montenegro), in which German forces participated. He was then again requested to
act in 2003, on account of the use of force in Iraq by US and British forces (German officials were allegedly
responsible for allowing US bases in Germany to be used for activities related to military actions against
Iraq). In both cases the Prosecutor declined to initiate investigations. On this matter, see C. Kress, in 2 JICJ
(2004), 245-64.
9 For instance, see the following provisions of criminal codes: Article 80 of the German Criminal
Code (‘Whoever prepares a war of aggression ([envisaged in Article 26 para 1 of the Basic Law] in which
the Federal Republic of Germany is supposed to participate and thereby creates a danger of war for the
Federal Republic of Germany, shall be punished with imprisonment for life or for no less that ten years’); of
Bulgaria (Article 409), the Russian Federation (Article 353); Ukraine (Article 437); Armenia (Article 384);
Uzbekistan (Article 151); Tajikistan (Article 395); Latvia (§72), Moldova (Article 139), Macedonia (Article
415); Montenegro. See www.legislationonline.org.
See also Article 1 of the Iraqi Law no. 7 of 17 August 1958 (which criminalizes ‘Using the country’s armed
forces against the brotherly Arab countries threatening to use such forces or instigating foreign powers to
jeopardize its security or plotting to overthrow the existing regime or to interfere in their internal affairs
against its own interest, or spending money for plotting against them or giving refuge to the plotters against
them or attacking in international fields or through publications their heads of state’).
10 Article 14 para. 3 of the 2005 Law establishing the Tribunal confers on the Tribunal jurisdiction over
‘The abuse of position and the pursuit of policies that may lead to the threat of war or the use of the Iraqi
 154 INTERNATIONAL CRIMINAL LAW

fact that the Statute of the ICC, while envisaging the crime of aggression in Article 5,
stipulates that the Court shall exercise jurisdiction over such crime once a provision
defining it is adopted through an amendment of the Statute. It is striking that in the
negotiations leading to the adoption in 1998 of the Statute of the ICC, no agreement
was reached on the definition of aggression. Indeed, many African and Arab coun­
tries wanted to hold to the 1974 Definition, and even broaden it, while other states
including Germany proposed solutions better tailored to suit the needs of criminal
law. It would seem, however, that the main bone of contention was about the role to
be reserved to the UN SC. It was a matter of discussion whether its determinations
were binding upon the Court, whether it could thus stop the Court from prosecuting
alleged cases of aggression, or whether the Court should instead be free to make its
own findings, whatever the deliberations of the supreme UN body. As stated above, in
the event states agreed on Article 5(2) that in fact put off the m atter until an amend­
ment to the Statute is adopted by the Assembly of States Parties.
Why there was no international follow-up to the criminalization of aggression after
1947 while other crimes were spelled out in various conventions, is not difficult to
grasp. There are many reasons for that.
First, in 1945-7 it was easy to penalize the leaders of the vanquished states: the war
was over, it was patent that it had been initiated in blatant disregard of international
treaties; it was felt necessary to react to it not only by resorting to the normal means
used by victors (reparation of the wrongful acts; that is, payment by the vanquished
states of huge sums of money as war reparations), but also more dramatically, by
m aking criminally accountable the single individuals that in some way had willingly
participated in the planning and waging of the war. The written provisions of the
Tribunals’ Statutes criminalizing aggression were held to be sufficient, supplemented
by general notions of criminal law (intent or knowledge as subjective ingredients of
the crime).
Secondly, in 1945 the UN Charter established for the future a system of bans and
permissions in the area of use of military force: such force was prohibited in inter­
national relations (Article 2, §4); it was instead allowed if used or authorized by the
Security Council (Articles 42-9 and 53 of the UN Charter) or in self-defence (Article
51). However, while the ban was crystal clear, the permission was in some respects
fuzzy. In particular, it soon became controversial whether anticipatory self-defence
was allowed, and if so, under what conditions. True, the better interpretation of
Article 51 seems to be that self-defence is lawful when an armed attack by another
state is im minent (pre-emptive self-defence, as in the case of Israel in 1967, when the
international community did not object to Israel’s attack to forestall the impending
invasion by some Arab countries); instead, anticipatory self-defence is unlawful when
the attack is launched to prevent a possible future aggression (preventive self-defence,

armed forces against an Arab country, in accordance with Article 1 of Law 7 of 1958’. For the text of that
Article 1, see above, n. 9.
For a view different from that set out here, see C. Kress in 2 JICJ (2004), 347-52.
 TORTURE AS A DISCRETE CRIME, AND AGGRESSION 155

as in the case of the Israeli 1981 attack on Iraq to destroy the Osirak nuclear reactor, an
attack the Security Council condemned by res. 487/1981). The fact, however, remains
that this interpretation is not upheld by all members of the international community.
This looseness of the international legal regulation of the exception to the ban per­
force impinged upon the ban: obviously, when self-defence is allowed, the prohibition
on military force is not breached and therefore a state may not be termed aggressor.
This grey area of international legal regulation, calculated to give states much leeway
in practice, a fortiori rendered the criminalization of aggression problematic, given
that ICL, as any corpus of criminal law, requires legal precision in the interest of the
accused.
Thirdly, the Cold War prompted members of the two blocs to refrain from flesh­
ing out the rules on the crime of aggression, for fear that they might be used in the
ideological and political struggle between the blocs. Furthermore, there was a general
hesitancy by all major powers to elaborate upon aggression, so as to retain as much
latitude as possible in the application of the rules on self-defence. Thus, the definition
of aggression remained to a large extent in abeyance.
Now that there seems to be a broad interest in reviving the notion and spelling out
its legal contours; it maybe of some interest to draw attention to some of the ‘acquis’ of
the past experience, so as to build on them.

7.3.2 T H E N E E D T O D I S E N T A N G L E C R I M I N A L L I A B I L I T Y O F
I N D I V I D U A L S F R O M S T A T E R E S P O N S I B I L IT Y : T H E T W O
D IF F E R E N T LEGAL R EG IM E S

To my m ind it would be fallacious to hold the view that, since no general agreement has
been reached in the world community on a treaty definition of aggression, perpetra­
tors of this crime may not be prosecuted and punished. The ruling in R. v. Jones et al.
issued in 2006 by the House of Lords bears out this view. The House unanimously held
that aggression is criminalized under customary international law. Lord Bighman of
Cornhill,11 as well as Lord Hoflfmann (§59) and Lord Mance (at §99) explicitly stated
that, contrary to what the C ourt of Appeals had held in the same case (§§24-30), the

11 His Lordship wrote the following: ‘It was suggested, on behalf of the Crown, that the crime of aggres­
sion lacked the certainty of definition required of any criminal offence, particularly a crime of this gravity.
This submission was based on the requirement in Article 5(2) of the Rome Statute that the crime of aggres­
sion be the subject of definition before the international court exercised jurisdiction to try persons accused
of that offence. This was an argument which found some favour with the Court of Appeal (in para. 43 of
its judgment). I would not for my part accept it. It is true that some states parties to the Rome statute have
sought an extended and more specific definition of aggression. It is also true that there has been protracted
discussion of whether a finding of aggression against a state by the Security Council should be a necessary
pre-condition of the court’s exercise of jurisdiction to try a national of that state accused of committing the
crime. I do not, however, think that either of these points undermines the appellants’ essential proposition
that the core elements of the crime of aggression have been understood, at least since 1945, with sufficient
clarity to permit the lawful trial (and, on conviction, punishment) of those accused of this most serious
crime. It is unhistorical to suppose that the elements of the crime were clear in 1945 but have since become in
any way obscure’ (§19; emphasis added).
 156 INTERNATIONAL CRIMINAL LAW

crime of aggression does not lack the certainty of definition required for a criminal
offence.12 True, as pointed out above, this is an area where states deliberately want to
retain a broad margin of discretion. Nevertheless, a few points are clear.
The basic point is that the custom ary rules and the treaty provisions (Articles 2(4)
read in conjunction with Articles 42-9, 51, and 53 of the UN Charter) that prohibit
the unlawful use of force as an international wrongful act are different from and
broader than the customary rules that criminalize aggression. The two legal regimes
of responsibility for aggression are different not only because each notion is linked
to a different ‘prim ary’ or substantive international rule of customary law, but also
with regard to the pre-conditions of responsibility and the legal consequences of such
responsibility.
First of all, aggression as an international wrongful act of a state embraces any ser­
ious and large-scale breach of Article 2(4) not justified by Articles 51 and 53 (and the
corresponding customary rules). As such, aggression is subject to the legal regime
governing the so-called aggravated responsibility of states.13
In contrast, the following are breaches of the ban on the use of military force that,
while constituting international wrongful acts giving rise to state responsibility, do not
amount to state aggression: (i) breaching Article 2(4) of the UN Charter by violating
through the use of force the territory or the air space or the independence of a state by
means of acts that are sporadic or in any event not large-scale; (ii) engaging in an armed
conflict in violation of international treaties proscribing resort to armed violence; (iii)
using force under the authority of the resolution of an international body or on hum ani­
tarian grounds but in contravention of the UN Charter; or (iv) resorting to self-defence
in disregard of the conditions laid down in Article 51 of the UN Charter (for instance,
individual self-defence not followed by a report to the SC, or collective self-defence
initiated without a request by the victim state nor followed by such state’s consent). All
these acts would be illegal state conduct not amounting to aggression proper.
Secondly, international rules on aggression as a wrongful act of state only envisage
and ban aggression by a state against another state. This is because traditionally, inter­
national rules tend to govern interstate dealings.
As for criminal law, international practice, particularly as evinced by the views
set forth by states within the UN (in particular on the occasion of the adoption of

12 If the above remarks are correct, it would follow that the contrary view propounded by a US delegate
in 2001 would be erroneous from the legal viewpoint (see 95 AJIL (2001), 400-1). The US representative of
the US State Department noted that ‘the [1974] Definition neither restated existing customary international
law’ nor generated such law, due to lack of subsequent practice and opinio juris. After noting that there was
no ‘opinio juris generalis', the US representative pointed out that there was no practice: ‘Obviously, there has
been no concordant practice based on the [General Assembly resolution 3314 on the definition of aggres­
sion], Just look at the records of the Security Council. And if anyone still had any doubts, the controversy
about Resolution 3314 in our own discussions, has clearly demonstrated the absence of opiniojuris generalis’
(at 400). Arguably this view is immaterial to the existence of the customary rules at issue, for it is an isolated
statement not supported by similar views of other states.
13 On the notion of ‘aggravated state responsibility’, I take the liberty of referring to my International
Law, 2nd edn (Oxford: Oxford University Press, 2005), 262-77.
 TORTURE AS A DISCRETE CRIME, AND AGGRESSION 157

the 1970 Declaration on Friendly Relations14 and of the 1974 Definition), seems to
bear out the following propositions. First, customary rules have evolved to the effect
that only serious and large-scale instances of use of force (not legitimized by the UN
Charter as collective enforcement or collective or individual self-defence) may be
regarded as amounting to international crimes involving the criminal liability of
those who planned, organized, and masterminded aggression. For example, it would
seem difficult to deny that the attack by Iraq on Kuwait in 1990 was not only an inter­
state breach of Article 2(4) of the UN Charter, not justified by self-defence, and thus
amounted to an aggression involving the responsibility of the state, but also consti­
tuted an international crime of aggression.
Secondly, ICL rules that prohibit and criminalize aggression also penalize aggres­
sive acts by non-state entities (such as terrorist armed groups, organized insurgents,
liberation movements, and the like) against a state. Since this body of law is geared to
penalizing individuals’ misconduct, one cannot see what would stand in the way of
extending criminal liability for aggression to individuals who do not belong to, nor
act on behalf of, a state. If the purpose of the relevant international rules is to protect
the world community from serious breaches of the peace, one fails to see why indi­
viduals operating for non-state entities should be immune from criminal liability for
aggressive conduct.
Thirdly, an additional subjective element is required by international criminal rules
for aggression, which instead is not envisaged for aggression as an international state
delinquency (see below).
There is another difference between the two classes of responsibility. Under the UN
Charter the UN SC is empowered to determine whether a state or non-state entity has
engaged in aggression, and also to adopt all the necessary measures to counter such
aggression. It can also adopt or authorize sanctions against either the delinquent state
or non-state entity, or against individuals participating in the aggression. The SC thus
enjoys considerable latitude in this matter. However, being a political body, its deter­
minations may not amount to a judicial finding of the criminal liability of individuals
for the crime of aggression. It follows that a decision of the SC condemning actions by
a state as aggression may have no direct legal effect on courts empowered to adjudicate
crimes of aggression. Courts are free to make any finding in this matter regardless of
what is decided by the SC.
It is thus clear that one of the merits of the distinction between two different regimes
of responsibility lies in, among other things, enabling courts that try persons accused
of aggression legitimately to embrace a judicial approach which may differ from the
political stand taken by international political bodies such as the UN SC. There may
be cases where one of those bodies does not consider that aggression has materialized,
while a national or international court may take a contrary position and consequently
find individuals criminally responsible for aggression. It remains nonetheless true,

14 GA res. 2625(XXV). Principle I (2) states that ‘A war of aggression constitutes a crime against peace, for
which there is responsibility under international law.’
 15« INTERNATIONAL CRIMINAL LAW

that when the SC concludes that in a particular instance acts committed by a state
am ount to aggression as an international wrongful act, it may sometimes prove eas­
ier for a national or international court to find that aggression as a crime was perpe­
trated and, therefore, to pronounce on the issue of whether the individuals involved
are criminally liable. For courts, pronouncements of the SC constitute im portant ele­
ments that may count, along with relevant evidence, for their making judicial findings
on criminal liability for the conduct at issue.

7.3.3 O B J E C T I V E A N D S U B J E C T I V E E L E M E N T S O F T H E C R I M E

(A) Objective Elements

Generally speaking, customary international law essentially prohibits some trad­
itional forms of aggression as either international wrongful acts or criminal acts,
or both. These instances of aggression, constituting the core of the notion normally
valid for both categories of responsibility, are basically those envisaged in terms in the
1974 Definition,15 and confirmed, at least in part, by the ICJ in Nicaragua16 (although
admittedly solely with respect to state responsibility).
Customary international law appears to consider as an international crime: the
planning, or organizing, or preparing, or participating in the first use of armed force
by a state or a non-state organization or other organized entity against the territor­
ial integrity and political independence of another state in contravention of the UN
Charter, provided the acts of aggression concerned are large-scale and produce ser­
ious consequences. It follows that single attacks that, albeit very serious in nature, are
limited in scope and time (such as, for example, that of Israel on Iraq in 1981) may

15 The objective element of aggression as an international crime may comprise various instances, if
they exhibit the necessary character of massiveness. Mention can be made of some instances, substantially
based on the 1974 Definition: 1. The invasion of or the attack on the territory of a state, or any military
occupation, however temporary, resulting from such invasion or attack, or any annexation by the use of
force of the territory or part of the territory of a state. 2. Bombardment, or use of any weapon or lethal
device, by the armed forces of a state or a non-state entity, against the territory of another state (as long
as such bombardment or use of weapons is not isolated or sporadic). 3. Blockade of the ports or coasts of
a state by the armed forces of another state or a non-state entity. 4. Large-scale attack on the land, sea, or
air forces, or marine and air fleets of a state. 5. The massive use of the armed forces of a state or a non-state
entity, which are within the territory of another state with the agreement of the receiving state, in blatant
contravention of the conditions provided for in the agreement and the customary rules on the use offeree
6. The sending by or on behalf of a state or a non-state entity of armed bands, groups, irregulars, or mer­
cenaries, which carry out acts of armed force against a state of such gravity as to amount to the acts listed
above, or its substantial involvement therein.
16 In addressing the element of aggression defined in Article 3(g) of the Definition, whereby aggression
includes the case where a state ‘sends or is substantially involved in sending into another state armed bands
with the task of engaging in armed acts against the latter state of such gravity that they would normally be
seen as aggression’, the Court held that ‘This description [...] may be taken to reflect customary international
law. The Court sees no reason to deny that, in customary law, the prohibition of armed attacks may apply
to the sending by a state of armed bands to the territory of another state, if such an operation, because of its
scale and effects, would have been classified as an armed attack rather than as a mere frontier incident had it
been carried out by regular armed forces’ (§195).
 TORTURE AS A DISCRETE CRIME, AND AGGRESSION 159

constitute blatant breaches of the ban on the use of force and consequently give rise to
the aggravated responsibility of the attacking state; nevertheless, they do not amount
to crimes. Instead, a massive attack such as that of 11 September 2001 against the
USA, while it is not a breach of the aforementioned ban (which only concerns states),
amounts to an international crime (more precisely, to both a crime against hum anity
and a crime of aggression), thus involving the criminal liability of its authors.
It is clear from the intrinsic features of aggression that such crime (i) is never perpe­
trated by single individuals acting severally; instead, it always results from some sort
of collective action ofaplurality of persons-, (ii) is an offence attributable to political and
military leaders and other senior state officials (or leading organs o f a non-state entity);
that is, those who mastermind, plan, or organize the crime. Instead, it may not involve
the personal criminal liability of low-level perpetrators (for instance, it would seem
difficult to charge with aggression the pilots carrying out air raids in foreign terri­
tory in execution of an aggressive plan, unless of course those pilots were fully aware
of the illegality and criminal nature of the acts). It follows that normally the mode of
responsibility for aggression is participation in a joint criminal enterprise to plan or
wage aggression.

(B) Subjective Elements

The crime also requires criminal intent. It must be shown that the perpetrator intended
to participate in planning or waging aggression, was aware of the scope, significance,
and consequences of the action taken and substantially contributed to ‘shaping’ or
‘influencing’ the planning or waging of aggression. A leader or high-ranking military
officer or senior state officials or leading private (for instance, an industrialist) may
also bear responsibility if he has knowledge of other leaders’ plans and willingly pur­
sues the criminal purpose of furthering the aggressive aims. International case law on
this matter is clear and consistent.17

17 See Goring and others, 279-80 (Goring), 282-4 (Hess), 285-6 (von Ribbentropp), 288-9 (Keitel), 291
(Kaltenbrunner), 294-5 (Rosenberg), 296 (Frank), 299-300 (Frick), 302 (Streicher), 304-5 (Funk), 307-10
(Schacht), 310-11 (Donitz), 315-16 (Raeder), 317-18 (von Schirach), 320 (Sauckel), 322-4 (Jodi), 325-7 (von
Papen), 328-30 (Seyss-Inquart), 330-1 (Speer), 333-6 (von Neurath), 336-7 (Fritzsche), 338-9 (Bormann).
In Krupp and others another US Military Tribunal noted that ‘the defendants were private citizens and
non-combatants [they were industrialists]’. The Tribunal went on to emphasize that ‘None of them had any
voice in the policies that led their nation into aggressive war; nor were any of them privies to that policy.
None had any control over the conduct of the war or over any of the armed forces; nor were any of them
parties to the plans pursuant to which the wars were waged and so far as appears, none of them had any
knowledge of such plans’ (488).
In Krauch and others (1G Farben case), a US Military Tribunal sitting at Nuremberg held that: ‘If the
defendants [senior staff or managers of the German company I. G. Farben specializing in synthetic rub­
ber, gasoline, nitrogen, and light metals, as well as explosives], or any of them, are to be held guilty under
either count one [planning, preparation, initiation, and waging of wars of aggression] or five [formulation
and execution of a common plan or conspiracy to commit crimes against peace] or both on the ground that
they participated in the planning, preparation, and initiation of wars of aggression or invasions, it must be
shown that they were parties to the plan or conspiracy, or, knowing of the plan, furthered its purpose and
objective by participating in the preparation for aggressive war’ (1108). The Court concluded that none of
the defendants were guilty of the crimes set forth in counts one and five (at 1128; see also 1124-7). The Court
 i6o INTERNATIONAL CRIMINAL LAW

As convincingly argued by a com m entator18aggression requires in addition a special

intent; that is, the will to achieve territorial gains, or to obtain economic advantages, or
deliberately to interfere with the internal affairs of the victim state (for instance, by top­
pling its government or bringing about a change in its political regime or ideological
leanings or in its international political alignment). It would seem that the standard
by which to evaluate whether an individual harbours that special intent can be found
in the General Treaty of Paris for the Renunciation of War (or Kellogg-Briand Pact) of
27 August 1928, which banned war as ‘an instrum ent of national policy’. In short, any
unlawful large-scale attack against a state intentionally carried out as an instrum ent
of national policy (hence, to acquire territory; or coerce the victim state to change its
government or its political regime, or else its domestic or foreign policy; or to appro­
priate assets belonging to the victim state) amounts to aggression as a criminal act.
The above considerations bear out that, as in the case of genocide (see above, 6.1-2),
the notion of aggression is split into two separate concepts, one valid for wrongful
acts of states (where no special intent would be required, for the purpose of banning
armed attacks amounting to a breach of Article 2.4 of the UN Charter), the other for

concluded that the accused lacked the required mens rea (at 1306). In his Concurring Opinion Judge Hebert
insisted on the need for knowledge and criminal intent for criminal liability for aggressive wars to arise.
He stated: ‘We are thus brought to the central issue of the charges insofar as the aggressive wars charges are
concerned. Acts of substantial participation by certain defendants are established by overwhelming proof.
The only real issue of fact is whether it was accompanied by the state of mind requisite in law to establish
individual and personal guilt. Does the evidence in this case establish beyond reasonable doubt that the acts
of the defendants in preparing Germany for war were done with knowledge of Hitler’s aggressive aims and
with the criminal purpose of furthering such aims?’ (1217).
In von Leeb and others (so-called High Command case) a US Military Tribunal held that 'There first must
be actual knowledge that an aggressive war is intended and that if launched it will be an aggressive war.
But mere knowledge is not sufficient to make participation even by high-ranking military officers in the
war criminal. It requires in addition that the possessor of such knowledge, after he acquires it shall be in a
position to shape or influence the policy that brings about its initiation or its continuance after initiation,
either by furthering, or by hindering or preventing it. If he then does the former, he becomes criminally
responsible; if he does the latter to the extent of his ability, then his action shows the lack of criminal intent
with respect to such policy (68). The Tribunal then noted the following: ‘The acts of commanders and staff
officers below the policy level, in planning campaigns, preparing means for carrying them out, moving
against a country on orders and fighting a war after it has been instituted, do not constitute the planning,
preparation, initiation and waging of war or the initiation of invasion that international law denounces
as criminal’ (490-1). The Tribunal also noted that ‘mere knowledge is not sufficient to make participation
even by high-ranking military officers in the war criminal. It requires in addition that the possessor of such
knowledge, after he acquires it, shall be in a position to shape or influence the policy that brings about its
initiation or its continuance after initiation, either by furthering, or by hindering or preventing it’ (488). ‘It
is not a person’s rank or status, but his power to shape or influence the policy of his state, which is the rele­
vant issue for determining his criminality under the charge of crimes against peace’ (489). A US Military
Tribunal took up the notion that it is necessary to show that a culprit has the power to shape or influence the
policy of an aggressor state, in Weizsäcker and others (Ministries case), at 425.
See also Araki and others (Tokyo trial), at 456-7, as well as the Röchling case (T„ at 4,7,10; ST, at 406-8).
18 S. Glaser, ‘Quelques remarques sur la definition de 1’agression en droit international penal’, in
Festschrift für Th. Rittler (Aalen: Verlag Scientia, 1957), 388-93; idem., ‘Culpabilite en droit international
penal’, 99 HR 1960-1, 504-5. Glaser’s views are taken up by G. Werle, Principles of International Criminal
Law (The Hague: Asser Press, 2005), 395.
 TORTURE AS A DISCRETE CRIME, AND AGGRESSION l6 l

individuals’ criminal offences (where instead the requisite subjective element of crime
includes special intent).

7.3.4 W H E T H E R C O N S P I R A C Y T O W A G E
A G G R E S S IO N IS C R I M I N A L I Z E D

The Statutes of both the Nuremberg IMT and the Tokyo Tribunal provided that, in
addition to aggression (planning, preparation, initiation, or waging of a war of aggres­
sion), also participation in a conspiracy to wage such a war was criminalized. The
indictments in both cases charged aggression as well as conspiracy to wage aggression
as a separate charge.
The Nuremberg IMT merged conspiracy with planning a war of aggression (at
225-6). It acquitted some defendants (e.g. Funk and Speer) of conspiracy (because
they had not participated in the early stages of the planning of aggression) and in the
event found no defendant guilty solely of conspiracy.
The Tokyo Tribunal tended instead to envisage the two charges as separate, and
indeed found one defendant (foreign minister Shigemitsu) guilty of waging a war of
aggression but acquitted him of conspiracy, whereas it held another defendant (ambas­
sador Shiratori) guilty of conspiracy but acquitted him of aggressive war.
In spite of the different attitudes taken by the two Tribunals and the lack of any fol­
low-up in subsequent case law, it would seem that conspiracy to wage a war of aggres­
sion may be regarded as a separate crime in ICL. Aggression is such a devastating
crime, with serious knock-on consequences for peace and the whole international
community, that it seems warranted to infer from the present system of ICL the
criminalization of the early stages of preparation of the crime, when more persons
get together and agree to put in place the necessary measures to engage in a war of
aggression. It is also notable that there is a parallel prohibition in the field of state
responsibility: that, laid down in Article 2(4) of the UN Charter, relating to the mere
threat of force. If such threat has been proscribed in interstate dealings so as to quench
any attempt or preliminary steps toward the actual use of force, it is only natural for
ICL to also criminalize the ‘preliminaries’ to the crime of aggression; that is, the get­
ting together of leaders and their agreeing to engage in aggression.
However, this inchoate crime (that is, prelim inary offence that has not been com­
pleted and has not yet caused any harm (see infra, 10.3 and 10.6.1)) is only crim inal­
ized per se if it is not followed up by the actual undertaking of aggression. If this
happens, aggression as a crime ‘absorbs’ the crime of conspiracy.
 8
TERRORISM AS AN
INTERNATIONAL CRIME

8.1 A CURRENT MISCONCEPTION: THE

ALLEGED LACK OF A GENERALLY AGREED
DEFINITION OF TERRORISM

Interminable polemical arguments have been exchanged between states since the 1970s
over what should be meant by terrorism. The bone of contention is twofold: could ‘free­
dom fighters’ engaged in national liberation movements be classified as terrorists?1

1 Three different positions of states and other authorities maybe identified, positions that do not necessar­
ily exclude one another, and in some instances overlap.
The first is that of states stubbornly insisting on any act by peoples or organizations engaged in wars of
self-determination being exempt from the label of terrorism (even when they engage in attacks against civil­
ians). These states, however, do not clarify what law would govern such acts or whether, and more simply, these
acts should be held to be authorized under international law. This stand was taken, for instance, by Pakistan
in 2002 when acceding to the 1997 Convention for the Suppression of Terrorist Bombing. The Convention
excludes from its scope activities of armed forces, including freedom fighters, in armed conflict, keeping such
activities subject to the legal regulation of international humanitarian law. Pakistan entered a reservation that
can be held to be at least ambiguous. A very similar position is taken by other states, which purport to exclude
the application of anti-terrorist conventions to armed conflict, without, however, clarifying whether the use
of force by freedom fighters against civilians in such conflicts must be covered by international humanitarian
law. This stand was taken by Egypt, Jordan, and Syria in the reservation they made in 2003-2005 when ratify­
ing, or acceding to, the Convention for the Suppression of the Financing of Terrorism.
The second position is that of states or authorities which hold that, while any act performed by freedom
fighters in wars of national liberation is not covered at all by the body of international law on terrorism, it
remains nevertheless governed by the international humanitarian law of armed conflict. It would seem that
this view was implicitly taken by the Secretary-General of the Arab League, Mr A. Moussa. On at least two
occasions, he clearly asserted that the legitimacy of the Palestinian struggle for self-determination did not
imply that innocent civilians (be they Palestinian or Israeli) might be attacked. By this he clearly meant
to say that Palestinians legitimately fighting in the occupied territories against the foreign Occupant were
not allowed by international law deliberately to attack civilians. Similarly the member states of the Islamic
Conference participating in the UN negotiations for the elaboration of a Comprehensive Convention on
Terrorism have proposed a draft provision encapsulating the famous exception to the notion of terrorism.
However, this time the proposal spells out the hitherto ambiguous formula used by Arab and Islamic coun­
tries. It is now specified that actions undertaken in the course of an armed conflict ‘including in situations
of foreign occupation’ are not covered by the Convention, hence may not be classified as ‘terrorist acts’.
Nevertheless—and here comes the novelty—it is now added that those actions remain covered by other rules
of international law (in particular, humanitarian law). It logically follows that, if such actions are contrary
 TERRORISM AS AN INTERNATIONAL CRIME 16 3

Should the working out of international rules on terrorism be made contingent upon
delving into the root causes of this phenomenon? Many states have asserted that as long
as no agreement is reached on these two contentious issues, no consent could evolve on
the very notion of terrorism either.
As a consequence, treaty rules laying down a comprehensive definition have not yet
been agreed upon. However, over the years, under the strong pressure of public opin­
ion and also in order to come to grips with the spreading of terrorism everywhere,
in fact widespread consensus on a generally acceptable definition of terrorism has
evolved in the world community, so much so that the contention can be made—based
on the arguments I shall set forth below—that indeed a customary rule on the object­
ive and subjective elements of the crime of international terrorism in time of peace has
evolved. The requisite practice (usus) lies in, or results from, the converging adoption
of national laws, the handing down of judgments by national courts, the passing of
UN GA resolutions, as well as the ratification of international conventions by a great
num ber of states (such ratifications evincing the attitude of states on the matter). In
contrast, disagreement continues to exist on a possible exception to such definition:
whether to exempt in time o f armed conflict from the scope of the definition acts that,
although objectively and subjectively falling within the definition’s purview, accord­
ing to a number of states are nevertheless legitimized in law by their being performed
by ‘freedom fighters’ engaged in liberation wars.

to those rules, their authors may be prosecuted under other relevant rules of international law. Translated
into ‘contemporary’ terms, this means that, for instance, Palestinians’ deliberate attacks on Israeli civilians
in the West Bank (occupied territory), while they could not be termed terrorist acts, would amount to war
crimes, in particular to ‘crimes the primary purpose of which is to spread terror among the civilian popula­
tion’; their perpetrators would be liable to be punished under national and international law for such crimes.
If this is so, it becomes clear that now the intent of Islamic states is simply to remove the label of ‘terrorism’
from any action of ‘freedom fighters’ contrary to international law. The fact remains however, that even
those states now concede—or, at least, it would seem so—that the authors of those actions may be prosecuted
and punished for their criminal conduct. The diplomatic contention then boils down to an essentially ideo­
logical dispute over how to further term an act that is undisputedly criminal: as a terrorist act or as a war
crime (intended to spread terror)? This difference in ideology and social psychology is not, however, the end
of the matter. For, classifying an act as terrorist may trigger the use by the relevant national police of a set of
investigative powers normally not authorized for any ordinary crime or for any war crime. It follows that,
if agreement emerges on assigning acts performed by freedom fighters in armed conflict to the regulation
of international humanitarian law alone, the whole range of investigative powers and consequent measures
accruing to enforcement agencies under domestic law may no longer be applied with regard to them.
A third, middle-of-the-road position, has also emerged, which combines the application of international
rules on terrorism with international humanitarian law. This view is enshrined in the UN Convention for
the Suppression of the Financing of Terrorism and is shared by 150 out of the 153 current parties to the
Convention. The same view is laid down in Canadian legislation on terrorism and has also been put forward
by some Italian courts, as well as the Israeli foreign minister. It would seem plausible to contend that this
stand is shared bytheUNSG.The supporters of this position hold that attacks by freedom fighters and other
combatants in armed conflict, if directed at military personnel and objectives in keeping with international
humanitarian law, are lawful and may not be termed terrorism. If instead they target civilians, they amount
to terrorist acts (not, therefore, to war crimes) if their purpose is to terrorize civilians. Thus the conduct of
hostilities is not left to the exclusive legal dominion of international humanitarian law. Principles and rules
on terrorism reach out to armed conflict, in that they apply to acts that are not consonant with international
humanitarian law.
 i6 4 INTERNATIONAL CRIMINAL LAW

It would appear that, generally speaking, the question of investigating the histor­
ical, social, and economic causes of terrorism has instead been put on the backburner,
although very recently the UN SG has again drawn attention to the need to address
conditions conducive to exploitation by terrorists’.2

8.2 FACTORS P O IN T IN G TO A GENERALLY

AGREED D E F IN IT IO N OF TERRORISM IN
TIM E OF PEACE

Many factors are indicative of the formation of substantial consensus on a definition

of terrorism in time of peace.
First, the Conventions on terrorism adopted by the Arab League, the Organization
of African Union, and the Conference of Islamic States, while providing in terms for
the aforementioned exception, nevertheless lay down a definition that is to a large
extent in line with that enshrined in other international instrum ents.3

2 See his Report to the General Assembly of 27 April 2006 (A/60/825), Uniting against Terrorism:
Recommendations for a Global Counter-Terrorism Strategy, at §§20-37.
3 Article 1(2) of the Arab Convention for the Suppression of Terrorism, of 22 April 1998, defines terror­
ism as ‘Any act or threat of violence, whatever its motives or purposes, that occurs in the advancement of an
individual or collective criminal agenda and seeking to sow panic among people, causing fear by harming
them, or placing their lives, liberty or security in danger, or seeking to cause damage to the environment or
to public or private installations or property, or to [sic] occupying to seizing them, or seeking to jeopardize a
natural resources [sic]’(text online: www.al-bab.com/ arab/docs/league/terrorism98.htm).
Article 1(2) of the Convention of the Organization of the Islamic Conference on Combating International
Terrorism, of 1 July 1999, provides that ‘“Terrorism” means any act of violence or threat thereof notwith­
standing its motives or intentions perpetrated to carry out an individual or collective criminal plan with the
aim of terrorising people or threatening to harm them or imperilling their lives, honour, freedoms, security
or rights or exposing the environment or any facility or public or private property to hazards or occupying
or seizing them, or endangering a national resource, or international facilities, or threatening the stabil­
ity, territorial integrity, political unity or sovereignty of independent States’ (text online: www.oic-un.org/
26icfm/c.html).
Article 1(3) of the OAU Convention on the Prevention and Combating of Terrorism, of 14 July 1999 pro­
vides that: ‘“Terrorist act” means:
(a) any act which is a violation of the criminal laws of a State Party and which may endanger the life,
physical integrity or freedom of, or cause serious injury or death to, any person, any number or group
of persons or causes or may cause damage to public or private property, natural resources, environ­
mental or cultural heritage ands is calculated or intended to:
(i) intimidate, put in fear, force, coerce or induce any government, body, institution, the general
public or any segment thereof, to do or abstain from doing any act, or to adopt or abandon a par­
ticular standpoint, or to act according to certain principles; or
(ii) disrupt any public service, the delivery of any essential service to the public or to create a public
emergency; or
(iii) create general insurrection in a State.
(b) any promotion, sponsoring, contribution to, command, aid, incitement, encouragement, attempt,
threat, conspiracy, organizing, or procurement or any person, with the intent to commit any act
referred to in paragraph (a)(i) to (iii).’
 TERRORISM AS AN INTERNATIONAL CRIME 16 5

Secondly, both the 1999 UN Convention for the Suppression of the Financing of
Terrorism4 and various UN General Assembly resolutions contain a similar notion,5
which is also shared in the Draft Comprehensive Convention on Terrorism that is still
being negotiated.6
Thirdly, most national laws,7 as well as national case law take the same approach.8
The elements of this definition on which there is general consent are as follows:
terrorism consists of (i) acts normally criminalized under any national penal system,
or assistance in the commission of such acts whenever they are performed in time
of peace; those acts must be (ii) intended to provoke a state of terror in the popula­
tion or to coerce a state or an international organization to take some sort of action;
and finally (iii) are politically or ideologically motivated; that is, are not based on the
pursuit of private ends.
These are the rough elements of a generally accepted definition. Let us consider how
they can be translated into a rigorous articulation within international law. Thereafter,
it will be appropriate briefly to look at the contentious exception.

4 Article 2 (1) (b) provides that terrorism is ‘Any [...] act intended to cause death or serious bodily injury
to a civilian, or to any other person not taking an active part in the hostilities in a situation of armed conflict,
when the purpose of such act, by its nature or context, is to intimidate a population, or to compel a govern­
ment or an international organization to do or to abstain from doing an act’. So far, 160 states have become
parties to the Convention.
5 Since 1994 the UN General Assembly has adopted resolutions including the following proposition:
‘Criminal acts intended or calculated to provoke a state of terror in the general public, a group of persons or
particular persons for political purposes are in any circumstance unjustifiable, whatever the considerations
of a political, philosophical, ideological, racial, ethnic, religious or any other nature that may be invoked to
justify them.’ See §3 of the Declaration on Measures to Eliminate International Terrorism, annexed to res.
49/10 adopted on 9 December 1994; §2 of the subsequent resolutions 50/53 (11 December 1995), 51/110 (17
December 1996), 52/165 (15 December 1997), 53/108 (8 December 1998), 54/110 (9 December 1999), 55/158
(12 December 2000), 56/88 (12 December 2001), 57/27 (19 November 2002), 58/81 (9 December 2003), 69/46
(16 December 2004.
6 See Article 2, in Report of the Ad Hoc Committee established by General Assembly Resolution 51/210 of
17 December 1996, Sixth Session (28 January-1 February 2002), A/57/37, Annex II (at 6).
7 For instance see the US Iran and Libya Sanction Act of 1996 (Public Law 104-72, 5 August 1996); the
US Antiterrorism and Effective Death Penalty Act of 1996; the UK Terrorism Act 2000, Section 1; Article
83.01(1) of the Canadian Criminal Code (which defines terrorism as a criminal offence that is commit­
ted ‘(A) in whole or impart for a political, religious or ideological purpose, objective or cause, and (B) in
whole or in part with the intention of intimidating the public, or a segment of the public, with regard to its
security, including its economic security, or compelling a person, a government or a domestic or an inter­
national organization to do or to refrain from doing any act, whether the public or the person, government
or organization is inside or outside Canada’. See also Article 15 of the Italian law (law-decree 27 July 2005,
no. 144, passed as law on 31 July 2005, as law no. 155), which adds Article 270 sexies to the provisions of the
Italian Criminal Code.
8 See for instance the decision of the Supreme Court of Canada in Suresh (online at www.scc-csc.gc.ca),
where the Court held that the definition of terrorism laid down in Art. 2(l)(b) of the UN Convention for the
Suppression of Financing of Terrorism ‘catches the essence of what the world understands by “terrorism” ’
(§98; see also §93).
 16 6 INTERNATIONAL CRIMINAL LAW

8.3 THE INGREDIENTS OF INTERNATIONAL

TERRORISM AS A DISCRETE INTERNATIONAL
CRIME IN TIME OF PEACE
(i) The Objective Element
A first element of international terrorism (as distinguished from, i.e. not necessarily coin­
ciding with, terrorism under national legislation) relates to conduct. The terrorist act con­
sists of conduct that is already criminalized under any national body of criminal law:
murder, mass killing, serious bodily harm, kidnapping, bombing, hijacking, and so on.
This conduct may, however, be, in some exceptional instances, lawful per se: for instance,
financing of an organization. It becomes criminal if the conduct has the requisite connec­
tion to terrorism, for example if the organization to which money is provided or chan­
nelled, or on whose behalf it is collected, is terrorist in nature. In that case, the character
of the organization taints the otherwise lawful action with criminality.
Furthermore, the conduct must be transnational in nature; that is, not limited to
the territory of one state with no foreign elements or links whatsoever (in which case it
would exclusively fall under the domestic criminal system of that state).9
As for the victims of crim inal conduct, they may embrace both private individuals
or the civilian population at large and also state officials, including members of state
enforcement agencies.

(ii) The Subjective Element

A second distinguishing trait of terrorism is the purpose of terrorist acts. A num ­
ber of international instrum ents and national laws provide that terrorists pursue the
objective of either spreading terror among the population or compelling a govern­
ment or an international organization to perform or abstain from performing an act.101
Other instrum ents also envisage a third possible objective: to destabilize or destroy
the structure of a country.11
One can understand that, both for descriptive purposes and also in order to cover
the whole range of possible crim inal actions, these treaties, laws or other legal instru­
ments enumerate a wide set of terrorist aims. In addition, expressly contemplating
various alternative purposes pursued by terrorists may prove useful to prosecutors

9 The transnational nature of international terrorism is pithily caught in Article 3 of the Convention for
the Suppression of the Financing of Terrorism (‘This Convention shall not apply where the offence is com­
mitted within a single State, the alleged offender is a national of that State and is present in the territory of
that State and no other State has a basis [...] to exercise jurisdiction [...]’).
10 See, for instance §3 of SC res. 1566 (2004) adopted on 4 October 2004; Art. 83.01(1) (B) of the Canadian
Criminal Code.
11 See, for instance, Article 1(2) of the 1999 Convention of the Organization of the Islamic Conference
on Combating International Terrorism; Article 1 of the EU Framework Decision on Combating Terrorism
(which refers to the aim o f‘seriously destabilising or destroying the fundamental political, constitutional,
economic or social structures of a country or an international organization’).
 TERRORISM AS AN INTERNATIONAL CRIME 167

and other enforcement agencies when the demands of terrorist groups are not clear or
are not made with regard to a specific terrorist attack. In these cases, in order to classify
the conduct as terrorist, it may suffice to determine that at least the immediate aim of
terrorists was to spread panic among the population. This indeed may greatly facilitate
the action of prosecutors in applying national laws against terrorism. However, close
scrutiny and legal logic demonstrate that in fact the primary goal of terrorists is always
that of coercing a public (or private) institution to take a certain course of action. The
spreading of deep fear or anxiety is only a means for compelling a government or
another institution to do (or not to do) something; it is never an end in itself. Also, the
destabilization of the political structure of a state is a means of m aking the incumbent
government take a certain course of action. To be sure, in some instances the terror­
ists’ goal is not set forth in so many words either before or after the terrorist action.
For instance, the 11 September attack on the Twin Towers and the Pentagon was not
accompanied by specific demands of the terrorist organization that had planned the
attack. Yet, even in these cases the murder, bombing, kidnapping is not made for its
own sake; it is instrum ental in inducing a public or private authority to do or refrain
from doing something. The 11 September attack was clearly intended to prom pt the
US government to change its overall policy in the Middle East, in particular by pulling
out its m ilitary forces there and reversing its policy vis-a-vis Israel.
Hence it can be said that ultimately terrorism always pursues one prim ary and
essential purpose, that of coercing a public authority (a government or an international
organization) or a transnational private organization (for instance, a m ultinational
corporation) to take (or refrain from taking) a specific action or a certain policy. This
is the hallm ark of any terrorist action.
The purpose in question can be attained through two possible modalities; first, by
spreading fear or anxiety among civilians (for instance, by blowing up a theatre, kid­
napping civilians, planting a bomb in a train, in a bus, or in a public place such as a
school, a museum, a bank). Clearly, the aim of terrorists is to induce the scared popu­
lation to put pressure on the government authorities. Secondly, the purpose may be
achieved by engaging in criminal conduct against a public institution (e.g. blowing
up, or threatening to blow up, the premises of Parliament, the M inistry of Defence,
or a foreign embassy) or else against a leading personality of a public or private body
(for instance, the head of government, a foreign ambassador, the president of a m ulti­
national corporation, and so on).
Another element unique to terrorism regards motive. The criminal conduct is not
taken for a personal end (for instance, gain, revenge, or personal hatred). It is based
on political, ideological, or religious motivations. Motive is im portant because it
serves to differentiate terrorism as a manifestation of collective criminality from crim ­
inal offences (murder, kidnapping, and so on) that are instead indicative of individ­
ual criminality. Terrorist acts are normally performed by groups or organizations, or
by individuals acting on their behalf or somehow linked to them. A terrorist act, for
instance the blowing up of a disco, may surely be performed by a single individual not
belonging to any group or organization. However, that act is terrorist if the agent was
 i68 INTERNATIONAL CRIMINAL LAW

moved by a collective set of ideas or tenets (a political platform, an ideology or a body

of religious principles) thereby subjectively identifying himself with a group or organ­
ization intent on performing similar acts. It is this factor that transforms the m urder­
ous action of an individual into a terrorist act.
Let us now translate the above into rigorous legal language. For terrorism to
materialize, two subjective elements (mens rea) are required: first, the subjective
element (intent) proper to any underlying criminal offence: the requisite psychological
element of murder, wounding, kidnapping, hi-jacking, and so on (dolus generalis); sec­
ond, the specific intent of compelling a public or a prominent private authority to take,
or refrain from taking, an action (dolus specialis).
Motive in criminal law is normally immaterial (‘an actor’s ultimate reason for acting
may not bear on his liability’12), although it is sometimes taken into account under some
specific conditions in a few national legal systems.13 Motive exceptionally becomes
relevant here: as noted above, criminal conduct must be inspired by non-personal
inducements. Hence, if it is proved that a criminal action (for instance, blowing up
a building) has been motivated by non-ideological or non-political or non-religious
considerations, the act can no longer be defined as international terrorism, although
it may of course fall under a broader notion of terrorism upheld in the state where
the act has been accomplished. This for instance holds true for cases similar to the
McVeigh case, a criminal action that lacked, however, the transnational element proper
to international terrorism. I am referring to Timothy McVeigh’s blowing up in 1995 of
a public building in Oklahoma City, with the consequent death of 168 persons; report­
edly that action was carried out in revenge for the killing, by the FBI, of members of
a religious sect at Waco, Texas. Similarly, if bandits break into a bank, kill some cli­
ents, and take others hostage for the purpose of escaping unharm ed with the loot, this
action cannot be classified as terrorism, although the killing and hostage-taking are
also intended to spark terror among civilians and compel the authorities to do or not
to do something. Here the ideological or political motive is lacking. Consequently, the
offence is one of armed robbery aggravated by murder and hostage-taking, not terror­
ism. Let us take another example, namely the episode at the Los Angeles International
airport (where on 4 July 2002 an Egyptian fired at and killed some tourists who were
about to take a plane bound for Israel, and was eventually shot down by enforcement
officers). To determine whether this was a terrorist act or simply multiple murder, one

12 G. P. Fletcher, Rethinking Criminal Law, at 452.

13 For instance, according to the penal provisions applying in Italy before 1981, voluntary murder of
either a female spouse, daughter or sister (guilty of having unlawful sexual intercourse with another person)
or of their sexual partner, perpetrated in an outburst of anger caused by the offence to the perpetrator’s hon­
our or to the ‘family honour’, was punished with a much lighter penalty (3 to 7 years’ imprisonment) than
any ordinary murder (not less than 20 years’ imprisonment). See Article 587 of the 1930 Italian Criminal
Code, repealed by the Italian law no. 442 of 5 August 1981.
At present, political motives are taken into account for the purpose of defining a crime as political and con­
sequently attributing jurisdiction over such crime to Italian courts. Under Article 8 of the Italian Criminal
Code a national ora foreigner committing a crime abroad may be brought to trial before an Italian court if,
among other things, the crime was ‘determined, in whole or in part, by political motives’.
 TERRORISM AS AN INTERNATIONAL CRIME 169

ought to inquire into the possible motives of the killer (in that case motives could have
been inferred from his life, his possible statements, his criminal record, any links he
might have had with terrorist groups, and so on).
Of course, motive by itself may not suffice for the classification of a criminal act as
terrorist. To clarify this point I shall give an example (although it again relates to ter­
rorist groups that were not involved in transnational terrorism, it may nevertheless
be useful for illustration purposes). In the 1970s some terrorist groups in Italy and
Germany (respectively, the Red Brigades and the Rote Armee Fraktion) carried out
armed robberies against banks to replenish the organization’s funds. Here the motive
of the criminal act was not personal (to acquire a private gain), but collective (to boost
the organization’s cash). Yet the action was not terrorist in nature, but an ordinary
criminal offence, because another crucial element proper to terrorism was lacking (the
purpose of compelling through criminal conduct an authority to take a certain stand).
However, this conclusion does not exclude that individual national criminal systems
may consider that, since the aforementioned acts were performed to support a terror­
ist organization, the crimes involved must be characterized as terrorist at least for such
purposes as jurisdiction, the use of special investigative methods, and so on.
The legal relevance of motive for determining whether one is faced with a terrorist
offence does undoubtedly pose serious problems for any prosecutorial agency or crim ­
inal court. It may admittedly prove hard to find the reasons that inspired the agent,
and to disentangle the specific basis for his action from the intricacies of his possible
motivations. In particular, it may be laborious to establish whether he acted out of
political, ideological, or religious motivations. In addition to this factual difficulty, it
may also be hard to decide in a particular instance whether a set of ideas or aspirations
make up a political credo, an ideology, or a religion. One easy way out could consist
of ascertaining whether the agent only acted out of strictly personal reasons, in which
case one could rule out that his acts be termed terrorist.

8.4 SPECIFIC SUB-CATEGORIES OF

INTERNATIONAL TERRORISM AS A
DISCRETE INTERNATIONAL CRIME

At the time when ideological clashes mired the international discussion on terrorism,
preventing the achievement of general consensus on the matter, in order to break the
deadlock states opted for the passing of international conventions on specific categories
of conduct. They thus agreed upon a string of conventions through which they imposed
on contracting parties the obligation to make punishable and to prosecute in their
domestic legal orders certain classes of actions. These actions were defined in each
convention by indicating the principal outward elements of the offence. The conven­
tions refrained from terming the conduct terrorist, nor did they point to the purpose
 170 INTERNATIONAL CRIMINAL LAW

of the conduct or motive of the perpetrators. Instead, they confined themselves to set­
ting out the objective elements of prohibited conduct.
This applies to (i) acts that, whether or not they are offences under national law, may
or do jeopardize the safety of aircraft, or of persons or property therein, or which jeop­
ardize good order and discipline aboard;14 (ii) the unlawful taking control, by force or
threat thereof or by any other form of intimidation, of an aircraft in flight;15 (iii) acts
of violence against persons on board an aircraft in flight or against the aircraft;16
(iv) murder and other violent acts against internationally protected persons or their
official premises, private accommodation, or means of transport;17 (v) unlawful pos­
session, use, transfer, or theft of nuclear material as well as threat to use it;18 (vi) taking
control of a ship by force or threat thereof or any other form of intimidation or acts
of violence against persons aboard or against the ship;19 (vii) taking control over a
fixed platform by force or threat thereof or any other form of intimidation, or acts
of violence against persons on board or against the platform;20 (viii) acts of violence
against persons at an airport serving international civil aviation or against the facil­
ities of the airport;21 (ix) the manufacture, or the movement into or out of a territory, of
unmarked plastic explosives;22 (x) the delivery, placing, discharging, or detonation of
explosive or other lethal device in a place of public use, a state or government facility,
a public transportation system, or an infrastructure facility.23
Other Conventions, instead, besides setting out the objective elements of criminal
conduct, also place emphasis on the purpose pursued by the perpetrators. This holds
true for the 1979 Montreal Convention against the Taking of Hostages, as well as the
1999 Convention for the Suppression of the Financing of Terrorism. Both legal instru­
ments characterize the terrorist actions they deal with as intended to compel a state or
an international organization to do or to abstain from doing any act; in addition, the
latter Convention contemplates the purpose of intim idating a population.24

14 Art. 1(b) of the 1963 Tokyo Convention on Offences and Certain Other Acts Committed on Board
Aircraft.
15 Article 1 (a) of the 1970 Hague Convention for the Suppression of Unlawful Seizure of Aircraft.
16 Article 1(1) of the 1971 Montreal Convention for the Suppression of Unlawful Acts against the Safety
of Civil Aviation.
17 Article 2(1) of the 1973 Convention on the Prevention and Punishment of Crimes against Internationally
Protected Persons, Including Diplomatic Agents.
18 Article 7 of the 1979 Vienna Convention on the Physical Protection of Nuclear Material.
19 Article 3(1) of the 1988 Rome Convention for the Suppression of Unlawful Acts against the Safety of
Maritime Navigation.
20 Article 2 of the 1988 Rome Protocol for the Suppression of Unlawful Acts against the Safety of Fixed
Platforms Located on the Continental Shelf.
21 Article II of the 1988 Montreal Protocol for the Suppression of Unlawful Acts of Violence at Airports
Serving International Civil Aviation.
22 Articles II and III of the 1991 Montreal Convention on the Marking of Plastic Explosives for the
Purpose of Detection.
23 Article 2(1) of the 1998 International Convention for the Suppression of Terrorist Bombings.
24 Article 1(1) of the Convention on the Taking of Hostages provides that Any person who seizes or
detains or threatens to kill, to injure or to continue to detain another person (hereinafter referred to a the
“hostage”) in order to compel a third party, namely a State, an international intergovernmental organization,
 TERRORISM AS AN INTERNATIONAL CRIME 171

It is warranted to contend that for the whole range of aforementioned conduct the
hallmarks of international terrorism as a discrete crime in time of peace, outlined
above, were considered implicit in the banning of such conduct. Indeed, the prim ary
purpose of those Conventions was to put a stop to terrorist conduct belonging to each
category of action banned by the Conventions and increasingly ubiquitous when those
legal texts were drafted.
Nevertheless, as the classes of action prohibited by the first ten Conventions m en­
tioned above are very broad, one cannot exclude from the scope of such Conventions
conduct that, although clearly banned by them, does not fall under the category of
terrorism for lack of the requisite elements. For instance, the hijacking of a plane by a
robber that aims at obtaining a huge sum of money as a ransom or the release of some
fellow criminals in exchange for saving the passengers, plainly falls under the 1970
Hague Convention, without, however, constituting an act of international terrorism
proper.

8.5 INTERNATIONAL TERRORISM IN ARMED

CONFLICTS: A SUB-CATEGORY OF WAR CRIMES
At present, both IHL and ICL already cover acts of terrorism performed during an
international or internal armed conflict.
International rules indisputably ban terrorism in time of armed conflict. Art. 33(1)
of the Fourth Geneva Convention of 1949 prohibits ‘all measures [...] of terrorism’
against civilians. Although the provision was primarily calculated to forestall terror­
ism by Occupying Powers or, more generally, by belligerents,25 terrorist acts are also
prohibited if perpetrated by civilians or organized groups in occupied territories or
in the territory of a party to the conflict. Thus, Article 33(1) is a provision of general
purport, applicable in any situation (whether terrorism is resorted to in the territory of
one of the belligerents, in the combat area, or in an occupied territory).

a natural or juridical person, or a group of persons, to do or abstain from doing any act as an explicit or impli­
cit condition for the release of the hostage commits the offence of taking hostages (“hostage-taking”) within
the meaning of this Convention.’
Article 2 (1) (b) of the Convention on the Financing of Terrorism provides that a person commits an
offence within the meaning of the Convention if that person provides or collects funds to carry out among
other things any act ‘intended to cause death or serious bodily injury to a civilian, to any other person not
taking an active part in the hostilities in a situation of armed conflict, when the purpose of such act, by its
nature or context, is to intimidate a population, or to compel a government or an international organization
to do or to abstain from doing any act’.
25 According to the ICRC Commentary, Article 33(1) was aimed primarily at forestalling a common
practice, that of belligerents resorting to ‘intimidatory measures to terrorise the population’ with a view
to preventing hostile acts (see ICRC, Commentary, Fourth Geneva Convention (Geneva, ICRC, 1958), at
225-6).
 17 2 INTERNATIONAL CRIMINAL LAW

A similar provision is contained in the Second Additional Protocol of 1977. Article

4(2)(d) prohibits acts of terrorism’ against all persons who do not take a direct part or
have ceased to take part in hostilities, whether or not their liberty has been restricted’
(Article 4(1)).
The two Protocols also spell out the general prohibition of terrorism. Article 51(2) of
the First Protocol prohibits acts or threats of violence the prim ary purpose of which
is to spread terror among the civilian population’. Article 13(2) of the Second Protocol
repeats this prohibition word for word. It can be safely contended that all these provi­
sions reflect, or have turned into customary law.26
Thus, IHL proscribes terrorism both in international and internal armed conflicts.
The question, however, arises of whether, in addition to addressing its prohibition to
states, international customary and treaty law also criminalize terrorism in armed
conflict. An ICTY TC convincingly proved in 2003 in Galic that already in 1992 (when
the facts at issue in that case occurred) a serious violation of the prohibition against
terrorizing the civilian population entailed, at least under treaty law, the individual
criminal responsibility of the person breaching the rule (§§113-29).27
Contrary to this holding one could object that the Statute of the ICC, which care­
fully and extensively lists in Article 8 the various classes of war crimes subject to the
ICC jurisdiction, fails to mention resort to terror against civilians. This argument
would not, however, be compelling. Indeed, as noted above (1.4.3) the various provi­
sions of the ICC Statute are not intended to codify existing customary rules; this is
borne out by Article 10 of the Statute (‘Nothing in this Part shall be interpreted as
limiting or prejudicing in any way existing or developing rules of international law for
purposes other than this Statute’), as well as by the fact that some specific provisions
of the Statute concerning the crimes over which the Court has jurisdiction go beyond
customary or previous treaty rules, whereas others only partially take account of cus­
tomary law (see above, 4.6; 5.7; 6.6).
Support for the criminalization of terrorist acts in the course of armed conflict can
be found in various normative developments. The relevant provisions of the Statutes
of the ICTR and the SCSL, in granting these two criminal tribunals jurisdiction over

26 In Strugarand others (Decision on Interlocutory Appeal) the ICTY AC held that ‘the principles prohib­
iting attacks on civilians and unlawful attacks on civilian objects stated in Articles 51 and 52 of Additional
Protocol 1 and Article 13 of Additional Protocol II are principles of customary international law’ (§10).
It is notable that in 1977, at the close of the Geneva Diplomatic Conference on the Reaffirmation of
International Humanitarian Law, the United Kingdom stated that Art. 51(2) was a ‘a valuable reaffirm­
ation of existing customary rules of international law designed to protect civilians (CDDH, Official records,
vol. VI, at 164, §119).
The 2004 British Manual of the Law of Armed Conflict, in referring to the prohibition o f‘terror attacks’,
seems clearly based on the assumption that this rule is general in nature (§5.21.1).
The important research work undertaken by the ICRC on customary law also concludes that a custom­
ary rule has evolved on this matter (see J.-M. Henckaerts and L. Doswald-Beck, Customary International
Humanitarian Law, vol. 1 (Cambridge: Cambridge University Press, 2005) at 8).
27 It would seem, however, that the TC’s finding that the prohibition of terror in armed conflict was crim­
inalized was essentially limited to the case at issue and to the accused standing trial. In addition, the TC left
open the question of the possible criminalization of terror under customary international law (see §138).
 TERRORISM AS AN INTERNATIONAL CRIME 173

violations of international rules of hum anitarian law, include ‘acts of terrorism ’.28 This
proves that the drafters of those Statutes took the view that such acts may am ount to
war crimes. Also Article 20(f)(iv) of the 1996ILC Draff Code of Crimes against Peace
and Security of M ankind considers that acts of terrorism’ committed in internal con­
flicts constitute war crimes. Furthermore, it seems significant that Art. 2(l)(b) of the
1999 Convention of the Financing of Terrorism explicitly refers to ‘a situation of armed
conflict’, thus implying that terrorist acts can be committed in such a ‘situation’. Of
course the Convention is only binding on the contracting parties. Nevertheless, so far
the Convention has been ratified or acceded to by 153 states (only three of which—
Egypt, Jordan, and Syria—have entered reservation to the relevant treaty stipulation);
the provision at issue is therefore indicative of the generally held view that terrorism is
also criminalized in time of armed conflict.
In summary, attacks on civilians and other ‘protected persons’ in the course of an
armed conflict, aiming at spreading terror, may amount to war crimes (although not
to grave breaches of the Geneva Conventions,29 with the consequence that the Geneva
provisions on mandatory universal jurisdiction over such crimes do not apply, such
universal jurisdiction being simply authorized by the Geneva Conventions30).
Tet us now consider the constitutive elements of terrorism as a war crime. It would
seem that in IF1T terrorism as a war crime has a narrower scope than the notion con­
templated by the whole body of general international law of peace.
First of all, the prohibited conduct arguably consists of any violent action or threat of
such action against civilians or other persons not taking a direct part in armed hostil­
ities (wounded, shipwrecked, prisoners of war). It can be inferred both from the whole
spirit and purpose of international hum anitarian law and also from the wording of
Articles 4(1) and (2)(d) of the Second Additional Protocol (a rule that, it is submit­
ted, codifies a general principle applicable to any armed conflict)31, which attacks on
combatants not being actively engaged in armed hostilities can also amount to terror­
ism: for instance, attacks (or threats of attack) on officers attending a mass or praying

28 Article 4(d) of the 1994 Statute of the ICTR provides that the Tribunal has jurisdiction over violations
of common Article 3 of the Geneva Conventions and the Second Additional Protocol and explicitly provides
for jurisdiction over ‘acts of terrorism’. Article 3(d) of the 2000 Statute of the Special Court for Sierra Leone
grants the Court jurisdiction over ‘acts of terrorism’.
29 The relevant provisions of the 1949 Geneva Conventions (Articles 50/51/130/147) do not include ‘acts
or measures of terrorism’ among the offences amounting to grave breaches.
30 As rightly held, with regard to the war crime of torture in armed conflict, by the Hague District
Court in the Afghani cases. The Court rightly emphasized the importance of the common provision of the
Conventions (Articles 49(3): 50(3); 129(3); 146(3)) stipulating that ‘Each High Contracting Party shall take
measures necessary for the suppression of all acts contrary to the provisions of the present Convention other
than the grave breaches defined in the following Article.’
31 This provision is simply an expansion and elaboration of common Article 3 of the four Geneva
Conventions of 1949, which the International Court of Justice held in 1986 in Nicaragua (merits) to consti­
tutes ‘a minimum yardstick’ applicable to any armed conflict (at §218). For the state practice and the practice
of international organizations that can corroborate the proposition set out above in the text, seethe wealth of
material collected in J.-M. Henckaerts and L. Doswald-Beck, Customary International Humanitarian Law,
vol. 1 cit. supra, n. 39, at 306-83. See also the Report of the International Commission of Inquiry on Darfur,
UN doc. S/2005/60 (25 January 2005), at §§154-67.
174 INTERNATIONAL CRIMINAL LAW

in a mosque, a church or a synagogue, or military personnel taking their children to

the cinema. This proposition is borne out by the aforementioned Article 2(l)(b) of the
Convention for the Suppression of the Financing of Terrorism, which includes among
the possible victims of terrorist acts in time of armed conflict any other persons [than
civilians] not taking an active part in the hostilities’. Furthermore, as rightly stressed by
the Italian Court of cassation in Bouyahia and others (at 18-19), acts may be held to be
terrorist if they are carried out against servicemen engaged not in warlike actions but
hum anitarian operations, or if such acts are performed against civilians and military
alike (for instance, if an attack is made against a military vehicle in a crowded market,
with the consequence that both civilians and military are among the victims).
The violent action or threat thereof can also be directed against a civilian object, even
if it is empty (for instance, a square, a private building, a theatre), as long as the goal
pursued in taking such action is that of terrorizing the population. As rightly noted in
the 2004 British Manual, the rule prohibiting terror attacks ‘would apply, for instance, to
car bombs installed in busy shopping streets, even if no civilians are killed or injured by
them, their object being to create panic among the population’ (§5.21.1). As for threats,
again the British Manual rightly pointed out that ‘threats of violence would include,
for example, threat to annihilate the enemy’s civilian population’(ibid.). In contrast, the
prohibition on terror does not cover terror caused as a by-product of attacks on military
objectives or as a result of genuine warning of impending attacks on such objectives’.32
We can thus move to the subjective element of the action or threat of action. Articles
51(2) of the First Protocol and 13(2) of the Second Protocol, which, as I stated above,
can be taken to spell out in many respects the terse content of other provisions on
humanitarian law on terrorism, make it clear that terrorist acts in armed conflict are
acts calculated to ‘spread terror’ among the civilian population or other protected per­
sons. Here, then, the purpose of coercing a public (or private) authority to take a cer­
tain course of action disappears or, at least, wanes. The only conspicuous purpose
appears to be that of terrorizing the enemy. In other words, in IHT terrorist acts are
performed within the framework of the general goal of defeating the enemy. Their
ultimate purpose is to contribute to the war effort. Instead of simply attacking civil­
ians, a belligerent carries out actions (for instance, random killing of persons pass­
ing through a bridge, or haphazard blowing up of civilian installations, or systematic
shelling of an empty place in a populated area) designed to beget profound insecurity
and anxiety in the population (and consequently in the enemy belligerent).
It is thus clear that also in time of armed conflict international crim inal law requires
intent, with the consequence that, as rightly emphasized in Galic, simple dolus
eventualis or recklessness must be ruled out.33

32 See M. Bothe, K.-J. Partsch, W. Solf, New Rulesfor the Victims ofArmed Conflicts (The Hague, Boston,
London, M. NijhoffPublishers, 1982), at 301; The Manual of the Law of Armed Conflict, op. cit. at §5.21.1.
33 The Trial Chamber noted that ‘the Prosecution is required to prove not only that the Accused accepted
the likelihood that terror would result from the illegal acts—or, in other words, that he was aware of the
possibility that terror would result—but that that was the result which he specifically intended. The crime of
terror is a specific-intent crime’(§136).
 TERRORISM AS AN INTERNATIONAL CRIME 175

In addition, motive becomes immaterial in terrorist acts qua war crimes. In time of
armed conflict, actions designed to spread terror in the enemy are always ‘public’ in
nature and any personal motive (for instance, desire for revenge, racial or ethnic hat­
red, anger, and so on) of the officer or the leader of an armed group ordering such acts
does not acquire any legal relevance.
In summary, during an armed conflict, belligerent acts of terrorism, being prohib­
ited and criminalized, are covered both by IHL and ICL. They may also be covered by
rules on terrorism as a discrete crime to the extent that a state fighting terrorism is
bound by an international convention that addresses terrorism both in time of peace
and in time of war. In this event there would be a twofold legal characterization of
the same conduct or the combined simultaneous application o f two different bodies
o f law to the same conduct or set of acts. A case in point is the Convention on the
Financing of Terrorism. If a state is party to such Convention, it may apply its pro­
visions to the financing of terrorist acts performed or planned in a foreign country
where an armed conflict is underway. It would consequently punish the financing of
violent acts abroad directed against persons not taking an active part in armed hostil­
ities (whereas it would not consider unlawful the financing of groups solely aimed at
attacking enemy armed forces in the foreign country concerned). It may also bring to
trial for war crimes the perpetrators of terrorist actions abroad, who benefited from
the financing in question.

8.6
INTERNATIONA L TERRORISM AS A
SUB-CATEGORY OF CRIMES AGAINST H U M A N IT Y

Terrorist acts can amount to crimes against humanity, subject to a num ber of
conditions.
First of all, it can be inferred from the relevant international rules and case law on
crimes against humanity that terrorist acts may fall under this category of crimes,
whether they are perpetrated in time of war or peace. Furthermore, they must cause
(or consist of) the following conduct: (i) murder; or (ii) great suffering; or (iii) serious
injury to body or to mental or physical health; or else take the form of (iv) torture;
(v) rape; or even (vi) enforced disappearance of persons; namely.
a r r e s t , d e t e n t i o n o r a b d u c t i o n o f p e r s o n s b y, o r w i t h t h e a u t h o r i z a t i o n , s u p p o r t o r a c q u i e s ­
c e n c e o f, a s t a te o r a p o l i t i c a l o r g a n i z a t i o n , f o llo w e d b y a r e f u s a l t o a c k n o w le d g e t h a t d e p r i v ­
a t i o n o f f r e e d o m o r to g iv e i n f o r m a t i o n o n t h e fa te o r w h e r e a b o u t s o f t h o s e p e r s o n s , w i t h t h e
i n t e n t i o n o f r e m o v i n g t h e m f r o m t h e p r o t e c t i o n o f t h e la w f o r a p r o l o n g e d p e r i o d o f t i m e .34

34 This is the definition of enforced disappearance of persons’ set out in Article 7(2)(i) of the ICC Statute,
which can be taken to be declaratory of an existing (or emerging) rule of customary international law banning
and criminalizing that offence. See also the International Convention for the Protection of All Persons from
Enforced Disappearances, recently adopted by the UN Human Rights Council. It applies in all circumstances
(see online: http://daccessdds.un.org/doc/UNDOC/LTD/G06/125/78/PDF/G0612578.pdffOpenElement).
176 IN T E R N A T IO N A L C R IM IN A L LAW

Terrorist acts must also meet the basic requirements of the category of crimes under
discussion. Consequently (i) terrorist action must be part of a widespread or system­
atic attack against a civilian population conducted with the support or tolerance or
acquiescence of a state or a non-state entity (even if terrorist acts are performed against
persons or state officials in another state); (ii) the perpetrator, in addition to mens rea
required for the underlying offence (murder, torture, etc.) must also have knowledge
that his action is part of a widespread or systematic attack.
It would seem that, as in the case of terrorism as a discrete crime, also when terror­
ist acts are such as to am ount to crimes against humanity, the victims may embrace
both civilians and state officials including members of armed forces. Admittedly, the
Statutes of international crim inal tribunals, in granting jurisdiction to these tribu­
nals over crimes against humanity, stipulate that the victims of such crimes must
be civilian. However, this limitation cannot be found in customary international
law, which to my m ind provides instead that crimes against hum anity may also be
perpetrated against military personnel and members of other enforcement agencies
(see above, 5.6) Generally speaking, it would be contrary to the whole spirit and logic
of modern international hum an rights law and hum anitarian law to limit to civil­
ians (especially in time of peace) the international protection of individuals against
horrendous and large-scale atrocities. This, I believe, also holds true for terrorism
as a crime against humanity. For instance, it would not make sense to suggest that
the 11 September 2001 attacks against the Twin Towers in New York, housing almost
exclusively civilians, amounted to a crime against humanity, whereas the crashing of a
civilian aircraft into the Pentagon in Washington D.C. constituted a different category
of crime because the victims were not civilians but primarily state officials (mostly
even military personnel) at work.
In the case of terrorist acts, what matters from the point of view of law is not so
much to prevent civilians from becoming the target of grave crimes. W hat is crucial is
to avoid (and punish) crim inal action, whomever its victims, taken to compel a pub­
lic or private entity to do or not to do something. In a way, the victims play almost a
secondary role in the criminalization of conduct. W hat clearly emerges from current
international law is that the widespread or systematic attack required as the necessary
context of a crime against hum anity must be one that targets the civilian population.
This is only logical, for a widespread or systematic attack against members of armed
forces in time of peace would simply constitute part and parcel of an armed conflict
(internal if the attackers are within the territory, international if they come from out­
side); in time of war, depending upon the circumstances, that attack could, or could
not, amount to a string of large-scale breaches of hum anitarian law. If instead a wide­
spread or systematic attack is undertaken against the civilian population, for such
atrocities to amount to a crime against hum anity one set of atrocities (for instance,
torture, rape, or other inhum ane acts of similar gravity) may also be directed against
m ilitary personnel.
Some examples can illustrate the above propositions. In time of peace a group of
terrorists, in addition to conducting attacks on civilians, engages in such atrocities
 T E R R O R IS M AS A N IN T E R N A T I O N A L C R I M E 177

against military or police personnel as bombing barracks, blowing up police stations,

destroying a major building of the defence ministry, or else kidnapping servicemen
and subjecting them to torture or rape. These acts (murder, imprisonment, torture,
rape, and so on) should be classified as crimes against humanity. Similarly, it may hap­
pen that in time of armed conflict an armed group or organization (or even a state),
besides indiscriminately and violently attacking on a large scale civilians and other
persons not taking an active part in hostilities, captures, rapes, or tortures enemy
combatants for the purpose of spreading terror among the enemy belligerents or to
obtain from them the release of imprisoned members of the group, organization (or
state). These acts, which normally would be classified as war crimes, may acquire the
magnitude of a crime against humanity.
It is clear from the above that, in addition to the aforementioned objective ele­
ments, it is also necessary for the author of terrorist acts to entertain the specific intent
required for terrorism as a discrete crime, namely the purpose of compelling a public
or private authority to take, or refrain from taking, a certain course of action, a p ur­
pose that may be achieved either by generating fear and anxiety among the public or
by other criminal actions (see above).
In summary, terrorism as a crime against hum anity substantially constitutes an
aggravated form of terrorism as a discrete crime.

8.7 SUMMING UP
International law defines and regulates international terrorism. International terror­
ism as a discrete international crime perpetrated in time of peace exhibits the following
requisites: (i) it is an action normally criminalized in national legal systems ; (ii) it is
transnational in character, i.e. not limited in its action or implications to one country
alone; (iii) it is carried out for the purpose of coercing a state or an international organ­
ization to do or refrain from doing something; (iv) it uses for this purpose two possible
modalities: either spreading terror among civilians or attacking public or eminent
private institutions or their representatives; (v) it is not motivated by personal gain but
by ideological or political aspirations.
In time of peace international terrorism may also exhibit the hallmarks of a crime
against humanity. This happens when it is part of a widespread or systematic attack
against civilians (although terrorist conduct as such may be taken against state offi­
cials or even combatants) and in addition takes the shape of certain categories of crim ­
inal conduct (murder, causing great suffering or serious injury to body or to mental or
physical health, torture, rape, or enforced disappearance).
In time of armed conflict, terrorism is criminalized when it consists of violent action
(or even threat of such action) taken (i) against civilians or any other person not taking
an active part in armed hostilities; and (ii) has as its prim ary purpose that of spreading
terror among the civilian population.
178 IN T E R N A T IO N A L C R IM IN A L LAW

At present it is this legal classification relating to the use of terror in time of armed con­
flict that poses most problems at the political level and has indeed become the political
albatross for final agreement on a treaty definition of terrorism. The classification is in
fact rejected by some states. These states, while in fact opposing the application of some
general and treaty rules of hum anitarian law, do not propose any alternative to the
simple violation of (or disregard for) existing law. Other states are instead prepared to
accept that classification subject, however, to the condition that the same acts covered
by international hum anitarian law be not also classified as terrorism under general or
treaty rules on terrorism, or be classified as such with a limited interference of rules on
terrorism in time of peace.
On account of the current discussions on terrorist conduct by ‘freedom fighters’, the
evolution of the legal regulation of terrorism in time of armed conflict might lead to
the formation of a distinct category of warlike terrorist acts. In other words, a process
could take place similar to that leading to the formation within the broad category of
crimes against humanity, of the class of genocide. Genocide, initially a sub-category
of crimes against humanity, gradually became a discrete class of international crimes,
with distinct requisites different from those of the ‘parent class’ of crimes against
humanity. Similarly, terrorism in time of armed conflict, currently a sub-category of
war crimes, might gradually become a discrete class of international crimes as a result
of the combined application of hum anitarian law and general norms on terrorism.

8.8 THE GENERAL QUESTION OF

MULTIPLICITY OF OFFENCES
At the end of the general exposition of the various classes of international crimes, we
must discuss a general issue that concerns all such crimes: whether a person may be
charged with more than one crime, and if so, under what conditions and with what
legal consequences.
This problem, which arises in any legal system, is particularly im portant in ICL. As
pointed out above, international crimes normally have an ‘international context’ in
that they involve or presuppose other criminal conduct, or else they are carried out by
or implicate a multitude of persons. In addition, there are no written rules endowed
with general binding force (i.e. not customary in nature) that neatly define the legal
contours of each category of crime (see on this subject the compelling considerations
of Judges Hunt and Bennouna in their Opinion in Delalic and others (AJ, 2001), quoted
infra, in fn. 38). Thus the legal question arises of how to classify and categorize the
various criminal offences and whether each and all of them may be charged or instead
one or more ‘absorb’ others.
In particular, in ICL there may be instances of multiple offences (for instance, rape
followed by murder of the same victim), or of offences that simultaneously affect
 T E R R O R I S M AS A N I N T E R N A T I O N A L C R I M E 179

numerous victims (for example, bombing a civilian house), or of offences consisting of

the simultaneous breach of many rules (this, for instance, occurs when by the same act
a military officer breaches the rule prohibiting the use of certain weapons and that ban­
ning the killing or wounding of innocent civilians). It is therefore appropriate to try to
distinguish and classify all these instances, both from a theoretical viewpoint (in order
logically to categorize the various cases that may occur) and for practical purposes.
At the national level, these questions have prim arily been discussed and explored in
countries belonging to the Romano-Germanic tradition, where the legal literature has
reached considerable theoretical conclusions. In common law countries the matter
has been left to courts, which have gradually developed a set of criteria for at least the
most im portant distinctions.
Based on the jurisprudential and scholarly contributions of national law and
international case law, a few distinctions will be set out below.

8.8.1 D IF F E R E N T IA T IN G C LA SSES OF M U L T IP L E O FFE N C E S

W hen a person, by a set of closely linked but separate actions, perpetrates several
crimes against various victims, his conduct may be easily classified. Take, for instance,
the case of a soldier in charge of a detention camp, who in a brief time-span beats up
a prisoner of war, rapes another, then takes part in an execution squad charged with
shooting a third prisoner of war allegedly responsible for war crimes (but not regu­
larly tried and sentenced). Clearly, in this case the soldier is accountable for breaches
of different rules of international law, against different persons. No particular problem
arises with regard to the charging of the offender and his sentencing by a court: he
will be accused of a set of different war crimes (ill-treatment, rape, unlawful killing of
prisoners of war); if found guilty, he will be sentenced for each of these crimes. As an
ICTY TC stated in Kupreskic and others, we are faced here with an accumulation of
separate acts, each violative of a different provision (§678c).
A person may instead breach the same rule against various persons: for instance, a
soldier in a spray of gunfire murders ten innocent civilians in a combat area. In this
case only one rule is breached, that prohibiting the unlawful killing of civilians, but
the offence is committed against several victims.
The legal literature has termed these two sets of cases a ‘real concurrence o f offences’.
They do not pose any major problem of charging: the accused will be charged with
three different war crimes, in the first case, and with as many war crimes in the form
of murder, as there are victims, in the second.
Matters become complicated in other instances. Think for example of the case
where a person, by a single act or transaction, simultaneously violates several rules.
These cases are defined as an ‘ideal concurrence of offences’ (‘ideal’, of course, simply
as opposed to ‘real’). Here again one ought to distinguish between various categories
ofbreaches.
First, it may happen that the same act violates one rule in some respects and another
rule in other respects, the two rules covering different matters. In such cases the same
i8o IN T E R N A T IO N A L C R IM IN A L LAW

criminal conduct simultaneously breaches two different rules and amounts to two dif­
ferent crimes. For example, a serviceman, by using such a weapon as a flame-thrower,
burns out a civilian building in occupied territory housing some innocent civilians,
thus causing their death; here the soldier by the same act becomes responsible for both
m urder of civilians and arson. Or, to mention the example suggested by an ICTY TC
in Kupreskic and others, take the case of ‘the shelling of a religious group of enemy
civilians by means of prohibited weapons (e.g. chemical weapons) in an international
armed conflict, with the intent to destroy in whole or in part the group to which those
civilians belong’. Here this ‘single act contains an element particular to [...] genocide
to the extent that it intends to destroy a religious group, while the element particular
to Article 3 [of the ICTY Statute] (on war crimes) lies in the use of unlawful weapons’
(§679a). Clearly, when faced with these cases the prosecution must charge the defend­
ant with two different crimes. Similarly, if it is satisfied that the accused is guilty of the
breach of both rules, the court ought to sentence him for both breaches (although, it
would seem, the sentences should run concurrently).
Another problem may arise when we are faced with a single conduct or transaction
that successively breaches two different rules and may thus amount in theory to two
offences, one lesser than the other. For instance, a soldier seriously wounds an enemy
prisoner of war, thereby causing his death: we are here faced with the crimes of griev­
ous bodily harm and murder. Or a soldier sexually harasses a civilian woman and then
rapes her; we have here sexual assault and rape. Or a serviceman pillages private prop­
erty in occupied territory, but in appropriating goods belonging to enemy civilians,
faced with their resistance, uses force against them; we have here theft and robbery. In
these cases the common law doctrine of the ‘lesser included offence’ and the civil law
‘principle of consumption’ coincide. Under both doctrines the more serious offence
prevails over and absorbs, as it were, the other. Hence, the charge and conviction may
be only for the more serious offence.
Thirdly, it may happen that an act or transaction simultaneously breaches vari­
ous rules covering the same subject matter. A typical instance of such occurrence in
national criminal law is that of serious sexual violence by a m an against his daughter:
here the same conduct simultaneously amounts to rape and incest. In ICL the rape
of a civilian woman by a soldier may be classified both as a war crime and as a crime
against humanity. On the basis of which principles or criteria should one decide under
which of these two classes a specific rape falls? The answer to this query is important
not only for courts, but also for prosecutors, when they decide how to charge a person
suspected of international crimes.
One may deduce the criteria for settling these last issues from the principles of
criminal law common to the major legal systems of the world as well as international
case law.
One test or criterion seems to commend itself. It is known in common law countries
as the Blockburger test (based on a famous decision by the US Supreme C ourt delivered
in 1932 in Blockburger (at 304) and confirmed by the US Supreme Court in Rutledge
(1996) (at 297)) and in civil law countries as ‘the principle of reciprocal speciality’.
 T E R R O R IS M AS A N I N T E R N A T IO N A L C R I M E 18 1

Under this test, that has been taken up by the ICTY Kupreskic and others (TJ) and in
Delalic and others (AJ),35 it must be established whether each of the two or more provi­
sions that appear at first sight to be breached requires an element that the other does
not. If both provisions cover the same conduct, but one of the provisions requires an
objective or objective element not envisaged in the other provision, then the former
provision will apply. This prevalence is grounded in the principle of speciality (as an
ICTY TC held in Kupreskic and others (§683) and the ICTY AC authoritatively con­
firmed in Delalic and others (AJ) (§340)). On the strength of this principle one ought to
conclude that the rule providing for the special requirement shall prevail.36
In practice two eventualities may occur.
First, it may happen that, although each of two rules at stake sets out different
requirements, the offence in fact meets all the requirements of both rules. In this case,
the crime will amount to an offence under both rules.
For instance, the rule on rape of civilians as a crime against hum anity requires an
objective element (the act must be part of a widespread or systematic practice) that
the rule on rape as a war crime does not require. This last rule, in its turn, requires
an objective element (that the rape be connected with an international or an internal
armed conflict) that the other rule does not require (at least under customary inter­
national law). Hence, if the rape has been perpetrated within an internal armed con­
flict as part of a systematic practice, the offence may be regarded as both a war crime
and a crime against humanity.37 Another example can be found in Kupreskic and
others (TJ). The TC envisaged the case where a person is charged both with murder as
a crime against humanity and with persecution (including murder) as a crime against
humanity. It then stated
In this case the same acts of murder may be material to both crimes. This is so if it is proved
that (i) murder as a form of persecution meets both the requirement of discriminatory intent
and that of the widespread or systematic practice of persecution, and (ii) murder as a crime
against humanity fulfils the requirement for the wilful taking of life of innocent civilians
and that of a widespread or systematic practice of murder of civilians. If these require­
ments are met, we are clearly faced with a case of reciprocal speciality or in other words the

35 As the ICTY AC pithily put it in Delalic and others (AJ): reasons of fairness to the accused and the consid­
eration that only distinct crimes may justify multiple convictions, lead to the conclusion that multiple criminal
convictions entered under different statutory provisions but based on the same conduct are permissible only
if each statutory provision involved has a materially distinct element not contained in the other. An element is
materially distinct from another if it requires proof of a fact not required by the other’ (§412).
See also Jelisic (AJ), §82.
36 As was stated in Kupreskic and others, ‘the rationale behind the principle of speciality is that if an
action is legally regulated both by a general provision and by a specific one, the latter prevails as most appro­
priate, being more specifically directed towards that action. Particularly in case of discrepancy between the
two provisions, it would be logical to assume that the law-making body intended to give pride of place to the
provision governing the action more directly and in greater detail’ (TC, §684).
37 However, things may be different if, as under the ICTY Statute, the requirement of being linked to an
armed conflict is common to all the crimes falling under the Tribunal’s jurisdiction, with the exception
of genocide. In that case, a murder committed in an armed conflict as part of a widespread or systematic
practice may only be considered as a crime against humanity.
 18 2 IN T E R N A T IO N A L C R IM IN A L LAW

requirements of the Blockburger test are fulfilled. Consequently, murder will constitute an
offence under both provisions of the Statute (Article 5(h) and (a)) (§708).

It would seem that in these cases the prosecution should charge cumulatively, that
is, under both heads; similarly, if found guilty of both crimes, the accused should be
sentenced for both (but the two sentences should be served concurrently).
Secondly, it may happen that each of the two rules covering the same matter requires
different ingredients, and the offence meets the requirements prescribed by one of the
rules but not those demanded by the other.
An illustration of this eventuality can be found in Delalic and others (AJ 2001). The
Prosecutor had charged, for the same facts, some defendants with both murder as a war
crime (covered by Article 3 of the ICTY Statute) and wilful killing as a grave breach of
the Geneva Conventions (pursuant to Article 2 of the same Statute). The AC held that
since the provision on grave breaches provided for an element not envisaged in the pro­
vision on war crimes, the defendants could only be convicted of a grave breach
The definition of wilful killing under Article 2 contains a materially distinct element not
present in the definition of murder under Article 3: the requirement that the victim be a pro­
tected person. This requirement necessitates proof of a fact not required by the elements of
murder, because the definition of a protected person includes, yet goes beyond what is meant
by an individual taking no active part in the hostilities. However, the definition of murder
under Article 3 does not contain an element requiring proof of a fact not required by the ele­
ments of wilful killing under Article 2. Therefore, the first prong of the test is not satisfied,
and it is necessary to apply the second prong. Because wilful killing under Article 2 contains
an additional element and therefore more specifically applies to the situation at hand, the
Article 2 conviction must be upheld, and the Article 3 conviction dismissed (§315).

In these cases the prosecution should charge the accused with the commission of
one crime only. However, as a precaution, in case it then does not succeed in proving
in court the specific requirement or legal ingredient, the prosecution may deem it
expedient to also charge the crime, either alternatively, as suggested in Kupreskic and
others (TJ, §727) (an option that would be more rational and legally well-founded), or
cumulatively, as held in Delalic and others (AJ) (§400) (an option more advantageous
from a practical viewpoint).38

38 In this last-mentioned case the AC ruled as follows: 'Cumulative charging is to be allowed in light of the
fact that, prior to the presentation of all of the evidence, it is not possible to determine to a certainty which of
the charges brought against an accused will be proved. The TC is better poised, after the parties’presentation
of the evidence, to evaluate which of the charges may be retained, based upon the sufficiency of the evidence.
In addition, cumulative charging constitutes the usual practice of both this Tribunal and the ICTR’ (§400).
Judges Hunt and Bennouna, in their Separate and Dissenting Opinion, concurred with the majority on
this point (§12). They offered, however, a more convincing reasoning: 'As a practical matter, it is not reason­
able to expect the Prosecution to select between charges until all of the evidence has been presented. It is
not possible to know with precision, prior to that time, which offences among those charged the evidence
will prove, particularly in relation to the proof of differing jurisdictional pre-requisites [...] Further [...]
the offences in the Statute do not refer to specific categories of well-defined acts, but to broad groups of
offences, the elements of which are not always clearly defined and which may remain to be clarified in the
Tribunal’s jurisprudence. The fundamental consideration raised by this issue is that it is necessary to avoid
 T E R R O R IS M AS A N I N T E R N A T IO N A L C R I M E 183

8.8 .2 T H E IM PA C T OF M U L T IP L IC IT Y OF C R IM E S O N SE N T E N C IN G

O f course, when a court satisfies itself that the accused is guilty on several heads for
the same conduct, it should impose a sentence that reflects the whole of the culpable
conduct. The principles governing this matter were appropriately set out by the ICTY
AC in Delalic and others (AJ) (§430).39 The approach was, however, rejected in the
ICC Statute, Article 78(3) of which provides that when a person has been convicted of
more than one crime, the Court shall pronounce a sentence for each crime and a joint
sentence specifying the total period of imprisonment. This, it would seem, implies that
a specific penalty should be attached to each offence, and the C ourt may not impose a
single term of imprisonment for a variety of offences.

any prejudice being caused to an accused by being penalised more than once in relation to the same conduct.
In general, there is no prejudice to an accused in permitting cumulative charging and in determining the
issues arising from accumulation of offences after all of the evidence has been presented [... j [However] there
may be specific examples of obviously duplicative cumulative charging, where there is no reason in the par­
ticular circumstances that the Prosecution needs to see how the evidence turns out before selecting the most
relevant charge. In those circumstances it maybe oppressive to allow cumulative charging’ (§12 and n. 14).
39 ‘The overarching goal in sentencing must be to ensure that the final or aggregate sentence reflects the
totality of the criminal conduct and overall culpability of the offender. This can be achieved through either
the imposition of one sentence in respect of all offences, or several sentences ordered to run concurrently,
consecutively, or both. The decision as to how this should be achieved lies within the discretion of the TC.’
 SECTION II

MODES OF CRIMINAL
LIABILITY
 9
PERPETRATION AND JOINT
CRIMINAL ENTERPRISE

9.1 GENERAL

As in any national legal system, also in ICL responsibility arises not only when a person
materially commits a crime but also when he or she engages in other forms or modal­
ities of criminal conduct. In the following paragraphs I shall set out these different
modalities of participation.
Before I do so, it may however prove fitting to discuss briefly the position in national
legal systems. They converge in holding that, where a crime involves more than one
person, all performing the same act, all are equally liable as co-perpetrators, or prin­
cipals. In contrast, national legal orders differ when it comes to the punishm ent of
two or more persons participating in a crime, where these persons do not perform the
same act but in one way or another contribute to the realization of a criminal design.
For instance, A draws up plans for a bank robbery, B provides the weapons, C per­
forms the actual robbery, D acts as a lookout, E drives the getaway car, and F hides the
loot and in addition gives shelter to the robbers. Many systems (for instance those of
the US, France, Austria, Uruguay, and Australia) do not make any legal distinction
between the different categories of participant and mete out the same penalty to each
participant, whatever his role in the commission of the crime. As the California Penal
Code provides at §31, all those concerned in the commission of a crime’ including
those who aid and abet the crime, are to be held liable as principals.
In spite of this legal regulation, for classificatory purposes and to aid analysis, legal
commentators and courts use descriptive terms to distinguish between the various
categories of participant: in the example given above, A is an accessory before the
fact’ (he is not a principal’ for he was not present when the robbery was perpetrated),
B is an aider and abettor (or an ‘accessory before the fact’), C is a ‘first degree prin­
cipal’, D and E are ‘second degree principals’, and F is an ‘accessory after the fact’.
However, as noted above, under the general sentencing tariff no distinction is made
between these different categories of person. It is only provided that for accomplices
or accessories extenuating circumstances may be taken into account if their participa­
tion in the offence is less serious than that of the principal or principals. In fact, for the
 188 IN T E R N A T IO N A L C R IM IN A L LAW

purposes of sentencing, judges often draw a distinction between principals, instiga­

tors, and aiders and abettors.
In other national legal systems (for instance, Germany, Spain, and Russia) the
law draws instead a normative distinction between two categories—principals, and
accomplices or accessories—and provides in terms that the persons falling under the
latter category must be punished less severely. Thus, for instance, in German law, the
scale of penalties for accomplices (at least in the case of aiders and abettors, Gehilfe) is
less harsh than for the perpetrator (Täter).1
We will see that in international law neither treaties nor case law (as indicative of
customary rules) make any legal distinction between the various categories, at least
as far as the consequent penalties are concerned. This lack of distinction follows both
from: (i) the absence of any agreed scale of penalties in ICL; and from (ii) the general
character of this body of law; that is, its still rudim entary nature and the ensuing lack
of formalism (see supra, 1.2).
Consequently, the differentiation between the various classes of participation in
crimes, which I shall set out below, is merely based on the intrinsic features of each
modality of participation. It serves a descriptive and classificatory purpose only. It is
devoid of any relevance as far as sentencing is concerned. It is for judges to decide in
each case on the degree o f culpability of a participant in an international crime and
assign the penalty accordingly, whatever the modality of participation of the offender
in the crime.

9.2 PERPETRATION
Whoever physically commits a crime, either alone or jointly with other persons, is
criminally liable. For instance, the soldier who kills a war prisoner or an innocent
civilian is liable to punishm ent for a war crime. Similarly, the serviceman who rapes
an enemy civilian as part of a widespread or systematic attack on civilians is account­
able for a crime against humanity.
Perpetration is thus the physical carrying out of the prohibited conduct, accompan­
ied by the requisite psychological element.12

1 In two cases, the Extraordinary Courts Martial established in the Ottoman Empire to try persons
accused of participating in massacring Armenians in 1915 and plundering their possessions, applied the
‘Imperial Military Penal Code’, which drew a normative distinction between principals and accessories. The
Court therefore made a point of distinguishing between the ‘principal perpetrators’ and the ‘accessories’,
and assigning a different sentence to each category of defendant. In Kemal and Tevfik it sentenced the prin­
cipal perpetrator to death and the accessory to 15 years of hard labour (at 5-6, or 157-8); in Bahaeddin §akir
and others, the majority of judges held that two defendants were accessories, while three dissenting judges
held that they ‘were equally guilty of having been principal co-perpetrators’ (at 4 and 8 or 171 and 173).
2 In some cases courts have minimized the role of perpetrators executing illegal orders. This for instance
holds true for Alfons Gdtzfrid, which concerns mistreatment at the Majdanek camp. The Stuttgart Court
(Landgericht) held that ‘According to established case-law [...], the offender or accomplice is defined as
one whose thoughts and actions coincide with those of the author of the crime, who willingly gives in to
 PE R PET R A T IO N A N D JO IN T C R IM IN A L E N T ER PR ISE 189

9.3 C O -PER PE TR A TIO N

Crimes are often committed by a plurality of persons. If all of them materially take part
in the actual perpetration of the same crime and perform the same act (for instance,
they are all members of an execution squad shooting innocent civilians), we can speak
of co-perpetration. All participants in the crime partake of the same criminal conduct
and the attendant mens rea.

9.4 PARTICIPATION IN A JOINT

CRIM IN AL EN TERPRISE TO COM M IT
INTERNATIONA L CRIMES

9.4.1 IN T R O D U C T IO N

International crimes such as war crimes, crimes against humanity, genocide, torture,
and terrorism share a common feature: they tend to be expression of collective crim ­
inality, in that they are perpetrated by a multitude or persons: m ilitary details, para­
military units or government officials acting in unison or, in most cases, in pursuance
of a policy. W hen such crimes are committed, it is extremely difficult to pinpoint the
specific contribution made by each individual participant in the criminal enterprise
or collective crime, on two grounds.
First, not all participants may have acted in the same manner, but rather each of
them may have played a different role in planning, organizing, instigating, coordinat­
ing, executing or otherwise contributing to the criminal conduct. For instance, in the
case of torture one person may order the crime, another may physically execute it, yet
another may watch to check whether the victim discloses any significant information,
a medical doctor may be in attendance to verify whether the measures for inflicting
pain or suffering are likely to cause death, so as to stop the torture just before the
measures become lethal, another person may carry food for the executioners, and so
on. The question arises as to whether all these participants are equally responsible for

incitement to political murder, silences his conscience and makes another persons criminal aims the basis
of his own conviction and his own action or who sees to it that orders of that kind are ruthlessly carried out,
or who in so doing otherwise displays consenting enthusiasm or who exploits State terror for his own pur­
poses. Accordingly, the accused could only be shown to have an attitude denoting guilt if, over and above the
activity he was instructed to carry out, he had performed some contributory act on his own initiative beyond
the call of duty, shown particular enthusiasm, had acted with particular ruthlessness in the extermination
operation or had shown a personal interest in the killings. These conditions cannot be shown to exist in the
case of the accused. He was at the end of the chain of command, had no power of decision himself and no
authority to act [...] Similarly, there is no evidence that the accused had any personal interest in the killings.
He merely wanted to carry out the order which had been issued to him’(67, b).
 190 I N T E R N A T IO N A L C R I M I N A L LAW

the same crime, torture. Similarly, in the case of deportation of civilians or prisoners
of war to an extermination camp, a commander may issue the order, several officers
may organize the transport, others may take care of food and drinking water, others
may carry out surveillance over the inmates so as to prevent their escape, others may
search the detainees for valuables or other things before deportation, and so on.
Secondly, the evidence relating to each individual’s conduct may prove difficult,
if not impossible, to find. It would, however, be not only immoral, but also contrary
to the general purpose of criminal law (to protect the community from the deviant
behaviour of its members that causes serious damage to the general interests) to let
those actions go unpunished. These considerations a fortiori apply to crimes such as
m urder or aggravated assault committed by a whole crowd; in such cases, it may prove
even more difficult to collect evidence about the exact participation of members of the
crowd in the crimes. The same considerations also hold true for cases where crimes
are institutionally committed within organized and hierarchical units such as intern­
ment, detention, or concentration camps, where it is difficult to pinpoint the grada­
tions of culpability of the various persons working within and for the organization.
As in most national legal systems, also in ICL all participants in a common criminal
action are equally responsible if they (i) participate in the action, whatever their position
and the extent o f their contribution; and in addition (ii) intend to engage in the common
criminal action. Therefore they are all to be treated as principals ,3 although of course
the varying degree of culpability may be taken into account at the sentencing stage.4

3 However, some courts of common law countries have taken the view that participants in a common
criminal design may play the role of, and be regarded as, accessories. Thus, for instance, in Einsatzgruppen,
with regard to common design, the Prosecutor T. Taylor, in his closing statement noted that ‘the elementary
principle must be borne in mind that neither under Control Council Law No. 10 nor under any known sys­
tem of criminal law is guilt for murder confined to the man who pulls the trigger or buries the corpse. In line
with recognized principles common to all civilized legal systems, §2 of Article II of Control Council Law No.
10 specifies a number of types of connection with crime which are sufficient to establish guilt. Thus, not only
are principals guilty but also accessories, those who take a consenting part in the commission of crime or
are connected with plans or enterprises involved in its commission, those who order or abet crime, and those
who belong to an organization or group engaged in the commission of crime. These provisions embody no
harsh or novel principles of criminal responsibility’ (372).
4 In this connection one may mention, by way of example, a decision of the Supreme Court of Bosnia
and Herzegovina in Tepez, delivered on 1 October 1999: ‘The appeal by the defence counsel argues that
the contested judgment has not individualised the criminal responsibility of the accused and his personal
involvement in actions characteristic of a war crime against the civilian population. For this crime to exist
it is necessary to “commit murder, torture, inhumane acts, inflict severe suffering, physical and mental
injuries on civilians, destroying their health and physical integrity”. The disposition does not include these
essential elements of this criminal act and therefore represents a major violation of the provisions of crim­
inal procedure. This Court finds these allegations groundless. The appeal fails to note that the contested
judgment states that the accused carried out these actions with three other named individuals (as well as
others), which means that he perpetrated the crime for which he has been pronounced guilty in complicity
with others. It further means that in cases of this kind where it is not possible to isolate individual actions
and their consequences or to distinguish the degree to which each person was involved in their execution,
it suffices that these actions complement each other and together form a single entity, which the accused
[Tepe] wishes to achieve by being involved. Therefore it was neither possible nor necessary for the court of
first instance to separate only the actions of the accused. It suffices that the accused participated in executing
these actions, even if it had only been one or two actions of personal involvement in the beating of civilians.
 PER PET R A T IO N A N D JO IN T C R IM IN A L E N T ER PR ISE 19 1

The notion of joint criminal enterprise (JCE) denotes a mode of criminal liability that
appears particularly fit to cover the criminal liability of all participants in a common
criminal plan. At the same time, this notion does not run contrary to the general princi­
ples of criminal law. As in national legal systems, the rationale behind this legal regulation
is clear: if all those who take part in a common criminal action are aware of the purpose
and character of the criminal action and share the requisite criminal intent, they must
perforce share criminal liability, whatever the role and position they may have played in
the commission of the crime. This is the case because: (i) each of them is indispensable for
the achievement of the final result; and on the other hand (ii) it would be difficult to dis­
tinguish between the degree of criminal liability, except for sentencing purposes.
Thus, it is by now widely accepted by international criminal courts that in the case of
collective criminality where several persons engage in the pursuit of a common crim ­
inal plan or design, all participants in this common plan or design may be held crim ­
inally liable for the perpetration of the criminal act, even if they have not materially
participated in the commission of said act; in addition, they may also be held respon­
sible, under a number of well-defined conditions, for criminal conduct that, although
not originally envisaged in the common criminal design, has been undertaken by one
of the participants and may to some extent be regarded as a natural and foreseeable
consequence of such a common plan. It is also widely accepted that at the international
level this mode of criminal liability can take three different forms. It was the ICTY AC
that first articulated in Tadic (AJ 1999) the doctrine of JCE as a fully fledged legal con­
struct of modes of criminal liability. However, the doctrine had already been upheld
at the national or international level by various courts, if only in passing. In Tadic (AJ
1999) the ICTY AC spelled out the three categories I will refer to below.

9.4.2 L I A B I L I T Y F O R A C O M M O N I N T E N T I O N A L P U R P O S E

(A) The notion

The first and more widespread category of liability is responsibility for acts agreed
upon when m aking the common plan or design. Here all the participants share the
same intent to commit a crime, and all are responsible, whatever their role and pos­
ition in carrying out the common criminal plan (even if they simply vote, in an assem­
bly or in a group, in favour of implementing such a plan). In addition to shared intent,
dolus eventualis (i.e. recklessness or advertent recklessness) (see supra, 3.7) may also
suffice to hold all participants in the common plan criminally liable. For instance, if a

However, the court of first instance has established that the accused personally beat up many individuals on
many occasions’ (2).
Also the decision of a Canadian court in Moreno deserves mentioning: ‘In reaching this conclusion, I am
influenced by one commentator’s view that the closer a person is involved in the decision-making process
and the less he or she does to thwart the commission of inhumane acts, the more likely criminal responsi­
bility will attach [...] of course, the further one is distanced from the decision makers, assuming that one
is not a “principal”, then it is less likely that the required degree of complicity necessary to attract criminal
sanctions, or the application of the exclusion clause, will be met’ (18). See also Ramirez (6-9).
 19 2 IN T E R N A T IO N A L C R I M I N A L LAW

group of servicemen decides to deprive civilians of food and water in order to compel
them to build a bridge necessary for military operations or to disclose the names of
other civilians who have engaged in unlawful attacks on the military, and then some
civilians die, the servicemen should all be accountable not only for a JCE to commit
the war crimes of intentionally starving civilians and compelling the nationals of the
hostile party to take part in operations of war directed against their own country’; they
should also be held guilty of murder. Indeed, even if the servicemen did not intend to
bring about the death of the civilians, the death was the natural and foreseeable con­
sequence of their common criminal plan and the follow-up action.
Society—in our case the world community—must defend itself from this collective
criminality by reacting in a repressive m anner against all those who, in some form,
took part in the criminal enterprise. Society may not indulge in distinctions between
the different roles played by each of the participants when trying to uproot or, better,
punish this form of collective criminality. All actors are guilty, even though in some
instances the mens rea (for example, intent to murder) is not attended by the corre­
sponding conduct (for example, stabbing or firing a gun); this applies to all those who,
while sharing the criminal intent, do not carry out the prim ary crime (for example,
the driver or the look-out in an armed robbery involving murder). However, the differ­
ing degrees of culpability can be taken into account at the stage of sentencing.

(B) Case law

In Ponzano, a case concerning the unlawful killing of four British prisoners of war by
German troops, the Judge Advocate adopted the approach suggested by the Prosecutor,
and stressed
the requirem ent that an accused, before he can be found guilty, m ust have been concerned in
the offence [... T]o be concerned in the com m ission of a crim inal offence [...] does not only
m ean that you are the person who in fact inflicted the fatal injury and directly caused death,
be it by shooting or by any other violent means; it also m eans an indirect degree of partici­
pation [... I]n other words, he m ust be the cog in the wheel of events leading up to the result
w hich in fact occurred. He can fu rth er th at object not only by giving orders for a crim inal
offence to be com m itted, but he can fu rth er th at object by a variety of other m eans (at 7).

The Judge Advocate also underlined that the accused should have knowledge of the
intended purpose of the crim inal enterprise.5

5 Georg Otto Sandrock et al. (also known as the Almelo Trial) can also be cited. Three Germans hitd
killed a British prisoner of war; it was clear that they had all had the intention of killing the British soldier,
although each of them played a different role. The British Court found all of them guilty of murder under
the doctrine of ‘common enterprise’ (at 35,40-1). In Holzer and others, brought before a Canadian military
court, in his summing up the Judge Advocate emphasized that the three accused (Germans who had killed
a Canadian prisoner of war) knew that the purpose of taking the Canadian to a particular area was to kill
him. The Judge Advocate spoke of a ‘common enterprise’ with regard to that murder (at 341, 347, 349). In
Jepsen and others a British court had to pronounce upon the responsibility of Jepsen and others for the death
of inmates of a concentration camp in transit to another concentration camp. The Prosecutor argued that
‘[I]f Jepsen was joining in this voluntary slaughter of eighty or so people, helping the others by doing his
share of killing, the whole eighty odd deaths can be laid at his door and at the door of any single man who
 PER PET R A T IO N A N D JO IN T C R IM IN A L EN TER PR ISE 193

In 2001, in Krstic, an ICTY TC held that the defendant had participated in a JCE to
commit genocide. The Court explained at length that initially Krstic had only taken
part in a common plan to forcibly expel Muslims from the area of Srebrenica; however,
later on, when it became apparent that the various military leaders in fact were plan­
ning the killing of thousands of military-aged men, the defendant showed, through
his various acts and behaviour, that he shared the ‘genocidal intent to kill the men
(§§621-45). The Chamber therefore found Krstic guilty of genocide and sentenced
him to 46 years in prison. The AC held instead that Krstic was only guilty of complicity
in genocide, for he had not shared the genocidal intent but simply aided and abetted
genocide. It reduced his sentence to 35 years’ imprisonment.
In 2003, in Blagojevic, Simic and others an ICTY TC held that the three accused,
Bosnian Serbs operating in the municipalities of Bosanski Samac and Odzak in Bosnia
Herzegovina, committed various crimes there. The main defendant, Simic (who, at the
time of the conflict, was the President of the Municipal Assembly and of the Crisis
Staff, later renamed ‘the War Presidency’), participated in a basic form of JCE. He
shared with others the intent to execute a common plan of persecution of non-Serb
civilians in the Bosanski Samac municipality. According to the TC, Simic, as the high­
est-ranking civilian in the municipality, acted in unison with others to execute a plan
that included: the forcible takeover of the town of Bosanski Samac, and the persecu­
tions of non-Serb civilians in the area, which took the form of unlawful arrests and
detention, cruel and inhumane treatment including beatings, torture, forced labour
assignments, and confinement under inhum ane conditions, deportations and forcible
transfers. The Chamber held that he was a participant in the JCE, while no evidence
permitted the conclusion that the other two defendants were also participants (TC,
2003, §§144-60, 983-1055).6
The ICTY took an important stand in Brdanin in 2004. In the indictment, the
Prosecution had alternatively pleaded the defendant’s criminal responsibility pursu­
ant to the first and third categories of JCE (on this third category see infra, 9.4.4). W ith
respect to the first category, the Prosecution alleged in the various counts that ‘[t]he
purpose of the joint criminal enterprise was the permanent forcible removal of Bosnian

was in anyway assisting in that act’. The Judge Advocate did not rebut the argument (at 241). In Schonfeld the
Judge Advocate stated that: ‘if several persons combine for an unlawful purpose or for a lawful purpose to be
effected by unlawful means, and one of them in carrying out that purpose, kills a man, it is murder in all who
are present [...] provided that the death was caused by a member of the party in the course of his endeavours
to effect the common object of the assembly’ (68).
6 It should be noted that the ICTR upheld the doctrine at issue as well. In Rwamakuba (Decision on
Interlocutory appeal) the AC held that the Tribunal had jurisdiction to try the appellant on a charge of
genocide through the mode of liability of JCE (§§9-39). In Elizaphan Ntakirutimana and Gerard
Ntakirutimana the AC relied upon the first category of JCE, but found that the TC had been correct in not
applying the doctrine to the case at issue (§§462,466,468-84). In Simba, in 2005, an ICTR TC held that the
accused was guilty of JCE to commit genocide and extermination (§§386-96, 411-19, 420-6). In another
case where the Prosecution had similarly charged a person with JCE to commit genocide and extermination
(Mpambara), an ICTR TC held instead that no proof beyond a reasonable doubt had been tendered that the
accused possessed the intent to be part of a JCE. It consequently acquitted him on all counts of the indict­
ment (§§13-4, 38-40, 76,113,164).
 194 IN T E R N A T IO N A L C R IM IN A L LAW

Muslim and Bosnian Croat inhabitants from the territory of the planned Serbian state
by the commission of the crimes alleged’. The alternative pleading of the third category
specified that ‘[the defendant] [was] individually responsible for the crimes enumer­
ated in [various counts] on the basis that these crimes were natural and foreseeable
consequences of the acts’ of deportation and forcible transfer of civilians. The Chamber
noted that for both categories of JCE to materialize, it was required to prove not only the
existence of a common criminal plan, but also that the crimes had been perpetrated by
one or more participants in such common plan. However, in the case at issue the crimes
had been committed by members of the army, police, and para-military groups that
had not participated in the criminal plan or enterprise (§345)7 The Chamber there­
fore dismissed the applicability of the notion of JCE to those crimes (§§351 and 355).
However, the AC reversed the TC decision on this issue, taking the contrary view. After
reviewing post-Second World War case law it concluded that such case law recognizes
the imposition of liability upon an accused for his participation in a common criminal
purpose, where the conduct that comprises the criminal actus reus is perpetrated by
persons who do not share the common purpose and that in addition it does not require
proof that there was an understanding or agreement to commit that particular crime
between the accused and the principal perpetrator of the crime. The AC thus held that
[W]hat matters in a first category ICE is not whether the person who carried out the actus
reus of a particular crime is a member of JCE, but whether the crime in question forms part of
the common purpose. In cases where the principal perpetrator of a particular crime is not a
member of the JCE, this essential requirement may be inferred from various circumstances,
including the fact that the accused or any other member of the JCE closely cooperated with
the principal perpetrator in order to further the common criminal purpose. In this respect,
when a member of the JCE uses a person outside the JCE to carry out the actus reus of a crime,
the fact that the person in question knows of the existence of the JCE—without it being estab­
lished that he or she shares the mens rea necessary to become a member of the JCE—may be a
factor to be taken into account when determining whether the crime forms part of the com­
mon criminal purpose. However, this is not a sine qua non for imputing liability for the crime
to that member of the JCE (§410). [...] Considering the discussion of post-World War II cases
and of the Tribunal’s jurisprudence above, the Appeals Chamber finds that, to hold a member
of the JCE responsible for crimes committed by non-members of the enterprise, it has to be
shown that the crime can be imputed to one member of the joint criminal enterprise, and that
this member—when using a principal perpetrator—acted in accordance with the common
plan. The existence of this link is a matter to be assessed on a case-by-case basis (§413).

The AC clinched the point by adding, always in light of post-Second World War
jurisprudence, that when the principal perpetrator is not part of the JCE, for the
accused to be held liable for the crime perpetrated, an understanding or an agreement
between the accused and the principal perpetrator of the crime is not necessary. It may
suffice that the crime at issue be part of the common criminal purpose (§§415-19) and
the accused uses’ the principal perpetrator to further that purpose (§§430-1).7

7 The TC had set out the same view in a previous decision in the same case (Brdanin, Decision on Form of
Further Amended Indictment and Prosecution Application to Amend, §44).
 PER PET R A T IO N A N D JO IN T C R IM IN A L E N T ER PR ISE 195

For the reasons set out below (§§9.4.5), it is respectfully submitted that this broad­
ening of the notion under discussion is excessive and raises doubts about its consist­
ency with the nullum crimen principle and the principle of personal responsibility.
The AC’s ruling in Brdanin seems all the more objectionable because in the same case
the Chamber also held that the doctrine of the JCE extends to ‘large-scale cases’ or in
other words covers instances where crimes are perpetrated on a large scale by indi­
viduals who are remote from the accused (§§420-5).

9.4.3 L I A B I L I T Y F O R P A R T I C I P A T I O N I N A C O M M O N C R I M I N A L
PLA N W IT H IN A N IN ST IT U T IO N A L FR A M EW O R K

(A) The notion

The second modality of liability is that of responsibility for carrying out a task within
a criminal design that is implemented in an institution such as an internment, deten­
tion, or concentration camp. In one such camp where inmates are severely ill-treated
and even tortured, not only the head of the camp, but also his senior aids and those
who physically inflict torture and other inhum an treatment bear responsibility for
those acts. In addition to those who physically carry out the misdeeds, also those who
discharge administrative duties indispensable for the achievement of the camp’s main
goals (for example, to register the incoming inmates, record their death, give them
medical treatment, or provide them with food) may incur criminal liability.
They bear this responsibility so long as they (i) are aware of the serious abuses being
perpetrated (knowledge); (ii) willingly take part in the functioning of the institution
(intent); and (iii) make an im portant contribution to the pursuit of the institution’s
goals. That they should be held responsible is only logical and natural: by fulfilling
their administrative or other operational tasks, they contribute to the commission of
crimes. W ithout their willing support, crimes could not be perpetrated. Thus, how­
ever peripheral their role, they may constitute an indispensable cog in the murdering
machinery. The m an who, upon arrival of new trains at Auschwitz, separated the men
and the women from the children and the elderly, knowing that this served to establish
who should be a forced labourer and who should instead be sent immediately to gas
chambers, was instrum ental in the perpetration of extermination. Had he intended to
shirk criminal responsibility, he should have asked to be relieved of his duties and to
discharge other duties elsewhere. This decision was possible and was sometimes made
(although it often involved being sent to combat zones on the Eastern Front). Similarly,
the locomotive driver of a train that carried hundreds of detainees to Auschwitz could
have been held criminally liable for his participation in extermination, so long as he
knew what would happen to the persons he was transporting and showed to share the
intent to exterminate those persons by willingly continuing to fulfil his role (instead
of asking to be exempted from this horrible task).
It can thus be noted that for this mode of liability no previous plan or agreement is
required. Nevertheless, one can legitimately hold that each participant in the criminal
 196 IN T E R N A T I O N A L C R I M I N A L LAW

institutional framework not only is cognizant of the crimes in which the institution
or its members engage, but also implicitly or expressly shares the criminal intent to
commit such crimes. It cannot be otherwise, because any person discharging a task
of some consequence in the institution could refrain from participating in its crim ­
inal activity by leaving it. As pointed out above, for criminal liability to arise it is also
necessary that the person at issue make a substantial contribution to the joint criminal
enterprise. It follows that those who, for example, merely sweep the streets or clean the
laundry should not incur criminal liability for their action, although they may both be
aware of the criminal purpose pursued by the whole institution and share it.
Clearly, this mode of responsibility is very close to that of criminal organizations
laid down in the IMT Charter annexed to the London Agreement of 8 August 1945
(Articles 9-11), and upheld in some respects by the IMT at Nuremberg (see infra, 2.2).
Indeed, in both cases belonging to and operating for an organization (or an institu­
tional framework) that prim arily or at least in part pursues criminal purposes involves,
subject to certain conditions, the personal guilt of a member. However, the conditions
for personal liability of a member to arise are only partially similar. True, in both cases
membership as such is not punishable. In both cases it is necessary for the member
to have knowledge of the criminal acts being committed or be personally implicated
in the commission of such acts.8 However, in the case of criminal organization this
would be sufficient, for the assumption is that the organization as such institutionally
pursues a criminal purpose (e.g. extermination of a racial or religious group). Instead,
in the JCE under consideration, since the institutional aims are not per se criminal
(the camp has been established to detain prisoners of war, or intern enemy civilians,
etc.), but the institution is incidentally used for criminal purposes (torture, murder,
extermination, rape, etc.), it is also necessary for a member to make a substantial con­
tribution to the furtherance of criminal purposes, for his liability to arise.

(B) Case law

One can find a particularly clear and significant illustration of this category of
criminality in Alfons Klein and others (the Hadamar trial), heard by a US Military
Commission sitting at Wiesbaden. It is fitting to dwell on this case at some length,
because it best shows how the category of criminality at hand works.
The accused were seven Germans. Between July 1944 and April 1945, they killed
over 400 Polish and Russian nationals, who had been obliged to work in Germany for
the German war effort and were suffering from tuberculosis or pneumonia. Brought
to Hadamar, in Germany, where there was a hospital or institution originally designed
to care for the mentally unsound, but with no medical facilities to treat persons ill with
tuberculosis or pneumonia, they were told that they would be given medication. In
fact they were killed by injections of poisonous drugs; afterwards the relevant medical

8 In Goring and others the IMT held that the definition or criminal organization ‘should exclude persons
who had no knowledge of the criminal purposes or acts of the organization and those who were drafted
by the State for membership, unless they were personally implicated in the commission of acts declared
criminal by Article 6 of the Charter as members of the organization’ (at 256).
 PE R P E T R A T IO N A N D JO IN T C R IM IN A L EN T E R PR ISE 197

records and death certificates were falsified. It would seem that the prim ary purpose
of these killings was to make space in hospitals for German war victims. The accused
comprised Klein, the administrative head of the hospital, a local Nazi Party leader who
made all the arrangements leading to the perpetration of the atrocities; W ahlmann, a
physician specializing in mental diseases, the Institution’s only doctor (he participated
in the conferences designed to plan the murders, knew what was going on at the hos­
pital, and acquiesced in it); three nurses, Ruoff, Willig, and Huber, who administered
the poisonous drugs; Merkle, the institution’s book-keeper (who registered incoming
patients for the purpose of recording dates and causes of death, actually falsifying
these documents); and Blum, a doorman and telephone switchboard operator, who
also served as caretaker of the cemetery, charged with burying the victims in mass
graves (but he sometimes walked through the wards to inspect the victims before they
were taken, dead, to his cellars a few hours later).
The charge for all of them was of Violation of international law’, namely, as the
Prosecutor specified in his opening argument, breach of the laws of warfare (at 202).
The specification stated that the seven accused ‘acting jointly and in pursuance of a
common intent’ ‘did [...] wilfully, deliberately and wrongfully aid, abet and partici­
pate in the killing of hum an beings of Polish and Russian nationality’. Thus, in add­
ition to the notion o f‘participation in killing based on common intent’ also the notion
o f‘aiding and abetting’ was used. However, in his Opening Argument the Prosecutor,
when setting out the applicable law (there was no Judge Advocate), emphasized that
all those who participate in a common criminal enterprise are equally guilty as
‘co-principals’ whatever the role played by each single participant. Referring to the
case of m urder committed by several persons, he pointed out that
Every single one of those who participated in any degree towards the accom plishm ent of
th at result [murder] is as m uch guilty of m urder as the m an who actually pulled the trigger
[...] That is why under our [that is, US] Federal Law all distinctions between accomplices,
between accessories before the fact and accessories after the fact, have been completely elim ­
inated. Anyone who participates in the com m ission of any crim e, w hether formerly called as
an accessory or not, are now co-principals and have been so for several years (203).

Moving then to the case at bar, the Prosecutor in fact offered an eloquent illustration
of the rationale behind the legal notion he was invoking:
At this H adam ar mill there was operated a production line of death. N ot a single one of
these accused could do all the things th at were necessary in order to have the entire scheme
of things in operation. For instance, the accused Klein, the adm inistrative head, could
n ot m ake the initial arrangem ents, receive those people, attend to undressing them , make
arrangem ents for their death chamber, and at the same tim e go up there and use the needle
th at did the d irty work, and then also tu rn around and haul the bodies out and b u ry them,
and falsify the records and the death certificates. No, when you do business on a wholesale
production basis as they did at the H adam ar Institution, that m urder factory, it means that
you have to have several people doing different things of that illegal operation in order to
produce the results, and you cannot draw a distinction between the m an who may have
initially conceived the idea of killing them and those who participated in the com mission
 198 IN T E R N A T IO N A L C R IM IN A L LAW

of those offences. Now, there is no question but that any person who participated in that
matter, no matter to what extent, technically is guilty of the charge that has been brought
[...] every single one of the accused has overtly and affirmatively participated in this entire
network that brought about the illegal result (205-7).

The defence counsel did not dispute these concepts, but in their arguments preferred
to rely upon the notions of necessity and superior orders, or argued that German law
rather than US or international law should apply. The Court upheld the charge. The
administrative head of the hospital and two nurses were sentenced to death; the phys­
ician (a 70-year-old man) to life im prisonment and hard labour; the book-keeper to 35
years and hard labour; the third nurse to 25 years and hard labour; the doorm an and
caretaker to 30 years and hard labour (at 247).
Courts also applied this notion of JCE in cases where the crimes had allegedly been
committed by members of m ilitary or administrative units running concentration
camps; that is, by groups of persons acting pursuant to a concerted plan.9In such cases
the accused held some position of authority within the hierarchy of the concentra­
tion camps. Normally, the defendants were charged with having acted in pursuance
of a common design to kill or m istreat prisoners and hence to commit war crimes.10
W hen found guilty, they were regarded as co-principals in the various crimes of ill-
treatment, because of their objective ‘position of authority’ within the concentration
camp system and because they had ‘the power to look after the inmates and make their
life satisfactory’ but failed to do so. In these cases, as the ICTY AC pointed out in Tadic
(AJ, 1999)

the required actus reus was the active participation in the enforcement of a system of repres­
sion, as it could be inferred from the position of authority and the specific functions held by
each accused. The mens rea element comprised: (i) knowledge of the nature of the system
and (ii) the intent to further the common concerted design to ill-treat inmates. It is import­
ant to note that, in these cases, the requisite intent could also be inferred from the position of
authority held by the camp personnel. Indeed, it was scarcely necessary to prove intent where
the individual’s high rank or authority would have, in and of itself, indicated an awareness
of the common design and an intent to participate therein. All those convicted were found
guilty of the war crime of ill-treatment, although of course the penalty varied according to
the degree of participation of each accused in the commission of the war crime (§203).

Later on an ICTY TC invoked this mode of responsibility in 2001 in Kvocka and

others. The Chamber found that the five defendants had occupied positions or roles

9 See, for instance, such cases as Dachau Concentration Camp, brought before a US Tribunal under
Control Council Law no. 10 (at 5,14), Nadler and others, decided by a British Court of Appeal under Control
Council Law no. 10 (at 132-4), Auschwitz Concentration Camp, decided by a German Court (at 882), as well
as Belsen, decided by a British military court sitting in Germany (121).
10 In his summing up in the Belsen case, the Judge Advocate took up the three requirements set out by the
Prosecution as necessary to establish guilt in each case: (i) the existence of an organized system to ill-treat
the detainees and commit the various crimes alleged; (ii) the accused’s awareness of the nature of the system;
and (iii) the fact that the accused in some way actively participated in enforcing the system, i.e. encouraged,
aided, and abetted or in any case participated in the realization of the common criminal design (637-41).
 PER PET R A T IO N A N D JO IN T C R IM IN A L EN TER PR ISE 199

in the operation of a detention camp at Omarska, where various crimes were com­
mitted (persecution, murder, and torture). Kvocka had been the camp commander’s
right hand; Kos was a guard shift commander; Radic was a shift commander. Zigic,
who was a taxi driver in the Prijedor area during the period of 26 May to 30 August
1992, used to enter Omarska as well as other two camps for the purpose of abusing,
beating, torturing, and killing prisoners. Finally, Prcac was de facto a deputy camp
commander. According to the Chamber, the Omarska camp was a JCE, a facility used
to interrogate, discriminate against, and otherwise abuse non-Serbs from Prijedor
and which functioned as a means to rid the territory of or subjugate non-Serbs’ (§323).
The Chamber held that the continuous perpetration of crimes in the camp was com­
mon knowledge to anybody living there (§324). It held that all the accused formed part
of a JCE to commit the crimes ascribed to them, and sentenced all of them to varying
sentences. The AC confirmed the convictions and sentences.
It is worth stressing that the TC rightly emphasized the need for the participation of
a person in an institutionalized JCE to be ‘significant’; that is, through an act or omis­
sion that makes an enterprise efficient or effective; e.g. a participation that enables
the system to ru n more smoothly or without disruption’ (§309). It then wisely went
on to note that the significance of the contribution is to be determined on a case-by­
case basis, taking into account a variety of factors (§311). On this point the AC took a
slightly different stand.11
In other cases the Chamber has stressed the need for the contribution of each partici­
pant in a JCE to be ‘substantial’.112For instance, in Limaj and others an ICTY TC found that
the Prosecution had not proved that the three accused persons (members of the Kosovo
Liberation Army) were liable for a joint criminal enterprise to commit in 1998 such crimes
as torture, ill-treatment, and murder in a prison camp in Kosovo (§§665-70).
It bears noting that the requirement that the contribution of a participant in a JCE
should be ‘substantial’ had not been envisaged by the ICTY AC in Tadic (AJ, 1999,
§227). This requirement seems to the present writer to be indispensable.

9.4.4 IN C ID E N T A L C R IM IN A L L IA B IL IT Y B A SE D O N FO R E SIG H T
A N D V O L U N T A R Y A S S U M P T IO N OF R ISK

(A) The notion

The third mode of responsibility concerns those participants who agreed to the main
goal of the common criminal design (for instance, the forcible expulsion of civilians
from an occupied territory) but did not share the intent that one or more members of
the group entertained to also commit other crimes incidental to the main concerted

11 It held that ‘in general, there is no specific legal requirement that the accused make a substantial contri­
bution to the JCE. However, there may be specific cases which require, as an exception to the general rule, a
substantial contribution of the accused to determine whether he participated in the JCE. In practice, the sig­
nificance of the accused’s contribution will be relevant to demonstrating that the accused shared the intent
to pursue the common purpose (§97). However, the Chamber subsequently held that in some exceptional
cases the ‘substantial’ character of a participant’s contribution is needed (§599).
12 Ibid., §667.
 200 IN T E R N A T IO N A L C R IM IN A L LAW

crime (for instance, killing or wounding some of the civilians in the process of their
expulsion). This mode of liability only arises if the participant who did not have the
intent to commit the ‘incidental’ offence, was nevertheless in a position to foresee its
commission and willingly took the risk.
A clear example in domestic criminal law of this mode of liability is that of a gang
of thugs who agree to rob a bank without killing anyone, and to this end agree to use
fake weapons. In this group, however, one of the members secretly takes real weapons
with him to the bank with the intent to kill, if need be. Suppose another participant in
the common criminal plan sees this gang member stealthily carrying those weapons.
If the armed m an then kills a teller or bank officer during the robbery, the one who saw
him take the real weapons may be held liable for robbery and murder, like the killer
and unlike the other robbers, who will only be liable for armed robbery. Indeed, he
was in a position to expect with reasonable certainty that the robber who was armed
with real weapons would use them to kill, if something went wrong during the rob­
bery. Although he did not share the mens rea of the murderer, he foresaw the event and
willingly took the risk that it might come about (plainly, he could have told the other
robbers that there was a serious danger of a m urder being committed; consequently,
he could either have taken the weapons away from the armed robber or withdrawn
from the specific robbing expedition or even dropped out of the gang).
To clarify the matter, one should perhaps distinguish between an abstract and a con­
crete foreseeability of the unconcerted crime. Arguably, for criminal liability under the
third category of JCE to arise it is necessary for the unconcerted crime to be abstractly
in line with the agreed-upon criminal offence; in addition, it is also essential that the
‘secondary offender’ had a chance of predicting the commission of the unconcerted
crime by the ‘prim ary offender’. For instance, if a param ilitary unit occupies a village
with the purpose of detaining all the women and enslaving them, a rape perpetrated
by one of them would be in line with enslavement, since treating other hum an beings
as objects may easily lead to raping them. It would, however, also be necessary for the
‘secondary offender’ to have specifically envisaged the possibility of rape (a circum­
stance that should be proved or at least inferred from the facts of the case), or at least
to have been in a position to predict the rape.
Furthermore, we should ask ourselves whether the mens rea requirement for this JCE
is the ‘secondary offender’s’ subjectiveforesight of the likelihood of the crime being com­
mitted by the ‘primary offender’ (i.e. the ‘secondary offender’ actually foresaw that the
offence would be committed), or instead objective foreseeability of that likelihood (i.e. he
ought to have foreseen that the crime was likely to be perpetrated). As the Supreme Court
of Canada rightly pointed out in two celebrated decisions concerning ‘constructive m ur­
der’ (i.e. murder imputed to a person by law from his course of actions, though his deeds
taken severally do not amount to voluntary murder), R. v. Vaillancourt (1987) and R. v.
Martineau (1990), objective foreseeability constitutes a lower threshold.13This threshold

13 See R. v. Vaillancourt, judgment of 3 December 1987, [1987] 2 S.C.R 636 (online: www.scc.lexum.umon-
treal.ca/I987/1987rcs2-636, at 24-29) and R. v. Martineau, judgment of 13 September 1990, [1990] 2. S.C.R 633
(online at: www.scc.lexum.umontreal.ca/1990/199rcs2-633, at 16-20). The facts in Vaillancourt are interesting
 PE R P E T R A T IO N A N D JO IN T C R IM IN A L EN T ER PR ISE 201

the Court in Vaillancourt considered admissible in cases of constructive m urder’,

whereas in Martineau the same C ourt held the subjective test to be more conson­
ant with principles of fundamental justice. Probably the later ruling was also dictated
by the fact that under Canadian legislation a finding of murder entails a mandatory
sentence of life imprisonment; it was therefore felt necessary to raise the threshold
of culpability for any such finding. Be that as it may, it would seem that at the inter­
national level the lower requirement of objective foreseeability is upheld by case law,
as proved by the cases that I will consider below. In other words, at the international
level what is required is not that the ‘secondary offender’ actually predicted that the
‘prim ary offender’ would engage in unconcerted criminal conduct; the test is rather
whether a m an of reasonable prudence would have forecast that conduct, under the
circumstances prevailing at the time. Three reasons seem to warrant the acceptance of
a lower threshold at the international level. First, the crimes at issue are massive and
of extreme gravity; moreover, they are normally perpetrated under exceptional cir­
cumstances of armed violence. Under these circumstances one can legitimately expect
that combatants and other persons participating in armed hostilities or involved in
large-scale atrocities be particularly alert to the possible consequences of their actions.
Secondly, the gravity of the crimes at issue makes it necessary for the world commu­
nity to prevent and punish serious misconduct to the maxim um extent allowed by
the principle of legality. Thirdly, in ICL there is no fixed scale of penalties; courts are
therefore free duly to appraise the level of culpability of the accused and impose a con­
gruous sentence accordingly.
Some commentators have noted that the foreseeability standard on which this form
of liability is based is unreliable, so much so that through such a standard—it has been
claimed—the doctrine introduces a ‘form of strict liability’. It has also been contended
that this category of criminal enterprise disregards the necessity that a person be held
guilty only if his culpability has been proven; or in other words, that the causal link
between his conduct and mens rea on the one side, and the crime, on the other, be
proved. Based on that doctrine, one would find a person guilty of, say, murder, even if
that person lacked the requisite subjective element (intent or dolus) proper to the crime
and only entertained a lesser form of mens rea (foreseeability plus willingly taking the
risk that the crime be perpetrated; that is dolus eventualis). It would follow that the
causal link between mens rea and conduct on the one side and the event or crime, on the
other, would be lacking. Thus—so the objection continues—under certain conditions,

During an armed robbery, appellants accomplice shot and killed a client. He then escaped but the appellant
was arrested and convicted of second degree murder (i.e. unlawful taking of human life with malice but with­
out deliberation or premeditation) as a party to the offence. However, the two had previously agreed to commit
the robbery armed only with knives; when on the night of the robbery the accomplice arrived with a gun, the
appellant insisted that it be unloaded; the accomplice removed three bullets from the gun and gave them to the
appellant, whose glove containing the three bullets was later recovered by the police at the scene of the crime. The
Court upheld the appeal against conviction and ordered a new trial. As Judge L’Heureux-Dube later noted in his
dissenting opinion in Martineau, ‘The facts themselves in Vaillancourt negated mens rea [...] Given these facts,
it seems unlikely that Vaillancourt, or any reasonable person in his position, had reason to foresee that anyone
would be killed in the course of the robbery’ (at 29).
 202 IN T E R N A T IO N A L C R IM IN A L LAW

one would place on a par the person who deliberately brought about the death of the
victim with an individual who instead did not intend to cause such effect.
This objection is indisputably im portant,14 and can be met by propounding three
arguments.
First, the foundation of this mode of responsibility is to be found in considera­
tions of public policy; that is the need to protect society against persons who (i) band
together to take part in criminal enterprises; and (ii) while not sharing the criminal
intent of those participants who intend to commit more serious crimes outside the
common enterprise, nevertheless are aware that such crimes may be committed; and
(iii) do not oppose or prevent them. These policy considerations were aptly spelled
out by the House of Lords in 1997, in two cases decided jointly, Regina v. Powell
and another and Regina v. English15, although the cases concerned crimes committed

For critical remarks about JCE, see in particular J. D. Ohlin, ‘Three Conceptual Problems with the
Doctrine of Joint Criminal Enterprise, 5 JICJ (2007), 69-90, in particular 75-88 (this paper is, however, marred
by the insistence on the concept of conspiracy and a misapprehension of the relevant international case law); E.
van Sliedregt, ‘Joint Criminal Enterprise as a Pathway to Convicting Individuals for Genocide’, ibid., 184-207,
particularly 187-91; K. Ambos, ‘Joint Criminal Enterprise and Command Responsibility’, ibid., 159-83.
Less significant is the objection frequently heard whereby the category of JCE under discussion in fact
amounts to, or is equally objectionable as, the common law concept of 'felony-murder. Such concept, still
widespread (albeit on the wane) in such countries as the UK, some states of the USA, New Zealand and cer­
tain Australian states, is substantially different from JCE. As first enunciated by Coke in 1797 (E. Coke, The
Third Part of the Institutes o f the Laws o f England, London: Clarke and Sons, 1817, at 56), the concept entails
that if an unlawful act involves the perpetration of murder, then the individual is guilty of murder (in the
celebrated example by Coke, if a person (A), intending to steal a dear in the park of another person (B),
throws an arrow at the dear but in so doing kills a boy hidden in a bush, he is guilty of murder ‘for that act
was unlawful, although A. had no intent to hurt the boy, nor knew not of him’). The concept has been widely
criticized for it equates manslaughter (involuntary killing) to murder i.e. intentional killing of another per­
son. In the case of the JCE we are discussing the secondary offender not only is involved in a common crim­
inal plan or purpose to commit some crimes and has the intention to commit those crimes, but also actually
foresees (or is in a position to foresee) the likely perpetration of a further crime by a member of the criminal
group, and nevertheless deliberately accepts the risk of such likelihood. There is therefore here a mental
element present with regard to the perpetration of the ‘extra crime’ (dolus eventualis) that is instead absent in
the felony-murder or, if present, then only in the attenuated form of culpa (negligence). In the case of‘felony­
murder’ the agent does not figure out at all the possibility of killing a person as a result of his engaging in an
unlawful action such as theft; instead in the category of JCE we are discussing the agent is aware (or at least is
fully in a position to be aware) that a crime may be perpetrated by another person and deliberately omits to
take action (i.e. to stop or prevent that person from perpetrating the crime, or to disassociate himself from
that criminal conduct). In addition, the concept of JCE can only be relied upon on condition that the lesser
culpability of the secondary offender shall be taken into account at the sentencing stage.
15 In the first case, P„ D., and a third man went to the home of a dealer in cannabis. As soon as he opened
the door, one member of the group shot him and he died shortly afterwards. The defendants were charged
with murder on the basis ofjoint enterprise. At the trial P. gave evidence and claimed that he was present at the
scene only to buy cannabis. D. did not give evidence, but it was submitted on his behalf that he was unaware
of the presence of the gun until it was used and that P. was responsible for the shooting. Both defendants were
convicted of murder. The Court of Appeal (Criminal Division) dismissed both defendants’ appeals.
In the second case, the defendant, E., aged 15 at the time of the offence, and W. were convicted of the mur­
der of a police sergeant on the basis ofjoint enterprise. Both the defendant and W. had attacked the deceased
with wooden posts. At the trial it was the Crown’s case that the defendant was present when W. produced the
knife with which the fatal injuries were inflicted. It was maintained on the defendant’s behalf that there was
evidence that he had fled the scene before W. produced the knife. The Court of Appeal (Criminal Division)
dismissed, however, E.’s appeal.
 P E R P E T R A T IO N A N D JO IN T C R IM IN A L EN T E R PR ISE 203

at the domestic level. The speeches of Lords Steyn16 and Sutton17 are enlightening. In
their view by punishing the ‘secondary offender’ the law intends to convey the message
that he should have opposed or impeded the crime of the ‘prim ary offender’.
The second argument is more germane to strictly legal considerations. Generally
speaking, one should not neglect an im portant factor: incidental criminal liability
based on foresight and risk is a mode of liability that is consequential on (and inci­
dental to) a common criminal plan; that is, an agreement by a m ultitude of persons
to engage in illegal conduct. The ‘extra crime’ we are discussing is the outgrowth of
previously agreed or planned criminal conduct for which each participant in the com­
mon plan is already responsible. This ‘extra crime’ is rendered possible by the prior
joint planning to commit the agreed crime(s) other than the one ‘incidentally’ or add­
itionally’ perpetrated. Thus, what is at stake here is not the responsibility arising when
members of a group (for instance, a military unit) engage in lawful action (for example,
overpowering by military force an enemy fortification) and in the course of combat
one of the combatants kills a civilian or rapes a woman—a crime for which of course
he alone must bear criminal responsibility. Our discussion here turns, rather, on cases

16 His Lordship stated the following: ‘At first glance there is substance in the third argument [of counsel
for the Appellants] that it is anomalous that a lesser form of culpability is required in the case of a secondary
party, viz. foresight of the possible commission of the greater offence, whereas in the case of the primary
offender the law insists on proof of the specific intention which is an ingredient of the offence. This general
argument leads, in the present case, to the particular argument that it is anomalous that the secondary party
can be guilty of murder if he foresees the possibility of such a crime being committed while the primary
can only be guilty if he has an intent to kill or cause really serious injury. Recklessness may suffice in the
case of the secondary party but it does not in the case of the primary offender. The answer to this supposed
anomaly, and other similar cases across the spectrum of criminal law, is to be found in practical and policy
considerations. If the law required proof of the specific intention on the part of a secondary party, the utility
of the accessory principle would be gravely undermined. It is just that a secondary party who foresees that
the primary offender might kill with the intent sufficient for murder, and assists and encourages the pri­
mary offender in the criminal enterprise on this basis, should be guilty of murder. He ought to be criminally
liable for harm which he foresaw and which in fact resulted from the crime he assisted and encouraged. But
it would in practice almost invariably be impossible for a jury to say that the secondary party wanted death
to be caused or that he regarded it as virtually certain. In the real world proof of an intention sufficient for
murder would be well nigh impossible in the vast majority of joint enterprise cases. Moreover, the proposed
change in the law must be put in context. The criminal justice system exists to control crime. A prime func­
tion of that system must be to deal justly but effectively with those who join with others in criminal enter­
prises. Experience has shown that joint criminal enterprises only too readily escalate into the commission
of greater offences. In order to deal with this important social problem the accessory principle is needed and
cannot be abolished or relaxed. For these reasons I would reject the arguments advanced in favour of the
revision of the accessory principle’ (8).
17 ‘My Lords, I recognise that as a matter of logic there is force in the argument advanced on behalf of the
appellants, and that on one view it is anomalous that if foreseeability of death or really serious harm is not
sufficient to constitute mens rea for murder in the party who actually carries out the killing, it is sufficient
to constitute mens rea in a secondary party. But the rules of the common law are not based solely on logic
but relate to practical concerns and, in relation to crimes committed in the course of joint enterprises, to the
need to give effective protection to the public against criminals operating in gangs. As Lord Salmon stated
in Reg. V. Majewski [1977] A.C. 443,482e, in rejecting criticism based on strict logic of a rule of the common
law, “this is the view that has been adopted by the common law of England, which is founded on common
sense and experience rather than strict logic”. In my opinion there are practical considerations of weight and
importance related to considerations of public policy which justify the principle stated in Chan Wing-Siu v.
The Queen [1985] A.C. 168 and which prevail over considerations of strict logic’ (15).
 204 IN T E R N A T IO N A L C R IM IN A L LAW

where a plurality of persons agrees to perpetrate one or more crimes for which they
all bear responsibility and in addition one of them commits a further crime. Here,
it is plain, the additional crime is premised on the existence of a concerted criminal
purpose. In other words, there exists a causal link between the concerted crime and
the ‘incidental’ crime: the former constitutes the preliminary sine qua non condition
and the basis of the latter (although, with regard to the latter, only the participant that
evinced knowledge and risk-taking shares the liability of the other participant who
perpetrated the ‘additional’ offence). To clarify further the nexus between the two
categories of crimes at issue, it could perhaps prove useful to insist on the distinction
between abstract and concrete (or specific) foreseeability, suggested above (9.4.4).18

18 The fact that the incidental crime may be based on a nexus with the concerted crime was clearly empha­
sized by various courts. Suffice it to mention here the decision of the Italian Court of Cassation in D’Ottavio
and others (decision of 12 March 1947).Two former Yugoslav war prisoners, who had escaped from a concen­
tration camp, were suddenly surrounded by four local individuals near an Italian village. While one of them
managed to flee, the other man was hit by two gunshots fired by D’Ottavio with his hunting rifle. The four
aggressors then immediately left the scene. The injured man later died. The Teramo Court of Assize held that
the accused had not intended to kill. With regard to the defendants other than D’Ottavio, it applied Article
116 of the Italian Criminal Code, providing that ‘Where the crime committed is different from that willed by
one of the participants, also that participant answers for the crime, if the fact is a consequence of his action or
omission. If the crime committed is more serious than that willed, the penalty is decreased for the participant
who willed the less serious offence.’ On appeal, the Court of Cassation held that: ‘The complaint concerning
the application of Article 116 is also without merit. By virtue of this provision, where the crime committed is
other than the one willed by one of the participants, also that participant is accountable for the crime if the
criminal result is a consequence of his action or omission. In order for a criminal event to be held to constitute
the consequence of the participant’s action, it is necessary that there be a causation nexus—which is not only
objective but also psychological—between the fact committed and willed by all the participants and the dif­
ferent fact committed by one of the participants. This is so because the participant’s responsibility envisaged
in Article 116 is grounded not in the notion of collective responsibility [...] but in the fundamental principle
of concurrence of interdependent causes, upheld and specified in Articles 40 and 41 of the Criminal Code.
By virtue of the latter principle all the participants answer for a crime both where they are the direct cause of
the crime and where they are the indirect cause, in accordance with the canon causa causae est causa causati
[the cause of a cause is also the cause of the thing caused; i.e. whoever voluntarily creates a situation bringing
to, or resulting in, criminal conduct is accountable for that conduct whether or not he willed the crime]. It is
this concurrence of causes that also in this particular case of participation re-establishes the requirement of
legal identity of the fact that is the precondition of the cooperation “in the commission of same crime". This
identity is at least generic if not specific in that all the defendants have effectively contributed to the first crime
that was the cause of the second. Here lies the nexus of objective causation: all participants have directly
cooperated in the crime of attempted illegal detention of persons (provided for in Article 605 of the Criminal
Code) by surrounding and chasing two fugitive prisoners of war, armed with a gun and a musket for the pur­
pose of unlawfully capturing them. This crime was the indirect cause of the subsequent and connected event
consisting of the rifle shot that D’Ottavio alone fired at one of the fugitives, a rifle shot that caused a wound
followed by death (see Article 584 on manslaughter). There also exists a psychological causation in that all the
participants shared the conscious will to engage in an attempt to unlawfully detain a person while foreseeing
a possible different crime, as can be inferred from the use of weapons: it was to anticipate that one of them
might have shot at the fugitives with a view to achieving the common purpose of capturing them.’
It would seem that the Court rightly stressed the causal link between the concerted and the not-envisaged
crime, by pointing to the fact that this causal link related to the objective element of the crime at issue.
However, there is ultimately a link with regard to the subjective element as well. The participant in the JCE to
commit a specific crime or set of crimes is put in the position to foresee the further, unconcerted crime, on
account of his joining the criminal enterprise to commit the agreed upon crime. Although he did not share
the intent of the participant that engaged in the further criminal conduct, he had predicted that conduct and
 PE R P E T R A T IO N A N D JO IN T C R IM IN A L E N T ER PR ISE 205

The third response to the objections under discussion is directed to emphasize that
the basic proposition suggested here on the basis of existing case law (that any partici­
pant in a JCE is also guilty for acts by another participant, under the conditions set out
in the case law) is premised on the proposition that at the sentencing stage one must,
however, take into account the different degrees of culpability of the participants. The
lesser form of mens rea of the ‘secondary offender’ shall be taken into account by met­
ing out a lighter sentence than that inflicted on the participant who materially perpe­
trated the offence not envisaged in the criminal plan. Both participants are guilty, but
the one who did not materially perpetrate the further crime must receive a less stiff
sentence on account of his lesser culpability.

(B) Limitations of the category at issue

There exist two im portant qualifications to the application of the third class of JCE
under discussion.
First, resorting to such class would be intrinsically ill-founded when the crime com­
m itted by the ‘prim ary offender’ requires special or specific intent (dolus specialis);
that is, the crime charged is one of genocide, persecution, or aggression (it is com­
mon knowledge that for genocide the intent to destroy a ‘protected group’ in whole
or in part is required; persecution presupposes the intent to discriminate on one of
the requisite grounds; aggression, at least in the opinion of some commentators,19 is
grounded in the intent to appropriate a foreign territory or to obtain economic advan­
tages, or to interfere with the internal affairs of the victim state; see above, 7.3.3(B)). In
these cases the ‘secondary offender’ may not share—by definition—that special intent
(otherwise one would fall under the first and second class of JCE), even though enter­
taining such intent is a sine qua non condition for being charged with the crime. He
may therefore not be accused of such crime under the doctrine at issue. This propos­
ition is based on two grounds. First, on a logical impossibility: one may not he held
responsible for committing a crime that requires special intent (in addition to the
intent needed for the underlying crime) unless that special intent can be proved, what­
ever mode of responsibility for the commission of crimes is relied upon (this leaves out
aiding and abetting, where it suffices to prove that the offender has made a substantial
contribution to the commission of the crime by others, had knowledge of the crime,
and intentionally provided assistance to its perpetration). Secondly, admittedly who­
ever is liable under the third category of JCE has a distinct mens rea from that of the
‘prim ary offender’; nevertheless, as the ‘secondary offender’ bears responsibility for
the same crime as the ‘prim ary offender,’ the ‘distance’ between the subjective element
of the two offenders must not be as drastic as in the case of crimes requiring special

willingly taken the risk that it might occur. There lies his culpability. He could have prevented the further
crime, or disassociated himself from its likely commission. His failure to do so entails that he too must be
held guilty. See also Mannelli and others.
19 S. Glaser, ‘Quelques remarques sur la definition de l’agression en droit international penal’in Festschrift
für Ih.Rittler (Aalen: Verlag Scientia, 1957), at 388-93 ; Idem ‘Culpabilite en droit international penal’, in 99
Hague Recueil (1960-1), at 504-5.
 206 IN T E R N A T I O N A L C R I M I N A L LAW

intent. Otherwise the crucial notions of personal culpability’ and ‘causation’ would
be torn to shreds.20 For such crimes the ‘secondary offender’ could only be charged—it
is submitted—with aiding and abetting the main crime (needless to say, subject to the
condition that the requirements of aiding or abetting the commission of one of the
three classes of aforementioned crimes are met).
Let us now consider the second qualification to the application of the third class of
JCE under discussion. Mature legal systems make it possible to take account of the
lesser degree of culpability of the ‘secondary offender’ by qualifying his culpability-
through a charge less than that against the ‘prim ary offender’. If the latter has engaged
in murder while conducting a concerted unlawful deportation of civilians, the ‘sec­
ondary offender’ could be accused of manslaughter. This different charge would take
into account the lesser degree of culpability of that offender.
Unfortunately ICL is a rudimentary body of law, which allows for such sophisticated
distinctions or gradations only to a very limited extent. In short, one cannot charge a
lesser offender with an offence belonging to a different category of international crimes;
for instance, one cannot charge the ‘primary offender’ with murder as a crime against
humanity and the ‘secondary offender’ with murder as a war crime. This would indeed
be erroneous, for the two categories show different features; the offences at issue belong
either to one category (for instance, crimes against humanity) if the requisite conditions
are met (chiefly, the existence of a context of widespread or systematic practice), or to the
other. Furthermore, laying different charges within the same category of international
crime is logically possible only with regard to some classes of underlying offences. As
classes of offences where a gradation is possible, one can mention: murder and m an­
slaughter (as a war crime, or a crime against humanity); wilful killing (as a grave breach);
and unlawful killing (as a war crime in an international armed conflict);21 rape and

20 In 2004 the ICTY AC took a contrary view in Brdanin, with regard to genocide. In its Decision on
Interlocutory Appeal of 19 March 2004 it held that ‘provided that the standard applicable to that head of
liability [the third category of JCE], i.e. “reasonably foreseeable and natural consequences” is established,
criminal liability can attach to an accused for any crime that falls outside of an agreed upon joint criminal
enterprise (§9). It went on to say that "The Trial Chamber erred by conflating the mens rea requirement of
the crime of genocide with the mental requirement of the mode of liability by which criminal responsibility
is alleged to attach to the accused’ (§10). The AC thus reversed a prior decision of the TC (.Brdanin, Decision
for Acquittal Pursuant to Rule 98 bis, 28 November 2003), which had held (correctly, in my opinion) that
the specific intent required for genocide cannot be reconciled with the mens rea required for a conviction
pursuant to the third category of JCE. The latter consists of the Accused’s awareness of the risk that genocide
would be committed by other members of the JCE. This is a different mens rea and falls shortofthe threshold
needed to satisfy the specific intent required for a conviction for genocide under Article 4(3)(a) [of the ICTY
Statute] (§57). In 2005, in Kvocka and others, the same AC limited the need for sharing the special intent to
the first category of JCE. It ‘affirmed’ ‘the Trial Chamber’s conclusion that participants in a basic or systemic
form of joint criminal enterprise must be shown to share the required intent of the principal perpetrators.
Thus, for crimes of persecution, the Prosecution must demonstrate that the accused shared the common dis­
criminatory intent of the joint criminal enterprise. If the accused does not share the discriminatory intent,
then he may still be liable as an aider and abettor if he knowingly makes a substantial contribution to the
crime’ (§110). This proposition was taken up by an ICTR TC in Simba (at §388).
21 This proposition is based on the assumption that grave breaches may only be committed in inter­
national armed conflicts, a position taken in 1995 by the ICTY AC in Tadic (IA), but probably no longer valid
under current international customary law.
 PE R P ET R A T IO N A N D JO IN T C R IM IN A L EN T ER PR ISE 207

sexual violence (as a war crime or a crime against humanity); and torture and inhum an
or degrading treatment (as a war crime or a crime against humanity). For other under­
lying offences it would seem difficult to apply such gradations of culpability and hence
of charging.

(C) Case law

The first case where this category of JCE was raised is Tadic (AJ, 1999). According to
the Prosecution the TC had erred in finding that the accused could not be charged
with the killing of five men in the village of Jaskici, when he participated in the attack
on that village and the village of Sivci on 14 June 1992, because there was no evidence
showing that he had killed or taken part in the killing of those five men. For the
Prosecution ‘the only conclusion reasonably open from all the evidence is that the
killing of five victims was entirely predictable as part of the natural and probable con­
sequences of the attack on the villages of Sivci and Jaskici on 14 June 1992’ (§175). The
Defence argued instead that the TC correctly found that ‘it was a possibility that the
five victims in Jaskici were killed by another, distinct group of armed men, especially
as nothing [was] known as to who shot the victims or in what circumstances’ (§176).
As for the Prosecution’s common purpose submission, the Defence contended that it
would have to be shown that the common purpose in which the Appellant allegedly
took part included killing as opposed to ethnic cleansing by other means’ (§177).
The AC upheld the Prosecution’s submissions after engaging in an elaborate outline
of the notion of common purpose or JCE in ICL.22 Based on this notion, the AC found
that in the case at issue the defendant had taken part in a common plan to commit
inhum ane acts against the non-Serb civilian population in the Prijedor region in 1992.
He was an armed member of the armed group that took part in the attack and com­
mitted several crimes. He must have been aware that the actions of the group of which
he was a member were likely to lead to [...] killings, but he nevertheless willingly took
that risk’ (§232). The AC therefore found the defendant guilty. Subsequently the TC,
to which the case had been remitted for sentencing purposes, held that for the murder
of the five Muslims, Tadic was simultaneously guilty of a grave breach, a war crime,
and a crime against humanity. It sentenced him to 24 years imprisonment for the
grave breach and the war crime and 25 years for the crime against humanity, with the
sentences to be served concurrently23 (in its previous judgment, where the murder of

22 ‘With regard to the first category, what is required is the intent to perpetrate a certain crime (this being
the shared intent on the part of all co-perpetrators). With regard to the second category [...], personal know­
ledge of the system of the treatment is required (whether proved by express testimony for a matter of reason­
able in view inference from the accused’s position of authority), as well as the intent to further this common
concerted system of ill-treatment. With regard to the third category, what is required is the intention to
participate in and further the criminal activity or the criminal purpose of a group and to contribute to the
joint criminal enterprise or in any event to the commission of a crime by the group. In addition, responsibil­
ity for a crime other than the one agreed upon in the common plan arises only if, under the circumstances of
the case, (i) it was foreseeable that such a crime might be perpetrated by one or are the members of the group
and (ii) the accused willingly took that risk' (§228).
23 ICTY, TC, Sentencing Judgment, §§15-18, 27-9 and 32 E and G.
 208 INTERNATIONAL CRIMINAL LAW

the five Muslims had not been imputed to Tadic, the TC had sentenced him to 20 years’
imprisonment).24 The AC subsequently reduced the sentence to 20 years’ imprison­
ment, both because it held the previous sentence to be excessive with regard to the
relatively minor position of the accused, and because in its view ‘there is in law no dis­
tinction between the seriousness of a crime against hum anity and that of a war crime’.
It was consequen tly wrong to consider the same offence as more grave if regarded as a
crime against hum anity than as a war crime.25
The question of this category of criminal liability arose again in Krstic, although
only tangentially, before the TC.26 The essential features of the category, as set out in
Tadic (AJ, 1999) were restated by the ICTY AC in Vasiljevic (§99), Kvocka and others
(§83), as well as in Babic (§27). In Stakic the AC, after reversing the TC’s ruling based on
the notion of co-perpetratorship’, held that the accused, in holding important positions
such as President of the Crisis Staff, had participated in a JCE to commit crimes of per­
secution, forced displacement, and ill-treatment in detention camps against Muslims
in the Prijedor area in Bosnia-Herzegovina. It then held that the accused bore criminal
liability under the third category of JCE for crimes not agreed upon, namely killings in
detention camps, transportation to camps of the non-Serb civilian population, and kill­
ings by the Serb armed military and police forces. The AC concluded that the accused
was responsible under the third head of JCE for the crimes of murder (as a war crime and
a crime against humanity) and extermination as a crime against humanity. It is notable
that the Chamber insisted on the requirement of dolus eventualis and held, based on the
findings of the TC, that this form of mens rea did exist in the case at issue (§§93-7).
An interesting application of the third category of JCE was made by an ICTY TC in
Blagojevic and Jokic. After noting that where the objective of a JCE changes in time, a
new and distinct JCE m aybe established, the TC pointed out that, with the establish­
ment of such new JCE a participant in the enterprise shall not incur responsibility for
criminal acts beyond the scope of the enterprise in which he had agreed to participate,
but only for those acts that are ‘natural and foreseeable consequences’, thereby falling
under the third category of JCE (§701).

24 ICTY, TC, Sentencing Judgment. 25 ICTY, AC, §§55-8,69 and 76(3).

26 As pointed out above, the Chamber held that the defendant had participated in a JCE to commit geno­
cide. Nevertheless, the Chamber relied upon the third category of criminal enterprise with regard to some
crimes committed against the persons who had escaped the massacre. It held that it was not proved that vari­
ous crimes committed against Muslims fleeing Srebrenica had been agreed upon in the criminal plan. They
were nevertheless to be imputed to the defendant—so held the Chamber—because they were the foreseeable
consequence of the policy of forcible expulsions that was part of the criminal plan: ‘The Trial Chamber is
not, however, convinced beyond reasonable doubt that the murders, rapes, beatings and abuses committed
against the refugees at Potocari were also an agreed upon objective among the members of the joint criminal
enterprise. However, there is no doubt that these crimes were natural and foreseeable consequences of the
ethnic cleansing campaign. Furthermore, given the circumstances at the time the plan was formed, General
Krstic must have been aware that an outbreak of these crimes would be inevitable given the lack of shelter,
the density of the crowds, the vulnerable condition of the refugees, the presence of many regular and irregu­
lar military and paramilitary units in the area and the sheer lack of sufficient numbers of UN soldiers to pro­
vide protection. In fact, on 12 July, the VRS organised and implemented the transportation of the women,
children and elderly outside the enclave; General Krstic was himself on the scene and exposed to firsthand
knowledge that the refugees were being mistreated by VRS or other armed forces’ (§616).
 PERPETRATION AND JOINT CRIMINAL ENTERPRISE 209

Finally, it should be mentioned that the ICTY AC has placed a broad interpretation
on the category of JCE at issue. In 2004 in Brdanin it held that this category of JCE can
also apply when acts of genocide are committed by the ‘prim ary offender’.27 In 2006,
in Karemera and others the ICTR AC held that this category of crim inal liability can
also cover crimes committed by fellow participants in a vast joint criminal enterprise
where crimes committed by the fellow participants are ‘structurally or geographic­
ally remote from the accused’.28 The same view was taken in 2007 by the ICTY AC in
Brdanin with regard to the category of JCE we are discussing (AJ, §§420-5).

9.4.5 T H E Q U E S T I O N OF W H E T H E R T H E ‘P H Y S IC A L P E R P E T R A T O R
S H O U L D A L S O BE P A R T O F T H E JCE

As we saw above (9.4.2(B)), in Brdanin the issue was raised of the relations between
members of a JCE and persons not part to the JCE who nevertheless carry out crimes
in execution of the JCE (deportation and forcible transfer of Bosnian Muslim or
Croat civilians). The question is as follows: do such perpetrators (henceforth physical
perpetrators) need to share the joint crim inal purpose for the members of the JCE
to be answerable for the crimes perpetrated? The TC answered in the negative (TJ,
§§344- 56), while the AC in the affirmative (AJ, §§410-19, 426-32). It is therefore
appropriate to dwell on the question of the relations between members of a JCE and
organized groups that commit crimes in execution of a common criminal purpose.
Normally members of a JCE make up fairly small groups and are persons operating
at the same level, even though in different capacities. Hence no serious problem arises:
each of them is responsible for the concerted criminal actions, even if such actions are
performed only by one member ofthe JCE. However, there may be cases where the mem­
bers of the JCE constitute a larger group and form part of a hierarchically constituted
organization or structure. This is typically the case for JCE II (participation in a com­
mon criminal plan within an institutional framework). Here, however, only those who
knowingly make a substantial contribution to the pursuit of common criminal purposes
are personally liable. Hence for all of them it is required that they be part to the JCE.
The problem becomes complicated when the criminal plan is agreed upon by a num ­
ber of members of a political or military group, and one of these members carries out
the common criminal purpose by ordering or instigating subordinate military units
outside the JCE to commit some or all of the crimes envisaged in the JCE.
One should distinguish between the legal position of (a) the member of the JCE that orders
or instigates outsiders to commit the crimes; and (b) that of the other members of the JCE.
To my mind the member of the JCE ordering or instigating the commission of crimes
may be responsible under two distinct heads of liability. He is responsible for (1) the
JCE to commit other crimes that may have been perpetrated by himself as well as other
members of the JCE; and for (2) ordering and instigating the crimes perpetrated by
the subordinates. These subordinates need not, of course, share the common criminal

27 Brdanin, Decision on Interlocutory Appeal, at §§9-10.

28 ICTR AC, Karemera and others, Decision on Jurisdictional Appeals: Joint Criminal Enterprise, at §§11-18.
 210 INTERNATIONAL CRIMINAL LAW

purpose (this is what occurred in Brdanin, according to the TC, which rightly found the
defendant guilty of ordering and instigating the crime of deportation and forced expul­
sion of Bosnian Muslims and Croats, perpetrated by the army: §§359-69). If brought to
trial, such subordinates are liable for the perpetration of the crime at issue.
Let us now move on to situation (b). Here the following question must be asked:
does a member of the JCE other than the member that orders or instigates subordinate
troops or paramilitary units or police officers (not part to the JCE) to perpetrate crimes
in consonance with the criminal purposes agreed by members of the JCE, bear respon­
sibility for the crimes perpetrated by the executioners? The answer may only be given
in light of general principles of international criminal law, in particular the principle
of personal criminal responsibility (indeed the judicial precedents relied upon by the
AC in Brdanin (AJ §§393-404) are not germane to the question under discussion).29 In
accordance with these principles the member of the JCE may only be held responsible
for those crimes if (i) when concerting the crime to be perpetrated in execution of the
JCE he had agreed to the physical perpetration of crimes by persons who, albeit outside
the JCE, could, however, act upon the orders of one of the members of the JCE (in this
case JCE I would be applicable); or (ii) he anticipated the risk that another member of
the JCE might order or instigate persons outside the JCE to perpetrate crimes and will­
ingly ran that risk (JCE III). It would not be sound to hold the member at issue liable

29 They are two cases brought before US Military Tribunals sitting at Nuremberg: Altstotter and others
(so-called Justice case) and Greifelt and others (so-called RUSHA case). As the AC admitted in Brdanin (AJ,
§393), in neither case did the Tribunals use the expression ‘JCE’. What matters, however, is that neither
judgment relied upon the notion of JCE. In the former, faced with crimes planned, ordered or committed by
members of the Ministry of Justice, the Tribunal adopted traditional notions of criminal responsibility, as is
apparent from the following passage: The defendants are not now charged with conspiracy as a separate and
substantive offense, but it is alleged that they participated in carrying out a governmental plan and program
for the persecution and extermination of Jews and Poles, a plan which transcended territorial boundaries as
well as the bounds of human decency. Some of the defendants took part in the enactment of laws and decrees
the purpose of which was the extermination of Poles and Jews in Germany and throughout Europe. Others,
in executive positions, actively participated in the enforcement of those laws and in atrocities, illegal even
under German law, in furtherance of the declared national purpose. Others, as judges, distorted and then
applied the laws and decrees against Poles and Jews as such in disregard of every principle of judicial behav­
ior. The overt acts of the several defendants must be seen and understood as deliberate contributions toward
the effectuation of the policy of the Party and State. The discriminatory laws themselves formed the subject
matter of war crimes and crimes against humanity with which the defendants are charged. The material
facts which must be proved in any case are: (1) the fact of the great pattern or plan of racial persecution and
extermination; and (2) specific conduct of the individual defendant in furtherance of the plan. This is but
an application of general concepts of criminal law. The person who persuades another to commit murder, the
person whofurnishes the lethal weapon for the purpose of its commission, and the person who pulls the trigger
are all principals or accessories to the crime (1063; emphasis added). The reason why the Tribunal did not
discuss the mental state of those who executed death sentences and other criminal acts agreed upon and
planned by the defendants is simply that those executioners so acted following orders by the defendants: who
were hence responsible not for JCE to commit persecution but for ordering persecution. Similarly in Greifelt
and others the Tribunal convicted the defendants of ordering and instigating the kidnapping of children of
foreign nationals, taking away foreign infants, executing in concentration camps foreigners and so on. As,
the Tribunal put it: ‘[i]t is no defense for a defendant to insist, for instance, that he never evacuated popula­
tions when orders exist, signed by him, in which he directed that the evacuation should take place. While in '
such a case the defendant might not have actually carried out the physical evacuation in the sense that he did
not personally evacuate the population, he nevertheless is responsible for the action, and his participation by
instigating the action is more pronounced than that of those who actually performed the deed’ (153).
 PERPETRATION AND JOINT CRIMINAL ENTERPRISE 211

even when the agreement (or consent) or the anticipation and deliberate taking of risk
are lacking. In such case the basic pre-condition of liability for JCE would be lacking,
and to hold the member responsible for the crimes committed by the physical perpet­
rator would be contrary to the principle of personal criminal responsibility.3031
Of course, also in the case I have just discussed the member of the JCE that ordered
or instigated subordinates is responsible for ordering and instigating the crimes,
although he did so in consonance with or in execution of a JCE (which in this respect
would not be relevant to the establishment of guilt of the accused, whereas it might
perhaps have some relevance to the setting of penalty).

9.4.6 T H E D I F F E R E N C E B E T W E E N JC E A N D A I D I N G A N D A B E T T I N G

It has been objected that the doctrine of JCE does not clearly distinguish between the
responsibility of a participant in JCE and that of an aider and abettor. Moreover, that
doctrine would even go so far as to foist a greater weight upon a person responsible for
aiding and abetting than on a participant in a JCE.
In fact a major difference between the two categories of persons does exist. It lies in
their respective mens rea (as for actus reus, in both cases a substantial contribution
is required, as I shall point out below with regard to JCE). The participant in a JCE
(i) takes part in a common criminal plan or purpose and shares a common intent to
perpetrate a crime (murder, forced expulsion, persecution, and so on); or (ii) by will­
ingly and knowingly participating in an institutional criminal framework, expressly
or implicitly evinces his sharing the criminal conduct in which that institutional
framework engages; or else (iii) in addition to adhering to a criminal plan and shar­
ing the intent to commit a crime, willingly runs the risk that another participant may
intentionally perpetrate a further crime that the former had foreseen.
In contrast, as we shall see when discussing aiding and abetting (see infra, 10.1), he
who aids and abets does not share, either at the outset or later, the criminal intent of the
perpetrator, although he is cognizant that the perpetrator intends to commit a crime;
the aider and abettor only intends to assist the perpetrator in the commission of a crime.
This is why, in principle, the criminal liability of the aider and abettor is more tenuous (or
less weighty) than that of a participant in a common criminal enterprise. As the ICTY
AC put it in a number of cases, aiding and abetting ‘generally involves a lesser degree of
individual criminal responsibility than co-perpetration in a joint criminal enterprise.
It should be added that, according to ICTY case law, it would be wrong to speak of
‘aiding and abetting a JCE’, for whenever a person intends to assist in the commission
of crimes by a group of persons involved in a JCE, that person should more correctly
be held liable for participation in the JCE.32

30 For a similar view, see the Partly Dissenting opinion of Judge Shahabuddeen in Brdanin (AJ) (§§4-13).
The contrary view is advanced by Judge Meron in his Separate Opinion in the same case (§§3—8).
31 Krnojelac (AJ, §75), Vasiljevic (AJ, §102), Kvocka and others (AJ, §92).
32 ICTY AC, Milutinovic and others, Decision on Dragoliuh Ojdanics Motion Challenging Jurisdiction-
Joint criminal Enterprise, §20; Kvocka and others (AJ, §91).
 212 INTERNATIONAL CRIMINAL LAW

9.4.7 T O W H A T E X T E N T C A N T H E IC C RELY U P O N T H E
D O C T R I N E O F JCE?

Hie ICC Statute does not contain a provision that regulates JCE in detail as a mode
of responsibility. That such form of criminal liability is implicitly perm itted under
the Statute can however be inferred from Article 25(1),33 which generically states that
criminal responsibility for any of the crimes covered by the Statute is incurred by any­
body committing a crime’ ‘jointly with another person. This provision, in addition to
co-perpetration (the same crime is committed by a plurality of persons, who perform
the same criminal act; see above, 9.3), also covers JCE. However, the ICC Statute goes
further, for, although in envisaging a different mode of liability (outsider’s contribu­
tion to a JCE; see below), it explicitly refers to the commission or attempted commis­
sion of such a crime [within the jurisdiction of the Court] by a group of persons acting
with a common purpose’ (Article 25(3)(d)).
As for the mens rea required for JCE under the Statute, one can refer to the general
provision of Article 30 (on the mental element of the crimes covered by the Statute),
which requires ‘intent or knowledge’. Should one hold the view that consequently the
Statute of the ICC always requires intent as the necessary subjective element necessary
for a finding of criminal liability, whatever the mode of responsibility, it would follow
that the ICC, while generally empowered to rely upon the doctrine of JCE, would be
barred from applying the third category referred to above.34
However, Article 30, before setting out the two mental elements of intent or know­
ledge, contains a general clause (‘unless otherwise provided’) that leaves other subject­
ive frames of mind unaffected, so long as they are provided for or required by other
provisions of the Statute or by customary international law.35 Hence the contention
can be made that dolus eventualis or recklessness for the third form of the JCE is not
excluded by the ICC Statute.
This interpretation would be justified by the need to punish criminal conduct that
otherwise would not be regarded as culpable. In addition, it would not be contrary to
the principle of personal culpability, for in any case the person at issue (i) would be

33 This provisions stipulates that:

‘In accordance with this Statute, a person shall be criminally responsible and liable for punishment for a
crime within the jurisdiction of the Court if that person:
(a) Commits such a crime, whether as an individual, jointly with another or through another person,
regardless of whether that other person is criminally responsible;
(b) Orders, solicits or induces the commission of such a crime which in fact occurs or is attempted;
(c) For the purpose of facilitating the commission of such a crime, aids, abets or otherwise assists in its
commission or its attempted commission, including providing the means for its commission;
(d) In any other way contributes to the commission or attempted commission of such a crime by a group of
persons acting with a common purpose. Such contribution shall be intentional and shall either:
(i) Be made with the aim of furthering the criminal activity or criminal purpose of the group, where
such activity or purpose involves the commission of a crime within the jurisdiction of the Court; or
(ii) Be made in the knowledge of the intention of the group to commit the crime’ (emphasis added).
34 It would seem that this is the view taken by the ICC Pre-Trial Chamber in Lubanga (§§322-67).
35 See G. Werle and F. Jessberger, ‘Unless Otherwise Provided’ in 3 JIC} (2005), 35-55.
 PERPETRATION AND JOINT CRIMINAL ENTERPRISE 2 13

guilty of intentionally participating in a crim inal purpose or plan; (ii) his mens rea
concerning the additional, not previously concerted crime, would have to be proved
by the Prosecution; and (iii) his lesser culpability would have to be taken into account
at the sentencing stage.
It should be added that, contrary to what various authors, including the present
one, have either implicitly or expressly contended,36 the gist of Article 25(3)(d) is the
regulation not of JCE but rather of a different mode of responsibility. This consists in
the fact that a person outside the crim inal group com m itting (or attem pting to com­
mit) a crime contributes to the perpetration of such crime without being a member
of the crim inal group. It would seem that such contribution is different from aiding
and abetting. Indeed, the aider and abettor intends to assist in the commission of a
crime by others but does not share the crim inal intent of the perpetrator (see 10.1
and 9.4.6). Here, instead, the outside contributor’ either (a) intends to further the
crim inal action (hence is aware of and shares the crim inal intent of the group), or
(b) simply knows, that is, is aware of, the crim inal intent of the group. In the former
instance, the outside contributor’, by sharing the criminal intent of the group only
distinguishes him self from members of the JCE in that he is not part of the criminal
agreement (neither at the moment when such agreement is made nor later). In the
latter instance, that is in the category (b), the outside contributor’ distinguishes him ­
self from the aider and abettor only in that he aides and abets a whole crim inal group
(that is, a multiplicity of persons) and not a single perpetrator. Otherwise, there is no
distinction between the two classes of persons assisting in the commission of crimes
by others.
Probably the inclusion of this new mode of liability is justified by its origin, namely
the fact that the provision was taken up from Article 2(3) of the 1997 International
Convention on the Suppression of Terrorist Bombing. The needs of the fight against
widespread and increasingly dangerous terrorist criminality warranted the expansion
of responsibility to these forms of external assistance’. The ICC Statute rather uncrit­
ically restated that provision of the Terrorist Bombing Convention.37

36 For instance see W.A. Schabas, An Introduction to the International Criminal Court, 3rd edn
(Cambridge: Cambridge University Press, 2007), at 211-13; K. Ambos, ‘Joint Criminal Enterprise and
Command Responsibility’ in 5 JICJ (2007), 172-3; A. Cassese, ‘The Proper Limits of Individual Responsibility
under the Doctrine of Joint Criminal Enterprise’, ibidem, at 132. See also the ICTY AJ in Tadic (1999, at §222).
On Article 25(3)(d) see also K. Ambos in O. Triffterer (ed), Commentary, at 483-6 as well A. Eser, in
Cassese, Gaeta, Jones (eds), The Rome Statute, I, 802-3.
37 The category o f‘outsider contributor’ to JCE is in some respects not dissimilar from the category of
‘external participation in mafia crimes’ (concorso esterno in associazione mafiosa), set forth by Italian courts
(see P. L. Vigna, ‘Fighting organized Crime, with particular reference to Mafia Crimes in Italy, in 4 JICJ
(2006), 526-7; according to this author the criminal offence at issue ‘covers cases where a person, although
not a part and parcel of the structure of a criminal organization and free from any link of subjection to
the association, nevertheless provides the association with a contribution which is specific, conscious and
voluntary. Such contribution must however be causally relevant to the strengthening of the criminal associ­
ation and aimed at the implementation (albeit partial) of the criminal plan.’ (ibidem).
 10
OTHER MODES OF LIABILITY

10.1 AIDING AND ABETTING

A person may participate in a crime without sharing the criminal intent of the prin­
cipal perpetrator, but simply by assisting him in the commission of a crime. In aiding
and abetting the objective element is constituted by practical assistance, encourage­
ment, or moral support, by the accessory to the principal (namely the author of the
main crime); in addition, such assistance, support, etc. must have a substantial effect
on the perpetration o f the crime.
Thus, unlike some instances that we will consider infra (see, for instance, 10.2),
aiding and abetting does not necessarily presuppose that the aider and abettor shares
a common plan or purpose with the principal or his criminal intent or other form of
mens rea; as stated by the ICTY AC in Tadic, ‘the principal may not even know about
the accomplice’s contribution’ (§229). W hat is required is that the person supporting
or assisting in the crime be aware that his action helps the perpetrator in the com­
mission of the crime, and intend to encourage such commission (on this subjective
element, see further below).
The substantial assistance required for aiding and abetting may be provided in the
form of positive action or omission before, during or after perpetration of the crime
(see e.g. Aleksovski, AJ, §62; Blaskic, AJ, §48). Furthermore, the assistance may be
physical (or tangible) or moral and psychological (Furundzija, TJ, §231). An example
of intangible assistance was given by the Advocate General in Schonfeld and others.
The defendants had been charged with being ‘concerned in the killing’ of three Allied
airmen who had been hiding in the home of a member of the Dutch resistance. In out­
lining the role of accessories not present at the scene of a crime, the Advocate General
gave the following example:

if he [the accessory] watched for his companions in order to prevent surprise, or remained
at a convenient distance in order to favour their escape, if necessary, or was in such a situ­
ation as to be able readily to come to their assistance, the knowledge of which was calculated
to give additional confidence to his companions, he was, in contemplation of law, present,
aiding and abetting (70).

An interesting legal issue is whether mere presence at the commission of a crime

may involve aiding and abetting such crime. The case law has rightly set out the notion
 OTHER MODES OF LIABILITY 215

that mere presence may only imply aiding and abetting when such presence involves
substantial encouragement to the crime on account of the authority of the onlooker,
with the consequence that the perpetrator draws moral and psychological support or
a legitimizing effect from that presence (if, for instance, such person is a superior to
the perpetrator, or has an im portant status in society or in the military hierarchy).
As a an example of an approving spectator’ whose mere presence involved his aid­
ing and abetting a crime, one can mention the Synagogue case (case against K. and
A, at 56), decided in 1948, under the terms of Control Council Law No. 10, by the
German Supreme Court in the British Occupied Zone. One of the defendants accused
was found guilty of a crime against hum anity (the devastation of a synagogue in 1938
in Germany), although he had not physically taken part in it, nor planned or ordered
it. The court of first instance and then the Supreme Court held that his interm ittent
presence on the crime scene, together with his status as a long-time militant of the
Nazi party, as well as his knowledge of the criminal enterprise, were sufficient to con­
vict him. Instead, as an example of presence not involving any liability for aiding and
abetting, mention can be made of the Pig-cart parade case (L. and others case), also
from the German Supreme Court in the British Occupied Zone. The defendant P. had
attended, as a spectator in civilian dress, a ‘parade’ of Nazi ‘assault troops’ in which
two political opponents of the Nazi party were exposed to public humiliation. P. had
followed the ‘parade’without taking any active part. According to the court (pronoun­
cing in 1948), P.’s conduct could not ‘even with certainty be evaluated as objective or
subjective approval. Furthermore, silent approval that does not contribute to causing
the offence in no way meets the requirements for criminal liability’ (234). Hence he
was acquitted.1
The subjective element of aiding and abetting resides in the accessory having know­
ledge that ‘his actions assist the perpetrator in the commission of the crime’.12 Thus,
this subjective element consists of two requirements:
(i) awareness that the principal will be using, is using, or has used the assistance for
the purpose of engaging in criminal conduct. It is not required that the accessory be fully
cognizant of the specificities of the crime that will be, is being, or has been committed

1 See also Furundijia, TJ, §203. The ICTY AC held in Brdanin that ‘an accused can be convicted for aid­
ing and abetting a crime when it is established that his conduct amounted to tacit approval and encourage­
ment of the crime and that such conduct substantially contributed to the crime. This form of aiding and
abetting is not, strictly speaking, criminal responsibility for omission. In the cases where this category was
applied, the accused had a position of authority, he was physically present on the scene of the crime, and his
non-intervention was seen as tacit approval and encouragement’ (AJ, §273). See also ICTR, Kayishema and
Ruzindana, TJ, §200-3, and AJ, §§201-2.
2 This, with convincing arguments, an ICTY TC held in Furundzija TJ, §§190-249. See also Aleksovski
TJ, §63; Kunarac and others T J, §391; Vasiljevic, TJ, §70; Kvocka, TJ, §254; Detalti, AJ, §352; Tadic AJ, §229;
Blaskic AJ, § 46; Krnojelac, AJ, §52; see also, Akayesu TJ, §484; Kayishema and Ruzindana, TJ, §§203-7;
Musema (§126), Ntakirutimana, TJ, §787; Kajelijeli, TJ, §763; Kamuhanda, TJ, §597.
In the ICC Statute aiding and abetting is envisaged in Article 25(3)(c), whereby a person is responsible if
‘For the purpose of facilitating the commission of such a crime [i.e. a crime within the jurisdiction of the
Court], aids, abets or otherwise assists in its commission or its attempted commission, including providing
the means for its commission’.
 2 16 INTERNATIONAL CRIMINAL LAW

by the perpetrator, let alone of the specific criminal intent of the perpetrator. The aider
and abettor is required to be aware either of the criminal intent of the perpetrator or
at least of the risk for the perpetrator to engage in criminal conduct.3 In other words, it
may suffice for the accomplice to entertain recklessness (dolus eventualis) with regard
to the behaviour of the principal. As an SCSL TC put it in Brima and others,
the mens rea required for aiding and abetting is that the accused knew that his acts would
assist the commission of the crime by the perpetrator or that he was aware of the substantial
likelihood that his acts would assist the commission of a crime by the perpetrator (§776).
This seems to correspond to fundamental principles of criminal law: if I lend a gun
to a well-known thug, already convicted of burglary or armed robbery, without know­
ing what specific crime he intends to perpetrate but in the knowledge that he will use
it to engage in criminal conduct, I am answerable for aiding and abetting whatever
crime he may later have committed by using that weapon (murder, armed robbery,
serious bodily harm, etc.); it is not necessary for me to be fully aware of the specific
crime he intends to perpetrate and the required mental element of that crime;4

3 In Furundzija an ICTY Trial Chamber held that ‘it is not necessary that the aider and abettor should
know the precise crime that was intended and which in the event was committed. If he is aware that one of a
number of crimes will probably be committed, and one of those crimes is in fact committed, he has intended
to facilitate the commission of that crime, and is guilty as an aider and abettor’ (TJ, §247). Another Trial
Chamber supported this proposition in Blaskic (TJ, §287), and the Appeals Chamber concurred in it in its
judgment in Blaskic (AJ, §50).
4 The issue arose in van Anraat before the Hague Court of Appeal. The Court was faced with a case of aid­
ing and abetting. The accused had provided to Iraq, between 1980 and 1988, the chemical raw material TDG
(Thiodiglycol) necessary for the manufacture of the mustard gas that the Iraqi Government had then used
against the Kurds in 1987-8. The Court discussed whether, faced with this case of aiding and abetting a war
crime, it was to apply Article 48 of the Dutch Criminal Code, on complicity as an accessory, or rather ICL.
The Court held appropriate to apply Dutch law, which in its view was cleared on the matter. Dutch law does
not require that the assistance provided by the accessory be indispensable or make a ‘causal contribution’ to
the main offence; it simply requires that ‘the assistance offered by the accessory [should] promote the offence
or [make] it easier to commit that offence’ (§12.4).
The Court first found that the accused knew that the chemicals he provided would be used to produce
mustard gas (‘The fact that TDG, in the quantities as supplied by the defendant—more than eleven hun­
dred (1,100) tons altogether—could only serve for the production of mustard gas and not—as continuously
argued by the defendant and his defence—for use in the textile industry, has been stated by expert witness
[A], among others, during the court session of 4 April 2007. [A] confirmed his earlier statement of 30 May
2006 before the examining magistrate in which he said that it is totally unthinkable that during the 1980’s
TDG was used in Iraq as textile ‘additive’ and that in Iraq not one factory had been found that was equipped
for the production of textile paint or printing ink. Also witness [head of the Iraqi team that set up the FFCD],
who was in charge of quality control of mustard gas and who was head of the team that set up the already
mentioned Full Final and Complete Disclosure (FFCD) stated mid-2005 before the examining magistrate: ‘If
one speaks about tons of TDG, then there is only one possible application: mustard gas’ §11.10).
The Court then found that the accused was aware of the high risk of use of the mustard gas in war (‘From
the defendant s awareness of the fact that his supplies of TDG served for the production of mustard gas in a
country that was involved in a long lasting war with a neighbouring country and of the efforts to conceal the
supplies of a precursor of that gas and the production of the poison gas itself, follows defendant’s awareness
that the mustard gas was going to be used by Iraq in the war [...]. Through his conscious contribution to the
production of mustard gas in a country at war, the defendant knew under those circumstances that he was
the one who supplied the material and created the occasion for the actual use of that gas, in the sense that
he was very aware of the fact that in the given circumstances the use of this gas could not and would not fail
to materialise. In different words: the defendant was very aware of the fact that—‘in the ordinary course of
 OTHER MODES OF LIABILITY 2 17

(ii) furthermore, the aider and abettor must willingly aim to help or encourage another
person in the commission of a crime; in this respect intent is therefore required.

Among the various cases where the notion was applied,5Akayesu can be cited, not
so much for outlining the legal contours of the notion (the TC at one point stated that
‘complicity’ was to be defined in the light of the Rwandan Penal Code: §537), as for the
legal findings on this matter. The TC found that
Akayesu, in his capacity as bourgmestre [mayor], was responsible for maintaining law and
public order in the commune of Taba and [...] had the effective authority over the communal
police. Moreover, as ‘leader’ of Taba commune, of which he was one of the most prominent
figures, the inhabitants respected him and followed his orders. Akayesu himself admitted
before the Chamber that he had the power to assemble the population and that they obeyed
his instructions. It has also been proved that a very large number of Tutsi were killed in
Taba between 7 April and the end of June 1994, while Akayesu was bourgmestre of the
Commune. Knowing of such killings, he opposed them and attempted to prevent them only
until 18 April 1994, after which date he not only stopped trying to maintain law and order
in his commune, but was also present during the acts of violence and killings, and some­
times even gave orders himself for bodily or mental harm to be caused to certain Tutsi, and
endorsed and even ordered the killing of several Tutsi [...] The Chamber holds that the fact
that Akayesu, as a local authority, failed to oppose such killings and serious bodily or mental
harm constituted a form of tacit encouragement, which was compounded by being present
[during] such criminal acts (§§704-5).
The Chamber added that Akayesu was present during numerous incidents of rape
and sexual violence against Tutsi women and, by his attitude and utterances, encour­
aged such acts, thus giving ‘tacit encouragement’ to the rapes being committed. The
Court concluded that he was criminally responsible ‘for having abetted in the prepar­
ation or execution of the killings of members of the Tutsi group and the infliction of
serious bodily and mental harm on members of the said group’ (§§706-7).

events’—the gas was going to be used. In this respect the Court assumes that the defendant, notwithstand­
ing his statements concerning his relevant knowledge, was aware of the—also then known—unscrupulous
character of the then Iraqi regime’ (§11.16; emphasis added).
5 Such cases include Schonfeld and Rohde, both heard by British military courts (at 64 and 56, respect­
ively), Zykion B, also heard by a British court (at 93), Einsatzgruppen, brought before by a US Military
Tribunal sitting at Nuremberg (at 569-85), S. and others (Hechingen Deportation case), brought before a
German court in the French Occupied Zone (at 484-90). However, in most of these cases the notion of aiding
and abetting was not clearly defined as distinct from that of participation in a common purpose.
TCs of the ICTR and the ICTY have made a better jurisprudential contribution to the outlining and enun­
ciation of the concept in Akayesu (dealing with the notion of complicity in genocide’; §§525-48), in Tadic
(§§688-92), and in Furundzija (§§190-249).
One may also mention a Canadian case involving torture: in Moreno (decision of 14 September 1993) the
Court held that ‘Presence at the commission of an offence can be evidence of aiding and abetting if accom­
panied by other factors, such as prior knowledge of the principal offender’s intention to commit the offence
or attendance for the purpose of encouragement [...] While mere presence at the scene of a crime (torture)
is not sufficient to invoke the exclusion clause [of the Refugee Convention], the act of keeping watch with a
view to preventing the intended victim from escaping may well attract criminal liability’ (at 16-17). See also
Ramirez (at 5-9).
 2 18 INTERNATIONAL CRIMINAL LAW

In Furundzija an ICTY TC found that the accused, an officer of the Bosnian Croat
armed forces, was present while the victim was being raped by another officer, and
interrogated her. It held that in this way he had given assistance, encouragement, or
moral support, having a substantial effect on the crime by the other officer, with the
knowledge that these acts assisted the commission of the offence. The TC therefore
found the defendant guilty of aiding and abetting outrages upon personal dignity,
including rape (§§270-5).6

10.2 INCITEMENT AS A FORM OF PARTICIPATION

IN INTERNATIONAL CRIMES
Incitement to commit a crime is some form of instigation, inducement, encourage­
ment, or persuasion to perpetrate the crime. Incitement does not necessarily presup­
pose a hierarchical position. It simply means taking all those psychological or physical
steps designed to prompt somebody else to commit a crime. It also requires the intent
to have the crime perpetrated.
Both positive acts and omissions may constitute incitement.7 Furthermore, incite­
ment is a crime only under certain conditions: (i) it must be direct and explicit;
(ii) commission of the crime by other persons must follow up. In other words, incite­
ment is not punished per se, but only if it leads to the perpetration of a crime (as we
shall see, at 10.8, ICL provides for an exception to this rule, in the case of genocide).
Thus, a causal connection is necessary between the instigation and the criminal con­
duct of the persons that committed the crime.8
The requisite subjective element may be set out as follows: (i) the person intended
to induce the commission of the crime by another person, or in other words ‘directly
intended to provoke the commission of the crime’;9 or (ii) the person was at least aware
of the likelihood that commission of the crime would be a consequence of his action;
in addition, (iii) the person must possess the mens rea concerning the crime he is
instigating.10
In Kurt Mayer, tried by a Canadian Military Court sitting at Aurich in Germany,
the accused, a Commander of the 25 SS Panzer Grenadier Regiment, was among other

6 In Aleksovski the TC found that the defendant ‘By being present during the mistreatment, and yet not
objecting to it notwithstanding its systematic nature and the authority he had over its perpetrators, the
accused was necessarily aware that such tacit approval would be construed as a sign of his support and
encouragement. He thus contributed substantially to the mistreatment. Accordingly, the accused must be
held responsible for aiding and abetting under Article 7(1) in the physical and mental abuse which detainees
were subjected to during the body searches on 15 and 16 April 1993’ (§87).
7 Blasktf TJ, §280 Kordic and Cerkez, TJ, §387.
8 See Blasktf,, TJ, §280; Kordic and Cerkez TJ, §387; Kvocka, TJ, §252; Akayesu, TJ, §482; Semanza, YJ,
§381; Rutaganda, TJ, §38; Musema, TJ, §120, Kajelijeli, TJ, 762; Kamuhanda, TJ, §593.
9 Blasktf, TJ, §278; Kordic and Cerkez, TJ §387; Bagilishema, TJ, §31.
10 Kvocka, TJ, §252; Naletiltf and Martinovic, TJ, §60.
 OTHER MODES OF LIABILITY 2 19

things, charged with having incited and counselled troops under his command to deny
quarter to Allied troops in 1943-4 in Belgium and France. The Judge Advocate stated
As it is an offence to deny quarter to prisoners I think an officer may be convicted of a war
crime if he incites and counsels troops under his command to deny quarter, whether or
not prisoners were killed as a result thereof. It would seem to be common sense to say that
not only those members of the enemy who unlawfully kill prisoners may be charged as war
criminals, but also any superior military commander who incites and counsels his troops to
commit such offences (At 840).11

10.3 INCHOATE CRIMES: GENERAL

Many legal systems punish not only consummated criminal offences (for instance,
murder, theft, etc.), but also ‘inchoate’, that is prelim inary or ‘just begun’ criminal
wrongdoings. These are acts that: (i) are preparatory to prohibited offences; (ii) have
not been completed, therefore have not yet caused any harm; and (iii) are punished on
their own; that is, in spite of the fact that they have not led to a completed offence.
The rationale behind criminalization of such offences is clear: the legal system
intends to protect society as far as possible; therefore, in addition to punishing offences
already perpetrated, it endeavours to prevent the commission of potential transgres­
sions. It consequently intervenes with its prohibitions at an early stage, before crimes
are completed; that is, at the stage of their preparation, so as to forestall the consum­
mation of the harmful consequences of actual crimes.
In many national legal systems (particularly in common law countries) three cat­
egories of such crimes are envisaged: attempt, conspiracy, and incitement. In inter­
national law, while attempt is regarded as admissible as a general class of inchoate
crimes, conspiracy and incitement are only prohibited as ‘prelim inary’ (not consum­
mated) offences when connected to the most serious crime, genocide. In addition, ICL
also punishes planning or ordering the commission of international crimes.
The very limited acceptance of conspiracy is probably due to the fact that this class
of criminal offence is not accepted in most civil law countries; hence it has been con­
sidered admissible at the international level only with regard to the most heinous and
dangerous international crime, genocide, which aims at destroying groups as such.
As for incitement, as we have seen above (10.2), in ICL it is prohibited only if it
leads to the actual perpetration of the crime; that is, as a form of participation in a
crime, probably because states and courts have felt that prohibiting incitement per se
in connection with any international crime, including war crimes and crimes against
humanity would excessively broaden the range of criminal conduct, the more so
because of the difficulty of clearly delineating the notion of incitement. Incitement
as such has been exceptionally prohibited, subject, however, to some stringent condi­
tions, in connection with genocide.1

11 Cases where incitement to commit war crimes was punished include Falkenhorst (at 23 and 29-30).
 220 INTERNATIONAL CRIMINAL LAW

As for planning and ordering, the rationale behind the tendency of international
law to punish them as inchoate crimes lies prim arily in this: the most serious and
large-scale international crimes result from careful preparation and concerted action
by many agents, or are the result of instructions and directives issued by m ilitary or
political leaders. In consequence international criminal rules aim to prevent or at least
circumscribe such conduct by stigmatizing it as criminal and making it punishable.
In summary, in ICL, within the general category of inchoate crimes one ought to
distinguish three subcategories:

1. Criminal conduct that is preparatory to a crime, but which by definition cannot

be followed by the intended crime. This subcategory encompasses attempt, where, by
definition, the subsequent offence is not consummated (because subjective or external
circumstances prevent consummation).
2. Criminal conduct that is preparatory to crimes proper, and is punished per se;
that is, even if it does not lead to the actual perpetration of the crime. However, when
this perpetration follows, the preparatory conduct is no longer punishable as such, as it
is absorbed’ into the actual crime (although it may nevertheless be taken into account
as an aggravating circumstance) if the author of the preliminary offence also is the per­
petrator of the actual crime. If, however, the two classes of agent differ, one agent will be
responsible for the inchoate crime, the other for the execution of the crime proper. This
subcategory embraces planning as well as conspiracy to wage aggression.
3. Criminal conduct that is punished per se, whether or not it is followed by the con­
summation of a crime; where a crime does follow, this conduct as well as the consum­
mated crime is punished. This subcategory includes incitement to commit genocide,
conspiracy to commit genocide, as well as ordering (which, if the order is executed,
involves the responsibility of the superior for ordering criminal conduct, while the
subordinate is liable for the execution of the unlawful order).

10.4 ATTEMPT

Attempt as a distinct criminal offence occurs whenever a person intending to commit

a crime tries to carry it out without, however, the normal outcome of his action com­
ing about.
One should distinguish between two different possibilities: (i) the perpetrator
takes the initial steps but is then stopped by others; or (ii) on account of circum ­
stances independent of his will, his action does not produce the effects he intended to
bring about; in other words, he performs all the necessary acts without, however, the
intended result following. An example of the first category is when a soldier starts to
beat a prisoner of war savagely with the intention of killing him, and is only prevented
from so doing by others, who drag him off his intended victim; or, within a general
context of massive attacks against civilians, a military officer orders his troops to blow
 OTHER MODES OF LIABILITY 221

up an internm ent camp where male civilians belonging to a particular ethnic group
are being held; however, the sudden and unexpected arrival of a superior officer who
is contrary to such acts of extermination at the last minute prevents the troops from
lighting the fuse of the dynamite. An example of the second category is when a sol­
dier shoots at a prisoner of war, intending to execute him, but the intended victim is
not fatally wounded and subsequently manages to escape. Another example is when a
military unit, following orders of an officer, launches a missile against a group of civil­
ian dwellings; the missile launcher, however, gets jammed and the missile is not fired,
or else the launcher misfires and the missile ends up in a nearby lake, where it does not
cause any victims or material damage. Yet another example is when an order for the
deportation of civilians is executed, w ithin the context of a systematic attack on civil­
ians, and all civilians detained in a camp are put on buses to be deported; however, an
air attack by the enemy belligerent prevents the buses from leaving and all the detain­
ees, taking advantage of the ensuing turmoil, manage to escape. A further example is
the case where the victim of an attempted m urder is already dead (without, of course,
the agent knowing this circumstance).12
Although in the category of attempt the intended harm is not caused to the vic­
tim, international law nevertheless makes attempt punishable, in order to prevent
breaches of international rules as far as possible. Consistently, this offence is punished
in various national laws on war crimes,13or is regarded as a distinct offence in national
case law on the same crimes.14
Some international tribunals have denied that either customary rules or the tri­
bunal’s Statute contemplated attem pt as a general category (Akayesu, TJ, §473;
Krnojelac, TJ, §432 nt. 1292).15 Probably this wrong conclusion is due partly to the

12 In Charles W. Keenan the accused had been ordered by his superior to ‘finish off’ a civilian woman at
whom the superior had already shot. A US Court of Military Appeal held that in the case at issue attempted
murder was to be ruled out only because the subordinate knew that she was no longer alive when he fired at
her (at 114). The Court stated that ‘so far as attempted murder is concerned, military law “has tended toward
the advanced and modern position” that holds one accountable for conduct which would constitute a crime
if the facts were as he believed them to be (see United States v. Thomas). Here the accused expressly testified
that he believed the woman was dead; and the board of review specifically refused to find that she was still
alive when the accused fired at her. Moore and Eakins also testified that they believed the victim was dead
before the accused fired. The board of review could, therefore, reasonably conclude that the accused knew he
was firing at a corpse. This conclusion necessarily absolves him of attempted murder’ (at 113).
To support its ruling the Court cited an important case, unrelated to war crimes, where the same Court had
extensively dealt with the notion at issue: Rodger D. Thomas, a case of attempted rape, which had offered the
Court the opportunity to discuss the requisite ingredients of the offence, with a reasoning that is along the
same lines as the notion propounded above for international criminal law (at 287-92). The Court held that the
elements of the offence of attempted rape were: ‘(i) an overt act; (2) specific intent; (3) more than mere prepar­
ation; (4) tending to effect the commission of the offence; and (5) failure to effect its commission’ (at 286).
13 See, for instance, the laws cited in UN Law Reports, vol. XV, at 89 (Norway, Yugoslavia, the
Netherlands).
14 See the cases reported in UN Law Reports, vol. VI, 120, as well as in UN Law Reports, vol. VII, at 73.
15 The TC stated that ‘The existence of a mistaken belief that the intended victim will be discriminated
against, together with an intention to discriminate against that person because of that mistaken belief, may
in some circumstances amount to the inchoate offence of attempted persecution, but no such crime falls
within the jurisdiction of this Tribunal’ (§242 n. 1292).
 222 INTERNATIONAL CRIMINAL LAW

fact that attempt as an inchoate offence rarely occurs in the case of crimes against
hum anity (while it is expressly prohibited for genocide), partly to the fact that where
attempted crimes were committed, prosecutors, followed by international tribunals,
misapprehended such offences and, instead of,charging the defendant with attempted
war crimes or crimes against humanity, wrongly charged him with other offences. A
case in point is Vasiljevic. The offender, a Bosnian Serb member of a param ilitary
group, together with three other persons, had allegedly taken seven Bosnian Muslim
civilians to the bank of the river Drina, forced them to line up and then opened fire
to kill all of them. Five men died, while two, pretending to be shot dead, jum ped into
the river and saved their life. The Prosecution charged the defendant with murder
(as a war crime and a crime against humanity) for the killing of the five Muslims,
while, for the attempted murder, it charged them with inhumane acts as a crime against
humanity and violence to life and persons as a war crime. The TC in the event convicted
the defendant o f ‘other inhum ane acts’ as a crime against humanity.16 The AC did not
reverse the decision on this specific issue. However, in another case where the pros­
ecutor had charged the defendant with murder as well as, for 12 attempted murders,
‘inhum an acts’ (Mrdja, Indictment, at 4), the TC rightly spoke o f‘attempted murder’ in
its sentencing judgment (§31).
The existence of a customary rule on attempt can be inferred, more than from the
fact that all penal systems of the world provide for attempt as a separate mode of crim ­
inal liability, from the existence of numerous cases where national courts have relied
upon the notion of attempt (normally attempted murder) in connection with war
crimes. In this respect one can cite numerous cases brought before German courts,
where the question revolved around the attempted killing of prisoners of war, civilians
or inmates in concentration camps;17 Canadian courts (for instance in Johann Neitz,
at 209, where the question at issue was the attempted murder of a prisoner of war),18as

16 On the decision, see the critical remarks of A. Cassese in 2 JICJ (2004), 265-74.
17 See, for instance, Friedrich Otto Köhler, at 274 (the defendant was a police officer charged with killing
German and foreign detainees in 1945); Kurt Köttner, at 682 (the accused, a member of the SS and head of
the security police, had committed war crimes in 1942 against persons detained in a concentration camp in
Poland); Otto Haupt and others, at 604 (the defendants had committed war crimes against prisoners of war
detained in a concentration camp); Karl Dietrich, at 485 (the issue was that of ill-treatment of Jews in occupied
territory; the Court of Assize ruled out attempted murder on the facts). Some cases concern aiding and abet­
ting attempted murder: see, for instance, W. J. F. Kleinhenn at, at 9 (in 1942 the accused had committed war
crimes against sick detainees in a concentration camp). Other cases concern attempted manslaughter; see for
instance S. case, at 505 (in early 1945 the defendant had committed war crimes against foreign workers).
18 The accused Johann Neitz, a German soldier, had shot twice at a member of the Royal Canadian Air
Force, who had been taken prisoner after his aircraft had been struck by flak. As a result of the shooting the
Canadian prisoner fell down but did not die. Neitz was charged both with committing a war crime, in that he
had fired, with intent to kill, two shots at the Canadian prisoner, and, alternatively, with a war crime in that
he had wounded the prisoner of war, in violation of the laws and usages of war (see the Prosecutor’s opening
address, at 13, and the Judge Advocate’s summing-up, at 195-205). The Court found Neitz guilty of the first
charge and sentenced him to be imprisoned for life (see ibid., 209).
 OTHER MODES OF LIABILITY 223

well as US courts (for instance, in Charles W. Keenan, at 114).19 In these cases national
courts made allowance for the war crime of attempted murder (or manslaughter).20

19 The US Court of Military Appeals allowed for the war crime of attempted murder, although it held that
in the case at issue the accused was not guilty of such crime. The facts were as follows. In 1966 a ten-man
squad of US servicemen entered a village in Vietnam where they suspected Vietcongs were hiding or were
being protected. One of them, corporal Luczko, shot twice at an unarmed woman; the accused, private
Keenan, asked the corporal whether he wanted him, Keenan, to finish her off. The corporal did not answer,
but fired a third shot at the woman. After that he asked Keenan to finish her off, and Keenan fired an auto­
matic burst. The Board of Review indicated that ‘it was not convinced beyond a reasonable doubt that the
victim was alive at the time the accused fired at her’ (113). On the basis of that finding ‘it absolved the accused
of all criminal responsibility in the death’ of the woman. Appellate Government counsel contended that the
Board of Review’s finding of fact did ‘not automatically render the evidence insufficient to affirm a finding
of guilt’ among other things ‘on the ground that he [the accused] committed the lesser included offense of
attempted murder’ (113). The Court of Military Appeals held, however, otherwise. It stated that ‘so far as
attempted murder is concerned, military law ‘has tended toward the advanced and modern position that
holds one accountable for conduct which would constitute a crime if the facts were as he believed them to be
[...] Here the accused expressly testified that he believed the woman was dead; and the board of review spe­
cifically refused to find that she was alive when the accused fired at her. Moore and Eakins [two members of
the squad] also testified that they believed the victim was dead before the accused fired. The Board of Review
could, therefore, reasonably conclude that the accused knew he was firing at a corpse. This conclusion neces­
sarily absolves him of attempted murder’ (113).
20 In contrast, some post-Second World War German courts pronouncing on cases of denunciation as a
crime against humanity held that attempt is conceptually not admissible if related to such category of crimes.
It is not clear whether the rather convoluted propositions of those courts may be construed as indicating that
such courts only referred to denunciations to the Gestapo (of Jews or political opponents or at any rate per­
sons contrary to the Nazi system) as crimes against humanity (that is, as ‘assaults on victims connected with
the Nazi rule based on violence and tyranny’). See P. case, at 15; V. case, at. 21 and O case, at 391-2.
It would seem that this exclusion of attempt is primarily due to the upholding by these German courts of
a very broad notion of crimes against humanity, a notion that also includes as part of the objective element
of the crime the ‘attempt to cause damage’. In this connection the first decision on the matter, namely the
aforementioned case P. is significant, although the Court’s reasoning is rather convoluted. The facts were
as follows: in 1933 the accused, a member of the SA (Sturmabteilungen) passed by the veranda of a young
man who was whistling the ‘International’ while shining shoes. The accused considered that he was being
provoked by the song and, after trying unsuccessfully to get into the house, returned some time later with a
police van. Having entered the house, he pushed himself forward between the police officials and punched the
young man in the face and kicked him in the legs. The police cam. to the aid of the young man by holding the
accused away from him, then arrested the young man and took him away to the police prison instead of the
SA barracks, as requested by the accused. A few days later the young man was released from police custody.
The court of first instance sentenced the accused to two months’ imprisonment for a crime against human­
ity pursuant to Control Council Law no. 10. The Supreme Court held that the crime of persecution has a the
relationship to violence and arbitrary rule, as was the case of the Nazi time with the persecution programme,
as one of its core elements ‘(14). ‘A connection must be established between the assault [on the victim] and the
system of violent and arbitrary rule prevailing’ (14). A harm or injury to the victim is also necessary. ‘All exter­
nal and inner harm which he [the victim] suffered at the hands of the perpetrators or their collaborators may
be considered for the objective characterization of a crime against humanity. Insofar as the injury has had an
effect, it belongs to the elements of the offence. Furthermore, a danger or threat caused by the perpetrator and
experienced by the victim, depending on the circumstances, may already signify or carry with it sufficient
harm for the victim (example: a person who is the subject of a dangerous denunciation commits suicide out of
fear, or flees into the woods and dies of hunger). Only from this perspective can what the perpetrator planned,
intended, prepared or sought, or what harm the act could have generated, be of interest for the objective def­
inition of the act. In addition, however, the ingredients of the act cannot be realised by what the victim did not
actually suffer but could easily have suffered (15). ‘This clarification makes it possible to state that, in respect
of crimes against humanity attempt is conceptually impossible in the German legal sense. Nonetheless, the
attempt as such to commit harm may by itself fall within the definition of a crime against humanity even if
 224 INTERNATIONAL CRIMINAL LAW

Other cases can also be mentioned.21

The customary rule has been codified in Article 25(3)(f) of the ICC Statute, whereby
a person is criminally responsible if he attempts to commit [a crime under the C ourt’s
jurisdiction] by taking action that commences its execution by means of a substantial
step, but the crime does not occur because of circumstances independent of the per­
son’s intentions’.
Clearly, what is required for the attempt to be punishable, is: (i) conduct consisting
of a significant commencement of the criminal action (to hold to the example given
above, it is not sufficient for the guard to take the prisoner out of his cell and possibly
even shout at, or abuse, him; it is necessary for the guard to start beating him savagely);
(ii) the clear intention to commit a crime; (iii) failure of that intention to take effect
owing to external circumstances.
The ICC Statute codifies international customary laws in another respect as well.
Article 25(3)(f) duly takes account of the cessation of the attempter’s criminal inten­
tion and leaves his initial steps unpunished

However, a person who abandons the effort to commit the crime or otherwise prevents
the completion of the crime shall not be liable for punishment under this Statute for
the attempt to commit that crime if that person completely and voluntarily gave up the
criminal purpose.

Thus, to continue with one of the aforementioned example, if the guard, after beat­
ing the prisoner for a while, suddenly decides not to carry through his initial purpose
and takes the prisoner back to his cell, he is not guilty of attempted m urder (although
he may well be guilty of other crimes). Similarly, if an officer gives an order to shoot
and kill a group of innocent civilians and then, just before the order is carried out,
changes his m ind and orders that their lives be spared, he is not considered criminally
liable for murder (although he may be guilty of inhum an treatment or even torture, if
he intended to carry out a mock execution).

the worst possible results did not occur’ (15). Turning to the specific offence at issue, the Supreme Court held
that it constituted dangerous bodily harm, which however did not fall under the category of crimes against
humanity. In its view, even if the act 'did prejudice the non-material value of the [young man] it did not do
so with an effect on humanity in general [...] The more serious harm that the accused had intended for the
[young man] did not materialise. The fact that the accused had tried to cause harm must not be taken into
consideration as regards the objective elements of the offence. Humanity as the bearer and protector of the
non-material values is not prejudiced by such an act; should this act—insofar as it was not punishable under
German criminal law—remain unpunished, it would not be unbearable for humanity (18).
See also V. (so called Nu case), at 21. (‘Certainly, the sense of repugnance [that the act would arouse in an
ideal observer] mentioned in the submission ruling can already occur when the person affected has only
been exposed to the danger of harm. Endangerment by denunciation can only be a crime against human­
ity by virtue of the typical hardship of inhumanity that the denunciation has effected. An attempted crime
against humanity does not even come into question, since, as explained in Sts 3/48 [P. case] an attempt is
inconceivable with this type of crime.’) Another relevant case is 0., at 391-2 (the Berlin Court of assize held
that there was no evidence of the harm to the victim; it added that ‘in the action of the accused one could
see the beginning of execution of the crime against humanity’; however, according to the Court, the notion
of attempt had already been excluded from the notion of crimes against humanity by the Supreme Court in
Cologne; the accused was therefore to be acquitted of the charge of crime against humanity: at 392).
21 See the cases reported in Law Reports o f Trials of War Criminals (UN War Crimes Commission), vol. VI,
at 120 and vol. VII at 73.
 OTHER MODES OF LIABILITY 225

As for the mens rea required for attempt, it may be noted that in common law coun­
tries as well as in many civil law systems, what is normally required is the intention to
carry out the offence (recklessness is not enough). It would seem that also in ICL the
subjective element required is intent.

10.5 PL A N N IN G

Planning consists of devising, agreeing upon with others, preparing, and arranging for
the commission of a crime. Think, for instance, of planning an air attack on civilians
or the use of such prohibited arms as chemical or bacteriological weapons, or the indis­
criminate killing of civilians as part of a widespread or systematic attack on civilians.
As stated by various courts, planning implies that one or several persons contemplate
designing the commission of a crime at both the preparatory and executory phases .22
Given the nature and features of international crimes, it is often the higher military
or civilian authorities that carry out the planning.
Whoever takes part in the planning of an international crime is liable to punish­
ment for the relevant crime, whatever his rank in the hierarchy and the level of his
participation (although of course the rank and role may be germane to punishment; it
is evident that the higher the status of the planner and the intensity of his participation
in the planning, the harsher his penalty should be).
The subjective element required is the intent to carry out the crim inal conduct; the
offender ‘directly or indirectly intended that the crime in question be com m itted’.23
A difficult question is whether planning an international crime is punishable per se,
regardless of whether or not it leads to the actual commission of the crime planned, or
instead is only punishable if planning is followed up by perpetration of the crime. TCs
of the ICTR opted for the latter solution in a number of cases.24 They grounded this
conclusion on the works of the International Law Commission and on the interpret­
ation of the relevant rule of the ICTR Statute (Article 6(1)) laying down the principle of
individual criminal responsibility, which ‘implies that the planning or the preparation
of a crime actually must lead to its commission’ (Musema, §115).
It may be noted that prosecuting someone for planning, where the planning is not
put into effect, comes close to prosecuting conspiracy (although with conspiracy there
must be an agreement of two or more persons, whereas planning may be carried out by
one person alone, and if done by more persons, no agreement is required). The ICTY
and ICTR Statutes allow conspiracy for genocide, but not for crimes against hum an­
ity and war crimes. (This was also the position of the IMT at Nuremberg: conspiracy
to commit crimes against peace was held admissible, whereas conspiracy to commit
crimes against hum anity and war crimes was not.)

22 Akayesu TJ, §480; Blaskic TJ, §279; Kordic and Cerkez, T J, §386; Brima and others, TJ, § 765.
23 Blaskic TJ, §278; Kordic and Cerkez, TJ, §386; Bagilishema, TJ, §31; Brima and others, TJ, §766.
24 Akayesu (§475), Rutaganda (§34), and Musema (§115).
226 INTERNATIONAL CRIMINAL LAW

An ICTY TC, ruling in Kordic and Cerkez, propounded a contrary view. It held that
an accused may be held criminally responsible for planning alone’ (§386). The reason
for this conclusion is that ‘planning constitutes a discrete form of responsibility under
Article 7(1) of the Statute’. However, the TC set forth two caveats: first, a person found
to have committed a crime will not be found responsible for planning the same crime’;
secondly, an accused will only be held responsible for planning, instigating or order­
ing a crime if he directly or indirectly intended that the crime be com m itted’ (§386).
Although there is no consistent case law on this matter, it would seem that the grav­
ity of international crimes (or at least of the most serious among them) may warrant
the conclusion that planning the commission of one or more of such crimes is punish­
able per se even if the crime is not actually perpetrated. The rationale is that ICL aims
not only to punish persons found guilty of crimes, but also to prevent persons from
engaging in serious criminal conduct. Consequently, in case of doubt criminal rules
must be interpreted as being also designed as far as possible to prevent offences.
It would follow that planning an international crime is also punishable per se as
a distinct form of criminal liability, subject to a set of conditions that can be derived
from the general system of ICL:

1. Only the planning of serious or large-scale international crimes constitutes a

discrete offence: for instance, the planning of massive war crimes (such as the exter­
mination of a large num ber of prisoners of war, or the large-scale deportation of civil­
ians to extermination camps), or of crimes against humanity, or genocide. Since the
rules on planning do not specify the legal ingredients of this crime, it seems warranted
to m aintain that, for international crimes of lesser gravity (for instance, ill-treatment
of one prisoner of war, the taking by members of the Occupying Power of private
property belonging to civilians), those rules must be construed in such a way as to
favour the accused (favor rei). Consequently, the mere planning of those crimes of
lesser gravity may be held not to constitute a crime per se.
2. If planning is followed up by execution of the crime by the same person, plan­
ning is no longer punishable as a crime distinct from that resulting from its execution:
the perpetrator may not be convicted of both planning and committing the crime,
for the latter absorbs’ the former.25 (in this respect planning is different from, hence
may not be equated to, such ‘inchoate crimes’ as conspiracy to commit genocide and
incitement to genocide, to be discussed below). Nevertheless, the fact of planning the
criminal offence may be held to constitute an aggravating factor.26
3. As for the requisite mens rea, it is necessary for the author to intend that the
planned crime be committed, or else he must be aware of the risk that the planned
crime would be perpetrated by him or by someone else (recklessness or dolus
eventualis).

25 ICTY TC, Kordic, §386; Brdanin, §268; SPCS TC, Brima and others, §767.
26 jc t y TC, Stakic, §443; SPCS TC, Brima and others, §767.
 OTHER MODES OF LIABILITY 227

10.6 CONSPIRACY

10.6.1 g e n e r a l : t h e n o t io n o f c o n sp ir a c y

It is common knowledge that conspiracy is a form of crim inality punished in common

law systems, but either unknown to, or accepted to a very limited extent by, civil law
countries. Conspiracy is a group offence, consisting of the agreement of two or more
persons to commit a crime. It is punished even if the crime is never perpetrated. In add­
ition, if the crime is carried out, the perpetrators are held liable both for conspiracy and
for the substantive crime they commit. The mens rea element of conspiracy required for
each and every participant is twofold: (i) knowledge of the facts or circumstances mak­
ing up the crime the group intends to commit; (ii) intent to carry out the conspiracy
and thereby perpetrate the substantive offence. Plainly, the basic rationale behind the
prohibition of this crime is the need to prevent offences, especially when they involve
several persons and are thus more dangerous to the community.
As noted above, in international law no customary rule has evolved on conspiracy
on account of the lack of support from civil law countries for this category of crime.
(In civil law systems, entering into agreement to commit a crime is not punishable per
se, unless it leads to the perpetration of the crime; only exceptionally, and for such cat­
egories of serious offences as those aimed at underm ining state security or at setting
up associations or organizations systematically bent on crim inal conduct in various
areas, is conspiracy as such prohibited). In 2006 in Hamdan the US Supreme Court,
per Justice Stevens, rightly ruled out that conspiracy is criminalized as a war crime in
international law (at 9-12,46-56).
Treaty rules on conspiracy can only be found in the London Agreement of 1945. In
Article 6 it made punishable persons ‘participating in a Common Plan or Conspiracy
for the accomplishment’ of any crime against peace and in addition made ‘leaders,
organizers, instigators or accomplices participating in the formulation or execution of
a Common Plan or Conspiracy to commit any of the foregoing crimes [that is, crimes
against peace, war crimes, and crimes against humanity] responsible for all acts per­
formed by any persons in execution of such plan’. This provision laid down ex post
facto law. However, as it referred to conspiracy to commit a crime against peace, it
punished persons who had conspired to wage the war that had just ended. In add­
ition, to the extent that it referred to other crimes, it also made conspiracy punishable
for acts already accomplished. In other words, in the end conspiracy was held to be
punishable to the extent that any plan or agreement to commit an international crime
had been actually carried out. (Strikingly, Control Council Law no. 10 only referred to
conspiracy to commit crimes against peace: see Article 2(l)(a).)
Nevertheless, generally speaking, both the IMT at Nuremberg and the Military
Tribunals sitting at Nuremberg took a restrictive view of conspiracy: see, in particular,
Goring and others (at 224-6) and Alstotter and others (at 289-90). In the former case
228 INTERNATIONAL CRIMINAL LAW

the influence of the French Judge Donnedieu de Vabres, and his insistence on the novel
nature of conspiracy in international law, were indisputably decisive.27
As noted above (7.3.4) it is warranted to hold that conspiracy to engage in a war
of aggression is criminalized by a customary rule of international law. However, the
getting together or more persons and their agreeing to undertake a war or other acts
of aggression is punishable only if such concerting measures for acts of aggression is
not followed up by the actual waging of aggression. If aggression is subsequently car­
ried out, this crime ‘absorbs’ the crime of conspiracy, unless those who conspired are
different from the persons who in fact undertook the aggression (in which case the
former are responsible for conspiring and the latter for aggression).

10.6.2 T H E O F F E N C E OF C O N S P I R A C Y T O C O M M I T G E N O C I D E

The only treaty rule on conspiracy currently in force is Article 3(b) of the 1948 Genocide
Convention, which, on the grounds and motivations set out above, makes ‘conspiracy
to commit genocide’ punishable (genocide was deemed to be such an odious crime
that even the mere agreement to commit it or its planning without any practical fol­
low-up that is, execution of the crime, were banned and criminalized). It would seem
that, like most other substantive provisions of the Convention, it has turned into cus­
tomary law. Among other things it has been taken up in the Statutes of both the ICTY
and the ICTR (but, strikingly, not in Article 6 of the ICC Statute, which consequently
differs in this respect from international customary law).
In Musema an ICTR TC held that conspiracy to commit genocide ‘is to be defined as
an agreement between two or more persons to commit the crime of genocide’ (§191).28In
Nahimana and others the ICTR TC added some interesting remarks on the modalities of
reaching agreement, as part of the conspiracy to commit genocide. It noted that:
Conspiracy to commit genocide can be comprised of individuals acting in an institu­
tional capacity as well as or even independently of their personal links with each other.
Institutional coordination can form the basis of a conspiracy among those individuals
who control the institutions that are engaged in coordination action. The Chamber consid­
ers the act of coordination to be the central element that distinguishes conspiracy from
‘conscious parallelism’, the concept put forward by the Defence to explain the evidence in
this case (§1048).

As for mens rea, it

rests on the concerted intent to commit genocide, that is, to destroy, in whole or in part, a
national, ethnical, racial or religious group, as such. Thus [...] the requisite intent for the

27 H. Donnedieu de Vabres, ‘Le proces de Nuremberg devant les principes modernes du droit
penal international’ in 70 HR (1947-1), 528-42. Among other things he held the view that in the event
Article 6 (in fine), of the Nuremberg Charter upheld the French notion of ‘complicity (at 541). He also
emphasized that, with regard to crimes against peace, the IMT ultimately avoided holding that there was a
general conspiracy (at 541-2).
28 See also Kajelijeli, TJ, §788; Nahimana and others, TJ, §1047.
 OTHER MODES OF LIABILITY 229

crime of conspiracy to commit genocide is ipso facto the intent required for the crime of
genocide, that is the dolus specialis of genocide (Musema, TJ §192).

As noted above, the crime of conspiracy to commit genocide is punishable even if it

fails to lead to its result; that is, even if genocide is not perpetrated.29

10.7 INCITEMENT TO GENOCIDE

Incitement is only prohibited in ICL with regard to genocide. The perceived gravity of
genocide accounts for this legal exception. Genocide is held to be such a heinous crime
involving the annihilation of entire hum an groups, that any act or conduct leading to,
or pushing towards, its perpetration is banned and criminalized.
However, incitement to genocide, to be punishable, must be not only direct but also
public (for clarification of these two notions, see below). At the same time incitement is
criminalized as such; that is, even if it is not followed by the commission of genocide.30
Incitement must be public: the fact of inducing or provoking other persons to
engage in acts of genocide must be performed in a public place (for instance, a square)
or in a public gathering, through speeches, ‘shouting or threats’(Akayesu TJ, §559)
or ‘through the sale or dissemination, offer for sale or display of written material or
printed matter [...] or the public display of placards or posters’ (ibid.), or else through
such means as radio or television capable of reaching the general public at large.31
Incitement made in private, for instance to small and selected groups, may, however,
amount to conspiracy to commit genocide (Akayesu, TJ, §556; AJ, §480).
Incitement must also be direct; that is, it must specifically provoke or induce other
persons to engage in genocide. In other words, it must not consist of vague and indir­
ect suggestions. Nevertheless, even implicit messages or utterances may am ount to
incitement, as long as the addressees immediately grasp the implications of the mes­
sage in light of its cultural and linguistic content.32 For instance, the use of the term
‘cockroaches’ referring to Tutsis in the Rwandan context, as possible targets of geno-
cidal action, could amount to incitement (Akayesu, TJ §§557-8).
As for the subjective element of the crime, an ICTR TC held that
[it] lies in the intent to directly prompt or provoke another to commit genocide. It implies a
desire on the part of the perpetrator to create by his actions a particular state of mind neces­
sary to commit such a crime in the minds of the person(s) he is so engaging. That is to say
that the person who is inciting to commit genocide must have himself the specific intent to
commit genocide, namely, to destroy, in whole or in part, a national, ethnical, racial or reli­
gious group, as such (Akayesu, TC, §560).

29 Musema, TJ, §194; Niyitegeka, TJ, §423; Nahimana and others, TJ, §1044; Kajelijeli, TJ, §788.
30 Akayesu, TJ, §§561-2; Musema, TJ, §§193-4; Kajelijeli,T], §855; Nahimana, TJ, §1029.
31 Akayesu, TJ, §§556 and 559; Kajelijeli, TJ, §851; Nahimana, TJ, §431.
32 Akayesu, TJ, §558; Kajelijeli, TJ, §853; Nahimana, TJ, §§1004-6.
230 IN T E R N A T IO N A L C R IM IN A L LAW

An interesting case is Ruggiu, the journalist of ‘Radio Mille Collines’ accused by

the ICTR Prosecutor o f ‘direct and public incitement to commit genocide and crimes
against hum anity (persecution)’. He pleaded guilty. An ICTR TC found that
when examining the acts of persecution which have been admitted by the accused, it is pos­
sible to discern a common element. Those acts were direct and public radio broadcasts all
aimed at singling out and attacking the Tutsi ethnic group and Belgians on discriminatory
grounds, by depriving them of the fundamental rights to life, liberty and basic humanity
enjoyed by members of wider society. The deprivation of these rights can be said to have as
its aim the death and removal of those persons from the society in which they live alongside
the perpetrators, or eventually even from humanity itself (§22).

10.8 ORDERING

Ordering is a mode of responsibility entailing that a dejure or de facto superior (within

a military or civilian hierarchy) issues a command to a subordinate to the effect that
he or she must take a certain course of action that is contrary to law and amounts to a
criminal offence.33As an ICTY TC rightly held in Kordic and Cerkez (§388), ‘no formal
superior-subordinate relationship is required for a finding of “ordering” so long as it
is demonstrated that the accused possessed the authority to order’. This proposition,
albeit not supported by any legal reason in the judgment, is warranted because ICL is
not a formalistic body of law geared to legal technicalities but aims at proscribing and
punishing crimes whatever the modalities of their commission.
As held by ICTY TCs,34 there is no need for the order to be given in writing or in
any particular form. In addition, the existence of an order may be proved through
circumstantial evidence.
If the order issued by a superior authority is passed on by a subordinate authority
down the chain of command, the latter authority, depending upon the circumstances,
may also be held to be responsible for ordering an illegal act (Kupreskic, TJ, §§827,862).
It would seem that it is not necessary for the order to be executed. An officer or any
other higher authority issuing a criminal order may be found guilty even if the order is
not carried out by the subordinates, if the superior intended the order to be executed and
knew that the order was illegal, or else the order was manifestly illegal. Thus, in General
Jacob H. Smith, in 1902 a US Court Martial held that General Smith was guilty of order­
ing that no quarter should be given to the enemy in the Philippines, even though in fact
his troops did not comply with this order (at 799-813). In many other cases courts have
convicted officers for issuing criminal orders, even if such orders were not executed.35

33 See for instance Krstic , TJ, §601; Naletilic an d M artinovid , TJ, §61; Stakic, TJ, §445; Sem anza , TJ, §382;
Bagilishema, TJ, §31; Rutaganga , TJ, §39; K a m u h a n d a , TJ, §594.
34 Blaskic (§281) and Kordic a n d C erkez (§388). See also SCSL TC, B rim a a n d others , §772.
35 See, for instance, High C om m and (at 118-23), The Hostages Trial (at 118-23), K u rt M ayer (at 98 and
108), Falkenhorst (at 18,23,29-30), H ans W ickm ann (at 133). In Tzofan and others v. IDF A dvocate and others
 O T H E R M OD ES OF LIA BILIT Y 231

If the internationally unlawful order is executed, the person issuing the order is
criminally liable qua co-perpetrator of the crime carried out by the subordinate.
Also for this category of criminality the requisite mental element is the intent to
have the crime committed,36 at least, as long as the order is specific; that is, instructs
to perform a specific crime. However, when the order is generic, recklessness or even
gross negligence may be considered sufficient.37
So far we have addressed the issue of orders that impose to engage in criminal con­
duct. Plainly, if a superior issues a lawful order (for example, bombing m ilitary instal­
lations near civilian houses, after taking all the necessary precautions imposed by
international hum anitarian law) and the subordinates, in partial non-compliance
with the order, commit a war crime (for instance, deliberately bombing some civilian
dwellings as well, or else failing to take the necessary precautions), the ordering official
is not criminally liable for that war crime.
However, there may be cases where even a lawful order may involve the responsibility
of the superior. This occurs when it can be proved that the superior was aware that the
execution of his or her order was most likely to lead to the commission of a war crime and
nevertheless willingly took this risk. A case in point is Blaskic. An ICTY TC had held that
the defendant had ordered artillery fire against some villages; a massacre of civilians had
ensued; according to the Chamber, even if doubt were still cast in spite of everything on
whether the accused ordered the attack with the clear intention that the massacre would
be committed, he would still be liable under Article 7(1) of the [ICTY] Statute for order­
ing the crimes’ (§474). Hence the TC held Blaskic guilty for this order as well. On appeal,
the AC held, however, that the test of recklessness had not been rightly set out. In its view,
while it may be correct to require a culpable mental state lower than intent and thus admit

(Yehuda Meir case), Judge D. Levin (concurring) held that ‘the higher the rank of the commanding officer
and the more comprehensive and more decisive his authority, the greater the responsibility incumbent upon
him to examine and determine the justification and legality of the order’ (at 745).
It should be noted that ordering is sometimes treated as a species of instigation, for instance ordering that
no quarter be given may be regarded as the same thing as inciting troops to commit war crimes.
36 In Jung and Schumacher, decided by a Canadian Military Court sitting at Aurich in Germany, the
Judge Advocate, in discussing the position of the defendant Jung, who had ordered the other defendant to
shoot and kill a Canadian war prisoner, noted the following: ‘The Court may find that the accused uttered the
words or some words to do harm to the prisoner, but it must be found that he uttered them with the expect­
ation and intention that they should be acted upon by someone who heard them, including Schumacher.
In this event he would have either incited, counselled or procured the acts to have been done, and so be
concerned [in the crime]. Now, if you find that the accused Jung handed the prisoner over to Schumacher,
knowing or expecting he would be killed, then again he would be concerned [in the killing of the Canadian
POW]’ (at 219-20).
On the requisite of intent in ordering, see Blaskic, TJ, §278; Kordic and Cerkez, TJ, §386; Stakic, TJ, §445;
Bagilishema, TJ, §31.
37 In one case a Canadian Court Martial held that the defendant was guilty of negligence for issuing
unlawful orders (he had instructed his subordinates that prisoners ‘could be abused’): see Major A.G. Seward
(at 1079-81). Interestingly, the defendant was acquitted on another count, namely of having caused bodily
harm to the Somali civilians beaten up, tortured, and killed by his subordinates. The Court Martial Appeals
Court of Canada noted in this regard that by this acquittal the defendant ‘must be taken to have been found
neither to have intended nor to have been capable or reasonably foreseeing that any of his subordinates
would mistreat unto death any Somalian [sic] prisoner’ (at 1082).
232 IN T E R N A T IO N A L C R IM IN A L LAW

recklessness (or dolus eventualis), the mental element required by the TC was too low a
standard: under such standard any military commander who issues an order would be
criminally responsible, because there is always a possibility that violations could occur’
(AJ, §41). The correct standard, according to the AC, is awareness of a substantial likeli­
hood of risk plus a volitional element, namely acceptance that the risk may ensue.38

38 According to the AC ‘The knowledge of any kind of risk, however low, does not suffice for the impos­
ition of criminal responsibility for serious violations of international humanitarian law. The Trial Chamber
does not specify what degree of risk must be proven. Indeed, it appears that under the Trial Chamber’s
standard, any military commander who issues an order would be criminally responsible, because there
is always a possibility that violations could occur. The Appeals Chamber considers that an awareness of a
higher likelihood of risk and a volitional element must be incorporated in the legal standard’ (AJ, §41). The
Chamber went on to hold that ‘a person who orders an act or omission with the awareness of the substantial
likelihood that a crime will be committed in the execution of that order, has the requisite mens rea for estab­
lishing liability under Article 7(1) pursuant to ordering. Ordering with such awareness has to be regarded
as accepting that crime’ (§42).
 11
CRIMINAL LIABILITY
FOR OMISSIONS

II.1 GENERAL

International criminal liability may arise not only as a result of a positive act or con­
duct (killing an enemy civilian, unlawfully destroying works of art, etc.) but also from
an omission; that is, failure to take action. Omission is only criminalized when the law
imposes a clear obligation to act and the person wilfully or recklessly fails to do what
was legally required.
It took a long time for a general rule on this matter to evolve in international crim ­
inal law. The reason for this state of affairs is clear. The first body of substantive rules
restraining conduct in war, namely traditional international hum anitarian law,
tended to prohibit action; in other words, it imposed on combatants the obligation
not to engage in conduct contrary to some international standards (killing civilians,
raping women, shelling hospitals, etc.). By the same token it refrained, as a rule, from
imposing positive obligations to do something. The purpose of this body of law was to
ensure respect for a modicum of legal standards by belligerents. Law largely respected
the autonomy of states, leaving them free to pursue their ends and purposes in war,
and only banned (and later criminalized) glaring breaches of the most fundamental
standards of behaviour. The law did not go so far as also to require that belligerents
should take some kind of positive action to protect civilians and other victims of war­
fare. International law-makers did not deem it expedient to restrict states’ conduct by
establishing obligations requiring states to do a particular thing under some specific
circumstances.
Progress was made after the Second World War, when an ‘interventionist’ attitude
in international hum anitarian law, intended to broaden the protection of war victims,
gradually replaced the previous liberal ‘laissez-faire approach’, substantially geared
to freedom of states subject to some exceptional prohibitions. As we shall soon see,
many provisions of the 1949 Geneva Conventions clearly laid down the duty to do
something and considered failure so to act as criminal.
As we shall see below, after the Second World War one particular class of responsi­
bility by omission, that is, superiors’ responsibility, has taken on distinct features and
evolved as a discrete and im portant form of this category.
 234 IN T E R N A T I O N A L C R I M I N A L LAW

11.2 RULES IMPOSING A POSITIVE

OBLIGATION TO ACT

Some provisions of the Geneva Conventions lay down unconditional (in other
words, unqualified) positive obligations. For instance, this holds true for Article
16(4) of the First Geneva Convention (on the wounded and sick arm ed forces in
the field), which contains positive prescriptions concerning the preparation and
transm ission by one belligerent to the other, of death certificates or lists of the
dead. It also holds true for Article 17 of the same Convention, which provides for
the burial or cremation of the dead.1O ther provisions lay down positive obligations
that are, however, very sweeping and therefore leave to C ontracting States a fairly
broad margin of appreciation. This applies for instance to Article 14(2) of the Third
Geneva Convention (on prisoners of war), concerning the duty to protect prisoners
of war against acts of violence or intim idation; Article 15 of the same Convention,
requiring m aintenance of prisoners of war free of charge; and Article 29 of the same
Convention, on the duty to take all sanitary measures necessary to ensure the clean­
liness and hygiene of detention camps. Similarly, Article 36 of the First Additional
Protocol obliges states studying, developing, acquiring, or adopting new weapons
to ascertain whether these weapons are prohibited by international law. Articles
76 and 77 of the same Protocol protect women and children, respectively, against
various forms of assault by im posing on states broad positive obligations. Articles
82 and 83 of the Protocol similarly lay down positive obligations concerning the
provision and availability of legal advisers, and dissem ination of the Conventions
and Protocol respectively.
As stated above, some provisions contain qualified obligations. For instance, Article
12(5) of the First Geneva Convention provides that a party to the conflict compelled to
abandon wounded or sick to the enemy must leave with them a part of its medical per­
sonnel as well as material, ‘as far as military considerations perm it’. Similarly, Article
12(2) of the Second Geneva Convention provides that enemy wounded, sick, or ship­
wrecked ‘shall not be wilfully left without medical assistance and care, nor shall condi­
tions exposing them to contagion or infection be created’ (emphasis added). Article 60
of the Third Geneva Convention imposes on the Detaining Power the duty to grant all
prisoners of war a monthly advance of pay, and specifies the amount of advance each

1 Similar provisions may be found in Articles 19 and 20 of the Second Geneva Convention (on the
wounded, sick, and shipwrecked at sea); as well as in Article 32(5) of the First Geneva Convention (on the
treatment of neutral personnel lending assistance to a belligerent and falling into the hands of the adversary
belligerent); in Articles 69-77 of the Third Geneva Convention (on prisoners of war), relating to relations
of prisoners of war with the external world; Article 118 of the same Convention, concerning release and
repatriation of prisoners of war at the close of hostilities (violation of this provision amounting to a grave
breach, pursuant to Article 85(4)(b) of the First Additional Protocol); Article 121 of the same Convention,
concerning the duty to establish an official inquiry into the death or serious injury of prisoners of war; and
Article 122 of the same Convention, providing for the establishment, by each belligerent, of an information
bureau concerning prisoners of war.
 C R IM IN A L L IA B IL IT Y FO R OMISSIONS 235

class of prisoner must obtain (depending on their rank); however, the provisions go on
to state that this amount may be modified by special agreement between the parties
to the conflict, or by the Detaining Power, subject to some conditions. Articles 55 and
56 of the Fourth Geneva Convention (on civilians), relating to provision of food and
medical supplies and hygiene and public health, respectively, are qualified by the pro­
viso ‘to the fullest extent of the means available’ to the Occupying Power.2

11.3 CULPABLE OMISSION OF AN ACT MANDATED

BY AN INTERNATIONAL CRIMINAL RULE

Serious violations of many of the above positive obligations (for instance, those enjoin­
ing to protect women and children from assault), as well as others laid down in other
provisions,3amount to an international crime, more specifically to war crimes, as held
by the ICTY AC in Tadic (IA): Article 7(1) of the ICTY Statute, on individual crim ­
inal responsibility also covers ‘the culpable omission of an act that was mandated by a
rule of criminal law (§188). Of course, it is necessary for the conditions set forth in the
same decision in Tadic (IA) (§§94-5), determining which violations may be regarded
as war crimes, to be met. However, in some instances, specified in the relevant provi­
sions, a serious violation may amount to a ‘grave breach’, with the attendant conse­
quences with regard to the mandatory character of ‘universal’ judicial repression at
the national level (see infra, 16.1).
As in the case of crimes consisting of positive conduct, criminal omission also may
only be punished if accompanied by a certain subjective frame of mind. As in those
cases, this mental element may vary, depending on the requirements of international
rules. Normally intent is required. However, in some cases the relevant rules or provi­
sions of international criminal law may require a less demanding subjective element,
such as recklessness.4 It would seem admissible to hold the view that, at least in some

2 Similarly, Article 69(1) of the First Additional Protocol imposes upon the Occupying Power the obliga­
tion to provide to the civilian population means for satisfying its basic need, ‘to the fullest possible extent
of the means available’ 'o that Power. Article 70 of the same Protocol provides for relief actions in favour
of the civilian population in occupied territories ‘subject to the agreement of the Parties concerned in such
relief actions’.
3 See, for instance, Sumida Haruzo and others (at 228-9, 278, and 280-2) for the breach of the duty
to provide food and care to detained civilians as a war crime. In Gozawa Sadaichi and others it was
held that the lack of food and medical supplies, as well as the existence of bad conditions for prisoners
of war, amounted to a crime of which the detaining authorities were guilty (at 200-1, 210-11, 222-3,
and 227-31). See also Schmitt (decision of the Antwerp Court Martial, at 751-2, and the subsequent
decision of the Cour militaire de Bruxelles, at 752, nt. 89 bis), as well as Koppelmann Ernst (decision of
the Brabant Court Martial, at 753-4, and of the Belgian Court of Cassation, at 185-6). In both cases
the courts dealt with the positive obligations of the commanders of detention camps for prisoners
of war.
4 In Ntagerura and others an ICTR TC set out the necessary subjective and objective elements of the
crime as follows: ‘The TC finds that in order to hold an accused criminally responsible for an omission as a
 236 IN T E R N A T IO N A L C R IM IN A L LAW

instances, and subject to strict conditions, even culpable negligence might suffice for
criminal liability to arise.5

11.4 THE RESPONSIBILITY OF SUPERIORS

11.4.1 T H E E M E R G E N C E O F T H E D O C T R I N E

Although it was adumbrated after the First World War,6 it was in the aftermath of
the Second World War that there evolved in international law the notion of criminal
responsibility of superiors for failure to prevent or punish crimes perpetrated by their
subordinates. The gradual evolution of ICL on the matter can be roughly divided up
into various phases.
At the outset law-makers and courts considered that m ilitary commanders were
to be held criminally liable for failure to prevent or punish, for in so acting they in
some way aided and abetted the crimes of their underlings. Some national laws set
out the notion tersely and conceived of such responsibility as a form o f complicity.7 For
instance, the French Order on War Crimes of 28 August 1944 provided in Article 4
that

principal perpetrator, the following elements must be established: (a) the accused must have had a duty to act
mandated by a rule of criminal law; (b) the accused must have had the ability to act; (c) the accused failed to
act intending the criminally sanctioned consequences or with awareness and consent that the consequences
would occur; and (d) the failure to act resulted in the commission of the crime’ (§658).
See also Blaskic (AJ,§ 663), Galic (AJ, § 175), and Brdanin (AJ, §§274-5).
5 A case where it would seem that a British court considered culpable negligence sufficient is Heinrich
Gerike and others (the Velpke Baby Home trial). The defendants were charged with war crimes for violat­
ing Article 46 (on respect for family honour and life by the Occupant) of the 1907 Hague Rules, for leaving
without food and care the children of Polish female workers compulsorily separated from their parents and
brought to a home for infants in Velpke; as a result of lack of care many children had died. The Prosecutor,
Major Draper (a Judge Advocate being absent) argued that the staff in charge of the children ‘were so grossly
and criminally negligent that they did in fact cause the death of something over 80 children in six months’
(at 326). He then noted that one of the questions arising in the case was whether ‘that neglect [was] more than
something that was gross and reckless, or was [...] wilful disregard of consequences to such an extent that
the party or parties responsible are deemed to have intended the natural and probable consequences of their
act, namely, that death would result’ (at 326). He then pointed out that ‘In either event it is the contention of
the Prosecution that they are within the charge which is laid before this Court, namely, that the accused are
concerned between the relevant dates in the killing by wilful neglect of a number of children, Polish nation­
als’ (ibid.). He then cited Archbold on gross negligence and recklessness (at 336-7), noting that his proposi­
tions were ‘in point in this case’ (at 337). The Court found two defendants not guilty (neither of them had
been entrusted with the care of the children; one had consistently disapproved of the running of the Home
and consequently decided to keep aloof, the other had tried unsuccessfully to have the Home removed); it
sentenced the remaining four either to death by hanging or to imprisonment (at 339-43).
6 See the proposals of the 1919 International Commission on the Responsibility of the Authors of the War
and on Enforcement of Penalties, in 14 AJIL (1920), at 121.
7 See the French Law of 1944 and the Chinese Law of 1946 quoted in UN Law Reports, vol. IV, at 87-8.
 C R IM IN A L L IA B IL IT Y FO R OMISSIONS 237

where a subordinate is prosecuted as the principal perpetrator of a war crime and his
hierarchical superiors may not be investigated as co-perpetrators, they shall be held to be
accomplices to the extent that they have organized or tolerated the criminal offences of their
subordinates (emphasis added).

Here the notion was clearly set out that a m ilitary commander is criminally liable
as an aider and abetter, if he tolerated—that is, failed to stop or repress—the commis­
sion of war crimes by his subordinates. A slightly broader notion was embodied in the
Chinese law of 24 October 1946 on the trial of war criminals, which, however, like the
French law, regarded culpable commanders as accomplices of the subordinates com­
mitting crimes.8 In 1949-50 two Belgian Courts M artial took the same approach in
Schmitt, although they stressed the notion that a commander is under a set of obliga­
tions, the breach of which may entail his crim inal liability.9
A further step in the evolution of the doctrine can be seen in a leading, if controver­
sial, US case, Yamashita (1946). In this case the first fully fledged enunciation of the
doctrine was propounded, again with regard to military commanders. The court did
not base itself on the notion of complicity but only stressed that command responsibil­
ity is consequent upon the breach of the duties incumbent upon commanders. Given
the importance of the case a few words of explanation prove necessary.
The Japanese general Yamashita had been Commanding General of the Japanese
Army in the Philippines between 1943 and 1945. His soldiers had massacred a large
part of the civilian population of Batangas Province and inflicted acts of violence, cru­
elty, and m urder upon the civilian population and prisoners of war, as well as wholesale
pillage and wanton destruction of religious monuments in the occupied territory. The
US authorities accused the General, before a US M ilitary Commission, of breaching

8 ‘Persons who occupy a supervisory or commanding position in relation to war criminals and in their
capacity as such have not fulfilled their duty to prevent crimes from being committed by their subordinates
shall be treated as the accomplices of such war criminals/
9 The Antwerp Court Martial dealt with the case of a German head of a prisoners of war camp at
Breedonck where many inmates died either of fatigue for the forced labour to which they had been subjected
or of starvation, whereas some 32 inmates were killed by some of the prison guards. The Court applied
Art. 66 of the Belgian Penal Code (which made liable for a crime both the perpetrators and aiders and abet­
ters). It stressed that the defendant, as head of the camp ‘had the positive duty to protect prisoners in his
custody’ (at 751). The Court therefore found that he was accountable, as co-perpetrator, for the killing for the
32 inmates, whereas for the death of the inmates resulting from excessive fatigue or starvation he was liable
as an accomplice, on account of his breach of his duty ‘since he had rendered such assistance that without it
the crimes could not be committed’; ‘he had seriously breached his duty as head of the camp and hence vol­
untarily and consciously cooperated to the criminal activity of the Sicherheitsdienst [the SS branch whose
members were in charge of the camp at his orders] (ibid.). On appeal, a Military Court of Appeal upheld the
decision and noted that the defendant’s action was twofold: ‘positive’, where he imposed exhausting labour
and ordered the destruction of food parcels, and ‘negative’, where he refrained to step in to prevent cruel acts.
The appellant, the Court went on to hold, must be punished for both classes of conduct. As for the latter, he
was punishable for the breach of his duty to see to it that ‘the inmates in his camp be adequately nourished
and treated’ so as not to ‘become physically exhausted and unable to work’; this duty, the Court noted,
was similar to ‘that incumbent upon a person charged with nourishing another person unable to attend to
himself, and who gets him to starve’. In this case ‘the failure to act constitutes the material act sufficient to
evidence criminal intent’ (at 752).
238 IN T E R N A T IO N A L C R IM IN A L LAW

his duty as an army commander to control the operations of his troops ‘by permitting
them to commit’ extensive and widespread atrocities. The Commission upheld these
submissions by setting out a new doctrine as follows:
Clearly, assignment to command military troops is accompanied by broad authority and
heavy responsibility. This has been true in all armies throughout recorded history. It is
absurd, however, to consider a commander a murderer or rapist because one of his soldiers
commits a murder or a rape. Nevertheless, where murder and rape and vicious, revenge­
ful actions are widespread offences, and there is no effective attempt by a commander to
discover and control the criminal acts, such a commander may be held responsible, even
criminally liable, for the lawless acts of his troops, depending upon their nature and the
circumstances surrounding them (1597).

The US Supreme Court, to which the case had been brought by the defendant by way
of a petition for habeas corpus agreed. It held that commanders had a duty to take such
appropriate measures as are within their power to control the troops under their com­
m and for the prevention of violations of the laws of warfare. It derived this duty from
a number of provisions of such laws: Articles 1 and 43 of the Regulations annexed
to the Fourth Hague Convention of 1907 (under the former, combatants, to be rec­
ognized as legitimate belligerents, must ‘be commanded by a person responsible for
his subordinates’; pursuant to the latter, the commander of a force occupying enemy
territory ‘shall take all the measures in his power to restore, and ensure, as far as pos­
sible, public order and safety, while respecting, unless absolutely prevented, the laws
in force in the country’); Article 19 of the Tenth Hague Convention of 1907, relating
to bombardment by naval vessel and providing that commanders-in-chief of the bel­
ligerent vessels ‘must see that the above Articles are properly carried out’; Article 26 of
the 1929 Geneva Convention on the wounded and sick, which made it the duty ‘of the
commanders-in-chief of the belligerent armies to provide for the detail of execution of
the foregoing Articles [of the Convention] as well as for unforeseen cases’. The C ourt’s
majority held that these provisions made it clear that the accused had
an affirmative duty to take such measures as were within his power and appropriate in the
circumstances to protect prisoners of war and the civilian population. This duty of a com­
manding officer has heretofore been recognized, and its breach penalized by our own mili­
tary tribunals (13).

However, two judges, Murphy and Rutledge, forcefully (and rightly) disagreed and
set forth their views in im portant Dissenting Opinions. They noted among other things
that the Court’s majority had not shown that Yamashita had ‘knowledge’ of the gross
breaches perpetrated by his troops (at 31, 36, 48-9, 50) or had any ‘direct connection
with the atrocities’ (at 36), or could be found guilty of ‘a negligent failure [...] to dis­
cover’ the atrocities (at 49) or in other words, had ‘personal culpability’ (at 36-79).10

10 Justices Murphy and Rutledge did not only dissent on the application of the law to the facts by the
Commission—they also objected to the whole notion of command responsibility as a matter of law. Justice
Murphy stated: ‘The recorded annals of warfare and the established principles of international law afford
 C R IM IN A L L IA B IL IT Y FOR OMISSIONS 239

This is therefore a case where the principle was affirmed, based (as the two dissenting
judges rightly noted), on a novel interpretation of existing rules of IHL, as well as a
questionable application of the principles to the case at bar, in addition to total disre­
gard for the required mental element for the crime.
Although case law thus started off on the wrong foot, soon other decisions handed
down after the Second World War followed suit and fleshed out the doctrine. Unlike
Yamashita, these decisions, which can be considered as a third step in the formation
of the doctrine at issue, emphasized the need for the commander to have knowledge
of the crimes committed by his underlings, in some instances also requiring criminal
intent for the commander’s liability to arise. They all neglected the notion of compli­
city. Furthermore, in some cases the doctrine was extended to civilian leaders.
In Karl Brand and others (Doctors case), a US Military tribunal sitting at Nuremberg
under Control Council Law no. 10 held the German top medical staff liable for the
killings perpetrated by their subordinate doctors, stressing that those leaders had
knowledge of what was going on.11 In Pohl and others, a US M ilitary Tribunal held that
‘the law of war imposes on a military officer in a position of command an affirmative
duty to take such steps as are within his power and appropriate to the circumstances to
control those under his command for the prevention of acts which are violations of the
law of war’ (at 1011). The Tribunal required ‘actual knowledge’ of the misdeeds of sub­
ordinates (1011-2). The same doctrine was set out in a subsequent case, Wilhem List
and others (Hostages case), where another US Military tribunal sitting at Nuremberg

not the slightest precedent for such a charge. This indictment in effect permitted the military commission to
make the crime whatever it willed dependent on its biased view as to the petitioner’s duties and his disregard
thereof, a practice reminiscent of that pursued in less respected nations in recent years’ (327 US, at 28).
11 After citing the Yamashita case, the Tribunal stated: ‘This decision is squarely in point as to the crim­
inal responsibility of those defendants in this dock who had the power and authority to control the agents
through whom these crimes were committed. It is not incumbent upon the prosecution to show that this or
that defendant was familiar with all of the details of all of these experiments. Indeed, in the Yam ashita case,
there was no charge or proof that he had knowledge of the crimes [...] But we need not discuss the require­
ment of knowledge on the facts of this case. It has been repeatedly proved that those responsible leaders of
the German medical services in this dock not only knew of the systematic and criminal use of concentration
camp inmates for murderous medical experiments, but also actively participated in such crimes. Can it be
held that Karl Brandt had no knowledge of these crimes when he personally initiated the jaundice experi­
ments by Dohmen in the Sachsenhausen concentration camp and the phosgene experiments of Bickenbach?
Can it be found that he knew nothing of the criminal Euthanasia Program when he was charged by Hitler
with its execution? Can it be said that Handloser had no knowledge when he participated in the conference
of 29 December 1941 where it was decided to perform the Buchenwald typhus crimes, when reports were
given on criminal experiments at meetings called and presided over by him? Was Rostock an island of
ignorance when he arranged the program for and presided over the meetings at which Gebhardt and Fischer
lectured on their sulfanilamide experiments, when he classified as “urgent” the criminal research of Hirt,
Haagen, and Bickenbach? Did Schroeder lack knowledge when he personally requested Himmler to supply
him with inmates for the sea-water experiments? Can it be found that Genzken had no knowledge of these
crimes when the miserable Dr. Ding was subordinated to and received orders from him in connection with
the typhus experiments in Buchenwald, when his office supplied Rascher with equipment for the freezing
experiments? Was Blome insufficiently informed in the face of proof that be collaborated with Rascher in the
blood coagulation experiments, issued a research assignment to him on freezing experiments and to Hirt
on the gas experiments, as well as performed bacteriological warfare and poison experiments himself? No,
it was not lack of information as to the criminal program which explains the culpable failure of these men to
destroy this Frankenstein’s monster. Nor was it lack of power’ (934-5).
 240 INTERNATIONAL CRIMINAL LAW

applied it to 12 high-ranking German officers charged, among other things, with the
unlawful killing of hostages by way of reprisal. In this case the Tribunal stressed that, to
pronounce a guilty verdict, it required proof of a causative, overt act or omission from
which a guilty intent can be inferred’ (1261). Turning to the liability of the defendants
for their failure to prevent or punish, the Tribunal noted that, for this form or c r i m i n al
liability to arise, knowledge by the army commander of the crimes committed by the
subordinates was required. Furthermore, the Tribunal emphasized that a commander
has the duty to require reports about occurrences taking place in the area under his
control, failing which he may be accused o f ‘dereliction of duty’ (at 1271-2).12 These
notions were taken up and elaborated on by another US Tribunal sitting at Nuremberg
in Wilhelm von Leeb and others (High Command case). The Tribunal noted that a com­
m ander’s criminal responsibility is personal. The act or neglect to act must be volun­
tary and crim inal’ (at 543). It went on to note that
there must be a personal dereliction. That can occur where the act is directly traceable to
him [the commander] or where his failure to properly supervise his subordinates constitutes
criminal negligence on his part. In the latter case it must be a personal neglect amounting to
a wanton, immoral disregard of the action of his subordinates amounting to acquiescence.
[...]. the occupying commander must have knowledge of these offences [by his troops] and
acquiesce or participate or criminally neglect to interfere in their commission and [...] the
offences committed must be patently criminal (543-5).13

The doctrine was not only embraced by US tribunals. The International Tokyo
Tribunal also upheld it in Araky and others (at 29-31). In dealing with responsibility
for war crimes against prisoners of war, the Tribunal insisted on the liability of com­
manders on account of their ‘negligence or supineness’ (at 30) if a commander that had
the duty to know ‘knew or should have known’ the commission of crimes but failed to
stop them or to take ‘adequate steps’ ‘to prevent the occurrence of [...] crimes in the
future’ (at 31).14 Similarly, the doctrine was enunciated by an Australian-US Military

12 'If he fails to require and obtain complete information, the dereliction of duty rests upon him and he is
in no position to plead his own dereliction as a defence’ (1271).
13 The Tribunal also noted the following: ‘Military subordination is a comprehensive but not conclusive
factor in fixing criminal responsibility. The authority, both administrative and military, of a commander
and his criminal responsibility are related but by no means co-extensive. Modern war such as the last war,
entails a large measure of de-centralization. A high commander cannot keep completely informed of the
details of military operations of subordinates and most assuredly not of every administrative measure.
He has the right to assume that details entrusted to responsible subordinates will be legally executed. The
President of the United States is Commander-in-Chief of its military forces. Criminal acts committed by
those forces cannot in themselves be charged to him on the theory of subordination. The same is true of
other high commanders in the chain of command. Criminality does not attach to every individual in this
chain of command from that fact alone. There must be a personal dereliction. That can occur only where the
act is directly traceable to him or where his failure to properly supervise his subordinates constitutes crim­
inal negligence on his part’ (at 543).
14 The Tribunal found the highest-ranking defendant, prime minister Hideki Tojo, guilty of acts of omis­
sion in that ‘He took no adequate step to punish offenders [who had ill-treated prisoners and internees] and
to prevent the commission of similar offences in the future. His attitude towards the Bataan death March
gives the key to his conduct towards these captives. He knew in 1942 something of the conditions of that
 CRIMINAL LIABILITY FOR OMISSIONS 24 1

Tribunal, in Soemu Toyoda (at 5005-6) and by a Chinese War Crimes Tribunal in
Takashi Sakai (at 1-7). It is notable that in Soemu Toyoda the Tribunal, besides insist­
ing on the need for knowledge as a requirement of command responsibility, also held
that such knowledge m aybe either actual’, as in the case of an accused who sees’ the
commission of the subordinates’ crimes or ‘is informed thereof shortly after’, but also
‘constructive knowledge’, which can be asserted to exist when there is
the commission of such a great number of offences within his command that a reasi nable man
could come to no other conclusion than that the accused must have known the offences or of
the existence of an understood and acknowledged routine of their commission (5005-6).

It can thus be held that in a matter of few years after the Second World War the
doctrine of command responsibility crystallized into an international customary rule
(i) imposing on military commanders as well as civilian or civilian leaders15 the obli­
gation to prevent or repress crimes by their subordinates if they knew or should have
known that the troops were about or were committing or had committed crimes; and
(ii) criminalizing the culpable failure to fulfil this obligation, albeit without clearly
outlining the mental element of such criminal liability. That such a rule (the existence
of which was authoritatively asserted in Delalic and others, TJ, §343) evolved so quickly
should not surprise. In modern times international criminality increasingly tends to
be planned, organized, ordered, or condoned or tolerated by superior authorities. In
other words, a clear trend is emerging in the world community towards commission
of crimes either by high-level military or political leaders or by low-level officials or
military personnel, who, however, perpetrate crimes because superior authorities (be
they military or civilian) do not prevent, or tolerate or at any rate fail to repress them.
Hence, the issue of superior responsibility has gradually acquired enormous im port­
ance in international criminal law.
Subsequently the customary criminal rule was enshrined in the Statutes of the
ICTY, ICTR, and the ICC and has been relied upon in many cases brought before the
ICTY and the ICTR. It covers superior responsibility for any international crime com­
mitted by subordinates; that is, not only war crimes but also crimes against humanity,
genocide, etc.
It is notable that after the establishment of the ICTY and the ICTR the doctrine was
gradually refined by case law, also under the influence of the 2002 German Code of

march and that many prisoners had died as a result of these conditions. He did not call for a report on the
incident. When in the Philippines in 1943 he made perfunctory inquiries about the march but took no
action. No one was punished.[... ] Thus the head of the Government of Japan knowingly and wilfully refused
to perform the duty which lay upon the Government of enforcing performance of the Laws of War’ (at 462).
15 For this latter category of cases see in particular, Azaky and others (the Tokyo trial), heard by the Tokyo
International Tribunal (vol. 20, at 791, 816, 831), Flick and others, brought before a US Military Tribunal sit­
ting at Nuremberg (at 1202-12), Röchling and others, heard by a French court in the French Occupation Zone
in Germany (at 8, or 403-4), and Delalic and others (§§370,377-8).
 242 INTERNATIONAL CRIMINAL LAW

Crimes Against International Law16 and some leading commentators.17 As a conse­

quence, the criminal liability of the superior was increasingly seen as a consequence of
his own culpability, not necessarily linked by means of a causal nexus to the responsi­
bility of the subordinates (see below).
It should be added that in 2003 the ICTY AC rightly set out the notion that com­
m and responsibility also applies in time of internal armed conflict, basically because
also with regard to such conflicts a general principle of international law assumes that
there must be an organized m ilitary force: military organization implies responsible
command and [...] responsible command in turn implies command responsibility’.18
It is striking that in this area there has been an inversion of the normal process
whereby states first develop an international rule binding upon them, namely an inter­
state rule imposing a certain behaviour, and then this rule gradually evolves as a penal
rule criminalizing any conduct contrary to the standards imposed by the interstate
rule (see above, 1.2 at point 4). In this case there first emerged a criminal law rule
(admittedly based on a general principle of IHL concerning ‘responsible com m and’)
that addressed itself to individuals (military commanders or civilian or political lead­
ers); then a written rule was agreed upon by states imposing on them to see to it that
their commanders prevent or repress crimes by their subordinates. This is Article 87 of
the First Additional Protocol, which is addressed to the Contracting parties and to the
Parties to a conflict, and spells out as well as codifies the principle on responsible com­
m and mentioned above.19This rule is accompanied by Article 86 of the same Protocol,
which in §2 restates the customary crim inal law rule.
As noted above, this class of responsibility is different from the others considered
so far, in that it is responsibility by omission: the person is criminally liable not for an
act he has performed, but for failure to perform an act required by international law.
In other words, he is responsible for the breach of an international obligation incum ­
bent upon any commander or superior authority, to prevent or suppress crimes by
subordinates.20

16 The Code is important for it draws a clear distinction between three different hypotheses: Responsibility
of superiors (Section 4), Violations of the duty of supervision (Section 13), and Omission to report a crime
(Section 14), thus identifying the distinct mens rea required for each of these classes.
17 See, in particular, M. Damaska, ‘The Shadow Side of Command Responsibility’, 49 Am. J. Comparative
Law (2001), 455-96.
18 Hadzihasanovic, Alagic and Kubura, Decision on Interlocutory Appeal Challenging Jurisdiction in
Relation to Command Responsibility, §17 and see §§11-36; see also in the same case the TC Decision on Joint
Challenge to Jurisdiction, §§67-179.
19 ‘The High Contracting Parties and the Parties to the conflict shall repress grave breaches, and take
measures necessary to suppress all other breaches of the Conventions or of this Protocol which result from
a failure to act under a duty to do so.’
20 According to the ICTY AC in Delalic and others (AJ) (§240) there is no duty, incumbent upon military
or civilian authorities, to ascertain that their subordinates are not committing crimes. This proposition is
questionable, in light of the abundant case law on the matter, as well as some clear treaty provisions and pro­
visions of important Military Manuals. With regard to international rules, it may suffice to mention Article
87 of the First Additional Protocol of 1977, on ‘Duty of commanders’. The obligation in question is clearly
set out in many national Military Manuals, for instance, those of Switzerland, Reglement (1987), Article
196 (‘Les commandants doivent informer la troupe de ses obligations aux termes des Conventions. Ils sont
 CRIMINAL LIABILITY FOR OMISSIONS 243

11.4.2 C R IM IN A L C A T E G O R IE S IN T O W H IC H T H E G E N E R A L
N O T IO N M A Y BE S U B D IV ID E D

International rules tend to lump together various classes of superior responsibility,

without drawing any distinction. This, for instance, holds true both for Articles 7(3)
and 6(3) of the Statutes of the ICTY and ICTR, respectively and for Article 28 of the
ICC Statute. These provisions are essentially of a descriptive nature, in that they indi­
cate the prohibited conduct by enumerating the various forms this conduct may take.
They do not, however, differentiate between the various categories of liability that can
be logically identified, nor do they attach any legal relevance to conduct falling under
one particular category rather than another.
Nonetheless, it is both logically appropriate and also relevant for the practical p ur­
poses of sentencing to draw a distinction between different classes. It is not sound, for
instance, to hold that a commander who fails to punish subordinates who previously,
unbeknown to him, have perpetrated acts of genocide, is responsible for genocide, if
only as an accomplice. Plainly, in this case the requisite conduct and the mens rea of
the superior are neatly different from those required for the perpetrators of genocide,
or for persons aiding and abetting genocide. Only when the superior in some way
knows of the crime being or about to be perpetrated and willingly fails to check or
prevent its commission, may he be deemed to participate in some way in the crime
(according to some commentators, as a co-perpetrator or accomplice), for in this case
there is a clear nexus of causality between the superior’s omission and the crime.

responsables du fait que leurs troupes respectent les Conventions et de punir d’eventuelles infractions’);
Russia’s Military Manual (1990), Part VII, §§a and b (commanders of all grades must call to account per­
sons who committed violations of the rules of international humanitarian law defined by Articles 85-7 of the
First Additional Protocol’); Germany, Military Manual (1992), ch. 1, no. 138; New Zealand, Military Manual
(1992), ch. 16, s. 2, §1603-2 (‘It is incumbent upon a commanding officer to ensure that the forces under his
command behave in a manner consistent with the laws and customs of war [...] and it is part of his responsi­
bility to ensure that the troops under his command are aware of their obligations’); Australia, Defence Force
Manual (1994), §1304 (‘Military commanders of all Services and at all levels bear responsibility for ensuring
that forces under their control comply with the Law of Armed Conflict’); Benin, Military Manual (1995),
ch. V (‘Chaque chef militaire est responsable du respect du droit de la guerre dans sa sphere de commande-
ment [...] il est particulierement responsable de l’instruction du droit de la guerre afin de communiquer
a sa troupe un comportement conforme au droit’); Canada, Law of Armed Conflict Manual (1999), at 15-1
and 16-1 (‘Commanders have responsibility to ensure that forces under their command are aware of their
responsibilities’); and France, Manual on the Law of Armed Conflict (2001), Introduction, at 14, para. 7 (‘Le
commandement [... ] doit s’assurer que les membres des forces armees connaissent leurs droits et appliquent
les obligations qui en sont le parallele. II est a ce titre responsable de leur instruction’).
As for case law, one may recall, in addition to Yamashita (see supra, 11.4.1), the instructions a Judge
Advocate issued to a US Court Martial in Medina; he stated: ‘In relation to the question pertaining to the
supervisory responsibility of a Company Commander, I advise you that as a general principle of military
law and custom a military superior in command is responsible and required, in the performance of his com­
mand duties, to make certain the proper performance by his subordinates of their duties as assigned by him.
In other words, after taking action and issuing an order, a commander must remain alert and make timely
adjustments as required by a changing situation’ (1732).
 244 INTERNATIONAL CRIMINAL LAW

This approach, which seems logically and theoretically correct and also consonant
with general principles of justice (because of its consequences at the level of senten­
cing), leads to distinguishing three categories:

1. A commander or superior breaches his duty to prevent his subordinates from

engaging in criminal conduct. He knows that an offence is about to be, or is being,
committed by his subordinates and willingly fails to stop the crime. In this case, the
superior has knowledge of the crime and its omission is deliberate (intent).
According to one view the offender should be legally treated as a co-perpetrator,
although the crimes are physically committed by the subordinates, or at least as an
accomplice. In a way the superior participates in the crime, for if he had acted to stop
it, the delinquency would not have been perpetrated. There is therefore a causal link
between the superior’s attitude and the commission of crimes. Arguably this view was
reflected to some extent not only in the French and Chinese laws mentioned above,
and in Araki and others (Tokyo trial), at 30-1, but also in the German Law on Crimes
against International Law (Section 4).21
Under a different view (that would seem to be reflected in e.g. Halilovic, TJ, §54;
Hadzihasanovic and Kubura, TJ, §75), the superior is responsible for violating his own
duty to prevent or stop misconduct by his subordinates; both the objective and the sub­
jective elements of his crime are different from those of the subordinates. For instance,
if the underlings have committed large-scale rape within a context of systematic attack
on civilians, the conduct (sexual assault) and the mens rea (intent to sexually abuse
a civilian in a grave manner, plus awareness of the systematic nature of the attack on
civilians) are different from those of the commander, who may be accused of failure to
act (conduct) with knowledge that crimes were being or were about to be perpetrated,
and intent not to stop or forestall them. True, the subjective and objective elements of
the criminal offence attributable to the superior are not far from those of aiding and
abetting: in both cases the person other than the perpetrator does not share the crim ­
inal intent of the perpetrator, but knows the crime that the perpetrator is committing
or will commit, and in both cases the person at issue provides assistance to the perpet­
rator (in the case under discussion by not preventing the commission of his crimes).
Nevertheless, the fact remains that the aider and abetter, by his action or omission,
intends to further the act of the perpetrator, and this element must be proved by the
prosecution; instead, in the case of command responsibility that intent is not a legal
requirement, and consequently need not be proved in court, although it may happen
that the commander by his inaction aimed in fact at furthering the crime of the subor­
dinate. Therefore, although for sentencing purposes the conduct of the superior may

21 Section 4(1) of the Code provides that the commander ‘shall be punished in the same way as the per­
petrator of the olfence committed by [the] subordinate’. And the Explanatory Memorandum of the German
Government states that ‘from a theoretical viewpoint’ ‘the negligence’ of the superior ‘could be classified as
mere complicity’ (at 39).
 CRIMINAL LIABILITY FOR OMISSIONS 245

be as blameworthy as that of the subordinates, the legal ingredients of the crimes are
different.2223
2. A commander or a superior breaches his duty properly to supervise the con­
duct of his troops or underlings. He intentionally or negligently omits to monitor the
actions of his subordinates, where he could have become cognizant of the imminent
commission of the offence or of the fact that the offence was being committed, and
therefore prevented it. Here the superior does not know that the subordinate is about
to commit or is committing a crime: he lacks knowledge. However, his failure to know
derives from his negligent or deliberate breach of his duty of supervision, with the con­
sequence that he does not impede the perpetration of crimes that he could foresee and
avoid. In these cases the offence imputable to the superior is arguably different from
and less serious than that perpetrated by the subordinate, in that it merely consists of
the deliberate or negligent dereliction o f supervisory duties.23.
However, a different view is also admissible, although it is arguably less persuasive.
One can contend that failure by the superior to exercise his duty of supervision has a
causal link with the drime, in that by breaching his supervisory duty he has in some
way contributed to bringing about the offence. In other words, the superior’s conduct
may be considered as serious as that of the subordinate; the former could therefore be
punished by a sentence similar to that of the subordinate.
3. A superior breaches his duty to report to the appropriate authorities crimes com­
mitted by his subordinates unbeknownst to him. Here the superior knows that a crime
has been perpetrated and fails immediately to draw the attention of the body respon­
sible for the investigation or prosecution of the crime. In this case, the superior is liable
to be punished for the specific crime of failure to report. His offence is plainly different
from that of his subordinatesdhe is responsible if, upon becoming cognizant of the
crimes of his subordinates, he deliberately or with culpable negligence fails to report

22 In Hadzihasanovic and Kubura, the TC held that there must be a link or nexus between the superior’s
omission and the crimes in the sense that the superior’s omission created or heightened a real and reason­
ably foreseeable risk that those crimes would be committed, a risk he accepted willingly’, a risk that ‘mate­
rialised in the commission of those crimes. In that sense, the superior has substantially played a part in the
commission of those crimes. [... ] it is presumed that there is such a nexus between the superior’s’ omission
and those crimes’ ( TJ, §§193).
23 With respect to the supervisory duties of a commander, the holding of the US Military Tribunal in
Wilhelm List and others (Hostage case) is instructive. Since the defence of List (commander in chief of the
German armed forces in 1941-2), had alleged that he had no knowledge of the killings of civilians in occu­
pied territory, the Tribunal noted the following: ‘A commanding general of occupied territory is charged
with the duty of maintaining peace and order, punishing crime, and protecting lives and property within the
area of his command. His responsibility is coextensive with his area of command. He is charged with notice of
occurrences taking place within that territory. He may require adequate reports of all occurrences that come
within the scope of his power and, if such reports are incomplete or otherwise inadequate, he is obliged to
require supplementary reports to apprize him of all the pertinent facts. If he fails to require and obtain com­
plete information, the dereliction of duty rests upon him and he is in no position to plead his own dereliction
as a defence’ (1271; emphasis added).
 246 INTERNATIONAL CRIMINAL LAW

them to the appropriate authorities for punishment. Here the superior’s conduct may
not be held to have caused, or contributed to cause, the criminal offence.24
In a case brought before the ICTY (Hadzihasanovic, Alagic and Kubura) the AC
dismissed the proposition (upheld instead by the TC: Decision on Joint Challenge
to Jurisdiction, §§197-202) that a commander may also be responsible for failure to
report crimes committed before he took command of the relevant unit. The main rea­
son for this holding is that there is no practice or opinio juris to support the propos­
ition (Decision on IA Challenging Jurisdiction in Relation to Command Responsibility,
§§37-56). It would seem instead that the proposition is correct (as was rightly opined
by Judges Hunt and Shahabuddeen in their dissenting opinions appended to the deci­
sion of the AC). It is not necessary to search for a specific customary rule on the mat­
ter. The duty to report follows, as in the case of crimes committed by the underlings
while the commander was in control, from the general principles on superior respon­
sibility set out by the AC in the same case (see §§12-18). If international law imposes
on a military commander the obligation to report to the appropriate authorities any
crime committed by his subordinates, clearly this obligation applies whether or not
the crimes have been committed when he was the commander. The purpose of the
obligation incumbent upon any person in a position of command to make his subor­
dinates criminally accountable is twofold: (i) to ensure military discipline and respect
for IHL; and (ii) to avoid the troops interpreting any inaction by the superior as an
implicit approval of their misconduct. It does not matter at all whether the crimes
were perpetrated when he was in control of the troops or prior to that date: this cir­
cumstance is immaterial to the fulfilment of the obligation. The contrary view is based

24 The various categories are instead merged in Toyoda. The Tribunal stated the following: ‘The Tribunal
considers the essential elements of command responsibility for atrocities of any commander to be: 1. That
offenses, commonly recognized as atrocities, were committed by troops of his command; 2. The ordering of
such atrocities. In the absence of proof beyond a reasonable doubt of the issuance of orders, then the essential
elements of command responsibility are: 1. As before, that atrocities were actually committed; 2. Notice of
the commission thereof. This notice may be either: a. Actual, as in the case of an accused who sees their com­
mission or who is informed thereof shortly thereafter; b. Constructive; that is the commission of such a great
number of offenses within his command that a reasonable man could come to no other conclusion than that
the accused must have known of the offenses or of the existence of an understood and acknowledged routine
for their commission. 3. Power of command; that is, the accused must be proved to have had actual author­
ity over the offenders to issue orders to them not to commit illegal acts, and to punish offenders. 4. Failure
to take such appropriate measures as are within his power to control the troops under his command and to
prevent acts which are violations of the laws of war. 5. Failure to punish offenders. In the simplest language
it may be said that this Tribunal believes the principle of command responsibility to be that, if this accused
knew, or should by the exercise of ordinary diligence have learned, of the commission by his subordinates,
immediate or otherwise, of the atrocities proved beyond a shadow of a doubt before this Tribunal or of the
existence of a routine which would countenance such, and, by his failure to take any action to punish the per­
petrators, permitted the atrocities to continue, he has failed in his performance of his duty as a commander
and must be punished’ (5005-6).
In Bagilishema the ICTR AC rightly insisted on the fact that the information about the crimes must be
specific, namely specifically related to the crimes by subordinates. It stated that it was necessary ‘to make a
distinction between the fact that the Accused had information about the general situation that prevailed in
Rwanda at the time, and the fact that he had in his possession general information which put him on notice
that his subordinates might commit crimes’ (§42). See also Krnojelac, TJ, §§312-13, AJ, §§165-71.
 CRIMINAL LIABILITY FOR OMISSIONS 247

on a misapprehension of the various categories of command responsibility and the

fact that ‘failure to report’ is a distinct category from the others, where the breach lies
in a dereliction of the duty to inform other authorities of the crimes so th?t they take
action to punish the perpetrators. In addition, as Judge Shahabuddeen rightly empha­
sized in his dissenting opinion (§14), one of the consequences of the AC ruling is that
the subordinates’ crimes may go unpunished: if the crimes were committed ‘shortly
before the assumption of duty of the new commander—possibly, the day before, when
all those in previous command authority disappeared', and were not reported by the
then commander, and the new commander were not obliged to report them even if he
knows that the crimes were committed, the crimes would not be punished by anyone.
This would clearly be contrary to the notion that superiors are legally bound to make
their subordinates criminally accountable.

11.4.3 G E N E R A L C O N D IT IO N S O F S U P E R IO R R E S P O N S IB IL IT Y

Before trying to identify the mental element required for each of these three categories,
it may be helpful to set out the general conditions required for all these categories.
Superior authorities, whether military or civilian, bear responsibility for crimes
committed by their subordinates if the following cumulative conditions are met:

1. Commission of international crimes by troops or other subordinates. It is not

necessary for the troops or the other subordinates to have physically perpetrated
the crimes. They may have engaged in criminal conduct under any head of liability,
namely perpetration, co-perpetration, aiding and abetting, joint criminal enterprise
to commit crimes, etc.25
2. Effective command and control over the subordinates. It is not necessary for there
to be a formal hierarchical structure. Individuals in positions of authority, whether
within civilian or military structures, may incur criminal responsibility under the
doctrine of command responsibility on the basis of their de facto or de jure position
as superiors. The mere absence of formal legal authority to control the actions of
subordinates should therefore not be understood to preclude the imposition of such
responsibility.26
Control must be effective. Thus in von Weizsäcker and others (Ministries case) a
US Military Tribunal sitting at Nuremberg held that the mere appearance of an offi­
cial’s name on a distribution list attached to an official document could only provide
evidence that it was intended that he be provided with the relevant information, and
not that ‘those whose names appear on such distribution lists have responsibility for,
or power and right of decision with respect to the subject-matter of such docum ent’

25 See ICTY, Boskovski and Tarculovski,TC,Decisionon the Prosecution’s Motion to Amend the Indictment,
§§18-20; Oric, TJ, 297-8.
26 Delalic and others (TJ, §§377-8), AJ (§§197-8); Kordic and Cerkez (TJ, §§405-7).
 248 INTERNATIONAL CRIMINAL LAW

(at 693).27 In Blaskic an ICTY TC held that ‘what counts is his material ability [of a
superior to control the subordinate], which instead of issuing orders or taking discip­
linary actions may entail, for instance, submitting reports to the competent authorities
in order for proper measures to be taken’ (§302). In Kordic and Cerkez an ICTY TC
provided some im portant examples.28 And in Cappellini and others the Milan Court of
Cassation held that a superior who in fact had been deprived of his authority, although
he still was formally vested with his position, could not be held responsible for crimes
perpetrated by his subordinates unbeknownst to him or even in breach of his orders,
for lack of the required intent (at 86-7).
It bears noting that in Kordic and Cerkez, the TC found that one of the accused,
Kordic, a civilian leader and politician having ‘tremendous influence’ and playing an
im portant role in military matters, nevertheless did not possess the authority to pre­
vent the crimes that were being committed, or to punish the perpetrators. It therefore
acquitted the accused of charges involving command responsibility, while nonethe­
less convicting him of various offences on the basis of perpetration under Article 7(1)
of the Statute (§§838-41).
A question that can also arise is how effective the control over subordinates must
be when crimes are perpetrated by irregular armies or rebel groups. The question was
convincingly discussed by an SCSL TC in Brima and others.29

27 See, e.g., Delalic and others, §354-78; Delalic and others (Appeal), §§192-5; Blaskic, §§295-303; Kordic
and Cerkez, §§405-17).
28 It thus stated that: ‘For instance, a government official who knows that civilians are used to perform
forced labour or as human shields will be held liable only if it is demonstrated that he has effective control
over the persons who are subjecting the civilians to such treatment. A showing that the official merely was
generally an influential person will not be sufficient. In contrast, a government official specifically in charge
of the treatment of prisoners used for forced labour or as human shields, as well as a military commander in
command of formations which are holding the prisoners, may be held liable on the basis of superior respon­
sibility because of the existence of a chain of command’ (§415).
In addition, with reference to civilian authorities, the same TC stated in the same case: ‘Evidence that
an accused is perceived as having a high public profile, manifested through public appearances and state­
ments, and thus as exercising some authority, may be relevant to the overall assessment of his actual
authority although not sufficient in itself to establish it, without evidence of the accused’s overall behav­
iour towards subordinates and his duties. Similarly, the participation of an accused in high-profile inter­
national negotiations would not be necessary in itself to demonstrate superior authority. While in the case
of military commanders, the evidence of external observers such as international monitoring or humani­
tarian personnel may be relied upon, in the case of civilian leaders evidence of perceived authority may not
be sufficient, as it may be indicative of mere powers of influence in the absence of a subordinate
structure’ (§424).
29 According to the TC, in a conflict involving irregular armies or rebel groups, 'the traditional indicia
of effective control provided in the jurisprudence may not be appropriate or useful’(§787). Such indicia
include ‘that the superior had first entitlement to the profits of war, such as looted property and natural
resources; exercised control over the fate of vulnerable persons such as women and children; the superior
had independent access to and/or control of the means to wage war, including arms and ammunition and
communications equipment; the superior rewarded himself or herself with positions of power and influ­
ence; the superior had the capacity to intimidate subordinates into compliance and was willing to do so;
the superior was protected by personal security guards, loyal to him or her, akin to a modern praetorian
guard; the superior fuels or represents the ideology of the movement to which the subordinates adhere; and
the superior interacts with external bodies or individuals on behalf of the group’(§788). The TC, however,
 CRIMINAL LIABILITY FOR OMISSIONS 249

Another question has arisen before international courts: whether commanders of a

unit engaged in joint combat with other units may be held liable for acts of these other
units formally not under their de jure command. In Hadzihasanovic and Kubura an
ICTY TC held that ‘mere participation in joint combat operations is not sufficient
to find that commanders of different units exercise effective control over all partici­
pants in battle. Although such cooperation might be an indicator of effective control,
it is appropriate to determine on a case-by-case basis what authority an accused com­
mander actually had over the troops in question’ (TJ, §84).
3. Knowledge (or constructive knowledge, namely knowledge that can be inferred
from or implied by the conduct of the persons involved, the surrounding circum ­
stances, etc.) or breach of the obligation to acquire knowledge. The superior knew, or
had information which should have enabled him to conclude in the circumstances at
the time, that crimes were about to be, or were being committed or had been commit­
ted. The superior is also criminally liable if, owing to the circumstances prevailing at
the time, he should have known and consciously disregarded information indicating
that his subordinates were going to commit (or were about to commit, or were com­
mitting, or had committed), international crimes. The case law has clarified that the
superior need not know the exact identity of the subordinates, it being sufficient that
he should know the category’ of the subordinates engaging in crim inal conduct (this
can be inferred from the fact that this is what courts have required prosecutors to
prove).30
4. Failure to act. The superior failed to take the action necessary to prevent or
repress the crimes, thereby breaching his duty to prevent or suppress crimes by his
subordinates.

11.4.4 S P E C IF IC A T IO N O F T H E S U B JE C T IV E E L E M E N T IN T H E
V A R IO U S C L A SSE S O F O M IS S IO N

The objective element of the crime is apparent from what has just been set out.
It is clear from the above that command responsibility, or responsibility by omis­
sion of superior authorities, is not a form of strict or objective liability; that is, liability
for offences for which one may be convicted without any need to prove any form or
modality of mens rea.31 Even for this category of crimes a mental element is required.
First of all, one ought to distinguish between the mens rea required for the crimes
perpetrated by the subordinates (normally intent, as in the case of killing of civilians,
rape, use of unlawful weapons, torture, etc.) and that required for the superior. This

conceded that the traditional indicia of control remain crucial, including the superior’s power to issue orders
and take disciplinary action (§789).
30 Krnojelac, decision on the defence motion on the form of indictment, §46; Hadzihasanovic and
Kubura, TJ, §90.
31 Recently ICTY Trial Chambers rightly took this view in Delalic and others, §239, and in Kordic and
Cerkez, §369.
 250 INTERNATIONAL CRIMINAL LAW

follows from the fact that in the case of superior responsibility the superior is crim in­
ally liable for his own culpability, which follows from his own breach of obligation;32
he is not responsible for the crimes committed by his subordinates, which may require
a different actus reus and mens rea, although there may be a causal link between those
crimes and the responsibility of the superior.
That law should admit for the superior a less culpable mental element as sufficient
for his liability to arise (for instance, gross negligence instead of the intent required
for the subordinates), is justified by his hierarchical position, the obligation attendant
upon this position to control the subordinates and ensure that they comply with the
law of international armed conflict, and the consequent need to make him account­
able for the conduct of his subordinates.
It would seem that intent is not always required for the superior to be held crim in­
ally liable.33 Rather, one should distinguish various situations:

1. The superior knows that crimes are about to be committed or are being com­
mitted by his subordinates and nonetheless takes no action. Here international rules
require for culpability (i) knowledge, that is awareness that the crimes are being or
are about to be committed;34 and (ii) intent, that is the will not to act or, in other
words, the conscious decision to refrain from preventing or stopping the crimes of the
subordinates (this intent is clearly different from that required for such crimes of the
subordinates as murder, torture, rape, etc., as well as the further subjective ingredient
of crimes against humanity, if any, namely awareness of the existence of a widespread
or systematic practice).35

32 Halilovic, TJ, §54; Hadzihasanovic and Kubura, TJ, §75.

33 In Baba Masao, the Judge Advocate summed up the law for the Australian Military Court trying the
case: ‘In order to succeed [in proving charges of command responsibility] the prosecution must prove [...]
that war crimes were committed as a result of the accused’s [Commanding General of the Japanese Army
in Borneo] failure to discharge his duties as a commander, either by deliberately failing in his duties or by
culpably or wilfully disregarding them, not caring whether this resulted in the commission of a war crime
or not’ (at 207).
34 In Maltauro and others the Court of Assize of Milan held in 1952 that the head of police, being cog­
nizant of the massacre that was about to be carried out by partisans, failed to prevent it. He was therefore
held responsible as a co-perpetrator of the massacre (at 176-7). The massacre took place in a prison where
numerous fascists, previously arrested by partisans on 28 April 1945 (the day when Schio, the small town in
northern Italy, had been liberated), were being held.
See also Sumida Haruzo and others (at 260-1).
35 See, for instance, Cappellini and others (at 86-7), Leoni (Milan Court of Cassation, decision of 31 July
1945, at 128), Bonini (Court of Cassation, decision of 3 March 1948, at 1137-8), Tabellini (Rome Military
Tribunal, decision of 6 August 1945, at 394-8). This last case is particularly interesting: the defendant was a
colonel of the Carabinieri, accused of having allowed, in October 1943, at the request of the German occu­
pying forces, the disarming and transfer of the Carabinieri stationed in Rome to Northern Italy; they had
been subsequently deported to Germany and detained in concentration camps. The Court found that the
defendant was not guilty of failure to prevent the commission of a crime. He was not aware of the real reasons
for the transfer and believed that it was done in the exercise of the Occupant’s power to transfer civil servants
and police forces; according to the Court ‘he lacked the requisite intent, because he carried out the execution
of the order believing that such order was not inconsistent with his duties and those of the police forces to
which he belonged, pursuant to international law’ (at 398).
 CRIMINAL LIABILITY FOR OMISSIONS 251

2. The superior has information which should enable him to conclude in the cir­
cumstances at the time that crimes are being or will be committed, and fails to act, in
breach of his supervisory duties.36 Or he does not pay attention to reports concern­
ing crimes about to be committed or being perpetrated by subordinates, and con­
sequently fails to prevent or stop those crimes. Here either recklessness or gross or
culpable negligence (culpa gravis) may be held sufficient. The former mental element
consists of awareness that failure to prevent the action of subordinates risks bringing
about certain harmful consequences (commission of the crimes), and nevertheless
ignoring this risk.37 The latter state of mind, as pointed out above (3.8), may be found
when: (i) the commander is required to abide by certain standards of conduct or to
take certain specific precautions (for example, to request reports on the conduct of his
underlings, or to exact that reports submitted to him be more accurate and specific); in
addition (ii) he contemplates the risk of harm and nevertheless takes it, for he believes
that the risk will not materialize.38 It should be clear that a conviction for command
responsibility can only be predicated on gross negligence; that is, if the military or
civilian commander’s conduct glaringly falls short of the standard set by the reason­
ably prudent and competent commander test.
3. The superior should have known that crimes were being or had been committed.
Here again gross or culpable negligence (culpa gravis) is sufficient.39

36 According to Delalic and others, this is the case when the commander or the superior authority ‘had in
his possession information of a nature, which at least, would put him on notice of the risk of such offences
[by his subordinates] by indicating the need for additional investigation in order to ascertain whether such
crimes were committed or were about to be committed by his subordinates’ (§383).
37 In Notomi Sueo and others a Temporary Court Martial in the Netherlands East Indies, in dealing with
the responsibility of the commander of a prisoner of war camp in Celebes, held in 1947 that: ‘Even though
a particular act had been neither ordered nor condoned by a superior, who might even [have] been unaware
of it, he must nevertheless be held responsible for the outrages of those under his command, on the ground
that as a Commander he was bound to prevent their occurrence, the more so as he could reasonably foresee
that they would be committed’ (at 209).
38 In Sumida Haruzo and others the Prosecutor stated that, ‘with respect to the torture inflicted by the
members ofhis unit [on the prisoners], this may be attributed to his [of Sumida Haruzo] neglect in exercis­
ing sufficient supervision, and he may, as a result, be condemned on a charge arising out of responsibility for
supervision, which is entirely different from being condemned on criminal responsibility’ (at 235).
In Delalic and others the ICTY AC upheld the interpretation given by the TC to the standard ‘had reason
to know’; that is, ‘a superior will be criminally responsible through the principles of superior responsibil­
ity only if information was available to him which would have put him on notice of offences committed by
subordinates. This is consistent with the customary law standard of mens rea as existing at the time of the
offences charged in the Indictment’ (§241). The AC specified, however, that the information available to the
commander ‘may be written or oral, and does not need to have the form of specific reports submitted pursu­
ant to a monitoring system’. Furthermore, this information ‘does not need to provide specific information
about unlawful acts committed or about to be committed. For instance, a military commander who has
received information that some of the soldiers under his command have a violent or unstable character, or
have been drinking prior to being sent on a mission, may be considered as having the required knowledge’
(§238). See also Bagilishema, AJ (ICTR), §28; Krnojelac, AJ, at §59; Blaskic, AJ, §62.
39 In Röchling and others a French court stated that the 'lack of knowledge’ alleged by the defendant was
culpable because he had the authority to stop the odious practices to which forced labourers were subjected
and instead showed utter indifference to the plight of those labourers (at 8). In Soemu Toyoda a US Military
Commission held that the accused ‘should have known, by use of reasonable diligence, the perpetration of
252 INTERNATIONAL CRIMINAL LAW

4. The superior becomes cognizant that crimes have been com m itted and fails
to repress them by punishing the culprits. Here, knowledge and intent or culpable
negligence would seem to be required for crim inal liability.

atrocities by his troops’ (at 5006). The Commission went on to point out that ‘In determining the guilt or
innocence of an accused, charged with dereliction of his duty as a commander, consideration must be given
to many factors. The theory is simple, its application is not. [...] His guilt cannot be determined by whether
he had operational command, administrative command, or both. If he knew, or should have known, by use of
reasonable diligence, of the commission by his troops of atrocities and if he did not do everything within his
power and capacity under the existing circumstances to prevent their occurrence and punish the offenders,
he was derelict in his duties. Only the degree of his guilt would remain’ (5008).
A Canadian Court Martial relied upon the notion of negligence in Sergeant Boland. The defendant had
failed to prevent two subordinates from torturing and beating to death a Somali civilian taken prisoner (at
1075-8). See also Medina (cited above).
In DelaliC and others an ICTY TC held that ‘from a study of these decisions [of post World War II tribu­
nals], the principle can be obtained that the absence of knowledge should not be considered a defence if, in
the words of the Tokyo judgement, the superior was “at fault in having failed to acquire such knowledge’”
(§388). In Blaskic an ICTY TC held that ‘after World War II, a standard was established according to which
a commander may be liable for crimes by his subordinates if “he failed to exercise the means available to
him to learn of the offence and, under the circumstances, he should have known and such failure to know
constitutes criminal dereliction”’ (§322).
 SECTION III

CIRCUMSTANCES
EXCLUDING CRIMINAL
LIABILITY
 12
JUSTIFICATIONS
AND EXCUSES

12.1 THE DISTINCTION BETWEEN

JUSTIFICATIONS AND EXCUSES

It is widely accepted in most national criminal systems, particularly in civil law coun­
tries, that it is necessary to draw a distinction between two categories of defences:
justifications and excuses.
W hen the law provides for a justification, an action that would per se be considered
contrary to law because it causes harm or damage to individuals or society is regarded
instead as lawful and thus does not am ount to a crime.1 Society, and the legal system
it has created, positively wants a person to do the otherwise illegal act, in that (i) the
act, though criminal, is the lesser of two evils (for instance, when one kills in lawful
self-defence, the death of the attacker is regarded as a lesser evil than that of the person
unlawfully attacked); or (ii), in the case of execution of a sentence of imprisonment or
of the death penalty the taking of liberty or life is a measure positively required by law.
Take also the case of lawful belligerent reprisals (for example, the use of prohibited
weapons). The commander ordering the reprisals as well as those carrying them out
do not act contrary to law, although the weapons used are prohibited by international
law. Resort to those weapons is warranted by the need to stop gross breaches of inter­
national law by the adversary, or to respond to those breaches with a view to prevent­
ing their recurrence. In these and other similar cases, society and its legal system make
a positive appraisal of what would otherwise be misconduct. Society wants the person
so to behave, because in weighing up two conflicting values (the obligation not to use
prohibited weapons and the necessity to impose on the enemy belligerent compliance
with law) it gives pride of place to one of them, although this entails the exceptional
infringement of the legal rules designed to satisfy the other need. The person acting

1 For an example of justifiable homicide, one can mention the US Manual for Courts Martial (1951),
§197(b): ‘A homicide committed in the proper performance of a legal duty is justifiable. Thus executing a
person pursuant to a legal sentence of death, killing in suppression of a mutiny or riot, killing to prevent
the escape of a prisoner if no other reasonably apparent means are adequate, killing an enemy in battle, and
killing to prevent the commission of an offence attempted for force or surprise such as burglary, robbery, or
aggravated arson, are cases of justifiable homicide.’
 256 INTERNATIONAL CRIMINAL LAW

under a justification intends to attain the result caused by his action and is aware that
by undertaking the conduct he will bring about that result (for instance, he intends
to cause the death or wounding of enemy combatants through the use of prohibited
weapons). However, this frame of m ind is not considered culpable mens rea; that is,
intent to murder, for that action and the attendant mental element are deemed to be
legally authorized.
By contrast, excuses may be raised in defence when, although the law regards as
unlawful an action that causes harm and is contrary to a criminal norm, the wrongdoer
is nevertheless not punished. Here the positive appraisal of the conduct excused is less
strong than that relating to conduct covered by a justification. In other words, the value
judgment enshrined in law is not so favourable as to consider the conduct as author­
ized. Indeed, although conduct is blameworthy and unlawful, the agent is not punished
because account ought to be taken of special circumstances. Furthermore, in the case of
excuses the required subjective element of the crime is lacking. Think, for instance, of
the following case: a captain acts under a mistake of fact, in that he orders the shooting
of a number of civilians in occupied territory who, he had been told, had committed war
crimes and had been duly court-martialled, whereas in fact they either had not commit­
ted the crimes or had not been duly tried, as required by international humanitarian law.
In this case the agent believes himself to be engaged in conduct (lawful execution of war
criminals) different from that prohibited by the criminal rule (execution of enemy civil­
ians not duly tried and sentenced). The actus reus cannot be called into doubt, whereas
mens rea is lacking (he did not intend to kill enemy civilians). True, he was aware that by
his order he would bring about their death. However, he did not mean to act contrary to
international prescriptions and therefore lacked the requisite culpable mental element. In
short, he intended to bring about the lawful death of those civilians, not their murder.
The distinction between the two categories at issue should be clear, although in both
cases the author performing an act that per se would fall foul of the law is not pun­
ished. The difference, as stated above, is as follows: (a) justifications set greater store
by the values underpinning the act, hence consider that ultimately such act should
be held to comport with law; (b) excuses, instead, deem that on balance the act must
still be held contrary to law for it offends against values protected by law; neverthe­
less, the author of the act may exceptionally be relieved of punishment, for he or she
acted under circumstances that would render such punishm ent unjust. Plainly, the
law-making bodies of each com m unity (national or international) choose between a
justification and an excuse on the basis of an appraisal of the various values at stake.
Although the basic practical consequence of each of the two categories does not dif­
fer, generally speaking in legal theory the characterization of a defence as a justification
or as an excuse may, however, entail some specific practical consequences from the point
of view of substantive law. In particular, three consequences should be pointed out:

1. Where the defendant successfully pleads an excuse, any aider and abettor may
nevertheless be responsible for the excused crime (if, of course, none of them was also
entitled to invoke an excuse). Take the case of a military officer who, knowing that
 JUSTIFICATIONS AND EXCUSES 25 7

a superior had ordered to shoot spies without trial (un unlawful action under IHL),
provides trucks for transportation of the spies to the place of execution where a fellow
officer will command the execution squad without being aware that the order was
unlawful (for he believed that the spies had been duly court-martialled). In this case
the officer in question may be charged with aiding and abetting a war crime, while the
officer commanding the execution squad may plead mistake of fact as an excuse.
In contrast, when a justification is urged (for instance, lawful execution after trial of
enemy civilians who had engaged in war crimes), there can be no aider or abettor, for
the simple reason that the conduct at issue (the execution of civilians qua war crim i­
nals) is not unlawful.
2. Any action in self-defence by the victim of criminal conduct is allowed (provided,
of course, it is in compliance with the requirements of self-defence) if for such criminal
conduct an excuse may be raised in defence. Self-defence is permissible because the
criminal conduct to which the actor acting in self-defence intends to react, although
excused by law, nevertheless remains unlawful per se. For example, a person, realizing
that he is about to be killed by a man involuntarily intoxicated, shoots at him to pre­
vent being murdered. This action is justified as self-defence, even if the act by the other
person would have not be punished, being legally excused (on involuntary intoxica­
tion, see infra, 12.5.2).
Instead, self-defence by the ‘victim’ of criminal conduct is not warranted when such
conduct is covered by a justification, because that conduct must be regarded as lawful
from the outset (whereas self-defence is only admissible to repel unlawful violence
by another subject). For instance, a belligerent may not invoke self-defence and thus
indiscriminately shell an area where an enemy military installation is located from
where enemy combatants were about to take lawful reprisals (an action covered by jus­
tification) against the belligerent at issue. Another example is that of a prisoner of war
who, trying to escape (i.e. accomplishing an act not prohibited by international law,
but unlawful under national law), kills a prison guard who was about to shoot at him
to prevent his escape. The prisoner of war may not urge self-defence as a justification,
for the guard was about to perform a lawful act.
In short, self-defence may be relied upon when a person is faced with an act by
another person that is either criminal or may be excused; in contrast, self-defence
may not be urged against an action that, although seemingly illegal, is made lawful by
another justification.
3. A defendant that successfully urges an excuse may be liable to pay compensation
for any damage resulting from his misconduct. For instance, in the case of murder
committed by a minor or by a person suffering from mental disorder, he will not be
punished, but his parents or relatives may have to pay compensation to the victim’s
relatives.
In contrast, if the behaviour is legally justified, no such obligation arises, for the
action is not considered unlawful.
 258 INTERNATIONAL CRIMINAL LAW

To identify the way ICL regulates justifications and excuses; that is, what defences
it subsumes under each of the two categories and with what legal and practical conse­
quences, one has, of course, to look to international customary and treaty rules for the
appropriate legal characterization of each defence (see, however, 12.2).

12.2 ICL: GENERAL

ICL envisages both justifications of some actions inherently contrary to law and
excuses for persons engaging in criminal conduct, although it is not yet clear whether
it draws a legal distinction between these two categories.
Among defences that may be logically and legally classified as justifications, one can
mention the following: (a) lawful punishment of enemy civilians or combatants guilty
of war crimes or other international crimes such as crimes against hum anity (e.g. the
execution, after conviction and sentencing by a duly constituted C ourt Martial, of
civilians who had engaged in prohibited attacks on the belligerents, amounting to
war crimes or other international crimes); (ii) lawful belligerent reprisals against war
crimes (as stated before, they may include the use of prohibited weapons as a response
to a serious violation of international hum anitarian law by the adversary, for instance,
the killing of prisoners of war or the intentional shelling of civilians); (iii) self-defence
(see infra, 12.3).
It is doubtful that consent may be raised as well in defence as a justification.
International crimes normally involve unlawful attacks on the life, body, or dignity
of hum an beings, and consent would be inadmissible as a justification, because these
values are protected by international norms that have the rank of jus cogens i.e. per­
emptory norms, and are therefore not derogable by either states or individuals.
Far more numerous are the classes of defence that may be defined as excuses from
the point of view of legal logic: mental disease, state of intoxication, mistake of fact,
mistake of law, duress, physical compulsion.2 It is doubtful whether force majeure is
admissible (the existence of an irresistible force or an unforeseen external event beyond
the control of a belligerent which makes it absolutely and materially impossible for the
belligerent to comply with a rule of hum anitarian law: for instance, non-compliance
with some rules on the treatment of prisoners of war on account of an earthquake, or
of a famine not caused by the belligerent); this excuse, if admissible, should, however,
be strictly construed to avoid abuse by combatants. In contrast, it is certain that under
customary international crim inal law neither superior order nor im m unity for acting
as a state official (the so-called ‘act of state doctrine’) may ever am ount to an excuse
(see, however, 13.2.6 on the provision of the ICC Statute on superior orders).

2 According to the British 2004 Manual of the Law of Armed Conflict, criminal responsibility is not
incurred by a person for such acts as he is physically compelled to perform against his will and in spite of his
resistance’ (§16.46). The 1958 British Manual of Military Law added:‘thus, if A by force takes the hand of B
in which is a weapon, and therewith kills C, A is guilty (of murder), but B is excused’ (§628).
 JUSTIFICATIONS AND EXCUSES 259

An im portant point should be made: until now case law has not highlighted any of
the three aforementioned specific practical distinctions between the two categories of
defences. Generally, international prosecutors and courts or, in the case of national
proceedings, prosecutors and courts of common law countries, confine themselves to
respectively requesting the culpability of the accused, or satisfying themselves either
that he is culpable or that he may rightly plead a defence. They seldom go so far as to
claim or order that the accused, if found not guilty on account of an excuse, should
also pay compensation. Similarly, it would seem that to date there have not been cases
where self-defence has been invoked as a response to another person’s criminal con­
duct which, although by itself contrary to international hum anitarian law, was never­
theless not punishable in that it was covered by an excuse. Nor have there been cases
where aiders and abettors have assisted the author of a crime, and a court found that he
was excused, whereas they were not covered by the same or another excuse.
Nevertheless, it would seem that this state of affairs in no way detracts from the
soundness of both the distinctions made above and the different specific practical
consequences following from each category of defence.

12.3 SELF-DEFENCE

12.3.1 C U S T O M A R Y IN T E R N A T IO N A L LA W

A person may plead self-defence whenever he commits an international crime in order

to prevent, or put an end to, a crime by another person against the agent or a third per­
son. Self-defence is lawful provided it fulfils the following requirements: (i) the action
in self-defence is taken in response to an imminent or actual unlawful attack on the life
of the person or of another person;3 (ii) there is no other way o f preventing or stopping
the offence; (iii) the unlawful conduct of the other has not been caused by the person
acting in self-defence; (iv) the conduct in self-defence is proportionate to the offence to
which the person reacts.
Examples of self-defence include the killing by a prison guard of an enemy prisoner
of war who was about to murder the guard, or the wounding of an enemy serviceman
by a civilian woman in the hands of the enemy occupant, for the purpose of preventing
or halting torture or rape.4
Plainly this defence must not be confused with self-defence under public
international law. The latter relates to conduct by states or state-like entities, whereas

3 For a treaty deviation from customary international law, see Article 31(1) (c) of the ICC Statute.
4 In Kordic and Cerkez an ICTY TC held that self-defence as a ground for excluding criminal responsi­
bility is one of the defences that ‘form part of the general principles of criminal law which the International
Tribunal must take into account in deciding the cases before it’ (§449). It went on to note that the ‘principle of
self-defence’ enshrined in Article 31(l)(c) of the Statute of the ICC ‘reflects provisions found in most national
criminal codes and may be regarded as constituting a rule of customary international law’ (§451).
 2Ö 0 INTERNATIONAL CRIMINAL LAW

the former concerns actions by individuals against other individuals. This confu­
sion is often made. For instance, in Kordic and Cerkez defence counsel argued that
the Bosnian Croats engaging in armed action under the authority of the two accused
were acting in self-defence, to react to a policy of aggression by Muslim forces (§448).
The ICTY TC rightly rejected the argument, noting that ‘military operations in self-
defence do not provide a justification for serious violations of international hum ani­
tarian law’ (§452).
In a number of cases courts discussed this justification, even if often they did not
uphold it on the facts.5
The plea was successful in Erich Weiss and Wilhelm Mundo, before a US Military
C ourt in Ludwigsburg. An American airm an who in May 1944 had safely parachuted
from his military aircraft over Germany was captured and turned over to two police­
men; when, during an air raid, a crowd gathered around them demanding that the
prisoner be killed, he suddenly moved his right hand in his pocket; the two policemen
fired at him and he was instantly killed. The two defendants pleaded that they had felt
threatened by the prisoner’s movement of his hand in his pocket and had fired in self-
defence. The US Court upheld the plea (149-50).

5 That self-defence may validly be put forward was held in obiter dicta by a US Tribunal sitting at
Nuremberg in Alfried Felix Alwyn Krupp and others (at 1438). Also the Judge Advocate in the trial of Willi
Tessmann and others by a British Military Court sitting at Hamburg accepted that self-defence could be
pleaded subject to certain strict conditions (177). In the former case the US Tribunal in obiter dicta, after not­
ing that ‘self-defence excuses the repulse of a wrong’, insisted on the mental attitude of the person invoking
the defence; it emphasized that ‘the mere fact that [...] a danger was present is not sufficient. There must be
an actual bona fide belief in danger by the particular individual’ (1438).
The plea also failed in Yamamoto Chusaburo, brought before a British Military Court sitting in Kuala
Lumpur. A Japanese sergeant, charged with a war crime for killing a Malayan civilian who was stealing
rice from a military store, claimed among other things that he had acted in self-defence: after arresting the
civilian, he had been surrounded by a hostile crowd; fearing a grave danger to life and property, the more
so because he was in pitch darkness, he had lost control of himself and in a rage killed the civilian with a
bayonet. The Prosecutor rebutted that there was evidence that the act had not been committed in defence of
property or person while the civilian was in the process of looting; it had been committed after the civilian
had been taken from his house into custody (76-9).
Similarly, in Frank C. Schultz, heard in 1969 by a US Court of Military Appeals, the plea, while implicitly
admitted in theory, failed on the facts. Schultz, a US marine, was a member of a four-man patrol commonly
referred to as a hunter-killer team designed to ambush and kill Viet Cong. He killed an innocent Vietnamese
farmer in a Vietnamese village. Before the Appellate Court he pleaded that he believed that the individual
killed was a member of the Viet Cong, or that he was in communication with the enemy and was signalling
the enemy and attempting to lead the appellant and his patrol into an ambush; he claimed that he did ‘what
he was instructed to do and what he felt he had to do to survive’. The Court rejected the defence, noting that
"The testimony of the accused shows his actions to be intentional. Thus removed is the possibility that death
of the victim resulted from accident or misadventure [...] Moreover, self-defence is unavailable for it is a plea
of necessity not available, normally speaking, to one who is an aggressor’ (136-8). It is worth noting that in
this case the Court relied upon another case, not dealing with a war crime, namely Carl D. O’Neal, where
the same Court had ruled that ‘a person cannot provoke an incident, and then excuse himself from respon­
sibility for injury inflicted by him upon another in the course thereof, on the ground of self-defence [...] A
plea of self-defence is a plea of necessity [...] It is generally not available to one who engages with another in
mutual combat’ (193).
 JUSTIFICATIONS AND EXCUSES 26l

12.3.2 T H E IC C S T A T U T E

Article 31(l)(c) of the ICC Statute envisages self-defence as a ground excluding respon­
sibility in the following terms:
[A person shall not be crim inally responsible if, at the tim e of th at person’s co n d u ct...]
the person acts reasonably to defend him self or herself or another person or, in the case of
war crimes, property which is essential to the survival of the person or another person or
property w hich is essential for accom plishing a m ilitary mission, against an im m inent and
unlaw ful use of force in a m anner proportionate to the degree of danger to the person or
the other person or property protected. The fact that the person was involved in a defensive
operation conducted by forces shall not in itself constitute a ground for excluding crim inal
responsibility under this subparagraph.

To a large extent Article 31 codifies a rule of customary international law, as an ICTY

TC rightly emphasized in Kordic and Cerkez (§451). However, the provision also con­
tains a clause that is extraneous to customary international law. This is the clause
whereby the defence applies to cases where one intends to defend property ‘which is
essential to the survival of the person or another person or property which is essen­
tial for accomplishing a military mission.’ Under this provision an officer may plead
self-defence if he has ordered his unit to use prohibited weapons against enemy com­
batants that are about unlawfully to attack and blow up a weapons depot or a military
compound containing tanks and other m ilitary equipment needed for an im m inent
expedition. Similarly, with a view to saving from an impending unlawful enemy attack
a m ilitary compound containing food, water, and other supplies indispensable to the
survival of servicemen, an officer could lawfully order to launch an indiscriminate
attack against the enemy combatants and the civilian population.
The provision at issue6lends itself to a num ber of major objections. First, it unjusti­
fiably departs from customary law, which does not cover these and similar eventual­
ities. That body of law only affords self-defence as a justification for saving the life and
limb of the person acting in self-defence or other persons whose life and limb was in
im minent danger.
Secondly, there is no rule in international hum anitarian law prohibiting attacks
against property which is essential for accomplishing a military mission or indispens­
able for the survival of combatants. International hum anitarian law only contains rules
protecting objects indispensable to the survival of the civilian population (customary

6 According to K. Ambos ('Other Grounds for Excluding Criminal Responsibility’, in Cassese, Gaeta,
Jones, ICC Commentary, I, at 1033), ‘The property defence was promoted by the United States and Israel, the
former invoking constitutional provisions and insisting that “the defence of one’s home can be perfectly
legitimate”. The US delegation even proposed an equal treatment of defence of life and physical integrity,
on the one hand, and property, on the other. This position did not find much sympathy, and the final text of
subparagraph (c) shows that protection of property is limited to war crimes situations in which the prop­
erty is “essential for the survival of the person or another person” or “essential for accomplishing a military
mission”. Even in this limited form, the protection of property was difficult to accept for many delegations;
it became the “real cliff-hanger” in the negotiations of the Working Group’. See also A. Eser, in Triffterer,
ICC Commentary, at 548.
 2Ö2 INTERNATIONAL CRIMINAL LAW

rules on the matter evolved out of Article 54 of the First Geneva Protocol of 1977). The
provision under discussion admittedly requires for self-defence to be resorted to, that
the im minent enemy attack be unlawful (as a US Tribunal sitting at Nuremberg pithily
put it in Alfried Alwyin Krupp and others, ‘self-defence excuses the repulse of a wrong’
(at 1438; emphasis added)). Hence the clause of Article 31(l)(c) we are discussing may
chiefly refer to an attack by the enemy that is unlawful either because it involves the
use of prohibited weapons (for instance, chemical or bacteriological means of warfare)
or because it causes disproportionate casualties among civilians. In both hypotheses
it seems, however, unrealistic to envisage the possibility for a belligerent to anticipate
that the enemy will use unlawful weapons; it is a fortiori unrealistic to demand that
the belligerent should anticipate that the enemy will bring about disproportionate col­
lateral damage. The contention is therefore warranted that this broadening of the jus­
tification of self-defence is either unworkable or likely to lead to serious abuses.
Thirdly, the clause at issue serves to justify through a legal defence war crimes com­
m itted solely to pursue m ilitary objectives. It is contrary to the very spirit of ICL, for
it eventually ‘covers’ and legitimizes crimes perpetrated for the sole sake of protecting
m ilitary exigencies, whereas in ICL justifications are provided for with a view to taking
into account fundamental values such as hum an life and dignity—values that under
certain circumstances ICL regards as such as to override m ilitary requirements.
Fourthly, the clause is unsustainable, for in fact it tries surreptitiously to introduce,
through a criminal rule, a new substantive legal standard into IHL; this standard aims
at protecting property that serves the m ilitary or military operations, a property that
traditionally is instead a legitimate m ilitary objective. However, since the provisions of
the ICC Statute, as those of the statute of any other tribunal, only apply to the judicial
institution at issue and to its jurisdiction (in this case to the ICC), the introduction
of the provision under discussion into IHL or even ICL could only occur through a
possible gradual turning of the provision into customary international law. I hope this
process will not materialize, on the grounds set out above.

12.4 EXCUSES: TWO MAIN CATEGORIES

Let us now move on to excuses. W ithin this category of circumstances precluding
crim inal liability one ought to distinguish between two classes.
The first includes those instances where, on account of his (transient or perm an­
ent) psychological conditions (insanity or intoxication), the person is not possessed o f
individual autonomy; that is, is not endowed with the capacity and free will to decide
upon his conduct. It is because of this incapacity to freely choose his actions or omis­
sions that the person lacks mens rea and cannot therefore be regarded as culpable if he
engages in a criminal offence. For such cases it is held in some national legal systems
that the person is ‘criminally not imputable’.
 JUSTIFICATIONS AND EXCUSES 263

A second class of excuses embraces instances where the person may not be held
culpable because, although he is fully possessed of his individual autonomy and may
in theory freely choose a course of action, he nonetheless lacks a criminal frame of
mind on account o f outside circumstances. This may be because (i) he is under a non-
culpable misapprehension about the facts or about the applicable rules; it follows that
he intends to bring about conduct that is different from that which actually occurs
and is prohibited by the criminal rule (for instance, he was made to believe that he was
engaging in the lawful execution of enemy war criminals duly tried and sentenced,
whereas in fact he was executing enemy civilians without any prior trial). Or else,
(ii) although he is aware of the consequences of his conduct (for instance, killing a
prisoner of war, torturing a civilian), he does not will those consequences but is obliged
by another person to carry out the prohibited act through an unavoidable and serious
threat to his life or limb.

12.5 EXCUSES BASED ON LACK OF

INDIVIDUAL AUTONOMY

12.5.1 IN S A N IT Y O R M E N T A L D IS O R D E R

Insanity or serious mental disorder, or mental incapacitation, or mental disease, may

be urged as an excuse whenever this state of m ind entails that the person is deprived of
the mental capability necessary for deciding whether an act is right or wrong. The plea
may be urged when at the time of commission of the crime the accused was unaware
of what he was doing and hence of forming a rational judgment about his conduct.
As a consequence, the accused, while committing an unlawful act, lacks the requisite
mens rea and may not be held responsible for his behaviour (see Article 31(1) (a) of the
ICC Statute).
A case that supports the availability of this defence in international hum anitar­
ian law is Stenger and Crusius, decided by the Leipzig Supreme C ourt in 1921. The
German Captain Crusius, commander of a company, had been accused of passing on^
to his subordinates, in the battle of 26 August 1914 against the French troops in the
forest near Sainte Barbe (Alsace), an order from Major General Stenger not to take
prisoners of war, or in other words to kill all captured enemy soldiers. The Supreme
C ourt of Leipzig found that he had m isunderstood that order, and hence wrongly
ordered the killing of prisoners of war. The C ourt also found, on the basis of the tes­
timony of various witnesses and expert witnesses, that the accused had undergone a
complete mental and psychological collapse, that is a state of utter mental confusion
[...] induced by a psychopathic disposition and by the particular disturbance’ of the
battles of the previous days [...] which would unequivocally preclude responsibility
 264 INTERNATIONAL CRIMINAL LAW

pursuant to crim inal law’ (at 2571-2).7 The Court therefore acquitted Crusius on that
count (2568-72).
In contrast, in the same case the Court rejected the plea with respect to a previous
episode. Captain Crusius had been accused of transm itting unlawful orders of Major
General Stenger to his subordinates, on 21 August 1914. The Court found that in fact
he had misunderstood the superior orders; when he passed them on to his subordi­
nates, he was in extreme agitation and psychological suffering’; however, his mental
state was not such as to preclude his ‘free determination of will’ (at 2567). Crusius was
therefore found guilty on that count.
Similarly, the plea was rejected in other cases, for instance in Kotälla, by a Special
Criminal Court of Amsterdam8 as well as in Frank C. Schultz, a case heard by a US
Court of Military Appeal.9
National laws and courts have upheld the notion that, in addition to insanity, there
may exist other forms of abnormality of m ind that may have a bearing on, and dim in­
ish, responsibility. In some states (in particular common law countries, notably Great
Britain) the plea, if successful, entails reducing the gravity of the offence with which a
defendant pleading the defence might be charged (for instance, reducing m urder to
manslaughter, whenever there is a mandatory sentence for murder, namely death or
life imprisonment). In other states (chiefly civil law countries), if the plea is successful,
the accused qualifies fo r mitigation o f sentence.
As an instance of national cases where courts, when adjudicating war crimes,
applied national law, one can mention Calley, a case brought in 1971 before a US Court

7 The Court admitted that this mental state only emerged gradually in the afternoon: ‘at around the
time when the accused, distraught, with a bright red face and swollen eyes, came running out of the forest,
screaming and rushing towards Dr. Döhner [another German serviceman, who testified in court], grabbing
his arm, desperately uttering calls, and leaving the overall impression of a maniac [...] this state did not
occur suddenly and abruptly but rather gradually worsened after having developed from an already existing
nervous condition induced by a psychopathic disposition and by the particular disturbance’ of the battles of
the previous days. The Court found that when the supposed superior order was passed on to his subordinates
‘the accused was suffering from a mental disorder rendering him incapable of forming a rational intention’
(at 2572). After suffering from ‘so-called diminished responsibility’ (verminderte Zurechnungsfähigkeit,
ibid.), he then found himself in a state of mind ‘precluding responsibility’.
8 The Court rejected the plea of mental disorder invoked by the accused (who had been charged with war
crimes and crimes against humanity). It held that it had established, ‘on the basis of its own observations
at the hearing and further information presented [...] at the hearing, that the accused [did] not suffer from
such a limited development of his mental faculties or mental disorder which could result in the offences
committed by him not being attributed to him or being attributed to him to a lesser extent’ (at 6). See also the
decision delivered in the same case by the Dutch Special Court of Cassation on 5 December 1949 (at 13).
9 The defence had raised the issue of insanity. The appellant had been accused and then convicted of
premeditated murder, for having killed an innocent Vietnamese civilian. The Court rejected the plea of
insanity. After noting that the testimony of two psychiatrists, one for the government, the other for the
defence, showed that the accused had suffered from probable mental impairment, the Court referred to
two previous cases unrelated to war crimes, Michael F. Kunak (354-66) and Vadis Storey (426-30), and
approvingly cited their holding whereby ‘More than partial mental impairment must be shown in order to
raise the issue. There must be evidence from which a court-martial can conclude that an accused’s mental
condition was of such consequences and degree as to deprive him of the ability to entertain the particular
state of mind required for the commission of the offence charged’ (138). See also Sergeant W. (decision of the
Military Court, at 2).
 JUSTIFICATIONS AND EXCUSES 265

Martial. The accused had been charged with premeditated m urder in violation of
Article 118 of the Uniform Code of M ilitary Justice (the charge was of killing a num ­
ber of Vietnamese civilians in the village of My Lai (4) in South Vietnam). The defence
raised, among other things, the issue of mental capacity. In his instructions to the
Court, the Judge Advocate accepted that a serviceman could be found to be ‘suffering
from a mental impairment or condition of such consequence and degree that it deprived
him of the ability to entertain the premeditated design to kill required in the offence of
premeditated murder’.10The US Army Court of Military Review took the same stand in
its judgment of 16 February 1973 on the same case, but rejected the plea.11
A Special Court in Amsterdam took a different approach, typical of civil law coun­
tries, in Gerbsch. Between 1944 and 1945 the accused was a guard at a penal camp in
Zoeschen, Germany, and there he ill-treated many detainees, in particular Dutchmen
and other persons transferred from the Netherlands. The Court found him guilty of a
crime against humanity, but took into account as a mitigating circumstance the fact
that his ‘mental faculties were defective and undeveloped’ when the crime was com­
mitted, as well as at the time of trial (at 492).12

10 The Judge Advocate stated the following: ‘The law recognizes that an accused may be sane and yet,
because of some underlying mental impairment or condition, be mentally incapable of entertaining a pre­
meditated design. You should therefore consider, in connection with all other relevant facts and circum­
stances, all evidence tending to show that Lt. Calley may have been suffering from a mental impairment or
condition of such consequence and degree that it deprived him of the ability to entertain the premeditated
design to kill required in the offence of premeditated murder. The burden of proof is upon the government
to establish the guilt of Lt. Calley beyond a reasonable doubt. Unless, in light of all the evidence, you are sat­
isfied beyond a reasonable doubt that Lt. Calley, on 16 March 1968, in the village of My Lai (4), at the time of
each of the alleged offences, was mentally capable of entertaining, and did in fact entertain, the premeditated
design to kill required by law, you must find him not guilty of each premeditated murder offence for which
you do not find premeditated design. You may, however, find Lt. Calley guilty of any of the lesser offences in
issue [unpremeditated murder or voluntary manslaughter], provided you are convinced beyond a reasonable
doubt as to the elements of the lesser offence to which you reach a guilty finding, bearing in mind all these
instructions’ (at 1716).
11 The two defence psychiatrists had asserted that the accused was acting automatically and did not
have capacity to premeditate because he was effectively without ability to reflect upon alternative courses of
action and choose from them; he did not have the mental capacity to contrive’ the deaths of the villagers.
The Court noted, however, that both psychiatrists agreed that Calley had capacity to perceive and predict,
the two functions essential to the pertinent mens rea. Appellant knew he was armed and what his weapon
would do. He had the same knowledge about his subordinates and their arms. He knew that if one aimed
his weapon at a villager and fired, the villager would die. Knowing this, he ordered his subordinates to
“waste” the villagers at the trail and ditch, to use his own terminology; and fired upon the villagers himself.
These bare facts evidence intent to kill, consciously formed and carried out’ (1178). The Court concluded (at
1178-9) that Calley had acted with premeditation.
Some international cases can also be mentioned. In Delalic and others an ICTY TC, based on national
legislation, admitted that there might be an impairment of mind affecting criminal liability (at §§1166 and
1186). The AC convincingly clarified the matter in the same case (Delalic and others, AJ): ‘The Appeals
Chamber recognises that the rationale for the partial defence provided for the offence of murder by the
English Homicide Act 1957 is inapplicable to proceedings before the Tribunal. There are no mandatory
sentences. Nor is there any appropriate lesser offence available under the Tribunal’s Statute for which the
sentence would be lower and which could be substituted for any of the offences it has to try. The Appeals
Chamber accepts that the relevant general principle of law upon which, in effect, both the common law and
the civil law systems have acted is that the defendant’s diminished mental responsibility is relevant to the
 266 INTERNATIONAL CRIMINAL LAW

It would seem that, in any case, uncontrollable fits of temper may not be considered
as falling under this category of possible excuses. At the most, and under strict condi­
tions, it might prove appropriate to take them into account, if need be, as extenuating
circumstances.13

12.5.2 IN T O X IC A T IO N

Being in an intoxicated state as a result of taking alcohol, drugs, or other intoxicants

may amount to an excuse only under very strict conditions: (1) the intoxication is so
serious as to negate mens rea (that is, it alters the agent’s mental state to such a point
that he is not in a position to be aware of his actions and to appraise the unlawfulness
of his conduct); (2) in the case of voluntary intoxication, the person has not become
voluntarily intoxicated knowing the risk that, as a result of his state, he was likely to
engage in criminal action (see Article 31(l)(b) of the ICC Statute).
As an example of an acceptable excuse, the case can be mentioned of a soldier to
whom a medical doctor or a nurse administers powerful sedatives or pain-killers
which seriously alter his mental state; as a result, he kills or wounds or rapes a prisoner
of war or an enemy civilian.
There are a few cases on this matter. In Yamamoto Chusaburo a British Military
Court sitting at Kuala Lampur rejected a plea of drunkenness.14In Kvocka and others,

sentence to be imposed and is not a defence leading to an acquittal in the true sense. This is the appropriate
general legal principle representing the international law to be applied in the Tribunal. Rule 67(A)(ii)(b) [of
the ICTY Rules of Procedure and Evidence] must therefore be interpreted as referring to diminished mental
responsibility where it is to be raised by the defendant as a matter in mitigation of sentence. As a defendant
bears the onus of establishing matters in mitigation of sentence, where he relies upon diminished mental
responsibility in mitigation, he must establish that condition on the balance of probabilities—that more
probably than not such a condition existed at the relevant time’ (§590).
13 In Erhard Milch one of the judges serving on a US Military Tribunal sitting at Nuremberg, Judge
Phillips, in his concurring opinion implicitly conceded that uncontrollable temper might be taken into
account, although he did not specify for what legal purposes. Nonetheless, in the case at issue he rejected a
defence claim that the accused (Field-Marshal in the German Luftwaffe, Aircraft Master General, Member
of the Central Planning Board and State Secretary in the Air Ministry), who had been charged with war
crimes and crimes against humanity involving deportations of civilian populations, forced labour and
illegal experiments, had made violent statements due to uncontrollable temper, overwork, and head injuries.
The judge noted that: ‘If but only a few of such remarks could be attributed to the defendant, his protestations
might be given some credence; but when statements such as appear in the documents have been persistently
made over a long period of time, at many places and under such varying conditions, the only logical con­
clusion that can be reached is that they reflect the true and considered attitude of the defendant toward the
Nazi foreign labour policy and its victims and are not mere aberrations brought on by fits of uncontrollable
anger’ (47).
14 The accused, a sergeant of the Japanese Army, had been charged with a war crime for killing a civilian
who had stolen rice from the army store. He pleaded, among other things, that he had acted under the influ­
ence of alcohol. According to the summary of the UN War Crime Commission, the Prosecutor said that
‘drunkenness in itself was not an excuse for crime, but where intention was of the essence of the offence,
drunkenness might justify a court in awarding a lesser punishment than the offence would otherwise have
deserved or it might reduce the offence to one of a less serious character. In such a case the man must be in such
a state of drunkenness as to make him incapable of formulating any intention to commit the offence, and such
a state would clearly affect the degree of killing of which the Court would find the accused guilty’ (at 78).
 JUSTIFICATIONS AND EXCUSES 267

one of the accused, Zigic, had beaten up and brutalized inmates in some detention
camps, abusing and humiliating his victims; the defence argued that he had been
intoxicated during many of the ‘incidents’ (§§616 and 680); at the sentencing stage the
Trial Chamber rejected the claim that intoxication was a mitigating factor, and found
instead that it was an aggravating circumstance. However, as the Prosecutor had not
previously raised the matter, the Chamber declined to treat intoxication as a fact ger­
mane to sentencing in the case at issue (§748).

12.5.3 M IN O R S

In many national legal systems it is normally considered that persons under a certain
age do not possess full individual autonomy and therefore may not freely decide how to
act: a child is regarded as unable to entertain criminal intent. Consequently, children
are normally considered exempt from criminal responsibility if they engage in criminal
conduct. At present in some countries the threshold has been lowered. For instance, in
Britain, children aged ten and above may be held accountable in some respects and
liable to conviction. Trials for children are normally held before special courts.
In international criminal law no customary rule has emerged on this matter.
However, a provision on the issue can be found in the ICC Statute: under Article 26,
‘The Court shall have no jurisdiction over any person who was under the age of 18 at
the time of the alleged commission of a crime.’ Plainly, this provision is couched in
terms referring to the C ourt’s jurisdiction, and not as a substantive rule of criminal
law whereby minors may not be held criminally responsible. It follows that, under that
provision, it would be lawful for a contracting state to bring to trial before its national
courts persons under eighteen (say, of fifteen) for allegedly committing war crimes, if
this is allowed under the relevant national legislation.
The aforementioned provision may appear to be somewhat at variance with another
provision of the same Statute, Article 8(2)(e)(vii), whereby ‘conscripting or enlisting
children under the age of fifteen years into armed forces or groups or using them to
participate actively in hostilities’ may am ount to a war crime. It follows that a state, a
national liberation movement, or insurgents may lawfully enlist children of sixteen or
seventeen—but if these children engage in criminal conduct, they are not amenable
to judicial process before the ICC (although of course they could be brought to trial
before national courts, assuming such courts have jurisdiction over them).
 13
OTHER EXCUSES: SUPERIOR
ORDER, NECESSITY, DURESS,
AND MISTAKE

13.1 EXCUSES WHERE LACK OF MENS REA

DERIVES FROM EXTERNAL CIRCUMSTANCES
In the preceding chapter I have discussed, among other things, those classes of excuses
where the lack of mens rea is consequent upon the agent lacking individual auton­
omy and consequently not being ‘imputable’. I will now consider those categories of
excuses where the absence of the requisite subjective element derives from external
circumstances.
On this score the first question to be discussed is whether one of the most invoked
excuses in national and international criminal trials, that is, superior orders, may be
legally considered, under international criminal law, as a circumstance precluding
criminal liability.

13.2 THE QUESTION OF SUPERIOR ORDERS: MAY

THEY BE PLEADED AS A DEFENCE?

13.2.1 N O T IO N

The basic assumption of the whole question of superior orders is that a subordinate
may be faced with a dilemma: (i) to respect military hierarchy and consequently exe­
cute whatever order he is given, whether or not lawful; or (ii) to refuse to carry out a
patently unlawful order in order to comply with the morally exacting demand that
one ought not to breach the law. As the great English constitutionalist Dicey put it, a
soldier may thus be caught in a grievously conflictual situation: he ‘may be liable to be
 OTHER EXCUSES: SUPERIOR ORDER, NECESSITY, DURESS, AND MISTAKE 269

shot by a court-martial if he disobeys an order, and to be hanged by a judge and jury

if he obeys it’.1
The legal condition of superior orders was not crystal clear before the Second World
War. In most national legal systems the principle of obedience tended to prevail,
with the consequent application of the respondeat superior principle (only the super­
ior should be held accountable). Exceptions were made only with regard to enemies.
In this respect the Wirz case, brought in 1865 before a US Military Commission is
instructive.2 In some states, however, the notion prevailed that a subordinate also bore
responsibility if he executed an unlawful order. For instance, in Germany a Criminal
Military Penal Code expressly contemplated such eventuality. In 1916 a German
Court in the Brussels case disregarded the fact that war crimes committed by a British
commander of a merchant vessel had been committed on superior orders.3 German
courts applied the same principle to German servicemen as well. After the First
World War in a few war crimes cases the Supreme C ourt in Leipzig held that superior
orders could not be urged as a defence when they were unlawful.4 It is notable that, in

1 A. V. Dicey, Introduction to the Study of Law of the Constitution, 10th edn (London: Macmillan, 1959),
at 303.
2 During the American Civil War (1861-5) Captain Henry Wirz, a Swiss doctor who had emigrated
to Louisiana and, ‘carried away by the maelstrom of excitement’ (as he was to write later), had joined the
Confederate army of the Southern states, was given the command of a prison camp in Andersonville
(Georgia). Here he contravened the laws of war by ill-treating the prisoners of war and keeping them in
appalling conditions. In 1865, when the war was over, he was tried by a Military Commission in Washington.
He defended himself by saying that he had acted on superior orders, being merely ‘the medium, or better, the
tool in the hands of his superiors’. The Judge Advocate objected that when an order is illegal both the super­
ior officer and his subordinate are guilty. As he put it: ‘A superior officer cannot order a subordinate to do an
illegal act, and if the subordinate obey such an order and disastrous consequences result, both the superior
and the subordinate must answer for it. General Winder [the officer above Wirz] could no more command
the prisoner [Wirz] to violate the laws of war than could the prisoner do so without orders. The conclusion
is plain, that where such orders exist both are guilty [...].’ (at 796). The Military Commission accepted the
argument and sentenced Wirz to death by hanging (ibid., at 797-8).
3 In 1916 Fryatt, the British commander of the merchant vessel Brussels flying the Union Jack, crossed
the path of a German submarine, which ordered him to heave to and identify himself. Instead of obeying,
Fryatt, an, orders from the Admiralty to all merchant vessels are in similar circumstances, bore down at full
speed on the enemy submarine and tried to ram it. The submarine moved off to avoid collision and Fryatt got
away. On another voyage, however, the Brussels was captured and Fryatt tried and sentenced for war crimes
(he was regarded as afranc tireur, that is an illegal combatant), even though he had merely been carrying out
superior orders (at 407-13 and n. 1 at 411).
4 In Dover Castle the Court found the accused not guilty for sinking, upon superior orders, a British hos­
pital ship that was suspected of transporting troops. The Court applied Article 47 of the German Military
Penal Code, whereby a subordinate is not accountable for carrying out superior orders, unless the order
is illegal or the subordinate goes beyond what has been ordered (at 2557/707-8). Similarly, in the Robert
Neumann case the same Court found that the defendant was entitled to plead superior orders for attacking
English prisoners of war, ‘who were often refractory’. According to the Court, ‘As matters stood, there could
be no doubt of the legality of the order’ (at 2554/699). In the Llandovery Castle case the two defendants had
submitted that they had fired on the lifeboats of the British steamer Llandovery Castle (a naval hospital that
according to the Germans had in fact been used for the transport of troops and had thus been sunk by the
German submarine) following the orders of their superior, Captain Patzig. The Court rejected the claim,
noting the following: ‘Patzig’s order does not free the accused from guilt. It is true that according to Section
47 of the Military Penal Code, if the execution of an order in the ordinary course of duty involves such a vio­
lation of the law as is punishable, the superior officer issuing such an order is alone responsible. According
 270 INTERNATIONAL CRIMINAL LAW

contrast, until 1944 the British and US Military Manuals upheld the contrary prin­
ciple respondeat superior.
The turning point occurred after the Second World War, when international instru­
ments (in particular the Charter of the Nuremberg IMT and Control Council Law
no. 10) laid down in unequivocal terms the notion that the defence at issue could not
be urged for international crimes. Subsequent case law and the Statutes of ad hoc
International Criminal Tribunals clearly show that a customary rule has evolved in
international law whereby wherever the conduct of a subordinate amounts to a ser­
ious violation of IHL or ICL, the perpetrator must be held accountable, whatever the
category of crime (a war crime, a crime against humanity, or another crime such as
torture). However, the fact that he acted following superior orders may be urged in
mitigation of punishment.
The rationale for this evolution of the law is clear. Modern crimes, whether com­
m itted in time or war or peace, tend to be large-scale and to involve a high num ber of
persons, both among the perpetrator and the victims. They also tend to be organized
crimes; that is, offences that are committed not by single individuals acting under per­
sonal impulse or out of greed or individual aggressiveness. Instead, they are normally
crimes perpetrated by m ilitary units, by organized groups, or with the support or at
least the acquiescence of state authorities. Things being so, to absolve subordinates
from crimes committed upon superior orders would mean (i) to disregard the moral
and psychological involvement of so many subordinates in the commission of crimes,
and thereby ignore the principle of moral responsibility; (ii) to leave unpunished hun­
dreds of persons who have physically perpetrated the most horrendous offences. This
rationale was efficaciously set out by the IMT in 1946. Nelte, one of the counsel for
defence, had well set forth a common argument by stating that his client Keitel and the
other accused had been ‘merely mouthpieces or tools of an overwhelming will’ (vol. 18,
at 6). The Tribunal generally dismissed the argument, by saying that:
Hitler could not make aggressive war by himself. He had to have the co-operation of states­
men, military leaders, diplomats, and business men. When they, with knowledge of his
aims, gave him their co-operation, they made themselves parties to the plan he had initi­
ated. They are not to be deemed innocent because Hitler made use of them, if they knew

to para. 2, however, the subordinate obeying an order is liable to punishment, if it was known to him that the
order of the superior involved the infringement of the civilian or military law. This applies in the case of the
accused. It is certainly to be urged in favour of the military subordinates that they are under no obligation to
question the order of their superior officer, and they can count upon its legality. But no such confidence can
be held to exist; if such an order is universally known to everybody, including the accused, to be without any
doubt whatever against the law. This happens only in rare and exceptional cases. But this case was precisely
one of them, for, in the present instance, it was perfectly clear to the accused that killing defenceless people
in the lifeboats could be nothing else but a breach of the law. As naval officers by profession they were well
aware, as the naval expert Saalwachter has strikingly stated, that one is not legally authorised to kill defence­
less people. They well knew that this was the case here. They quickly found out the facts by questioning the
occupants in the boats when these were stopped. They could have gathered, from the order given by Patzig,
that he wished to make use of his subordinates to carry out a breach of the law. They should, therefore, have
refused to obey. As they did not do so, they must be punished’ (at 2586/ 721-2).
 OTHER EXCUSES: SUPERIOR ORDER, NECESSITY, DURESS, AND MISTAKE 271

what they were doing. That they were assigned to their tasks by a dictator does not absolve
them from responsibility for their acts. The relation of leader and follower does not preclude
responsibility here any more than it does in the comparable tyranny of organised domestic
crime (226).
The general rules of ICL on superior order apply to orders of both military and
civilian authorities, and whatever the rank of the superior authority, provided (i) the
subordinate is under a legal obligation to obey (otherwise he would not face a clash of
obligations); (ii) the authority issuing the order wields formal and substantial control
over that subordinate; and (iii) the order involves the perpetration of an international
crime.

13.2.2 T H E IN T E R N A T IO N A L A N D N A T IO N A L LA W T E S T S

The question of superior orders is often framed in different terms: both national legis­
lation and national judgments frequently state that, whenever a subordinate executes a
superior order that is contrary to international rules, he is responsible when the order
is manifestly unlawful. In this respect one can mention national legislation.5 Some
national Military Manuals also take the same stand.6 There is abundant case law on
this matter, starting with a case decided by the Austrian Supreme Military Tribunal
on 30 March 1915 (case of the Russian prisoner of war J. K„ at 20).
This legal regulation may appear to be somewhat inconsistent with the defin­
ition of superior order set out above. However, as a commentator has conclusively
demonstrated,7 this seeming contradiction does not exist. National military manuals
and laws approach the issue in the m anner just outlined because they intend to cover
any violation o f military law, whether or not it amounts to an international crime.
International rules, instead, only regulate the more limited question of international
crimes, and take it for granted that any such crime is manifestly unlawful, with the

5 For example, the French Penal Code (Articles 122-4) of 1994, the Spanish Military Criminal Code of
1985 (Article 21), the Criminal Code of Sweden (1999) (Chapter 24, Section 8), the Israeli law (Section 19(B)
of the Criminal Code Ordinance, 1936), the Peruvian Code of Military Justice (Article 19, para. 7).
6 See, for instance, the US Field Manual of 1956 (§509), the Canadian Manual for Courts Martial (1999,
at 16-5) and the US Manual for Courts Martial (2002 edn), Rule 916(d).
7 Gaeta, ‘The Defence of Superior Orders: the Statute of the International Criminal Court v. Customary
International Law’ 10 EJIL (1999), at 172-91.
See also Hass and Priebke (Appeal) (decision of 15 April 1998). Discussing the acts of German military in
Rome in 1943, and the claim of the defendants that they had executed civilians as a reprisal and upon super­
ior order, the Court noted that Article 40 of the Italian Military Penal Code in Time of Peace was applicable,
whereby an order must be executed unless it is manifestly illegal. The Court went on to state that Article 8
of the Statute establishing the Nuremberg Tribunal had not derogated from that provision; indeed, by laying
down that a superior order could not excuse an order, [Article 8] simply took away from the judge the task of
verifying the concrete manifest illegality of the order and was based on the presumption that such illegality
existed whenever the offence ordered and executed amounted to a war crime or at any rate to a crime subject
to the jurisdiction of the Tribunal. This standard of appraisal was patently grounded on the very essence
of war crimes: these crimes are envisaged for the purpose of protecting fundamental values endowed with
absolute character and valid for the whole of mankind; hence they are laid down regardless of any particular
viewpoint, are clear in their essence and intend to criminalize highly condemnable conduct (at 52-3).
 272 INTERNATIONAL CRIMINAL LAW

consequence that a subordinate executing an order to commit the crime is no dess

responsible than his superior. The only exception is where the law on a particular mat­
ter is obscure or highly controversial, in which case another defence, that of mistake
of law (see infra, 13.5), may be raised.
In practice, two different approaches are taken in case law in determining whether
or not the plea at issue may be upheld. International courts and tribunals, applying
the customary rule outlined above, as well as the provisions of their statutes restating
that rule, simply satisfy themselves that: (i) the act at issue amounted to a war crime or
any other international crime over which the court or tribunal has jurisdiction; (ii) the
order was given by a superior authority and the subordinate carried it out.8
In contrast, many national courts, relying upon the test of manifestly unlawful
orders’, take a different approach. They ask themselves whether: (ij-the order con­
cerned the performance of an act that was undisputedly unlawful, for it constituted
an international crime; and (ii) the person executing the order knew or should have
known the order to be manifestly unlawful.
The best illustration of how a national court proceeds under this text may be found in
the instructions issued by the Judge Advocate to a US Court Martial in Calley. The Judge
Advocate first spelled out the test. He noted that the standard by which to appreciate
whether the subordinate knew the order to be unlawful was whether ‘a m an of ordin­
ary sense and understanding would, under the circumstances, know [the order] to be
unlawful (at 1722). He then specified that, unless the jurors found beyond reasonable
doubt that the accused had acted with actual knowledge that the order was unlawful1,
they would have to infer knowledge of the unlawful nature of the order from a set of
circumstances, such as the rank and educational background of the subordinate, his
military training, and his military experience in prior military operations.9

8 However, also various national courts have taken this stand. For instance, in SIPO-Brussels the Brussels
Court Martial clearly stated in 1951 that whenever the execution of an order involves a war crime, the order
may not amount to a defence and the subordinate is punishable. As the Court pointed out, ‘the execution of
such orders, particularly the gruesome slaughter [of Resistance fighters, by members of the SIPO (Sicherheit
Polizei) at Gangelt, should be considered as a flagrant breach of the laws and customs of war and the laws of
humanity and should be punished as such’ (at 1522).
9 The Judge Advocate then instructed the Court as follows: ‘[U] nless you find beyond a reasonable doubt
that he [Calley] was not acting under orders directing him in substance and effect to kill unresisting occu­
pants of My Lai (4), you must determine whether Lt. Calley actually knew those orders to be unlawful
Knowledge on the part of any accused, like any other fact in issue, may be proved by circumstantial evidence,
that is by evidence of facts from which it may justifiably be inferred that Lt. Calley had knowledge of the
unlawfulness of the order which he has testified he followed. In determining whether or not Lt. Calley had
knowledge of the unlawfulness of any order found by you to have been given, you may consider all relevant
facts and circumstances, including Lt. Calley’s rank, educational background, OCS schooling, other train­
ing while in the Army, including Basic Training, and his training in Hawaii and Vietnam, his experience
on prior operations involving contact with hostile and friendly Vietnamese, his age, and any other evidence
tending to prove or disprove that on 16 March 1968, Lt. Calley knew the order was unlawful. If you find
beyond reasonable doubt, on the basis of the evidence, that Lt. Calley actually knew the order under which
he asserts he operated was unlawful, the fact that the order was given operates as no defence [...] Think
back to the events of 15 and 16 March 1968. Consider all the information which you find to have been given
Lt. Calley at the company briefing, at the platoon leaders’briefing, and during his conversation with Captain
Medina before lift-off. Consider the gunship “prep” and any artillery he may have observed. Consider all
 OTHER EXCUSES: SUPERIOR ORDER, NECESSITY, DURESS, AND MISTAKE 273

It is clear from this and other cases that the application of the national law test
is more cumbersome and, what is more important, leaves more discretionary power
to the judicial body. In contrast, the international test is straightforward and easy to
apply. This point was made in 1998 by the Italian Military Court of Appeal in Hass and
Priebke (at 52-3).
Nonetheless, we shall see that, in practice, both tests have led to the same results.
Indeed, the national law test has mostly been applied in cases where the act performed
was in blatant breach of universally recognized rules of international law. In those
instances where instead the plea of superior order has been upheld by national courts
under the national law test, either the courts erred, or the relevant international rules
were absolutely unclear and probably international courts and tribunals would also
have reached the same conclusion.

13.2.3 C A SE LA W R E JE C T IN G T H E P L E A O F S U P E R IO R O R D E R

National courts and international courts or tribunals have dismissed this plea in numer­
ous cases, in particular in cases concerning the killing or ill-treatment of: (i) defenceless
shipwrecked persons;10(ii) innocent civilians in occupied territory;11 (iii) prisoners of war;12
(iv) non-combatants detained in the combat area;13 and also in cases concerning (v) the
taking of illegitimate reprisals against civilians;14 or (vi) unlawfully punishing civilians
who are acting on behalf of, or collaborating with, the enemy;15or (vii) refusing quarter.16

the evidence which you find indicated what he could have heard and observed as he entered and made his
way through the village to the point where you find him to have first acted causing the deaths of occupants,
if you find him to have so acted. Consider the situation which you find facing him at that point. Then deter­
mine, in light of all the surrounding circumstances, whether the order, which to reach this point you will
have found him to be operating in accordance with, is one which a man of ordinary sense and understanding
would know to be unlawful. Apply this to each charged act which you have found Lt. Calley to have commit­
ted. Unless you are satisfied from the evidence, beyond reasonable doubt, that a man of ordinary sense and
understanding would have known the order to be unlawful, you must acquit Lt. Calley for committing acts
done in accordance with the order’ (1723-4).
10 See, for instance, Llandovery Castle (at 2580-6); Peleus (at 128-9).
11 See, for instance, Schintholzer and others (Military Tribunal of Verona, 21 February 1989, unpublished,
p. 44 of the typescript); Josef Kramer and others (the Belsen trial), at 631-2; Heinrich Gerike and others (the
Velpke Baby Home trial), at 338; Sipo-Brussels case (at 3-10); Gdtzfrid (at 62-6).
12 gee, for instance, Gozawa Sadaichi and others (at 225, 229, 231); Sumida Haruzo and others (at 232,
240-1,258), Strauch and others (at 562-3).
13 See, for instance, Lages (at 2); Zuhlke (at 133-4); Ranter (at 157-9); Zimmermann (at 30-1); Bellmer (at
543); Thomas L. Kinder (at 770-4); Walter Grijfen (at 587-91); Frank C. Schultz (at 137); Charles W. Keenan (at
114-19); Michael A. Schwarz (at 859-61); William L. Calley (US Army Court of Military Review, at 1180-2);
US Court of Military Appeals, at 541-5); Sergeant W. (Brussels War Council, at 3, and Military Court, at 2),
Sablic and others (at 120-1); M. andG. (at 989-90); Major Shmuel Malinki and others (at 88-132).
See also a case where, in an obiter dictum, the Court held that the plea was not applicable in a civil war
(Nwaoga, at 3).
14 See Wagener and others (Rome Military Tribunal, at 52-3; High Military Tribunal, at 746); Neubacher
Fritz (at 39-41).
15 See for instance WolfgangZeuss and others (at 206-7,216).
16 See for instance Nikolaus von Falkenhorst (at 226-7,237).
 274 INTERNATIONAL CRIMINAL LAW

13.2.4 C A SE LA W U P H O L D IN G T H E P L E A

In a few cases courts have upheld the plea because, in their view, either the order was
lawful,17 or the accused lacked the requisite mens rea due to: (i) absence of freedom of
judgment; or (ii) mental disorder; or (iii) mistake of law. In other cases courts found
it necessary first to appraise whether a generic order was lawful, and then to deter­
mine whether the execution of the order by the subordinate was in keeping with inter­
national legal standards.18
I shall discuss here only the cases where, according to the courts, the defendant
lacked the requisite subjective element (mens rea).

17 See for instance Neumann. Upon the orders of a superior officer, the accused had taken part in an
attack on prisoners of war who had refused to work, and had in addition ‘belaboured a prisoner with his
fists and feet’. The German Supreme Court at Leipzig held that the accused could not be held responsible for
these events, for there could be no doubt as to the legality of the order (at 2554). The Court went on to state
that ‘Unless there is to be irreparable damage to military discipline, even in a body of prisoners, disorderly
tendencies have to be nipped in the bud relentlessly and they have to be stamped out by all the means at the
disposal of the commanding officer and if necessary even by the use of arms. It is of course understood that
the use of force in any particular case must not be greater than is necessary to compel obedience. It has not
been established that there was any excessive use of force here. The accused has been charged with having
continued to belabour [the Scottish prisoner of war] Florence when he was lying on the ground and after
the resistance of the prisoners generally had already been overcome. For this, however, no adequate proof
has been forthcoming’ (at 2553-4; at 699 for the English translation). It is notable that in the same case the
Court also ruled out the defence being available to the accused with regard to other instances where he had
ill-treated prisoners of war using what the Court held to be excessive force, not justified by the order (ibid, at
2554-6 and 699-704, for the English translation).
In von Falkenhausen the Brussels Court Martial (Conseil de guerre) held that the superior orders concern­
ing the execution of reprisals against the population could amount to an admissible plea to the extent that
the reprisals were necessary to ensure the security of the Occupant; indeed, according to the Court at the
time these reprisals were carried out, under international law such reprisals could not be regarded as a ‘fla- ■
grant violation of the laws of warfare’ (at 868-70).
18 Reference can be made to V. J. F. G. (Korad Khalid v. Paracommando soldier), brought in 1995 before a
Belgian Military Court. In 1993 a member of the Belgian military troops in Somalia had wounded a Somali
child who was trying to enter the safety area, through barbed wire fencing guarded by the accused. The
Court found that the order ‘to defend and prevent anyone from penetrating into the cantonment of various
Belgian military units’ was lawful (at 1064-6). It then considered how the defendant had carried it out. It
noted that ‘on observing the child creep through the concertina and thus arrive in the immediate vicinity
of the bunker, he [the defendant] first gave the necessary verbal warning in both Somali and English [...] he
then fired two warning shots into the ground about 50 cm away from the child, who still showed no reaction,
[...] he finally decided to fire an aimed shot [...] at non-vital organs, viz. the legs [...] the procedure fol­
lowed by the accused was the only possible one to fulfil his defensive duties [...] he was physically incapable
of catching the intruder (in view of the special position of the bunker, which was accessible only from the
rear along an aperture in the cantonment wall) [...] and [in addition] it was unrealistic to call upon other
reserve facilities, e.g. the picket; [furthermore] in view of the possible imminent attack, the reaction had to
be prompt and this reaction was also commensurate; [...] all being considered, there was no other action
suitable in the circumstances which could be taken to prevent further penetration [... and] the force used
was unmistakably proportional to the nature and extent of the threat’ (at 1066-7).
A similar case is D. A. Maria Pierre (Osman Somow v. Paracommando Soldier). A Belgian Military Court
held that the order was lawful and that, in accidentally causing the death of a Somali civilian, the Belgian
soldier on guard duty who had executed the order was not responsible for he had not failed 'to exercise fore­
sight and care’ when firing a warning shot which by ricochet had fatally wounded the Somali (at 1069-71).
 OTHER EXCUSES: SUPERIOR ORDER, NECESSITY, DURESS, AND MISTAKE 275

A. Lack of Mens Rea Due to P urported Absence of Freedom of Judgment

In Kappler and others (1948) the Rome Military Tribunal dealt with the unlawful
reprisals ordered by Hitler for the m urder in Rome of 32 members of an SS unit. The
SS Lieutenant Colonel Kappler, besides carrying out those orders, decided to kill ten
more Italians because meanwhile another SS had died as a result of the bombing. In
addition, he had five more Italians killed by mistake: a total of 335 persons. The Court
held that the reprisals were unlawful, and Kappler was guilty of ordering the shooting
often persons plus the additional five people. However, it found that he was not guilty
for the killing of 320 persons ordered by Hitler.19 The C ourt’s reasoning is highly
questionable and, indeed, was ‘reversed’ in subsequent judgments of Italian courts.20
If one were to share this approach, in all cases where a superior gives an unlawful
order the subordinate would be relieved of responsibility. Indeed, one could easily
prove that the superior authorities’ widespread practice of issuing unlawful orders,
together with the great clout of such authorities, bring about a frame of m ind whereby
the subordinate forfeits his awareness and will, and hence lacks the requisite mental
element for the commission of the crime.

B. Lack of Mens Rea Due to Alleged M ental Disorder

There are also other instances where, according to some national courts, the execution
of unlawful superior orders, while not constituting per se a valid defence, may bring
about such a state of m ind in the subordinate, that in the end he lacks any autonomous
will, as well as the intent necessary for criminal responsibility to arise. In such cases
courts have held that the accused found himself in such state of confusion that he was
unable to entertain the mental attitude required for mens rea.
A case in point is Caroelli and others decided by the Italian Court of Cassation
on 10 May 1947. In northern Italy, in the area under the control of the Republic set
up in 1943-5 by Italian fascists with the support and under the control of Germans
(the so-called Repubblica Sociale Italiana), the provincial representative of the govern­
ment {prefetto) had ordered the head of the National Guard, M r Caroelli, to execute
ten partisans by way of reprisal following the killing of a National Guard officer. The
reprisal, in addition to being unlawful, was absolutely arbitrary (the prefetto had been
informed by one of his subordinates that the killing of the National Guard officer was
due more to jealousy than to political motives, and at any rate the police were about to

19 It held that: ‘The mental habit of prompt obedience that the accused had developed working in an
organization based on very strict discipline, the fact that orders with the same content had been previously
executed in the various areas of military operation, the fact that an order from the Head of State and Supreme
Commander of the armed forces, owing to the great moral force inherent in it, cannot but diminish, espe­
cially in a serviceman, that freedom of judgment which is necessary for an accurate appraisal, all these are
elements which lead this Court to believe that it may not be held with certainty that Kappler was aware and
willed to obey an unlawful order’ (at 30). The Supreme Military Tribunal upheld the judgment by a decision
of 25 October 1952, at 97-118.
The Court applied the same reasoning to the four other accused, who had executed Kappler’s order, and
found them not guilty (at 51).
20 See, in particular, Hass and Priebke (Appeal), 15 April 1998, at 52-4.
 276 INTERNATIONAL CRIMINAL LAW

ferret out the perpetrator). The case was brought before the Court of Assize of Padua
which acquitted Caroelli, his deputy, and another officer, on the strength of Article 51,
last paragraph, of the Italian Criminal Code, whereby ‘whoever executes an unlawful
order is not punishable, whenever the law does not allow him to scrutinize the law­
fulness of the order’. On appeal from the Prosecutor, the Court of Cassation held that
reliance upon that provision was wrong, because the order was patently unlawful and
arbitrary, and the subordinates were not bound to carry it out, pursuant to Article
40 of the Military Criminal Code applicable in Time of War. Nevertheless, the three
accused were acquitted, because they ‘lacked freedom of will, in the conduct ordered
by their superior’. The Court emphasized that, when the order was given, Caroelli
tried to oppose it ‘in two agitated talks’ with theprefetto and, when he left theprefetto’s
office, he had a cadaverous appearance’ and ‘could hardly stand on his feet’. According
to the Court this showed that the order brought about in Caroelli a state of ‘psychic
confusion that was also accompanied by clear physical manifestations’ and this ‘con­
fusion was transm itted to his aides’.21
The above reasoning does not comport with the relevant rules and principles of
international law. In any event, assuming that the legal grounds set out by the court
were correct, it remains that in this and similar cases the excuse the defendant might
validly raise is not superior order, but mental disorder (see supra, 12.5.1). In addition,
in all such cases, it would of course be necessary for the courts to be extremely cautious
in establishing the facts and the credibility of witnesses, lest the plea of superior orders
should become a general pretext for negating criminal responsibility.

C. Lack of Mens Rea Due to Uncertainty of Law

In addition to some cases I will cite below—that is Wagener and others (13.4.2) and
Thomas L. Kinder (13.5.3)—it is worth mentioning a significant case, brought before a US
Military Tribunal sitting at Nuremberg. In Wilhelm von Leeb and others (High Command
case), the Tribunal discussed the question of whether the field commanders under trial,
by obeying an order issued by their superior authorities to use prisoners of war for the
construction of fortifications, had complied with an unlawful order and were therefore
guilty. The Tribunal rightly held that the order was not patently illegal because the law
on the matter was unclear; consequently the accused were not responsible under this
count.22

21 According to the Court, ‘when the manifestation of will contrary to the criminal action ordered by
the superior is such as to cause clear physical troubles and a psychic confusion that nullifies the subordin­
ate’s freedom of decision, clouding a clear vision of hierarchical relations, evidently there does not exist that
integrity of awareness and will required for making up a generic criminal intent, and even more the specific
criminal intent necessary for the crime at issue’ (at 2).
22 It is worth quoting the Tribunal’s reasoning: ‘One serious question that confronts us arises as to the use
of prisoners of war for the construction of fortifications. It is pointed out that the [IV] Hague Convention
[of 1907] specifically prohibited the use of prisoners of war for any work in connection with the operation of
war, whereas the later Geneva Convention [of 1929] provided that there shall be no direct connection with
the operations of war. This situation is further complicated by the fact that when the proposal was made to
definitely specify the exclusion of the building of fortifications, objection was made before the [Geneva]
 OTHER EXCUSES: SUPERIOR ORDER, NECESSITY, DURESS, AND MISTAKE 277

Arguably, in this and other similar cases, the defence that can be validly raised is
not that of superior order but of mistake of law.

13.2.5 W H E T H E R U N L A W F U L O R D E R S M A Y R E L IE V E OF
R E S P O N S IB IL IT Y IF G IV E N O N T H E B A T T L E F IE L D

In some cases courts have denied in the case at bar that the execution of an unlawful
order could amount to a defence, while conceding in obiter dicta that, however, this
might have been the case had the order been given in the heat of the battle, when the
subordinate had no time for reflection.
Thus, in Kotdlla, in its judgment of 14 December 1948, an Amsterdam Special
Criminal Court rejected the claim of the accused, an SS commander of the security
staff in the police-run transit camp at Amersfoort, that he had acted under superior
orders, when ill-treating, torturing, and murdering inmates in the period 1942-45.23
Similarly, in Calley, the Court of Military Appeals held, per Judge Quinn, that ‘In
the stress of combat, a member of the armed forces cannot reasonably be expected
to make a refined legal judgment and be held criminally responsible if he guesses
wrong on a question as to which there may be considerable disagreement’ (at 543-4).
Admittedly, it is more difficult for a subordinate to make up his m ind and refuse to
obey an illegal order in the midst of battle. Nonetheless, even under those extraor-

conference to that limitation, and such definite exclusion of the use of prisoners was not adopted. There is
also much evidence in this case to the effect that Russia used German prisoners of war for such purposes. It
is no defence in the view of this Tribunal to assert that international crimes were committed by an adversary,
but as evidence given to the interpretation of what constituted accepted use of prisoners of war under inter­
national law, such evidence is pertinent. At any rate, it appears that the illegality of such use was by no means
clear. The use of prisoners of war in the construction of fortifications is a charge directed against the field
commanders on trial here. This Tribunal is of the opinion that in view of the uncertainty of international law
as to this matter, orders providing for such use from superior authorities, not involving the use of prisoners
of war in dangerous areas, were not criminal upon their face, but a matter which a field commander had the
right to assume was properly determined by the legal authorities upon higher levels’ (at 534; see also 535).
Another case in point is E. van E., decided after the Second World War by a Dutch Special Court of
Cassation. In April 1945 a Dutch unit of resistance fighters in occupied Netherlands, recognized by Royal
decree as members of the Dutch armed forces, shot and killed four members of the Dutch Nazis (NSB) they
had captured. The order to kill them, given by the commander B., was executed by van E. with two other
members of the unit. The Court found that ‘given the circumstances in which the order was given, the
accused was entitled to assume in good faith that his commanding officer was authorized to give that order
for the liquidation of the prisoners, and that this order was within the scope of his subordination’. The Court
therefore found van E. not criminally liable and acquitted him (in Neder}., 1952, 514-16). To better grasp
the purport of this decision, it must be recalled that in the case against the commander, B., the same Court
held that he was not guilty for ordering to shoot and kill the prisoners, because the law was unclear and he
committed a pardonable error of law.
23 The Court held that: according to a universal sense of justice, orders to carry out acts—which, as has
been proven in this case, bear the stamp of inhumanity and unlawfulness—do not simply absolve a subor­
dinate of responsibility under criminal law and the latter remains personally responsible [...] This is all the
more compelling in this case where the issue in question does not under any circumstance concern the kind
of orders that are given in action and on the battlefield, which in themselves must be obeyed immediately,
but, rather, acts of lengthy duration on numerous occasions during which the accused could have given more
sincere signs of his own goodwill and a sense of responsibility’ (at 24/9).
 278 INTERNATIONAL CRIMINAL LAW

dinary circumstances he is required to appraise the legality or illegality of the order,

provided the legal regulation of the matter is universally accepted and clear beyond
doubt. Various US Courts M artial rightly recognized this principle. Calley, as well as
two cases related to another episode, Schwarz and Green, stand out.
In the first case, a US C ourt M artial convicted Calley of the ‘premeditated murder’
of 22 infants, children, women, and old men in the village of My Lai in South Vietnam
on 16 March 1968. First Lieutenant Calley, a platoon leader, claimed that he had acted
upon orders from Captain Medina, given before the occupation of the village and as
soon as the village was invaded. Medina allegedly ordered the killing of every living
thing—men, women, children, and animals—adding that under no circumstances
were they to leave any Vietnamese behind them as they passed through the villages
en route to their final objective. W hether or not this order was truly given (Medina
always denied having issued it), the C ourt M artial held that the deliberate killing of
unarm ed persons offering no resistance and under the control of armed forces was
manifestly illegal. It thus took the view that even in the heat of battle, manifestly illegal
orders must not be executed and, if complied with, may not be used in defence.24 The
C ourt of Military Appeals took the same stand (at 538-46).25
In short, the particular circumstances under which the order is given and executed
should be taken into account not, however, for the purpose of relieving the subor­
dinate of his responsibility in patently illegal cases, but as a possible extenuating
circumstance.26

24 See the Instructions from the military judge to the Court Martial, March 1971 (at 1720-4).
25 See also Schwarz and Green. In 1970 a five-man US Marines patrol in South Vietnam had been sent
out, overnight, to search out, locate, and kill Viet Cong. In a small hamlet called Son Thang they came across
sixteen civilians, women and children, in three huts, and killed all of them upon order of the team leader. The
plea entered by two members of the team, to have acted upon orders and under conditions of extreme tension
and stress for fear of ambushes, was rejected. In the first case (Schwarz), the Navy Court of Military Review
held that ‘the accused could not have honestly and reasonably believed that Herrod’s [the team leader] order
to kill the apparently unarmed women and children was legal [...] The record [...] before us shows beyond
any doubt that Herrod’s orders to kill the unarmed women and children were patently illegal and were rec­
ognized as being so by members of the patrol including private Schwarz’ (at 860, 863). The same view was
taken in Green (see NCMR 70-3811,19 May 1971).
26 In the war crime case of Major Shmuel Malinki and others, an Israeli court, in applying Paragraph
19(B) of the Israeli Criminal Law, drew a distinction between ‘sudden and unexpected orders’ and ‘other
orders’. It stated that ‘A soldier [...] is educated and trained to use his weapon in two types of activities—
independently and in a group framework. In a group framework he is trained to act most mechanically with
general reliance on the commander’s order, without hesitation. He is trained to act quickly and immediately,
as automatically as possible, in order to fulfil his task in the framework suitably. In training and in the daily
routine the soldier is educated towards battle activity, where there is no time for deliberation, no place for
independent thoughts on the part of the private who forms part of a unit, where the results of the battle and
the fate of the soldier and his comrades might depend on his unquestioned obedience to his commander’s
orders and his speed in operating his weapon before the enemy. The modern and sophisticated weapon of
our era adds and obliges educating the soldier in speed and maximum automatism in its use [...] The soldier
who operates within a framework and obeys a sudden and unexpected order to fire from his commander,
will in general be relieved of criminal responsibility for the results in taking a man’s life through his actions,
since the necessary training of the soldier to respond immediately and almost automatically to orders of this
kind deprives him of the possibility that he consider the circumstances under which the order was given and
forces him to rely on the commander regarding the reason for using his weapon’ (at 134-5).
 OTHER EXCUSES: SUPERIOR ORDER, NECESSITY, DURESS, AND MISTAKE 279

13.2.6 T H E IC C S T A T U T E

Article 33 of the ICC Statute surprisingly combines the international approach

that rules out the possibility of urging superior order as a defence to the commis­
sion of all international crimes, and the national approach, based on the ‘manifestly
unlawful order’ standard. Indeed, the provision excludes that superior order may
be relied upon where crimes against hum anity or genocide are at stake, although
it does so by using the terminology typical of the national test (Article 33(2) con­
siders that orders to commit genocide or crimes against hum anity’ shall always be
deemed as manifestly unlawful’). Instead, for war crimes and for this category alone,
Article 33(1) uses both the terminology and the substance of the national test: it stip­
ulates that superior orders shall not relieve a subordinate of criminal responsibility
unless three conditions are met: the person (a) was under a legal obligation to obey
orders of the Government or a superior; (b) did not know that the order was unlawful;
and (c) the order was not manifestly unlawful.
The provision in question is at odds with customary international law, for it does
not include war crimes in the category of offences with regard to which superior order
enjoining their commission are always manifestly unlawful.
This inconsistency is all the more striking because, while one could consider that
traditional international law was not crystal clear about the list of prohibited war
crimes, at present the ICC Statute enumerates in detail the various classes of such
crimes. In addition, the ‘Elements of Crime’ specify the various subjective and object­
ive ingredients of each individual class of crimes. Hence, at present any serviceman is
expected and required to know whether the act he is about to commit falls under the
category of war crimes; he therefore must (or is expected to) be aware of whether or not
the execution of a superior order involves the commission of such a crime. Besides, the
ICC Statute provides in its preamble that it intends to address ‘the most serious crimes
of concern to the international community as a whole’. Hence also the war crimes over
which the ICC has jurisdiction are held to be ‘serious’ and ‘of concern’ to the whole
international community. In addition, Article 8(1) of the Statute stipulates that, with
regard to war crimes, the Court has jurisdiction over such crimes ‘in particular when
committed as part of a plan or policy or as part of a large-scale commission of such
crimes’. One can hardly see how, given all these limitations on the class of war crimes
over which the Court may pronounce, there might be cases where a subordinate could
validly claim that he was not aware of the illegality of the order and that the order was
not ‘manifestly unlawful’.
As a result of the wording of Article 33, an odd condition might arise. In a particu­
lar case the Prosecutor in his indictment may prefer to characterize the same offence
as both a war crime and a crime against hum anity (cumulative charges are allowed,
under existing case law (see 8.8.1)). It could consequently happen that the defence of
superior orders urged—with respect to the offence as a crime against humanity, would
be dismissed out of hand, whereas it could hold, and eventually even be upheld, should
the same offence be classified as a war crime. This result would hardly prove consistent
 280 INTERNATIONAL CRIMINAL LAW

with the object and purpose of the Statute and its intent ‘to put an end to im punity for
the perpetrators’ o f‘the most serious crimes’.
The inconsistency between customary and treaty law should arguably prompt the
interpreters, and in particular the Court, to construe Article 33 strictly, so as to make it
as consonant as possible with customary international law. In other words, when deal­
ing with serious violations of IHL perpetrated on superior orders, the Court should start
from the assumption that an order to commit such violations is by definition ‘manifestly
unlawful’, unless one is faced with the exceptional (and far-fetched) occurrence that the
substantive law on the matter (that is, a particular provision of the ICC Statute) is unclear
and the agent may usefully plead the defence of mistake of law (see infra, 13.5).

13.3 NECESSITY AND DURESS

13.3.1 G E N E R A L

Necessity or duress may be urged as a defence when a person, acting under a threat of
severe and irreparable harm to his life or limb, or to life and limb of another person, perpe­
trates an international crime. The person under threat, although he breaches an inter­
national rule and consequently commits an international crime, is not punishable.
Duress is often termed ‘necessity’, both in national legislation and in cases relating
to war crimes and crimes against humanity. However, there are some im portant dif­
ferences between these two categories of defences:

1. Necessity designates threats to life and limb emanating from objective circum­
stances. As pointed out in the 1958 British Manual of Military Law, necessity proper
covers situations other than those where one is faced with threats or compulsion of a
third party. Necessity denotes, for instance, the condition where a person ‘in extrem
ity of hunger kills [another person] to eat him ’ (§630, no. I).27
Instead, duress always involves threat or psychological compulsion by one or more
persons.28

27 In the 2004 UK Manual necessity is defined as ‘duress by circumstances’ (see §16.42.1).

28 Duress is different from physical compulsion, which occurs when a person, for example, constrains
by force another person who had a weapon in his hand to shoot at a third person, who dies. This compul­
sion is also a defence (according to the 2004 UK Manual criminal responsibility is not incurred by a person
for such acts as he is physically compelled to perform against his will and despite his resistance’, §16.46). A
definition of duress is laid down in Article 19(l)(d) of UNTAET Regulation 2000/15 of 6 June 2000 on the
‘Establishment of Panels with Exclusive Jurisdiction Over Serious Criminal Offences’: [A person shall not
be criminally responsible if, at the time of that person’s conduct’] the conduct which is alleged to consititute
a crime within the jurisdiction of the panels has been caused by duress resulting from a threat of imminent
death or of continuing or imminent serious bodily harm against that person or another person, and the per­
son acts necessarily and reasonably to avoid this threat, provided that the person does not intend to cause
a greater harm than the one sought to be avoided. Such a threat may either be: (i) made by other persons; or
(ii) constituted by other circumstances beyond that person’s control.’ This provision was applied by the SPSC
 OTHER EXCUSES: SUPERIOR ORDER, NECESSITY, DURESS, AND MISTAKE 281

2. In the case of necessity the agent intends to cause an unlawful harm ful effect. In
other words, he does entertain the criminal intent required by the criminal rule: he is
not only aware that by his action he causes the death of another person but he indeed
wills that death, because achieving this result is the only means for him to avert a
serious im m inent threat to his own life. For instance, he wills the death of the other
shipwrecked person who is attempting to climb into the small boat capable of carry­
ing only one person. Nevertheless, the law considers that he must be excused by not
being punished.
In contrast, duress to a large extent negatives the subjective element of the person
under coercion (he does not will the death of the prisoner of war he is constrained by
another person to kill). The criminal intent of the person causing duress in a way sub­
stitutes for his mens rea. Hence, with duress, unlike necessity, a third person, that is,
the person threatening the agent, is held criminally responsible for the harm caused by
the person acting under duress (for instance, a lieutenant is responsible for the death
of an innocent civilian he has constrained a soldier to kill).
The requirements prescribed by international rules for each of these two defences
are, however, the same. The relevant case law (see below) is almost unanim ous in requir­
ing four strict conditions for duress and necessity to be upheld as a defence, namely:
(1) the act charged is done under an immediate threat of severe and irreparable harm
to life or limb; (2) there is no adequate means of averting such evil; (3) the crime com­
mitted is not disproportionate to the evil threatened (as would, for example, occur in
case of killing in order to avert a sexual assault). In other words, to be proportionate,
the crime committed under duress or necessity must, on balance, be the lesser of two
evils or an evil as serious as the one to be averted; (4) the situation leading to duress or
necessity must not have been voluntarily brought about by the person coerced.

13.3.2 N E C E S S IT Y

As stated above, generally speaking, necessity is a broader heading than duress. It des­
ignates threats to life and limb emanating from objective circumstances and not from
another person.
It would seem that international law admits this defence,29 albeit under strict
conditions.30
The law on necessity was clearly set out in Krauch and others (I. G. Farben case, at
1174-9). The defendants had claimed that the utilization of slave labour in I. G. Farben

in Jose Valente; the court found, however, that there was no evidence that the defendant, accused of killing
pro-independence supporters, had been compelled to kill under threat to his life (at 10).
29 The law on necessity (and duress, treated on the same footing) is summarized in vol. XV of the UN Law
Reports, at 174. For the relevant case law see, in particular, Ohlendorfand others (Einsatzgruppen case), at 471
and 480-1; the High Command case, at 509; the Trial of Gustav Alfred Jepsen and others (.Jepsen case), at 357;
the Fullriede case, at 549; Eichmann (Appeal) at 318; Gotzfrid, at 68-70; Zuhlke (134-5); and Finta, at 837. See
also the Case of the Gestapo members, at 112.
For these conditions see Erdemovic, Dissenting Opinion of Judge Cassese, §§14-16.
 282 IN T E R N A T IO N A L C R IM IN A L LAW

plants was the necessary result of compulsory production quotas imposed upon them
by the government as well as the obligatory governmental measures requiring them
to use slave labour to achieve such production. The US M ilitary Tribunal sitting at
Nuremberg summed up the conditions under which necessity is admissible by noting
that an order of a superior officer or a law or governmental decree could not be urged
as necessity unless, in its operation, this order was ‘of a character to deprive the one
to whom it [was] directed of a moral choice as to his course of action’. Consequently
necessity was not available ‘where the party seeking to invoke it was, himself, respon­
sible for the existence or execution of such order or decree, or where his participa­
tion went beyond the requirements thereof, or was the result of his own initiative’ (at
1179).
An interesting case concerning necessity is Stanislaus Bednarek, brought before the
Austrian Supreme M ilitary Tribunal (judgment of 9 September 1916). The accused, a
Russian subject, while being on territory under Russian control, had reported to the
Russian police that three German soldiers were in hiding; as a result they had been
arrested; later on, captured by the Austrian army, the Russian had been accused of
treason and sentenced on 11 October 1915 by an Austrian military court. On appeal,
the Supreme M ilitary Tribunal found that he was not guilty. The General Military
Prosecutor, in submissions of 26 July 1916, had noted that the accused, being subject
to Russian law, was obliged to report the three German soldiers to the police, pursu­
ant to §164 of the ‘New Russian Criminal Law’ of 22 March 1903; he had therefore
acted under ‘irresistible coercion’ (at 4). The Supreme Military Tribunal upheld this
submission and found that the accused had found himself in a condition akin to ‘state
of necessity’; there existed therefore a ‘circumstance excluding culpability’, namely
‘irresistible coercion’ (at 2).
Another case where necessity was upheld is Flick and others. The question was
whether some defendants (Steinbrinck, Burkart, Kaletsch, and Terberger), managers
of various companies belonging to the ‘Flick Konzern’ were guilty of having employed
conscripted foreign workers, concentration camp inmates, or prisoners of war allo­
cated to them through the slave-labour programme of the German Government. The
defendants claimed that they had done so ‘under the circumstances of compulsion
under which such employment came about’. The US M ilitary Tribunal upheld the
plea. It noted that the defendants lived ‘in a reign of terror’. ‘The Reich, through its
hordes of enforcement officials and secret police, was always “present”, ready to go
into instant action and to mete out savage and immediate punishm ent against any­
one doing anything that could be construed as obstructing or hindering the carry­
ing out of governmental regulations or decrees.’ The Tribunal therefore found that
the defendants had acted ‘under clear and present danger’ and acquitted them (at
1199-202). In contrast, the Tribunal rejected the plea of necessity urged by two other
defendants (Flick and Weiss), for they had taken steps not initiated in governmental
circles but in the plant management; they therefore had acted not as a result of com­
pulsion or fear ‘but admittedly for the purpose of keeping the plant as near capacity
production as possible’ (at 1202).
 OTHER EXCUSES: SUPERIOR ORDER, NECESSITY, DURESS, AND MISTAKE 283

Necessity was also accepted as a defence in Veit Harlan. In 1950 the Court of Assizes
of Hamburg found that the accused, the film director who in 1940, at the request and
instigation of Goebbels and under his constant control, had directed the infamous film
Jud Süss (The Jew Süss), had indisputably perpetrated a crime against hum anity pursu­
ant to Control Council Law no. 10: both the objective and subjective elements of such
a crime were present (at 65-8). Indeed, by making the film, which had been viewed by
16 million Germans (at 9), he had significantly contributed to the persecution of Jews.
Nevertheless he was not guilty, for he had acted under necessity. The Court found that
the accused had produced the film on the direct orders of Goebbels. He could not have
refused to obey such orders, for such a refusal ‘since the beginning of the war would
have been regarded as refusal to execute a military order’, and would have been pun­
ishable ‘with the most severe penalties and even the death sentence’. Hence, according
to the Court, ‘the possibility of an open refusal was a priori ruled out’ (at 69). In add­
ition, the Court noted that ‘a great num ber of distinguished persons enjoying wide
consideration were removed from their influential positions, taken to concentration
camps, pushed to commit suicide or executed, on many occasions without even the
outward appearance of legal process. All these facts make it clear that Goebbels, like
the other Nazi leaders, did not shy away from any violent action in order to put into
effect his purposes and intentions’ (at 69-70). The Court went on to state that the pro­
duction of propaganda films had become of crucial importance to Goebbels’ policy of
anti-Semitic persecution, and for that purpose he closely monitored the production
of the film at issue. The Court concluded that, considering all these circumstances, it
was not to be ruled out that, in case of an open or ‘hidden’ refusal, there could arise
a ‘danger for the body and life’ of the film director (at 70). The Court then examined
whether the accused could have avoided executing Goebbels’ order to produce the film
by, for example, pretending to be taken ill, or escaping abroad, and concluded that
none of these means would have profited him (at 71-84). In summary, the accused had
executed Goebbels’ order ‘under threat of danger to body and life’ and was therefore
not culpable.31
A case of necessity (termed ‘duress by circumstances’) is envisaged in the 2004 UK
Manual: ‘where a country suffers a severe food shortage, the commander of a prisoner

31 When appraising these cases, one should always consider them against their historical background. As
far as German cases relating to the Second World War are concerned, one should be mindful of the fact that,
according to academic research (carried out, among other things, by perusing the investigative or judicial
documentation available at Ludwigsburg), in almost no cases where subordinates refused to carry out illegal
orders did the superior authorities take punitive measures. See in particular H. Jäger, Verbrechen unter
totalitärer Herrschaft—Studien zur nationalsozialistichen Gewaltkriminalität (Frankfurt: Suhrkamp, 1982),
83-160; D. Goldhagen, ‘The “Cowardly” Executioner: On Disobedience in the SS’, 19 Patterns of Prejudice
(1985), 19-32; D. H. Kitterman, ‘Those Who Said “No!”: Germans Who Refused to Execute Civilians during
World War IT, in 11 German Studies Review (1988), 241-54. See also A. Rückerl, The Investigation of Nazi
Crimes 1945-1978—A Documentation (Heidelberg, Karlsruhe: C. F. Müller, 1979), at 80-4. It is apparent
from these studies that most cases of refusal to obey unlawful orders did not result in any negative conse­
quence for the subordinate but in some cases refusal resulted in lack of promotion, or demotion. It would
seem that there were no cases where, following a refusal, the subordinate suffered loss of life or limb (see
Jäger, op. cit., 158-60; Kitterman, op. cit., 251-2).
 284 IN T E R N A T IO N A L C R IM IN A L LAW

of war camp who cannot obtain relief supplies or assistance may have no alternative
other than to put prisoners of war on rations that are below the m inim um standards
laid down in Geneva Convention III’ (§16.42.3).

13.3.3 U N A V A IL A B IL IT Y O F N E C E S S IT Y T O M E M B E R S OF
S P E C IA L U N IT S B E N T O N D IS R E G A R D IN G LAW

It is worth emphasizing the fourth requirement mentioned above, in order to high­

light its particular relevance to war-like situations. According to case law, which is
consonant with general principles of ICL, necessity cannot excuse from criminal
responsibility a person who freely and knowingly chooses to become a member of a
unit, organisation, or group institutionally intent upon actions contrary to IHL.32 In
other words, if a person has voluntarily joined a military or param ilitary unit whose
m ain purpose is to engage in crim inal action, he is not allowed to plead in defence to
the crimes perpetrated in that capacity that he acted under threat to his life or limb.
Indeed, when he chose to acquire membership in that unit he knew or should have
known that its prim ary purpose was to perpetrate criminal offences.33

13.3.4 D U R E S S A N D S U P E R IO R O R D E R

In the case law, duress is commonly raised in conjunction with superior orders.
However, there is no necessary connection between the two. Superior orders may be
issued without being accompanied by any threats to life or limb. In these circumstances,

32 In addition to Einsatzgruppen (at 91) and Erhard Miich (at 40), both decided by US courts sitting at
Nuremberg, some cases brought after the Second World War before German courts are particularly sig­
nificant in this respect, for those courts also acted on the strength of Control Council Law no. 10. Thus,
in T. and K„ a case decided by the German Supreme Court in the British Zone, the two accused had been
members of the National-Socialist party, one being Colonel (Standartenführer) of the SA, the other a com­
mittee member of the NSDAP (Nazi party). They had participated in attacks on synagogues on 10 November
1938 (Kristallnacht), and in arson. They claimed that they acted upon superior orders and in addition under
duress (Notstand). The Court dismissed the claim, pointing out that: ‘As an old member of the [National-
Socialist] Party T. knew the programme and the fighting methods of NSDAP. If he nevertheless made him­
self available as official Standartenführer, he had to count from the start that he would be ordered to commit
such crimes. Nor, in this condition of necessity for which he himself was to blame, could he have benefited
from a possible misapprehension of the circumstances that could have misled him as to the condition of
necessity or compulsion’ (at 200-1). See also the decision of the Oberlandesgericht of Freiburg im Breisgau
in the Gestapo informer case, at 200-3, as well as the decision the German Supreme Court in the British
Occupied Zone in H. and others (at 129-30).
A number of cases brought before the Italian Court of Cassation can also be mentioned: see, e.g., the deci­
sion in Spadini (at 354), in Toller (at 920), and in Fumi (at 380). The same position was taken by the Court of
Appeal of Versailles in Touvier (at 341).
33 Interestingly, in the Sipo-Brussels case the Brussels Court Martial took into account voluntary par­
ticipation in a criminal organization, not from the viewpoint of duress, but with regard to the relevance of
superior orders. In restating a decision in previous cases, it held that ‘superior orders cannot be considered
to provide extenuating circumstances, at least in the case where the accused has voluntarily and consciously
joined such an organization [ie a criminal organization such as the Gestapo or the SD]’ (at 1519). On superior
orders given within a ‘criminal organization’, see also Sch. O. at 306-7.
 OTHER EXCUSES: SUPERIOR ORDER, NECESSITY, DURESS, AND MISTAKE 285

if the superior order involves the commission of an international crime (or, under the
different heading referred to above, is manifestly illegal under international law), the
subordinate is under a duty to refuse to obey the order. If, following such a refusal,
the order is reiterated under a threat to life or limb, then the defence of duress may be
raised, and the superior order loses any legal relevance. Equally, duress may be raised
independently of superior order, for example, where the threat issues from a fellow
serviceman, or even a subordinate.34
In evaluating the factual circumstances that may be relevant to duress, according
to a trend discernible in the case law there may arise the need to distinguish between
the various ranks of the military or civilian hierarchy. Clearly, the lower the rank of
the recipient of an order accompanied by duress, the less likely it is that he enjoyed any
real moral choice.

13.3.5 M A Y D U R E S S BE A D E F E N C E T O K IL L IN G ?

In some cases, under the influence of English criminal law courts have set forth the
proposition that for crimes involving killing duress cannot be admitted as a defence,
but may only be urged in mitigation. That this defence is not available if the offence
charged is murder is a principle that goes back to Blackstone35 (who, however, only
adumbrated it) and was eloquently justified by the English criminal lawyer J. F. Stephen
in 1883.36 In short, this principle is grounded in the notion that hum an life is such
a sacred asset that its taking may never be justified, not even when the person that
takes the life of another is under a very serious threat to his own life. For the sake of
safeguarding the value of hum an life, English law therefore prefers to consider guilty a
person acting under duress, although it then attenuates the harshness of this approach
by considering duress a mitigating circumstance and consequently meting out a very
lenient sentence, as occurred in the celebrated case of Dudley v. Stephens (also called

34 One of the first cases where the issue of duress was raised in connection with superior orders is
Llandovery Castle. After finding that the two defendants were guilty of a war crime for they had carried
out the illegal order of their captain Patzig to fire on shipwrecked persons, the court noted that ‘the defence
finally points out that the accused must have considered that Patzig would have enforced his orders, weapon
in hand, if they had not obeyed them. This possibility is rejected. If Patzig had been faced by refusal of the
part of his subordinates, he would have been obliged to desist from his purpose, as then it would have been
impossible for him to attain his object, namely, the concealment of the torpedoing of the Llandovery Castle.
This was also quite well known to the accused, who had witnessed the affair. From the point of view of neces­
sity (Section 52 of the [German] Penal Code) they can thus not to claim to be acquitted’ (at 2586/ 722-3). See
also Jose Valente, at 10.
35 Commentaries, Book IV, at 30,
J. F. Stephen, History of the Criminal Law o f England (1883), ii (New York: B. Franklin, 1964), at 107-9.
He wrote the following: ‘It is of course a misfortune for a man that he be placed between two fires but it
would be a much greater misfortune for society at large if criminals confer immunity upon their agents by
threatening them with death or violence if they refuse to execute their commands. If immunity could be so
secured a wide door would be open to collusion and encouragement would be given to male-factors secret
or otherwise [...] these reasons lead me to think that compulsion by threats ought in no case whatever be
admitted as an excuse for crime though it may and ought to operate in mitigation of punishment in most,
though not all, cases5(at 108-9).
 286 IN T E R N A T I O N A L C R I M I N A L LAW

the Mignonette case),37where the court also set out the basic rationale for this attitude.38
This balancing of values has been called moralistic’ and ‘hypocritical’39 and assailed
for absurdly requiring men to act as heroes (‘To require a person to die so that another
(though innocent) man may be saved will be to invoke in him a standard of heroism that
can hardly be expected’, has written a distinguished Nigerian criminal lawyer).40 In the
case of killing, a choice between the two possible options (duress as a defence or as an
extenuating circumstance) may of coifrse only be based on policy considerations.
Let us consider how courts have dealt with this delicate m atter when pronouncing
upon international crimes, and whether a correct solution can be reached by way of
interpretation of existing law.
An im portant case is Holzer and others, decided by a Canadian Military Court sit­
ting at Aurich, Germany, and applying Canadian law. In March 1945 three Canadian
airmen abandoned their disabled aircraft near Opladen, in Germany, and were cap­
tured by German soldiers. One of the Canadians, who was wounded, was subsequently
killed by the three German defendants. These three raised the defence of superior
orders, as well as that of duress, claiming that they had been compelled at gunpoint
by Lieutenant Schaefer (not among the accused) to kill the wounded airman. Holzer’s
defence counsel insisted on this plea of duress, both in his Opening Address and in his
Closing Address. He relied generally on international law, but on the issue of duress
he quoted German law and in particular Articles 52 and 54 of the German Criminal
Code. Also the defence counsel for the other two accused insisted on this plea. The
plea was, however, assailed by the prosecutor in his Closing Address: citing English

37 Three seamen and a cabin boy of seventeen or eighteen had been cast away in a storm on the high seas,
and compelled to put into an open boat that soon went drifting on the ocean; after eighteen days, being with­
out food and water, two of the seamen, namely Dudley and Stephens, decided to kill the boy and eat him,
while the third dissented; one of the two then killed the boy, and they, with the third seaman, fed on his flesh
for four days. On the fourth day a passing vessel picked up the boat, and the men were rescued, still alive but
‘in the lowest state of prostration’. They were carried to a British port and committed for trial. The Court
held that the defendants were guilty of murder and sentenced them to death. However, the Crown afterwards
commuted the sentence to six months’ imprisonment (at 608).
38 The Court, among other things, stated that ‘Though law and morality are not the same, and many
things may be immoral which are not necessarily illegal, yet the absolute divorce of law from morality would
be of fatal consequence; and such divorce would follow if the temptation to murder in this case were to be
held by law an absolute defence of it. It is not so. To preserve one’s life is generally speaking a duty, but it may
be the plainest and the highest duty to sacrifice it [...] The duty, in case of shipwreck, of a captain to his crew,
of the crew to the passengers, of soldiers to women and children [...]; these duties impose on men the moral
necessity, not of the preservation, but of the sacrifice of their lives for others, from which in no country, least
of all, it is to be hoped, in England, will men ever shrink, as indeed they have not shrunk. It is not correct,
therefore, to say that there is any absolute or unqualified necessity to preserve one’s life’ (at 607).
39 See for instance H. L. Packer, The Limits of Criminal Sanction (Stanford, CA.: Stanford University
Press, 1968), at 118.
40 See K. S. Chkkol, The Law of Crimes in Nigeria (Zaria: Kola, 1989), at 152, as well as, more generally,
the sharp reflections set out at 150-8. This author also notes the following: ‘True, the notion of sacrifice of
one’s own life has in fact some religious foundations when it is remembered that according to Christian the­
ology it was the sacrifice made by Jesus of his own life that has redeemed mankind so that as the Bible tells us
“whoever believeth in him shall not perish but have everlasting life”. However, no matter how grandiose the
notions of sacrifice or heroism may sound it must be realized that an average man or woman can hardly be
expected to be overwhelmed by them when faced with the threat of death’ (at 152).
 OTHER EXCUSES: SUPERIOR ORDER, NECESSITY, DURESS, AND MISTAKE 287

law he excluded duress as a defence in the case of the taking of innocent lives. In stat­
ing the law to the members of the court, the Judge Advocate took the same position as
the prosecutor and he too relied on English law. The court sentenced both Holzer and
another accused (Weigel) to death, while it sentenced the third accused (Ossenbach)
to 15 years’ imprisonment.41 In summary, in this case the prosecutor and the Judge
Advocate clearly upheld—unquestionably by way of ratio decidendi—the traditional
common law position. However, the weight of this decision is belittled by the fact that
in his summing up the Judge Advocate explicitly stated that the court should apply the
Canadian War Crimes Regulations and Canadian law, not international law.
The majority of the ICTY AC took a similar stand in Erdemovic in 1997 (Judges
A. Cassese and Sir Ninian Stephen dissenting).42
Other cases support instead the proposition that duress may be a defence even
where the underlying offence involves the killing of innocents. Ohlendorf and
others (Einsatzgruppen case), decided by the United States Military Tribunal II sitting
at Nuremberg, deserves to be mentioned.43 Indeed, this Tribunal acted under Control
Council Law no. 10, and therefore its decisions are more indicative of international law
than the ones by national courts acting under national legislation.
The defence counsel for Ohlendorf, the lead defendant, submitted in his opening
statement that the question of duress (or necessity, as he termed it) was to be looked at
on the basis of three legal systems: United States law (as the law of the state adm inister­
ing justice in the case at issue), German law (as the law of the defendant), and Soviet
law (as the law of the place where the alleged crimes had been committed). He then
expounded the position under the three legal systems and concluded that necessity was
applicable. The Military Tribunal, in dealing with the plea of duress, cited both Soviet
law and German law, and held that duress could be urged as a defence even in cases of
unlawful killing, provided certain requirements were met.44 In the event, the defence
of duress was rejected on the facts and all the accused but one were convicted.

41 Vol. I, at 289-99, 304, 312,315, 338, 345-6; vol. II, 1-4.

42 In July 1995 the accused, a member of a Bosnian Serb military unit, had participated in the shooting
and killing of many unarmed Bosnian Muslims as a member of an execution squad. Before the ICTY he
pleaded guilty and claimed that he had refused to shoot at the civilians, because he felt sorry for them, but his
commander had told him ‘If you are sorry for them, stand up, line up with them and we will kill you too.’
The Appeals Chamber’s majority held that ‘duress does not afford a complete defence to a soldier charged
with crimes against humanity and/or a war crime involving the killing of innocent human beings’ (§19). The
same majority consequently held that in such cases duress could only be used in mitigation of punishment.
The view propounded in Holzer and others and taken up in Erdemovic by the majority of the ICTY Appeals
Chamber had already been upheld in the provisions of two military manuals. One is the British Military
Manual, para. 629 of which provides: ‘No criminal responsibility is incurred by a person for an act per­
formed by him under an immediate and well-grounded fear for his own life, provided that the act does not
involve the taking of innocent life. Otherwise threats afford no defence to a person accused of a war crime
but may be considered in mitigation of sentence.’ The other is the United States Manual for Courts Martial,
of 1984, whereby duress is a defence ‘to any offence except killing an innocent person’.
43 See Ohlendorf and others (Einsatzgruppen case) at 56-9, 61-82,462-3,471-2,480-1.
44 It is worth quoting the most important passage of the judgment: ‘Let it be said at once that there is no
law which requires that an innocent man must forfeit his life or suffer serious harm in order to avoid com­
mitting a crime which he condemns. The threat, however, must be imminent, real and inevitable. No court
288 IN T E R N A T IO N A L C R IM IN A L LAW

Other cases, and in particular Gustav Alfred Jepsen and others, may be cited in sup­
port of the proposition that duress may excuse the taking of hum an life.45
In spite of this contradictory case law, it would seem that, generally speaking, the
customary rule of international law on duress does not exclude the applicability of
this defence to war crimes and crimes against hum anity whose underlying offence
is murder or unlawful killing. However, as the right to life is the most fundamental
hum an right, the rule demands that the general requirements for duress be applied
particularly strictly in the case of killing of innocent persons. The following proposi­
tions seem to commend themselves.
First, it is extremely difficult to meet the requirements for duress where the offence
involves killing of innocent hum an beings. Indeed, courts have rarely allowed the
defence to succeed in those cases, even where they have in principle admitted its applic­
ability. But for the two cases cited above, plus some Italian and German decisions,46
which stand out as exceptional, the only cases where national courts have upheld the

will punish a man who, with a loaded pistol at his head, is compelled to pull a lethal lever. Nor need the peril
be that imminent in order to escape punishment. But were any of the defendants coerced into killing Jews
under the threat to be killed themselves if they failed in their homicidal mission? The test to be applied is
whether the subordinate acted under coercion or whether he himself approved of the principle involved in
the order’ (at 480).
45 Gustav Alfred Jepsen and others related to the killing of six internees by Jepsen, a Dane who worked as
a guard in a German concentration camp. In April 1945, as the Allied troops were approaching, the German
authorities ordered that the internees be transferred to another camp, those who were fit on foot, those who
were ill by train. Jepsen was one of the guards escorting the train. During the transfer there were various
air raids and a large number of internees died, many of illnesses or starvation. At one point it was ordered
that 52 internees still alive should be shot ‘to avoid typhus’. Jepsen participated in the shooting by killing
six internees. Although in his deposition made under oath he had not mentioned duress, during the trial
proceedings, and then before sentencing, he claimed that the German Obermaat Engelmann who had given
the order to kill all the internees, had compelled him at gunpoint to participate in shooting the internees. His
defence counsel pleaded, among other things, the state of necessity (Notstand) as provided for in Section 54
of the German Criminal Code. The Judge Advocate, in his summing up, stated that duress could be invoked
in the case, provided the requisite conditions were met. The court found Jepsen guilty but, as the Judge
Advocate put it, since there was ‘an element of doubt as to whether or not [he] acted under some degree of
compulsion’, he was sentenced to life imprisonment rather than to death (at 222-4, 233-51,357-9,363).
Also some judgments of German and Italian courts are particularly important (they are quoted in Judge
Cassese’s Dissenting Opinion in Erdemovic). In many of them the plea of duress was not upheld across the
board, as it were, but only with regard to some of the defendants, while it was rejected with respect to other
defendants, i.e„ it was applied with discrimination. It must be noted that almost all of these cases concern
execution squads or execution groups and duress was upheld with regard to minor executants, whereas it
was ruled out with respect to those who had issued orders or to senior officials who, following orders from
the highest authorities, had in their turn ordered the execution of innocent persons.
Finally, oneshouldmentionothercases.wherethecourtconcededthepossibilityofraisingduressasadefence
to a charge of killing innocent people, although the defence failed on the facts. See for instance Llandovery
Castle by the German Supreme Court of Leipzig (at 722-3); Eichmann by an Israeli court (at 340); Müller and
others brought first before the Belgian Military Court of Brussels and then the Belgian Court of Cassation
(see 400-3); Touvier (at 340-1) and Papon (at 151), by French courts; Priebke, by an Italian court (at 55-7);
Retzlaffand others by a Soviet court (at 118-20), as well as a string of German cases and a case recently dealt
with by a Military Court of Belgrade (Sablic and others, at 73, 126). For references to these cases see Judge
Cassese’s Dissenting Opinion, §§31-4.
46 For detailed references to these cases see Judge Cassese’s Separate and Dissenting Opinion in Erdemovic
CAJ), §§35-9.
 OTHER EXCUSES: SUPERIOR ORDER, NECESSITY, DURESS, AND MISTAKE 289

plea in relation to violations of IHL relate to offences other than killing. This bears
out the strong reluctance of national courts to make duress available for offences
involving killing. The reason for this restrictive approach no doubt has its roots in
the fundamental importance of hum an life to law and society. As the Court of Assize
of Arnsberg (Germany) pointed out in Wetzling and others (at 623), the right to life is
one of the most fundamental and precious hum an rights, and any legal system is keen
to safeguard it at the utmost. It follows that any legal excusing of attacks on this right
must be strictly construed and only exceptionally admitted.
Secondly, it is relevant to examine whether a crime would have been committed
in any case by a person other than the one acting under duress, in which case duress
seems admissible as a defence. In fact, where the accused has been charged with par­
ticipation in a collective killing which would have proceeded irrespective of whether
the accused was a participant, the defence has in principle been allowed. Thus the case
law makes an exception for those instances where—on the facts—it is highly probable,
if not certain, that if the person acting under duress had refused to commit the crime,
the crime in any event would have been carried out by persons other than the accused.
The best example is where an execution squad has been assembled to kill the victims,
and the accused participates, in some form, in the execution squad, either as an active
member or as an organizer, albeit only under the threat of death. In this case, if an
individual member of the execution squad first refuses to obey but has then to comply
with the order as a result of duress, he may be excused: indeed, whether or not he is
killed or instead takes part in the execution, the civilians, prisoners of war, etc., would
be shot anyway. Were he to comply with his legal duty not to shoot innocent persons,
he would forfeit his life for no benefit to anyone and no effect whatsoever apart from set­
ting a heroic example for m ankind (a task that the law cannot demand him to fulfil).
His sacrifice of his own life would be to no avail.

13.3.6 T H E IC C S T A T U T E

Article 31(l)(d) rightly lumps necessity and duress together as grounds for excluding
criminal responsibility. They are defined as follows:
[A person shall not be criminally responsible if, at the time of that person’s conduct], The
conduct which is alleged to constitute a crime within the jurisdiction of the Court has
been caused by duress resulting from the threat of imminent death or of continuing or
imminent serious bodily harm against that person or another person, and the person acts
necessarily and reasonably to avoid this threat, provided that the person does not intend
to cause a greater harm than the one sought to be avoided. Such a threat may either be
(i) ade by another person; or (ii) constituted by other circumstances beyond that person’s
control.

To a very large extent this provision codifies customary international law as inter­
preted above, in that, among other things, it does not exclude in principle the excuse at
issue in the event of a person under duress killing another person.
 290 IN T E R N A T IO N A L C R IM IN A L LAW

13.4 MISTAKE OF FACT

13.4.1 G E N E R A L

This excuse, as a defence to a criminal charge, m aybe invoked when, although there is
actus reus, that is conduct contrary to international criminal law, the requisite mens
rea is lacking because the person mistakenly was of the honest and reasonable b e lie f that
there existed factual circumstances making the conduct lawful (see Article 32(1) of the
ICC Statute).47
The erroneous belief about factual circumstances must be based on reasonable
grounds or in other words not be specious or far-fetched. More specifically, the mis­
take must not result from negligence. This proposition is grounded both on (i) the gen­
eral spirit of international criminal law, directed to ban as much as possible behaviour
contrary to international rules protecting hum an values; and (ii) the general principle
of interpretation whereby rules setting forth exceptions to general prohibitions must
be strictly construed.
The proposition is consonant with the general approach to the issue of mistake of
fact taken in most national legal systems (for instance, in France and the UK). It is also
supported by case law on international crimes.
The following example is provided in the US Manual for Air Forces;
A pilot attacks, admittedly in a negligent manner, and consequently misses his target, a
military objective, by several miles. The bombs fall on civilian objects unknown to the pilot.
No deliberate violation of international law has occurred. However, he might be subject
to possible criminal punishment under his own state’s criminal code for dereliction of
duty. He could not be charged with a violation of the law of armed conflict (AFP 110-31,
19 November 1976,15-16).

The plea of mistake of fact was urged by the defence, and acted upon by the court
in Michael A. Schwarz, a case brought before a US Court M artial (at 171-83). The
accused, a member of a five-man night patrol called ‘killer team ’, had gone out on
19 February 1970 to search out, locate, and kill enemy Viet Cong in South Vietnam.
They had soon entered a small hamlet, Son Thang, where there were three huts, occu­
pied by civilian women and children. The team killed 16 civilians. W hen they sur­
rounded the first hut, four women came out and lined up on the patio in front of it.
The accused went inside the empty hut to search it out. While inside he heard the team
leader yell outside ‘Shoot them, shoot them all, kill them.’ He jum ped up and ran out,
and participated in killing the four women. His defence counsel argued that Schwarz
had mistakenly believed that they were under attack because the people standing on
the patio were performing hostile acts. In his instructions to the jury the M ilitary Judge
told them that if they found that the accused mistakenly, but reasonably, believed that

47 ‘A mistake of fact shall be a ground for excluding criminal responsibility only if it negates the mental
element required by the crime.’
 OTHER EXCUSES: SUPERIOR ORDER, NECESSITY, DURESS, AND MISTAKE 291

be was returning fire, and shot the victims only by accident, they must acquit.48 The
Court Martial, while finding the accused guilty of the murder of the other twelve civil­
ians, found him not guilty of the four homicides at the first hut, accepting that he had
been inside when the firing had begun.49
A US Court Martial rejected the plea of mistake of fact in William L. Calley. Lieutenant
Calley, accused of summarily killing a number of defenceless civilians detained by US
troops in a Vietnamese village, claimed, among other things, that he genuinely thought
that the villagers had no right to live because they were enemy, and thus was devoid of
the requisite mens rea, namely malice’ because he was not conscious of the criminal
quality of his acts. The US Army Court of Military Review dismissed the argument.50

13.4.2 M IS T A K E O F FA C T A N D S U P E R IO R O R D E R

Mistake of fact may be urged in defence in particular when an unlawful order is exe­
cuted, any time the accused may prove that he was not aware that the order was unlaw­
fu l in point o f fact.
A telling example of such mistake of fact is given by the 2004 UK Manual: ‘if an
artillery commander is ordered to fire at an enemy command post in a particular
building and he does so believing that it is a command post but it later turns out that,
unbeknown to him, it was a school, he would not be guilty of a war crime because he
did not intend to attack a school’ (§16.45.1). Another example is provided by the US
Manual for Air Forces: if a hospital is selected as a target for attack, ‘[although the per­
son making the selection would be criminally responsible, a pilot given such coordi­
nates would not be criminally responsible unless he knew the nature of the protected
target’ (AFP 110-31,19 November 1976,15-16).

48 The Military Judge stated the following: "The court is advised that if the accused was of the honest belief
that he and his team-mates were being attacked by enemy forces he cannot be found guilty of any offence
charged or the lesser included offences thereto. Such belief no matter how unreasonable will exonerate the
accused. In determining whether the accused was of the belief that enemy forces were attacking him and
his team-mates you should consider the accused’s age, education, military training, and combat experience
together with all the other evidence bearing upon this issue. The burden is upon the prosecution to establish
the accused’s guilt of each offence charged by legal and competent evidence beyond a reasonable doubt. The
accused committed no crime unless he knew that the enemy forces were not attacking him.’ (Reported in
Michael A. Schwarz, Appeal, at 862-3.)
49 On appeal, defence counsel argued that the aforementioned instructions unduly restricted the mem­
bers of the court, for they limited the defence to the case where the accused believed that he was under enemy
attack, without extending it to the case where he believed that the ‘killer team’was attacking the enemy. The
Court of Military Review rejected the argument, ruling that: ‘In the setting of this case we are certain that
the instructions conveyed to the court the direction that the accused must be acquitted unless they found
beyond a reasonable doubt that he did not honestly believe that he was in immediate contact with the enemy
either offensively or defensively’ (at 862-3).
50 The Court held that ‘To the extent this state of mind reflects a mistake of fact, the governing principle
is: to be exculpatory, the mistaken belief must be of such a nature that the conduct would have been law­
ful had the facts actually been as they were believed to be [...] An enemy in custody may not be executed
summarily’ (at 1180).
 292 INTERNATIONAL CRIMINAL LAW

In many cases the accused raised this defence claiming that he had executed enemy
persons because he had reasonably believed that the victims had been properly tried
and sentenced to death, whereas it had then turned out that this was not the case. In a
number of cases courts admitted the defence in law, while however finding that in the
case at issue it was untenable on the facts (see for instance Almelo,51 Stalag Luft III52
and Wagener and others.)53
The Norwegian Supreme Court upheld the defence in Hans, overturning a contrary
decision of the Court of Appeal. The accused, an officer of the German Security Police,
had been charged with executing without trial Norwegian nationals during the bel­
ligerent occupation of Norway by Germany. He had claimed that the execution took
place on the orders of his superior, who had acted pursuant to a secret decree issued
by Hitler in June or July 1944 abolishing German tribunals in occupied territories and
vesting in the German secret police the authority to carry out executions for offences
considered to be of a political character. The Court of Appeal of Eidsivating (Norway)
found the accused guilty, for he had not taken steps to establish the legality of the exe­
cution orders. However, on appeal the Supreme Court of Norway reversed the decision,
among other things because ‘it was not sufficient to support a conviction for wilful
murder, [to hold] that the accused ought to have known the circumstances which made
his act unlawful’ (at 306); in addition the decision of the Court of Appeal ‘did not

51 In the Almelo case, in March 1945, four German members of a special security detachment had arrested
a British pilot in the Dutch village of Almelo. The pilot, after bailing out of a burning Lancaster, was hiding in
a Dutch house in civilian clothes, together with a Dutch civilian, who was hiding from the Germans to avoid
compulsory labour service in Germany. The German officer in charge of the detachment (not on trial) had
told the four accused that the British officer had been sentenced to death and was to be executed, together
with the Dutchman. The two were then shot dead. Defence counsel argued that, so far as the accused knew,
it was quite possible that the two victims were in fact liable to be shot. The Judge Advocate stated that if
the Court found that the accused honestly believed, or that a reasonable man might have believed, that the
British officer had been tried according to the law, and that they were carrying out a lawful execution, they
must acquit the accused. The Court found, however, that the accused were guilty (at 41 and 45).
52 In Stalag Luft III, tried by a British Court Martial in 1947, the accused were charged with killing some
fifty British officers who had escaped in March 1944 from a German internment camp (Stalag Luft III). Some
of the accused pleaded that they had shot the British officers upon superior orders and without knowing
that they were prisoners of war on the run, in the belief that they were liable to punishment and were to be
lawfully executed. Thus, in the case of the accused Jacobs, he claimed that he had been told that the British
officers were 'parachute sabotage agents’ who had been sentenced to death but had then escaped and killed
two German officials during the break-out. The Judge Advocate admitted that this defence was based on
mistake of fact and regarded it as admissible If the facts alleged by the accused were proved (at 15-16). In the
case of the accused Preiss, his plea, regarded as admissible by the Judge Advocate, was that ‘he thought this
was a legal execution and [...] he did not know for certain that Cochran [the British prisoner he shot dead]
was actually an escaped prisoner-of-war’ (at 23). Similarly, in the case of the accused Schulz, according to the
Judge Advocate his defence was that ‘he really believed that this was a legal shooting which was being carried
out in secret for some special purpose and that it was the shooting of two spies, although he knew they might
have been officers, on the orders of some high authority’ (at 27).
53 The Italian High Military Tribunal in Wagener and others admitted that this defence may be invoked
and found valid on its merits; it held that a military, notwithstanding the manifest criminal nature of an
order, may be relieved of responsibility when he makes a culpable mistake of fact’ (in the case at issue the
Tribunal found, however, that the defence was not available to the accused) (at 763-4). See also Buck and
others (at 39-44).
 OTHER EXCUSES! SUPERIOR ORDER, NECESSITY, DURESS, AND MISTAKE 293

disclose sufficiently clearly whether the accused had been aware of the unlawfulness of
his acts, a fact which the Court seemed to have taken for granted’ (at 306).54
In many cases German courts upheld the defence. For instance, in Wülfing and K.,
the two accused were respectively an officer and a sergeant of the German army, serv­
ing on the army’s Special Services. They were accused of a crime against humanity: on
13 April 1945, while the American troops were approaching the German town where
they were stationed, they had killed a German civilian opposed to national social­
ism, whom they considered guilty of instigation to desertion. Wülfing, the officer, had
ordered the other accused (K.) and a non-commissioned officer to execute the civil­
ian; the officer then finished him off with his pistol. In a decision of 4 August 1947 the
District Court of Hagen, acting by virtue of Control Council Law no. 10, found that
the offence, ‘m urder’ (Mord), was a crime against hum anity and therefore sentenced
Wülfing to life imprisonment (at 613-21). In contrast it found that K„ who in any case
might have been held responsible for ‘criminal homicide’ (Totschlag) only55, was in fact
not guilty because he had acted under mistake: he had believed that he was participat­
ing in the execution of a death sentence passed by a regular court (the C ourt held in
addition that he had acted under duress (Notstand) for he had feared that, if he did not
carry out the order to shoot, he himself would be killed by the officer who was standing
by, pistol in his hand; at 618-20).
Similarly, in a decision of 12 December 1950, the German Federal Court of Justice
(.Bundesgerichtshof) upheld the excuse in the Polish prisoner of war case. On 10 October
1940 the accused, commander of a detachment of border guards also entrusted with
assignments by the Gestapo, on the orders of the Gestapo officer superior to him
commanded the execution squad that carried out a death sentence by hanging of K., a
Polish prisoner of war (who in fact had not been court-martialled and duly sentenced).
The Court of Assize of Flensburg acquitted the accused. It found that he lacked the
intent or culpable negligence required for the charge of ‘deliberate killing’ (vorsät­
zliche Tötung). The Federal Court of Justice dismissed the Prosecutor’s appeal and
upheld the acquittal. It noted that it was not necessary to establish whether or not
awareness of the unlawful character of his action was part of the mens rea, and more
specifically was part of the intent required for the crime. W hat mattered was that in
the case at issue the accused had not entertained any doubt that a death sentence had
been issued by a competent authority; hence he was not aware of the unlawfulness of
his act and consequently lacked intent.56

54 Interestingly, in another case, Flesch, both the Court of Appeal of Frostating (Norway) and the
Supreme Court of Norway convicted an officer of the German Security Police because he knew that the
executed persons (Norwegian nationals and Russian prisoners of war) had not been sentenced to death by
a court (at 307).
55 In Germany Section 212 of the Criminal Code provides for Totschlag as distinguished from Mord (mur­
der), provided for in Section 211. Totschlag is roughly equivalent to the second-degree murder of the US.
56 The Federal Court noted that at the time of the offence the accused knew that offices of the Gestapo
imposed and executed penalties against nationals of eastern peoples. At the time of the offence he also
assumed that K. had been sentenced to death in this manner. As the Court of assize explicitly stated, ‘the
accused had no doubt that the judgment had been issued by the competent and appropriate authority in
 294 IN T E R N A T IO N A L C R IM IN A L LAW

13.5 MISTAKE OF LAW

13.5.1 G E N E R A L

Like most national legal systems, international law does not consider ignorance of law
as a ground for excluding crim inal responsibility. Article 32(1) first sentence of the ICC
Statute (‘A mistake of law as to whether a particular type of conduct is a crime within
the jurisdiction of the court shall not be a ground for excluding crim inal responsibil­
ity’) may be held to codify existing customary law.
The rationale behind the principle ignorantia legis non excusat (ignoring the law
may not am ount to a justification for the commission of a crime) is self-evident. The
law is a body of rules which are normally fairly deep-rooted (for in most legal systems
legal rules are consonant with the fundamental moral, or religious values obtaining

accordance with properly conducted proceedings and was legally binding and final’. The accused had based
his conviction on a telex he had received in October 1944 from the Regional office (Gau) of the Gestapo. The
telex was signed by the head of the office, a senior governmental official (Regierungsrat) and stated more or
less that the Polish K. had been sentenced to death for a violation of the ‘Order for the Protection of Law
Enforcement Agents’ and that the main office of the Reich Security Service had ordered that the execution
should take place in the district where the offence had been committed. Such an Order did not in fact exist;
but that is immaterial. The crucial fact is that the accused [...] had believed in some type o f judgment’based
on legal requirements (even if not rendered by a regular court). Furthermore, he knew that fully qualified
lawyers were employed at the higher office of the Gestapo. Ina prior conversation in the Gestapo office it had
been explicitly pointed out that the conviction of foreigners by the Gestapo occurred with the participation
of fully qualified lawyers in something akin to ‘Chamber of Judges’ (at 234).
The Federal Court also dismissed the Prosecution’s ground of appeal against the lower court’s finding that
the accused had not acted ‘negligently in either a factual or legal sense’ in his assumption that the Pole had been
legally sentenced to death and that the task of executing the sentence, entrusted upon him, was lawful. The
Federal Court concluded that the accused was not guilty, because he lacked either intent (Vorsatz) or negligence
(Fahrlässigkeit). It stated that the international law question of whether or not the legal Regulations issued by
the Nazi authorities (on the punishment of Polish war prisoners by Gestapo officers, outside of any regular
trial proceedings) were valid, could be left undecided. In any case, at least at the time of the offence, the legal
issue was still dubious, as was held by a United States Military Tribunal at Nuremberg in the Wilhelm von Leeb
and others case (see Law Reports of Trials of War Criminals 1949, vol. 12, at 86, where a contrary conclusion is
reached). Rather, what really mattered was to establish whether the accused, ‘based on his personal circum­
stances, could and should have recognized the possible legal invalidity’ of those Regulations.
The Court of Assize had answered this question in the negative, after admitting the illegality of those
Regulations. It had pointed out that at the time of the offence one could not expect that the accused, ‘based
on his personal circumstances’ ‘could recognize that possibly those Regulations were contrary to inter­
national law and consequently the death sentence issued by the Gestapo, with whose execution he had been
tasked, was legally invalid’. This was all the more true because the accused, based on the record of police
interrogation of the Pole, ‘was convinced and could be convinced that the Pole had attacked the Police
Superintendent Sch. and seized him by the neck, thereby committing a criminal offence punishable by
death. Although the accused may have been a particularly capable, knowledgeable and experienced official
within the group of criminal investigators to which he had belonged for 24 years, nevertheless, according to
the legally incontrovertible evidence presented to the Court of assize, he did not have the knowledge neces­
sary for appraising these legal issues of public law and international law. There are no indications that, at the
time of the offence, the accused had any reason to mistrust the academically trained head of the Gestapo
office. Moreover, the accused had learned from experience prior to the offence that the administration of
criminal justice against Poles had passed from the hands of the judiciary to the offices of the Gestapo and
that, according to his observations, generally Public Prosecutors and ordinary courts had not opposed this
development’ (at 234-5). See also Schemer Z„ at 712-15, as well as the Case of the Gestapo members, at 112.
 OTHER EXCUSES: SUPERIOR ORDER, NECESSITY, DURESS, AND MISTAKE 295

in society). In addition, legal rules are normally accessible to everybody. Hence all
those living under a legal system are bound to know the law; were one allowed to
successfully plead that he committed a crime because he ignored that that conduct
was prohibited, the road would be open to general non-compliance with the law. The
foundations of society would be undermined. In addition, (i) if ignorance of law were
admitted as a defence, the applicability of criminal norms would differ from person
to person, depending on their degree of knowledge of law; (ii) the admission of such a
defence would eventually constitute an incentive for persons to break the law, by sim­
ply proving thereafter that in fact they were not aware of the existence of a legal ban.
Nevertheless, in ICL there may be cases where a mistake of law may become relevant
as an excuse. This occurs not when the offender was unaware of the unlawfulness of
his conduct, but when: (i) he had no knowledge of an essential element o f law referred
to in the international prohibition of a certain conduct; (ii) this lack of knowledge did
not result from negligence; and (iii) consequently the person, when he took a certain
action, did not possess the requisite mens rea.
These concepts were applied back in 1921 by the Leipzig Supreme Court in
Llandovery Castle. The captain of a German submarine, after sinking a British hos­
pital ship suspected of transporting troops, had ordered to fire on some lifeboats full
of shipwrecked marines. The court discussed the issue of whether ignorance of law
could be raised as a defence to this war crime. It dismissed the defence, on account of
the clear illegality of the conduct at issue. The court noted that:
The fact that his [the Captain’s] deed is a violation of international law must be well-known
to the doer, apart from acts of carelessness (Fahrlässigkeitsvergehen), in which careless igno­
rance (fahrlässige Unkenntnis) is a sufficient excuse. In examining the question of the exist­
ence of this knowledge, the ambiguity of many of the rules of international law, as well as
the actual circumstances of the case, must be borne in mind, because in war time decisions
of great importance have frequently to be made on very insufficient material. This consid­
eration, however, cannot be applied to the case at present before the court. The rule of inter­
national law, which is here involved, is simple and is universally known. No possible doubt
can exist with regard to the question of its applicability. The court must in this instance
affirm [Captain] Patzig’s guilt of killing contrary to international law (at 2585/721).

13.5.2 E X C E P T IO N S T O T H E IN A P P L IC A B IL IT Y O F T H E E X C U SE

In those areas of IHL or ICL where the rules are clear, incontrovertible, and universally
recognized, one is barred from invoking the plea (or, if one puts forward the defence,
the court must dismiss it out of hand). Many cases support this proposition.57

57 In Jung and Schumacher, a case brought before a Canadian Military Court sitting at Aurich in Germany,
the Judge Advocate, after discussing the legal position of the two defendants (one had ordered the other to
execute a Canadian prisoner of war), noted: ‘Both Jung and Schumacher have admitted that they knew the
killing of a prisoner to be wrong. If I am wrong in this, the Court will correct me since they find the facts. In
any event, ignorance of the law is no excuse’ (at 221).
Similarly, in Buhler, the accused (Secretary of State and Deputy Governor General of that part of Poland
occupied by German armed forces and known as the Government-General), charged with war crimes and
 296 IN T E R N A T IO N A L C R IM IN A L LAW

However, legal certainty and clarity are not commonly found in ICL. As was pointed
out above (1.2), this body of law has grown gradually, in a somewhat haphazard man­
ner, and largely consists of customary, that is, unwritten rules. Often some of these
rules of a customary nature are loose, opaque, or ambiguous. In the case of treaty rules,
frequently they are not couched in clear and detailed terms. In addition, state agents
normally tend to behave in accordance with their own national law, ignoring the legal
commands deriving from international law. National law implementing international
rules may contain gaps, or be unclear, or refrain from explicitly referring to inter­
national rules on points not covered by it. Case law has undoubtedly elucidated many
obscure points, but is still far from clarifying all the main areas of ICL. Furthermore,
it is indeed true what the Judge Advocate said in his summing up in Peleus; that is that
‘no sailor and no soldier can carry with him a library of international law’.58 As a court-
tervailingfactor to theseflaws of ICL, courts therefore tend to attach to mistake of law a
greater weight than the one most national legal systems attribute to the same excuse,59
A generally balanced approach to the delicate legal issue can be found in the Dutch
jurisprudence. Some cases in particular need to be mentioned. In Wintgen the Special
Court of Cassation upheld the defence. The accused, a member of the German Security
Police in occupied Netherlands, acting under orders set fire to a number of houses
near Amsterdam as a reprisal for acts of sabotage perpetrated by unknown persons on

crimes against humanity, had pleaded ignorance of international law; the Polish Supreme National Tribunal
sitting in Cracow rejected the plea on the grounds that as a doctor of laws the accused must have possessed
sufficient knowledge of the rights and duties of an Occupying Power and of the general principles of crim­
inal law common to all civilized countries (at 682). In Enkelstrohth a Dutch Special Court at Arnhem held
that the accused, a German police officer, must know that the shooting without previous trial even of a spy
caught red-handed was contrary to the Hague Regulations, the more so because several German Ordinances
promulgated in occupied Netherlands had enacted precise rules for the trial of saboteurs; according to the
court the shooting in question was so clearly at variance with international law that even a police officer of
inferior rank must have known that it was unlawful (at 685-6).
Similarly, in William L. Calley a US Court of Military Review held that the accused could not rely upon
the defence of mistake of law for he willingly had summarily executed enemy civilians in custody. The Court
stated that ‘Mere absence of a sense of criminality is [...] not mitigating, for any contrary view would be an
excrescent exception to the fundamental rule that ignorance of the very law violated is no defence to violat­
ing it. The maxim ignorantia legis neminem excusat applies to offences in which intent is an element [...] “It
matters not whether appellant realized his conduct was unlawful. He knew exactly what he was doing; and
what he did was a violation [...] of a nature which had to be shown to be knowing and wilful]. He intended to
do what he did, and that is sufficient” (United States v. Gris, at 864)’ (at 1180).
58 Addressing the question of superior orders, he stated the following: ‘It is quite obvious that no sailor
and no soldier can carry with him a library of International law, or have immediate access to a professor in
that subject who can tell him whether or not a particular command is a lawful one’ (at 129). With specific
regard to the case at bar (alleged killing of shipwrecked persons), the Judge Advocate noted that ‘If this were
a case which involved the careful consideration of questions of International Law as to whether or not the
command to fire at helpless survivors struggling in the water was lawful, you might well think it would not
be fair to hold any of the subordinates accused in this case responsible for what they are alleged to have done;
but it was not fairly obvious to you that if in fact the carrying out of Eck’s command involved the killing of
these helpless survivors, it was not a lawful command, and that it must have been obvious to the most rudi­
mentary intelligence that it was not a lawful command, and that those who did that shooting are not to be
excused for doing it upon the ground of superior orders?’ (at 129).
59 This point was well made in 1921 by the German Supreme Court in Llandovery Castle as cited above.
 OTHER EXCUSES'. SUPERIOR ORDER, NECESSITY, DURESS, AND MISTAKE 297

a nearby railway line. The Court held that his action amounted to a war crime, for it
was contrary to Article 50 of the Hague Regulations of 1907 providing that ‘No general
penalty, pecuniary or otherwise, shall be inflicted upon the population on account
of the acts of individuals for which they can not be regarded as jointly and severally
responsible.’ Nevertheless, according to the court the accused could not be punished,
for he was not aware that his conduct constituted a war crime. The court held that the
force of the plea of mistake of law depended on the intellectual status and military
position of the individual concerned and on the nature of the acts committed. The
accused held a very subordinate rank in the Security Police and the destruction of
property was generally held to be morally a less grave offence than, for example, the
killing of innocent civilians or prisoners of war (at 484-6).
In B„ a case brought before a Dutch Court Martial, the C ourt again upheld the
defence. The accused, B„ was commander of a unit of the Dutch Resistance movement
which had been granted the status of armed forces as part of the Royal Dutch Army,
by a Dutch royal decree of 1944. In April 1945 the unit joined with a detachment of
French parachutists who had landed in the Netherlands. Shortly thereafter the group
took prisoner four Dutch Nazis, members oftheNSB, in civilian clothes; they regarded
those Dutch Nazis asfranc-tireurs and traitors. One of them escaped. As Germans had
surrounded the group, there was a danger that, with the help of the escaped prisoner,
the Germans would attack them. Under these circumstances the presence of the pris­
oners (meanwhile the group had captured other Dutch Nazis and released some oth­
ers) presented a serious danger to the unit. B. consulted with the French commander,
who did not instruct him to kill the prisoners. However,
he [B.] did gather from his behaviour, and also from that of the other French parachutists
who were present, consisting of pointing to their Sten guns and drawing their hands across
their throats, that, in his position, they would have proceeded to liquidate [the prisoners].

B. then ordered v. E. to kill the prisoners with the assistance of other members of
the unit. W hen the case was brought before a Dutch Field Court M artial in 1950,
the Prosecuting Officer, in his statement, argued that the conduct of the accused was
unlawful. However, with regard to the accused’s defence that he was mistaken as to the
unlawfulness of the offence, he stated that this was ‘not in itself sufficient to relieve him
of responsibility; for that, the error must also have been pardonable. Only if there was
no intent and no negligence as to the unlawfulness, is the accused not liable criminally.’
In conclusion he asked the Court to find the accused guilty of being an accomplice to
manslaughter and to sentence him to six months’ imprisonment. The Court Martial
agreed with the Prosecuting Officer that the action by B. was unlawful, for, although
the prisoners’ legal status was ‘even inferior to that of franc-tireurs', they nevertheless
could not be shot and killed immediately after being caught. However, the Court noted,
the views among the Dutch unit were that the shooting and killing of the Dutch pris­
oners was not unlawful. This conclusion resulted from the instructions issued to those
units. In addition, it was ‘general knowledge that the broadcasts of Radio Orange from
England were intended to give the impression that members of the NSB were to be
 298 INTERNATIONAL CRIMINAL LAW

regarded as traitors and that it was unnecessary to show them any consideration, nor
would they be shown any’. Consequently, according to the Court, ‘the accused believed
that he was entitled to act as he did and [his] intent was not therefore directed at the
unlawfulness of his actions’. He ‘had to take his decision without being able to consult
a superior, he was placed in a position for which he was not trained and in circum­
stances in which it was practically impossible quietly to consider the relative merits
of the various interests’. The Court concluded that the accused was ‘mistaken as to the
unlawfulness of his actions’, hence was ‘not criminally liable’ and must be acquitted (at
516-25).60
In principle, and as can be inferred from these cases, a court should take into
account various factors:

(i) whether the international rule allegedly breached is universally admitted and
recognized or laid down in written rules of which the defendant is apprised
(on this issue mention should be made of the decision of the ICC P-TC in
Lubanga),61 or is instead controversial, or obscure, or open to glaringly
differing interpretations;
(ii) the intellectual status including the education, training, etc. of the person
relying upon this defence;

60 Another Dutch case where the court upheld the defence of excusable mistake of law is Arlt, decided on
7 November 1949 by the Special Court of Cassation. The accused, a German judge, had been charged with
a war crime for having sentenced to death a Dutchman who had participated in a strike. The Court held
that the establishment by the civil administration of the German Occupying Power of a summary Court
Martial (Polizeistandgericht) was contrary to international law. Nevertheless, it stated that: ‘With regard to
the question of what penalty the accused deserves—perhaps even to the question of whether he deserves at all
punishment on the ground of excusable error in law—the judgment should take into account the manner in
which he, within the established framework, has discharged his judicial functions’ (at 2).
In contrast, in Zimmermann the same Court held that the defence under discussion was not available
to the accused. Zimmermann, during the German occupation of the Netherlands, was a German official
attached to the Dutch Provincial Labour Office of Meppel; in this capacity he was responsible for the deport­
ation of many Dutch workers to Germany for forced labour there. His reliance, in the appeal to the Court,
on ‘his alleged ignorance of the criminal nature of the German deportation of Dutch men to slave labour
to Germany’ was of no avail. The Court stressed that ‘similar practices applied by Germany on a much
smaller scale in the First World War in Belgium and Northern France gave rise to general outrage and even
prompted attempts at intervention on the part of neutral countries [...]; such responsible German officials
as the then Head of the Political Department and Representative of the Foreign Office in Belgium, von der
Laneken, and the then Governor-General, von Bissing, opposed this measure as a violation of international
law or as a dangerous error [...] [hence] it must be regarded as a matter of general knowledge that public
opinion condemned these practices’ (at 30-2).
61 The defendant had claimed that at the time of commission of the crime of which he stood accused he
had not known that to recruit children under 15 and have them actively participate in armed hostilities
amounted to a war crime. The Pre-Trial Chamber rightly noted that various elements proved that instead
he was aware of the criminal nature of that conduct. (A few months before the commission of the crime the
Democratic Republic of Congo (DRC) had ratified the ICC Statute, which subjects that crime to the ICC
jurisdiction; that state had already ratified the Geneva Conventions and Protocols on IHL, which prohibit
the use of children under 15; in addition, even before ratification by the DRC of the ICC Statute, the commu­
nities living in the area where the armed conflict in which the defendant had been engaged were cognizant
of the ICC Statute and of the prohibitions it upheld; finally, a witness had testified that the defendant had
discussed with him questions raised by the ICC Statute and the need to protect children): see §§304-16).
 OTHER EXCUSES: SUPERIOR ORDER, NECESSITY, DURESS, AND MISTAKE 299

(iii) his position within the military hierarchy (clearly, a commander is expected
and required to know the laws of war and more generally, international
prohibitions, while a subordinate, particularly if he ranks very low, may not
be required to possess such knowledge);
(iv) the importance of the value protected by the rules allegedly breached
(normally such values as life and dignity of a hum an being are universally
protected, even under national criminal law, and one may therefore be more
demanding with regard to these values).62

13.5.3 M IS T A K E O F LA W A N D S U P E R IO R O R D E R

A subordinate executing an unlawful order is relieved of crim inal responsibility if

he can prove his ignorance of law, under the conditions set out above with regard to
this defence. Clearly, if the rules of international law on a particular matter, instead of
being universally and clearly established, are confused or exceedingly controversial,
the subordinate may not be aware that the order he has to carry out is contrary to IHL
or to ICL. This mistake of law may be such as to negate mens rea. If that is the case, the

62 Other examples of cases where the excuse in question might be raised can be mentioned. For instance,
the Occupying Power, while it may normally be appropriate the produce of public immovable property (land
and buildings) belonging to the occupied state, under Article 56 of the Hague Regulations may not appro­
priate (i) the produce of those immovable assets belonging to the occupied state that have been set aside for
religious purpose, for the maintenance of charitable or educational institutions or for the benefit of art and
science; or (ii) the produce of the immovable property belonging to municipalities. Hence, if it can be proved
that the officer of an Occupying Power selling the produce of foreign immovable property bonafide ignored
that a certain immovable of the enemy state had been set aside for educational purposes, or that it belonged
to a municipality, and mistakenly believed that it instead belonged to the enemy state, this mistake of law
might be relied upon as an excuse if it could also be proved that, as a consequence of that ignorance, the offi­
cer lacked the requisite criminal intent.
Similarly, under the Third Geneva Convention of 1949 prisoners of war may only be punished for offences
against the law in force in the armed forces of the Detaining Power after a trial has been conducted before
a court offering all the essential guarantees of justice. If the officer of the Detaining Power charged with
enforcing the penalties meted out by courts of that Power ignores that in particular cases the international
prescriptions on the proper conduct of trials against prisoners of war have not been complied with, he may
raise his ignorance as a defence provided he can prove that as a result of his mistake of law he did not have
the requisite mens rea when executing the penalty.
Another example may be taken from Hinrichsen, brought before the Dutch Special Court of Cassation in
1950. Article 53(2) of the Hague Regulations of 1907 provides that the Occupant may seize ‘all the appliances
[...] adapted [...] for the transport of persons or things, even if they belong to private individuals’, but then
must ‘restore’ them and ‘fix compensation’ when peace is made. In the spring of 1945 Hinrichsen, a member
of the German Frontier Customs Guard, seized in occupied Netherlands two privately owned motorcycles
without payment or receipt. After the war he pleaded before a Dutch Criminal Court that his action was not
at variance with international law. The Special Court of Cassation held, on the contrary, that his action was
contrary to Article 53(2), for the accused did not provide the means for later verification of the seizure. It
added however that in determining the penalty it was appropriate to take into consideration the fact that,
unlike the case of requisition under Article 52 of the Regulations, giving a receipt was not expressly pre­
scribed for seizure of means of transport; consequently, the punishment must not be severe (at 486-7). This
is clearly a case where international law is not absolutely clear and unambiguous and therefore invocation of
the defence at issue might be regarded as admissible.
 300 INTERNATIONAL CRIMINAL LAW

subordinate is not criminally liable, not however because the order is lawful, but sim­
ply because the law on the matter is not straightforward and universally recognized,
and the subordinate is not required to settle controversial legal issues when deciding
whether or not to execute an order. This is the approach taken by most courts (for some
exceptions where in contrast the plea was upheld, see supra, 13.2.4(C)).
In Wagener and others in 1950 the Italian High Military Tribunal upheld the plea in
theory but rejected it in the case at issue.63 Other cases worth mentioning are Grumpelt
(■Scuttled U-Boats case)64 and Thomas L. Kinder, heard by a US C ourt M artial in 1954.
In the latter case the defendant, a US airm an serving in a US airbase in Korea situated
south of the actual battle line, and assigned to the air police section to perform guard
duty at a bomb dump, had been accused of killing a detained Korean civilian, who had
been apprehended near the base, and whose legal status was uncertain. In addition to
invoking superior orders, the defence counsel also urged on behalf of the accused a
mistake of law both as to (i) the legality of the order of the superior officer; and as to
(ii) whether or not the airm an was required to obey all orders without exception of a
superior officer. The US Air Force Board of Review, on appeal from the General Court
Martial, admitted the plea in principle, but dismissed it on the facts.65

63 According to defence counsel, General Wagener, when obeying the order to take reprisals against
Italian internees in territory occupied by Germany, had erred, not, however, about criminal law, but about
international law (as far as the lawfulness of reprisals was concerned) and constitutional and international
law (with regard to the power to issue military proclamations). The Court, while implicitly conceding the
admissibility of the defence, rejected it in the case at bar, noting first, that the violation of the laws of war­
fare entailed criminal punishment and, second, that a military may not invoke as a defence ignorance of
the duties inherent in his military status. The commander of a big unit in time of war may not ignore inter­
national obligations deriving from the laws of war, the more so when these obligations coincide with the
principles, prevailing in any law, directed to safeguard the life and limb of individuals’ (at 763).
64 In this case Grumpelt, an officer in the German Navy, had scuttled two German U-boats after the bel­
ligerents had signed the terms of surrender, providing, among other things, that all German vessels would
be handed over to the British Command on 5 May 1945. A few hours after the signature of the Instrument of
Surrender but before the cessation of hostilities, the German Naval Command had issued a coded order that
all U-boats must be scuttled. A few hours later the same Command issued another order countermanding
the first. The accused claimed that (i) he had received the first order but not the second; and (ii) when he had
decided to scuttle the two submarines he was not apprised of the terms of surrender; had he known them,
he would have been able to refrain from obeying the first order. He thus implied that he lacked mens rea, for,
ignoring the terms of surrender, he honestly believed that the (first) order was legal. The Judge Advocate put
the question to the Military Court as follows: ‘Are you satisfied that the man’s state of mind at the time in
question was this: “I honestly believed I had an order: I did not know anything about any surrender; it was
not for me to inquire why the higher command should be scuttling submarines; I honestly, conscientiously
and genuinely believed I had been given a lawful command to scuttle these submarines and I have carried
out that command and I cannot be held responsible”? Gentlemen, that is a matter for you to consider’ (at 70).
The Court found the accused guilty of the charge of committing a war crime.
65 It first cited paragraph 154«(4) of the 1951 Manual for Courts Martial, whereby ‘As a general rule,
ignorance of law [...] is not an excuse for a criminal act. However, if a special state of mind on the part of
the accused, such as specific intent, constitutes an essential element of the offence charged, an honest and
reasonable mistake of law, including an honest and reasonable mistake as to the legal effects of known facts,
may be shown for the purpose of indicating the absence of such a state of mind’ (at 775). The Court then went
on to say that ‘As the offence of murder charged in the instant case involves a specific intent to kill, “mistake
of law” is in principle an applicable defence to negative the unlawfulness of the element of the specific intent
to kill.’ Turning to the case at issue, the Court pointed out the following: ‘However, viewing the defence of
 OTHER EXCUSES: SUPERIOR ORDER, NECESSITY, DURESS, AND MISTAKE 301

As is clear from this case law, courts only admit mistake of law as a defence to the
execution of an illegal superior order when it m aybe proved that the subordinate acted
under the honest and reasonable belief that the law allowed the execution of that supe­
rior order. It follows that this defence is not admissible when the law on the m atter is
clear or should be known to any serviceman engaged in armed conflict (or, more gen­
erally, to any person of average intelligence and education).

mistake of law as based on a claim in the instant case that the accused was mistaken in law as to the legality
of the order of the superior officer, the defence fails for a prerequisite of such defence is that the mistake of
law was an honest and reasonable one and as pointed out in the preceding paragraph the evidence not only
does not raise a reasonable doubt as to whether or not the accused possessed an honest and reasonable belief
that the order was legal, but justifies the inference that the accused was aware of the illegality of the order.
Viewing the defence of mistake of law as based on a claim that the accused mistakenly believed the law to be
that a soldier must without exception obey every order of a superior officer, we must also reject the defence
for not [only] is such a view unreasonable, but is so absurd as to render unbelievable an honest belief by the
accused that he entertained such an opinion of the law. The absurdity of such a belief can be illustrated by
innumerable examples such as a superior officer’s orders to commit rape, to steal for him, for the subordinate
to cut off his own head, etc. Accordingly, under the circumstances of the instant case, we find no merit to a
defence based on the principle of mistake of law’ (at 775-6).
 14
IMMUNITIES

14.1 GENERAL: VARIOUS CLASSES

OF IMMUNITIES
One of the possible obstacles to prosecution for international crimes may be consti­
tuted by rules intended to protect the person accused by granting him im munity from
prosecution.
There exist two categories of immunities that may in principle come into play and
be relied upon.

1. Those accruing under international law. They may relate either to conduct of state
agents acting in their official capacity (so-calledfunctional immunities), or protect the
private life of the state official (personal immunities). The former immunities apply, on
the strength of the so-called Act of State doctrine, to all state agents discharging their
official duties. In principle, an individual performing acts on behalf of a sovereign
state may not be called to account for any violations of international law he may have
committed while acting in an official function. Only the state may beheld responsible
at the international level. The latter category of immunities (personal im m unities) are
granted by international_customary.or treaty rules to some c a te g o r ies of in d i v id u a l s
on account of their fnnction.s-a.nd are intended to protect both their private and they-
.public life, or in other words to render them inviolable while in office. Such individuals
comprise Heads of State, prime ministers or foreign ministers, diplomatic agents, and
high-ranking agents of international organizations. They enjoy these immunities so as
to be able to discharge their official mission free from any impairment or interference.
These immunities end with the cessation of the agent’s official duties.
All these immunities may be invoked by a state official beforeforeign courts or other
foreign organs (for example, enforcement agencies).
2. The immunities provided for in national legislation and normally granted to
the Head of State, members of cabinet, and members of Parliament. They normally
cover the acts of the individuals concerned and involve exemption from nqtinnnl j ur­
isdiction. In addition, they also often include im munity from national prosecution for
ordinary crimes having no link with the function and committed either before or dur­
ing the exercise of the functions. However, such immunity terminates as soon as the
functions come to an end, although normally the individual remains immune from
jurisdiction for any official act performed during the discharge of his functions.
 IMMUNITIES 303

The rationale behind these national immunities is grounded in the principle of

separation of powers and in particular the need to protect state officials (say, the Head
of State) from interference by other state organs (say, courts) that could jeopardize
their independence or political action.
All these categories of immunity normally apply to ordinary crimes. Do they also
apply to international crimes? To answer this question one must of course establish
whether there are international customary or treaty rules that cover this matter.

14.2 FUNCTIONAL AND PERSONAL IMMUNITIES

PROVIDED FOR IN INTERNATIONAL
CUSTOMARY LAW

Let us now return to and dwell upon an issue that is of great importance for our
purposes: the distinction between two categories of immunities laid down in inter­
national law; that is, functional (or ratione materiae or organic) immunities and per­
sonal (or ratione personae) immunities. One ought always to distinguish between
these two categories when discussing the question of, among other things, exemption
from foreign jurisdiction.
The first category is grounded in the notion that states must respect other states’
internal organization and may not therefore interfere with the structure of foreign
states or the allegiance a state official may owe to his own state. Hence no state agent is
accountable to other states for acts undertaken in an official capacity and which there­
fore must be attributed to the state. ——
The second category is predicated on the need to avoid a foreign state either infrin­
ging sovereign prerogatives of statest or interfering with the official functions of
a state agent under the pretext of dealing with an exclusively private act (ne impediatur
legatio, i.e. the immunities are granted to avoid obstacles to the discharge of diplo­
matic functions).
This distinction, based on state practice1 as well as some recent judicial
decisions,12is important. Organic orfunctional immunities: (i) relate to substantive law,

1 With regard to the first class of immunities, suffice it to refer to the famous McLeod incident and
the Rainbow Warrior case. For the McLeod case, see British and Foreign Papers, vol. 29, at 1139, as well
as Jennings, ‘The Caroline and McLeod Cases’, 32 AJIL (1938), 92-9; see also the decision of 1841 of the
New York Supreme Court in People v. McLeod, at 270-99. For the Rainbow Warrior case, see UN Reports
of International Arbitral Awards, XIX, at 213. See also the Governor Collot case, in J. B. Moore, A Digest of
International Law, vol. II (Washington: Government Printing House, 1906), at 23.
2 One can mention the judgment rendered by the Supreme Court of Israel in Eichmann (at 308-9),
that handed down by the German Supreme Court (Bundesgerichtshof) in Scotland Yard, at 1101-2
(the Director of Scotland Yard was not amenable to German civil jurisdiction for he had acted as a state agent).
See also the judgment delivered by the ICTY AC in Blaskic (subpoena) (at §§38 and 41).
For other cases see in particular M. Bothe, ‘Die strafrechtliche Immunität fremder Staatsorgane’, in 31
Zeit. Ausl. Off. Recht Volk (1971), at 248-53.
304 INTERNATIONAL CRIMINAL LAW

that is, amount to a substantive defence (although the state agent is not exonerated from
compliance with either international law or the substantive law of the foreign coun­
try—if he breaches national or international law, this violation is not legally imputable
to him but to his state);3 (ii) cover official acts of any de jure or de facto state agent;
(iii) do not cease at the end of the discharge of official functions by the state agent (the
reason being that the act is legally attributed to the state, hence any legal liability for it
may only be incurred by the state); (iv) are erga omnes; that is, may be invoked towards
any other state.
In contrast, personal immunities: (i) relate to procedural law, that is, they render
the state official immune from civil of criminal jurisdiction (a procedural defence);
(ii) cover official or private acts carried out by the state agent while in office, as well
as private or official acts performed prior to taking office; in other words, they assure
total inviolability; (iii) are intended to protect only some categories of state officials,
namely diplomatic agents, Heads of State, heads of government, foreign ministers
(under the doctrine set out by the International Court of Justice in its judgment in the
Case Concerning the Arrest Warrant of 11 April 2000, at §§51-5); (iv) come to an end
after cessation of the official functions of the state agent; (v) may not be erga omnes (in
the case of diplomatic agents they are only applicable with regard to acts performed as
between the receiving and the sending state, plus third states through whose territory
the diplomat may pass while proceeding to take up, or to return to, his post, or when
returning to his own country: so-called jus transitus innoxii, i.e. the right to move
from one place to another without hindrance).
The above distinction permits us to realize that the two classes of immunity
coexist and somewhat overlap as long as a state official who may also invoke personal
or diplomatic immunities is in office. While he is discharging his official functions,
he always enjoys personal immunity.4 In addition, he enjoys functional immunity,
subject to one exception that we shall see shortly, namely in the case of perpetration of
international crimes. Nonetheless, the personal im munity prevails even in the case of
the alleged commission of international crimes, with the consequence that the state
official may be prosecuted for such crimes only after leaving office.

3 Nevertheless, it would seem that if the state official acting abroad has breached criminal rules of the
foreign state, he may incur criminal liability and be liable under foreign criminal jurisdiction (at least, this
is what happened both in McLeod and in the Rainbow Warrior case). Be that as it may, it seems certain, how­
ever, that the state official in question will not in any case be asked to pay for any damage his act may have
caused. The state for which he acted remains internationally responsible for that act and will have to bear all
the legal consequences of such responsibility.
4 For a recent departure from this rule, see the 2002 decisions of the European Union concerning the
freezing of the private assets of Mugabe (head of state in Zimbabwe): see Council Common Position of
18 February 2002 concerning restrictive measures against Zimbabwe (2002/145/CFSP), in Official Journal
of the European Communities, 21.32.2002, L50/1; Council Regulation (EC) No. 310/2002 of 18 February
2002 on the same matter, ibid., L50/4; Council Common Position of 22 July 2002 amending Common
Position 2002/145/CFSP, ibid., L195/1; Commission Regulation no. 1643/2002 of 13 September 2002, ibid.,
L247/22; and Council decision of 14 September 2002 implementing Common Position 2002/145/CFSP, ibid.,
L247/56.
 IMMUNITIES 305

14.3 THE CUSTOMARY RULE LIFTING FUNCTIONAL

IMMUNITIES WITH RESPECT TO
INTERNATIONAL CRIMES

(A) T H E Q U E S T I O N O F I M M U N I T Y F R O M P R O S E C U T I O N

The traditional rule whereby senior state officials may not be held accountable for acts
performed in the discharge of their official duties was significantly undermined after
the Second World War, when international treaties and judicial decisions upheld the
principle that this ‘shield’ no longer protects those senior state officials accused of war
crimes, crimes against peace, or crimes against humanity. More recently, this prin­
ciple has been extended to torture and other international crimes.
It seems indisputable that by now an international general rule has evolved on the
matter. Initially this rule only applied to war crimes and covered any member of the
military of belligerent states, whatever their rank and position. W hen the major provi­
sions of the London Agreement of 8 August 1945 (setting forth the Statute of the IMT)
gradually turned into customary law, Article 7 (‘The official position of defendants,
whether as Heads of State or responsible officials in Government departments, shall
not be considered as freeing them from responsibility or mitigating punishm ent’) has
also come to acquire the status of a customary international rule.
National case law proves the existence of such a rule. Many cases where state m ili­
tary officials were brought to trial demonstrate that state agents accused of war crimes,
crimes against humanity, or genocide may not invoke before national courts their
official capacity as a valid defence. Even if we leave aside cases where tribunals adjudi­
cated on the strength of international treaties or Control Council Law no. 10, a string
of significant judgments where courts applied national law should be mentioned.5
Admittedly, in most of these cases the accused did not challenge the court’s jurisdic­
tion on the ground that he had acted as a state official. The fact remains, however, that
the courts did pronounce on acts performed by those officials in the exercise of their
functions. The defendants’ failure to raise the ‘defence’ of acting on behalf of their
state shows that they were aware that such defence would have been of no avail. In add­
ition, in some cases the defendant did plead that he had acted in his official capacity
and hence was immune from prosecution. This, for example, happened in Eichmann,

5 One may recall, for instance, Eichmann in Israel (at 277-342), Barbie in France (see the various judg­
ments in 78 ILR, 125ff, and 100 ILR 331ff), Kappler (193-9), and Priebke in Italy (959ff), Rauter (526-48),
Albrecht (747-51), and Bouterse in the Netherlands (Amsterdam Court of Appeal), Kesserling (9ff) before a
British Military Court sitting in Venice, and von Lewinski (called von Manstein) before a British Military
Court in Hamburg (523-4), Pinochet in the UK (see infra, n. 7), Yamashita in the USA (1599ff), Buhler before
the Supreme National Tribunal of Poland (682), Pinochet and Scilingo in Spain (at 4-8 and 2-8, respect­
ively), and Miguel Cavallo in Mexico (by Judge Jesus Guadalupe Luna authorizing the extradition of Ricardo
Miguel Cavallo to Spain).
 ЗОб INTERNATIONAL CRIMINAL LAW

where the accused raised the question o f ‘A ct of State’. Although the Court used that
terminology, which could be misleading, in essence it took the right approach to
the question at issue and explicitly held that state agents acting in their official cap­
acity may not be immune from criminal liability if they commit international crimes
(at 309-12).
It can also be conceded that most of the cases under discussion deal with military
officers. However, it would be untenable to infer from that fact that the customary rule
only applies to such persons. It would indeed be odd that a customary rule should have
evolved only with regard to members of the military and not for all state agents who
commit international crimes.
Besides, it is notable that the Supreme Court of Israel in Eichmann (at 311) and more
recently various Trial Chambers of the ICTY have held that the provisions of, respect­
ively, Article 7 of the Charter of the IMT at Nuremberg and Article 7(2) of the Statute of
the ICTY (both ofwhich relate to any person accused of one of the crimes provided for in
the respective Statutes) ‘reflect a rule of customary international law’.6 In 2002 in Letkol
Inf. Soedjarwo the Indonesian Ad Hoc Court on Human Rights held that the relevant
provision of the ICC Statute has ‘developed’into ‘a legal principle’ (at 23). Furthermore,
Lords Millet and Phillips of W orth Matravers in the House of Lords’ decision of
24 March 1999 in Pinochet took the view, with regard to any senior state agent,
that functional im m unity cannot excuse international crimes.7 The ICTY Appeals
Chamber had already set out this legal proposition in Blaskic (subpoena) (§41) (see
also SCSL, TC, Taylor (Decision on the im munity from prosecution), §§52-3).
In addition, im portant national M ilitary Manuals, for instance those issued in 1956
in the USA and in 1958 (and then in 2004) in the UK,8 expressly provide that the fact
that a person who has committed an international crime was acting as a government
official (and not only as a serviceman) does not constitute an available defence.
It is also significant that, at least with regard to one of the crimes at issue, genocide,
the ICJ implicitly admitted that under customary law official status does not relieve
responsibility (see Reservations to the Convention on Genocide, at 24).9

6 See Karadzic and others (§24), Furundzija (§140), and Slobodan Milosevic (decision on preliminary
motions) (§28).
7 See at 171-9 (Lord Millet) and 186-90 (Lord Phillips of Worth Matravers). Instead, according to Lord
Hope (at 152), Pinochet lost his immunity ratione materiae only because of Chile’s ratification of the Torture
Convention. In other words, for him the unavailability of functional immunity did not derive from custom­
ary law; it stemmed from treaty law.
8 See the US Department of the Army Field Manual, The Law of Land Warfare (July 1956), §§498 and 510.
See also the British manual, The Law of War on Land (1958), at §632 and the 2004 Manual of the Law of
Armed Conflict (UK Ministry of Defence, London: Oxford University Press, 2004), at 16.38.1. (‘Heads of
State and their ministers are not immune from prosecution and punishment for war crimes. Their liability is
governed by the same principles as those governing the responsibility of civilian authorities’).
9 One should also recall that on 11 December 1946 the UN General Assembly unanimously adopted
Resolution 95, whereby it affirmed’ ‘the principles recognized by the Charter of the Nuremberg Tribunal
and the judgment of the Tribunal’. These principles include Principle III as formulated in 1950 by the UN
International Law Commission. This Principle provides as follows: “The fact that a person who committed
an act which constitutes a crime under international law acted as Head of State or responsible Government
official does not relieve him from responsibility under international law.’ See YILC (1950-11), 192. All the
 IMMUNITIES 307

Arguably, while each of these elements of practice, on its own, cannot be regarded
as indicative of the crystallization of a customary rule, taken together they may be
deemed to evidence the formation of such a rule (a rule, it should be added, on whose
existence legal commentators seem to agree, although admittedly without producing
compelling evidence concerning state or judicial practice,10 and which the Institut
de droit international recently restated, at least with regard to Heads of State or
government).11
Let me emphasize that the logic behind this rule, which was forcefully set out as
early as 1945 by Justice Robert H. Jackson in his Report to the US President on the
works for the prosecution of major German war criminals,12 is in line with present-
day trends in international law. Today, more so than in the past, it is state officials,
and in particular senior officials, that commit international crimes. Most of the time
they do not perpetrate crimes directly. They order, plan, instigate, organize, aid and
abet, or culpably tolerate or acquiesce, or willingly or negligently fail to prevent or
punish international crimes. This is why ‘superior responsibility’ has acquired such
importance since Yamashita (1946) (see above, 11.4). To allow these state agents to go
scot-free only because they acted in an official capacity, except in the few cases where

Nuremberg Principles, Israel’s Supreme Court noted in Eichmanti, ‘have become part of the law of nations
and must be regarded as having been rooted in it also in the past’ (at 311).
It is notable that the UN SG took the same view of the customary status of the Genocide Convention (or,
more accurately, of the substantive principles it lays down), a view that was endorsed implicitly by the UN
Security Council (see Report of the Secretary-General Pursuant to Para. 2 of Security Council Resolution
808 (1993), UN Doc. S/25704, §45) and explicitly by a TC of the ICTR in Akayesu (§495) and of the ICTY in
Krstic (§541).
A further element supporting the existence of a customary rule having a general purport can be found in
the pleadings made by the two states (the Congo and Belgium) that were in dispute before the International
Court of Justice in the aforementioned Case Concerning the Arrest Warrant of 11 April 2000. In its Memoire
of 15 May 2001, the Congo explicitly admitted the existence of a principle of ICL, whereby the official status
of a state agent cannot exonerate him from individual responsibility for crimes committed while in office;
the Congo also added that on this point there was no disagreement with Belgium (Memoire, at 39, §60).
10 See, e.g.,S. Glaser,‘L’Acted’Etatetleproblemedelaresponsabiliteindividuelle’.Revwedeciroifpefialefcfe
criminologie (1950), Iff.; S. Glaser, Introduction, 71-6; M. Bothe, supra n. 2,254-7; Y. Dinstein, ‘International
Criminal Law’, 5 IYHR (1975), 82-3; A. Bianchi, ‘Immunity versus Human Rights: The Pinochet Case’,
10 EJIL (1999), 269-70.
11 See the Resolution on ‘Immunities from Jurisdiction and Execution of Heads of State and of
Governments in International Law’ adopted at the Session of Vancouver (August 2001), Article 13(2).
12 In his Report to the US President of 6 June 1945, Justice R. H. Jackson (who had been appointed by
President Roosevelt as ‘Chief Counsel for the United States in prosecuting the principal Axis War Criminals’)
illustrated as follows the first draft of Article 7 of the London Agreement (whereby ‘The official position of
defendants, whether as Heads of State or responsible officials in Government departments, shall not be
considered as freeing them from responsibility or mitigating punishment), contained in a US memoran­
dum presented at San Francisco on 30 April 1945: ‘Nor should such a defence be recognized as the obsolete
doctrine that a head of state is immune from legal liability. There is more than a suspicion that this idea is
a relic of the doctrine of the divine right of kings. It is, in any event, inconsistent with the position we take
toward our own officials, who are frequently brought to court at the suit of citizens who allege their rights to
have been invaded. We do not accept the paradox that legal responsibility should be the least where power
is the greatest. We stand on the principle of responsible government declared some three centuries ago to
King James by Lord Justice Coke, who proclaimed that even a King is still “under God and the law’” (in
International Conference on Military Trials, 47).
 308 INTERNATIONAL CRIMINAL LAW

an international criminal tribunal has been established or an international treaty is

applicable, would mean to bow to traditional concerns of the international community
(chiefly, respect for state sovereignty). In the present international community respect
for human rights and the demand that justice be done whenever human rights have been
seriously and massively put in jeopardy, override the traditional principle of respect for
state sovereignty. The new thrust towards protection of human dignity has shattered the
shield that traditionally protected state agents.13

(B) T H E Q U E S T I O N O F E X E M P T I O N F R O M T H E D U T Y T O
A S S IS T C O U R T S

An important issue related to that we have just discussed is the extent to which the
current removal of state officials’ im munity from prosecution for international crimes
also sets aside their right not to appear before an international court to testify, or at any
rate to assist the court. To put it differently, may a state official that an international
criminal court, through the issuance of a binding order or subpoena, has ordered to
appear before the court either to give testimony or to deliver probative material, refuse
to do so? Or is he instead legally bound to comply with the order? Clearly, the question
does not turn on answering for international crimes, but on giving or handing over
evidence about crimes committed by others. In Blaskic (subpoena) Croatia contended
that under international law the ICTY was not allowed to issue binding orders to state
organs acting in their official capacity; hence it asked the AC to quash the subpoena
duces tecum (a judicial injunction to hand over evidence, accompanied by a threat of
penalty in case of non-compliance) issued by an ICTY TC to the Croatian defence
Minister, which ordered him to produce military documents or, alternatively, appear
before the Chamber to show cause of non-compliance with the order. The TC relied
upon Article 18(2) of the ICTY Statute, which grants the Prosecutor express author­
ity to deal with state authorities and therefore implies, according to the TC, a general
power of the Tribunal directly to approach’ the relevant state officials. It consequently
held that state officials could be directly addressed by the Tribunal by means of
compelling orders (Blaskic, Decision on the Objection o f Croatia to the Issuance o f sub­
poena duces tecum, §§67-9). The AC held instead that the general customary rule on
functional immunities of state officials, though set aside by another customary rule
where such officials are accused of international crimes, was still applicable when it
came to the question of state cooperation with international criminal courts. These
courts face states, so did the AC argue, and have therefore to address themselves to

13 A recent deviation from the rule should, however, be stressed. In 2007, in Ibrahim Matar and others
v. Avraham Dichter, the US District Court for the Southern District of New York dismissed a civil action
brought before US courts under the Alien Torts Statute against a former Israeli agent who, in his capacity as
head of the Shin Beth, had allegedly authorized, planned, and directed the bombing on 22 July 2002 of an
apartment building in Gaza City housing a Palestinian terrorist (the bombing caused many deaths and other
casualties among civilians, and was termed in the petition a war crime). The US District Court, applying the
US Foreign Sovereign Immunities Act, held that that action was covered by immunity (at 4-15).
 IMMUNITIES 309

states, not to individual state officials, if they intend to order the production of docu­
ments, the seizure of evidence, etc. (§§42-3). The AC buttressed this legal argument by
noting that in any case, were the state to refuse to deliver documents, the state official
concerned would be bound by such refusal, and his appearing in court publicly to
explain such refusal would serve little purpose (§44).
It would seem that the AC laid too much emphasis on state sovereignty and tradi­
tional international law. The contention is warranted that at present the expansion
of the hum an rights doctrine and the thrust towards international criminal justice
involve a significant erosion of traditional tenets. The duty of states to cooperate with
international criminal courts that they have either voluntarily accepted or to which
they are subjected on the strength of binding resolutions of the UN SC, entails that
these courts are authorized to issue binding orders or subpoenas directly to state offi­
cials (hence not through designated state channels), whenever they need the handing
over of probative material necessary for the administration of justice. If the highest
state authorities refuse to deliver the documents requested and consequently oblige
the subpoenaed state official to behave accordingly, it is nevertheless im portant for
such official to appear before the international court in order formally and publicly to
set out the reasons for such refusal.
Similarly, international criminal courts are authorized to compel incumbent (and
a fortiori former) state officials to testify in court, by issuing a subpoena ad testifican­
dum. This is borne out by case law.14

14.4 INTERNATIONAL PERSONAL IMMUNITIES

(A) D O T H E Y I N V O L V E I M M U N I T Y F R O M P R O S E C U T I O N ?

The problem of international personal immunities arises with regard to state offi­
cials accused of international crimes when they are abroad: may they be arrested and
brought to trial for the alleged crimes? As we shall see, the problem can be differently
framed and solved when the state official is in his own country; the question then
arises whether under national (or international) law national courts are empowered to
take proceedings against him.
The conflict between international rules granting personal immunities and the cus­
tomary rules proscribing international crimes may be settled as indicated by the ICJ in
its judgment in the Case Concerning the Arrest Warrant of 11 April 2000 (§§51-7). The
Court logically inferred from the rationale behind the rules on personal immunities

14 See Krstic (Decision on application for subpoenas) (ICTY, AJ, §§23-8); Milosevic (Decision on applica­
tion for interview and testimony of Tony Blair and Gerhard Schröder), ICTY, AJ, §§12-33; and Norman and
others (Decision on interlocutory appeal against Trial Chamber decision refusing to subpoena the President of
Sierra Leone) (SCSL, AC, §§8-29). It should be noted that in the last two cases the court declined to issue the
subpoena only because it held that the testimony of the dignitaries at issue was not material to the defence
case.
 3 io INTERNATIONAL CRIMINAL LAW

of such senior state officials as Heads of State or government (plus foreign ministers
and diplomatic agents), that these immunities must perforce prevent any prejudice to
the effective performance’ of their functions. They therefore bar any possible interfer­
ence with the official activity of such officials. It follows that an incumbent senior state
agent (belonging to one of the categories mentioned above) is immune from jurisdic­
tion, even when he is on a private visit or acts in a private capacity while holding office.
Clearly, not only the arrest and prosecution of such a state agent while on a private visit
abroad, but also the mere issuing of an arrest warrant, may seriously hamper or jeop­
ardize the conduct of international affairs of the state for which that person acts.
In summary, even when accused of international crimes, the state agent entitled to
personal immunities is inviolable and immune from prosecution on the strength of
the international rules on such personal immunities. This proposition is supported by
some case law (for instance, Pinochet15 in the UK and Fidel Castro16 in Spain, which
relate to a former and an incumbent Head of State, respectively).
If the allegations about international crimes committed by foreign state officials are
known before they enter a foreign territory, the territorial state may ask the foreign state
official to refrain from setting foot in the territory; if that official is already on the ter­
ritory, the state may declare him persona non grata and request him to leave forthwith.
Of course, it may be that an international treaty on specific international crimes
implicitly or expressly prescribes that personal immunities may not relieve officials of
responsibility for the international crimes they envisage. Many treaty rules, although
couched in general terms, may be interpreted to this effect. On this score one can
mention the Genocide Convention of 1948 (Article IV), the 1984 Convention on
Torture (Article 4), as well as a num ber of treaties on terrorism. To these treaties one
should add the Statutes of the ICTY and ICTR. Both contain a provision (respectively,
Articles 7(2) and 6(2)), whereby ‘The official position of any accused person, whether
as Head of State or Government or as responsible Government official, shall not relieve
such person of criminal responsibility nor mitigate punishment.’ The strictness of this
provision can be construed to the effect that it rules out the possibility of invoking

15 See, e.g., the speech of Lord Browne-Wilkinson, in R. v. Bow Street Stipendiary Magistrate and others,
ex parte Pinochet Ugarte, judgment of 24 March 1999, at 112-15. See also the speeches of Lord Hope of
Craighead, at 145-52, Lord Saville of Newdigate, at 169-70, Lord Millet, at 171-91, and Lord Phillips of
Worth Matravers, at 181-90.
16 See Order {auto) of 4 March 1999 (no. 1999/2723). The Audiencia Nacional held that the Spanish Court
could not exercise its criminal jurisdiction, as provided for in Article 23 of the Law on the Judicial Power,
for the crimes attributed to Fidel Castro. He was an incumbent Head of State, and therefore the provisions
of Article 23 could not be applied to him because they were not applicable to Heads of State, ambassadors,
etc. in office, who thus enjoyed immunity from prosecution on the strength of international rules to which
Article 21(2) of the same Law referred (this provision envisages an exception to the exercise of Spanish juris­
diction in the case of ‘immunity from jurisdiction or execution provided for in rules of public international
law’); see Legal Grounds nos 1-4. The Court also stated that its legal finding was not inconsistent with its rul­
ing in Pinochet, because Pinochet was a former Head of State, and hence no longer enjoyed immunity from
jurisdiction (see Legal Ground no. 5). For the (Spanish) text of the order, see the CD-Rom, EL DERECHO,
2002, Criminal case law.
 IMMUNITIES 311

personal immunities as a legal ground for not being prosecuted or tried.17 The same
interpretation could be advanced with regard to the 1984 Convention on Torture,
Articles 1-4 of which are so strict as to warrant such interpretation. However, the only
treaty that explicitly excludes the right to rely upon personal immunities is the ICC
Statute (Article 27(2)).
Certainly, there is still resistance to this trend favourable to lifting personal immunities
in the case of international crimes. For example, in March 2000 the US State Department
allowed a Peruvian alleged torturer to go free on the grounds that he enjoyed personal
(that is, diplomatic) immunity.18
The question must nevertheless be raised as to whether a customary rule has evolved
in the international community removing personal immunities for alleged international
crimes, at least when jurisdiction over such crimes is granted to international criminal
courts orTribunals, This question is not only theoretical, but also has a practical dimen­
sion. For instance, the STT, unlike the Statutes of other international criminal courts
and tribunals referred to above, does not provide in terms for the lifting of the immunity
under discussion. Can we nevertheless hold the view that the Tribunal is not barred from
prosecuting and trying state officials enjoying personal immunities (including inviol­
ability and immunity from foreign criminal jurisdiction)? In other words, is a Head
of State, a prime minister, a foreign minister or a diplomat, charged by the Tribunal’s
Prosecutor with the crime of terrorism, precluded from claiming personal immunity?
It is submitted that the above question must be answered in the affirmative, on three
grounds.
First, the judgment of the ICJ on Arrest warrant does not exclude either explicitly or
implicitly that a customary rule on the matter has evolved with regard to international
criminal courts and tribunals. It held that ‘the immunities enjoyed under international

17 Therefore, it would seem that one ought to reject as unfounded the claim made by the Serbian author­
ities of the FRY that some of the co-accused of Mr Slobodan Milosevic, in particular the former foreign min­
ister of the FRY and incumbent president of Serbia, Mr M. Milutinovic, could not be arrested and handed
over to the ICTY because they enjoyed immunities under the national or federal Constitution. Assuming
this were correct under national law, the rules of the ICTY Statute would prevail, because those rules were
enacted by the Security Council under Chapter VII of the UN Charter, and therefore override contrary
treaties, customary rules, and also national legislation pursuant to Article 103 of the UN Charter.
18 In the above example, Major Tomas Ricardo Anderson Kohatsu, a retired official of Peru’s notorious
Army Intelligence Service, was alleged by the US State Department to have perpetrated ‘horrendous crimes’
in 1997. In early March 2000 the Peruvian authorities sent him to the US to appear before a hearing of the
Inter-American Commission on Human Rights in Washington. When he was about to leave the US to return
to Peru, FBI agents detained him, pursuant to the 1984 UN Convention against Torture, duly ratified by the
US. However, a few hours later he was released following a decision by the Under-Secretary of State, Thomas
Pickering. According to Pickering, Anderson was entitled to diplomatic immunity because he held a G-2
visa, granted to accredited members of the staff of the Peruvian Mission to the Organization of American
States. Consequently, he could not be arrested or prosecuted (on-line: at www.windos\temp\center for
constitutional rights.htm).
It was pointed out by M. Ratner, (US Center for Constitutional Rights), that Anderson had not in fact been
accredited to the Peruvian Mission. More importantly, the 1984 Convention on Torture does not permit
exemption for diplomatic immunity. In any case, it was for the US courts to determine the matter. As Ratner
pointed out, unlike Pinochet, ‘despite serious doubts as to Anderson’s claimed immunity, the decision to
allow him to return to Peru was made by the State Department and not the courts’ (see ibid., at 2, §3).
 312 INTERNATIONAL CRIMINAL LAW

law by an incumbent or former Minister for Foreign Affairs do not represent a bar to
criminal prosecution in certain circumstances’ (§61). It then enumerated among such
instances the case where ‘an incumbent or former Minister for Foreign Affairs may be
subject to criminal proceedings before certain international criminal courts, where
they have jurisdiction (ibid., emphasis added). The Court then mentioned the ICTY, the
ICTR, and the ICC, noting that the ICC Statute expressly provides, in Article 27(2), that
“[ijmmunities or special procedural rules which may attach to the official capacity of a
person, whether under national or international law, shall not bar the Court from exer­
cising its jurisdiction over such a person” ’ (ibid.). It is thus clear that the ICJ did not make
the lifting of personal immunities before international criminal courts contingent upon
the express or implicit contemplation of such lifting in the relevant court’s statute. It
instead held that the non-invocability of personal immunity before international courts
was admissible to the extent that the relevant court or tribunal had jurisdiction over the
international crime with which the state official at stake was charged.
Secondly, the rationale for foreign state officials being entitled to urge personal
immunities before national courts does not apply to international courts and tribu­
nals. That rationale resides in the need for foreign state officials not to be exposed to
the prosecution by national authorities that might use this means as a way of inter­
fering with the foreign state officials’ activity, thereby unduly impeding or limiting
their international action. In many states judicial authorities are not independent of
the political power; they could therefore decide to prosecute foreign state officials on
grounds that have little to do with their legal or illegal conduct and indeed amount to
a way of unduly interfering with the action of those state officials. This danger of abuse
does not arise instead with regard to international criminal courts and tribunals, which
are totally independent of states and subject to strict rules of impartiality. In addition,
these courts and tribunals are much better equipped than national courts to deal with
international crimes, because they are ‘specialized’ in this area and their judges are
selected on account of their particular competence or experience in the matter.
Thirdly, the current thrust of international law is to broaden as much as possible
the protection of hum an rights and, by the same token, to make those who engage
in heinous breaches of such rights criminally accountable. The very logic of the pre­
sent trends of international law therefore fully warrants the subjection of state officials
to the judicial scrutiny of international independent bodies, whenever such officials
(i) are accused of serious criminal offences against basic values of the world commu­
nity; and (ii) there is no risk that such judicial scrutiny be surreptitiously used as a
means of unduly restraining the official activity of the state agent concerned.
In summary, it seems justified to hold that under customary international law
personal immunities of state officials may not bar international criminal courts and
tribunals from prosecuting and trying persons suspected or accused of having com­
m itted international crimes, or at any rate the criminal offences over which the rele­
vant international court or tribunal has jurisdiction.
All this applies to incumbent senior state officials. As soon as the state agent leaves
office, he may no longer enjoy personal immunities and, in addition, becomes liable
 IMMUNITIES 313

to prosecution for any international crime he may have perpetrated while in office (or
before taking office), pursuant to the aforementioned customary rule lifting functional
immunities in the case of international crimes.

(B) D O T H E Y E X E M P T S E N I O R S T A T E O F F I C I A L S F R O M T H E D U T Y
T O T E S T I F Y O R H A N D O V E R E V ID E N C E ?

We must now briefly discuss the question of whether an incumbent senior state official
belonging to one of the four abovementioned categories is entitled to invoke personal
immunity in order to refuse either to testify before an international court or tribunal
dealing with international crimes, or to hand over documents needed by the court or
tribunal.
Plausibly, a Head of State may not be compelled to testify before a foreign court,
not even with regard to an international crime: an order to testify issued by a national
court to a foreign Head of State (or prime m inister or foreign minister or diplomat)
would run counter to international rules protecting personal immunities (as for the
rationale behind this legal regulation, see above). Arguably here traditional notions
relating to state sovereignty still apply and have not yet been set aside by the demands
of international justice.
Does the same hold true for orders issued by international courts exercising jur­
isdiction over international crimes? It would seem that the rationale applying to the
lifting of personal immunity from prosecution for those crimes, mentioned above,
should also apply to the right of one of those senior officials to refrain from testifying;
it follows that such right may not be invoked. Here the param ount demands of inter­
national justice, together with the absence of any possible risk that the international
court may interfere with the state agent activity or abuse its powers, override the rights
of senior state officials deriving from traditional notions of respect (by other states or
state organs) for their sovereign prerogatives. It follows that an international criminal
tribunal is empowered to compel a senior state official (belonging to one of the four
categories) to testify (subpoena ad testificandum) or to hand over im portant docu­
ments (subpoena duce tecum).
Interestingly, the ICTY AC in Krstic (Decision on application for subpoenas decision)
(§27) and anICTRTC in Bagosora (Decision on requestfor a subpoena for Major}. Biot,
at §4) affirihed this authority of international criminal courts, stating that they may
compel senior state agents to testify, whether or not such agents witnessed the rele­
vant facts in their official capacity. Other courts have in fact eschewed pronouncing
on the merits of this matter. In Fofana and others, in 2006 a SCSL TC did not grant a
request to issue a subpoena ad testificandum against the incumbent President of Sierra
Leone, for it found that the requirements set out in Rule 54 (on the power to issue such
orders ‘as may be necessary for the purposes of an investigation or for the preparation
or conduct of the trial’) were not met in the case at issue. The AC upheld the decision
(§§8-39). However, a member of the TC, Judge T. Thompson, issued a forceful opinion
clearly showing that the law allowed the issuance of the order (Dissenting Opinion,
3H INTERNATIONAL CRIMINAL LAW

§§14-30); similarly, in the AC Judge Robertson appended an opinion along the same
lines, providing reference to previous case law (Dissenting Opinion, §§10-50). In
Milosevic Slobodan (Decision of 9 December 2005) the AC declined to call to testify
the incumbent British prime minister.

14.5 NATIONAL PERSONAL IM M U N IT IE S

The question of whether a national court is authorized to start proceedings against

a national accused of international crimes, who happens to be a senior state official
enjoying immunities under national law (for instance, the Head of State, a member of
cabinet, a member of parliament) must be looked at from the viewpoint of international
and national law.
Customary international law, it would seem, does not contain any rule imposing
upon a state the obligation to disregard national legislation on immunities. However,
treaty rules may impose the obligation to punish the authors of international crimes.
If this is the case, any national legislation granting im munity would be in conflict with
the treaty obligation.
National law may contain general rules granting immunity from prosecution for
any crime, including international crimes. It very much depends on each particular
national system. However, after the entry into force of the ICC Statute, those states that
are gradually ratifying such Statute are no longer allowed to rely upon any possible
national legislation on immunities. The national implementation of the ICC Statute
requires that states change their legislation (including their constitutional provisions,
if any) on immunities, removing any such immunities for the international crimes
that fall under the jurisdiction of the Court.
 PART III

PROSECUTION
AND PUNISHMENT
BY INTERNATIONAL
COURTS
 15
THE ESTABLISHMENT OF
INTERNATIONAL CRIMINAL
TRIBUNALS

The idea of setting up an international criminal court goes back to the aftermath of the
First World War. The attainment of that goal has been slow and painstaking. The pro­
cess toward the eventual adoption of a Statute for a permanent International Criminal
Court and the adoption of statutes of various mixed or hybrid (that is, composed both
of national and international judges) criminal tribunals, can be conceptualized in
terms of various distinct phases: (i) abortive early attempts (1919-45); (ii) criminal
prosecutions in the aftermath of the Second World War: the Nuremberg and Tokyo
Tribunals (1945-47); (iii) elaboration by the ILC of the Statute of a permanent Court;
(iv) the post-Cold War ‘new world order’: the development of the two ad hoc Tribunals
(1993-94); (v) the drafting of the ICC Statute (1994-98); (vi) the adoption of statutes of
ad hoc hybrid criminal tribunals.

15.1 ABORTIVE EARLY ATTEMPTS ( 1919 - 1945)

The period immediately following the First World War is notable for numerous
attempts to establish a variety of international criminal institutions, all of which
ended in failure. For instance, in 1919 the ‘Commission on the Responsibility of the
Authors of the War and on Enforcement of Penalties’ proposed the establishment of
a ‘high tribunal composed of judges drawn from many nations’.1 In the same year the
victors had agreed upon a few provisions of the peace treaty with Germany, signed at
Versailles, which provided for the punishm ent of the leading figures responsible for
war crimes committed during the war and went so far as to lay down in Article 227 the
responsibility of the German Emperor (Wilhelm II) for ‘the supreme offence against
international morality and the sanctity of treaties’. The same provision envisaged the
establishment of ‘a special tribunal’, composed of five judges (to be appointed by the

1 See the Report of the Commission, in 14 AJIL (1920), at 116. As for the objections of the US delegates,
 3 i8 INTERNATIONAL CRIMINAL LAW

USA, Great Britain, France, Italy, and Japan) and charged with trying the Emperor.
The Allies were clearly motivated by their outrage at the atrocities perpetrated by the
vanquished Powers, in particular Germany, and wished to set an example. However,
the accused would have been judged by their erstwhile opponents; this would have
thrown doubt on the fairness of the proceedings and the impartiality of the tribunal.
In any case, the Netherlands, where the German Emperor had taken refuge, refused to
extradite him, chiefly because the crimes of which he was accused were not contem­
plated in the Dutch Constitution.2 In addition, the aforementioned provisions of the
Versailles Treaty were harshly criticized by some eminent publicists, among them the
Italian leading jurist and politician V. E. Orlando.3
As for the trials of German m ilitary personnel alleged to have committed war crimes,
no international court was set up, nor were they tried by courts of the Allies, as had
been envisaged in Articles 228-30 of the Versailles Treaty. Eventually, out of the 895
Germans accused (who comprised various generals and admirals including the Chief
of Staff of the Army, General E. Ludendorff, General Paul von Hindenburg, later Chief
of Staff of the Army, as well as the former Chancellor Bethmann-Hollweg), the Allies
selected only 45 cases for prosecution.4 Ultimately 12 m inor indictees were brought to
trial in 1921, and before a German court, the ‘Imperial Court of Justice’ (Reichsgericht,
sitting at Leipzig). Six of the 12 indictees were acquitted. Thus, the attempts to establish
some sort of international criminal justice ended in failure. However, some of the
judgments delivered by the Leipzig C ourt set significant precedents, chiefly because of
the high legal quality of those judgments. The attempts to bring to justice the ‘Young
Turks’ responsible for the massacres of the Armenians in 1915-16 were generally
partial failures; some Extraordinary Courts M artial in Istanbul brought to trial a few
m inor accused (plus major defendants, but in absentia).5

2 On the non-implementation of Article 227, see, inter alia, A. Merignhac and E. Lemonon, Le Droit des
gens et la guerre de 1914-1918, II (Paris: Pedone, 1921), 580ff.
The Dutch diplomatic note of 21 January 1920 to the Allies stated that Or, ni les lois Constituantes du
Royaume qui sont basees sur des principes de droit universellement reconnus, ni une respectable tradition
seculaire qui a fait de ce pays de tout temps une terre de refuge pour les vaincus des conflits internationaux,
ne permettent au Gouvernement des Pays-Bas de deferer au desir des Puissances en retirant a l’ex-empereur
le benefice de ces lois et cette tradition’ (see the text of the Dutch diplomatic notes in A. Merignhac, ‘De la
responsabilite penale des actes criminels commis au cours de la guerre 1914-1918’ in 47 Revue de droit inter­
national et de legislation comparee (1920), 37-45. According to a distinguished author, B. Swart, ‘A rrest and
Surrender’ in Cassese, Gaeta, and Jones (eds), ICC Commentary, II, at 1643, ‘Given the fact that the former
Article 4 of the Dutch Constitution permitted extradition on the basis of a treaty only, that the acts alleged
did not constitute criminal offences according to Dutch law or to extradition treaties concluded with the
Allied and Associated Powers, and that the Constitution did not permit the conclusion of an extradition
treaty for the surrender of one person only, it is hard to see that the Dutch government could have reacted
in a different way.’
3 V. E. Orlando, Tl processo del Kaiser’ (1937), reprinted in Scritti varii di diritto pubblico e scienza polit­
ico (Milan: Giuffre, 1940), 97ff. For an English translation of this paper see 5 JICJ (2007), at n. 1015-1028.
4 See C. Mullin, The Leipzig Trials—An Account of the War Criminals’ Trials and a Study of German
Mentality (London: Witherby, 1921), at 27.
5 See, in particular, Rental and Tevfik (at 1-7), Bahäeddin §akir (at 1-8), Mehmed ‘A ll Bey and others
(at 177-84), Said Halim Pa?a and others (at 353-64) as well as the other cases cited in 1.5. More generally,
see T. Akcam, Armenien und der Völkermord: die Istanbuler Prozesse und die Türkische Nationalbewegung
 THE ESTABLISHMENT OF INTERNATIONAL CRIMINAL TRIBUNALS 319

In 1920, the ‘Advisory Committee of Jurists’, summoned to prepare the project

for the Permanent Court of International Justice, proposed that the ‘High C ourt of
International Justice’ to be established should also ‘be competent to try crimes con­
stituting a breach of international public order or against the universal law of nations,
referred to it by the Assembly or by the Council of the League of Nations’.6 However,
a few months later the Assembly of the League of Nations rejected the proposal out of
hand as being ‘premature’.7Thereafter, draff statutes of an international criminal court
were adopted by non-governmental organizations such as the Inter-Parliamentary
Union, in 1925,8 and by scholarly bodies such as the International Law Association, in
1926.9 None of these draffs, however, led to anything concrete.10
Such early attempts were laudable for their far-sighted recognition of the need for an
international organ of criminal jurisdiction. Nevertheless, these initiatives could not
bear fruit in a period which placed an exceptionally high premium upon considerations
of national sovereignty. Although new values had emerged which transcended narrow
nationalistic concerns (such as the gradual elaboration of principles seeking to limit
the methods of warfare, or the protection of workers through the establishment of the
International Labour Organization, or the protection of minorities through the numer­
ous treaties entered into after the First World War), state sovereignty was nevertheless
still very much the bedrock of the international community. The practical import of this
was that no feasible mechanism could be brought into being enabling a state official—let
alone a Head of State—accused of war crimes or other outrages to be tried.

15.2 THE NUREMBERG AND TOKYO

TRIBUNALS (1945-1947)
It was nevertheless this scenario (an international community dominated by state
sovereignty) that led to the successful establishment, in the immediate post-war
period, of the Nuremberg and Tokyo Tribunals. These Tribunals were a response to the

(Hamburg: Hamburger Edition, 1996), 192-207, 353-64; V. N. Dadrian, ‘The Documentation of the World
War I Armenian Massacres in the proceedings of the Turkish Military Tribunal’, in 23 International Journal
of Middle East Studies (1991), 549-76; V. N. Dadrian, ‘The Turkish Military Tribunal’s Prosecution of the
Authors of the Armenian Genocide: Four Major Court-Martial Series, in 7 Holocaust and Genocide Studies
(Spring 1997), 28-59. See also G. Lewy, ‘Revisiting the Armenian Genocide, in Middle East Quarterly (2005),
online: www.meforum.org/article/748.
6 See the text of the Second Resolution adopted by the Advisory Committee in Lord Phillimore, ‘An
International Criminal Court and the Resolutions of the Committee of Jurists’, 3 BYBIL (1922-3), 80.
7 Ibid., at 84.
8 See the text of the draft in B. Ferencz (ed.), An International Criminal Court—A Step Toward World
Peace—A Documentary History and Analysis, vol. I (London, Rome, New York: Oceana, 1980), 244ff.
9 Text reproduced in Ferencz, ibid., at 252fF.
10 A Convention for the creation of an International Criminal Court to try terrorist offences was
also adopted on 16 November 1937 by the League of Nations, but never entered into force. See generally
V. V. Pella, ‘Towards an International Criminal Court’, 44 AJIL (1950), 37-68.
 320 INTERNATIONAL CRIMINAL LAW

overwhelming horrors of the Nazi genocide in Europe and the Japanese crimes perpe­
trated during the wartime occupation of large parts of many South East Asian nations
(for instance, the so-called rape of Nanking, biological experiments in Manchuria,
the fall of Singapore and the extensive loss of life there, and other crimes). It took the
full extent of the atrocities committed during the war to demonstrate the pernicious
consequences that could follow from the pursuit of extreme notions of state sover­
eignty and to jolt the international community out of its complacency. The conviction
gradually emerged that tyranny and the attendant disregard for hum an dignity could
no longer be allowed to go unchecked and unpunished.
It is worthwhile to consider what, in particular, induced the Allies to hold trials
of the Germans and their collaborators after the Second World War and what, more
recently, has persuaded governments to hold similar trials for war crimes and crimes
against humanity.
After the defeat of Germany, the British, led by Churchill, stated that it was enough
to arrest and hang those prim arily responsible for determining and applying Nazi pol­
icy, without wasting time on legal procedures; m inor criminals, they suggested, could
be tried by specially created tribunals.11 However, neither President F. D. Roosevelt,
nor Henry Stimson, the US Defense Secretary, agreed; nor, indeed, did Stalin. In the
end, they prevailed, and the International Military Tribunal was set up in Nuremberg
to try the ‘great Nazi crim inals’, while lesser Allied tribunals in the four occupied
zones of Germany were to deal with m inor criminals. The Americans advanced vari­
ous arguments to support their view, later accepted by the other Allies.
First, how could a defeated enemy be condemned without due process of law? To
hang them without trial would mean to do away with one of the mainstays of democ­
racy: no one can be considered guilty until his crimes have been proved in a fair trial.
To relinquish such a fundamental principle would have put the Allies on a par with the
Nazis who had ridden roughshod over so many principles of justice and civilization,
when they had held mock trials, or punished those allegedly guilty without even the
benefit of judicial process.
Secondly, those who set up the Nuremberg Tribunal felt that the dramatic rehearsal
of Nazi crimes—and of racism and totalitarianism —would make a deep impression
on world opinion. Thus, the trial was designed to render tragic historical phenomena
plainly visible.
The third reason was a desire on the part of the Allied powers to act for posterity.
The crimes committed by the Third Reich and its Nazi officials were so appalling
that some detailed record had to be left. A trial held on a grand scale would allow the
Tribunal to assemble a massive archive useful not only in court, but also to historians
and to the generations to come. The trial would also serve as a lesson in history for future
generations.

11 See F. Smith (ed.), The American Road to Nuremberg: the Documentary Record, 1944-1945 (Stanford,
Cal.: Hoover Institution Press, 1982), 31-3,155-7.
 THE ESTABLISHMENT OF INTERNATIONAL CRIMINAL TRIBUNALS 321

In addition, for the Americans there was a particular motivation behind the estab­
lishment of an international tribunal. It was eloquently set forth by Justice Robert
H. Jackson (the special representative of the US President to the London Conference
and later the US Chief Prosecutor at Nuremberg) in 1945, when he stressed that the
trial would have rendered visible and indeed ‘authenticated’ in the USA, a country not
devastated by war, the Nazi crimes.12
A further rationale for the Nuremberg trial was the collective character of the Nazi
crimes. The massacre of civilians and prisoners of war, the persecution of Jews, gyp­
sies, and political opponents were not only large-scale phenomena but, in addition,
indicative of a policy pursued assiduously by the highest echelons of the Nazis and
applied by the whole military and bureaucratic apparatus. The crimes commissioned
by the directives of the Nazi leaders belonged to collective or system criminality’: such
was their nature that it would have been impossible to punish them by using the courts
of the state to which the perpetrators belonged. In consequence, and as mentioned
above, only an adversary (together with neutral states, as had been suggested)13 could
have made sure that justice was done, upon winning the war.
In the summer of 1945, the ‘Big Four’ (the United Kingdom, France, the United
States, and the Soviet Union) convened the London Conference to decide by what
means the world was to punish the high-ranking Nazi war criminals. The resultant
Nuremberg Charter established the IMT to prosecute individuals for ‘crimes against
peace’, ‘war crimes’, and ‘crimes against hum anity’. The IMT met from 14 November
1945 to 1 October 1946. In addition, in occupied Germany, the four major Allies, pur­
suant to Control Council Law no. 10, prosecuted through their own courts sitting
in Germany, in their respective zones of occupation, the same crimes committed by
lower-ranking defendants.
On 26 July 1945, two weeks before the conclusion of the London Conference, the
‘Big Four’ issued the Potsdam Declaration announcing, to the surprise of many, their

12 In a Memorandum he submitted on 30 June 1945, together with a Redraft of the US proposals for the
new International Tribunal, to the representatives of the UK, France, and the Soviet Union participating in
the London Conference on military trials, he wrote the following: "The Unites States [...] has conceived of
this case as a broad one. It must be borne in mind that Russian, French, English and other European peoples
are familiar with the Hitlerite atrocities and oppressions at first-hand. Our country, three thousand miles
away, has known of them chiefly through the press and radio and through the accusations of those who have
suffered rather than through immediate experience. German atrocities in the last war were charged. The
public of my country was disillusioned because most of these charges were never authenticated by trial and
conviction. If there is to be continuing support in the United States for international measures to prevent
the regrowth of Nazism, it is necessary now to authenticate, by methods which the American people will
regard as of the highest accuracy, the whole history of this Nazi movement, including its extermination of
minorities, its aggressions against neighbors, its treachery and its barbarism’ (International Conference on
Military Trials, at 126).
13 See, for instance, C. C. Hyde, ‘Punishment of War Criminals’, Proceedings of the American Society of
International Law, Thirty-Seventh Annual Meeting, 1943 (1943), 43-4. See also H.Kelsen, Peace Through
Law (Chapel Hill: University of North Carolina Press, 1944), at 111; H. Kelsen, ‘Will the Judgment in the
Nuremberg Trial Constitute a Precedent in International Law?’, 1 International Law Quarterly (1947),
at 170.
 322 INTERNATIONAL CRIMINAL LAW

intention to prosecute leading Japanese officials for these same crimes.14Subsequently,

on 19 January 1946, General Douglas MacArthur, Supreme Commander for the Allied
Powers in Japan, approved, in the form of an executive order, the Tokyo Charter, set­
ting forth the constitution, jurisdiction, and functions of the International Military
Tribunal for the Far East (IMTFE). Like the Nuremberg Charter, the Tokyo Charter,
which was issued on 26 April 1946, included the newly articulated crimes against
peace and humanity.15
By and large, the Tokyo Charter was modelled on the Nuremberg Charter. However,
there were some differences between the two texts and the way they regulated the
structure of the Tribunals and the charges that could be brought against the defend­
ants.16 It is also notable that the bench comprised persons from newly independent
countries, such as India and the Philippines.
The Tokyo Trial (which commenced on 3 May 1946, and lasted for approximately
two-and-a-half years) was the source of much controversy both during and after the
event. Some have claimed that the trial was either a vehicle for America’s revenge for
the treacherous attack on Pearl Harbor, or a means of assuaging American national
guilt over the use of atomic weapons in Japan. Others, defence counsel at the trial
included, attacked the trial’s legitimacy on legal grounds.17
Whereas the post-First World War experience showed the extent to which inter­
national justice can be compromised for the sake of political expedience, the post-Sec-
ond World War experience revealed, conversely, how effective ‘international’ justice
could be when there is political will to support it as well as the necessary resources.
These sets of experiences were nevertheless one-sided, as everybody knows. They
imposed Victors’ justice’ over the defeated. The major drawback of the two ‘inter­
national’ Tribunals was that they were composed of judges (4 and 11, respectively)
appointed by each of the victor Powers; the prosecutors too were appointed by each of
those Powers and acted under the instructions of each appointing state (at Tokyo there
was a chief prosecutor, or ‘Chief of Counsel’ as he was called, namely the American
Joseph B. Keenan, and 10 associate prosecutors). Thus, the view must be shared that
the two Tribunals were not independent international courts proper, but judicial bod­
ies acting as organs common to the appointing states.18

14 Some of the Allies in the Pacific theatre prosecuted the Japanese for ‘war crimes’ under their respect-
ive military laws: see, inter alia, R. John Pritchard, ‘War Crimes Trials in the Far East,’ in R. Bowring and
P. Kornick (eds), Cambridge Encyclopedia of Japan (Cambridge: Cambridge University Press, 1993), 107.
15 The Charter had been drafted by the Americans only, essentially by Joseph B. Keenan, Chief Prosecutor
at the Tokyo Trial, and the other Allies were only consulted after it was issued: B. V. A. Roling and A. Cassese,
The Tokyo Trial and Beyond (Cambridge: Polity Press, 1993), 2.
16 For a summary of the principal differences see ibid., 2-3.
17 For instance, the legal categories of the crimes against peace and humanity have been criticized as ex
postfacto legislation, in that these crimes did not exist in international law prior to 1945 (ibid., 3-5).
18 The Nuremberg IMT admitted this legal reality when it stated that ‘The making of the Charter [of the
1MT] was the exercise of the sovereign legislative power by the countries to which the German Reich uncon­
ditionally surrendered; and the undoubted right of these countries to legislate for the occupied territories
has been recognised by the civilised world [...] The Signatory Powers created this Tribunal, defined the law
it was to administer, and made regulations for the proper conduct of the Trial. In doing so, they have done
 THE ESTABLISHMENT OF INTERNATIONAL CRIMINAL TRIBUNALS 323

However, the IMTs were im portant in many respects. First, they broke the ‘m onop­
oly’ over criminal jurisdiction concerning such international crimes as war crimes,
until that moment firmly held by states. For the first time non-national, or m ulti­
national, institutions were established for the purpose of prosecuting and punishing
crimes having an international dimension and scope.
Secondly, new offences were envisaged in the London Agreement and made punish­
able: crimes against hum anity and crimes against peace. W hether or not this was done
in breach of the principle of nullum crimen sine lege, it is a fact that since 1945 those
crimes gradually became the subject of international customary law prohibitions.
Thirdly, while until that time only servicemen and m inor officers had been pros­
ecuted, now for the first time military leaders as well as high-ranking politicians and
other civilians were brought to trial.
Fourthly, the statutes and the case law of the IMT and the IMTFE and the various
tribunals set up by the Allies in the aftermath of the Second World War contributed to
the development of new legal norms and standards of responsibility, by providing, for
example, for the elimination of the defence of obedience to superior orders’. Finally, a
symbolic significance emerged from these experiences in terms of their moral legacy.19

15.3 THE WORK OF THE ILC ( 195O -I 954)

In order to build on the positive dimensions of the establishment of the IMT and
IMTFE, and perhaps stung by the inevitable association of these Tribunals with
‘victor’s justice’, the United Nations system in the late 1940s commenced its quest
to establish more permanent and impartial mechanisms for dispensing international
criminal justice.
The efforts of the UN in this respect can be traced along two separate tracks:
codification of international crimes and the elaboration of a draff statute for the
establishment of an international court.
Pursuant to a request by the General Assembly on 21 November 1947 (res. 177/
II), the International Law Commission (ILC) commenced the formulation of the
principles recognized in the Charter of the Nuremberg Tribunal, to prepare a draff
code of offences against the peace and security of mankind. Concurrently, the task of
formulating a draff statute for the establishment of an international criminal court
was assigned to another special rapporteur, who submitted his first report to the ILC
in March 1950.20

together what any one of them might have done singly; for it is not to be doubted that any nation has the right
thus to set up special courts to administer law’ (at 218).
19 M. Lippman, ‘Nuremberg: Forty-Five Years Later’, 7 Conn. ]. Int. L. (1991), 1.
20 Report of the International Law Commission on the Question of International Criminal Jurisdiction,
UNGAOR, 5th Sess., UN Doc. A/CN.4/15 (1950).

j
 324 INTERNATIONAL CRIMINAL LAW

The late 1940s and the 1950s were characterized by much work by a variety of inter­
national bodies on tasks that, while designed to be complementary and interlocking,
were nevertheless poorly co-ordinated. The ILC special committee charged with pre­
paring a draft statute for an international criminal court produced a text in 1951 that
was revised in 1953.21 However, neither the early discussions in the Commission, nor
the provisions of Article VI of the 1948 Genocide Convention referring to a (future)
‘international penal tribunal’ were translated into reality. The 1953 Draff Statute of the
Court was shelved because the definition of aggression, which had been entrusted to
another body, was not completed. That result was expected since there were differing
bodies working separately at different venues (Geneva and New York), and producing
different texts at different times. It was, therefore, easy for the GA to put off discussion
of each text successively because the one or the other was not then ready. The lack of
synchronization was not entirely fortuitous; it was the result of a political will to delay
the establishment of an international criminal court, due to the fact that the world was
then sharply divided and frequently at risk of war.22

15.4 THE PO ST-C O LD WAR NEW W ORLD ORDER

AND TH E ESTABLISHM EN T OF AD HOC
TRIBUNA LS ( 1 9 9 3 - 1 9 9 4 )

15.4.1 GENERAL

Various factors led to the establishment of international criminal tribunals in the

early 1990s.
The end of the Cold War proved to be of crucial importance. It had significant effects.
For one thing, the animosity that had dominated international relations for almost half
a century dissipated. In its wake, a new spirit of relative optimism emerged, stimulated
by the following factors: (i) a clear reduction in the distrust and m utual suspicion
that had frustrated friendly relations and co-operation between the Western and the
Eastern bloc; (ii) the successor states to the USSR (the Russian Federation and the
other members of the Confederation of Independent States) came to uphold greater
respect for international law; (iii) as a result there emerged unprecedented agreement
in the UN SC and increasing convergence in the views of the five permanent members,
with the consequence that this institution became able to fulfil its functions more
effectively.

21 Report of the Committee on International Criminal Jurisdiction, UNGAOR, 7th Sess., Supp. No. 12
at 21, UN Doc. A/26645 (1954).
22 M. C. Bassiouni, The Statute of the International Criminal Court—A Documentary History (Ardsley,
NY: Transnational Publishers, 1998), 13-15.
 T H E E S TA B LISH M E N T OF IN T E R N A T IO N A L C R IM IN A L T R IB U N A L S 3^5

Another effect of the end of the Cold War was no less im portant. Despite the prob­
lems o fthat bleak period, during the Cold War era the two power blocs had managed
to guarantee a modicum of international order, in that each of the Superpowers had
acted as a sort of policeman and guarantor in its respective sphere of influence. The
collapse of this model of international relations ushered in a wave of negative con­
sequences. It entailed a fragmentation of the international com m unity and intense
disorder which, coupled with rising nationalism and fundamentalism, resulted in a
spiralling of mostly internal armed conflicts, with much bloodshed and cruelty. The
ensuing implosion of previously multi-ethnic societies led to gross violations of inter­
national hum anitarian law on a scale comparable in some respects to those committed
during the Second World War.
A further crucial factor contributing to an enlarged need for international crim ­
inal justice was the increasing importance of the hum an rights doctrine, which soon
became a sort of ‘secular’ religion. As the few available international mechanisms for
monitoring respect for hum an rights had proved deficient, the notion gradually took
hold that the best way of ensuring compliance with those rights was to prosecute and
punish those individually responsible for their breach. This begot the quest for, or at
least gave a robust impulse to, international criminal justice.
This period is thus characterized by the development of institutions empowered to
prosecute and punish serious violations of international hum anitarian law.

15.4.2 T H E T W O A D H O C T R I B U N A L S F O R
Y U GO SLA V IA A N D RW ANDA

The conflicts which erupted in, amongst other places, the former Yugoslavia and
Rwanda served to rekindle the sense of outrage felt at the closing stage of the Second
World War.23 Thus, the UN SC set up ad hoc Tribunals pursuant to its power to decide
on measures necessary to m aintain or restore international peace and security: in
1993 the International Criminal Tribunal for the former Yugoslavia (ICTY), and in
1994 the International Criminal Tribunal for Rwanda (ICTR).
The former was empowered to exercise jurisdiction over grave breaches of the
Geneva Conventions, violations of the laws and customs of war, genocide, and crimes
against hum anity allegedly perpetrated in the former Yugoslavia since 1 January 1991.
The latter was called upon to adjudicate genocide, crimes against humanity, and viola­
tions of Article 3 common to the Geneva Conventions and of the Second Additional
Protocol, allegedly perpetrated in Rwanda (or in ‘the territory of neighbouring states
in respect of serious violations of international hum anitarian law committed by
Rwandan citizens’) between 1 January and 31 December 1994.

23 See, for example, the letters to A. Cassese of Lawrence Eagleburger of 8 May 1996 (‘the United States
could no longer remain silent on the issue of war crimes [... A] cts against humanity could not and would not
be ignored’) and Elie Wiesel of 28 June 1996 (not to prosecute the criminals would amount to condoning
their crimes. In extreme situations, speaking out is a moral obligation) reprinted in The Path to the Hague:
Selected Documents on the Origins of the ICTY (UN: ICTY, 1996), at 89 and 91, respectively.
326 IN T E R N A T IO N A L C R IM IN A L LAW

The response of the international community to the conflict in Yugoslavia had been
tardy and lukewarm, due to impotence at the military and political levels. The estab­
lishment of a Tribunal was thus seized upon during the conflict not only as a belated
face-saving measure but also in the pious hope that it would serve as a deterrent to
further crimes.24 As the UN SC itself noted, the ICTY was established in the belief that
an international tribunal would contribute to ensuring that such violations are halted
and effectively redressed’.25
The SC established the ICTY in its Resolution 827 of 25 May 1993.26 A striking fea­
ture of this Resolution was that the SC determined that the situation in the former
Yugoslavia, and in particular in Bosnia and Herzegovina—where there were ‘reports
of mass killings, massive, organised and systematic detention and rape of women and
[... ] the practice of “ethnic cleansing” ’—constituted a threat to international peace
and security under Chapter VII of the UN Charter.27
The setting up of the ICTY gave rise to many objections.28 In brief, the principal
criticisms were that: (i) the Tribunal was established to make up for the impotence
of diplomacy and politics; (ii) by establishing the Tribunal the SC exceeded its pow­
ers under the Charter, adopting an act that was patently ultra vires; (iii) by the same
token, by creating a criminal court dealing only with crimes allegedly committed in a

24 See in this regard the letter of Lawrence Eagleburger of 8 May 1996 to A. Cassese: ‘There can be—and
are—arguments about the wisdom of external armed intervention in the tragedy that is Bosnia [...] Of far
greater precedential significance is the UN’s decision to try accused war criminals before an International
Tribunal especially created for that purpose [... T]hese trials will serve to put potential future war criminals
on notice that the international community will not tolerate crimes against humanity’ (ibid.).
25 See in this regard UNSC Resolution 827 of 25 May 1993.
26 The resolution was adopted following consideration of the Secretary-General’s Report (S/25704,3 May
1993), submitted pursuant to Security Council Resolution 808. The Secretary-General’s Report proposed a
Statute for the ICTY, which was unanimously adopted without amendment.
In terms of the drafting of the Statute of the ICTY, it appears that the first draft was prepared by a group
of three rapporteurs appointed by the Conference on Security and Cooperation in Europe (CSCE). In a
letter of 24 November 1992 the British Government, then holding the Presidency of the European Union,
proposed ‘to draft a convention establishing an ad hoc tribunal to deal with war crimes and crimes against
humanity committed in the former Yugoslavia’. The Ministers of Foreign Affairs of the CSCE, meeting in the
CSCE Council, responded favourably on 15 December 1992. The three rapporteurs then produced a draft on
9 February 1993. On 16 January 1993 the French Foreign Minister, Roland Dumas, appointed a Commission
of Experts with the task of drafting a statute of an ad hoc international tribunal. Various drafts were subse­
quently submitted by a number of states and international bodies to the UN Secretary-General and used by
him in his drafting of the Statute of the ICTY after the Security Council, at the proposal of France, adopted
on 22 February 1993 Resolution 808 (1993), by which it decided to establish an international Tribunal (The
Path to the Hague, cit., at 13).
27 In operative paragraph 2 of Resolution 827 of 25 May 1993, the Security Council decided 'to establish
an international tribunal for the sole purpose of prosecuting persons responsible for serious violations of
international humanitarian law committed in the territory of the former Yugoslavia between 1 January 1991
and a date to be determined by the Security Council upon the restoration of peace and to this end to adopt
the Statute of the International Tribunal annexed to the above-mentioned [Secretary-General’s] report’. The
Security Council amended the ICTY Statute by resolution 1166 (1998) on 13 May 1998 to add a third Trial
Chamber and three new judges. Likewise, the Statute of the ICTR was amended by the Security Council in
its Resolution 1165 of 30 April 1998 to provide for a third Trial Chamber.
28 See, in particular, G. Robertson, Crimes against Humanity—The Struggle for Global Justice (London:
Penguin, 2000), 300ff.
 T H E ESTA B LISH M E NT OF IN T E R N A T IO N A L C R IM IN A L T R IB U N A L S 327

particular country, instead of granting to the new court jurisdiction over crimes com­
mitted everywhere in the world, the SC had opted for ‘selective justice’.
The first criticism is right. However, half a loaf is better than pie in the sky. As long as
an international criminal court endowed with universal jurisdiction was lacking, the
establishment of ad hoc tribunals proved salutary. The Tribunal’s Appeals Chamber in
Tadic (IA) proved the second criticism to be wrong (see §§9-40 of that judgment). As
for the attack on ‘selective justice’, one could answer that such justice, however objec­
tionable, is better than no justice at all.
The ICTR was established in like fashion to the ICTY in response to the civil war
and genocide in Rwanda. While many of the factors mentioned above with regard
to the former Yugoslavia were also motivations for the establishment of the ICTR,
the overwhelming magnitude of the crimes committed there and the fact that they
assuredly amounted to genocide lent particular urgency to the establishment of the
ICTR. Sensitive to criticism that the establishment of the ICTY represented yet another
illustration of the disproportionate attention paid to the problems of Europe vis-a-vis
the developing world, the international community was also anxious to establish a
Tribunal for Rwanda so as to assuage its conscience and shield itself from accusa­
tions of double standards. An additional feature leading up to the establishment of the
ICTR was that, in the early stages at least, the proposal to establish an international
Tribunal was an initiative of the new Rwandan government. As they set about their
task of post-war reconstruction, the new government had initially felt that one means
of attracting international blessing for the new regime would be through a national
process of self-examination and international judicial condemnation of the worst
abuses that had occurred during the civil war.29
The SC adopted the Statute and judicial mechanism for the Rwanda Tribunal by
SC Resolution 955 of 8 November 1994, after having determined that ‘this situation
continues to constitute a threat to international peace and security’.30
Even though the Statutes for the ICTY and the ICTR differ, the Tribunals share an
Appellate Chamber and (at least initially) a Prosecutor. This may appear to be a curious
formula for distinct ad hoc Tribunals; but it demonstrates the need for ensuring some
uniformity in administering international criminal justice.

29 In July 1994, the Security Council passed Resolution 935, using the precedent of the former Yugoslavia
as a model, to establish a commission of experts to investigate violations committed during the Rwandan
civil war (see SC Res. 935, UNSCOR, 49th Sess., 3400th mtg 1, UN Doc. S/RES/935 (1994)). The Rwandan
commission lasted only four months, which was not long enough for it to perform its task effectively. On
1 October 1994, the Rwandan commission submitted its preliminary report to the Secretary-General,
and a final report on 9 December 1994 (see Preliminary Report of the Independent Commission of Experts
Established in accordance with Security Council Resolution 935 (1994), UNSCOR, UN Doc. S/1994/1125
(1994); Final Report of the Commission of Experts Established pursuant to Security Council Resolution 935
(1994) and Annex, UNSCOR, UN Doc. S/1994/1405 (1994).
30 Article 1 of the Statute of the ICTR thus declared that the ICTR ‘shall have the power to prosecute
persons responsible for serious violations of international humanitarian law committed in the territory of
Rwanda and Rwandan citizens responsible for such violations committed in the territory of neighbouring
states, between 1 January 1994 and 31 December 1994, in accordance with the provisions of the present
Statute’.
328 IN T E R N A T IO N A L C R IM IN A L LAW

After the decision to create the Rwanda Tribunal, which took much time and effort
to establish and function, the SC arguably reached a point of ‘tribunal fatigue-.31
Indeed, the logistics of setting up the two ad hoc Tribunals had strained the capabil­
ities and resources of the UN and consumed the SC time. The SC found itself frequently
seized with issues and problems concerning these Tribunals and their administration,
and as a result became less inclined to establish other similar organs. Furthermore, it
did not consider that other international conflicts deserved the establishment of an
ad hoc tribunal. After 1994, at least for some time, the SC simply did not see fit to take
the same approach with regard to situations that were meanwhile arising in the world.

15.5 THE DRAFTING AND ADOPTION OF THE

STATUTE OF THE ICC ( 1994- 1998)

15.5.1 G E N E R A L

It was only in 1989, once the Cold War had drawn to a close, that the General Assembly
once again requested the ILC ‘to address the question of establishing an international
criminal court’.32
The question of an international criminal court came back on to the United Nations’
agenda by an unexpected route in 1989 after a hiatus of 36 years, following a sugges­
tion in the GA by Trinidad and Tobago that a specialized international criminal court
be established to deal with the problem of drug trafficking. In response to the GA’s
mandate arising out of the 1989 special session on drugs, the ILC in 1990 completed a
report which was submitted to the 45th session of the GA. Though that report was not
limited to the drug trafficking question it was, nonetheless, favourably received by the
General Assembly, which encouraged the ILC to continue its work. The ILC produced
a comprehensive text in 1993, which was modified in 1994.33

31 A term aptly coined by David Scheffer, then Senior Counsel and Advisor to the US Permanent
Representative to the UN; cited in M. C. Bassiouni, The Statute of the International Criminal Court: A
Documentary History, cit., 10, n. 50.
32 UN General Assembly Resolution 44/39 of 4 December 1989. In addition, a proposal to establish a
criminal court dealing with international crimes such as aggression and war crimes did appear to be revived
again in August 1990, in response to the Iraqi invasion of Kuwait and to hostage-taking of foreigners and
atrocities allegedly committed in Kuwait (see various dispatches cited in The Path to the Hague, op. cit., at
7,9,11). However, it is unclear to what extent it was envisaged that the court would have a truly international
character (see The Times, 26 September 1990, ‘Echo of Nuremberg Trials in Iraq’). In any case, these steps
did not lead to any proposal at the international level, although moves towards the establishment of an inter­
national tribunal to prosecute and punish war crimes committed by Iraqi forces in Kuwait seems once more
to be gaining momentum. (See in this regard A. Cassese, ‘On Current Trends towards Criminal Prosecution
and Punishment of Breaches of International Humanitarian Law’, 9 EJIL (1998), 8-9.)
33 Report of the International Law Commission, 46th Sess., 2 May-22 July 1994, UN GAOR, 49th Sess.,
Supp. No. 10, UN Doc. A/49/10 (1994).
 T H E E STA B LISH M E N T OF IN T E R N A T IO N A L C R IM IN A L T R IB U N A L S 329

The judicial institution envisaged in the 1994 ILC Draft to a very great extent took
account of the concerns of states and in particular of major Powers. Among the salient
features of the ICC delineated in the Draft, the following should be emphasized: (i) the
Court had ‘automatic jurisdiction’ (that is, jurisdiction following from the mere fact
of ratifying the Statute) solely over genocide; for other crimes such as war crimes and
crimes against humanity the Court could exercise its jurisdiction only if such jurisdic­
tion had been accepted by the custodial state, the territorial state, as well as any other
state seeking jurisdiction over the accused (Article 21); (ii) only states parties or the
SC could initiate proceedings (Articles 23 and 25); the Prosecutor had no such power;
(iii) the SC had extensive powers with regard to prosecution of cases relating to situa­
tions falling under Chapter VII of the UN Charter (threat to the peace, breach of the
peace, or act of aggression); under Article 23(3), in these cases a prosecution could not
be commenced except in accordance with a decision of the SC.

15.5.2 T H E PREPA RA TO RY C O M M IT T E E ( 19 9 5 - 19 9 8 )
A N D T H E R O M E D I P L O M A T I C C O N F E R E N C E O F 19 9 8

Although the two ad hoc Tribunals were limited both temporally and geographic­
ally to the conflicts in the former Yugoslavia and Rwanda, respectively, their over­
all successes provided a final spur to the emergence of the ICC, an organ of global
jurisdictional reach and thus potentially able to respond to violations occurring
anywhere. Furthermore, much jurisprudence had accumulated regarding the
interpretation of the offences punishable in terms of the new Statute. Those seeking a
permanent, effective, and politically uncompromised system of international criminal
justice drew upon all these factors.
The GA established in 199634 a Preparatory Committee on the Establishment of
an International Criminal Court (PrepCom). This Committee submitted to the
Diplomatic Conference at Rome (15 June-17 July 1998) a Draft Statute and Draft Final
Act consisting of 116 articles contained in 173 pages of text with some 1,300 words in
square brackets, representing multiple options either to entire provisions or to some
words contained in certain provisions.
Both in the works of the PrepCom and in the Rome negotiations, three major
groupings of states emerged.
The first was the group of so-called Like-Minded States, which included countries
from all regions of the world and was to a large extent led by Canada and Australia.
This group favoured a fairly strong Court with broad and ‘automatic jurisdiction’, the
establishment of an independent prosecutor empowered to initiate proceedings, and

34 The 1994 ILC report on the Draft Statute for an International Criminal Court was submitted to the
49th session of the General Assembly, which resolved to consider it at its 50th Session, but first it set up an ad
hoc committee to discuss the proposal. This committee, referred to as the 1995 Ad Hoc Committee for the
Establishment of an International Criminal Court, met inter-sessionally for two sessions of two weeks each
from April to August 1995 (ibid.).
 330 IN T E R N A T IO N A L C R IM IN A L LAW

a sweeping definition of war crimes embracing crimes committed in internal armed

conflicts.
A second group comprised the permanent members of the SC (P-5) however, minus
the UK (which during the preparatory negotiations joined the Like-Minded States)
and France (which also joined the Like-Minded States in Rome). The three remaining
permanent members, and in particular the USA, were opposed to automatic jurisdic­
tion’ and to granting to the prosecutor the power to initiate proceedings. By the same
token they were eager to assign extensive tasks to the SC. This body was to be empow­
ered both to refer matters to the C ourt and to prevent cases from being brought to the
Court. In addition, these states were opposed both to envisaging aggression among
the crimes subject to the C ourt’s jurisdiction, and to including any reference to the use
of nuclear weapons among the violations of hum anitarian law over which the Court
was to exercise jurisdiction.
The third grouping embraced members of the non-aligned-movement (NAM). They
insisted on envisaging aggression among the crimes provided for in the Statute; some
of them (Barbados, Dominica, Jamaica, and Trinidad and Tobago) pressed for the
inclusion of drug trafficking, whereas others (India, Sri Lanka, Algeria, and Turkey)
supported providing for terrorism. They strongly opposed the assignment of any role
to the SC and opposed any jurisdiction over war crimes committed in internal armed
conflicts. In contrast, they insisted on the inclusion of the death sentence among the
possible penalties.
A group of distinguished diplomats, and in particular the Canadian Philippe
Kirsch, who chaired the Committee of the Whole (where the major points of the draft
Statute were substantially negotiated) must be credited with having been able skilfully
to devise and suggest a num ber of compromise formulae that in the event permitted
the Conference to adopt the Statute by 120 votes to 7 (USA, Libya, Israel, Iraq, China,
Syria, Sudan) with 20 abstentions.

15.6 OTHER CRIMINAL TRIBUNALS

15.6.1 T H E E S T A B L I S H M E N T OF I N T E R N A T I O N A L I Z E D
O R M IX E D COU RTS OR T R IB U N A LS

In the late 1990s and early 2000s the UN SC considered the situations in, among other
places, Sierra Leone, Cambodia, and East Timor as being suitable for the establish­
ment of ad hoc international courts.
In the case of Sierra Leone, it actively dealt with the matter. Eventually, in October
2000, at its request the Secretary-General drafted the statute of a Special Tribunal,
which became part of the Agreement of 16 January 2002 between the UN and Sierra
 T H E ESTA B LISH M E NT OF IN T E R N A T IO N A L C R IM IN A L T R IB U N A L S 331

Leone.35 The Special Court for Sierra Leone (SCSL) has a mixed composition (being
made up of nationals of Sierra Leone and international judges and staff) and has jur­
isdiction over crimes against humanity, violations of common Article 3 to the Geneva
Conventions and the Second Additional Protocol, as well as other serious violations of
IHL, and some criminal offences under Sierra Leonean law.
As for East Timor, Section 10 of UNTAET Regulation 2000/11 (as amended by
Regulation 2001/25) conferred on the Special Panels for Serious Crimes (SPSC), which
were a part of the Dili District Court, jurisdiction over genocide, war crimes, and
crimes against humanity, as well as m urder and sexual offences, provided that these
offences were committed between 1 January 1999 and 25 October 1999.
Furthermore, following negotiations with the UN the Cambodian Parliament
adopted in 2001 (and amended in 2004 in light of an agreement of 2003 with the UN)
a law establishing the Extraordinary Chambers in the Courts of Cambodia (ECCC)
for prosecuting crimes committed during the period of Democratic Kampuchea
(1975-9). The Chambers are composed partly of Cambodian judges, partly of inter­
national judges.
In addition, following an agreement with the Government of Bosnia and
Herzegovina, in 2005 the High Representative for Bosnia and Herzegovina set up
a Section for War Crimes in the Criminal and Appellate Divisions of the C ourt of
Bosnia and Herzegovina; this section is mixed in composition. The establishment of
the War Crimes Chamber (WCC) was considered necessary to enable effective war
crimes prosecutions in Bosnia. The WCC is part of the State C ourt and exercises jur­
isdiction over the most serious war crimes in Bosnia, while the cantonal and district
courts can handle other war crimes cases.36

35 See UN Doc. S/2000/915. See also SC Res. 1315 (2000). For an overview, see M. Frulli, 'The Special
Court for Sierra Leone: Some Preliminary Comments’, 11 EJIL (2000), 857-69.
36 The WCC tries cases concerning lower- to mid-level perpetrators’ referred to it by the ICTY pursuant
to Rule 11 E>i5of the ICTY RPE. In this respect, the WCC represents an important component of the comple­
tion strategy of the ICTY. Furthermore, the WCC is responsible for those cases submitted to it by the Office
of the Prosecutor (OTP) of the ICTY where investigations have not been completed. The ICTY AC referred
the first case to the WCC on 1 September 2005 (Radovan Stankovic). Mr Stankovic was transferred to Bosnia
on 29 September 2005, to stand trial before the WCC for charges of crimes against humanity, including
enslavement and rape. The ICTY has since referred other cases to the WCC.
The WCC also has jurisdiction over ‘Rules of the Road’ cases. The ‘Rules of the Road’ procedure was first
established in response to the widespread fear of arbitrary arrest and detention immediately after the con­
flict in Bosnia. Under this procedure, the authorities in Bosnia were required to submit every war crimes
case proposed for prosecution in Bosnia to the OTP of the ICTY to determine whether the evidence was
sufficient by international standards before proceeding to arrest. This process of review reduced incidents
of arbitrary arrest in Bosnia. The ICTY ceased reviewing cases on 1 October 2004. The review function was
subsequently assumed by the Special Department for War Crimes within the Office of the Prosecutor of the
State Court.
The WCC includes national as well as international judges and prosecutors, defense counsel, experts in
witness protection and support, as well as other officials engaged in providing substantive and administra­
tive support.
The WCC has both Trial and Appeals Chambers. There are at present five judicial panels allocated to the
WCC. Panels include two international judges and one local judge, who is the presiding judge. According
to the transition strategy of the WCC, between August 2006 and December 2007 the configuration of
332 IN T E R N A T IO N A L C R IM IN A L LAW

On 30 May 2007 the UN SC established a Special Tribunal for Lebanon, by a reso­

lution (1757-2007) adopted on the strength of Chapter VII of the UN Charter. The
Tribunal, which consists of a Trial Chamber and an Appeals Chamber, has jurisdiction
over terrorist attacks committed in Lebanon since 14 February 2005 and only applies
Lebanese criminal law on terrorism. All the organs of the Tribunal have a mixed
composition, the Lebanese nationals being outnumbered by international
professionals.37

15.6.2 T H E R A T I O N A L E B E H I N D T H E E S T A B L I S H M E N T
OF TH ESE CO U R TS A N D T R IB U N A L S

As pointed out above, in recent years, faced with emergency situations involving the
commission of large-scale atrocities, states have preferred to resort neither to national
nor to international crim inal courts, but rather to establish courts that are mixed
in their composition, and the statutes and rules of which combine aspects of inter­
national law and municipal law. Such courts have been set up for Sierra Leone, East
Timor, Kosovo, Cambodia, and Lebanon.38
These courts aim at improving on the two Ad Hoc Tribunals, which were perceived
as being m arred by four essential flaws: (i) their costly nature; (ii) the excessive length
of their proceedings; (iii) their remoteness from the territory where crimes have
been perpetrated and consequently the limited impact of their judicial output on the
national populations concerned; (iv) the unfocused character of the prosecutorial tar­
gets resulting in trials of a num ber of low-ranking defendants. The attempt was there­
fore made to establish lean and agile courts sitting in the territory where crimes had

the judicial panels will shift to two national judges and one international judge. By the end of 2009, it is
anticipated that there will no longer be any international judge within the WCC.
37 See UN doc. S/2006/893 15 November 2006, containing the Report of the UN SG to the SC as well as
the Statute of the Tribunal and the Agreement between the UN and Lebanon on the establishment of the
Tribunal.
38 In Cambodia, after years of pressure by the international community, and after a UN Commission had
proposed the establishment of an international criminal tribunal, Cambodian authorities have opted for the
creation of special Cambodian courts with mixed composition. On 2 January 2001 Cambodia’s Parliament
(the National Assembly) passed a Law on the Establishment of Extraordinary Chambers in the Courts of
Cambodia for the Prosecution of Crimes Committed during the Period of Democratic Kampuchea.
In East Timor the UN provisional administration (UN Transitional Administration in East Timor, or
UNTAET) adopted Regulation 2000/11 in 2000 setting up mixed panels within the District Court of Dili
and Regulation 2000/15 established Panels with exclusive jurisdiction over ‘serious criminal offences’.
In Sierra Leone, after the drafting of a Statute of a Special Court in 2000, on 16 January 2001 the UN and
the Sierra Leone Government signed an agreement establishing a mixed court, accompanied by a Statute.
In Kosovo the UN provisional administration (UNMIK or UN Interim Administration in Kosovo) passed
a Regulation on the appointment of international judges to serve on Kosovar courts (UNMIK Regulation
no. 2000/64 of 15 December 2000). This Regulation provided for the establishment of panels of three judges,
composed of two international judges (one of them presiding over the panel) and a local one. It also provided
for the appointment of international prosecutors and investigating judges.
UNMIK has appointed international judges and prosecutors to district courts throughout Kosovo and to
the Supreme Court. The international judges have worked on criminal cases involving alleged war crimes or
inter-ethnic violence, and on property issues.
 T H E ESTA B LISH M E NT OF IN T E R N A T IO N A L C R IM IN A L T R IB U N A L S 333

been committed (and also including prosecutors and judges from this country), which
would be relatively inexpensive and only tasked with prosecuting and trying those
most responsible for the crimes perpetrated. Thus, some ‘m ixed’ or, as they are often
termed, ‘internationalized\courts and tribunals were set up.
This notion encompasses judicial bodies that have a mixed composition. There may
be two versions of these courts and tribunals. First, they may be organs of the rele­
vant state, being part of its judiciary. This applies to the Cambodian Extraordinary
Chambers as well as the courts in Kosovo and the ‘Special Panels for Serious Crimes’
in East Timor. Alternatively, the courts may be international in nature: they may be
set up under an international agreement and not be part of the national judiciary. This
holds true for the SCSL and the STL.
A multitude of historical and practical reasons combine to warrant the establish­
ment of courts that are neither national nor international, but mixed.
First, as a result of an emergency situation (armed conflict, civil strife, strong reli­
gious and ethnic tension), serious and widespread crimes are committed. W hen the
emergency situation is over, it is felt that bringing to trial those responsible for serious
crimes may help in the post-conflict peace-building process and may also serve to
deter the future commission of large-scale offences.
Secondly, as a result of the emergency situation a breakdown of thejudicial system may
have come about. This may have been caused by civil war (as in East Timor and Sierra
Leone), possibly followed by an international conflict (as in Kosovo). Alternatively,
even if a very long time has elapsed since the crimes were perpetrated, and a stable
government has taken root, as a result of a series of historical factors the judiciary may
not be capable of administering justice in an unbiased and even-handed manner: this
is what has happened in Cambodia, where the presence in the government of persons
who allegedly are closely linked to the perpetrators of genocide, together with the lack
of a really independent judiciary, might lead to unfair trials. It may also happen that
the population prevents or hampers the conduct of fair trials. In Kosovo, the ethnic
biases of Kosovo Albanians and Serbs rendered the presence of international judges
indispensable for administering justice.39
Thirdly, it is, however, considered that the judicial response must not lie in the
establishment o f an international tribunal. This option is normally ruled out because
of the combination of two factors: (i) lack of political will of the relevant organs of
the international organization that should set up the international tribunal, for the
country or the situation at issue are regarded as either inconsequential in geo-political

39 As the UN Secretary-General pointed out in his Report of 6 June 2000, ‘Despite the appointment of
more than 400 judges, prosecutors and lay judges [of whom 46 were non-ethnic Kosovo Albanians, 7 of
whom Kosovo Serbs] and the increased capacity of the courts, the unwillingness of witnesses to testify and
the ethnic bias and risk of intimidation of some judicial personnel have hampered the administration of
justice’ (see S/2000/538, §57). The detainees of the Mitrovica detention centre who had gone on a hunger
strike in protest over the length of their pre-trial detention, stopped the strike only when the UN Special
Representative promised that he would ensure that a Kosovo Serb or an international judge would preside
over their cases, in addition to Kosovo Albanian judges (ibid., §59).
 334 IN T E R N A T IO N A L C R IM IN A L LAW

terms, or likely to spawn further friction and thus to embroil the international
organization in a never-ending conflict; and (ii) lack of will of major Powers to fund
the international tribunal.
Fourthly, it is felt that by holding trials in the territory where the crimes have been
perpetrated, the local population is exposed to past atrocities, with the twofold advan­
tage of making everybody cognizant of those atrocities, including those who sided
with the perpetrators, and bringing about a cathartic process in the victims or their
relatives, through public stigmatization of the culprits and just retribution; thus,
exposure of past misdeeds to the local population contributes to the process of grad­
ual reconciliation.
Finally, those in charge of finding a solution feel that using the national judiciary
under some sort of international scrutiny, or even control, may prove advantageous
and useful in many other respects.40

15.6.3 T H E M A I N P R A C T I C A L P R O B L E M S T H A T M A Y A R I S E

Nevertheless, one should not underestimate the practical problems and difficulties
that may arise with mixed or hybrid, international-national courts.
The first problem is to ensure that the national and international component of the
prosecution work in close, constructive, and constant agreement.41
Another, no less serious problem may be to ensure the smooth cooperation of the
national and international components of the bench. There may be differences in m en­
tality, language, experience, and legal philosophy.42

40 In short:
1. It assuages the nationalistic demands of local authorities, loath to hand over to international bodies an
essential prerogative of sovereign Powers, the administration of justice.
2. It involves, in rendering justice, persons (the local prosecutors and judges) familiar with the mentality,
language, habits, and so on of the accused.
3. It may expedite prosecution and trials without compromising respect for international standards and
international law in general.
4. It may produce a significant spill-over effect, in that it may contribute to gradually promoting the
democratic legal training of local members of the prosecution and the judiciary.
41 This may not prove easy, because the local prosecutors may either tend to be over-zealous, when the
accused belong to an ethnic or religious group to which they are hostile, or instead to engage in dilatory
tactics, or even to create obstacles to prosecution, when the accused belong to their own group.
Article 20 of the Cambodian Draft Law provides that in case of disagreement between the two prosecutors
(one national, the other international) the matter is referred to a Pre-Trial Chamber of five judges, two inter­
national and three national, deciding by an affirmative vote of at least four judges; if there is no majority, for a
decision, ‘the prosecution shall proceed’. The same procedure applies to possible disagreements between the
two investigating judges (see Article 23). At least in some respects this solution appears to be fairly sensible,
for in the final analysis it ensures that a prosecution will be instituted. However, as has been pointed out, it
is indeed unusual and contrary to the principled distinction between prosecution and bench, for a panel of
judges to settle disputes between co-prosecutors concerning prosecutorial strategy at trial.
42 Particularly where international members of the Tribunal are the majority (as is the case in Sierra
Leone and East Timor), they may be perceived by the local judges as intrusive and overwhelming, with the
consequence that the local judges may seek to obstruct or otherwise hamper the process of administering
justice. On the other hand, the action of the international component may be thwarted by the attitude of
 T H E E STA B LISH M E N T OF IN T E R N A T IO N A L C R IM IN A L TR IB U N A LS 335

Two other practical problems may prove crucial: funding and security. Making
financial resources available is a sine qua non for a court to function.43As for security,
it is obvious that the dangers following from the hatred, resentment, and social con­
flict festering in those countries may pose a serious risk to those working in the judi­
cial process. These seem to be the reasons why the trial against the former Liberian
President Charles Taylor is being held before the Special Court for Sierra Leone, not at
the seat of the Court (Freetown), but at The Hague.

the local judges. Probably the main reasons that the internationalized courts operating in Kosovo have
been assailed for their alleged flaws in ensuring fair trials, and full respect for the rights of the defence, are
linked to this problem. The matter is further complicated where, as in the case of Cambodia, the relevant
law provides that the majority belongs to national judges (three to two), but a decision may only be taken if
four judges are in favour. This entails that if two national judges are not agreeable, no decision is made, and
the court becomes deadlocked.
Things may become even worse, for under Article 46 of the Cambodian law, in cases of last resort
Cambodians may be appointed to take the place of foreign judges (and prosecutors); hence, if there is a
deadlock and the subsequent withdrawal of the international judges, judicial action may continue with all-
Cambodian staff. It follows that ultimately the national component may gradually push the international
component to withdraw, so as to take complete control of the judicial process.
43 Money is needed not only to pay salaries to prosecutors and judges, but also to fund interpretation
and translation from and to English and French, one of these two languages being the vehicular means
of the international component. For instance, in his Report of 16 January 2001 on East Timor, the UN
Secretary-General pointed out that a shortage of skilled translators had hampered efforts at all judicial
levels (see S/2001/42, §23). In his Report of 18 October 2001 the Secretary-General noted that the lack of
resources, including interpreters, had led to delayed hearings and unduly prolonged detention of suspects
(see S/2001/983, §20).
Money is also needed to provide those authorities with books and documents. (For instance, in Kosovo,
the destruction of law libraries among other things has resulted in a dearth of relevant legal texts.)
 16
INTERNATIONAL VERSUS
NATIONAL JURISDICTION

16.1 PRIMACY AND COMPLEMENTARITY

The establishment of international criminal courts and tribunals has posed the tricky
problem of how to coordinate their action with that of national courts: whenever both
classes of courts are empowered to pronounce on the same crimes, which should take
precedence, and under what conditions? Obviously, the problem does not arise in the
area where those courts do not have concurrent jurisdiction; that is, with regard to
crimes that fall under the exclusive jurisdiction of national courts (for example, pir­
acy, large-scale drug trafficking, slave trade, serious instances of international terror­
ism falling short of crimes against humanity, war crimes, or terrorism as a discrete
crime, etc.).
The problem only arises when one or more states may assert their criminal jurisdiction
over a specific crime on the basis of one of the accepted heads of jurisdiction: territoriality,1

1 The principle is grounded on ideological and political reasons: the need to affirm territorial sovereignty,
which evolved in the age of reason and was linked to the consolidation of modern states. The principle
has numerous advantages. First, the locus delicti commissi (the place where the offence has allegedly been
committed) is usually the place where it is easiest to collect evidence. It is therefore considered the forum
conveniens, or the appropriate place of trial, as was restated in Eichmann (at 302-3). Secondly, it is nor­
mally the place where the rights of the accused are best safeguarded, for—if he is not a foreigner fleetingly
residing there—he is expected to know the law of the territory, hence he is likely to know the criminal law
in force there as well as his rights as a defendant in a criminal trial. In addition, unless he is a non-resident
foreigner, he knows and speaks the language in which the trial unfolds. Thirdly—and this applies in par­
ticular to international crimes, whose gravity may have serious repercussions on the society within which
the crime has been committed—if the prosecution and punishment occur on the territory where the crime
was perpetrated, it is more likely for the cathartic process of criminal trials to have effect: the victims and
their families relive their tragedies, the whole society becomes aware of what has happened and is thus put
in a position to better come to terms with, hence to psychologically overcome, past crimes. Moreover, the
judges, jury, and advocates, being members of the community where the crimes took place, are aware of
local feelings about the crimes and conscious of the press and publics close scrutiny of their administration
of justice; they are thus broadly accountable to the community for the manner in which they dispense just­
ice. Finally, by administering justice over crimes perpetrated in the territory, the territorial state affirms its
authority over attacks on law and order within its bounds; by the same token it helps to deter the commission
of future offences.
However, in the case of international crimes, a major obstacle to the territoriality principle is posed by
the fact that these crimes are often committed by state officials or with their complicity or acquiescence.
 INTERNATIONAL VERSUS NATIONAL JURISDICTION 337

active personality or nationality,2 passive personality (or nationality),3 or

Consequently state judicial authorities may be reluctant to prosecute state agents or to institute proceedings
against private individuals that might eventually involve state organs.
2 This principle is implemented in one of two forms. In some states, courts have jurisdiction over certain
criminal offences committed by their nationals abroad. This is so, whether or not those offences are criminal
under the law of the territorial state; that is, the state in which the conduct constituting the offences under
the law of the state of nationality were committed. In this case the underlying motivation is the will of a
state that its nationals comply with its own law, whether at home or abroad, regardless of what is provided
for in the foreign state when the crime is committed. In other countries jurisdiction over crimes committed
by nationals abroad is subordinated to the crime being punishable under the law of the territorial state as
well (this, for instance, holds true for Egypt). In this case the essential rationale behind the principle is the
desire—or constitutional prohibition in many cases—of the state of nationality not to extradite its nation­
als to the state where the crime has been perpetrated. Hence the law of the state of active nationality must
provide for the possibility of trying the accused in that state, so that he does not escape justice altogether.
Indeed, it is striking that countries such as the UK, which have no constitutional or other prohibition on the
extradition of their nationals, normally do not provide for active nationality as a basis for jurisdiction.
All this holds true, generally speaking, for criminal offences. As for international crimes, states that
uphold this ground of jurisdiction do so in order to bow to international dictates; that is to make inter­
national law effective by complying with its commands. Thus, they normally do not require that the offence
be also punishable by the territorial state, as it is sufficient for the offence to be regarded as an international
crime by international rules (be they customary or treaty provisions).
3 By virtue of this principle states may exercise jurisdiction over crimes committed abroad against their
own nationals. Plainly, the principle is grounded both on: (i) the need to protect nationals living or residing
abroad-, and (ii) a substantial mistrust in the exercise of jurisdiction by the foreign territorial state.
Normally states invoking this ground of jurisdiction also provide that, whenever the accused is abroad,
a ‘double incrimination’ is required for prosecuting a crime, namely that the offence be considered as such
both in the territorial state and in the state of the victim. ‘Double criminality’ is usually considered a pro­
cedural requirement of extradition: to extradite from state X to state Y, the crime in question must be an
offence in both states. Normally, where passive nationality is exercised, the state will have to seek extradition
of the perpetrator as he will be abroad. Of course, that is not always the case. If the perpetrator, for example
a Chilean accused of murdering a Spaniard, is in the state exercising jurisdiction, e.g. Spain, then Spain will
be able to proceed against him without going through extradition proceedings and therefore without having
to worry about ‘double criminality’.
Outside extradition law the requirement at issue is intended to avoid prosecuting a person for an act that is
not considered a criminal offence by the state where it has been performed; in other words, the rationale for
this requirement maybe found in the general principle oflegality (nullum crimen sine lege) which is common
to all national legal systems, in addition to being a general principle of international criminal law (see 2.3).
However, as far as international crimes are concerned, this requirement is replaced by the requirement that
the offence be considered as an international crime by international law, whatever the content of the legal
regulation in the territorial state. In this connection, the decision of the Supreme Court of Argentina delivered
in Priebke on 2 November 1995, concerning the extradition to Italy of a German national who had allegedly
committed crimes in Italy and subsequently acquired Argentinian nationality, is pertinent: the Court expli­
citly held that as the offence of which the defendant stood accused, namely a war crime, was internationally
regarded as an international crime, this sufficed for the purpose of the double incrimination principle.
There has been frequent resort to this ground of jurisdiction to prosecute war crimes, particularly after
the cessation of hostilities and by the victor state against the vanquished (former) enemies. (Notable depar­
tures based on the active nationality principle are the trials instituted in 1902 by US Courts Martial against
American servicemen who had fought in the Philippines, the Leipzig trials against Germans, imposed upon
Germany by the Allies, and the various trials before US Courts Martial for crimes committed in Vietnam.)
More recently courts have relied upon this jurisdictional ground with regard to crimes against human­
ity and torture. Significant in this respect are some cases tried in absentia: Astiz, a case brought before
French courts (an Argentinian officer had tortured two French nuns in Argentina), as well as some cases
brought before Italian courts against Argentinian officers for crimes allegedly perpetrated against Italians
(or Argentinians also having Italian nationality) in Argentina (see for instance Sudrez Mason and others
338 INTERNATIONAL CRIMINAL LAW

universality,4 and at the same time an international tribunal is empowered to adjudi­

cate the same crime by virtue of its Statute.
There are no general international rules determinative of this matter, just as there
are no international customary rules designed to resolve the question of concurrent

(at 8-15), Furthermore, this ground of jurisdiction has been laid down in national legislation with regard to
terrorism, for instance in the United States (see §§2331 and 2332 of the Federal Criminal Code), in France
(Articles 113-17 of the Criminal Code), and in Belgium. It is also stipulated in a number of international
conventions against terrorism and in the 1984 Convention against Torture (Article 5(1)).
Resort to the passive personality principle is, however, particularly incongruous in the case of inter­
national crimes such as for instance those against humanity, and torture. By definition, these are crimes
that injure humanity, that is, our sense of humanity; in other words our concept of respect for any human
being, regardless of the nationality of the victims. As a consequence, their prosecution should not be based
on the national link between the victim and the prosecuting state. This is indeed a narrow and nationalistic
standard for bringing alleged criminals to justice, based on the interest of a state to prosecute those who
have allegedly attacked one of its nationals. The prosecution of those crimes should instead reflect a univer­
sal concern for their punishment; it should consequently be based on such legal grounds as territoriality,
universality, or active personality.
It follows that, as far as such crimes as those against humanity, torture, and genocide are concerned, the
passive nationality principle should only be relied upon as afail-back, whenever no other state (neither the
territorial state, nor the state of which the alleged criminal is a national, nor other states acting upon the
universality principle) is willing or able to administer criminal justice. Perhaps this is the reason why in
international conventions such as that on torture this ground ofjurisdiction, unlike those just mentioned, is
envisaged not as an obligation of contracting states but simply as an authorization to prosecute (see Article
5(l)(c) of the 1984 Convention against torture).
Conversely, the head of jurisdiction under discussion may prove appropriate for such offences as war
crimes or terrorism as a discrete offence, where the need to protect national interests and concerns acquires
greater relevance.
4 Under this principle any state is empowered to bring to trial persons accused of international crime
regardless of the place of commission of the crime, or the nationality of the author or of the victim. This jur­
isdictional ground was included in the 1949 Geneva Conventions, the 1984 Convention against Torture, and
a string of international treaties on terrorism. The universality principle has been upheld in two different
versions, both predicated on the notion that the judge asserting universal jurisdiction so acts in order to sub­
stitute for the defaulting territorial or national state: the narrow notion (conditional universal jurisdiction)
and the broad notion (absolute universal jurisdiction).
Under the former notion, only the state where the accused is in custody may prosecute him or her (the
so-called forum deprehensionis, or jurisdiction of the place where the accused is apprehended). Thus, the
presence of the accused on the territory is a condition for the existence of jurisdiction. This class of jurisdic­
tion is accepted, at the level of customary international law, with regard to piracy. At the level of treaty law
it has been upheld with regard to grave breaches of the 1949 Geneva Conventions and the First Additional
Protocol of 1977, torture (under Article 7 of the 1984 Torture Convention), as well as terrorism (see the vari­
ous UN-sponsored treaties on this matter).
Under the latter notion of universality a state may prosecute persons accused of international crimes
regardless of their nationality, the place of commission of the crime, the nationality of the victim, and even
of whether or not the accused is in custody or at any rate present in the forum state. However, as many legal
systems do not permit trials in absentia, the presence of the accused on the territory is then a condition for
the initiation of trial proceedings. Clearly, this conception of universality allows national authorities to
commence criminal investigations of persons suspected of serious international crimes, and gather evi­
dence about these alleged crimes, as soon as such authorities are seized with information concerning an
alleged criminal offence. They may thus exercise criminal jurisdiction over such persons, without requiring
that the person first be present, even temporarily, in the country.
It is notable that in Eurico Guterres (at 12) an Indonesian court asserted that national courts are endowed
with universal jurisdiction over crimes against humanity.
 INTERNATIONAL VERSUS NATIONAL JURISDICTION 339

jurisdiction of two or more states, by giving pride of place to one legal ground of
national jurisdiction (say, territoriality) over another such ground (say, passive nation­
ality). Luckily, while not even treaty rules have settled the possible conflict between
states claiming jurisdiction over the same person accused of the same crime, possible
conflicts between national and international criminal courts have been resolved by
treaty rules or otherwise binding resolutions. In short, in the case of the ICTY and the
ICTR, primacy has been given to the international tribunal, whereas in the ICC sys­
tem national courts take precedence over the Court, under certain specific conditions.
Let us now dwell at some length on this matter, to consider both the legal implications
and the underlying political motivations.

16.2 THE PRIMACY OF INTERNATIONAL

AD HOC TRIBUNALS

The Statutes of the ICTY and ICTR, at Articles 9 and 8 respectively, provide that each
Tribunal shall have concurrent jurisdiction with national courts to prosecute persons
for serious violations of IHL, but add in paragraph 2 that the Tribunal ‘shall have pri­
macy over national courts’:
At any stage of the procedure, the International Tribunal may formally request national
courts to defer to the competence of the International Tribunal in accordance with the pre­
sent Statute and the Rules of Procedure and Evidence [RPE] of the International Tribunal.

The reasons for proclaiming the Tribunals’ primacy were clear. In the case of the
former Yugoslavia, the ongoing armed conflict among the successor states and the
deep-seated animosity between the various ethnic and religious groups made national
courts unlikely to be willing or able to conduct fair trials. It was considered that the
authorities would have hesitated to bring their own people (Muslims, Croats, or Serbs)
to book, whereas, had they initiated proceedings against their adversaries, probably
such proceedings would have been highly biased. As for other states, the experience
built up until that time showed that they shied away from bringing to trial alleged
perpetrators of crimes committed elsewhere. Hence the need was felt to affirm the
overriding authority of the International Tribunal. Similar considerations held true
for Rwanda, where in addition the national judicial system had collapsed and conse­
quently seemed unable to render justice.
However, the Statutes do not specify on what conditions and how primacy is to be
exercised. In his Report to the SC elaborating upon the draff Statute of the ICTY, the
UN SG simply stated that ‘The details of how the primacy will be asserted shall be
set out in the rules of procedure and evidence of the International Tribunal’ (§65).
The judges of the ICTY skilfully drew up a set of rules on primacy, which were subse­
quently taken up by the judges of the ICTR. These rules do not lay down the absolute
primacy of the Tribunal; rather, they provide that the concurrent jurisdiction of the
 340 INTERNATIONAL CRIMINAL LAW

Tribunal and national courts may lead to the prevalence of national courts, and even
that the Tribunal may divest itself of a case when it considers that the case may more
appropriately be tried by a national court (Rule llbis of the ICTY RPE). Thus, judges
worked out a mechanism whereby a case could be referred back to national courts
whenever they deemed it appropriate. However, the Rules provide that at the request
of the Prosecutor the Tribunal may assert its primacy in three cases:
1. When a national prosecutor investigates an international crime or a national
court conducts proceedings with regard to the criminal offence not as an international
crime, but as ‘ordinary criminal offence’ (for instance, genocide is being investigated
or tried as ‘multiple m urder’, or serious ill-treatment of prisoners of war is handled
as ‘assault’ and not as a war crime). In this case, the classification of the offence as
an ordinary crime presupposes a deliberate (or unconscious) proclivity to misrepre­
sent the very nature, hence to belittle the seriousness, of international crimes. In other
words, the national court shows that, either intentionally or unwittingly, it is not cog­
nizant of both the international dimension and the gravity of the criminal offence.
2. W hen a national court proves to be unreliable-, this happens where it is proved,
under Rule 9(ii) of the ICTY RPE, that there is ‘a lack of impartiality or independence’,
or ‘the investigations or proceedings are designed to shield the accused from inter­
national criminal responsibility’, or else ‘the case is not diligently prosecuted’. Clearly,
in all these instances national authorities may not be trusted because they are intent
on ‘protecting’ the accused or else take a patently persecutory attitude to him.
3. When, although the relevant national court appears to be reliable and able to
conduct a fair trial, nonetheless the case is closely related, or may be relevant, to other
cases being tried by the International Tribunal. Under Rule 9(iii), ‘what is in issue is
closely related to, or otherwise involves, significant factual or legal questions which
may have implications for investigations or prosecutions before the Tribunal’. Plainly,
these cases are of such overriding significance or general im port or wide ramifications,
that it appears appropriate for them to be brought before an international court.5
By and large, the scheme adopted by the judges of the ICTY (and of the ICTR) seems
wisely to reconcile (i) the need not to overload international institutions with relatively
m inor cases, leaving them to national courts; as well as (ii) the demands of state sov­
ereignty; with (iii) the requirement that international courts should replace national
institutions when these prove unreliable or unfair, and in addition should deal with
major international crimes of relevance to the international community as a whole.

5 Interestingly the first two exceptions are also exceptions for non bis in idem, (or in other words, prohib­
ition of double jeopardy) and therefore can be justified by reference to the Statute’s provisions on non bis in
idem. (There is a clear parallel between conditions for exercising primacy and non bis in the sense that what
justifies the court retrying an accused are surely also grounds for taking over the proceedings before they
have reached their conclusion.) Instead, case (3) is somewhat unusual and dubious in that it has no basis
in the Statute. Moreover, it will nearly always apply—any national war crimes prosecution in the former
Yugoslavia is bound to have implications for the ICTY’s proceedings. This may therefore seem to be too
all-encompassing.
 INTERNATIONAL VERSUS NATIONAL JURISDICTION 341

So far the two Tribunals have occasionally relied upon their primacy.6 In recent
times, the two ad hoc Tribunals have emphasized the importance of national courts
dealing with the crimes falling under the Tribunals’jurisdiction. There seem to be two
grounds behind this new trend. First, national courts of the states concerned (those of
the successors to the former Yugoslavia, and those of Rwanda) are now better struc­
tured, more efficient, and less prone to bias. Secondly, the workload of the international
Tribunals has increased and it therefore proves appropriate for national courts grad­
ually to share the burden and even start to take over the job from the Tribunals, the
more so in light of the so-called ‘completion strategy’ adopted by the UN SC for the
two ad hoc Tribunals and aimed at closing down their activity in a few years. A number
of cases have indeed been passed on by the ICTY to the War Crimes Chamber of the
High Court in Bosnia and Herzegovina, as well as to courts in Croatia and Serbia.7
While the ICTY and the ICTR enjoy primacy over any national court, the SCSL
and the STL have been granted primacy only over the courts of Sierra Leone and
Lebanon, respectively. Under Article 8(2) of the SCSL Statute ‘The Special Court shall
have primacy over the national courts of Sierra Leone. At any stage of the procedure,
the Special Court may formally request a national court to defer to its competence in
accordance with the present Statute and the Rules of Procedure and Evidence.’ Under
Article 4 of its Statute the STL has primacy over Lebanese courts.8It follows that these

6 See, for instance, the Tadic case, as far as the ICTY is concerned: on 8 November 1994 the ICTY made a
request for deferral to the Federal Republic of Germany, whose authorities were investigating Tadic’s alleged
crimes; Germany immediately complied with the request and surrendered the accused to the Tribunal: see
Tadic, Decision on deferral to the competence of Tribunal. See also the order of 4 October 2002 in Republic
of Macedonia (Prosecutor’s Request for Deferral and Motion for Order to the Former Yugoslav Republic of
Macedonia), §6-53).
7 In his statement to the UN SC of 18 June 2007, in dealing with the referral of cases involving inter­
mediate and lower-ranking accused to competent national jurisdictions by the International Tribunal (as
authorized by Security Council resolution 1534 of 2004 and stipulated under Rule I Ibis of the Rules), the
ICTY President pointed out that the impact of the referrals already processed on the overall workload of the
International Tribunal had been substantial: ‘Ten accused have been transferred to the Special War Crimes
Chamber of Bosnia and Herzegovina, two accused have been transferred for trial before the domestic courts
of Croatia, and one accused has been transferred to Serbia for trial. Only two accused remain to have their
transfer finalized. Of the cases referred by the International Tribunal, two trial proceedings have been com­
pleted by the Sarajevo Special War Crimes Chamber. The International Tribunal is satisfied that the trials of
both of these accused respected international norms of due process. Unfortunately, one of the accused con­
victed and sentenced to 20 years’ imprisonment, Radovan Stankovic, escaped from the custody of the Bosnia
and Herzegovina authorities on 25 May 2007. The International Tribunal is extremely concerned about
this escape and has requested a full report from the Bosnia and Herzegovina authorities. The International
Tribunal is hopeful that those authorities, and other states, will do all in their power to return Stankovic to
custody. A failure to do so may impact upon the future integrity of the 11bis referral process. With respect to
the Ademi and Norac case referred to Croatia on 14 September 2005, the trial which has suffered some delays
is expected to commence today, 18 June and I hope that it will proceed expeditiously.
8 Article 8 provides as follows:
T. The Special Tribunal and the national courts of Lebanon shall have concurrent jurisdiction. Within its
jurisdiction, the Tribunal shall have primacy over the national courts of Lebanon.
2. Upon the assumption of office of the Prosecutor, as determined by the Secretary-General, and no later
than two months thereafter, the Special Tribunal shall request the national judicial authority seized
with the case of the attack against Prime Minister Rafiq Hariri and others to defer to its competence.
The Lebanese judicial authority shall refer to the Tribunal the results of the investigation and a copy of
the court’s records, if any. Persons detained in connection with the investigation shall be transferred
to the custody of the Tribunal.
 342 INTERNATIONAL CRIMINAL LAW

two Tribunals do not enjoy any primacy over national courts of other states, for which
the Tribunal’s Statute is a res inter alios acta (a legal instrum ent made by others and
only binding upon others), with the consequence that (i) in principle those national
courts may bring to trial persons who stand accused before one of the international
Tribunals without breaching the Statute; (ii) the international Tribunal may not oblige
the relevant national court to defer to its jurisdiction; (iii) the international Tribunal
may not issue to state officials (other than from Sierra Leone and Lebanon, respect­
ively) or to third states binding requests for judicial cooperation (for instance, for the
carrying out of searches or seizures to gather evidence, for allowing or enabling a wit­
ness to be questioned, for the apprehension of a suspect).
This drawback would seem to be less serious in the case of the STL, because its
Statute has been adopted by virtue of a legally binding SC resolution (res. 1757-2007)
passed on the strength of Chapter VII of the UN Charter; it would follow that, in
principle, no member state of the UN may refuse to acknowledge the existence and
functioning of the Tribunal and should cooperate with it, by virtue of the general duty
of cooperation with UN organs (or bodies set up by the UN or upon its authorization)
underlying UN membership.
On the other hand, when a trial against one of the persons falling under the inter­
national Tribunal’s jurisdiction is held in a ‘third’ country, the ban on ne bis in idem
(or double jeopardy) does not apply, for the international Tribunal may try again that
person, if the national trial did not prove to be fair and effective.9

16.3 THE COMPLEMENTARITY OF THE ICC

In contrast, the ICC is based on the principle of complementarity whereby the Court
is subsidiary or complementary to national courts. These courts enjoy priority in the

3. (a) At the request of the Special Tribunal, the national judicial authority seized with any of the other
crimes committed between 1 October 2004 and 12 December 2005, or a later date decided pursuant
to Article 1, shall refer to the Tribunal the results of the investigation and a copy of the court’s records,
if any, for review by the Prosecutor;
(b) At the further request of the Tribunal, the national authority in question shall defer to the competence
of the Tribunal. It shall refer to the Tribunal the results of the investigation and a copy of the court’s
records, if any, and persons detained in connection with any such case shall be transferred to the cus­
tody of the Tribunal;
(c) The national judicial authorities shall regularly inform the Tribunal of the progress of their investiga­
tion. At any stage of the proceedings, the Tribunal may formally request a national judicial authority
to defer to its competence.’
9 Under Article 9(2) of the SCSL Statute A person who has been tried by a national court for the acts
referred to in articles 2 to 4 of the present Statute may be subsequently tried by the Special Court if; (a) The act
for which he or she was tried was characterized as an ordinary crime; or (b) The national court proceedings
were not impartial or independent, were designed to shield the accused from international criminal respon­
sibility or the case was not diligently prosecuted.’
Under Article 5(2) of the STL Statute, A person who has been tried by a national court may be subse­
quently tried by the Special Tribunal if the national court proceedings were not impartial or independent,
were designed to shield the accused from criminal responsibility for crimes within the jurisdiction of the
Tribunal or the case was not diligently prosecuted.’
 INTERNATIONAL VERSUS NATIONAL JURISDICTION 343

exercise of jurisdiction except under special circumstances, when the ICC is entitled
to take over and assert its jurisdiction. There are two underlying reasons for this
approach.
First, states saw a practical ground: they considered it inappropriate for the Court
to be flooded with cases from all over the world. The Court, having a limited number
of judges and limited financial resources and infrastructure, would be unable to cope
with a broad range of cases. It is healthy, it was thought, to leave the vast majority of
cases concerning international crimes to national courts, which may properly exercise
their jurisdiction based on a link with the case (territoriality, nationality) or even on
universality. Among other things, these national courts may be in a better position to
collect the necessary evidence and to lay their hands on the accused.
Secondly, there was perhaps a principled motivation, namely the intent to respect
state sovereignty as much as possible.
Complementarity is laid down in paragraph 10 of the Preamble as well as in
Article 1 of the Statute (whereby the ICC ‘shall be complementary to national criminal
jurisdictions’) and is spelled out in Articles 15, 17, 18, and 19. In short, the C ourt is
barred from exercising its jurisdiction over a crime whenever a national court asserts
its jurisdiction over the same crime and (i) under its national law the state has
jurisdiction; (ii) the case is being duly investigated or prosecuted by its authorities
or these authorities have decided, in a proper manner, not to prosecute the person
concerned; and (iii) the case is not of sufficient gravity to justify action by the Court
(ex Article 17). In addition, the Court (iv) may not prosecute and try a person who has
already been convicted of or acquitted for the same crimes, if the trial was fair and
proper (Articles 17(c) and 20).
The Court is instead authorized to exercise its jurisdiction over a crime, even if a
case concerning the crime is pending before national authorities, and thus to override
national criminal jurisdiction, whenever: (i) the state is unable or unwilling genuinely
to carry out the investigation or prosecution, or its decision not to prosecute the per­
son concerned has resulted from its unwillingness or inability genuinely to prosecute
that person; and (ii) the case is of sufficient gravity to justify the exercise of the C ourt’s
jurisdiction.
The question, of course, arises as to what is meant by ‘unwillingness’ or ‘inability’
of a state to prosecute or try a person accused or suspected of international crimes.
These two notions are spelled out in Article 17(2) and (3). A state may be considered
as ‘unwilling when: (i) in fact the national authorities have undertaken proceedings
for the purpose of shielding the person concerned from criminal responsibility; or
(ii) there has been an ‘unjustified delay’ in the proceedings showing that in fact the
authorities do not intend to bring the person concerned to justice; or (iii) the proceed­
ings are not being conducted independently or impartially or in any case in a m anner
showing the intent to bring the person to justice. A state is ‘unable’ when, chiefly on
account of a total or partial collapse of its judicial system, it is not in a position: (i) to
detain the accused or to have him surrendered by the authorities or bodies that hold
him in custody; or (ii) to collect the necessary evidence; or (iii) to carry out criminal
 344 INTERNATIONAL CRIMINAL LAW

proceedings. One should also add cases where the national court is unable to try a per­
son not because of a collapse or m alfunctioning of the judicial system, but on account
of legislative impediments, such as an amnesty law, or a statute of limitations, making
it impossible for the national judge to commence proceedings against the suspect or
the accused.
Complementarity applies not only with regard to the states parties to the ICC Statute
but also with respect to states not parties (see Article 18(1)). Thus, for instance, if the
national of a state not party (A) has committed an international crime on the territory
of a state party (B) and then escapes to another state not party (C), and this state asserts
its jurisdiction on the ground that the crime is provided for in an international treaty
and the suspect is present on its territory (the forum deprehensionis principle) or on
the ground of universality, the ICC may not exercise jurisdiction if it is proved that
state C is willing and able to conduct a proper and fair trial.
Complementarity applies whatever the trigger mechanism of the C ourt’s proceedings;
that is, both when the case (i) has been brought to the Court by a state party (Articles
13(a) and 14); or (ii) has been initiated by the Prosecutor motuproprio, and the Prosecutor
has been authorized by the Pre-Trial Chamber to commence a criminal investigation
(Articles 13(c) and 15), and when (iii) it is the UN Security Council that has referred to
the Court a situation in which one or more of [... the] crimes [falling under the Court’s
jurisdiction] appears to have been committed’ (Articles 13(b) and 52(c)).

16.4 THE NUREMBERG SCHEME VERSUS

THE ICC SCHEME

It may useful to discuss the proper role to assign to an international criminal court vis-a-
vis national courts. It would seem that two major models have so far been worked out.
One is that adopted in Nuremberg.10 Under this scheme an international court was
entrusted with the task of dealing with the major leaders accused of international crimes,
whereas national courts were called upon to handle the criminal offences of minor cul­
prits (after the Second World War, German courts were requested to adjudicate upon
crimes committed by Germans against other Germans, while national courts of the
Allies pronounced on crimes perpetrated by Germans against foreign nationals).
Interestingly, the two ad hoc International Criminal Tribunals set up for the former
Yugoslavia and Rwanda respectively, were based, as pointed out above (16.2) on the
principle of their primacy over national courts because initially they were intended as
a substitute for the national courts of states deemed unable or unwilling to dispense

10 This scheme was first drawn up in 1944 by an American officer (Lieutenant Colonel Murray C. Bernays,
chief of the Special Projects Office of the Personnel Branch at the US War Department), was then refined by
other staff, subsequently upheld by the US Secretary of War Henry L. Stimson, and finally accepted by the
other three Great Powers in London in 1945.
 INTERNATIONAL VERSUS NATIONAL JURISDICTION 345

justice. Nonetheless, from the outset the ICTR concentrated on military and civil­
ian leaders, leaving to Rwandan courts the task of trying m inor offenders. The ICTY,
whose first Prosecutor did not envisage at the prosecutorial level a distinction between
major and m inor offenders, gradually moved towards the Nuremberg scheme, in that
(i) it has now firmly decided to concentrate on major cases, concerning political and
military leaders or other major defendants;11 and (ii) it has increasingly asked national
courts of the states concerned to try lesser accused.112
A different division of labour is provided for in the Statute of the ICC. As emphasized
above (16.3), all crimes may be brought before national courts, whatever the magnitude
of the crime or the status, rank, or importance of the accused. The ICC steps in only when
such courts prove unable or unwilling to do justice, and provided the case is of sufficient
gravity’ to justify action by the Court (ex Article 17(l)(d) of the Court s Statute).
It would seem that the Nuremberg model still has much merit. It is logical and
consistent for very serious international crimes allegedly perpetrated by leaders to be
adjudicated by an international court offering all the advantages that will be outlined
infra (see 21.2). Trials held in the country where the crime has been committed or
where the victims or their relatives live, may arouse animosity and conflict; by the
same token, it may turn out to be difficult forjudges to remain impartial. In particular,
when crimes are very serious and large scale (think, for example, of grave instances of
genocide or crimes against humanity) and have been committed either by the central
authorities or with their (tacit or explicit) approval or acquiescence, it will be difficult
for a national court to prosecute the alleged planner or perpetrators, unless there is a
change in government. However, even if this is the case, there may be a risk of witch
hunting’ or of using the criminal courts for settling political accounts, a situation

11 At present the Tribunals’ Bureau (consisting of the President, the Vice-President, and the presiding
Judges), exercises a screening of the indictments, so as to exclude those that deal with minor defendants.
Under Rule 28(A), ‘On receipt of an indictment for review from the Prosecutor, the Registrar shall consult
with the President. The President shall refer the matter to the Bureau which shall determine whether the
indictment, prima facie, concentrates on one or more of the most senior leaders suspected of being most
responsible for crimes within the jurisdiction of the Tribunal. If the Bureau determines that the indictment
meets this standard, the President shall designate one of the permanent Trial Chamber Judges for the review
under Rule 47. If the Bureau determines that the indictment does not meet this standard, the President shall
return the indictment to the Registrar to communicate this finding to the Prosecutor.’
In res. 1329 (2000) the UN SC had taken note of the position expressed by the International Tribunals that
civilian, military and paramilitary leaders should be tried before them in preference to minor actors’.
12 See among other things the recently amended Rule llfois of the RPE. Under this provision, concerning
‘Referral of the Indictment to Another Court’:
(A) If an indictment has been confirmed, irrespective of whether or not the accused is in the custody of
the Tribunal, the President may appoint a Trial Chamber for the purpose of referring a case to the
authorities of a state (i) in whose territory the crime was committed; or (ii) in which the accused was
arrested, so that those authorities should forthwith refer the case to the appropriate court for trial
within that state.
(B) The Trial Chamber may order such referral proprio motu or at the request of the Prosecutor, after hav­
ing given to the Prosecutor and, where applicable, the accused, the opportunity to be heard.
(C) In determining whether to refer the case in accordance with paragraph (A), the Trial Chamber shall,
in accordance with Security Council Presidential Statement S/PRST/2002/21, consider the gravity of
the crimes charged and the level of responsibility of the accused.
 346 INTERNATIONAL CRIMINAL LAW

which cannot contribute to the fair and impartial administration of justice. Hence,
international courts are by definition better suited to pronounce upon large-scale and
grave crimes allegedly perpetrated by political or military leaders. For such cases the
rule of complementarity laid down in the Statute of Rome may appear to be question­
able. However, since the draftsmen of the Statute have opted for that model, one can
only hope that the Court will interpret and apply the relevant rules of the Statute in
such a way as to assert the C ourt’s jurisdiction whenever cases in that category are
brought before the Court.

16.5 THE NEED FOR STATE COOPERATION

For international criminal tribunals, state cooperation is crucial to the effective­
ness of judicial process. Only other bodies or entities, that is, national authorities or
international organizations, can enforce the decisions, orders, and requests of such
criminal tribunals. Unlike domestic criminal courts, international tribunals have no
enforcement agencies at their disposal: without the intermediary of national author­
ities, they cannot seize evidentiary material, compel witnesses to give testimony,
search the scenes where crimes have allegedly been committed, or execute arrest war­
rants. For all these purposes, international courts must turn to state authorities and
request them to take action to assist the courts’ officers and investigators. W ithout the
help of these authorities, international courts cannot operate.
Admittedly, this holds true for all international institutions, which need the support
of states to be able to operate. However, international criminal courts are much more
in need of such support, and more urgently, because their action has a direct impact on
individuals living on the territory of sovereign states and subject to their jurisdiction.
Trials must be expeditious; evidence must be collected before it becomes stale; inter­
national tribunals must be able to summon witnesses to testify at short notice.

16.6 MODELS OF COOPERATION

In deciding upon how to regulate the cooperation of states with an international crim ­
inal court, the framers of the Statute of international criminal tribunals may choose
between two possible models: the ‘horizontal’ and the vertical’ model, as the ICTY
Appeals Chamber put it in Blaskic (subpoena) (§§47 and 54).
Under the former model the relations between states and the international court
are shaped on the pattern of interstate judicial cooperation in criminal matters. As
aptly stated by a distinguished commentator,13 interstate judicial cooperation and

13 See B. Swart, ‘International Cooperation and Judicial Assistance—General Problems’, in Cassese,

Gaeta, and Jones, ICC Commentary, II, 1590-2. The present author relies heavily on the excellent paper by
Swart (1592-1605).
 INTERNATIONAL VERSUS NATIONAL JURISDICTION 347

assistance shows the following hallmarks: (i) it has a consensual basis, being grounded
on treaty relations; (ii) treaties normally require that the offence for which extradi­
tion is requested be considered such in both the requesting and the requested state;
(iii) often treaties provide for exceptions to extradition, relating to certain offences
(for example, political or fiscal offences) or to some categories of persons (for instance,
nationals of the requested state), or for some sentences (for instance, extradition is
often excluded when the requesting state may impose the death penalty); (iv) extradi­
tion may be refused also when the requested state can assert its jurisdiction over the
offence; (v) judicial assistance or cooperation may normally be refused on grounds of
security, public order, overriding national interests, etc.; (vi) as a rule the collection
of evidence, search, and other investigatory actions requested by a state may not be
undertaken by the authorities of that state, but only by those of the requested state,
through the system o f‘letters rogatory’; normally a foreign country may not enter into
direct contact with individuals subject to the sovereignty of the requested state.
If this model is applied to international courts, it follows that the court has no super­
ior authority over states except for the legal power to adjudicate crimes perpetrated
by individuals subject to state sovereignty. Otherwise, the international court cannot
in any way force states to lend their cooperation, let alone exercise coercive powers
within the territory of sovereign states.
The second model could be termed ‘vertical’ or ‘supra-state’. It departs from the
traditional setting of state-to-state judicial cooperation, where by definition all cooper­
ating states are on an equal footing. This more progressive scheme presupposes that
the international judicial body is vested with sweeping powers not only vis-ä-vis indi­
viduals subject to the sovereign authority of states, but also towards states themselves.
The international court is now empowered to issue binding orders to states and, in
case of non-compliance, may set in motion enforcement mechanisms. W hat is no less
im portant, the international court is given the final say on evidentiary matters: states
are not allowed to withhold evidence on grounds of self-defined national interests or
to refuse to execute arrest warrants or other courts’ orders. In short, the international
court is endowed with an authority over states that markedly differentiates it from
other international institutions.

16.7 COOPERATION OF STATES UNDER THE ICTY

AND ICTR SCHEME

The ICTY and the ICTR incarnate the coercive ‘supra-state’ model, both because they
have the Chapter VII authority of the Security Council behind them, and on account of
the practice developed by judges. Indeed, the law of the ad hoc international Tribunals
as it concerns state cooperation is largely judge-made. Article 29 of the ICTY Statute,
and the corresponding Article 28 of the ICTR Statute, simply provide in a general way
348 INTERNATIONAL CRIMINAL LAW

that ‘states shall cooperate with the International Tribunal’ and ‘shall comply without
undue delay with any request for assistance or an order issued by a Trial Chamber’.
However, the specific practice relating to arrest warrants and orders for transfer of an
accused, requests for assistance, the question of the persons or entities to whom sub­
poenas may be addressed, that of the breadth and specificity of the ‘sanctions’ against
non-cooperative states, and many related questions, were left to the judges to define.
This happened in due course in the Blaskic (subpoena) case. A TC had issued subpoe­
nas to Croatia and one of its senior ministers. On appeal the AC overturned the deci­
sion, confining subpoenas to individuals acting in a private capacity, while allowing
binding orders to be directed to states.
It is apparent from the provisions of the ICTY Statute as developed and spelled out
by the judges in the aforementioned case that the relations between the ICTY (and the
ICTR) and states are shaped as follows: (i) the Statutes of the Tribunals impose upon
states an obligation to cooperate; this obligation is at the same time sweeping (for it
embraces any matter where the Tribunal may need the cooperation of a state), and
strict (for it is assisted by the sanctioning powers of the SC in case of non-compliance
by a state); (ii) it follows from that obligation that states are not allowed to rely upon
such traditional clauses for refusing cooperation or extradition as ‘double crim inal­
ity’, political offence, nationality of the person requested for surrender, etc.; (iii) the
Tribunal is endowed with broad and binding powers, for it can issue binding orders to
states (for the handing over of evidence, arrest of suspects, etc.), or subpoenas to indi­
viduals acting in a private capacity; (iv) although states may invoke national security
concerns as a ground for refusing the transmission of documents and other evidence,
this is subject to strict limitations, and the Tribunal may have the final say on the mat­
ter (see Blaskic (subpoena), §§61-9); (v) the collection of evidence m aybe carried out
by the authorities of the relevant state, but the Tribunal’s Prosecutor is authorized to
undertake investigations and gather evidence directly (that is, without going through
the official channels) on the territory of the states of the former Yugoslavia, as well as
on the territory of those states which have passed implementing legislation author­
izing such Tribunal’s activity (see Blaskic (subpoena), §§53-4); (vi) in case of non-
compliance by a state with the obligation to cooperate stemming from the Statutes,
the Tribunals may make a judicial finding of failure to cooperate, and the President
is then authorized to submit it to the SC. This organ, according to Blaskic (subpoena),
§§36-7, is then legally bound by that finding; that is, it may not contest that that par­
ticular state has indeed failed to cooperate, as found by the Tribunal (the SC is, how­
ever, free to take, or not to take, sanctions); if it is an individual who fails to cooperate,
the Tribunal may hold him in contempt and initiate contempt proceedings, even in
absentia (Blaskic (subpoena), §§57-60).
The mechanisms the AC set out in Blaskic (subpoena) have subsequently been
codified in ICTY Rule 54(D-I).
r INTERNATIONAL VERSUS NATIONAL JURISDICTION 349

16.8 COOPERATION OF STATES UNDER

THE ICC SCHEME
In contrast to the ICTY and ICTR, which are creatures of the Security Council moulded
into their present shape in large part by the judges, states have had the opportunity, in
drawing up the ICC Statute (an international treaty that states may or may not ratify,
given its consensual nature), to express themselves, in no uncertain terms, about how
they wish international justice to work, and they have adopted a mostly state-oriented
approach, which however, as has been rightly been pointed out, may be held to be ‘a
m ixture of the “horizontal” and the “vertical” [model]’.14
A few points are relevant in this regard. First, the Statute does lay down in Article
86 a general obligation to cooperate. However, this obligation essentially serves as a
general statement that is specified and spelled out in a number of specific provisions;
as has been stated, in the ICC Statute ‘the choice has been made to list the specific
obligations of the states parties exhaustively and to indicate their scope and contents
as precisely as possible’.15 Plainly, this specific enumeration is intended to restrict as
much as possible the judicial power of interpretation of the duty to cooperate, and by
the same token to lay down extensive ‘legislative’ safeguards for states.
Secondly, the ICC Statute does not specify whether the taking of evidence, exe­
cution of summonses and warrants, etc. is to be undertaken by officials of the
Prosecutor with the assistance, when needed, of state authorities, or whether instead
it will be for state enforcement or judicial authorities to execute those acts at the
request of the Prosecutor. Judging from the insistence in the Statute on the need to
comply with the requirements of national legislation, however, the conclusion seems
warranted that the framers of the Statute intended the latter. It would seem that this
conclusion is borne out by the implementing legislation so far adopted by some of
the states parties.16
Thirdly, both the principle of complementarity (with the ensuing power of the states
concerned to decide whether they intend to exercise their jurisdiction over a crime)
and the general right of states to challenge the C ourt’s jurisdiction may create obstacles
to, or slow down or even hamper, states’ cooperation and the C ourt’s jurisdiction.

14 B. Swart, cit., at 1594.

15 Ibid., at 1595.
16 See, for instance, the Canadian Crimes Against Humanity and War Crimes Act 2000, the Norwegian
Act on Implementation of the Rome Statute of 15 June 2001, the Swiss Federal Law on Cooperation with
the ICC of 22 June 2001, and the French Law on the same matter of 26 February 2002. See, however, Article
2(1) of Part II of the United Kingdom International Criminal Court Act 2001 (C.17) whereby the British
Secretary of state transmits any ICC request for arrest or surrender to ‘the appropriate judicial officer’, with­
out exercising any scrutiny (see also Article 5(5) and (8)).
 350 INTERNATIONAL CRIMINAL LAW

Fourthly, the possible surrender of persons to the Court is subject to a condition

typical of interstate judicial cooperation: the principle of speciality, laid down in
Article 101(1), whereby:
A person surrendered to the Court under this Statute shall not be proceeded against, pun­
ished or detained for any conduct committed prior to surrender, other than the conduct
or course of conduct which forms the basis of the crimes for which that person has been
surrendered.

Fifthly, in case of competing requests for surrender or extradition, i.e. a request

for arrest and surrender of a person, emanating from the Court, and a request for
extradition from a state not party, the request from the Court does not automatically
prevail. Under Article 90(6) and (7), a state party may decide between compliance
with the request from the Court and compliance with the request from a non-party
state with which the state party is bound by an extradition treaty. This seems odd, for
one would have thought that the obligations stemming from the Rome Statute should
have taken precedence over those flowing from other treaties. Arguably, this prior­
ity would follow both from the primacy of a Statute establishing a universal criminal
court over bilateral treaties (or multilateral treaties binding on a group of states) and
from the very purpose of the Statute, to administer international justice in the inter­
est of peace. It seems instead that the Statute, faced with the dilemma of international
versus national justice, has left the option to the relevant states.
Furthermore, as regards the protection of national security information, the Statute
substantially caters to state concerns by creating a national security exception to
requests for assistance. Article 93(4) provides that ‘a state Party may deny a request
for assistance, in whole or in part, only if the request concerns the production of any
documents or disclosure of evidence which relates to its national security’. Admittedly,
Article 72, to which this provision refers, does envisage a complex mechanism designed
to induce a state invoking national security concerns to disclose as much as possible of
the information it wishes to withhold. This mechanism is largely based on the Blaskic
(subpoena) decision of the ICTY Appeals Chamber. However, the various stages of this
mechanism are turned in the Statute into formal modalities that will be cumbersome
and time-consuming.17 In addition, in Blaskic (subpoena) the emphasis was on the
obligation of states to disclose information; only in exceptional circumstances were

17 Indeed, Article 72 (‘Protection of national security information’) establishes a three-step procedure

when a state—or individual—invokes national security. Article 72 is triggered when a state is of the opinion
that ‘disclosure of information [requested by the Court or Prosecutor] would prejudice its national security
interests’. First, cooperative means are employed to reach an amicable settlement, e.g. modification of the.
request, a determination by the Court of the relevance of the information sought or agreement of conditions
under which the assistance could be provided. Secondly, if cooperative means fail, and the state decides
against disclosure, it must notify the Court or Prosecutor of the specific reasons for its decision, unless a
specific description of the reasons would itself necessarily result in a prejudice to the state’s national security
interests. The Court may then hold further consultations on the matter, if need be ex parte and/or in camera.
The third step, in the event that the state is found not to be complying with its obligations, is for the Court to
refer the matter to the Assembly of States Parties or, if the Security Council originally referred the matter to
the Court, to the Security Council.
 INTERNATIONAL VERSUS NATIONAL JURISDICTION 351

states allowed to resort to special steps for the purpose of shielding that information
from undue disclosure to entities other than the Court. In Article 72 emphasis is
instead laid on the right of states to deny the C ourt’s request for assistance.
Finally, in the event of failure of states to cooperate, Article 87(7) provides for
the means substantially enunciated by the ICTY in the AC decision in Blaskic (sub­
poena), namely, ‘the Court may make a finding to that effect and refer the matter to
the Assembly of states Parties or, where the Security Council referred the matter to the
Court, to the Security Council’. However, the ICC could arguably have gone further
and articulated the consequences of a C ourt’s finding of non-cooperation by a state.
The Statute could have specified that the Assembly of States Parties might agree upon
countermeasures, or authorize contracting states to adopt such countermeasures, or,
in the event of disagreement, that each contracting state might take such counter­
measures. In addition, it would have been appropriate to provide for the possibility of
the SC’s stepping in and adopting sanctions even in cases where the matter had not
been previously referred by this body to the Court: ope fails to see why the SC should
not act upon Chapter VII if a state refuses to cooperate and such refusal amounts to a
threat to the peace, even in cases previously referred to the C ourt by a state or initiated
by the Prosecutor proprio motu. Of course, the ICC Statute does not exclude this pos­
sibility, but it would also have been a good idea expressly to include it.

16.9 THE QUESTION OF SURRENDER

OF NATIONALS
The constitution or the laws of many civil law countries lay down the principle that
nationals may not be extradited for prosecution abroad.18 This principle is clearly the
remnant of a bygone era when states intended to protect their nationals as much as
possible against any foreign interference. It is a typical expression of the Westphalian
international community of sovereign states that distrust one another and lack any
common values of universal scope and purport. In today’s international community,
where respect for hum an rights counts among the universal values that all states are
required to abide by, this rule has become a relic of the past. Provided that the rights
of the accused are safeguarded and the trial is fair and expeditious, why should a
state have the right to refuse the extradition of one of its nationals to another state
on the territory of which the individual has committed a crime? The outdated nature
of this legal tradition has come to light in the course of the work of the international

18 The principle can be found, for example, in the Constitutions of Brazil (1988, revised in 1996; Article
50), the German Federal Republic (1949, Article 16(2), now amended to allow the handing over of Germans
to the ICC and other members of the European Union), the Federal Republic of Yugoslavia (1992, Article
17-3), Poland (1997, Article 55-1) and Slovenia (1991, revised in 2000, Article 47). The principle finds legis­
lative recognition in France, in Article 3 of the Law of 10 March 1927, as well as in most bilateral treaties
concerning mutual judicial assistance.
 352 INTERNATIONAL CRIMINAL LAW

criminal tribunals. Often, when the Prosecutor in The Hague requests a state to hand
over a national accused of international crimes, the state takes refuge behind a con­
stitutional provision forbidding extradition. This refusal is all the more absurd in
such cases, because it leads to the effective impunity of the persons in question, by
protecting them against crim inal prosecution for very serious crimes which transgress
universal values.
The reaction of international tribunals has been twofold. First of all, they have
held that, by virtue of a well-established principle of international law, states may
not invoke their national legislation, even of constitutional rank, to evade an inter­
national obligation.19 This objection is, of course, traditional. The other is, however,
innovative. It posits that, at the very most, the constitutional rules in question only
apply to relations between sovereign states, and not to relations between a state and an
international court. The former relations are based on the principle of formal equality,
whereas the latter are hierarchical in nature. Thus, while one may speak of extradition
of the accused from one state to another, it is more appropriate to speak of transfer of
the accused from a state to an international criminal tribunal.20 Furthermore, given
that the rights of the accused are fully respected before international judicial bodies,
the protection of the national state no longer makes any sense. International judges
have thus introduced a new legal concept, dissociated from obsolete principles, where
the vision of the international com m unity propounded by Immanuel Kant prevails
over that extolled by Hobbes.

19 For this argument see, although on different matters, the Order of the ICTY President in Blaskic
(3 April 1996, §7), the annual report for 1996 of the ICTY President to the General Assembly (A/51/292,
16 August 1996), at §182, as well as Milosevic (decision on preliminary motions) (§47).
20 The argument is, for example, set forth in the ICTY President’s report to the UN General Assembly
on the activities of the Tribunal during 1977 (UN Doc. A/RES/52/375,18 September 1997, §§186-9; also in
ICTY Yearbook 1997, at 145).
 17
THE ADOPTION OF THE
ESSENTIAL FEATURES OF THE
ADVERSARIAL SYSTEM

17.1 THE ADVERSARIAL VERSUS THE

INQUISITORIAL SYSTEM: GENERAL

Most national legal systems based on the Romano-Germanic (or civil law’) tradition
tend to apply the inquisitorial system, while in common law countries the adversar­
ial model is preferred.1 These two models do not constitute watertight categories. In
fact neither model can ever be found in its ‘pure’ form anywhere, because in practice
historical circumstances, local traditions, and the influence of the other model have
led to an adaptation of each system, to take account of requirements specific to each
country. Therefore, the models at issue must be taken to be some sort of abstract intel­
lectual constructs’ similar to the ‘ideal types’ propounded by Max Weber (who used
this notion for such categories as feudalism, mysticism, etc.).12

1 There are differences in terminology, some authors speaking of ‘adversarial’ others of accusatorial’
systems for common law countries. See the views of such distinguished authors as M. R. Damaska (The
Faces ofJustice and State Authority—A Comparative Approach to the Legal Process (New Haven and London:
Yale University Press, 1986), at 3-6, 69, 88, 97-8) and W. R. LaFave, J. H. Israel, and N. King (Criminal
Procedure, 3rd edn (St Paul, Minn.: Hornbook, 2000), at 30-3). According to the latter commentators: ‘The
key to an adversary system is the division of responsibilities between the decision-maker and the parties.
An adversarial system of adjudication vests decision-making authority, both as to law and fact, in a neutral
decision-maker who is to render a decision in light of the materials presented by the adversary parties [...]
The adversary model gives to the parties the responsibility of investigating the facts, interviewing possible
witnesses, consulting possible experts, and determining what will or will not be told [...] The judge and jury
are then to adjudicate impartially the issues presented by the opposing presentations’ (at 31); ‘The American
criminal justice process is designed to be accusatorial as well as adversarial. The concepts of adversarial
adjudication and accusatorial procedure complement each other, but are not virtual equivalents. The adver­
sarial element assigns to the participants the responsibility for developing the legal and factual issues of the
case, while the accusatorial element allocates burdens as between the parties with respect to the adjudication
of guilt. An accusatorial procedure requires the government to bear the burden of establishing the guilt of
the accused, as opposed to requiring the accused to bear the burden of establishing his innocence’ (at 33).
2 See, for instance, M. Weber, ‘Religious Rejections of the World and Their Directions’ (1915), in
H. H. Gerth and C. Wright Mills (eds), From Max Weber—Essays in Sociology (London: Routledge & Kegan
Paul, 1970), 323ff. (‘Such constructions make it possible to determine the typological locus of a historical
 354 INTERNATIONAL CRIMINAL LAW

According to anthropologists, the adversarial system was the first substitute for
private vengeance. Social groups agreed that the wronged person must no longer use
force to do justice by himself. He (as well as, later on, his relatives or members of his
clan or group) could instead accuse the alleged culprit. An arbiter, referee, or judge was
selected to hear the evidence. Each party (the accuser and the accused) autonomously
searched for and collected all the available evidence, respectively to support the charges
and to rebut them to prove the innocence of the accused. The ‘judges’ (often a groups
of persons from the community where the alleged crime had been committed) simply
acted as referees in the contest between the parties. The proceedings were public and
oral. One reason for this was that most participants were illiterate. The trial normally
took place before the popular assembly, or at any rate in a public place. Thus, under
this system the initiation of criminal action was left to the aggrieved party; no public
official had the right to institute proceedings. This system can be found in the Greek
polis3 as well as in Rome, in the Republican period. Later on some typical features
of the inquisitorial model were incorporated: notably, public officials (the prosecu­
tor) replaced the private accuser. In addition, the new notion of popular justice led to
the firm establishment of juries proper. At present, the model can be found, in vari­
ous forms, in many countries including England, the USA, Canada, Australia, New
Zealand, and some African countries, as well as in Ireland and (since 1988) in Italy.
The inquisitorial model emerged in ancient Rome, at the time of the Empire, and
was in full bloom in the Middle Ages, at least from the thirteenth century. The kings
and princes gradually adopted the system the Catholic Church used as its standard
method for investigating and prosecuting offences against clergymen and, later on,

phenomenon. They enable us to see if, in particular traits or in their total character, the phenomena approxi­
mate one of our constructions: to determine the degree of approximation of the historical phenomenon to
the theoretically constructed type. To this extent, the construction is merely a technical aid which facilitates
a more lucid arrangement and terminology’ (at 324).)
3 The trial of Socrates, in 399 B.C. is indicative of the system prevailing in the Greek polis. He was accused
by three Athenian citizens (Meletus, Anytus, and Lycon)) of three crimes (not recognizing the gods wor­
shipped in Athens, introducing new deities, and corrupting the young). The prosecution was conducted by
the three accusers, whereas the defence was made by Socrates himself: there was no professional prosecu­
tor or defence counsel. Socrates was tried by 501 citizens (chosen by lot out of the roughly 25,000 citizens)
who acted as jurors, no separate judge presiding over the proceedings. The proceedings were divided up
into two parts: the establishment of whether Socrates was guilty or innocent, and the issuance of the pen­
alty. After the first round of speeches by the accusers and the defendant, the jury took a vote on the issue
of guilt or innocence and decided by 280 votes to 221 that Socrates was guilty of the crimes with which he
had been charged. Then Meletos requested the death penalty, whereas Socrates suggested as ‘an appropri­
ate penalty which is strictly in accordance with justice’ that he be given ‘free maintenance at the state’s
expense’. After the jury decided to uphold Meletos’ request and sentenced him to death, Socrates made a
third speech, where he explained that death was not an evil. Whereas normally sentences were carried out at
once, that of Socrates could not be executed forthwith, for the day before Socrates’trial was the first day of an
important ceremony (the Annual Mission of the state galley to Delos to commemorate the deliverance of the
Cretan Minotaur by Theseus). As the absence of the galley lasted a month, Socrates awaited his execution in
prison, which was then carried out by having him drink hemlock. See Plato, Apology, in Plato, The Last Days
o f Socrates (ed. by H. Tredennick, Penguin,1969) 45-76; Xenophon, ‘Socrates’ Defence’, in Xenophon, in
R. Waterfield, (ed.) Conversations of Socrates (Penguin, 1990), 41-67. See also C. Mosse, Leproces de Socrate
(Bruxelles; Editions Complexe, 1987), 89-114.
 THE ADOPTION OF THE ESSENTIAL FEATURES OF THE ADVERSARIAL SYSTEM 355

persons accused of heresy. Under this model investigations were conducted in secret by
an official who questioned the suspect, the victim, and any witness, recorded in writing
their statements and then decided on the guilt or innocence of the person; thus, the
same person investigated the offence and adjudicated it. Although it paid scant respect
to the rights of the suspect or of the accused, in one respect this system was a more
advanced step than the adversarial system on the path towards criminal justice: it was
based on the notion that justice was not a private business but must be administered by
‘public’ officials, whose principal task was to collect evidence to establish whether the
accused was guilty. Initially inquisitorial proceedings were secret and written, for they
were ru n by cultured functionaries and based on evidence collected by the investigat­
ing judge. At a later stage it was provided that the investigating judge should make his
own pronouncement on the charges and later submit the file to a court.
At present, after the introduction of many improvements borrowed from the
adversarial system, including emphasis on the rights of the accused, the system
prevails in such countries as France, Belgium, Spain, all Latin American countries,
French-speaking African countries, China, Japan, and so on.

17.2 HOW THE TWO MODELS WORK:

A COMPARISON

I shall now try succinctly to set out the main traits distinguishing the two models.

17.2.1 I N IT IA T IO N O F IN V E S T IG A T IO N , P R O S E C U T IO N ,
A N D T R IA L P R O C E E D IN G S

As stated above, in countries that adopted the adversarial system, initially private citi­
zens (normally the victim or his relatives) set in motion trial proceedings. Gradually,
however, the centralized organs, or at any rate public officials, that is the police, were
entrusted with the task of investigating criminal offences. Proceedings may be ini­
tiated by the prosecutor, who normally enjoys discretion (for instance, in England
and Wales); that is, may choose the cases he considers worthy of being tried in court.
Once proceedings are started, the victim or other private citizens do not play any role
(except as possible witnesses).
In inquisitorial systems, investigations are set in motion either by the police
(normally a special branch of enforcement officials, called judicial police (‘police
judiciaire)’ and subordinate to the prosecutor) or by the victim (who reports to the
police), or autonomously by the public prosecutor as soon as a criminal offence is
reported to him. In such countries as Germany (subject to some exceptions) and Italy,
the prosecutor is duty bound to begin investigations as soon as he becomes cognizant
of the possible commission of a criminal offence. If after investigating he considers
 356 INTERNATIONAL CRIMINAL LAW

that there are sufficient elements for prosecuting, he is legally bound to prosecute; in
other words he does not enjoy any discretion (such discretion does, however, exist in
other countries, for instance in France and Belgium). In some countries, for instance
in France, proceedings may be started by the victim or even by private organizations
(trade unions, associations, etc.) if they claim to have been prejudiced by the alleged
criminal offence; they normally undertake the so-called ‘constitution departie civile,
by which they claim compensation from a criminal court if the court finds the accused
to be guilty. Bringing a civil action before a criminal court serves two purposes:
‘to obtain a ruling on the guilt of the person and to obtain compensation for the
damage suffered’.4

17.2.2 H O W E V ID E N C E IS G A T H E R E D

In the adversarial model each party (prosecution and defence) gathers the evi­
dence autonomously on its own behalf by means of investigators, although often the
prosecution is also under the obligation to look for exculpatory evidence. In general,
each party must disclose its evidence to the other party before the hearings. More
specifically, the prosecutor must hand over the evidence to the defence before the case
for the prosecution starts (so-called ‘discovery’), while, at the end of the case for the
prosecution and prior to presenting its case for the accused, the defence may be under
an obligation to disclose the evidence it has gathered.
In the inquisitorial system, after the prosecutor has initiated proceedings against a
suspect and collected the evidence indicating that he may have committed an offence,
an investigating judge (juge d ’instruction), acting in the interest of the whole com­
m unity as an ‘organ of justice’, gathers the evidence for both the prosecution and
the defence by questioning witnesses, ordering the search of premises, the seizure of
documents, etc. Generally speaking, this phase of the proceedings is in writing (not
oral), in that the pieces of evidence and the decisions of the investigating judge are put
together so as to constitute a case file (dossier de la cause). In addition, the procedure
is not public, since all occurs in camera mainly to keep at bay the public at large: wit­
nesses are not able to become cognizant of the testimony given by other witnesses,
and decisions rendered by the investigating judge are not made public. Nevertheless
the suspects as well as civil petitioners (parties civiles), if any, may be apprised of the
unfolding of the proceedings through their counsel to whom the case file is made
known at any stage.
The investigating judge may then either dismiss the charge and close the case or, if
he considers that there are reasons for believing that a prima facie case has been made
out and the case is ‘ripe’ for trial, he may turn over to the court and the parties the case
file (dossier de la cause) containing all the results of pre-trial investigations and the
evidence collected, together with all the decisions he has made.

4 See V. Dervieux, ‘The French System’, in Delmas-Marty and Spencer, European Criminal Procedures,
at 227.
 THE ADOPTION OF THE ESSENTIAL FEATURES OF THE ADVERSARIAL SYSTEM 357

The scheme hinging on the investigating judge was introduced by Napoleon in

France in the code of criminal procedure adopted in 1808. The major advantage of
this system is that an independent judge searches for the evidence in the interest of
justice, thereby alleviating the task of the suspect or defendant, who, when indigent or
at any rate not well-off, may not afford to pay for competent investigators and lawyers.
Thus, the system tends to protect impecunious persons. However, in addition to the
proceedings before the investigating judge often being excessively lengthy, the princi­
pal drawback of this scheme is that the exercise by such judge of his extensive powers
(including that of seizing evidence or ordering searches and in particular the power
to place suspects in police custody) may lead to abuses. Indeed, that judge may be
inclined to put pressure on a suspect by deciding to detain him in police custody for
the purpose of questioning. The investigating judge was therefore abolished in some
countries such as Germany (in 1974) and Italy (in 1988). In France since 15 June 2000
the investigating judge has been deprived of the power to decide whether to remand
a suspect in custody or grant him bail. It is now an independent judge (called juge des
libertes et de la detention) who decides on remand in custody. He also rules on any
request by a defendant held in custody to be set free.

17.2.3 C O M P O S IT IO N O F T H E C O U R T

In the adversarial model hearings are normally presided over by a judge, with the
guilty or not guilty verdict being decided by a jury consisting of people elected, or
chosen or at least selected from the general public. Thus, in common law countries
proceedings are divided into two stages: finding of guilt (a matter on which the jury
decides) and sentencing, if a guilty verdict is handed down by the jury (a matter for the
judge). If the accused enters a guilty plea (see infra, 17.2.5), there is no trial proper but
only sentencing proceedings are conducted before a judge.
The notion of popular justice, going back to ancient Greece, was revamped in
England in the late Middle Ages, when people living in the place where the alleged
offence had been committed would be asked to answer under oath whether the person
was guilty or innocent (the underlying idea being that the neighbours, knowing the
accused, would be in a better position to appraise his guilt or innocence: justice was
administered by the representatives of the community to which the alleged culprit
belonged). Gradually there began a process of selecting those people so as to make up
a jury proper. The jury later came to be predicated on the notion that judgment should
not be arbitrarily passed by the king or by the special courts set up by him, but by the
accused person’s ‘peers’ or equals. Article 39 of the ‘Magna Carta’ (1215) proclaimed
a set of fundamental rights for ‘freemen’ (that is, for members of the aristocracy and
the middle classes and for them only, as opposed to the monarch and members of the
lower classes), including the right of every ‘freeman’ to stand trial for criminal offences
‘by the lawful judgment of his peers and by the law of the land’.5

5 In the English translation: ‘No freeman shall be arrested, or detained in prison, or deprived of his free­
hold, or outlawed, or banished, or in any way molested; and we will not set forth against him nor send against
him, unless by the lawful judgment of his peers and by the law of the land.’
 358 INTERNATIONAL CRIMINAL LAW

The thinkers of the Age of Reason and the French Revolution, adm iring the jury
system existing in England, im ported it into France and bolstered the notion by
appealing to ‘people’s justice’. Flowever, the appalling abuses of the French revolution­
ary courts led to the abandonment of this system in continental Europe (except for
such countries as Belgium) and its replacement by the system of assize courts (‘cours
d ’assises’) (see below).
In the inquisitorial model no jury proper exists. Normally the court is composed
of professional judges (one, or three, or five, depending on the degree of jurisdiction).
For some particularly serious offences, Assize Courts may be competent and they are
made up of both lay judges and professional judges. For instance, in France, under
Article 231ff. of the Code of Criminal Procedure, Assize Courts (Cours d ’assises) have
jurisdiction over ‘crimes’ or serious criminal offences (as opposed to delits or misde­
meanours). They are composed of three professional judges (including the presiding
judge) and a ‘jury’ of nine persons, chosen by lot. They pronounce upon both the facts
and the law, on guilt or innocence and, in case of conviction, on the sentence; any
decision against the accused m ust be taken by a m inim um of eight votes to four. In
many civil law systems, decisions taken by such courts are reasoned (they are drafted
by the presiding judge or any of the professional judges serving on the court along
with the lay judges). A similar system applies, for instance, in Italy, where Assize
Courts have jurisdiction over serious crimes, as provided for in Article 5 of the Code
of Criminal Procedure. They are made up of two professional judges (one of them as
a presiding judge) and a jury composed of six persons (Article 3(1) of the law no. 287
of 10 April 1951).

17.2.4 T R IA L P R O C E E D IN G S

In the adversarial system when hearings commence, no file is submitted to the court. In
general the oral nature of proceedings prevails.6The documents collected by the parties
are tendered to the court after commencement of trial and only admitted into evidence
after being introduced by a witness. Nothing is considered as ‘evidence’ until it has been
heard orally at trial. The same applies to testimonial evidence; the statements, if any,
made prior to trial proceedings by a witness usually have no probative value unless they
are scrutinized in oral testimony given in open court, and admitted into evidence.
Under the inquisitorial model the trial phase is based on the principle that writ­
ten proceedings prevail. The information collected by the investigating judge, once
recorded in the case file, becomes ‘evidence’ and is then submitted in court for scrutiny
by both parties and the judges. In other words, any piece of information, document, or
exhibit that is in the case file constitutes ‘evidence’ before being submitted to the court
in oral proceedings. In practice, witnesses testifying in court often are simply asked
whether they confirm the testimony they gave to the investigating judge.

6 In England there are provisions, quite commonly used, for introducing written statements into evi­
dence without the witness being called. See section 9 of the British Criminal Justice Act 1967.
 THE ADOPTION OF THE ESSENTIAL FEATURES OF THE ADVERSARIAL SYSTEM 359

17.2.5 G U IL T Y P L E A

In the adversarial system at the beginning of trial proceedings the court asks the
defendant whether he pleads guilty or not guilty. If the defendant pleads guilty, there
are no further trial proceedings and sentencing proceedings ensue, designed to estab­
lish the appropriate sentence. The ‘guilty plea’ procedure is sometimes the result of
‘plea-bargaining’ between the prosecution and defence counsel (or the accused): the
defendant pleads guilty in return for the reduction of charges or the promise of a
request for a lenient sentence.7
The main merit of such procedure is to arrive at a sentence acceptable to both par­
ties, and, what is even more important, to avoid trial proceedings proper. In this
way in many common law countries most criminal cases are resolved through plea­
bargaining, thus significantly reducing the workload of courts.8 In England only some
1 or 2 per cent of all cases are finally disposed of by jury trial, the routine m ethod being
summary trial in the magistrates’ courts (which normally simply record the guilty
plea entered by the accused and the sentence inflicted).9
This system does not exist in civil law countries, where trials are initiated even when
the accused admits to being guilty. However, in these cases trials are relatively short.
In addition, in inquisitorial systems the screening of cases and the consequent reduc­
tion in the num ber of cases brought to trial, is carried out by the investigating judge: it
is for this judge to select the cases where the charges are such as to make a prim a facie
case against the accused, and those which instead must be shelved, for the evidence is
flimsy or inconsistent.
Whereas, as stated above, in common law systems trial proceedings are divided into
two stages (the jury is the trier of fact while the judge decides on the sentence), in civil
law systems trial proceedings are unified, as it were: the proceedings for pronouncing
on the guilt or innocence of the defendant are, should he be convicted, combined with
those on sentencing.

7 It would seem that in the UK the majority of guilty pleas are not the result of plea-bargaining. Rather,
what usually happens is that the accused realizes that the evidence against him is overwhelming and there is
no point fighting it. Or he just wants to get the whole thing over with. There is usually ‘credit’for a guilty plea,
i.e. it is taken into account in mitigation, but that is not the same thing as plea-bargaining. Plea-bargaining
most commonly occurs in the USA with drug offences. An accused is found with a drug. He can plead guilty
to possession and the prosecution will drop the charge of possession with intent to supply (which is much
more serious). But if the accused does not plead guilty to possession, the prosecution will go ahead with
charges of both possession and possession with intent to supply. That is the ‘bargain’—dropping one (ser­
ious) charge in exchangefor a guilty plea on the other (less serious) charge.
8 See, for the United Kingdom, the details provided by A. Ashworth, The Criminal Process—An Evaluative
Study, 2nd edn (Oxford: Oxford University Press, 1998), 268-84. According to Ashworth, ‘in the magis­
trates’ courts the rate of guilty pleas is well over 90 per cent’ (at 268).
9 See J. R. Spencer, ‘Introduction’, in Delmas-Marty and Spencer (eds), European Criminal Procedures,
at 18.
 36o INTERNATIONAL CRIMINAL LAW

17.2.6 T H E P O S IT IO N O F T H E A C C U S E D

Traditionally in the adversarial model trials in absentia are prohibited. It is permitted

for the defendant not to be in court only when by his behaviour he disrupts the hear­
ings or he escapes after the beginning of trial. The logic behind this principled position
is clear: as the trial consists of a contest between two parties (the prosecutor and the
defendant), if one of them is missing no trial proper may start. In addition, as it falls to
each party to gather the evidence, supporting the charges and exculpating the accused,
respectively, if the latter is absent, no one can play that crucial role in his stead.
The defendant may be held in custody (on remand) or free on bail (there is a right
to bail’, that is, a presumption or expectation of bail, but bail is easily refused for a
repeat offender who has a history of failing to surrender to custody). That a defendant
is free is substantially justified by the fact that in this position he is in a better position
to prepare his defence and also on other basic grounds: the presumption of innocence
and the right to liberty. However, if at the end of trial the accused is found guilty
and sentenced to imprisonment, and an appeal is lodged, the convicted person often
remains in jail pending the appellate proceedings (the right to bail is forfeited upon
conviction).
While in court, the accused may only be heard (and examined and cross-examined)
if he decides to appear as a witness on his own behalf, but then he is required to take
an oath or make a solemn declaration that he will tell the truth. He enjoys the funda­
mental right to silence (he is entitled not to incriminate himself, for he is presumed
innocent). In addition, no negative inference may be drawn by the court from his
choosing to remain silent and not to give evidence: since he is entitled to a presum p­
tion of innocence, there is no requirement that the accused contribute to the discovery
of truth, this task exclusively falling to the prosecution.10
In contrast, in many systems based on the Romano-Germanic tradition trials in
absentia are admissible; that is the defendant may be tried even if he has never appeared
in court and is on the run or in hiding. (This holds true for such countries as France,
Belgium, Italy, Greece, the Netherlands, most Latin American countries, and China;
among the exceptions are Spain and Germany.) The reason why this class of trials is
not quintessentially regarded as inadmissible is twofold. First, the investigating judge
gathers evidence not only for the prosecution but also for the defence. Hence, when
trial proceedings begin, the court has exculpatory evidence available, and is therefore
in a position to evaluate both the evidence against and that in favour of the accused.
Second, in most civil law systems (including those such as Italy where the accusatorial
system has been adopted) it is considered that the public interest of the community in

10 However, in 1964 a law passed in the UK, the Criminal Justice and Public Order Act, authorized trial
courts to draw adverse inferences against defendants who, in some well-defined conditions, fail to give evi­
dence. Equally, the 5th Amendment to the US Constitution does not prevent adverse inferences being drawn
from a defendant’s silence or failure to testify.
Furthermore, in John Murray v. UK the European Court of Human Rights held that the drawing of infer­
ences from the refusal of a suspect or accused, when arrested, to give an account of his presence in a certain
place could not be regarded as ‘unfair or unreasonable in the circumstances’ (§54).
 THE ADOPTION OF THE ESSENTIAL FEATURES OF THE ADVERSARIAL SYSTEM 361

adjudicating alleged criminal offences should prevail over the right of the accused to be
present in court, at least whenever the accused voluntarily tries to evade justice (none­
theless, we will see infra, 18.5, that trials in absentia are only admissible if a set of safe­
guards are respected, as repeatedly stated by the European Court of Hum an Rights).
In inquisitorial systems normally the suspect (during investigation by the investi­
gating judge) or accused (once committed for trial) is in jail when pre-trial im prison­
ment is warranted either by the gravity of the crime or when (i) the accused is likely to
tam per with evidence; or (ii) may try to escape; or (iii) may engage again in criminal
conduct. However, even in inquisitorial systems the principle applies that pre-trial
detention must not be the general rule. Therefore release (often on bail) is provided for.
In any case, if at the end of trial the accused is convicted and enters an appeal (or an
appeal is lodged by the Prosecutor), he may be released on bail, unless he must be kept
in custody by law, but then only within certain time-limits.
W hen the defendant appears in court, he may be questioned, without taking an oath,
by the presiding judge (or by any other member of the court, often with the explicit or
tacit consent of the presiding judge), or by the parties, on any issue relevant to the trial.

17.2.7 T H E R O L E O F V IC T IM S

In the adversarial system victims may bring a private prosecution, but if the prosecu­
tor decides to take up the case and commence investigations and subsequently bring
proceedings, they no longer play a role. Victims only appear in court as witnesses. If
they want to claim compensation, they must institute civil proceedings separately,
after the criminal trial.
In contrast, as pointed out above, in the inquisitorial system victims may institute
proceedings or take part in the crim inal proceedings initiated by the prosecution,
through the so-called constitution de partie civile (application to join criminal pro­
ceedings as a civil petitioner) aimed at claiming compensation. ‘Civil petitioners’ or
civil parties’ participate in the proceedings: they have access, through their lawyers,
to the case file (dossier) during the investigation by the investigating judge (instruction
preparatoire) and may call for certain investigations; at trial, they can call evidence,
question witnesses, and set out their legal views as to the guilt of the accused. Thus, in
contrast with the adversarial system, crim inal and civil proceedings are not separate
but may be merged.

17.2.8 T H E R O L E O F T H E C O U R T

Under the adversarial scheme the judge, faced with a jury entitled to establish whether
the accused is guilty or innocent, tends to play a rather passive role, as a sort of umpire
or referee between two contending parties. W hen trial proceedings begin, the judge
does not know the results of the investigations conducted by the prosecution: he only
knows the charges preferred against the defendant, the facts as concisely set out in
the indictment, and the law the prosecution intends to invoke, as indicated in the
 362 INTERNATIONAL CRIiMINAL LAW

indictment. The judge becomes cognizant of the facts of the case only through pres­
entation of evidence in court (the same applies to the jury, which before the beginning
of trial is totally unaware of the evidentiary material available to the prosecution and
only comes to know it through its production in court).
The principal role of the judge is to conduct the hearings, uphold or reject objections
by a party to submissions of the other party, and grant or dismiss motions by the par­
ties. As stated by A. Orie,

It is logical in a common-law system, where the jury is entirely dependent on the evidence
presented to it at trial, that the trial judge be called by the parties to intervene whenever
necessary in order to prevent the information on which the jury will base its verdict being
polluted by improper elements.11

At the end of the proceedings and before the jury retires for deliberation, the judge
then sums up the relevant facts (in the UK, but not in the USA), and (in trials in many
countries, including England and Wales) the applicable law as well as possible defences.
The jury will then decide on the guilt or innocence of the accused.
Generally speaking, the court is not required actively to seek the truth but simply to
decide whether the evidence produced in court by the prosecution is sufficient to sub­
stantiate its charges and prove beyond a reasonable doubt that the defendant is guilty.
Under the inquisitorial model the court plays instead an active role in seeking the
truth. When trial proceedings begin, the court is already cognizant of both the facts
of the case and the law invoked by the parties: the court acquires this knowledge by
reading the case file (dossier d ’instruction, dossier de la cause) which, as stated above,
contains the evidence collected by the investigating judge for both the prosecution
and the accused, plus any other relevant document or record as well as all the rul­
ings made by the investigating judge. The presiding judge is a dominating figure in
the conduct of proceedings. He may question both the witnesses and the accused (of
course the parties may question the witnesses as well). In particular, the presiding
judge questions the accused from the outset and tends to ask him questions on many
occasions, each time an im portant witness gives evidence in court. Thus a continuing
dialogue between the presiding judge and the defendant unfolds during the trial. The
role witnesses play in court hearings is less im portant than in common law countries:
as pointed out above, often they are simply asked to confirm or deny in court the state­
ments made to the investigating judge. In addition, the court may call evidence prop-
rio motu (on its own initiative).
The different philosophy behind the two systems can perhaps be summed up as
follows: in the inquisitorial system the court aims to actively discover the tru th —
the defendant contributes to this discovery by answering questions in court; in an
adversarial system, instead, the defendant, chiefly through his defence counsel, plays

11 A. Orie, ‘Adversarial v. Inquisitorial Approach in International Criminal Proceedings Prior to

the Establishment of the ICC and in the Proceedings before the ICC’, in Cassese, Gaeta, and Jones, ICC
Commentary, at 1427.
 THE ADOPTION OF THE ESSENTIAL FEATURES OF THE ADVERSARIAL SYSTEM 363

in some respects a greater role in contributing, as much as the prosecution, to the

establishment of the truth, for the whole trial hinges on a contest between the parties.

17.2.9 R U L E S O N A D M IS S IO N A N D E V A L U A T IO N O F E V ID E N C E

In the adversarial system any statement made before trial or any testimony, affidavit or
exhibit collected at the stage of investigation does not per se constitute evidence proper:
it may come into evidence only at trial, once tendered by a party and if the adversary does
not object to it (or if the objection is rejected by the judge). Here again orality prevails:
a written document, an affidavit or any other object produced by a party may legally
become a piece of evidence only if, through oral examination and cross-examination,
it is admitted into evidence. It follows, among other things, that a statement made by a
witness to an investigator prior to trial is only admissible for the purpose of either cor­
roborating the testimony made by the witness at trial or, instead, of undermining his
credibility by showing the inconsistency of that testimony with the prior statement. The
truthfulness or instead mendacity of the prior statement are not important per se, but
only if used to test the veracity of the witnesses’ testimony in court.
Furthermore, there exist strict and detailed rules of evidence. In particular, hear­
say evidence is inadmissible (subject to a string of exceptions that in some common
law countries run to 40). The rationale behind the setting out of such rules is that the
evaluation of the evidence produced in court falls to the jury; that is a group of laymen
who are not normally familiar with legal technicalities and therefore do not know
what weight to give to each piece of evidence.
There are also strict rules for the order in which evidence is presented in court.
Normally the prosecution starts, making its case, and of course the defence may cross-
examine prosecution witnesses; once the prosecution case is closed, the defence may
take over; at the end the prosecution may sometimes present evidence in rebuttal, and
then the defence may present evidence in rejoinder. Closing statements by the pros­
ecution and then the defence bring the trial proceedings to an end.
The test for establishing guilt is that the accused must be found guilty of the offences
charged ‘beyond a reasonable doubt’.
In the inquisitorial system the principle of freedom in the type of evidence employed
(liberte des preuves) is upheld.12 Rules of evidence are normally flexible and general,
for it is professional judges who apply them (even in assize courts, the presence of two
or three such judges makes it possible to dispense with strict and extremely detailed
rules of evidence). Pre-trial statements or affidavits may be taken into account by the
investigating judge and held to amount to evidence if he considers that they are trust­
worthy. At trial, what ultimately matters is that the court forms its intimate convic­
tion’ as to the guilt or innocence of the defendant.

12 On the limitations of the principle in, for instance, the French system, see, however, G. Stefani,
G. Levasseur, B. Bouloc, Procedure Penale, 16th edn (Paris: Dalloz, 1996), at 34-8.
364 INTERNATIONAL CRIMINAL LAW

Also, the order of presentation of evidence is different from that prevailing in com­
mon law countries. In the inquisitorial system no strict distinction is made between
the prosecution and the defence case with regard to the order in which the evidence is
presented in court: in any case, the evidence is that contained in the case file submitted
by the investigating judge.

17.2.10 A P P E L L A T E P R O C E E D IN G S

In the adversarial system, resort to appeal proceedings is quite restricted. First of all,
the prosecutor may not appeal against an acquittal. Secondly, appellate proceedings
do not entail a retrial. The appellate court consists of professional judges, without any
jury. Normally appellate judges pronounce on issues of law (for instance, on whether
the trial judge gave wrong instructions to the jury). Facts may not be the subject of pro­
ceedings, unless the appellant claims that facts have been so grossly misrepresented by
the judge in his instructions to the jury as to result in a miscarriage of justice.
As we shall see (20.1), the range of appellate proceedings is restricted in many ways.
The rationale behind this approach is twofold. First, there is a fundamentally ideo­
logical reason. As explained by J. Spencer, ‘In England, the jury was introduced as a
substitute for the judgment of God pronounced through the ordeal, and like the judg­
ment of God it was not open to challenge on the ground that it had given an answer
that was wrong.’13 It was thought that it would be improper and illogical to ask an
‘appellate’ jury to pass judgment again on guilt or innocence (unless, on account of
gross errors of law the Court of Appeal remits a case to the trial court for a new trial).
In other words, the verdict of the jury at the trial level is final, unless it is invalidated
by serious mistakes made by the judge in his instructions to the jury. Hence, there is
no jury in the appeal court. The second reason is economic: reducing the number of
cases appealed alleviates the burden of appeal courts.
In the inquisitorial system both parties may lodge an appeal against conviction or
sentence, but the Prosecutor may also appeal against an acquittal. Furthermore, appel­
late proceedings may entail a sort of retrial, in that the same evidence may be scru­
tinized a second time and legal arguments reheard. In short, appellate proceedings
consist of a full rehearing of the case. According to a leading authority, M. R. Damaska,
the reason behind this civil law approach is twofold.14 First, prosecution is conceived
of as an ongoing process, while in common law countries there is a tendency to see
it as a one-off event. Secondly, members of the judiciary (often both prosecutors
and judges) are professionals working in a hierarchical system; it is therefore taken
for-granted that higher courts should normally re-do what has allegedly been done
badly by inferior courts.

13 Spencer, op. cit., at 28. See also p. 7.

14 M.R. Damaska, The Paces of Justice and State Authority—A Comparative Approach to the Legal Process,
op. cit., 48-50; Idem, ‘Models of Criminal Procedure’, 51 Zbornik, Collected Papers of Zagreb Law School
(2001), 495-6.
 THE ADOPTION OF THE ESSENTIAL FEATURES OF THE ADVERSARIAL SYSTEM 365

17.2.11 T H E M A IN F E A T U R E S O F E A C H M O D E L

The gist of the adversarial model is that it is based on two prim ary considerations:
(i) to leave the establishment of judicial truth to a contest between the parties, to be
essentially settled by a group of laymen representing the defendant’s ‘peers’; and (ii)
to protect the rights of the accused as much as possible by laying down a set of strict
procedural safeguards that act as the ultimate bulwark against a judiciary that may be
prone to abuses, or a potentially tyrannical executive.
It is generally argued that this model has various merits. Essentially, it reduces the
risk of errors, both (i) because the pitting of a prosecutor against a defence counsel
ensures that the two sides of the story are both aired under the guidance of a neutral
judge; and (ii) by separating the establishment of facts (reserved to the jury) from the
settlement of legal issues (entrusted to the judge). In addition, the adversarial model
enhances respect for the rights of the parties involved in the trial proceedings: since it
consists in a joust in open court, regulated by strict rules of procedures enforced by a
neutral judge, the chances that the rights of everybody, in particular of the defendant,
are fully respected, are significantly bolstered.
The major drawback of the model is, however, that the impecunious defendant who
cannot afford to pay effective investigators for the collection of evidence and a compe­
tent trial attorney is destined to succumb or to engage in bargaining a plea for reduc­
tion of the sentence. That recently the institution of the Public Defender has been
introduced in the US system so as to reduce the imbalance between the parties, essen­
tially bears out the weakness of this model.
Another limitation of the system is the fact that before and during trial proceedings
proper, the parties are allowed to file motions on matters of procedure, thereby slow­
ing down the pace of the proceedings.
The essence of the inquisitorial model lies instead in the strong emphasis on the
public interest in prosecuting and punishing all those who offend against societal val­
ues enshrined in criminal rules. Consequently, public institutions such as the pros­
ecutors and investigating judges play a significant role in administering justice, whilst
lesser emphasis is placed on the role and the rights of the accused.
The main merit of this system is that in principle, (i) the judicial institution is bent
on ensuring that the public interest in dispensation of justice be impartially and effi­
caciously served. It is thus inclined to pay no attention to the respective weight of the
parties, and in addition not to depend on the strength of either the prosecution’s trial
attorneys or the defence counsel. It is also notable that (ii) the right of the victims to
take part in criminal proceedings as civil petitioners (parties civiles) in order to claim
compensation for the damage they have allegedly received from the criminal offence
may prove a potent incentive to the establishment of facts and the finding of guilt, if
any. Furthermore, (iii) the settlement at the judicial investigation stage of all the pro­
cedural questions on which the parties raise objections or file motions entails that trial
proceedings proper are not encumbered by those motions. Generally speaking, trial
proceedings are more expeditious than in the accusatorial system, for the evidence is
366 INTERNATIONAL CRIMINAL LAW

available in the case file and all the procedural motions have already been disposed of.
This, however, may be counterbalanced by the excessive length of the phase of judicial
investigation (instruction preparatoire). Another advantage of the inquisitorial system
is that (iv) the conduct of the proceedings by the presiding judge and the pro-active
role he normally plays enables him to cut short possible attempts by the parties to slow
down the proceedings, or any other delaying tactic.
Notable weaknesses of this system should not, however, be underestimated: (i) the
lack of plea-bargaining entails that there is no means of forestalling trial proceedings,
whenever the charges are prim a facie well-founded and therefore the investigating
judge is barred from closing the case before trial; (ii) the central role played by the
judge, coupled with the lack (or at least the scant role) of two pitted advocates, each
pleading on behalf of one of the parties, may result in not attributing to the rights and
concerns of the accused the importance they would deserve; and (iii) the fact that
appellate proceedings in fact am ount to a re-trial unduly prolongs the procedure.
It should finally be reiterated that the aforementioned characteristics of the two
models are only intended to point to trends, for often individual countries incorpor­
ate elements of the system prevailing in other legal systems. One never finds in one
country a ‘pure’ system, but most of the time adversarial systems with some inquisi­
torial features or inquisitorial systems with some adversarial traits. Moreover, the two
systems seem to be converging in many ways.

17.3 THE TRANSPOSITION OF THE

ADVERSARIAL MODEL ON TO THE
INTERNATIONAL LEGAL LEVEL

17.3.1 T H E N U R E M B E R G IM T

In international proceedings the adversarial system has basically prevailed, but without
a jury. This system has been predominant since 1945, when the procedure for the IMT
was being discussed in London. The US and UK delegates managed to make common
lawnotions prevail, in spite of some resistance from the Soviet delegate. No major objec­
tions were raised by the French representatives, who, however, expressed misgivings.15
In retrospect, the upholding at the international level of the adversarial model was
a sensible and wise move, particularly in light of the contradictory proposals put for­
ward in the London negotiations by the French and Soviet delegations, proposals
which on the whole were not apt to meet the requirements of a fair trial and not suit­
able for duly protecting the rights of the accused.

15 For the history of the London negotiations leading to the acceptance of the adversarial model, see the
first edition of this book (2003), at 376-81.
 THE ADOPTION OF THE ESSENTIAL FEATURES OF THE ADVERSARIAL SYSTEM 367

Nevertheless, these two delegations won, as it were, in other areas: (i) the power
o f the court to play an active role, in particular by calling witnesses and questioning
witnesses and the accused;16 (ii) the right of the accused to make an unsworn state­
ment at the end o f trial;17 and (iii) the rules o f evidence.18 In addition, on the question
(iv) whether the court should have the power to try the accused in absentia, one of the
two delegations in question had the merit of clarifying, and giving final shape to, the
initial tentative American proposal to this effect,19 so that such power was eventually
laid down in the Statute, in Article 12.

16 The power of the court to ask questions of the accused and the witnesses was insisted upon by the
Soviet delegate, with French support, and, in spite of the misgivings of Justice Jackson (International
Conference on Military Trials, at 257, 262-4), found no opposition in the British delegate, who noted that
after all the court was ‘a complete master of the situation’ (ibid., 263). The matter was settled fairly easily, in
light of the generally shared intent to conduct the future trial as expeditiously and effectively as possible,
and also on account of the equally shared feeling that it was necessary to avoid any dilatory or obstructive
tactics by the defence.
17 The two delegations from common law countries accepted that the accused, in addition to giving
evidence on their own behalf under oath, be entitled to make a final statement, without taking any oath,
although this system is well known to Continental Europe but unknown in the USA and the UK.
18 Also with regard to evidentiary requirements, the USA and the UK easily accepted the idea that rules
of evidence should be simplified, the more so because there was no jury and the common law rules of evi­
dence constituted, as Jackson put it, ‘a complex and artificial science to the minds of Continental lawyers,
whose trials usually are conducted before judges and do not accord the jury the high place it occupies in our
system’ (ibid., xi).
19 A provision of the initial US draft tackled the issue without taking any definite position; nevertheless,
the US draft did not rule out trials in absentia. It was worded as follows: ‘The Tribunal shall determine to
what extent proceedings against defendants may be taken without their presence’ (ibid., 25). Plainly, this
remarkable departure from the US traditional opposition to trial in absentia was due to the extreme gravity
of the crimes committed and the exceptional circumstances the negotiators were facing. The provision was
restated in the subsequent drafts (ibid., 58, 123 (with a slight change), 179, 183). Subsequently, a draft was
proposed by the Soviet delegation, whereby ‘The Tribunal shall have the right to take proceedings against
persons charged with the crimes, set out in Article 2 of this Agreement, in the absence of the defendant, if
the defendant should be hiding or if the Tribunal should for other reasons find it necessary to conduct the
hearing in the absence of the defendant’ (ibid., 183).
The British delegation decided to incorporate and improve upon this Soviet draft, and proposed a new
text (ibid., at 206 and 353) that, subject to minor drafting changes, eventually became the final Article 12 of
the Statute.
At Nuremberg, discussing the defence motion that proceedings should be suspended against the accused
Krupp von Bohlen und Halbach (because of his incapacity to stand trial) and whether otherwise he would be
tried in absentia, the US Prosecutor Jackson noted that, ‘Of course, trial in absentia has great disadvantages.
It would not comply with the constitutional standard for citizens of the United States in prosecutions con­
ducted in our country. It presents grave difficulties to counsel under the circumstances of this case. Yet, in
framing the Charter, we had to take into account that all manner of avoidances of trial would be in the inter­
ests of the defendants, and therefore, the Charter authorized trial in absentia when in the interests of justice,
leaving this broad generality as the only guide to the Court’s discretion.’ (Trial of the Major War Criminals,
vol. II, at 5.) He went on to say that ‘the Court should not overlook the fact that of all the defendants at this
Bar, Krupp is unquestionably in the best position, from the point of view of resources and assistance, to be
defended. The sources of evidence are not secret. The great Krupp organization is the source of most of the
evidence that we have against him and would be the source of any justification. When all has been said that
can be said, trial in absentia still remains a difficult and an unsatisfactory method of trial, but the question is
whether it is so unsatisfactory that the interests of these nations in arraigning before your Bar the armament
and munitions industry through its most eminent and persistent representative should be defeated’ (at 6).
In setting out the court’s decision on the matter, to the effect that proceedings should be postponed for the
368 INTERNATIONAL CRIMINAL LAW

17.3.2 T H E T O K Y O IN T E R N A T IO N A L T R IB U N A L (iM T F E )

Unlike the Statute of the IMT, the C harter of the IMTFE was not negotiated
between the Allies, but drafted by the Americans, chiefly J. B. Keenan, subse­
quently US Chief Prosecutor at the trial. It was then issued as an executive decree
of US General D. M acArthur. In essence, the C harter reproduced the substance of
the Statute of the IMT; the procedure was very sim ilar to that which unfolded in
Nuremberg.
It would seem that the application of the adversarial system proved unfair to the
defence, because all the documents and materials likely to be used in evidence before
the court were in the hands of the prosecution, and defence counsel were not allowed
to inspect the prosecution’s files. The relative unfairness of the procedure was com­
pounded by the authoritarian conduct of business by the President, the Australian
Judge Webb, who in addition to sometimes treating witnesses in a derogatory manner,
did not allow his fellow judges to question the witnesses directly.
Some judges expressed misgivings about the conduct of trial. The Indian judge, Pal,
in his lengthy Dissenting Opinion, criticized the inconsistency of procedural deci­
sions taken by the majority (at 629-56). The French judge, Bernard, in his Dissenting
Opinion, also assailed the Tribunal’s procedure, arguing that the rights of the defend­
ants had not been safeguarded.20 The Dutch Judge Roling as well voiced criticisms
of the manner in which the trial had been conducted, not however in his Dissenting
Opinion, but in scholarly writings.21

accused, the Tribunal’s President (Lord Justice Lawrence) stated that: ‘It is the decision of the Tribunal that
upon the facts presented the interests of justice do not require that Gustav Krupp von Bohlen be tried in
absentia. The Charter of the Tribunal envisages a fair trial, in which the Chief Prosecutors may present the
evidence in support of an indictment and the defendants may present such defense as they may believe them­
selves to have. Where nature rather than flight or contumacy has rendered such a trial impossible, it is not in
accordance with justice that the case should proceed in the absence of a defendant’ (ibid., 21).
20 He stated: ‘The Defendants, in spite of the fact that the charges concerned crimes of the most serious
nature, proof of which [involved] the greatest difficulties, were directly indicted before the Tribunal and
without being given an opportunity to endeavour to obtain and assemble elements for the defense by means
of a preliminary inquest conducted equally in favour of the Prosecution as of the Defence by a magistrate
independent of them both and in the course of which they would have been [sic] benefited by the assistance
of the defence counsel. The actual consequences of this violation of principle have been, in my opinion, par­
ticularly serious at the present case’ (at 494). The French judge also criticized the fact that, the Tribunal hav­
ing no power of review of the action taken by the Prosecutors, the prosecution had not been exercised ‘in an
equal and sufficiently justified manner regarding all justiciable’ (sic). In particular, the judge regretted that
the Emperor of Japan had not been indicted. In his view, the Emperor’s ‘absence from the trial, while making
one wonder whether, if his case is measured by a different standard, international justice would merit to be
exercised, was certainly detrimental to the defense of the Accused’ (494).
21 See, for instance, B. V. A. Roling and A. Cassese, The Tokyo Trial and Beyond, cit., at 50-5.
 THE ADOPTION OF THE ESSENTIAL FEATURES OF THE ADVERSARIAL SYSTEM 369

17.3.3 T H E A C C E P T A N C E O F T H E A D V E R S A R IA L
M O D E L IN I 9 9 3 - 2 OOO

The two ad hoc Tribunals established by the UN SC in 1993 (the ICTY) and in 1994
(the ICTR) also embodied the essentials of the adversarial system, probably because
of (i) the intellectual and psychological appeal of the Nuremberg and Tokyo model;
and (ii) the prevailing influence, among the draftsmen, of persons with a common law
background.22 In addition, it was perhaps felt that (iii) the adversarial system better
safeguarded the rights of the accused.
The ICTY Statute laid down the fundamental features of the adversarial system, in
that it entrusted the Prosecutor with the conduct of investigations and the submis­
sion of indictments. After the possible confirmation of the indictment by a review
judge, trial proceedings could commence. The Statute thus discarded the model based
on an investigating judge responsible for gathering evidence on behalf of both par­
ties. However, it stopped there, without further setting out the details of the proced­
ural system. It fell to the judges to decide, when drafting the Rules of Procedure and
Evidence (RPE), how to fashion the system.
In their first text of the Rules, the judges in essence adopted a system very close
to the US Memorandum circulated among judges by the US Department of Justice
and containing proposal for draft Rules. Thus, the court was conceived of as a sort of
referee, which could become cognizant of the material supporting the indictment or
of the probative material produced by the defence only after commencement of trial
proceedings, and subject to the production of such material in court. Nevertheless,
the C ourt was granted extensive powers in matters of evidence, on the Nuremberg
model (Rules 89, on general provisions, and 94, on judicial notice). In addition, it was
empowered to order either party to produce additional evidence and summon wit­
nesses (Rule 98). It was also provided that when appellate proceedings were instituted
additional evidence could be submitted by the appellant (Rule 115). These were signifi­
cant ameliorations of the adversarial system, essentially derived from the inquisitorial
model. In short, the Court would be vested with fairly extensive powers, so as to be in
control of the proceedings.
However, fairly soon it became clear that, to expedite proceedings which, being
grounded in the adversarial model, were rather lengthy, it was necessary to depart
from the common law scheme. In particular, it was necessary to move away from the
approach whereby the court has no knowledge of the case before commencement of
trial, and even during trial only becomes cognizant of the probative material that the
parties tender into evidence.
It was in 1997 that a TC first took a step in this direction in Dokmanovic (follow­
ing a Scheduling Order in Mrksic and others). In an Order of 28 November 1998 the
TC decided that the Prosecutor m ust deliver to it witnesses’ statements taken from

22 The drafters of the ICTY Statute were headed by a British lawyer, and consisted of two US nationals, a
Liberian and an Israeli. Hence all of them were from common law countries. However, the then UN Legal
Counsel, to whom the team reported for final approval, was a German.
370 INTERNATIONAL CRIMINAL LAW

witnesses the Prosecutor intended to call for trial and any other m aterial on which
it intended to rely at trial; in addition, the prosecution m ust file a pre-trial brief
clarifying the allegations in the indictm ent, setting out the details of the case and
identifying the points in issue; it was also ordered to deliver, to both the TC and the
defence, at least one week prior to commencement of trial, a copy of the proposed
opening statement. Similar and parallel obligations were imposed on the defence.
As it was breaking new ground, the TC took two precautionary measures. First, it
prom oted, and was given in a status conference sum m oned by a scheduling order
of 20 November 1997, the agreement of the parties to the procedural measures it
envisaged. Secondly, it set out the reasons explaining both the rationale behind
this departure from a strict view of the adversarial system and the limits of such
departure.23
This precedent was followed in other cases and then led to the adoption of a string
of new Rules of Procedure and Evidence designed to enable the court to know the
case file in advance of commencement of trial and thus better to control and conduct
proceedings: in particular, Rule 65ter, on the pre-trial judge, Rule 73bis, on pre-trial
conferences, and Tbter, on pre-defence conferences. (It should, however, be empha­
sized that in these instances the ‘case file’ is different from that the investigating judge
hands over in the inquisitorial system to the court and the parties. Now, the file is
only made up of the ‘supporting m aterial’ handed by each party, namely the set of
witness statements gathered by them; the file does not include, instead, the appraisal
of the ‘evidence’ by the investigating judge, nor the decisions made by him on pro­
cedural issues).
The same approach was taken in the drafting of the ICC Statute. In essence, the
draftsmen opted for the adversarial model, but introduced some significant qualifica­
tions, thanks primarily to the strong diplomatic efforts of the French delegation. Thus,
for instance the role of victims was greatly enhanced (see infra, 17.4(B)).

23 As for the first point, the TC stated that it would benefit from having access to the documen­
tation requested, because such access would have promoted ‘better comprehension of the issues and
more effective management of the trial’. In addition, the rationale behind the Rule then applicable,
that is Rule 15(C) whereby the judge reviewing an indictment was disqualified from sitting on the
case, did not prevent the TC from examining material supporting the indictment. This was because,
as the European Court of Human Rights had held in Hauschildt, ‘suspicion and a formal finding
of guilt are not to be treated as being the same [...] the mere fact that a [...] Judge has also made
pre-trial decisions in the case [__ ] cannot be held as in itself justifying fears as to his impartiality
(at 2). As for the limits within which the court intended to consider the material it was to receive, the
TC pointed out that it would not regard that material ‘as evidence [...] unless and until submitted in
the course of the trial’ (ibid.).
 THE ADOPTION OF THE ESSENTIAL FEATURES OF THE ADVERSARIAL SYSTEM 371

17.4 THE PRINCIPAL ELEMENTS OF THE

INQUISITORIAL MODEL INCORPORATED INTO
INTERNATIONAL PROCEDURE
In spite of the basic incorporation of the adversarial system into international pro­
ceedings, in the procedure before the ICTY, other international tribunals (including
those of mixed composition) and the ICC some elements of the inquisitorial system
have been added so as to reduce some of the major disadvantages of the other system.

(A) The ICTY, ICTR, and the SCSL

In the ICTY and ICTR before the beginning of trial the prosecution (later on the
defence) hands over to the judges a file with the essentials about the case (see below).
Thus, the judges are in possession of a dossier enabling them better to control the
case.
Furthermore, the TC may exercise a fairly broad measure of control on the case
presented by the prosecution so as to shorten the possible length of proceedings.
First of all, after the initial appearance of the accused, the TC may designate a pre­
trial judge from among its members, who, under the authority and supervision of the
TC, shall coordinate communication between the parties during the pre-trial phase
and ensure that the proceedings are not unduly delayed’. He shall also ‘take any meas­
ure necessary to prepare the case for a fair and expeditious trial’ (Rule 65ter ICTY).
Secondly, upon order of the pre-trial judge, before commencement of the trial
the prosecution must file with the TC a set of documents that make it possible for
the judges to form a clear picture of the case. In particular, the Prosecution must file
(i) the final version of its pre-trial brief including, for each count, a sum m ary of the
evidence; (ii) the list of witnesses the Prosecutor intends to call, with, among other
things, a sum m ary of the facts on which each witness will testify; (iii) the total num ­
ber of witnesses and the number of witnesses who will testify against each accused
and on each count; (iv) the estimated length of time required for each witness and the
total time estimated for presentation of the Prosecutor’s case; (v) the list of exhibits the
Prosecutor intends to offer. Similar documents must be filed by the Defence after the
prosecution case rests.
Under Rule 78bis (ICTY), adopted in 1998, the TC is entitled to exercise a range
of extensive powers prior to commencement o f the trial. Under this Rule (as subse­
quently amended), the TC is authorized (i) to call upon the Prosecutor to reduce the
num ber of witnesses; and, if need be, (ii) to shorten the estimated length of the exam-
ination-in-chief of some witnesses; besides (iii) determining the time available to the
Prosecutor for presenting evidence. More recently, the Rule was amended so as to
grant even larger powers to the TC. Now, after having heard the Prosecutor, the TC,
‘in the interest of a fair and expeditious trial’, may invite him or her to reduce the
num ber of counts charged in the indictment. The TC may also direct the Prosecutor,
372 INTERNATIONAL CRIMINAL LAW

after hearing the parties and ‘in the interest of a fair and expeditious trial’, to select the
counts in the indictment on which to proceed (however, any decision taken by the TC
may be appealed as of right by a party). In addition, the TC may ‘fix a num ber of crime
sites or incidents comprised in one or more of the charges’ which ‘are reasonably rep­
resentative of the crimes charged’. Plainly, the main purpose of these extensive powers
accruing to the TC is to oblige the Prosecution to drop the less im portant charges and
focus on those which are really crucial, so as to expedite trial proceedings.24
Another im portant change made by the ICTY judges to the adversarial system is the
authorization to admit written documents in lieu of oral testimony. Rule 89 F grants in
general terms the power to admit written evidence (a TC ‘may receive the evidence of
a witness orally or, where the interests of justice allow, in written form’). Under Rule
92bis (ICTY RPE) a TC Chamber may dispense with the attendance of a witness in
person, ‘and instead admit, in whole or in part, the evidence of a witness in the form
of a written statement or a transcript of evidence’ given by a witness in proceedings
before the Tribunal, ‘in lieu of oral testimony which goes to proof of a matter other
than the acts and conduct of the accused as charged in the indictm ent’.25
An im portant provision that does not constitute a deviation from the adversarial
system as upheld in some common law countries, and may prove exceedingly useful
to save time, is that on the power of the court to take judicial notice of adjudicated
facts. Under Rule 94(B) ICTY RPE, at the request of a party or proprio motu, a TC,
after hearing the parties, ‘may decide to take judicial notice of adjudicated facts or

24 For a decision applying the Rule at issue see ICTY TC, Milutinovic and others, Decision on Application
o f Rule 73 bis. On this and other innovative Rules see the important paper by O-Kwon, "Ihe Challenge of an
International Criminal Trial as Seen from the Bench’, 5 JICJ (2007), at 363-75.
25 The Rule goes on to provide as follows: ‘(i) Factors in favour of admitting evidence in the form of a writ­
ten statement or transcript include but are not limited to circumstances in which the evidence in question:
(a) is of a cumulative nature, in that other witnesses will give or have given oral testimony of similar facts;
(b) relates to relevant historical, political or military background; (c) consists of a general or statistical ana­
lysis of the ethnic composition of the population in the places to which the indictment relates; (d) concerns
the impact of crimes upon victims; (e) relates to issues of the character of the accused; or (f) relates to factors
to be taken into account in determining sentence, (ii) Factors against admitting evidence in the form of a
written statement or transcript include but are not limited to whether: (a) there is an overriding public inter­
est in the evidence in question being presented orally; (b) a party objecting can demonstrate that its nature
and source renders it unreliable, or that its prejudicial effect outweighs its probative value; or (c) there are
any other factors which make it appropriate for the witness to attend for cross-examination.(B) If the TC
decides to dispense with the attendance of a witness, a written statement under this Rule shall be admissible
if it attaches a declaration by the person making the written statement that the contents of the statement are
true and correct to the best of that person’s knowledge and belief and (i) the declaration is witnessed by: (a) a
person authorised to witness such a declaration in accordance with the law and procedure of a state; or (b) a
Presiding Officer appointed by the Registrar of the Tribunal for that purpose; and (ii) the person witnessing
the declaration verifies in writing: (a) that the person making the statement is the person identified in the
said statement; (b) that the person making the statement stated that the contents of the written statement
are, to the best of that person’s knowledge and belief, true and correct; (c) that the person making the state­
ment was informed that if the content of the written statement is not true then he or she may be subject to
proceedings for giving false testimony; and (d) the date and place of the declaration. The declaration shall
be attached to the written statement presented to the TC. (C) The TC shall decide, after hearing the parties,
whether to require the witness to appear for cross-examination; if it does so decide, the provisions of Rule
92ter shall apply.’
 THE ADOPTION OF THE ESSENTIAL FEATURES OF THE ADVERSARIAL SYSTEM 373

documentary evidence from other proceedings of the Tribunal relating to matters at

issue in the current proceedings’. This Rule enables the TCs to avoid repetitive tes­
timony whenever the court has to deal with many cases that in part hinge on some
common or similar facts.
It is also notable that the court may call evidence and summon witnesses proprio
motu.26
In summary, by and large, the court is in control of the proceedings, or at any rate may
exercise powers that are not always granted to courts in most common law systems
(sometimes not even in civil law countries).
The two stages of determination of guilt and sentencing have been merged (except of
course when at the outset the accused enters a guilty plea).
Rules of evidence tend to be very flexible; furthermore, affidavits are admissible,
albeit to a limited extent (see infra, 19.7(A)).
While awaiting trial and during trial the defendant is normally in detention (after
his arrest, for security reasons or on political grounds the state hosting the Tribunal
as well as other states are often reluctant to detain him or to ensure his presence at
trial). However, there is an increasing tendency to free the accused on bail, provided
guarantees are given by the relevant state authorities that he will not be allowed to
escape.
In the appellate proceedings additional evidence may be heard when this is deemed
to be in the interest of justice (see infra, 20.3).

(B) The ICC

First of all, unlike the ICTY and ICTR (where the Prosecutor must collect evidence
against the accused but is legally obliged to hand over to the defence any exculpa­
tory evidence so found) under Article 54(l)(a) of the ICC Statute he is entrusted
with establishing the tru th ’. For this purpose he must investigate both ‘incrim inat­
ing’ and ‘exonerating’ circumstances ‘equally’. As has been rightly emphasized,27 the
Prosecutor has thus been conceived of truly as an ‘organ of justice’. It is also notable
that in the ICC system the Pre-TC plays a major role in scrutinizing and monitoring
the action of the Prosecutor, in particular for the purpose of safeguarding both respect
for the rights of the suspect or accused and the correct conduct of business by the
Prosecutor. In many respects this Pre-TC plays the role that, in some civil law systems
which largely borrow from the common law tradition, is entrusted to a judge, who,
however, is not the ‘investigating judge’ of civil law systems. This is a judge deprived
of the power to conduct investigations and collect evidence. He only acts as a judicial
guarantee of full respect for law by the prosecuting authorities and for the issuing of

26 See, for instance, the ICTY decision in Stakic (Decision summonsing Mr Baltic proprio motu to appear
as a witness), at 2-3.
27 See S. Zappala, Human Rights in International Criminal Proceedings (Oxford: Oxford University Press,
2003), 29-45.
374 INTERNATIONAL CRIMINAL LAW

any order requested by those authorities, which may entail curtailm ent of the rights of
the suspect or other persons.28
Articles 58 and 60 of the ICC Statute are based on the principle o f‘no deprivation of
liberty unless necessary on certain grounds’.
In addition, while in proceedings before the ICTY and the ICTR victims do not
play any autonomous role, as they may only appear in court as witnesses if called
by one of the parties (normally the Prosecution) or the Court itself, in the Statute of
the ICC they have been given several roles: although they are not entitled as of right
to address Chambers, if admitted to do so they may submit briefs, attend the hear­
ings and examine or cross-examine witnesses. However, the legal institution typical
of civil law countries, namely the right to join criminal proceedings as a private peti­
tioner (‘constitution departie civile), with all the attendant rights and powers, has not
been fully upheld, not even before the ICC.

(C) Innovative Features of the Extraordinary Chambers in the C ourt of

Cam bodia (ECCC) and the Special Tribunal for Lebanon (STL)
Significant procedural novelties can be found in the Statute and the Rules of the ECCC
as well as in the Statute of the STL. In both cases the incorporation of typical features
of the civil law procedure into the adversarial scheme has been motivated by (i) the
influence of the system prevailing in the country concerned (Cambodia and Lebanon
are both based on the Romano-Germanic tradition passed on to those countries
through French influence); and (ii) the need to avoid the pitfalls of the adversarial
system (chiefly the excessive length and the costly nature of proceedings).
The principal innovations in the Extraordinary Chambers for Cambodia are as
follows:

(i) The Cambodian Law on the Establishment of the Extraordinary Chambers, of

27 October 2004 (the Law) envisages both prosecutors and investigating judges.
(ii) The former are two (one Cambodian and one international). They (a) initiate
criminal proceedings, either on their own initiative or on the basis of complaints (Rule
49 of the Internal Rules); (b) conduct preliminary investigations’ (Rules 13(4)(a) and
50) with the assistance of the ‘judicial police’ and investigators (Rules 15(2) and 16); (c)
file ‘submissions’ with the investigating judges; (d) conduct the prosecution in court;
and (e) may appeal the verdicts of the Chambers.
(iii) Under Article 23 of the Law two investigating judges (one Cambodian and one
foreign) conduct investigations, question suspects and victims, hear witnesses, and
generally speaking collect evidence. Under Rule 55(5) they must ‘ascertain the tru th ’
and for that purpose ‘they shall conduct their investigation impartially, whether the
evidence is inculpatory or exculpatory’. They also may ‘order the provisional deten­
tion of a charged person after an adversarial hearing’ (Rule 63(1)). They wind up the

28 One may think, for instance, of the institution of the ‘giudice delle indagini preliminari’ or ‘the judge
supervising preliminary investigations’ in the Italian system.
 THE ADOPTION OF THE ESSENTIAL FEATURES OF THE ADVERSARIAL SYSTEM 375

investigation by issuing a closing order, which either indicts a person charged and
commits him or her for trial, or dismisses the case (Rule 67).
(iv) If the accused person is committed for trial, the investigating judges hand over
a case file both to the judges and the parties, as well as to the civil petitioners (civil
parties’) (Rule 86).
(v) At trial, the judges are empowered to ensure that the proceedings are expeditious
and fair and to this effect the presiding judge may take all the necessary measures.
(vi) Victims may participate in proceedings as civil parties’ and to this effect they
may exercise a host of rights before the investigating judges as well as in court (Rule 23).
(vii) Trial proceedings are conducted by the presiding judge, who, together with
the other judges, may ask the accused ‘any questions which they consider to be con­
ducive to ascertaining the tru th ’ (Rule 90(1)). After the questioning by the judges,
the accused may be questioned by the co-prosecutors ‘and all the other parties’ (Rule
90(2)). Witnesses are then questioned by the judges and the various parties (Rule 91).
(viii) The TC, in its judgment, may also decide on any claim by ‘civil parties’, grant­
ing reparation when appropriate (Rule 100).

The procedure envisaged for the Special Tribunal for Lebanon is less innovative,
although it shows many significant novelties. Under the Tribunal’s Statute:

(i) A Pre-Trial Judge (pursuant to Articles 8 and 18) can not only scrutinize and con­
firm or dismiss the indictment, but also, at the request of the Prosecutor, issue orders
and warrants necessary for the proper ‘conduct of the investigation’ and ‘the prepa­
ration of a fair and expeditious trial’; clearly, this judge, who must be international
and not Lebanese, is called upon to play the crucial role of supervising the prosecu­
tor’s activity and ensuring the rights of the suspects and witnesses; in addition he is
empowered to put in place all those measures that make it possible for the subsequent
trial proceedings to be absolutely fair, streamlined, and rapid. It should be emphasized
that the pre-trial judge provided for in the STL is different both from the similar insti­
tution existing within the ICTY, the ICTR, and the SCSL, as well as from the inves­
tigating judge of the Roman-Germanic tradition. He differs from existing pre-trial
judges for, as rightly stressed by the UN S.-G. in his Report on the Tribunal (§28):
unlike the pre-trial judge of either of the ad hoc tribunals for the former Yugoslavia and
Rwanda, or of the Special Court for Sierra Leone, who is designated by the presiding judge of
a trial chamber from among the members of that chamber, the pre-trial judge of the special
tribunal for Lebanon is a dedicated, single international judge serving as a pre-trial judge
only and not as a member of any of the chambers.
Arguably it follows that now the pre-trial judge enjoys greater freedom and is vested
with more extensive powers, given that he is not restrained by the fact that he will
have subsequently to pronounce on the facts. The STL pre-trial judge also differs from
the investigating judge of civil law systems in that the latter is empowered to gather
376 INTERNATIONAL CRIMINAL LAW

the evidence available on behalf of both the prosecution and the defence, whereas the
former may only supervise the collecting of evidence.
(ii) An independent Defence Office is provided for in Article 13 to act as a counter­
vailing factor to the Office of the Prosecutor; thus an attempt is made to remove the
imbalance between the prosecution and the defence that normally besets the adver­
sarial approach (the former being endowed with extensive financial means, facilities
and personnel, and the latter at a clear disadvantage for lack of similar wherewithal).
(iii) Victims may take part in proceedings, articulating their views and concerns
according to modalities to be specified in the RPE (Article 17).
(iv) Under Article 23 trials in absentia are allowed when the accused has either
waived his right to be present, or absconds, or is not surrendered by the state where he
finds himself; however, the Tribunal’s power to conduct such proceedings is hedged
around with a string of safeguards for the absent defendants.
(v) In the conduct of proceedings the TC may ‘take strict measures to prevent any
action that may cause unreasonable delay’ (Article 21(1)); in addition, it can allow
the presentation of evidence ‘in written form’ ‘where the interest of justice allow’ it
(Article 21(3)). Thus, measures are envisaged designed to expedite proceedings and do
away with possible delaying tactics of the parties.
(vi) Under Article 20(2) the exam ination of witnesses ‘shall commence with
questions posed by the presiding judge, followed by questions posed by other m em ­
bers of the TC, the Prosecutor and the Defence’; this procedure is not, however,
rigid, for it applies ‘unless otherwise decided by the TC in the interest of justice’.
Thus, in essence, the typical way of holding hearings adopted in civil law countries
may now replace that of com m on law systems. It remains to be seen how the pos­
sible adm ixture of the two models in the conduct of hearings will work out.

17.5 TOWARDS A ‘MIXED’ PROCEDURAL MODEL

In summary, over the years there has been a gradual incorporation of significant fea­
tures of the inquisitorial model into the procedural system of international criminal
tribunals, initially based on the adversarial scheme. The need to speed up proceed­
ings has been the prim ary rationale for this gradual change. Subsequently, when the
Statutes of the ICC and other tribunals such as the ECCC and STL were being elabo­
rated, the draftsmen took these developments into account. Thus, they inserted into
the procedural scheme of the ICC and the STL further elements of the inquisitorial
model; that is, a larger role for victims, expanded functions for the Prosecutor (who
under the ICC Statute is called upon to act as an organ of justice, and consequently
is charged with gathering evidence on behalf of both parties—the prosecution and
the defence), and a crucial role for the Pre-TC (in the ICC) and for the pre-trial judge
in the STL, entrusted with the task of scrutinizing the activity of the Prosecutor. The
 THE ADOPTION OF THE ESSENTIAL FEATURES OF THE ADVERSARIAL SYSTEM 377

Cambodian Law on the ECCC goes even further, for it also envisages investigating
judges. Furthermore, in both the ECCC and the STL the conduct of trial hearings is
substantially modelled on the typical criminal procedure of civil law countries.
This significant progress towards a procedural scheme still largely adversarial but
enshrining im portant features of the other model needs, however, further honing
with the passage of time and increased experience. Not surprisingly, some attempts
by judges with a civil law background to inject elements of the inquisitorial system
into the essentially adversarial procedure of the ICTY have given rise to controversy
in the course of trial: for instance, in the Prlic and others trial the defence objected
to the frequent questioning of the witnesses by the bench.29 It has also been pointed
out by a leading authority that it is problematic to merge the two systems because
each of them rests on its own logic and has its own inherent implications, difficult
to combine.30
Nonetheless future efforts to test new procedures are necessary, with a view
to m eeting the essential demands of international trials; that is, fairness and
expeditiousness.

29 See Transcript of hearings, 15825-15839 (15 March 2007). The Presiding Judge (the French Judge
Antonetti) very effectively explained the sense and rationale for the need of Judges to ask questions to wit­
nesses (see 15829-32). Defence counsel objected to the Judges’ approach (see Counsel Murphy, 15832-36;
Counsel Alaburic, 15837-39; this counsel stressed that also in Croatia ‘Judges seek the material truth regard­
less of the disposition of the parties in the proceedings’, at 1538). Also defence counsel Scott objected, as fol­
lows: ‘I am well familiar with the oft-made statement that it [the ICTY procedure] is a hybrid system and not
based on any one national system or any one legal approach. [...] I am no generic defender or advocate of the
adversarial common law system even though I come from that system. Like every system, it has its strengths
and its weaknesses. I think there is much that common lawyers can learn from the civil law system. However,
I am also no generic defender or advocate of the civil law system. Like every system, it has its strengths and
its weaknesses. [...] Whether one likes it or not, the adversarial system is primarily a party-driven system.
Each party, the Prosecution and the Defence, have both the responsibility, and weighty responsibilities,
and the opportunity to present its case. Not the Court’s case, not the Judges’ case, its case. [...] The Trial
Chamber cannot and should not try to perform all roles at once, to be Prosecutor, Defence counsel, Judge,
jury and executioner. If the Trial Chamber begins to cast itself in the role of the Prosecution and to shape
the Prosecution’s case to prove the Prosecution’s case against the accused, then I assume that my brethren
on the Defence side will object to that. Indeed, I will object to that. Likewise, if the Trial Chamber casts itself
as Defence counsel to the accused directing the Defence cases, putting forward and protecting the Defence
cases, then indeed I, the Prosecution, will object to that, and presumably, at least in some circumstances, the
Defence counsel will object to that. [...] To use an American sports example, [...] to use a baseball example
you call the [sic: bats] and the strikes, but you don’t determine—you don’t tell how—the batter when to
swing and you don’t tell the pitcher when to pitch. You do not direct or present the Prosecution’s case or the
Defence case. Now, specifically on questioning witnesses. In light of what I’ve just said, a party must be given
a reasonable opportunity to conduct its examination according to his or her own plan. A plan is disrupted,
it is made more difficult, confusing, takes more time if there are constant interruptions and detours. For
example, taking a witness off in a particular direction before the examiner, him or herself, is ready to go in
that direction himself or herself’ (Transcript, hearing of 19 March 2007, at 15853-55).
30 See M. Damaska, ‘Problematic Features of International Criminal Procedure’, in A. Cassese (ed.),
Oxford Companion to International Criminal Justice (Oxford: Oxford University Press, forthcoming).
 18
GENERAL PRINCIPLES
GOVERNING INTERNATIONAL
CRIMINAL TRIALS

18.1 THE NATURE AND ROLE OF PRINCIPLES

There do not yet exist international general rules on international criminal proceed­
ings. Each international court has its own Rules of Procedure and Evidence (RPE).
Probably, with the gradual winding down of the judicial activity of the ICTY, the ICTR
and the Special Court for Sierra Leone (SCSL) and the contemporaneous consolida­
tion of the ICC, at least some rules of procedure of this Court may come to be generally
accepted by states and then turn into general international rules. This is, however, a
process that is likely to take a num ber of years.
Nonetheless, one may set out some general principles governing international
trials. They may be extracted by way of generalization both from the Statutes of the
current Tribunals and the ICC and their RPE, and the Charters of the two previous
ad hoc Tribunals (the IMT and the IMTFE), as well as from judicial practice. In other
words, they may be drawn from the relevant rules governing proceedings before
international criminal tribunals, as well as existing case law and the general principles
of law on the criminal process. These principles, which also give rise to basic hum an
rights of the defendant (as well as, whenever appropriate, of the victims and the wit­
nesses), lay down: (i) the requirement that the accused be tried by an independent and
impartial court; (ii) the presumption of innocence (that is, the right of defendants to
be presumed innocent until proved guilty); (iii) the requirement that the trial be fair
and expeditious; and (iv) the right of the accused to be present during trial (that is, the
prohibition of trial in absentia or its acceptance under very strict conditions).
Such principles also reflect fundamental standards on hum an rights laid down in
international treaties,1as well as the general principles on criminal law upheld in most
countries of the world.

1 See, for instance, the 1950 European Convention on Human Rights, the 1966 UN Covenant on Civil
and Political Rights, the 1969 American Convention on Human Rights, and the 1981 African Convention
on Human and Peoples’ Rights.
 GENERAL PRINCIPLES GOVERNING INTERNATIONAL CRIMINAL TRIALS 379

The principles in question may play a crucial role. They dictate the manner in which
crim inal proceedings must unfold, and by the same token confer fundamental rights
that their beneficiaries may invoke and vindicate before the court, if need be by appeal­
ing decisions infringing such rights. The principles may also serve as a useful tool for
the proper construction of procedural rules and regulations, whenever the latter are
not clear or lend themselves to conflicting interpretations.

18.2 THE PRINCIPLE THAT JUDGES MUST BE

INDEPENDENT AND IMPARTIAL
This principle is firmly embedded in all legal systems. It gives rise to one of the most
fundamental hum an rights.2
The independence and im partiality of judges is ensured only by: (i) adopting selection
mechanisms that make it possible to choose persons who are not only competent, of
moral integrity and unbiased, but also independent of any political or governmental
authority; (ii) prohibiting judges from seeking or receiving instructions from outside
authorities or being in any way involved in the interests or concerns of the parties;
(iii) setting up monitoring procedures that prevent judges from practising or showing
bias and, if they are found to be partial or slanted, removing judges from a case or even
from the court.
In international proceedings the authorities that might in some way interfere with
the judges’ im partiality are either the appointing authority or states (in particular the
judge’s national state). Hence, election by a ‘parliamentary’ body is the best mechan­
ism for appointing judges. This system is provided for in the Statutes of the ICTY and
the ICTR (judges are elected by the UN GA, upon proposal of the UN SC, which short­
lists candidates nominated by member states). The Statute of the ICC provides for the
election by the Assembly of States parties to the Statute of candidates nominated by
each contracting state. A wise m anner of further strengthening the independence of
judges is that envisaged in Article 36(9)(a) of the ICC Statute, whereby judges are not
eligible for re-election after the expiry of their mandate.
Once judges have been elected, various means exist for ensuring that they remain
independent. First, they are duty bound to refrain from engaging in activity that might
jeopardize their independence or affect confidence in their independence (see for
instance Article 40(1) of the ICC Statute). Secondly, they enjoy privileges and immun­
ities including immunity from states’ jurisdiction; such immunities are, among other
things, designed to shield them from any undue interference by states. Thirdly, the
internal regulations of the Tribunals provide for methods and procedures for ensuring
judges’ independence. These also apply to any specific case where a judge may feel that he

2 See ICTY Furundzija (AJ), §§177,189, and 191.

38o INTERNATIONAL CRIMINAL LAW

may not be impartial, or may be seen as involved in, or concerned with, a particular case.
These mechanisms are those for the disqualification, recusal, or self-recusal of judges.
A number of rules set out the standards and mechanisms for ensuring the disquali­
fication of a judge from sitting on a particular case. In this respect ICTY Rule 15(A)
can be mentioned. A much more elaborate and detailed provision can be found in the
ICC Rules (Rule 34). Normally, if the judge does not disqualify himself or herself, the
m atter may be brought to the Court or the relevant Chamber, or even to the Tribunal’s
Bureau (this holds true for the ICTY and the ICTR), where a majority of judges shall
pronounce on the matter, in the absence of the judge concerned.3

18.3 THE PRESUMPTION OF INNOCENCE

All major national legal systems proclaim the principle that an accused is presumed
innocent until proved guilty. The provisions of the Statutes of the ICTY (Article 21(3),
the ICTR (Article 20(3)), and the ICC (Article 66) also clearly set out the principle.
It is generally agreed that the presumption of innocence specifically entails that:
(i) the person charged with a crime must be treated as being innocent until proved
guilty; (ii) the burden of proof, that the accused is guilty of the crimes with which he
is charged, is on the Prosecutor; the defendant may limit himself to rebutting the evi­
dence produced by the Prosecutor, but does not have to prove his innocence;4 (iii) in
order to find the accused guilty of the crimes charged, the court must be convinced

3 To date the issue of disqualification has been raised in a number of cases: Delalic and others (decisions
of the Bureau of 4 September 1998 and 1 October 1999), Kordic and Cerkez (decision of the Bureau of 4 May
1998), Brdanin and Talic (Decision of 18 May 2000), Furundzija (AJ). It is notable that so far motions for
recusal have never been granted.
In Delalic and others (decision of the Bureau of 4 September 1998), the Bureau held that the fact of hav­
ing been elected second Vice-President of Costa Rica did not disqualify Judge Odio Benito because she had
pledged not to assume any function in the Costa Rican Government before the completion of her mandate
as a judge, and the commitment had been confirmed by the President of Costa Rica. Subsequently, some of
the defendants filed with the Appeals Chamber a motion for disqualification of three judges sitting on that
Chamber on their appeal against conviction; they claimed that these judges, by participating in the plenary
session of the ICTY judges which found that Judge Odio Benito was not disqualified from sitting on the
case at the trial level, were disqualified from sitting on the appeal. The Tribunal’s Bureau, by a decision of
25 October 1999, dismissed the motion. It found that the three judges had participated in an administrative
decision concerning the general question of whether Judge Odio Benito was entitled to continue to exercise
her functions as a judge; they had not participated in any judicial decision on the specific question of whether
Judge Odio Benito should be disqualified from sitting in Delalic and others (§14).
In Furundzija (AJ) the appellant had recused Judge Mumba because, before being elected judge of the ICTY,
she had been a delegate of her government in the UN Commission on the Status of Women, where the defin­
ition of rape (the offence submitted to the TC in the case at issue) had been discussed; in addition, she had met
persons who were later involved in the trial, namely three authors of the amicus curiae briefs submitted in the
case, as well as one of the prosecutors. The Appeals Chamber found that the link of the judge with these person
had been ‘tenuous’ (§194) and in addition her membership in the UN body, her sharing the goals of that body,
and her concern for the protection of the rights of women had not created any bias in her (§§195-215).
4 See ICTY, Delalic and others, TJ, §§599,601.
 GENERAL PRINCIPLES GOVERNING INTERNATIONAL CRIMINAL TRIALS 381

of his guilt according to a certain standard of p roof which in civil law countries is
normally ‘the judge’s innermost conviction’ (Vintime conviction dujuge), whereas in
common law countries is ‘finding the accused guilty beyond a reasonable doubt’.
As for the treatment o f the accused as presumed innocent, three major issues arise.
First, there is the question of when a person should enjoy the presumption of inno­
cence: from the moment he is formally charged with crimes, thereby becoming
accused, or from the moment he becomes a suspect and may therefore be investigated
and questioned before being formally charged? The text of the relevant provisions sup­
ports the former construction, for such provisions tend to link the presumption under
discussion to the status of ‘accused’. However, as has been sensibly noted,5 the pre­
sumption must also apply during pre-trial investigations and should be considered
even stronger with regard to a person against whom ‘not even a prima facie case has
been confirmed’.6 It follows, among other things, that ‘the investigating authorities
must also investigate in favour of the suspect in order to exclude any reasonable doubt
from their suspicion’.7
Another issue that constitutes a sore point in international criminal proceedings is
the media coverage of the detention and trial of the accused. Faced with such horren­
dous offences as war crimes, crimes against humanity, and genocide, the media tend to
take it for granted that persons accused of such crimes are guilty, and even to portray
them as ‘monsters’ or ‘butchers’. At the national level, especially in common law coun­
tries (where the media are often prohibited from publishing any evidence or specific
information against the accused in advance of trial, and the media or photographers
are frequently denied access to the courtroom during trial proceedings), courts may
resort to safeguards against such intrusions. In any case, the accused is offered the
right to sue the media for libel. In contrast, no such remedy is available in international
proceedings. Ultimately, the question is one of self-restraint on the part of the pros­
ecutor and the media. However, some sort of remedy should be devised against the
excessive ‘presence’ of the media and their trampling underfoot of the presumption
of innocence.
Thirdly, one ought to stress the main consequences of the application of the pre­
sumption of innocence to trial proceedings. If the accused refuses to enter a plea of
guilty or not guilty, the TC must enter a not guilty plea, precisely because of that pre­
sumption. Furthermore, the accused has the right not to incriminate himself and, more
generally, to remain silent (the same right applies to suspects). The provisions covering
this matter ‘substantially aim at protecting the right of the accused [or the suspect] to
refuse to answer questions, because he or she is presumed innocent and, hence, has
no duty to contribute to the proceeding’.8 It would seem that another consequence is

5 See C. J. M. Safferling Towards an International Criminal Procedure (Oxford: Oxford University Press,
2001), at 67-75, Zappala op. cit., 84-5.
6 Zappala, at 84.
7 Safferling, at 73.
8 Zappala, at 90.
382 INTERNATIONAL CRIMINAL LAW

that the accused has no obligation to give evidence in court; in addition, no adverse
consequence may be drawn from his decision not to testify on his own behalf.
W ith regard to the burden of proof, it is for the Prosecutor to prove that the accused
is guilty. If the Prosecutor does not produce convincing evidence to this effect, the
charges must be dismissed.9 The charges may be thrown out even before the end of
the trial, that is when the prosecution case rests, either at the request of the defence or
by the court acting proprio motu.101Furthermore, as provided in Article 67(l)(i) of the
ICC Statute, no reversal of the burden of proof is admissible. This provision, it is sub­
mitted, enshrines or codifies a general principle.11
One of the consequences of the principle under discussion is that the defendant
has the right to a finding of guilt or innocence on all the charges preferred against
him by the prosecution, unless they are cumulative. In particular, the defendant is
entitled to have the charges against him considered by the Court if they are framed
as alternative charges. Clearly, were the court to leave in abeyance the question of
whether or not one of the charges was well founded, this would be prejudicial and
unfair to the accused. For such a stand would imply that some charges would remain
undetermined, thus leaving open the question of whether the defendant was guilty
or innocent on those charges. Consequently, that position would run counter to
the presumption of innocence. A British Court of Appeal set out these notions in
Paul Hermann (on multiple charges, some of them preferred as alternative)12 and in

9 In WolfgangZeuss and others (the Natzweiler trial) in his summing up the Judge Advocate insisted on the
point that ‘the onus of proving the charge which is made against these accused rests upon the Prosecution,
and they have to satisfy you beyond reasonable doubt of the guilt of any of the accused before such accused
can be convicted’ (at 199).
10 In Jelisic an ICTY TC, based on Rule 98bis, at the end of the Prosecutor’s case acquitted the accused
of some of the charges, namely those concerning genocide (see §§16-17), although this decision was much
criticized on appeal. See Jelisic (AJ), §§30-77.
11 It follows that, for instance, the reversal of the burden of proof provided for in Rule 92 of the ICTY
Rules of Procedure and Evidence (‘a confession by the accused given during questioning by the Prosecutor
shall [...] be presumed to have been free and voluntary unless the contrary is proved’) is perhaps of doubtful
legality. See Zappala, at 94.
12 The defendant, a Gestapo official, was accused of having been concerned in the shooting of a large
number of Jews in Poland. Sixteen charges were made against him. The first (murder of a Jew) was framed
under Control Council Law no. 10 as a crime against humanity; the same killing was charged as murder
under the German Criminal Code; the third charge (killing a German woman) was again charged as a crime
against humanity and, in charge no. 4, as a murder under German criminal law. The same held true for the
other charges, concerning the killing of other Jews, and all presented as alternatives. At the request of the
prosecution, and with the consent of the defence, the Court commenced by trying the first two charges. The
defendant was found guilty on the first charge, and the Court made no finding on the alternative charge.
The Court then proceeded to. try the third and fourth charges, convicting the accused on the third charge,
again without making any finding on the alternative fourth charge. The prosecution did not wish to proceed
with the other charges; although the defence objected, asking that they should be tried, they were left on the
file; the defendant was sentenced to death. On appeal the defence counsel stated that he had not been aware
that the trial would come to an end after dealing with the four charges only. He wished the other charges to
be dealt with. The defence was that the accused ‘was the victim of a conspiracy, and this could not be fully
brought out in the charges which were dealt with’ (at 164). He therefore asked that the case be remitted to
the High Court. The prosecutor argued that the first and third charges ‘were complete in themselves and
could properly be tried separately. Conviction on either of these would justify the sentence imposed, and it
 GENERAL PRINCIPLES GOVERNING INTERNATIONAL CRIMINAL TRIALS 383

Felix Schwittkowski (on alternative charges).13

Finally, the judgments of international courts must provide the legal reasons for the
court’s findings. These courts must set out the grounds on which they have appraised
the available evidence in such a m anner as to reach the conclusion that the accused
is guilty beyond reasonable doubt of the crimes charged. The need for courts’ reasons
constitutes an im portant safeguard for the rights of the accused; it enables him to
appeal against any conviction he may consider to be based on an erroneous evaluation
of the evidence leading to a miscarriage of justice.14

18.4 THE PRINCIPLE OF FAIR AND

EXPEDITIOUS TRIAL
That trials must be fair is by now a universally accepted principle of international law.
It was laid down in hum an rights treaties (for instance, the UN Covenant on Civil and
Political Rights (Articles 14(1) and 26) and in the European Convention on Human
Rights (Article 6) and the American Convention on Hum an Rights (Article 8)), with
regard to national trials. International case law, in particular that of the European Court

was unnecessary for the Prosecution to proceed further’ (at 164). He added that, if however the Court felt
‘that there is any possibility of there having been any injustice to the Appellant’ he would not resist an order
for some or all of the charges to be tried. The Court of Appeal held that the trial judge had been right in
pronouncing on only two charges: ‘After the convictions on the 1st and the 3rd charges it would have been
unnecessary and almost inhumane to have gone on with a long string of capital charges; the Defence had no
right to demand that the Prosecution should have gone on further’ (at 164). In the Court’s view, the defence
counsel should have made a complete defence in respect of each of the 16 charges. The Court however added:
‘On the other hand, the Court is sensible to the difficulties in which the Defence was placed, and we feel that,
if there is any likelihood of a miscarriage having occurred, we will do what we can to cure it, whether or not
there was any technical mistake on the part of the Defence [...] we feel that it would be more satisfactory if
the Defence were allowed to have some of the other charges tried, as has been requested’ (at 166). The Court
therefore adjourned the appeal sine die and remitted the case to the trial judge.
13 The appellant had stabbed a Norwegian officer in a cafe in Flensburg. He had been charged with
assaulting a member of the Allied Forces in Germany contrary to a Military Government Ordinance, and
on an alternative charge with causing dangerous bodily injury, contrary to the German Criminal Code. The
trial judge found that the accused did not know that the man he had stabbed was a Norwegian officer, and
convicted him on the second charge, adding, ‘We do not think that if charges are framed as alternatives there
should be conviction under both’; he consequently pointed out that there would be no verdict on the first
charge, ‘but it will of course remain on the Record’ (at 20-2). The Court of Appeal held instead that, ‘In a case
where there are alternative charges, if a Court finds the necessary facts proved which would justify a convic­
tion on each of the alternative charge (sic), we consider that it is a proper course to convict on one charge, but
to record no formal finding in respect of the other; should an Appellate Court subsequently find it necessary
to quash the conviction on the one charge, it would then be open to that Court, in a case where all the neces­
sary elements constituting the alternative offence have been clearly stated by the Trial Court to have been
proved, to substitute a conviction on that alternative charge. If, however, as in the present case, the Court
finds that a material element constituting the offence has not been proved, we consider that the accused is
entitled to a formal finding of not guilty in respect of the charge which has not been proved’ (at 22).
■14 In Kupreskic and others (AJ), the ICTY AC found that the appraisal of the evidence by the TC had been
so fallacious as to generate a miscarriage of justice (§§21-76).
 384 INTERNATIONAL CRIMINAL LAW

of Hum an Rights, has upheld and spelled out the principle.15 Ih e principle has then
come to acquire a fundamental value also with regard to international proceedings.
It has been laid down in various international instruments establishing international
criminal tribunals16 and has been upheld by case law.17 It seems indisputable that by
now it belongs to the category of customary norms of international law. Arguably the
principle is even endowed with the force of a peremptory norm (jus cogens); that is,
may not be derogated from by treaty; this proposition, absent state practice and case
law on this specific issue, seems nevertheless warranted by the insistence of all states
on the importance of fair and expeditious trials.
The principle is articulated into three main standards: equality of arms; publicity of
proceedings; and expeditiousness of proceedings.

18.4.1 E Q U A L IT Y O F A R M S

There are two different notions of equality of arms.

First, there is the concept developed in the case law of the European C ourt of
H um an Rights over the years. It implies that the accused may not be put at a ser­
ious procedural disadvantage with respect to the prosecutor. This applies to the
accused only: hum an rights treaties do not grant the prosecutor the right to be put
on a par with the defence. On the other hand, hum an rights treaties do not forbid,
and sometimes even require, the accused to be put in a ‘better’ or more advanta­
geous position than the prosecution in order to preserve an overall balance in the
proceedings (the prosecutor norm ally being better equipped than the defence for
the collection of evidence).
Secondly, equality of the parties is an essential ingredient of the adversarial struc­
ture of proceedings, based on the notion of the trial as a contest between two parties.
Under this approach, it is indispensable for both parties to the proceedings to have
the same rights; otherwise, there is no fair fight between the two contestants’, and the
spectators will not be convinced by the outcome. Here, fairness works both ways and
therefore also the prosecutor is entitled not to be put in a disadvantageous position.

15 See, in particular, such cases as Artico (§32), Barbera, Messegue and Jabardo (§67), Edwards (§36),
Raffineriesgrecqu.es (§49), including the more recent Birutis and others (§§26-35) and Beckles (§§48-66).
The requisites of a fair trial in international law were discussed, with regard to war crimes, by the Supreme
Court of Norway in Latza and others (at 52-85).
Seethe Charter of the IMT (Article 16) and that of the IMTFE (Article 9); it is set forth in the Statutes of the
ICTY (Article 21(2)), the ICTR (Article 20(2)), and the ICC (Article 67, also on the rights of the accused).
16 See, for instance, Article 21 of the ICTY Statute; Article 20 of the ICTR Statute; and Articles 64,66, and
67 of the ICC Statute; Article 17 of the SCSL Statute; and Articles 2 and 6 of the ETSPs Regulation 2000/30
as amended by Regulation 2001/15.
17 For instance, see ICTY, Tadic, Decision on the Prosecution’s Motion requesting Protective Measuresfor
Witness R, at 4; Brdanin, Decision on Third Motion by Prosecution for Protective Measures, §13; Jelisic, AJ,
at §27; Kraijsnik, Decision on Prosecutions’ Motion for Judicial Notice of Adjudicated Facts and Admission
of Written Statements Pursuant to Rule 92bis, §20; SCSL, Brima and others, Decision on the Prosecution’s
Motion for Concurrent Hearing of Evidence Common to Cases, §§35,47.
 GENERAL PRINCIPLES GOVERNING INTERNATIONAL CRIMINAL TRIALS 385

Similar worries do not exist in inquisitorial systems of justice, where proceedings are
conceived of as an official inquiry’.18
Here, I shall focus on the first notion of equality of arms. It acquires particular
importance in cases concerning international crimes. Since these crimes are complex,
may involve multiple defendants, the evidence may be scattered over many countries
and the legal issues at stake may prove complicated, it is crucial for the defendant fully
to exercise a host of fundamental rights and in particular to be assisted by competent
counsel or even a robust team of such counsel.19
In light of the notion of equality of arms under discussion, the defence must be
ensured the following:

1. The defendant is entitled to know full particulars specifying the charges pre­
ferred against him in the indictment.
2. In addition, within the shortest delay, he has the right to examine the evidence
gathered by the prosecution in support of the charges. The ‘discovery’ process is regu­
lated in detail so as to guarantee the defence as far as possible. The ‘disclosure’ of this
evidentiary material must be done by the prosecutor within a set time-period (30 days
after the initial appearance of the accused, under the Rules of the ICTY and ICTR); in
addition, no later than 60 days before the date set for trial, the prosecutor must com­
municate to the defence ‘copies of the statements of all witnesses whom the Prosecutor

18 It would seem that there is some confusion over the two concepts of procedural equality and some
misunderstanding of the case law of the European Court of Human Rights in the case law of the ICTY, e.g.
with regard to the admissibility of evidence and to disclosure of evidence. For instance, in Zlatko Aleksovsi
(Decision on Prosecutor’s Appeal on Admissibility of Evidence) (§§22-8), the AC refused to apply more lenient
standards of admissibility to (hearsay) evidence presented by the defence, stating that the prosecution is also
entitled to a fair trial within the meaning of human rights conventions. This confusion in case law may be
due to the fact that the two different conceptions of equality may, in certain situations, clash.
On the notion that the excessively succinct and summary nature of many defence witness statements vio­
lated the principle of equality of arms, see the ICTY decision in Kupreskic and others of II January 1999, at 2.
19 The question may also arise before national courts dealing with international crimes. It bears men­
tioning in this respect the decision of the Hague Court of Appeal in van Anraat. The defence had claimed
that the principle of equality of arms had been breached, for their financial means in the case had been
largely insufficient (the defendant was assisted by his two counsels on the basis of assignment of legal assist­
ance). The defence claimed that the lack of financial means was serious because the case concerned offences
allegedly committed approximately 20 years earlier in an other part of the world with a totally different cul­
ture and because the investigation had been carried out in many countries all over the world; for that reason,
the defence had not had a reasonable chance to conduct an independent investigation. The Court dismissed
the claim, noting that the defence counsel had had ample opportunity to make investigations and develop
their legal points. The Court conceded that ‘the present criminal case has exceptional proportions, partly
because of its international dimensions and the fact that the offences (serious international crimes) would
have taken place decades ago and mainly in a non-European country. In hearing such a case, especially when
the police and the Public Prosecution Service apparently have ample (extra) financial means available for the
execution of their tasks, one should make sure that the defence does not end up in a relatively disadvanta­
geous position. This could be true if the present rules for financed legal aid should not acknowledge the spe­
cial nature of this case. According to the Court, from this special nature arises the need for a defence carried
out by two counsels working closely together, which indeed they did, also during the hearings. Moreover the
defence brought forward, in general terms, a number of other aspects that hindered them in the perform­
ance of their duties, for lack of financial room’ (§6.1).
386 INTERNATIONAL CRIMINAL LAW

intends to call to testify at trial’ (Rule 66(A)(ii) ICTY).20 The defence has the right to
inspect all books, documents, photographs, and tangible objects in the Prosecutor’s
custody upon request (Rule 66(B) ICTY), but if the defence makes this request, it trig­
gers reciprocal disclosure obligations on the part of the prosecution (see Rule 67(C)).
3. The defendant has the right to appoint one or more defence counsel; if he is indi­
gent, he has the right to a counsel appointed and paid by the Tribunal; in addition, he
has the right to appoint or have appointed by the Tribunal, in case of indigence, one or
more investigators for the purpose of collecting evidence.21
4. The defendant has the right to call witnesses and to cross-examine any witness
called by the prosecution. He also has the right to request depositions. In addition, he
has the right to obtain the attendance of witnesses (for instance, by asking the Court
to subpoena witnesses, or to call as court witnesses persons who would be reluctant to
testify on behalf of the defence).22 Furthermore, the accused may request the granting
of safe-conducts to witnesses who might fear for their liberty, as well as the taking of
testimony by videoconference, whenever witnesses refuse to attend court proceedings
at the C ourt’s seat.

18.4.2 P U B L IC IT Y O F P R O C E E D IN G S

That the proceedings must be public, subject to some exceptions, is a general principle
of m odern criminal law. Publicity of the hearings is a means of better ensuring that the

20 See among other things the Decision in Kupreskic and others on Order ofPresentation of Evidence, at 2-3.
21 It would seem that the court must also make sure that the accused is adequately defended by coun­
sel. In this connection it is interesting to mention Kottsiepen, a case brought after the Second World War
before a British Court of Appeal sitting in Germany. The Court held that defence counsel had not sufficiently
assisted the accused and noted the following: ‘The case came before the Court on 30th March. Counsel who
appeared on behalf of the Appellant stated that he had only been instructed on the previous day, owing to
the illness of Counsel who drew the Notice of Appeal. He was offered an adjournment by the Court, but he
stated that he had read the papers, interviewed his client, and was ready to proceed. It soon became appar­
ent, however, that Counsel had not read the Record of trial which had been supplied to the Appellant, and
he did not appear to appreciate the nature of the proceedings before the Court of Appeal. In spite of repeated
invitations to argue the questions of law raised in the Notice of Appeal, he persisted in irrelevancies’ (at 110).
The Court noted that ‘As no assistance was forthcoming from this source [the defence counsel], the Court
called upon the Director of Prosecutions to deal with the questions of law involved, in the hope that Counsel
would then appreciate what were the issues before the Court. Counsel was then again invited to argue these
questions, but at that stage would only address the Court in mitigation of sentence. It such circumstances
it was obviously unfair to the Appellant to proceed with the case, and it was adjourned to enable him to be
adequately represented. [...] We cannot but deplore that a lawyer should be so lacking in respect for the
Court or for his client’s interests as to appear to argue a case without familiarizing himself with the issues
involved, or with the procedure of the Court’ (at 108-12). Since the Prosecutor was not opposing the quash­
ing of the conviction made by the Trial Court, the Court quashed it, considering that ‘on the findings of the
learned Trial judge, the conviction on the charge as it was framed’ could not be supported (at 112).
See also Hermann, generally on the rights of the defence, and more specifically on the right of the accused
to a finding on an alternative charge (at 164-6). For more recent cases, see for instance the decision of an
ICTY TC in Martic Milan (Decision on Appeal against Decision of Registry), at 2-8.
22 This was repeatedly done, for instance, in Kupreskic and others. See for instance the Decision on Defence
Motion to Summon Witnesses (at 2-3).
 GENERAL PRINCIPLES GOVERNING INTERNATIONAL CRIMINAL TRIALS 387

trial, being under public scrutiny, is fair, in particular that the rights of the accused are
not infringed and that the court conducts the proceedings impartially.23
Nevertheless, the conduct of in camera hearings is provided for whenever this is
required by the need to protect the victims and witnesses (Articles 22 (ICTY Statute),
21 (ICTR Statute), and 68 (ICC Statute)). The decision to hold hearings behind closed
doors is taken by the Court, at the request of one of the parties or proprio motu.

18.4.3 E X P E D IT IO U S N E S S O F P R O C E E D IN G S

One of the obvious requirements of a fair trial is that trial proceedings be as speedy
as possible. Plainly, as the accused enjoys the presumption of innocence until found
guilty (see supra, 18.3), it is only rational and appropriate to establish whether he is
innocent or guilty as rapidly as possible.
This principle, laid down both in treaties on hum an rights and in the Statutes of the
ICTY, the ICTR, the ICC, the SCSL and so on, acquires special importance in inter­
national criminal proceedings, on the following main grounds.
Firstly, very often, for practical reasons, it is difficult or inappropriate for international
courts to release the accused person on bail; hence, frequently the accused is in prison,
from his arrest until conviction or acquittal (see, however, infra, 21.4). This feature of
international trials renders expeditiousness of proceedings all the more necessary.
Secondly, defendants tend not to plead guilty (see also infra, 21.4). Were they to use
this means, they would avoid commencement of trial proceedings proper (or term in­
ate them, if the guilty plea is entered after commencement of trial). They do so either
because they are innocent, or because, even if guilty, they prefer to take the chance of
a trial. They hope that the evidence will be insufficient to establish their guilt; thus,
being acquitted, they will avoid the stigma attaching to international crimes, which
renders perpetrators more odious than authors of common offences. Particularly at
the beginning of the ad hoc Tribunals’ activity pleading guilty being a relatively rare
occurrence, it follows that in most cases the court must conduct trial proceedings. It
should be added that in international proceedings entering a guilty plea is not neces­
sarily part of, nor does it necessarily result from, a process of plea-bargaining (that is
an agreement between prosecution and defence whereby the defendant pleads guilty to
a reduced charge in exchange for the prosecution’s undertaking to drop other charges
or to seek a relatively lenient sentence).
In proceedings before the ICTY TC, so far defendants have entered a guilty plea
on a few occasions.24 At the ICTR, guilty pleas, such as were entered in Kambanda

23 This requirement was not set out in the Charters of the IMT and the IMTFE. It is, however, laid down
in the Statutes of the ICTY (Article 21(2)), the ICTR (Article 20(2)), and the ICC (Article 67(1)), as a funda­
mental right of the accused. It is also proclaimed with regard to national trials, in the European Convention
on Human Rights (Article 6(1)), the UN Covenant on Civil and Political Rights (Article 14(1)), and the
American Convention on Human Rights (Article 8(5)).
24 In Erdemovic the accused pleaded guilty to murder as a war crime (before he had pleaded guilty to
murder as a crime against humanity), and was sentenced to five years’imprisonment (Erdemovic, Sentencing
388 INTERNATIONAL CRIMINAL LAW

(Judgment and sentence), in Serushago (Sentence), and in Ruggiu, have always been
considered as a mitigating circumstance.25
Thirdly, it is notable that often in international criminal trials both questions of fact
and those relating to law prove extremely complex, thereby requiring much time for
their proper consideration. In particular, it may prove necessary to call a great num ­
ber of witnesses, coming from different countries. Furthermore, international courts
must rely on state cooperation for investigations, the gathering of evidence, the appre­
hension of accused, and so on. All this necessarily complicates and slows down the
whole process. In addition, language barriers prolong the proceedings, as normally
the language of the witnesses and the accused is different from that of the court, and,
on top of that, the court is bound to employ more than one official language. Often the
defendant contributes to the length of proceedings by filing many procedural motions,
as he is entitled to do, but this inevitably delays the outcome of the trial.
As noted above (17.4), international criminal tribunals have worked out various
mechanisms and procedures for shortening the length of proceedings. For instance,

Judgment).The Court considered his entering a guilty plea as a mitigating circumstance. It stated that: ‘An
admission of guilt demonstrates honesty and it is important for the International tribunal to encourage
people to come forth, whether already indicted or as unknown perpetrators. Furthermore, this voluntary
admission of guilt which has saved the International Tribunal the time and effort of a lengthy investigation
and trial is to be commended’ (at 16).
In Jelisic the accused pleaded guilty to war crimes and crimes against humanity; he was convicted of plun­
der, cruel treatment, and inhumane acts as well as murder. The prosecution asked the TC to pronounce a life
sentence on the accused (§119). The court sentenced him to 40 years’ imprisonment. (The TC considered the
guilty plea as a mitigating factor, but gave it relatively little weight since the accused had shown no remorse
for his crimes: §127.)
In Todorovic the accused first pleaded not guilty to all charges. Subsequently, when he filed a motion chal­
lenging the legality of his arrest, he reached an agreement with the Prosecutor whereby he would: (i) plead
guilty to some charges; (ii) withdraw all his motions including those concerning the legality of his arrest;
and (iii) cooperate with the prosecution. Under the agreement, the Prosecutor undertook to recommend to
the TC a sentence of not less than 5 years’ and not more than 12 years’ imprisonment. Both parties under­
took not to appeal against any sentence imposed by the TC within that range. The TC held that a guilty plea
‘should, in principle, give rise to a reduction in the sentence that the accused would otherwise have received’,
adding that a guilty plea facilitates the work of the Tribunal by avoiding a possible lengthy trial with all the
attendant difficulties, and ‘relieves victims and witnesses of the necessity of giving evidence with the attend­
ant stress which this may incur’ (§80). It also stated that it was in no way bound by the agreements between
the prosecution and the defence, noting that it was ‘the Chamber’s responsibility to determine an appropri­
ate sentence in this case’ (§79). The TC sentenced Todorovic to 10 years’ imprisonment.
In Sikirica and others (Sentencing judgment) the three accused made a plea agreement with the prosecu­
tion after the prosecution case had ended and the defence case had commenced. The accused admitted a
number of facts and the prosecution agreed to recommend a reduced sentence. The TC sentenced Sikirica to
15 years’ imprisonment, Dosen to 5 years, and Kolundzija to 3 years (§245).
25 Kambanda, after reaching an agreement with the prosecution, pleaded guilty to all counts in the
indictment; the agreement entered into with the prosecution expressly stated that no agreements, under­
standings, or promises had been made between the parties with respect to sentence, which remained at the
discretion of the TC (Kambanda, §48). The court sentenced the accused to life imprisonment (§62). The AC
confirmed the sentence (Kambanda (AJ) (§126). In Serushago (Sentence), after entering into an agreement
with the prosecution, the accused pleaded guilty to four of the five counts in the indictment (they included
genocide and crimes against humanity). The TC sentenced him to 15 years’ imprisonment (§42). The sen­
tence was confirmed by the AC (Serushago (Appeal, reason for judgment), §34).
 GENERAL PRINCIPLES GOVERNING INTERNATIONAL CRIMINAL TRIALS 389

the following measures taken by the ICTY can be mentioned: (i) a pre-trial judge as
well as pre-trial conferences have been provided for; (ii) time limits for the filing of pro­
cedural or preliminary motions have been set; (iii) provision has been made for admis­
sion of written evidence and in particular for the filing of affidavits. Furthermore, (iv)
the num ber of judges has been increased, in particular through the election of ad litem
judges (that is, non-permanent judges, or not-full-time judges, who only sit in one or
two cases), which required an amendment to the ICTY Statute by SC resolution no.
1329 (2000) of 30 November 2000 (a similar amendment was adopted for the ICTR).

18.5 THE PRINCIPLE THAT THE ACCUSED

SHOULD BE PRESENT AT HIS TRIAL

Common law countries always require that the accused be present at trial, for trial
proceedings to be commenced. As noted above (16.2), this requirement is chiefly dic­
tated by the adversarial nature of the common law trial: being substantially a ‘duel’
between the prosecution and the defence, it must of necessity presuppose the attend­
ance of both adversaries, so that each of them may gather evidence on his own behalf
and cross-examine the witnesses called by the adversary. Probably it is also felt that a
trial where the accused is absent may lend itself to abuses, particularly in authoritarian
countries. (Indeed on many occasions undemocratic states have used trials in absentia
to convict and sentence political dissidents living abroad.) Also, international courts
have emphasized the importance of the defendant’s attendance at trial, stressing in
particular that his presence serves to verify the accuracy of his statements and com­
pare them with those of the victim and other witnesses.26
In some countries constitutional reasons underpin this approach. For instance, in
the United States, the Supreme Court ruled in Gagnon (at 1484) that the ban on trials
in absentia was rooted in both the ‘Confrontation Clause’ of the Sixth Amendment
and the ‘Due Process Clause’ of the Fourteenth Amendment (the former grants every
person the right ‘to be confronted with the witnesses against him’, while the latter
provides that a state shall not ‘deprive any person of life, liberty, or property, with­
out due process of law’). Nevertheless, even countries banning trials in absentia do
allow criminal trials to go on if, after appearing in court, the defendant deliberately
absconds to avoid trial: this is for instance admitted by the US Supreme Court (see
Taylor (at 18-29) and Crosby (at 748-53)).
In contrast, many civil law countries perm it such trials (see 17.2.6). This institu­
tion is not connected with the inquisitorial system prevailing in civil law countries,
as proved by the fact that such a country as Italy, which embraced the accusatorial

26 This was emphasized by the European Court of Human Rights in Poitrimol v. France §35 and reiterated
in Krombach v. France (§86).
390 INTERNATIONAL CRIMINAL LAW

system a few years ago, allows trials in absentia, whereas countries such as Spain and
Germany, which have adopted the inquisitorial system, rule them out.27
The rationale behind trials in absentia is that one should not allow justice to be
thwarted by the accused, when he chooses to escape instead of standing trial. In these
systems, the defendant has a legal entitlement to be present at his trial; if he absconds
and flees the jurisdiction even before commencement of trial, he implicitly waives that
entitlement. Were judges to be barred from proceeding by the defendant’s absconding,
this would mean that ultimately criminal justice could be kept at bay by the accused.
In addition, as the European Court of Human Rights put it in Colozza v. Italy, ‘the
impossibility of holding a trial by default may paralyse the conduct of criminal pro­
ceedings, in that it may lead, for example, to dispersal of the evidence, expiry of the
time-limit for prosecution or a miscarriage of justice’ (§29).
Moreover, in countries upholding a system whereby the evidence on behalf of the
accused is gathered either by the investigating judge or by the prosecutor, at least in
principle the defendant is not put at a disadvantage. Holding trial proceedings is not
detrimental to the defendant, for his case may be made in court by a defence counsel
(appointed by the court, if need be) on the basis of the evidence impartially collected
by the investigating judge (or the prosecutor, if he is requested under the legal system
to act in the objective interests of justice and not solely as a party to the subsequent
proceedings). Hence, under these systems, it is admissible that criminal trials be con­
ducted in the absence of the defendant, provided some basic safeguards, forcefully
spelled out by the European Court of Hum an Rights, are respected: (i) the accused
m ust have been formally notified of the charges against him; (ii) there is evidence that
he is deliberately absconding or at any rate absenting himself from the proceedings;
(iii) the accused must be granted the right to be defended in court by counsel (Lala
and Pelladoah v. The Netherlands, §§33-4,40-1; Krombach v. France, §§88-91); (iv) the
accused has the right to appear in court at any moment and request that proceedings
be commenced again, even if he has already been convicted; in other words, he has the
right to obtain from a court that has heard his case ‘a fresh determination of the merits
of the charge’ (Colozza v. Italy, §29; Krombach v. France, §§85,87); indeed, ‘the author­
ities have a positive obligation to afford the accused the opportunity to have a complete
rehearing of the case in his or her presence’ (Krombach v. France, §87).
As national legal systems differ on this matter, can one find an international rule
prohibiting or allowing such trials? Indisputably, there is no international treaty
provision prohibiting them. In addition, the existing treaty rules on trial proceed­
ings, contained in such treaties as the UN Covenant on Civil and Political Rights

27 Probably this is so because they regard the right of the accused to be present as an overriding human
right, or because they consider it ‘uneconomic’ to hold trials that are not directed at attaining the ultim­
ate purpose of any criminal trial, namely, to do justice, and therefore put the accused in jail, if convicted.
However, in Germany another ground is advanced: as C. Roxin (Strafverfahrensrecht, Munich: Beck, 1991,
at 405) put it, ‘the trial judge must personally see the accused in front of him, in order to arrive at the
right picture of his personality’. See also B. Swart, ‘La place des criteres traditionnels de competence dans la
poursuite des crimes internationaux’, in A. Cassese and M. Delmas-Marty, Juridictions nationales et crimes
internationaux (Paris: PUF, 2002), 581-3.
 GENERAL PRINCIPLES GOVERNING INTERNATIONAL CRIMINAL TRIALS 391

(Article 14(3)) and the European Convention on Hum an Rights (Article 6(1)) have been
interpreted by the relevant international bodies as not ruling out trials in absentia.28
W hat should then be the condition of international criminal proceedings? Some
international treaties regulate the matter. Thus, Article 12 of the Charter of the IMT
explicitly allowed trials in absentia ‘in the interests of justice’.29 In contrast, the Statute
of the ICC, at Article 63(1), makes the presence of the accused a basic requirement for
the commencement of trials (with the exception, normally provided for in common
law countries, that the defendant’s presence is not required if he disrupts the trial: see
Article 63(2) and Rule 124 If the ICC RPE).30 Also Article 5 of the SPET Regulations
2000/30 (as amended by regulation 2001/25) and Rule 80 of the ECCC exclude trials
in absentia.
More recently, however, Article 22 of the Statute of the STL takes a contrary stand: it
allows trials in absentia, subject to a set of conditions.31 Similarly, trials in the absence
of the accused are provided for in Rule 56 of the RPE of the IHC, which refers to Iraqi

28 Thus, the UN Human Rights Committee held in Daniel Mbenge v. Zaire (1983) (§§13-14.2), in Hiber
Conteris v. Uruguay (1985) (§§9.2-10) and in Dieter Wolfv. Panama (1992) (§6.5) that the relevant state had
breached Article 14(3) (d), but only because trial proceedings had commenced without the accused having
knowledge of the proceedings against him. In Raphael Henry v. Jamaica (1991) (§8.3) the Committee held
that the state had not breached that provision because the accused had opted for representation by coun­
sel and therefore could not claim that his absence during the appeal hearing constituted a violation of the
Covenant. Furthermore, in Daniel Mbenge v. Zaire the Committee stated that Article 14(3)(d) ‘and other
requirements of due process enshrined in Article 14 cannot be construed as invariably rendering proceed­
ings in absentia inadmissible irrespective of the reasons for the accused person’s absence. Indeed, proceed­
ings in absentia are in some circumstances (for instance, when the accused person, although informed of the
proceedings sufficiently in advance, declines to exercise his right to be present) permissible in the interest of
the proper administration of justice’ (§14.1).
The European Court of Human Rights held that the right of the accused to participate in the trial,
although not expressly laid down in Article 6 of the European Convention on Human Rights, is a right
‘whose existence is shown by the object and purpose of the Article taken as a whole’ (Colozza v. Italy, §27,
Brozicek v. Italy, §45). Nonetheless, the Court did not rule that trials in absentia were as such incompatible
with Article 6: see, for instance, Colozza v. Italy (§§27.30), Poitrimolv. France (1993, §§30-39) and Krombach
v. France (§§82-91).
29 Indeed, one of the accused, Martin Borman, was tried and sentenced in his absence (vol. I, at
338-41).
30 On this matter see, among other things, AC, Milosevic, Decision on Interlocutory Appeal of the Trial
Chamber Decision on the Assignment of Defence Counsel, §20; ICTR, TC Barayagwiza, Decision of defence
Counsel Motion to Withdraw, §§5-7; SCSL, Gbao, Ruling on the Issue of the Refusal of the Third Accused,
Augustine Gbao, to Attend Hearing of the Special Court for Sierra Leone, §8; Sesay, Ruling on the Issue of
the Refusal of the Accused Sesay and Kallon to Appear for their Trial, §15. See also ICTY, TC, Naletilic and
Martinovic, Decision on Prosecutor’s Motion to Take Depositions for Use at Trial, at 1-2.
31 T. The Special Tribunal shall conduct trial proceedings in the absence of the accused, if he or she:
a. has expressly and in writing waived his or her right to be present;
b. has not been handed over to the Tribunal by the State authorities concerned;
c. absconded, or otherwise cannot be found and all reasonable steps have been taken to secure his
or her appearance before the Tribunal and to inform him or her of the charges confirmed by the
Pre-Trial Judge.
2. When hearings are conducted in the absence of the accused, the Special Tribunal shall ensure that:
a. the accused has been notified, or served with the indictment, or notice has otherwise been given
of the indictment through publication in the media or communication to the State of residence or
nationality;
392 I N T E R N A T I O N A L C R I M I N A L LAW

law (pursuant to which defendants can be tried in their absence if they may not have
been arrested, or escape after being taken into custody.)32
The relevant provisions of the Statutes of the ICTY, the ICTR, and the SCSL do not
enshrine any specific provision on the matter. They provide that the accused has the
right ‘to be tried in his presence’ (Articles 21(4)(d), 20(4)(d), respectively). However,
nothing is said about the case where the accused implicitly waives this right by
absconding before trial proceedings commence: one could easily construe this pro­
vision to the effect that the trial may be conducted in the absence of the accused if he,
after being duly notified of the indictment and the charges contained therein, were to
flee in order to evade criminal justice.
The question therefore arises of how international courts should behave when
international treaty provisions or similar rules, regulations or provisions are silent on
the matter.
A convincing solution may be found by both drawing upon general principles and
looking at the specificities of international criminal proceedings.
In favour of trials in absentia one can argue that it would be contrary to law and
justice to authorize the alleged perpetrator of gruesome crimes to make a mockery of
international justice by preventing trials through his deliberate absence. In addition,
the contention is warranted that the rights of the accused are not jeopardized if pro­
vision is made for his right to a fresh trial, should he decide to surrender after being
tried and convicted in his absence. Furthermore, if the procedure before the relevant
court is shaped in such a m anner as to ensure that the judges are in a position fully to
consider and appraise the evidence, in particular exculpatory evidence, the absence of
the accused does not necessarily taint the proceedings as unfair.
However, whenever international crim inal proceedings are substantially based on
the adversarial system, in case of doubt the specificities of international trials may
warrant the view that the accused must be present at trial, before trial proceedings
are commenced. The combination of the demands and implications of the adversar­
ial system with the unique features of international criminal proceedings may make
this requirement indispensable. In international trials the search for and collection of
evidence may prove extremely difficult, because: (i) as a rule the court is headquartered
in a country far away from the place of the crime; (ii) witnesses may be scattered over
many countries; (iii) there is no investigating judge charged with collecting evidence

b. the accused has designated a defence counsel of his or her own choosing, to be remunerated either
by the accused, or if proven to be indigent, by the Tribunal;
c. whenever the accused refuses or fails to appoint a defence counsel, such counsel has been assigned
by the Tribunal’s Defence Office with a view to ensuring full representation of the interests and
rights of the accused.
3. In case of conviction in absentia, the accused, if he or she had not designated a defence counsel of his
or her choosing, shall be retried in his or her presence before the Special Tribunal, unless he or she
accepts the judgement.’
32 Para. 135 of the Iraqi Criminal Proceedings Law, with Amendments no. 23 of 1971.
 GENERAL PRINCIPLES GOVERNING INTERNATIONAL CRIMINAL TRIALS 393

on behalf of both the prosecution and defence; under the accusatorial system each
party must search for and find the necessary evidence; (iv) the lack of an international
body of investigators under the control of the prosecutor endowed with the power
of freely going wherever the evidence may happen to be, to question witnesses and
collect material evidence, makes investigations undertaken for the purpose of sup­
porting or dismissing charges extremely complex. This task is all the more difficult
for the defence, given that, in addition to the problems facing the prosecution, the
defence encounters a further hurdle: it may not count on the considerable num ber of
investigators normally available to the prosecution, nor on the court’s power to order
investigators to collect evidence on the defence’s behalf; (v) in most cases matters are
further complicated by the fact that the crimes with which the accused is charged have
been perpetrated years before the court proceedings start.
As a result of these features of international criminal trials, it proves crucial for the
fair conduct of proceedings that the accused be present when the fundamentals of the
adversarial model are upheld. If present, the defendant may, first, issue instructions
to his defence counsel, or consult with them, on the evidence that could be collected
to support his case, so as to make counsel able to muster all the necessary probative
materials in rebuttal of the prosecution’s case. Secondly, the presence of the accused
is im portant for the cross-examination of prosecution witnesses, for he can suggest to
his defence counsel points or issues on which to conduct cross-examination. Thirdly,
the accused may prove of great importance for the court in its findings, because he
may decide to testify in court, and in any case his behaviour and appearance might
be of some relevance to the court in establishing whether he may be found guilty or
innocent.
It is doubtful whether one should infer from the above that, if the defendant escapes
after the commencement of trial, one should opt for the extreme solution of staying
proceedings until he is arrested and brought to trial again, or voluntarily surrenders.
Arguably the characteristics of international criminal proceedings emphasized above
should rule out the solution normally adopted at the national level by countries that
ban trials in absentia; that is, continuation of trial if the accused absconds after the
initiation of proceedings (see for instance Rule 43 of the US Federal Court Rules of
Procedure). It could be argued that at the national level this solution is justified by
the need both to safeguard the public interest in dispensation of justice and to pre­
vent individuals from evading adjudication. At the international level—so it could
be held—the need to ensure a fair trial and to avoid any miscarriage of justice should
instead prevail.33 In addition, there would be no problem with regard to statutes of

33 When it seemed that states were intent on refusing to cooperate with the Tribunal and therefore no
trial could be held for lack of indictees at The Hague, the ICTY adopted an imaginative measure designed
to respect the principle whereby the accused must be present for a proper trial to be conducted, while at
the same time taking some action to react to the lack of cooperation of states and their consequent refusal
to detain persons accused of appalling crimes. This measure was Rule 61 of the Rules of Procedure and
Evidence (‘Procedure in Case of Failure to Execute a Warrant’). Under this Rule, if an arrest warrant has
not been executed, the confirmation judge may order that the Prosecutor submit the indictment to the TC
394 IN T E R N A T IO N A L C R IM IN A L LAW

limitations (another rationale for holding trials in absentia in civil law countries),
since international crimes do not fall under any statute of limitations.
However, it seems to me that this extreme legal solution should be avoided, because
otherwise a defendant absconding after commencement of trial would be in a position
to stultify international justice outright. Moreover, as most international criminal
courts are not permanent, a possible stay of proceedings might result in the defendant
evading justice for good, once the court is terminated.
It should be added, nevertheless, that even courts which in principle rule out
in absentia trials have held that under certain circumstances in absentia proceed­
ings may exceptionally be warranted. The ICTY AC envisaged such a case in Blaskic
(Subpoena). It held that if an individual does not comply with a subpoena or order
issued by the Tribunal, he can be held in contempt of the Tribunal and the specific con­
tempt procedure can therefore be set in motion. Should the individual also fail to attend
contempt proceedings, ‘in absentia proceedings should not be ruled out’ (§59).34

of which the judge is a member. The Prosecutor then submits the indictment in open court, together with
the evidence available, and may call and examine witnesses. The TC may conclude that ‘there are reasonable
grounds for believing that the accused has committed all or any of the crimes charged in the indictment’,
whereupon it: (i) issues an international arrest warrant to be transmitted to all states; and (ii) may freeze the
assets of the accused; if the absence of the accused is due to lack of cooperation of a state, it (iii) may request
the Tribunal’s President to notify the SC of the failure of the relevant state to comply with Article 29 of the
ICTY Statute (on the obligation of states to cooperate). Clearly, this procedure is not a trial proper. It is a
sort of fall-back, designed to stimulate states to arrest indictees, by making public and exposing the charges
preferred by the Prosecutor against some persons.
The procedure was resorted to on five occasions between 1995 and 1996 (Nicolic; Martic; Mrskic, Radic,
and Slijvancanin; Raijc; Karadzic and Mladic). Since then it has not been relied upon, on account of the
increasing number of arrests made either by states or by the NATO forces in Bosnia and Herzegovina, which
enabled the Tribunal’s Chambers to conduct trials proper.
34 The AC held that, although it was not appropriate to hold such proceedings against persons falling
under the primary jurisdiction of the Tribunal (that is, accused of one of the international crimes over which
the Tribunal had jurisdiction under its Statute), ‘By contrast, in absentia proceedings may be exception­
ally warranted in cases involving contempt of the International Tribunal, where the person charged fails
to appear in court, thus obstructing the administration of justice. These cases fall within the ancillary or
incidental jurisdiction of the International Tribunal’ (§59). The AC added, however, that all the necessary
judicial safeguards should be offered to the absent defendant (§59).
It should be noted that the Chamber’s judgment was unanimous; this means that the three judges from
common law countries sitting on the AC concurred with the Chamber’s ruling on the question of in absentia
proceedings as well.
 19
STAGES OF INTERNATIONAL
PROCEEDINGS IN OUTLINE

I P R E -T R IA L A ND T RIA L

(A) PROSECUTORS INVESTIGATIONS

AND PRE-TRIAL PROCEEDINGS

19.1 GENERAL
It may prove useful to summarize the unfolding of international proceedings, from
the investigations initiated by the Prosecutor to appeal or revision proceedings. I will
undertake the exposition of these proceedings on the basis of the relevant provisions
of the Statutes and the Rules of Procedure and Evidence (RPE) of the various inter­
national tribunals as well as their judicial practice, if any.

19.2 THE SETTING IN MOTION OF

INTERNATIONAL CRIMINAL INVESTIGATIONS
(A) ICTY and other tribunals
The Statutes of these tribunals provide that the Prosecutor alone decides whether or not
to commence investigations, either ex officio or on the basis of information obtained
from any source, particularly from governments, UN organs, intergovernmental, or
non-governmental organizations. The Prosecutor assesses the information received or
obtained and decides whether there is a sufficient basis to proceed.1 Thus, no formal

1 Articles 18(1) ICTY Statute and 17(1) ICTR Statute, Article 15(1) of the SCSL Statute, Article 11(1) of
the STL Statute.
396 I N T E R N A T I O N A L C R I M I N A L LAW

right of complaint is granted to the alleged victims nor is the power of governments
to set in motion investigations provided for, although both the victims, governments,
and international organizations are endowed with a similar power.
The prosecutor enjoys very broad discretion as to whether or not to initiate inves­
tigations and against whom. It is not clear why any decision as to the initiation of
investigations into a specific case is left to the Prosecutor alone. Probably, at least in the
case of the ICTY, ICTR, SCSL, and the STL the UN SG (and the UN SC) considered
that: (i) the existence of numerous reports about the alleged crimes perpetrated made
it superfluous to grant a right of complaint proper; in addition (ii) such a right would
have triggered the proceedings even when the alleged crimes were of m inor im port­
ance and it was therefore not appropriate for them to be amenable to international
justice; furthermore, (iii) to grant a right of complaint to governments might have
enabled states to act on political grounds or at any rate might have prompted politic­
ally motivated states to make use of criminal justice for their own ends.

(B) ICC
The Statute provides that investigations may be initiated: (i) at the request of a state
party to the Statute; or (ii) by the Prosecutor proprio motu; or (iii) at the request of the
UN Security Council acting under Chapter VII of the UN Charter. The alleged victims
of crimes or non-governmental organizations acting on their behalf have no right to
refer a case to the Court, but the Prosecutor may make use of the information and
allegations they might submit. Probably it was felt that to grant such a right would
have resulted in the Court being flooded with innumerable complaints, most of them
probably frivolous or unfounded. This should not apply to the right of a state to lodge
a complaint: states are expected carefully to screen allegations of crimes made by the
victims or by private organizations, with a view to ascertaining whether they are sup­
ported by reliable evidence.
A second notable feature of the ICC system is that the Statute clearly draws a distinc­
tion between preliminary probing, as we shall term it here, and investigation proper. The
need for preliminary scrutiny to precede investigations is provided for in Article 15
with exclusive regard to cases where the Prosecutor decides to take proceedings prop­
rio motu. The probing consists of a search for information or the gathering of evi­
dence about an alleged crime, for the purpose of establishing if there is a reasonable
basis to proceed with an investigation’. The initiation of the prelim inary scrutiny by
the Prosecutor on his own initiative is based on any relevant information he may
have received from any reliable source, as well as ‘written or oral testimony at the
seat of the C ourt’ (Article 15(1)). If and when he has established that such ‘reason­
able basis’ exists, the Prosecutor must submit to the P-TC a ‘request for authorization
of an investigation’. If the P-TC grants the request, the Prosecutor may commence
the investigation. By contrast, if a ‘situation’ is referred by a state or by the UN SC,
the Prosecutor’s request to the P-TC for authorizing an investigation is not required.
Clearly, it is assumed that the national authorities of the referring state have already
undertaken an inquiry, and that the SC, through one of its subordinate bodies, has
 ST A G E S O F I N T E R N A T I O N A L P R O C E E D I N G S I N O U T L I N E 397

already made a first screening of information relating to the possible perpetration of

crimes (this happened in the case of Darfur, where the SC referred the situation to the
ICC by its resolution 1593(2005), based on the report of the International Commission
of Inquiry on Darfur contained in UN doc. S/2005/60).
Article 14 ICC Statute provides for the initiation of investigations at the request of
a state into a ‘situation’ in which one or more crimes under the C ourt’s jurisdiction
appear to have been committed. It stipulates that as far as possible’ the state should
‘specify the relevant circumstances’ and provide the necessary ‘supporting docum en­
tation’. Interestingly, the Statute does not set out any further requirements; in particu­
lar, it does not require that the requesting state be the national state of the victim or
the alleged perpetrator, or the state on whose territory the crime has been allegedly
committed. Hence, any contracting state, even a state that has no link whatsoever with
the crime, may file the request. This regulation is clearly based on the principle that the
crimes under the C ourt’s jurisdiction are of universal concern; consequently any state
possessing the relevant information may bring them to the attention of the Prosecutor.
So far no state has made use of this power (which was to be expected, for it is a fact of
international life that states rarely accuse officials or nationals of other states of direct
or indirect involvement in serious international crimes).
Commencement of investigations at the request of the SC may be undertaken by
the Prosecutor whenever that UN body, ‘acting under Chapter VII of the UN Charter’
(Article 13(b)), refers to the Prosecutor ‘a situation in which one or more’ crimes under
the jurisdiction of the ICC appear to have been committed. Plainly, the SC may only
submit to the ICC ‘situations’ involving serious crimes the perpetration of which
amounts to a ‘threat to the peace’ (or a ‘breach of the peace’)—these are, however,
broad notions involving a wide discretionary power of the SC in their interpretation.

19.3 CONDITIONS THE PROSECUTOR

MUST FULFIL BEFORE INITIATING
AN INVESTIGATION
(A) ICTY and other tribunals
Under the Statutes of the ICTY, ICTR, and SCSL, the Prosecutor not only has absolute
freedom to decide whether or not to initiate investigations and against whom, but also is
free to carry out investigations outside any judicial scrutiny (although he must, of course,
comply with a set of obligations regarding the conduct of investigations and the rights of
the suspects, and in addition has to turn to a judge to request the issuance of arrest war­
rants or any other orders as may be required for the conduct of the investigation or the
preparation of trial). Judicial scrutiny of the probative material collected and charges pre­
ferred by the Prosecutor is only made at the end of the investigations, when the Prosecutor
submits an indictment to a reviewing judge, who may admit or dismiss it.
398 IN T E R N A T IO N A L C R IM IN A L LAW

However, in practice the huge discretionary power by the Prosecutor has been grad­
ually limited by suggesting that he or she should prosecute those who appear to bear
the greatest responsibility for international crimes; that is, the leaders and organizers
of those crimes. This need to concentrate on the major suspects is justified by both
practical reasons (international tribunals are costly and may not afford to bring to trial
all the perpetrators of international crimes, including the so-called ‘small fry’), and by
the very rationale for the establishment of such tribunals, which is to dispense justice
with regard to the most serious crimes affecting the whole international community
on account of their gravity.
In the ICTY as early as 1995 the Judges unanimously urged the Prosecutor to work
out a programme of indictments against those most responsible for serious crimes.2 In
2004 the UN SC passed resolution 1534 which, among other things, called upon both
the ICTY and the ICTR ‘in reviewing and confirming any new indictments, to ensure
that any such indictments concentrate on the most senior leaders suspected of being
most responsible for crimes within the jurisdiction of the relevant Tribunal’. In light
of this resolution the ICTY Judges adopted a rule granting an administrative body of
the Tribunal (the Bureau, consisting of the President, Vice-president and the Presiding
judges of trial chambers) the task of establishing whether or not indictments issued by
the Prosecutor concerned ‘the most senior leaders’ suspected of crimes.3
This new trend has subsequently been taken into account when adopting the stat­
utes of the most recent tribunals. Thus, for instance, Article 1 of the SCSL Statute
provides that the Special C ourt ‘shall [...] have the power to prosecute persons who
bear the greatest responsibility for serious violations of international hum anitarian
law [...] including those leaders who, in committing such crimes, have threatened the
establishment of and implementation of the peace process in Sierra Leone’. In Brima
and others the SCSL held that the requirement that the Court should prosecute ‘per­
sons who bear the greatest responsibility’ for the crimes over which the C ourt has
jurisdiction ‘solely purports to streamline the focus of the prosecutorial strategy’
(§653). The TC thus rightly disagreed with the other TC that in Fofana Decision on the

2 On 30 January 1995 the Judges adopted a resolution expressing their concern about the urgency with
which appropriate indictments should be issued’ and then recalled that ‘the security Council, in establishing
the International Tribunal by resolutions 808 and 827, intended expressly to entrust it with the historic mis­
sion of bringing to trial those responsible for “mass killings”, “organized and systematic detention and rape
of women”, and “the practice of ethnic cleansing in the territory of the former Yugoslavia” in order thereby
to “contribute to the restoration and maintenance of peace’”. The resolution went on to state that ‘Due to the
gravity and historic dimension of that mission, the Judges are anxious that a programme of indictments
should effectively meet the expectations of the Security Council and of the world community at large’ (ICTY,
CC/PIO/003-E, 1 February 1995). The Prosecutor indicated that he shared that concern and a few months
later issued indictments against some major leaders.
3 Rule 28 (A) provides as follows: ‘On receipt of an indictment for review from the Prosecutor, the Registrar
shall consult with the President. The President shall refer the matter to the Bureau which shall determine
whether the indictment, prima facie, concentrates on one or more of the most senior leaders suspected of
being most responsible for crimes within the jurisdiction of the Tribunal. If the Bureau determines that the
indictment meets this standard, the President shall designate one of the permanent Trial Chamber Judges
for the review under Rule 47. If the Bureau determines that the indictment does not meet this standard, the
President shall return the indictment to the Registrar to communicate this finding to the Prosecutor.’
 ST A G E S O F I N T E R N A T I O N A L P R O C E E D I N G S I N O U T L I N E 399

preliminary Defence motion on the lack o f personal jurisdiction had held instead that,
the issue of personal jurisdiction being a jurisdictional requirement, ‘in the ultimate
analysis, whether or not in actuality the Accused is one of the persons who bears the
greatest responsibility [...] is an evidentiary matter to be determined at the trial stage’
(§44). The TC in Brima and others rightly noted that it could not accept the idea that
the drafters of the Statute had made the requirement at issue ‘a jurisdictional threshold
which, if not met, would oblige a Trial Chamber to dismiss the case without consider­
ing the merits’ (§653).

(B) ICC
The ICC Statute limits the power of the Prosecutor by providing that the Court shall
only have jurisdiction over the gravest international crimes, thereby ruling out crimes
by m inor perpetrators or of scant gravity: pursuant to preambular paras 4 and 9 and
Article 5(1) it shall only deal with ‘the most serious crimes of concern to the inter­
national community as a whole’ (see also Articles 1 and 8(1)).
Furthermore, the actions of the Prosecutor are subject to a set of conditions, when­
ever the initiation of investigations (i) has been requested by a state; or (ii) has been
made by the Prosecutor on his own initiative. (These conditions do not apply when the
investigations have been requested by the SC.)
In the case of a state request, as well as in the case of the Prosecutor acting proprio
motu, the Prosecutor, as pointed out above, must first of all determine whether ‘there
would be a reasonable basis to commence an investigation’ (Articles 18(1) and 53(1)).
If he is satisfied that there is such a basis, he must notify ‘all States Parties and those
States which, taking into account the information available, would normally exercise
jurisdiction over the crimes concerned’ (Article 18(1), which formally speaking only
applies to referrals by states).
If there has been a referral by the Security Council, the Prosecutor may find other
means of enabling states to become cognizant of the referral and of the possible initiation
of court investigations. Only if no state concerned is investigating the alleged crime, nor
has the alleged author been brought to trial before a national judge, or instead a state
concerned is investigating or conducting judicial proceedings, but is clearly ‘unwilling
or unable’ to do justice, may the Prosecutor initiate investigations proper.
As stated above, if a preliminary probing has been initiated by the Prosecutor prop­
rio motu, in addition to these conditions, it is necessary for the Prosecutor to submit to
the P-TC a request for authorization of an investigation, together with any supporting
material collected (Article 15(3)—(5)). The Prosecutor may initiate investigation only
after obtaining such judicial authorization (and also if the other conditions mentioned
above are fulfilled).
The differentiation between the three instances of initiation of proceedings can be
easily explained. In the case of referral to the Court by the SC, it has been considered
that a ruling by the P-TC was not necessary. The fact that the crimes submitted by the
SC to the Court involve a threat or even a breach of the peace has been considered of
param ount importance and at any rate sufficient to remove that condition.
400 IN T E R N A T IO N A L C R IM IN A L LAW

The condition, imposed only for cases where the Prosecutor initiates the inquiry on his
own motion, that the P-TC should first authorize the conduct of investigation, is aimed
at limiting the power of the Prosecutor. The same condition has not been regarded as
necessary when the initiative is taken by a state. Clearly it has been thought that before
submitting a case to the ICC a state party gives due consideration to the importance
and significance of the step it takes in bringing a situation before the Court.
It is, however, important to emphasize that, conversely, no distinction is made among
the three categories of instances as far as the possibility of challenging the admissibility
of a case is concerned. W hether the case is brought to the Court by a state or the SC, or is
initiated by the Prosecutor, the accused or other persons involved as well as states have
the right to challenge the admissibility of the case prior to confirmation of the indict­
ment containing the charges; that is, during investigation (see Article 19).

19.4 CONDUCT OF INVESTIGATIONS BY

THE PROSECUTOR

(A) ICTY and other tribunals

As stated above, as a rule the adversarial system prevails. Hence, the situation is dif­
ferent from that in civil law countries (such as France, Belgium, or Spain) which still
have an investigating judge (juge d ’instruction). There, it is first the Prosecutor who,
through his staff, conducts prelim inary investigations and gathers the available evi­
dence against the suspect; then, if he considers that he has a prima facie case, he hands
the case over to the investigating judge. It falls to this judge to search out and collect
evidence, both that against the accused and the evidence exculpating him. Such evi­
dence, together with all decisions made by the judge on procedural and other matters,
is then handed over to the court in a case file (dossier).
Conversely, in international proceedings the Prosecutor gathers evidence against
the suspect in his investigations. It is primarily for the defence to look for and collect
evidence aimed at refuting the charges (although if the Prosecutor finds exculpatory
evidence, he is duty bound to disclose it to the defence).
In the conduct of investigations, the Prosecutor may need to have a suspect arrested.
Under Rule 40(A) ICTY, he will have to request the relevant state to arrest the suspect
and place him in custody. However, under Rule 40(B), if a state is prevented from keep­
ing the suspect in custody or is unable or unwilling to take the measures necessary to
prevent his escape, the Prosecutor may apply to a judge of the ICTY designated by the
Tribunal’s President for an order to transfer the suspect to the seat of the Tribunal or
any other place decided upon by the Tribunal’s Bureau.
Rule 40bis also provides for the possibility of the Prosecutor requesting a judge of
the Tribunal to issue an order for the transfer to, and provisional detention of a suspect
in, the detention unit of the Tribunal.
 ST A G E S O F I N T E R N A T I O N A L P R O C E E D I N G S I N O U T L I N E 401

Similar provisions apply at the ICTR and the SCSL. In contrast, as we saw above
(17.4(C)) the ECCC provides for a civil law system based on investigating judges, who,
once they are requested by the prosecutors to deal with a case, gather the evidence on
behalf of both the prosecution and the defence and, where they do not dismiss the
prosecutors’ submissions, prepare a case file proper, which is then handed over to both
parties and the Court.

(B) ICC
I have already mentioned above that in the ICC the system is in some respects closer
to the inquisitorial model and perhaps more attuned than the ICTY and ICTR scheme
to the specific requirements of international criminal trials. Under Article (54)(l)(a) of
the ICC Statute, the Prosecutor is under the obligation to gather evidence both against
and in favour of the suspect or accused. It is for the P-TC to issue orders of arrest and
other orders requested by the Prosecutor (Article 58 of the ICC Statute).
Interestingly, under Article 56 of the ICC Statute, upon request of the Prosecutor
and while the investigation is being conducted by him, the P-TC may take measures
to collect or preserve evidence, whenever such evidence might not be available subse­
quently for the purposes of trial.

19.4.1 T H E N E E D F O R C O O P E R A T I O N BY STA TES

The conduct of investigations involves the search for and collection of evidence. To do
so, any international prosecutor perforce needs to rely upon the cooperation of states.
Indeed, the suspects, the victims, or any witnesses are on the territory of a sovereign state.
The Prosecutor has no power or authority to carry out his functions on such territory. To
discharge his mission he therefore needs the cooperation of all the relevant countries.
Such cooperation may take two different forms: (i) at the request of the Prosecutor,
the national authorities (prosecutors or investigating judges, depending on the national
legislation) may carry out all the actions required by the Prosecutor, for instance, ques­
tion suspects, victims, or witnesses, conduct on-site investigations, seize documentary
evidence or other evidentiary material; (ii) they may authorize the international pros­
ecutor to carry out investigations on national territory, if need be with the assistance
of specially designated national authorities (judges, prosecutors) or of the national
authorities that are territorially competent.
Clearly, the second form of cooperation is by far more internationally oriented and
favourable to the expansion of the Prosecutor’s powers.

(A) ICTY and other tribunals

The choice between the two different modes of cooperation very much depends on the
attitude of individual states. For instance, as far as the ICTY is concerned, in their imple­
menting legislation some states (such as Australia, France, Italy, New Zealand, and Spain)
tend to attribute to national judicial authorities the power to collect evidence and per­
form other acts necessary for the Prosecutor’s investigations. Other states (for instance,
402 IN T E R N A T IO N A L C R IM IN A L LAW

Austria, Finland, Germany, and Switzerland), on the other hand, tend to authorize the
Prosecutor to fulfil at least some tasks autonomously on the national territory.
In Blaskic (Subpoena) the ICTY AC held that normally the International Tribunal
m ust turn to the relevant national authorities for the collection of evidence, the seiz­
ure of evidentiary material, etc. However, the Tribunal’s Prosecutor was authorized
directly to carry out such activities on the territory of a state in two situations: (i) when
the state was one of the former belligerents or entities of the former Yugoslavia (§53);
and (ii) when such investigative activity was authorized by national implementing
legislation (§55). In addition, according to the AC, the Tribunal was authorized to
reach out directly to private individuals living on the territory of a state when such
individuals were needed to testify in court or deliver a particular document and the
state concerned had refused to comply with an order of the Tribunal; in such instances
the Tribunal could directly summon a witness or order an individual to hand over
evidence or appear in court (§§55-6).4
O f course, whenever international police or m ilitary forces are available which
are lawfully stationed on the territory of a state where evidence may be found,
the Prosecutor may tu rn to them for assistance in the gathering of evidence. This
happened in Kordic and Cerkez (Decision on defence motion to suppress evidence),
where an ICTY TC held that the search and seizure of docum ents in Bosnia and
Herzegovina by members of the Office of the Prosecutor accom panied by forces
of SFOR was ‘perfectly w ithin the powers of the Prosecution provided for in the
[ICTY] Statute’ (at 4).

(B) ICC
The problem of cooperation by states proves of special importance for the ICC, on vari­
ous grounds. First, the Statute’s provisions on cooperation are numerous and detailed

4 The AC justified the first exception as follows: ‘The first class encompasses States: (i) on the terri­
tory of which crimes may have been perpetrated; and in addition, (ii) some authorities of which might be
implicated in the commission of these crimes. Consequently, in the case of those states, to go through the
official channels of identifying, summoning and interviewing witnesses, or to conduct on-site investiga­
tions, might jeopardize investigations by the Prosecutor or defence counsel. In particular, the presence of
state officials at the interview of a witness might discourage the witness from speaking the truth, and might
also imperil not just his own life or personal integrity but possibly those of his relatives. It follows that it
would be contrary to the very purpose and function of the International Tribunal to have state officials pre­
sent on such occasions. The states and Entities of the former Yugoslavia are obliged to cooperate with the
International Tribunal in such a manner as to enable the International Tribunal to discharge its functions.
This obligation (which, it should be noted, was restated in the Dayton and Paris Accords), also requires
them to allow the Prosecutor and the defence to fulfil their tasks free from any possible impediment or
hindrance’ (§53).
As to the third exception, the AC found that it was justified on the following grounds: ‘In the above-
mentioned scenarios [that is, if the national authorities refuse to co-operate and therefore prevent an indi­
vidual from testifying or handing over evidence] the attitude of the State or Entity may jeopardize the
discharge of the International Tribunal’s fundamental functions. It is therefore to be assumed that an inher­
ent power to address itself to those individuals inures to the advantage of the International Tribunal. Were it
not vested with such a power, the International Tribunal would be unable to guarantee a fair trial to persons
accused of atrocities in the former Yugoslavia’ (§55).
 ST A G E S O F I N T E R N A T I O N A L P R O C E E D I N G S I N O U T L I N E 403

(Articles 86-102) and, in essence, impose upon contracting states both general and
specific obligations to cooperate. However, in case of refusal or failure to comply with
such obligations, the Court can only ‘make a finding to that effect’ and refer the mat­
ter to the Assembly of States Parties (or to the UN SC, where a ‘situation’ had been
referred by this organ to the Court) (Article 87(7)). If the failure to cooperate comes
from a state not party to the Statute that had entered into an ad hoc agreement or
arrangement with the Court, the Court may inform the Assembly of States Parties or,
depending on whether the matter had been referred to it by the SC, such organ.
Secondly, the general scheme of relations between the C ourt and states is substan­
tially based on a ‘horizontal’ approach: states are not subordinate to the Court but on
its level, as it were (see supra, 16.6 and 8). It follows that if a state decides not to cooper­
ate, the Court (through the Assembly of States Parties) can only fall back on the usual
international law mechanisms for inducing compliance with international obliga­
tions. It lacks any special authority, or power, or means of putting into effect its orders,
or generally discharging its mission, on the territory of a recalcitrant state party.5
Thirdly, the Court acts upon the principle of complementarity. In other words, it
only adjudicates cases where national prosecutorial or judicial authorities are unable
or unwilling to deal with a case (see supra, 16.3). One of the consequences is that,
except where the relevant state consents to the exercise of the C ourt’s jurisdiction,
the Court substitutes for national authorities. Proceedings commence before the ICC
only if national authorities have been labelled by the Court as ‘unwilling or unable
genuinely to prosecute’, or as having held trial proceedings ‘not conducted independ­
ently or impartially’ or ‘inconsistent with the intent to bring the person concerned to
justice’ (Article 17). Whenever this is so, it follows that those national authorities are
most unlikely to be prepared to cooperate with the ICC, for instance in the collection
of evidence, service of documents, execution of searches, and seizures.

19.4.2 R I G H T S OF S U S P E C T S A N D O T H E R P E R S O N S
INV O LV ED IN IN V ESTIG A TIO N S

Any person involved in investigations, for instance suspects (that is, any person about
whom there are grounds to believe that he may have committed an international
crime) or persons questioned as witnesses (whether or not they may become suspects)
possesses under customary and treaty law a set of fundamental rights. Such rights are
protected, albeit implicitly, in the Statutes of the various international tribunals. They
are laid down in much detail in the Statute of the ICC.

5 However, under Article 57(3)(d) of the ICC Statute the P-TC may authorize the Prosecutor 'to take
specific investigative steps within the territory of a State Party without having secured the cooperation of
that state under Part 9 [on International Co-operation and Judicial Assistance] if, whenever possible hav­
ing regard to the views of the state concerned, the P-TC has determined in that case that the state is clearly
unable to execute a request for cooperation due to the unavailability of any authority or any component of
its judicial system competent to execute the request for cooperation under Part 9’.
404 IN T E R N A T IO N A L C R IM IN A L LAW

Under Article 55(1) of the ICC Statute these rights are granted to suspects even
when investigations and other prelim inary activities are carried out by state author­
ities at the request of the Prosecutor. These rights include the right: (i) to be questioned
in a language that the person understands or to be assisted by an interpreter without
payment; (ii) not to be subjected to any form of coercion or threat; (iii) not to be sub­
jected to any form of cruel, inhum an, or degrading treatment; (iv) not to incriminate
him self or to confess guilt; and (v) not to be arbitrarily deprived of liberty.
Persons suspected of an international crime possess in addition the following rights
(laid down in Article 55(2) of the ICC Statute as well as in customary law): (i) to be
informed, prior to questioning, that there are grounds to believe that they have com­
mitted an international crime; (ii) to be cautioned that any statement the suspect
makes shall be recorded and may be used in evidence; (iii) to remain silent, without
their silence creating a presumption of guilt; (iv) to be legally assisted by a person
freely chosen or assigned by the court’s registry, at the court’s expense; (v) to be ques­
tioned in the presence of counsel.

19.4.3 S U B M IS S IO N O F T H E I N D I C T M E N T O R C H A R G E S

(A) ICTY and other tribunals

In the ICTY, ICTR, SCSL, and STL system the outcome of investigations may be the
drawing up, by the Prosecutor, of an indictment, containing ‘a concise statement of the
facts of the case and of the crime with which the suspect is charged’. The Prosecutor
may proceed to take such a step whenever he ‘is satisfied in the course of an inves­
tigation that there is sufficient evidence to provide reasonable grounds for believing
that a suspect has committed a crime within the jurisdiction of the Tribunal’ (Rule
47(B) of the ICTY RPE). The indictm ent must set forth the name and particulars of
the suspect; it must also contain a concise statement of facts and an indication of the
charges preferred against the suspect. The Prosecutor may withdraw the indictment
without prior leave, at any time before its confirmation by a judge; thereafter, the
withdrawal needs the leave of a judge or the TC (see, for instance, Rule 51(A) of the
RPE of the ICTY). Similarly, the Prosecutor may amend the indictm ent without prior
leave, before confirmation, whereas thereafter he can do so only with the leave of a
judge or, depending upon the case, of the competent TC (see for instance Rule 50 of
the RPE of the ICTY).
Of course the system prevailing in the ECCC is different, due to the presence of
investigating judges (see supra, 17.4(C)).

(B) ICC
Under the ICC Statute, at the end of investigations the Prosecutor submits charges
setting out the facts and the crimes of which the suspect is accused (see Article 61).
Interestingly, the Prosecutor does not enjoy a discretional power to conclude that,
upon investigation of a case, there is insufficient basis for a prosecution. Whenever he
 ST A G E S O F I N T E R N A T I O N A L P R O C E E D I N G S I N O U T L I N E 405

reaches this conclusion, he must so inform the P-TC, stating the reasons therefor, as
well as, if it is a state or the SC that has referred a situation, that state or the SC (Article
53(1) and (2) of the ICC Statute). Either at the request of the state or the SC or, depend­
ing upon the case, on its own initiative, the P-TC may review the Prosecutor’s decision
and request him to reconsider it (Article 53(3)).6 This entails that the Prosecutor’s dis­
cretionary power is not unqualified but subject to judicial scrutiny.
This legal regulation appears to be meritorious in that: (i) it sets out the general
standards by which the Prosecutor may decide whether or not to prosecute a case; (ii)
obliges the Prosecutor to give reasons for his deciding not to prosecute, and in add­
ition (iii) empowers the P-TC to reverse his decision. Thus,, the otherwise unlettered
powers of the Prosecutor are significantly restricted and any abuse is forestalled or, in
any case, may be checked.

(C) The question of cumulative charging

A question that has arisen many times is whether the charges made by the Prosecutor
may be cumulative for the same act (for instance, the same m urder is characterized
and charged both as a war crime and as a crime against humanity). In Akayesu an
ICTR TC held that cumulative charges could be made in three instances (where the
offences charged had different legal ingredients; where the relevant provisions pro­
tected different interests; and where it proved necessary to record a conviction for
multiple offences in order fully to describe what the accused had done) (§§461-70).
In a decision in Tadic an ICTY TC held that the m atter was only theoretical and had
no practical relevance.7 In Kupreskic and others, an ICTY TC held that two seem­
ingly conflicting requirements must be reconciled, namely ‘the requirement that the
rights of the accused be fully safeguarded’ and the requirem ent that ‘the Prosecutor
be granted all the powers consistent with the [ICTY] Statute to enable her to fulfil
her mission efficiently and in the interests of justice’ (§724). In the light of these
requirements, the Chamber concluded that the Prosecutor should make cumula­
tive charges if the facts charged violated simultaneously two or more provisions of
the Statute, and charge in the alternative when an offence appeared to be in breach
of more provisions, one of them being special to the other (for instance, a m urder
could be charged as a crime against hum anity and in the alternative, for the event of
the widespread or systematic practice not being proved, as a war crime) (§§720-7).
However, both another TC, in Brdanin and Talic (Decision on the amended indict­
ment) (§§29-43), and the ICTY AC, in Delalic and others (§400), disallowed this
view (see supra, 8.8.1).

6 However, under Article 53(3)(b) the P-TC may, on its own initiative, review a decision of the Prosecutor
not to proceed only if such decision has been made on the grounds that the prosecution would not serve the
interests of justice on account of the gravity of the crime, or the interests of the victims, or the age or infirm­
ity of the alleged perpetrator, and his or her role in the alleged crime.
1 Tadic (Decision on the Defence motion on the Form of the indictment), at 10.
4o6 IN T E R N A T IO N A L C R I M I N A L LAW

19.4.4 C O N F I R M A T I O N P R O C E E D I N G S

(A) ICTY and other tribunals

The procedure for reviewing indictments adopted by the ICTY and other Tribunals
differs from that of the ICC. The RPE of these Tribunals do not provide for what is con­
sidered an essential safeguard in adversarial systems of justice, namely a ‘preliminary
hearing’ in which the Prosecutor must, in an open and (normally) adversary hearing,
establish that there is a prim a facie case (As a rule such preliminary judicial screening
is intended to prevent discriminatory or ‘vindictive’ prosecution.) This is probably due
to the fact that originally the RPE envisaged confirmation of the indictment as a pre­
requisite for issuing an arrest warrant.
In the ICTY and ICTR system the Prosecutor must submit the indictment, together
with the ‘supporting material’ (that is, all the material designed to corroborate the
charges brought by the Prosecutor) to the judge designated by the President8 (under
the Statute of the STL to the Pre-Trial Judge: Articlel8(l)). The judge reviews the
indictment and supporting material and may: (i) confirm the indictment; (ii) dismiss
it; (iii) request the Prosecutor to present additional material; or (iv) adjourn the review
so as to give the Prosecutor the opportunity to modify the indictment. The reviewing
judge performs these alternative acts in a hearing that is ex parte; that is, without the
suspect or his counsel, and is held in camera. Pursuant to Article 19(1) of the ICTY
Statute, the standard for confirmation of the indictment is that there must be a prima
facie case,9 a standard that would seem to be more exacting than that required for the
Prosecutor’s appraisal (which is that ‘there is sufficient evidence to provide reasonable
grounds for believing that a suspect has committed a crime’, Rule 47(B) ICTY).10
Upon confirmation of the indictment, the suspect acquires the status of accused.
In addition the judge may, at the request of the Prosecutor, issue one or more arrest
warrants or any other order sought by the Prosecutor, which the judge deems
appropriate to make in the interests of justice.

(B) ICC
In contrast, in the ICC system the Prosecutor submits the charges to the P-TC, which
holds a public hearing in the presence of the ‘person charged’ (unless such person
waives his right to attend, or absconds and may not be detained, and the Chamber
decides nonetheless to hold the hearing), or his counsel. The purpose of the hearing is
to enable the Chamber to ‘determine whether there is sufficient evidence to establish

8 As for the ICTY, see however Rule 28 (A) mentioned above.

9 Judge McDonald, in her decision in Kordic and others (Review of the indictment) held that, for the pur­
poses of confirmation, the prima facie case standard meant that there was a credible case which would, if not
contradicted by the defence, be a sufficient basis to convict the accused on the charge (at 1123).
10 In his Separate Opinion in the decision in Rajic (Review of the indictment), Judge Sidhwa held that this
requirement means that it is sufficient for the Prosecutor to point to ‘such facts and circumstances as would
justify a reasonably or ordinary prudent man to believe that a suspect has committed a crime’ (at 1065).
For a detailed discussion of Rule 47(B), see D. Hunt, ‘The Meaning of a Prima Facie Case’, in R. May et al.
(eds.), Essays on ICTY Procedure and Evidence (The Hague: Kluwer, 2001), 137-49.
 ST A G E S O F I N T E R N A T I O N A L P R O C E E D I N G S I N O U T L I N E 407

substantial grounds to believe that the person committed each of the crimes charged’
(Article 61(7)). The P-TC may: (i) confirm the charges or some of them, and commit the
person to a TC for trial; (ii) decline to confirm the charges ‘in relation to which it has
determined that there is insufficient evidence’; or (iii) adjourn the hearing to enable
the Prosecutor to submit further evidence or amend the charges (Article 61(7)).n
Upon confirmation of the charges the P-TC may issue an arrest warrant or a sum ­
mons to appear (Article 58). (Under the same provision the P-TC may issue arrest war­
rants even before a person is formally charged, so long as the investigation has been
initiated and if there are reasonable grounds to believe that the person has committed
a crime’ within the C ourt’s jurisdiction.)

19.5 PR E-T R IA L PROCEEDINGS

(A) ICTY and other tribunals

Pre-trial proceedings commence with the initial appearance of the accused and his
entering a plea of guilty or not guilty. In this initial hearing the court reads to the
accused the indictment or the charges previously confirmed by a reviewing judge,
satisfies itself that the accused understands the nature of the charges and is assisted by
defence counsel, and then asks him whether he pleads guilty or not guilty.
If the accused pleads guilty, the TC enters a finding of guilt—provided the requisite
conditions are met.112The TC also instructs the Registrar to set a date for the sentencing
hearing (see e.g. Rule 62bis ICTY). If instead the accused pleads not guilty, the TC:
(i) shall ensure that the Prosecutor discloses to the defence, within the prescribed
time limit (30 days after the initial appearance of the accused, under Rule 66(A)
(i) of the ICTY) the supporting material which accompanies the indictment or the
charges at the time of confirmation; and (ii) appoints a pre-trial judge, charged with
coordinating communication between the parties during the pre-trial phase. It may
also convene a status conference to organize exchanges between the parties so as to
ensure expeditious preparation for trial.
In the period before initiation of trial proper, the parties may file prelim inary
motions, namely motions which: (i) challenge the court’s jurisdiction; or (ii) allege
defects in the form of the indictment or charges; or (iii) seek severance of counts
joined in one indictment or separate trials, or instead seek joinder of trials; or
(iv) raise objections concerning the assignment of counsel. The parties may also file
other motions that are not preliminary in nature (for instance, motions for the provi­
sional release of the accused, for the disqualification or recusal of a judge, etc.).

11 See also Rules 121-6.

12 These conditions were first laid down jurisprudentially in the Erdemovic (AJ) and then codified in
paragraphs (i) to (iv) of Rule 62bis ICTY.
408 IN T E R N A T IO N A L C R IM IN A L LAW

Motions are considered and pronounced upon by the TC and, subject to certain
conditions, may be appealed to the AC.
Once the Prosecutor’s disclosure of evidence is completed and prelim inary motions,
if any, are disposed of, the pre-trial judge orders the Prosecutor to file within a certain
time limit: (i) a pre-trial brief addressing the factual and legal issues; (ii) admissions
by the parties and a statement of matters which are not in dispute; (iii) a statement of
contested matters of fact and law; (iv) a list of witnesses the Prosecutor intends to call,
with among other things a sum m ary of the facts on which each witness will testify;
(v) a list of exhibits the Prosecutor intends to offer stating where possible whether the
defence has any objection as to authenticity. The pre-trial judge may convene one or
more status conferences, where measures are taken to ensure expeditious preparation
for trial, and in addition the status of the accused is reviewed.
As soon as the Prosecutor has completed all his filings and all the motions have been
disposed of, the pre-trial judge submits to the TC a complete/he (which however may
not be compared to the ‘case file’ of inquisitorial systems, where the investigating judge
records the evidentiary material gathered for both the prosecution and the defence, as
well as all his decisions on preliminary motions and other issues). This file includes all
the documents filed by the parties, transcripts of the status conferences, and minutes of
meetings held by the judge with the parties (Rule 65ter (L)( i) ICTY). It enables the TC
to hold a Pre-Trial Conference where, if the TC ‘considers that an excessive number of
witnesses are being called to prove the same facts’, it may call upon the Prosecutor to
reduce the number of witnesses he intends to call, or to shorten the estimated length of
the examination-in-chief for some witnesses (Rule 73 bis (C) ICTY).
W ith the assistance of the aforementioned ‘complete file’ the TC is thus in a position
to commence trial.

(B) ICC
Also in the ICC system pre-trial proceedings commence with the initial appearance
of the accused and his entering a plea of guilty or not guilty. Such proceedings unfold
before the P-TC. The Statute allots extensive powers to this Chamber. After the arrest
of the accused and his surrender to the Court, the Chamber must satisfy itself that
the accused has been informed of both the charges against him and his rights under
the Statute (Article 60(1)). He has, among other things, the right to apply for interim
measures pending trial. In particular, he may apply for provisional release. In this
phase the Prosecutor may amend the charges after giving notice to the accused and
provided that the P-TC has authorized such amendment. It is also in this phase that
the Prosecutor must proceed to the disclosure of the ‘evidentiary’ materials he has col­
lected (this matter is regulated by Rules 76-84).
Interestingly, under Rule 121(10) the Registry must keep a ‘full and accurate record’
of the proceedings before the Chamber, ‘including all documents transm itted to the
Chamber’, hence also the ‘supporting material’ on which the charges preferred by the
Prosecutor are based. The Prosecutor, the person subject to an arrest warrant or to a
summons, and victims and their legal representatives are entitled to consult this record.
The record is then transmitted to the TC before trial proceedings open (Rule 131).
 ST A G E S O F I N T E R N A T I O N A L P R O C E E D I N G S I N O U T L I N E 409

It would seem that in this way the ICC system goes much further than the two ad
hoc Tribunals in making available evidentiary’ material to the TC before trial. Thus,
it would seem that an important feature of the inquisitorial system has been to some
extent incorporated into the ICC procedure.

(B) TRIAL PROCEEDINGS

19.6 CASE PRESENTATION

(A) ICTY and other tribunals

In the ICTY, ICTR, and SCSL, normally the trial begins with an opening statement of
the Prosecutor, where he sets out the main elements of charges and outlines his case.
The defence may, if it so wishes, also make an opening statement, although normally
defence counsel prefer to make such a statement when the Prosecutor’s case rests, that
is at the beginning of the defence case (see e.g. Rule 84 ICTY).
The accused has the right to make a statement ‘under the control of the TC’, without
taking an oath. This statement is not testimony, hence the accused is not examined and
cross-examined upon it. However, the TC may decide to give it probative value (Rule
84bis ICTY). Allowing the accused to make a statement that is not a piece of evidence
is an exception to the normal scheme of the adversarial system, where the accused
may, if he so wishes, give evidence on his own behalf, but is then examined and cross-
examined. The IMT Charter, in Article 24(j), had already envisaged this departure;
however, that rule provided for the defendants to make a statement at the end of the
trial, after the closing statements of the prosecution and defence—this was of course
in addition to their testifying as witnesses. (Thus defendants effectively spoke after
the case was closed; that is, after the trial had finished and all the evidence had been
produced; so the IMT could probably tolerate this departure because in effect it was
happening ‘outside the trial proper’.) The reason behind this departure from most
common law systems is that this is the only opportunity for defendants freely—that
is, without being cross-examined—to set out their general views and explain their
motivations or why they consider they are innocent.
The Prosecutor then presents his case. To this end, he calls witnesses and produces
exhibits. Witnesses are first examined-in-chief by the Prosecutor, then cross-examined
by defence counsel, and subsequently re-examined by the Prosecutor.13

13 As happens in the practice of some common law countries (but not in England, where it is strictly for­
bidden), also in international criminal tribunals before examining witnesses in court, each party is entitled
to undertake ‘witness preparation, also called ‘witness proofing; that is, to rehearse the examination-in­
chief, by asking the witness all the necessary questions. In this way the whole testimony is rehearsed. It is
commonly stated that, as a rule of thumb, a prosecutor or defence counsel should never ask a question in
4 io IN T E R N A T IO N A L C R IM IN A L LAW

The usual rules on the nature and limits of examination-in-chief (or direct
examination, in American terminology), cross-examination, and re-examination (or
redirect, in American terminology) apply. Thus, in examining witnesses-in-chief the
Prosecutor or the Defence must refrain from asking leading questions (that is, questions
that suggest the answer, such as: ‘Was the car yellow?’), whereas such questions may be
put in cross-examination.14
The traditional reason behind the prohibition of leading questions in examination-
in-chief is that in criminal trials held in common law countries the jury is required to
hear the full information, or the account of events, directly from the witness, without
any interference from prosecutors or defence counsel. In short, jurors are expected
to hear information about facts not through the prosecutor or defence counsel but as
directly as possible from the witness. In contrast, in cross-examination the prosecutor
and defence counsel are allowed to ask leading questions so as to put their case to the
witness and to cast doubt on the acceptability or credibility of the witness.
However, judges sitting on international courts tend to be more flexible or at
any rate are not bound by strict rules. They may therefore allow leading questions in
examination-in-chief, thereby overruling objections by the other party, if they consider
that such questions may justifiably be put in the interests of justice, in particular to speed
up the proceedings. Even in common law systems, leading questions are allowed where
the witness is dealing with matters that are not contentious, or are agreed between the
parties, and it seems appropriate to apply a similar rule to international tribunals.
Cross-examination ‘shall be limited to the subject-matter of the evidence-in-chief
and matters affecting the credibility of the witness and, where the witness is able to
give evidence relevant to the case for the cross-examining party, to the subject-matter
of that case’ (Rule 90(H)(i) of the ICTY). A further requirement is set out for cross-
examination: under Rule 90(H)(ii) of the ICTY:

examination-in-chief or cross-examination without previously knowing the answer. This practice, which
could sound odd or unfair to lawyers of civil law countries, is among other things aimed at: (i) focusing, in
the questions and answers, on the key issues of testimony; (ii) reducing the witness’s anxiety about his tes­
timony in court and at the same time building in him a feeling of security and confidence; and (iii) putting
the witness in a proper frame of mind to be effective in his testimony or with a view to avoiding receiving
a surprising answer which could be damaging to one’s client (especially in cross-examination, although in
examination-in-chief it could be partly also a way of making the witness comfortable). It is also important
to ‘control’ the witness, for instance asking short, specific questions, so that the witness does not go ranting
off on other subjects.
Often, prosecutors and defence counsel also simulate cross-examination, so as to better prepare the wit­
ness to questions from the other side.
In Kupreskic and others (Decision on communications between the parties and their witnesses) the Trial
Chamber ruled that, once a witness had made the ‘solemn declaration’provided for in Rule 90(1) he could no
longer communicate with the party that had called him, except with the leave of the Chamber (at 3).
14 A legal provision on leading questions can be found in Rule 611(C) of the 2001 US Federal Rules of
Evidence (‘Leading questions should not be used on the direct examination of a witness except as may be
necessary to develop the witness’ testimony. Ordinarily leading questions should be permitted on cross-
examination. When a party calls a hostile witness, an adverse party, or a witness identified with an adverse
party, interrogation may be by leading questions.’)
 ST A G E S O F I N T E R N A T I O N A L P R O C E E D I N G S I N O U T L I N E 411

In the cross-examination of a witness who is able to give evidence relevant to the case for
the cross-examining party, counsel shall put to that witness the nature of the case of the
party for whom that counsel appears which is in contradiction of the evidence given by the
witness.
This is an im portant requirement. For example, a witness may testify that he saw
the accused commit a killing and the defence does not put to the witness in cross-
examination that he could have been mistaken; then in his closing speech the defence
argues that the witness could have been mistaken as it was night-time at the time of
the killing. There is then a basic unfairness. 'The witness should have been given the
opportunity to respond to that suggestion when he was on the witness stand, when he
might have been able to give a convincing rebuttal (e.g. ‘Nonsense, it was broad day­
light’ or ‘But I was standing only a metre away’).
A party may also ask the TC to authorize questions relating to additional matters;
that is, matters not raised in examination-in-chief.15
Judges may at any stage put questions to the witnesses (Rule 85(B) ICTY). In the
practice of the ICTY, if judges ask questions at the end of re-examination and these
questions are not directly related to matters raised in examination-in-chief or cross-
examination, then the parties are authorized to examine and cross-examine the wit­
ness on those specific matters.
Generally speaking, judges have broad powers in directing the examination of wit­
nesses. Their guiding principle is that they must conduct business in the interest of
justice so as to ensure a fair trial. They therefore enjoy considerable latitude. They
exercise their powers by ruling on possible objections by the counter-party to specific
questions put by the Prosecutor or defence counsel, as the case may be. Also, and more
generally, they direct the case presentation by deciding what measures should be taken
to facilitate the testimony of vulnerable witnesses, by ruling on the admissibility or
relevance of evidence, controlling the m anner of questioning to avoid any harassment
or intimidation of the witness, and by deciding on written or oral motions submitted
by the parties with respect to the questioning of witnesses.16
After the close of the Prosecution’s case, the pre-trial judge orders the defence to
file a list of the witnesses it intends to call, with a summary of the facts on which each
witness will testify and the estimated length of time required for each witness, plus
a list of exhibits (Rule 65fer (G) ICTY). In addition, the TC may hold a Pre-Defence
Conference, where it may call upon the defence to reduce the number of witnesses
it intends to call to prove the same facts or to shorten the estimated length of the
examination-in-chief for some witnesses (Rule 73fer ICTY). The defence then makes

15 Under Rule 90(H)(iii) ICTY RPE, "Ihe Trial Chamber may, in the exercise of its discretion, permit
enquiry into additional matters.’ See also ICTY, Decision in Kupreskic and others on limitation of scope of
cross-examination of character witnesses, at 2.
16 Under Rule 90(F) of the ICTY RPE, ‘The Trial Chamber shall exercise control over the mode and order
of interrogating witnesses and presenting evidence so as to (i) make the interrogation and presentation
effective for the ascertainment of the truth; and (ii) avoid needless consumption of time/
41 2 IN T E R N A T IO N A L C R IM IN A L LAW

an opening statement and calls witnesses, who are questioned in accordance with the
rules set forth above, and may of course also produce exhibits.
At the end of the defence case, the Prosecutor may present evidence in rebuttal, then
the defence may submit evidence in rejoinder, and the Court may have evidence ordered
by it to be presented (court evidence). Normally witnesses called by the court are ques­
tioned first by the judges, then cross-examined by the Prosecutor and subsequently by
defence counsel; this sequence is established in the interest of the defence, which has
thus the opportunity to first hear the questioning by the judges and the prosecution.
The Court may then re-examine the witnesses, with the usual caveat that if in so doing
it raises matters not previously considered in examination or cross-examination, the
Prosecutor and the Defence have the right to cross-examine on such matters.
Once all the evidence has been presented, the Prosecutor makes a closing argument,
followed by a closing statement by the Defence. In these arguments both parties, in
addition to summing up their appraisal of the evidence and to setting out their main
arguments on points of fact and law, are obliged to address sentencing matters in their
closing speeches (ICTY Rule 86(C)).17
It should be noted that while the above procedure, substantially based on the
adversarial system, prevails in international criminal tribunals, a slightly different
procedure is adopted by the Statute of the STL, Article 20(2) of which provides that
‘Unless otherwise decided by the Trial Chamber in the interests of justice, examination
of witnesses shall commence with questions posed by the presiding judge, followed
by questions posed by other members of the Trial Chamber, the Prosecutor and the
Defence.’ Here the Statute—it would seem—adopts a m ixture of the inquisitorial
system (where witnesses are first questioned by the Judges and then, if need be, by the
parties, but cross-examination proper is absent), and the adversarial system (where
witnesses are always examined and cross-examined by the parties and the Judges
tend to play a passive role). The advantage of the former system’s approach is that the
Judges, in their search for judicial truth, tend to go to the heart of the case and refrain
from putting questions that may lead to a distortion of the evidence; in addition, they
prove to be fairer than the contending parties to vulnerable witnesses. However, if the
accused feels that the questions put by the Judges to a witness are not fair and tend
to damage him, he has no means of putting things right (unless his defence counsel
then asks the witness questions that aim at supporting the defendant’s case). As for
the adversarial system’s approach, it no doubt ensures a more thorough and in-depth
questioning of witnesses. However, as has been rightly noted, it can easily distort the

17 This rule has been much attacked by the defence, who say: how can we address sentencing matters
when (a) our position at that point is that the accused is completely innocent (‘Your Honours, my client is
completely innocent, but if you find him guilty, please bear in mind that he only beat the victims with his
fists and not with a stick’!); and (b) we do not know what factual findings the Chamber will make. It is like
the old schoolboy plea, when charged with breaking the window in the headmaster’s study: (i) first, there is
no witness in the headmaster’s study; (ii) if there is a window, it is not broken; (iii) if it is broken, I did not do
it; (iv) if I did it, it was an accident. This does not sound very convincing as a closing speech—the protestation
of innocence is undercut by what sounds like admissions by the accused.
 ST A G E S O F I N T E R N A T I O N A L P R O C E E D I N G S I N O U T L I N E 413

evidence, because the people who ask the questions do so in the hope of obtaining
answers that fit the case they are putting forward’.18
A provision similar to that of Article 20(2) of the STL Statute cited above can be
found in Rule 90 of the ECCC. There, the order of questioning of the witnesses is,
however, more consistent with the inquisitorial system generally adopted by the
Extraordinary Courts, which hinges on the crucial role of investigating judges (who
hand over to the Court as well as the parties a case file proper, namely a file containing
‘the written records of the investigative action undertaken in the course of a prelimin­
ary investigation [by the Prosecutors] or a judicial investigation [by the investigating
judges], together with the application by the parties, written decisions and any attach­
ments thereto at all stages of the proceedings, including the record of proceedings
before the Chambers’ (Glossary appended to the Rules).

(B) ICC
The Statute and the Rules of the C ourt tend to leave much latitude to TCs on the
m ethod of case presentation. Article 64(8) (b) provides that ‘the presiding judge may
give directions for the conduct of proceedings’ and that, subject to such directions,
‘the parties may submit evidence in accordance with the provisions of this Statute’.
Rule 140 confirms this flexibility for it provides that each party is entitled to exam­
ine a witness it has called and to cross-examine a witness called by the adversary, but
it falls to the TC to decide whether to question a witness ‘before or after’ a witness is
questioned by the prosecution or the defence (Rule 140 (2)(c)). It would seem that a
mixed m ethod (adversarial-inquisitorial) is therefore suggested, but it is ultimately
left to the TC to decide whether to place greater emphasis on the adversarial approach
(by asking the prosecution and defence to examine and cross-examine the witnesses
first, followed by the court) or for the inquisitorial approach (in which case the witness
is first questioned by the Judges and then by the parties, who may also cross-examine
him). One point is, however, set out clearly and may not be departed from by the court:
‘the defence shall always have the opportunity to speak last’.
Furthermore, it has wisely been provided that matters relating to sentencing be sep­
arately addressed by the parties, before the end of trial, in ‘additional hearings’ (see
Article 76 and Rule 143).

19.7 RULES OF EVIDENCE

(A) ICTY and other tribunals

The most fundamental principle relating to the taking of evidence in international
trials is common to all systems based on the adversarial model and is indeed inherent
in this model. Under this principle any written or oral testimony, document, or other

—18 J. R. Spencer, ‘Evidence’, in Delmas-Marty and Spencer, European Criminal Procedures, at 629.
 4H I N T E R N A T I O N A L C R I M I N A L LAW

piece of documentation only become evidence if admitted in court after being the sub­
ject of arguments by the parties. In other words, no evidence proper exists outside court
proceedings. Also affidavits and any exhibit or evidentiary material may only become a
piece of evidence after being presented in a court hearing by the party concerned, being
discussed or agreed between the parties, and declared admissible by the court.
A second fundamental principle, which is not shared with common law systems
and is unique to international proceedings, is that courts are not bound by strict and
‘technical’ rules of evidence but enjoy great flexibility and should be guided, rather
than by formal standards, by general principles of fairness.
That in international trials rules of evidence should be simplified as much as pos­
sible was first proposed in 1945 by the US delegate to the London Conference. In illus­
trating paragraphs 17 and 18 of the American draff of the Proposed Agreement19 to
the representatives of the other three Powers, Justice Robert H. Jackson, the US repre­
sentative, stated

We do not want technical rules of evidence designed for jury trials to be used in this case to
cut down what is really and fairly of probative value, and so we propose to lay down as a part
of the statute [of the future IMT] that utmost liberality shall be used [...] The idea may have
more significance to British and American lawyers than it does to Continental lawyers.20

As provided in Article 19 of the IMT Charter, an international court may adopt and
apply ‘to the greatest possible extent expeditious and non-technical procedure’, and
admit ‘any evidence which it deems to have probative value’. This regulation of the
administration of evidence is premised on the notion that: (i) there is no jury consist­
ing of lay people without any expert knowledge; the court is made up of professional
judges, who are in a position to appraise the probative value of each piece of evidence;
(ii) the specific features of international criminal proceedings require courts to be
flexible and to be guided prim arily by the need to ensure a fair and expeditious trial.
It follows that, among other things, a TC ‘may exclude evidence if its probative
value is substantially outweighed by the need to ensure a fair trial’ (Rule 89(D) of
the ICTY).21
W ithin this general context, some specific rules on certain matters have evolved
and may be held to be customary in nature; consequently, under general principles
of international law, they may be derogated from by courts if their statutes or rules of
procedure and evidence so require (see supra, 1.4.1-3).
In international criminal proceedings it now seems accepted that the standard of
proof should be that judges must be convinced beyond a reasonable doubt of the guilt
of the accused before they may convict. This standard is laid down in Rule 87(A) of

19 International Conference on Military Trials, at 59.

20 Ibid., at 83.
21 In Blaskic a TC insisted on the general principle of liberal admission of evidence: ‘The principle
embodied by the case-law of the Trial Chamber on the issue is the one of extensive admissibility of evi­
dence-questions of credibility or authenticity being determined according to the weight given to each of
the materials by the Judges at the appropriate time’ (§34).
 ST A G E S O F I N T E R N A T I O N A L P R O C E E D I N G S I N O U T L I N E 415

the ICTY and in Article 66(3) of the ICC Statute. It has also been upheld in the ICTY
case law.22
Courts are authorized to exclude evidence that has been gathered in breach of fun­
damental principles of law, for instance in violation of fundamental human rights safe­
guards (such as evidence obtained from a person who has previously been subjected to
inhum an or degrading treatment), or by dubious or devious methods (for example, by
surreptitiously obtaining the piece of evidence at issue).23
Other rules are designed to expedite trial proceedings and avoid waste of time. For
instance, international courts may take judicial notice of facts of common knowledge
or of public documents (such as UN records, records of other proceedings of the same
court, etc.). That means that neither party is required to provide evidence that such
facts occurred or that the documents are authentic (for a case where the court took
judicial notice of UN reports, see for instance Akayesu, §157). Furthermore, instead
of calling expert witnesses, their statements may be filed with the court, so that, if the
other party does not object to the statement and does not wish to cross-examine the
expert witness, the statement is admitted into evidence without calling the witness to
testify in person (see Rule 92bis of the ICTY).24 In addition, affidavits (that is, formal
written statements signed by a witness in front of a public official or in accordance
with another procedure provided for in national legislation) may be admitted into
evidence only, however, (i) to corroborate the testimony of a witness who appears in
court; (ii) if the affidavit is filed prior to the testimony of that witness; and (iii) if the
opposing party does not object to the filing of such affidavit.25
There are also rules on evidence relating to cases of sexual assault. In such cases, in
light of current practices, rules of evidence tend to protect the victim. Consequently:
(i) no corroboration of the victim’s testimony is required; (ii) consent of the victim is
not allowed as a defence if the victim was subjected to or threatened with or had reason
to fear violence, duress, detention, or psychological oppression, or reasonably believed
that if she or he did not submit, another person might be so subjected, threatened, or
put in feaf; (iii) the prior sexual conduct of the victim may not be admitted into evi­
dence (see Rule 96 of the ICTY).

22 See for instance Jelisic (§108), Kunarac (Decision on motion for acquittal) (§3); Kvocka (Decision on
defence motionsfor acquittal) (§12); Delalic and others (AJ) (§434); Jelisic (AJ) (§§34-7).
For a case where the AC found that the standard had not been correctly applied, see ICTR, AC, Akayesu,
§§171-2.
23 As stated in Rule 95 of the ICTY RPE, ‘No evidence shall be admissible if obtained by methods which
cast substantial doubt on its reliability or if its admission is antithetical to, and would seriously damage, the
integrity of the proceedings.’ See also Article 69(7) of the ICC Statute.
24 See on this Rule the TC’s Decision in Milosevic on Prosecutor’s request to have written statements
admitted, §§4-30, the decision in Milosevic on Prosecution application to admit evidence pursuant to Rule
92bis without cross-examination, at 2. See also the TC’s decision in Galic on the Prosecution request for
admission of Rule 92bis statements, at 4-19, as well as on Admission into evidence of a written statement by
a deceased witness, at 2-6.
25 On the legal value of affidavits and the need to consider them carefully see among other cases, Josef
Kramer and others (the Belsen trial), at 636.
 4 i6 IN T E R N A T I O N A L C R I M I N A L LAW

Rules also require that communications between lawyer and client be treated as
privileged’. Consequently, they require that such communications be not subject to
disclosure at trial, with some exceptions (if the client consents to such disclosure or
has voluntarily disclosed the content of such communications to a third party, and
such party then gives evidence of that disclosure); see Rule 97 of the ICTY.26
Finally, special rules deal with the delicate question of evidence affecting national secur­
ity o f states. Plainly, in international trials, particularly when crimes linked to armed con­
flict are at stake, important evidence may be in the possession of military officers or other
state agents who rely upon sources affecting national security. Courts have therefore to
strike a balance between the need to respect the legitimate security concerns of states and
the demands of justice. This in particular applies to states when the documents raising
national security concerns are in their custody: as the ICTY Prosecutor rightly argued in
her Brief in Blaskic (Subpoena), to grant a state a blanket right to withhold, for security
purposes, documents necessary for trial might jeopardize the very function of an inter­
national criminal tribunal and ‘defeat its essential object and purpose’ (§§70—3).27
A related problem may arise when the source of information is confidential and a wit­
ness provides documents or information only on condition that the source not be dis­
closed. This is a frequent occurrence in the case of international criminal proceedings,
for intelligence organizations may hold documents and other information of great rele­
vance to a trial, but may not be prepared to go public’. In these cases, the possibility that
the person could be compelled by the court to disclose his source may prompt him to
refuse to testify. The ICTY RPE takes this possibility into consideration in Rule 70.28
Regard is also taken of some categories of potential witnesses who could not testify
without breaching their official duties of confidentiality (this in particular applies to
staff members of the International Committee of the Red Cross). These staff members,

26 On this matter see the decision of 27 November 1996 in Tadic (Decision on prosecution motion for
production of defence witness statements), at 2, and Separate Opinion of Judge Stephen, at 3-7.
27 In Blaskic (Judgment on the request of Croatia) the Appeals Chamber suggested some general criteria
for the situation where the documents are in a state’s custody. In its view: (i) the Court must establish
whether the state is acting in good faith; (ii) (Judge Karibi-Whyte dissenting) the state at issue may be invited
to submit the relevant documents to the scrutiny of one judge designated by the TC; this measure ‘should
increase the confidence of the state that its national security secrets will not accidentally become public’;
(iii) if the documents need to be translated into one of the working languages of the Tribunal, such translation
may be carried out by the state itself; (iv) the documents will then be scrutinized by the judges in camera,
in ex parte proceedings, and no transcript is made of the hearing; (v) the documents considered not rele­
vant will be returned to the state, whereas those that are material to the case may be redacted by the state
concerned. The Chamber added that in exceptional cases a state may be allowed, subject to some stringent
conditions, to withhold documents of great relevance to national security while at the same time of scant
relevance to the trial proceedings (§68).
28 This Rule provides that: (i) the TC may not order the party to produce additional evidence received
from the entity or person providing the initial information; (ii) it may not summon (or compel to appear in
court) the person or a representative of the entity for the purpose of obtaining that additional evidence; in
addition, (iii) if the party concerned calls a witness to introduce the information at issue, the TC may not
compel him to answer questions relating to the information or its origin, if the witness declines to answer
on grounds of confidentiality. It should be added that in any case (iv) the Rule in no way detracts from the
power of the TC to exclude evidence ‘if its probative value is substantially outweighed by the need to ensure
a fair trial’.
 ST A G E S O F I N T E R N A T I O N A L P R O C E E D I N G S I N O U T L I N E 417

if called by either party to testify, may decline to do so, and the court will not compel
them to give evidence. This right for officials of the ICRC was envisaged in Simic.29

(B) ICC
Article 69 of the ICC Statute substantially takes up and restates the practice of inter­
national criminal tribunals.
The Statute contains a provision (Article 72) which regulates the delicate issue of pro­
tecting national security information in detail. It emphasizes the need for the parties con­
cerned to take all reasonable steps [...] to resolve the matter through cooperative means’, so
as to achieve solutions acceptable to both the parties and the Court. If solutions cannot be
agreed and the Court holds that the documents are relevant and necessary for establish­
ing the guilt or innocence of the accused, provision is made for such measures as hearings
in camera and ex parte, the drawing of inferences, as well as orders for disclosure.
This right for officials of the ICRC to refuse to testify is formally set out in Rule
73(4-6) of the ICC RPE.

19.8 CONTROL OF PROCEEDINGS

(A) ICTY and other tribunals

International courts, like any court, have an inherent power to control their pro­
ceedings. This, among other things, entails that TCs are authorized to decide when
an exception should be made to the basic principle whereby proceedings are public.
Whenever the need to protect victims or witnesses, public order or morality, security,
or the interests of justice so require, a TC may, either on its own initiative or at the
request of either party, order that a hearing or part thereof be held in closed session
(see e.g. Rule 79 of the ICTY). It may also order that, for the sake of protecting wit­
nesses, witnesses give testimony though image- or voice-altering devices or closed-
circuit television (see, for instance, Rule 75(B) of the ICTY).
In addition, the TC may order that a person be excluded from the courtroom when­
ever this proves necessary to ensure a fair trial to the accused or to m aintain the dig­
nity and decorum of the proceedings; it can also order the removal of the accused if he
engages in disruptive conduct (see for instance Rule 80 of the ICTY).
The Court may also initiate contempt proceedings against a witness when he or she
contumaciously refuses or fails to answer a question’ (Rule 77(A) of the ICTY).

(B) ICC
In the ICC system the powers of the Court to control proceedings are regulated by
Article 64 in a manner not dissimilar from that envisaged in other international

29 Simic (Decision on the prosecution motion under Rule 73 for a ruling concerning the testimony of a
witness), 27 July 1999, §§34-80.
4 i8 I N T E R N A T I O N A L C R I M I N A L LAW

crim inal tribunals. Detailed provisions are contained in Article 70 on the issue of
offences against the adm inistration of justice and in the relevant Rules of the ICC RPE
implementing or spelling out the Statute’s provisions.

19.9 DELIBERATIONS

W hen both parties have completed their presentation of the case, the C ourt declares
the hearings closed and retires to deliberate in private.
A major issue is that of the standard of proof required for a court to determine
whether the accused is guilty. It is common knowledge that in common law systems the
standard of proof varies depending on whether the proceedings are criminal or civil,
whereas in countries of Romano-Germanic tradition the standard of proof in crim ­
inal cases is rather loose, but is formally pre-established by law for civil litigation.
In common law countries the standard normally required in criminal proceedings
is that facts must be proved ‘beyond a reasonable doubt’. This means that the facts
must be proved in such a way that a court satisfies itself without hesitation that the
accused is guilty; in other words, the court must find that the accused is guilty without
entertaining a doubt that would cause any reasonable and prudent person to hesitate
before reaching a definite conclusion. As the European Court of Hum an Rights put
it in Barbera, Messegue and Jabardo, ‘any doubt should benefit the accused’ (§77). In
1947 Lord Denning set out a clear definition of the standard of proof under discussion
in Miller v. Minister o f Pensions. He pointed out that
the degree of cogency as is required in a criminal case before an accused person is found
guilty [... ] is well settled. It need not reach certainty, but it must carry a high degree of prob­
ability. Proof beyond reasonable doubt does not mean proof beyond the shadow of a doubt.
The law would fail to protect the community if it admitted fanciful possibilities to deflect
the course of justice. If the evidence is so strong against a man as to leave only a remote pos­
sibility in his favour which can be dismissed with the sentence ‘of course it is possible, but
not in the least probable’, the case is proved beyond reasonable doubt, but nothing short of
that will suffice (3).30

30 In Heinz Heck and others, (the Peleus trial) the Judge Advocate in his summing up explained the
‘beyond a reasonable doubt’ test as follows: ‘A reasonable doubt does not mean some fanciful or imaginary
doubt such as a weak mind may grasp if it is struggling to avoid an honest conclusion on evidence that is
plain. It means the kind of doubt that might affect you in the conduct of some important affair of your own.
If, having considered this case as I know you will, most anxiously, you are left with a reasonable doubt such
as I have described, then it is your duty to give to any accused person as to whom you entertain such a doubt
the benefit of it and to acquit him. If, on the other hand, the evidence that you have heard drives your minds
to the conclusion that he is guilty, it is equally your duty to say so without regard to the consequences of this
finding’ (at 123). In WolfgangZeuss and others (the Natzweiler trial) the Judge Advocate, in his summing up,
stated that ‘reasonable doubt means just such an inquiry as you would make into any affairs of your own in
your everyday life. Probably there are few things in the world about which we can be utterly and completely
certain. In most things there is some doubt—some little doubt—in one’s mind, but you are not obliged
to take into account any sensitive doubt—anything which would not affect your judgment in you own
 ST A G E S O F I N T E R N A T I O N A L P R O C E E D I N G S I N O U T L I N E 419

In contrast, less stringent requirements are provided for in non-crim inal proceed­
ings; that is, civil actions, for instance tort cases: issues must be proved ‘by a prepon­
derance of the evidence’ (that is, evidence showing, as a whole, that the fact sought
to be proved is more probable than not, evidence that is more convincing than that
offered in opposition to it), or, under a test requiring a higher degree of proof, ‘by clear
and convincing evidence’ (that is, by evidence that is clear and explicit and is suffi­
cient to make out a prima facie case). The upholding of different standards of proof for
criminal and civil proceedings accounts for the possibility that trial proceedings may
be term inated with the acquittal of the defendant and be followed by proceedings for
damages in tort law (as for instance in the famous O. /. Simpson case). It should, how­
ever, be noted that in some instances a rule different from that prevailing is applied
even in criminal proceedings: for instance, under Rule 850(a)(b) of the US Uniform
Code of Military Justice, ‘The accused has the burden of proving the defense of lack of
mental responsibility by clear and convincing evidence.’31
Also in many countries of continental Europe and most other civil law countries the
law draws a distinction between private law proceedings and criminal proceedings.
For the former category the law defines both the classes of admissible evidence and the
requirements for their admissibility, and their respective probative value. In contrast,
in criminal proceedings, the principle of the free evaluation of evidence obtains: the
court freely evaluates the evidence and freely decides what weight to give to each piece
of evidence. W hat matters is that the judge reaches the ‘conviction’ that the accused is
guilty or innocent.32

affairs. What you have to do is to be satisfied beyond reasonable doubt. That means you must not be left, hav­
ing decided that a person is guilty, feeling that perhaps you were wrong about that’ (at 199).
In its decision of 6 December 1988 in Barbera, Messegue and Jabardo v. Spain, the ECHR held that ‘all rea­
sonable doubts must be silenced’ (Series A146, §77). See also Safferling, at 259-60.
31 The standard of proof, even under Rule 850, is still ‘beyond a reasonable doubt’, e.g. if the accused is
charged with murder under the US Uniform Code, the prosecution will have to prove that he committed
the murder beyond a reasonable doubt. The question is—what happens if the accused then turns around
and says, yes I killed intentionally, but I was insane or otherwise deranged at the time? It would be very
hard for the prosecution to prove 'beyond a reasonable doubt’ that the person is not insane. So what the law
does instead is to put a burden on the defence to prove that he was insane. But it would equally be too harsh
on a defendant to require him to prove ‘beyond a reasonable doubt' that he was insane. Insanity is a tricky
question and no certainties exist. Therefore the law imposes on the defendant this intermediate standard
of ‘clear and convincing evidence’. This is known as a reversible burden of proof and is often imposed for
‘special defences’.
Interestingly, the ICC Statute forbids any reversing of burdens of proof. See Article 67(l)(i) of the Rome
Statute.
32 For instance, in French law Articles 1315tf. and 1341-8 of the Civil Code set out the modes of evidence
and the probative force of each class of admissible evidence in civil proceedings, whereas Articles 353, 427,
and 536 of the Code of Criminal Procedure lay down the standards of the ‘intime conviction of the judges
and provide that the court need not explain why they have attached value to one piece of evidence rather than
to another. Article 192(1) of the Italian Code of Criminal Procedure is stricter: ‘The judge appraises evidence
and gives account in the judgment’s legal grounds of the conclusions reached and the criteria adopted.’
Arguably the provisions in French law on standards of proof in criminal trials do not force the jury or the
court to convict a person whenever a certain amount of evidence is available (hence the word ‘intime’); on
the other hand, as a matter of principle, they may, in reaching a decision of guilty, use all evidence available,
 420 I N T E R N A T I O N A L C R I M I N A L LAW

It would seem that the two standards of proof required in common law and civil law
countries, respectively (the ‘beyond reasonable doubt’ test and the test of the ‘intime
conviction; that is, the inner conviction of the judge) are not identical, the latter being
more loose and, it would seem, broader (but other commentators have advanced a
contrary view33).
The judgments delivered by international criminal courts and tribunals must always
provide a statement of the facts as found by the court and the legal reasons for the
court’s findings. Judges who do not concur with the majority may append their sep­
arate or dissenting opinions to the judgment. On this score, criminal courts uphold
the system prevailing in both common law systems and international civil’ (that is,
interstate) courts, such as the International Court of Justice.

19.10 SENTENCING

(A) ICTY and other tribunals

Whenever sentencing does not constitute a procedure per se as a result of the accused
entering a guilty plea, the sentence is part of the verdict.
Penalties are not provided for in an accepted tariff o f penalties. International provi­
sions only rule out the death sentence. Otherwise they provide very general indications
to courts. For instance, the ICTY Statute stipulates that ‘in determining the terms of
imprisonment, the TCs shall have recourse to the general practice regarding prison
sentences in the courts of the former Yugoslavia’ (Article 24(1)). The ICTR Statute pro­
vides similarly in Article 23(1), referring of course to the courts of Rwanda. By the same
token. Article 19 (1) of the SCSL refers to the practice of the ICTR and the national courts
of Sierra Leone; Rule 10.1 of the Regulation no. 2000/15, s. 10 of the ETSP provides
that the Special Panels ‘shall have recourse to the general practice regarding prison
sentences in the courts in East Timor’ (it is, however, specified that imprisonment ‘may
not exceed a maximum of 25 years’); Article 24(1) of the Statute of the STL enjoins the
Tribunal to take the sentencing practice of Lebanese courts into account.
In short, it is for each TC to establish the prison sentence it considers appropriate, in
view of the gravity of the crime.
In Delalic and others (AJ) (§806) and in Aleksovski (AJ) (§185) the ICTY Appeals
Chamber held that retribution and deterrence ought to constitute the main guiding
principles in sentencing for international crimes. TCs have normally taken the same

unless otherwise provided by law, and attach to it the value it deserves in their eyes (this is called in German
legal literature ‘freie Beweiswiirdiging’).
In the Netherlands and Germany the courts must be convinced beyond reasonable doubt.
33 For, instance, Pradel (at 474). According to J. R. Spencer, ‘Evidence’ cited above at note 18, ‘It is ques­
tionable whether the actual level of certainty the two tests require is really different’ (at 601). Indeed, the
question arises of whether one can really have an ‘(intime) conviction if one is not convinced beyond reason­
able doubt that the accused is guilty.
 ST A G E S O F I N T E R N A T I O N A L P R O C E E D I N G S I N O U T L I N E 421

approach.34 Thus, for example, in Todorovic (SJ) an ICTY TC took the view that those
two notions or, as it termed them, ‘purposive considerations’, merely formed the back­
drop against which the sentence of an individual accused must be determined (§28).35
In other cases, ICTY TCs have also considered reprobation and stigmatization as
among the main purposes of sentencing.36 Furthermore, some cases TCs have men­
tioned the purpose of rehabilitating the accused, particularly when he was of young age.37
Also reconciliation is sometimes indicated as one of the objectives pursued in punishing
the perpetrators of serious crimes (see for instance SPSC, Consta Nunes, §85; Cloe, §22).
Some courts have tried to set out some general considerations warranting their sen­
tencing policy.38

34 See, for instance, ICTR, Akayesu, SJ, §19; Kayishema and Ruzindana, SJ, §2; Kambanda, SJ, §28;
ICTY, Furundzija (§288); SPSC, Joni Marques and others (Los Palos case), at §310; Manuel Gonsalves
Leto Bere, at 12.
35 The Chamber went on to say that the principle of retribution ‘must be understood as reflecting a fair
and balanced approach to the exaction of punishment for wrongdoing. This means that the penalty imposed
must be proportionate to the wrongdoing: in other words that the punishment be made to fit the crime’ (§29).
As for deterrence, it held that it meant that ‘the penalties imposed by the International Tribunal must, in
general, have sufficient deterrent value to ensure that those who would consider committing similar crimes
will be dissuaded from doing so’. The Chamber went on to say that, ‘Accordingly, while the Chamber recog­
nises the importance of deterrence as a general consideration in sentencing, it will not treat deterrence as a
distinct factor in determining sentence in this case’ (§30).
36 See, for instance, Erdemovic (SJ) (§65); Furundzija (§289); and Blaskic. (§§763-4).
In Erdemovic (SJ) the ICTY TC held that: "The International Tribunal sees public reprobation and stigma­
tisation by the international community, which would thereby express its indignation over heinous crimes
and denounce the perpetrators, as one of the essential functions for a prison sentence for a crime against
humanity’ (§65).
37 For instance, in Furundzija (§291) an ICTY TC stated that none of the various purposes of punishment
such as retribution, deterrence, and stigmatization was to detract ‘from the TC’s support for rehabilita­
tive programmes in which the accused may participate while serving his sentence; the Trial Chamber is
especially mindful of the age of the accused in this case’.
In the same case the TC also stated that it was to be guided in its determination of the sentence by the
principle proclaimed as early as 1764 by Cesare Beccaria (An Essay on Crimes and Punishment, 1775,
reprinted (Brookline Village, Ma: Brandon Press Inc., 1983)), namely that ‘punishment should not be harsh,
but must be inevitable’. It went on to state that 'It is the infallibility of punishment, rather than the severity
of the sanction, which is the tool for retribution, stigmatisation and deterrence. This is particularly the case
for the international tribunal; penalties are made more onerous by its international stature, moral authority
and impact upon world public opinion, and this punitive effect must be borne in mind when assessing the
suitable length of sentence’ (§290). This proposition, while it seems correct in that it stresses the particular
stigma attaching to punishment by an international tribunal, as well as the need for the penalties not to
be excessively harsh, could appear questionable in another respect: it does not seem that inevitability of
punishment is a major feature of international courts; these courts must of necessity concentrate on major
instances of gross violations of international criminal law and therefore cannot but be selective; it follows
that in many instances perpetrators will not be punished, unless they are brought before national courts.
See also Delalic and others (AJ, §806); Obrenovic (SJ, §53); ICTR, Kayishema and Ruzindana, SJ, §26.
38 For instance, in Nadler and others a British Court of Appeal acting under Control Council Law
no. 10 stated that ‘Upon the conviction of any person of a crime against humanity under Law 10, a capital
sentence is, in the opinion of this Court, the appropriate sentence where such person has unlawfully and
maliciously killed another or where the inhumane conduct of such person has materially contributed to the
death of another. While in cases that fall within neither of these two classes a sentence of death will usually
be excessive, there may, nevertheless, be other cases where the conduct of the convicted person is so grossly
or persistently inhumane, on such a scale or so serious in its consequences that a capital sentence is proper
although it is not proved that such conduct has either caused or contributed to a death’ (at 134-6).
422 IN T E R N A T IO N A L C R IM IN A L LAW

It should be added that suspended sentences have been held to be admissible (see,
for instance, Bulatovic, ICTY, SJ, at §19).
In addition to imprisonment, some Statutes provide for the imposition of fines or
for forfeiture of proceeds, property, and assets derived directly from the crime (see,
for instance, Article 19(3) of the SCSL Statute, and Rule 10(1) of the ETSP Regulation
no. 2000/15, s. 10).

(B) ICC
Articles 77 and 78 of the ICC Statute, although less terse than the provisions of other
international criminal tribunals, do not provide the ICC TCs with any definite guide­
line concerning the determination of sentence.
However, this Statute makes much headway in the area of penalties. For, in add­
ition to providing for imprisonment, it also stipulates, in Article 77(2), that besides
imprisonment the Court may order ‘a fine under the criteria provided for in the Rules
of Procedure and Evidence’ and ‘a forfeiture of proceeds, property and assets derived
directly or indirectly from the crime, without prejudice to the rights of bona fide third
parties’. The Court may order that money and other property collected through fines
or forfeiture be transferred to a trust fund established by decision of the Assembly of
States Parties for the benefit of the victims or their families (Article 79).

19.11 REPARATION OR COMPENSATION

TO VICTIMS
(A) ICTY and other tribunals
The ICTY and ICTR Statutes only provide for the right of victims to restitution. Articles
24(3) of the ICTY Statute and 23(3) of the ICTR Statute stipulate that in addition to
imprisonment, a TC ‘may order the return of any property and proceeds acquired by
criminal conduct, including by means of duress, to their rightful owners’. Nonetheless,
the Rules of Procedure and Evidence make allowance for compensation, to be granted
by the competent national court.39

39 Rule 105 of the ICTY RPE regulates restitution in detail. It stipulates that after a judgment of convic­
tion containing a specific finding of unlawful taking of property, at the request of the Prosecutor or proprio
motu the TC may hold a special hearing on the question of restitution. If such property or its proceeds are in
the hands of third parties not otherwise connected with the crime, they will be summoned before the TC and
given the opportunity to justify their claim to the property or its proceeds. The TC, if it is able to determine
the rightful owner ‘on the balance of probabilities’, orders its restitution or the restitution of its proceeds.
If instead it is unable to determine ownership, it requests the competent national authorities to do so, and
orders thereafter the restitution of the property or of its proceeds.
Rule 106, on compensation to victims, cannot of course grant victims a right to compensation, absent
any provision on the matter in the Statute. Nonetheless it provides that the Registrar shall transmit to the
relevant national authorities the judgment finding the accused guilty of a crime that has caused injury to a
victim. It will be for the victim to claim compensation before the competent national court. For this purpose,
 ST A G E S O F I N T E R N A T I O N A L P R O C E E D I N G S I N O U T L I N E 423

(B) ICC
The ICC Statute is more favourable to victims. First, Article 75 provides for various
forms of reparations (restitution, compensation, and rehabilitation). Secondly, as
pointed out above, Article 79 stipulates that the Assembly of States Parties shall estab­
lish a trust fund for the benefit of the victims and their families. Rules 94-9 of the ICC
RPE regulate the matter in some detail.
It appears from Article 75 and the Rules just mentioned that the proceedings for
determining reparations may be initiated either by a victim, under Rule 94, or by the
TC on its own motion, pursuant to Article 75(1) and Rule 95. Victims and the con­
victed persons may take part in the proceedings and be heard by the court. The TC
may appoint experts to assist it in determining the damage, loss, or injury and sug­
gesting ‘the appropriate types and modalities of reparation. Interestingly, under Rule
97(2), ‘The Court shall invite, as appropriate, victims or their legal representatives, the
convicted person as well as interested persons and interested states to make obser­
vations on the reports of experts.’ The TC grants reparation by ordering an ‘award
against a convicted person’.

‘the judgment of the Tribunal shall be final and binding as to the criminal responsibility of the convicted
person for such injury’. (The final and binding nature of the Tribunal’s findings seems to be an aspect of the
ICTY’s primacy).
See also Article 25 of the Statute of the STL.
 20
APPEALS AND
ENFORCEMENT

20.1 GENERAL

The right of defendants to appeal against conviction or sentence is normally regarded

as a fundamental hum an right. Subject to some exceptions, it is basically predicated
on the notion of fair trial. At present this right is laid down in numerous international
treaties on hum an rights, as well as in the Statutes of international courts. In contrast,
before the blossoming of the hum an rights approach, the right was not considered
fundamental, as evidenced by the fact that the Statutes of the IMT and the Tokyo
Tribunal did not contemplate a right of appeal.
Alongside this right in some legal systems provision is also made for the power of
Prosecutors to appeal against acquittals. Clearly, here the rationale is no longer to
reaffirm a fundamental hum an right, but rather to ensure the proper administration
of justice, by enabling the Prosecutor to file an appeal to a higher court when he con­
siders that the acquittal of the accused amounts to a miscarriage of justice.
The notion and purpose of appellate proceedings vary, however, in national
systems (see supra, 17.2.10). Subject to a num ber of specifications and exceptions,
in civil law countries, that is countries of Rom ano-Germ anic legal tradition, these
proceedings am ount largely to a retrial by a court of appeal. Very often both law
and facts are brought before this court, for the appellant may claim that the trial
court has misapplied or m isunderstood the relevant law, or wrongly established the
facts. Hence the court of appeal hears the case anew, if need be by adm itting or call­
ing the same or new witnesses, and confirms, reverses, or quashes the trial court’s
judgm ent or sentence. The right of appeal inures to both the convicted person and
the Prosecutor, who may also appeal against acquittal or sentence (however, usu­
ally any increase in the sentence by the appeals court is only adm itted w ithin strict
limits).
In contrast, in most common law countries appellate proceedings do not lead to a
retrial, for the appellate court consists of professional judges who may not substitute
themselves for a jury, the only body entitled to make findings of fact. Hence appeals
courts, which do not have any jury, do not review facts, but decide on the basis of the
trial record. Only exceptionally do they receive evidence, provided it would have been
 APPEALS A ND ENFORCEM ENT 425

admissible at trial and was not adduced at that stage (for instance, because it was not
then available). The reason why courts of appeal only exceptionally hear evidence is
that the jury is the sole trier of fact and all the relevant evidence must be put before,
and evaluated by, the trial court. The court of appeal may dismiss the appeal, quash the
judgment, or request a retrial by a trial court.
In addition, appellate proceedings in common law jurisdictions normally exhibit two
features designed to reduce the number of appeals and thus shorten the total length of
proceedings. First, normally the Prosecutor may not appeal against acquittal (it is felt that
such appeal would compromise the acquitted defendant’s right to be tried by a jury). Nor
is he allowed to appeal against sentence. Secondly, subject to an exception to be mentioned
below, the accused may not appeal automatically against conviction or sentence, but only
if granted leave to appeal by a judge sitting on the appeals court. This holds true for the
vast majority of appeals, those made on grounds of mixed law and fact (for instance, on
the ground that the trial judge issued wrong instructions to the jury, misdirecting it in
his summing up about the elements of the offence). The purpose of requesting that the
defendant be granted leave to appeal is to avoid frivolous, vexatious, or unmeritorious
appeals: the single judge that pronounces upon the request for leave to appeal functions as
a sort of filter, granting leave only if he holds that there is an arguable point in the appeal.
However, no leave to appeal is required when the appellant challenges the conviction or
sentence on the ground of pure law (for example, on the ground that the indictment was
defective on its face; that the trial court lacked jurisdiction to try the offence because the
offence had been committed abroad; that the facts and evidence relied upon by the pros­
ecutor did not amount to the offence of which the appellant had been convicted; that a
defence submission of no case to answer had been wrongly rejected by the judge; or that
admissible evidence had been excluded). W hen appealing on the ground of pure law the
appellant may bring the case before the court of appeal as of right.
In international criminal proceedings neither the common law system nor the civil
law model have been upheld. Rather, a mixed system has been accepted, as we shall see
below.

(A) APPEALS

20.2 APPEALS AGAINST

INTERLOCUTORY DECISIONS

(A) ICTY and other tribunals

International criminal courts, like their national counterparts, may issue interlocu­
tory decisions, either on preliminary motions (for instance, those which challenge
the jurisdiction of the court, or allege defects in the form of the indictment, or seek
 426 IN T E R N A T IO N A L C R IM IN A L LAW

severance of counts or separate trials, or raise objections based on refusal of a request

for assignment of counsel; Rule 72(B) of the SCSL adds objections ‘based on abuse of
process’ to the usual list of objections); or on any other motion (for instance, motions
for provisional release).
In drafting the Rules of Procedure on appeals against interlocutory decisions,
judges of the ICTY and the ICTR went through different stages. In a first phase they
took a rather restrictive approach in providing for such interlocutory appeals; they
subsequently broadened the range of cases where such appeals could be lodged; finally,
faced with an increasing number of such appeals and fearing that they might unduly
delay trial proceedings, they adopted a restrictive attitude aimed at significantly limit­
ing the number of appeals likely to be filed.
At present, TCs’ decisions on prelim inary (or non-preliminary) motions may not be
the subject of interlocutory appeal except in two cases:

(i) if the motion challenged the court’s jurisdiction;

(ii) (a) within the ICTY system,
if certification is granted by the TC that the TC’s decision ‘involves an issue that
would significantly affect the fair and expeditious conduct of the proceedings or
the outcome of the trial and for which, in the opinion of the TC, an immediate
resolution by the AC may materially advance the proceedings’ (Rule 72(B) and
73 (B) ICTY);1 or
(b) under Rule 72 ICTR,
if a bench of three AC Judges decides that the appeal is such as to satisfy the
requirements for jurisdiction laid down in Rule 72(D) ICTR (a system similar
to that of the ICTR is provided for in Rule 72(D) SCSL).

Thus, only for interlocutory decisions on jurisdiction does the appeal lie as of right.
For some of these appeals, i f ‘certification’ or leave to appeal is granted, the AC may
decide to apply an expedited appeals procedure’, which, among other things, involves
that the appeal is determined entirely on the basis of: (i) the original records of the TC;
(ii) written submissions by each party, without a second exchange of briefs in reply;
and (iii) without any hearing (see Rule 116 bis ICTY and Rule 117 ICTR).

(B) ICC
Interlocutory appeals are regulated in a detailed m anner in Article 82. Either party
may appeal any of the following decisions: (i) decisions by a TC on jurisdiction or
admissibility; (ii) decisions by a TC granting or denying release of a person being
investigated or prosecuted; (iii) decisions by a P-TC, acting on its own initiative pursu­
ant to Article 56(3), to take testimony or a statement by a witness or to examine, collect

1 This ‘certification’ system may be criticized since a TC, fearing reversal by the AC, may simply decide
not to certify the appeal, in which case the avenue of appeal is completely cut off. Probably the reply would be
that ultimately the issue can be resolved by final Appeals, after the TC judgment has been rendered.
 APPEALS A ND ENFORCEM ENT 427

or test evidence which may not be available subsequently for the purpose of trial;
(iv) decisions by a P-TC or a TC involving an issue that would significantly affect the
fair and expeditious conduct of proceedings or the outcome of the trial.
Furthermore, the state concerned or the Prosecutor may, subject to leave by the
P-TC, file an appeal against a decision of the P-TC authorizing under Article 57(3) (d)
the Prosecutor to take ‘specific investigative steps within the territory of a State Party
without having secured the cooperation of that State’. The appeal shall be heard on an
expedited basis (Article 82(2)).

20.3 APPEALS AGAINST JUDGMENT

OR SENTENCE

(A) ICTY and other tribunals

The relevant provisions of the Statutes confer the right of appeal on both the defend­
ant and the Prosecutor, thus departing from the common law system. Both parties may
appeal against conviction or sentence, and the Prosecutor may appeal against acquittal.
Another departure lies in the range of grounds on which one may appeal, which is
much broader. These grounds are: (i) an error of law so serious as to invalidate the judg­
ment; (ii) an error of fact so serious as to entail a miscarriage of justice. The Statute of the
SCSL provides that appeals may also be lodged for ‘procedural errors’ (Article 20(1)(a)).
However, as in common law systems, international appellate proceedings do not
involve a re-trial. On many occasions the ICTY AC has emphasized that an appeal is
not an opportunity for the parties to re-argue their cases.2
As for errors of fact, the standard for appraising them was clearly enunciated by an
ICTR AC in Bagilishema (AJ, §13) and an ICTY AC in Brdanin (AJ, §13): ‘when con­
sidering an appeal by the Prosecution, as when considering an appeal by the accused,
the Appeals Chamber will only hold that an error of fact was committed when it deter­
mines that no reasonable trier of fact could have made the challenged finding’.3 In
Brdanin (AJ) the ICTY AC suggested an interesting way of expediting consideration

2 See Furundzija (AJ), §40; Kupreskic and others (AJ) §22; Kordic (AJ), §21; Kvocka and others (AJ), §14.
3 Both ACs had premised that 'The same standard of unreasonableness and the same deference to factual
findings of the Trial Chamber apply when the Prosecution appeals against an acquittal.’ In Bagilishema the
AC specified however that ‘For the error to be one that occasioned a miscarriage of justice, it must have been
“critical to the verdict reached”. Because the Prosecution bears the burden at trial of proving the guilt of the
accused beyond a reasonable doubt, the significance of an error of fact occasioning a miscarriage of justice is
somewhat different for a Prosecution appeal against acquittal than for a defence appeal against conviction.
An accused must show that the Trial Chamber’s factual errors create a reasonable doubt as to his guilt. The
Prosecution faces a more difficult task. It must show that, when account is taken of the errors of fact commit-
ted by the Trial Chamber, all reasonable doubt of the accused’s guilt has been eliminated.’ (§14).
The basic assumption of the ACs with regard to the their pronouncements on alleged errors of fact made
by TCs, is that in principle a TC is in a better position to appraise the evidence. As the ICTY AC put it in
Kupreskic (AJ, §32) and then repeated in Kunarac (AJ, §40) 'The Trial Chamber has the advantage of observ­
ing witnesses in person and so is better positioned than the Appeals Chamber to assess the reliability and
credibility of the evidence. Accordingly, it is primarily for the Trial Chamber to determine whether a witness
 428 I N T E R N A T I O N A L C R I M I N A L LAW

of allegations of errors of fact: it held that eight categories of alleged errors could be
summarily dismissed.4
A third departure from the common law system is that the possibility to hear fresh
evidence is much wider. Under Rules 115 of the ICTY and ICTR RPE a party to the
appellate proceedings may lodge a motion asking that additional evidence, which was
not available to this party at trial, be presented. The AC may authorize the presentation
of such evidence ‘if it considers that the interests of justice so require’. Plainly, if such
authorization is granted, the AC will have to hear the new evidence and may therefore
have to reconsider some of the facts. Of course, the same rules governing the presenta­
tion of evidence before Trial Chambers will also apply before the AC.5
The AC may dismiss the appeal, or acquit the appellant, or order that the accused be
retried, or change the sentence.

(B) ICC
The ICC Statute provides that appeals may be lodged both by the convicted person
and by the prosecutor. They may appeal against either the judgment or the sentence,
or both (Articles 81 and 83).
Under Article 81 of the ICC Statute, an appeal may be lodged by the Prosecutor on
the ground of an error of fact, an error of law, or a ‘procedural error’. Instead, the con­
victed person (or the Prosecutor on that person’s behalf) may appeal not only on one
of those grounds but also ‘on any other ground that affects the fairness or reliability of
the proceedings or decision’. Furthermore, either party may appeal against a sentence
‘on the ground of disproportion between the crime and the sentence’.

is credible and to decide which witness’testimony to prefer, without necessarily articulating every step of the
reasoning in reaching a decision on these points.’
4 The AC held that it could summarily dismiss alleged errors which (1) challenge factual findings that do
not constitute the basis of a conviction; (2) misrepresent the TC’s factual findings or ignore other relevant
factual findings; (3) constitute mere assertions that the TC failed to consider relevant evidence; (4) constitute
mere assertions that the TC could not have reasonably inferred a particular conclusion from circumstantial
evidence; (5) are clearly irrelevant or lend support to the challenged finding; (6) challenge the TC’s reliance
or lack of reliance on one piece of evidence without explaining why the finding should not stand on the basis
of the remaining evidence; (7) are contrary to common sense; (8) relate to factual findings whose relevance
is unclear (§§19-31).
5 In Tadic (Appeal, decision on admissibility o f additional evidence) (§§27-74) the AC held that the
unavailability of the evidence at trial must not result from lack of due diligence on the part of relevant
defence counsel; in addition, the interests of justice required admission of evidence only if the evidence: (i)
was relevant to a material issue; (ii) was credible; and (iii) was such that it would probably show that the con­
viction was unsafe. See also Delalic and others (Appeal, order on motion o f Landzo), at 2-3; Jelisic (Appeal,
decision on additional evidence) at 3; as well as Kupreskic and others (AJ) (§§48-76).
On many occasions the ICTY AC has rejected motions for the admission of additional evidence. This, for
instance, happened in Tadic (Appeal, decision on the admission of additional evidence), as well as in Jelisic
(Appeal) (§§20-1). In other cases additional evidence has been admitted; see for instance Delalic and others
(Order on motionfor the extension of the time-limit an admission o f new evidence, of 31 May 2000, at 8-9, and
of 14 February 2000, at 2-3), in Kupreskic and others (Appeal, decision on the motions ofDrago Josipovic and
others), at 7). See also Akayesu (Appeal, decision o f 22 August 2000) (at 5-6). In Kupreskic and others (AJ) the
ICTY AC, after considering and weighing additional evidence (§§263-302) concluded that the findings of
the Trial Chamber had resulted in a miscarriage of justice (§§303-4).
 APPEALS A N D EN FORCEM EN T 429

(B) REVIEW

20.4 REVIEW OF JUDGMENT OR SENTENCE

(A) ICTY and other tribunals

Under Article 26 of the ICTY Statute and Article 25 of the ICTR Statute, whenever a
new fact is discovered which (i) was not known to the party concerned at the time of
trial or appellate proceedings; and (ii) could have been a decisive factor in reaching the
decision, the convicted person or the Prosecutor may apply for review of the judgment.
Rule 119 ICTY and Rule 120 ICTR add a further condition: the new fact ‘could not have
been discovered through the exercise of due diligence’.6 As the ICTR AC put it in 2000
in Barayagwiza (Decision on Prosecutor’s Requestfor Review or Reconsideration):
it is clear from the Statute and the Rules that, in order for a Chamber to carry out a review,
it must be satisfied that four criteria have been met. There must be a new fact; this new fact
must not have been known by the moving party at the time of the original proceedings; the
lack of discovery of the new fact must not have been through the lack of due diligence on the
part of the moving party; and it must be shown that the new fact could have been a decisive
factor in reaching the original decision (§41).7

The rationale behind this review procedure is evident: although a judgment or sen­
tence may be endowed with the legal force of res judicata (that is, the force of a binding
and final judicial decision), it would be contrary to elementary principles of justice
not to revise it whenever a new fact emerges that was unknown at the time of trial and
which, if known, would have led to a totally different decision. In the case of review
proceedings, what must be new is a fact, not evidence of a fact known at the time of
trial. As the ICTY AC rightly held in Tadic (Appeal on admission o f additional evi­
dence), ‘The mere subsequent discovery of evidence of a fact which was known at trial
is not itself a new fact within the meaning of Rule 119 of the Rules’ (§32).
Under the ICTY and ICTR systems the convicted person may at any time file an
application for review, while the Prosecutor has a time limit of ‘one year after the final
judgment has been pronounced’. The motion for review does not automatically lead to
a new trial: it is necessary for the Chamber which delivered the judgment to conclude
that the new fact, if proved, could have been ‘a decisive factor in reaching the decision’.
This examination is ‘preliminary’ in nature. If the Chamber’s conclusion is affirma­
tive, then the relevant Chamber commences a new trial, and its judgment may then be
appealed (Rules 120-1 of ICTY and 121—2 of ICTR).

6 On this issues ee Barayagwiza (Decision on Prosecutor’s Requestfor Review or Reconsideration), §65 and
Tadic (Decision on Motion for Review), §27.
7 This holding was taken up by the ICTY AC in Delic (Decision on Motion for Review), at 7; Jelisic (Decision
on Motion for Review), at 3; Tadic (Decision on Motion for Review), §20; Josipovic (Decision on Motion for
Review), at 2-3.
430 IN T E R N A T IO N A L C R I M I N A L LAW

(B) ICC
In Article 84 the ICC system broadens the category of persons entitled to apply for review
(termed by the Statute ‘revision’). It grants the right to apply not only to the convicted
person and the Prosecutor (who must only act on behalf of the convicted person, hence
may not seek review against him, or contra reuni) but also, after the death of this person,
to spouses, children, parents, or ‘one person alive at the time of the accused’s death who
has been given express written instructions from the accused to bring such a claim’.
In addition, the Statute broadens the classes of requirements necessary for applying
for review. These conditions include not only (i) the discovery of a decisive fact; but also
(ii) the discovery that decisive evidence was false, forged, or falsified; or (iii) the fact that
one or more judges sitting on trial committed an act of serious misconduct or a serious
breach of duty justifying the removal of that or those judges from office pursuant to Article
46 of the ICC Statute. Under the same Statute, the motion for review is submitted to the
AC, which, if it considers it meritorious, may (i) reconvene the original Trial Chamber; or
(ii) constitute a new Trial Chamber; or (iii) retain jurisdiction over the matter.

20.5 REVIEW OF OTHER FINAL DECISIONS

In Barayagwisa (Appeal on request fo r review or reconsideration) the question arose of
whether a decision of the ICTR AC of 3 November 1999, which dismissed the indict­
ment against the appellant and terminated the proceedings, could be the object of
review before the AC at the request of the Prosecutor.8 The defence submitted that
Articles 24 (on appellate proceedings) and 25 (on review proceedings) of the ICTR
Statute, in granting the right of appeal or review to a convicted person (besides the
Prosecutor) presupposed that the right to request review was only available after con­
viction. The AC dismissed the argument. It pointed out that
If the Appellant were correct that there could be no review unless there has been a con­
viction, it would follow that there could be no appeal from acquittal for the same reason.
Appeals from acquittals have been allowed before the AC of the ICTY. The Appellant’s logic
is not therefore correct (§47).

Furthermore, the Chamber stated that it considered it im portant to note

that only a final judgment may be reviewed pursuant to Article 25 of the Statute and to
Rule 120. The parties submitted pleadings on the final or non-final nature of the Decision
[of 3 November 1999] in connection with the request for reconsideration. The Chamber
would point out that a final judgment in the sense of the above-mentioned articles is one

8 The AC in its Decision of 3 November 1999 noted that the Appellant had been detained for a total
period of 11 months before being notified of the charges against him. It stressed that the Prosecutor had
thus breached her duty of prosecutorial due diligence, and applied the abuse of process doctrine’ (§§73-7,
85-6, 92-6, 100-1). It should be emphasized that such doctrine is upheld in such common law countries as
the US, Canada and the UK, while in countries having a Romano-Germanic tradition it is unknown (except
perhaps for the Netherlands).
 APPEALS A N D EN FO R C EM EN T 431

which terminates the proceedings; only such a decision may be subject to review. Clearly,
the Decision of 3 November 1999 belongs to that category, since it dismissed the indictment
against the Appellant and terminated the proceedings (§49).

Hence, according to this case law, in the ICTY and ICTR systems a review motion
m aybe filed even against a decision that is not stricto sensu a judgment, provided such
decision puts an end to the proceedings.

(C) E N F O R C E M E N T OF S E N T E N C E S

20.6 PLACE OF IMPRISONMENT

International courts do not have any prison available in which to detain convicted per­
sons. Consequently they must of necessity turn to states to see whether they may hold
those persons in jail. Article 27 of the ICTY Statute provides that ‘imprisonment shall
be served in a state designated by the International Tribunal from a list of states which
have indicated to the Security Council their willingness to accept convicted persons’.
Article 29(1) of the Statute of the STL is almost identical. Articles 26 of the ICTR and
22 of the SCSL have a fairly similar tenor, providing that im prisonment shall be served
in Rwanda or Sierra Leone respectively, or in any other state that has concluded an
agreement on the matter with one of these countries.
Similarly, Article 103(l)(a) of the ICC Statute provides that ‘a sentence of im prison­
ment shall be served in a state designated by the Court from a list of states which have
indicated to the Court their willingness to accept sentenced persons’.
So far the ICTY and the ICTR have entered into agreements with a num ber of indi­
vidual states, which have agreed to hold persons convicted by the Tribunals in their
national prisons.9

20.7 CONDITIONS OF DETENTION

O f course, imprisonment of convicted persons must be in conformity with the general
laws and regulations applicable in the relevant state. However, conditions of detention

9 The UN has made agreements for the ICTY with Italy (6 February 1997), Finland (7 May 1997), Norway
(24 April 1998), Sweden (23 February 1999), Austria (23 July 1999), France (25 February 2000), Spain (28 March
2000), Germany (17 October 2000), Denmark (19 June 2002) and the UK (11 March 2004). The agreement with
Spain differs in many respects from the other agreements. Among other things, it provides that Spain will only
consider the enforcement of sentences pronounced by the ICTY where the duration of the sentence imposed
does not exceed the highest maximum sentence for any crime under Spanish law (currently 30 years). The UN
has entered into agreements for the ICTR with Benin, France, Italy, Mali, Swaziland, Sweden and Tanzania.
 432 I N T E R N A T I O N A L C R I M I N A L LAW

of those persons must also accord with international standards. This requirement,
although not explicitly laid down in the Statutes of the ICTY, the ICTR, and other
tribunals, is implicit in the whole system of international courts: these judicial bodies
are bound to respect international standards on human rights and in particular those
on the rights of the accused, victims and witnesses. It follows that they may hand over
convicted persons to states for their detention only on condition that such states them ­
selves abide by those standards. It is therefore quite natural for the ad hoc agreements
concluded by these tribunals with states willing to hold convicted persons expressly to
require full respect for those standards,10 and in addition for the tribunals to envisage
and make provision for, in agreement with states, international oversight of condi­
tions of detention (see infra, 20.9).
The ICC Statute makes the above requirements explicit. Article 106(2) provides that
‘the conditions of imprisonment [...] shall be consistent with widely accepted inter­
national treaty standards governing treatment of prisoners’.

20.8 REDUCTION OR COMMUTATION

OF SENTENCE AND PARDON

International provisions stipulate that the state where the convicted person serves his
sentence is not allowed to reduce or change the penalty, or release the person, before
expiry of the sentence pronounced by the international tribunal (see, for instance,
ArticlellO(l) and (2) of the ICC Statute). Only the international tribunal may decide
upon any change in the sentence.
However, conflicts may arise between the general legislation of the state enforcing
the penalty and international prescriptions. It may happen that in the state at issue
detainees are entitled to a reduction of sentence, or to early release, or to special treat­
ment (for instance, parole) after serving the sentence for a certain number of years, or
in case of good behaviour. If these conditions are not applied to persons convicted by
an international tribunal, this might be deemed to constitute discrimination against
international convicts.

(A) ICTY and other tribunals

This difficult issue has been settled in a flexible m anner in agreements concluded
between the Tribunals and states. For instance, in the first such agreement with the
ICTY, which also served as a model for subsequent agreements, that with Italy of

10 In the various Agreements concluded by International Tribunals with states for the enforcement of
sentences it is provided that conditions of detention shall be compatible with the Standard Minimum Rules
for the Treatment of Prisoners, the Body of Principles for the Protection of All Persons under Any Form of
Detention or Imprisonment and the Basic Principles of the Treatment of Prisoners’ (Article 3(5) of the vari­
ous Agreements).
 APPEALS A N D E N FO R C EM EN T 433

6 February 1997, it is provided that ‘if pursuant to the applicable national law of
the requested state, the convicted person is eligible for non-custodial measures or
working activities outside the prison, or is entitled to benefit from conditional release,
the Minister of Justice shall notify the President of the Tribunal’ (Article 3(3)). The
provision then stipulates that, if the President of the Tribunal, in consultation with the
judges, does not consider those national measures appropriate, the convicted person
shall be transferred to the international tribunal, presumably for the purpose of being
transferred to another state willing to have him serve the remainder of his sentence. A
similar provision (Article 8) covers the issue of pardon or commutation of sentence.
The question of pardon is particularly difficult. In most states only the Head of State
may grant pardon. States are extremely jealous of this prerogative accruing to their
supreme national organ. Article 28 of the ICTY Statute (similarly to the corresponding
Article 27 of the ICTR Statute as well as Article 23 of the SCSL’s Statute and Article 30
of the Statute of the STL) provides that
If, pursuant to the applicable law of the State in which the convicted person is imprisoned,
he or she is eligible for pardon or commutation of sentence, the State concerned shall notify
the International Tribunal accordingly. The President of the Tribunal, in consultation with
the judges, shall decide the matter on the basis of the interests of justice and the general
principles of law.
On the face of it, the matter is ‘decided’ by the President of the Tribunal in consult­
ation with judges. The power of pardon would thus seem ultimately to belong to the
international body, in contrast to the regulation of most national constitutions.
The judges of the ICTY skilfully smoothed out the problem in the RPE. Under
Rule 123, if the person is eligible for pardon or commutation of sentence under national
legislation, the state concerned shall notify the Tribunal, and then the Tribunal’s
President, in consultation with the judges ‘shall [...] determine whether pardon or
commutation is appropriate’ (Rule 124), on the basis of a set of criteria laid down in
Rule 125. Thus, the international body only decides on the appropriateness of pardon
(or commutation), and the final decision is left to the relevant national authority (Rules
124-6 ICTR are identical in content).

(B) ICC
The ICC Statute does not make any provision for the granting of pardon. Under Article 110
it is for the Court alone to take any decision on the reduction of sentences (whereas
Rule 211(2) makes provision for the eligibility, under national law, for a prison pro­
gramme or benefit entailing ‘some activity outside the prison facility’, and simply pro­
vides that the Court must be notified and shall exercise its supervisory activity). It is
therefore probable that a solution similar to that set out by the ICTY and ICTR in their
Rules will be opted for, the more so because Article 104 of the ICC Statute provides
that the Court ‘may, at any time, decide to transfer a sentenced person to a prison of
another state’.
 434 I N T E R N A T IO N A L C R I M I N A L LAW

20.9 SU PERVISION OF IM PR ISO N M EN T

(A) ICTY and other tribunals

The Statutes of the ICTY and the ICTR provide that imprisonment served in a state
designated by the Tribunal shall be ‘subject to the supervision of the International
Tribunal (Articles 27 and 26, respectively). The ICTY entered into an agreement with
the International Committee of the Red Cross, authorizing the Committee to make
inspections not only in the Detention Unit in The Hague (where accused are held pend­
ing trial or appeal) but also, subject to the consent of the relevant state, in the country
where the sentence is enforced. Indeed, almost all Agreements on the enforcement of
sentences provide for inspection by the ICRC.11

(B) ICC
Under Rule 211 of the ICC RPE the C ourt’s Presidency ‘may [...] request any infor­
mation, report or expert opinion from the State of enforcement or from any reliable
source’. The Presidency may also delegate a judge or a staff member of the court to
supervise the conditions of detention.112

11 For instance, the Agreement with Italy of 6 February 1997 stipulates in Article 6(1) that the ICRC
may carry out inspections ‘at any time and on a periodic basis’; ‘the frequency of visits [is] to be determined
by the ICRC’. The ICRC submits a ‘confidential report based on the findings of these inspections’ to the
Italian Minister of Justice and the President of the ICTY, who will consult each other on those findings. The
Tribunal’s President may then request the Italian Minister of Justice ‘to report to him any changes in the
conditions of detention suggested by the ICRC’.
The Agreement with Spain differs from the other Agreements on the enforcement of sentences in that it
provides for inspections of the conditions of detention and treatment of the convicted persons by a Parity
Commission instead of by the ICRC.
12 Under Rule 211(l)(c) the judge or the staff member ‘will be responsible, after notifying the State
of enforcement, for meeting the sentenced person and hearing his or her views, without the presence of
national authorities’.
 21
THE SPECIFICITY OF
INTERNATIONAL TRIALS

21.1 THE NEED FOR

INTERNATIONAL TRIALS

In appraising international trials, one should predicate such assessment on the notion
that these tribunals have been established only to fill in for national courts, which
tend to refrain from prosecuting and trying persons suspected of those so widespread,
large-scale and organized atrocities that are international crimes. W ithout the (rela­
tive) inertia or the reluctance of national courts, there would be no need for inter­
national criminal tribunals.
A few words on the fundamental reasons for the failure of municipal courts promptly
and effectively to react to international atrocities may perhaps be fitting.
The normal response to atrocities would be to bring the alleged perpetrators to just­
ice in the courts of the state where the crimes were perpetrated, or at least of the state
of which the alleged perpetrator is a national. Plainly, this response would not target
the state as such for blame, but rather the individuals (state officials or persons act­
ing in a private capacity) who allegedly perpetrated the atrocities (in contrast to the
approach taken when international enforcement agencies seek to impose respect for
international values upon the state where atrocities have been committed, in which
case the state itself is stigmatized and sanctioned).
However, in the case of international crimes, there may be a major obstacle to the
territoriality principle (see 16.1): these crimes are often committed by state officials
or with their complicity or acquiescence. For example, war crimes are committed by
servicemen, or torture is perpetrated by police officers, or genocide is carried out by
state officials or param ilitary groups or at any rate with the tacit approval of state
authorities. It follows that domestic judicial authorities may be, and in fact are, loath to
prosecute state agents or to institute criminal proceedings against private individuals,
which might eventually involve state organs.
These considerations also hold true with regard to another head of jurisdiction,
based on the nationality of the alleged offender (so-called active personality principle
(see 16.1)). Here again it often happens that municipal courts tend to turn a blind eye
to crimes committed by nationals, particularly when they are members of the military
 436 IN T E R N A T I O N A L C R I M I N A L LAW

and the crimes allegedly perpetrated are part of a pattern of conduct that national
authorities implicitly instigate, or countenance, or at least tolerate.
O ther reasons prompt national courts to be rather disinclined to institute pro­
ceedings for crimes that lack a territorial or national link with the state. Here the
motivation behind the courts’ attitude is the lack of meta-national interests; that is,
the absence of a real concern about repression of crimes perpetrated elsewhere. It is a
fact that until 1994, when the establishment of the ICTY gave a great impulse to the
prosecution and punishment of alleged war criminals, the criminal provisions of the
1949 Geneva Conventions had never been applied. Domestic courts are still loath to
search for, prosecute, and try foreigners who have committed crimes abroad against
other foreigners. For them, the short-term objectives of national concerns seem still
to prevail.
The lack of national prosecution of crimes committed abroad is also due in part to
the frequent failure of national parliaments to pass the necessary legislation imple­
menting treaties that provide for the prosecution of international crimes or, more
specifically, grant national courts universal jurisdiction over such crimes.1 In this
respect the implementation in the USA of the 1949 Geneva Conventions is indicative:
the relevant US Statute only provides for jurisdiction over grave breaches of those
Conventions where the perpetrator or the victim has US nationality; in this manner
the universality principle proclaimed in the Conventions has been deprived—within
the US legal system—of its enormous innovative scope.12

1 For instance, Egypt, while it has ratified many international treaties on international crimes, has then
refrained from enacting the necessary implementing legislation, particularly with regard to genocide and
the 1949 Geneva Conventions (but on 23 January 2000 a decree (no. 149) was passed by the Prime Minister
establishing a National Committee for International Humanitarian Law). In China, only the Geneva
Conventions of 1949 and the Convention on Torture of 1984 have been translated into national legisla­
tion. By contrast, no such legislation exists with regard to the crimes of genocide, crimes against humanity,
aggression, or terrorism. In Morocco, neither the four 1949 Geneva Conventions nor the 1948 Convention
on Genocide have ever been published in the Bulletin officiel (it is as a result of such publication that inter­
national treaties take effect in the national legal system). In Brazil, the 1949 Geneva Conventions were rati­
fied in 1957, but no implementing legislation has yet been passed (however, on 27 November 2003 a decree
was passed setting up a National Commission for the dissemination and implementation of international
humanitarian law). In Italy, no detailed legislation implementing the 1949 Geneva Conventions has ever
been passed. (A legislative ‘order’ imposing compliance with the Conventions was enacted, but no legisla­
tion has been approved rendering non-self-executing provisions of the Conventions susceptible to appli­
cation.) Furthermore, the 1984 Convention on Torture was duly ratified and implementing legislation was
passed, but no legislative provision has been enacted on the definition of torture (which is not provided for
as a distinct crime in the Italian Criminal Code). Similarly, such states as Morocco, Tunisia, Jordan, and
Kuwait have not enacted legislation defining torture, whereas Algeria has passed a definition that the UN
Committee against Torture held to be incomplete (see CAT, 8 November 1996, UN doc. A/52/44, §74).
2 After defining the crimes over which US courts have jurisdiction as a result of implementing the four
Geneva Conventions of 1949, that is war crimes including grave breaches of the Geneva Conventions as well
as violations of common Article 3, the US legislation (War Crimes, US Code, Title 18, Chapter 18, passed on
21 August 1996) sets forth the grounds on which US courts may assert jurisdiction over these crimes, as fol­
lows: ‘(a) Offense. Whoever, whether inside or outside the United States, commits a war crime, in any of the
circumstances described in subsection (b), shall be fined under this title or imprisoned for life or any term of
years, or both, and if death results of the victim, shall be subject to the penalty of death, (b) Circumstances. The
 T H E SPEC IFICITY OF IN T E R N A T IO N A L T R IA L S 437

In other instances national courts clearly show a marked reluctance domestically

to apply international treaties binding upon the state. A clear illustration of this trend
can be seen in the Hissene Habre case, brought before a court of Senegal and decided in
2001,3 as well as in a 1995 decision of the Amsterdam Court of Appeal in Pinochet.4
Furthermore, problems specific to ICL may arise before national courts when the
alleged perpetrator of a crime is a state official enjoying im munity from prosecution
under national legislation (for instance, the Head of State, the head or a senior member
of the government, or a member of parliament). Clearly, if this is the case, domestic
courts are barred from instituting criminal proceedings against the accused, because
the latter enjoys personal immunity (rationepersonae): see 14.5. It may also be that the
alleged perpetrator, whatever his official status, is covered by an amnesty law. In this
case, again, the national authorities of the state where the amnesty was granted may be
precluded from taking judicial action.
Another reason militating against national courts is that they may not be legally
equipped to deal with international crimes, or may not have the moral authority to pro­
nounce upon such crimes. W hen Eichmann was brought to trial before the Jerusalem
District Court—a national court indisputably made up of independent and impartial

circumstances referred to in subsection (a) are that the person committing such war crime or the victim of
such war crime is a member of the Armed Forces of the United States or a national of the United States.’
Thus, while the Geneva Conventions lay down the universality principle, this principle is replaced in the
USA, in blatant breach of the Conventions, by the traditional principles of active and passive personality.
3 On 20 March 2001 the Supreme Court of Senegal refused to apply Article 6 of the 1984 Convention on
Torture (which imposes on every state party on whose territory an alleged torturer is present a duty to insti­
tute criminal proceedings against him). This Convention had been ratified by Senegal on 16 June 1986 and
published in the Journal officiel on 9 August 1986, thus becoming part of Senegalese legislation pursuant to
Article 79 of the Constitution. When proceedings were initiated against the former Chadian dictator Habre,
then residing in Senegal, for torture allegedly sponsored when he led his country, the Supreme Court, to
which the case had been brought, held that Article 4 of the Convention (obliging every state party to ensure
that all acts of torture are offences under its criminal law’) had not been fully implemented in Senegal. The
Senegalese legal system had been made to comply with the Convention by providing in Article 295-1 of the
Criminal Code (passed by law no. 96 of 28 August 1996) that torture was a crime. Nevertheless, Article 669
of the Code of Criminal Procedure had not been changed. (Under this provision any foreigner who has com­
mitted a crime abroad against state security or counterfeited the national seal may be prosecuted in Senegal
if arrested in Senegal or extradited to that country.) Hence, according to the Court, no provision of criminal
procedure confers universal jurisdiction on Senegalese courts for the prosecution and judgment of persons
who happen to be on the territory of Senegal and are allegedly perpetrators or accomplices of acts coming
within the purview of the law of 28 August 1996 designed to implement Article 4 of the Convention, if such
acts have been performed by foreigners abroad’ (at 7). This reasoning appears to be specious, for arguably the
ground of jurisdiction provided for in Article 6 of the Convention (the principle offorum deprehensionis) did
not need to be translated into an express provision of national legislation to become operational. At the most
the court could have argued that it was Article 5(2) of the Convention that had not been implemented (under
this provision, ‘Each State Party shall likewise take such measures as may be necessary to establish its jurisdic­
tion over such offences in cases where the alleged offender is present in any territory under its jurisdiction and
it does not extradite him pursuant to Article 8 to any of the states mentioned in paragraph 1 of this Article’).
4 Although the Chilean leader was staying in Amsterdam, the Prosecutor refused to apply the 1984
Convention on Torture, ratified by the Netherlands, arguing, among other things, that Dutch courts did not
have jurisdiction. On appeal by the complainants, the Court held that it was evident that prosecution of
Pinochet by the Dutch Public Prosecutions Department would encounter so many legal and practical prob­
lems that the Public Prosecutor was perfectly within his rights to decide not to prosecute’ (at 365).
438 IN T E R N A T IO N A L C R IM IN A L LAW

judges and capable of dealing with the large-scale crimes with which the defendant
was charged—the German philosopher Karl Jaspers, a staunch anti-Nazi since 1933,
seriously objected. He stressed that
the crime against the Jews is at the same time a crime against mankind. Hence, judgment on
this crime can only be passed by a judicial body representing mankind [...] these crimes do
not concern only Jews, but all and everybody, for it is mankind itself that has been attacked
through the Jews[...] It seems to me [Jaspers went on to say] that the essence and whole
range of consequences of this matter are trivialised—however odd this may sound—if one
leaves the judgment of these monstrous crimes to the court of a single state, and mankind
sets its mind at rest with that.5

Faced with this and other similar national legal conditions, international courts
are obviously called upon to play the crucial role of replacing national courts. In other
words, in the field of ICL national courts may not play the role they normally fulfil, of
acting simultaneously as state judicial organs and guardians of the international legal
order—a role first highlighted by the great French international lawyer Georges Scelle,
who characterized this phenomenon as role-splitting’ {dedoublementfonctionnel); he
emphasized that since the world com m unity lacks a judiciary exercizing compulsory
jurisdiction, often it falls to domestic courts to establish with binding force whether in
a specific case the law (including international law) is breached; domestic courts thus
operate as a surrogate for international tribunals.6 In contrast, in the area covered by
ICL this phenomenon does not materialize, on the grounds set out above.

21.2 MERITS OF INTERNATIONAL

CRIMINAL JUSTICE

International tribunals present a num ber of advantages over domestic courts, particu­
larly those sitting in the territory of the state where atrocities have been committed.
First, international courts proper may be more impartial than domestic courts, for
they are made up of judges having no link with the territory or the state where the
crimes were perpetrated. W hen national courts conduct proceedings, national feelings,
political ideologies, widespread resentment among the population, or possible public
reaction to the verdict may seriously interfere with the task of judges. International
judges may more easily ignore the possible future reaction of the public or the media to
their judicial determinations. Even the so-called mixed or ‘internationalized’ courts
(such as the SCSL, the ETSP, the Cambodian Extraordinary Courts or the STL) may
avoid the pitfalls of national territorial tribunals, for the international component in

5 K. Jaspers, ‘Karl Jaspers zum Eichmann-Prozess- Ein Gespräch mit Francois Bondy’in 13 Der Monat (May
1961) at 16. English translation by A. Cassese (‘Who should have Tried Eichmann?’) in 4 JICJ (2006) 855-6.
6 For references to Seelies writings, see A. Cassese, ‘Remarks on Scelle’s Theory of “Role Splitting”
(dedoublementfonctionnel) in International Law’, in 1 EJIL (1990), 210-31.
 T H E SP EC IFICITY OF IN T E R N A T IO N A L T R IA L S 439

these courts ensures the needed impartiality. International judges may be in a better
position to be unbiased, or at any rate more even-handed, than national judges who
have been caught up in the milieu in which the crime in question was perpetrated. The
punishment by international tribunals of alleged authors of serious crimes normally
meets with less resistance than national punishment, as it injures national feelings to a
lesser degree. The independence and impartiality of international courts is also guar­
anteed by their composition and the way judges are selected.
True, international criminal tribunals can be faulted for not being vested with jur­
isdiction over the two classes of international crime that are now so widespread: ter­
rorism and aggression. This, however, is a failing the blame for which should be laid at
the door of the states that have set up those tribunals. It would be injudicious to assail
the tribunals themselves for being deprived of jurisdiction over those two classes of
crime. In addition, steps are being taken to fill this gap: in 2007 the STL was vested
with jurisdiction over crimes of terrorism (albeit as defined by Lebanese criminal law),
and efforts are being made within the ICC to reach a definition of aggression as an
international crime that would be acceptable to all and thus make Article 5(2) of the
ICC Statute operational in 2009, when a Review Conference is held to consider any
amendments to the ICC Statute.
Secondly, international judges, being selected on account of their competence in the
area of international hum anitarian and criminal law, are better suited to pass judg­
ments over crimes that markedly differ from ordinary’ criminal offences such as theft,
murder, assault, etc. More than domestic courts, they are able to adjudicate on large-
scale organized criminality (genocide, crimes against humanity, war crimes), as well
as on the responsibility of military and political leaders. Domestic courts are often not
well equipped to look into responsibility for ‘system criminality’.
Also, international courts, more easily than national judges, are able to try crimes
with ramifications in many countries. Often witnesses reside in different countries,
and other evidence needs to be collected, requiring the cooperation of several states.
In addition, special expertise is needed to handle the often tricky legal issues arising
from the various national legislations involved.
A nother m erit of these tribunals and courts is that they act on behalf of the whole
international com m unity and are therefore entitled to pronounce upon crimes that
offend universal values; that is, values recognized by the whole world community.
Such crimes do not only ru n counter to m oral and legal values prevailing in the
local com m unity directly affected by them. They also infringe values that are trans­
national and of concern to the world com m unity as a whole. Hence—as Jaspers
had noted back in 1961: se supra 21.1—only international courts representing this
com m unity can appropriately pronounce on such crimes.
Furthermore, such courts, as they apply international principles and rules, are not
bound by national approaches and traditions. They may therefore ensure some kind
of uniformity in the application of international law, whereas proceedings conducted
before national courts may lead to disparity both in the interpretation and application
of that law and the penalties given to those found guilty.
 440 IN T E R N A T IO N A L C R IM IN A L LAW

Finally, as international trials are by definition more visible than national criminal
proceedings, holding international trials signals the will of the international commu­
nity to break with the past, by punishing those who have deviated from acceptable
standards of human behaviour. In delivering punishment, the international commu­
nity’s purpose is not so much retribution as stigmatization of deviant behaviour, in the
hope that this will have a deterrent effect.

21.3 THE MAIN TRAITS OF INTERNATIONAL

CRIMINAL PROCEEDINGS
It may be appropriate briefly to sum up the specific features of international criminal
proceedings, as outlined in the previous chapters.

1. It is the Prosecutor who sets proceedings in motion (however, in the ICC system,
the Prosecutor may also act at the request of a state or the UN SC). The Prosecutor is
not duty bound to initiate proceedings any time he becomes cognizant of an inter­
national crime. He enjoys broad discretionary power in selecting the crimes on which
to concentrate. In this respect, it is for him or her to decide which crimes under the jur­
isdiction of the relevant international court are so serious and of concern to the world
community as to deserve to be brought before it. In the case of the ICC, the Prosecutor
enjoys a broader discretionary power; he chooses the ‘situation’ ‘in which one or more
crimes appear to have been com m itted’; in other words, he also chooses the country
where crimes have allegedly been perpetrated, besides the various categories of crime
against which to proceed.
However, in the ICTY the Prosecutor’s indictments are subjected to a sort of pre­
lim inary scrutiny by an administrative body, the Tribunal’s Bureau, whereas, more
appropriately, in the ICC system the Prosecutor acts under the scrutiny of the P-TC.
2. As a rule there is no international body such as an investigating judge charged
with the collection of evidence on behalf of both the prosecution and the defence. It
falls to the Prosecutor, before preferring charges against the accused, to search for and
gather the evidence. Although in principle the defendant could refrain from collecting
exculpatory evidence and confine himself to challenging the veracity and credibility
of the evidence led by the prosecution, in fact it is for the defence to seek out evidence
to refute the prosecution’s charges.
However, at least under the ICC Statute, the Prosecutor is also bound to search
for, gather, and pass on to the defence any evidence exonerating the accused.7 In
other words, he acts as an ‘organ of justice’; thus, at least within the ICC system, the
Prosecutor, unlike his counterpart in many national law systems, is not merely a party

7 Under Rule 68 of the ICTY and ICTR Rules, the prosecutor is only obliged to hand over to the defence
any exculpatory evidence he may have found.
 T H E SP EC IFICITY OF IN T E R N A T IO N A L T R IA L S 441

to trial. Furthermore, in a marked departure from the adversarial model prevailing in

international courts and tribunals, the ECCC system envisages the active role of inves­
tigating judges, who gather evidence on behalf of both parties, and in the event decide
whether to close the case or instead commit the accused for trial.
3. In order to collect evidence (interview witnesses, search premises, seize material
evidence, etc.) and arrest suspects, both the Prosecutor and the Defence must tu rn to
national authorities, in particular to the authorities of the state where the witness, or
material evidence, or the suspect are located. W ithout state cooperation international
trials become deadlocked or come to a standstill. True, in fact both the prosecution
and the defence may sometimes interview witnesses without having to turn to any
national authorities (sometimes to contact witnesses they simply telephone them or
send them an e-mail). The same is true of gathering evidence—certain evidence may
be garnered without turning to national authorities at all. However, anything involv­
ing coercive powers (obliging a crucial witness to testify in court, enjoining a person
to hand over a crucial piece of evidence, ordering a search of premises, compelling an
indictee to surrender to the court, etc.) depends on help from the authorities, as does
consulting archives, etc. which belong to a state.
4. There is no jury responsible for evaluating the facts. International courts consist
only o f professional judges. The underlying reasons for this regulation are that: (i) it
would be difficult to establish the criteria for appointing jurors (for instance, what
nationality should they be? How could they be selected?); (ii) international crimes are
complex offences; their appraisal requires extensive legal knowledge of both public
international law and criminal law. In addition, the facts are often extremely compli­
cated and the evidence may prove difficultto evaluate. In short, only experiencedjudges
possessing wide legal expertise may be in a position to adjudicate these crimes.
5. Although, strictly speaking, trials in absentia are not prohibited, they tend not to
be conducted. The adversarial model imposes the presence of the accused, except for
cases where he explicitly or implicitly waives his right to be present after appearing in
court (but the Charter of the Nuremberg IMT and the Statute of the STL make express
provision for trials in absentia, the latter Statute providing, however, for a set of condi­
tions on which the initiation of a trial in absentia is made contingent).
6. The guilty-plea procedure, adopted at the international level, does not yield the
same results as in common law systems, where it often makes it possible drastically
to cut the number of criminal trials. At the international level various factors tend to
discourage defendants from making use of this procedural mechanism. Nevertheless,
in recent times there seems to be an increasing trend towards greater resort to plea­
bargaining. Among the reasons behind this new tendency one may perhaps discern
the need to reduce the num ber of trials and thereby ease the current gridlock of inter­
national criminal proceedings.
7. International trials are not a contest between two parties, overseen by a neutral
and passive referee. International judges play an active role in directing the proceedings.
In particular, they possess extensive powers with regard to evidence. They may call
 442 IN T E R N A T IO N A L C R IM IN A L LAW

witnesses. They may also summon as court witnesses persons that a party would like
to call to testify, but is loath to do so on a num ber of grounds.
8. Victims do not play a major role except as witnesses. However, in the ICC sys­
tem (and under the Statute of the STL) they may set forth their views and concerns in
court, and, although they may not call witnesses, they are entitled to examine or cross-
examine witnesses called by either party (but they have no access to the evidence gath­
ered by the parties, nor can they lodge an appeal). Under Rules of the ECCC civil
parties’ (that is victims whose application to become ‘private petitioners’ participating
in the proceedings has been granted by the investigating judges or the TC) are instead
entitled to full participation in the proceedings, in keeping with the tradition of
inquisitorial systems.
9. As the principle nulla poena sine lege finds only limited application at the inter­
national level, courts enjoy broad powers in sentencing convicted persons.
10. As a rule appellate proceedings are not aimed at a retrial, but are designed to
verify whether the trial court erred in law or misapprehended facts in such a serious
m anner as to bring about a miscarriage of justice (however, in some limited cases
courts of appeal may hear new evidence).

21.4 MAIN PROBLEMS BESETTING INTERNATIONAL

CRIMINAL PROCEEDINGS
For all the merits of international trials, one should not, however, be blind to the
numerous and grave problems plaguing, and indeed sometimes m arring such trials.
The crucial problem international criminal courts face is the lack of enforcement
agencies directly available to those courts, for the purpose of collecting evidence,
searching premises, seizing documents, or executing arrest warrants and other judi­
cial orders (see above, 16.5). It follows that, as I have already emphasized many times,
international courts must rely heavily on the cooperation of states. They are totally
dependent on international diplomacy and states’ good will. As long as states refuse
outright to assist those courts in collecting evidence or arresting the accused persons,
or do not provide sufficient assistance, international criminal justice can hardly fulfil
its role. This, of course, also applies to those cases, such as that of the ICTY, where a
multilateral force established under the aegis of the UN provides assistance in execut­
ing arrest warrants (I am referring of course to the NATO forces operating in Bosnia
and Herzegovina and, more recently, to the EU forces operating in Kosovo).
In addition, there exists a need for international criminal courts to amalgamate the
approach of judges, each with a different cultural and legal background: some judges
come from common law countries, others from states with a Romano-Germanic tra­
dition; some are criminal lawyers, others are primarily familiar with international
law; some have previous judicial experience, others do not.
 T H E SP EC IFICITY OF IN T E R N A T IO N A L T R IA L S 443

Another serious problem is the excessive length of international criminal proceed­

ings. It results primarily from two factors: (i) the inherent difficulties of international
trials; and (ii) adoption of the adversarial system.
The intrinsic difficulties of international proceedings are linked to (a) the complex­
ity of such crimes as genocide, war crimes, and crimes against humanity, which nor­
mally are a manifestation of organized criminality and involve more than one person;
(b) the difficulty of collecting evidence that may be scattered over large territories
or more than one state; (c) the need to prove some special ingredients of the crimes
charged such as, for instance, the existence of a widespread or systematic practice (for
crimes against humanity) or to look into numerous crimes perpetrated by troops in
the field, to establish one of the objective elements of command responsibility (see
supra, 11.4); (d) language problems; at the national level proceedings are normally
conducted in only one language; before international courts in at least two, and pos­
sibly in three or more languages, with the consequence that documents and exhibits
need to be translated into all these languages.
The adversarial model further protracts international criminal proceedings. It
requires that all the evidence be scrutinized orally through examination and cross-
examination (whereas in many inquisitorial systems the evidence is selected and
appraised beforehand by the investigating judge). The adversarial system was con­
ceived of, and adopted, in most common law countries as a fairly exceptional alter­
native to the principal policy choice, namely avoidance of trial proceedings through
plea-bargaining. In fact, on account of this feature, the adversarial model works suffi­
ciently well in most countries. However, in international crim inal proceedings defend­
ants tend not to plead guilty, because of the serious stigma attaching to international
crimes: often they prefer to stand trial, in spite of the time involved in examination
and cross-examination of witnesses.
The excessive length of international trials is often coupled with and compounded by
the need to uphold a feature of the inquisitorial system that can be found in a num ber of
civil law countries, namely keeping the accused in custody both in the pre-trial phase
and during trial and appeal. At the international level this need is warranted by the
scant reliance that international tribunals and courts may have on the cooperation of
the relevant states forcefully to ensure that the defendants will appear again in court,
once summoned to resume their participation in trial or appellate proceedings.8

8 It should, however, be noted that a new trend is emerging in ad hoc international tribunals towards
granting provisional release to defendants. Thus recently ICTY’s judges granted requests for provisional
release to nine accused in three different cases.
On 20 June 2007 the AC in the Hadzihasanovic and Kubura case granted the motion on behalf of Enver
Hadzihasanovic for provisional release to Bosnia and Herzegovina pending the hearing of his appeal. The
AC found that a series of conditions for provisional release, including the fact that the time he spent in deten­
tion amounted to approximately two-thirds of his sentence, had been met. His release was subject to specific
terms and conditions which included the obligation to surrender his passport and report to the local police
regularly. On 15 March 2006, the TC had sentenced Hazihasanovic to five years’ imprisonment for murder
and cruel treatment of Bosnian Croat and Bosnian Serb civilians and prisoners of war.
444 IN T E R N A T I O N A L C R I M I N A L LAW

All this makes for a state of affairs that is hardly consistent with the right to a ‘fair
and expeditious trial’ and the presumption of innocence accruing to any defendant.
Another major flaw of international trials is that international criminal courts and
tribunals must perforce confine themselves to prosecuting and trying those who bear
the heaviest responsibilities for international crimes, the leaders or the high-ranking
m ilitary officers. They may not or cannot try the thousands of people who have
physically carried out murder, torture, rape, and other heinous acts. However, it is
precisely these perpetrators that the survivors and the relatives of the victims would
like to see in the dock.
Finally, some international criminal tribunals suffer from the ‘Nuremberg syn­
drome’, the tendency to try the ‘vanquished’, while the ‘victors’ remain sheltered from
any judicial scrutiny. It is a fact that the accusations widely made against NATO airmen
attacking Serbia in the 1999 war, or against some members of the Tutsi leadership for
the 1994 genocide, have never been verified through judicial inquiry. It is, however, to
be noted that other international courts or tribunals do not seem to be m arred by this
deficiency. In particular, the way the ICC has been structured entails that in principle
it should not be tainted by that notable drawback.

The TC in the Milutinovic and others case granted on 18 June 2007 the request by Vladimir Lazarevic for
temporary release to Serbia from 26 June to 2 July 2007 and the request by Nebojsa Pavkovic for temporary
release to Serbia from 4 July to 10 July 2007. Both provisional releases were subject to the specific terms and
conditions as detailed in the TC’s decision.
In the Prlic and others case the TC ruled on II June 2007 to accept the defence motions to provisionally
release Jadranko Prlic, Bruno Stojic, Slobodan Praljak, Milivoj Petkovic, Valentin Coric, and Berislav Pusic.
The TC decided to keep the dates and the specific terms and conditions confidential.
 IN D E X

absentia (trial in) convergence with inquisitorial analogy, ban on 47-50

international tribunals 378, systems 366 apartheid 13
389-94,441 court, role of appeals
national systems 360-1 judge, passive role of 361-2 generally 424-5
Nuremberg IMT 367,391-2 drawback of 365 interlocutory decisions
STL 391 evidence ICC 426-7
abuse of process doctrine 430 admission, rules of 363 ICTY and other
accused, right of 380-3, evaluation of 363 tribunals 425-6
384-94,403-4 gathering of 356 judgments
act of state doctrine 303-4 testimonial evidence 358 ICC 428
active personality principle 337 witness, introduction of ICTY and other
ad hoc tribunals documents by 358 tribunals 427-8
adversarial system of guilty plea 359 sentence
hearings 369-70 history of 354 ICC 428
flaws of 332 initiation of investigation by ICTY and other
generally 324-5 victims or relatives 355 tribunals 427-8
mixed or internationalized inquisitorial model armed conflict
Bosnia and compared 355-64 distinction between internal
Herzegovina 331 merits of 365 and international in ICC
Cambodia 330-5 plea bargaining 359 Statute 96
East Timor 330-5 transposition onto internal
establishment of 330-2 international legal level war crimes and 81,90
Lebanon 332 ad hoc tribunals 369-70 international 81
problems of 334-5 Nuremberg tribunal 366-7 modern weapons 95
rationale for establishment Tokyo international protected persons 89
of 332-4 tribunal 368 international terrorism in
Sierra Leone see Special trial proceedings subcategory of war
Court of Sierra Leone oral nature of 358 crimes 171-5
‘new world order’ and trials in absentia, prohibition armed resistance movements
establishment of 324-8 of 360 war crimes and 88
primary jurisdiction, victims, role of 361 Armenian massacres
assertion of 340 aggression (1915) 101-2,103
Rwanda see International conspiracy to wage 161 trials 318
Criminal Tribunal for criminal liability of individuals arms trade, unlawful 12
Rwanda or state responsibility arrest 346-51
spur to emergence of disentanglement of 155-8 attempt 220-5
International Criminal emergence of the notion a u td e d e re a u t judicare
Court 329-30 of 152-5 principle 338
Yugoslavia see International exclusion from international
Criminal Tribunal for jurisdiction 148 bail 360
the former Yugoslavia international crime, as 12,148 belligerent reprisals (taking of,
adversarial system objective elements of 158-9 as a justification) 258
accused self-defence and 154-5 Bernard Henry 368
bail and the 360 subjective elements of 159-61 Bernays, Murray C. 34
heard on oath 360 aiding and abetting beyond reasonable doubt
right to silence 360 joint criminal enterprise, standard 363,414-15
appellate proceedings 364 distinguished 206,211 biological experiments
composition of the court objective element 214-15,256-7 war crimes 92
judge, presided over by 357 subjective element 215-18 Bosnian Muslims, massacres of
jury and finding of guilt 357 see also superiors’ as genocide 135-6
sentencing 357 responsibility burden of proof 382,383
INTERNATIONAL
CRIMINAL LAW
This market leading textbook is written by an expert In the field of International
criminal law. The author uses his broad academic expertise and judicial experience to
provide a uniquely personal approach to the subject, with thought-provoking analysis
that also brings the political and human contexts to life.
International Criminal Law established itself as the most popular textbook in the
field in its first edition. The second edition continues to provide a clear and concise
introduction to both substantive international criminal law and international criminal
procedure. The book offers a stimulating explanation of the fundamentals of these
two branches of international criminal law and provides a theoretical framework to
the rules, principles, concepts and legal constructs that are central to the subject.
The book covers the various international courts and tribunals as well as mixed or
hybrid courts or such national courts as the Iraqi High Tribunal. In addition to using
the case law of the International Tribunals for the former Yugoslavia and Rwanda
and of recent mixed tribunals, it uniquely draws upon national jurisprudence, in
particular Belgian, Dutch, English, French, German, Italian, Spanish and US case law
on International crimes.

N EW TO T H IS ED IT IO N :
• N ew ch ap te rs on jo in t crim inal e n terp rise and com m and re sp o n sib ility
e nsure the book is up to date and co ve rs to p ics that have acquired
cru cial im p ortan ce in the international legal scene
• Referen ces to case law have been updated and exp an d ed in lig h t of the
rapid develo pm ents in international ju risp ru d e n ce

Antonio Cassese is Professor of International law, University of Florence; former

President of the Council of Europe Committee for the Prevention of Torture
(1989-93), former Judge and President of the International Tribunal for the
Former Yugoslavia (1993-2000); Chairman of the UN International Commission of
Inquiry into violations of Human Rights and Humanitarian Law in Darfur (2004-5),
Independent Expert tasked by the UN Secretary-General to Review the Efficiency
of the Special Court for Sierra Leone (2006); since 1995 member o! the Institut
de Droit International He was winner ol the Wolfgang Friedmann Memorial
Award for 2007 for outstanding contributions to the field of international law He
gives readers a unique perspective of international justice, based on his judicial
experience, which includes substantial contributions to leading cases including:
Tadic (1995 and 1999), Blasklc (1997), Erdemovic (1997) and Furundzfia (1998)

online resource centre

www.oxfordtextbooks.co.uk/orc/casseseicl2e/
A companion website provides:
• A wealth of biographical references
• All the major legal documents relevant to the subject

Cover linage 0 2005 Photo Scala. Florence/ Bildarchiv Preusstscher Kuliurbesitz Berlin