Jump to ContentJump to Main Navigation
The Hidden Histories of War Crimes Trials$

Kevin Heller and Gerry Simpson

Print publication date: 2013

Print ISBN-13: 9780199671144

Published to Oxford Scholarship Online: January 2014

DOI: 10.1093/acprof:oso/9780199671144.001.0001

Show Summary Details
Page of

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (oxford.universitypressscholarship.com). (c) Copyright Oxford University Press, 2021. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use. date: 23 January 2021

The Crime of Aggression

The Crime of Aggression

From the Trial of Takashi Sakai, August 1946, to the Kampala Review Conference on the ICC in 2010

(p.387) 19 The Crime of Aggression
The Hidden Histories of War Crimes Trials

Roger S. Clark

Oxford University Press

Abstract and Keywords

This chapter turns to the August 1946 trial of the Japanese general — Takashi Sakai — by a domestic tribunal in Nanking, the Republic of China, in order to reflect on the meaning of the concept of a ‘crime against peace’ or the ‘crime of aggression’, and the exercise of national jurisdiction. It raises three basic questions: What is the crime of aggression? Who can commit it? Where may they be prosecuted?

Keywords:   war crimes trials, Japanese war criminals, Takashi Sakai, crimes against peace, national jurisdiction

(I) Introduction

At the Review Conference on the International Criminal Court (ICC) held in Uganda 31 May to 11 June 2010, the States Parties to the Rome Statute brought to a next stage the efforts to define the crime of aggression and to set out the ‘conditions’ for the Court’s exercise of jurisdiction over the crime.1

I want to use the August 1946 trial of the Japanese general, Takashi Sakai, by a domestic tribunal in Nanking, the Republic of China, not so much to think about the exercise of jurisdiction by an international tribunal like the ICC or the International Military Tribunal for the Far East as to think both about the meaning of the concept of a ‘crime against peace’ or the ‘crime of aggression’2 and about the (p.388) exercise of national jurisdiction. My three basic questions go to (a) what is the crime of aggression; (b) who can commit it (politicians? Colonels? Foot soldiers? The janitor at the Ministry of Foreign Affairs?); and (c) where may they be prosecuted (only in an international tribunal? In an aggressor state? In a victim state? In any state that volunteers for the task?). In raising the set of questions about ‘national’ jurisdiction, I thus have in mind not only jurisdiction by a ‘territorial’ or ‘victim’ state, such as China was, but also those ‘third party states’ who might claim to exercise some version of universal jurisdiction.3 Now that the main unfinished business of Rome has been brought closer to completion,4 it is time to consider more carefully issues such as these which are highly relevant to what states need to do, or may do, when they accept the Statute as amended.

I do not suggest for a moment that Sakai provides a definitive precedent for anything, but I do find it enormously suggestive. The account of the trial as it appears in the Reports of the United Nations War Crimes Commission5 is brief and conclusory. I have been able to supplement that account. Professor Suzannah Linton kindly shared with me a document in English entitled Summary Translation of the Proceedings which she located in the National Archives in London. It appears to be the immediate source of the published report.6 At my request, Jingsi Wang, (p.389) a graduate student at Beijing Normal University, College of Criminal Science, located several Chinese secondary sources mentioning this and other Nanking prosecutions. He also discovered that China’s Second Historical Archives in Nanking have the Sakai judgment but that the archives are not open to the public. He was, however, successful in obtaining what appears to be a copy of all except a page or two of the whole judgment (in Mandarin) in the archives in Taipei. Yiqiang Lin was kind enough to translate this version for me.7 It is fairly similar to the two derivative documents, except that it contains a lengthy and very interesting ‘Attachment’ that summarizes the evidence involving Sakai. Neither the determinative facts nor the legal theories relied upon by the Tribunal are articulated with total clarity in any of the versions. But that is perhaps the beauty of the trial to the speculative mind.

Essentially, here is what happened. Sakai, a Japanese military officer, who eventually retired as a general in 1943, spent a period as a military commander in China during the war commencing in 1939, and prior to that during the Sino-Japanese hostilities from 1931 onwards. For the most part, his activities in the 1930s involved consolidating and expanding Japanese control in an undeclared war, rather than engaging in wholly new invasions. While not a great deal is made of it in the proceedings as a crime against peace, the highlight of his career as a leader seems to have been taking charge of the invasion of Hong Kong in December 1941,8 an operation that did not go as smoothly as his superiors had hoped.9 He was a joint holder of the post of Governor of Hong Kong for a while. A decade earlier, in 1931, he was responsible for creating terrorist disturbances in what are now known as the cities of Beijing and Tianjin and in the province around them (Hobei, or ‘Hopei’ at the time). In 1934, following assorted assassinations, he threatened to attack Beijing and Tianjin by artillery and air and effectively placed Hobei under Japanese control. Aside from Hong Kong, it is not entirely clear what new incursions he was involved (p.390) in after 1939, although it appears from the judgment that he was commander of the army stationed in Guangzhou (Canton) and Hainan in southern China when he was sent to take Hong Kong; and some of his war crimes were also alleged to have taken place in the south, on Hainan Island and in ‘Kwantung’—which in context seems to be another English transliteration of Guangzhou/Canton.10 He was convicted of crimes against peace, war crimes, crimes against humanity11 and, it appears, ‘offences against the internal security of the State [which] should be punished in accordance with the Criminal Code of the Republic of China’.12 Sentenced to death, he was duly executed by a firing squad on 30 September 1946.13

In a project like this, dedicated to unearthing the obscure, I suspect that someone will prove me wrong, but here goes: Sakai was the first Japanese accused to be convicted of a crime against peace14 and the first—and perhaps the only one—to be shot for it.15 His conviction and execution pre-dated the decisions of both the Nuremberg and Tokyo Tribunals.16

(p.391) (II) What is the Crime of Aggression?

If the applicable Chinese Rules governing the Trial of War Criminals17 contained a definition of crimes against peace, it is not reproduced in the Report. Instead, we learn that the Rules made some kind of reference to international law. The Notes on the case assert that the verdict on crimes against peace ‘was made with regard, though without express reference, to rules which were explicitly formulated in the latest development of international law in this sphere’.18 (How the ‘regard’ was made known is not established. None of the materials give any indication that the Tribunal referred to these documents.)19 The Notes then quote three ‘definitions’ that would later be carefully considered in the ICC drafting exercise: those in the (p.392) Nuremberg Charter, those in the Tokyo Charter and those in Control Council Law No. 10.20

The Nuremberg Charter reads:

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.21

The Tokyo Charter provides:

Crimes against peace: namely, the planning, preparation, initiation or waging of a declared or undeclared war of aggression, or a war in violation of international law, treaties, agreements or assurances, or participation in a common plan or conspiracy for the accomplishment of any of the foregoing.22

Control Council Law No. 10, the second to last chronologically23 (before Tokyo) is the most detailed of the three. It reads:

Crimes against peace: Initiation of invasions of other countries and wars of aggression in violation of international laws and treaties, including but not limited to planning, preparation, initiation or waging of a war or 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.24

None of these instruments was as detailed in defining the crime as the Kampala amendments. Article 8 bis of the Kampala amendments, which defines aggression, (p.393) makes a careful drafting distinction between a ‘crime of aggression’ and an ‘act of aggression’, the later being an element of the former. ‘Crime of aggression’, as defined in Kampala for the purpose of the Rome Statute, ‘means the planning, preparation, initiation or execution, by a person in a position effectively to exercise control over or to direct the political or military action of a State, of an act of aggression25 which, by its character, gravity and scale, constitutes a manifest violation of the Charter of the United Nations’.26 The definition goes on to explain ‘act of aggression’ in these words:

2. For the purpose of paragraph 1, ‘act of aggression’ means the use of armed force by a State against the sovereignty, territorial integrity or political independence of another State, or in any other manner inconsistent with the Charter of the United Nations. Any of the following acts, regardless of a declaration of war, shall, in accordance with United Nations General Assembly resolution 3314 (XXIX) of 14 December 1974, qualify as an act of aggression:

  1. a) The invasion or attack by the armed forces of a State of the territory of another State, or any military occupation, however temporary, resulting from such invasion or attack, or any annexation by the use of force of the territory of another State or part thereof;

  2. b) Bombardment by the armed forces of a State against the territory of another State or the use of any weapons by a State against the territory of another State;

  3. c) The blockade of the ports or coasts of a State by the armed forces of another State;

  4. d) An attack by the armed forces of a State on the land, sea or air forces, or marine and air fleets of another State;

  5. (p.394) e) The use of armed forces of one State which are within the territory of another State with the agreement of the receiving State, in contravention of the conditions provided for in the agreement or any extension of their presence in such territory beyond the termination of the agreement;

  6. f) The action of a State in allowing its territory, which it has placed at the disposal of another State, to be used by that other State for perpetrating an act of aggression against a third State;

  7. g) The sending by or on behalf of a State of armed bands, groups, irregulars or mercenaries, which carry out acts of armed force against another State of such gravity as to amount to the acts listed above, or its substantial involvement therein.27

How consistent is this with the Sakai prosecution? I put aside for the moment the question whether Sakai fitted into an appropriate place in the structure of the Japanese Government. That is the subject of the next section of this chapter. For the moment, though, consider the ways in which the depredations in which he was engaged fit the (now detailed) categories that can constitute an ‘act of aggression’. Consider Article 8bis (2), (a)–(g). At some points, there must have been an ‘invasion28 or attack by the armed forces of another State on the territory of another State’; a ‘military occupation, however temporary, resulting from such invasion or attack’; an ‘annexation by the use of force of the territory of another State or part thereof’; ‘bombardment by the armed forces of a State against the territory of another State or the use of any weapons by a State against the territory of another State’; a ‘blockade of the ports or coasts of a State by the armed forces of another State’; an ‘attack by the armed forces of a State on the land, sea or air forces, or marine and air fleets (p.395) of another State’; ‘use of armed forces of one State which are within the territory of another State with the agreement of the receiving State, in contravention of the conditions provided for in the agreement or any extension of their presence in such territory beyond the termination of the agreement’;29 and ‘sending by or on behalf of a State of armed bands, groups, irregulars or mercenaries, which carry out acts of armed force against another State of such gravity as to amount to the acts listed above, or its substantial involvement therein’.

In short, every action except possibly item (f) on the list seems to be plainly involved. In fact, with a little creativity, even (f) is perhaps implicated in Japanese activities (although Sakai was not involved in all of them). Consider, for example, the invasion of Hong Kong and the drive south to Burma, Malaysia and Indonesia, some of which was conducted from the Chinese mainland—the only real issue is whether China had ‘placed [the territory] at [Japan’s] disposal’ in any meaningful way.

To recapitulate, the actions of the Japanese State in China are encompassed by the Kampala definition of an ‘act of aggression’. But what of Sakai’s individual responsibility?

(III) Who can Commit the Crime of Aggression?

Sakai appears to have been convicted in some generalized reliance on the three 1940s sources of law reproduced above, on the theory that he was guilty of ‘participating in the war of aggression’.30 The note on the case asserts:

In the Nuremberg Charter the range of persons liable to prosecution and punishment for crimes against peace is defined in the first and last paragraph of Article 6. It includes any person (p.396) implicated in its commission whether as an individual or a member of organisations, or as a leader, organiser, instigator or accomplice. The same follows from the Far Eastern Charter and Law No. 10.31

‘The’ war32 rather begs the question about the events of the 1930s. What exactly was the nature of that conflict? There is, moreover, no discussion in the Report of why ‘participating’ is the appropriate verb (or why the notes on the case instead use the word ‘implicated’) or whether the theory was that he was part of a broad conspiracy. ‘Participating’ (or ‘implicated’) represents a pretty open-ended explanation of the conduct explaining who can commit the crime, and could catch those well down the food chain—although Sakai no doubt regarded himself as of some importance, as did his Chinese captors.33 Would the language used be enough to encompass the sergeants and the foot soldiers as well as the top generals and cabinet ministers who were prosecuted at Tokyo?34 The International Military Tribunal at Nuremberg tried to rein in the generality of the language35 and the subsequent Nuremberg tribunals made some more efforts in this direction.36

Think back to the conduct words in the Kampala definition. A perpetrator must be involved in the ‘planning, preparation, initiation or execution’ of an act of aggression. Sakai, if he fitted this, was probably mostly on the ‘execution’ end of the list. I doubt that even a high level person who carries out someone else’s policy, without more, (p.397) makes the cut, today at least. The drafters of the amendments adopted in Kampala tried to refine a little further in their definition who might be guilty by adding a circumstance element37 to the conduct one. To this end, the important language in their efforts is the requirement that the perpetrator be ‘a person in a position effectively to exercise control over or to direct the political or military action of a State’.38

The crime of aggression is thus a ‘leadership’ crime, a proposition captured by the element that the perpetrator has to be in a position effectively to exercise control over or to direct the political or military action of a state. There was vigorous discussion in the work leading up to Kampala about how this applies to someone like an industrialist who is closely involved with the organization of the state but not formally part of its structure.39 Some support was shown for clarifying the matter by choosing language closer to that used in the United States Military Tribunals at Nuremberg, namely ‘shape and influence’ the political or military action of a state, rather than ‘exercise control over or to direct’.40 Nevertheless, the quoted language remains in the Kampala amendments. Sakai does not appear from the material I have located to have been one who was able to ‘shape and influence’, let alone to ‘exercise control over or to direct’. He apparently acted on someone else’s policy decisions. Of course, one can see why the Chinese populace cheered at his execution—he was the hated person on the spot, not the remote one in Tokyo. I very much doubt, however, that he would have been convicted of a crime against peace had he lived to be indicted in the Tokyo Trial.

Perhaps it should be added that one can, in principle, be liable for most crimes either as a principal or as an accessory. It is fair to say that Sakai’s analysis runs together both the conduct words in the early part of the Nuremberg Charter’s definition41—words which are most applicable to principals—and the accomplice language that is found in the latter part of Article 6 of the Nuremberg Charter.42 This latter language applies on its face to all three Nuremberg offences, but fits a little awkwardly with aggression. If it really is a leadership crime, then it is more intellectually satisfying to characterize its perpetrators as principals rather than accomplices—the accomplice category applies more readily to underlings. It was (p.398) thoughts such as these that led in the Kampala negotiations not only to the definition of the crime which is found in Article 8bis, but also to an amendment to Article 25 of the Rome Statute. Article 25(3), found in the general part of the Rome Statute, deals with the various ways in which primary and secondary parties can be associated with an offence. Relatively early drafts of the aggression amendments simply excluded the application of Article 25(3), in toto,43 but some of the participants in the negotiation thought that this went too far and insisted on language designed to assert the leadership point again (perhaps from an abundance of caution)44 and at the same time leave open the possibility that the Court might find it useful to divine some residual possibilities in the Article for those close to the centres of power.

At all events, it is difficult to argue with the comments made by Lord Wright, Chairman of the United Nations War Crimes Commission, in the Foreword to the second of the two volumes containing a report of the Sakai trial:

[The case] included charges of crimes against peace as well as war crimes and crimes against humanity. I need only comment on the first of these charges: the main current of thought and decisions on crimes against peace which have been given since the end of the war have been that such crimes can only be committed as a matter of legal principle by accused individuals who may be described as acting on the policy-making level. In this particular case, however, it is difficult to see that the accused came with that category. I do not think that this decision can be relied on as substantially affecting the general current authority on this matter.45

One has thus to leave the issue of ‘who can’ with an expression of extreme doubt. Without knowing more about where he fitted in the Japanese hierarchy, is most unlikely that Sakai would be guilty of aggression under the Kampala definition. He was just not close enough to the centre of power as expressed there.

(p.399) (IV) Jurisdiction

Jurisdiction is a fundamental issue of international criminal law. I think that if an officious bystander had asked the members of the Tribunal that tried Sakai where it got its competence/jurisdiction from, it would have answered that its competence came (immediately at least) from the Chinese Rules governing the trial of war criminals.46 As a matter of positive law, it would probably have referred to Article I of the Rules in force at the time:

In the trial and punishment of war criminals, in addition to rules of international law, the present Rules shall be applied; in cases not covered by the present Rules, the Criminal Code of the Chinese Republic shall be applied.

In applying the Criminal Code of the Chinese Republic, the Special Law shall as far as possible be applied, irrespective of the status of the delinquent.47

Now, if one probes a little deeper, the Tribunal might add that there are various ways in which one might state questions of jurisdiction, especially in terms of jurisdiction to prescribe (that is to apply one’s laws to the situation) and in terms of jurisdiction to enforce (that is to apply the law to this particular person). It will be noted that China apparently domesticated the Nuremberg/Tokyo crimes in some fashion (as a matter of substantive law) but left open the application of other rules of existing domestic criminal law (both the general and the special parts) as needed. What might be the Tribunal’s specific theory when it came to applying international law, the details of which were apparently not spelled out in the legislation? (p.400) Was it simply a domestic tribunal exercising jurisdiction over what happened in the territory on an effects, passive personality or protective theory?48 Or is it to be regarded as a tribunal acting for the international community, exercising some sort of universal jurisdiction?

These thoughts flow inevitably into an examination of some of the arguments that have arisen concerning the ICC and the extent to which international law contemplates domestic jurisdiction in respect of the crime of aggression. One aspect of the terms of this debate is to look at how aggression and the fundamental principle of complementarity enshrined in the Rome Statute work together. This matter received some, albeit cursory, examination during the life of the SWGCA, whose work product paved the way for the Kampala amendments on aggression.49 Recalling that discussion is a good way into an appreciation of the issues. The discussion occurred in the context of the application of the concept of complementarity to the crime of aggression. Complementarity in the Rome Statute is the fundamental principle that deals with the implications of the proposition that there is concurrent jurisdiction as between a national jurisdiction, or set of national jurisdictions, and the ICC itself.

Article 17 of the Rome Statute teases out the basic rule of complementarity as one of ‘admissibility’. It provides, in relevant part, that:

  1. 1. Having regard to paragraph 10 of the Preamble and article 1, the Court shall determine that a case is inadmissible where:

    1. (a) The case is being investigated or prosecuted by a State which has jurisdiction over it, unless the State is unwilling or unable genuinely to carry out the investigation or prosecution;

    2. (b) The case has been investigated by a State which has jurisdiction over it and the State has decided not to prosecute the person concerned, unless the decision resulted from the unwillingness or inability of the State genuinely to prosecute;

    3. (c) The person concerned has already been tried for conduct which is the subject of the complaint, and a trial by the Court is not permitted under article 20, paragraph 3.50

(p.401) Two essential issues remain to be decided, probably by the Court in due course: (1) what is meant by ‘a State which has jurisdiction’? (2) Does this include a state acting on the basis of a universal jurisdiction theory?

A key phrase in both subparagraph (a) and subparagraph (b) is thus ‘a State which has jurisdiction over it’.51 There are three possible ‘states’ to which this might refer: an aggressor state; a victim state52 and a third (or ‘bystander’) state. ‘Has’ is perhaps interesting here. It must refer to the ground level requirement that the national law of the relevant state authorizes the prosecution and adjudication of the case53 (p.402) and to an absence of procedural obstacles to domestic prosecution.54 But does it also include some concept of the legitimacy of the exercise of jurisdiction by reference either to some considerations in the Statute itself or to some questions of general international law? For the purposes of complementarity in this respect there does not appear to be anything ‘special’ about aggression concerning aggressor state or victim state jurisdiction. They are relatively straightforward examples of states having nationality and territorial (or perhaps ‘protective’) jurisdiction.55 The more difficult case is that of universal jurisdiction by a bystander state. Two questions arise: (1) Does Article 17 even contemplate that the Court might defer to a bystander state on complementarity grounds? (2) Is aggression the subject of (p.403) universal jurisdiction under customary law? The first question applies to all of the crimes within the jurisdiction of the Court; the second is specific to aggression.

(1) Does a ‘State which has jurisdiction’ include one claiming universal jurisdiction?

As to the first question on how Article 17’s ‘has jurisdiction’ should be interpreted on universal jurisdiction, the position is frankly mysterious. One might have thought that the issue would be the subject of some useful preparatory work somewhere. If there is, it has not come to my attention. The author represented the Government of Samoa in the negotiations in New York and Rome leading to conclusion of the Statute. He asked on several occasions for an explanation about whether complementarity applied to a state contemplating acting on a basis of universal jurisdiction. He does not recall ever having received a useful answer. The subsequent literature is surprisingly sparse.56 For example, in the leading commentary of the Statute edited (p.404) by Otto Triffterer, the authors of neither the commentary on Article 1757 nor that on Article 18,58 address the issue. On the other hand, the late Christopher Hall, the author of the commentary of Article 19, discusses ‘State which has jurisdiction over a case’59 in these words:

Since all States under international law may exercise universal jurisdiction over the crimes within the Court’s jurisdiction, it is likely that paragraph 2(b) meant only to include those States which had provided their own courts with jurisdiction under national law over the case under the relevant principle of jurisdiction, whether based on territory, the protective principle, the nationality of the suspect or the victim or universality.60

The assumption here is that complementarity applies to situations in which it is appropriate to exercise universal jurisdiction (assuming the necessary legislation is in (p.405) place). The same assumption appears in an important article by Professor Darryl Robinson,61 a member of the Canadian delegation leading up to and in Rome who describes himself as ‘the drafter of the text that became Article 17’.62 That Robinson believes that ‘a State which has jurisdiction’ includes one operating on the basis of universality appears from his interesting discussion of ‘burden-sharing’ between the ICC and national courts. He invites the reader to:

Consider the scenario of a ‘third State’. The ICC has investigated a situation, and one of the persons most responsible has fled to a third State. Although the third State could initiate its own proceedings (for example under universal jurisdiction), and indeed the State would have the right to do so under complementarity, the ICC and the third State may agree that the ICC is the most efficient and effective forum to prosecute that person, because it has already amassed the necessary evidence.63

Neither the Hall comment,64 nor that by Robinson,65 addresses specifically the second question, to which we now turn, whether universal jurisdiction is appropriate for aggression.

(2) Is universal jurisdiction over aggression recognized in contemporary international law?

As to the question of the propriety of exercising universal jurisdiction over another State’s aggression there is fairly widespread agreement that aggression is a crime under international law.66 But does that translate into a crime for which there is universal jurisdiction? It is very doubtful that under current customary law it can be asserted unequivocally that aggression ‘is’ subject to universal jurisdiction.67 The (mostly) non-governmental authors of the Princeton Principles on (p.406) Universal Jurisdiction contend that there is universal jurisdiction over a person ‘duly accused of committing serious crimes under international law’.68 ‘For the purposes of these Principles’, they contend, ‘serious crimes under international law include: (1) piracy; (2) slavery; (3) war crimes; (4) crimes against peace; (5) crimes against humanity; (6) genocide; and (7) torture’.69 Yet one has to have some doubts about how far the inclusion of crimes against peace in this list is supported by State practice. Other authorities are much more equivocal.70 Putting aside Sakai and the other cases brought in victim states,71 the only prosecutions clearly for the crime against peace were those under the Nuremberg and Tokyo Charters and under Control Council Law No. 10 in the 1940s and these were before tribunals best described as ‘international’ or ‘victor’s’ (or both). I know of no disinterested third-party prosecution for aggression. On the other hand, Astrid Reisinger Coracini’s careful research has located what she describes as ‘statutory provisions relating to aggression as a crime under international law in some twenty-five countries, predominantly Eastern European and Central Asian States’.72 Some, but by no means all, of those twenty-five (and there may be a few others) contemplate universal jurisdiction.73 But as practice, this is pretty thin!

Then there is the work of the International Law Commission on the Draft Code of Crimes against the Peace and Security of Mankind, commencing in the 1940s and concluding in 1996.74 The work had proceeded on the basis that the (p.407) Nuremberg and Tokyo crimes—war crimes, crimes against humanity and crimes against peace—were part of a more general category, the details of which would emerge from rational analysis. That hope ultimately proved illusory and the work-product in 1996 included essentially the Nuremberg crimes, amplified by the addition of genocide, which had emerged from crimes against humanity, and crimes against United Nations and associated personnel. Article 16 of the Draft Code notes75 the crime of aggression; Article 17 the crime of genocide; Article 18 crimes against humanity; Article 19 crimes against United Nations and associated personnel; and Article 20 war crimes. For present purposes, the interesting issue is the jurisdictional theories addressed by the Commission. Article 8, entitled ‘Establishment of jurisdiction’, provides:

Without prejudice to the jurisdiction of an international criminal court, each State Party shall take such measures as may be necessary to establish its jurisdiction over the crimes set out in articles 17, 18, 19 and 20, irrespective of where or by whom those crimes were committed. Jurisdiction over the crime set out in article 16 shall rest with an international court. However, a State referred to in article 16 is not precluded from trying its nationals for the crime set out in that article.76

Notably absent is any reference to a requirement for the exercise of universal jurisdiction over the crime of aggression. The International Law Commission’s (ILC’s) draft can thus be read as an assertion that general international law supports universal jurisdiction for the other crimes to which the Draft Code refers, but that aggression is different—jurisdiction rests there with an international court or the courts of the aggressor state. Indeed, by implication, the ILC even casts doubt on whether there is jurisdiction in the courts of the victim State.77

(p.408) The United States did not participate in the work of the SWGCA. However, it sent a large delegation to the final meetings of the Assembly of States Parties that took place before the Review Conference, expressing doubts there and in a number of informal meetings early in 2010 about the jurisdictional principles applicable to the crime of aggression. It pursued the matter in Kampala. At its insistence, the following paragraphs were asserted in the ‘understandings’ annexed to the resolution adopting the aggression amendments, beneath the heading ‘Domestic jurisdiction over the crime of aggression’:

4. It is understood that the amendments that address the definition of the act of aggression and the crime of aggression do so for the purpose of this Statute only. The amendments shall, in accordance with article 10 of the Rome Statute, not be interpreted as limiting or prejudicing in any way existing or developing rules of international law for purposes other than this Statute.

5. It is understood that the amendments shall not be interpreted as creating the right or obligation to exercise domestic jurisdiction with respect to an act of aggression committed by another State.78

Paragraph 4 can hardly be regarded as saying anything more than Article 10 of the Rome Statute says about leaving principles of international customary law free to develop for purposes other than the Statute. (This is not to say that the Statute does not, inevitably, contribute to the development of customary law. Believing to the contrary is like believing that King Canute could really order the waves back.) It is perhaps best understood as an effort to insist on the hardly earth-shattering proposition that, whatever customary law is on the meaning of ‘act of aggression’ and ‘crime of aggression’, it might not be exactly the same as in the 2010 amendment. Paragraph 5 is likely to have more practical effect. It is plainly aimed at discouraging states from exercising jurisdiction over the crime of aggression based on universal jurisdiction, and possibly even on the basis of being a victim State.79 It is, however, hardly a forceful proposition about the right to exercise these jurisdictions, other than to assert that the right does not rest on the amendments—it has presumably to be found in existing customary law.

(p.409) (V) Conclusion

In my introduction, I raised three basic questions about the crime of aggression: What is it? Who can commit it? Where (other than in an international court or tribunal) may the perpetrators be prosecuted? I suggested that Sakai, while it has no definitive answers to any of these questions and hardly stands as authority for anything, is very suggestive of the issues and of potential answers. These issues were all on the table over five decades later at Kampala and are bound to suck up the lives of many trees (and their electronic equivalents) in decades to come.

As to the what is it? question, it has always been clear that, for purposes of criminal responsibility, the elements of the crime of aggression entail a combination of something that a State80 as an entity ‘did’ and something that an individual did on behalf of that state. Both sets of elements are necessary.

As to the State: at Nuremberg and Tokyo and, I think, in Nanking, there was little discussion of what acts by a State could, in general, amount to aggression. The assumption was that the Axis powers had done ‘it’ (subject to any possible justifications such as self-defence that tended inexorably to fail on the facts); the question was whether the accused was a person who should bear the responsibility. Kampala, using the General Assembly’s definition as a basis, has come up with a list of things that a state might do to bring itself within the aggressor category. Using the Kampala definition of ‘act of aggression’ as a checklist, I have suggested that Japan’s actions fitted most of those ‘acts’ on the list.

As to the individual and the who question: Sakai acted on behalf of Japan, although his plea that he was only acting on orders and that Japan alone should therefore be responsible was not acceptable.81 Denying the accused a plea of ‘act of state’ was again a major move made at Nuremberg, Tokyo, and Nanking. Nanking described Sakai as a ‘leader’ but did not further explain the category. Kampala, echoing Nuremberg, speaks of a ‘person in a position effectively to exercise control over or to direct the political or military action of a State’. As in any legal category, there will be easy cases and more difficult cases when it comes to apply this formula, but prosecutors I have spoken to assure me that it can be applied, with (p.410) little more difficulty than many other criminal categories. I doubt that Sakai would fit the category, but more information about his place in the Japanese hierarchy would perhaps be helpful.

Then there is the question of where (in addition to an international tribunal) the crime may be prosecuted. Nanking China had legislation in place (although I do not have the exact text) that permitted the trial of a defeated enemy. No one seems to have had the slightest doubt at the time about the appropriateness of this. As we have seen, there was, however, some debate about the propriety of victim state jurisdiction both in the ILC and surrounding the Kampala amendments. As for universal jurisdiction, there are legislative precedents for taking such jurisdiction, and there are arguments against it. This is no doubt a work in progress and we shall see what happens as states adopt the Kampala amendments as part of their domestic legal structures.


(1) Article 5(1) of the Rome Statute of the International Criminal Court, adopted in 1998, included aggression within the crimes over which the Court has jurisdiction, but Article 5(2) required further work leading to a ‘definition’ and ‘conditions for the exercise’ of that jurisdiction. This became the main item on the agenda of the 2010 Review Conference. A comprehensive resolution of the aggression issues left over from 1998 is contained in the Review Conference’s Resolution 6, ICC Doc. RC/Res. 6 (2010) (Kampala Amendments). See Roger S. Clark, ‘Amendments to the Rome Statute of the International Criminal Court Considered at the first Review Conference on the Court, Kampala, 31 May–11 June 2010’, Goettingen J. Int’l L., 2 (2010), 689.

(2) The terms are interchangeable. Generally speaking, the term ‘crimes against peace’ was the more common usage in the 1940s and 50s; by the time of the International Law Commission’s (ILC)’s Draft Statute for an International Criminal Court in 1994, United Nations (UN) General Assembly Official Records, 49th Sess., Supp. No. 10, UN Doc. A/49/10 (1994), ‘crime of aggression’ had largely captured the day. An early analysis of the crime, which influenced the negotiations on the Nuremberg Charter and thinking on the subject at the time of Sakai, is A.N. Trainin, Hitlerite Responsibility Under Criminal Law (n.d. probably 1945 English trans.). Professor Trainin, using both terms, set the stage for much of the later discussion with these words, at 35:

The direct and most dangerous form of offence against peace is the attack of one State on another—aggression—which directly breaks the peace, and forces war on the peoples. Aggression is, therefore, the most dangerous international crime. In the interests of the struggle for peace, the penalty for the crime must fall not only on those guilty of carrying out aggression, but also on those who try to fan the flame of war, who prepare aggression. Activities preparing the ground for aggression must comprise the conclusion of blocs and agreements having the aim of aggression (as, for example, the ‘Axis’ Treaty between Germany and Italy); the infringement of treaties which serve the cause of peace; the provoking of international conflicts by all kinds of means; the propaganda for aggression.

(Footnote emphasizing that aggression does not ‘refer to just wars, wars of liberation’ omitted.)

In spite of such arguments, preparation for war alone never gained much traction as an inchoate offence, although the Kampala amendments catch in the criminal net those who prepare or plan for an aggression which is in fact completed. See text at n 25 below.

(3) It is common, in the Anglo-American literature at least, to speak as though there is something of a closed list of bases of national jurisdiction that are acceptable in at least some circumstances: territorial (including ‘objective territorial’ or ‘effects’), nationality (or active personality), passive personality, protective and universal. See generally Ellen S. Podgor and Roger S. Clark, Understanding International Criminal Law (New York, NY: LexisNexis, 2nd edn, 2009), 25–6. As we note there, State practice is richer than that; it includes other bases that seem to be acceptable such as ‘landing state jurisdiction’ (the place where planes or ships arrive following criminal activity) and ‘transferred jurisdiction’. There is no need to pursue exotic alternatives here, since the classic categories cover the jurisdictional field for at least the issues with aggression that were current in the 1940s and even now.

(4) The ICC will finally be able to exercise its jurisdiction over the crime of aggression one year after thirty States have ratified the Kampala amendments and there is a further decision from the Assembly of States Parties (ASP) bringing everything into effect, whichever is later. The ASP decision may not take place before 1 January 2017.

(5) Trial of Takashi Sakai, United Nations War Crimes Commission, XIV Law Reports of Trials of War Criminals 1, Case No. 83, Chinese War Crimes Military Tribunal of the Ministry of National Defence, Nanking, 29 August 1946 (the Report). Philip R. Piccigallo, The Japanese on Trial: Allied War Crimes Operations in the East, 1945–51 (Austin, TX: University of Texas Press, 1979), 164–5 has some newspaper references that reinforce the material in the report. Sakai was one of a number of trials held in Nanking, but the only one reported in the United Nations collection. Piccigallo’s book is the most comprehensive discussion of the Nanking Trials that I have found in English.

(6) The summary translation (Summary Translation) is organized under two headings, ‘Facts’ and ‘Reasons’. The material therein appears (somewhat edited and re-organized) as ‘Facts and Evidence’, ‘Defence of the Accused’ and ‘Findings and Sentences’ of the Tribunal in the Report. The report also contains four pages of ‘Notes on the Case’, supplied by the UN War Crimes Commission editors. The Notes include some matter in the judgment and the Summary Translation that is omitted from the Report.

(7) ‘The judgment’. I hope that it will be published.

(8) The Hong Kong invasion seems to be treated, almost in passing, as one aspect of Sakai’s involvement in the aggression against China. On the other hand, the Attachment to the Judgment contains twenty-one paragraphs of findings about war crimes and crimes against humanity attributed to him in respect of atrocities committed by his troops in Hong Kong, especially against prisoners of war and medical personnel. There is no discussion in the materials of the (then) international status of Hong Kong as British rather than Chinese territory. Does the tribunal assume (a) that Hong Kong was part of China, or (b) that, even if it was to be regarded as within the sovereignty of Great Britain, there is jurisdiction over such crimes anyway in a domestic tribunal under some kind of universal theory, or (c) neither of the above? The trial in question was conducted by the Nationalist or ‘Republic of China’ authorities that currently occupied the Chinese seat at the recently established United Nations. The Republic apparently regarded Hong Kong as part of China but did not resist the British placement of it on the General Assembly’s list of non-self-governing territories. After the People’s Republic took up the Chinese seat in New York, it insisted successfully that it be removed from the list, as it was to be considered part of China. See Letter of March 8, 1972 from the Permanent Representative of China to the United Nations addressed to the Chairman of the Special Committee on decolonization, UN Doc. A/AC.109/396 (1972).

(9) In a passage omitted in the later material, the judgment records that he led a large number of troops that attacked Hong Kong by surprise on 8 December 1941, ordering his troops to take it within two days. The siege continued until the 17th because of unexpected resistance ‘which made [Sakai] turn to revenge by brutality’.

(10) ‘Kwantung Army’ is also a term used to describe the Japanese crack forces that occupied Manchuria in the north.

(11) The material offers no indication of what the Tribunal regarded as the legal basis of a crime against humanity. The Tribunal is merely quoted as saying:

In inciting or permitting his subordinates to murder prisoners of war, wounded soldiers, nurses and doctors of the Red Cross and other non-combatants, and to commit acts of rape, plunder, deportation, tortures and destruction of property, he had violated the Hague Conventions concerning the Laws and Customs of War on Land and the Geneva Convention of 1929. These offences are war crimes and crimes against humanity.

Report, above n 5, (Notes) 7.

The ‘source’ of crimes against humanity as understood at Nuremberg was general principles of law, rather than anything to be found in the black letter of the Geneva or Hague Conventions. There is, however, an echo of the Martens clause in preamble to the 1907 Fourth Hague Convention in the concept of crimes against humanity. The big difference is that Hague and Geneva required a ‘war’ while crimes against humanity did not necessarily require an armed attack. None of this is addressed in Sakai.

(12) Report, above n 5, 5. It may be that this was an alternative to convicting him for the crime against peace in respect of (some of) his activities in the mid-30s. On the other hand, it may be that the reference to the Chinese Criminal Code is to the provisions allowing for the trial and punishment of crimes against peace. The material is ambiguous. A domestic tribunal has the advantage over a typical international tribunal that it may well have competence to exercise jurisdiction not only over international crimes but also over domestic ones committed on the national territory. There are some nice questions that are finessed here about the extent of a combatant’s privilege to kill in a messy situation like that in China in the 1930s. Was it a military occupation to which the rules of armed conflict apply to the exclusion of territorial law? Were the Japanese occupiers some early example of unlawful combatants subject to domestic law? Some wonderful questions, but not much enlightenment!

(13) As Piccigallo puts it: ‘The court condemned Sakai to be shot. After ratification by Chiang Kai-Shek, the execution took place as scheduled, before a large, approving public audience.’

Piccigallo, above n 5, 165. Close your eyes, dear reader, and imagine the cheering throngs at the shooting.

(14) At least one German, Arthur Greiser, preceded him to conviction and public execution (that time by hanging) by a few weeks. See Mark Drumbl’s contribution to this collection. (There must be a good article out there somewhere on the esthetics of shooting rather than hanging.)

(15) It was not the last Chinese trial including aggression. Using newspaper sources, Piccigallo, above n 5, 163–4, discusses at least two later Nanking prosecutions in which aggression was one of the charges, those of Generals Isogai (life imprisonment) and Tani (executed by hanging). I have not located any comprehensive account of the Chinese trials.

(16) The Sakai trial was concluded on 27 August 1946 (Summary Translation, above n 6, and Judgment, above n 7) or on the 29th (the Report, above n 5). It is not clear when it began, although paragraph 19 of the Attachment to the Judgment has a witness testifying on 11 June 1946. The Nuremberg verdict did not come until 1 October 1946. General MacArthur promulgated the Charter of the International Military Tribunal for the Far East (Tokyo Tribunal) on 19 January 1946; that trial began on 29 April 1946 but was not completed until November 1948. Paragraph 4 of the Attachment, in the context of the 1935 threats to attack Beijing and Tianjin, refers to evidence given before the Far East Tribunal on 9 July 1946 by a Major-General Tanaka. So the Nuremberg and Tokyo Charters were evidently available to the Chinese judges, as was some of the early evidence given in Tokyo. But the Nuremberg and Tokyo judgments were not. By the same token, Control Council Law No. 10 (CC Law No. 10), under which later American trials, one French trial, and a handful of Soviet trials took place, was promulgated on 20 December 1945 and is referred to in the Notes on the case in the published account of the Chinese decision, although it is not apparent what use the Chinese Tribunal itself made of CC Law No. 10, if any. The Report, the Summary Translation and the Judgment make no mention of the relevant Charters or CC Law No. 10, but the Report and the Summary do refer to Article 1 of the Nine-Power Treaty of 1922 and to Article 1 of the Pact of Paris 192 (Kellogg–Briand Pact). In the Nine-Power Treaty (Treaty Between the United States of America, Belgium, the British Empire, China, France, Italy, Japan, the Netherlands, and Portugal, signed at Washington, February 6, 1922) the Contracting Powers undertook in Article I(1) ‘[t]‌o respect the sovereignty, the independence, and the territorial and administrative integrity of China’. The Notes on the case, at 7, refer to these two treaties and point out that the Tribunal ‘had thereby stressed’ Sakai’s guilt in taking part in ‘a war in violation of international treaties’. None of the earlier British, US, Australian or Dutch trials that I have found included aggression (or later ones, for that matter), although the Australian law permitted such a charge. The Soviet Union had instituted war crimes trials against German military (and Soviet collaborators) beginning with the December 1943 Kharkov trial. See George Ginsburgs, Moscow’s Road to Nuremberg: The Soviet Background to the Trial (Boston, MA and Leiden: Martinus Nijhoff, 1996), 52. While Soviet scholars had written since the 1930s about the criminality of aggression, it is not clear which of these early trials included that offence and I am aware of no trials conducted by other states that definitely included aggression before this one, other than the Polish prosecution of Greiser. Nonetheless, over the next few years the Soviet Union convicted a large number of German military for illegal activity that included crimes against peace. This is definitely an area begging for more research. Pioneering archival research is contained in Irina V. Bezborodova, Generaly Vermakhta v plenu (Generals of the Wehrmacht in Captivity) (Moscow: Rossiski gos. gumanitarny universitet, 1998), carefully reviewed in George Ginsburgs, ‘Light Shed on the Story of Wehrmacht Generals in Soviet Captivity,’ Crim. L. Forum,11 (2000),101–20.

(17) The Notes to the Report, at 3 n 1, after quoting part of the Rules, says that ‘[t]‌he above Rules were later replaced by a Law Governing the Trial of War Criminals of 24th October, 1946, an account of which will be found in the Annex to this Volume [Vol. 14 at 152]’. Since we are not given most of the text of either document, it is hard to know how they differ. On some apparent textual differences between the two sets of Rules, see below n 47.

(18) Report, above n 5, 3.

(19) Compare the explicit references to the Nine-Power Treaty and the Kellogg–Briand Pact, above n 16.

(20) The last of these gave rise to the second round of United States prosecutions in Nuremberg and to the French prosecution of Roechling. Unlike the Nuremberg trials, which did not permit an appeal, the French system did, luckily for Roechling. His conviction was reversed on appeal on the basis that ‘his vanity perhaps allowed him to attribute more authority to himself than he was actually entitled to’. An embarrassing way to escape punishment! See the UN Secretariat’s Historical review of developments relating to aggression, UN Doc. PCNICC/2002/WGCA/L. 1, 83.

(21) Nuremberg Charter, Article 6(a). Article 6 concludes with the following, applicable to all the Nuremberg crimes:

Leaders, organizers, instigators and accomplices participating in the formulation or execution of a common plan or conspiracy to commit any of the forgoing crimes are responsible for all acts performed by any persons in execution of such plan.

It has always been a hard question who beyond leaders, organizers and instigators can be held responsible for aggression.

(22) Tokyo Charter, Article 5(A). Words in italics do not appear in the Nuremberg Charter. The words clarify that a declaration of war is irrelevant. They also contain a reference to international law (adding custom to treaties) that would in theory be helpful to the prosecution. In practice, the Kellogg–Briand Pact of 1928 provided the main legal source for prosecution at Nuremberg and Tokyo (and it may well be of Sakai).

(23) Above n 16.

(24) Control Council Law No. 10, Article II(1)(a). Words in italics are not in the Nuremberg Charter. The reference to ‘invasions’ finesses an argument that took place in the International Military Tribunal at Nuremberg and in the drafting of the Kampala amendments. The invasion of Austria, Bohemia and Moravia occurred as a result of threats rather than actual fighting (would such fighting be necessary for a ‘war’?) and the Tribunal stopped short of calling those actions wars of aggression. CC Law No. 10 would encompass such actions, although no one was ultimately convicted on such a theory. (Prosecutions failed on the merits.) The mysterious meaning of ‘war’ was a major reason why the Kampala amendments avoided the use of that term, against the assertion that Nuremberg required that there be a war. CC Law No. 10 at least muddied those waters. CC Law No. 10 also included the insidious drafting words ‘including but not limited to’. There was a great deal of discussion in the drafting of the Kampala amendments about whether the list of ‘acts’ therein were the only ones that could amount to aggression—was the list, ‘open’, ‘closed’ or ‘semi-closed’? The received wisdom is that the Kampala list is (semi-)open, but that any additions must be ejusdem generis with those named specifically.

(25) One major challenge with defining the crime of aggression since Nuremberg has been how to explicate the intricate relationship between what the aggressor state does (for which there is state responsibility) and what the individual actor does. The drafters of the Kampala amendments used a drafting convention under which what the state does is defined as an ‘act of aggression’ and what the individual does is the ‘crime of aggression’. The definition of ‘act of aggression’ in the amendment is derived substantially from the General Assembly’s famous Definition of Aggression, GA Res. 3314 (XXIX) of 14 December 1974, stripped of much of its indeterminacy. Resolution 3314 is, in turn, substantially derived from the Convention for the Definition of Aggression signed in London, 3 July 1933. During the drafting of the Nuremberg Charter, Justice Jackson endeavoured to incorporate some of the detail from this treaty, pointing out quite correctly that the Soviet Union was its main proponent. The USSR was not so keen, at that point, on something that could apply generally in the future. Thus the Nuremberg and Tokyo Charters took for granted that ‘it’ was what their adversaries did. See Roger S. Clark, ‘Nuremberg and the Crime against Peace’, Wash. U. Global Stud. L. Rev., 6 (2007), 529–36.

(26) Article 8bis(1). Article 8bis represented the drafting work of the Court’s Special Working Group on the Crime of Aggression (SWGCA), open to all states. Between 2003 and 2009 the SWGCA had picked up work begun before Rome and in the Preparatory Commission for the Court which functioned between 1998 and 2002. Article 8bis and its accompanying Elements of Crime (which spell out exactly what the prosecution must prove) were adopted with no change whatsoever in Kampala. The Group’s Article 15bis, which dealt with the ‘conditions’ for the exercise of jurisdiction, was less refined and far more contentious. In the end, a somewhat different pair of Articles, 15bis and 15ter, emerged in Kampala. Article 15bis deals with complaints triggered by states or by the Prosecutor acting proprio motu; 15ter deals with Security Council referrals: Clark, above n 1.

(27) One of the debates that took place during the drafting of the ICC provisions on aggression was whether the public law ‘defences’, such as self-defence of the state, authorization of the Security Council or (more controversially) humanitarian intervention, should be specifically included. Ultimately, they were not, the matter being left to be dealt with under the requirement that a breach of the Charter be ‘manifest’ and/or by the material dealing with defences in Article 31 of the Rome Statute. Due process requires that an accused be entitled to raise such issues. There was an interesting example of this in Sakai, best explained in the Summary Translation, above n 6. In respect of the events in the 1930s, the Summary states that the ‘defendant pleaded on the grounds that when he demanded the withdrawal of Chinese troops from Hopei and the dismissal of Chinese Administrative Heads in Hopei, he acted within the stipulations of the Final Protocol of 1901’. (The Protocol represented the settlement with the ‘Powers’, including Japan, at the conclusion of the Boxer Rebellion.) The Tribunal responded, implicitly agreeing that the argument was one that might be made, but nonetheless dismissing it on the merits:

[T]‌here is no stipulation in the Final Protocol of 1901 which prohibits the Chinese Government from stationing Chinese troops in Hopei and it does not give Japan the right to demand the dismissal of Chinese Administrative Heads in Hopei. The argument that the Exchange of Notes appended to the Protocol provides a limit of 20 li, separating the stationing of Japanese troops in Tientsin from Chinese troops, is untenable. It is clear that this plea is a deliberate mis-interpretation of the said Protocol.

(Summary Translation, above n 6, 2.)

The Notes, at 5–6, discuss this argument as though it applied to charges of the breach of Chinese domestic law; I think it more likely from the context in the Summary and the judgment that it related to an aggression theory.

(28) Notice the reference to an ‘invasion’ in Article 8bis(2)(a), thus sidestepping the Austria/Bohemia/Moravia problem of lack of a ‘war’: see above n 24. Threats by Sakai to attack Beijing and Tianjin with artillery and air force led to the withdrawal of Chinese troops and effective control by Japan. Was this similar to Goering’s threat to bomb Prague off the map, followed by an incursion: an ‘invasion’ rather than a ‘war’?

(29) Consider, in respect of this ‘act’, the creeping nature of the activities in which Sakai engaged, bearing in mind that ‘consent’ of some kind (‘not real’?) was extracted from China from time to time.

(30) Report, above n 5, 2, 6. In summarily denying the accused’s defence of superior orders, the Tribunal is also reported as saying: ‘Granted that the defendant participated in the war on the orders of his Government, a superior order cannot be held to absolve the defendant from liability for the crime’ (at 5). Participation’ appears in the Nuremberg text both in the basic definition of aggression which includes also a separate inchoate crime of a ‘common plan or conspiracy’ to commit a crime against peace and also in the final paragraph of the definitional Article. In the latter, it relates to complicities and is applicable to all crimes. The basic definition of crimes against peace reads:

namely the planning, preparation, initiation or waging of a declared or undeclared war of aggression, or a war in violation of international law, treaties, agreements or assurances, or participation in a common plan or conspiracy for the accomplishment of any of the foregoing.

The final paragraph of Article 6 of the Nuremberg Charter deals generally with individual responsibility for each of the three Nuremberg crimes. It reads:

Leaders, organizers, instigators and accomplices participating in the formulation or execution of a common plan or conspiracy to commit any of the foregoing crimes are responsible for all acts performed by any persons in execution of such plan.

The discussion of the General’s connection to the crimes of his subordinates did not add anything to our understanding of the rules for command responsibility. It did, however, flag the issues for later analysis. He was apparently charged on the basis that he ‘incited’ or ‘permitted’ his subordinates to commit the relevant war crimes and crimes against humanity (which were not precisely distinguished). Report, above n 5, 7. He defended on the basis that ‘he could not be held responsible for the…violations because they were perpetrated by his subordinates and he had no knowledge of them’ (at 7). His argument must have raised the nice question about what element of culpability applied to a commander. Was it a case for strict liability? Negligence? Recklessness? Knowledge? Intent? Rather than entering into a debate on the law, the Tribunal found against him on the facts: ‘All the evidence goes to show that the defendant knew of the atrocities committed by his subordinates and deliberately let loose savagery upon civilians and prisoners of war’ (at 7). Knowledge accompanied by deliberate inaction was evidently enough. Compare the standard for military leaders in Article 28 of the Rome Statute (‘knew or should have known’).

(31) Report, above n 5, 4.

(32) Above n 24.

(33) The Summary Translation, above n 6, 2, describes Sakai as ‘one of the leaders who came to China to carry out Japanese aggressive policy’. The Report, above n 5, 1, renders this as ‘one of the leaders who were instrumental in Japan’s aggression against China’. Yiqang Lin’s translation of the judgment says ‘he was one of those chiefly responsible for implementing the Japanese war of aggression.’ The differences are subtle and the Report’s version perhaps ascribes to Sakai a little more policy-making capacity than the others.

(34) Article 5 of the Tokyo Charter limited jurisdiction of the Tribunal to ‘the power to try and punish Far Eastern war criminals who as individuals or as members of organizations are charged with offenses which include Crimes against Peace’. A judgment call had to be made by the prosecution and some of the accused had a tenuous connection to the aggression, yet all were ultimately convicted of it.

(35) See generally, Clark, above n 25, 527.

(36) See Kevin Jon Heller, ‘Retreat from Nuremberg: The Leadership Requirement in the Crime of Aggression’, EJIL, 18 (2007), 477.

(37) The Elements of Crimes, echoing Article 30 of the Rome Statute, treat the material elements of a crime as being conduct, consequence or circumstance elements.

(38) Above n 25.

(39) See Heller, above n 36.

(40) See Informal Intersessional meeting of the Special Working Group on the Crime of Aggression, held at the Liechtenstein Institute on Self-Determination, Woodrow Wilson School, Princeton University, from 11 to 14 June 2007, Doc. ICC-ASP/6/SWGCA/INF.1 (2007) at 3. Towards the end of the work of the Preparatory Commission for the Court, Samoa made a proposal for the Elements of Aggression which included a statement that the perpetrator ‘need not formally be a member of the government or military’ but the group thought it unnecessary to spell this out. See Elements of Crime of Aggression—Proposal Submitted by Samoa, UN Doc. PCNICC/2002/WGCA/DP.2. On ‘shape and influence’ as the standard in the Tokyo Trial, see Neil Boister & Robert Cryer, The Tokyo International Tribunal: A Reappraisal (Oxford: Oxford University Press, 2008), 151–3.

(41) Above n 21.

(42) American federal law, and the law of some states, has two concepts of conspiracy, one as an inchoate crime, and one as a mode of complicity. The inchoate version was apparently what was contemplated in Article 6(a) of the Nuremberg Charter, above n 21; ‘common plan or conspiracy’ also found its way into the concluding (complicity) words of Article 6 about leaders and organizers and the rest.

(43) See final Discussion paper by the Coordinator on the Crime of Aggression at the Preparatory Commission for the Court, UN Doc. PCNICC//2002/WGCA/RT.1/Rev. 2 (2002). This drafting model had been carried over from the work of the ILC which had not really addressed itself to general part issues and aggression. At this point, it was widely assumed that the aggression definition would stand on its own (later called a ‘monist’ approach) without reference to the general provisions on criminal responsibility in Part 3 of the Statute. In the Rome process from late 1995 onwards there was an understanding that there would be heavy reliance for definitions of responsibility in respect of genocide, war crimes and crime against humanity, on a general part as well as on the provisions on specific crimes, but this was not carried forward into the drafting of the provision on aggression, I think by oversight. Later in the negotiations, it was agreed that the drafting should follow the model of the other three crimes and rely on the general part—a default rule that could be set aside where appropriate. This drafting approach came to be known as the ‘differentiated’ approach.

(44) Para 3bis is added to Article 25, paragraph 3 of the Statute which deals with individual responsibility:

In respect of the crime of aggression, the provisions of this Article shall apply only to persons in a position effectively to exercise control over or to direct the political or military action of a State.

(45) Law Reports of Trials of War Criminals, Selected and Prepared by the United Nations War Crimes Commission. Vol. XIV (1949) at x–xi. Lord Wright added that ‘there was abundant evidence against the accused of peculiarly atrocious offences in the nature of war crimes and crimes against humanity which would have justified the sentence’. The Attachment to the Judgment (above n 7) confirms this.

(46) China was going through a revolutionary upheaval at the time. I am not sure how this legislation was promulgated, but all concerned appear to have accepted it as legitimate under the domestic system at the relevant time. The Tribunal was obviously created after the event, but the Nuremberg and Tokyo decisions would argue that the substantive law already existed, as treaty or customary law, and that there was no ex post facto problem with setting up a mechanism for trial in such circumstances. Some twenty-first century sensitivities would be more squeamish about the military aspects of the trial.

(47) Language of rule as quoted in the report. An Annex to Vol. XIV of the UN War Crimes Commission Reports contains a lengthy discussion of a successor law, Law governing the Trial of War Criminals, of 24 October 1946 (United Nations Day—coincidence or deliberate?). Unfortunately, neither of the actual texts (or a translation thereof) is reproduced and I do not have access to a complete text of either. Thus it is not clear to me whether the definition of crimes (other than those crimes already contained in the Chinese code) had any specific detail or whether it relied simply on the reference to ‘international law’ quoted above, which could be interpreted as incorporating the Nuremberg and Tokyo Charters by reference. The Rules adopted later in 1946, as discussed in the Annex, are quoted as containing essentially the same language but also as having the following content on crimes against peace (not explicitly so labelled):

A person who commits an offence which falls in any one of the following categories shall be considered a war criminal:

  1. (1) . Alien combatants or non-combatants who, prior to or during the war, violate an International Treaty, International Convention or International Guarantee by planning, conspiring for, preparing to start or supporting, an aggression against the Republic of China, or doing the same in an unlawful war.

[There follows a general reference to other crimes: to violations of the ‘Laws and Usages of War’, to assorted crimes that would fit the crimes against humanity category (including a China-specific item of ‘distributing, spreading, or forcing people to consume, narcotic drugs or forcing them to cultivate plants for making such drugs’) and another general reference to ‘Chinese Criminal Law’.]

(48) Note that what Sakai did was mostly done with him located in China (or Hong Kong). Some Japanese aggressors (including many of those tried in Tokyo) never set foot on Chinese soil. As to them, a territorial theory would have to be based on an ‘effects’ or ‘objective territorial’ analysis: above n 3. The 24 October 1946 Law, above n 47, required that a defendant be an ‘alien combatant or non-combatant’ but was silent on where the perpetrators had to be at relevant times. Local collaborators apparently would be (and were) prosecuted under local law. There is no reason in principle why a high-level collaborator, located in and a national of a country that is occupied with his assistance, could not be guilty for, say, aiding and abetting the aggression, but I suppose that treason or some similar crime would always be enough to trigger appropriate punishment.

(49) Report of the inter-sessional meeting of the Special Working Group on the Crime of Aggression, held at the Liechtenstein Institute on Self-Determination, Woodrow Wilson School, at Princeton University, New Jersey, United States, from 21 to 23 June 2004, Doc. ICC-ASP/3/25 (2004) (concluding that relevant Articles of Rome Statute on complementarity could be applied to crimes of aggression without modification).

(50) Rome Statute of the International Criminal Court, Article 17(1). Paragraph 1 concludes with ‘(d) The case is not of sufficient gravity to justify further action by the Court.’ This appears to be an issue of admissibility that, unlike the three preceding subparagraphs quoted above, is distinct from complementarity. Paragraphs 2 and 3 of Article 17 define ‘unwillingness’ and ‘inability’. Article 20(3), on ne bis in idem, referred to in para 1(c) above, precludes trial in the ICC of a person tried in another court for the same conduct, unless the other proceedings were (a) for the purpose of shielding the person concerned from criminal responsibility for crimes within the jurisdiction of the ICC, or (b) were otherwise not conducted independently and impartially or in accordance with the norms of due process recognized by international law and were conducted in a manner which, in the circumstances, was inconsistent with an intent to bring the person concerned to justice. On complementarity in general, see especially Mohamed M. El Zeidy, The Principle of Complementarity in International Criminal Law: Origin, Development and Practice (Boston, MA and Leiden: Martinus Nijhoff, 2008).

(51) Article 19(2)(b), using similar language, permits a challenge to jurisdiction or admissibility by ‘a State which has jurisdiction over a case’. Article 18(1), which deals with preliminary rulings regarding admissibility, uses slightly different language. It requires the Prosecutor at an early stage to ‘notify all States Parties and those States which, taking into account the information available, would normally exercise jurisdiction over the crimes concerned.’ (Emphasis added.) Paragraph 2 of Article 18 states that, within a month ‘a State may inform the Court that it is investigating or has investigated its nationals or others within its jurisdiction with respect to criminal acts which may constitute crimes referred to in Article 5…’ (Emphasis added.) ‘Would normally exercise’ (Article 18(1)) is perhaps a nod in the direction of territorial or nationality jurisdiction; ‘within its jurisdiction’ (Article 18(2)) is neutral.

(52) As a practical matter, absent a successful extradition request (made almost certainly to a third state to which the alleged perpetrator had travelled), a victim state would probably have not only to fend off the aggression but also get its hands on the perpetrators by advancing to the aggressor’s capital or capturing leading military in the field. See Nicolaos Strapatsas, ‘Complementarity and Aggression: A Ticking Time Bomb’, in C. Stahn and L. van den Herik (eds), Future Perspectives on International Criminal Justice (The Hague: TMC Asser Press, 2009), 460. I know not how China obtained custody of Sakai, who seems to have been enjoying his retirement back in Japan from late 1943. Somebody, presumably the occupation authorities, must have handed him over. (His ‘residence’ is described in the Summary Translation and the Judgment as ‘Tokyo’.) At its First Session earlier in 1946, the UN General Assembly adopted GA Res. 3 (I) of 13 February 1946, in which the Assembly:

Recommends that members of the United Nations forthwith take all the necessary measures to cause the arrest of those war criminals who have been responsible for or have taken a consenting part in [war crimes, crimes against humanity and crimes against peace], and to cause them to be sent back to the countries in which their abominable deeds were done, in order that they may be judged and punished according to the laws of those countries.

This language echoes that in the October 1943 ‘Statement on Atrocities’ made by Roosevelt, Churchill and Stalin concerning punishment of responsible Germans. (The drafting was largely Churchill’s.) China participated in the 1943 ‘Four Nation’ meeting in Moscow, but there was no similar statement on Japanese atrocities then. China was also an active member of the United Nations War Crimes Commission which, no doubt, helps explain why Sakai is one of the cases selected for publication. But why this and not others?

(53) This will almost certainly be pursuant to legislation criminalizing aggression in the particular jurisdiction. The decision of the House of Lords in R v Jones [2006] UKHL 16 is suggestive of the way in which national courts approach such problems. Just before hostilities began in the Second Gulf War in 2003, several people entered military bases in the United Kingdom, committing damage in an endeavour to disrupt preparations for war. Charged with offences including criminal damage, aggravated trespass and attempted arson, they sought to justify their actions on the basis that the pending actions by the US and UK Governments would amount to the crime of aggression. They might therefore, they argued, lawfully use force in an attempt to prevent that offence from taking place. The House held that, while aggression was recognized in customary international law as a crime, it was not a crime in English law absent action by the legislature. The power that English courts once had to create common law crimes no longer existed. See Roger S. Clark, ‘Aggression: A Crime Under Domestic law?’ NZLJ, [2006], 349. The House of Lords does not offer any thoughts about universal jurisdiction here—given the procedural stance in Jones, the relevant criminal activities would have been taking place in England. Jurisdiction would have been based on territoriality.

(54) Pål Wrange comments:

At the domestic level, there would be certain difficulties, such as the question of immunities of foreign leaders. Procedural immunity is enjoyed by some types of officials as long as they hold office, and it will prevent a State from prosecuting, even for international crimes (International Court of Justice (ICJ), Arrest Warrant case). While immunity does not apply before the International Criminal Court, the renunciation of immunity inter partes in the Rome Statute probably does not have effect on domestic prosecutions. Domestic prosecutions might also apply national immunities protecting officials from prosecutions before their own courts. Such domestic immunity would however not be a valid excuse to not prosecute cases falling under the Rome Statute.

Pål Wrange, ‘The principle of complementarity under the Rome Statute and its interplay with the crime of aggression’, in summary of Conference on International Criminal Justice held in Turin, Italy, 14–18 May 2007, Doc. ICC-ASP/6/INF.2 at 37 (2007). See also Wrange’s discussion of potential problems of ‘act of State’ and ‘executive privilege’. The reference to the Arrest Warrant or ‘Yerodia’ case is to Case Concerning the Arrest Warrant of 11 April 2000 (D.R. Congo v Belgium), 2002 ICJ, which contains an extensive discussion of the jurisdictional and immunity issues involved in prosecuting foreign leaders. If there is immunity, there is an inability to prosecute nationally and the case should go to the ICC where immunities do not apply (Article 27 of the Rome Statute). Such problems afflict prosecutions in aggressor states, victim states and universal state jurisdictions, but perhaps not equally. In the aggressor state, it is at least possible that the immunity problems will have been removed in Rome Statute implementation legislation. That is, however, (absent a change of regime) the place where prosecution is least likely to occur. In the victim state or a third state, there will be the problem of whether one formerly eligible for immunity ratione personae as a head of state or senior official continues to have immunity ratione materiae after leaving office for involving the state in an act of aggression. The Pinochet case in the UK stands as authority for the proposition that there are some activities like torture that are not part of a leader’s function for the purposes of producing immunity. It is not clear whether this is consistent with dicta of the ICJ in the Yerodia Case and with the direction in which the ILC is proceeding Immunity of State officials from Foreign Criminal Jurisdiction. See generally, Roman A. Kolodkin, Special Rapporteur, Second Report on Immunity of State Officials from Foreign Criminal Jurisdiction, UN Doc. A/CN.4/631 (2010). These authorities suggest that at least some examples of immunity ratione materiae continue to apply after the actor leaves office. Sakai made a plea based on superior orders but does not seem to have argued that he had immunity (either before or after his 1943 retirement).

(55) The SWGCA discussed theories of jurisdiction not in the context of Article 17 but in the context of Article 12 of the Rome Statute. Article 12 requires, as a ‘precondition’ to the Court’s exercise of jurisdiction, that either the territorial state or the state of nationality be a party to the Statute. What of a situation where a victim state (where the effects of the aggression were apparent) is a party but the aggressor state (where all the preparations took place) is not? There was wide support in the Special Working Group for the proposition that the victim state’s acceptance should be sufficient to meet he requirements of Article 12. One of the SWGCA’s Reports commented:

Given that the conduct of a leader responsible for the crime of aggression would typically occur on the territory of the aggressor States, the question was raised whether the crime could also be considered to be committed where its consequences were felt, namely on the territory of the victim State. The answer to that question had important consequences for the application of article 12, paragraph 2 (a), which linked the Court’s jurisdiction to ‘the State on the territory of which the conduct in question occurred’. Broad support was expressed for the view that concurrent jurisdiction arises where the perpetrator acts in one State and the consequences are felt in another, while some delegations required more time to consider the issue. While some delegations expressed the possible need for clarifying language, possibly in the elements of crime, several stated that the Rome Statute was sufficiently clear and that ‘over-legislating’ should be avoided. The reference to ‘conduct’ in article 12 encompassed also the consequences of the conduct. The decision in the Lotus case supported this reasoning.

Report of the Special Working Group on the Crime of Aggression, ICC-ASP/7/SWGCA/1 (26 November 2008). Compare possible different inferences from the work of the ILC, below n 77. The reference to Lotus is, of course, a reference to The SS Lotus (France v Turkey), (1927) PCIJ, Ser. A, No. 10, at 4 (negligence on board French ship leading to deaths on Turkish ship—Turkey has jurisdiction, concurrent with France, at least on basis of the effects on Turkish territory but perhaps not on a passive personality theory). The Special Working Group’s result was rejected, although the logic of its reasoning was perhaps underscored in Article 15bis (5) of the Kampala Amendments, above n 1. It asserts that ‘[i]‌n respect of a State that is not party to this Statute, the Court shall not exercise its jurisdiction over the crime of aggression when committed by that State’s nationals or on its territory.’ The implications of Lotus are avoided by the treaty language.

(56) The issue does not appear to be discussed in El Zeidy, above n 50, the leading work on complementarity. Nor is it mentioned in the account of the negotiations by John Holmes, the Canadian diplomat who chaired the Article 17 discussions at the Preparatory Committee in New York and then in Rome, John Holmes, ‘The Principle of Complementarity’, in Roy Lee (ed), The Making of the Rome Statute, Issues, Negotiations, Results (The Hague: Kluwer Law International, 1999), 41. His main concern in the negotiations was that states would need to legislate in order to ‘have’ jurisdiction. He says, at 67:

It was not enough that a State had instituted national proceedings, it must establish to the Court that it had jurisdiction in such a case. This addition was intended to forestall situations where a State could challenge (and delay) the Court from proceedings on the ground that it was investigating when in fact the investigation was sure to fail because the State lacked jurisdiction even as far as its own courts were concerned.

In a later book chapter, Holmes comments:

Of course, in reality there is a need for the ICC, since States may be unwilling to exercise jurisdiction over international crimes, despite a duty to do so, especially when the nexus between the State and the crime is limited. Universal jurisdiction exists with respect to many of the crimes included in the Statute, but States have been reluctant to exercise it.

J.T. Holmes, ‘Complementarity: National Courts versus the ICC’, in A. Cassese, P. Gaeta & J.R.W.D. Jones (eds), The Rome Statute of the International Criminal Court: A Commentary (Oxford: Oxford University Press, 2002). This can be read as assuming that complementarity applies to universal jurisdiction, but as saying nothing about whether aggression is one of the ‘many’ over which there is universal jurisdiction.

(57) Sharon A. Williams and William A. Schabas, ‘Article 17, Issues of Admissibility’, in Otto Triffterer (ed), Commentary on the Rome Statute of the International Criminal Court—Observers’ Notes, Article by Article (Oxford: Hart Publishing, 2nd edn, 2008), 605.

(58) Daniel D. Ntanda Nsereko, ‘Article 18, Preliminary Rulings regarding Admissibility’, Triffterer, above n 57, 627. Judge Nsereko’s discussion of the notification requirement in Article 18(1) underscores that complementarity issues can arise both in respect of state parties to the Statute and non-parties.

(59) The issue here is who might make a challenge to admissibility. Can a ‘volunteer’ make such a claim? It will be noted that there are actually two issues not resolved by the language of Article 19. Does a volunteer have ‘standing’ to raise the admissibility issue and insist that it wants to proceed; what are the criteria by which the Court will determine whether to accede to its claim as a matter of ‘substance’ or not?

(60) Christopher K. Hall, ‘Challenges to the jurisdiction of the Court or the admissibility of a case’, above n 57, 649. Hall’s references are perhaps telling. He writes:

There have been three comprehensive global studies that have described state practice concerning universal jurisdiction over crimes under international law.

See ‘Harvard Research in International Law,’ Am. J. Int’l L. Supp., 29 (1935), 435; Amnesty International, Universal Jurisdiction: The Duty of States to Enact and Implement Legislation (Sep. 2001, AI Index: IOR 53/022—018/2001) (Second Edition forthcoming); J.-M. Henckaerts/L. Doswald-Beck, Customary International Humanitarian Law, Vol. 1, 604–7; Vol. 2, 3383–940 (2005).

None of these sources deals with the crime of aggression—while at least the later ones support universal jurisdiction for breaches of humanitarian law, it is not clear how far they go to support Hall’s proposition that all states ‘may exercise jurisdiction over the crimes within the Court’s jurisdiction’. Hall was one of the authors of the Amnesty International study that he cites; it contains nothing indicating that there is universal jurisdiction over aggression—the word is barely mentioned and then only in passing. Amnesty was, indeed, one of a handful of major non-governmental organizations (NGOs) that was singularly unsupportive of the effort to complete the left-over mandate from Rome on aggression.

(61) Darryl Robinson, ‘The Mysterious Mysteriousness of Complementarity,’ Crim. L. Forum, 21 (2010), 67. Robinson is at pains to correct a (common) misconception of Article 17, namely that the main issue is whether a state is ‘able and willing’. From the plain meaning of Article 17, it is apparent that the basic rule is that a case is admissible whenever there are no domestic proceedings taking place. It is only when there are such proceedings that it is necessary to go to the second prong, that is whether the state in question is able and willing to investigate or prosecute genuinely. Many commentators see the ‘able and willing issue’ as the primary one, rather than the ‘is being’ or ‘has been’ one.

(62) Robinson, above n 61, 68.

(63) Robinson, above n 61, 97. This excellent hypothetical was floated earlier, but without the reference to universal jurisdiction, in a 2003 Informal Expert Paper, The Principle of Complementarity in Practice, produced for the ICC Office of the Prosecutor by a distinguished group coordinated by Professor Robinson, found at <http://www.icc-cpi.int/iccdocs/doc/doc654724.pdf> (accessed 25 February 2013).

(64) Above n 60.

(65) Above n 61.

(66) Podgor and Clark, above n 3, 160–1; R. v Jones, above n 53.

(67) Something could, in principle, be a crime under international law without that necessarily carrying with it a right to exercise universal jurisdiction. It may entail simply an obligation to penalize it at the national level or only a right to do so. The sixth preambular paragraph to the Rome Statute of the International Criminal Court reads:

Recalling that it is the duty of every State to exercise its criminal jurisdiction over those responsible for international crimes,…

Writing shortly after Rome, Slade and Clark said this about the paragraph:

It is delightfully ambiguous. Does the State obligation of which it speaks ‘to exercise its criminal jurisdiction over those responsible for international crimes’ refer to jurisdiction over crimes within the territorial ‘jurisdiction’ of the State? Or is it referring to a much broader ‘jurisdiction’ of the ‘universal’ kind, regardless of where the events occurred? In this respect, perhaps, it perhaps reflects the ambiguity of some negotiators about whether the International Court should derive its jurisdiction on some theory of universal jurisdiction or whether it should be narrower, based on the consent of the territorial State or the State of nationality.

Tuiloma Neroni Slade and Roger S. Clark, ‘Preamble and Final Clauses’, in Lee, above n 56, 427.

(68) Princeton Project on Universal Jurisdiction, The Princeton Principles on Universal Jurisdiction (2001), principle 1(2). Principle 3 asserts that with respect to serious crimes under international law ‘national judicial organs may rely on universal jurisdiction even if their national legislation does not specifically provide for it’. There appears to be no existing example of such reliance. If Sakai turns on a universal theory, the tribunal has its empowering legislation to rely upon.

(69) Princeton Project on Universal Jurisdiction, above n 68, principle 2(1).

(70) I could find no reference to aggression in the most important recent study of universal jurisdiction, Luc Reydams, Universal Jurisdiction: International and Municipal Legal Perspectives (Oxford: Oxford University Press, 2003). Cedric Ryngaert, Jurisdiction in International Law (Oxford: Oxford University Press, 2008) has a very erudite discussion of the universality principle at 100–26 in which he discusses numerous candidates for universal jurisdiction, especially over what he calls ‘core crimes’—but aggression does not even make it into the discussion. See also Holmes’s opinions, above n 56.

(71) Above n 16.

(72) Astrid Reisinger Coracini, ‘Evaluating domestic legislation on the customary crime of aggression under the Rome Statute’s complementarity regime,’ in Carsten Stahn and Göran Sluiter (eds), The Emerging Practice of the International Criminal Court (Boston, MA and Leiden: Martinus Nijhoff, 2008), 734 (footnote omitted).

(73) In a note to the author, Dr. Reisinger Coracini said that almost all those with legislation espoused at minimum a territorial theory of jurisdiction. She added that universal jurisdiction appeared to be asserted by, at least, Bulgaria, Croatia, Hungary (but in respect of incitement only) and Moldova. Others, like Armenia, Bosnia, Estonia, Georgia and Kazakhstan, adopt some kind of passive personality or protective principle which supports victim state jurisdiction.

(74) Draft Code of Crimes against the Peace and Security of Mankind, Yearbook of the International Law Commission, 1996, Vol. II (Part Two). For some history of the Draft Code, see Timothy L.H. McCormack & Gerry J. Simpson, ‘The International Law Commission’s Draft Code of Crimes against the Peace and Security of Mankind: An Appraisal of the Substantive Provisions,’ Criminal Law Forum, 5 (1994), 1; Rosemary Rayfuse, ‘The Draft Code of Crimes against the Peace and Security of Mankind: Eating Disorders at the International Law Commission,’ Criminal Law Forum, 8 (1997), 43. The exclusion of universal jurisdiction over aggression is especially telling in the context of the ILC’s concurrent work on state responsibility. In its commentary on the obligations of states arising from a breach of a peremptory norm, the Commission notes widespread agreement that peremptory norms (jus cogens) extend at least to the prohibitions of aggression, slavery and the slave trade, genocide, racial discrimination and apartheid. See James Crawford, The International Law Commission’s Articles on State Responsibility (Cambridge: Cambridge University Press, 2002), 284. Jus cogens norms are often equated with those over which there is universal criminal jurisdiction for the individuals responsible, but the ILC, certainly in relation to aggression, did not make that precise connection.

(75) The 1996 Draft Code makes no serious attempt to ‘define’ in any detail either the specific offences or the general part of the Code.

(76) The 1996 Draft Code, Article 8. See also Article 9, which proceeds on a similar basis of distinction: ‘Without prejudice to the jurisdiction of an international criminal court, the State Party in the territory of which an individual alleged to have committed a crime set out in article 17, 18, 19 or 20 is found shall extradite or prosecute that individual.’ The discussion of these Articles by the ILC in Yearbook of the International Law Commission, 1996, Vol. I at 49–53, supports the present author’s caution about universal jurisdiction for aggression.

(77) Cf. the comments in the Special Working Group on the Crime of Aggression, above n 55. The matter was not without controversy in the ILC. In the final discussion of the jurisdictional issues there, the Special Rapporteur on the Draft Code commented:

[T]‌he text clearly raised a number of problems. Mr Kabatsi had proposed that a court of the country of the author of the crime of aggression should be considered competent. In that case, why not a court in the victim country? Given the large number of proposals made, the Commission was in danger of adopting an unsatisfactory provision if it was too hasty in its decisions. A small informal group should look into the question.

Yearbook of the International Law Commission, 1996, Vol. I, at 50 (summary records). Nothing further happened on the record other than the final adoption of the text as described above. Strapatsas, above n 52, 454–455, points out that there is some post-World War II practice of states, such as Australia, China, Denmark, Greece, Poland and the United States (with regard to the Pacific theatre of war) claiming jurisdiction on the basis of being a victim state and, in the case of China [Sakai] and Poland [Greiser and Koch], actually exercising it. He gives short shrift to the argument that there is no victim state jurisdiction. (See also the practice of the USSR in respect of German generals, above n 16.)

(78) Understandings to Kampala Amendments, above n 1.

(79) In arguing in Kampala for such an understanding, Harold Koh, Legal Adviser to the US State Department, asserted in a Statement dated 4 June 2010:

Even if states incorporate an acceptable definition into their domestic law, it is not clear whether or when it is appropriate for one state to bring its neighbour’s leaders before its domestic courts for the crime of aggression. Such domestic prosecutions would not be subject to any of the filters under consideration here, and would ask the domestic courts of one country to sit in judgment upon the state acts of other countries in a manner highly unlikely to promote peace and security.

(80) It is certainly possible to conceive of aggression being committed by a non-state entity (a terrorist organization, or a dissident group that spreads across borders, for example) but that is not what was prosecuted at Nuremberg and Tokyo and in Nanking. Nor is it what is contemplated in the Kampala Amendments. Other legal categories of crime (terrorism, war crimes, crimes against humanity) are necessary to capture such activities. Professor Larry May, whose path-breaking book suggests a different conceptualization in general from that adopted in Kampala, argues forcefully that ‘some terrorist groups can wage war and…terrorist leaders can be prosecuted for waging aggressive war as long as such prosecutions are subject to the rule of law’: Larry May, Aggression and Crimes against Peace (Cambridge: Cambridge University Press, 2008), 297. Given that the ICC will not have jurisdiction in such cases, at least in the foreseeable future, should such developments be encouraged by (another) treaty? By unilateral practice?

(81) Above n 30.