State Responsibility for Genocide

‘Crimes against international law are committed by men, not by abstract entities, and only by punishing individuals who commit such crimes can the provisions of international law be enforced.’

Nuremberg Judgment, at 41

In the past years international law has made strides in establishing individual responsibility for crimes against international law as one of its most fundamental principles. This year, however, provides us with the first opportunity for adjudication on state responsibility for genocide in the case brought before the International Court of Justice by Bosnia and Herzegovina against Serbia and Montenegro. This article attempts to provide a methodological approach for deciding the many issues raised by this case, mainly by focusing on a strict separation between primary and secondary rules of international law, with this approach being both theoretically and practically desirable. The article also deals with the question of state responsibility for acts of non-state actors on the basis of state de facto control. By applying this general methodology to the facts of the Genocide case, the article will show that the principal difficulties the Court will face if it decides to use this approach will not be in applying the relevant substantive law, but in establishing the facts and assessing the available evidence.

1Introduction

This year truly marks a milestone in the development of international law: a sovereign state will for the first time in history be standing trial for genocide before the International Court of Justice.¹ Despite numerous procedural and practical difficulties, the ICJ seems likely to decide on the merits of the Genocide case² brought before the Court 13 years ago by Bosnia and Herzegovina³ against Serbia and Montenegro,⁴ on the latter’s responsibility for genocide committed during the Bosnian conflict following the break-up of the former Socialist Federal Republic of Yugoslavia (SFRY).

Historically, international law was concerned only with actions of states, while the individuals through whom states acted remained almost entirely outside its purview. Yet, since the end of the Cold War the international legal system has seen a resurgence of the driving ideas behind the Nuremberg and Tokyo trials – that of establishing individual criminal responsibility for grievous atrocities and mass violations of human rights and the laws of armed conflict, under international law, and before international courts, if municipal judicial systems are unwilling or unable to prosecute the offenders. The establishment of the ad hoc international criminal tribunals for the former Yugoslavia (ICTY) and Rwanda (ICTR) by the UN Security Council and the development of hybrid, internationalized criminal tribunals such as the one for Sierra Leone finally culminated in the creation of a new, permanent International Criminal Court (ICC).⁵ Yet, all these developments of the past decade dealt with individual and not state responsibility for international crimes. Indeed, it was precisely the need to remove from culpable individuals the protective shield of the state which caused the development of individual responsibility in international law,⁶ as is shown by the quotation from the Nuremberg judgment at the beginning of this article. This article will endeavour to show that this quotation does not exhaust the inquiry into the responsibility of states for genocide and other international crimes in contemporary law, nor would such responsibility of states debar or diminish the criminal responsibility of individuals.

After World War II, Germany and Italy did pay some compensation to the victims of Nazi and fascist atrocities, but this only happened after a long and arduous process.⁷ Furthermore, what was mainly in dispute were the modalities and amounts of reparations, not the responsibility for atrocities as such. The importance of the Bosnia v. Serbia case is simply that it provides us with the first opportunity for international adjudication of state responsibility for the crime of genocide. The judges of the ICJ will therefore have to deal with many complex issues of international law, and their judgment will have far-reaching consequences. But it is also important to note that the ICJ is neither the sole, nor the primary, mechanism of establishing state responsibility. Of all the situations in international relations involving the legality of the use of force, for example, the ICJ has only dealt with very, very few, and it may well be that it will decide on an equally minuscule proportion of cases when it comes to state responsibility for genocide. It must also be noted, however, that the importance of the ICJ’s jurisprudence lies not in the number of cases it decides, but in the principles of international law and the basic legal reasoning which it sets out in its judgments, which can then be applied outside the halls of the Peace Palace itself.

The purpose of this article is to provide some analysis of the broader questions on which the current Genocide case will turn, and which will probably recur in any future cases of state responsibility for genocide. My principal thesis is that applying a proper methodology of state responsibility, and particularly maintaining the distinction between primary and secondary rules of international law, will show that many of the more controversial legal questions are actually easily resolved, while the principal problems international courts will face in adjudicating state responsibility for genocide will be of a factual, evidentiary nature.

The chief purpose of this article is therefore methodological, as its goal is to develop a structure which should be followed in any case involving a state’s responsibility for genocide, whether its responsibility is analysed by an international lawyer working in an international court or other dispute settlement body, or in a foreign ministry, or merely when writing a journal article, as I am doing now. Following the approach outlined below will provide some legal clarity to what may be the most emotionally and morally demanding task in international law. Obtaining some factual clarity, however, is a very different challenge, as is, for that matter, overcoming the enormous practical difficulties which stand in the way of a state’s responsibility for genocide actually being invoked, even if the facts of genocide are clear for all to see. This is particularly true for any sort of collective, institutional reaction of the international community to the commission of genocide, which is additional to the mechanisms of state responsibility. The realities of international relations being what they are, it is far too easy to accuse states of utter hypocrisy and simply leave it at that – the example of some members of the United Nations Security Council who because of their selfish interests fail to put an end to the ongoing atrocities in the Sudan comes to mind – when it is only us, the ordinary people, who can through our action compel states to do the right thing. Only when, and if, the overwhelming majority of human beings make a final, moral choice that human society is to root out the very idea of genocide and be prepared to do whatever it takes to achieve this goal will genocide become a thing of the past. To this fundamental moral question international law can provide no definitive answers, for all its role as ‘the gentle civilizer of nations’.⁸

That being said, this article will not be a moral tractate, nor will it try to answer the question of why or why not states invoke the responsibility of another state for genocide from the perspective of international relations. The question I try to answer is not how to shame states into action, but how to lay down a methodological foundation that an international lawyer should follow when states finally do act and a state’s responsibility for genocide is actually invoked. To that effect, Section 2 of the article will very briefly deal with the concept of genocide in international law and its differentiation from other international crimes. Section 3 will examine several specific problems genocide poses for the international law of state responsibility. Section 4 will deal with the issue of state responsibility for acts of non-state actors, an issue which is not entirely specific to genocide but which will be central in the Bosnia v. Serbia case, and probably in future cases as well. Section 5 will provide an analysis of the Genocide case itself.

2Concept of Genocide in International Law

The word ‘genocide’, coined by the Polish Jewish lawyer Raphael Lemkin,⁹ immediately invokes a set of memories of the horrors of the Holocaust and World War II and the methodical, organized extermination of millions of human beings. The moral and social stigma carried by this word is one of the primary reasons why states and other political actors use it in international discourse, or more often desperately try not to use it.¹⁰ However, there is a marked difference between the ordinary, lay meaning of the word ‘genocide’, or even the concept of genocide in anthropology or other social sciences, and the legal concept of genocide, or better yet, the concept of genocide in international law.¹¹ To the average lay person, genocide is any organized, planned mass murder of human beings on account of their race, ethnicity, religion or other personal characteristic. Yet, to the international lawyer genocide is a term of art, with a meaning strictly defined by Article II of the 1948 Genocide Convention, and reflected in customary law:¹²

[G]enocide means any of the following acts committed with intent to destroy, in whole or in part, a national, ethnical, racial or religious group, as such:

• Killing members of the group;

• Causing serious bodily or mental harm to members of the group;

• Deliberately inflicting on the group conditions of life calculated to bring about its physical destruction in whole or in part;

• Imposing measures intended to prevent births within the group;

• Forcibly transferring children of the group to another group.

The five objective acts which comprise the actus reus of genocide must not only be committed with intent, but with a specific intent to destroy, in whole or in part, a protected group as such in order for them to constitute genocide. Genocide is not simply a crime against individuals because of their membership in a group, but an attack on the existence of protected human groups.¹³ The list of protected groups is also exhaustive – the destruction of political or other social groups,¹⁴ for instance, is not genocide according to international law.¹⁵ This has been criticized by some authors as the ‘blind spot’ of the Genocide Convention,¹⁶ which somehow needs to be rectified through customary law.¹⁷ While this is a proposition which might have made some sense 60 years ago, as genocide was the only international crime which did not require a nexus with an armed conflict, it is completely beside the point under modern international law, as the destruction of other human groups and collectives not covered by the Genocide Convention simply qualifies as murder or persecution-type crimes against humanity which are also of a jus cogens nature and are punishable under international criminal law.¹⁸ It must also be emphasized in the strongest terms that genocide per se is not a worse crime than crimes against humanity, as is best shown by the Khmer Rouge atrocities¹⁹ against the members of their own ethnic group and the fruitless and legally pointless attempts to stretch the legal definition of genocide to encompass the Cambodian killing fields.²⁰

Genocide and crimes against humanity are simply different as a matter of international law, while the moral condemnation every sane person must attach to such atrocities should not depend on the outcome of legal academic debates. An international court, however, should not adopt either an expansive or a restrictive definition of genocide, but the legally correct one,²¹ which is in conformity both with the text and the preparatory work of the Genocide Convention and the jurisprudence of international tribunals.²² There would otherwise be no point in making the distinction in the first place, yet this is exactly what states did when they adopted the Convention in 1948. It is the extreme mens rea of genocide which draws the distinction between genocide and crimes of humanity as a genus; murder, extermination or deportation are all crimes against humanity when committed as part of a widespread or systematic attack directed against any civilian population and with knowledge of the attack.²³ They only become genocide if the perpetrator commits them with the intention of physically or biologically²⁴ destroying a protected group, in whole or in part.²⁵

To sum up, the focus of the law on the criminal state of mind and genocidal intent of the perpetrator and not the outward manifestations of his or her behaviour such as organization, planning or number of deaths make genocide legally distinct from crimes against humanity. Such a high degree of required intent carries great probative difficulties – but these probative difficulties can be overcome precisely by drawing inferences from concrete, objective circumstances.²⁶

Genocidal intent should also be clearly distinguished from motive. Genocide, like any other crime, can be committed with a variety of motives, ranging from ethnic or religious hatred, revenge or fear to the acquisition of material gain, power or territory. Motive, as generally in criminal law, will serve a probative purpose, and is not by itself an element of the crime of genocide. Acts of genocide, such as murder, can be committed with the motive of establishing control over a territory, as can the acts of persecution-type crimes against humanity. But these acts only become genocide if they are committed with the specific intent of destroying the group as such, not displacing it. The specific intent of genocide which differentiates it from crimes against humanity essentially goes to the method (destruction as opposed to displacement of a group) of bringing about the goal or motive, which may be common to both types of crimes.²⁷

Proving genocide and distinguishing between genocide and crimes against humanity will be a major part of the Bosnia v. Serbia case before the ICJ, as Bosnia argues that the totality of all crimes committed during the conflict amounts to genocide, while Serbia claims that only crimes against humanity and war crimes were committed during the war, and not genocide. The specific issues will be dealt with in more detail below, in Section 5.

3Specific Problems Genocide Poses to the Methodology of State Responsibility

AState Responsibility for a Crime

1Basic Methodology of State Responsibility

One of the primary contributions of the codification effort of the International Law Commission to developing a methodology of state responsibility is its separation of the primary rules of substantive international law from the secondary rules of state responsibility, and its focus on the latter.²⁸ The pre-World War II rules of state responsibility were intimately connected to issues surrounding diplomatic protection, state treatment of foreign nationals and the development of the ‘international minimum standard’,²⁹ generally detested in the developing world.³⁰ The ILC’s codification project of the law on state responsibility lasted for decades and went through several iterations, finally culminating in the 2001 Draft Articles on Responsibility of States for Internationally Wrongful Acts.³¹ The ILC did its best to formulate a separate body of rules which are applicable to a wide variety of situations, and it therefore produced the present Articles, which are logically structured in such a way as to remind us of a civil code in a European legal system. Roberto Ago, one of the ILC’s most influential Special Rapporteurs on state responsibility, saw the Draft Articles as specifying:

[T]he principles which govern the responsibility of States for internationally wrongful acts, maintaining a strict distinction between this task and the task of defining the rules that place obligations on States, the violation of which may generate responsibility . . . [I]t is one thing to define a rule and the content of the obligation it imposes, and another to determine whether that obligation has been violated and what should be the consequences of the violation.³²

By creating this rather bland piece of codification, the ILC has managed to relegate extremely controversial issues to the area of primary rules, and thereby achieve both a wider acceptance of the Articles and a sound methodological structure.³³ Some of the more delicate questions of state responsibility, such as whether fault or damages comprise a necessary element of responsibility, have also been relegated to the area of primary rules.³⁴ Even though this solution is not entirely accepted by some commentators,³⁵ it was in my view a very wise decision. The matter of fault is especially pertinent when state responsibility for genocide is at issue, as the very existence of genocide depends on subjective elements. The basic postulate of analysis in this article is that state responsibility is per se neither strict nor subjective as to any required element of fault by a state: the precise nature of the responsibility, and the required degree of culpability, be it specific intent, negligence or knowledge, purely depend on the primary rules. It is quite possible to disagree with this basic starting position adopted by the ILC, as state practice can be interpreted in various ways. The ILC’s work is certainly not gospel and its authority, as well as that of the ICJ for that matter, does not place it or the ICJ beyond criticism. The basic distinction between primary and secondary rules is but one starting point – however appealing it is to this author, it has been criticized, and in a very powerful way.³⁶ It is also quite easy to just pay lip service to this distinction, and thus fail to draw all of the consequent methodological conclusions. Nevertheless, this article will show that it is precisely the question of state responsibility for genocide which demonstrates both the theoretical and the practical wisdom of adopting such an approach.

I firmly believe that maintaining, as much as possible, a distinction between primary and secondary rules is the only way in which we can preserve a semblance of methodological sanity, and preserve the concept of the law of state responsibility as a body of general rules, applicable in all areas except those regulated in whole or in part by a special regime.³⁷ As we shall see, the secondary rules of state responsibility actually do not bring much specifically regarding genocide; on the contrary, it is the primary rules on genocide, and the interaction of these rules with the ones regulating state responsibility which make the whole matter exceedingly complicated. It is in the area of primary rules where we will find answers on issues such as intent or burden of proof, while the purpose of the secondary rules is mainly to define a proper standard for attribution, which would apply even in cases not involving genocide.

2A Special Regime?

As is evident from the previous exposition on genocide, its definitional elements reside primarily in the area of international criminal law. Yet, while it is a crime for the existence of which there must be a criminally responsible individual, genocide is also an internationally wrongful act, in the sense that its commission can give rise to state responsibility under the customary rules of attribution of individual acts to a state. If the proper standard of attribution is fulfilled, individual and state responsibility run concurrently, as both a state and an individual could be held responsible for a single act under international law.³⁸ However, the purposes served by these two types of responsibility would be different – the criminal responsibility of an individual would serve the general goals of punishment, deterrence and prevention, while state responsibility would be remedial and reparatory, not punitive.³⁹

The question of state responsibility for an international crime became extremely controversial during the ILC’s codification process. The previous draft of the Articles, specifically Draft Article 19,⁴⁰ envisaged a more robust system and laid down a fundamental distinction between ordinary international delicts and international crimes of states. The exact scope and consequences of this distinction have been a subject of major contention within the ILC itself, with some members, such as Special Rapporteur Gaetano Arangio-Ruiz proposing an even stronger special régime,⁴¹ while others, such as James Crawford, the last ILC Special Rapporteur on state responsibility, were vehemently opposed to the whole idea,⁴² as were a number of influential states⁴³ and controversy also raged outside the chambers of the ILC itself.⁴⁴ A similar debate took place during the drafting process of the Genocide Convention, with states having widely diverging views on the nature of state responsibility for genocide.⁴⁵

So, can a state commit a crime? Yes and no. A state can do nothing by itself – it can only act through individuals, who would in the overwhelming majority of cases be its de jure organs. But just as individuals can commit international crimes when acting in their official capacity,⁴⁶ so can their criminal acts be attributed to a state. Yet, this state responsibility for an international crime is by its nature not criminal, but remains ‘civil’ (however inapt that term might be), i.e. ‘normal’ state responsibility,⁴⁷ despite some authors’ advocacy to the contrary.⁴⁸ This is clear both from the ILC’s rejection of the concept of state crimes and (arguably) the possibility of awarding punitive damages during the second reading of the Draft Articles and from the total lack of any state practice or opinio juris which would indicate acceptance of the concept of the criminal responsibility of a state.

What genocide, as an emblematic serious breach of an obligation under a peremptory norm of international law, does entail is an aggravated regime of state responsibility, according to Articles 40 and 41 of the final ILC Articles, a heavily watered down version of the previous draft. The purpose of this aggravated regime is mainly to make a statement – that there are some obligations under international law which are simply fundamental, as they enshrine universal, communitarian values.⁴⁹ Unfortunately, the aggravated regime of state responsibility for violations of jus cogens obligations actually brings very little substance to our present discussion, as it deals more with obligations of third states in responding to a violation⁵⁰ than with the responsibility of the state actually committing the breach.⁵¹ It should nevertheless be noted that the conception of the aggravated regime adopted in the final ILC Articles is not simply a terminological substitute for state criminality, as to make it palatable to states and sceptical authors: it is a fundamental rejection of the notion of a different kind of responsibility of states, whose legal nature remains the same even if the norms violated form the very firmament of the international legal order. Other consequences, such as collective enforcement action, may of course attach to violations of these obligations, and the values enshrined in them are certainly more important, but the secondary rules of state responsibility make a single, uniform system.

Naturally, the usefulness of mechanisms of state responsibility is limited when a state is perpetrating genocide against its own citizens or within its boundaries, and not against the people of another state. Many of the historical examples of genocide were precisely those of a state committing atrocities against its own nationals, such as the massacre of its Armenian population by Turkey during World War I, the 1994 genocide in Rwanda or the ongoing massacres in Darfur, Sudan.⁵² However, the prohibition of genocide, as all peremptory norms, is undoubtedly also an obligation erga omnes, owed to the international community as a whole.⁵³ State obligations under the Genocide Convention are in essence the same in nature as those arising from (other) human rights treaties: even though the primary beneficiaries of the state obligations are individuals, these obligations still run between the state parties to the Genocide Convention.⁵⁴ Accordingly, states which are not themselves directly injured can still invoke the responsibility of a state committing genocide against its own people, and can in all likelihood do so even before the ICJ,⁵⁵ according to the principles on invoking responsibility for violations of obligations erga omnes.⁵⁶ It would still, however, be necessary to establish a jurisdictional basis for any proceedings before the ICJ, as the fact that the ICJ would be called upon to decide on a matter involving jus cogens and erga omnes obligations does not obviate the need for consent by the state parties to the ICJ’s jurisdiction.⁵⁷ What erga omnes obligations do establish is the legal interest of a state, even if it is not directly injured, to invoke the responsibility of another state, whether before the ICJ or in some other setting.

Non-injured states could claim from the responsible state, inter alia, the cessation of the wrongful act and the specific performance of its international obligation, and could even ask the Court to issue provisional measures, however unlikely it may be that a genocidal state would actually abide by such measures.⁵⁸ Non-injured states cannot claim compensation as a form of reparation to themselves, as no damage was caused to them, but they might be able to claim compensation on behalf of the victims of the responsible state, possibly by setting up a compensation fund for which the non-injured state could act as a trustee. Article 48(2)(b) of the ILC Articles does seem to allow for a claim by a non-injured state for reparations in the interest of the beneficiaries of the erga omnes obligation, but this was admittedly done in progressive development of the law.⁵⁹ Such a development has also been supported by commentators and the Institut de droit international,⁶⁰ but it remains to be seen whether it will be accepted in state practice or before international dispute settlement bodies, as it certainly should.

3State Responsibility Under Article IX of the Genocide Convention

A state’s responsibility for other international crimes and mass human rights violations, which are defined by treaty and customary law can come before the ICJ through the mechanism of the optional clause under Article 36 para. 2 of the ICJ Statute or a specific compromisory clause, and these issues are usually a part of a wider dispute, typically involving the legality of the use of force. Such was the case, for example, in Nicaragua v. United States,⁶¹ which is primarily a case on self-defence and other aspects of the jus ad bellum, while the issue of the alleged responsibility of the US for the violations of international humanitarian law by the contras is dealt with rather incidentally. The recently decided Congo v. Uganda⁶² case also mostly deals with use of force matters, while violations of human rights and humanitarian law have a more summary treatment.⁶³ The Genocide case will in that sense be the ICJ’s first true human rights case.

State responsibility for genocide can also be invoked under customary law, either before the ICJ or in some other forum, but genocide is unique in that Article IX of the Genocide Convention explicitly confers upon the ICJ jurisdiction to adjudicate ‘[d]isputes between the Contracting Parties relating to the interpretation, application or fulfillment of the present Convention, including those relating to the responsibility of a State for genocide or for any of the other acts enumerated in article III.’ In the current Genocide case pending before the ICJ, the Court has found that it has jurisdiction only pursuant to Article IX.⁶⁴ Serbia raised several issues as to the scope and nature of state responsibility under the Convention, arguing that the Convention contemplates only state responsibility for a state’s failure to perform its obligations under Articles V, VI and VII⁶⁵ of the Convention, and not responsibility for the actual commission of genocide. The Court rejected this argument in the preliminary objections stage, finding that the Convention does not exclude any type of state responsibility.⁶⁶ A similar argument was raised during the oral hearings on the merits by Professor Brownlie, arguing on behalf of Serbia, who also stated, for example, that the general law of state responsibility is not applicable in a dispute under Article IX of the Convention.⁶⁷ The Court should emphatically reject this argument. It is true that the preparatory work of the Convention shows extreme confusion among the states participating in the drafting process on the matter of state responsibility for genocide.⁶⁸ But it is also true that in the end states have used very clear language in Article IX, which gives the ICJ jurisdiction over disputes ‘including those relating to State responsibility for genocide’. Any ambiguities in the travaux – which is actually not simply ambiguous but is in a state of near-total confusion⁶⁹ – are overridden by the plain meaning of the text of the Convention, while bearing in mind that recourse to the travaux is justified in the first place only when the ordinary meaning of the text as established by primary means of interpretation is ambiguous.⁷⁰ It is clear, therefore, that as a state can be responsible for the commission of genocide under general international law, so it can be held responsible before the ICJ under Article IX of the Convention. Additionally, general principles of international law such as those dealing with treaty interpretation, state responsibility or the concept of obligations erga omnes can be applied by the ICJ in any case involving the responsibility of a state for breaches of its treaty obligations – the Genocide Convention does not set up a special, fragmented legal regime, which would be out of touch with the development of general international law.

It is also important to determine whether the primary rules for the purpose of establishing state responsibility for genocide can only be found in the Genocide Convention. I would submit that it is necessary for the ICJ to have recourse not only to the Convention itself and, if needed, its travaux préparatoires, but also to general principles of international criminal law when interpreting the Convention. These principles would help the Court, as we will later see, in defining concepts such as complicity or incitement, and would also draw light on issues such as standards of proof.

Of the 133 states parties to the Genocide Convention, 15 still have reservations to Article IX, although 11 such reservations, mostly by former socialist countries, have been withdrawn.⁷¹ In its well-known Advisory Opinion on Reservations to the Genocide Convention,⁷² in which it precipitated the modern law of reservations, the ICJ opined that reservations to the Genocide Convention are permissible, unless they are contrary to the object and purpose of the treaty. Subsequently, several states have objected to reservations made to Article IX of the Convention, while the whole topic of reservations to human rights treaties has attracted much controversy.⁷³

In it most recent judgment in the Congo v. Rwanda case,⁷⁴ the ICJ found that it lacked jurisdiction to hear the case, inter alia because Rwanda (of all states) submitted a reservation to Article IX when it acceded to the Convention in 1975. The ICJ thereby confirmed the validity of reservations to Article IX, as it only deals with the jurisdiction of the ICJ itself, which is always based on the consent of the parties.⁷⁵ However, in their joint separate opinion Judges Higgins, Kooijmans, Elaraby, Owada and Simma caution the Court of the new developments in international law regarding reservations, and advise the Court to revisit the issue of compatibility of reservations on Article IX with the object and purpose of the Genocide Convention, and of the consequences of such incompatibility.

At the moment, beside the Bosnian case, the ICJ has one more pending case involving state responsibility for genocide, namely that of Croatia v. Serbia and Montenegro. The outcome of this case will to a great extent depend on the Bosnian one, though it must be noted that the Croatian case is fraught with even more severe jurisdictional difficulties, as the Court has not decided on its jurisdiction before its 2004 decision in the NATO cases.⁷⁶ Regardless, it seems unlikely that these will be the last cases the ICJ will hear on state responsibility for genocide.

I will now turn to the issue arguably most specific to genocide, which, as we have seen, is a crime which can be distinguished from other international crimes only by the specific, genocidal intent to destroy a protected group as such, in whole or in part. So, if a state can be responsible for genocide, how should we proceed with attributing genocidal intent to a state? Can a state even have intent, and what does it mean to say that a state has intent or not? Or, as the following section will show, it could be that all of these questions are the wrong questions to ask.

BAttributing Intent to a State

A state can be responsible for genocide, though it cannot commit genocide as such – that can be done only by individuals, and their acts can be attributed to a state. As to how a state can posses genocidal intent, one view was repeatedly stated during the oral arguments in the Bosnia v. Serbia case:⁷⁷ the political or military leadership of the state has to possess specific genocidal intent, as evidenced by the existence of a genocidal plan, in order for a state to be responsible for genocide.⁷⁸ In the opinion of this author, the answer to the question of how genocidal intent is to be attributed to a state is a very simple one: one does not have to. This answer is again the consequence of a fundamental methodological distinction between primary and secondary rules.⁷⁹ Genocide is an internationally wrongful act like any other, and the same basic rules of state responsibility should apply.

As for all other internationally wrongful acts, the issue of fault⁸⁰ in the commission of such acts is purely a matter for primary rules.⁸¹ Genocidal intent need only be shown for those individuals whose acts are being attributed to a state, and who are not necessarily a part of the state leadership or its de jure organs, while the process of attribution itself should be totally unaffected. For example, a state is responsible for an act of genocide committed by its de jure organs, say a military unit, even if these organs act ultra vires or in contravention of instructions.⁸² In such a case the state leadership would not possess genocidal intent – far from it – but the state would still be responsible for the commission of genocide. The issue does get more complicated when we are dealing with de facto control as a basis for attribution, which will be dealt with in more detail below, but the same principle still applies. It is the fact of control that is dispositive, not any shared intent, the proof of which would be even more difficult than establishing the individual criminal responsibility of the direct perpetrators.⁸³

The soundness of this approach seems apparent when applied to the facts of the Genocide case: in order to prevail in the case, Bosnia should not be expected to prove that the Serbian leadership possessed genocidal intent – what it needs to show is that the perpetrators on the ground or their superiors in the Bosnian Serb chain of command possessed such intent, while Serbia exercised the required amount of control over them. Nor is Bosnia to be expected to show that the crimes being attributed were committed as a part of a coherent genocidal plan – nor could it probably do so. To the best of our knowledge, there is no equivalent of the Wannsee Conference⁸⁴ for the Bosnian war, no ‘smoking gun’, as it were, which clearly shows the Serbian leadership plotting the extermination and destruction of the Bosnian Muslims/Bosniaks as a group. What Bosnia could show, and arguably is showing though there is no legal requirement for it do so, is a general purpose of both the Bosnian Serb and the Serbian/Yugoslav leadership in pursuing the conflict in Bosnia – that of the creation of a ‘Greater Serbia’, i.e., the addition of Serb-dominated parts of Bosnia and Croatia to the FRY.⁸⁵ That plan, though criminal to any right-thinking person, was not necessarily genocidal, nor should Bosnia be expected to prove that it was so, even though it did imply ethnic violence on a large scale. Its purpose is to show the reasons and motives behind the actions of the Bosnian Serbs as a matter of evidence and to corroborate any genocidal intent, not to conclusively establish responsibility.

If, as we shall see in the next section of this article, a state can be responsible for acts other than the direct commission of genocide under the Genocide Convention, such as complicity or direct and public incitement to genocide, then the degree of mens rea of individuals perpetrating these acts should also be found in general principles of international criminal law. Any genocidal intent on the part of the Serbian leadership, though it is certainly relevant, is dispositive only for their individual criminal responsibility.⁸⁶ The only dispositive factor for the purposes of state responsibility is control, and all other considerations have a purely evidentiary purpose.

CState Responsibility for Breaches of Ancillary Obligations

1Defining Ancillary Obligations

When we discuss state responsibility for genocide what first comes to mind is state responsibility for the actual commission of genocide. Both theoretically and practically the duty of states not to commit genocide is of the most fundamental nature, and state responsibility for a breach of this obligation is of the greatest significance. However, under the Genocide Convention states not only have a fundamental duty not to commit genocide, but also have a number of other, ancillary obligations, such as the duties to prevent and punish genocide. This brings us to the issue of whether states can be held responsible for not performing these obligations and what would be the practical significance, if any, of invoking a state’s responsibility for breaches of these obligations. The text of the Genocide Convention seems pretty clear on the first of these issues: Article IX unambiguously states that ‘[d]isputes between the Contracting Parties relating to the interpretation, application or fulfillment of the present Convention, including those relating to the responsibility of a State for genocide or for any of the other acts enumerated in article III’ (emphasis added) shall fall within the jurisdiction of the ICJ. Article III additionally stipulates that any of ‘[t]he following acts shall be punishable: (a) Genocide; (b) Conspiracy to commit genocide; (c) Direct and public incitement to commit genocide; (d ) Attempt to commit genocide; (e) Complicity in genocide’.

The primary purpose of the Genocide Convention was establishing individual criminal responsibility, yet this does not change the fact that the wording and subsequent textual interpretation of Article IX are clear: states can be responsible for genocide, and for any acts enumerated in Article III. The additional fact that the categories dealt with in Article III, such as conspiracy, incitement or complicity, pertain to the area of individual criminal responsibility also does not bring any severe conceptual difficulties regarding state responsibility. The actions of a state organ that engages in direct and public incitement to genocide can be attributed to the state using the same mechanism of attribution of acts of state organs directly committing genocide.

Article I of the Convention prescribes additional obligations: ‘[the] Contracting Parties confirm that genocide, whether committed in time of peace or in time of war, is a crime under international law which they undertake to prevent and to punish’ (emphasis added), while Article V imposes a duty on states to ‘enact, in accordance with their respective Constitutions, the necessary legislation to give effect to the provisions of the present Convention, and, in particular, to provide effective penalties for persons guilty of genocide or any of the other acts enumerated in article III’. The usage of the word ‘including’ in Article IX of the Convention suggests that state responsibility for the breaches of these obligations can also fall within the jurisdiction of the ICJ, as Article III obligations are mentioned only as an example. We will now, therefore, examine the scope of the ancillary obligations under the Genocide Convention in more detail.

2State Responsibility for Failing to Prevent or Punish Genocide

All state parties to the Genocide Convention have an obligation to prevent genocide, and this duty, coupled with those from Article 41 of the ILC Articles, inasmuch they (start) to reflect customary law, to cooperate in order to bring to an end a situation created by a breach of peremptory norm, is truly an obligation erga omnes. But what exactly is the content of the obligation to prevent genocide? Did, for example, the United States, or France, or Belgium have an obligation to prevent and/or stop the Rwandan genocide? It would be extreme to argue that states have assumed an obligation to prevent genocide outside their territories under Article I of the Convention, and there is absolutely no state practice to support such a contention.⁸⁷ However, what can be done under positive international law is to distinguish between the responsibility of states who through their passivity do nothing to prevent a genocide, and states who perform actions which indirectly support the commission of genocide, e.g., by selling weapons to the génocidaires, or allowing them to use their territory as a base of operation. It could be argued that a due diligence, Corfu Channel type responsibility would arise under Article I, which could be distinguished from state obligations under Article III of the Convention.

Despite numerous calls for a robust regime of genocide prevention, it does not seem that there is such a sweeping duty in the lex lata of international law.⁸⁸ The Genocide Convention does not contribute anything else in that regard: states have a duty to prevent and punish genocide in exactly the same way as they have to prevent and punish crimes against humanity or other massive human rights violations, however flimsy that duty might be. The Convention basically only allows for a jurisdictional possibility of submitting a case to the ICJ, though it would seem that such a possibility would remain largely theoretical unless state responsibility is also invoked for other acts prohibited under the Genocide Convention.

As for the obligation to punish genocide, it can be observed at several levels. First of all, under Article V of the Convention states have a duty to incorporate its provisions in their own criminal law. Secondly, states have an obligation to prosecute any persons for whom there are reasonable grounds for believing that they have committed genocide. Failure to perform any of these obligations will give rise to state responsibility, but any remedies an international court could grant for such a breach would be limited. The responsible state would be obliged to perform its obligation, i.e. prosecute suspected perpetrators of genocide, or extradite them to a state able and willing to prosecute or surrender them to a competent international tribunal, per the principle aut dedere aut judicare.⁸⁹

3State Responsibility for Conspiracy, Direct and Public Incitement and Attempt to Commit Genocide

Article III of the Convention defines four different forms of participation in genocide, in addition to its direct perpetration. The first three, namely conspiracy, direct and public incitement and attempt are inchoate or incomplete crimes, whose existence is separate from genocide itself. For example, direct and public incitement exists even if no persons were actually incited to commit genocide, while attempt will in fact be punishable only if no genocide was committed.⁹⁰ Precisely because of their inchoate nature these three categories should not be confused with general modes of participation in a criminal offence, such as complicity or ‘regular’ incitement.⁹¹

The inchoate concept of conspiracy adopted in the Convention is a common law one, and has no direct counterpart in civil law systems.⁹² To establish conspiracy, it must be proven that two or more persons agreed upon a common plan to perpetrate genocide, while sharing genocidal intent.⁹³ Even though a genocidal conspiracy is, as a matter of fact, an almost inevitable part of every instance of genocide, it is still extremely hard to prove, as it would involve not only documentary evidence but also cooperation and testimony by some of the conspirators.⁹⁴

Direct and public incitement to commit genocide is also a very specific, inchoate offence.⁹⁵ Its purpose is to punish extreme propaganda which facilitates the commission of genocide, by inciting ethnic hatred and dehumanizing a particular group. Some countries have criminalized hate speech or similar types of incitement of ethnic hatred, such as Holocaust denial. Some, such as the United States, believe that criminalizing hate speech would severely impair the freedom of speech and would in the long run cause more damage than good. They therefore criminalize only direct, clear-cut incitement by one person of another to commit a specific crime. Yet, the prohibition against direct and public incitement in the Genocide Convention is not targeted at mere hate speech, but at its most extreme variants. Incitement to genocide must be direct, in the sense of specifically urging other individuals to take immediate criminal action. It must also be public, in the sense that it is conveyed using some form of mass media or before an open audience. It need not be particular, in the sense of inciting a specific person, but it must be clear and unambiguous as to the criminal activity which is being promoted.⁹⁶ And, of course, statements which incite to genocide are usually the best evidence of genocidal intent in relation to acts under Article II of the Convention, such as murder.

It is clear what kind of activity the drafters of the Convention wanted to punish: the memories of Nazi genocidal propaganda were still very fresh in 1948. Julius Streicher, who published the main Nazi newspaper Der Stürmer and several other hatemongering publications, was sentenced to death by the International Military Tribunal (IMT) at Nuremberg for crimes against humanity. And his crimes were indeed depraved – some of his worst statements were reproduced in the IMT judgment:⁹⁷ ‘[a] punitive expedition must come against the Jews in Russia. A punitive expedition which will provide the same fate for them that every murderer and criminal must expect. Death sentence and execution. The Jews in Russia must be killed. They must be exterminated root and branch’ as well as ‘[i]f the danger of the reproduction of that curse of God in the Jewish blood is to finally come to an end, then there is only one way – the extermination of that people whose father is the devil.’⁹⁸

Neither the IMT Nuremberg trial, nor the subsequent criminal proceedings under Control Council Law No. 10⁹⁹ did specifically deal with direct and public incitement to genocide, as the crime of genocide itself was not yet defined. More recently, the International Criminal Tribunal for Rwanda (ICTR) delivered its judgment in the so-called Media Case,¹⁰⁰ in which it found the owners and editors of the Radio Télévision Libre des Mille Collines (RTLM) and the Kangura newspaper criminally responsible for direct and public incitement to genocide. The RTLM broadcasts were particularly conducive for the commission of the Rwandan genocide, as they, for instance, equated Tutsis with cockroaches that need to be exterminated, and even read out lists of people who were supposed to be massacred.¹⁰¹

As for establishing state responsibility for direct and public incitement to genocide, it again poses no conceptual difficulty. For instance, if the RTLM radio was located not in Rwanda itself, but in Congo/Zaire or some other neighbouring country, and was in fact controlled by the government of that country, there would be no legal obstacle for a non-complicit Rwandan government to invoke the responsibility of that state, and demand the cessation of the wrongful act or engage in proportionate counter-measures. Rwanda would not even have to prove a causal nexus between the broadcasts and the commission of genocide,¹⁰² unless it sought compensation for damages under Article 36 of the ILC Articles as a form of reparation.

There is also no conceptual difficulty in establishing state responsibility for an attempt to commit genocide. Bearing in mind the realities of international relations, however, it would appear extremely unlikely for a state’s responsibility to be invoked solely for attempting to commit genocide. Attempt basically remains within the paradigm of individual criminal responsibility,¹⁰³ while state responsibility for attempt is simply a theoretical possibility.

4State Responsibility for Complicity in Genocide

State responsibility for complicity in genocide could prove to be of far greater practical import. Complicity itself can take a variety of forms, ranging from aiding and abetting to specific incitement, or even covering up the evidence of the crime. Most relevant is aiding and abetting as a form of complicity in genocide,¹⁰⁴ as the required level of mens rea for this type of participation in the offence is arguably lower than that for perpetration. Namely, according to the jurisprudence of the two ad hoc tribunals, an aider and abettor need not possess specific, genocidal intent for his criminal responsibility to accrue, but must only have knowledge of the commission of genocide (as well as actively assist it), which would include knowledge of the immediate perpetrators’ genocidal intent.¹⁰⁵ A textbook example would be a businessman selling poison gas to Auschwitz, while knowing full well that the gas would be used to exterminate the Jews.¹⁰⁶ The only two convictions for genocide in Srebrenica by the ICTY, for example, have been for aiding and abetting genocide. Both Krstić and Blagojević were found not to have possessed genocidal intent, but they did have knowledge of such intent of their superiors and were therefore guilty as accomplices.¹⁰⁷

DConcluding Remarks

As this section has shown, state responsibility for genocide is not by its nature criminal, though this concept of responsibility does not divest genocide of its nature as a crime under international law, and individual criminal responsibility runs concurrently with state responsibility. This section has also shown that it is not necessary for genocidal intent to be attributed to a state in any special way, as the issue of fault is a matter of primary rules. Genocidal intent must be shown only in relation to the actual perpetrators of genocide. The sole dispositive element of a state’s responsibility for genocide is its control over those who commit it, and the following section will attempt to establish the proper standard(s) of control whose fulfilment would lead to attribution of private acts to a state. Not only is this approach to attributing genocidal intent methodologically the right one, but it is also the most practicable: proving genocidal intent of high-ranking state leadership with any reasonable degree of certainty would often be an impossible task, because of the frequent lack of any direct evidence.

States can be held responsible not only for the commission of genocide, but also for breaches of several ancillary obligations under the Genocide Convention: failing to prevent or punish genocide; conspiring, directly and publicly inciting or attempting to commit genocide; and for being complicit in genocide. State responsibility for complicity would seem to be of most practical import, while state responsibility for conspiracy and attempt would be largely theoretical, especially when bearing in mind the practical difficulties.

4State Responsibility for Acts of Non-State Actors

ARelevance to Genocide

The previous sections of this article examined several matters which, to a greater or a lesser extent, pose specific problems regarding state responsibility for genocide. This part will for two basic reasons deal with an issue which is not entirely specific to genocide – state responsibility for acts not committed by its de jure organs, which follows from state control over persons or groups of persons committing such acts, generally outside the state’s territory.¹⁰⁸ Firstly, state responsibility for acts of non-state organs will be a central issue in the pending Bosnia v. Serbia case, and its exposition here is necessary for further analysis of the issues arising in that case. Secondly, the issue of state responsibility for acts of non-state actors is highly likely to recur in many future cases involving the application of the Genocide Convention. Though genocide has always been a ‘state crime’, in the sense that it was usually orchestrated and executed by states on a massive scale, states today find many ways, some more and some less successful, of trying to cover up their tracks when on a genocidal rampage. Genocidal states and dictators du jour have learned from the example of Nazi Germany, albeit unfortunately not what a reasonable person would have expected them to learn. They no longer draw up minutes of conferences at which they had plotted the destruction of fellow human beings, and are generally not arrogant enough to leave ample documentary trace or any other direct evidence linking them to the crime.¹⁰⁹ Genocide by proxy has become the order of the day.

The other case concerning the application of the Genocide Convention pending before the ICJ, that of Croatia v. Serbia,¹¹⁰ will (if it ever reaches the merits stage) also revolve around the issue of control that Serbia exercised over the Croatian Serbs. Even in the most obvious recent example of genocide, that of Rwanda, the massacres were for the most part not committed by regular Rwandan armed forces, but by the Interahamwe militia, organized and controlled by the extremist Hutu regime. The most brutal of the ongoing crimes in Darfur, Sudan, are also not being committed by regular Sudanese armed forces, but by the Janjaweed militia, allegedly under the control of Khartoum.¹¹¹

As stated earlier, what is necessary to establish state responsibility for genocide is not any genocidal intent of the state leadership, but instead their control over those who did engage in genocide (and did possess genocidal intent), or in some of the other acts prohibited under the Genocide Convention, such as direct and public incitement or complicity. This part will attempt to ascertain the proper standard of control necessary for the responsibility of a state to arise, by analysing the relevant case law, the ILC Articles and other relevant sources of international law. The purpose of this section is therefore to give more specific content to the secondary rules, as it is not enough to simply say that state control over non-state actors will lead to that state’s responsibility – what must be established is the precise kind and degree of control required for attribution. To that end, it is necessary to examine in some detail two cases, Nicaragua and Tadić, which seem to have come up with different answers to the same question, and to see whether these two approaches are reconcilable or are actually mutually exclusive, a true symptom of the fragmentation of international law.¹¹²

BNicaragua and Tadić

The ICJ case most closely resembling the present Genocide case as to the application of the law of state responsibility is undoubtedly Nicaragua v. United States.¹¹³Nicaragua is the quintessential ICJ case on the international law on the use of force, particularly the legality of the use of force in self-defence. It, however, laterally deals with an issue of the greatest relevance to this discussion: the responsibility of the United States for violations of international humanitarian law by the contras, an armed group in rebellion against the government of Nicaragua.¹¹⁴ As the contras were not de jure organs of the US, the Court had to formulate the tests for the attribution of the acts of the contras to the US, in the course of which it focused on the issue of control. I will note at the outset that several authors talk about the Nicaragua test of control,¹¹⁵ but that in fact the Court formulated not one, but two such tests.

The first Nicaragua test can be found in paragraphs 109 and 110 of the judgment, and will for our present purposes be called the test of ‘complete control’, which can be stated as follows: in order for the acts of a non-state organ de jure, other actor or paramilitary group to be attributed to a state, the relationship between the state and any such group must be so much one of dependence on one side and control on the other that it would be right to equate the group either with an organ of the state or as acting on the state’s behalf. The Court examined several factors which could help establish such dependence and control: (i) whether the non-state organ, actor or group was in fact created by the state;¹¹⁶ (ii) whether the involvement of the state amounts to more than providing training or financial assistance; in other words, even if the support provided by a state to the group is crucial for its activities, that is not enough – the relationship must be one of complete dependence;¹¹⁷ (iii) whether, even though a state has a degree or potential for control inherent in its providing support to a group, the state in fact exercises that control;¹¹⁸ (iv) whether the state selected, installed or paid the political leaders of the group.¹¹⁹

In essence, the test of complete control requires the dependent group to have no real autonomy from the controlling state.¹²⁰ The type and degree of control must qualitatively be the same as the control a state exercises over its own organs, forces or territory, and the state would accordingly be responsible for any acts committed by such a group, even if a specific act was committed ultra vires or against explicit instructions. The controlled non-state actor would become a state organ de facto, as the only thing missing for such an actor to actually be considered a proper organ of the state would be a determination to that effect by the state’s own domestic law.

The second Nicaragua test, that of effective control and formulated in paragraph 115 of the judgment, was subsidiary in the Court’s analysis, as it found that it could not establish the United States’ responsibility under the first one. According to the Court, general control, or a high degree of dependency are insufficient to impute all of the acts of the contras to the US – that control would have to be complete, and the dependency total for such kind of attribution to occur. However, this also means that at least some of the acts of the group could be attributed to the state, if the state had effective control over the particular operation in the course of which violations of international human rights or humanitarian law have been committed.

The Court in Nicaragua therefore envisages a two-fold paradigm of state responsibility for acts not committed by its de jure organs. Its first test, that of complete control, is a general one as it deals with the control a state exercises over the entire functioning of a group, and its fulfilment for all intents and purposes equates de facto controlled groups with state organs. The second test, that of effective control, is a specific and a subsidiary one, as it deals with the state’s control over the conduct of a specific operation in the course of which violations have been committed, when the conditions of the first test are not met.¹²¹ Both tests undoubtedly have an exceptionally high evidentiary threshold, but they seem to have been accepted by all of the judges in the case. Even the US Judge Schwebel, who submitted a powerful dissenting opinion in which he disagreed with almost all of the conclusions of the Court, accepted its position as to state responsibility for violations of international humanitarian law.¹²² Even more importantly, the Court’s reasoning as to the test of effective control was emphatically endorsed by Judge Roberto Ago, at the time the Special Rapporteur of the International Law Commission on state responsibility.¹²³

The issue of state responsibility as expounded by the ICJ in Nicaragua then came quite unexpectedly before the ICTY, a court whose domain is the determination of individual criminal responsibility. The ICTY has no competence to pronounce upon the responsibility of any state, yet it used Nicaragua to answer a question which undoubtedly fell within its purview: how to draw a distinction between an international and a non-international armed conflict. This distinction is (still) essential for determining which rules of international humanitarian law apply in a particular case, as different sets of rules apply for different types of armed conflict,¹²⁴ and as this has an impact on individual criminal responsibility, particularly for the grave breaches of the 1949 Geneva Conventions, which apply only in international armed conflicts.¹²⁵

The ICTY had to deal with this question when defining the character of the armed conflict in Bosnia, which prima facie appeared to be a three-sided civil war, i.e. a non-international armed conflict, between the Bosnian Muslims, Croats and Serbs. However, the intense involvement of Serbia (FRY) and Croatia in the conflict posed the question of whether this non-international armed conflict had become ‘internationalized’ through an agency relationship, which would bring the full rules of international humanitarian law to bear.

Tadić was a guard in one of the horrific Bosnian Serb prison camps in the Prijedor municipality of Bosnia, and was charged for his participation in war crimes and crimes against humanity committed in those camps. The Trial Chamber therefore had to establish whether grave breaches of the Geneva Conventions, defined as war crimes in Article 2 of the ICTY Statute, applied to the accused, i.e., whether a state of international armed conflict existed at the material time. To do so, the Trial Chamber invoked Nicaragua,¹²⁶ relying on (unrelated) paragraphs of that judgment which stated that the conflict within Nicaragua was not international¹²⁷ and which define the tests of control for the purposes of state responsibility, as explained above,¹²⁸ without, however, clearly distinguishing the two tests of attribution used by the ICJ. Additionally, the ICJ in Nicaragua had never used the tests of control it developed regarding state responsibility in order to determine whether the conflict in Nicaragua was international or not.¹²⁹ In any case, the Trial Chamber proceeded to examine whether the forces of the Bosnian Serbs on the whole have remained the agents of the FRY after the (supposed) withdrawal of Yugoslav troops from Bosnia in May of 1992. It found that the relationship of the Republika Srpska (RS) to the FRY was one of an ally, albeit a highly dependent ally,¹³⁰ that the RS could not be considered a de facto organ or agent of the FRY, and that therefore the applicable law in this internal conflict was Common Article 3 of the Geneva Conventions, and not their grave breaches regime.¹³¹ In an impressive dissent, Presiding Judge Gabrielle Kirk McDonald disputed the majority’s finding as to the nature of the relationship of the Bosnian Serbs and the FRY, concluding that in fact the RS was under control of the FRY.¹³² But, more importantly, Judge McDonald clearly distinguishes between the two tests the ICJ adopts in Nicaragua,¹³³ and finds that an agency relationship exists between the RS and the FRY under the first test of complete dependency and control, not under the second test of effective control.¹³⁴

In its judgment, the Appeals Chamber of the ICTY¹³⁵ affirmed the relevance of Nicaragua and an agency relationship, while reversing the Trial Chambers application of Nicaragua and initiating a direct conflict of jurisprudence with the ICJ. On appeal, the Prosecution argued that the general international law of state responsibility is not relevant for establishing the character of an armed conflict, and that a different test should be devised.¹³⁶ The Appeals Chamber disagreed, finding that the use of irregular forces implies the need for a determination of a standard of control.¹³⁷ The Appeals Chamber then proceeded to analyse Nicaragua in some depth,¹³⁸ but its analysis of the law of state responsibility seems to be flawed from the very beginning: the only Nicaragua test it deals with is the one of effective control, set out in paragraph 115 of the judgment, while it completely dismisses the test of complete control as set out in paragraphs 109 and 110 of the Nicaragua judgment,¹³⁹ and does so contrary both to the submissions of the Prosecution and the dissent of Judge McDonald.¹⁴⁰

The Appeals Chamber found this sole Nicaragua test unpersuasive on two grounds.¹⁴¹

Firstly, the Chamber asserts that the Nicaragua test goes against the very logic of state responsibility. It quite correctly states that the purpose of attribution of acts of private individuals to a state is to avoid the possibility of a state acting de facto through such individuals but distancing itself from them through its own law when a question of its responsibility arises. According to the Chamber, the requirement of international law for the attribution to states of acts performed by private individuals is that the state exercises control over the individuals, but the degree of control may, however, vary according to the factual circumstances of each case.¹⁴² A single private individual would need specific instructions from a state in order for his or her acts to be attributable to a state, while an organized and hierarchically structured group, such as a military unit, would not need such specific control, but it would be sufficient to require that the group as a whole be under the overall control of the state for attribution to occur.¹⁴³ Secondly, the Appeals Chamber states that the Nicaragua test is at variance with state practice, as states were held responsible for acts committed by paramilitary units when they had overall control over them.¹⁴⁴

Finally, the Chamber articulates what it holds is the proper standard, that of overall control.¹⁴⁵ To the Appeals Chamber the crucial point is the distinction between state control over a particular individual, in the case of which a proof of specific instructions is necessary, and control over organized armed groups, which requires only a proof of overall state control.¹⁴⁶ The Chamber also briefly articulates a third test, whose relevance or application is not readily apparent in the Chamber’s analysis, which states that private individuals acting within the framework of, or in connection with, armed forces, or in collusion with state authorities may be regarded as de facto state organs.¹⁴⁷ This third test appears to target the same type of control or interdependence the ICJ contemplated in its first Nicaragua test,¹⁴⁸ but the Appeals Chamber does not expound on it in any sort of detail. The Chamber then proceeded to analyse the facts regarding the relationship between the RS and the FRY, and found that the requirements of its test of overall control were satisfied, and that consequently the conflict between the forces of the Bosnian government and the Bosnian Serbs should be regarded as international.¹⁴⁹

Before examining other relevant developments in the international law of state responsibility since Nicaragua, it would prove helpful to provide some initial remarks on the Tadić case. Firstly, though there is much that is compelling in the Court’s discussion, it is all based on a demonstrably false basic premise: that the ICJ in Nicaragua said that effective control is the only way a state can assume responsibility over acts of non-state actors. This is simply not true. Before expounding on the test of effective control in paragraph 115 of the judgment, the ICJ clearly contemplated the test of complete control in paragraphs 109 and 110 of the judgment, as was well noted both by the Prosecution and Judge McDonald in her dissent in the Tadić case. Like the Chamber’s test of overall control, the ICJ’s test of complete control is general, in that it is not targeted at a specific situation, but at the relationship of agency, control and dependency between a state and a non-state actor. The test of effective control, on the other hand, is specific, as it requires the showing of control over a particular operation.¹⁵⁰ If the conditions of the test of complete control are satisfied, it is not necessary to examine whether a particular act was committed under specific instructions of a state in order for that act to be imputed to the state, just as with the application of the Chamber’s test of overall control. Where the two tests differ, however, is in the degree of control which they require for attribution. The ICJ’s test demands a showing of complete dependence,¹⁵¹ while the ICTY’s test will allow attribution even for a state coordinating or helping in the general planning of the non-state actor’s military activity.¹⁵² The issue of which of these two tests is the correct one from the standpoint of state responsibility will be examined further below, but one still cannot help the impression that the ICTY Appeals Chamber dramatically misread Nicaragua, even though it still might have articulated a proper standard of state responsibility.¹⁵³

A second issue is whether the ICTY has misapplied Nicaragua, regardless of whether it has misread it in the first place. In other words, was going into the whole matter of state responsibility actually necessary to determine whether an armed conflict is international or not, and was it actually necessary for the ICTY to directly challenge a ruling of the ICJ?¹⁵⁴ The full answer to this question is beyond the scope of this article, but it will be touched upon in more detail below.

CWork of the International Law Commission

The ILC went through several iterations of the Draft Articles on State Responsibility, its attempt to codify and progressively develop the general customary international law of state responsibility. As stated earlier, one of the primary methodological contributions of the ILC is the separation of primary rules of international law and secondary rules which regulate state responsibility.

The ILC establishes several paradigms of attribution of individual conduct to a state in its examination of customary law, the most important of which is the conduct of state organs, under Article 4¹⁵⁵ of the Draft Articles, which defines organs as (primarily) persons or entities having that status under the internal law of the state. Can we therefore make a distinction between de jure organs, i.e. organs within the meaning of Article 4, and de facto state organs, i.e. persons or entities which for all intents and purposes act as state organs, even though that status is not recognized by the internal law of the state? In other words, can a state escape responsibility merely by shielding the acts committed by its de facto organs through the deceptions of its own internal law?¹⁵⁶ The ILC Commentaries quite rightly answer this question in the negative: ‘[a] State cannot avoid responsibility for the conduct of a body which does in truth act as one of its organs merely by denying it that status under its own law. This result is achieved by the use of the word “includes” in paragraph 2 [of Article 4]’.¹⁵⁷ The issue of a test or standard for establishing whether a particular person or entity is a de facto organ of a state is not discussed in the ICL Commentaries, but the basic idea is readily apparent: a de facto state organ must act in essentially the same way as a de jure state organ.¹⁵⁸

Article 8 of the ILC Articles deals with the attribution of conduct under instructions of or directed or controlled by a state: ‘[t]he conduct of a person or group of persons shall be considered an act of a State under international law if the person or group of persons is in fact acting on the instructions of, or under the direction or control of, that State in carrying out the conduct.’ The previous draft of the Articles did not contain specific requirements of instructions, direction or control but used an abstract formulation on attribution of acts of persons or groups of persons in fact acting on behalf of the state.¹⁵⁹ The final Articles themselves do not define a standard of control for attribution, but the ILC approvingly cites the second Nicaragua test, that of effective control, and explicitly dismisses Tadić (though it tries to distinguish it),¹⁶⁰ while in their submissions to the ILC states have not made any comments regarding the proper standard of control.¹⁶¹

It would appear that the ILC Articles, like Nicaragua, offer two distinct paradigms of state responsibility relevant for this analysis. A state could either be responsible under Article 4 for all of the acts of its de facto organs or agents, even those committed ultra vires under Article 7, or it could be responsible under Article 8 for acts of persons or entities under its direction and control, or acting on its instructions. The ILC endorses the effective control test for responsibility under Article 8, and dismisses the test of overall control. The ILC does not discuss the issue of control required under Article 4, though it seems to favour a more restrictive approach, such as the one in the first Nicaragua test of complete control.¹⁶²

DAnalysis: A Distinct Method, if Not a Distinct Discipline

I will now proceed to analyse the tests of control developed in Nicaragua and Tadić and try to establish whether they are mutually exclusive and, if so, which are the correct ones under international law. But, before doing so, I would like to return to a theme of the previous part of this article: the methodological need to separate primary from secondary rules. This theme crops up rather suddenly in the discussion in this part, as it is ostensibly simply about determining the content of secondary rules, i.e. the proper standards of control. However, the jurisprudence outlined above, particularly Tadić, shows a tendency of again blurring this important distinction, by conflating the concepts of state responsibility with those of the primary rules of humanitarian law and human rights law, while a similar tendency can also be noticed in recent scholarly discussions on the legality of the use of force in self-defence in response to the 9/11 terrorist attacks. So, in addition to discussing the standards of control necessary for attribution of individual acts to a state, this section will also examine the relationship between the law of state responsibility, and the jus ad bellum, the jus in bello, and the concept of state jurisdiction in international human rights law.

Quite relevant for the issue of state responsibility for genocide is the claim made by the US administration after the traumatic events of 11 September 2001, and in the subsequent ‘war on terror’ that state responsibility now ensues not only for control, be it complete, overall or effective, but for the mere fact of harbouring a terrorist (or, arguably, genocidal) non-state actor.¹⁶³ As well noted by Ratner, this American position, and the way in which it is either accepted or rejected by other members of the international community, has profound implications on the jus ad bellum, the jus in bello and the law of state responsibility.¹⁶⁴ This view does, however, tend to follow the unfortunate tendency of conflating the rules of state responsibility with those of primary international law.

A state may well harbour terrorists (or genocidaires) and it would certainly bear state responsibility for its own act of harbouring these persons. The jus ad bellum may even allow a state attacked by these terrorists to respond against the harbouring state, though neither state practice nor the ICJ¹⁶⁵ provide much clarity on the issue of self-defence to attacks by private actors.¹⁶⁶ But this does not mean that the harbouring state has automatically assumed state responsibility for all acts committed by the terrorists – the law on state responsibility and the law on the use of force have their own separate logic and can simply develop independently.¹⁶⁷ By way of example, if we accept that the US actions in Afghanistan were lawful as being in self-defence,¹⁶⁸ this does not inevitably lead to the conclusion that the state of Afghanistan must have been responsible for the 9/11 terrorist attacks. It is responsible for its own actions, that is allowing the terrorists to operate from its territory; it is not, however, responsible for the acts of the terrorists themselves unless it controls them. In the case of genocide as opposed to terrorism, this situation could certainly lead to the state’s responsibility for complicity in genocide, under Article III of the Genocide Convention, but not to its responsibility for the commission of genocide, under Article II of the Convention.

Another area of confusion between primary and secondary rules is the convergence between the rules of humanitarian law and the law of state responsibility. This relationship is quite apparent from the ICTY’s jurisprudence on determining whether an armed conflict was international or non-international in its character, as best seen in Tadić. The question posed and resolved in Tadić was whether a prima facie internal armed conflict can become ‘internationalized’ through a relationship of agency of one of the belligerents with a foreign power. This type of internationalization of an armed conflict also seems perfectly reasonable and logical. What is not logical, however, except in an extremely strange circular sense, is the use of the international law of state responsibility for fashioning a test for this type of agency, as in Tadić the tests of state responsibility were used to establish which part of the law of armed conflict applied in the first place. How a test of attribution can determine the content of a legal obligation is completely beyond me.

The chief methodological problem of the ICTY’s jurisprudence on the classification of armed conflicts is an unwarranted search for a precise, artificial test which can be used to determine the nature of a conflict. No such test can be fashioned, as legal tests by their very nature attempt to extract a dispositive point from the ‘totality of the circumstances’, and it is precisely this totality one needs to look at in order to see an armed conflict for what it truly is.¹⁶⁹ But even if the need for a test is so great, and even if the test of overall control developed by the Appeals Chamber is actually the correct one from the standpoint of international humanitarian law,¹⁷⁰ the legal nature and source of authority of that test is certainly not to be found in the international law of state responsibility.¹⁷¹

Unlike the law on the use of force or the law of armed conflict, the law of state responsibility is not a distinct discipline of public international law, but has general application. Even so, it does have a distinct method and purpose. It rose from the muddied waters of diplomatic protection and treatment of aliens, and should now try to resist being dragged back into the methodological mud made up of different substantive primary rules. The answer to whether an internal armed conflict has become international is not to be found in the law of state responsibility – it is hard to escape the impression that the ICTY’s ruling in Tadić simply tries to do too much and that the Appeals Chambers unnecessarily attempted to deal with questions which were immaterial to its basic purpose of assigning individual criminal responsibility.¹⁷² The law of state responsibility is the answer, however, to the questions of attribution of individual acts to states for the purposes of assessing their responsibility for violations of either international humanitarian law or human rights law, including the Genocide Convention.

Even if Tadić is right as to what type of agency can internationalize a non-international armed conflict,¹⁷³ it is simply wrong in its interpretation of the law of state responsibility. The cases the Appeals Chamber relies on do not support its test of overall control: the Yeager case¹⁷⁴ before the Iran–US Claims Tribunal which it cites¹⁷⁵ dealt with the responsibility of a state for ‘outsourcing’ to private actors the functions of its governmental authority – this type of attribution is similar to that dealt with by Articles 5 and 9 of the ILC Articles, and does not require a state’s control over the entity in question. All it takes for the attribution of the acts of such a para-statal entity to the state is the fact that the state allows the entity to exercise such functions – as in Yeager, where the revolutionary ‘Komitehs’ de facto acted as security officers in Iran. The Yeager case itself is dealt with by the ILC as an example of attribution of conduct by a group, which even if not explicitly authorized by a state to do so, did exercise elements of governmental authority in the absence of the official authorities, which leads to the attribution of these acts to the state under the rule now codified in Article 9 of the ILC Articles.¹⁷⁶ This type of attribution does not deal with the actions of an entity outside the territory of the state, which does not purport to exercise governmental functions on behalf of that state, but on its own behalf. Such a case would require the existence of control as the guiding principle of attribution.

On the other hand, the cases of the European Court of Human Rights cited by the Appeals Chamber¹⁷⁷ do mention ‘effective overall control’ – but they are actually dealing with a different concept, which is specific to human rights treaties: establishing state jurisdiction over a person within a certain territory as a prerequisite for going into the question of state responsibility for a violation of that person’s human rights, which makes their reliance on Tadić¹⁷⁸ inadequate. When read in the light of the subsequent cases before the European Court dealing with the extraterritorial application of the European Convention on Human Rights, such as Banković,¹⁷⁹Issa¹⁸⁰ and Ilascu,¹⁸¹ these cases show that the concept of state jurisdiction has little to do with attribution as an element of an internationally wronful act of a state.¹⁸² Jurisdiction is actually the threshold criterion for the existence of the positive international obligation of a state party to the treaty to ‘secure’ the human rights of persons ‘within its jurisdiction’.¹⁸³ These cases of the European Court do not revolve around the general law on state responsibility, but deal with a concept of state jurisdiction which is specific to human rights treaties, as it is a prerequisite for their application. This is not to argue that human rights treaties are ‘self-contained’ regimes bereft of contact with general international law¹⁸⁴ – the European Court certainly should (and does)¹⁸⁵ apply the general law on state responsibility when it actually deals with issues of state responsibility. State jurisdiction, however, is not one of them, and the reliance by Tadić on these cases is misplaced.

Furthermore, Tadić cannot be distinguished from Nicaragua, either on the facts or as a matter of law, as the ILC tries to do in its commentary on Article 8.¹⁸⁶ The fact that the ICTY is dealing with individual and not state responsibility is irrelevant, as the Appeals Chamber quite clearly poses the problem as a lex specialis one – it explicitly says that as it cannot find an answer to what constitutes an ‘internationalized’ non-international armed conflict in the lex specialis of international humanitarian law, it must try to answer this question by recourse to the general international law of state responsibility.¹⁸⁷ But that is precisely the problem: international humanitarian law is in no way lex specialis to the law of state responsibility: the issue of the qualification of the legal nature of an armed conflict is solely one for international humanitarian law, and has nothing to do with the law of state responsibility, even though the factual patterns on which these branches of law operate might be very similar.

To conclude, inasmuch as it deals with the interpretation of the law of state responsibility, Tadić cannot be distinguished from Nicaragua – they cannot both be right.¹⁸⁸ And the right answer as to attribution based on state control over private actors is, predictably, given by the ICJ in Nicaragua and later by the ILC. A state can be held responsible for all of the acts of its non-de jure organs, even those committed ultra vires, pursuant to Article 4 of the ILC Articles, as de facto organs under the complete control of a state per the test defined by the ICJ in paragraphs 109 and 110 of its Nicaragua ruling. In making an assessment as to whether a non-state actor is actually a de facto organ of a state it is necessary to take into account numerous factors, but the basic principle is apparent: a de facto organ must have essentially the same qualities as a de jure organ, except the veneer of legality provided by the state’s own domestic law. An act of a non-state actor can also be attributed to a state under Article 8 of the ILC Articles if a state issues specific instructions, or the acts are committed under its direction or control. The proper standard of control under Article 8 would be the effective control test from paragraph 115 of Nicaragua. This conclusion is also in conformity with the ICJ’s most recent pronouncement on state responsibility in Congo v. Uganda, in which it seems to have adopted this model of responsibility, although it did not discuss it in great detail.¹⁸⁹ In state practice there is no case using the overall control test as stated by Tadić,¹⁹⁰ nor have states indicated any support for a standard of attribution under Article 8 other than effective control.

5Bosnia v. Serbia: Analysis

ABasic Facts of the Case

As stated at the beginning of this article, this year marks the first time in history that a state is on trial for genocide before an international court. The Genocide case brings many complicated legal and factual issues, and conclusively resolving most of them is certainly not a matter within the scope of this article. The basic facts are well known. The war in Bosnia and Herzegovina from 1992 to 1995 following the break-up of the former Yugoslavia was the most violent European conflict in the past 60 years, and it raised spectres of ethnic hatred thought long buried by most Europeans. The conflict in Bosnia, itself a multiethnic microcosm within the former Yugoslavia, was the bloodiest by far of all the Yugoslav wars. Estimates range from one hundred thousand¹⁹¹ to two hundred and fifty thousand dead,¹⁹² mostly civilians, but no reliable and conclusive figure exists.¹⁹³ The war ended with the American and European-brokered Dayton Peace Accords, which not only put an end to the fighting itself but also established the new constitutional framework of Bosnia.¹⁹⁴

One of the main features of the war was the principal method of waging it – ethnic cleansing. War crimes, crimes against humanity and possibly even genocide were perpetrated on such a scale as to invite the intervention of the international community and initiate a new phase in international criminal justice with the creation of the ICTY. The almost total disrespect for the laws of armed conflict and the eruption of ethnic violence among yesterday’s neighbours, widely intermarried and connected by the same language, history and culture is almost inexplicable to a casual observer.

Of the many crimes committed in the course of the conflict one surely stands out: the July 1995 massacre of more than 7,000 Bosnian Muslim men and boys at the hands of the Army of Republika Srpska in the small East Bosnian town of Srebrenica is widely regarded as being the single worst act of mass murder committed in Europe after the Second World War, and has been legally qualified as genocide by the ICTY.¹⁹⁵ Yet, as with all major crimes, Srebrenica itself is still a subject of two differing points of view. From the perspective of Bosnian Muslims (and the majority of global opinion), it is seen as the most grievous crime committed during the Bosnian war and a symbol of Serbian aggression, while for many Serbs, both in Bosnia and in Serbia, the crime itself is either completely fabricated or at the very least blown out of proportion. As could be heard at a recent gathering at the Belgrade University Faculty of Law, Srebrenica was in fact ‘liberated’ and the Muslims should remove their false graveyard from Serbian land which could be put to some good agrarian use.¹⁹⁶

As the Genocide case progresses before the ICJ, it is readily apparent that the general publics of both Bosnia and Serbia do not see it in the legalistic, formalistic fashion in which state responsibility for genocide has been presented in this article. Rather, the case is seen as the final adjudication on the nature of the totality of the conflict, of whether it was a Serbian war of aggression or still in essence a civil war, albeit with extensive external involvement, in which no party was innocent.¹⁹⁷ Within Bosnia, the Republika Srpska is actively opposed to the Bosnian lawsuit and does everything in its power to obstruct it. Additionally, any outcome of the case is perceived by the public as affecting the future, post-Dayton constitutional arrangement of Bosnia. For example, the Republika Srpska is regarded by many Bosnian Muslims as a ‘genocidal creation’, which has no legitimacy and which should be abolished,¹⁹⁸ while many Bosnian Serbs feel that their own sufferings have been diminished and that they are again the victims of injustice, and that crimes committed against them will not be addressed by the ICJ.¹⁹⁹

The case itself will revolve around three main groups of issues: jurisdiction, establishing whether genocide in fact happened in Bosnia, and establishing whether Serbia is responsible as a state for that genocide. My purpose here is not to present the Genocide case in great detail – that would require a much more detailed examination of the facts and evidence, a great deal of which is not publicly available. The purpose of this section is to apply the methodology outlined in the preceding sections to the basic facts of the case, and to show that the ICJ will not face any significant legal obstacles in applying this methodology but that its chief problem will be evaluating the evidence and establishing the relevant facts.

BJurisdiction and Brief Procedural History

In 1993 Bosnia and Herzegovina instituted proceedings before the ICJ against the FRY pursuant to the 1948 Genocide Convention, arguing that the Serbian aggression in Bosnia amounted to genocide.²⁰⁰ In 1996 the Court rendered a judgment on preliminary objections filed by the FRY, finding that it had jurisdiction to hear the case (exclusively) under Article IX of the Genocide Convention.²⁰¹

After the fall of Milošević, the new, democratically-elected authorities of the FRY (now Serbia and Montenegro) withdrew the counter-claims made by the previous government²⁰² and applied for revision of the 1996 judgment,²⁰³ claiming that in 1996 the FRY was not a member of the United Nations and the Statute of the Court (as was found in 2004 by the Court itself in the Legality of Use of Force cases (‘NATO cases’)²⁰⁴) nor a party to the Genocide Convention and that therefore the ICJ manifestly lacked jurisdiction. In 2003 the Court rejected the request for revision of its earlier judgment.²⁰⁵ However, after its 2004 decision in the NATO cases the government of Serbia and Montenegro has again raised the jurisdictional issue,²⁰⁶ which was elaborated in the oral hearings, lasting from February to May 2006.²⁰⁷

As much as the case is complicated factually and legally as to state responsibility, it is at least as complicated when it comes to jurisdictional matters. The issue of the FRY’s UN membership was fraught with such needless complications and legal difficulties that it could be only called by that ghastly appellation of failed taxonomy: sui generis. Serbia’s extensive procedural manoeuvring, equivocation and strategic litigation have led the Court to a very real possibility of finding that its own jurisprudence is in conflict with itself – a situation certainly not appreciated by the judges of the ICJ.²⁰⁸ The legal arguments on jurisdiction will not be analysed here. Suffice it to say that the ICJ will have the opportunity to clarify its doctrine on the issues such as access to the Court,²⁰⁹ automatic succession regarding human rights treaties,²¹⁰res judicata²¹¹ and forum prorogatum.²¹²

As is apparent from this brief list, the jurisdictional question is very complicated and is in some ways the result of the Court’s own conflicting jurisprudence. I will happily not involve myself in any further discussion of the issue, except to remark that as the legal issue is more and more unclear, so will elements of judicial policy and prudence begin to prevail. Though it cannot be said that Serbia’s legal case on jurisdiction is a frivolous one and entirely without merit, there are significant policy considerations going against it: if the Court finds that it does not have jurisdiction, it might lead to a repeat of the South West Africa fiasco and to a significant loss of prestige of the Court, to introducing an element of doubt in the finality of the Court’s pronouncements, and the Court might be seen as allowing itself to be badgered in its decision-making process by a state’s disavowal of its previous position and the adoption of a completely new litigation strategy. To this end, some use of estoppel²¹³ by the Court might come into play – though the issue of Serbia’s lack of access to the Court, due to its concomitant lack of UN membership, depends on objective criteria,²¹⁴ and does not seem to depend on the separate question of Serbia’s consent to the Court’s jurisdiction.²¹⁵ In any event, the odds are that the case will proceed to the merits, though the possibility of the Court finding that it lacks jurisdiction is not to be excluded.

CProving Genocide in Bosnia

1A General or a Particular Approach?

The first burden Bosnia has to discharge in order to prevail in the case is to prove that the Bosnian Serbs actually committed genocide during the conflict. As stated at the outset, I have no problems in labelling any mass murder of human beings as genocide in the lay sense or as defined by some other social science, but what Bosnia needs to prove is that genocide occurred in the sense of Article II of the 1948 Genocide Convention and other general principles of international criminal law.

What is readily apparent from both Bosnia’s written and oral pleadings is that it takes a general approach to the issue of proving genocide. It is basically arguing that the totality of all crimes during the conflict amount to genocide, and is using specific situations as illustrations of this general idea.²¹⁶ So, for instance, it is not focusing specifically on the massacre in Srebrenica, but is also treating the horrendous siege of Sarajevo²¹⁷ or the detention camps in the Prijedor municipality as examples of genocide.²¹⁸ Basically, Bosnia argues that in the same way that one can say that the atrocities perpetrated in Rwanda in 1994 were genocide, so can one say that the crimes committed during the Bosnian conflict from 1992 to 1995 also, on the whole, amount to genocide. Now, it should be noted that, even if this fails as a litigation strategy, it is still good for Bosnia to have its day in Court, and for all these crimes to be presented as a part of a bigger picture, which they undoubtedly are. One should never forget the immense human pain and suffering caused by the Balkan warlords in the intensely legalistic and formalistic setting of the International Court of Justice.

However, this still begs the question of whether Bosnia’s approach to proving genocide is legally the correct one. For one thing, it is in almost total conflict with the jurisprudence of the ICTY, on which Bosnia otherwise extensively relies. According to the Tribunal, the ‘only’ instance of genocide committed during the Bosnian war was Srebrenica, and all other crimes, including the siege of Sarajevo²¹⁹ and the Prijedor camps,²²⁰ have been qualified either as war crimes or as crimes against humanity. Additionally, ‘ethnic cleansing’²²¹ has generally been treated by the ICTY as persecution, a crime against humanity, and not as genocide.²²² According to the Tribunal and distinguished commentators, the intent (not simply motive) behind ethnic cleansing is the dislocation of a particular group, and the purpose of acts of murder and rampage is to induce this dislocation, while the specific intent behind genocide can only be the physical and biological destruction of a group as such, in whole or in part.²²³ The fact that a significant number of people are killed in order to forcibly achieve dislocation does not in itself allow for the legal qualification of the killings as genocide. This approach to genocidal intent is crucial to distinguishing genocide from crimes against humanity, and has always been followed by international tribunals. It must also be noted that deportation can become an actus reus of genocide, if it is conducted in such a way as to deliberately inflict on the group conditions of life calculated to bring about its physical destruction in whole or in part.²²⁴ For example, this is precisely the way in which genocide was conducted by Turkey in 1915–1917 against its Armenian population by forcing them to march endlessly through deserts and other difficult terrain, without food, water or shelter, and these marches alone caused massive casualties.²²⁵ Deportation by itself is not enough, however, to constitute genocide – it must be followed by genocidal intent, which cannot be inferred just from the act of deportation. As well stated by Schabas, genocide is the last resort of the frustrated ethnic cleanser.²²⁶

2Standards of Proof

The principal reason why the ICTY qualified ‘only’ Srebrenica as genocide in Bosnia is the high standard of evidence required before the Tribunal: proof beyond reasonable doubt. The question arises whether the ICJ should also adopt this standard in its approach to proving genocide in this case. After all, as Bosnia argues, proof beyond reasonable doubt is relevant only for the issue of individual criminal responsibility, and not for state responsibility before the ICJ in what is essentially a ‘civil’ proceeding.²²⁷ However, I believe that the answer should still be in the affirmative, and that it is again a product of a firm distinction between primary and secondary rules.

It is well known that there is a general problem in the treatment of evidence before international tribunals in general and the ICJ in particular, which can be seen as the result of a clash between two legal cultures.²²⁸ Namely, the common law tradition has a strict classification of different standards of proof, such as proof beyond reasonable doubt, clear and convincing evidence or preponderance of the evidence, and these standards are the product of several features of the common law system, such as trial by jury and a general rigidity of procedure. On the other hand, the civil law tradition has no different standards of proof: the judges can almost always freely evaluate evidence and assign it probative value according to their own, personal conviction²²⁹ (or intime conviction, as the French would put it).²³⁰ But, this freedom of evaluating evidence requires civil law judges to provide reasoning behind their factual conclusions, which common law juries need not do, while a regular appeal to a higher court usually consists of a trial de novo, with the appellate judges giving little or no deference to the factual conclusions of the trial court.²³¹ The ICJ, in which civil lawyers generally dominate in numbers, has therefore never ventured into expounding on rigid standards of proof in its judgments, and this has frequently drawn criticisms from common lawyers.²³²

Why should the Court then adopt the beyond reasonable doubt standard in this case? Well, it is quite easy to see why Serbia would make such an argument, and why it would be opposed by Bosnia, but there are several objective reasons for taking such an approach.

Firstly, even though genocide is an internationally wrongful act as any other, it still remains a crime under international law. As stated earlier, the primary rules regarding genocide consist not only of the Genocide Convention (which, of course, says absolutely nothing about standards of proof), but also of general principles of international criminal law which the ICJ needs to utilize when interpreting the Convention.²³³ And one of these principles is undoubtedly the requirement of proof beyond reasonable doubt: not only is it stipulated in the rules of procedure and evidence, as well as the jurisprudence of the ICTY²³⁴ and the ICTR,²³⁵ it is now also explicitly stated in the Rome Statute of the ICC.²³⁶ International criminal law is precisely the one area of international law which has relatively settled rules and standards of evidence.²³⁷ Divesting the primary rules on genocide of this standard of proof would lead to exactly the same consequence as divesting them of the requirement of proving specific, genocidal intent – it would lead to the creation of a ‘tort’ of ‘civil’ genocide, which we could not distinguish from crimes against humanity as a genus and which is certainly not contemplated by the Genocide Convention.²³⁸ The fact that proceedings before the ICJ are ‘civil’ and not criminal is irrelevant, as the ICJ must establish whether a particular crime was actually committed by a person or a group of persons, according to the primary rules of international criminal law, though it need not conduct a criminal trial. Genocide does not lose its status as a crime simply because its existence is assessed in a judicial proceeding on state responsibility.

Secondly, precisely because individual and state responsibility run concurrently it is necessary to maintain the same evidentiary standard in criminal trials and in proceedings before the ICJ. It would be rather strange for the ICTY to say repeatedly that a particular crime did not qualify as genocide, and for the ICJ to say that it actually did. It is therefore imperative to avoid creating a situation which I can only call a possible O.J. Simpson moment in international law. Avoiding different interpretations of the definition of genocide and its different application to the same facts is mandated by the basic principle of legal certainty and the need for preventing, if possible, the fragmentation of international law. Strangely, it is the ICJ which is now in the position to fragment the general framework of international law: as the ICJ must resist the attempt of the ICTY Appeals Chamber in Tadić to modify the rules of state responsibility without any basis in state practice, so should the ICJ show some deference to the ICTY in what is the Tribunal’s primary work – interpreting international criminal law and applying it to situations and cases arising from the ashes of the former Yugoslavia.

An additional and related issue is exactly how the ICJ will treat the judgments of the ICTY.²³⁹ Even though the legal conclusions of the ICTY are generally not res judicata for the ICJ, the evidence collected by the Tribunal and subjected to a fair, adversarial process should be given broad deference by the ICJ, and would generally need exceptionally relevant and credible evidence to rebut it.²⁴⁰ Most interesting will be the ICJ’s treatment of the ICTY’s determinations on issues which have both factual and legal components, such as whether a particular person or group of persons possessed genocidal intent.²⁴¹

3Preliminary Conclusion as to Genocide in Bosnia

If the ICJ adopts, either explicitly or implicitly, proof beyond reasonable doubt as the proper standard of evidence in relation to whether genocide actually occurred in Bosnia, as I believe the primary rules on genocide warrant, it is hard to see how Bosnia can manage to prove any instance of genocide except the atrocious Srebrenica massacre. Even though it is my firm belief that Srebrenica is a case of genocide, it must be noted that even the ICTY struggled to reach such a conclusion, and that Srebrenica is in many ways an exceptional case²⁴² in the mass of ethnic cleansing perpetrated in Bosnia, chiefly by the Bosnian Serbs. If the ICJ decides not to adopt rigid evidentiary standards to prove genocide, or to apply a less strict definition of genocide, it is possible that it might find other grave crimes committed in Bosnia, such as those in the Prijedor prison camps, to amount to genocide.²⁴³ It would, however, be wrong for the ICJ to treat the totality of crimes committed in Bosnia as one, single crime of genocide. We can certainly say that there was a genocide in Rwanda, or a genocide during the Holocaust – but we can say that because there was a clear genocidal policy and plan in the commission of these crimes and because the overwhelming number of specific crimes can legally qualify as genocide. But, Bosnia is significantly different as to the facts – again, the one instance of genocide in Bosnia found by the ICTY is Srebrenica. There certainly was a general criminal policy of the Bosnian Serbs (and Belgrade) – a true joint criminal enterprise. But it was nevertheless a policy of forcible territorial acquisition and ethnic violence, and was not clearly a policy of genocide – at least that is this author’s appraisal of the facts. The fact that the ICJ is adjudicating on state responsibility and not on individual criminal responsibility for genocide is, again, irrelevant: it must establish the commission of genocide by applying the principles of international criminal law, as an international criminal court would do, without, of course, assigning individual responsibility. Some sort of proof by induction which would either eliminate or lessen the burden of proving every specific act of genocide is simply unwarranted if neither a general, very clear genocidal plan or policy, nor a demonstrable pattern of specific acts of genocide can be proven beyond reasonable doubt.

Even if Bosnia is unable to prove that genocide other than Srebrenica was committed during the conflict, it need not prove genocide at all for some of the inchoate crimes under the Genocide Convention, such as attempt, conspiracy and direct and public incitement. However, severe evidentiary difficulties will still remain.

DAttributing the Acts of the Republika Srpska to Serbia

After establishing the instances in which genocide was committed in Bosnia by the Bosnian Serbs, the ICJ must decide whether the acts of the Republika Srpska are attributable to Serbia. As stated previously, Bosnia does not need to show that the Serbian leadership had genocidal intent. What it needs to show is the control Serbia had over those having such intent and committing genocide or some of the other acts prohibited by the Genocide Convention.²⁴⁴ Additionally, while there are primary rules which warrant a strict standard of evidence regarding the proof of whether genocide was in fact committed, no such standard exists as to the facts pertaining to the relationship between Serbia and the RS. In other words, Bosnia does not need to persuade the ICJ judges beyond reasonable doubt that Serbia actually controlled the RS, or any particular military operation, and the judges are free to adopt a strict evidentiary standard or not do so at all, though a requirement of convincing and credible evidence would be for the best.

As discussed earlier, there are two separate bases of imputability which could apply in this case. Serbia could be responsible under Article 4 of the ILC Articles for the acts of the RS forces as its de facto organs – for this type of responsibility, it would be necessary for Bosnia to show complete control of Serbia over the RS under the first Nicaragua test – in essence the same kind, if not degree, of control Serbia has over its own de jure organs. If Bosnia can prove this relationship of total control and dependency, all of the acts of the RS would be attributable to Serbia, even those committed in operations not directly controlled by Belgrade or against its explicit instructions.²⁴⁵

Alternatively, if Bosnia cannot satisfy the demanding test of complete control, Serbia could still be responsible under Article 8 of the ILC Articles for the acts of the RS forces committed in the course of operations over which Serbia had effective control, under the second Nicaragua test.

Historical scholarship is generally divided as to the precise amount of control Serbia exercised either over the Bosnian or the Croatian Serbs. Evidence available to the general public is of such nature that reasonable persons can disagree over whether a relationship of complete control on the one side and total dependency on the other existed at the relevant time.

Some of the most important facts which would support the thesis of a relationship of complete control between Serbia and the Bosnian Serbs would be:

• The enormous amount of financial and logistical support given by Serbia to the RS;

• Unlike in Nicaragua or in Congo v. Uganda, it can actually be said that the Army of the RS (VRS) was in fact ‘created’ by the (former) Yugoslav National Army (JNA), which later became the FRY’s army (VJ). The fact of creation was deemed very relevant in both of the ICJ cases.²⁴⁶ However, both of these cases could be distinguished as to the facts, as the circumstances of a forcible break-up of a federal state are not the same as those of one state simply intervening in another;

• Many of the VRS officers and generals, including Mladić and Krstić, were at the same time generals of the Yugoslav Army, received their salaries and promotions from Belgrade through the so-called 30th Personnel Centre, and after the end of the conflict actually retired in Serbia.²⁴⁷ News reports indicate, for instance, that the decree of retirement of General Mladić was actually signed by FRY President Koštunica, while the family of General Krstić was given an apartment by the Army. One could say that this is not only proof of responsibility and control, but also proof of remarkable stupidity;

• Serbia did not only intervene in Bosnia through the RS. It had also sent, or had allowed the sending of numerous paramilitary (or ‘volunteer’) units, most of which had direct links with either the military or the state security agencies/secret police. The most notorious of these where the Serb Volunteer Guard, commanded by Željko Ražnatović ‘Arkan’;²⁴⁸ the ‘Šešeljevci’, a group commanded by the Serbian Radical Party and its leader Vojislav Šešelj;²⁴⁹ the Red Berets, which were later made official as the Special Operations Unit ( JSO) and a part of the secret police, commanded by Franko Simatović, later deputy head of Serbian state security police;²⁵⁰ and the ‘Scorpios’, a unit again tied with the Serbian police, whose members are shown on a recent graphic video while killing several Bosniak youths from Srebrenica, as filmed by the executioners themselves.²⁵¹

Some of the most important facts which would support the thesis of the relationship between Serbia and the Bosnian Serbs being one of alliance, albeit with Serbia being the senior partner would be:

• There is no evidence of the VRS or the RS political leadership taking orders from Belgrade. However, they might not have needed to, as the final political goal was the same, and as Belgrade put all possible resources at their disposal. This raises the significant issue of whether the proof of complete control requires that it be shown that it was in fact exercised at some point;²⁵²

• There is some evidence that the leadership of the RS openly defied Serbia, such as the public humiliation of Miloševićc when the Bosnian Serbs refused to accept the Vance-Owen and the Contact Group peace plans he was advocating, and the subsequent (limited) sanctions imposed by Milošević against the RS.²⁵³ However, some authors have suggested that these sanctions were in fact a sham designed to fool the West.²⁵⁴

• The RS was not a simple para-military group or even a classical insurrectional movement. It was organized as a territorial, para-statal entity, though it had never achieved statehood, primarily because of the lack of recognition by the international community. Its political leaders were elected at popular elections, not appointed by Serbia, it collected taxes, ran a bureaucracy and a judicial system, and continues to do so to this very day. All this suggests that by its very nature the RS had more independence than a mere para-military group.

As already stated, the issue of whether the Bosnian Serbs/RS were under complete control of Serbia and therefore its de facto organs is a complex one, on which objective, reasonable persons can disagree. This is foremost the result of a (deliberate) lack of specific, reliable evidence. Milošević, now dead, obviously denied that he had such control,²⁵⁵ while Radovan Karadžić and Ratko Mladić are still fugitives from justice. The testimony of other high-level Bosnian Serb officials, such as Biljana Plavšić, can support either thesis. My view is that complete control probably existed initially, almost certainly in 1992, but that it lessened over time, and that by 1995 Serbia did not have the kind of influence over the RS as it had before. This is, however, completely open to debate.²⁵⁶

In the case the Court does not find that Serbia had complete control over the RS in the relevant period, then it is incumbent upon Bosnia to prove that Serbia had effective control over a particular operation during which an act of genocide was committed – Srebrenica first of all. The Krstić judgment seems to support the view that Belgrade was not directly involved in the extermination of the Muslim men of Srebrenica, but some other evidence, such as the recent video, shows forces arguably under Serbia’s control, or which at least operated from Serbia directly participating in the massacre, albeit not in the town of Srebrenica itself.

EPreliminary Remarks on the State Responsibility of Serbia

As has been shown, the issue of state responsibility of Serbia for the commission of genocide is factually a highly complex one, and will provide extensive probative difficulties to the Bosnian side. However, state responsibility of Serbia for complicity in genocide might be easier to prove, as the intense involvement of various paramilitary groups in the crimes of the Bosnian war is much easier to attribute to Serbia. Genocide must still be proven, though, as must be the knowledge of Serbia’s agents of the genocidal intent of the genocidaires themselves.²⁵⁷ Serbia might also be held liable for direct and public incitement to genocide, though it would seem at first glance that most of the venom spewed by Belgrade’s propaganda machinery did not reach the qualitative level (especially the directness) of the genocidal propaganda in Rwanda or Nazi Germany.

Serbia certainly should be held responsible for failing to prevent genocide and failing to punish genocide, even if the ‘only’ instance of genocide in Bosnia is the Srebrenica massacre. The content of the duty to prevent genocide is unclear, however, as are the consequences for a state which fails to do so, and it is incumbent upon the ICJ to provide some clarity on the matter: the most reasonable course would be to decide on a breach of a due diligence obligation of Serbia not to allow paramilitary forces to enter Bosnia and commit criminal acts, if it could be reasonably expected that these acts would amount to genocide. This type of a due diligence obligation arising from Article I of the Genocide Convention would be similar in nature to the positive obligation of states to secure human rights to persons within their jurisdiction found in human rights treaties. The distinction between the state obligation to prevent genocide and its duty not to be complicit in genocide under Article III of the Convention is somewhat difficult to draw, except by saying that complicity in genocide implies some sort of positive involvement in the commission of the crime, while failing to prevent genocide is an act of omission.

On the other hand, Serbia is undoubtedly guilty of sheltering genocidaires such as Mladić, who is still at large, and his VRS henchmen, and has absolutely no reasonable legal defence for such behaviour.²⁵⁸ Serbia, of course, remains internationally responsible for every single act of intervention in the Bosnian conflict, but the ICJ unfortunately lacks jurisdiction to adjudicate on these matters.

6Conclusion

This article has tried to tackle issues specific to state responsibility for genocide, as well as problems of state responsibility that are not so specific, but which are likely to arise in almost all cases involving the responsibility of a sovereign state for the crime of crimes. But, we should bear in mind that this appellation of genocide as the ‘crime of crimes’ is in some ways misleading: as stated before, genocide is not inherently worse than crimes against humanity – it is simply different. Even though the word ‘genocide’ carries special moral opprobrium, it is far more fitting to extend this same moral condemnation to the concept of crimes against humanity, than to fuse these two concepts legally. The specificity of genocide lies, as we have seen, in the extreme mental element of the crime itself, and it is necessary to maintain this distinction before the ICJ in the same way as before international criminal tribunals, in order to prevent any further fragmentation and inconsistency in international law. For the most part, the methodological principles stated in this article would apply not only to state responsibility for genocide, but also to state responsibility for other grave international crimes. What crimes against humanity lack, and this is also a unique characteristic of genocide, is the compromisory clause in Article IX of the Genocide Convention and the ensuing possibility of bringing a genocide case before the ICJ.

As stated repeatedly in this article, it is positively crucial to maintain the methodological distinction between primary and secondary rules when analysing a state’s responsibility for genocide, in accordance with the codification effort of the ILC. This distinction is not a panacea – it will not solve every legal problem of state responsibility for genocide, but as this article shows it will solve a great deal, and it would at this point be useful to summarize some of the most relevant conclusions:

• State responsibility for genocide is not by its nature criminal, though this concept does not divest genocide of its nature as a crime under international law, and individual criminal responsibility runs concurrently with state responsibility. The aggravated regime of state responsibility for serious breaches of peremptory norms of international law, which does apply to genocide, is by itself of little practical value.²⁵⁹

• It is not necessary for genocidal intent to be attributed to a state in any special way, as the issue of fault is a matter of primary rules. Genocidal intent must be proven beyond reasonable doubt, but only in relation to the actual perpetrators of genocide. The sole dispositive element of a state’s responsibility for genocide is its control over those who commit it. Not only is this the methodologically right thing to do, but it is also the most practicable: proving genocidal intent of high-ranking state leadership with any reasonable degree of certainty would often be an impossible task, because of the frequent lack of any direct evidence. Milošević would be a case in point.²⁶⁰

• States can be held responsible not only for the commission of genocide, but also for breaches of ancillary obligations under the Genocide Convention: failing to prevent or punish genocide; conspiring, directly and publicly inciting or attempting to commit genocide, and for being complicit in genocide. State responsibility for complicity would seem to be most practicable, while state responsibility for conspiracy and attempt would be largely theoretical. State responsibility for violations of obligations flowing from the Genocide Convention does not exist in a vacuum: general principles of international criminal law are a part of primary law which should be applied, while other principles of international law, especially those on obligations erga omnes and possible remedies are a part of the relevant secondary rules.²⁶¹

• If the responsibility of a state is invoked for acts not committed by its de jure organs, the state can most often be held responsible either under Article 4 or under Article 8 of the ILC Articles.

• The acts of a non-state actor can be attributed to a state under Article 4 if the actor is under the state’s complete control, as contemplated by paragraphs 109 and 110 of the ICJ’s Nicaragua judgment. If such control exists, the non-state actor becomes a state organ de facto, and all of its acts, even those committed against explicit state instructions, would be attributable to the state. The practical implication of attribution under Article 4 is that it would not be necessary to prove the state’s involvement in a specific operation during which genocide was committed by its de facto organ.

• It would be necessary, however, to prove the state’s involvement in a specific operation if the degree of control required under Article 4 cannot be proven, and attribution is therefore sought under Article 8 of the ILC Articles. In that case, a state’s instructions, direction or control over a specific operation must be shown, under the ICJ’s effective control test from paragraph 115 of the Nicaragua judgment. The ICTY’s overall control test from Tadić is simply wrong as a matter of state responsibility, though it may be correct if used for its primary purpose, the qualification of an armed conflict under international humanitarian law. A state’s responsibility for genocide also cannot occur if the state merely harbours genocidaires, and does not control them. However, harbouring could lead to the state’s responsibility for failing to prevent and punish genocide and/or for complicity in genocide.²⁶²

• Though it is necessary to prove genocide beyond reasonable doubt, it is not necessary to prove the facts relevant for assessing the relationship of control between the state and the non-state actor beyond reasonable doubt. However, reliable and credible evidence should still be required.²⁶³

The Bosnia v. Serbia case currently before the Court shows that the primary difficulties which the ICJ will face when dealing with state responsibility for genocide are not strictly legal, if the Court follows the proper methodology, but are ones of facts and evidence. Not only proving genocide, but also proving the link between the genocidaires and the political leadership of a state is a long and arduous process. It is difficult for the ICTY which is equipped with an enormous budget and a formidable investigatory apparatus to deal with these issues, let alone the ICJ. It is almost paradoxical that Bosnia might have done better if it had waited a little while longer (considering that the case took 13 years before it reached the oral hearings) for more of the proceedings before the ICTY to come to an end. Any judgment against Perišić,²⁶⁴ the former Chief of Staff of the Yugoslav Army, Serbian secret police chief Stanišić or in the new, consolidated Srebrenica case²⁶⁵ would have been useful for the ICJ, much less any judgments against Karadzcaron;ić and Mladić, when they are apprehended, or against Milošević, which will now unfortunately never come to pass. It is disappointing, to say the least, that the principal architects of the bloody downfall of the former Yugoslavia, first and foremost Milošević, but also the Croatian president Tuđman and the Bosniak president Izetbegović, have died without criminal punishment. Yet, some progress has certainly been made in what is now the new, post-Cold War system of international criminal justice.

But the old system of state responsibility under international law is still very much alive, and it is especially relevant to genocide. Genocide is indeed a state crime: there is not a single instance of genocide in recorded history which was not committed either directly by a state, or by a state through one of its proxies. State responsibility for genocide can therefore serve a remedial purpose, by alleviating some of the persisting consequences of atrocities through compensation to the victims, but it can also have a wider, systemic purpose, by showing that genocide is indeed a crime committed by states, which goes against the public order of international law.²⁶⁶ Even though such state responsibility remains ‘civil’, it must be emphasized that the consequences of a serious breach of a peremptory norm of international law are not exhausted by the regime of state responsibility: they can, and should, provoke a much wider, institutional reaction, such as Chapter VII action by the Security Council or enforcement actions by regional organizations.²⁶⁷ Whether states actually choose to invoke the responsibility of another state for genocide is a different matter, as is whether the permanent members of the Security Council will actually intervene to stop a genocide when their own interests are involved.²⁶⁸ The world failed to stop the genocide in Rwanda, as it is failing now with the ongoing massacres in Darfur. But, as stated at the beginning of this article, this is not a problem the law can solve by itself: the true solution requires a fundamental moral commitment and not pages upon pages of legal analysis.

And so we are again at a historic moment of international law: as Milošević was the first head of state on trial for genocide, so is Serbia now the first state to be so tried by the ICJ. And even though these new developments in international law might not provide much deterrence, they may still provide at least some justice, and that is far, far better than anything we have had until now.

From September 2006 to May 2007 the author will serve as law clerk to Judge Thomas Buergenthal of the International Court of Justice.

1

Hereinafter ICJ, or the Court.

2

Application of the Convention on the Prevention and Punishment of the Crime of Genocide (Bosnia and Herzegovina v. Serbia and Montenegro), General List No. 91 [hereinafter Bosnia v. Serbia or the Genocide case. Oral and written pleadings in the case are available at http://www.icj-cij.org/icjwww/idocket/ibhy/ibhyframe.htm.

3

Hereinafter Bosnia.

4

Formerly the Federal Republic of Yugoslavia (FRY) [hereinafter Serbia].

5

See, e.g., S.R. Ratner and J.S. Abrams, Accountability for Human Rights Atrocities in International Law: Beyond the Nuremberg Legacy (2nd edn., 2001), at 187–225.

6

See, e.g., Ratner and Abrams, supra note 5, at 3–14; Dupuy, ‘International Criminal Responsibility of the Individual and International Responsibility of a State’, in A. Cassese, P. Gaeta and J.R.W.D. Jones (eds), The Rome Statute of the International Criminal Court: A Commentary, vol. 2 (2002), at 1085 et seq.

7

Of course, the modalities of compensation, and especially those of private as opposed to state claims remain controversial to this day. See, e.g., Buxbaum, ‘A Legal History of International Reparations’, 23 Berkeley J. Int’lL. (2005) 314. The bulk of compensation to individuals (more than $50 billion) went through the negotiation processes of the Conference on Jewish Material Claims Against Germany, see at http://www.claimscon.org/index.asp.

8

M. Koskenniemi, The Gentle Civilizer of Nations: The Rise and Fall of International Law 1870–1960 (2002).

9

See R. Lemkin, Axis Rule in Occupied Europe (1944, a 2005 reprint is also available); see also W.A. Schabas, Genocide in International Law: The Crime of Crimes (2000), at 14–50; S. Power, A Problem from Hell: America and the Age of Genocide (2002), at 1–85.

10

See, e.g., Power, supra note 9, at 358–364.

11

Some domestic legal systems may have adopted a broader definition of genocide, which differs from the one in international law. See, e.g., the order of the investigative judge of the Spanish Audiencia Nacional in the Pinochet case, No. 1/98, 5 Nov. 1998, where the definition of genocide was interpreted as encompassing political and other social groups, as the Spanish criminal legislation did not include crimes against humanity or use them as a basis for universal jurisdiction. See Carrasco and Fernandez, ‘In re Pinochet’, 93 AJIL (1999) 690. See also the judgments of German courts in the case of Bosnian Serb Nikola Jorgić, Oberlandesgericht Düsseldorf, 2 StE 8/96, judgement of 26 September 1997, Bundesgerichtshof, judgment of 30 April 1999; Bundesverfassungsgericht, judgment of 12 December 2000.

12

Convention on the Prevention and Punishment of the Crime of Genocide, 9 December 1948, 78 UNTS 277.

13

UN GA Res. 96 (I), 11 December 1946: ‘The victim of the crime of genocide is the group itself.’ See also Prosecutor v. Jelisić, IT-95–10, Trial Chamber judgment of 14 December 1999, para. 79. ICTY judgments can be found at http//:www.un.org/icty/.

14

See Schabas, supra note 9, at 102–150.

15

Even if the list of protected groups is exhaustive, it still needs to be determined how to define a national, ethnic, racial or religious group. In most cases this is relatively easy. In the case of Rwanda, however, the ICTR had to establish whether the Tutsi could be subsumed under one of these categories, and encountered great difficulties arising from the relatively artificial distinction made between the Tutsi and the Hutu. The ICTR therefore oscillated from an objective approach (i.e., whether ethnic or racial groups, for instance, are defined by certain objective characteristics), to more subjective (or more anthropological) approaches (i.e., how the group identifies itself, or how the perpetrator identifies the group). See, e.g., Prosecutor v. Rutaganda, ICTR-96-3, Trial Chamber judgment of 6 Dec. 1999; Appeals Chamber judgment of 26 May 2003. ICTR judgments can be found at http//:www.ictr.org.

16

See Van Schaack, ‘The Crime of Political Genocide: Repairing the Genocide Convention’s Blind Spot’, 106 Yale LJ (1997) 2259.

17

See also Ratner and Abrams, supra note 5, at 42–45.

18

See, e.g., A. Cassese, International Criminal Law (2003), at 64–95; Ratner and Abrams, supra note 5, at 69–79.

19

See, e.g., Ratner and Abrams, supra note 5, at 267–283.

20

See the discussion and an overview of the literature in Ratner and Abrams, supra note 5, at 284–290.

21

I do not share the concern expressed by Schabas that expanding the definition of genocide might trivialize the offence, by basically equating it with any systematic act of mass murder. See Schabas, supra note 9, at 150. That is precisely how lay persons actually see genocide. Admittedly, the fact that the killings of Cham Muslims or ethnic Vietnamese in Cambodia might legally qualify as genocide, while the (more numerous) killings of ethnic Khmer who were targeted by other ethnic Khmer because of their social status might not so qualify, could to the average lay person seem completely bizarre. Policy considerations aside, this distinction is nevertheless what the law requires.

22

A strict conception of genocidal intent and the ensuing distinction between genocide and crimes against humanity have been unambiguously accepted in the jurisprudence of the two ad hoc tribunals, particularly the ICTY. See Tournaye, ‘Genocidal Intent Before the ICTY’, 52 ICLQ (2003) 447.

23

See, e.g., Article 7 of the Rome Statute of the ICC; Article 5 of the Statute of the ICTY.

24

See, e.g., Prosecutor v. Krstić, IT-98–33, Appeals Chamber judgment of 19 April 2004, para. 25. So-called cultural genocide was explicitly excluded from the scope of the Genocide Convention during its drafting process. See Schabas, supra note 9, at 152 et seq.; Ratner and Abrams, supra note 5, at 31. However, these acts are again still punishable under international law as crimes against humanity, and can, in the right circumstances, be evidence of the intent to physically destroy a group and thus be probative of genocide stricto sensu. See Ratner and Abrams, supra note 5, at 36.

25

This has also been confirmed by the ICJ in its Advisory Opinion on the Legality of the Threat or Use of Nuclear Weapons, 8 July 1996, para. 26: ‘It was maintained before the Court that the number of deaths occasioned by the use of nuclear weapons would be enormous; that the victims could, in certain cases, include persons of a particular national, ethnic, racial or religious group; and that the intention to destroy such groups could be inferred from the fact that the user of the nuclear weapon would have omitted to take account of the well-known effects of the use of such weapons. The Court would point out in that regard that the prohibition of genocide would be pertinent in this case if the recourse to nuclear weapons did indeed entail the element of intent, towards a group as such, required by the provision quoted above. In the view of the Court, it would only be possible to arrive at such a conclusion after having taken due account of the circumstances specific to each case.’ (emphasis added)

26

See Prosecutor v. Kayishema and Ruzindana, ICTR-95-1, Trial Chamber judgment of 21 May 1999, para. 93: ‘The Chamber finds that the intent can be inferred either from words or deeds and may be demonstrated by a pattern of purposeful action. In particular, the Chamber considers evidence such as the physical targeting of the group or their property; the use of derogatory language toward members of the targeted group; the weapons employed and the extent of bodily injury; the methodical way of planning, the systematic manner of killing. Furthermore, the number of the victims from the group is also important.’

27

See Prosecutor v. Stakić, IT-97-24, Appeals Chamber judgment of 22 March 2006, paras. 43–45.

28

See, e.g., Rosenstock, ‘The ILC and State Responsibility’, 96 AJIL (2002) 792.

29

See, e.g., A Cassese, International Law (2nd edn., 2005), at 241–243.

30

See also Matsui, ‘The Transformation of the Law of State Responsibility’, 20 Thesaurus Acroasium (1993) 1.

31

The UN General Assembly took note of the Articles by its Resolution 56/83 of 12 December 2001. It has not yet been conclusively decided whether the Articles will be formally adopted by treaty, in the vein of the Vienna Convention on the Law of Treaties, or whether they will simply be allowed to germinate in state and international judicial practice. See Crawford and Olleson, ‘The Continuing Debate on a UN Convention on State Responsibility’, 54 ICLQ (2005) 959. See also Caron, ‘The ILC Articles on State Responsibility: The Paradoxical Relationship Between Form and Authority’, 96 AJIL (2002) 857.

32

ILC Commentaries to the Draft Articles, extract from the Report of the International Law Commission on the work of its Fifty-third session, Official Records of the General Assembly, Fifty-sixth session, Supplement No. 10 (A/56/10), chp.IV.E.2), available at http://untreaty.un.org/ilc/texts/instruments/english/commentaries/9_6_2001.pdf, at 59–60.

33

See Cassese, supra note 29, at 244–245.

34

See also M.N. Shaw, International Law (5th edn., 2003), at 698 et seq.

35

See Cassese, supra note 29, at 251. Cassese rightly notes that responsibility under Articles 17 and 18 of the Draft Articles entails an element of knowledge.

36

See Allott, ‘State Responsibility and the Unmaking of International Law’, 29 Harv. Int’l. LJ (1988) 1.

37

For an excellent discussion on the functional nature of the distinction, see Crawford, ‘The ILC’s Articles on Responsibility of States for Internationally Wrongful Acts’, 96 AJIL (2002) 874, 876–880.

38

See more Nollkaemper, ‘Concurrence Between Individual Responsibility and State Responsibility in International Law’, 52 ICLQ (2003) 615.

39

Ibid., 622–624.

40

The Draft Articles adopted on the first reading in 1996 are available at http://lcil.law.cam.ac.uk/ILCSR/articles_1996(e).doc.

41

See, e.g., the Seventh Report on State Responsibility by Prof. Arangio-Ruiz, UN Doc. A/CN.4/469, 9 May 1995, available at http://daccessdds.un.org/doc/UNDOC/GEN/G95/612/34/PDF/G9561234.pdf?OpenElement.

42

At least insofar as they considered that international crimes, even if they are a viable concept, do not entail a separate regime of state responsibility. See, e.g., the First Report on State Responsibility by Prof. Crawford, UN Doc. A/CN.4/490, 24 April 1998, esp. para. 81. ILC reports and discussions on state responsibility can be found online at http://untreaty.un.org/ilc/guide/9_6.htm .

43

See, e.g., the remarks of the United Kingdom, United States, France and Germany. For state opinions, see at http://lcil.law.cam.ac.uk/ILCSR/Statresp.htm#Comments. For a full overview of state comments to Draft Article 19, see International Law Commission, First Report on State Responsibility by Prof. James Crawford, Addendums 1 and 2, A/CN.4/490/Add.1&2, 1 May 1998.

44

See, e.g., J.H.H. Weiler, A. Cassese and M. Spinedi, International Crimes of State : A Critical Analysis of the ILC’s Draft Article 19 on State Responsibility (1989); Rosenne, ‘State Responsibility and International Crimes: Further Reflections on Article 19 of the Draft Articles on State Responsibility’, 30 NYU J. Int’lL. & Pol. (1997) 145; Abi-Saab, ‘The Uses of Article 19’, 10 EJIL (1999) 339; Gaja, ‘Should All References to International Crimes Disappear from the ILC Draft Articles on State Responsibility?’, 10 EJIL (1999) 365; Pellet, ‘Can A State Commit a Crime? Definitely, Yes!’, 10 EJIL (1999) 425.

45

See Schabas, supra note 9, at 418–424.

46

Only acts of individuals committed in their official capacity can in principle be attributed to a state. See, e.g., the ILC Commentary to the Draft Articles, at 84 et seq. Unfortunately, the ICJ has recently remarked that individuals entitled to immunities under international law can be tried for acts committed in their private capacity when their immunity expires, which would lead to the conclusion that they can never be tried for acts committed in their official capacity, i.e. that state responsibility and individual criminal responsibility could not run concurrently. See Case Concerning the Arrest Warrant of 11 April 2000 (Congo v. Belgium), Judgment of 14 February 2002, para. 61. To deal with this issue, a rather sordid fiction was developed by some courts and authors when applying the international law of immunities, the claim being that acts such as crimes against humanity or genocide can only be committed in a private capacity for the purpose of waiving immunity, which begs the question whether such acts are still ‘official’ for the purposes of state responsibility. Such a counterintuitive proposition is unfortunately supported by a reading of the ICJ’s dictum in the Arrest Warrant case. See also the discussion of Spinedi, ‘State Responsibility v. Individual Responsibility for International Crimes: Tertium Non Datur’, 13 EJIL (2002) 895.

47

See, e.g., Abi-Saab, supra note 44, 347 et seq. See also Schabas, supra note 9, at 441–446.

48

See generally N.H.B. Jørgensen, The Responsibility of States for International Crimes (2000). See also van der Vyver, ‘Prosecution and Punishment of the Crime of Genocide’, 23 Fordham Int’lLJ (1999) 286, at 295–298. For an outline of a more cautious approach, see Spinedi, ‘La Responsabilité de l’état pour “crime”: une responsabilité pénale?’ in H. Ascensio, E. Decaux and A. Pellet (eds), Droit international pénal (2000), at 91–114.

49

Even if the response to violations of these basic principles still remains rather traditional. See Gattini, ‘A Return Ticket to “Communitarisme”, Please’, 13 EJIL (2002) 1181 .

50

Pursuant to Article 41 all states have a positive duty to cooperate in order to bring to an end a serious breach of a peremptory norm, and a negative duty not to recognize the situation created by the breach as lawful. See also Shelton, ‘Righting Wrongs: Reparations in the Articles on State Responsibility’, 96 AJIL (2002) 833, at 841–844.

51

Cassese rightly points out that the role of the regime of aggravated state responsibility is relatively minor. What is necessary is the establishment of institutional mechanisms for dealing with serious breaches of fundamental norms of international law. See Cassese, supra note 29, at 269–277. See also Klein, ‘Responsibility for Serious Breaches of Obligations Deriving from Peremptory Norms of International Law and United Nations Law’, 13 EJIL (2002) 1,241, Tams, ‘Do Serious Breaches Give Rise to Any Specific Obligations of the Responsible State?’, 13 EJIL (2002) 1,161.

52

The US government has relatively consistently asserted that the crimes committed in Darfur can legally be qualified as genocide – see, e.g., the statement of then Secretary of State Colin Powell of 9 September 2004 at http://www.washingtonpost.com/wp-dyn/articles/A8364-2004Sep9.html and http://edition.cnn.com/2004/WORLD/africa/09/09/sudan.powell/. However, according to the UN Commission of Inquiry, established pursuant to SC Res. 1564, the atrocities in Darfur do not amount to genocide because of the lack of evidence of specific intent. The Commission’s Report of 25 January 2005 is available at http://www.un.org/News/dh/sudan/com_inq_darfur.pdf. See also G. Prunier, Darfur: The Ambiguous Genocide (2005); Schabas, ‘Darfur and the “Odious Scourge”: The Commission of Inquiry’s Findings on Genocide’, 18 Leiden JIL (2005) 871.

53

Case Concerning the Barcelona Traction, Light and Power Co., Ltd., ICJ Reports (1970), at 3 et seq, paras 33 and 34.

54

See B. Simma, ‘International Human Rights and General International Law: A Comparative Analysis’, Collected Courses of the Academy of European Law (1994, IV, 2) , at 195–200.

55

If it decides to treat the developments of erga omnes obligations as modifying its traditional rules on standing. Though the Barcelona Traction dictum can certainly be interpreted that way, states have not brought breaches of erga omnes obligations before the ICJ if they were not the directly injured state. In any case, the concept of erga omnes obligations does not affect the consensual nature of the ICJ’s jurisdiction, which must be established independently. See more, C.J. Tams, Enforcing Obligations Erga Omnes in International Law (2005), at 158–197.

56

See Articles 42 and 48 of the ILC Articles. See also Tams, supra note 55; Brown Weiss, ‘Invoking State Responsibility in the Twenty-First Century’, 96 AJIL (2002) 798.

57

See especially the most recent Case Concerning Armed Activities on the Territory of the Congo (New Application: 2002), (Democratic Republic of the Congo v. Rwanda), Judgment on Jurisdiction and Admissibility, 3 February 2006, para. 64 et seq.

58

As is arguably the case in Bosnia v. Serbia: one of the Bosnian claims is that Serbia failed to abide by the Court’s provisional measures.

59

ILC Commentaries, at 323.

60

See the Resolution on obligations erga omnes in international law, adopted by the Institut de droit international at its Krakow Session on 27 August 2005.

61

Military and Paramilitary Activities in and Against Nicaragua (Nicaragua v. United States), Judgment (merits), 27 June 1986, ICJ Reports (1986), at 14.

62

Case Concerning Armed Activities on the Territory of the Congo (Democratic Republic of the Congo v. Uganda), Judgment (merits), 19 December 2005.

63

Ibid., at paras 181–221.

64

Genocide case, Judgment (preliminary objections) of 11 July 1996.

65

I.e., obligations to enact legislation and prosecute offenders. See more infra.

66

Ibid., at para. 32: ‘[t]he Court now comes to the second proposition advanced by Yugoslavia, regarding the type of State responsibility envisaged in Article IX of the Convention. According to Yugoslavia, that Article would only cover the responsibility flowing from the failure of a State to fulfill its obligations of prevention and punishment as contemplated by Articles V, VI and VII; on the other hand, the responsibility of a State for an act of genocide perpetrated by the State itself would be excluded from the scope of the Convention. The Court would observe that the reference in Article IX to “the responsibility of a State for genocide or for any of the other acts enumerated in Article III”, does not exclude any form of State responsibility. Nor is the responsibility of a State for acts of its organs excluded by Article IV of the Convention, which contemplates the commission of an act of genocide by “rulers” or “public officials”.’ Three members of the Court did, however, entertain doubts as to whether state responsibility can exist concurrently with individual criminal responsibility. See the Joint Declaration of Judges Shi and Vereshchetin and the Declaration of Judge Oda.

67

Verbatim record of the public sitting held on Monday 13 March 2006 at 15.00, President Higgins presiding, paras 297–304.

68

See Schabas, supra note 9, at 418–424.

69

See, e.g., in Jørgensen, supra note 48, at 32–41.

70

See Art. 32, Vienna Convention on the Law of Treaties, 1155 UNTS (May 23, 1969) 311. Though the Vienna Convention as such does not apply to the interpretation of the Genocide Convention which preceded it, the rules of Art. 32 do reflect customary law.

71

See at http://www.unhchr.ch/html/menu3/b/treaty1gen.htm and http://www.preventgenocide.org/law/convention/reservations/.

72

Reservations to the Convention on the Prevention and Punishment of the Crime of Genocide, Advisory Opinion of 28 May 1951, ICJ Reports (1951), at 22.

73

See, e.g., Human Rights Committee, General Comment No. 24(52) Relating to Reservations, 2 Nov. 1993; F. Hampson, Reservations to Human Rights Treaties, working paper pursuant to Sub-Commission decision 1998/113, Sub-Commission on Prevention of Discrimination and Protection of Minorities, Fifty-first session, UN Doc. E/CN.4/Sub.2/1999/28 (28 June 1999); Tenth Report on Reservations to Treaties by Mr. Alain Pellet, Special Rapporteur, International Law Commission, Fifty-seventh session, UN Doc. A/CN.4/558/Add.2 (30 June 2005).

74

Case Concerning Armed Activities on the Territory of the Congo (New Application: 2002), (Democratic Republic of the Congo v. Rwanda), Judgment on Jurisdiction and Admissibility, 3 February 2006.

75

Ibid., paras 64–70.

76

See, e.g., Legality of Use of Force (Serbia and Montenegro v. Belgium), General List No. 105, Judgment of 15 December 2004. In these cases the ICJ decided by 8 votes to 7 that it did not have jurisdiction ratione personae to entertain the application of Serbia and Montenegro, as it was not a member state of the United Nations at the time the alleged violations of international law occurred.

77

See, e.g., the oral pleadings of Professor Ian Brownlie on behalf of Serbia, Verbatim record of the public sitting of the Court held on 16 March 2006, at 10 a.m., President Higgins presiding, CR 2006/21.

78

See also Schabas, supra note 9, at 444.

79

See also Nollkaemper, supra note 38, at 633.

80

For a discussion on the place of fault in the ILC’s codification effort see Gattini, ‘Smoking/No Smoking: Some Remarks on the Current Place of Fault in the ILC Draft Articles on State Responsibility’, 10 EJIL (1999) 397. See also Crawford, ‘Revising the Draft Articles on State Responsibility’, 10 EJIL (1999) 436.

81

See, e.g., the ILC Commentaries to the Draft Articles, at 69–73, where the ILC Special Rapporteur clearly mentions genocide as an example, and deals with the issue of fault.

82

See Article 7 of the ILC Articles.

83

Dupuy argues that it would be necessary to show the genocidal intent of the government official who actually orders the commission of genocide to establish state responsibility for genocide. See Dupuy, supra note 6, at 1095–1096. This may be true, but proving a direct order or instruction allows for almost an automatic inference of genocidal intent, whatever its relevance. In a great number of cases, however, it will be impossible to prove a direct order by the state to a non-state actor acting as its proxy, and it is precisely these cases which show that it is not necessary to prove the genocidal intent of a government, but merely its control of a non-state actor.

84

Held on 20 January 1942 in the Berlin suburb of Wannsee, at which high-ranking Nazi officials discussed the ‘Final Solution of the Jewish Question’. The minutes of the Conference (the Wannsee Protocol), drafted by Adolf Eichmann, have been saved and presented as evidence at the Nuremberg trials and are available at http://www.writing.upenn.edu/~afilreis/Holocaust/wansee-transcript.html, and have also been complemented by Eichmann’s testimony at his 1961 trial in Jerusalem. See more H. Arendt, Eichmann in Jerusalem: A Report on the Banality of Evil (1994); C.R. Browning, The Origins of the Final Solution (2005), esp. at 398–415.

85

It should, of course, be noted that the idea of the creation of a ‘Greater Serbia’ was not a coherent, unified programme but a jumbled mix of different approaches and nationalistic ideologies – from pan-Slavism to romantic nationalism to Serbian chauvinistic nationalism of the end of the 20th century. See, e.g., J.R. Lampe, Yugoslavia as History – Twice There Was a Country (1996), at 39, 40, 52; S. K. Pavlowitch, Serbia – The History behind the Name (2002), at 91, 95, 99, 108–110; J. Ridgeway and J. Udovički, Burn This House: The Making and Unmaking of Yugoslavia (1997), at 46.

86

Some authors argue that the applicability of the aggravated regime of state responsibility would depend on attributing a criminal act, which would also include intent, to a high-ranking official. See Nollkaemper, supra note 38, at 633. However, whether the aggravated regime applies or not depends solely on the nature of the violated norm, and intent may certainly be an element of the violation, but need not be an element of attribution.

87

See Toufayan, ‘The World Court’s Distress When Facing Genocide: A Critical Commentary on the Application of the Genocide Convention Case (Bosnia and Herzegovina v. Yugoslavia (Serbia and Montenegro)), 40 Tex. ILJ (2005) 233.

88

See also Schabas, supra note 9, at 447–455.

89

See, e.g., M.C. Bassiouni and E.M. Wise, Aut dedere aut judicare: The Duty to Extradite or Prosecute in International Law (1995).

90

The main purpose served by these separate crimes is preventive. See Schabas, supra note 9, at 257.

91

Ibid., at 258.

92

See Cassese, supra note 18, at 196–198.

93

See Schabas, supra note 9, at 265.

94

The Wannsee Protocol and Eichmann’s testimony during his in trial in Jerusalem would be a good example, see supra note 84.

95

Ordinary accessory-type incitement is still punishable under normal rules on complicity. The offence of direct and public incitement does not require the proof of a causal nexus between the act of incitement and the commission of genocide. It is of course quite unlikely that the responsibility of either a state or an individual would be invoked if no genocide had in fact occurred, but the lack of a requirement of proving causality is there simply to alleviate the already heavy evidentiary burden.

96

See Schabas, supra note 9, at 276–280.

97

The entire judgment is available at http://www.yale.edu/lawweb/avalon/imt/proc/judcont.htm. See also at http://www.law.umkc.edu/faculty/projects/ftrials/nuremberg/nuremberg.htm.

98

See at http://www.yale.edu/lawweb/avalon/imt/proc/judstrei.htm.

99

Control Council Law No. 10, Punishment of Persons Guilty of War Crimes, Crimes Against Peace and Against Humanity, 20 December 1945, 3 Official Gazette, Control Council for Germany (1946), at 50–55.

100

Prosecutor v. Nahimana, Ngeze and Barayagwiza, ICTR-99-52-T, Trial Chamber Judgment of 3 December 2003. The judgment is currently on appeal.

101

See ibid., esp. paras 390, 395 and 396. See also Schabas, ‘Hate Speech in Rwanda: The Road to Genocide’, 46 McGill LJ (2000) 142; J.-P. Chrétien (ed.), Rwanda: Les médias du génocide (1995).

102

As direct and public incitement is an inchoate crime.

103

See Schabas, supra note 9, at 280–285.

104

Though aiding and abetting has for peculiar reasons been distinguished from complicity in the jurisprudence of the ad hoc tribunals. Still, as stated by Schabas, it is hard to justify this distinction, as aiding and abetting is a classic form of complicity. See ibid., at 293.

105

But see ibid., at 300–303.

106

See in that regard United Kingdom v. Tesch et al. (‘Zyklon B case’) 1 L. Rep. Tr. War.Crim. (1947) 93, though, of course, Tesch could not have been prosecuted for genocide in 1947.

107

Prosecutor v. Krstić, Appeals Chamber judgment, para. 140 et seq. Prosecutor v. Blagojević and Jokić, IT-02-60, Trial Chamber judgment of 17 January 2005, para. 776 et seq.

108

Attribution under Article 5 of the ILC Articles, which deals with a similar issue, is irrelevant for the purpose of this essay, as it is clearly limited to entities which are empowered by internal law to exercise governmental authority, yet fall short from being an organ under Article 4, such as various public companies or other para-statal entities. See ILC Commentaries, at 94. Attribution under Articles 9–11 will also not be dealt with in this paper, nor will it deal with any other basis for state responsibility other than state control over non-state actors.

109

Though, on occasion, the ample arrogance of these people does get the better of them, and they do make public statements from which their (potential) genocidal intent can be inferred – for example, the statement of General Mladić to the cameras upon Bosnian Serb armed forces taking over the enclave of Srebrenica: ‘the moment has finally come to take revenge on the Turks’, see Prosecutor v. Krstić, Trial Chamber Judgment, para. 336 and n. 890; or the current statements of Iranian President Ahmadinejad about wiping Israel from the map, see at http://edition.cnn.com/2005/WORLD/meast/10/26/ahmadinejad/index.html.

110

Application of the Convention on the Prevention and Punishment of the Crime of Genocide (Croatia v. Serbia and Montenegro), General List No. 118.

111

See, e.g., at http://news.bbc.co.uk/2/hi/africa/3613953.stm.

112

See the 2004 ILC report by Martti Koskenniemi, The Function and Scope of the lex specialis Rule and the Question of ‘Self-contained Regimes’: An Outline, available at http://untreaty.un.org/ilc/sessions/55/fragmentation_outline.pdf.

113

Military and Paramilitary Activities in and against Nicaragua (Nicaragua v. United States), Merits, Judgment, ICJ Reports (1986), at 14.

114

Ibid., at para. 93 et seq.

115

See, e.g., Meron, ‘Classification of Armed Conflict in the Former Yugoslavia: Nicaragua’s Fallout’, 92 AJIL (1998) 236; Chase, ‘Legal Mechanisms of the International Community and the United States Concerning State Sponsorship of Terrorism’, 45 Va. J Int’lL (2004) 41.

116

Nicaragua v. United States, supra note 113, at paras 93 and 94.

117

Ibid., para. 110.

118

Ibid., paras. 109 and 110.

119

Ibid., para. 112.

120

Ibid., para. 114.

121

Ibid., para. 277.

122

Dissenting Opinion of Judge Schwebel, paras. 257–261. See also the Dissenting Opinion of Judge Jennings, para. 538.

123

Separate Opinion of Judge Ago (English translation), paras 14–19.

124

See, e.g., Y. Dinstein, The Conduct of Hostilities under the Law of International Armed Conflict (2004), at 14–16.

125

See, e.g., Greenwood, ‘International Humanitarian Law and the Tadić Case, 7 EJIL 265 (1996) 265, at 275–276.

126

Prosecutor v. Tadić, Trial Chamber judgment, paras 585 et seq.

127

Nicaragua, para. 219.

128

Ibid., at paras. 109 and 115.

129

The very fact that the relevant parts of the Nicaragua judgment are separated by more than a hundred paragraphs is an obvious indication.

130

Prosecutor v. Tadić, Trial Chamber judgment, paras 605–606.

131

Ibid., at para. 607.

132

Separate and Dissenting Opinion of Judge McDonald, at 292.

133

Ibid., at 295–296.

134

Ibid., at 299.

135

The Appeals Chamber did previously examine several quite important jurisdictional aspects of the case on interlocutory appeal, before the Trial Chamber’s judgment was delivered: Prosecutor v. Tadić, Decision on the Defence Motion for Interlocutory Appeal on Jurisdiction, 2 October 1995.

136

Prosecutor v. Tadić, Appeals Chamber judgment, at paras 83–89.

137

Ibid., at para. 97.

138

Ibid., at paras 99 et seq.

139

Ibid., at paras 111 and 112.

140

Ibid., at paras 100, 107–113.

141

Ibid., at paras 115.

142

Ibid., at paras 117.

143

Ibid., at paras 118–120.

144

Ibid., at paras 124 et seq.

145

Ibid., at paras 131: ‘In order to attribute the acts of a military or paramilitary group to a State, it must be proved that the State wields overall control over the group, not only by equipping and financing the group, but also by coordinating or helping in the general planning of its military activity. Only then can the State be held internationally accountable for any misconduct of the group. However, it is not necessary that, in addition, the State should also issue, either to the head or to members of the group, instructions for the commission of specific acts contrary to international law.’ (emphasis added)

146

Ibid., at para. 137.

147

Ibid., at para. 143.

148

See de Hoogh, ‘Articles 4 and 8 of the 2001 ILC Articles on State Responsibility, the Tadić Case and Attribution of Acts of Bosnian Serb Authorities to the Federal Republic of Yugoslavia’, 72 BYbIL (2001) 255, at 290.

149

Ibid., at paras 160–162.

150

Nicaragua, at para. 115.

151

Ibid, at para. 110.

152

Prosecutor v. Tadić, Appeals Chamber judgment, para. 131.

153

See also de Hoogh, supra note 148, at 278–281.

154

See Prosecutor v. Tadić, Appeals Chamber judgment, Separate Opinion of Judge Shahabuddeen, para. 4 et seq.

155

‘1. The conduct of any State organ shall be considered an act of that State under international law, whether the organ exercises legislative, executive, judicial or any other functions, whatever position it holds in the organization of the State, and whatever its character as an organ of the central government or of a territorial unit of the State. 2. An organ includes any person or entity which has that status in accordance with the internal law of the State.’

156

It should be noted that the law of treaties explicitly excludes this possibility – see Article 27 of the Vienna Convention on the Law of Treaties: ‘[a] party may not invoke the provisions of its internal law as justification for its failure to perform a treaty.’

157

ILC Commentaries, at 90–91.

158

See also the statement of 13 August 1998 by Professor Bruno Simma, then Chairman of the ILC Drafting Committee, commenting on then Draft Article 5 (now Article 4), available at http://lcil.law.cam.ac.uk/ILCSR/rft/Simma(98dcrep).rtf, at 5–6: ‘Paragraph 2 [of Article 5] recognizes the significant role played by internal law in determining the status of a person or an entity within the structural framework of the State. This role of internal law is decisive when it makes an affirmative determination that a person or an entity constitutes an organ of the State. The commentary will explain that the term “internal law” is used in a broad sense to include practice and convention. The commentary will also explain the supplementary role of international law in situations in which internal law does not provide any classification or provides an incorrect classification of a person or an entity which in fact operates as a State organ within the organic structure of the State.’

159

See First Report on State Responsibility by James Crawford, Addendum 5, UN Doc. A/CN.4/490, Add. 5, 22 July 1998, at 16 et seq.

160

ILC Commentaries, at 105–107.

161

See International Law Commission, Comments and observations received from Governments, A/CN.4/488, 25 March 1998.

162

In its submissions to the ILC on the first version of the Draft Articles, the United States and the United Kingdom have also objected to the idea of state organs having that status solely under internal law. Other governments made no submissions on the issue. See International Law Commission, Comments and observations received from Governments, A/CN.4/488, 25 March 1998, at 28–29; First Report on State Responsibility, supra note 159, at 7–9.

163

See the Address to the Nation by President Bush on September 11, 2001, available at http://www.whitehouse.gov/news/releases/2001/09/20010911-16.html, who inter alia stated that ‘We will make no distinction between the terrorists who committed these acts and those who harbor them.’

164

Ratner, ‘Jus ad Bellum and Jus in Bello after September 11’, 96 AJIL (2002) 905, at 907.

165

The ICJ’s jurisprudence can certainly be read in a way which would incorporate the rules of state responsibility into the jus ad bellum, as self-defence would be lawful under Article 51 of the UN Charter only in the case of an armed attack for which another state is responsible. See Case Concerning Oil Platforms (Iran v. United States), Judgment (Merits) of 6 November 2003, paras 57–61; Legal Consequences of the Construction of a Wall in the Occupied Palestinian Territory, Advisory Opinion of 9 July 2004, paras 138 and 139. The ICJ has passed up another opportunity to clarify its position on the matter in Congo v. Uganda – see the Judgment in that case, paras 146 and 147. This approach of the Court was quite correctly criticized by Judges Simma and Kooijmans in their separate opinions – see Separate Opinion of Judge Simma, paras 4–15; Separate Opinion of Judge Kooijmans, paras 16–35.

166

Nor has doctrine managed to make much more sense of the issue – for one of the earliest treatments of attacks by private actors and self-defence under Article 51 of the Charter, see Brownlie, ‘International Law and the Activities of Armed Bands’, 7 ICLQ (1958) 712.

167

In other words, the possible development of special primary rules on state sponsorship of terrorism does not depend on a simultaneous change of traditional rules of state responsibility. See Jinks, ‘State Responsibility for Sponsorship of Terrorist and Insurgent Groups: State Responsibility for the Acts of Private Armed Groups’, 4 Chi. J Int’lL (2003) 83.

168

A view which would certainly be supported by SC Resolutions 1368 and 1373 (2001), which ‘reaffirm the inherent right of individual or collective self-defence’ in response to terrorism (though these resolutions do not say self-defence against whom, and in what way).

169

See Meron, supra note 115, at 241–242.

170

The test of overall control has certainly been applied by the ICTY in several subsequent cases in order to determine the character of a particular episode of an armed conflict. See more Cullen, ‘The Parameters of Internal Armed Conflict in International Humanitarian Law’, 12 U Miami Int’l& Comp L Rev (2004) 189.

171

See also Meron, supra note 115, at 239–241.

172

See Sassòli and Olson, ‘International Decision: Prosecutor v. Tadić’, 94 AJIL (2000) 571, at 578.

173

And it now might be so by the sheer force of precedent, certainly within the ICTY – see, e.g., Prosecutor v. Zejnil Delalić et al., IT-96-21, Appeals Chamber Judgment, 20 February 2001, paras 6–27 – and it will possibly be treated as such by the ICC.

174

Kenneth P. Yeager v. Islamic Republic of Iran, 17 Iran-US Claims Tribunal Reports (1987) 92.

175

Tadić, Appeals Chamber Judgment, paras 126 and 127.

176

See ILC Commentaries, at 109–110.

177

Loizidou v. Cyprus, App. no. 15318/89, Judgment (preliminary objections) of 23 February 1995, paras 60–64; Judgment (merits) of 28 November 1996, paras 52–57.

178

Tadić, Appeals Chamber Judgment, para. 128.

179

Banković and Others v. Belgium and Others., App. no. 52207/99, Decision on admissibility of 12 December 2001, esp. paras 54–82.

180

Issa and Others v. Turkey, App. no. 31821/96, Judgment, 16 November 2004, esp. para. 69.

181

Ilascu and Others v. Moldova and Russia, App. No. 48787/99, Judgment of 8 July 2004, esp. paras 336–352, 376–394.

182

Article 2(a) of the ILC Articles.

183

Article 2(b) of the ILC Articles

184

See more Craven, ‘Legal Differentiation And The Concept Of The Human Rights Treaty In International Law’, 11 EJIL (2000) 489.

185

For example, the European Court used the ILC Draft Articles on State Responsibility in the Ilascu case, para. 319, when it discusses state responsibility for ultra vires acts and breaches which extend over time; or in Blečić v. Croatia, Application no. 59532/00, Grand Chamber Judgment of 8 March 2006, para. 48, on the same issue. It makes no mention of the ILC Articles in the section of the Ilascu judgment in which it discusses the jurisdiction of Moldova and Russia over the applicants, nor does it at any point in any of these cases discuss the different standards of control as expounded in Nicaragua and Tadić.

186

See ILC Commentaries, at 106–107. A similar attempt was made by Judge Shahabuddeen, a former ICJ judge himself, in his separate opinion in Tadić: see Appeals Chamber judgment, Separate Opinion of Judge Shahabuddeen, para. 4 et seq.

187

Tadić, Appeals Chamber Judgment, para. 98 et seq.

188

See also Wittich, ‘The International Law Commission’s Articles on the Responsibility of States for Internationally Wrongful Acts Adopted on Second Reading’, 15 Leiden JIL (2002) 891.

189

Congo v. Uganda, para. 160: ‘The Court concludes that there is no credible evidence to suggest that Uganda created the MLC. Uganda has acknowledged giving training and military support and there is evidence to that effect. The Court has not received probative evidence that Uganda controlled, or could control, the manner in which Mr. Bemba put such assistance to use. In the view of the Court, the conduct of the MLC was not that of “an organ” of Uganda (Article 4, International Law Commission Draft Articles on Responsibility of States for Internationally Wrongful Acts, 2001), nor that of an entity exercising elements of governmental authority on its behalf (Article 5). The Court has considered whether the MLC’s conduct was “on the instructions of, or under the direction or control of” Uganda (Article 8) and finds that there is no probative evidence by reference to which it has been persuaded that this was the case. Accordingly, no issue arises in the present case as to whether the requisite tests are met for sufficiency of control of paramilitaries (see Military and Paramilitary Activities in and against Nicaragua (Nicaragua v. United States of America), Merits, Judgment, I.C.J. Reports 1986, pp. 62–65, paras. 109–115).’

190

But see, however, the defence of Tadić by Professor Cassese, then judge in the ICTY Appeals Chamber and former President of the Tribunal, in Cassese, supra note 29, at 248–250.

191

This most recent figure is the product of research conducted for the International Criminal Tribunal for the Former Yugoslavia by the Sarajevo-based Investigation and Documentation Centre (IDC) led by Mirsad Tokaca. According to the IDC, 93,000 people were killed in the Bosnian war, mostly civilians, and of those 70 per cent were Bosniaks (Muslims), slightly under 25 per cent were Serbs, slightly under five per cent Croats and about one per cent of others. See at http://www.bosnia.org.uk/news/news_body.cfm?newsid=1985, accessed on 5 Jan. 2006.

192

As most recently stated by US Under-Secretary of State, Mr. Nicholas Burns, at a conference in Washington, DC on 21 November 2005, available at http://www.state.gov/p/us/rm/2005/57189.htm.

193

The number of people killed during the conflict is also a subject of a macabre dispute between Bosnia and Serbia in the Genocide case, both in the written pleadings and during the oral hearings. See, e.g., the pleadings of Mr. van den Biesen on behalf of Bosnia, Verbatim record of the public sitting of the Court held on 27 February 2006, at 10.30 a.m., President Higgins presiding, CR 2006/2, at 45 et seq, and of Mr. Obradovic on behalf of Serbia, Verbatim record of the public sitting of the Court held on 8 March 2006, at 10 a.m., President Higgins presiding, CR 2006/12, at 38 et seq.

194

The General Framework Agreement for Peace in Bosnia and Herzegovina, available at http://www.ohr.int/dpa/default.asp?content_id=380.

195

See the Trial Chamber judgment in Prosecutor v. Krstić, IT-98-33-T, 2 August 2001 and the judgment of the Appeals Chamber in the same case of 19 April 2004, as well as the recent Trial Chamber judgment in Prosecutor v. Blagojević, IT-02-60-T, 17 January 2005.

196

See, e.g., at http://www.b92.net/info/vesti/index.php?yyyy=2005&mm=05&dd=17&nav_id=168589 (in Serbian), accessed on March 30, 2006.

197

See, e.g., the numerous news articles and readers’ comments at two leading news outlets in Serbia (www.b92.net) and Bosnia (www.avaz.ba), available in Serbian/Bosnian.

198

For example, at the 10th anniversary commemoration of the Srebrenica genocide one of the speakers, Ibran Mustafić, once active in the Association of the Mothers of Srebrenica and unannounced by protocol, declared: ‘Long live Bosnia and Herzegovina, and death to all genocidal creations on this territory’, see at http://www.b92.net/info/vesti/index.php?yyyy=2005&mm=07&dd=11&nav_id=172392 (in Serbian).

199

For instance, while the tenth anniversary commemoration was being held in Srebrenica, a Bosnian Serb group organized a parallel event for Serb victims in the nearby village of Bratunac.

200

Application Instituting Proceedings, 20 March 1993.

201

Judgment of 11 July 1996 – Preliminary Objections.

202

See Order of 10 September 2001.

203

Application for Revision of the Judgment of 11 July 1996 in the Case concerning Application of the Convention on the Prevention and Punishment of the Crime of Genocide (Bosnia and Herzegovina v. Yugoslavia), Preliminary Objections (Yugoslavia v. Bosnia and Herzegovina), 24 April 2001.

204

See the Case Concerning Legality of Use of Force (Serbia and Montenegro v. Belgium), judgment of 15 December 2004, General List No. 105 (one of eight almost identical cases against NATO member states), esp. paras 51–79, 91.

205

Judgment of 3 February 2003. However, a similar genocide application by Croatia against Serbia and Montenegro will almost certainly be rejected by the Court, as it had not yet ruled on its jurisdiction, unlike in the Bosnian case.

206

Simultaneously with the application for revision in the Genocide case, Serbia submitted an Initiative to the Court to Reconsider Ex Officio Jurisdiction over Yugoslavia, 4 May 2001.

207

For a more detailed description of the procedural history of the case, see the Press Release of the ICJ of 27 February 2006, available at http://www.icj-cij.org. See also the analysis of the case by the IWPR, Serbia and Montenegro on Trial for Genocide, 24 February 2006, available at http://www.iwpr.net.

208

See in that regard the Joint Declaration of Vice-President Ranjeva and Judges Guillame, Higgins, Kooijmans, Al-Khasawneh, Buergenthal and Elaraby in the NATO cases.

209

See the oral pleadings of Mr. Vladimir Djeric on behalf of Serbia, verbatim record of the public sitting of the Court held on 9 March 2006, at 10 a.m., President Higgins presiding, CR 2006/13, at 11

210

See, e.g., General Comment No. 26 of the Human Rights Committee. See also the oral pleadings of Professor Tibor Varady, Mr. Vladimir Djeric and Mr. Andreas Zimmerman on behalf of Serbia, Verbatim record of the public sitting of the Court held on 9 March 2006, at 10 a.m., President Higgins presiding, CR 2006/13

211

See the pleadings by Prof. Alain Pellet on behalf of Bosnia, verbatim record of the public sitting of the Court held on 28 February 2006, at 10 a.m., President Higgins presiding, CR 2006/3.

212

Ibid., at 19.

213

See the oral pleadings of Prof. Thomas Franck on behalf of Bosnia, verbatim record of the public sitting of the Court held on 21 April 2006, at 3 p.m., President Higgins presiding, CR 2006/36, at 25 et seq.

214

See Article 35(2) of the ICJ Statute and UN SC Res. (1946) 9.

215

For a much more detailed analysis of the jurisdictional issue, see Vitucci, ‘Has Pandora’s Box Been Closed? The Decisions on the Legality of Use of Force Cases in Relation to the Status of the Federal Republic of Yugoslavia (Serbia and Montenegro) within the United Nations’, 19 Leiden JIL (2006) 105.

216

See, e.g., the oral pleadings of Mr. Phon van den Biesen on behalf of Bosnia, Verbatim record of the public sitting of the Court held on 27 February 2006, at 10.30 a.m., President Higgins presiding, CR 2006/2, at 28 et seq; oral pleadings of Professor Thomas Franck on behalf of Bosnia, verbatim record of the public sitting of the Court held on 1 March 2006, at 10 a.m., President Higgins presiding, CR 2006/5, at 19–20; Reply of Bosnia and Herzegovina, 23 April 1998, at 77–373.

217

See, e.g., the oral pleadings of Mr. Phon van den Biesen, verbatim record of the public sitting of the Court held on 28 February 2006, at 3 p.m., President Higgins presiding, CR 2006/4, at 21 et seq.

218

See, e.g., the oral pleadings of Ms. Magda Karagiannakis on behalf of Bosnia, verbatim record of the public sitting of the Court held on 1 March 2006, at 10 a.m., President Higgins presiding, CR 2006/5, at 33 et seq.

219

See Prosecutor v. Galić, IT-98-29, Trial Chamber judgment of 5 December 2003. General Galić was the VRS officer in charge of the siege of Sarajevo, and was not even charged with genocide, but was convicted for crimes against humanity.

220

See Prosecutor v. Stakić, IT-97-24, Trial Chamber judgment of 31 July 2003, paras 547–561, esp. para. 553: ‘[though the evidence shows that] the common goal of the members of the SDS in the Municipality of Prijedor, including Dr. Stakić as President of the Municipal Assembly, was to establish a Serbian municipality, there is insufficient evidence of an intention to do so by destroying in part the Muslim group’ as well as para. 554 in fine: ‘[t]he intention to displace a population is not equivalent to the intention to destroy it.’ See also the Appeals Chamber judgment in the same case of 22 March 2006, paras 14–57, esp. para. 56.

221

For the origin of the term ‘ethnic cleansing’ see Petrović, ‘Ethnic Cleansing – An Attempt at Methodology’, 5 EJIL (1994) 359.

222

See, e.g., the Trial Chamber judgments in the cases of Sikirica, paras 58, 89; Jelisić, paras. 68, 79; Krstić, para. 553; Stakić, para. 518, 519; and Tadić, para. 697; as well as the Appeals Chamber judgement in Tadić, para. 305.

223

See Cassese, supra note 18, at 98–100, 106–107; Schabas, supra note 9, at 194–201. See also supra, Section 1.

224

Article II(c) of the Genocide Convention.

225

For a bibliography on the Armenian genocide, see http://www.umd.umich.edu/dept/armenian/facts/gen_bib1.html.

226

Schabas, supra note 9, at 200–201. Even the Nazis at first wanted to deport the Jews from Europe en masse, with plans ranging from the Russian expanses to Madagascar, with the Final Solution becoming a fully genocidal programme once expulsion became impossible. See Browning, supra note 84, at 36–110.

227

See the Genocide case, Reply of Bosnia and Herzegovina, 23 April 1998, at 34 et seq.

228

See, e.g., Brower, ‘Evidence before International Tribunals: The Need for Some Standard Rules’, 28 Int’lL. (1994) 47.

229

As stated by a judge of the Iran–United States Claims Tribunal: ‘the burden of proof is that you have to convince me’, quoted by Brower, supra note 228, at 52.

230

See Cassese, supra note 18, at 374–376.

231

This is, of course, a broad generalization, as many civil law systems have quite specific procedural institutions. For more detailed comparisons of adversarial and inquisitorial systems, especially as seen through the prism of international criminal law, see Orie, ‘Accusatorial v. Inquisitorial Approach in International Criminal Proceedings prior to the Establishment of the ICC and in the Proceedings before the ICC’, in Cassese, et al., supra note 6, 1,439. For a very vivid appraisal of the French Courd’assises by a lawyer with an Anglo-American background, see Lettow Lerner, ‘The Intersection of Two Systems: An American on Trial for an American Murder in the French Cour d’Assises’, 3 U Ill L Rev (2001) 791. For an account of the very interesting experience of Italy, a country with a strong civil law tradition which adopted an adversarial system a decade ago, see Pizzi and Montagna, ‘The Battle to Establish an Adversarial Trial System in Italy’, 25 Mich J Int’lLaw (2004) 429.

232

Especially, for example, in the recent Oil Platforms and Wall cases – see, e.g., the Separate Opinion of Judge Buergenthal in the Oil Platforms case, paras 41–46; Separate Opinion of Judge Higgins in the same case, paras 30–39. The Court’s recent judgment in Congo v Uganda may be seen as a ray of hope in soothing the sensibilities of common lawyers, as it seems to engage in much more specific discussion of the rules of evidence than any of its previous judgments.

233

See supra 2A3.

234

Rule 87, ICTY Rules of Procedure and Evidence, as amended on 21 July 2005.

235

Rule 87, ICTR Rules of Procedure and Evidence, as amended on 21 May 2005.

236

Article 66 (3) of the ICC Statute.

237

It could also be argued that other branches of international law, such as the law on the use of force, require the establishment of a single evidentiary standard, such as clear and convincing evidence. See O’Connell, ‘Lawful Self-Defense to Terrorism’, 63 U Pitt. L Rev (2002) 889.

238

See Schabas, supra note 9, at 443.

239

Both parties in the Genocide case spent a significant amount of time on the proper role the ICJ should give to the evidence and jurisprudence of the ICTY. See the pleadings of Prof. Thomas Franck and Ms. Magda Karagiannakis on behalf of Bosnia, verbatim record of the public sitting of the Court held on 28 Feb. 2006, at 10 a.m., President Higgins presiding, CR 2006/3, at 29 et seq.

240

See also Nollkaemper, supra note 38, at 628–630.

241

See also Dupuy, supra note 6, at 1097–1099.

242

See Schabas, ‘Was Genocide Committed in Bosnia and Herzegovina? First Judgments of the International Criminal Tribunal for the Former Yugoslavia, 26 Fordham Int’lLJ (2001) 907.

243

The best starting point for the ICJ in this case would be the ICTY Trial Chamber in the Milošević case Decision on the Motion for Judgment of Acquittal, IT-02-54, 16 June 2004, paras 117 et seq., which lists several municipalities in which a reasonable Trial Chamber could find that genocide was committed, and do so beyond reasonable doubt. The legal weight of this decision – the only legal product of the entire Milošević trial – is lower than that of judgment, as it is based only on the evidence presented by the Prosecution (though subject to cross-examination), and as the Trial Chamber does not give its own assessment of the evidence, but discusses whether a reasonable, generic Trial Chamber could establish the existence of genocide. Nevertheless, this decision can still provide a useful baseline for possible instances of genocide in Bosnia other than Srebrenica.

244

See supra 2B.

245

This would, for example, have practical relevance for Serbia’s possible responsibility for the Srebrenica massacre. Thus, the Netherlands Institute for War Documentation (NIOD) Report on the Srebrenica genocide states that upon reports of the crimes committed upon the fall of the enclave, ‘the mood in Belgrade was one of incredulity and total disbelief ’, while ‘there is no evidence to suggest participation in the preparations for the executions on the part of Yugoslav military personnel or the security agency (RDB).’ See at http://213.222.3.5/srebrenica/toc/p4_c02_s020_b01.html. Be that as it may, if the Bosnian Serb forces were otherwise under the complete control of Serbia and were its de facto organs, than even these acts which might have been committed ultra vires would still be attributable to Serbia.

246

See Nicaragua, paras 93 and 94; Congo v. Uganda, para. 160.

247

See, e.g., the reports by B92 of 21 March 2005 regarding General Vinko Pandurevic, indicted before the ICTY for the Srebrenica genocide, available at http://www.b92.net (in Serbian).

248

Indicted before the ICTY, see Prosecutor v. Ražnatović, IT-97-27, Initial Indictement of 30 September 1997, but was assassinated in Belgrade on 15 January 2000.

249

Indicted before the ICTY and currently waiting trial, see Prosecutor v. Šešelj, IT-03-67, Modified Amended Indictement of 15 July 2005.

250

Simatović and his superior, Jovica Stanišić, were indicted before the ICTY, see Prosecutor v. Stanišić and Simatovic, IT-03-69, Second Amended Indictment of 20 December 2005, and are currently waiting trial. Members of the JSO later assassinated the first democratically elected prime minister of Serbia, Zoran Đinđić, and are currently standing trial in Belgrade.

251

The video, as presented at the Milošević trial, is available at http://jurist.law.pitt.edu/monitor/2005/06/srebrenica-killings-video-icty.php. Members of the ‘Scorpios’ are also currently standing trial in Belgrade.

252

As Nicaragua would indicate, but this is by no means certain.

253

See R. Holbrooke, To End a War (rev. edn., 1999), at 51 et seq.

254

See S.P. Ramet and B. Babel, The Disintegration of Yugoslavia from the Death of Tito to Ethnic War (1996), at 257–258.

255

Though he did admit to ‘giving aid’ to the Bosnian Serbs; see, e.g., the IWPR report at http://www.iwpr.net/?p=tri&s=f&o=165884&apc_state=henitri2003.

256

The fact that it was Milosevic who negotiated at Dayton on behalf of the Bosnian Serbs as well as Serbia would go against this conclusion, for example.

257

See supra, 2C4.

258

Political justifications of their inability to arrest Mladić by the Serbian Government are legally completely irrelevant. Serbia’s failure to surrender Mladić to the ICTY has recently led the European Union to suspend further talks on stabilization and association; see, e.g., at http://news.bbc.co.uk/2/hi/europe/4968104.stm.

259

See supra, 2A.

260

See supra, 2B.

261

See supra, 2C.

262

See supra, 3E.

263

See supra, 4C2.

264

Prosecutor v. Perišić, IT-04-81.

265

Prosecutor v. Popović et al., IT-05-88.

266

See, e.g., A. Nollkaemper, supra note 38, at. 624–627, 639–640.

267

See, e.g., Article 41(3) of the ILC Articles, which states that the consequences of serious breaches prescribed by the Articles are without prejudice to further consequences that such breaches may entail under international law.

268

See also the 2005 World Summit Outcome Document, A/RES/60/1, 24 October 2005, available at http://daccessdds.un.org/doc/UNDOC/GEN/N05/487/60/PDF/N0548760.pdf?OpenElement, paras 138 and 139, which stress the duty of all states to protect their own populations against genocide and other grave crimes, as well as the duty of the international community to take collective action when peaceful means prove inadequate. See also SC Res. 1647 (2006), 28 April 2006, which reaffirms the provisions of paras 138 and 139 of the World Summit Outcome Document.

I would like to thank Steven Ratner for his most valuable guidance, and Dino Kritsiotis, Vojin Dimitrijević, Andrea Gattini, Annalisa Ciampi, André J. J. de Hoogh, Carsten Hoppe, Jason Morgan-Foster, Vidan Hadži-Vidanović and Vladimir Pavlović for their helpful comments. All errors are my own.