1. Introduction
The recent judgment of the International Court of Justice (ICJ) in the Case Concerning the Application of the Convention on the Prevention and Punishment of the Crime of Genocide (Bosnia and Herzegovina v. Serbia and Montenegro) Footnote 1 was a historic moment for the global community. The question before the ICJ was whether SerbiaFootnote 2 could be held responsible for genocide on the basis of atrocities committed in Bosnia and Herzegovina following the 1992 break-up of the former Yugoslavia. The case represented the first time that a court had adjudicated whether a sovereign state could be held responsible for genocide in the almost sixty years since the Convention on the Prevention and Punishment of the Crime of Genocide was unanimously approved by the General Assembly of the United Nations.Footnote 3
The ICJ's 171-page judgment is likely to generate a plethora of journal articles as legal scholars assess the long-term impact of the judgment on previously uncharted territories of international jurisprudence. In this respect, one of the most interesting issues arising from the case involves the challenges presented when different international courts have to adjudicate on overlapping factual situations.
The general issue of the fragmentation of international law has already been the topic of much legal writing,Footnote 4 and the International Law Commission has conducted a working-group study on the potential for the fragmentation of jurisprudence with the proliferation of international adjudicative systems.Footnote 5 In the arena of international criminal law this discussion came to prominence following the creation of the first ad hoc criminal tribunal, the International Criminal Tribunal for the former Yugoslavia (ICTY) in 1993, and is likely to remain a topical issue with the advent of the permanent International Criminal Court (ICC).
Given that the ICTY has now produced over a decade's worth of jurisprudence on the atrocities committed in the former Yugoslavia, the ICJ's Bosnia v. Serbia judgment makes for a particularly good study of the issues that can arise when an international court with one mandate has to adjudicate on a situation that an international court with a different mandate has already adjudicated on. The ICJ is the principal judicial organ of the United Nations, and its work involves disputes between states. By contrast the ICTY was created by the United Nations in order to establish individual criminal responsibility for the atrocities committed since 1991 in the former Yugoslavia. The consequences of these differences in mandate and expertise come to the fore in Bosnia v. Serbia.
The overlap between the issues in Bosnia v. Serbia and the ICTY cases arises in terms of both the legal analysis and the fact-finding process. With respect to the former we look at the ICJ's legal test for the attribution of state responsibility, established in Military and Paramilitary Activities in and against Nicaragua (Nicaragua v. United States of America),Footnote 6 and restated in Bosnia v. Serbia,Footnote 7 as compared with the test proposed by the ICTY Appeals Chamber in Prosecutor v. Tadić.Footnote 8 We find that the ICJ's approach in Bosnia v. Serbia was a sensible one, and that the ICJ's concern not to disrupt the consistency of jurisprudence within the ICTY, and within its own court, should be taken into account by those who are interested in the project of generating principled guidelines for consistency between international bodies.
With respect to the issue of fact finding, we first show how the ICJ drew inferences not only from the findings of guilt made by the ICTY, but also – problematically – from the absence of certain ICTY convictions, and even charges, in order to find that genocide had not occurred in Bosnia and Herzegovina in any region other than in Srebrenica.Footnote 9 We then look at how the ICJ addressed the factual questions relating to whether Serbia could be held responsible for the Srebrenica genocide. We note that the failure of the ICJ to request unredacted versions of documents that Bosnia and Herzegovina believed would show that the genocide at Srebrenica could be attributed to the Serbian state, along with the ICJ's refusal to draw a negative inference from Serbia's failure to produce those documents, may turn out to be the this case's most troublesome legacy.Footnote 10
2. The legal standard for the attribution of state responsibility
The past decade has produced an ongoing debate among legal scholars about the extent of disjunct between the jurisprudence of the ICJ and ICTY on the question of the appropriate legal standard for imputing the actions of an individual, or group of individuals, to a state.Footnote 11 So far as the issue of state responsibility falls within the purview of international law, over which the ICJ has general jurisdiction, it is presumed to be more competent in this area than the ICTY, which has jurisdiction over issues of individual criminal responsibility. Indeed, in its decision in Nicaragua the ICJ established a test for the attribution of individual actions to a state almost a decade before the ICTY came into existence.Footnote 12
2.1. The ICJ's Nicaragua standard for attributing state responsibility
In Nicaragua the ICJ needed to determine whether the acts of a group of non-state actors, the Contras, in Nicaragua, could be attributed to the United States. Under Nicaragua the Court determined that the actions of actors who are not organs of the state under the internal laws of the state can still be considered de facto organs of the state if their relationship is ‘one of dependence on the one side and control on the other’.Footnote 13 When a relationship of such ‘complete dependence’ is found, all the acts of those formally ‘non-state’ actors can be imputed to the state. However, even when this standard of ‘complete dependence’ is not met, the conduct of individuals may still be attributable to the state if the latter had effective control over the specific operations in the course of which such conduct took place.Footnote 14
2.2. The ICTY's use of Nicaragua to determine the nature of an armed conflict
In 1997 the ICTY Trial Chamber introduced the Nicaragua decision into its jurisprudence in order to answer the question of whether or not the conflict in Bosnia and Herzegovina was an internal or an international armed conflict.Footnote 15 This question was relevant to the work of the ICTY because under international humanitarian law (IHL) individuals can only be charged with ‘grave breaches’ crimes of the Geneva Conventions (as set out in Article 2 of the Statute of the ICTY) committed during an international, but not internal, armed conflict,Footnote 16 and only if their victims are ‘protected persons’ under Article IV of the Geneva Conventions.Footnote 17
When an armed conflict involves two sovereign states, the conflict is obviously international. However, when the conflict involves a state and one or more non-state actors, even though operating across international borders, as in the case of the former Yugoslavia, it becomes necessary to impute the actions of the non-state actors to another sovereign state in order to determine that the conflict is international. Without such an attribution the conflict is viewed as internal and so the ‘grave breaches’ regime of the Geneva Conventions does not apply.Footnote 18
Prior to the break-up of the former Yugoslavia, what was then the Socialist Federal Republic of Yugoslavia (SFRY) maintained a national Yugoslav People's Army (JNA). During the break-up of the SFRY the political leadership in the Serb-controlled region of Bosnia and Herzegovina formed its own entity called the Serb Republic of Bosnia and Herzegovina (subsequently Republika Srpska). On 12 May 1992 the Army of Republika Srpska (VRS) was formed. Pursuant to a UN Security Council resolution,Footnote 19 the FRY formally withdrew JNA forces from Bosnia and Herzegovina on 19 May 1992. However, not only were troops of Bosnian Serb origin from throughout the JNA transferred into the VRS, but the FRY maintained control over the VRS.Footnote 20 The extent of that control became a key issue for the ICTY cases dealing with atrocities committed in Bosnia and Herzegovina after the formal withdrawal date.Footnote 21 If the control that the FRY had over the VRS in Bosnia and Herzegovina was sufficient to impute the acts of the VRS to the FRY, then the conflict in Bosnia and Herzegovina could be considered ‘international’ for the purposes of IHL and, accordingly, those brought before the Tribunal could be charged with crimes against protected persons under the ‘grave breaches’ regime of the Geneva Conventions.Footnote 22
2.3. The ICTY trial chamber's use of Nicaragua in Tadić
Duško Tadić was a café owner and karate expert who had access to the detention camps in Prijedor, in Bosnia and Herzegovina, and was brought before the ICTY on charges of war crimes and crimes against humanity that allegedly occurred in the camps in 1992. In Prosecutor v. Tadić, the trial chamber had to decide whether, after the formal withdrawal of the JNA from Bosnia and Herzegovina, the FRY nevertheless maintained sufficient control over the VRS that the VRS was essentially a de facto organ of the FRY.Footnote 23 If so, this relationship would ‘internationalize’ the conflict for the purposes of IHL, with the result that Tadić could be charged with ‘grave breaches’ crimes against those detained within the camps.
In making its analysis, the trial chamber invoked Nicaragua. Although it acknowledged that the key issue before the ICJ in Nicaragua was whether the accused state could be held responsible for violations of IHL, it went on to justify its invocation of Nicaragua on the grounds that the ICJ also had to decide which part of IHL to apply to the conflict.Footnote 24 As applied to the case before it, the trial chamber understood that for the ‘grave breaches’ regime to come into force, Nicaragua required the prosecution to prove that after the formal withdrawal of troops on 19 May 1992, the FRY nevertheless maintained ‘effective control’ over the specific operations of the newly formed VRS.Footnote 25
In applying the stringent ‘effective control’ test to the facts of the case, the trial chamber concluded that the standard was not met and that therefore the ‘grave breaches’ regime of the Geneva Conventions did not apply.Footnote 26
In a strong dissenting opinion, Presiding Judge McDonald argued that the majority had misinterpreted Nicaragua, and that the ‘effective control’ test is a subsidiary test to be applied only when no ‘agency’ relationship has been found between the state and the non-state actors in question.Footnote 27 What Judge McDonald referred to as the ‘agency test’ the court in Nicaragua had explained as a relationship that is ‘one of dependence on the one side and control on the other’.Footnote 28 In other words, if the non-state actor or group does not have such a relationship of complete dependency on the state to be considered its agent, only then should the question be asked as to whether the state exercised such effective control over specific operations of those groups that it might nevertheless be appropriate to hold the state accountable for violations committed in those specific operations. More fundamentally, Judge McDonald disagreed with the trial chamber majority that the Nicaragua case was even relevant in determining the nature of the conflict for the purposes of finding individual criminal responsibility: ‘The inapplicability of the Nicaragua standard of effective control is patent; it was neither designed for these factual circumstances nor is it an appropriate consideration.’Footnote 29
2.4. The ICTY Appeals Chamber's rejection of the ICJ's Nicaragua standard
In 1999 the ICTY Appeals Chamber entered the fray, creating an explicit conflict with Nicaragua.Footnote 30 The Appeals Chamber agreed with the trial chamber that Nicaragua was relevant to determining the nature of the armed conflict. Rejecting Judge McDonald's dissenting argument that Nicaragua should not be used to assess the nature of the conflict, the Appeals Chamber explained,
What is at issue is . . . the conditions on which under international law an individual may be held to act as a de facto organ of a State. Logically these conditions must be the same both in the case: (i) where the court's task is to ascertain whether an act performed by an individual may be attributed to a State, thereby generating the international responsibility of that State; and (ii) where the court must instead determine whether individuals are acting as de facto State officials, thereby rendering the conflict international and thus setting the necessary precondition for the ‘grave breaches’ regime to apply.Footnote 31
The Appeals Chamber also disagreed with Judge McDonald's assertion that the ‘effective control’ test in Nicaragua was a subsidiary test to what she had termed the ‘agency’ test.Footnote 32 However, it also disagreed with the trial chamber majority's acceptance of Nicaragua's ‘effective control’ test, and in an unexpected move it set itself up in direct conflict with the ICJ, stating that ‘the Appeals Chamber, with respect, does not hold the Nicaragua [effective control] test to be persuasive’.Footnote 33 Its justification for this attack on the jurisprudence of the ICJ was that the ‘effective control’ test did not comport either with the logic of the law of state responsibility,Footnote 34 or with judicial or state practice.Footnote 35
The Appeals Chamber held that international law proposes different tests of state responsibility which are applied on the basis of the type of actor whose behaviour is being assigned for attribution. The first test applies when the actor is an individual or part of a group that is not militarily organized; in this case evidence of specific instructions by the state must be proven, or the state must ratify the conduct.Footnote 36 The second test applies when the actor is part of an organized paramilitary structure; in this case a lesser standard of ‘overall control’, which the Chamber explains must ‘comprise more than the mere provision of financial assistance or military equipment or training’, must be shown.Footnote 37 The Chamber also makes a fairly cursory reference to a third test that seems akin to a notion of apparent authority, whereby the actions of individuals can be attributed to the state when those actions approximate the behaviour of a person who is part of an organ of the state.Footnote 38
The handling of Nicaragua in the Tadić appeal has been criticized, not only for the substance of what the Appeals Chamber said about the ‘effective control’ test,Footnote 39 but also for the fact that the Appeals Chamber even addressed the Nicaragua test in the first place.Footnote 40 It is not too difficult to see the intuition behind the ICTY's general inclination to look for an attribution test in trying to determine whether an external state exercises so much control over the actions of a group of non-state actors that the conflict is substantively an international one.Footnote 41 Nevertheless, the Appeals Chamber violated the basic principle of judicial economy by offering a wholesale criticism of the validity of the Nicaragua attribution test for the purposes of state responsibility, when the only issue it needed to decide in the case before it was whether the alleged acts with which Tadić was charged took place within an international or an internal armed conflict. It was therefore not surprising that when presented with a case that required the application of this test of attribution, the ICJ would take the opportunity to explicate its position on Nicaragua in the context of comments made by the ICTY in the Tadić appeal.
2.5. The ICJ's response to the ICTY's attack on Nicaragua
Given the aggressive attack that the ICTY Appeals Chamber had launched on the ICJ's Nicaragua attribution test, the ICJ was fairly measured in its response to the issue in Serbia v. Bosnia. Avoiding a tit-for-tat approach, the Court did not try to argue that the ICTY had been incorrect with respect to the issue that had been before it – namely how to determine the nature of the conflict. Specifically, the ICJ noted that
Insofar as the [ICTY's] ‘overall control’ test is employed to determine whether or not an armed conflict is international . . . it may well be that the test is applicable and suitable; the Court does not however think it appropriate to take a position on the point in the present case.Footnote 42
This was wise, since just as state responsibility is not within the ICTY's area of special expertise, neither is the characterization of the nature of a conflict for the purpose of applying IHL to individual actors within the ICJ's area of specialty. However in respect to the ICTY's further claim that the ‘overall control’ test should also be applied to questions of state responsibility for the acts of non-state organs, the Court was outright in its rejection of the Appeals Chamber's approach.Footnote 43 Directly rebutting the Tadić Appeal Chamber's argument that logic required a single test to answer both the question of state responsibility and the question of the nature of an armed conflict, the Court stated,
[T]he degree and nature of a State's involvement in an armed conflict on another State's territory which is required for the conflict to be characterized as international, can very well, and without logical inconsistency, differ from the degree and nature of involvement required to give rise to that State's responsibility for a specific act committed in the course of the conflict.Footnote 44
Accordingly, the final outcome of the interplay between the ICJ and ICTY on this question of attribution suggests that the ICJ will continue to use its Nicaragua test in assessing questions of state responsibility that come before it, and that the ICTY will continue to use its ‘overall control’ test in characterizing whether a conflict is internal or international for the purposes of applying the ‘grave breaches’ regime of IHL.
The ICJ's language of deference towards the ICTY with regard to the issues that come before that Tribunal give the appearance of a court that is being guided by principles of judicial restraint: the ICJ commented as much as was necessary for it to adjudicate the case before it, while reserving judgment on the appropriateness of the ICTY's approach to the questions it faces. However, the Court's approach might also be understood as having been driven by purely pragmatic concerns. One might observe the likely futility had the ICJ attempted to fight the ICTY's judgment on the issue of how to assess the nature of the conflict, given that since the Tadić appeal the ICTY has developed a solid body of jurisprudence that is grounded in the ‘overall control’ test for determining the nature of the conflict,Footnote 45 and to which the Pre-trial Chamber of the ICC has already referred approvingly.Footnote 46 To the extent that this more pragmatic view has any credence, it may just be the by-product of a current legal order in which there have not yet been any formal or enforced guidelines to govern the interrelationship between the ICJ and the ICTY. However, this possibility notwithstanding, the lesson from Bosnia v. Serbia may well be that any attempt to guide different international adjudicative bodies towards a principled system in order to avoid the potential for fragmented jurisprudence between the international bodies will need to be flexible enough to accommodate the concerns for consistency within each of the international bodies.
3. Fact finding
There were two heavily fact-dependent questions that the Court had to resolve in the case before it. Did genocide take place in Bosnia and Herzegovina? And if so, could Serbia be held responsible?
As a threshold issue, the Court first addressed Serbia's question of whether, as a matter of law, it would be possible to find a state responsible for genocide in the absence of an individual, over whom the state had control, having first been held criminally liable for genocide by a duly constituted court.Footnote 47 Answering in the affirmative, the Court pointed out that the Statute of the ICJ authorized it to undertake the task of deciding if genocide had been committed by a state, and that the Genocide Convention gave it jurisdiction over this task.Footnote 48 Furthermore, the Court observed that any alternative answer would lead to an outcome whereby there might be no legal recourse available to victims of genocide in a situation in which, due to political constraints, the individual perpetrators had not been held accountable: ‘The Court accordingly concludes that state responsibility can arise under the Convention for genocide or complicity, without an individual being convicted of the crime or an associated one.’Footnote 49 In our view, this is an entirely correct statement. However, as the following elucidates, the Court's fact-finding approach in this test case raises doubts as to whether, in practice, a state will ever be held responsible for genocide outside the parameters of the prior convictions of individual perpetrators.
3.1. Was genocide committed in Bosnia and Herzegovina?
To answer this question the Court needed evidence of both the actus reus of genocide (the acts set out in paragraphs (a) to (e) of Article II of the Genocide Convention, such as killing, causing serious bodily or mental harm, and forcibly transferring children) and the dolus specialis (that those acts were ‘committed with intent to destroy, in whole or in part, a national, ethnical, racial or religious group, as such’ – as set out in the chapeau of Article II of the Genocide Convention). In terms of the question of whether genocide had been committed in Bosnia and Herzegovina, the ICJ was highly deferential to the findings of the ICTY, which has competence to decide both these issues with respect to whether individual alleged perpetrators can be held liable for genocide.
The Court was careful to lay out the different weights it proposed to give to findings presented from different stages of the ICTY's process, from indictment through to appeal, recognizing that because no definitive rulings are made before the conclusion of a trial, it would not be appropriate for the Court to give weight to evidence elicited from pre-trial stages because the standard of proof at those stages is lower than the standard the Court was using in Bosnia v. Serbia.Footnote 50 However, this decision notwithstanding, the Court made a point of noting that while the inclusion of charges in an indictment could not be accorded evidentiary weight, the Prosecutor's decision to ‘not include or to exclude’ a genocide charge from an indictment might be ‘significant’.Footnote 51 The Court did not, however, offer any explanation for why this would be so.
Having laid out this framework, the Court then proceeded to consider the factual evidence presented to it under the categories established by Article II of the Genocide Convention. The subsequent 122 paragraphs of the judgment are largely devoted to laying out the evidence presented by the parties, and in particular summarizing the findings of the judgments of the ICTY. In the process, the Court found there to be persuasive and conclusive evidence that the material elements (actus reus) of genocide were committed with respect to Genocide Convention Article II(a)Footnote 52 (killing members of a protected group), Article II(b)Footnote 53 (causing serious bodily or mental harm to members of the protected group), and Article II(c)Footnote 54 (deliberately inflicting on the group conditions of life intended to bring about the destruction of the group in whole or in part), but not with respect to Articles II(d)Footnote 55 or II(e).Footnote 56
With regard to dolus specialis, however, the Court found that in all instances, except for the massacres at Srebrenica, the specific intent element required to reach a finding of genocide was not established.Footnote 57 The Court's primary basis for finding that the specific-intent element of genocide was established with respect to the massacres at Srebrenica was the ICTY trial and appellate judgments in Prosecutor v. Krstić.Footnote 58 In the Krstić judgments the ICTY Trial and Appeals Chambers found that genocide had been committed in Srebrenica, and the Court concluded that it had ‘no reason to disagree’ with the ICTY's findings on this issue.Footnote 59 Accepting its reasoned explanation for why it would give evidentiary significance to the judgments of the ICTY,Footnote 60 this part of the Court's judgment is unobjectionable. However the Court's basis for finding that specific intent had not been established in other situations in which the material element of genocide had been established was much less convincing.
3.1.1. Drawing inferences from the absence of ICTY convictions
With respect to the massive killings that the Court found had been committed in areas of Bosnia and Herzegovina other than Srebrenica, the Court's justification for finding that the specific-intent element had not been established was on the grounds that it ‘carefully examined the criminal proceedings of the ICTY and the findings of its chambers . . . and observes that none of those convicted were found to have acted with specific intent (dolus specialis)’.Footnote 61 The Court applies the same justification to its finding that the specific-intent element was not established with respect to the material acts it found had been committed against the protected group under Article II(c) of the Genocide Convention in a number of detention camps across Bosnia and Herzegovina. In explaining its inability to find specific intent, the Court states that ‘in none of the ICTY cases concerning the camps cited above has the Tribunal found that the accused acted with such specific intent’.Footnote 62
The problem with the Court's reasoning is that the question before it at that stage was whether genocide had occurred in Bosnia and Herzegovina, not whether genocide was committed by the relative handful of individuals who have to date been prosecuted by the ICTY. The ICTY does not have the resources or the mandate to investigate every possible charge of genocide arising out of the horrific crimes committed in Bosnia and Herzegovina since 1991. Furthermore, the ICTY was never judging whether genocide occurred at a given location or time, but rather whether an individual before it was responsible for a particular act of genocide or not. It is therefore inappropriate to draw inferences about whether genocide did or did not take place based on what the ICTY chambers have not found to be substantiated beyond a reasonable doubt, in respect of any given individual.
Further problematic is that the Court appears to give weight to the absence of a conviction for genocide occurring in a place other than Srebrenica in instances in which the accused died before or during proceedings (Milošević, Talić, Kovačević and Drljača),Footnote 63 or in which a case with an indictment that includes a genocide charge has not yet gone to trial (Karadžić and Mladić).Footnote 64 In doing so, the Court implies that had Milošević, for example, survived trial and been convicted of the crimes for which he was indicted, or had Karadžić and Mladić been arrested and brought to trial, then the Court might have reached a different outcome in the Bosnia v. Serbia case before it. This line of reasoning is likely to be particularly frustrating for Bosnian victims when in a different part of the judgment the Court holds Serbia responsible for failing to arrest Mladić.Footnote 65 While the availability of evidence will always have an effect on the outcome of judicial proceedings, the availability or not of judgments before one tribunal should not be determinative of the outcome of proceedings before another independent body, especially if other sources of evidence or avenues of inquiry remain open, and least of all on a matter of this gravity and importance.
The Court's deference to the ICTY's findings was sensible to the extent that it would have been unnecessarily duplicative for the ICJ itself to determine those facts which had already been authoritatively established by the ICTY, and especially when there is general acceptance regarding the rigour of the ICTY's own fact-finding process.Footnote 66 However, the inferences it appeared to draw from the absence of findings – in circumstances where that absence may not be probative on the issue of whether or not genocide had actually taken place (as opposed to the distinct issue of whether the ICTY had proved that a particular accused before it had committed genocide) – show the dangers of an uncritical approach to the incorporation of actions and findings of one adjudicative body by another.
3.1.2. Drawing inferences from the absence of ICTY charges
The Court reaches its conclusion that specific intent was lacking in all situations other than Srebrenica not only by drawing negative inferences from the absence of ICTY genocide convictions; it also gives weight to the absence of ICTY genocide charges. In one of the most questionable portions of the judgment the Court applies this belief in the significance of what is not in an indictment to refute Bosnia and Herzegovina's proposition that the pattern of atrocities, committed across many regions, demonstrated specific intent. In coming to the conclusion that specific intent could not be inferred from the pattern of atrocities, the Court not only relies on the ICTY findings, but also on ‘actions of the Prosecutor [of the ICTY], including decisions not to charge genocide offences in possibly relevant indictments, and to enter into plea agreements . . . by which the genocide-related charges were withdrawn’.Footnote 67
Giving evidentiary weight to the Prosecutor's decision not to include a genocide charge in any given indictment, or to negotiate a plea agreement that involves withdrawing a genocide charge, is troublesome. First, the Prosecutor's decision not to charge genocide in an indictment may have nothing at all to do with the absence of evidence that genocide was committed in any particular situation. The evidence might indeed be conclusive as to the actus reus but wanting with regard to the criminal liability of the particular individual accused person or persons before the Tribunal. Second, crucial evidence may have been obtained from a state intelligence source under the provisions of Rule 70B of the ICTY Rules of Procedure and Evidence. Reference to such evidence in an indictment would thus be precluded. And, with regard to a plea agreement, the acceptance by the Prosecutor of a confession of guilt to a less serious crime than genocide might well be driven by the advantages of avoiding a lengthy trial or the unavailability of essential evidence. There might, indeed, be other weaknesses in the prosecution case that are unrelated to whether or not genocide was actually committed.
3.2. Can the genocide at Srebrenica be attributed to Serbia?
Having found that, with respect to the first question, the massacre at Srebrenica, led by members of the VRS, was the only instance in which genocide took place in Bosnia and Herzegovina,Footnote 68 the Court then turned to the second question of whether the Serbian state could be held responsible for that case of genocide. Because the genocide at Srebrenica took place in July 1995, after the formal withdrawal of the JNA troops from Bosnia and Herzegovina,Footnote 69 the critical issue before the Court was whether the acts of the VRS at Srebrenica could be attributed to the Serbian state. In this respect, the extent of control that the FRY had over the VRS became a critical issue.
The question of whether the Serbian state can be held responsible for genocide at Srebrenica is not an issue that the ICTY has the jurisdiction to answer, because its Statute confines it to deciding issues of individual criminal responsibility.Footnote 70 Thus while evidence obtained through the ICTY could go some way towards answering the question of state responsibility, the Court was unable to rely directly on the conclusions that the ICTY had drawn from the facts before it with regard to this question. Instead, the Court applied its two-part Nicaragua standardFootnote 71 and the ILC Articles 4 and 8 on State ResponsibilityFootnote 72 to evidence presented by the parties from a range of sources.Footnote 73
3.2.1. Access to unredacted documents
The Court stated at the outset of its analysis that it is the applicant – in this case Bosnia and Herzegovina – who has the burden of proof in establishing its case and proving the facts it asserts.Footnote 74 Bosnia and Herzegovina accepted this as a general proposition, but put forth the argument that the burden should be reversed on the specific question of whether acts of genocide could be attributed to Serbia. Its rationale for this request was that Serbia had refused to provide unredacted versions of documents from the FRY's Supreme Defence Council (FRY Council) meetings that Bosnia and Herzegovina believed would be probative on the issue of state control over the massacre at Srebrenica.Footnote 75
The FRY Council consisted of the highest-ranking political and military officers of the FRY and was the constitutionally highest authority over the military during the events in question.Footnote 76 In oral argument, counsel for Bosnia and Herzegovina argued that the unredacted versions of the minutes of the FRY Council meetings could be expected to show orders given by the FRY to armed forces in Bosnia and Herzegovina and payment by the FRY to officers in the VRSFootnote 77 – both of which would have been central to proving that the FRY exercised effective control over the actions of the VRS at Srebrenica.
In what has since become one of the most controversial parts of the Court's judgment,Footnote 78 the Court not only refused to request unredacted versions of the documents from Serbia (as the Court acknowledged it has the authority to do through its proprio motu powers under Article 49 of the Statute of the ICJ and Article 62 of the Rules of the Court);Footnote 79 it also refused to draw any inferences on account of Serbia's failure to provide unredacted versions of those documents.Footnote 80
In attempting to justify its decision, the Court noted that Bosnia and Herzegovina had ‘extensive documentation and other evidence available to it, especially from the readily accessible ICTY records’.Footnote 81 This is hardly persuasive, given that Bosnia and Herzegovina's reason for requesting unredacted versions of these documents was that it believed these documents would provide evidence on the issue of attribution that was not clear from the documentation it already had available to it. The unredacted documents were not available to Bosnia and Herzegovina from the ICTY because of a confidentiality order imposed by the Tribunal at Serbia's request.
3.2.2. Was the VRS a de facto organ of the FRY?
Having refused Bosnia and Herzegovina's request to demand unredacted versions of the FRY Council documents from Serbia, the Court assessed the question of attribution on the basis of the redacted versions of the FRY Council documents and other evidence presented by the parties.
With regard to whether the actors involved in the genocide at Srebrenica were de jure organs of the Serbian state, the Court looked to facts elicited through the ICTY and to intercepted documents, presented by Bosnia and Herzegovina, referring to the paramilitary group, the Scorpions, as a ‘unit of the Minister of the Interior of Serbia’ and as ‘the MUP of Serbia’.Footnote 82 In what could only be called a cursory consideration of the evidence on this point, the Court concluded that there was not sufficient evidence that either the VRS or Republika Srpska satisfied the standard of being completely dependent on the FRY,Footnote 83 or that the Scorpions were acting in complete dependence on the FRY.Footnote 84
In another telling example of the Court's dependence on the ICTY for fact finding, the Court noted that the Stanišić and Simatović case pending before the ICTY might provide evidence of a relationship of complete dependence between the Scorpions and the Serbian state. However, it went on to say that ‘the Court cannot draw further conclusions as this case remains at the indictment stage’.Footnote 85 Thus the question lingers as to whether a more expeditious trial in the ICTY's Stanišić and Simatović case could have provided evidence for the ICJ that the Scorpions were completely dependant on the FRY, thereby generating a potentially different outcome in the Bosnia v. Serbia case. Furthermore, if a relationship of complete dependence between the Scorpions and the FRY does come out subsequently in the Stanišić and Simatović case, then the finality of the ICJ's finding on this issue may be questioned.
3.2.3. Did the FRY have effective control over the VRS?
Having found that the test of complete dependence was not met, the Court moved to the next test of whether the FRY nevertheless had ‘effective control’ over specific operations carried out by the actors involved in the Srebrenica massacre. The Court referenced reports by the Secretary-General of the United Nations,Footnote 86 the Netherlands Institute for War Documentation,Footnote 87 and the US Central Intelligence Agency.Footnote 88 Again, the Court concluded that the evidence was not sufficient to establish that Serbia had effective control over the acts constituting genocide at Srebrenica, observing that
The applicant has not proved that instructions were issued by Federal authorities, or by any other organ of the FRY, to commit such massacres, still less that any such instructions were given with the specific intent characterizing the crime of genocide, which would have to be present in order for the Respondent to be held responsible on this basis.Footnote 89
The Court denied that Serbia had effective control over the actors at Srebrenica because of the absence of evidence of instructions from the FRY to commit the massacre. However, it is exactly such instructions that Bosnia and Herzegovina claimed were likely to be found in the unredacted versions of the FRY Council documents.
3.2.4. The ICJ's inadequate response to the issue of unredacted documents
Given that the Court found insufficient evidence to attribute the genocide at Srebrenica to Serbia, the Court's failure to request unredacted versions of documents that may have been probative on the issue is likely to damage the legitimacy of the Court's judgment in the eyes of many. Of course, only Serbia, the judges of the ICTY trial chamber, and some members of the ICTY Office of the Prosecutor know how probative the unredacted versions of the Supreme Defence Council documents actually are. However, it is not an unreasonable assumption that Serbia's argument that it cannot provide unredacted versions on the grounds of its national security interests, some 15 years after the period in question, is but a veil to keep hidden evidence of a more explicit link between the FRY and the acts taken by the VRS at Srebrenica.Footnote 90 The unredocted documents may have also provided evidence of the specific-intent element that the ICJ failed to find regarding the commission of genocide in other areas of Bosnia and Herzegovina.Footnote 91
What is particularly frustrating is that this is an issue that the Court could so easily have avoided. Not only did it have the authority to request the documents, but, if Serbia had refused its request, it also had the authority to draw a negative inference from Serbia's refusal. As Vice-President Al-Khasawneh points out in his dissent, the Court itself has previously explained that there is an inherent inequity in asking a victim state to provide evidence of the accused state's direct control when the accused state retains control over the proof that would establish such control.Footnote 92 Accordingly, the Court in Corfu Channel (United Kingdom v. Albania) Footnote 93 explained that the victim state ‘should be allowed a more liberal recourse to inferences of fact and circumstantial evidence’.Footnote 94 However, the Court took neither of the options readily available to it.
Even if, in actuality, the unredacted versions of the documents would not have provided the explicit link between the FRY and genocide at Srebrenica that the Court needed to hold the Serbian state responsible, the perception of unfairness generated by the Court's refusal even to ask for the documents is a sad legacy to leave following 14 years of litigation on this case. At its best, an international judicial process – like any form of adjudication that is perceived to be legitimate – has the potential to lay contested issues to rest, thereby allowing those affected to move into a phase of healing and a more stable form of coexistence, if not complete reconciliation. By refusing, without any plausible justification, to request unredacted versions of the documents, the Court undermined its potential to play this much needed role in the region.
4. Conclusion
In Bosnia v. Serbia, the UN's principal adjudicative body held that, as a matter of law, it can find a state responsible for the commission of genocide. As the first case to be adjudicated by the ICJ on this question in the almost sixty years since the Genocide Convention was adopted, this alone makes for a historic moment in international law. However, the ICJ's approach to dealing with legal and factual issues that overlapped with the work of the ICTY will also become a focal point for those interested in the development of international law in the coming years.
The ICJ's approach in Bosnia v. Serbia to the challenge the ICTY raised against the Nicaragua standard for assessing state responsibility can be viewed as being guided by principles of judicial restraint; the ICJ commented as much as was necessary for it to adjudicate the case before it, while reserving judgment on the appropriateness of the ICTY's approach to the questions it faces. However, the Court's approach might also be understood as having been driven by purely pragmatic concerns that steered the Court away from upsetting already decided cases before the ICTY.
Those who are interested in establishing a framework through which to mitigate the potential problems arising from the fragmentation of international law are generally concerned with a lack of consistency between different international bodies. However, to the extent that the ICJ's approach to the question of the appropriate legal standard for assessing state responsibility in Bosnia v. Serbia was driven by pragmatic concerns, the case draws attention to the issue of jurisprudential consistency within different adjudicative bodies. Any attempt to standardize inter-court relations should be sufficiently flexible as to accommodate the – sometimes competing – need for intra-court jurisprudential consistency.
While the issues arising from the failure of one international court to defer to the previous judgments of another international court have been articulated at length in the academic literature,Footnote 95 our critique of the fact-finding approach taken by the ICJ in the Bosnia v. Serbia judgment suggests that scholars, judges, and litigants should perhaps be equally concerned about the potential problems arising from an inappropriate reliance on the findings of other courts. Furthermore, this is not a concern unique to the ICJ. The Pre-trial Chamber of the ICC recently relied on the ICJ's factual findings of the presence of Ugandan forces as an occupying power in IturiFootnote 96 in order to decide that, contrary to the changes presented by the prosecution, the nature of the conflict within which the alleged crimes took place was international, not non-international.Footnote 97 As the case load of the ICC grows, one can predict the question of the circumstances under which it is appropriate for one international body to adopt the factual findings of another international body will become a significant issue for those interested in international criminal cases.
However, it is in terms of the implications for future applicants trying to establish state responsibility for genocide before the ICJ that the fact-finding approach taken by the Court in this case raises the most concerns. If one is to accept the Court's position that it has the ability and jurisdiction to make a positive finding of state responsibility for genocide, even in the absence of a genocide conviction for any individual perpetrator,Footnote 98 then the Court should not use the absence of a genocide conviction to preclude it from finding that genocide was committed.Footnote 99 If its fact-finding approach in Bosnia v. Serbia is to set any precedent for how the ICJ will adjudicate future cases under the Genocide Convention, it is hard to see how the Court will ever make a positive genocide determination in the absence of a criminal court having already convicted individual perpetrators of genocide. On the basis of this case, the Court's approach seems to be that if another judicial body with jurisdiction over the events at issue has not already established genocide, the ICJ will not either. If this is correct, then the Bosnia v. Serbia judgment may have taken away with one hand what it has offered with the other – promising an international legal system that can hold states accountable for the commission of genocide, while simultaneously ensuring that in practice that promise will only rarely be fulfilled.
