Kevin Aquilina & Klejda Mulaj. Journal of Human Rights. Volume 17, Issue 1. Jan-Mar 2018.
Protection from genocide has been a common denominator in state rhetoric since 1948 when the Genocide Convention was adopted. However, state accountability for this archetypical crime of the state is virtually nonexistent. This article addresses a two-pronged puzzle: namely, (1) why, no government involved in the commission of genocide has to date been held responsible for it; and (2) how legal processes of the sole court that addresses states’ disputes regarding genocide, the International Court of Justice, condition and even limit the quality of decisions taken by the court with particular reference to state liability for this crime. The analysis contributes to an emerging debate on the application of state responsibility with reference to the protection from genocide by highlighting existing shortcomings pertaining to the interpretation and implementation of the Genocide Convention that, in turn, warrants a holistic revision of this treaty.
Genocide is no ordinary crime. It has been labeled as “contrary to the spirit and aims of the United Nations” and as “an odious scourge” in the preamble to the Convention on the Prevention and Punishment of the Crime of Genocide (United Nations: 278), hereinafter referred to as “the Genocide Convention” or “the convention,” as well as “the ultimate human rights problem” and “the crime of all crimes” (Schabas: 7, 15). Whilst most human rights conventions and declarations relate to the individual’s right to life, the Genocide Convention is concerned with human groups’ right to life. Resolution 96(I) adopted by the UN General Assembly in December 1946 draws this group versus individual rights comparison in most clear terms: “Genocide is a denial of the right of existence of entire human groups, as homicide is the denial of the right to live of individual human beings; such denial of the right of existence shocks the conscience of humankind, results in great losses to humanity in the form of cultural and other contributions represented by these human groups, and is contrary to moral law and to the spirit and aims of the United Nations” (United Nations: 188-189).
Nearly 70 years since its adoption, the Genocide Convention has been the subject of litigation between states and criminal prosecution of individuals before a number of international courts and tribunals such as the International Court of Justice (ICJ or “the court”), the International Criminal Tribunal for Rwanda (ICTR), and the International Criminal Tribunal for the Former Yugoslavia (ICTY). There are other international courts and tribunals currently seized of genocide cases that have not yet been determined. Such is the case of the International Criminal Court (ICC). The same applies equally to national courts, such as the Extraordinary Chambers in the Courts of Cambodia (Ciorciari and Heindel). However, notwithstanding the institution of these cases, few and far between have been those instances where an international court or tribunal has found that genocide has been perpetrated by an individual and virtually never by a state. Indeed, judgments delivered so far by international courts and tribunals with regard to the application of the Genocide Convention indicate that genocide has been found to have been committed in only a limited number of cases. In so far as the ICTR is concerned, there have been few such occasions when compared to victims’ suffering, the first two landmark decisions being Prosecutor v. Jean Kambanda (ICTR) and Prosecutor v. Jean-Paul Akayesu (ICTR). At the end of June 2016, the ICTY had determined four genocide decisions: Prosecutor v. Radislav Krstić (ICTY); Prosecutor v. Vujadin Popović et al. (ICTY); Prosecutor v. Zdravko Tolimir (ICTY); and Prosecutor v. Radovan Karadžić (2016). The ICJ, to date, has delivered only one genocide judgment, namely the Case Concerning Application of the Convention on the Prevention and Punishment of the Crime of Genocide, Bosnia and Herzegovina v. Serbia and Montenegro (ICJ), though the latter case was limited to the obligation to prevent genocide at Srebrenica in 1995 and the ICJ relied heavily on evidence produced before the ICTY to arrive at that conclusion.
It is puzzling that, although genocide is a collective crime against groups inconceivable without the involvement of the state, indeed considered as the archetypical crime of the state, no government involved in the commission of genocide has to date been held responsible for this heinous crime. Whilst individual responsibility for genocide has been established for a handful of perpetrators, the state on whose behalf they operated has not borne the brunt of its criminal enterprise. The Genocide Convention imposes some obligations upon states but it does not provide explicitly that states may be held responsible for the crime of crimes. The centerpiece of the genocide law, the Genocide Convention, posited from a criminal-justice perspective, is concerned, primarily, with prosecution of individual perpetrators rather than their master sponsor, their government. It was hoped that state responsibility for genocide would be finally discharged in the latest judgement decided by the ICJ. However, this was not the case. Indeed, as shown below, the provisions of the convention and its technical interpretation afforded by the court disadvantaged the victim state. Coupled with the legal culture of the court that is more reverential to the doctrine of state sovereignty rather than to emphasize the state’s assumption of responsibility for atrocities committed, these drawbacks are contributing to the Genocide Convention losing its efficacy.
The second aspect of the “puzzle” of this article concerns how the ICJ’s legal process conditions the (quality of) decisions taken by the court. The article considers specifically the ICJ genocide cases with a view to shedding light on the limitations inherent in the convention as well as in the ICJ’s legal culture that are contributing to make the infliction of punishment for genocide more difficult to achieve. These are noticeable omissions in the existing large interdisciplinary genocide literature, that are rendering a disservice to victims whilst molding the impression that genocidaries can, through legal contortions, avoid the dispensation of justice in their regard, thereby nullifying the convention’s objectives. Indeed, with regard to the ICJ, there are inherent difficulties in international law, as interpreted and applied by the court, which make it more problematic for the court to establish state responsibility for genocide. This article aims to address the limitations inherent in the Genocide Convention on attributing state responsibility for the worst crime known to humankind, by reference to genocide cases decided by the ICJ.
A review of the judgments delivered by the ICJ in its entire history indicates that the court has dealt with the crime of genocide in only two decided cases, namely Bosnia v. Serbia and the Case Concerning the Application of the Convention on the Prevention and Punishment of the Crime of Genocide (ICJ). In the former case, Bosnia and Herzegovina had requested the court to adjudge Serbia responsible for violations of the Genocide Convention in the context of the Bosnian War (1992-1995) (ICJ: 64). In the latter case, it was Croatia that made a similar request (ICJ: 50). Here, Serbia retorted through a counterclaim by requesting the court to declare Croatia responsible for perpetrating genocide against Croatia’s Serbs (ICJ: 51). In both cases, Serbia rejected the concept of state criminal responsibility and denied any violation of the Genocide Convention. It was only in Bosnia v. Serbia (ICJ) that the ICJ established state responsibility for failing to prevent genocide on Serbia’s side even if the court’s failure to find Serbia guilty of the principal crime of genocide (rather than its accessory to) or of failure to prevent genocide, did not go as far as the Bosnian party had requested and expected from the court’s ruling. This indeed is perceived by Bosnians (who adhere to the idea of Bosnian unity and statehood) as a weakness in the court’s decision that dented the kind of outcome that the Bosnian side expected from those judicial proceedings. However, in many ways this was a landmark finding in the history of the ICJ that such responsibility was attributed even in a limited way. Nevertheless, this was also a case that exposed impediments in attributing state responsibility for the commission of genocide. Reasons that can be adduced for limitations inherent in the convention as well as in the ICJ’s legal process pertain to the following: (1) ambiguities in the definition and meaning of genocide; (2) the adoption of a high standard of proof to ground state responsibility for genocide; (3) the exclusion by the Genocide Convention of the applicability of nontreaty sources of international law; and (4) the lack of a finding of state responsibility for offenses of a lesser aggravated character than that of genocide. The ensuing analysis begins by engaging with the meaning of state responsibility for genocide prior to exploring the abovementioned limitations to the attribution of state responsibility for genocide via a close reading of the convention and the ICJ’s related cases.
State Responsibility for Genocide
The concept of “state responsibility” has a fascinating history connected directly with the formation of the United Nations and UN discussions on the text of the Genocide Convention that make clear the political and legal intricacies inherent in the term. Before referring below to those discussions, it is necessary to clarify what we mean by “state” in the context of this contribution. At first glance, it may not appear obvious that there is an ambiguity pertaining to the definition of state itself because, in the wide use of the term in various disciplines such as political science, law, sociology, history, etc., this word is frequently taken for granted. Given its centrality to various disciplines, it is not surprising that its meaning may vary depending on the nature of the research question and the context of analysis. In general, there are two broad conceptions of state: a national-territorial concept according to which the state comprises the whole territory denoted on a map and all that is within it (people, government, resources) and a more limited, institutional concept of state. While both conceptions are heuristic abstractions, the most appropriate in the context of our analysis is the second, that is, the institutional conception of the state. Hence, with Theda Skocpol, we understand “state” to mean: “a set of administrative, policing and military organizations headed, and more or less well coordinated, by an executive authority” (Skocpol: 29). Therefore, state responsibility for genocide in the context of this article refers to responsibility of a state’s government, administration, or institutions that orchestrated, oversaw and directed the state’s genocide policy.
The issue of genocide was first included on the agenda of the UN General Assembly (UNGA) in its first session on October 1, 1946. Ten days after, the UNGA adopted Resolution 96(I) that recognized that genocide is a crime under international law, invited member states to enact legislation for the prevention and punishment of that crime and recommended the drawing up of a convention to be submitted to the General Assembly (United Nations). Work on the drafting of the convention was referred to the Sixth UNGA (Legal) Committee. Afterwards, for more than two years, the text of a draft resolution on genocide was debated in this committee, in the Ad Hoc Committee, and the General Assembly. From the inception of these discussions, controversy pertained to the referent of criminal responsibility for genocide. Notably, the representative of the United Kingdom, Sir Hartley Shawcross, consistently expressed the view that genocide “is an international crime for the commission of which principals and accessories, as well as States, are individually responsible” (United Nations: para. 57). The UK representative reiterated this view too in the Sixth Committee meeting of November 17, 1947 (United Nations). Moreover, in a subsequent meeting on October 16, 1948, the UK representative, Sir Gerald Fitzmaurice, suggested an amendment to Article V that made direct reference to the concept of state responsibility for genocide, apart from that of the individual: “Criminal responsibility for any act of genocide as specified in Articles II and IV shall extend not only to all private persons or associations, but also to States, Governments, or organs or authorities of the State or Government by whom such acts are committed” (United Nations). Sir Gerald Fitzmaurice argued that both the responsibility of governments, and that of private individuals, should be clearly expressed in Article V of the convention (United Nations).
The UN discussions on state responsibility for genocide show that the British view that responsibility for genocide ought to belong not only to individuals but also to states was endorsed by many governments, for example, those of Belgium, Bolivia, Cuba, Ecuador, Luxemburg, Syria, Venezuela, etc. (refer, for instance, to United Nations). However, the British view was contested by others. France, for example, disputed this view on the grounds that French law made no provision for criminal responsibility of states (United Nations). Retention of the concept of governmental responsibility in the text of the Genocide Convention was strongly opposed also by Brazil, China, Greece, India, Lebanon, Poland, etc. (United Nations). In the end, the UK amendment was rejected by a flimsy margin of only two votes (24 votes to 22) (United Nations). The ground for this rejection, however, seems to have to do more with the semantics of formulation of “state responsibility” rather than the idea itself (United Nations). However, although the United Kingdom’s amendment was defeated by a very thin majority, this does not mean that the concept of state responsibility for genocide has been laid to rest. On the contrary, the narrow vote indicates that there was agreement between states that the latter may be civilly liable therefor.
The final consideration of the draft convention in the Sixth Committee led to the adoption of the Genocide Convention on December 9, 1948. The adopted text envisaged individual criminal responsibility for genocide but did not provide for state responsibility for genocide. Disappointed, Sir Gerald Fitzmaurice opined that the convention “approached genocide from the wrong angle, the responsibility of individuals, whereas it was really governments that had to be the focus” (United Nations e: para. 85). Similarly, Sir Hartley Shawcross was disenchanted by the prospects of a convention excluding state responsibility for genocide. In his view, it was a “complete delusion to suppose that the adoption of a convention of the type proposed, even if generally adhered to, would give people a greater sense of security or would diminish existing dangers of persecution on racial, religious, or national grounds” (United Nations: para. 6).
Although the Genocide Convention does not provide explicitly that states may be held responsible for genocide, the idea that a state can be liable for this crime was initially supported by the International Law Commission in its 1976 draft principles on state responsibility where Article 19 thereof defines a state crime as “an internationally wrongful act which resulted from the breach by a state of an international obligation so essential for the protection of the fundamental interests of the international community that its breach was recognized as a crime by that community as a whole.” According to Article 19, “an international crime may result, inter alia, from … a serious breach on a widespread scale of an international obligation of essential importance for safeguarding the human being, such as those prohibiting … genocide.” For Law Professor James Crawford, this is “[t]he single most controversial element in the draft articles on State responsibility.” The commission reconsidered the issue of state crimes at its 1998 session, when it decided that it should “be put to one side” (United Nations: para. 331[a]) due to the legal difficulties in placing state crimes within the framework of state responsibility, in establishing the differences between individual and state responsibility, and in extending state liability within a civil law context. In the August 2001 version of the International Law Commission’s Draft Articles on Responsibility of States for Internationally Wrongful Acts, the words “state crime” do not feature at all, though James Crawford maintains that there exist international crimes such as aggression and genocide “which are committed mainly or only by State agencies” (Crawford: 19).
Nonetheless, the Draft Articles on Responsibility of States for Internationally Wrongful Acts (International Law Commission) hold states responsible for internationally wrongful acts (Crawford et al.). Such acts materialize when there is a breach of an international obligation irrespective of whether that obligation arises from a treaty, a custom, or a peremptory norm. According to Article 53 of the Vienna Convention on the Law of Treaties 1969, a peremptory norm of international law, also known as jus cogens, “is a norm accepted and recognized by the international community of States as a whole as a norm from which no derogation is permitted and which can be modified only by a subsequent norm of general international law having the same character” (United Nations). In the case of genocide, this crime is regulated by all the three above-listed sources of international law and, hence, state responsibility for the crime of genocide is threefold (though this article presents its analysis from a strictly conventional approach). A state manifests its conduct through a person or a group of persons who act on the “instructions of, or under the direction or control of, that State” (International Law Commission 2001: Article 8).
As the history of the debate on state responsibility in the process of drafting the Genocide Convention reviewed above shows, although the state has been considered to be the main perpetrator of genocide, the convention does not make explicit that the commission of that crime entails state responsibility, complicity, or involvement. The ICJ, nevertheless, has finally ruled that states may commit genocide and the other acts enumerated in Article III of the Genocide Convention and, therefore, incur the responsibility for this crime (ICJ: 168-169, 174 and 471, subpara.). Moreover, the court has stated that it “will have recourse not only to the Convention itself, but also to the rules of general international law … on responsibility of States for internationally wrongful acts” (ICJ: 149). In many ways, this is a landmark pronouncement that raises hopes for attribution of state responsibility for genocide, although these hopes did not fully materialize in the judgements referred to in this analysis.
In legal parlance, “responsibility” is frequently used interchangeably with “obligation” or “duty.” The United Nations Convention enshrines the concept of positive obligations when it refers to a state’s responsibility to prevent genocide. Human rights law — more recent to, and following on the path of, the Genocide Convention — also sanctifies the concept of “positive obligations” (Akandji-Kombe). States do not have only an obligation to ensure that they do not commit breaches of human rights law but also a positive obligation to take the necessary measures to safeguard human rights. A state has indeed a proactive duty to guarantee the enjoyment of human rights rather than restrict itself to simply punish a violation of human rights through coercive force after the crime would have been executed. In concrete form, the state’s positive obligation to prevent genocide takes the form of (1) the duty to protect life, (2) the taking of positive measures, and (3) the duty to investigate and prosecute (Harris et al.: 36-46, 48-52).
The Genocide Convention, to its credit, embraced positive obligations before this concept matured in human rights law and developed to the extent that it is known today (Mowbray; Xenos). The convention in Article I imposes a responsibility on states to prohibit genocide as evident in the convention’s title and also in the very opening provision of the convention whereby the contracting parties pledge to prevent genocide. Moreover, the Genocide Convention in Article VIII empowers contracting parties “to call upon the competent organs of the United Nations to take such action under the Charter of the United Nations as they consider appropriate for the prevention and suppression of acts of genocide or any of the acts enumerated in Article III” (ICJ: 425-432).
Central to Bosnia v. Serbia (ICJ) was the issue of whether a state could perpetrate genocide rendering itself liable and in breach of the convention. A detailed summary of Bosnia v. Serbia (ICJ) and Croatia v. Serbia (ICJ) cases is not easy to provide in the limited space available here. In brief, Bosnia contended that following the breakup of the Socialist Federal Republic of Yugoslavia (SFRY), Serbia had committed atrocities with the intent to destroy in whole or in part the protected group of Bosnian Muslims. Relying on ICTY evidence, the ICJ confirmed that the SFRY government in Belgrade had provided considerable support to Republika Srpska (the self-proclaimed Republic of the Serb People of Bosnia). The court found that massive killings were perpetrated during the Bosnian War but the specific genocidal intent on the part of the perpetrators was missing. It is only in the massacre of Srebrenica of July 11-19, 1995, where more than 7,000 Bosnian Muslim men and boys were killed by the Serb forces, that genocidal intent can be discerned, the court found. Hence, the Srebrenica massacre was qualified as genocide. Nevertheless, the specific intent was to destroy Muslims in the whole of Bosnia not conclusively established by the ICJ. The court decided that Serbia had not committed genocide, nor had conspired to commit genocide, nor had incited the commission of genocide nor was it complicit in genocide. But Serbia had violated the obligation to prevent genocide in Srebrenica (ICJ). In the other ICJ genocide case of Croatia v. Serbia (ICJ), Croatia maintained that Croat Serb forces and the Yugoslav National Army which opposed the independence of Croatia perpetrated genocide against Croat people living in the so-called Krajina Republic — comprising around one third of the Croatian territory controlled by the Serb military — between 1991 and 1995. As in Bosnia v. Serbia, evidence of genocidal intent was to be sought, first, in the state’s policy, but it could also be inferred from a pattern of conduct, where such intent is the only reasonable inference to be drawn from the acts in question. It was for the party alleging a fact, that is, genocide, to demonstrate its existence. The court established the actus reus of genocide but that the intentional element of genocide was lacking and, accordingly, rejected Croatia’s claim in its entirety (ICJ: 441).
Many observers and victims’ relatives had hoped that the ICJ would establish Serbia’s state responsibility for genocide in Bosnia, or at least in Srebrenica where genocide was judiciously proven to have occurred, especially since the ICJ affirmed that states can commit genocide. Moreover, the ICJ relied heavily on ICTY evidence that showed beyond a reasonable doubt the very large extent of military and financial aid Serbia granted her ethnic brethren in Bosnia. In finding Serbia guilty for not preventing genocide in Bosnia v. Serbia (ICJ), the ICJ ruled on an aspect of state responsibility —responsibility to prevent — that remains quite undeveloped, since the duties related thereto are not set out with precision by the convention.
A careful reading of the convention indicates that its main thrust is placed on punishment rather than on prevention. The various forms of genocidal acts, its mens rea, the types of genocidal crimes, and other matters related to the commission of the crime of genocide in its differing manifestations are set out in detail in the convention. Yet, the positive obligation consisting of state responsibility to prevent the perpetration of genocide is left in essence within the discretion of the contracting parties to determine how to apply it in practice. The court has contributed to the elucidation of this obligation as one of conduct, not one of success, in preventing genocide. As the ICJ has held, a “State does not incur responsibility because the desired result is not achieved” (ICJ 2007: 430). Further, states are, however, required to use their “capacity to influence effectively the actions of persons likely to commit, or [are] already committing, genocide” (ICJ: 430). But a state is held responsible for breaching the obligation to prevent genocide only when genocide actually takes place (ICJ: 431).
Although the duty of states to prevent genocide is mentioned in Article I, the convention is conspicuous by its absence of developing such obligation, leaving it within the discretion of contracting parties to implement this requirement. Should such parties fail to comply with this obligation, then it is upon the international society of states to carry forward this duty according to Article VIII of the convention. Preventive measures can take the form of nonjudicial and judicial measures, as enumerated in Article 33 of the United Nations Charter, but the convention does not make reference to such measures. Nonetheless, when preventive diplomacy fails, recourse to coercive measures can be had in accordance with the UN Charter, as dictated by Article VIII of the convention. But at this stage, the harm would have already been done, the breach of an international obligation sustained, and, in all probability, the injury suffered would be irremediable.
Ambiguities Related to, and Limitations in, the Definition of Genocide
This article contributes to an emerging debate on the application of state responsibility with reference to the protection from the crime of genocide by highlighting existing limitations pertaining to the interpretation and implementation of the Genocide Convention. One such difficulty lies precisely in the ambiguities raised by, and the limitations inherent in, the wording of the convention with regard to the definition of “genocide.” Genocide is defined in Article II of the convention as the following:
[A]ny of the following acts committed with intent to destroy, in whole or in part, a national, ethnical, racial or religious group, as such: (a) killing members of the group; (b) causing serious bodily or mental harm to members of the group; (c) deliberately inflicting on the group conditions of life calculated to bring about its physical destruction in whole or in part; (d) imposing measures intended to prevent births within the group; (e) forcibly transferring children of the group to another group. (United Nations)
This definition is ambiguous on various counts.
First, there is the problem of establishing the perpetrators’ special genocidal intent, the mens rea of genocide consisting in the dolus specialis, which has been interpreted by the ICTR (United Nations: 91) and by the ICJ in Bosnia v. Serbia as being required by the convention to ground genocide. The same applies to requiring the group’s destruction in part as meaning “considerable” when this interpretation is not evident from the convention’s wording (Kent: 577-578). The mental element consists in terms of Article II of the convention in the “intent to destroy, in whole or in part, a national, ethnical, racial or religious group as such” coupled with the intent to commit the individual acts concerned enumerated in Article II, Paragraphs (a) to (e) of the Genocide Convention. Yet, what seems at face value to be straightforward is far from being so. Difficulties in interpretation have arisen, even as lately as 2015, with regard to Article II of the Convention before the ICJ in Croatia v. Serbia on three main issues: “(1) on the meaning and scope of ‘destruction’ of a group, (2) on the meaning of destruction of a group ‘in part’, and finally (3) on what constitutes the evidence of the dolus specialis” (ICJ: 133). In this respect, the court was requested to rule on whether destruction was limited to physical or biological destruction of the group, thereby excluding “the intent to stop it from functioning as a unit” (ICJ: 133). The ICJ ruled in favor of this limitation. In doing so, the court relied on the decisions delivered by the ICTY and the ICTR not only with regard to the physical destruction of the group but also as to the targeted group, genocide and lesser crimes, the actus reus of genocide, and the definition of genocide. The court was also asked as to whether the extermination of the group was required, to which it ruled that it was difficult to establish the genocidal intent “on the basis of isolated acts” (ICJ: 139) and that there had to be destruction of “the group itself in whole or in part” (ICJ: 139). As to what constitutes the meaning of destruction of the group “in part” as opposed to “in full,” the court opined that the targeted part of a protected group had to be “substantial in relation to the overall group.” In establishing what constitutes a substantial part of the particular group, the court looked at “the quantitative element and evidence regarding the geographic location and prominence of the allegedly targeted part of the group” (ICJ: 198; ICJ: 142). The court also defined the target group positively, that is, not “negatively as the ‘non-Serb’ population” (ICJ: 196) and required the consideration of the opportunity available to the alleged perpetrator to commit genocide within a geographically limited area as being material to establish the dolus specialis (ICJ: 199). The court has adopted a qualitative criterion in establishing genocide: “If a specific part of the group is emblematic of the overall group, or is essential to its survival, that may support a finding that the part qualifies as substantial” (ICJ: 200). Finally, as to the evidence required of the dolus specialis, the court considers that it should be manifest in a state’s declared policy (Götz: 243-263 and 264-272; Browning: 374-415; Hochstadt: 5; ICJ: 143) or, in the absence of such declaration, is “established by indirect evidence, that is, deduced or inferred from certain types of conduct” (ICJ: 148). With regard to the former, that is, evidence of genocidal intent, it can be found in official state policy, as has been established in other genocide cases most prominently in the case of the Holocaust with reference to the Final Solution, Nazi Germany’s declared policy on the extermination of the Jews. As to the latter, the court “accepted the possibility of genocidal intent being established indirectly by inference” through the notion of reasonableness, that is, “it is necessary and sufficient that this is the only inference that could reasonably be drawn from the acts in question” (ICJ: 148).
All the above judicial interpretations of the genocidal intent are fundamental to the application of the Genocide Convention. Yet, such fundamental interpretations should not have been left to the court to determine but to the contracting parties to adopt them in full, or to subject them to such modification(s) that they deem necessary, in a protocol to the Genocide Convention. Notably, first, the Genocide Convention was concluded in 1948 and, since then, a number of international and national courts and tribunals have had the opportunity to interpret its provisions. But no holistic revision of the Genocide Convention has ever taken place in this period within the UN General Assembly, its creator, notwithstanding the various occurrences of genocide since the end of the Second World War. Second, the Genocide Convention does not quantify the targeted group, that is, the actual size thereof. If the group is atomized to the level of a town or village, this would have the potential disadvantage of classifying nongenocidal cases as genocides. The ICJ seems to have averted this incongruence by stating that the group has to be “substantial in relation to the overall group” (ICJ: 142). However, the “substantial” feature of the group is left undefined by the convention. The judicial classification of the group on a case-by-case basis does not bring about either the needed clarity or uniformity of the definition of the group.
Third, the Genocide Convention (or international law, for that matter) does not distinguish between genocide and lesser crimes such as ethnic cleansing. In terms of the convention, ethnic cleansing has no legal significance of its own. No universally agreed definition of “ethnic cleansing” therefore exists but scholars, journalists, and policymakers frequently use the term to refer to a deliberate policy pursuit with the aim of removing and dispossessing an undesirable indigenous population (defined on ethnonational terms) by means of force and/or intimidation (Mulaj: 4). Ethnic cleansing is not necessarily genocide, although the ICJ has admitted that there can be situations where ethnic cleansing may satisfy the constitutive elements of genocide. The court opines the following:
This is not to say that acts described as “ethnic cleansing” may never constitute genocide, if they are such as to be characterized as, for example, deliberately inflicting on the group conditions of life calculated to bring about its physical destruction “in whole or in part”, contrary to Article II, paragraph (c), of the Convention, provided such action is carried out with the necessary specific intent (dolus specialis), that is to say with a view to the destruction of the group, as distinct from its removal from the region…. In other words, whether a particular operation described as “ethnic cleansing” amounts to genocide depends on the presence or absence of acts listed in Article II of the Genocide Convention, and of the intent to destroy the group as such. (ICJ: 190)
Ethnic cleansing, contrary to genocide, cannot be prosecuted in its own right for international criminal law does not recognize an international crime of ethnic cleansing. However, some of its constitutive elements may fall under subdivisions of international criminal law such as crimes against humanity or war crimes and, where there is the genocidal intent, the actus reus of ethnic cleansing may fall under the crime of genocide despite the fact that the mens rea of ethnic cleansing does not.
Further, ethnic cleansing may, in certain situations, point towards the specific intent required to prove that genocidal acts have taken place. It is indeed here that international law appears to complicate matters for an applicant state before the ICJ attempting to obtain a declaration from the court of state responsibility for genocide by the respondent state because, although it recognizes genocide as the crime of all crimes, international law fails to recognize ethnic cleansing as a separate and distinct crime in its own right rather than as an appendage to the crime of genocide or to other international crimes. In addition, although the convention correctly penalizes genocide, it can be extremely difficult for the ICJ to attribute state responsibility for the crime of genocide, due to the doctrine of precedent adopted by the ICJ that requires it to adopt a high standard of proof to ground state responsibility for genocide, as explained in the following section. The contradiction thus lies in the fact that, whilst genocide is an international crime, state perpetrators may run scot free on two counts: first, because the probative standard to meet it is too high and, second, because if genocide cannot be proved due to the high probative standard required by the ICJ from the applicant state, the said state cannot rely on the lesser offense of ethnic cleansing once the latter conduct does not, in itself, amount to an international or domestic crime, and the Genocide Convention does not allow the ICJ to find state responsibility for a lesser offense even if the court finds in its judgment responsibility for such lesser offense but not for genocide.
Fourth, the actus reus of genocide has given rise to ambiguities in its interpretation that the ICJ has attempted to resolve. The convention, in Article II, stipulates that the actus reus may take the form of five distinct and separate acts. The court has had occasion, in its case law, to interpret some of these genocidal acts. Genocide is a collective crime in the sense that victims are part of a group defined, in the words of Article II of the convention, as “a national, ethnical, racial or religious group” (Van der Wolf: 29-32). The word “killing” in the genocidal act of “killing members of the group” in Article II(a) of the convention refers “to the act of intentionally killing members of the group” (ICJ: 186-187; and ICJ: 156). The issue that arose before the ICJ with regard to the second genocidal act is what constitutes “serious” bodily or mental harm to members of the group. The court interprets “serious” as referring to bodily or mental harm that contributes “to the physical or biological destruction of the group, in whole or in part” (ICJ: 157). Rape and other acts of sexual violence are considered by the court as falling under this genocidal act (ICJ: 159). Furthermore, the court’s view is that “the persistent refusal of the competent authorities to provide relatives of individuals who disappeared in the context of an alleged genocide with information in their possession, which would enable the relatives to establish with certainty whether those individuals are dead, and if so, how they died, is capable of causing psychological suffering,” but the said harm “must be such as to contribute to the physical or biological destruction of the group, in whole or in part” (ICJ: 160). As to the genocidal act of “deliberately inflicting on the group conditions of life calculated to bring about its physical destruction in whole or in part,” the ICJ has held that this covers “physical destruction, other than killing, whereby the perpetrator ultimately seeks the death of the members of the group” (ICJ: 161). It also considers forced displacements that take place “in such circumstances that they were calculated to bring about the physical destruction of the group” (ICJ: 163) as falling under this genocidal act. Another genocidal act is “imposing measures intended to prevent births within the group” that the ICJ has interpreted to mean that “rape and other acts of sexual violence” constitute the actus reus of genocide if “they are of a kind which prevents birth within the group. In order for that to be the case, it is necessary that the circumstances of the commission of those acts, and their consequences, are such that the capacity of members of the group to procreate is affected” (ICJ: 166). Once contracting parties have raised these ambiguities in the interpretation of the actus reus of genocide and the ICJ has attempted to clarify the convention, it would be appropriate if these clarifications are approved by the said parties through specific amendment to the convention to support and give more weight to the court’s interpretation.
Standard of Proof to Ground State Responsibility for Genocide
The standard of proof varies from one international court and tribunal to another. There is thus no consistency and uniformity in the standard of proof adopted by diverse international courts and tribunals. The ICJ follows a strict stare decisis approach and will not “depart from previous findings, particularly when similar issues were dealt with in its earlier decisions … unless it finds very particular reasons to do so” (ICJ: 125). This means it is very difficult, not to say well-nigh impossible, for the court to alter its standard of proof in genocide cases unless, of course, the matter is dealt with through an amendment to the Genocide Convention. It also means that the standard of proof is case-law driven, not imposed upon the court by the contracting parties to the Genocide Convention. In so far as the ICJ is concerned, it had established its standard of proof in the Corfu Channel Case (ICJ: 16-17). The ICJ’s standard of proof requires a threefold test to be met. It has (1) “to constitute decisive legal proof,” (2) to be “conclusive evidence,” and (3) once “[a] charge of such exceptional gravity against a State [is made, it] would require a degree of certainty that has not been reached here” (ICJ: 125). In sum, the three-pronged standard of proof established by the ICJ in the Corfu Channel Case and applied in that case and subsequent ones has to be decisive, conclusive, and certain and well beyond proof beyond reasonable doubt, simply because the international actors involved are states and because the ICJ follows a strict stare decisis approach without departing from previous findings in relation to the standard of proof. The ICJ has restated this standard of proof in its more recent case law on genocide using the term “fully conclusive” evidence to sum up these three ingredients into a comprehensive one (ICJ). The ICJ’s high standard applies to states, contrary to the international criminal courts and tribunals that apply a lower standard to individuals prosecuted before them. The considerations in the Corfu Channel Case should be understood in the light of the fact that this judgment was the very first case to be decided by the ICJ at a time when the court was still in the process of asserting its newly acquired authority vis-à-vis states and of adopting what today can be viewed as an overcautious approach to ensure that contracting parties to the ICJ Statute are put at ease and are made comfortable enough to participate in proceedings before the court.
All the three above-cited cases decided by the ICJ demonstrate that the convention has been interpreted by the court as requiring an extremely high standard of proof. Whilst the burden of proof is placed on the party who alleges a fact (ICJ: 173), or, as the court put it, “it is for the party alleging a fact to demonstrate its existence” (ICJ: 172), the standard of proof has been established not as one on a balance of probabilities, or proof beyond reasonable doubt, as is usually the case in national civil or criminal procedures, respectively, but at a much higher level of “evidence that is fully conclusive” (ICJ: 209).
In accordance with the ICJ’s statement that “[s]tate responsibility and individual criminal responsibility are governed by different legal regimes and pursue different aims” (ICJ: 129), the court enshrines this statement in its judgment when it passes on to apply a higher standard of proof to establish state responsibility than international criminal courts and tribunals do with regard to identifying individual criminal responsibility. In this connection, the court draws a distinction between state responsibility and individual criminal responsibility: “The former concerns the consequences of the breach by a State of the obligation imposed upon it by international law, whereas the latter is concerned with the responsibility of an individual as established under the rules of international and domestic criminal law, and the resultant sanctions to be imposed upon that person” (ICJ: 129).
Although this distinction is legally sound, the same cannot be said with regard to the standard of proof. Within a national court scenario, the standard of proof varies depending on the nature or subject matter of the dispute. If the case is a criminal one, the standard of proof to be resorted to is proof beyond reasonable doubt; if the case is a civil one, the standard of proof resorted to is proof on a balance of probability. The criminal law standard is by far more stringent than that of civil law. Nonetheless, such stringency does make sense within a criminal law context as the punishment to be meted out can include, and normally would include in the case of conviction for the crime of genocide, deprivation of liberty through imprisonment. In a civil case, if a state is found to be responsible, normally reparations would have to be made good to the other state through the liquidation of damages. In the realm of international law, however, whilst international criminal courts and tribunals adopt the reasonable-beyond-doubt standard on the same lines as national criminal courts do, the International Court of Justice as an international civil court does not apply the standard of proof on a balance of probability but proof by fully conclusive evidence. Furthermore, the ICJ has not limited such proof to “conclusive evidence” but “proof by fully conclusive evidence.” The standard of “proof by conclusive evidence” is already high, let alone “proof by fully conclusive evidence.” Whilst “proof by conclusive evidence” might be equated to proof beyond reasonable doubt, “proof by fully conclusive evidence” requires a higher standard of proof that does not allow any doubt or uncertainty as to state responsibility.
The standard of proof by fully conclusive evidence, adopted by the court in the 1949 Corfu Channel judgment, is neither established in the Genocide Convention nor in the ICJ Statute. Instead, it is entirely case-law driven. Moreover, given that proving a case by the applicant state against the respondent state constitutes an onerous task, applying “proof by fully conclusive evidence” might serve as a disincentive for a state to bring a dispute against another state before the ICJ. The number of cases brought before the ICJ is indeed on the low side and perhaps the time has arrived to revisit the standard of proof in order to remove barriers for states to make better use of the contentious jurisdiction of the court. In short, the standard of proof related to genocide varies from one international court to another. As explained, the ICJ adopts the “evidence by fully conclusive standard of proof.” But this is not the standard adopted before international judicial bodies such as the ICTY, the ICTR, and the ICC, which consistently adopt the proof-beyond-reasonable-doubt standard. The latter has been defined by Lord Denning as follows:
Proof beyond a reasonable doubt does not mean proof beyond a shadow of a doubt. The law would fail to protect the community if it admitted fanciful possibilities to deflect the course of justice. If the evidence is so strong against a man as to leave only a remote possibility in his favour, which can be dismissed with the sentence “of course it is possible but not in the least probable” the case is proved beyond reasonable doubt, but nothing short of that will suffice. (King’s Bench Division: 372)
The inconsistency regarding standard of proof runs the risk of limiting the attribution of state responsibility for genocide.
Moreover, although the raising of the evidentiary bar setting to extremely high standards of proof for genocide is characteristic of the ICJ, Gregory Kent observes when discussing the decisions of the ICTR and the ICTY, that “in some cases, there are ‘prosecutorial omissions and errors as well as a tendency on the part of the judges to require that the prosecution meet higher evidentiary standards in these cases than in other types of cases”‘ that are “illustrative of certain narrowing developments in case law” (Kent: 573-574). Furthermore, Kent notes that “the idea, in Bosnia at least, that a plan of action directed at the destruction of non-Serb groups needed further evidence to be proven beyond all reasonable doubt seems farcical to the victims and equally far-fetched to academic experts” (Kent: 576). Once more, an excessive standard of proof contributes directly to limiting attribution of responsibility for genocide.
Exclusions Made by the Genocide Convention Regarding Sources of International Law and the Finding of State Responsibility for a Lesser Offense
First, Article 38 of the ICJ Statute is considered to be declaratory of the sources of public international law. It provides the following:
The Court, whose function is to decide in accordance with international law such disputes as are submitted to it, shall apply:
- International conventions, whether general or particular, establishing rules expressly recognized by the contesting states;
- International custom, as evidence of a general practice accepted as law;
- The general principles of law recognized by civilized nations;
- Subject to the provisions of Article 59, judicial decisions and the teachings of the most highly qualified publicists of the various nations, as subsidiary means for the determination of rules of law. (United Nations)
Although such sources are not arranged in a hierarchical order, the first three sources of international law, that is, treaty law, custom, and jus cogens are a primary source of international law whilst the other two sources (judicial decisions and the writings of publicists) are a subsidiary source of international law. In so far as the primary sources are concerned, they are of equal value and are not placed in a hierarchical order amongst themselves. Hence, it cannot be argued that treaty law is superior to both customary law and jus cogens and that customary law is, in turn, superior to jus cogens. In other words, all three sources are of equal weight, although treaty law normally tends to be more ascertainable than the other two sources of international law. Nonetheless, the international crime of genocide is perhaps one of those very few international crimes that can claim to have as its source conventional law, customary law, and a peremptory norm status. The subsidiary sources of international law are also relevant for understanding this international crime. Although there are five sources of international law, the text of the Genocide Convention refers only to the primary source of treaty law as contained in the convention. The other primary and secondary sources of international law are not referred to by the Genocide Convention. This emanates from a reading of Article IX of the convention that stipulates the following:
Disputes between the Contracting Parties relating to the interpretation, application or fulfilment of the present Convention, including those relating to the responsibility of a State for genocide or for any of the other acts enumerated in Article III, shall be submitted to the International Court of Justice at the request of any of the parties to the dispute. (United Nations)
The above article refers to disputes “relating to the interpretation, application or fulfilment” not of the five sources of international law but only “of the present Convention.” It is only in this limited case that the ICJ has jurisdiction to decide “disputes between the Contracting Parties.” It is therefore not automatic for the court to decide a dispute between contracting parties on the basis of nontreaty provisions. On the contrary, once a treaty has been identified as applying to that dispute, the court will decide that dispute in terms of that treaty even if, for instance, there might have been a rule of customary international law that would have displaced that treaty or a peremptory norm that would have abrogated a treaty provision. This is because (1) Article IX allows the ICJ to settle a dispute by having recourse only to treaty law (“the present Convention”) and (2) Article 38 of the ICJ Statute does not apply any hierarchical relationship between the primary sources of international law inter se. This, it is contended, constitutes a serious limitation imposed upon the court by the contracting states to the convention. The Genocide Convention should be flexible enough to allow the court to have recourse to all the sources of international law, even though these might not be conventional law, provided that such sources do not run counter to conventional law but rather supplement it. The ICJ has both in Bosnia v. Serbia (2007) and Croatia v. Serbia (2015) correctly taken the view that its jurisdiction was limited to conventional law, thereby precluding it from resorting to other primary sources of international law (ICJ: 129). In sum, the court’s position is that “the text is quite clear that the jurisdiction for which it provides is confined to disputes regarding the interpretation, application or fulfillment of the Convention, including disputes relating to the responsibility of a State for genocide or other acts prohibited by the Convention” (ICJ: 88).
Second, the Genocide Convention does not allow the conviction in a criminal trial of an accused or the finding of state responsibility in the case of a state in a dispute with another state before the ICJ, of an offense of a less aggravated character than that of genocide. Indeed, the ICJ has held that it “has no power to rule on alleged breaches of other obligations under international law, not amounting to genocide, particularly those protecting human rights in armed conflict. That is so even if the alleged breaches are of obligations under peremptory norms, or of obligations which protect essential humanitarian values, and which may be owed erga omnes” (ICJ: 147). As can be seen from the two cases decided by the ICJ related to armed conflict in the former Yugoslavia, namely, those of Bosnia v. Serbia (ICJ) and Croatia v. Serbia (ICJ), the court limited itself to deciding whether there was a case of state responsibility for genocide. But the court did not consider — and, in fact, was precluded from doing so in terms of Article IX of the convention — whether the criminal conduct in question could have amounted to an offense of a lesser aggravated character or a lesser offense. This means that, although the court might be of the view that genocide might not have been committed in terms of the convention, it could still decide that a lesser offense might have been perpetrated if the impediment of Article IX of the convention is removed. For instance, in Croatia v. Serbia (ICJ), the court found that there was the actus reus of genocide but the special intention of genocide was lacking. Thus, where the court found that rape had occurred but could not find proof for the genocidal intent, rather than declaring that Serbia had not committed genocide, it could have declared that Serbia was still guilty of a lesser aggravated offense if there existed no such conventional hindrance in Article IX of the convention (ICJ: 158). In this way, although justice would be meted out to the respondent state that would end up not being declared responsible for genocide perpetration, justice would still be done for the victims of crime, because the respondent state would have been declared responsible for the commission of a crime of a lesser nature than that of genocide rather than being absolved of state responsibility, as happened in the instant case.
Although genocide is inconceivable without state complicity or involvement, no state involved in the occurrence of genocide has been held responsible to date, except for Serbia, which the ICJ found to have been in breach of its obligation to prevent genocide in Bosnia (ICJ: 471). The centerpiece of the genocide law — the Convention on the Prevention and Punishment of the Crime of Genocide — does not even explicitly provide that states may be responsible for genocide. Posited from a criminal-justice perspective the convention is aimed at individual perpetrators rather than states, although perpetrators act as agents of the state. We believe that this narrowing of the referent of responsibility is damaging for the punishment and prevention of genocide. Governments ought to answer for their genocidal actions.
Sir Hartley Shawcross and Sir Gerald Fitzmaurice were right to suggest during the UN discussions on the drafting of the convention that responsibility for genocide should be vested both with states and individuals and that for the convention to be effective it ought to approach genocide from the angle of governments not merely individuals (United Nations). As explained above, the narrow defeat of the amendment proposed by the United Kingdom to explicitly include in the text of the convention that states may be held responsible for genocide does not equate with the defeat of the idea of state responsibility. From a justice point of view, it is anathema to punish few state officials who perpetrated, or were complicit in, genocide without holding responsible the government that ordered it. At a time when the principle of Responsibility to Protect has made inroads into the UN and foreign policies of many countries (Bellamy; Bellamy and Dunne), it is ironic that responsibility for the worst crime known to humankind is not attributable to the state(s) involved in its commission. In our opinion, a revised convention should clarify that states may be held responsible for genocide, in addition to individuals, in view of the fact that in the vast majority of cases genocide is carried out with a state’s connivance or its direct solicitation. This article therefore contributes to an emerging debate on the application of state responsibility with reference to the protection from the crime of genocide by highlighting existing limitations pertaining to the interpretation and implementation of the Genocide Convention.
This article has considered, too, the correlation between court practice and legal content in the ICJ’s setting. It has paid attention to the ICJ’s legal procedure and its impact on the quality of the court’s decisions pertaining to state responsibility for genocide. Several limitations are identified that make difficult the attribution of state responsibility under the Genocide Convention.
A careful reading of the convention reveals shortcomings that hinder the proper prosecution and effective punishment of this crime. These deficiencies have been discussed above and, in sum, consist principally of the following: the lack of the Genocide Convention to articulate in detail the ingredients of positive responsibility, that is, state responsibility to prevent the commission of genocide, although the convention’s rendering of negative responsibility (punishment of genocide perpetrators) is better expressed; ambiguities related to, and limitations in, the precise definition of genocide; the adoption of a high standard of proof to ground state responsibility for this crime that in turn makes it highly improbable to attribute state responsibility for the commission of genocide, thus undermining its punishment (that is, attribution of negative responsibility); the exclusions made by the Genocide Convention regarding sources of international law whereby the convention specifically precludes the court from resorting to other primary sources of international law such as customary international law and jus cogens even if these make wider provision in relation to genocide; and the impossibility for the court under the Genocide Convention to find state responsibility for a lesser offense than that of genocide. For so long as the above-identified lacunae in the UN Genocide Convention continue to persist, it is very likely that genocidaires and their respective states will continue to evade justice and to violate fundamental human rights. It is of the essence that state leaders are not allowed to hide behind state immunity from prosecution before the ICC or other international and national tribunals or for their state to protect them by either not approving their extradition to face trial before an international criminal court or tribunal or not prosecuting them domestically. Moreover, the Genocide Convention needs to be amended to allow the ICJ to find state responsibility for a less serious offense than that of genocide such as ethnic cleansing, war crimes, and crimes against humanity. A new provision could be introduced into the convention to interpret the standard of proof to attribute state responsibility as one of “proof beyond reasonable doubt.” Otherwise, the pitching of the standard of proof at a very high level is tantamount to limiting the attribution of state responsibility for genocide.
Although this article is concerned primarily with the issue of agency pertaining to states, its analysis may contribute to open an interdisciplinary debate with the view of a possible broadening of agency beyond states and individuals in attributing criminal responsibility for genocide to include also violent nonstate actors that may be involved in the perpetration of this crime. Indeed, there is a gap in literature to be filled by future research in so far as the convention is missing recent empirical developments, for instance, where violent nonstate actors commit genocide. It appears that the so-called Islamic State of Iraq and the Levant (ad-Dawlah al-Islāmīyah fil ‘Irāq wa ash-Shām, known also as the “Islamic State”) has moved in that direction committing vicious, persistent attacks against civilians in Iraq and Syria. Such a situation cannot remain unaddressed by the convention. Violent nonstate actors have to be brought within the fold of the convention’s punitive provisions apart from punishing their leaders and fighters on the basis of individual criminal liability.
If the provisions of the Genocide Convention continue to make it difficult to attribute responsibility for the most heinous crime to a state, or violent group involved in its commission, this bodes worse for a state to implement its positive obligation to prevent genocide and its negative obligation to punish this crime in order to safeguard the right of existence of human groups. Unless the Genocide Convention is amended, it will continue to be very difficult for the ICJ to condemn a state for its responsibility in the perpetration of genocide thereby reducing the effective deterrent for states that may elect to walk that atrocious criminal route. This article has thus raised some concerns regarding interpretive and procedural deficiencies pertaining to the protection from the perpetration of genocide that deserve to be addressed in order to ensure that the worst crime known to humankind is punished with the full rigor of the law.