At its inception, the crime of genocide, which broadly concerns criminal conduct targeted at a group, was generally seen as somehow more culpable or aggravated than international crimes targeted at an individual. Critical opposition to that view exists (See Milanović on the Karadžić and Mladić Trial Chamber judgments). Contemporary application, however, of the law continues to consider genocide as “horrific in its scope” precisely because perpetrators identify “entire human groups for extinction” and “seek to deprive humanity of the manifold richness its nationalities, races, ethnicities and religions provide” (Krstić, Appeals Chamber judgment, para. 36).
The Appeals Chamber in Krstić has emphasized that the gravity of genocide is “reflected in the stringent requirements which must be satisfied before this conviction is imposed” (para. 37). This includes proving a specific intent to destroy a group such that the group targeted for destruction was either the whole “protected group”, or a “substantial” part of that whole (the “substantiality test”). Where the requirements are satisfied, the Appeals Chamber implores that “the law must not shy away from referring to the crime committed by its proper name” (para. 37).
My contention is that the law in fact has shied away from referring to the crime of genocide by its proper name. Part 1 considers the development of the law since the ICTY Krstić judgment apparently widened the scope of the crime by finding that perpetrators of prohibited acts in Srebrenica in 1995 had a specific intent to destroy a ‘’substantial part” of the Bosnian-Muslim protected group in Bosnia-Herzegovina (BiH). Part 2 shows that despite the Srebrenica precedent, chambers have been keen to preserve a set of “stringent requirements” rooted in a historical misunderstanding of the law by closely associating it with atrocities such as the Holocaust (see further, Jarvis and Tieger). I conclude by tentatively proposing a bifurcation in the law so that the substantiality test clearly delineates between large-scale mass atrocity crimes targeting substantially whole groups and those crimes, which tend to target smaller (sub-)subsets of protected groups.
Specific intent: when exactly does an intent to destroy ‘a part’ become ‘a substantial part’ of the whole?
The Karadžić and Mladić Trial Chamber judgments of 2016 and 2017, respectively, demonstrate the inconsistencies in the judicial interpretation of the ‘specific intent’ for genocide and, in particular, the substantiality test. There is a wide disparity between the substantiality test as perceived in the preparatory works to the 1948 Genocide Convention (GC) and the test as subsequently developed by the ICTY/ICTR jurisprudence: an intent to destroy ‘considerable’ numbers of a protected group (for instance, as applied to the Jewish population of Europe targeted by Nazi Germans) versus an intent to destroy a numerically small, although symbolically significant, (sub-)subset of a protected group (for instance, as applied to the Bosnian Muslims targeted in Srebrenica by Bosnian-Serb forces). (See also Krstić Appeals Chamber judgment, paras 6-14).
The Karadžić and Mladić Trial Chamber judgments held that in relation to the charge of genocide in Srebrenica, the surrounding facts and circumstances were such that the only reasonable inference was that Mladić/Karadžić and the physical perpetrators of prohibited acts had an intent to destroy the Bosnians Muslims of Srebrenica. Critically, the Bosnian Muslims of Srebrenica were seen as a substantial part of the protected group (i.e. Bosnian Muslims in BiH) because Srebrenica: had a strategic political significance; was designated a UN safe area; carried some emblematic significance; and was controlled by the physical perpetrators (para. 3553-3554). The Trial Chamber reiterated that neither the absolute nor the relative numbers were determinative and that chambers may consider any number of non-exhaustive criteria both objective and subjective (following guidelines in the Krstić AC judgment, paras 12-14).
The judgments relating to Srebrenica are to be contrasted with how the charges of genocide in the six municipalities were dealt with, particularly, in the Mladić judgment. Notably, the majority found that many of the physical perpetrators of the prohibited acts (killings and serious bodily or mental harm) in five of the six municipalities had an intention to destroy a part of the Bosnian-Muslim group as such (paras 3456-3526). The chamber, however, found that this intent was not to destroy a substantial part of the whole on the basis that there were other reasonable inferences, namely, that: (a) Bosnian Muslims were only targeted in their own municipalities; (b) physical perpetrators had limited geographical control; and (c) numbers of Bosnian Muslims under Bosnian-Serb control were relatively small (paras 3534-3535). None of those factors, however, seem determinative: (a) physical perpetrators can only commit prohibited acts in their areas of de facto activity and control with the means available to them; (b) each of the municipalities had well-documented strategic/symbolic significance especially, for instance, in the case of Prijedor; and (c) the systematic commission of prohibited acts by designated brigades attached to designated municipalities demonstrated a manifest pattern at the symbolic beginnings of the conflict.
The latter inferences are at least as persuasive as the positive substantiality inferences cited in respect of Srebrenica, which raises three further questions. First, was the satisfaction of the substantiality test in respect of Srebrenica the only reasonable inference? Second, does a part-positive finding of specific intent of the physical perpetrators allow us to infer something more about the overall intent of the senior perpetrators either physical or high-level (as Joint Criminal Enterprise members or otherwise)? Third, might the assessment of the substantiality test be different if viewed from the broader perspective of higher-level perpetrators when considering the five municipalities together than in isolation? That the same guideline-factors for substantiality are applied to justify opposite outcomes in one case perhaps affirms the ultimately nebulous nature of the test. (See Milanović’s contrasting analysis and Ambos’s astute critique of the Karadžić judgment).
Preserving the “stringent requirements”
Is it possible to reconcile an expansive interpretation of the substantiality test (for instance, as it was applied to the Srebrenica or Brčko cases – see Jelisić Appeals Chamber judgment, paras 66-72) with a more restrictive interpretation (for instance, as in the municipalities component of the Mladić Trial Chamber judgment or the Jelisić Trial Chamber judgment)?
Without the Srebrenica precedents, a significant lacuna in the law would exist in terms of group protection as the law would only apply to very large proportions of a protected group. Perpetrators could factor into their behaviour sophisticated, decentralised means of destroying groups just short of meeting the technical threshold of the required ‘part’. The issue, therefore, is less whether Srebrenica set an incorrect precedent but why the substantiality test appears so difficult to apply consistently.
Restrictive interpretations seems appealing because of a belief that ‘stringent requirements’ must not be diluted. That belief is mistaken precisely because ‘dilution’, or ‘elaboration’ as I would prefer to call it, was exactly what the Appeals Chamber in Krstić, in the absence of useful precedents, allowed in deciding that the Bosnian Muslims of Srebrenica constituted a substantial part, despite the fact that they represented less than 3% of the whole.
Conceptual clarity: is classifying genocide by degree a way forward?
Given the wide disparity in the law, there is arguably a case for a bifurcation of the crime of genocide: ‘first’ and ‘second’ degree genocide. Introducing degrees could allow for a clearer enunciation of the nature, scale, and scope of the mens rea element of the crime (although the actus reus will have a bearing when drawing inferences). Degrees could also bridge areas of contention where inference on the part of a court is required, namely: ascertaining the nature of the pre-defined protected group and its relationship to the alleged targeted group; determining the relevance of any unequivocal numerical proportions of ‘a part’; assessing which non-numerical factors objectively must be relevant (for instance, perpetrators’ activity and control, scale of prohibited and other culpable acts, or broader plans by superiors); and how far the subjective element can aid until it becomes meaningless.
Degrees may also help rescue the idea (and it is simply that) that the “crime of crimes” ought to be reserved for large-scale and systematic commission of prohibited acts against entire human groups. A ‘lesser’ degree could allow space for capturing alternative manifestations/modes of carrying out destructions of groups whilst still acknowledging ‘genocide’ given the devastating social, cultural, and historical consequences physical destruction can have for even (sub-)subsets of protected groups (Srebrenica being a case in point). Litigation contesting the border between degrees would be unlikely if the delineation between first and second degree is sufficiently wide.
Whilst any immediate statutory change would be fanciful, and the contours of any delineation of the scope of the crime hotly contested, at minimum, academic debate and judicial appreciation of the two degrees (perhaps tacit and/or through consideration at sentencing) may allow for much needed clarity, fair labelling, and allow for better application of now quite disparate conceptions of ‘genocide’ in the case law. The political contestation of the term may also usefully be narrowed; with a second degree allowing a more acceptable compromise. In any case, further discussion would certainly bring us closer to the hope that the law does not shy away from referring to the crime committed by its proper name.
Note: The judgments in the Karadžić and Mladić appeals before the MICT are expected to be issued at the end of 2018 and end of 2020 respectively.