Home EJIL Analysis What Lies Beneath the ‘G’ Word? Genocide-Labelling and Fact-Finding at the UN

What Lies Beneath the ‘G’ Word? Genocide-Labelling and Fact-Finding at the UN

Published on May 28, 2015        Author: 

In late 2013, the Special Adviser on the Prevention of Genocide warned that “there is a risk of genocide” in the Central African Republic (CAR). A year later, with thousands dead and hundreds of thousands displaced, a UN-mandated Commission of Inquiry (CoI) determined that genocide had not occurred because “the threshold requirement to prove the existence of the necessary element of genocidal intent ha[d] not been established…” (Executive Summary). Their answer seems clear, and yet this post will argue the Commission may have reached the wrong conclusion. In doing so, it will also draw attention to discrepancies between the UN’s classifications of genocide and raise questions about the powers of fact-finding bodies more generally.

It should be noted at the outset that the CoI left little doubt that serious crimes had been committed in CAR. Established at the request of the Security Council, the Commission had a mandate to investigate violations dating back to January 2013 when Séléka fighters began their march on CAR’s capital, Bangui. Though some of the worst violence took place on its watch, the Commission could not “establish with any degree of accuracy the number of people who were killed in the conflict.” Conceding that the available estimates “fail to capture the full magnitude of the killings that occurred”, it nevertheless concluded that “all the parties were involved in serious violations of international humanitarian law and gross abuses of human rights including rape and other gender based sexual offences and violations.”

What about genocide?

The CoI’s analysis of this key question begins with the applicable law, where it notes that genocide requires the actus reus (‘specific acts committed against specific groups’), the mens rea of specific (genocidal) intent, and – in line with the Rome Statute’s Elements of Crimes – ‘a manifest pattern of similar conduct directed against the targeted group’ (para. 450). Against this backdrop, the report establishes that the genocide label would prima facie apply only to acts committed by the Christian anti-balaka against CAR’s Muslims. Crucially, genocide would not be applicable to attacks committed by Muslims against Christians. The Commission then assesses the case law of several tribunals in order to distinguish ethnic cleansing from genocide.

This is where the legal analysis takes a perplexing turn. Before it has a chance to examine the legal elements of genocide, the CoI says (para. 452):

…the information available to it reveals repeated instances of crimes against humanity amounting to the fact pattern of ethnic cleansing committed by the anti-balaka in the areas in which Muslims had been living. In terms of criminal responsibility, however, the Commission is of the view that these acts of ethnic cleansing would best be prosecuted with (sic) under the rubric of crimes against humanity, which is the crime category that is explicitly recognized in the Rome Statute and in the relevant legislation of the CAR… [T]he facts of the situation indicated that… crimes against humanity… capture the full essence of the policy of ethnic cleansing that was pursued.

There are two problems with this conclusion. First, the Rome Statute and the relevant domestic laws ‘explicitly recognize’ genocide as a crime, so it is hard to see the relevance of the assertion that crimes against humanity ‘capture the full essence’ of what happened (though it is not a party to the 1948 Genocide Convention, CAR has incorporated the Rome Statute into its national legislation). Secondly, the Commission seems to be conflating its fact-finding role with prosecutorial discretion when it suggests ‘these acts of ethnic cleansing would best be prosecuted’ as crimes against humanity. Should the CoI be giving such prosecutorial advice? Or is its role limited to fact-finding and legal analysis?

The problems do not end there. After ruling out genocide, the Commission nevertheless examines the constitutive elements of the crime. Though it repeatedly conflates the actus reus and mens rea requirements, the report’s main finding seems to be that “…the Commission has not received sufficient evidence to establish reasonable grounds to believe that the perpetrators acted with specific intent to destroy the targeted group…” (para. 457). At first blush, this seems a plausible conclusion, notwithstanding the fact that the report provides some circumstantial evidence to the contrary.

The Commission acknowledges, however, that the inquiry cannot be limited to just the mens rea. Direct evidence of genocidal “intent to destroy, in whole or in part, a national, ethnical, racial or religious group” (Art. II, Genocide Convention) is extremely rare. Absent a ‘smoking gun’, the key question – as the CoI concedes – is whether genocidal acts can provide plausible evidence of genocidal intent. In other words, whether the mens rea can be inferred from the actus reus.

Drawing on the case law of international tribunals, the Commission sets out to assess “the extent to which attacks against a large proportion of a targeted group… can be indicative of specific intent to destroy that group” (para. 459, citing Muhimana). Yet, despite its own estimates that the Muslim population of Bangui fell from 100,000 to 1,000 inhabitants, the CoI disregards this circumstantial evidence of a plan to eliminate a protected group. Without any discussion of how such large scale violence can potentially constitute the actus reus of genocide and what this means for the question of inferring genocidal intent, the Commission simply re-affirms its earlier finding that “[i]n the absence of evidence demonstrating… specific intent to destroy the group as such, the Commission is unable to conclude that the crime of genocide took place” (para. 459, citing Stakic).


The Commission’s conclusions are troubling for several reasons. Although it claims to adhere to ‘a reasonable grounds to believe’ standard of proof (para. 456-457), the CoI’s insistence that it lacks evidence of genocidal intent seems to belie its own methodological choices. For the purpose of ascertaining genocide at this stage of fact-finding, the more interesting question seems to be whether there are reasonable grounds to believe that the available facts can be indicative of specific intent to destroy that group. Although its figures show that 99% of Bangui’s Muslims were forcibly removed (many of them possibly killed), the Commission invokes the ‘reasonable grounds’ standard to conclude that it is unreasonable to believe such widespread targeting could be indicative of specific intent to destroy that group. In doing so, the report relies on the ICC’s Al-Bashir decision to argue that “the finder of fact is open to draw conclusions relating to one particular reasonable conclusion, which may have a greater evidentiary basis” (para. 457). This is odd. In Al-Bashir the Appeals Chamber re-interpreted the ‘reasonable grounds’ standard to expand the types of inferences that can be made when evaluating inconclusive evidence of genocide. The CoI cites the case, but then does the exact opposite. Despite abundant evidence of widespread targeting of Muslims, it declines to make any inferences from those facts, choosing instead to fall back on the assertion that it could not find clear evidence of genocidal intent.

The CoI’s analysis is equally problematic when compared to how other fact-finding bodies have dealt with genocide. Two examples stand out. In a recent report, the Office of the High Commissioner for Human Rights (OHCHR) concluded that “[m]embers of ISIL may have perpetrated genocide against the Yezidi community… in the context of a manifest pattern of conduct aimed at the destruction of the group” (para. 78). Not surprisingly, the media picked up on the report’s use of the ‘G’ label. The problem is that the report itself provides no evidence of ISIL’s genocidal intent. The genocide label is actually grounded in a summary analysis of several attacks, which presumably constitute the actus reus of genocide (para. 17-19). In other words, unlike the CAR Commission, the OHCHR inferred the mens rea of genocidal intent from the actus reus of genocide.

The OHCHR’s ‘Mapping report’ on the Democratic Republic of Congo took a similar approach. Leaked to the press to forestall Rwandan efforts to have the ‘G’ label removed, the report argues that “the apparent systematic and widespread attacks described in this report reveal a number of inculpatory elements that, if proven before a competent court, could be characterized as crimes of genocide” (para. 31). Though it acknowledges that “a number of countervailing factors… could lead a court to find that the requisite intent was lacking, and hence… genocide was not committed” (para. 32), it is telling that the OHCHR used the ‘G’ word. On this occasion the UN chose to leave open the possibility of genocide, rather than ruling it out at the fact-finding stage.


What explains these discrepancies? Why are ‘genocidal acts’ enough in some situations (DRC, ISIL)? Why is ‘genocidal intent’ a requirement in others (CAR)? To be sure, genocide remains a contested concept, so some questions are bound to generate controversy. However, one cannot discount the possibility of deeper forces being at play. Although this post is limited to genocide (as opposed to other international crimes), a comparative perspective yields insights into the methodological choices of fact-finding bodies. It also raises questions about their fact-finding role. Are they only establishing facts and providing plausible legal interpretations? Or are they also laying the ground for prosecution, with pragmatic choices about, say, resource allocation factored into the analysis? Does devising prosecutorial strategy at the fact-finding stage compromise the objectivity of a commission’s findings? A separate question, given the controversial nature of the ‘G’ label, is whether fact-finding bodies should have a role to play in alerting the world to the threat of genocide (ISIL)? Or is this better left to purely political bodies (CAR)? Whatever the answers to these questions, there are still few agreed standards on how fact-finding bodies should operate. As the list of commissions of inquiry and fact-finding missions grows, the UN may want to harmonise its approach to these key issues.

Print Friendly

4 Responses

  1. Dear Sir

    While I agree with the core argument of your post in so far as the application of reasonable grounds to believe standard is concerned, I do believe that the finding of the CoI (before dealing with the evidence in relation to Genocide) must not be interpreted in the same way as proposed in the blog.

    The report of the CoI goes on to state that the crime of ethnic cleansing was evident from the acts of Christian anti-balaka. It is for this reason that the report states that such acts that constitute the crime of ethnic cleansing may be prosecuted under the head of Crimes against Humanity (CAH) as the Rome Statute and the domestic laws of CAR provide specifically for prosecution of CAH while not providing any explicit provision against Ethnic Cleansing.
    The aformention interpreation is strengthened by the use of the word “however” by the CoI when stating, “In terms of criminal responsibility, however, the Commission is of the view…”
    Moreover, this interpretation is in fact more in consonance with the subsequent inquiry of the CoI into the crime of Genocide.

    However as stated above the CoI’s invocation of the ‘reasonable grounds’ standard to conclude that it is unreasonable to believe that widespread targeting could be indicative of specific intent to destroy that group is indeed highly problematic in light of the Al-Bashir precedent which cleary held otherwise. And to that extent I would agree that the CoI may have erred.

  2. Patryk I. Labuda Patryk I. Labuda

    Dear Mr. Sharma,

    Many thanks for your comment. I have re-read the relevant paragraphs (452-453) but I am not entirely sure I understand your point. It seems clear to me that the word “however” is supposed to distinguish CAH from genocide. Ie, in terms of criminal responsibility, the CoI thinks CAH are more appropriate than genocide. This conclusion follows from paragraph 452, which compares genocide and ethnic cleansing, as well as para. 453 (1st and 2nd sentence read together).

    Where I think you may have a point is regarding “explicitly recognized…”. I think it is quite possible that here the CoI wanted to distinguish ethnic cleansing, which is not an international crime, from CAH, which are. That would make sense, even though that sentence is not exactly a model of clarity.

    But – again – that in no way changes what these two paragraphs and this part of the report are about. The CoI (para. 443-461) establishes that, in terms of criminal responsibility, the crimes it has catalogued “would best be” be prosecuted as CAH rather than genocide. If I understand the rest of your comment correctly, you seem to agree that this conclusion is unpersuasive in light of the ‘reasonable grounds’ test.

  3. Dhruv Sharma

    Dear Sir

    If I may clarify my point.
    Through an analysis of the evidence on board it was ‘revealed’ to the CoI that there was indeed a pattern of commission Ethnic Cleansing in the CAR. However, considering that ethnic cleansing is not a crime under the Rome Statute or Domestic law of CAR, the CoI felt it would be prudent to prosecute such crimes under the head of Crimes against Humanity. The word, ‘however’ is not used to rule out Genocide from the list of crimes that may have been committed but to state that while ethnic cleansing is also a crime committed, however, the acts may still be better prosecuted under the head of Crimes Against Humanity (as there was sufficient evidence of persecution and forcible transfer of population) for the lack of a specific legislation on ethnic cleansing.

    This is further clarified upon reading the first line of Para 454, which states, “it still remains to be considered whether these crimes could also constitute the necessary actus reus required for genocide…” This line conclusively shows that the CoI did not rule out the possibility of commission of Genocide but thought it prudent to first categorically state that in the very least the crime of Ethnic Cleansing/Crimes Against Humanity had been committed in the CAR. Upon the establishment of Ethnic Cleansing the CoI then went ahead to further analyse the fulfillment of the actus reus of Genocide.

    Thus I believe the CoI did not rule out Genocide from the list of crimes committed in the CAR through para 452-53 but only diverted in para 453 to first conclude that criminal responsibility could be attached at least for ethnic cleansing and crimes against humanity before continuing its analysis on Genocide.

  4. Thanks so much. Here is where the crux of our disagreement lies: “the CoI… thought it prudent to *first categorically* state that *in the very least* the crime of Ethnic Cleansing/Crimes Against Humanity had been committed in the CAR.”

    In my opinion, the CoI is putting the cart before the horse. We already know from the report’s other sections that CAH were committed. Here the question is whether the violence in CAR reached the threshold of genocide. The CoI is not — as you suggest — answering that “at the very least” CAH were committed. Rather, they are saying these violations would “best be prosecuted” as CAH. That is quite different from saying “at the very least”.

    How best to prosecute crimes is a question of prosecutorial policy. We all understand that the ICC Prosecutor or a domestic prosecutor would have to decide whether or not to actually bring charges of genocide, which are, arguably, harder to prove than CAH. The question is why a CoI is proffering such advice at the fact-finding stage? I am not sure this does them any favors. It just seems they have made a pragmatic decision about what is easier to prove, and – again – I think that’s the prosecutor’s job, not theirs.

    What reinforces my skepticism is that, contrary to your assertion, they reach this conclusion about ‘how best to prosecute’ ethnic cleansing before they even analyse the constitutive elements of genocide. One could argue that the sequence is irrelevant, ie it is just the way the report presents these issues. The problem is – and here we agree – that the legal analysis of genocide that follows is – to put it charitably – unconvincing. My reading of their analysis (454-459) is that the CoI is not really interested in exploring the possibility of genocide. Why? Because they have already determined that the violence would “best be prosecuted” as CAH.

    In other words, they’ve put the cart before the horse. The outcome was pre-determined, so there is no real need to discuss whether and to what extent genocidal acts are linked to the actual annihilation of a protected group, how and when forcible transfer can threaten the physical existence of a protected group, or how evidence of forcible transfer can contribute to the establishment of genocidal intent (for an excellent post that explores these complex issues, see All this is dismissed because the CoI did not find a smoking gun (while claiming to apply a ‘reasonable grounds to believe’ standard of proof!).

    But I think we agree on this last point.

    To summarize, I would agree with you if the CoI had said that “at the very least” CAH were committed and then not rule out the possibility of genocide. But that’s not what they do. First they say it’s best to prosecute these violations as CAH, and then, after the fact, they also rule out genocide. I find their approach unconvincing.