Hate Speech as Persecution: Tackling the Gordian Knot

Two recent incidents have reignited the question of whether hate speech can constitute the actus reus for persecution, giving rise to a successful charge of crimes against humanity under the Rome Statute of the International Criminal Court (ICC). First, the report of the UN International Fact Finding Mission on Myanmar (FFM) to the United Nations Human Rights Council highlighted the use of Facebook by Myanmar authorities to create “a climate in which individuals and groups may become more receptive to incitement and calls for violence” against the Rohingya population. Second, the ICC recently authorized an investigation into atrocities committed by the Myanmar government on the Rohingya population. This post examines the standard for hate speech to constitute persecution as a crime against humanity as set out by the International Criminal Tribunal for Rwanda (ICTR) in Prosecutor v. Nahimana, and assesses whether hate speech simpliciter can amount to persecution as a crime against humanity.

A Hazy Standard

Under Article 3 of the Statute of the ICTR, ‘persecution’ is a crime against humanity. In Prosecutor v. Nahimana, the ICTR Trial Chamber was confronted with the question of whether hate speech – both simpliciter and accompanied by calls for violence – could constitute an actus reus for the crime of persecution. The Trial Chamber resoundingly answered in the affirmative, holding that there did not have to be calls for violence for hate speech to constitute persecution (para. 1073). Note that the ICTR Trial Chamber’s decision in Nahimana was directly in conflict with the International Criminal Tribunal for the former Yugoslavia’s (ICTY) Trial Chamber’s decision in Prosecutor v. Kordic, which held that hate speech simpliciter does not rise to the same level of gravity as the other acts enumerated in Article 5. The Nahimana Trial Chamber’s judgment was appealed to the ICTR Appeals Chamber, which issued its final ruling on hate speech as persecution in 2007, holding that hate speech accompanied by calls for violence could amount to persecution as a crime against humanity. While the Appeals Chamber expressed doubts as to whether ‘mere’ hate speech unaccompanied by such calls for violence could amount to persecution, it ultimately decided it was not necessary to answer the question (paras. 986-988). This has left unresolved the question as to whether, if at all, ‘mere’ hate speech (as opposed to speech accompanied by calls to violence) can amount to persecution as a crime against humanity.  

Persecution as a Crime Against Humanity: Toward Clarity

The ICTY Trial Chamber in Prosecutor v. Kupreskic laid down an authoritative four prong test for persecution which requires: (1) a gross or blatant denial; (2) on discriminatory grounds; (3) of a fundamental right, laid down in international customary or treaty law; and (4) reaching the same level of gravity as other crimes against humanity (para. 621).

Thereafter, whereas the ICTY Trial Chamber in Kordic ruled that hate speech not calling for violence did not meet prongs three and four, the ICTR Trial Chamber in Nahimana boldly found that even hate speech not inciting violence could satisfy the test. As stated above, the Appeals Chamber in Nahimana, while applying Kupreskic’s gravity test, relied heavily on the fact that the hate speech concerned was accompanied by “calls for violence” to conclude that it met the requisite threshold, and cast doubt on ‘mere’ hate speech ever meeting that same threshold. Further, the Appeals Chamber (para. 986) reinforced this separation between hate speech simpliciter and hate speech explicitly calling for violence through truisms such as “speech cannot, in itself, directly kill members of a group, imprison or physically injure them.” Of course it cannot – but then, it does not have to.

The Appeal Chamber’s difficulty with speech not being able to directly “kill”, “imprison” or “physically injure” a person is misplaced. It has been clear for some time now in international criminal law (ICL) jurisprudence that acts do not have to amount to violence to qualify as persecution. Judgments such as those by the United States Military Tribunal in the Ministries case, the Kvocka decision at the ICTY, and the Bikindi decision at the ICTR, have made it clear that non-violent acts such as denying bank accounts, educational or employment opportunities, or even denial of proper medical care, for instance, can constitute underlying acts of persecution. It is this principle that the Nahimana Trial Chamber recognized as well, holding that there need not be a link between persecution and violence. In the words of the Kvocka Trial Chamber, “acts not inherently criminal may nonetheless become criminal and persecutorial if committed with discriminatory intent” (para. 186). The Kvocka Appeals Chamber fleshed this out further (para. 325), stating that:

“in the context in which they were committed and taking into account their cumulative effect, the acts of harassment, humiliation and psychological abuse…are acts which by their gravity constitute material elements of the crime of persecution.”

Judge Shahabudeen summed this up rather neatly in his Partial Dissent in the Nahimana Appeals Judgment, finding that:

“all that can be legitimately extracted from the post-World War II jurisprudence…is that the underlying acts must be sufficiently grave to affect the ‘life and liberty’ of the victims – though not necessarily a physical act against them.”

Therefore, it is evident that non-violent acts are not excluded from qualifying as persecution. So, even hate speech simpliciter would technically fit the bill. In his hate speech ‘spectrum’, atrocity speech crimes scholar Gregory S. Gordon categorizes hate speech as “general hate speech” that targets a community (e.g. “the Tutsis are rapists”); hate speech that directly targets the community (e.g. “You are vermin”); and finally, the least controversial – speech that calls for violence against the concerned community (see further: Gregory S. Gordon, Atrocity Speech Law: Foundation, Fragmentation, Fruition, 2017). Now, although speech crime analysis is largely muddled in ICL jurisprudence (see, e.g., Richard A. Wilson, Incitement on Trial: Prosecuting International Speech Crimes, p. 105, 2017), there is no dearth of judicial opinion in support of the proposition that even the former two types of hate speech (i.e. hate speech simpliciter, that does not contain specific calls to violence) can amount to persecution.

Precedent for this proposition exists right from the Nuremberg decisions where convictions were secured, for instance, on the basis of hate speech being a “poison … injected into the minds of thousands of Germans” (Streicher). In other words, the defendant’s speech was egregious enough to constitute a crime against humanity not because it had direct links with violence, but because “week after week, month after month… [it] infected the German mind with the virus of anti-Semitism.” Similarly, consider Dietrich, where speech was part of “a well thought-out, oft-repeated, persistent campaign to arouse the hatred of the German people against Jews.” No calls for violence were deemed necessary in Dietrich; it was enough that the purpose of the speech was to “enrage the Germans against the Jews.”

The ICTY in Kupreskic affirmed this understanding of Nuremberg jurisprudence, finding that the International Military Tribunal (IMT) considered both physical attacks on a person as well as other discriminatory acts to fall within the ambit of persecution (para 597). The Hartford Guidelines on Speech Crimes in International Criminal Law, formulated by a team of international criminal law experts, also support this conclusion, stating that hate speech can qualify as the crime against humanity of persecution, even in the absence of violence. Importantly, however, to qualify as persecution, this breach of fundamental rights must additionally reach a level of gravity similar to other crimes against humanity.

The Gravity Threshold

Under the Rome Statute, all charges brought before the ICC have to satisfy the threshold of “sufficient gravity” in order to be admissible (Article 17, Rome Statute). Specifically in the context of persecution, however, gravity has been deemed necessary by the ICTY Appeals Chamber (see, e.g., Krnojelac Appeal, para. 199). But what continues to confound tribunals is what kind of speech can meet this “gravity” threshold. As inconsistent as ICL jurisprudence is with respect to persecutory speech, it is particularly difficult to pinpoint any clear and replicable benchmark from relevant decisions for what the requisite gravity threshold looks like in the context of hate speech (accompanied by violence or otherwise). Judgments have either simply not dealt with this at all (see, e.g., Kordic, which simply states “most importantly, [the speech] does not rise to the same level of gravity as the other acts enumerated…” without explaining further) or have neglected to explain in any detail the precise circumstances in which hate speech may meet the gravity threshold (see, e.g., Nahimana Appeal Judgment).

Thus, it becomes necessary to revisit the original test for persecution laid down in Kupreskic. In explaining its conception of gravity in relation to persecution, Kupreskic held that the “overall consequences [of the persecutory act] must offend humanity in such a way that they may be termed ‘inhumane’” (para. 622). The ICTR Trial Chamber in Prosecutor v. Ruggiu, noting Kupreskic’s test, convicted the accused on the basis of public radio broadcasts, which singled out and attacked Tutsis and Belgians (para. 22). This nascent conception of the gravity threshold for persecutory hate speech was further clarified in the Nahimana Trial Judgment, where the ICTR Trial Chamber held that the act of denigrating a population on the basis of its ethnicity “in and of itself…can be an irreversible harm.” (para 1072). Although the Nahimana Appeals Chamber did not seem to agree, Judge Shahabudeen in his partial dissent appeared to echo the above decisions in stating that even discriminatory acts that harass or psychologically abuse victims could be grave enough to constitute persecution if the cumulative effect impaired “the quality of life”. Unfortunately, the Appeals Chamber’s judgment, which could have crystallized the law on this essential issue, proved a missed opportunity.

Further, contrary to some scholarly opinion (see, e.g., Audrey Fino on hate speech in ICL) that Seselj and other ICL jurisprudence posit violence as a sine qua non for persecution, it is arguable that the Seselj Appeal itself did not rule out that hate speech simpliciter can rise to the requisite gravity threshold for persecution. Note, for instance, that the Appeals Chamber (para. 163) stated that “Seselj’s speech denigrated the Croatians of Hrtkovci on the basis of their ethnicity, in violation of their right to respect for dignity as human beings.” Although it is lamentable that the Appeals Chamber in Seselj did inter alia rely on the nexus between the concerned speeches and ensuing violence, its finding was that Seselj had, in fact, “infected” the village with violence and hatred. It is unclear how the Appeals Chamber may have proceeded had the ‘infection’ been limited to hatred absent violence; however, one cannot conclude with certainty that the Appeals Chamber’s determination of hatred was irrelevant to its findings on persecution.

Hence, once hate speech simpliciter reaches the requisite gravity level (and satisfies the contextual elements for crimes against humanity), creating an additional requirement for it to call for violence amounts to adding an incorrect and unnecessary ingredient to speech that can constitute persecution as crime against humanity. It also has the result of leaving out of its ambit otherwise denigrating or dehumanizing discriminatory speech that is equally effective in creating a conducive climate for mass violence against the targeted group.

The ICC’s go-ahead to investigate atrocities in Myanmar provides an opportunity once more to clarify the extant law on this issue. Another dimension has been added to this problem, with much of the inflammatory speech which triggered mass atrocities against the Rohingya Muslims being disseminated on Facebook. The military not only used official government accounts with millions of followers on Facebook to consistently spread anti-Rohingya rhetoric, but even seemingly innocuous pages for entertainment, beauty and other informational pages were taken over to amplify the hate messaging. The use of Facebook, whose ubiquitous presence in Myanmar sometimes has its people equating it to the internet itself, as a platform for this messaging is no coincidence. Among other things, in its assessment of crimes against humanity, the IMT examined the reach of anti-Semitic speech, noting that Der Stürmer reached a “circulation of 600,000 in 1935.” In this regard, social media has certainly brought new dimensions to the highly contentious issue of atrocity speech crimes in ICL jurisprudence and assuming a successful indictment in the Myanmar case, it will be interesting to see the way the ICC grapples with this hot potato of an issue.