Defining Incitement to Genocide: A Response to Susan Benesch

[Gregory Gordon is Professor of Law, University of North Dakota School of Law.]



I would like to begin by thanking Opinio Juris for inviting us to have this important discussion here about the crime of direct and public incitement to commit genocide. I would also like to congratulate Susan Benesch on her excellent article regarding this verbal harbinger and prerequisite of mass atrocity. Professor Benesch provides a much needed exploration of the more complex facets of incitement that will afford jurists, advocates, and would-be offenders greater clarity in assessing the process by which permissible speech corrodes into forbidden exhortation.



But I cannot share in Professor Benesch’s conclusion that the crime of incitement remains “alarmingly” ill-defined. As I point out in my articles A War of Media, Words, Newspapers and Radio Stations: The ICTR Media Trial Verdict and a New Chapter in the International Law of Hate Speech, 45 VA. J. INT’L L. 139, 150 (2004) and From Incitement to Indictment? Prosecuting Iran’s President for Advocating Israel’s Destruction and Piecing Together Incitement Law’s Emerging Analytical Framework, 98 J. CRIM L. & CRIMINOLOGY (forthcoming June 2008), jurisprudence from the Rwandan incitement prosecutions has gleaned four criteria through which speech content regarding race or ethnicity can be analyzed as either legitimate expression or criminal advocacy: (1) purpose; (2) text; (3) context; and (4) the relationship between speaker and subject.



With respect to the “purpose” criterion, this jurisprudence has provided some examples of legitimate objectives: historical research, dissemination of news and information, and public accountability of government authorities. At the opposite end of the spectrum, explicit calls for violence would evince a clearly illegitimate purpose. The International Criminal Tribunal for Rwanda has given examples of discourse falling between these two extremes, including permissible speech focusing on ethnic animosity but geared toward raising ethnic consciousness, not provoking ethnic violence.



The “text” criterion, which entails a rigorous parsing of the words themselves, helps further reveal the purpose of the speech and provides an important piece of the contextual puzzle. The “context” criterion, arguably the linchpin of the entire analysis, mandates an examination of the circumstances external to and surrounding the text so that its true significance can be divined. This includes situating the words and their utterance within the relevant linguistic, social, economic and historical framework. In applying this criterion, we must, among other things, ask whether the speaker embraced the views espoused or distanced himself from them. We also have to consider whether the speaker is using code words or indirect means of inciting the audience in a way that will be grasped by listeners at that time and place.



Finally, the case law instructs the finder of fact to examine the relationship between the speaker and the subject. According to this part of the test, the analysis should be more speech-protective when the speaker is part of a minority criticizing the government or the country’s majority. In all due respect, the application of this four-part test certainly calls into question Professor Benesch’s conclusion that “a mere racist could be convicted of a crime tantamount to genocide, and speech may be unduly and dangerously restricted.”



Still, Professor Benesch refers to a poorly reasoned intermediate court decision in the Léon Mugesera case, subsequently overturned by the Canadian Supreme Court, to illustrate how “alarmingly” ill-defined the crime of incitement remains. In addition to being reversed by the Supreme Court, that decision was at odds with two lower court decisions finding Mugesera’s speech constituted incitement. The decision’s interpretation of Mugesera’s speech as being about “elections, love and courage” is instantly discredited by the language of the portions of the speech at issue. For example:

You know there are ‘Inyenzis’ [cockroaches] in the country who have taken the opportunity of sending their children to the front, to go and help the ‘Inkotanyis’ [Tutsi warriors, fierce fighters] …. Why do they not arrest these parents who have sent away their children and why do they not exterminate them? Why do they not arrest the people taking them away and why do they not exterminate all of them? … [We] must do something ourselves to exterminate this rabble…. I asked if he had not heard of the story of the Falashas, who returned home to Israel from Ethiopia? He replied that he knew nothing about it! [I] am telling you that your home is in Ethiopia, that we will send you by the Nyabarongo so you can get there quickly’…. Another important point is that we must all rise, we must rise as one man … if anyone touches one of ours, he must find nowhere to go.

Of course, the Canadian Supreme Court engaged in the kind of rigorous exegetical analysis necessary for evaluation of incitement allegations and included a contextual examination of the term “Inyenzi” as well the murderous metaphoric significance of transporting Tutsis to an ethnic-stereotype “homeland” via a non-navigable river traditionally used to dispose of corpses after ethnic massacres (even though it did not explicitly and systematically apply the four-part test). Thus, when seen in the larger context, the intermediate court’s decision represents a sui generis aberration in a string of decisions that had no difficulty finding Mugesera’s words constituted incitement. Such an anomaly is certainly not evidence of an “alarming” definitional deficit. (Professor Benesch comments that Mugesera’s speech preceded the Rwandan genocide by too long — seventeen months — but she ignores the essential point, made in the Canadian decisions finding incitement, that the speech itself was preceded and followed by large-scale ethnic violence – that was the context which permitted a finding of incitement.)



Nevertheless, as a solution to this perceived problem, Professor Benesch proposes a brand new test — that a speech be considered incitement to genocide if there is a “reasonable possibility” that genocide can occur when the speech was given. Although Professor Benesch acknowledges that causation has been rejected as a requirement for establishing incitement, I am afraid her proposed test creates enough of a conceptual fissure to let causation slide in through the back door. I submit that a retrospective actuarial assessment of the prospects for genocide is at odds with the fundamentally inchoate nature of the incitement crime. That an inchoate crime is committed prior to, and independently of, the object crime is axiomatic. The main purpose of punishing inchoate crimes is to allow the judicial system to intervene before an actor completes the object crime. The crime carries such a high risk for society that it must be punished without reference to subsequent acts, if any, of genocide. The crime is complete when the words are spoken in the proper context. And while it is true that incitement has never been prosecuted without a subsequent genocide occurring, adopting a test which would tend to perpetuate that pattern would needlessly cabin incitement law, which I believe should be used for its intended function — pre-atrocity deterrence, as opposed to mere post-atrocity punishment.



That said, the likelihood of subsequent violence given the circumstances surrounding a speech and the mental perspective of its listeners indirectly factor into the contextual analysis already called for in the law’s current iteration. Accordingly, the prongs of Professor Benesch’s six-part test do a wonderful job of fleshing out the existing four-part test. And so I view her analysis as an invaluable addition to incitement law.



In fact, I think it important to acknowledge that the analytic framework for incitement law is still developing. As I point out in my articles, the ICTR Media Case does not even explicitly set out a four-prong test – I argue that the test should be further refined to include four prongs, instead of what is formulated as two (purpose and context – although, as I indicate, the ICTR does actually engage in the analysis of text and relationship between speaker and subject without explicitly acknowledging it). And as I recognize above, the Mugesera Supreme Court decision was not sufficiently disciplined in its analysis to apply the test in a systematic, step-by-step way (although it does ultimately cover the essential components of the test). I also lament in my articles that the existing case law does not go far enough in identifying different types of incitement (such as “accusation in a mirror,” among others) and so I believe Professor Benesch’s contribution is timely and vital. I merely propose that her six-prong test be integrated into incitement law’s existing framework, which has been growing organically. In short, there is no sense in throwing the proverbial baby out with the bath water by adopting a new “reasonable possibility” test, which would likely lock incitement into its traditional role of retrospective punishment device and retard its recent evolution toward prospective deterrence mechanism. We must vigilantly protect free speech whenever and however we can but never at the cost of laying the groundwork for another genocide.