1. Introduction
We know that of the many things words can do,Footnote 1 one is cause significant harm. Words can cause direct psychological harm, but they can also directly or indirectly lead to physical violence. In the context of mass violence, words have been used to create and strengthen particular social relations conducive to neighbours killing neighbours.Footnote 2 As Lynne Tirrell notes, patterns of speech acts in Rwanda in the early 1990s developed into ‘linguistic practices that constitute[d] permissibility conditions for non-linguistic behaviors’, or mass violence.Footnote 3 What began as one social group using offensive terms to describe another social group ended in genocide. But when we try to impose criminal liability for speech-based participation in crimes related to mass violence, we are faced with a challenge in matching specific speech acts with specific crimes.
Incitement to genocide, for instance, remains a puzzling area of international criminal law due to at least two ongoing debates. First, we must balance a commitment to the principle of free speech, a principle with varying levels of deference in domestic legal systems, with a commitment to preventing the harms that can be caused by certain speech acts. Second, because incitement is an inchoate crime, we must determine how much what surrounds or follows a speech act should count in assigning criminal liability for that speech act. Neither debate has been settled thus far, despite the growing focus on hate speech in international criminal law. In light of these outstanding issues, I argue that we need a better understanding of the theoretical underpinnings of the debates surrounding hate speech and inchoate crimes.
In this article, I construct a theoretical basis for speech acts as incitement to genocide, distinguishing these speech acts from speech as genocide and speech denying genocide, by integrating international law with concepts drawn from speech act theory and moral philosophy.
I use the case drawn on by many commentators in this area of international criminal law, the trial of media executives for the roles they played in the Rwandan genocide through public speech acts by media entities insulting an ethnic group or advocating violence against an ethnic group.Footnote 4 Each of these men were institutional leaders and were charged with using their positions within Rwandan society to distribute what I call genocidal hate speech, genocidal incitement speech, and genocidal participation speech. I argue for a distinction between these three types of speech, and a difference in individual criminal liability for the dissemination of each type of speech. I also argue that there should be a difference in individual criminal liability for speech acts within the context of an ongoing or recent genocide, and speech acts that can be separated from a site of mass violence.
In Section 2, I sketch the landscape of incitement in domestic and international legal systems and I explore the conflict between free speech and hate speech in international criminal law. In Section 3, I consider an account of speech acts in the Rwandan genocide and demonstrate the need for distinctions between different types of speech acts in a genocide. I then introduce the tools I will use, from moral philosophy and speech act theory, and argue that these tools should inform our criminal liability structures. In Section 4, I analyze the charge of ‘direct and public incitement to commit genocide’ under international criminal law and through the lens of speech act theory. In Section 5, I discuss the charge of ‘genocide’ based on speech acts and distinguish genocide and incitement to genocide from other forms of hate speech that can occur during a genocide. In Section 6, I argue that existing genocide denial legislation cannot be justified by speech act theory in the same way that speech act theory can account for individual criminal liability in the context of an ongoing or recent genocide. By applying speech act theory and moral philosophy to the legal framework, I construct an account of speech acts and genocide that can ground a consistent understanding of incitement in international criminal law.
2. Incitement and international criminal law
Incitement means, generally, ‘encouraging or persuading another to commit an offence’.Footnote 5 It is considered to be an inchoate crime, meaning criminal liability for incitement does not require completion of the advocated crime. Black's Law Dictionary describes such an offence as a ‘step toward the commission of another crime, the step in itself being serious enough to merit punishment’.Footnote 6 But since the intended crime may or may not actually be committed, there are questions about whether and how much individual criminal responsibility should result from an inchoate crime like incitement.Footnote 7 The crime of incitement is thus handled differently by national legal systems and various human rights instruments, often based on how freedom of speech is balanced against goals of preventing harm. It is clear, however, that no legal system or human rights treaty recognizes an absolute right to free expression. In this section, I consider the treatment of incitement in domestic legal systems, international human rights law, and international criminal law.
2.1 Incitement and free speech in human rights law
International and regional human rights treaties contain provisions that limit free speech in order to prevent violence, as well as provisions that provide for freedom of expression. The International Covenant on Civil and Political Rights is a good example of the murky standard in international human rights law for how states should balance protection of free expression with prevention of violence and discrimination.Footnote 8 Article 19.2 reads:
Everyone shall have the right to freedom of expression; this right shall include freedom to seek, receive and impart information and ideas of all kinds, regardless of frontiers, either orally, in writing or in print, in the form of art, or through any other media of his choice.Footnote 9
This right is limited by Article 19.3, which notes that it may be restricted when necessary to protect the ‘rights or reputations of others’ or for the ‘protection of national security or of public order (ordre public), or of public health or morals’.Footnote 10 Article 20.2 offers the most relevant limitation, maintaining that ‘[a]ny advocacy of national, racial or religious hatred that constitutes incitement to discrimination, hostility or violence shall be prohibited by law’.Footnote 11
The European Convention for the Protection of Human Rights and Fundamental Freedoms has been interpreted by the European Court of Human Rights (ECtHR) to give broad scope to freedom of expression.Footnote 12 In Handyside v. The United Kingdom, the ECtHR determined that the principle of freedom of expression is, ‘applicable not only to “information” or “ideas” that are favourably received or regarded as inoffensive or as a matter of indifference, but also to those that offend, shock or disturb the State or any sector of the population’.Footnote 13 On the other hand, the International Convention on Elimination of All Forms of Racial Discrimination condemns, ‘the dissemination of ideas based on racial superiority or hatred, incitement to racial discrimination, as well as all acts of violence or incitement to such acts against any race or group of persons of another colour or ethnic origin’,Footnote 14 and does not attempt to balance this limitation with a corresponding commitment to freedom of expression.
Several international and regional human rights bodies apply the principle of proportionality in order to assess the acceptability of infringing on rights in order to prevent speech-based harm.Footnote 15 This principle, which maintains that any restriction on the freedom of expression ‘must be proportionate to the legitimate aim pursued by the limitation’,Footnote 16 has been applied in cases before the United Nations Human Rights Committee, the ECtHR, and the Inter-American Commission on Human Rights.Footnote 17 The precise scope of limitation on free expression is not clear in international human rights law, but it seems well-established that preventing incitement to violence or incitement to discrimination could be a permissible limitation. I now turn to how domestic legal systems handle balancing the prohibition of incitement with freedom of expression.
2.2 Incitement in domestic law
In the United States, the criminalization of incitement is understood narrowly and pitted directly against free speech. The courts tend to provide for broad protection of individual speech rights, regardless of the content of the speech.Footnote 18 The US Supreme Court consistently holds that government regulation of speech can only be permitted if it falls into one of the following categories:Footnote 19 fighting words,Footnote 20 true threats,Footnote 21 direct incitement,Footnote 22 obscenity,Footnote 23 child pornography,Footnote 24 or deliberate defamation or libel.Footnote 25 Thus, unlike in many human rights treaties, speech under US law cannot be limited merely due to its hateful content.Footnote 26
In 1969, the US Supreme Court declared unconstitutional laws that, ‘forbid or proscribe advocacy of the use of force or of law violation except where such advocacy is directed to inciting or producing imminent lawless action and is likely to incite or produce such action’.Footnote 27 Accordingly, there is an important distinction in the US between expression of protected opinions (that could, indirectly, contribute to someone eventually committing an unlawful act) and illegal incitement to immediate violent action. The circumstances surrounding the act of incitement are relevant in determining whether the incitement is likely to lead to immediate violent acts, but the actual commission of a crime following an act of incitement is not required.
Some civil law countries legislate along similar lines, criminalizing provocation (similar to incitement) as a specific form of participation in a criminal act.Footnote 28 Many countries with this type of legislation recognize the expressive force of the prosecution of the crime of incitement in preventing the development of conditions conducive to violent actions.Footnote 29
In the United Kingdom (UK) and in most civil law systems, there is less concern with the protection of free speech. In England and Wales, the Racial and Religious Hatred Act criminalizes the incitement of hatred against a person on the grounds of that person's religion.Footnote 30 British law on incitement recognizes a distinction based on whether or not there was actual commission of a crime following the alleged act of incitement. A person can be charged with incitement if no crime is committed as a result of the incitement. However, if a crime is committed, the person is not charged with incitement, but rather as complicit in the resultant crime, and thus is charged as an accessory to the crime or with the crime itself. Under this type of legislation, the inchoate crime of incitement is subsumed by any resultant crimes. Canadian law is similar. In 2005, the Supreme Court of Canada defined incitement in a way that does not require ‘a direct causal link between the speech and any acts of murder or violence’.Footnote 31
We now have several prominent views of how incitement should be prosecuted. The UK system and many civil law systems focus on the outcome of speech to determine whether to prosecute for incitement or for the resultant crime. The US laws on incitement allow for greater expression of opinions by assessing the likelihood that speech will actually lead to further crimes in determining whether to indict a speaker. Additionally, we have differing views on the expressive value of the detached crime of incitement. We will return to these legal schemes later in the article.
2.3 Genocide and incitement to genocide in international criminal law
The history of hate propaganda and violent speech in international conflicts begins with the trial of Julius Streicher at Nuremberg for his role as the publisher of Der Stürmer, an anti-Semitic weekly newspaper.Footnote 32 The Convention on Genocide was then adopted in 1948 in the wake of the Second World War and the Holocaust.Footnote 33 The definitions contained therein were the product of international consensus, but were largely influenced by the crimes that occurred in connection with the Holocaust. In this sub-section, I will focus on the specific crimes of genocide and incitement to genocide as they involve speech acts.
Article II of the Convention on Genocide defines ‘genocide’ and reads as follows:
In the present Convention, genocide means any 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.Footnote 34
The mens rea, or mental element, of the crime of genocide involves both the knowledge that circumstances conducive to the commission of genocide exist, and the ‘intent to destroy, in whole or in part, a national, ethnical, racial or religious group, as such’. For crimes of genocide, the prosecutor must prove that the offender had a ‘specific intent’, referred to as dolus specialis. Without dolus specialis, a crime cannot be punished as genocide.Footnote 35
In Article III, the acts of (a) Genocide; (b) Conspiracy to commit genocide; (c) Direct and public incitement to commit genocide; (d) Attempt to commit genocide; and (e) Complicity in genocide, are listed as punishable.Footnote 36 For purposes of this article, which focuses on the crimes of genocide and incitement to genocide in the context of the Media case brought before the International Criminal Tribunal for Rwanda, the applicable definition of genocide is identical.Footnote 37 The ad hoc committee that prepared a draft of the Convention on Genocide, and the members of the UN General Assembly who ultimately adopted the Convention on Genocide, debated at length the inclusion and wording of incitement as a punishable act.Footnote 38 Early drafts contained a qualifier specifically referring to the inchoate nature of the crime (‘. . . whether such incitement be successful or not’), with some delegations viewing the text as superfluous and others viewing the inclusion of the phrase as necessary to stress the preventive purpose of the Convention on Genocide.Footnote 39 The US delegation expressed concerns with regards to the inclusion of incitement as a punishable act, due to concerns about free speech limitation.Footnote 40 The UK delegation noted that since incitement would almost always result in conspiracy, attempt or complicity, the punishment of such an early stage of genocide was unnecessary.Footnote 41 Other delegations stressed the danger of incitement and the goal of preventing genocide as reasons for criminalizing incitement.Footnote 42 The final text of the Convention on Genocide (as cited above) was adopted unanimously and without abstentions by the UN General Assembly on 9 December 1948.Footnote 43
Despite the agreed-upon text of the Convention on Genocide (and the identical text in the statutes governing the International Criminal Tribunal for Rwanda and other ad hoc international criminal tribunals), we are still left with a problem. Prosecutors and judges must determine how to charge defendants who have been accused of crimes related to their speech within the context of a genocide. These prosecutors and judges need more than a definition of the crime of direct and public incitement to commit genocide in order to assess these speech acts, especially because they hail from different nations with different priorities related to free speech and criminal prosecution. The distinction between genocidal hate speech, genocidal incitement speech, and genocidal participation speech, which I will begin to explore in Section 3, addresses the problem of how to prosecute speech acts inciting violence.
3. The Media case, speech act theory, and responsibility
3.1 The Media case
In 2003, the International Criminal Tribunal for Rwanda (ICTR) rendered judgment for three Rwandan media executives who held leadership positions before and during the genocide of 1994 (the Media case). Ferdinand Nahimana, Jean-Bosco Barayagwiza, and Hassan Ngeze were convicted of genocide, direct and public incitement to commit genocide, conspiracy to commit genocide, and crimes against humanity, although all three convictions for genocide were overturned on appeal.Footnote 44 Nahimana and Barayagwiza were co-founders of the radio station Radio Télévision Libre des Mille Collines (RTLM), and Barayagwiza was also a founding member of the party Coalition pour la Défense de la République (CDR). Ngeze, a journalist, was the founder and editor-in-chief of the newspaper Kangura and a founding member of the CDR party.
I use the Media case as a paradigm case to explore the theoretical distinctions between different types of speech in the context of genocide. However, I do not provide a thorough background of the case itself, as this has been well-covered by many others.Footnote 45 Instead, I will draw out some key features of the Media case in order to demonstrate the need for philosophical concepts to distinguish between different types of speech related to genocide.
3.2 Hate speech as derogatory terms
In the article ‘Genocidal Language Games’, Lynne Tirrell argues that, ‘the derogatory terms used against the Tutsi during the Rwandan genocide were action-engendering’ and thus played a crucial role in the Rwandan genocide.Footnote 46 Her analysis focuses on the effects of what I will later define as ‘genocidal hate speech’, or speech used as propaganda in creating a climate of hatred and permissibility. Tirrell mistakenly relies on the Media Trial Chamber Judgement, rather than the Media Appeals Chamber Judgement, in claiming that the defendants in the Media case were convicted ‘of genocide, not only incitement’ for the roles they played at their respective institutions.Footnote 47 More importantly, because Tirrell's analysis fails to consider the legal repercussions of the speech acts, she fails to adequately distinguish between different types of speech acts made by the defendants in the Media case.Footnote 48 Even though she claims that the ‘action-engendering force of derogatory terms’Footnote 49 implies the successful completion of an act of violence, as a result of persuasion by the defendants in the Media case, Tirrell's account does not seem to focus on terms that explicitly incite or participate in genocide. Rather, Tirrell's analysis looks at the effects of ‘genocidal hate speech’ as propaganda in creating a climate of hatred and permissibility.
Tirrell focuses on what she calls ‘deeply derogatory terms’, which are ‘tied to systems of oppression’ and have five distinguishing characteristics.Footnote 50 First, they express the insider/outsider function, which means they assign ontological status in marking members of an outside group.Footnote 51 Second, they communicate a negative message that there is an essential difference between the inside group and the outside group, establishing a hierarchy.Footnote 52 Third, deeply derogatory terms are connected to social context, specifically as they include networks of oppression and discrimination.Footnote 53 Fourth, their function varies based on the particulars of the context.Footnote 54 Finally, deeply derogatory terms are action-engendering within a context.Footnote 55
While her analysis is useful in understanding the social atmosphere in Rwanda in 1994 and the psychological impact of the speakers on the hearers (who went on to commit genocide), it is not useful in considering the legal culpability of the Media case defendants, and it reveals that we need to better integrate concepts from moral philosophy, speech act theory, and international criminal law. Tirrell's analysis is insufficient for three main reasons. First, her conception does not consider the dolus specialis. Because genocide is a crime of specific or special intent, it necessarily involves a perpetrator who specifically targets victims on the basis of their group identity with a deliberate desire to inflict destruction upon the group itself. Tirrell's account of deeply derogatory terms addresses the role of implicit speech acts in licensing the genocide, but it does not consider the legal requirement of a special intent to commit (or incite) genocide on the part of the speaker. There is a distinction, at least in terms of criminal culpability, between contributing to the creation of permissibility conditions for genocide, and committing or inciting genocide, and this is largely captured by the dolus specialis requirement.
Second, Tirrell's comparison of the use of deeply derogatory terms in Rwanda with an instance of children calling a shirtless construction worker ‘sausage-face’ seems to minimize the moral and criminal culpability of a person advocating the destruction of an entire group of persons.Footnote 56 I recognize that she attempts to use a silly game to distinguish between mere insults and deeply derogatory terms, but this distinction is not helpful for understanding her analysis of the Rwandan genocide. It should be assumed that any speech acts that arguably contributed to genocide, are more serious than the unintentional speech acts of unaccountable children. Instead, Tirrell should have focused on the difference between one kind of speech, which may contribute to the permissibility conditions for genocide but not incur criminal culpability for the speaker, and other types of speech, which are actually ‘action-engendering’ in a more explicit way.
This leads me to my final concern. Tirrell's account does not adequately distinguish between the various types of genocidal speech I identify, of which only two lend themselves to criminal culpability. The speech acts Tirrell focuses on can only constitute genocidal hate speech. Any action on behalf of a hearer that results from a genocidal hate speech act, is merely the result of persuasion, not a criminal act of incitement or a criminal act of instigation. This does not change the negative impact of the hearer's actions, but it does control the individual criminal liability of the speaker.
Now that I have considered the difficulty in balancing free speech concerns with preventing violence in the international criminal law of incitement, as well as the trouble with lumping all hateful statements into the same category of speech acts, I turn to the resources I argue help construct a basis for understanding incitement to genocide.
3.3 Austinian speech act theory
To answer these puzzles, I begin with the tools of speech act theory, specifically J.L. Austin's work.Footnote 57 Austin distinguishes between locutions, illocutions, and perlocutions, although one speech act can have more than one of these functions. Locutions are acts of saying something, namely the actual noises made, words said in a particular order, and the conventional meaning of the words as stated. Illocutions are acts done in saying something, explaining in what way the locutionary act is used. For Austin, the success or failure of an illocutionary speech act depends on the satisfaction of certain felicity conditions, namely the context, circumstances, and conventions surrounding the speech acts. Perlocutions are acts done by saying something, or the effects of an illocutionary speech act on the hearer or audience. Like illocutionary acts, perlocutionary acts can succeed or fail, but their success depends on the uptake and response of other actors. A perlocutionary act is unsuccessful if there is no uptake or further action on the part of a separate actor.Footnote 58
This three-fold distinction between locutions, illocutions, and perlocutions sets us up to begin solving the problem of assessing speech advocating for genocide.Footnote 59 We should be able to label a speech act as one (or more) of these types of acts in order to determine what moral and/or legal responsibility should attach to a given speech act. Speech act theory can help us parse out what parts of an actor's statement constitute an action in itself, and distinguish this from consequences that rely on a separate actor or other circumstances. We will return to speech act theory in Section 4, where I use speech act theory to analyze direct and public incitement to commit genocide.
3.4 Moral and legal luck in responsibility
As I noted in the previous sub-section, speech act theory can help us separate the parts of a statement that are solely attributable to the speaker from those that involve consequences or circumstantial factors, or even luck. The presence of luck also plays an important role in assessing the moral and/or legal responsibility of a speaker, particularly in light of the uptake and felicity conditions discussed above. These two features of Austin's speech act theory allow us to see how two speakers might make the exact same statement, with the exact same intent to incite violence, and yet the circumstances surrounding the two statements result in two very different sets of consequences. For instance, Speaker A could suggest to her younger brother that he hit their older brother because she thought he had taken the last cookie from the jar, and it had previously been promised to her younger brother. Speaker A's belief could be true, her younger brother could agree, and then he could proceed to hit their older brother. But Speaker B, in the same scenario and with the same information, might be wrong about her older brother taking the last cookie, even though she successfully convinces her younger brother to hit her older brother. However, in this case, the violence was not tied to a claim of justification on the part of Speaker B. Or consider Speaker C, in the same scenario and with the same information as Speaker A, who might fail to convince her younger brother to hit her older brother because Speaker C's younger brother has, unbeknownst to her, become a pacifist. All three speakers acted in an identical manner, yet the outcomes differ.
We may think that these speakers should be morally assessed in the same way, or that the speakers should be legally assessed in the same way, or both, or neither. But we have to have a reason for assessing the speakers differently, despite the identical nature of their acts and their identical intentions. If we think that Speaker C is less morally culpable because no violence occurred due to her speech act, or that Speaker B is more morally culpable because unjust and inexplicable violence occurred due to her speech act, we accept moral luck. We think that circumstances surrounding our actions change our moral responsibility.
Or we may think that these distinctions are not relevant in the moral sense, and argue that an actor should be held morally responsible for reasonably foreseeable acts taken in response to the actor's own statements. We may think that this is the only fair way to assess the moral culpability of the three speakers, based on outcomes that an actor could have foreseen as the result of her speech acts. Thus we would say that Speaker A, Speaker B, and Speaker C are equally morally culpable for their statements and their intentions, despite the varying outcomes. If we claim that agents should be praised or blamed equally if the only difference between two scenarios is due to luck, then we deny moral luck. If we do so, then the same moral responsibility should be attributed to all identical actors, regardless of the differences in uptake on the part of listeners, or the differences in circumstances, and corresponding differences in actual consequences.
However, whether or not we deny moral luck, we must identify something other than moral luck that justifies differential punishment based on outcomes, because domestic and international criminal legal systems do punish actors differently based on different outcomes. If Speakers A, B, and C are now suggesting that their younger brothers poison their respective older brothers, and the older brothers of Speaker A and Speaker B die, the legal consequences for Speaker A and Speaker B will differ significantly from the legal consequences for Speaker C. Even if we morally assess all three speakers identically, Speaker C will not face any legal consequences (assuming the incitement remains a family matter), purely as a result of legal luck. Speaker A and Speaker B will likely face significant legal consequences. I now turn to two views of legal responsibility that could justify the legal luck characterizing existing punishment practices.
On a subjectivist view of legal responsibility, only the intentions and corresponding actions of an agent are sufficiently under the agent's control, and therefore the intentions and actions serve as the basis of legal responsibility.Footnote 60 According to this view, risk creation is the basis for criminal responsibility; attempts are thus seen in the same way as successfully completed acts.Footnote 61 Sanford Kadish argues for the subjectivist view based on the deterrent purpose of criminal law.Footnote 62 Because the occurrence of harm is not related to the intention of an actor, it does not serve as a deterrent. This view only justifies punishment aimed at deterring a specific offender, and punishment should ensue regardless of the actual harm created.Footnote 63
Alternatively, we could take an objectivist view of legal responsibility. This view identifies the responsibility of an actor based on the amount of harm caused by the actor, in addition to the actor's intention and action.Footnote 64 On this view, an actor has control over possible outcomes from a chosen course of action, assuming they are reasonably foreseeable, and therefore the results constitute a part of the activity.Footnote 65 Michael Moore argues that because actors can reasonably predict outcomes of their actions, any separation between actions and consequences is merely artificial.Footnote 66 The purpose of punishment, then, is general deterrence of similar future actions, but also expressing that the harm caused is not acceptable.
The objectivist view is more compelling, both in terms of justification and desirable legal policy. International criminal punishment will always serve as a specific deterrent, in that an individual is virtually guaranteed not to re-offend during their incarceration. Specific deterrence can work if the possibility of punishment actually affects actors’ decisions to participate in criminal activities. We have some evidence that the cases pending before the International Criminal Court (ICC) may be serving as a specific deterrent for government leaders or leaders of other organizations, since other leaders of these organizations have been held accountable for their actions before the ICC.Footnote 67
But given the collective nature of most international crime and the collective nature of crimes that result from incitement, it is hard to assess the long-term specific deterrent effect of international criminal law on individuals who seek to incite violence. Economic accounts of deterrence, premised on the idea that people engage in wrongdoing for personal gain, explain criminal behaviour as unrelated to the specific preferences of individuals.Footnote 68 These accounts suggest that people will be less likely to commit crimes if there is an additional cost to their wrongdoing, namely that they will face prison time if caught and convicted.Footnote 69 But an actor seeking to contribute to the destruction of another group is most likely not acting out of personal or economic self-interest, and thus may not be susceptible to a specific deterrent effect.Footnote 70 Finally, most evidence suggests that international criminal law does not appear to serve as a specific deterrent.Footnote 71 So, if we see the purpose of international criminal law as preventing harm, through general deterrence or through expression of what we see as harms worth preventing through the law, then we should take an objectivist view of legal responsibility. An actor should be held legally responsible for harms that are reasonably foreseeable as a result of her intended actions.
It is clear that the ICTR has seen the legal responsibility of inciters similarly, but to the end that hearers are nearly absolved of responsibility. The Media case Trial Chamber treated the RTLM and Kangura audiences as automatons, influenced mechanistically by the words of a speaker, rather than legally responsible under the objectivist view. But each actor is responsible for her own intentions and actions, and also has at least some control over possible outcomes from a chosen course of action. A particular outcome is not pre-determined if it relies on the uptake of another individual, as that individual has responsibility for her own intentions and actions, regardless of outside influence. As will become clear in Section 4, the relationship between an inciter and hearer is much more complex, at least as far as it relates to individual criminal responsibility.
Because speech act theory can help us isolate aspects of moral and legal luck in speech acts, it plays a crucial role in allowing us to establish objective legal responsibility under international criminal law. In Section 4, I look at how the distinction between different types of speech acts can help assess the inchoate crime of incitement.
4. Incitement to genocide
In this section, I focus on the conviction of the defendants in the Media case of the crime of direct and public incitement to commit genocide, and, in the following section, on the overturned genocide convictions. By looking at the Media case, we can analyze how the intent, performance, and results of various speech acts can lead to distinct indictments and convictions.
I briefly note that the mens rea requirement of dolus specialis, referred to above in the genocide definition, applies to both crimes. That is, a person convicted of direct and public incitement to commit genocide, must have the intent to cause the destruction, in whole or in part, of a national, ethnical, racial or religious group, as such. For the defendants in the Media case, who were each initially convicted of genocide in addition to direct and public incitement to commit genocide, it was not necessary to show the special intent to incite genocide because it was established that each of the defendants had the requisite intent to commit genocide. The Media case Trial Chamber considered the fact that genocide occurred ‘significant’ evidence of genocidal intent.Footnote 72 On appeal, when the Media case Appeals Chamber overturned the genocide convictions, the genocidal intent was established through other evidence.Footnote 73 There is no need to establish a specific ‘intent to incite’.
4.1 Causation and inchoate crimes
In the first judgment handed down by the ICTR, the Akayesu Trial Chamber found that there must be a connection between the dissemination of propaganda and the commission of a crime, stating that the prosecution has the burden of proving ‘a definite causation between the act characterized as incitement . . . and a specific offence’.Footnote 74 The Media case Trial Chamber cites the Akayesu Trial Chamber approvingly with respect to proof of causation,Footnote 75 and also makes a number of claims about the causal effect of the media on the genocide of the Tutsis.Footnote 76 However, the Media case Trial Chamber also chose to emphasize the ‘potential of the communication to cause genocide’Footnote 77 as the requirement for a finding of incitement, vacillating from the opinion that causation is a requirement for incitement to commit genocide, and the opinion that as an inchoate crime, causation is not a requirement for a finding of incitement. The inchoate nature of incitement to genocide was confirmed by the Media Case Appeals Chamber.Footnote 78
4.2 Hate speech v. Incitement speech
The Media case Appeals Chamber also distinguished between ‘hate speech in general (or inciting discrimination or violence)’ (which I refer to herein as genocidal hate speech Footnote 79 ) and direct and public incitement to commit genocide (referred to herein as genocidal incitement speech).Footnote 80 Genocidal hate speech dehumanizes, contributes to a culture of hate, and creates permissibility conditions for the commission of genocide. An example of genocidal hate speech would be an RTLM radio host stating that ‘Tutsis are cockroaches’.
Because it is likely that a speaker of genocidal hate speech intends to convey hatred and cause harm through the speech act, and may in fact do so, either directly or indirectly, genocidal hate speech can always be seen as a locutionary act.Footnote 81 However, genocidal hate speech does not provide instructions or call for any action in connection with hate propaganda, and this distinguishes genocidal hate speech (which creates permissibility conditions for a genocide to occur, or at least fails to prohibit the commission of genocide) from genocidal incitement speech (which is an explicit command to commit genocide). ‘Tutsis, because they are cockroaches, must be exterminated, so you listeners must go exterminate them’ is an example of genocidal incitement speech. This functions differently from genocidal hate speech, under the law, because it directs someone to commit acts of genocide.
4.3 Felicity conditions
Felicity conditions play a minor role in understanding the speech acts I discuss in most of this article, but it is important to note that in the case of speech that occurred during the Rwandan genocide, the context, circumstances, and conventions surrounding the speech were such that they contributed significantly to the success of speech acts. Tirrell's account of deeply derogatory terms, discussed in Section 3, succeeds in describing the social conditions that surrounded the Rwandan genocide, even though it does not adequately distinguish between different types of speech acts.Footnote 82 The speech that occurred prior to and during the 1994 genocide took place within the context of intense conflict, as there had been violent clashes occurring between Tutsis and Hutus since 1959. According to Human Rights Watch, there were small-scale massacres of Tutsis in 1992 and 1993, which ‘established patterns for the genocide of 1994’.Footnote 83 Based on the ongoing violence and conflict between the Hutus and Tutsis in April 1994 and the linguistic permissions that had been developing alongside the violence, I assume that for the speech acts considered in the Media case, conditions were felicitous.
4.4 Illocutions, perlocutions, and luck
To understand the categorization of the explicitly performative act of direct and public incitement to commit genocide, Austin's conception of illocutionary acts and perlocutionary acts is a useful starting point. First, we consider the idea that a speech act of incitement is a perlocutionary act, dependent upon the co-operation of the audience in order to be successful.Footnote 84 Perlocutionary acts include the speaker's purpose, the ‘perlocutionary object’, as part of their meanings.Footnote 85 Therefore, if the mental state of a speaker is directed at inciting to commit genocide, then the perlocutionary act would be the attempted commission of genocide by a hearer of the speech act. If the hearer does not commit or attempt to commit genocide, and the speaker's perlocutionary object is not achieved, then the perlocutionary act is unsuccessful.
But does the failed purpose negate the perlocutionary act? It seems that there could be any number of possible consequences of an attempt to incite a hearer to take an action, and therefore it might make sense to consider perlocutionary effects based on the effect intended by the speaker, rather than the success or failure of the perlocutionary act.Footnote 86 But under this conception, if the intent of the speaker is all that matters, and the actual uptake or effect of the speech act is irrelevant, there does not appear to be any value in distinguishing between illocutionary acts and perlocutionary acts. The speaker is responsible for the success of the persuasion, and the hearer is no longer an independent agent.Footnote 87 I call this the intention-based account.
Alternatively, as Yueguo Gu suggests, a successful perlocutionary act should result in a change in the ‘addressee's understanding of sense, his/her recognition of force and his/her inferring of implicature’.Footnote 88 I call this the uptake-based account. Here, we recognize the hearer as an independent agent within the perlocutionary act, and the success depends on both the successful persuasion of the speaker and the successful, as-intended-by-the-speaker, uptake of the hearer.
Consider, in the context of the Media case, a ‘successful perlocutionary act’, where a RTLM radio announcer successfully persuades a radio listener that the Tutsi people are cockroaches. The mindset of the hearer has been altered such that the hearer no longer considers Tutsi people to be humans. This could be considered a successful perlocutionary act of genocidal hate speech under either conception discussed above, as it acknowledges both the intent of the RTLM radio announcer to persuade a hearer to change his mental attitude, as well as the active participation and appropriate uptake on the part of the hearer.
However, when the RTLM radio announcer attempts to persuade a hearer to exterminate the Tutsi people based on their status as cockroaches, there is a second piece of uptake required on the part of the hearer, in order to make the perlocutionary act successful under the uptake-based account. Even if the hearer is persuaded by the speech act of the RTLM radio announcer that the Tutsi people are cockroaches, he must also be convinced to take physical action to murder Tutsi people, in order to make it a successful perlocutionary act of genocidal incitement speech. On the intention-based account, it is the intended perlocutionary effect that constitutes the perlocutionary act, therefore the success does not matter.
Remaining in the context of genocide incitement speech, consider an instance in which the RTLM radio announcer persuades a hearer that the Tutsi people are cockroaches, and further persuades the hearer that the Tutsi people should be exterminated, and the hearer is successfully convinced that he should kill his neighbour, who he knows is a Tutsi. However, the hearer cannot find his machete, and therefore he does not kill his neighbour. Following Gu and the uptake-based account, the hearer has been persuaded to change his mindset, and the uptake has even convinced the hearer to take an action. However, for a reason other than the successful persuasion by the RTLM radio announcer, the incitement to commit a criminal act has not been a successful perlocutionary act.
But it seems strange to distinguish this case from a situation in which the hearer found his machete and killed his neighbour, thus resulting in a successful perlocutionary act. This sense of strangeness comes from the presence of luck, because it appears that the RTLM radio announcer (and the neighbour) were just lucky that the hearer could not find his machete. If we want to deny luck, we should blame and punish the RTLM radio announcer in the same way regardless of the success of the perlocutionary act.
In this scenario, we can see why the distinction between persuasion and incitement matters in criminal law, and why the criminal concept of incitement involves more than mere persuasion. And we can also see that our intuitions for individual responsibility, at least moral responsibility, do not solely track the successful commission of genocide on the part of a hearer. As Wilson notes, ‘only the locutionary and illocutionary aspects of the speech act are entirely under the control of the speaker’.Footnote 89 This leaves us unsatisfied with the account of genocidal incitement speech as a perlocutionary act of an individual.Footnote 90
4.5 International law of incitement to genocide
Given the inchoate nature of the crime of direct and public incitement to commit genocide, as confirmed by the Media case Appeals Chamber, it makes the most sense to deem direct and public incitement to commit genocide to be an illocutionary act rather than a perlocutionary act. If the uptake of the hearer cannot be taken into account in evaluating the success or failure of the perlocutionary act, and, in this case, it cannot, then the uptake-based account of perlocutionary acts is not useful and we must return to the intention-based account of perlocutionary acts. And, as stated above, this account dissolves the legal distinction between a perlocutionary act and an illocutionary act. Dennis Kurzon notes that if incitement is an illocutionary act, it is similar to other illocutionary acts in that it is usually indirect, the hearer, ‘has to draw inferences to understand the speaker's meaning, [and] incitement entails using a number of illocutionary acts – making statements, promises, and requests, but not one act which may be glossed as “to incite”’.Footnote 91
Considering the various approaches to incitement in domestic law, it seems that the British legal approach to incitement, discussed in Section 2, supports the idea that there is no such thing as incitement as a perlocutionary act. This is, in fact, consistent with concerns expressed by the British delegation during the drafting of the Convention on Genocide. Incitement is an illocutionary act, but if the completed crime occurs, it becomes a perlocutionary act and the inciting speaker becomes an accessory to the crime rather than a mere inciter. However, in the Media case, this would mean that the defendants should only have been charged with genocide or conspiracy to commit genocide, based on the ICTR's assessment of their level of participation in the crime, and not with incitement to genocide.
Yet, international criminal laws against genocide assume (at least in practice) that genocide has occurred, for any of the laws to be invoked. International criminal laws are invoked by international criminal courts, and these courts generally establish at the beginning of a trial that genocide or some other atrocity has occurred for a case to go forward. This practice of judicial notice means that international criminal laws are not, or at least have not yet been, invoked by courts without first establishing the fact that atrocities occurred. It is possible that someone could be charged with direct and public incitement to commit genocide, based on speech acts advocating for the extermination of a group, if no violence takes place, but this seems highly unlikely. We are thus left with a question as to the value of the freestanding crime of direct and public incitement to commit genocide. Under the British conception of an unsuccessful act of incitement as an illocutionary act, the only use for the crime of direct and public incitement to genocide would be if someone tried to incite genocide, and was not heard or did not influence anyone to commit genocide, but genocide happened to occur anyway. This scenario too seems unlikely. We will now turn to a final possibility for the value of the crime of direct and public incitement to commit genocide.
4.6 Incitement and genocide prevention
The usefulness of the analysis of international criminal law using speech act theory is limited, depending on what we see as the purpose of punishment in international criminal law. As noted earlier, one benefit of using speech act theory to inform our legal analysis is that it helps us get clear on the location of responsibility by identifying the influence of luck on a given scenario. If punishment is intended to hook up solely with the actions and intentions of the speaker, we should be in the practice of denying moral luck as grounds for punishment, but this is not the only thing we are trying to capture with punishment.
We might think, following Larry May, that, ‘incitement is so potentially dangerous an activity that perhaps it should be treated as punishable in itself, just as “attempted murder” is punishable independently of whether the defendant murdered’. Footnote 92 It may be that the purpose of the Convention on Genocide (and the similarly-purposed ICTR Statute) goes beyond establishing culpability for crimes of genocide, as suggested by delegates from other nations charged with drafting the Convention on Genocide. If this is the case, we could see an expressive reason for punishment, such as the prevention of genocide. The prevention of genocide may not be a defensible purpose for domestic legal systems with free speech concerns, but in international criminal law, it may make sense to include direct and public incitement to genocide as a separate crime, regardless of the likelihood that a defendant could be convicted of active participation as well. If we think that genocide is so bad as to justify preventing any risk of that harm, we should deny both moral and legal luck as exculpating.
Genocide prevention seems to be a worthwhile initiative, and given the seriousness of genocide, a reasonable influence on international criminal law. However, it is still not clear why criminalizing direct and public incitement to genocide prevents genocide any better than the other crimes listed under the Convention on Genocide. Unless international bodies and tribunals intend to utilize the charge in prosecuting speakers who have not also been charged with active participation in genocide,Footnote 93 direct and public incitement to genocide merely increases the number of crimes, for which one can be charged. As in the Media case, one can be charged with genocide, conspiracy to commit genocide, and direct and public incitement to commit genocide, but what is really gained by adding this latter charge? As I argue, the speech acts can be sufficiently identified and punished under the crimes of genocide and conspiracy to commit genocide.
Accordingly, we could deny moral luck, dishing out moral blame for an actor who makes statements intended to incite others to commit genocide. Moral blame (in the form of political, religious, or social condemnation) also serves an expressive function. But we should punish this actor based on the incomplete crime, because we are not in an epistemic position to identify a complete crime, nor are we even in an epistemic position to perfectly identify the actor's intention. Even with potential outcomes as grave as genocide, this objectivist view seems to support legal luck in constructing laws related to inchoate crimes like incitement. We would still punish an actor who intends to incite others to commit genocide, but we would support differential punishment due to the lack of uptake and avoided harm. We punish to deter future crimes of incitement that are incomplete, deter future crimes of incitement with uptake that result in completed offences of genocide, and to express that such actions are morally and legally blameworthy. I now take up the completed offences of genocide in Section 5.
5. Genocide, as such
The defendants in the Media case were also initially convicted of the crime of genocide, based on their actions as leaders of media entities. The Media case Trial Chamber found that where the potential of communication to cause genocide (i.e., the direct and public incitement to commit genocide) was realized, both the crime of incitement to genocide and the crime of genocide had been committed.Footnote 94 The Media case Trial Chamber found that the language of RTLM, Kangura, and each of the individual defendants, met the high bar for genocidal intent.Footnote 95 The Media case Trial Chamber actually noted that while, ‘[t]he nature of media is such that causation of killing and other acts of genocide will necessarily be effected by an immediately proximate cause in addition to the communication itself’, this fact ‘does not diminish the causation to be attributed to the media, or the criminal accountability of those responsible for the communication’.Footnote 96
However, the Media case Appeals Chamber overturned each of the convictions for genocide. The Media case Appeals Chamber found that there was not sufficient evidence that any of the three defendants played an active part in the radio transmissions of the RTLM or the publishing of Kangura after the genocide began on 6 April 1994.Footnote 97 In order to be convicted for the crime of genocide based on a speech act, there must be clear causation between the speech act and an act of genocide, and the Media case Appeals Chamber determined that this was not the case.Footnote 98 Therefore, the following analysis of perlocutionary speech acts constituting genocide is hypothetical, such as the example given in which the relationship between the speech act and the genocidal act is clear.
5.1 Genocide as a perlocutionary act
Article 6 of the ICTR Statute provides (paragraph 1):
A person who planned, instigated, ordered, committed or otherwise aided and abetted in the planning, preparation or execution of a crime referred to in articles 2 to 4 of the present Statute, shall be individually responsible for the crime.Footnote 99
This definition of individual criminal responsibility is what distinguishes the crime of direct and public incitement to commit genocide from the crime of genocide, as such. The defendants in the Media case were initially found directly guilty of genocide, ‘[f]or their words and deeds – in a sense for their words as deeds that instigated the killing of Tutsi civilians as such’,Footnote 100 although ultimately the evidence did not support a finding under this standard.
The verbs found in Article 6 of the ICTR Statute are distinguishable from the verb ‘to incite’, in that they do not presume an inchoate crime of genocidal incitement speech. Rather, these verbs require a successful perlocutionary act of genocide, which is possible in part due to the felicitous circumstances in which the genocide took place. Due to the success requirement, the distinction between the illocutionary act and the perlocutionary act now proves helpful in analyzing speech acts as genocide. We can also distinguish speech acts to which these verbs apply from genocidal hate speech, labeling the speech acts associated with planning, instigating, ordering, committing or aiding and abetting in the commission of the crime of genocide, genocidal participation speech. This type of explicit performative is a basis for direct criminal liability.
Consider now a scenario in which an RTLM radio announcer produces genocidal participation speech. The RTLM radio announcer persuades a hearer that the Tutsi people are cockroaches, and then persuades the hearer that the Tutsi people should be exterminated, de dicto, as with genocidal incitement speech. However, in this scenario, the RTLM radio announcer gives a list of names of Tutsi people to be exterminated, de re. A hearer, who has been persuaded that Tutsis are cockroaches, and has been persuaded that they should be exterminated, hears the names of two of his neighbours read out in the list of Tutsi people. This time, he is able to find his machete, and he murders his Tutsi neighbours.
In the case of genocidal participation speech, Gu's uptake-based account of perlocutionary acts as joint ventures between the speaker and the hearer proves useful. The speaker's illocutionary act must contain the dolus specialis and the speaker must have an intent that the perlocutionary effect of the speech act be genocide. Due to the causation requirement, a successful change in the attitude on the part of the hearer, and the commission of a lethal act by the hearer are relevant for the success of the perlocutionary act of genocide. Should the proper uptake not be secured, the illocutionary act of genocidal participation speech would be considered an attempt to commit genocide, rather than a perpetration of genocide, because any concern with the successful uptake on the part of the hearer prevents the speech act from being charged as an inchoate crime like incitement.
5.2 Genocidal hate speech
Genocidal hate speech, as I have used the term throughout this article, is generally in line with Tirrell's conception of deeply derogatory terms used within a particular context. This type of speech could perhaps be categorized as a crime against humanity, but it should be distinguished from genocidal participation speech and genocidal incitement speech. Speech acts incurring this type of criminal liability should also be distinguished from protected offensive speech, but this designation allows for criminal liability that does not require a perlocutionary or illocutionary speech act.
Article III of the ICTR Statute (‘Crimes Against Humanity’) states that the ICTR:
shall have the power to prosecute persons responsible for the following crimes when committed as part of a widespread or systematic attack against any civilian population on national, political, ethnic, racial or religious grounds: (a) Murder; (b) Extermination; (c) Enslavement; (d) Deportation; (e) Imprisonment; (f) Torture; (g) Rape; (h) Persecutions on political, racial and religious grounds; (i) Other inhumane acts.Footnote 101
The Media case Trial Chamber categorizes instances of what I would call genocidal hate speech (and most of what Tirrell identifies as deeply derogatory terms) as ‘persecutions’ under the definition of crimes against humanity. The Media case Trial Chamber perceived hate speech in this implicit form as:
a discriminatory form of aggression that destroys the dignity of those in the group under attack. It creates a lesser status not only in the eyes of the group members themselves but also in the eyes of others who perceive and treat them as less than human.Footnote 102
Aligning with Tirrell's conception of deeply derogatory terms, persecution was seen as ‘conditioning’ a population and ‘creating a climate of harm’.Footnote 103
The classification of genocidal hate speech as a crime against humanity, rather than genocide, slightly lowers the bar for the speaker's mens rea to the requirement that the speech acts must be committed on discriminatory grounds. There is no special intent requirement to prove a crime against humanity,Footnote 104 and there is no concern about the successful uptake of any hearer. Furthermore, there is no expectation of a secondary actor as the hearer, only the object of the genocidal hate speech (i.e., the victim of the persecution). The actus reus requirement for a crime against humanity is the context of ‘a widespread or systematic attack against any civilian population on national, political, ethnic, racial or religious grounds’.Footnote 105 The context requirement would go a long way towards avoiding free speech concerns, since criminalizing ‘persecutions’ outside of such a context could limit the ability of individuals (or media outlets) to express their opinions.
However, crimes against humanity are an aspect of international criminal law that could be used more liberally than international criminal laws pertaining to genocide have been. These crimes could be charged, based on acts or speech acts that have taken place in areas where no genocide has taken place or been recognized. Therefore, the analysis of what constitutes an attack will play a bigger role, and the concerns about limiting free speech will remain controversial. As Catherine MacKinnon notes, the Media case Trial Chamber's analysis provides a novel harmonization between equality and speech rights, and will continue to play an important role in evaluating the distinction between protected speech and discrimination around the world.Footnote 106 In Section 6, I will see how this might play out in another realm of criminal law.
6. Genocide denial legislation
In this section, I argue that my normative account of hate speech in the context of genocide cannot justify, at least some, existing genocide denial legislation. By looking briefly at Holocaust denial legislation in Germany, we can see the practical importance of distinguishing the three specific types of speech I considered above from other types of hate speech. The goal of making the distinction between genocidal hate speech, genocidal incitement speech, and genocidal participation speech was to account for concerns about allowing for freedom of expression while preventing violence against vulnerable individuals, as well as assigning criminal liability for speech acts based on the actor rather than the surrounding circumstances. Using my account, in most circumstances, we cannot appropriately assign criminal liability for speech acts denying genocide.
6.1 Genocide denial
While a conspiracy is ongoing, perpetrators of a particular genocide cannot be permitted to deny that genocide.Footnote 107 When a community is still moving past recent atrocities, genocide denial may be highly likely to incite immediate violence. In Rwanda, where the ethnic tensions remain and a large portion of the population remembers the 1994 genocide, genocide denial legislation can be justified on these grounds.Footnote 108 In the case of Rwanda, ‘[g]iven the short passage of time and the existence of militant groups that deny the genocide, a statute specifically outlawing genocide denial [is an] important tool to help Rwanda overcome the racist attitudes that have fuelled its violent past’.Footnote 109
Yet, this justification cannot suffice for much of the European Union, Canada, and Australia, which have all criminalized denial of the Holocaust. These communities are not uniquely positioned for violence related to statements of denial, and free speech concerns should permit ‘deniers to espouse their beliefs, however bogus’, as it will tend to strengthen ‘society's sense that its individuals are autonomous and capable of making important personal choices’.Footnote 110 Germany and Israel also have Holocaust denial legislation. Israel's legislation may be justifiable based on the existing rates of ethnically-motivated violence. Germany, on the other hand, is many years past the Holocaust, despite its unique connection to the atrocities. In the next section, I will consider whether Holocaust denial legislation in Germany can be justified.
6.2 Case study: Germany and the Auschwitz lie
The aftermath of the Second World War and the Holocaust led to a German legal system less concerned with protecting free speech, and more concerned with protecting the values of individual human honour and dignity, than some other countries. In Germany, the Basic Law (Grundgesetz) guarantees freedom of speech and freedom of opinion, but these rights are expressly subject to limitations defined in ‘the general laws, the provisions of law for the protection of youth, and by the right to inviolability of personal honour’.Footnote 111 The Criminal Code, one type of the ‘general laws’, has been continually strengthened in response to neo-Nazi propaganda, particularly the ‘so-called “Auschwitz lie” – the claim that the extermination of European Jews by the National Socialist regime never took place, that such reports were a deliberate lie’.Footnote 112 It is important to note that while this focus on personal dignity over free speech would be highly contentious in the US, it is widely assumed in Germany that the law and the courts should play a significant role in preventing the ‘Auschwitz lie’ from being used as hate propaganda.Footnote 113 This German concept of incitement to popular hatred has been termed ‘Volksverhetzung’.Footnote 114
Article 130 of the German Criminal Code was motivated by the public interest in safeguarding public peace. The first section outlaws acts committed ‘in a manner capable of disturbing the public peace’, of incitement of ‘hatred against segments of the population or calls for violent or arbitrary measures against them’, or assaults on ‘the human dignity of others by insulting, maliciously maligning, or defaming segments of the population’.Footnote 115 The second section pertains to written materials:
which incite hatred against segments of the population or a national, racial or religious group, or one characterized by its ethnic customs, which call for violent or arbitrary measures against them, or which assault the human dignity of others by insulting, maliciously maligning, or defaming segments of the population or a previously indicated group, Footnote 116
and outlaws the dissemination, public display, supply to children, or physical production of such written materials, or the dissemination of a broadcast of such content by radio, television or other media.Footnote 117 The third section outlaws the public approval, denial or downplay of particular acts committed under the rule of National Socialism, ‘in a manner capable of disturbing the public peace’.Footnote 118 The fourth section outlaws disturbance of the public peace ‘in a manner that violates the dignity of the victims by approving of, glorifying, or justifying National Socialist rule of arbitrary force’.Footnote 119 Article 130a outlaws the attempt to cause the commission of the offences listed in Article 130.Footnote 120
6.3 Speech act theory and genocide denial legislation
The prohibited Volksverhetzung appears to be dissimilar to the concepts of genocidal hate speech, genocidal incitement speech, and genocidal participation speech discussed above, and therefore the German laws are not justifiable by the same theoretical means. The acts contemplated by Article 130 seem most similar to genocidal incitement speech, in that they focus on hateful speech acts that are ‘capable’ of disturbing the public peace rather than acts that actually disturb the public peace. This implies that Article 130 outlaws inchoate crimes, or illocutionary acts, since the perlocutionary success of the speech acts is irrelevant for prosecution. This is in line with my previous analysis of acts of incitement as illocutionary acts.
However, if the acts outlawed by Article 130 are considered illocutionary acts, then there must be felicitous conditions in place for them to succeed as perlocutionary acts of completed crimes of genocide. Unlike in Rwanda in 1994, it is not clear that a context exists in Germany such that incitement speech could constitute genocidal incitement speech. History and the fact that the Holocaust occurred do not provide enough evidence that social structures and linguistic permissions exist in Germany, at least today, that would permit such horrific acts to occur. The levels of ethnic, religious, and national hatred are simply not sufficient. The fact that these laws exist at all, looks more like evidence of a national German commitment to pluralism and tolerance, rather than evidence of a climate of hatred. Therefore, the context is unlikely to lead to successful perlocutionary acts of hatred, or even disturbance of the public peace. The public peace may be disturbed by persons who are actually opposed to the views of those propagating Volksverhetzung, but this would constitute an unintended perlocutionary effect.
The acts outlawed by Article 130 could still be seen as illocutionary acts if the speech acts are considered an illegal harm in and of themselves, and not attempts to incite genocide. It may be that the German law and courts have determined that Volksverhetzung or genocide denial should be punished for reasons other than the acts that might occur as a result of such speech. As noted above in Section 4.6, the seriousness of the crime of genocide may justify the imposition of criminal liability on persons who threaten the broad goal of genocide prevention.
However, these are not the reasons provided by the German laws, which explicitly focus on the capacity for these acts to disturb the public peace. The empirical question of how to balance the value of free speech against speech that may disturb the public peace or offend someone in Germany today, in a more minor fashion than mass killing, is a question beyond the scope of this article. My argument of this section, then, remains modest. The genocide denial laws in Germany cannot be justified on the same grounds as international criminal laws against genocidal hate speech, genocidal incitement speech, and genocidal participation speech. There must be a narrower purpose in outlawing genocide denial speech, and Article 130 does not currently provide enough justification for its stated goals.
7. Conclusion
The judgements handed down by the ICTR in the Media case established that certain types of speech can constitute or contribute to some of the most harmful crimes under international law. By distinguishing between genocidal hate speech, genocidal incitement speech, and genocidal participation speech, I have shown how speech act theory justifies the international criminal law that places individual criminal responsibility on the perpetrators of these forms of speech. My account responds to two debates that pervade the intersection of hate speech and international criminal law: namely, the balancing of freedom of expression with the prevention of violence, and the challenge in imposing individual criminal liability for the inchoate crime of incitement to genocide. I have also shown that while the speech act theory can help us understand the international criminal laws pertaining to genocide, and justify imposition of legal responsibility on individual speakers in the context of genocide, it cannot defend certain domestic genocide denial laws on the same grounds.
