Terrorist-Extremist Speech and Hate Speech: Understanding the Similarities and Differences

Abstract

The terms ‘hate’ and ‘hatred’ are increasingly used to describe the rationale of a kind of anti-Western terrorist-extremist speech. This discursively links this kind of terrorist-extremist speech with the well-known concept of ‘hate speech’, a link that suggests the two phenomena are more alike than they are unlike. In this article I interrogate the similarities and differences between anti-Western terrorist-extremist speech and hate speech as they manifest in Western liberal democratic states along two axes: to whom the speech is addressed, and how harm is occasioned. Relying on a combination of philosophical conceptions and public policy empirics, I demonstrate that there are significant differences between the two types of speech, especially in their mechanisms of harm. The implications of this analysis are that these differences should be better understood in order to respond appropriately to these two distinct types of harmful speech.

1 Introduction

A key feature of public discourse around anti-Western terrorist-extremist speech that occurs in Western liberal democratic states is that it tends to be referred to as purveying, and motivated by, ‘hate’. Examples include criticism of a convicted terrorist-extremist preacher in the United Kingdom who was described as a ‘hate preacher … advocating hate and division’ (Dodd 2016). The Head of the UK Metropolitan Police’s Counter-Terrorism Command condemned the ‘hate’ propagated by two Islamist advocates as having been influential in encouraging people ‘to join terrorist organisations’ (cited in Walker 2017: 534). Media coverage has suggested that the ‘ideology espoused by hate preachers’ has led to terrorist radicalisation (Dearden 2017). In November 2017 then UK Prime Minister Gordon Brown gave a ‘statement on anti-terrorism measures’ in which he described government initiatives to combat terrorism, including regional counter terrorism policing units dedicated to investigating ‘those who recruit terrorists and promote hate’. In discussing the work of mosques in building cohesion and encouraging communities to reject violent extremism, he suggested that ‘the governance of mosques could be strengthened to … challenge those who feed hate’ (Brown 2014).

In Australia, radical Islamist speakers have been described as ‘preaching their hate’ (Olding 2016) and giving ‘hate-filled sermons’ (Rubinsztein-Dunlop 2016). A prominent journalist described the 9/11 attacks and the 7/7 London attacks as being perpetrated by people because ‘they hate us’ (Sheridan, cited in Aly and Green 2010: 274). A scholar has termed terrorist-extremist speech ‘hate propaganda’, and regards efforts against terrorism to be rooted in ‘longstanding norms against hateful incitement directed against groups’ (Tsesis 2017a: 655, 676). Government documents use the terms hatred and radicalisation in combination, as though the latter constitutes an embodiment of the former (eg Australian Government 2018a; Australian Government 2018b; UK Government, 2015b: 11–12). A recent Australian government report identifies the ‘use of vilifying language’ and the use of ‘hateful ideology’ as being ‘signs of radicalisation’ and states that people who are becoming radicalised ‘increasingly use ideological language and “hate” rhetoric’ (Attorney-General’s Department 2015: 6–7). The government urges parents to look out for warning signs that their children are being radicalised, including downloading ‘violent or hate speech videos’. They suggest that deradicalisation of a person who is moving away from violent extremism might involve learning about ‘non-hateful ideologies’ (Attorney-General’s Department 2015: 14, 20).

The use of the terminology of ‘hate’¹ discursively links anti-Western terrorist-extremist speech with the more well-known concept of ‘hate speech’, which is typically understood as speech that is capable of harming its targets in speech-acts that constitute acts of discrimination and marginalisation.² So how similar are anti-Western terrorist-extremist speech and hate speech philosophically? Is it useful to group these two types of speech together as being both motivated and operationalised through ‘hate’, as though their commonalities exceed their differences?

In this article, I will analyse the concepts of anti-Western terrorist-extremist speech and hate speech by investigating the similarities and differences between them along two axes: to whom the speech is addressed, and how harm is occasioned by the speech. The use of this differentiation is heuristic; there is of course overlap between them, as will become clear from the discussion below. However they are helpful in disaggregating elements of the two types of speech, elements that occur simultaneously when the speech is uttered. The philosophical analysis in the paper will be interpolated with empirics from public policy. The analysis will show that the differences between anti-Western terrorist-related speech and hate speech are significant and that the two should not be elided. The primary point of differentiation lies in their mechanisms of harm. I will argue that because the differences between these two types of speech are so significant, using the nomenclature of ‘hate’ to describe anti-Western terrorist-extremist speech creates confusion and elides two very different types of harmful speech. This finding has implications for public policy around appropriate responses to these types of speech – since the mechanisms that render both types of speech harmful and warranting a regulatory response are so different, the choices around regulating them should recognise, and take into account, these differences.

For the purposes of the argument I will draw on empirical examples from two liberal democratic contexts – the United Kingdom and Australia. This is because both countries regulate hate speech, albeit through different mechanisms, and, as I shall show, both are engaged in counter radicalisation policies that attempt to counter the risks of terrorist-extremist speech. The similarities between their systems of government also render them good comparators (Tiffen and Gittins 2004: 1). It is important to note that the analysis in this article is limited to speech in Western liberal democratic states because there are complications in social hierarchies and the capacity of different types of speech to cause harm in other types of societies that require their own analysis, and render a cross-cultural argument unviable.

In the next section I will clarify the type of anti-Western terrorist-extremist speech with which this article is concerned, and what I mean by hate speech. In the third section, I will analyse the similarities and differences between these two types of speech, first focusing on the question of to whom such speech is addressed, which involves analysing the audience and targets for each type of speech, and secondly analysing how harm is occasioned by each type of speech. In section four I will consider two other kinds of extremist speech that are related to this discussion: those who engage in anti-Western terrorist-extremist speech and hate speech simultaneously, and right-wing extremist speech. In the conclusion I will draw out implications of the analytical findings for public policy in relation to these types of speech.

2 Understanding Anti-Western Terrorist-Extremist Speech and Hate Speech

A focus on all types of, or possibilities for, terrorist-extremist speech will not be possible in this article since to do so would require a much larger and more diverse study. The purpose of this section is to clarify the type of anti-Western terrorist-extremist speech with which this article is primarily concerned.

First, this article is not concerned with terrorist incitement, understood as speech that directly incites people to commit terrorism. There are, of course, difficulties in differentiating speech that indoctrinates, exhorts, and encourages terrorism from speech that incites terrorism (Chen 2017: 379; Tsesis 2017a). Indeed, terrorist-extremist speech of many kinds poses a particular difficulty in so far as it can create real and substantive risks to human life even when it falls short of direct incitement. It has been argued that this risk, especially combined with the extraordinary reach of the internet, means that the dissemination of all kinds of terrorist speech presents a heightened risk of harm. This has led some scholars steeped in First Amendment jurisprudence to question its understanding that speech can only be regulated as a form of incitement when the speaker intends to incite imminent lawless action, and when the speech is likely to succeed in doing so (Chen 2017: 384).³ For example, eminent free speech scholar Eric Posner has proposed criminalising accessing materials online that ‘glorify, express support for, or provide encouragement for ISIS’, distributing such materials, or encouraging others to access them (Posner 2015, cited in Chen 2017: 388). Sunstein has proposed expanding the conception of incitement to include speech that constitutes a genuine risk to public safety even in the absence of an imminent threat (Sunstein 2015, cited in Han 2017: 493). These arguments presuppose that the delineation between incitement on the one hand, and encouragement, indoctrination, persuasion, information, inspiration and legitimation on the other, is much more difficult to draw in the context of terrorist activities and in particular in the context of online speech. These scholars and others propose redrawing, or reconsidering (eg Tsesis 2017a: 685–688), incitement and true threat doctrines in the United States to respond to the risks of some terrorist speech.

However, in both the United Kingdom and Australia there exist discrete offences of intentionally inciting and encouraging someone to commit a crime.⁴ Under these provisions the direct incitement of the crime of terrorism is prohibited under criminal law.⁵ International law also recognises the need to prohibit incitement of terrorism specifically, defining incitement of terrorism as speech that combines an intention to incite with a ‘direct call’ to commit terrorism in the context of a likelihood that the speech will cause terrorism to occur (UN Secretary-General, cited in Tsesis 2017a: 678). I will therefore leave aside the phenomenon of speech that directly incites terrorism, for example by explicitly instructing people to commit terrorist offences, because these discrete criminal offences warrant their own detailed examination with respect to their mechanisms of harm.

The type of terrorist-extremist speech that this article seeks to examine in detail expresses an anti-Western ideology and encourages an atmosphere in which everyday, innocent citizens of Western liberal democracies are posited as complicit in and responsible for the inequalities and injustices that are understood to motivate terrorist attacks, and are therefore posited as legitimate targets of violence. Although some terrorists might choose to target members of a particular sub-group as well (police or soldiers, for example) in this type of speech this is still as a component of a broader anti-Western extremist message that implicates all citizens in culpability for the conditions that justify terrorism. Examples of anti-Western terrorist-extremist speech include Bin Ladin’s call to ‘punish the oppressor’ (Bin Ladin 2004), the Book of Jihad’s call to fight all non-believers, and the magazine Inspire which, among other things, describes the 9/11 attacks as ‘blessed raids’.⁶

This speech likely does not constitute direct incitement. Instead, it operates indirectly to ‘create an atmosphere conducive to terrorism and … popularise views which terrorists then exploit’ (UK Government, 2015a, cited in Shepherd 2017: 68) to convince listeners of the justifiability of engaging in, planning and instigating terrorist attacks.⁷ It has been described as a type of ‘indoctrination that influences and directs persons to engage in future terrorist operations’ (Tsesis 2017b: 369). It is intended to ‘inspire’, ‘inform’, ‘recruit’ (Leibowitz 2017: 811), ‘exhort’ (Han 2017: 490), ‘indoctrinate’ (Tsesis 2017a: 653), ‘mentor’ (Tsesis 2017a: 667), ‘influence’ and ‘persuade’ (Redish and Fisher 2017: 570, 577). It involves propaganda and misinformation (Conway 2006), and provides ‘legitimacy and encouragement’ to terrorism and potential terrorists (Cohen-Almagor 2017: 434).

This speech is the object of public policy outside of incitement provisions. UK law criminally prohibits directly and indirectly encouraging or inducing others to prepare or commit terrorist acts.⁸ Australian law criminally prohibits inter alia ‘advocating terrorism’,⁹ urging the overthrow of government,¹⁰ urging violence against groups on the basis inter alia of political opinion both where the violence threatens government¹¹ and where the violence does not,¹² and ‘recruiting for a terrorist organisation’ where to recruit is defined as to ‘induce, incite or encourage’.¹³ These provisions capture speech including, but not limited to (Gelber 2016), the anti-Western terrorist-extremist speech with which I am primarily concerned here.

Similarly, a focus on all understandings of hate speech will also not be possible within the confines of this article. For introductory purposes hate speech is to be understood as speech that is directed at a person to effect discrimination on the ground of their perceived membership of a marginalised group. For example, Parekh defines hate speech as ‘directed against a specified or easily identifiable individual or … a group of individuals based on an arbitrary and normatively irrelevant feature’, which ‘stigmatizes the target group by implicitly or explicitly ascribing to it qualities widely regarded as highly undesirable’, and treating ‘the target group … as an undesirable presence and a legitimate object of hostility’ (Parekh 2012: 40–41). Gelber and McNamara have defined hate speech as ‘expression that is capable of inciting prejudice toward, or effecting marginalisation of, a person or group of people on a specific ground’ (Gelber and McNamara 2015a: 631).¹⁴ Below, I will clarify further aspects of the distinctive operation of hate speech as it relates to the argument I am making.

I move now to investigate the similarities and differences between anti-Western terrorist-extremist speech and hate speech.

3 Similarities and Differences between Anti-Western Terrorist-Extremist Speech, and Hate Speech

In this article, I seek to show that while there are similarities between anti-Western terrorist-extremist speech and hate speech, there are also significant and important differences between them. The first point of discussion is the question of to whom the speech is addressed.

3.1 To Whom Is the Speech Addressed? Direct Audiences and Marginalised Groups

In discussing the audiences of both anti-Western terrorist-extremist speech and hate speech, it can be assumed that in both types of speech the speakers hope or intend that some among the population who become aware of its existence will become interested in the ideas inherent in them. Others are likely to find the ideas abhorrent. In this sense, both types of speech are consistent with the broad idea of public debate – that people share ideas publicly to attempt to persuade others to agree with their point of view, and that listeners decide with whom they will agree and disagree (Sunstein 1995: 21–23). In this sense, speakers of either anti-Western terrorist-extremist speech or hate speech are not doing anything unusual; they are participating in public debate in the marketplace of ideas. However, the fact that this speech is a contribution to persuasive public discourse is not what renders it vulnerable to policy regulation. So what does render these types of speech regulable?

Answering this question requires considering the direct audience of the speech. The policy literature suggests that from a regulatory point of view, the concern with anti-Western terrorist-extremist speech is its capacity to contribute to the radicalisation of those listeners¹⁵ who, as a result of being exposed to it, may become persuaded to adopt a worldview that is more likely to dispose them to commit terrorist attacks (Redish and Fisher 2017: 568; Shepherd 2017: 67–68; UK Government, 2015b: 10–13). Anti-Western terrorist-extremist speech can ‘justify the politics of inflicting terrorism or denigrate the legitimacy of the state in terms of its values of liberal democracy or the protection of all citizens as equals’ (Walker 2017: 524–5). It propagates the kinds of ideas that can play a role in encouraging listeners to view terrorist acts as justifiable, and ordinary citizens as legitimate targets for terrorist acts. This means that the direct audience of anti-Western terrorist-extremist speech can be understood as those who are targeted for radicalisation by the speech; those who become more likely to adopt a point of view that predisposes them to commit or justify acts of terrorist violence as a result of hearing it.

To be clear, I am not suggesting that all anti-Western terrorist-extremist speech is necessarily causal of terrorist violence. There is, on the one hand, recent evidence that several terrorist incidents were carried out by perpetrators who were radicalised, at least in part, by anti-Western terrorist-extremist speech (Han 2017: 487; Leibowitz 2017: 811; Tsesis 2017a: 654, 657). However, it is also important to recognise that a multitude of factors has been identified as being involved in radicalisation (Cherney et al. 2018: 194, 196; Shepherd 2017: 68) and that there is much that is poorly understood, and contested, about radicalisation processes (Dalgaard-Nielsen 2010; Heath-Kelly 2013: 396–402; Richards 2011). Nevertheless, in so far as governments seek to respond to terrorist-extremist speech by monitoring, challenging or suppressing it, the problem they are seeking to address is that of radicalisation; the conversion of some listeners into potential terrorist actors in a range of forms. This, for example, is the rationale for the UK government’s Prevent strategy which was introduced after the 2005 London bombings – to intervene as early as possible in radicalisation processes to prevent young people from becoming engaged in terrorist-related activities, including by ‘challenging extremist (and non-violent) ideas that are also part of terrorist ideology’ (UK Government 2011: 6). The Australian government, similarly, has designed a ‘living safe [sic] together’ program which is designed to help counter the threat of terrorism by inter alia assisting ‘individuals to disengage from violent extremist influences and beliefs’ (Australian Government 2018c).

What can be said of hate speech? I suggest that the same category of direct audience exists as in relation to anti-Western terrorist-extremist speech, although it is a different group of people. Hate speakers engage in speech that promulgates a world view that enables and facilitates discrimination against, and marginalisation of, minorities. The direct audience of such speech is those who may be persuaded to adopt the views being espoused and thereby become more disposed to promoting and practicing discriminatory attitudes and behavior towards marginalised minorities. Indeed, several sub-national hate speech provisions in Australia are directed precisely to counter this effect of hate speech by addressing the fact that the speech incites its audience to ‘hatred’ of a subgroup of people.¹⁶

However, it is clear that a hate speaker also has a second category of audience that does not appear in anti-Western terrorist-extremist speech. This category is the marginalised groups themselves. The literature on hate speech has established that hate speech is directed at harming marginalised targets who are vulnerable to discrimination. Hate speech is targeted at those whom the speaker identifies as belonging to a category the speaker wishes to impugn, whom the speaker has the (often informally derived) authority to impugn, and whom the speaker states to be deserving of verbal harm and exclusion. The potential targets of hate speech are those who are perceived by the speaker to be deserving of discrimination based on a specified, identifiable ground. Importantly, the targets of hate speech are vulnerable to harm by virtue of being members of marginalised groups (Gelber 2019; see also Barendt 2005: 171, 173–4; O’Neill, Rice, and Douglas, 2004: 519).¹⁷ ‘Hate speech’ cannot be promulgated against any potential target whom a speaker wishes to disparage – to understand hate speech in this way would be to broaden it to include virtually anyone about whom a speaker expresses dislike. This would lose its meaning. Rather, ‘hate speech’ is a narrow, specific category of speech that constitutes a discursive manifestation of the marginalisation, discrimination and exclusion suffered by those groups vulnerable to that status. This is borne out by the fact that, even though hate speech laws are almost invariably written in a facially neutral ways, in practice they operate to protect those identities that are marginalised and discriminated against in Western liberal democratic states (Bleich 2018; Gelber 2009: 286) such as racial minorities, and LGBTQI+ folk.

This, therefore, is an important point of differentiation between anti-Western terrorist-extremist speech and hate speech. Both anti-Western terrorist-extremist speech and hate speech are addressed to a direct audience in order to persuade its members of the worldview of the speaker. However, hate speech also has a second category of audience, minorities that are targets for discursive harm by virtue of being members of an institutionally marginalised group. Hate speech targets some people for harm who are vulnerable to that harm due to their membership of identifiable marginalised groups. Terrorist-extremist speech does not have this category of audience.

It might be suggested, in criticism of my argument, that anti-Western terrorist-extremist speech also targets people for harm; it targets everyday citizens for the harms of terrorism and for the fear associated with vulnerability to terrorist violence. I response to this I note, as will be discussed in the next section below, that although anti-Western terrorist-extremist speech engenders fear in everyday citizens of becoming a target for violence, this is not phenomenologically the same as being a member of a marginalised minority that is targeted through speech for discursive discrimination. Members of marginalised groups are vulnerable to the harms of hate speech by virtue of their membership of that group. In anti-Western terrorist-extremist speech the group that is targeted for harm is not a marginalised minority; it is the mainstream majority.

3.2 How Are the Harms Incurred?

This raises the next point of differentiation –how can people be harmed by these two types of speech? There is some conceptual overlap between the discussion that follows, and the previous discussion about at whom the speech is addressed. Nevertheless, parsing the argument out in the way is helpful in clarifying differences.

Let us look first at anti-Western terrorist-extremist speech. In the previous section, I argued that anti-Western terrorist-extremist speech risks the radicalisation of some of its audience. To this extent, the regulation of this speech seeks to reduce the risk of harm from subsequent terrorist attacks, which are many and wide-ranging. The violence promulgated by terrorist acts obviously are massive and include loss of life, short and long term injury, the destruction of property, risks to health, loss of work, and so on. These harms are not the focus of this article, which seeks to understand the harms of anti-Western terrorist-extremist speech.

There are other harms that result from anti-Western terrorist-extremist speech that are related to the mechanisms by which the speech itself (as opposed to a discrete, subsequent action) can harm. In anti-Western terrorist-extremist speech, the harms that regulation seeks to remedy are the intimidation and fear that they produce among the general population; the fear that anyone at any time might themselves become a victim of a terrorist attack through no action of their own and in their everyday places of work and leisure (Redish and Fisher 2017: 567–8).¹⁸ Terrorism is designed to inflict terror on populations who come to understand that they may be targeted for violence. This, indeed, is why terrorism has its own nomenclature, conceptual variation notwithstanding, and is no longer categorised under other criminal alternatives (Di Filippo 2014: 8–11; Schwenkenbecher 2012: 38–39). For example, the definition of terrorism in the United Kingdom explicitly refers to the use or threat of violent action involving serious violence that is inter alia designed to intimidate the public.¹⁹ Similarly, in Australia the definition of terrorism explicitly includes violent actions or threats of violence that are intended to influence the government by intimidation or to intimidate the public.²⁰ Federal government counter-terrorism strategy defines violent extremism as seeking change ‘through fear and intimidation’ (Australian Government 2017).

Some have argued that the fear of terrorism is disproportionate to the objective risk of being the victim of a terrorist attack (eg Braithwaite 2013: 98–99). Indeed, the ongoing fear of terrorism in liberal democratic states has been described as ‘delusional’ (Mueller and Stewart 2012). Nevertheless, other studies have shown that fear of terrorism is based on both tangible events and, importantly, their mediatisation (Nellis and Savage 2012). The fear anti-Western terrorist-extremist speech can produce can manifest in a fear of victimisation, of physical harm and of being unsafe (Nellis 2009). It can also result in a fear of losing civil liberties and a fear of the manipulation of the threat of terrorism (Aly and Green 2010: 274–5; Aly and Balnaves 2007).

The specific harm mechanism that needs to be taken into account when considering anti-Western terrorist-extremist speech is that such speech is designed to instill fear among the population that they risk becoming the victim of a terrorist attack. It contributes to the creation of a generalised sense of anxiety and fear among the population. Although this fear may be disproportionate to the objective risk to people’s everyday safety, it is commensurate with people’s perceptions of their risk, their lack of security and personal safety, and their experiences of vulnerability to harm.

Now let us look at hate speech from the same perspective. How are targets of hate speech harmed? As noted above, hate speech is to be understood as speech that targets individuals for harm on the basis of their perceived membership of a marginalised group. In particular, the hate speech literature recognises that hate speech can harm constitutively, rendering it a discursive mechanism of prejudice and harm (Gelber 2002: Chs 3–4). The fact that hate speech harms constitutively means that the harms of hate speech are not restricted to harms that occur consequently or causally, as a result of the hate speech (Maitra and McGowan 2012, 6; Matsuda et al., 1993). Examples of constitutive harm include subordination, ranking targets as inferior, silencing (Langton 1990, 1993; Maitra 2009; McGowan 2003, 2005; Maitra and McGowan 2012: 7–8), legitimating and justifying discrimination against targets, depriving them of powers, persecuting them, degrading them (Langton 2012: 76–80, 86–89; Maitra and McGowan 2007: 62), and undermining targets’ dignity (Williams 1991: 73; Waldron 2012). Silencing can occur not just in the sense that fear may lead targets not to speak back, but more comprehensively by disabling and rendering ‘unspeakable’ the views of the targets such that even if they do speak, their words are unable to have the effects they wish or intend them to have (Langton 1993).

Relatedly, those who experience constitutive harms can also experience casual harms from hate speech including existential pain, fear, a reduction of associative and communicative freedoms, disempowerment, withdrawal from public physical spaces and from expressive opportunities, direct and indirect silencing, exclusion, and dehumanisation (Gelber and McNamara 2015b: 5–13). These reported harms accord closely with empirical literature alleging that the harms of hate speech include psychological distress, the inducement of fear, and concomitant psychological and physical consequences, ranging from increased heart rate to anxiety, heart and stroke risks, PTSD and physical and mental health risks (Matsuda 1993; Anderson 2013; Meyer 2003; Victorian Health Promotion Foundation, 2012; Vijleveld, Scheepers, and Ellemers, 2012). These experiences can induce an unwillingness to go to public spaces, or a preparedness to restrict one’s life in material ways, such as loss of associative or linguistic liberties, that the target perceives may reduce their risk of being subjected to hate speech in future.

On one level these causal harms sound very similar to the causal harms resulting from the fear produced by anti-Western terrorist-extremist speech. Both groups experience fear, which can lead to the loss of associative liberties, freedom of movement, and so on. But to understand my argument only in this way would be to miss two crucial points. First, hate speech occasions both causal and constitutive harms. Second, the constitutive harms of hate speech explained above occur in relation to hate speech but not in the same way in relation to anti-Western terrorist-extremist speech. Anti-Western terrorist speech does not have the capacity to occasion these kinds of constitutive harms against marginalised subgroups in the manner I have outlined. Third, and relatedly, an understanding of how hate speech harms is linked to the way in which those targeted for harm are identified. Those targeted for harm in hate speech are identified on the basis of their membership of marginalised groups that suffer discrimination. This is what renders the hate speech capable of harming, and these are the people who suffer the constitutive harms of the speech. By contrast, those targeted for harm in anti-Western terrorist-extremist speech are identified randomly on the basis of living in Western societies. Anyone can become a victim, regardless of their ethnic or racial identity, sexuality, religion, age, and so on.

On the issue of how the two types of speech harm, then, they display strong differences. Although both are considered outside of the speech that requires protection in a democracy, their mechanisms of harm are substantively different. The mechanisms of harm of anti-Western terrorist-extremist speech in Western liberal democratic states lie, first, in creating the conditions in which some direct audience members may be persuaded to undertake discrete acts of non-speech harms and, second, in creating fear and intimidation among the population. Importantly the harms of fear and intimidation are occasioned on those who are perceived to belong to the Western mainstream – ordinary citizens going about their everyday lives.

The mechanisms of harm of hate speech lie, first, in creating the conditions in which some direct audience members may decide to undertake discrete acts of non-speech-based harms. Secondly, hate speech also creates fear and intimidation – but not among the entire population; rather, among members of marginalised groups. Additionally, hate speech directly harms those who are targeted by it, because and to the extent to which it is a discursive act of discrimination against identified groups, whose marginalisation renders them vulnerable to the discriminatory speech-act. The harms are incurred not on ordinary, mainstream citizens but on those identities who are marginalised and discriminated against. Hate speech derives its capacity to harm from the fact that it is targeted at these groups. Relatedly the harm is incurred constitutively in the act of speaking itself, in addition to causally due to subsequent, discrete acts of physical harm and discrimination. The speech-based harms of hate speech are, therefore, to be distinguished from those of anti-Western terrorist speech.

The range of similarities and differences identified in this discussion is summarised in Table 1.

Table 1 Comparison between anti-Western terrorist-extremist speech and hate speech

Arguably there are greater differences between these two types of speech than there are similarities. The differences include: 1) the message that the speakers are attempting to persuade the direct audience to adopt differs. The anti-Western terrorist-extremist message is directed against ordinary community members of Western societies. The hate speech message is directed against members of specific, targeted, marginalised groups. 2) There is a difference in who is harmed. In the case of anti-Western terrorist-extremist speech, the harm to the community is the creation among the general community of a fear of potentially becoming a victim of terrorism. By contrast, hate speech does not generate any equivalent generalised fear amongst the whole community of being targeted by hate speech. Rather, it generates fear in members of marginalised communities of being targeted by hate speech. 3) There is a difference in the nature of the harm. Anti-Western terrorist-extremist speech harms by making a whole population experience fear and anxiety. By contrast, hate speech harms constitutively in the utterance of the speech, and is a form of discrimination in and of itself, in addition to making its targets experience fear and anxiety. 4) Finally, there are indirect mechanisms of harm that may arise subsequently to the speech being uttered. Anti-Western terrorist-extremist speech may produce a heightened risk of a subsequent act of terrorist violence. Any such risk is attributable to the population as a whole, or perhaps subgroups identified on the basis of their relationship to the state such as police or soldiers. Hate speech may produced a heightened risk of subsequent acts of discrimination and violence, but only among members of marginalised groups susceptible to that discrimination.

This analysis has rendered clear the differences between these two types of speech. This raises the subsequent question of how to understand other types of terrorist-extremist speech. I will address two here.

4 Other Types of Terrorist-Extremist Speech

I suggested earlier that the discussion would enable further clarification of two other types of terrorist-extremist speech: firstly, anti-Western terrorist extremist speech that also engages in hate speech at the same time, and secondly, right-wing extremist speech.

There are undoubtedly many speakers who engage in anti-Western terrorist-extremist speech and who also engage in hate speech at the same time. That is to say, there are terrorist-extremists who share the motivations of hate speakers, which is to target the marginalised for harm (Leibowitz 2017: 811–812), while simultaneously trying to persuade audiences to adopt terrorist worldviews and creating fear amongst the general population of the likelihood of terrorism. Where this occurs the analysis here suggests that the harms of this type of speech are to be understood cumulatively. This means that both the terrorist-related harms outlined above and the hate speech related harms outlined above can be occasioned by the same speaker. This does not elide the differences between the two types of speech, rather it means that both occur at the same time.

An example of this is a Muslim public speaker, Zakir Naik, whose visa to visit the UK was revoked by the government in 2010. The decision to revoke his visa was based on a list of ‘behaviours unacceptable in the UK’ released in August 2005 by then Home Secretary Charles Clarke after the London bombings of July 2005. Interestingly the list was titled, ‘Tackling Terrorism’, and its opening paragraph situated the press release in the context of the government’s ‘ongoing work to tackle terrorism and extremism’ (Clarke 2005). The stated purpose of the list of unacceptable behaviours was to ‘deal with those who fomented terrorism and sought to provoke others to commit terrorist acts’²¹ – a clear and unequivocal counter-terrorism rationale. However the Secretary of State, Jacqui Smith, also stated that the list had been devised with a view to exclude ‘individuals who engaged in violence or hatred in support of their ideology’. This formulation extended the bases for exclusion to include the category of ‘hatred’. This context and these public statements demonstrate the tendency of government speakers to elide terrorist-extremist speech and hate speech under the aegis of preventing terrorist violence.

The list of unacceptable behaviours did differentiate between different types of views, relevantly listing views which ‘foment, justify or glorify terrorist violence’ as one basis for exclusion and views which ‘foster hatred which might lead to inter-community violence’ as another. It is interesting that the qualification of leading to inter-community violence was added as a requirement to the fomenting hatred category. Within the United Kingdom, the offence of racial hatred rests on ‘threatening, abusive or insulting’ conduct, and there is no requirement that the hatred has to lead to violence for it to be actionable under law, only that the speaker intends, or it is likely, that racial hatred will result (McNamara, 2007: 170). Therefore, domestic law recognises both the harms subsequent to hate speech as being of concern, and the direct harms incurred in the saying of hate speech itself. By contrast, in the exclusion provisions, although both types of speech are formally recognised as unacceptable, it is only the harms subsequent to the speech that are recognised as a sufficient basis on which to exclude a person from the United Kingdom.

Zakir Naik had made a range of public statements, which included justifying both terrorism and anti-Semitism. In informing him of his exclusion from the UK the authorities had stated that it was because he engaged in unacceptable behaviour ‘by making statements that attempt to justify terrorist activity and fostering hatred’,²² thus citing both types of speech as grounds.

The important thing about recognising that speakers can engage in both types of speech is that they are engaging in two types of speech simultaneously. That is to say, the fact that a speaker can engage in both does not imbue the two types of speech with the same mechanisms of harm as one another. They remain two types of speech, differentiated in the ways elucidated above, and not one. These kinds of speakers can encourage and justify terrorism and can also engage in speech that marginalises Jews, homosexuals, women, or people of colour. But this ought not to be taken to mean that all of the speech they engage in has the same targets and mechanisms of harm.

There is no room here to canvass the vast array of laws that can potentially be used to respond to hate speech in countries like the United Kingdom and Australia (see, eg, Brown 2015: 19–48). Nor is there room to summarise the array of preventative counter-radicalisation measures being undertaken, for example under the ‘Prevent’ strategy²³ in the UK, and as part of the ‘Living Safe Together’ strategy in Australia,²⁴ in a bid to stem the risks of radicalisation. The point to be made here is that even where and when the two types of speech intersect, they remain distinct problems with regard to their audiences and mechanisms of harm. Policymakers should therefore take care to differentiate between the two types of speech in addressing them.

The final, related phenomenon can be called right-wing extremist speech. This is an extreme version of hate speech, and therefore its mechanisms of harm are the same as those of hate speech as explicated above. It is speech motivated by and articulating hate speech (Rumney 2003: 143), and the attitudes that underpin it, as the basis for targetting vulnerable minorities for violent harm and explicit exclusion or expulsion. For this reason it is distinct phenomenologically from anti-Western terrorist-extremist speech. In right-wing extremist speech, speakers engage in hate speech while also engaging in speech that justifies extreme violence. This extreme version of hate speech generates fear among marginalized groups that they may become subjected to violence while carrying out their ordinary daily activities. This type of speech is therefore the same as hate speech in terms of the two axes address here: to whom the speech is addressed, and how it harms.

Right-wing extremist speech, and the radicalisation it potentially supports, are extant and growing threats. In August 2018 the former head of counter-terrorism for the UK police warned that the risk of right wing terrorism was being underappreciated (Busby 2018) by those tasked with protecting the community against terrorism, a comment that combines an understanding of right wing extremism with terrorism. The attack that took place in Christchurch, New Zealand, in March 2019 was an example of a right-wing extremist attack. It was described as ‘terrorist’ (eg Roy 2019) due to the random nature of the violence carried out against innocents going about their daily lives and the fear the attack induced among the targeted community, which in this case was Muslims. In these senses it is understandable that these kinds of incidents were described using the ‘terrorist’ nomenclature by politicians and the media. But it is important to note that the messages promoted and adopted by the perpetrator were those of hate speech. The analysis in this article has been focused on differentiating between different types of harm-inducing speech. This analysis suggests that when we discuss harm-inducing speech, great care needs to be taken to differentiate between who is targeted for harm, and how harm can occur to enable a better understanding of the phenomenon at stake.

5 Conclusion

Given the important differences outlined between anti-Western terrorist-extremist speech and hate speech in terms of to whom they are addressed and how they harm, it is important that policymakers do not elide the differences between the two when seeking to develop and implement policy designed to prevent harm. They are similar in so far as both present a risk of persuading audience members to adopt their views. Addressing that risk provides a justification for policy that limits the ability of speakers to reach their direct audience, or punishes them for doing so either by banning their speech or imposing a criminal or civil penalty. However, mitigating or remedying the harms incurred require quite different policies.

Mitigating or remedying the harm to the general population of anti-Western terrorist speech involves reducing the likelihood of fear being experienced. Therefore, the best policy remedies for this fear lie in limiting such speech, or imposing criminal or civil penalties. However, in the realm of mitigating the harms of hate speech, the argument I have presented implies a different policy approach – one that seeks to undo the constitutive harms of discrimination. This implies state support for anti-discrimination measures that are designed to overcome marginalisation. Hate speech remedies require a broad anti-discrimination response that seeks to overcome the marginalisation faced by members of groups that suffer from discrimination.

A similar pattern applies to constructing the best policy remedies for indirect harms that occur subsequently to each type of speech. The range of measures required to prevent the harms of subsequent acts of terrorism is capacious; anti-terrorism policy is vast and multifaceted, including provisions ranging from preventing the financing of terrorism to counter-radicalisation programs. The range of measures required to prevent the harms of subsequent acts of discriminatory violence against the targets of hate speech is very different from this and includes anti-discrimination laws and policies. The implications, therefore, for policy designed to remedy the harms of anti-Western terrorist-extremist speech and hate speech are significant.

The analysis here supports the argument that the persistent elision of these two types of speech provides sufficient grounds to be concerned and that they ought to be understood to be far from synonymous.