1 Introduction: Beyond the Free Speech Clash

Debates about hate speech and the law are dominated by an unavoidable and seemingly interminable conflict between the need to address the harms of hate speech on the one side and the right to freedom of expression on the other. While this chapter does not entirely escape from the so-called free speech dilemma, it questions this confrontational framing of issues and identity groups and seeks to look beyond it. The conflict is often presented in the abstract as though two equal forces slam together and where they meet in the middle is where the objective “balance” between them lies. Indeed, the need to strike “the correct balance” between rights is widely asserted and rarely questioned in discussions on hate speech. But are the two sides always so evenly weighted? Can a fair or neutral balance be struck when interests in regulating speech are presented as “minority rights” and interests in free speech are exalted as universal and fundamental? To what extent does this representation of distinct and inherently opposing interests obscure complex realities? What other issues might this abstract binary distract us from?

This chapter draws on the parliamentary debates that preceded the enactment of the stirring up hatred offences of England and Wales to flag a number of important considerations beyond the free speech dilemma for introducing stirring up gender hatred offences. The aim, therefore, is not to evaluate proposals and criticisms relating to the enactment of stirring up gender hatred offences, but rather to draw insights from research into the existing offences that can be used to contextualise and inform such evaluations. Currently, the use of threatening, abusive or insulting words or behaviour to stir up racial hatred is prohibited under Part III of the Public Order Act 1986 (POA86), and the use of threatening words or behaviour to stir up hatred on grounds of religion or sexual orientation is prohibited under Part IIIA of the same Act. In 2014, the Law Commission conducted a consultation on the possibility of extending these offences through the addition of hatred on grounds of disability and transgender identity but concluded that there was insufficient evidence that such offences were necessary. In this consultation, which also examined other areas of hate crime law, the issue of gender hatred was conspicuously absent. However, in 2018 the Law Commission commenced a new review of hate crime law, with consideration of gender hatred explicitly within its terms of reference. The remit of this review is considerably wider than its predecessor and includes not only whether the stirring up hatred provisions should be extended to include more types of hatred, including gender hatred, but also whether and how they might be reformed, and whether the scattered provisions of anti-hate law should be consolidated within a single Hate Crime Act.Footnote 1

Free speech has been a prominent concern in the Law Commission’s consultation, alongside various other issues primarily relating to the overall coherence of the provisions. These issues are briefly mapped out in relation to the stirring up hatred offences in the first section of this chapter so as to establish the legal context within which an offence of stirring gender hatred would be situated. The relationship between the stirring up hatred provisions and other areas of law that are premised on identity categories—hate crime law and anti-discrimination law—is also considered. Having established the uneven legal terrain of the existing offences, we then turn to their logics. Here, we question how the justifications advanced for the existing stirring up hatred offences might apply to gender hatred offences, including how they might fit within the logics of public order law. Through this analysis, another conflict is revealed: that between the preservation of public order and the pursuit of equality. Ultimately, I argue that this conflict and other underlying tensions need to be exposed and confronted in order for stirring up gender hatred offences to challenge—rather than reproduce—forms of inequality.

2 The Current Stirring up Hatred Offences

The offence of stirring up racial hatred was first enacted by the Race Relations Act (1965), under the subheading of “Public Order”. Although it was criticised for its limited scope and enforcement mechanisms, this act was groundbreaking as the first legal instrument of England and Wales that sought to address racism. An offence under Section 6 of the Race Relations Act (1965) required the publication, distribution or use in a public place of threatening, abusive or insulting words that were both intended and likely to stir up racial hatred.Footnote 2 The Race Relations Act (1976) relocated this offence to Section 5A of the Public Order Act (1936) and removed the requirement for intent to be proved. Then, Part III of the Public Order Act (1986, hereafter “POA86”) introduced intent as an alternative, rather than an additional requirement, to proving that racial hatred was likely to be stirred up and replicated the offence across a range of media, including publications, plays, recordings and broadcasting. I refer to this suite of offences (Sections 18-22 POA86) as the “stirring up racial hatred offences”. The first of these offences reads as follows:

A person who uses threatening, abusive or insulting words or behaviour, or displays any written material which is threatening abusive or insulting, is guilty of an offence if–

  1. a)

    he intends thereby to stir up racial hatred, or

  2. b)

    having regard to all the circumstances racial hatred is likely to be stirred up thereby. (Section 18(1) POA86)

Important features of these offences, which have not changed since 1986, are as follows: (1) they are inchoate (i.e. there is no requirement to prove that racial hatred has actually been stirred up), (2) they concern speech addressed to third parties who might be incited to hate, rather than speech addressed directly to its targets, and (3) they concern speech that targets a group of people defined by reference to their race, rather than individuals. These features are often overlooked or confused but are important for understanding the particular conduct that the offences prohibit and, therefore, how they are distinct from other provisions.

The Racial and Religious Hatred Act (2006) added offences of stirring up religious hatred to the POA86. Such an addition had been discussed for decades. Prior to the twentieth century, attempts to introduce a religious hatred provision tended to be in the form of a proposed amendment to a bill or a private members bill. Conversely, in the twentieth century the government sought to amend Part III of the POA86 so that the racial hatred provisions would be extended to include religious hatred. However, parity between the two offences was rejected by the House of Lords. Race and religion were distinct and should be treated as such, it was argued, because race is immutable, and it makes no sense to persuade someone to change their race, while religion is chosen and proselytising is often viewed as a religious duty (see Wintemute, 2002, pp. 137–138).Footnote 3 Consequently, the Lords imposed changes to the Racial and Religious Hatred Bill, with the effect that the religious hatred offences differ from the racial hatred offences in four ways:

  • The religious hatred offences were enacted within a new Part IIIA of the POA86, instead of being incorporated within Part III.

  • They encompass only words or material that are threatening, and not also that which is abusive or insulting.

  • They require both intent to stir up religious hatred and the likelihood that religious hatred would be stirred up to be proved, rather than either.

  • They include an additional free speech provision.Footnote 4

The religious hatred offences are therefore considerably narrower and more difficult to prosecute than the racial hatred provisions. Indeed, the necessity of proving intent was removed from the racial hatred offences by the Race Relations Act (1976) precisely because it was an excessively difficult criterion to meet (see Scarman, 1975, p. 35).

While different types of hatred might well require different responses (Goodall, 2009, p. 219), the discrepancies between the racial and religious hatred provisions is problematic due to the difficulty in meaningfully distinguishing between the two sentiments in practice and the particular approach that parliament has taken in its attempts to do so. Racial hatred is defined in the current provisions as “hatred against a group of persons defined by reference to colour, race, nationality (including citizenship) or ethnic or national origins” (Section 17, POA86). When the racial hatred offence was first enacted in 1965, in the shadow of the Holocaust, there was concern to ensure that antisemitism would be encompassed. It was claimed that explicit language in the text of the provision to clarify the matter was unnecessary and that Jews would of course, one way or another, be deemed a group of persons defined by the abovementioned criteria (see HC Deb 3 May 1965). The question of the scope of “race” was then considered in the 1983 case of Mandla v Dowell Lee. This case before the House of Lords concerned whether Sikhs could benefit from racial discrimination legislation. The Lords’ nuanced examination of the issue resulted in the establishment of seven criteria for ethnicity, including shared religion, and Lord Fraser explicitly ruled out the necessity of being “drawn from what in biological terms was common racial stock” (pp. 549 and 564). However, in subsequent debates on the stirring up hatred offences (e.g. Ramsay and Mackay, HL Deb 14 Mar 2005; Khan and Harris, HC Deb 21 June 2005), this judgement was reduced to the assertion that Sikhs (and Jews) benefit from racial hatred provisions because they are “mono-ethnic” religions, but other religious groups such as Muslims, Hindus and Christians do not because they are “multi-ethnic”.

There are two important things to note about the distinction between purportedly “mono-ethnic” and multi-ethnic religions (aside from the circularity of using ethnicity to classify religious groups when religion is a criterion for identifying ethnicity). Firstly, this distinction means that hatred against some religious groups can be prosecuted under the wider racial hatred provisions, while identical hatred against other religious groups can only be prosecuted under the narrower religious hatred provisions.Footnote 5 Secondly, the approach to determining the scope of the offences has focused on classifying the groups targeted by the hatred, rather than classifying the hatred. Thus, rather than considering whether a particular group is racialised within the speech at issue, it has been deemed appropriate to instead assess whether that group is racial, as though race is something that can be objectively and apolitically ascertained. This approach was demonstrated not only in relation to the distinction between mono- and multi-ethnic religions but also in a discussion in the Lords (HL Deb 23 October 1986) on the extent to which the stirring up of hatred against gypsies might fall within the scope of the racial hatred offence. Here, it was concluded that only hatred against “genuine” gypsies would be caught. In this way, the law racialises by asserting which groups are and are not “defined by reference to race” and distributes protections accordingly. There is therefore some contradiction within measures that are ostensibly enacted to facilitate integration and inclusion but that reproduce practices of differentiation and classification.

The provisions on stirring up hatred on grounds of sexual orientation (“SO hatred offences”Footnote 6) were enacted by the Criminal Justice and Immigration Act (2008), which amended Part IIIA of the POA86 to include SO hatred alongside religious hatred. Even though it was generally (although not unanimously) agreed that sexual orientation is, like race, an immutable characteristic, the inclusion of the SO hatred offences at the narrower threshold of the religious hatred offences was not questioned. Rather than any particular property of sexual orientation, the establishment of the SO hatred offences at the stricter threshold seems to reflect the extent to which homosexuality was viewed as a controversial and contestable topic. For example, Liberal Democrat MP Evan Harris rationalised the stricter threshold for SO hatred as follows:

it perhaps requires less protection because there is a great deal of sincerely held, often religious, opinion that extends to sexual orientation that does not—generally speaking, in this country, thank goodness—extend to race. (HC Deb 6 May 2008)

There is also a separate free speech provision in relation to the SO hatred offences, to which a second paragraph was added by the Marriage (Same Sex Couples) Act (2013).Footnote 7 Ignoring the fact that no such free speech provision exists in relation to the racial hatred offences, advocates of an SO free speech provision argued that it was necessary for consistency with the religious hatred provisions, while opponents argued that it was a “wrecking amendment” (Bercow, HC Deb 26 Jan 2009) that neutered the SO hatred offences. After a decade of being in force, there has been only one prosecution for stirring up hatred on grounds of sexual orientation (R v Ali, Ahmed and Javed, 2012).

3 Stirring up Gender Hatred

Consideration of the current law on stirring up hatred demonstrates that there is no straightforward option to simply add gender to the existing offences. It would need to be decided if gender hatred offences should be enacted at the wider or the narrower threshold, that is, should intent be a necessary element of the offences and should they encompass language that is abusive and insulting as well as that which is threatening? Also, could an argument for an explicit free speech provision be made in relation to gender hatred offences and to what extent might this undermine them? Furthermore, if the scope of the current offences was replicated for gender hatred, there is a risk that this could lead to problematic classificatory decisions regarding what counts as “a group defined by reference to gender”. For example, might threatening speech against feminists escape from the offences on the basis that not all feminists are women? While these considerations would apply to the addition of any identity category to the stirring up hatred provisions, this section briefly sets out some further considerations and contexts that are specific to gender hatred.

3.1 Gender Hatred and Misogyny

Technically, the stirring up hatred provisions do not protect particular identity groups. Despite being enacted with certain minority groups in mind (Jews, Commonwealth immigrants, Muslims, homosexuals), the stirring up hatred offences address hatred on grounds of characteristics that may be common to either a minority or a majority group (Bell, 2002, p. 181). For example, a person may not stir up hatred against a group defined by reference to race, regardless of whether their race is black, white, British, Pakistani, Welsh and so on. Nevertheless, protection on certain grounds is likely to be far more important to persons who comprise a minority in relation to that characteristic and where there is a history of that minority being oppressed and persecuted (Neller, 2018, p. 77). This is reflected in advocacy of the stirring up hatred provisions that emphasises the need to protect particularly victimised groups. There is therefore some discrepancy between the justification of a provision and the text of the enacted legislation, where the former recognises structural inequalities and the latter does not (Mason, 2014b, p. 166). Thus, there is considerable debate as to whether the incorporation of gender within anti-hate legislation should replicate this veil of neutrality or should refer specifically to misogyny (see Campbell, 2018, p. 34; Mullany & Trickett, 2018, pp. 28–30; Sloan Rainbow, 2017, pp. 63–64). I refer throughout this chapter to prospective “stirring up gender hatred offences”, not to advocate the former approach, but to draw attention to this feature of the existing offences.

However, while several scholars have pointed to the ways in which misogyny, like racial hatred, aims to reinforce the structural dominance of one group over another (see Gill and Mason-Bish, 2013, p. 4), the notion of gender hatred nevertheless “questions our conception of hate crime victims as a minority group” (Mason-Bish, 2014, p. 178). This could be a double-edged sword: on the one hand, it may be more difficult to evoke the required levels of group vulnerability, deservingness and sympathy for half the population (Mason, 2014a, pp. 83–84); on the other hand, misogyny cannot so easily be marginalised or deprioritised as a “minority” interest. Here, understandings of how gender intersects with other identity categories can add important nuance that has often been lacking in parliamentary debates. Femaleness, like other characteristics, should not be viewed as a uniform marker of vulnerability, thereby allowing the existence of successful and powerful women to be touted as evidence that misogyny is not a problem. Equally, the existence of misogyny—evidence of which abounds, from domestic violence and femicide to “pick-up artists” and the “manosphere” (e.g. Ging, 2017)—should not be used to attach fearfulness and risk to femaleness. Rather, gender hatred should be seen as an element that accumulates unevenly, combining with and exacerbating other forms of hatred, disadvantage and exclusion (Williams Crenshaw, 1993; Mason-Bish, 2014, p. 177; Sloan Rainbow, 2017, p. 73). Indeed, Hannah Mason-Bish (2014, p. 173) suggests that attention to gender within the hate crime paradigm might illuminate a need to move away from approaches that focus on membership of one identity group at a time (see also Mason, 2005; Moran & Sharpe, 2004).

3.2 The Absence of Gender in UK Anti-Hate Law

In her examination of how federal hate crime laws were constructed in US Congress, Valerie Jenness (1999) argues that sex was incorporated within US hate crime law relatively easily. In the US, gender was incorporated within federal hate crime law by the Violence Against Women Act 1994, with advocacy of this law presenting violence against women as a civil rights issue. Additionally, violence against women was found to fit within the logic of hate crime law through the notion that acts of gender-based violence are acts of discrimination, the effects of which extend beyond the individual victim and are seen to disadvantage women as a group (ibid., p. 562). While there were debates in the US about how feasible it would be to include violence against women within hate crime law, Jenness states:

There was no debate about the legitimacy of sex as a status in need of protection and legal consideration, in large part because the legitimacy of ‘sex’ as a line of stratification resulting in discrimination had long since gained legal and extralegal currency. (Ibid., p. 563–564)

However, the acceptance of sex as a prohibited ground for discrimination has not led to the category’s smooth incorporation into either hate crime law or hate speech law in England and Wales. Ten years after the Race Relations Act (1965), parliament passed the Sex Discrimination Act (1975). This Act drew upon the racial discrimination provisions but sought “to avoid a number of weaknesses which experience had revealed in the enforcement provisions of the race relations legislation” (Home Office, 1975, p. 11). Then, in the drafting of the Race Relations Act (1976), the government (ibid.) aimed to harmonise the legislation on racial and gender equality but, as noted above, ousted the stirring up racial hatred provision to the Public Order Act (1936). Thus, the commensurability of the categories of race and sex as prohibited grounds of discrimination was well established in UK law, but stirring up hatred was excluded from this frame; there was no discussion of the possibility of including a stirring up hatred provision within the Sex Discrimination Act (1975).

Hate crime law was enacted in the UK with the introduction of racially and religiously aggravated offences under the Crime and Disorder Act (1998, as amended by the Anti-Terrorism, Crime and Security Act, 2001) and the requirement to enhance sentences where an offence is motivated by hostility on grounds of race, religion, sexual orientation, disability or transgender identity under the Criminal Justice Act (2003, as amended by the Legal Aid, Sentencing and Punishment of Offenders Act, 2012). Again, the read-across from anti-discrimination law to anti-hate legislation described by Jenness in the US context has not occurred in England and Wales. Whereas in the US the rationales of hate crime law were used to introduce provisions on violence against women, in the UK provisions on violence against women preceded the enactment of hate crime law. Thus, gender hatred was already afforded “special” protections (Mason-Bish, 2014). Indeed, difficulties in the passage of the stirring up religious and SO hatred provisions demonstrate the need to clearly articulate the legal gap which new offences will fill, along with the difficulty of grasping and effectively communicating the intricate legal specificities of such a gap in an area of law that developed in an uneven and piecemeal fashion (Law Commission, 2020, p. 170).

4 Rationalising the Addition of a New Category

So, without a consensus that characteristics protected from discrimination should be protected from hatred, how might gender hatred be deemed to fit within the logics of the existing stirring up hatred provisions? In Chap. 10 of the 2020 consultation paper, the Law Commission has comprehensively reviewed approaches established by academics, stakeholders and other jurisdictions for determining which categories should be included in anti-hate legislation. The Law Commission concluded that three criteria should be used in the selection of categories for protection: demonstrable need, additional harm and suitability (Law Commission, 2020, p. 208). A greater understanding of these criteria and how they have operated in practice can be gained by looking back at the arguments made around the addition of hatred on grounds of religion and sexual orientation to the stirring up hatred offences.

Firstly, as might be expected, demonstrable need has indeed been essential to the justification of previous expansions. The question of need was at the heart of previous debates on extending the stirring up hatred offences and, as noted above, the Law Commission recommended in 2014 that the offences should not be extended to encompass hatred on grounds of disability or transgender identity due to a lack of evidence that such problems were sufficiently pressing. Secondly, additional harm requires “evidence that criminal targeting based on hostility or prejudice towards the characteristic causes additional harm to the victim, members of the targeted group, and society more widely” (Law Commission, 2020, p. 208). The notion of additional harms reflects the situation in relation to hate crime law, which provides for aggravated offences or enhanced sentences for existing crimes, rather than the situation of the stirring up hatred provisions, which create entirely new offences. However, the last aspect of this criterion that requires harm to be caused to “society more widely” has proved problematic in relation to the stirring up offences: this framing draws an all-too-common distinction between targeted groups and society, suggesting that their interests are separate and that harm caused to some groups somehow does not constitute harm to society. Thus, only those harms capable of being framed as a threat to the interests of legislators, and not only as a “minority rights” issue, have been deemed sufficient to warrant a legislative response. In other words, there must be a degree of “interest convergence” (Bell, 1980), which ultimately ensures that existing power relations are maintained.Footnote 8 Finally, the third criterion, suitability, appears to be so broad that it undermines any point in specifying criteria: any proposed characteristic will undoubtedly face objections on the grounds that it is “unsuitable”, a poor fit with the existing offences, a poor use of criminal justice resources and inconsistent with the rights of others.

These criteria will now be considered in relation to their potential to justify the addition of gender to the stirring up hatred offences by examining how they were deployed in the justification and shaping of the racial and religious hatred offences and in the addition of sexual orientation.

4.1 Fitting Gender Hatred into the Logics of the Racial and Religious Hatred Offences

The stirring up racial hatred offence was presented as embroiled within a conflict between minority interests in the amelioration of hatred on the one side and universal interests in freedom of expression on the other. However, the original racial hatred provisions, their later amendments and the religious hatred provisions were also enacted amidst concerns about “race riots” and the segregation and “ghettoization” of communities. Indeed, the public order framing of the provisions belies concerns not only to protect “victims” but also to prevent their retaliation and to facilitate their integration. For example, in presenting the original Race Relations Bill, Home Secretary Frank Soskice stated:

Basically, the Bill is concerned with public order. Overt acts of discrimination in public places, intensely wounding to the feelings of those against whom these acts are practised, perhaps in the presence of many onlookers, breed the ill will which, as the accumulative result of several such actions over a period, may disturb the peace. (HC Deb 3 May 1965)

This concern with public order and “the peace” provides the interest convergence element for justifying the provisions: racial hatred is a “minority” problem until it threatens the peace of wider (whiter) society. The conflict that carried greater weight in the impetus to legislate was therefore between free speech and public order—the Hobbesian clash between freedom and security—with the sense that the racial hatred legislation was not so much required to protect vulnerable minorities as to mitigate a “clash of civilizations”.

In contrast, gender hatred presents no such risk of inter-communal conflict and violent retaliation; there is no territorial divide between genders, and cumulative acts of misogyny are unlikely to lead women to breach the peace.Footnote 9 Therefore, public order, as it was conceived in relation to racial hatred, does not seem to provide adequate justification for a stirring up gender hatred offence. Indeed, we might question whether it is desirable to include a stirring up gender hatred offence under the auspices of public order law, not least due to the extent to which the protection of public order has been associated with pacification and the protection of the status quo, rather than the promotion of a more just and equal society (Neocleous, 2000, p. 110).

A more helpful precedent for advocating stirring up gender hatred offences can be found in relation to another, albeit interconnected, justification for the racial hatred offences: the symbolic function of legislating. Here, the argument in relation to racial hatred (and later religious hatred) was that enacting stirring up hatred legislation would reassure racialised groups that their concerns were recognised and taken seriously by the government. This rationale was apparent in the extent to which the Race Relations Act (1965) was presented as a counterbalance to measures designed to increase restrictions on immigration from the Commonwealth nations. For example, a 1965 Government White Paper set out the approach to race relations as follows:

This policy has two aspects: one relating to control on the entry of immigrants so that it does not outrun Britain’s capacity to absorb them; the other relating to positive measures designed to secure for the immigrants and their children their rightful place in our society.Footnote 10

This technique of counterbalancing measures that disproportionately disadvantage minorities with the enactment of “minority rights” legislation in order to ameliorate any backlash was also evident decades later in discussions on introducing hate crime legislation. In a letter to the Lord Privy Seal regarding the police powers to stop and search that became Section 60 of the Criminal Justice and Public Order Act (1994), Home Secretary Michael Howard wrote:

The proposed new powers are already being described in the minority press as recreating the discredited ‘sus’ law and there are serious implications for both community relations and to public order if we are unable to present any positive counter-balance. It is therefore important that the Government take the initiative on racial crimes if it is to counteract the belief amongst ethnic minority communities that we do not take their concerns equally seriously. (Cited by Ruddock, HC Deb 12 April 1994)

This fits with the public order rationale that racial hatred provisions will reduce the likelihood of inter-communal conflict (and conflict between racialised communities and the police). However, rather than viewing the law as effective to the extent to which it is enforced against those who breach it, this approach also locates the efficacy of the law in its communication of “a message” to both potential perpetrators and victims. This function was also clearly endorsed by advocates of stirring up religious hatred provisions in the twenty-first century. For example, Labour MP Chris Mullin referred to a religious hatred clause as “a small step, but it will send a signal that we take seriously the kind of abuses to which some of our constituents are being subjected” (HC Deb 26 November 2001; see also Whitaker, HL Deb 27 November 2001), and Labour MP Frank Dobson stated: “If we do not take this opportunity to declare that incitement to hatred of people on the grounds of their religion is wrong, we will declare that we tolerate its continued existence” (HC Deb 21 June 2005).

Thus, while there is no pressing risk of rioting or outbreaks of mass violence stemming from gender hatred, advocates of a stirring up gender hatred offence could easily argue that it is necessary in order to send a clear message that such hatred is wrong and is taken seriously. Such an argument can be supported by evidence from a trial by Nottinghamshire Police of recording incidents of misogyny. While the incidents recorded here predominantly involve behaviour targeted at an individual, and therefore would not qualify as stirring up hatred, research into the trial concludes that the policy was valued for challenging the normalisation of misogynistic behaviour and empowering victims (Mullany & Trickett, 2018, pp. 24–26). It therefore seems reasonable to suggest that the enactment of national stirring up gender hatred offences could have a similar declarative and de-normalising effect. This corresponds with academic work on the moralising effect of criminal law (Mason, 2014a; Bottoms, 2002, pp. 25–26) and the importance of legal recognition for self-esteem (Thompson, 2012). However, it is an unfortunate irony that the discrepancies between the racial and religious hatred provisions currently undermine their capacity to communicate a commitment to equality. While the Law Commission (2020) has proposed equalising these provisions, the full scope of what they communicate and whether a declarative function is sufficient should be carefully considered.

4.2 Fitting Gender Hatred into the Logics of the Sexual Orientation Offences

Like gender hatred, hatred on grounds of sexual orientation does not present a risk to public order in the conventional sense. While gay and queer spaces have materialised and references are made to “the gay community”, sexual orientation lacks the aspects of heredity, segregation and foreignness that have been recognised as producing risks to public order and “the peace”: as with gender hatred, there is little risk that victims of SO hatred will seek violent retribution. So how were the SO hatred offences understood as belonging alongside the racial and religious hatred offences within the Public Order Act 1986?

Firstly, the argument that sexual orientation is commensurate to race was easier to make and more widely accepted than the argument that religion is commensurate to race. In the extensive debates on stirring up religious hatred, it was repeatedly argued, as noted above, that race is an immutable characteristic, whereas religion is not, and that this distinction warrants differential treatment. Advocacy of enacting SO hatred offences was therefore aided by a broad consensus that stirring up hatred against an immutable characteristic is morally wrong. The same substantive argument could be applied in advocacy of enacting gender hatred offences. However, the claim that special protections are required to combat hatred on grounds of immutable characteristics permits no space for understandings of race, sexual orientation or gender as socially constructed or variable phenomena; there is no space for understanding processes of racialisation, sexual discovery, gendering or any kind of fluidity in identification. Such an argument would therefore leverage “the law’s propensity to classify” (Grabham, 2009, p. 186) and replicate assumptions about “true” identities as fixed, distinct and discernible. Thus, if the enactment of stirring up gender hatred offences is to pursue an equality agenda, advocacy based on the parity of gender with identity categories that are afforded protection under the existing offences may first need to challenge how those protected categories are conceptualised.

Arguments about the symbolic importance of legislating also played a role in the passage of the SO hatred offences. However, the need to reassure sexual minorities that hatred against them was taken seriously and to utilise the declarative power of the law to affirm their status in society was far less prominent in the debates preceding the enactment of the SO hatred offences than it was in those preceding the enactment of the religious hatred offences. Presumably, this again reflects the fact that the reassurance of sexual minorities was not deemed important for the maintenance of public order. This calls into question the extent to which advocacy of stirring up gender hatred offences might succeed through a justification based on the need to empower women and affirm the principle of gender equality. While such objectives seem undeniably worthwhile, could gender hatred offences lack the necessary interest convergence to render such legislation not only desirable but also essential in the eyes of legislators?

The question of interest convergence in relation to the SO hatred offences provides insight into how stirring up hatred can be viewed as a public order issue where there are no discernible risks of riots. First, however, it is relevant to consider the interest divergence that produced considerable opposition to the SO hatred offences. Whereas the debates on the racial and religious hatred provisions pitched “minority interests” against the universal value of free speech, but were lent support by public order concerns, opponents of the SO hatred offences pitched “minority interests” in stymying expressions of homophobia against both the universal value of free speech and the particular rights of Christians to express their religious beliefs. Essentially, it was suggested that the message communicated by the SO hatred offences could not affirm the valued status of homosexual and bisexual persons without simultaneously denouncing Christians who expressed their belief that homosexuality is a sin. Thus, while sexual orientation was likened to race as an immutable identity category, it was differentiated through a distinction between being homosexual—which should not be disparaged—and engaging in same-sex sexual activities—which is fair game for moral judgement. This distinction, which is expressed explicitly in the SO hatred free speech provision (Section 29JA, POA86), creates a tension between the right of religious groups not to have hatred stirred up against them on the basis of their religious practices (e.g. wearing religious attire) and the absence of such a right for groups who might have hatred stirred up against them on the basis of their sexual practices.

Yet, in the representation in parliament of a clash between the interests of homosexuals and Christians, the issue was depoliticised as an unfortunate conflict between discrete, fixed and relatively homogeneous identity groups.Footnote 11 However, the side of Christianity was bolstered by associations with both free speech and “tradition”, from which homosexuals were implicitly excluded by their history of being criminalised.Footnote 12 While homosexuals were surely able to stand up for themselves (HL Deb 22 January 2008), persons who might be criminalised by the SO hatred offences were repeatedly described as fearful pensioners, who were merely expressing their “legitimate” and “sincere” beliefs. Lord Waddington expressed this Christian exceptionalism particularly clearly in his statement that “it should be clear in the Bill … that a Christian expressing strong views will not be caught” (HL Deb 3 March 2008). Despite such special dispensation, speakers presented the whole of Christianity as under attack from a new “orthodoxy” that was using law to assert itself (Johnson & Vanderbeck, 2014). In the framing of a conflict between minorities, then, Christianity benefited from a revered history of moral authority and valued traditions (which were never associated with the persecution and oppression of homosexuals), while homosexuals were supposedly representatives of a new orthodoxy that threatened vulnerable Christians. Such a framing can hardly be deemed conducive to a “neutral” balancing of interests.

Despite this, a form of interest convergence was achieved and the SO hatred provisions were passed into law (albeit at the narrow threshold of the religious hatred provisions). The extra ingredient that made this possible was the invocation of a more familiar and less revered type of perpetrator: black rap artists, whose violently homophobic lyrics were cited as evidence in support of the SO hatred offences. In contrast to the “sincere” beliefs of Christians, freedom of expression and sincerity of belief were not raised in relation to the criminalisation of homophobic lyrics (Johnson & Vanderbeck, 2014, p. 163). Significantly, the lyrics were all in Jamaican patois and were therefore recognisably “non-British”: the image of the young, black, Caribbean rapper is opposite in almost every way to the white, British, Christian, middle-class pensioner, and provided a means of locating homophobia outside of what is conventionally considered “respectable” society. The lyrics

emerge from a culture, mostly in the Caribbean, that is deeply homophobic and where violence and murders on such grounds are commonplace. (Turner, HL Deb 9 July 2009)

Thus, hatred was again presented as a public order issue through the representation of a “clash of civilisations”. A “righteous white moralising” (Williams Crenshaw, 2017, p. 57) narrative of good versus evil, wherein respectable, tolerant society is at risk from uncivilised, intolerant others, provided the interest convergence that transformed minority interests into societal interests.

In order for the stirring up SO hatred offences to make sense within public order law, public order needed to be reconceptualised beyond risks of riots and intercommunal violence. However, the achievement of this through the image of the tolerant society that must be protected from foreign vulgarity reinforced the existing privileges and exclusions of society. Homosexuals were “folded into” (Puar, 2007) this society through a narrative that affirmed simultaneously the virtuous status of white Christians and the familiar “problem status” of black youth (Hall et al., 1987, pp. 200–201). The risk that the fight against gender hatred could be similarly co-opted to reinforce such narratives and differentiations should be taken seriously. Comparable evidence of violent misogyny in rap lyrics abounds and tropes about how “they” treat “their” women are a well-worn means of differentiating and subordinating racialised minorities (see Farris, 2017; Spivak, 1988). Questions should therefore be asked about who it is anticipated that stirring up gender hatred offences would criminalise and the extent to which such offences are justified—and may therefore be implemented—on the basis of the supposed deviance of racialised and already-marginalised groups.

5 Conclusion: Pursuing Gender Equality

This chapter sets out the ways in which adding gender hatred offences to the Public Order Act (1986), as it currently stands, could replicate several problematic dynamics and cannot be assumed to represent a straightforward advancement towards equality. The passage of the current stirring up hatred offences suggests that the free speech objection can only be overcome where interests in stymying a particular type of hatred converge with interests in protecting public order. However, the public order that is to be protected has been conceptualised as the status quo, complete with contemporary biases and exclusions. Uninterested in change, a public order framing erases histories of oppression to produce depoliticised “clashes of civilisations” that require appeasement and pacification (Jackson, 2013), rather than the redress of structural biases (Spade, 2011). Protecting public order and pursuing equality therefore appear as incompatible objectives, such that the capacity of stirring up gender hatred offences to challenge the existing order may be curtailed by their enactment under the auspices of public order law. Alternatively, the foregrounding of an equality agenda could be used to reformulate the minority/majority dynamics of the free speech dilemma. By prioritising equality over “order”, ameliorating expressions of hatred can be viewed as a universal interest, and interests in propagating such hatred can be viewed as minority interests. Here, the burden of proof is reversed so that it is the “free speech” interests of those who would stir up hatred that must demonstrate convergence with the universal value of promoting a fair and equal society in order to be taken seriously.

Moreover, care should be taken to look beyond these dramatic, binarising conflicts and to listen to the persistent dissonance that surrounds the law’s approach to identity, not least because such dissonance could significantly distort the desired symbolic function of the offences: to present the stirring up hatred offences as a desirable model to be replicated in regard to gender could be to endorse the ways in which they currently stratify and circumscribe identity. If the offences continue to be stratified, adding gender hatred offences alongside either race in Part III of the POA86 or religion and sexual orientation in Part IIIA would affirm a hierarchy of victims that requires different criteria to be met depending on who the hatred targets. Additionally, the inflexible, taxonomic understanding of identity has enabled identity groups to be pitched against each other and their interests to be treated as zero sum. Advocates of stirring up gender hatred offences should therefore make a case for a clear break from this approach, unless they also advocate legal pronouncements on who does and does not count as a group defined by reference to gender and unless they are willing to risk essentialising representations of women’s interests.

To meaningfully pursue gender equality—for all diverse, fluid and intersecting identity groups who are disadvantaged by gender hatred—a holistic approach is required. In an area of law that addresses hatred against groups, rather individuals, it is essential that basic third-wave feminist precepts do not get drowned out in the clamour for legal recognition. The empowerment of women cannot be complete until prejudices and hatreds on grounds of race, religion, sexual orientation, disability, transgender identity, class and all other such grounds are eradicated alongside sexism and misogyny. This is not to say that the law should never address any one of these issues singularly, but that legislation should be formulated in such a way that recognises rather than negates intersections and fluidities. The law cannot effectively address inequality if the inequalities of the law are not confronted.