• 1 Introduction

  International human rights law (IHRL) was established in the aftermath of World War II with the aim of preventing a reoccurrence of the atrocities committed in the name of fascism. The need to protect the other from rights violations committed by the majority was a central concern of the drafters of IHRL treaties in the post-War period and was recognised as key to preventing the commission of future atrocities. It is no coincidence that the first three IHRL treaties drafted under the auspices of the United Nations addressed genocide, refugees and racial discrimination, respectively.¹x Convention on the Prevention and Punishment of the Crime of Genocide, 9 December 1948, 78 UNTS 277; Convention relating to the Status of Refugees, 14 December 1950, 189 UNTS 150; International Convention on the Elimination of All Forms of Racial Discrimination, 21 December 1965, 660 UNTS 195 (hereinafter ICERD). Yet it is increasingly apparent that the mere existence of IHRL is insufficient to prevent violations of the rights of minorities and that mass atrocities including ethnic cleansing and genocide have not been confined to history.
  This article takes as its starting point that the creation of IHRL treaties and corresponding monitoring mechanisms should be viewed as the first step towards protecting the rights of out-groups. It argues that IHRL monitoring mechanisms must both recognise and seek to address dehumanisation as a root cause of human rights violations, if IHRL is to achieve its purpose. Thus, they must develop the preventative part of their mandates and elaborate a positive obligation for States to disrupt the process of dehumanisation and change societal attitudes towards out-groups.
  The term ‘out-group’ is used in this article as a catch-all term to denote a group bound by a common identity, distinct from that of the majority – in-group – population, that is used as a pretext for the commission of rights violations.²x This includes ethnic, linguistic or religious and national minorities, as recognised in international minority rights law, but also encompasses other groups that may be identified as ‘other’ by the in-group such as sexual minorities (sexual orientation or gender identity), persons with disabilities, migrants, refugees and political minorities. It is not relevant for the purposes of this article if the societal out-group self-identifies on the basis of this identity, as long as the in-group views the out-group as ‘other’ and this has the potential to result in human rights violations. While it is human nature for members of in-groups to stereotype or be prejudiced towards out-groups,³xSee further, I. Tourkochoriti, ‘How Far Should the State Go to Counter Prejudice? On the Legitimacy of the Role of the State to Change Hearts and Minds’, in this special edition. this becomes problematic when it is used to legitimise the ill-treatment of these out-groups, particularly at a societal level. The concept of dehumanisation (and infrahumanisation) has been developed within social psychology, and the associated field of genocide studies, in order to explain the social process that underpins the commission of harm against out-groups. In contrast, concepts such as prejudice, stereotyping and intolerance, which are perhaps more familiar to a legal audience, form just one stage in the process of dehumanisation. The concept of dehumanisation has, further, informed academic literature that has explored how these social processes can be countered or prevented in practice. Thus, the conceptual framework provided by dehumanisation allows this article to expose the social processes that legitimise human rights abuses and reveal how these processes can be countered through the elaboration of a positive State obligation.
  IHRL scholarship has not previously engaged in detail with the insights provided by social psychology, and related fields, in relation to the process of dehumanisation. By adopting the lens of dehumanisation, this article sheds new light on why the IHRL project has not been able to achieve its stated aim of protecting out-groups from rights violations and how the current IHRL framework can be repurposed and reframed to address dehumanisation as the root cause of these rights violations.
  In order to provide a comprehensive picture of current IHRL practice, this article explores four IHRL treaties, and the practice of their respective monitoring bodies in relation to European States. The European Convention on Human Rights (ECHR)⁴x Convention for the Protection of Human Rights and Fundamental Freedoms, 4 November 1950, 213 UNTS 222 (hereinafter ECHR). and the International Covenant on Civil and Political Rights (ICCPR)⁵x International Covenant on Civil and Political Rights, 16 December 1966, 999 UNTS 171 (hereinafter ICCPR). are generally applicable instruments that seek to prevent a range of rights violations, including those most obviously connected to mass atrocities, such as the right to life and the prohibition of torture. In contrast, the Framework Convention for the Protection of National Minorities (FCNM)⁶x Framework Convention for the Protection of National Minorities, 1 February 1995, CETS No 157 (hereinafter FCNM). and the International Convention on the Elimination of all forms of Racial Discrimination (ICERD) are targeted instruments, which seek to protect the rights of out-groups. The mandates and working practices of each instrument’s monitoring mechanism notably impact their ability to address dehumanisation as a root cause of rights violations. The European Court of Human Rights (ECtHR) serves an entirely judicial function, whereas the Advisory Committee to the FCNM (AC-FCNM) is limited to a State reporting process and issuing interpretative guidance in the form of Thematic Commentaries. In contrast, the two treaty bodies, the Committee on the Elimination of Racial Discrimination (CERD) and the Human Rights Committee (HRC), consider State reports, issue interpretative guidance through General Recommendations/Comments and serve a quasi-judicial function. Nonetheless, this article asserts that it is possible for all four mechanisms considered here to establish and elaborate a positive State obligation to counter dehumanisation, albeit to varying degrees.
  Following this introduction, Section 2 of this article draws on literature from Social Psychology and Genocide Studies in order to introduce the concept of dehumanisation and demonstrate how dehumanisation impacts the realisation of rights. Section 3 explores whether it is possible for IHRL to be interpreted to imply a positive State obligation to counter or prevent dehumanisation. Here it is revealed that pre-existing positive State obligations to prevent hate (and other forms of intolerant) speech and to create tolerant societies have the potential to address both implicit and explicit dehumanisation. Sections 4 and 5 analyse whether the current practice of IHRL monitoring mechanisms is sufficient to respond to the threat posed by dehumanisation to the human rights of out-groups. Section 4 focuses on whether IHRL monitoring mechanisms have sufficiently recognised that dehumanisation undermines the realisation of rights. Section 5 draws on Social Psychology, and related fields, to evaluate whether current interpretations of the positive State obligations identified in Section 3 are sufficient to prevent or counter dehumanisation and to ascertain how these interpretations can be strengthened in practice.

• 2 Dehumanisation as a Cause of Rights Violations

  Drawing on research from Social Psychology and Genocide Studies, this section sets out the premise of this article: dehumanisation facilitates and legitimises the violation of the human rights of out-groups. Consequently, it identifies the key characteristics of dehumanisation and establishes the connection between dehumanisation and human rights violations. Examples from the AC-FCNM’s Opinions on States Reports are used to demonstrate the contemporary relevance of dehumanisation as a cause of human rights violations in Europe, specifically in relation to migrants, Muslims and Roma.
  Dehumanisation, broadly defined, involves the categorisation of an out-group as lacking human characteristics. Categorisation does not need to be overt and explicit; it can also be unconscious and implicit.⁷x N. Haslam and S. Loughan, ‘Dehumanization and Infrahumanization’, 64 Annual Review of Psychology 399, at 405 (2014). The process is closely related to phenomena including prejudice, stereotyping, othering and delegitimisation.⁸x D. Bar-Tal, ‘Delegitimization: The Extreme Case of Stereotyping’, in D. Bar-Tal, C.F. Grauman, A. Kruglanski & W. Stroebe (eds.), Stereotyping and Prejudice: Changing Conceptions (1989) 169, at 169. See further, Tourkochoriti, above n. 3. Dehumanised groups may be likened to animals, diseases⁹x Bar-Tal, above n. 8, at 172; N.S. Kteily and E. Bruneau, ‘Darker Demons of Our Nature: The Need to (Re)Focus Attention on Blatant Forms of Dehumanization’, 26 Current Directions in Psychological Science 487, at 488 (2017). or ‘superhuman creatures such as demons, monsters, and satans’.¹⁰x Bar-Tal, above n. 8, at 172. In less blatant forms of dehumanisation, the humanity of the out-group may not be denied outright. Instead the out-group will be categorised as less human than the in-group or as having undesirable characteristics (infrahumanisation).¹¹x Haslam and Loughnan, above n. 7, at 405. Notably, dehumanisation is observed primarily in relation to ‘low-status/disadvantaged targets’.¹²x Kteily and Bruneau, above n. 9, at 488.
  Both blatant and less blatant forms of dehumanisation of out-groups, which fall into the category of ‘low-status/disadvantaged’, can be observed in Europe. For example, in the United Kingdom, the description of migrants as cockroaches in a tabloid newspaper was singled out for criticism by the HRC.¹³x HRC, Concluding observations on the seventh periodic report of the United Kingdom of Great Britain and Northern Ireland, UN doc. CCPR/C/GBR/CO/7 (2015), at para. 10. The AC-FCNM has criticised the portrayal of Roma as ‘inadaptable’,¹⁴x AC-FCNM, Fourth Opinion on the Czech Republic adopted on 16 November 2015 ACFC/OP/IV(2015)004, at para. 53. ‘asocial’,¹⁵xIbid. ‘lazy’¹⁶x AC-FCNM, Fourth Opinion on the Slovak Republic adopted on 3 December 2014 ACFC/OP/IV(2014)004, at para. 35. and ‘criminal’,¹⁷xIbid. all of which suggest infrahumanisation. While these are all human characteristics, infrahumanisation results in these characteristics being attributed to the entire out-group rather than to individuals belonging to the out-group. Specifically, such classifications may lead an out-group to be perceived as being ‘outside the boundaries of the commonly accepted groups, and … thus excluded from the society’.¹⁸x Bar-Tal, above n. 8, at 171. Accordingly, the AC-FCNM has expressed concern that the instrumentalisation of xenophobia by political parties has led to the stratification of society in Cyprus:

  members of the predominant Greek Cypriot linguistic and religious community are viewed as “first class citizens”, EU citizens and wealthy immigrants are regarded as coming second, and Turkish Cypriots, Roma, refugees and asylum-seekers are considered as falling into a third category.¹⁹x AC-FCNM, Fourth Opinion on Cyprus adopted on 18 March 2015 ACFC/OP/IV(2015)001, at para. 34.

  Here, the latter category is viewed as less human than the first two categories and, therefore, as excluded from society. Similar exclusion from society has been observed by the AC-FCNM in relation to Roma, who are frequently subject to ‘social rejection’ and viewed as ‘foreigners’,²⁰x AC-FCNM, Fourth Opinion on Germany adopted on 19 March 2015 ACFC/OP/IV(2015)003, at para. 54. and, in the Netherlands, where younger Muslims have reported feeling that ‘they are seen as members of an ethnic and religious group first and citizens of the Netherlands second’.²¹x AC-FCNM, Third Opinion on the Netherlands adopted on 6 March 2019 ACFC/OP/III(2019)003, at para. 55.
  While for Bar-Tal, dehumanisation can occur in ‘any context of intergroup relations: international, interreligious, intercultural, or interideological’,²²x Bar-Tal, above n. 8, at 179. it appears to require facilitating factors, which support the construction of the out-group as a threat to the in-group.²³xIbid., at 171. Thus, adverse societal conditions,²⁴x S. Opotow, ‘Moral Exclusion and Injustice: An Introduction’, 46 Journal of Social Issues 1, at 4 (1990). a perceived conflict of interests²⁵x J. Vaes, M.P. Paladino, L. Castelli, J.P. Leyens & A. Giovanazzi, ‘On the Behavioral Consequences of Infrahumanization: The Implicit Role of Uniquely Human Emotions in Intergroup Relations’, 85 Journal of Personality and Social Psychology 1016, at 1031 (2003). or the presence of conflict²⁶x Kteily and Bruneau, above n. 9, at 490. have been identified as potential motivating factors behind dehumanisation. Again, this is borne out in Europe, where migrants are currently blamed for a range of societal ills, ranging from ‘the economic situation and high unemployment’,²⁷x AC-FCNM Cyprus, above n. 19, at para. 34. to ‘austerity policies to public health and security’.²⁸x AC-FCNM, Fourth Opinion on Italy adopted on 19 November 2015 ACFC/OP/IV(2015)006, at para. 58. Language that portrays migrants as an ‘alien invasion’,²⁹xIbid. Roma as criminals³⁰x AC-FCNM Slovak Republic, above n. 16, at para. 35. and Muslims as terrorists³¹x AC-FCNM, Fourth Opinion on Denmark adopted on 20 May 2014 ACFC/OP/IV(2014)001, at para. 64. serves to heighten the sense of threat.
  The value of dehumanisation as a concept, for the purposes of this article, derives from the social process it reveals. This perception of threat combined with the denial of the humanity of the victim out-group, allows the in-group to legitimise and rationalise human rights violations. As observed by Haslam and Loughnan ‘[d]ehumanization has also been shown to predict forms of aggression that are perceived as reactive and retaliatory – and often righteous – by the perpetrator’.³²x Haslam and Loughnan, above n. 7, at 415. Specifically, Bar-Tal suggests that dehumanisation reduces prosocial and increases antisocial behaviour towards out-groups.³³xIbid., at 414. The reduction of prosocial behaviour, or collective helping, has the potential to result in discrimination and reduce the mobility of dehumanised out-groups³⁴x Opotow, above n. 24, at 9. on the basis that they are perceived ‘as less worthy of help, forgiveness, and empathy’.³⁵x Haslam and Loughnan, above n. 7, at 416. In the case of migrants in Hungary, the perceived threat posed by immigrants to the State has been linked to a lack of support for the admission of asylum seekers.³⁶x AC-FCNM, Fourth Opinion on Hungary adopted on 25 February 2016 ACFC/OP/IV(2016)003, at paras. 75-76.
  In contrast, antisocial behaviour underpinned by the dehumanisation of the out-group is likely to include acts of aggression and punitive behaviours.³⁷x Haslam and Loughnan, above n. 7, at 415. Goff and others suggest that ‘[d]ehumanization is a method by which individuals and social groups are targeted for cruelty, social degradation, and state-sanctioned violence’.³⁸x P.A. Goff, J.L. Eberhardt, M.J. Williams & M.C. Jackson, ‘Not Yet Human: Implicit Knowledge, Historical Dehumanization, and Contemporary Consequences’, 94 Journal of Personality and Social Psychology 292, at 305 (2008). As a result, dehumanisation might underpin discrimination or punitive criminal justice legislation.³⁹x Kteily and Bruneau, above n. 9, at 490. See further Opotow, above n. 24, at 9. At the extreme end of the scale, Kteily and Bruneau emphasise that

  the depiction of groups such as Africans, Native Americans, Tutsis, the Roma, and Jews (alongside others) as apes, savages, or vermin not only accompanied colonization, slavery, and extermination but facilitated these atrocities.⁴⁰x Kteily and Bruneau, above n. 9, at 487, 490. See also Bar-Tal, above n. 8, at 176.

  Significantly, for Stanton and Bar-Tal, dehumanisation is one stage in a larger process that facilitates the commission of mass atrocities.⁴¹x D. Bar-Tal, ‘Causes and Consequences of Delegitimization: Models of Conflict and Ethnocentrism’, 46 Journal of Social Issues 65 at 66 (1990); G.H. Stanton, ‘Could the Rwandan Genocide Have Been Prevented?’, 6 Journal of Genocide Research 211, at 213-17 (2004).
  In the context of Europe, both violent and non-violent anti-social behaviour has been observed by the AC-FCNM. Specifically, the AC-FCNM has linked physical attacks against Roma in the UK and Italy,⁴²x AC-FCNM, Fourth Opinion on the United Kingdom adopted on 25 May 2016 ACFC/OP/IV(2016)005, at para. 72; AC-FCNM Italy, above n. 28, at para. 59. ‘[t]he heinous fatal stabbing of an Eritrean man in Dresden’⁴³x AC-FCNM Germany, above n. 20, at para. 56. and ‘physical attacks… against local reception centres for immigrants from the Middle East and Africa’⁴⁴x AC-FCNM Italy, above n. 28, at para. 58. in Italy to prejudice in these societies. In Spain ‘persisting discrimination against Roma in all fields of daily life, including in private-law relations such as access to goods and services, employment or housing’ has also been linked to prejudice.⁴⁵x AC-FCNM, Fourth Opinion on Spain adopted on 3 December 2014 ACFC/OP/IV(2014)003, at para. 40. Thus, the dehumanisation of out-groups has the potential to result not only in discrimination and violations of identity rights but also in the denial of core human rights found in the ECHR and ICCPR, such as the right to life and the prohibition of torture.
  However, as the process of dehumanisation allows the in-group to morally legitimise these extreme behaviours, out-groups are frequently not recognised as victims of rights violations or blamed for their own treatment. Opotow explains, ‘[t]hose who are morally excluded are perceived as nonentities, expendable, or undeserving; consequently, harming them appears acceptable, appropriate, or just’.⁴⁶x Opotow, above n. 24, at 1. This can also be observed in Europe, where, for example, the AC-FCNM has expressed concern that in the UK, ‘Gypsies and Travellers are often portrayed as perpetrators and a “criminal” group rather than as victims’,⁴⁷x AC-FCNM United Kingdom, above n. 42, at para. 72. and in Spain that ‘large parts of Spanish society do not recognise as unacceptable the notion that individuals may be insulted or treated less well because of their Roma ethnic origin’.⁴⁸x AC-FCNM Spain, above n. 45, at para. 40. Thus, dehumanisation is how the in-group not only rationalises anti-social behaviour against out-groups but also allows the in-group to deny that out-groups are the bearers of rights in the first place.
  Dehumanisation can occur at the individual, societal and institutional levels. As a result, the potential human rights violations that flow from dehumanisation can be perpetrated by private individuals, acting alone or in concert with others, or by societal institutions, including organs of the State.⁴⁹x Opotow, above n. 24, at 12. Of particular concern in the European context is the institutionalisation of dehumanisation, within societal institutions such as the mainstream media. This has the potential to have more far-reaching consequences than individualised dehumanisation ‘because institutionalized harm occurs on a much larger scale’.⁵⁰xIbid., at 13. However, institutions are able to legitimise harming out-groups only if dehumanisation has been first normalised and accepted at an individual level.⁵¹xIbid. As explained by Opotow, ‘[m]oral exclusion emerges and gains momentum in a recursive cycle in which individuals and society modify each other’.⁵²xIbid., at 11. Whereas, historically, negative portrayals of out-groups have frequently been associated with far-right political parties and extreme elements in Europe, IHRL monitoring mechanisms have expressed concern at the adoption of divisive and intolerant rhetoric by the mainstream press and politicians.⁵³x AC-FCNM, Fourth Opinion on Croatia adopted on 18 November 2015 ACFC/OP/IV(2015)005rev, at para. 41; AC-FCNM United Kingdom, above n. 42, at para. 73; HRC United Kingdom, above n. 13, at para. 10; HRC, Concluding observations on the fourth periodic report of Switzerland, UN doc. CCPR/C/CHE/CO/4 (2017), at para. 20; CERD, Concluding observations on the combined eighteenth to twenty-fifth periodic reports of Hungary, UN doc. CERD/C/HUN/CO/18-25 (2019), at para. 16; CERD, Concluding observations on the combined twenty-third and twenty-fourth periodic reports of Norway, UN doc. CERD/C/NOR/CO/23-24 (2019), at para. 12. This has led dehumanisation to become increasingly acceptable within European societies.⁵⁴x AC-FCNM the Netherlands, above n. 21, at para. 54. As institutional, societal and individual dehumanisation interact and are mutually reinforcing, this has the potential to gradually legitimise more extreme responses to the perceived threat posed by out-groups.
  Consequently, out-groups, including migrants, Muslims and Roma, have been and continue to be dehumanised in Europe. Dehumanisation requires, first, the categorisation, be it explicit or implicit, of the out-group as not human or less human than the in-group; second, the perception that the out-group poses a threat or undermines the interests of the in-group; and, finally, that these factors are used to legitimise interferences with the rights of the out-group. Dehumanisation not only facilitates discrimination and violations of identity rights but, once institutionalised, has the potential to legitimise widespread and coordinated rights violations by organs of the State. In Europe, the dehumanisation of Roma has long been institutionalised in Central Europe and has legitimised discrimination, ethnic cleansing and genocide.⁵⁵x C. Cahn, ‘CERD and Discrimination Against Roma’, in D. Keane and A. Waughray (eds.) Fifty Years of the International Convention on the Elimination of All Forms of Racial Discrimination: A Living Instrument (2017), at 107-108. While the dehumanisation of migrants and Muslims is less ingrained, it is increasingly institutionalised – most clearly, in Hungary.⁵⁶x CERD Hungary, above n. 53, at para. 16; HRC, Concluding observations on the sixth periodic report of Hungary, UN doc. CCPR/C/HUN/CO/6 (2018), at para. 17. If IHRL law is to achieve its purpose, then IHRL monitoring mechanisms must recognise that dehumanisation underpins violations of the rights of these out-groups.

• 3 A Positive State Obligation to Counter Dehumanisation under IHRL

  Dehumanisation is a social process that is reinforced by interactions at an institutional, societal and individual level. As identified earlier, prior to giving rise to rights violations, dehumanisation requires the categorisation of out-groups as not human or less human than the in-group alongside the categorisation of out-groups as a threat to the interests of the in-group. However, it is not clear whether IHRL is equipped to counter the social processes behind dehumanisation. While dehumanisation undermines the realisation of rights, the social processes underpinning it are not necessarily rights violations in themselves, for example, the categorisation of out-groups may be unconscious or unspoken. Thus, it is not enough for IHRL to simply require that States refrain from actively violating rights. If dehumanisation is to be addressed, IHRL must require that States adopt positive measures to disrupt the process of dehumanisation at a societal level. Specifically, they must seek to change societal attitudes towards out-groups.
  The existence of positive obligations derived from IHRL treaties has been clearly established by their monitoring mechanisms and in academic literature.⁵⁷xSee generally, L. Lavrysen, Human Rights in a Positive State: Rethinking the Relationship between Positive and Negative Obligations under the European Convention on Human Rights (2016). Notably, within the UN system, States are under an obligation not only to respect rights by not actively violating them, but also to protect individuals from rights violations perpetrated by private actors. Specifically, the HRC’s General Comment No. 31 establishes that States must ‘take appropriate measures or to exercise due diligence to prevent, punish, investigate or redress the harm caused by such acts by private persons or entities’ [emphasis added].⁵⁸x HRC, General Comment No. 31 on The Nature of the General Legal Obligation Imposed on States Parties to the Covenant, UN doc. CCPR/C/21/Rev.1/Add.13 (2004), at para. 8. If States are ‘to prevent… the harm’, then it follows that they must adopt measures to challenge the root causes of this harm. This interpretation is supported by the UN’s ‘respect, protect, fulfil’ framework: the obligation to fulfil requires that States proactively adopt measures to facilitate ‘the full realisation of rights’.⁵⁹x Lavrysen, above n. 57, at 12; Committee on Economic Social and Cultural Rights, General Comment No. 12 on The right to adequate food (art. 11), UN doc. E/C.12/1999/5 (1999), at para. 15. Similarly, the ECtHR has emphasised that ‘the Convention is intended to guarantee not rights that are theoretical or illusory but rights that are practical and effective’,⁶⁰xAirey v. Ireland, ECHR (1979) Series A. No. 32, at para. 24. in order to legitimise reading positive State obligations into the ECHR.⁶¹x Lavrysen, above n. 57, at 6. As dehumanisation is a root cause of rights violations perpetrated against out-groups, it follows that States must counter or prevent dehumanisation in order to both protect out-groups from private actors and fulfil their human rights obligations by removing obstacles to the realisation of rights.
  However, the substance of States’ positive obligations differs between instruments and between rights. As a result, this section establishes the scope of States’ positive obligation to prevent or counter dehumanisation under the ECHR, FCNM, ICCPR and ICERD. Notably, while an explicit obligation to counter or prevent dehumanisation has not been recognised, all four instruments establish two types of positive obligations that, in combination, have the potential to serve the same purpose: the obligation to prevent intolerant and/or hate speech and the obligation to create tolerant societies.
  The categorisation of out-groups as not human or less human than the in-group is central to the process of dehumanisation. While not all forms of categorisation are explicit or overt, when they are, it is possible for States to intervene by prohibiting forms of expression that categorise the out-group. Article 6(2) FCNM and Article 20(2) ICCPR both establish a positive obligation for States to prevent ‘discrimination, hostility or violence’ motivated by the identity of the out-group. While the ICCPR explicitly requires that incitement to such acts ‘shall be prohibited by law’, the FCNM requires that States ‘take appropriate measures to protect’, allowing space for broader measures at a societal level to address the root causes of hate speech. In contrast, although ICERD does not expressly establish a positive obligation to prevent hate speech, this obligation has been read into the Convention by the CERD.⁶²x CERD, General recommendation No. 35 on Combating racist hate speech, UN doc. CERD/C/GC/35 (2013). Significantly, in elaborating the content of this positive obligation, all three bodies have focused on ensuring the accountability of perpetrators of hate speech, hate crimes or discrimination, through appropriate legal frameworks, prosecution and punishment.⁶³x AC-FCNM, Fifth Opinion on Finland adopted on 27 June 2019 ACFC/OP/V(2019)001, at para. 105; AC-FCNM Hungary, above n. 36, at para. 103; HRC, Concluding observations on the fifth periodic report of the Netherlands, UN doc. CCPR/C/NLD/CO/5 (2018), at para 16(c); HRC, Concluding observations on the third periodic report of Bosnia and Herzegovina, UN doc. CCPR/C/BIH/CO/3 (2017), at para. 22; CERD General recommendation No. 35 (2013), above n. 62, at para. 9; CERD, Concluding observations on the combined second to fifth periodic reports of Serbia, UN doc. CERD/C/SRB/CO/2-5 (2018), at para. 14.
  The ECHR does not contain a provision that expressly requires that States adopt positive measures to give effect to their rights obligations. However, in practice, the ECtHR has pointed to Article 1 ECHR, which requires that States ‘secure to everyone within their jurisdiction the rights and freedoms’ [emphasis added], in conjunction with substantive convention rights as the basis of positive obligations.⁶⁴xAssenov v. Bulgaria, ECHR (1998) Reports 1998-VIII, at para. 102; Vgt Verein Gegen Tierfabriken v. Switzerland, ECHR (2001) Reports 2001-VI, at para. 45; Assanidzé v. Georgia, ECHR (2004) Reports 2004-II, at para. 182. See further, J.-F. Akandji-Kombe, Positive Obligations Under the European Convention on Human Rights: A Guide to the Implementation of the European Convention on Human Rights (2007), at 8-9. While the ECtHR initially focused on legislative change when elaborating the content of States’ positive obligations under the Convention,⁶⁵x M. Bossuyt, International Human Rights Protection: Balanced, Critical, Realistic (2018), at 139. it has increasingly read a variety of positive obligations into almost all of the Convention rights.⁶⁶x Lavrysen, above n. 57, at 60. Thus, although no express obligation to prevent hate speech exists in the ECHR, the ECtHR has established that

  as a matter of principle it may be considered necessary in certain democratic societies to sanction or even prevent all forms of expression which spread, incite, promote or justify hatred based on intolerance.⁶⁷xErbakan v. Turkey, ECHR (2006) Application no. 59405/00, at para. 56. Translation from ECtHR, Factsheet on Hate Speech, February 2020, www.echr.coe.int/Documents/FS_Hate_speech_ENG.pdf.

  Further, in the case of Karaahmed v. Bulgaria, while the ECtHR accepted that an appropriate legal framework existed, the failure to investigate instances of hate speech that incited violence against a religious community amounted to a violation of Article 9 ECHR.⁶⁸xKaraahmed v. Bulgaria, ECHR (2015) Application no. 30587/13, at para. 110. Thus, a positive obligation exists under the ECHR for States to ensure not only that hate speech is prohibited in law but also that this law is implemented in practice.
  These positive obligations have the potential to both protect out-groups from rights violations perpetrated by individuals and prevent additional violations that are legitimised by the explicit dehumanising portrayal of the out-group. However, legal regulation alone is insufficient to address the societal causes of rights violations.⁶⁹xSee further, A. Böcker, ‘Can Non-discrimination Law Change Hearts and Minds?’, in this special edition. Significantly, while all four bodies have focused on the legal prohibition of hate speech, they have also suggested that States are under a positive obligation to attempt to address the societal root causes of such speech. Thus, the AC-FCNM, CERD and HRC have all advised that States should introduce ‘awareness-raising campaigns’, as part of their strategy to address hate and other forms of intolerant speech.⁷⁰x AC-FCNM, Fourth Opinion on Sweden adopted on 22 June 2017 ACFC/OP/IV(2017)004, at para. 54; AC-FCNM, Fourth Opinion on Switzerland adopted on 31 May 2018 ACFC/OP/IV(2018)003, at para. 64; CERD Norway, above n. 53, at para. 12(e); CERD, Concluding observations on the combined twelfth and thirteenth periodic reports of Czechia, UN doc. CERD/C/CZE/CO/12-13 (2019), at para. 11(b); HRC the Netherlands, above n. 63, at para. 16; HRC United Kingdom, above n. 13, at para. 10. Similarly, in Nachova and Others v. Bulgaria, the ECtHR established that

  the authorities must use all available means to combat racism and racist violence, thereby reinforcing democracy’s vision of a society in which diversity is not perceived as a threat but as a source of enrichment [emphasis added].⁷¹xNachova and others v. Bulgaria, ECHR (2005) Reports 2005-VII, at para. 145.

  Thus, it appears that a positive obligation exists under the ECHR, FCNM, ICCPR and ICERD for States to adopt not only legal measures to prohibit hate speech but also non-legal measures to counter the societal attitudes that underpin such hate speech.
  However, the categorisation that underpins dehumanisation is not always articulated through speech. Stereotypes may be so ingrained that they no longer require articulation.⁷²x Goff, Eberhardt, Williams & Jackson, above n. 38, at 304-306. Further, implicit or even unconscious forms of categorisation may legitimise structural discrimination or undermine the realisation of the rights of out-groups.⁷³xIbid., at 305; Kteily and Bruneau, above n. 9, at 492. If implicit categorisation is to be addressed, then measures are required to challenge societal attitudes towards out-groups, more generally. Significantly, under all four instruments, States are also under a general obligation to create tolerant societies. For example, Article 6(1) FCNM requires that States parties ‘encourage a spirit of tolerance and intercultural dialogue’ and ‘take effective measures to promote mutual respect and understanding and co-operation among all persons living on their territory’. In interpreting the purpose of Article 6(1) FCNM, the AC-FCNM has established that States should ‘enhance the majority population’s openness towards diversity’,⁷⁴x AC-FCNM Croatia, above n. 53, at para. 46. promote ‘an overall positive attitude towards diversity and societal integration’,⁷⁵x AC-FCNM, Fourth Opinion on Austria adopted on 14 October 2016 ACFC/OP/IV(2016)007, at para. 34. and equip their populations ‘with the knowledge and understanding to identify and combat intolerance and prejudice’.⁷⁶x AC-FCNM Germany, above n. 20, at para. 60.
  Similarly, Article 2 ICERD requires that States adopt ‘other means of eliminating barriers between races, and … discourage anything which tends to strengthen racial division’ [emphasis added]. Further, Article 7 ICERD requires that States ‘adopt immediate and effective measures, particularly in the fields of teaching, education, culture and information’ [emphasis added]. Notably, the text of Article 7 ICERD not only requires that States seek to change societal attitudes in order to counter existing racial discrimination but also establishes that such measures must be pre-emptive insofar as they must be adopted ‘with a view to combating prejudices which lead to racial discrimination’ [emphasis added]. Through its Concluding Observations on State Reports, the CERD has emphasised that the purpose of these provisions is to ‘combat stereotypes’,⁷⁷x CERD, Concluding observations on the combined fifth to ninth reports of Ireland, UN doc. CERD/C/IRL/CO/5-9 (2019), at para. 24(b); CERD, Concluding observations on the combined twentieth to twenty-second periodic reports of Bulgaria, UN doc. CERD/C/BGR/CO/20-22 (2017), at para. 20(c). ‘promote tolerance and understanding…’⁷⁸x CERD, Concluding observations on the combined twenty-second to twenty-fourth periodic reports of Poland, UN doc. CERD/C/POL/CO/22-24 (2018), at para. 16(c); CERD, Concluding observations on the twenty-third periodic report of Finland, UN doc. CERD/C/FIN/CO/23 (2017), at para. 23. and ‘address the root causes of prejudices’.⁷⁹x CERD, Concluding observations on the combined tenth and eleventh periodic reports of the Republic of Moldova, UN doc. CERD/C/MDA/CO/10-11 (2017), at para. 13(c).
  In contrast to the targeted instruments, the ICCPR does not contain a provision that expressly requires that States adopt positive measures to create tolerant societies. However, Article 2(2) ICCPR explicitly requires that States ‘adopt such laws or other measures as may be necessary to give effect to the rights’ [emphasis added]. This suggests that the drafters foresaw the need for States to adopt a range of positive measures, beyond the adoption of legislation, to give full effect to the provisions of the treaty. The text of the preamble to the ICCPR recognises that it is not enough for rights to be enshrined in law, but that they ‘can only be achieved if conditions are created’, suggesting that societal change may be necessary if these rights are to be enjoyed in practice. Further, Article 2(1) ICCPR requires that States ‘undertake to respect and ensure to all individuals within its territory and subject to its jurisdiction the rights recognized in the present Covenant, without distinction of any kind’. Here ‘ensure’ has been interpreted by the HRC to require proactive steps by the State to prevent human rights violations by private persons,⁸⁰x HRC General Comment No. 31 (2004), above n. 58, at para. 8. an interpretation that also aligns with the ‘respect, protect, fulfil framework’.⁸¹x Although initially developed in relation to socio-economic rights, this framework has subsequently been acknowledged to apply more generally. For example, Committee on Economic Social and Cultural Rights General Comment No. 12, above n. 59, at para. 15; UN General Assembly, Interim Report of the Special Rapporteur on Freedom of Religion or Belief, UN doc. A/71/269 (2016), at para. 23. Through its practice, the HRC has reaffirmed this interpretation by elaborating the content of a positive obligation for States to address intolerance and prejudice. Much like the AC-FCNM and CERD, the HRC has required that States adopt positive measures to ‘promote tolerance and respect for diversity’, ⁸²x HRC Hungary, above n. 56, at para. 18; HRC, Concluding observations on the sixth periodic report of Italy, UN doc. CCPR/C/ITA/CO/6 (2017), at para. 13. ‘respect for human rights’⁸³x HRC, Concluding observations on the seventh periodic report of Sweden, UN doc. CCPR/C/SWE/CO/7 (2016), at para. 17; HRC United Kingdom, above n. 13, at para. 10(b). and to ‘eradicate stereotyping and discrimination’.⁸⁴x HRC Hungary, above n. 56, at para. 18; HRC Sweden, above n. 83, at para. 17.
  Finally, while no explicit obligation to foster tolerant societies exists under the ECHR, Lavrysen has identified a ‘cluster of cases … where the Court has imposed obligations on the State under a variety of Convention provisions to act as a guarantor of pluralism within society’.⁸⁵x Lavrysen, above n. 57, at 94. Specifically, the ECtHR has recognised that States have a positive obligation to address the societal causes of rights violations, insofar as ‘[t]he role of the authorities is not to remove the cause of tension by eliminating pluralism, but to ensure that the competing groups tolerate each other’.⁸⁶xSupreme Holy Council of the Muslim Community v. Bulgaria, ECHR (2004) Application No. 39023/97, at para. 96. Further, in S.A.S. v. France, the ECtHR emphasised that the State ‘has a duty … to promote tolerance’.⁸⁷xS.A.S. v. France, ECHR (2014) Reports 2014, at para. 149. This again suggests that in order to discharge their duties under the ECHR, States are under a positive obligation to adopt non-legal measures to foster tolerant societies.
  Under all four treaties considered here, a positive State obligation to counter or prevent dehumanisation has not been recognised. However, States are under a positive obligation to adopt effective legal and societal measures to prevent hate speech. This has the potential to reduce dehumanisation, by limiting forms of expression that overtly categorise out-groups and by signalling that such categorisation is unacceptable at a societal level. Further, the obligation to take measures to foster tolerant societies establishes an obligation for States to address the root causes of intolerance, including implicit or unconscious forms of categorisation.
  These positive obligations are more clearly articulated in the text of some instruments than others. Further, the mandates of their respective monitoring bodies have impacted their ability to elaborate positive State obligations. Through the State reporting processes as well as the adoption of General Comments or Thematic Commentaries, the AC-FCNM, CERD and HRC have been able to elaborate the measures that States are required to take in order to prevent rights violations. In contrast, as a court, the ECtHR is limited to hearing the facts of the case before it, after the alleged violation has occurred. As a result, it does not serve the same preventative function as the other mechanisms considered here. Nonetheless, the two identified positive obligations allow all four mechanisms to require that States adopt measures to address dehumanisation as a root cause of rights violations, albeit to varying degrees.

• 4 Recognising Dehumanisation as a Cause of Rights Violations in Practice

  This article has identified that under IHRL, States are under a positive obligation to both address hate speech and create tolerant societies. This should allow IHRL monitoring mechanisms to not only require that States counter dehumanisation but to also elaborate the content of this obligation, through their monitoring practice. However, if they are to do so, these mechanisms must first recognise that dehumanisation undermines the realisation of rights. While these mechanisms have not expressly engaged with dehumanisation as a concept, it is possible to ascertain the extent to which they have engaged with the factors that contribute to dehumanisation. This section thus focuses on the extent to which the AC-FCNM, CERD, ECtHR and HRC have expressed concern about the explicit categorisation of out-groups through hate speech and related phenomena such as prejudice, intolerance and stereotyping as well as the explicit portrayal of the out-group as a threat. Further, the extent to which these mechanisms have connected the categorisation of out-groups, including implicit and unconscious categorisation, to other human rights violations reveals whether they recognise dehumanisation to be a cause of rights violations. The practice of these mechanisms is again illustrated with reference to the situation of migrants, Muslims and Roma in Europe.
  As has been illustrated earlier, the AC-FCNM has consistently expressed concern at the treatment of migrants, Muslims and Roma in Europe. It has explicitly identified discourse that stigmatises or stereotypes minorities as problematic⁸⁸x AC-FCNM Czech Republic, above n. 14, at paras. 53-54; AC-FCNM Slovak Republic, above n. 16, at paras. 35-37. and has expressed concern at the increased acceptability of xenophobic discourse⁸⁹x AC-FCNM the Netherlands, above n. 21, at para. 54; AC-FCNM Italy, above n. 28, at para. 58. and the role of mainstream media and politicians ‘in spreading intolerant and racially hostile narratives’.⁹⁰x AC-FCNM United Kingdom, above n. 42, at para. 73; AC-FCNM the Netherlands, above n. 21, at para. 54. In so doing, the AC-FCNM has identified the danger of not only hate speech but also the role that ‘stigmatization and stereotyping’ plays in feeding hostility towards out-groups.⁹¹x AC-FCNM, Fourth Opinion on the Republic of Moldova adopted on 25 May 2016 ACFC/OP/IV(2016)004, at para. 6. Furthermore, the AC-FCNM has recognised that minorities may be scapegoated with the aim of ‘nurturing and instrumentalising xenophobic sentiments in the population’⁹²x AC-FCNM Cyprus, above n. 19, at para. 34. or for political gain.⁹³x AC-FCNM Austria, above n. 75, at para. 31; AC-FCNM Italy, above n. 28, at para. 59.
  Significantly, the AC-FCNM has expressed concern about the impact of these forms of categorisation on broader societal conditions and, specifically, the potential for them to impact out-groups’ enjoyment of rights. Thus, it has highlighted the impact of xenophobia and intolerance on ‘society’s understanding of minority identities and issues’,⁹⁴x AC-FCNM Denmark, above n. 31, at para. 65. ‘a climate in which Muslims and persons with a migration or minority background feel unsafe’⁹⁵x AC-FCNM Germany, above n. 20, at para. 56. as well as ‘an attitude of impunity in which the far right extremists feel emboldened to stage anti-Roma demonstrations and physical attacks’.⁹⁶x AC-FCNM Italy, above n. 28, at para. 59. All of this has been explicitly connected by the AC-FCNM to rights violations, including hate crime,⁹⁷x AC-FCNM Austria, above n. 75, at para. 36. discrimination⁹⁸x AC-FCNM Spain, above n. 45, at para. 40. and access to rights,⁹⁹x AC-FCNM Czech Republic, above n. 14, at para. 57. including freedom of religion or belief.¹⁰⁰x AC-FCNM Spain, above n. 45, at para. 42. Furthermore, the AC-FCNM has identified how xenophobia and the construction of out-groups as a threat combine in order to legitimise rights violations:

  Anti-gypsyism and Islamophobia are reported to be growing in particular on social media, and the negative public debate fed by stereotypes and the construction of enemy images has also led to more frequent violent attacks.¹⁰¹x AC-FCNM Austria, above n. 75, at para. 36.

  Thus, through its practice, the AC-FCNM has identified explicit dehumanisation as a cause of rights violations. However, it tends not to engage with the impact of unconscious or implicit categorisation on the realisation of rights. This is perhaps because it is much easier to identify the resultant rights violations than it is to identify implicit or unconscious categorisation as their underlying cause.
  Although the CERD, like the AC-FCNM, is a targeted mechanism, its approach to dehumanisation aligns more closely with that of the HRC. The Concluding Observations of the CERD and HRC since 2015 reveal that both treaty bodies recognise express forms of categorisation, such as hate speech and intolerant speech, as constituting rights violations, especially when such speech is linked to hate crime. Thus, in relation to Switzerland, the HRC expressed concern ‘about the increasing prevalence of hate speech and acts of hatred against the Muslim, Jewish and Roma communities’.¹⁰²x HRC Switzerland, above n. 53, at para. 20. Similarly, in relation to Poland, the CERD expressed concern at ‘the prevalence of racist hate speech against minority groups … which fuels hatred and intolerance’.¹⁰³x CERD Poland, above n. 78, at para. 15. While both bodies tend to focus on speech that meets the threshold of hate speech, they have also suggested that less explicit forms of categorisation such as stereotyping,¹⁰⁴x CERD Czechia, above n. 70, at para. 11. See also CERD Hungary, above n. 53, at para. 22; HRC Italy, above n. 82, at para. 74. prejudice,¹⁰⁵x CERD Czechia, above n. 70, at para. 11. See also CERD Hungary, above n. 53, at para. 22. stigmatisation¹⁰⁶x HRC Italy, above n. 82, at para. 74. See also CERD, Concluding observations on the combined nineteenth and twentieth periodic reports of Italy, UN doc. CERD/C/ITA/CO/19-20 (2017), at para. 14. and ‘chronic negative portrayal’¹⁰⁷x HRC Sweden, above n. 83, at para. 16. constitute rights violations. The proliferation of hate and intolerant speech by the media and politicians has been singled out as particularly problematic by both treaty bodies.¹⁰⁸x HRC Switzerland, above n. 53, at para. 20; HRC Sweden, above n. 83, at para. 16; CERD Poland, above n. 78, at para. 15; CERD, Concluding observations on the combined twenty-first to twenty-third periodic reports of the United Kingdom of Great Britain and Northern Ireland, UN doc. CERD/C/GBR/CO/21-23 (2016), at para. 15.
  Significantly, both mechanisms have explicitly recognised that hate speech may result in human rights violations, insofar as it legitimises hate crime, violence and ‘acts of intimidation’ towards out-groups.¹⁰⁹x CERD United Kingdom, above n. 108, at para. 15; CERD Poland, above n. 78, at para. 15; HRC Switzerland, above n. 53, at para. 20. Further, the CERD has recognised that hate speech serves the function of excluding out-groups from societal membership, a practice that is recognised by social psychologists as legitimising rights violations: ‘racist hate speech … seeks to degrade the standing of individuals and groups in the estimation of society’.¹¹⁰x CERD General recommendation No. 35 (2013), above n. 62, at para. 10. Yet, with the exception of hate crime and violence, these bodies tend not to explicitly recognise the connection between categorisation, on the one hand, and rights violations, on the other.
  In some Concluding Observations, this link is made implicitly, insofar as the negative portrayal of an out-group is mentioned in the same paragraph as other rights violations. For example, in relation to Sweden, the CERD mentioned ‘stereotypical representation of Muslims’, ‘reports of racist hate crimes and hate speech against Muslim ethno-religious minority groups’, ‘reports of attacks against mosques’ and ‘difficulties … in accessing employment and housing outside of minority-populated areas’.¹¹¹x CERD, Concluding observations on the combined twenty-second and twenty-third periodic reports of Sweden, UN doc. CERD/C/SWE/CO/22-23 (2018), at para. 18. This suggests that the CERD is aware that these are not unrelated issues, but it does not expressly connect the rights violations with the underlying cause. However, in other instances, these mechanisms have failed to make this connection even when societal debates surrounding the adoption of laws that violate rights, such as bans on building minarets¹¹²x HRC Switzerland, above n. 53, at para. 42. or wearing religious clothing,¹¹³x HRC, Concluding observations on the fifth periodic report of France, UN doc. CCPR/C/FRA/CO/5 (2015), at para. 22. have explicitly categorised out-groups.¹¹⁴x See, for example, AC-FCNM, Second opinion on Switzerland adopted on 29 February 2008 ACFC/OP/II(2008)002, at para. 89; European Commission against Racism and Intolerance, ECRI Report on Switzerland (fourth monitoring cycle) adopted on 2 April 2009 CRI(2009)32, at para. 120; A. Vakulenko, Islamic Veiling in Legal Discourse (2012) at 17-18; H. Elver, The Headscarf Controversy: Secularism and Freedom of Religion (2012), at 117-120. This is significant, as only when this link is identified by the CERD and HRC, in the same paragraph of Concluding Observations, do they require that States adopt positive measures to address intolerance as a cause of rights violations.¹¹⁵x CERD Sweden, above n. 111, at paras. 18-19; HRC Italy, above n. 82, at paras. 12-13.
  Further, both Treaty Bodies have struggled to connect rights violations to broader societal conditions, when the categorisation of out-groups is not expressly articulated. Thus, although the CERD has singled out the forced sterilisation of Roma women in the Czech Republic and Slovakia to be of particular concern,¹¹⁶x CERD, Concluding observations on the combined eleventh and twelfth periodic reports of Slovakia, UN doc. CERD/C/SVK/CO/11-12 (2018), at para. 23; CERD, Concluding observations on the combined tenth and eleventh periodic reports of the Czech Republic, UN doc. CERD/C/CZE/CO/10-11 (2015), at para. 22; CERD Czechia, above n. 70, at para. 19. it has not explicitly linked this to the broader moral exclusion and implicit categorisation of Roma in these societies. In contrast, the HRC has connected patterns of societal exclusion, in the form of ‘rejection, exclusion and violence’ faced by Roma in France, to broader rights violations in the form of discrimination in relation to ‘access to health care, social benefits, education and housing which is compounded by forced evictions from settlements and a frequent lack of resettlement solutions’.¹¹⁷x HRC France, above n. 113, at para. 13. However, this is the exception in the monitoring practice of the CERD and HRC. Notably, both bodies have consistently expressed concern about discrimination against migrants, foreigners, ethnic minorities and Roma in the employment, education, housing and healthcare sectors,¹¹⁸x HRC, Concluding observations on the sixth periodic report of Spain, UN doc. CCPR/C/ESP/CO/6 (2015), at para. 9; HRC, Concluding observations on the seventh periodic report of Norway, UN doc. CCPR/C/NOR/CO/7 (2018), at para. 8. the failure to provide sufficient stopping sites for travellers and Roma,¹¹⁹x HRC Switzerland, above n. 53, at para. 50. and forcible evictions.¹²⁰x HRC Italy, above n. 82, at para. 14. Widespread or structural discrimination, as disclosed by these patterns of exclusion, suggest that categorisation is implicit or even unconscious. However, the Treaty Bodies do not identify dehumanisation, if the categorisation of the out-group is not explicitly articulated. This directly impacts whether IHRL mechanisms are able to ask States to address the societal conditions that have facilitated these rights violations.
  In contrast to the mixed approach of the CERD and HRC, the ECtHR not only fails to identify the relevance of dehumanisation to interferences with the rights of migrants, Muslims and Roma, but has also allowed States to explicitly dehumanise migrant and Muslim applicants in order to legitimise interferences with their rights. Under Article 8 ECHR, migrants are frequently accepted by the ECtHR to be a threatening ‘other’, whose rights must give way to immigration control¹²¹xAbdulaziz, Cabales and Balkandali v. the United Kingdom, ECHR (1985) Series A No. 94, at para. 67; Boujlifa v. France, ECHR (1997) Reports 1997-VI, at para. 42; Nunez v. Norway, ECHR (2011) Application No. 55597/09, at para. 71. or the economic well-being of the State.¹²²xBerrehab v. the Netherlands, ECHR (1988) Series A No. 138, at para. 26. This has led scholars to criticise the ECtHR for endorsing the portrayal of migrants as less human than citizens and, therefore, for accepting that migrants do not have the same entitlement to rights as members of the in-group.¹²³x C. Costello, The Human Rights of Migrants and Refugees in European Law (2016), at 128; A. Desmond, ‘The Private Life of Family Matters: Curtailing Human Rights Protection for Migrants under Article 8 of the ECHR?’, 29 European Journal of International Law 261, at 265 (2018). M.-B. Dembour, When Humans Become Migrants: Study of the European Court of Human Rights with an Inter-American Counterpoint (2015), at 504.
  Similarly, the ECtHR has been accused of relying on stereotypes of Islam and of unfavourably comparing Islam and, by extension, Muslims with ‘European values’ and ‘Europeans’, in order to legitimise not finding a violation of the applicant’s rights under Article 9 ECHR.¹²⁴xSee further, K. Henrard, ‘State Obligations to Counter Islamophobia: Comparing Fault Lines in the Supervisory Practice of the HRC/ICCPR, the ECtHR and the AC/FCNM’, in this special edition. See, also, C. Evans, ‘The “Islamic Scarf” in the European Court of Human Rights’, 7 Melbourne Journal of International Law 52, at 65-67 (2006); C.H. Skeet, ‘Orientalism in the European Court of Human Rights’, 14 Religion and Human Rights 31 (2019); K. Henrard, ‘Integration Reasoning at the ECtHR: Challenging the Boundaries of Minorities’ Citizenship’, 38 Netherlands Quarterly of Human Rights 55, at 70-72 (2020). For example, the ECtHR has accepted that the visible presence of Islam poses a threat to the in-group, insofar as it has a ‘proselytising’ effect,¹²⁵xDahlab v. Switzerland, ECHR (2001) Reports 2001-V. challenges the secular foundations of the State¹²⁶xIbid.; Leyla Șahın v. Turkey, ECHR (2005) Reports 2005-XI, at para. 115; Dogru v. France, ECHR (2008) Application No. 27058/05, at para. 72. and undermines societal cohesion.¹²⁷xS.A.S., above n. 87, at para. 141; Osmanoğlu and Kocabaş v. Switzerland, ECHR (2017) Application No. 29086/12, at paras. 96-97. For a more detailed discussion of this point, see generally, Henrard (2020), above n. 124. Thus, by accepting the in-group’s portrayal of Muslim applicants as less human and, therefore, as less deserving of rights, the ECtHR has not only failed to recognise the connection between categorisation and the realisation of rights but has also participated in this process. Nonetheless, in S.A.S. v. France, concerning the so-called French ‘burqa ban’, the ECtHR did express concern at the institutionalisation of dehumanisation, insofar as

  a State which enters into a legislative process of this kind takes the risk of contributing to the consolidation of the stereotypes which affect certain categories of the population and of encouraging the expression of intolerance, when it has a duty, on the contrary, to promote tolerance.¹²⁸xS.A.S., above n. 87, at para. 149.

  However, by deferring to the State through the margin of appreciation in this case, the ECtHR signalled its unwillingness to challenge the majority’s perception of threat and effectively endorsed the dehumanisation of the Muslim applicant.
  In contrast to its treatment of migrants and Muslims, the ECtHR has explicitly recognised that travellers and Roma may be subject to rights violations as a direct result of intolerance and prejudice linked to their identity.¹²⁹xChapman v. United Kingdom, ECHR (2001) Reports 2001-I, at para. 96. Nonetheless, the ECtHR has historically been slow to recognise that violations of the rights of Roma or traveller applicants are enabled by widespread dehumanisation, in the absence of explicit articulations of discriminatory motives.¹³⁰xSee further, R. Sandland, ‘Developing a Jurisprudence of Difference: The Protection of the Human Rights of Travelling Peoples by the European Court of Human Rights’, 8 Human Rights Law Review 475, at 485 (2008). See for example, Buckley v. the United Kingdom, ECHR (1996) Reports 1996-IV; Velikova v. Bulgaria, ECHR (2000) Reports 2000-VI; Anguelova v. Bulgaria, ECHR (2002) Reports 2002-IV. While, more recently, the ECtHR has begun to identify violations of Article 14 ECHR, the prohibition of discrimination, in cases concerning Roma,¹³¹xSee, for example, Nachova, above n. 71; Moldovan and Others v. Romania (No. 2), ECHR (2005) Reports 2005-VII; Stoica v. Romania, ECHR (2008) Application No. 42722/02; D.H. and others v. the Czech Republic, ECHR (2007) Reports 2007-IV. its approach has been inconsistent.¹³²xBalogh v. Hungary, ECHR (2004) Application No. 47940/99; Bekos and Koutropoulos v. Greece, ECHR (2005) Reports 2005-XIII; Cobzaru v. Romania, ECHR (2007) Application No. 8254/99. Thus, in V.C. v Slovakia, concerning the forced sterilisation of a Roma woman, the ECtHR found a violation of Articles 3 and 8, but not Article 14 ECHR as it was unconvinced ‘that it was part of an organised policy or that the hospital staff’s conduct was intentionally racially motivated’.¹³³xV.C. v. Slovakia, ECHR (2011) Application No, 18968/07, at para. 177. Despite a wealth of evidence from the AC-FCNM, CERD and HRC¹³⁴x S.E. Berry, ‘The Siren’s Call? Exploring the Implications of an Additional Protocol to the European Convention on Human Rights on National Minorities’, 23 International Journal on Minority and Group Rights 1, at 30-31 (2016). and a dissenting opinion of Judge Mijovic, which emphasised that this was a specific issue faced by Roma women in Slovakia that had been legitimised by the broader societal context,¹³⁵xV.C. (2011), above n. 133, at dissenting opinion Judge Mijovic. the ECtHR failed to recognise the role played by dehumanisation in this rights violation. By individualising human rights violations committed against Roma, the ECtHR fails to recognise that the implicit categorisation of this out-group by the broader society underpins individual rights violations.
  Thus, while the AC-FCNM, CERD and HRC have recognised that the explicit categorisation of out-groups results in rights violation, the ECtHR has not only failed to recognise the significance of dehumanisation but has also contributed to this process itself. To some extent, this pattern can again be attributed to the mandates and working methods of these mechanisms. The AC-FCNM, CERD and HRC all monitor State reports, a process that allows them to obtain a broad understanding of the situation prevailing in a State and how this pertains to the realisation of the rights of out-groups. This facilitates the identification of dehumanisation at a societal and institutional level. In contrast, as the ECtHR does not monitor State reports, its competence is restricted to the facts of the case before it. As the facts of the case are, inevitably, individualised, this restricts the ECtHR’s ability to identify whether the interference with the applicant’s rights was a result of the dehumanisation of the out-group.
  However, in D.H. and others v. the Czech Republic, the ECtHR demonstrated that it is able to find a violation of Article 14 ECHR, when the facts of the case before it form part of a broader pattern of discrimination or intolerance against an out-group.¹³⁶xD.H. and others (2007), above n. 131, at paras. 191-95. Thus, moving forward, there are opportunities for the ECtHR to strengthen its work in this area. The ECtHR could, for example, solicit information from third party interveners in order to inform its decisions in cases where broader societal intolerance appears to have undermined access to rights, rather than individualising violations that are clearly structural. The ECtHR could also, through obiter dicta, engage with the impact of dehumanisation on the realisation of rights and more clearly establish the scope of States’ positive obligation to counter such dehumanisation. Further, States are frequently afforded a margin of appreciation in cases concerning the rights of persons belonging to minorities.¹³⁷x Berry (2016), above n. 134. The ECtHR could make recognition of this margin of appreciation contingent on the State’s compliance with its positive obligation to foster tolerant societies, in cases where patterns of discrimination or intolerance appear to underpin the interference with the applicant’s rights or where the actions of the State have increased intolerance towards out-groups, as the ECtHR explicitly recognised in SAS v. France.¹³⁸xS.A.S., above n. 87, at para. 149.
  Despite the existence of positive States obligations to prevent hate speech and to foster tolerant societies under all instruments, the IHRL mechanisms explored in this section have yet to fully appreciate the impact of dehumanisation on the realisation of rights. In particular, while they are aware of the connection between hate speech and hate crime, they are much less aware of the impact of categorisation on a wider range of rights, especially when categorisation is implicit or unconscious. This directly impacts whether the recommendations of these mechanisms require that States address dehumanisation as a cause of rights violations. Consequently, if IHRL is to achieve its purpose and protect out-groups from rights violations, then IHRL monitoring bodies must explicitly recognise the root causes of these violations.

• 5 The Content and Scope of a Positive State Obligation to Counter Dehumanisation

  Under IHRL, States are required to both adopt measures to restrict forms of speech that facilitate dehumanisation and address the societal intolerance that underpins dehumanisation. However, in practice, IHRL monitoring mechanisms have yet to fully appreciate the impact that dehumanisation has on the realisation of rights. Drawing on social psychology and related fields, this section analyses whether monitoring mechanisms’ interpretation of the scope of States’ obligations to prevent hate and/or intolerant speech and to create tolerant societies is sufficient to counter dehumanisation as a cause of rights violations. Further, it identifies how the current practice of these mechanisms can be strengthened in order to encourage States to adopt a more robust response to dehumanisation. Significantly, despite the existence of these obligations under the ECHR, in practice, the ECtHR has rarely found a violation in cases where the State has failed to ensure tolerance of out-groups¹³⁹xSee, for example, S.A.S., above n. 87, at dissenting opinion of Judges Nussberger and Jäderblom, at para. 14. Cf. Supreme Holy Council of the Muslim Community, above n. 86, at para. 96. and has not elaborated the content of these obligations. Consequently, this section focuses exclusively on the practice of the AC-FCNM, CERD and HRC. It is revealed that while a comprehensive interpretation of States’ obligation to prevent hate and/or intolerant speech has been developed by these mechanisms, they must elaborate the substance of the obligation to create a tolerant society in more detail, if States are to effectively counter dehumanisation.

  5.1 Preventing Hate and/or Intolerant Speech

  Intolerant speech, including hate speech, explicitly categorises out-groups. Further, it has the potential to reinforce and strengthen the dehumanisation of out-groups, particularly when such expressions are legitimised by those with authority, either expressly, through repetition, or implicitly, by failing to condemn.¹⁴⁰x W.A. Donohue, ‘The Identity Trap: The Language of Genocide’, 31 Journal of Language and Social Psychology 13 (2011). Consequently, reducing the space for intolerant speech in the public sphere has the potential to reduce the dehumanisation of out-groups. The AC-FCNM, CERD and HRC have identified two main components of the positive obligation to prevent intolerant and/or hate speech: first, an obligation to adopt effective laws to prohibit hate speech, and, second, an obligation to regulate the speech of individuals who have the ability to influence public opinion, such as politicians and the media. In developing the content of these obligations, the three mechanisms have provided States with specific guidance and have sought to balance the need to restrict hate speech with the needs of democratic society.
  At the most fundamental level, the AC-FCNM and CERD have stressed that States must ensure that domestic legislation prohibiting hate speech is comprehensive¹⁴¹x CERD General recommendation No. 35 (2013), above n. 62, at para. 9; AC-FCNM Finland, above n. 63, at para. 105; AC-FCNM Germany, above n. 20, at paras. 61 and 70. and is implemented in practice,¹⁴²x CERD Ireland, above n. 77, at para. 22(a); CERD Norway, above n. 53, at para. 14(b). through proper investigation, prosecution and sanctions.¹⁴³x CERD United Kingdom, above n. 108, at para. 16(a); CERD Poland, above n. 78, at paras. 22(e) and 24(e); AC-FCNM Austria, above n. 75, at para. 39; AC-FCNM Finland, above n. 63, at para. 57. See also, HRC United Kingdom, above n. 13, at para. 10(d); HRC Hungary, above n. 56, at para. 18. Beyond this basic standard, the AC-FCNM and CERD have emphasised that law enforcement, prosecutors and the judiciary should receive appropriate and regular training to ensure the effectiveness of these laws.¹⁴⁴x CERD, Concluding observations on the twentieth to twenty-second periodic reports of Greece, UN doc. CERD/C/GRC/CO/20-22 (2016), at para. 17(b); CERD Sweden, above n. 111, at para. 11(c); AC-FCNM Finland, above n. 63, at para. 104; AC-FCNM Cyprus, above n. 19, at para. 39. In order to improve reporting, States should seek to raise out-groups’ awareness of the existence of hate speech legislation¹⁴⁵x AC-FCNM the Netherlands, above n. 21, at para. 66; AC-FCNM, Fourth Opinion on Norway adopted on 13 October 2016 ACFC/OP/IV(2016)008, at para. 58; CERD Greece, above n. 144, at paras. 17(d) and (e); CERD United Kingdom, above n. 108, at para. 16(c). See also, HRC, Concluding observations on the fifth periodic report of Austria, UN doc. CCPR/C/AUT/CO/5 (2015), at para. 16. and improve trust between out-groups and law enforcement authorities,¹⁴⁶x AC-FCNM Finland, above n. 63, at para. 57; AC-FCNM Germany, above n. 20, at para. 77; CERD United Kingdom, above n. 108, at para. 16(c); CERD Greece, above n. 144, at paras. 17(e). including by increasing diversity in the police force.¹⁴⁷x AC-FCNM Finland, above n. 63, at para. 57; AC-FCNM Germany, above n. 20, at para. 77. Thus, in addition to adopting laws to prohibit hate speech, monitoring mechanisms have also required that States ensure these laws are effective in practice.
  The AC-FCNM, CERD and HRC have also required that States adopt measures to reduce the impact of hate or intolerant speech by those who have the potential to influence public opinion. As noted by Donohue, ‘[c]learly, public language matters; it creates a context for how people interact with one another’.¹⁴⁸x Donohue, above n. 140, at 28. Here, the media and politicians have the potential to lead public opinion in a way that individual acts of intolerance do not. Within social psychology and genocide studies, both the media and the politicians have been recognised as playing a key role in the dehumanisation of out-groups¹⁴⁹x Goff, Eberhardt, Williams & Jackson, above n. 38, at 304; Donohue, above n. 140, at 14; 25; Stanton, above n. 41, at 214-15; Opotow, above n. 24, at 13; A. De Buyse, ‘Words of Violence: “Fear Speech,” or How Violent Conflict Escalation Relates to the Freedom of Expression’, 36 Human Rights Quarterly 779, at 779- 81 (2014). and have been implicated in the commission of mass atrocities.¹⁵⁰x Donohue, above n. 140, at 13, 14, 25; Stanton, above n. 41, at 214-15. Thus, it is clear that if IHRL is to achieve its purpose, politicians and the media cannot be permitted to spread hate or intolerance without intervention. However, IHRL – particularly in Europe – is premised on the understanding that democracy and human rights are mutually reinforcing.¹⁵¹x S.E. Berry, ‘Democracy and the Preservation of Minority Identity: Fragmentation within the European Human Rights Framework’, 24 International Journal on Minority and Group Rights 205, at 212-13 (2017). Given the central role that politicians and the media play in ensuring effective democracy, the regulation of hate and intolerant speech is a complex area for IHRL mechanisms to navigate. Significantly, CERD has sought to adopt a nuanced approach and has explicitly acknowledged that whether speech constitutes hate speech is context specific and that factors including ‘the economic, social and political climate’, ‘the position or status of the speaker’, and ‘the reach of the speech’ must be taken into account.¹⁵²x CERD General recommendation No. 35 (2013), above n. 62, at para. 15.
  The AC-FCNM and CERD have both emphasised the need for the authorities to publicly condemn acts of hate speech and related phenomena such as racist propaganda and ‘derogatory and intolerant language’,¹⁵³x AC-FCNM Norway (2016), above n. 145, at para. 53. particularly when perpetrated by politicians or others in the public eye.¹⁵⁴x CERD United Kingdom, above n. 108, at para. 16(d); CERD, Concluding observations on the combined nineteenth to twenty-second periodic reports of Germany, UN doc. CERD/C/DEU/CO/19-22 (2015), at para. 9; AC-FCNM Austria, above n. 75, at para. 40; AC-FCNM Finland, above n. 63, at para. 53. Such condemnation serves to prevent intolerant speech from being normalised in society through the silence or acquiescence of those in authority. Further, if the authorities challenge the categorisation of the out-group, this also has the potential to break the recursive cycle whereby society and the public authorities legitimise the adoption of increasingly extreme measures in response to a perceived threat posed by out-groups. Significantly, both the AC-FCNM and CERD have recognised that such condemnation is a necessary component of ‘promoting a culture of tolerance and respect’,¹⁵⁵x CERD General recommendation No. 35 (2013), above n. 62, at para. 37. See also CERD United Kingdom, above n. 108, at para. 16(d); AC-FCNM United Kingdom, above n. 42, at para. 76; AC-FCNM Spain, above n. 45, at para. 48. thus reaffirming the mutually reinforcing nature of measures to restrict the impact of hate speech and measures to foster tolerance.
  Beyond condemnation, both the AC-FCNM and CERD have recommended that States adopt measures to restrict hate and intolerant speech in political discourse and the media. In relation to hate speech in political discourse, the AC-FCNM has asked States to call ‘on all political parties to refrain from using it’,¹⁵⁶x AC-FCNM United Kingdom, above n. 42, at para. 76. and take steps ‘to combat stereotypes and prejudice in political discourse’.¹⁵⁷x AC-FCNM Czech Republic, above n. 14, para. 57. Similarly, the CERD has asked that the authorities ‘call upon politicians to ensure that their public statements do not contribute to intolerance, stigmatization or incitement to hatred’.¹⁵⁸x CERD Finland, above n. 78, at para. 11(c). Further, the CERD has explicitly emphasised the need to apply legislation on hate speech to politicians and public officials.¹⁵⁹x CERD Italy, above n. 106, at paras. 15(a) and (g); CERD Norway, above n. 53, at paras. 14(a) and (c). See also, AC-FCNM Austria, above n. 75, at para. 38. Significantly, neither mechanism has required that States legislate to prohibit forms of intolerant political speech that do not meet the threshold of hate speech but, nonetheless, have the potential to categorise out-groups.
  This is perhaps where these mechanisms have sought to strike a balance between protecting out-groups from speech that categorises, on the one hand, and allowing space for democratic debate, on the other. The prohibition of forms of speech that the in-group perceives to be legitimate would not only remove the opportunity for such ideas to be contested but would also run the risk of reducing confidence in the democratic process. Once dehumanisation has been institutionalised and/or normalised in political discourse, as it has in many European States, then the in-group is unlikely to respond positively to the condemnation of speech that it perceives to be legitimate.¹⁶⁰x Haslam and Loughnan, above n. 7, at 416. This is because, as noted by Haslam and Loughnan, ‘people actively resist information that challenges them’ and the self-identification of the in-group is based on negative comparisons with the out-group.¹⁶¹xIbid. Consequently, the condemnation of intolerant speech in public discourse is more likely to be effective if it is adopted to prevent rather than counter dehumanisation.
  Both the AC-FCNM and CERD have expressed concern that the portrayal of minorities by the media has the potential to perpetuate intolerance.¹⁶²x CERD Czechia, above n. 70, at para. 11(c); CERD Hungary, above n. 53, at para. 16; AC-FCNM Denmark, above n. 31, at para. 64; AC-FCNM Italy, above n. 28, at para. 58. Media expressions that negatively categorise out-groups but do not constitute hate speech pose particularly complex issues for monitoring mechanisms. Thus, the AC-FCNM and CERD have emphasised the need for media professionals to undertake training to improve reporting on minority groups and diversity.¹⁶³x AC-FCNM Austria, above n. 75, at para. 40; AC-FCNM Germany, above n. 20, at para. 65; AC-FCNM United Kingdom, above n. 42, at para. 76; CERD Czechia, above n. 70, at para. 12(c). While both bodies have recognised the need for some form of press regulation,¹⁶⁴x AC-FCNM United Kingdom, above n. 42, at para. 97; CERD General recommendation No. 35 (2013), above n. 62, at para. 39. the AC-FCNM has emphasised that measures should not impact the freedom or independence of the press.¹⁶⁵x AC-FCNM Germany, above n. 20, at para. 65; AC-FCNM, Fifth Opinion on Denmark adopted on 7 November 2019 ACFC/OP/V(2019)003, at para. 76. In contrast, the CERD has recommended formal regulatory measures, through legislation,¹⁶⁶x CERD Slovakia, above n. 116, at para. 14(a). professional codes of conduct or professional ethics¹⁶⁷x CERD Italy, above n. 106, at para. 15(f). and media supervisory mechanisms.¹⁶⁸x CERD Greece, above n. 144, at para. 17(c). However, some forms of reporting that reinforce negative stereotypes tend to avoid regulation, for example when newspapers report only the ethnicity of minority criminals. In this respect, rather than formal regulation, the CERD has suggested that the ‘[m]edia should avoid referring unnecessarily to race, ethnicity, religion and other group characteristics in a manner that may promote intolerance’.¹⁶⁹x CERD General recommendation No. 35 (2013), above n. 62, at para. 40. Further, the AC-FCNM has called on the authorities in the UK

  to engage with media outlets to promote a more nuanced understanding and reporting of facts to avoid fuelling intolerant and ethnically hostile behaviour while promoting the use of less derogatory language.¹⁷⁰x AC-FCNM United Kingdom, above n. 42, at para. 76.

  While this is a complex area for IHRL monitoring mechanisms to navigate, the AC-FCNM and CERD have clearly sought to balance the need to counter dehumanisation with the requirements of a democratic society. These mechanisms have sought to develop precise and nuanced guidance for States that requires the prohibition of hate speech but also recognises the dangers of the over-regulation of intolerant speech, especially if the negative categorisation of out-groups is already ingrained within society.

  5.2 Creating Tolerant Societies

  While regulation and condemnation have the potential to reduce the influence of speech that categorises, they do not address the societal root causes of dehumanisation nor do they address implicit or unconscious categorisation. In order to address these issues, States must seek to change societal attitudes and create societies that are tolerant of diversity. Significantly, the AC-FCNM, CERD and HRC have recommended that States adopt measures to address the societal root causes of hate speech as well as other forms of intolerant speech through, for example, ‘awareness-raising campaigns’.¹⁷¹x AC-FCNM Norway (2016), above n. 145, at para. 53; AC-FCNM Switzerland, above n. 70, at para. 64; CERD Czechia, above n. 70, at para. 12(b); HRC the Netherlands, above n. 63, at para. 16; HRC Hungary, above n. 56, at para. 18. IHRL also establishes a positive State obligation to create tolerant societies. This section draws on social psychology, and the related fields of interculturalism and genocide studies, in order to analyse whether IHRL mechanisms’ interpretation of States’ positive obligation in this respect is sufficient to improve societal tolerance and, thereby, reduce the dehumanisation that legitimises rights violations.
  During State reporting processes, the AC-FCNM, CERD and HRC have frequently recommended that States ‘promote tolerance and understanding’ or develop integration policies and strategies,¹⁷²x AC-FCNM, Fourth Opinion on Malta adopted on 14 October 2016 ACFC/OP/IV(2016)009, at para. 25; AC-FCNM the Netherlands, above n. 21, at para. 57. without elaborating what this entails in practice. More specific recommendations focus on the adoption of awareness-raising activities and educational campaigns to promote tolerance¹⁷³x AC-FCNM Czech Republic, above n. 14, at para. 57; AC-FCNM Finland (2016), above n. 63, at para. 52; CERD Poland (2018), above n. 78, at para. 16(c); CERD Norway, above n. 53, at para. 12(e); HRC Hungary, above n. 56, at para. 18; HRC Italy, above n. 82, at para. 13; HRC Sweden, above n. 83, at para. 17. or eliminate prejudice and counteract stereotypes.¹⁷⁴x CERD Czechia, above n. 70, at para. 12(b); CERD Norway, above n. 53, at para. 12(e); HRC Hungary, above n. 56, at para. 18. However, these recommendations tend not to elaborate the form such activities should take. The most detailed guidance provided by the AC-FCNM and CERD concerns the design of inclusive education curricula to increase knowledge of the history and culture of out-groups.¹⁷⁵x Art. 12 FCNM; CERD General recommendation No. 35 (2013), above n. 62, at paras. 34-35. For example, the CERD has required that Italy ‘ensure that the school curriculum includes the history of the State party’s colonial past in order to convey the consequences and the continued impact of racially discriminatory policies’.¹⁷⁶x CERD Italy, above n. 106, at para. 26(e). See also, CERD United Kingdom, above n. 108, at para. 35(c).
  From the perspective of social psychology, while inclusive school curricula and public awareness campaigns have the potential to improve societal cohesion,¹⁷⁷x L. Bekemans, ‘Educational Challenges and Perspectives in Multiculturalism vs. Interculturalism: Citizenship. Education for Intercultural Realities’, in M. Barrett (ed.), Interculturalism and Multiculturalism: Similarities and Differences (2013), at 177, 183; G. Bouchard, ‘Interculturalism: What Makes it Distinctive?’, in M. Barrett (ed.), Interculturalism and Multiculturalism: Similarities and Differences (2013), at 107. they are insufficient to create tolerant societies.¹⁷⁸x W.G. Stephan and C.W. Stephan, ‘An Integrated Threat Theory of Prejudice’, in S. Oskamp (ed.), Reducing Prejudice and Discrimination (2000), at 40-41. This is because knowledge alone is unlikely to counter dehumanisation, especially if groups tend not to interact with one another or when such interactions are primarily negative.¹⁷⁹x A. Rodríguez Pérez, N. Delgado Rodríguez, V. Betancor Rodríguez, J.P. Leyens & J. Vaes, ‘Infra-humanization of Outgroups throughout the World. The Role of Similarity, Intergroup Friendship, Knowledge of the Outgroup, and Status’, 27 Anales de Psicología 679, at 685-86 (2011). If the out-group has already been dehumanised, the in-group is likely to view stereotypes as legitimate, not see the need to address intolerance against out-groups, and/or have a vested interest in the negative categorisation of the out-group.¹⁸⁰x Haslam and Loughnan, above n. 7, at 416. In this case, the in-group is unlikely to engage with educational activities that actively challenge their beliefs. Consequently, measures that seek to counter pre-existing dehumanisation are less likely to be successful than measures that seek to prevent dehumanisation in the first place. In order to counter pre-existing dehumanisation, IHRL mechanisms must adopt more robust recommendations that require that States combine educational measures with more wide-ranging measures designed to create tolerant societies.
  Building on Allport’s Contact Theory,¹⁸¹xSee generally, G.W. Allport, The Nature of Prejudice (1954). social psychologists and interculturalists have suggested that increased interactions between different societal groups, with the aim of developing affective ties and intergroup friendship, are necessary to reduce the prejudice that underpins dehumanisation.¹⁸²x Rodríguez Pérez, Delgado Rodríguez, Betancor Rodríguez, Leyens & Vaes, above n. 179, at 681; P. Loobuyck, ‘Towards an Intercultural Sense of Belonging Together: Reflections on the Theoretical and Political Level’, in N. Meer, T. Modood & R. Zapata-Barrero (eds.), Interculturalism and Multiculturalism (2016), at 225; T. Cantle, ‘Interculturalism as a New Narrative for the Era of Globalisation and Super-Diversity’, in M. Barrett (ed.), Interculturalism and Multiculturalism: Similarities and Differences (2013), at 69, 83. See also, Tourkochoriti, above n. 3, for a discussion of how instrumental intergroup interventions can reduce prejudice. Similarly, within genocide studies, Donohue highlights the potential for everyday interactions in the workplace to reduce dehumanisation: ‘[u]nderlying these efforts could also be attempts to initiate dialogue groups that allow individuals from different sides to simply become more comfortable with one another’.¹⁸³x Donohue, above n. 140, at 27. The cohesion strand of interculturalism likewise requires the creation of spaces and opportunities for intercultural interactions to take place¹⁸⁴x Cantle, above n. 182, at 79. and the removal of barriers to successful interactions,¹⁸⁵x M. Barrett, ‘Intercultural Competence: A Distinctive Hallmark of Interculturalism?’, in M. Barrett (ed.), Interculturalism and Multiculturalism: Similarities and Differences (2013), at 157. with the aim of breaking down ‘prejudices, stereotypes and misconceptions of others’ and generating ‘mutual understanding, reciprocal identification, societal trust and solidarity’.¹⁸⁶x Loobuyck, above n. 182, at 230.
  Notably, monitoring mechanisms have been able to interpret States’ positive obligation to create tolerant societies to encompass an obligation to foster affective ties and intergroup friendships. While ‘awareness-raising activities’ have been the default recommendation of monitoring mechanisms, both the AC-FCNM and CERD have occasionally highlighted the need for States to facilitate interactions between different groups in society through ‘trust-building activities’ and the creation of platforms to facilitate dialogue between different groups.¹⁸⁷x AC-FCNM Norway, above n. 140, at para. 53; AC-FCNM Cyprus, above n. 19, at para. 36; CERD Norway, above n. 53, at para. 24; CERD, General recommendation No. 27 on discrimination against Roma, UN doc. A/55/18, annex V (2000), at para. 9. Further, in the educational setting, the AC-FCNM has emphasised the importance of ‘bringing together pupils’¹⁸⁸x AC-FCNM, Fourth Opinion on Estonia adopted on 19 March 2015 ACFC/OP/IV(2015)002, at para. 84. from different backgrounds and organising ‘classes and school activities in ways that facilitate intercultural exchanges and the development of friendships’.¹⁸⁹x AC-FCNM Cyprus, above n. 19, at para. 59. It has also highlighted examples of best practice during the State reporting process, such as the ‘BookEdu’ programme in Copenhagen, which promotes intercultural dialogue in schools.¹⁹⁰x AC-FCNM Denmark (2019), above n. 166, at para. 91. See also, AC-FCNM Austria, above n. 75, at para. 31. These activities have the potential to facilitate meaningful contact between the in-group and out-groups. However, these recommendations are rare, and when they are made, the terminology used, such as ‘trust building exercises’, is vague. Guidance for States is specifically needed because the in-group is likely to resist measures that aim to facilitate intergroup contact if they perceive that the out-group poses a threat to its well-being.¹⁹¹x Stephan and Stephan, above n. 178, at 38; Haslam and Loughnan, above n. 7, at 416. If participation in intercultural activities is not voluntary and does not respect the rights of all members of society, including freedom of association, it risks breeding resentment and becoming counterproductive.
  Further, if measures to facilitate intercultural contact are to be successful, IHRL mechanisms must require that States address structural discrimination. Interculturalists have emphasised that structural discrimination poses barriers to successful interactions, by reducing the opportunities for everyday interactions between the in-group and out-groups.¹⁹²x Barrett, above n. 185, at 157. Desegregation in the educational context has the potential to facilitate interactions between pupils of different backgrounds and presents the opportunity for sources of intergroup tension to be directly addressed.¹⁹³xSee further, A. Kok, ‘The Promotion of Equality and Prevention of Unfair Discrimination Act 4 of 2000: Proposals for Legislative Reform to Promote Equality Through Schools and the Education System’, in this special edition. Significantly, both the AC-FCNM and CERD have consistently highlighted the need for States to adopt a range of measures to tackle societal segregation, specifically in relation to Roma in the context of education, employment and housing, under rights relating to non-discrimination, equality and education.¹⁹⁴x CERD Germany, above n. 154, at para. 13(c); CERD Czechia, above n. 70, at para. 18; CERD, Concluding observations on the combined twelfth and thirteenth periodic reports of Bosnia and Herzegovina, UN doc. CERD/C/BIH/CO/12-13 (2018), at para. 23(b); CERD Slovakia, above n. 116, at para. 12(b); CERD Serbia, above n. 63, at para. 23; CERD Finland, above n. 78, at para. 13. Significantly, neither body has recognised the central role played by measures to counter segregation and structural discrimination in fostering societal tolerance. Nonetheless, the elaboration of States’ obligations in this respect has the potential to facilitate the creation of tolerant societies.
  However, there is a danger that measures intended to remove structural disadvantage by creating mixed neighbourhoods, for example, run the risk of serving an assimilatory function and violating the rights of out-groups.¹⁹⁵x S.E. Berry, ‘Aligning Interculturalism with International Human Rights Law: “Living Together” without Assimilation’, 18 Human Rights Law Review 441, at 459 (2018). Notably, this appears to have been anticipated by the AC-FCNM and CERD, insofar as they have emphasised that States must consult with out-groups in the development of policies or strategies that pertain to their own social inclusion.¹⁹⁶x AC-FCNM Austria, above n. 75, at para. 34; AC-FCNM Spain, above n. 45, at para. 50; CERD Poland, above n. 78, at para. 22; CERD General recommendation No. 27 (2000), above n. 187, at para. 9.
  Finally, Haslem and Loughnan suggest that a ‘way to reduce dehumanization is to promote a common or superordinate identity, thereby emphasizing the similarities and shared fate of different subgroups and de-emphasizing their boundaries’.¹⁹⁷x Haslam and Loughnan, above n. 7, at 416. As the creation of a common identity aims to reduce prejudice, it is possible for this to fall within States’ positive obligation to create tolerant societies. However, national identity is often a politically sensitive subject, especially if the in-group believes that the out-group poses a threat to its cultural existence.¹⁹⁸x G. Bouchard, ‘What is Interculturalism?’, 56 McGill Law Journal 435, at 445 (2011). Thus, official attempts to create an inclusive identity may be viewed as a threat to national identity, heighten the sense of threat that underpins dehumanisation and may even be counterproductive.¹⁹⁹x Haslam and Loughnan, above n. 7, at 416. It is, then, perhaps unsurprising that IHRL mechanisms have rarely recommended that States seek to create an inclusive identity, with the exception of the AC-FCNM in relation to Moldova. Here, the AC-FCNM recommended that the authorities

  implement a long-term strategy for the formation of a civic identity that is inclusive and firmly based on respect for ethnic and linguistic diversity as an integral part of Moldovan society [emphasis added].²⁰⁰x AC-FCNM Moldova, above n. 91, at 1.

  It is, however, possible for IHRL mechanisms to recommend less divisive measures that, nonetheless, have the potential to facilitate the creation of an inclusive superordinate identity. Here, interculturalist Zapata-Barrero suggests ‘redesigning institutions and policies in all fields to treat diversity as a potential resource and a public good, and not as a nuisance to be contained’.²⁰¹x R. Zapata-Barrero, ‘Theorising Intercultural Citizenship’, in N. Meer, T. Modood & R. Zapata-Barrero (eds.), Interculturalism and Multiculturalism (2016), at 64. Notably, both the AC-FCNM and CERD have recognised the potential for the public authorities to develop a ‘positive political culture’²⁰²x CERD General recommendation No. 27, above n. 187, at para. 11. and send a positive message about diversity.²⁰³x AC-FCNM Moldova, above n. 91, at para. 38; AC-FCNM Slovak Republic, above n. 16, at para. 37. Further, the AC-FCNM has recommended that States seek to create a sense of societal belonging for all groups²⁰⁴x AC-FCNM, Third Opinion on Poland adopted on 28 November 2013 ACFC/OP/III(2013)004, at 2; AC-FCNM, Fourth Opinion on Bosnia and Herzegovina adopted on 9 November 2017 ACFC/OP/IV(2017)007, at para. 73. and adopt strategies to ensure the integration of society as a whole, rather than focusing only on the integration of out-groups.²⁰⁵x AC-FCNM Italy, above n. 28, at para. 95; AC-FCNM Malta, above n. 172, at para. 25. It has also stressed the importance of an inclusive public discourse for both negotiating space for diversity within society and ensuring that such negotiations do not become a source of conflict.²⁰⁶x AC-FCNM Austria, above n. 75, at para. 34. Thus, a number of the measures suggested by the AC-FCNM have the potential to create an inclusive superordinate national or civic identity, but its measured approach has the potential to offset the divisiveness of the subject.
  The AC-FCNM and CERD have recognised that a range of measures are required to create tolerant societies. While their recommendations broadly correspond with those identified within social psychology and interculturalism, these recommendations must be strengthened if dehumanisation is to be successfully countered. Specifically, monitoring mechanisms must explicitly recognise that educational measures alone are insufficient to foster a tolerant society. Recommendations should regularly emphasise that education must be bolstered with measures to facilitate intercultural dialogue alongside measures to create an inclusive superordinate identity. Barriers to tolerance, such as structural discrimination, must be removed as part of these efforts.
  Significantly, the AC-FCNM has elaborated the content of States’ positive obligation to foster a tolerant society in the most detail. This can be attributed to the existence of an explicit obligation in Article 6(1) FCNM but also to the fact that its Opinions on State Reports are far more detailed than Treaty Bodies’ Concluding Observations. However, if dehumanisation as a root cause of rights violations is to be successfully addressed and IHRL is to achieve its purpose, then all IHRL monitoring mechanisms must urgently develop their practice in this area. The potential for the ECtHR to elaborate the content of a positive obligation to foster tolerant societies has been considered above. However, the other three IHRL monitoring mechanisms considered here serve a more preventative function than the ECtHR, and as a result have greater opportunities to elaborate the content of this obligation through State reporting processes and General Comments/Recommendations or Thematic Commentaries. While it is not the role of IHRL monitoring mechanisms to prescribe how States meet their obligations, these three mechanisms can provide detailed, non-prescriptive, guidance for States that draw on best practices and elaborate the purpose of different types of activities and the prerequisites for their success. This guidance is all the more important as agents of the State frequently perpetrate or are complicit in rights violations that are underpinned by the dehumanisation of the out-group. States should be given discretion regarding how they meet their obligation to create tolerant societies, not if.

• 6 Conclusion

  Dehumanisation requires the categorisation of an out-group as not human or less human than the in-group and as a threat to the in-group. This serves to legitimise the violation of the rights of the out-group. These rights violations are not limited to hate speech, acts of discrimination and violations of identity rights but also extend to the commission of mass atrocities. By adopting the lens of dehumanisation, this article has demonstrated that if IHRL is to achieve its purpose, it is imperative that all IHRL monitoring mechanisms seek to address dehumanisation as a root cause of rights violations. To date, IHRL monitoring mechanisms have insufficiently recognised that dehumanisation undermines the realisation of rights, particularly when dehumanisation is implicit or unconscious.
  The insights provided by social psychology have allowed this article to demonstrate how IHRL monitoring mechanisms can interpret the pre-existing IHRL framework to address dehumanisation through their monitoring practice. Specifically, pre-existing positive State obligations to prevent hate speech and foster tolerant societies, in theory, should be sufficient to counter dehumanisation as a cause of rights violations. Significantly, the AC-FCNM, CERD and HRC have clearly elaborated the content of States’ positive obligation to prevent hate speech and have struck a balance between competing rights in this respect. However, the interpretation of the positive State obligation to foster tolerant societies requires strengthening and further elaboration by all IHRL mechanisms if out-groups are to be protected from rights violations. This obligation is central to challenging unconscious and implicit dehumanisation as well as the societal conditions that allow dehumanisation to occur. IHRL mechanisms must require that States not only educate their societies about out-groups but also create opportunities for intercultural interactions to take place, during which friendship and affective ties can be forged. The removal of structural barriers is central to the success of these measures. Finally, States must be required to create a positive public culture, which recognises out-groups as an integral part of society. Significantly, all of these measures have a greater prospect of success if adopted to prevent rather than counter dehumanisation.
  The most detailed elaboration of the content of the positive State obligation to create tolerant societies has, perhaps unsurprisingly, originated from targeted mechanisms. However, dehumanisation not only results in discrimination and violations of identity rights but also underpins serious and widespread rights violations, including the commission of mass atrocities. Consequently, the HRC and ECtHR must also engage with the impact of dehumanisation on the realisation of rights and require that States take measures to address dehumanisation, if absolute rights, such as the prohibition of torture, are to be guaranteed.
  If IHRL is to achieve its purpose and protect out-groups from rights violations, IHRL monitoring mechanisms must seek to strengthen States’ positive obligation to create tolerant societies within their respective frameworks and provide non-prescriptive guidance regarding how this can be achieved in practice.

Noten

• * The author would like to thank Prof. Kristen Henrard, Prof. Alexandra Xanthaki and the two anonymous reviewers for their insightful feedback on previous drafts of this article. She would also like to thank Prof. Henrard and all participants in the international conference ‘Positive state obligations concerning fundamental rights and “changing the hearts and minds””, 30-31 January 2020, at Erasmus University Rotterdam, for their constructive comments and feedback. All errors and omissions remain the author’s own.

• 1 Convention on the Prevention and Punishment of the Crime of Genocide, 9 December 1948, 78 UNTS 277; Convention relating to the Status of Refugees, 14 December 1950, 189 UNTS 150; International Convention on the Elimination of All Forms of Racial Discrimination, 21 December 1965, 660 UNTS 195 (hereinafter ICERD).

• 2 This includes ethnic, linguistic or religious and national minorities, as recognised in international minority rights law, but also encompasses other groups that may be identified as ‘other’ by the in-group such as sexual minorities (sexual orientation or gender identity), persons with disabilities, migrants, refugees and political minorities. It is not relevant for the purposes of this article if the societal out-group self-identifies on the basis of this identity, as long as the in-group views the out-group as ‘other’ and this has the potential to result in human rights violations.

• 3 See further, I. Tourkochoriti, ‘How Far Should the State Go to Counter Prejudice? On the Legitimacy of the Role of the State to Change Hearts and Minds’, in this special edition.

• 4 Convention for the Protection of Human Rights and Fundamental Freedoms, 4 November 1950, 213 UNTS 222 (hereinafter ECHR).

• 5 International Covenant on Civil and Political Rights, 16 December 1966, 999 UNTS 171 (hereinafter ICCPR).

• 6 Framework Convention for the Protection of National Minorities, 1 February 1995, CETS No 157 (hereinafter FCNM).

• 7 N. Haslam and S. Loughan, ‘Dehumanization and Infrahumanization’, 64 Annual Review of Psychology 399, at 405 (2014).

• 8 D. Bar-Tal, ‘Delegitimization: The Extreme Case of Stereotyping’, in D. Bar-Tal, C.F. Grauman, A. Kruglanski & W. Stroebe (eds.), Stereotyping and Prejudice: Changing Conceptions (1989) 169, at 169. See further, Tourkochoriti, above n. 3.

• 9 Bar-Tal, above n. 8, at 172; N.S. Kteily and E. Bruneau, ‘Darker Demons of Our Nature: The Need to (Re)Focus Attention on Blatant Forms of Dehumanization’, 26 Current Directions in Psychological Science 487, at 488 (2017).

• 10 Bar-Tal, above n. 8, at 172.

• 11 Haslam and Loughnan, above n. 7, at 405.

• 12 Kteily and Bruneau, above n. 9, at 488.

• 13 HRC, Concluding observations on the seventh periodic report of the United Kingdom of Great Britain and Northern Ireland, UN doc. CCPR/C/GBR/CO/7 (2015), at para. 10.

• 14 AC-FCNM, Fourth Opinion on the Czech Republic adopted on 16 November 2015 ACFC/OP/IV(2015)004, at para. 53.

• 15 Ibid.

• 16 AC-FCNM, Fourth Opinion on the Slovak Republic adopted on 3 December 2014 ACFC/OP/IV(2014)004, at para. 35.

• 17 Ibid.

• 18 Bar-Tal, above n. 8, at 171.

• 19 AC-FCNM, Fourth Opinion on Cyprus adopted on 18 March 2015 ACFC/OP/IV(2015)001, at para. 34.

• 20 AC-FCNM, Fourth Opinion on Germany adopted on 19 March 2015 ACFC/OP/IV(2015)003, at para. 54.

• 21 AC-FCNM, Third Opinion on the Netherlands adopted on 6 March 2019 ACFC/OP/III(2019)003, at para. 55.

• 22 Bar-Tal, above n. 8, at 179.

• 23 Ibid., at 171.

• 24 S. Opotow, ‘Moral Exclusion and Injustice: An Introduction’, 46 Journal of Social Issues 1, at 4 (1990).

• 25 J. Vaes, M.P. Paladino, L. Castelli, J.P. Leyens & A. Giovanazzi, ‘On the Behavioral Consequences of Infrahumanization: The Implicit Role of Uniquely Human Emotions in Intergroup Relations’, 85 Journal of Personality and Social Psychology 1016, at 1031 (2003).

• 26 Kteily and Bruneau, above n. 9, at 490.

• 27 AC-FCNM Cyprus, above n. 19, at para. 34.

• 28 AC-FCNM, Fourth Opinion on Italy adopted on 19 November 2015 ACFC/OP/IV(2015)006, at para. 58.

• 29 Ibid.

• 30 AC-FCNM Slovak Republic, above n. 16, at para. 35.

• 31 AC-FCNM, Fourth Opinion on Denmark adopted on 20 May 2014 ACFC/OP/IV(2014)001, at para. 64.

• 32 Haslam and Loughnan, above n. 7, at 415.

• 33 Ibid., at 414.

• 34 Opotow, above n. 24, at 9.

• 35 Haslam and Loughnan, above n. 7, at 416.

• 36 AC-FCNM, Fourth Opinion on Hungary adopted on 25 February 2016 ACFC/OP/IV(2016)003, at paras. 75-76.

• 37 Haslam and Loughnan, above n. 7, at 415.

• 38 P.A. Goff, J.L. Eberhardt, M.J. Williams & M.C. Jackson, ‘Not Yet Human: Implicit Knowledge, Historical Dehumanization, and Contemporary Consequences’, 94 Journal of Personality and Social Psychology 292, at 305 (2008).

• 39 Kteily and Bruneau, above n. 9, at 490. See further Opotow, above n. 24, at 9.

• 40 Kteily and Bruneau, above n. 9, at 487, 490. See also Bar-Tal, above n. 8, at 176.

• 41 D. Bar-Tal, ‘Causes and Consequences of Delegitimization: Models of Conflict and Ethnocentrism’, 46 Journal of Social Issues 65 at 66 (1990); G.H. Stanton, ‘Could the Rwandan Genocide Have Been Prevented?’, 6 Journal of Genocide Research 211, at 213-17 (2004).

• 42 AC-FCNM, Fourth Opinion on the United Kingdom adopted on 25 May 2016 ACFC/OP/IV(2016)005, at para. 72; AC-FCNM Italy, above n. 28, at para. 59.

• 43 AC-FCNM Germany, above n. 20, at para. 56.

• 44 AC-FCNM Italy, above n. 28, at para. 58.

• 45 AC-FCNM, Fourth Opinion on Spain adopted on 3 December 2014 ACFC/OP/IV(2014)003, at para. 40.

• 46 Opotow, above n. 24, at 1.

• 47 AC-FCNM United Kingdom, above n. 42, at para. 72.

• 48 AC-FCNM Spain, above n. 45, at para. 40.

• 49 Opotow, above n. 24, at 12.

• 50 Ibid., at 13.

• 51 Ibid.

• 52 Ibid., at 11.

• 53 AC-FCNM, Fourth Opinion on Croatia adopted on 18 November 2015 ACFC/OP/IV(2015)005rev, at para. 41; AC-FCNM United Kingdom, above n. 42, at para. 73; HRC United Kingdom, above n. 13, at para. 10; HRC, Concluding observations on the fourth periodic report of Switzerland, UN doc. CCPR/C/CHE/CO/4 (2017), at para. 20; CERD, Concluding observations on the combined eighteenth to twenty-fifth periodic reports of Hungary, UN doc. CERD/C/HUN/CO/18-25 (2019), at para. 16; CERD, Concluding observations on the combined twenty-third and twenty-fourth periodic reports of Norway, UN doc. CERD/C/NOR/CO/23-24 (2019), at para. 12.

• 54 AC-FCNM the Netherlands, above n. 21, at para. 54.

• 55 C. Cahn, ‘CERD and Discrimination Against Roma’, in D. Keane and A. Waughray (eds.) Fifty Years of the International Convention on the Elimination of All Forms of Racial Discrimination: A Living Instrument (2017), at 107-108.

• 56 CERD Hungary, above n. 53, at para. 16; HRC, Concluding observations on the sixth periodic report of Hungary, UN doc. CCPR/C/HUN/CO/6 (2018), at para. 17.

• 57 See generally, L. Lavrysen, Human Rights in a Positive State: Rethinking the Relationship between Positive and Negative Obligations under the European Convention on Human Rights (2016).

• 58 HRC, General Comment No. 31 on The Nature of the General Legal Obligation Imposed on States Parties to the Covenant, UN doc. CCPR/C/21/Rev.1/Add.13 (2004), at para. 8.

• 59 Lavrysen, above n. 57, at 12; Committee on Economic Social and Cultural Rights, General Comment No. 12 on The right to adequate food (art. 11), UN doc. E/C.12/1999/5 (1999), at para. 15.

• 60 Airey v. Ireland, ECHR (1979) Series A. No. 32, at para. 24.

• 61 Lavrysen, above n. 57, at 6.

• 62 CERD, General recommendation No. 35 on Combating racist hate speech, UN doc. CERD/C/GC/35 (2013).

• 63 AC-FCNM, Fifth Opinion on Finland adopted on 27 June 2019 ACFC/OP/V(2019)001, at para. 105; AC-FCNM Hungary, above n. 36, at para. 103; HRC, Concluding observations on the fifth periodic report of the Netherlands, UN doc. CCPR/C/NLD/CO/5 (2018), at para 16(c); HRC, Concluding observations on the third periodic report of Bosnia and Herzegovina, UN doc. CCPR/C/BIH/CO/3 (2017), at para. 22; CERD General recommendation No. 35 (2013), above n. 62, at para. 9; CERD, Concluding observations on the combined second to fifth periodic reports of Serbia, UN doc. CERD/C/SRB/CO/2-5 (2018), at para. 14.

• 64 Assenov v. Bulgaria, ECHR (1998) Reports 1998-VIII, at para. 102; Vgt Verein Gegen Tierfabriken v. Switzerland, ECHR (2001) Reports 2001-VI, at para. 45; Assanidzé v. Georgia, ECHR (2004) Reports 2004-II, at para. 182. See further, J.-F. Akandji-Kombe, Positive Obligations Under the European Convention on Human Rights: A Guide to the Implementation of the European Convention on Human Rights (2007), at 8-9.

• 65 M. Bossuyt, International Human Rights Protection: Balanced, Critical, Realistic (2018), at 139.

• 66 Lavrysen, above n. 57, at 60.

• 67 Erbakan v. Turkey, ECHR (2006) Application no. 59405/00, at para. 56. Translation from ECtHR, Factsheet on Hate Speech, February 2020, www.echr.coe.int/Documents/FS_Hate_speech_ENG.pdf.

• 68 Karaahmed v. Bulgaria, ECHR (2015) Application no. 30587/13, at para. 110.

• 69 See further, A. Böcker, ‘Can Non-discrimination Law Change Hearts and Minds?’, in this special edition.

• 70 AC-FCNM, Fourth Opinion on Sweden adopted on 22 June 2017 ACFC/OP/IV(2017)004, at para. 54; AC-FCNM, Fourth Opinion on Switzerland adopted on 31 May 2018 ACFC/OP/IV(2018)003, at para. 64; CERD Norway, above n. 53, at para. 12(e); CERD, Concluding observations on the combined twelfth and thirteenth periodic reports of Czechia, UN doc. CERD/C/CZE/CO/12-13 (2019), at para. 11(b); HRC the Netherlands, above n. 63, at para. 16; HRC United Kingdom, above n. 13, at para. 10.

• 71 Nachova and others v. Bulgaria, ECHR (2005) Reports 2005-VII, at para. 145.

• 72 Goff, Eberhardt, Williams & Jackson, above n. 38, at 304-306.

• 73 Ibid., at 305; Kteily and Bruneau, above n. 9, at 492.

• 74 AC-FCNM Croatia, above n. 53, at para. 46.

• 75 AC-FCNM, Fourth Opinion on Austria adopted on 14 October 2016 ACFC/OP/IV(2016)007, at para. 34.

• 76 AC-FCNM Germany, above n. 20, at para. 60.

• 77 CERD, Concluding observations on the combined fifth to ninth reports of Ireland, UN doc. CERD/C/IRL/CO/5-9 (2019), at para. 24(b); CERD, Concluding observations on the combined twentieth to twenty-second periodic reports of Bulgaria, UN doc. CERD/C/BGR/CO/20-22 (2017), at para. 20(c).

• 78 CERD, Concluding observations on the combined twenty-second to twenty-fourth periodic reports of Poland, UN doc. CERD/C/POL/CO/22-24 (2018), at para. 16(c); CERD, Concluding observations on the twenty-third periodic report of Finland, UN doc. CERD/C/FIN/CO/23 (2017), at para. 23.

• 79 CERD, Concluding observations on the combined tenth and eleventh periodic reports of the Republic of Moldova, UN doc. CERD/C/MDA/CO/10-11 (2017), at para. 13(c).

• 80 HRC General Comment No. 31 (2004), above n. 58, at para. 8.

• 81 Although initially developed in relation to socio-economic rights, this framework has subsequently been acknowledged to apply more generally. For example, Committee on Economic Social and Cultural Rights General Comment No. 12, above n. 59, at para. 15; UN General Assembly, Interim Report of the Special Rapporteur on Freedom of Religion or Belief, UN doc. A/71/269 (2016), at para. 23.

• 82 HRC Hungary, above n. 56, at para. 18; HRC, Concluding observations on the sixth periodic report of Italy, UN doc. CCPR/C/ITA/CO/6 (2017), at para. 13.

• 83 HRC, Concluding observations on the seventh periodic report of Sweden, UN doc. CCPR/C/SWE/CO/7 (2016), at para. 17; HRC United Kingdom, above n. 13, at para. 10(b).

• 84 HRC Hungary, above n. 56, at para. 18; HRC Sweden, above n. 83, at para. 17.

• 85 Lavrysen, above n. 57, at 94.

• 86 Supreme Holy Council of the Muslim Community v. Bulgaria, ECHR (2004) Application No. 39023/97, at para. 96.

• 87 S.A.S. v. France, ECHR (2014) Reports 2014, at para. 149.

• 88 AC-FCNM Czech Republic, above n. 14, at paras. 53-54; AC-FCNM Slovak Republic, above n. 16, at paras. 35-37.

• 89 AC-FCNM the Netherlands, above n. 21, at para. 54; AC-FCNM Italy, above n. 28, at para. 58.

• 90 AC-FCNM United Kingdom, above n. 42, at para. 73; AC-FCNM the Netherlands, above n. 21, at para. 54.

• 91 AC-FCNM, Fourth Opinion on the Republic of Moldova adopted on 25 May 2016 ACFC/OP/IV(2016)004, at para. 6.

• 92 AC-FCNM Cyprus, above n. 19, at para. 34.

• 93 AC-FCNM Austria, above n. 75, at para. 31; AC-FCNM Italy, above n. 28, at para. 59.

• 94 AC-FCNM Denmark, above n. 31, at para. 65.

• 95 AC-FCNM Germany, above n. 20, at para. 56.

• 96 AC-FCNM Italy, above n. 28, at para. 59.

• 97 AC-FCNM Austria, above n. 75, at para. 36.

• 98 AC-FCNM Spain, above n. 45, at para. 40.

• 99 AC-FCNM Czech Republic, above n. 14, at para. 57.

• 100 AC-FCNM Spain, above n. 45, at para. 42.

• 101 AC-FCNM Austria, above n. 75, at para. 36.

• 102 HRC Switzerland, above n. 53, at para. 20.

• 103 CERD Poland, above n. 78, at para. 15.

• 104 CERD Czechia, above n. 70, at para. 11. See also CERD Hungary, above n. 53, at para. 22; HRC Italy, above n. 82, at para. 74.

• 105 CERD Czechia, above n. 70, at para. 11. See also CERD Hungary, above n. 53, at para. 22.

• 106 HRC Italy, above n. 82, at para. 74. See also CERD, Concluding observations on the combined nineteenth and twentieth periodic reports of Italy, UN doc. CERD/C/ITA/CO/19-20 (2017), at para. 14.

• 107 HRC Sweden, above n. 83, at para. 16.

• 108 HRC Switzerland, above n. 53, at para. 20; HRC Sweden, above n. 83, at para. 16; CERD Poland, above n. 78, at para. 15; CERD, Concluding observations on the combined twenty-first to twenty-third periodic reports of the United Kingdom of Great Britain and Northern Ireland, UN doc. CERD/C/GBR/CO/21-23 (2016), at para. 15.

• 109 CERD United Kingdom, above n. 108, at para. 15; CERD Poland, above n. 78, at para. 15; HRC Switzerland, above n. 53, at para. 20.

• 110 CERD General recommendation No. 35 (2013), above n. 62, at para. 10.

• 111 CERD, Concluding observations on the combined twenty-second and twenty-third periodic reports of Sweden, UN doc. CERD/C/SWE/CO/22-23 (2018), at para. 18.

• 112 HRC Switzerland, above n. 53, at para. 42.

• 113 HRC, Concluding observations on the fifth periodic report of France, UN doc. CCPR/C/FRA/CO/5 (2015), at para. 22.

• 114 See, for example, AC-FCNM, Second opinion on Switzerland adopted on 29 February 2008 ACFC/OP/II(2008)002, at para. 89; European Commission against Racism and Intolerance, ECRI Report on Switzerland (fourth monitoring cycle) adopted on 2 April 2009 CRI(2009)32, at para. 120; A. Vakulenko, Islamic Veiling in Legal Discourse (2012) at 17-18; H. Elver, The Headscarf Controversy: Secularism and Freedom of Religion (2012), at 117-120.

• 115 CERD Sweden, above n. 111, at paras. 18-19; HRC Italy, above n. 82, at paras. 12-13.

• 116 CERD, Concluding observations on the combined eleventh and twelfth periodic reports of Slovakia, UN doc. CERD/C/SVK/CO/11-12 (2018), at para. 23; CERD, Concluding observations on the combined tenth and eleventh periodic reports of the Czech Republic, UN doc. CERD/C/CZE/CO/10-11 (2015), at para. 22; CERD Czechia, above n. 70, at para. 19.

• 117 HRC France, above n. 113, at para. 13.

• 118 HRC, Concluding observations on the sixth periodic report of Spain, UN doc. CCPR/C/ESP/CO/6 (2015), at para. 9; HRC, Concluding observations on the seventh periodic report of Norway, UN doc. CCPR/C/NOR/CO/7 (2018), at para. 8.

• 119 HRC Switzerland, above n. 53, at para. 50.

• 120 HRC Italy, above n. 82, at para. 14.

• 121 Abdulaziz, Cabales and Balkandali v. the United Kingdom, ECHR (1985) Series A No. 94, at para. 67; Boujlifa v. France, ECHR (1997) Reports 1997-VI, at para. 42; Nunez v. Norway, ECHR (2011) Application No. 55597/09, at para. 71.

• 122 Berrehab v. the Netherlands, ECHR (1988) Series A No. 138, at para. 26.

• 123 C. Costello, The Human Rights of Migrants and Refugees in European Law (2016), at 128; A. Desmond, ‘The Private Life of Family Matters: Curtailing Human Rights Protection for Migrants under Article 8 of the ECHR?’, 29 European Journal of International Law 261, at 265 (2018). M.-B. Dembour, When Humans Become Migrants: Study of the European Court of Human Rights with an Inter-American Counterpoint (2015), at 504.

• 124 See further, K. Henrard, ‘State Obligations to Counter Islamophobia: Comparing Fault Lines in the Supervisory Practice of the HRC/ICCPR, the ECtHR and the AC/FCNM’, in this special edition. See, also, C. Evans, ‘The “Islamic Scarf” in the European Court of Human Rights’, 7 Melbourne Journal of International Law 52, at 65-67 (2006); C.H. Skeet, ‘Orientalism in the European Court of Human Rights’, 14 Religion and Human Rights 31 (2019); K. Henrard, ‘Integration Reasoning at the ECtHR: Challenging the Boundaries of Minorities’ Citizenship’, 38 Netherlands Quarterly of Human Rights 55, at 70-72 (2020).

• 125 Dahlab v. Switzerland, ECHR (2001) Reports 2001-V.

• 126 Ibid.; Leyla Șahın v. Turkey, ECHR (2005) Reports 2005-XI, at para. 115; Dogru v. France, ECHR (2008) Application No. 27058/05, at para. 72.

• 127 S.A.S., above n. 87, at para. 141; Osmanoğlu and Kocabaş v. Switzerland, ECHR (2017) Application No. 29086/12, at paras. 96-97. For a more detailed discussion of this point, see generally, Henrard (2020), above n. 124.

• 128 S.A.S., above n. 87, at para. 149.

• 129 Chapman v. United Kingdom, ECHR (2001) Reports 2001-I, at para. 96.

• 130 See further, R. Sandland, ‘Developing a Jurisprudence of Difference: The Protection of the Human Rights of Travelling Peoples by the European Court of Human Rights’, 8 Human Rights Law Review 475, at 485 (2008). See for example, Buckley v. the United Kingdom, ECHR (1996) Reports 1996-IV; Velikova v. Bulgaria, ECHR (2000) Reports 2000-VI; Anguelova v. Bulgaria, ECHR (2002) Reports 2002-IV.

• 131 See, for example, Nachova, above n. 71; Moldovan and Others v. Romania (No. 2), ECHR (2005) Reports 2005-VII; Stoica v. Romania, ECHR (2008) Application No. 42722/02; D.H. and others v. the Czech Republic, ECHR (2007) Reports 2007-IV.

• 132 Balogh v. Hungary, ECHR (2004) Application No. 47940/99; Bekos and Koutropoulos v. Greece, ECHR (2005) Reports 2005-XIII; Cobzaru v. Romania, ECHR (2007) Application No. 8254/99.

• 133 V.C. v. Slovakia, ECHR (2011) Application No, 18968/07, at para. 177.

• 134 S.E. Berry, ‘The Siren’s Call? Exploring the Implications of an Additional Protocol to the European Convention on Human Rights on National Minorities’, 23 International Journal on Minority and Group Rights 1, at 30-31 (2016).

• 135 V.C. (2011), above n. 133, at dissenting opinion Judge Mijovic.

• 136 D.H. and others (2007), above n. 131, at paras. 191-95.

• 137 Berry (2016), above n. 134.

• 138 S.A.S., above n. 87, at para. 149.

• 139 See, for example, S.A.S., above n. 87, at dissenting opinion of Judges Nussberger and Jäderblom, at para. 14. Cf. Supreme Holy Council of the Muslim Community, above n. 86, at para. 96.

• 140 W.A. Donohue, ‘The Identity Trap: The Language of Genocide’, 31 Journal of Language and Social Psychology 13 (2011).

• 141 CERD General recommendation No. 35 (2013), above n. 62, at para. 9; AC-FCNM Finland, above n. 63, at para. 105; AC-FCNM Germany, above n. 20, at paras. 61 and 70.

• 142 CERD Ireland, above n. 77, at para. 22(a); CERD Norway, above n. 53, at para. 14(b).

• 143 CERD United Kingdom, above n. 108, at para. 16(a); CERD Poland, above n. 78, at paras. 22(e) and 24(e); AC-FCNM Austria, above n. 75, at para. 39; AC-FCNM Finland, above n. 63, at para. 57. See also, HRC United Kingdom, above n. 13, at para. 10(d); HRC Hungary, above n. 56, at para. 18.

• 144 CERD, Concluding observations on the twentieth to twenty-second periodic reports of Greece, UN doc. CERD/C/GRC/CO/20-22 (2016), at para. 17(b); CERD Sweden, above n. 111, at para. 11(c); AC-FCNM Finland, above n. 63, at para. 104; AC-FCNM Cyprus, above n. 19, at para. 39.

• 145 AC-FCNM the Netherlands, above n. 21, at para. 66; AC-FCNM, Fourth Opinion on Norway adopted on 13 October 2016 ACFC/OP/IV(2016)008, at para. 58; CERD Greece, above n. 144, at paras. 17(d) and (e); CERD United Kingdom, above n. 108, at para. 16(c). See also, HRC, Concluding observations on the fifth periodic report of Austria, UN doc. CCPR/C/AUT/CO/5 (2015), at para. 16.

• 146 AC-FCNM Finland, above n. 63, at para. 57; AC-FCNM Germany, above n. 20, at para. 77; CERD United Kingdom, above n. 108, at para. 16(c); CERD Greece, above n. 144, at paras. 17(e).

• 147 AC-FCNM Finland, above n. 63, at para. 57; AC-FCNM Germany, above n. 20, at para. 77.

• 148 Donohue, above n. 140, at 28.

• 149 Goff, Eberhardt, Williams & Jackson, above n. 38, at 304; Donohue, above n. 140, at 14; 25; Stanton, above n. 41, at 214-15; Opotow, above n. 24, at 13; A. De Buyse, ‘Words of Violence: “Fear Speech,” or How Violent Conflict Escalation Relates to the Freedom of Expression’, 36 Human Rights Quarterly 779, at 779- 81 (2014).

• 150 Donohue, above n. 140, at 13, 14, 25; Stanton, above n. 41, at 214-15.

• 151 S.E. Berry, ‘Democracy and the Preservation of Minority Identity: Fragmentation within the European Human Rights Framework’, 24 International Journal on Minority and Group Rights 205, at 212-13 (2017).

• 152 CERD General recommendation No. 35 (2013), above n. 62, at para. 15.

• 153 AC-FCNM Norway (2016), above n. 145, at para. 53.

• 154 CERD United Kingdom, above n. 108, at para. 16(d); CERD, Concluding observations on the combined nineteenth to twenty-second periodic reports of Germany, UN doc. CERD/C/DEU/CO/19-22 (2015), at para. 9; AC-FCNM Austria, above n. 75, at para. 40; AC-FCNM Finland, above n. 63, at para. 53.

• 155 CERD General recommendation No. 35 (2013), above n. 62, at para. 37. See also CERD United Kingdom, above n. 108, at para. 16(d); AC-FCNM United Kingdom, above n. 42, at para. 76; AC-FCNM Spain, above n. 45, at para. 48.

• 156 AC-FCNM United Kingdom, above n. 42, at para. 76.

• 157 AC-FCNM Czech Republic, above n. 14, para. 57.

• 158 CERD Finland, above n. 78, at para. 11(c).

• 159 CERD Italy, above n. 106, at paras. 15(a) and (g); CERD Norway, above n. 53, at paras. 14(a) and (c). See also, AC-FCNM Austria, above n. 75, at para. 38.

• 160 Haslam and Loughnan, above n. 7, at 416.

• 161 Ibid.

• 162 CERD Czechia, above n. 70, at para. 11(c); CERD Hungary, above n. 53, at para. 16; AC-FCNM Denmark, above n. 31, at para. 64; AC-FCNM Italy, above n. 28, at para. 58.

• 163 AC-FCNM Austria, above n. 75, at para. 40; AC-FCNM Germany, above n. 20, at para. 65; AC-FCNM United Kingdom, above n. 42, at para. 76; CERD Czechia, above n. 70, at para. 12(c).

• 164 AC-FCNM United Kingdom, above n. 42, at para. 97; CERD General recommendation No. 35 (2013), above n. 62, at para. 39.

• 165 AC-FCNM Germany, above n. 20, at para. 65; AC-FCNM, Fifth Opinion on Denmark adopted on 7 November 2019 ACFC/OP/V(2019)003, at para. 76.

• 166 CERD Slovakia, above n. 116, at para. 14(a).

• 167 CERD Italy, above n. 106, at para. 15(f).

• 168 CERD Greece, above n. 144, at para. 17(c).

• 169 CERD General recommendation No. 35 (2013), above n. 62, at para. 40.

• 170 AC-FCNM United Kingdom, above n. 42, at para. 76.

• 171 AC-FCNM Norway (2016), above n. 145, at para. 53; AC-FCNM Switzerland, above n. 70, at para. 64; CERD Czechia, above n. 70, at para. 12(b); HRC the Netherlands, above n. 63, at para. 16; HRC Hungary, above n. 56, at para. 18.

• 172 AC-FCNM, Fourth Opinion on Malta adopted on 14 October 2016 ACFC/OP/IV(2016)009, at para. 25; AC-FCNM the Netherlands, above n. 21, at para. 57.

• 173 AC-FCNM Czech Republic, above n. 14, at para. 57; AC-FCNM Finland (2016), above n. 63, at para. 52; CERD Poland (2018), above n. 78, at para. 16(c); CERD Norway, above n. 53, at para. 12(e); HRC Hungary, above n. 56, at para. 18; HRC Italy, above n. 82, at para. 13; HRC Sweden, above n. 83, at para. 17.

• 174 CERD Czechia, above n. 70, at para. 12(b); CERD Norway, above n. 53, at para. 12(e); HRC Hungary, above n. 56, at para. 18.

• 175 Art. 12 FCNM; CERD General recommendation No. 35 (2013), above n. 62, at paras. 34-35.

• 176 CERD Italy, above n. 106, at para. 26(e). See also, CERD United Kingdom, above n. 108, at para. 35(c).

• 177 L. Bekemans, ‘Educational Challenges and Perspectives in Multiculturalism vs. Interculturalism: Citizenship. Education for Intercultural Realities’, in M. Barrett (ed.), Interculturalism and Multiculturalism: Similarities and Differences (2013), at 177, 183; G. Bouchard, ‘Interculturalism: What Makes it Distinctive?’, in M. Barrett (ed.), Interculturalism and Multiculturalism: Similarities and Differences (2013), at 107.

• 178 W.G. Stephan and C.W. Stephan, ‘An Integrated Threat Theory of Prejudice’, in S. Oskamp (ed.), Reducing Prejudice and Discrimination (2000), at 40-41.

• 179 A. Rodríguez Pérez, N. Delgado Rodríguez, V. Betancor Rodríguez, J.P. Leyens & J. Vaes, ‘Infra-humanization of Outgroups throughout the World. The Role of Similarity, Intergroup Friendship, Knowledge of the Outgroup, and Status’, 27 Anales de Psicología 679, at 685-86 (2011).

• 180 Haslam and Loughnan, above n. 7, at 416.

• 181 See generally, G.W. Allport, The Nature of Prejudice (1954).

• 182 Rodríguez Pérez, Delgado Rodríguez, Betancor Rodríguez, Leyens & Vaes, above n. 179, at 681; P. Loobuyck, ‘Towards an Intercultural Sense of Belonging Together: Reflections on the Theoretical and Political Level’, in N. Meer, T. Modood & R. Zapata-Barrero (eds.), Interculturalism and Multiculturalism (2016), at 225; T. Cantle, ‘Interculturalism as a New Narrative for the Era of Globalisation and Super-Diversity’, in M. Barrett (ed.), Interculturalism and Multiculturalism: Similarities and Differences (2013), at 69, 83. See also, Tourkochoriti, above n. 3, for a discussion of how instrumental intergroup interventions can reduce prejudice.

• 183 Donohue, above n. 140, at 27.

• 184 Cantle, above n. 182, at 79.

• 185 M. Barrett, ‘Intercultural Competence: A Distinctive Hallmark of Interculturalism?’, in M. Barrett (ed.), Interculturalism and Multiculturalism: Similarities and Differences (2013), at 157.

• 186 Loobuyck, above n. 182, at 230.

• 187 AC-FCNM Norway, above n. 140, at para. 53; AC-FCNM Cyprus, above n. 19, at para. 36; CERD Norway, above n. 53, at para. 24; CERD, General recommendation No. 27 on discrimination against Roma, UN doc. A/55/18, annex V (2000), at para. 9.

• 188 AC-FCNM, Fourth Opinion on Estonia adopted on 19 March 2015 ACFC/OP/IV(2015)002, at para. 84.

• 189 AC-FCNM Cyprus, above n. 19, at para. 59.

• 190 AC-FCNM Denmark (2019), above n. 166, at para. 91. See also, AC-FCNM Austria, above n. 75, at para. 31.

• 191 Stephan and Stephan, above n. 178, at 38; Haslam and Loughnan, above n. 7, at 416.

• 192 Barrett, above n. 185, at 157.

• 193 See further, A. Kok, ‘The Promotion of Equality and Prevention of Unfair Discrimination Act 4 of 2000: Proposals for Legislative Reform to Promote Equality Through Schools and the Education System’, in this special edition.

• 194 CERD Germany, above n. 154, at para. 13(c); CERD Czechia, above n. 70, at para. 18; CERD, Concluding observations on the combined twelfth and thirteenth periodic reports of Bosnia and Herzegovina, UN doc. CERD/C/BIH/CO/12-13 (2018), at para. 23(b); CERD Slovakia, above n. 116, at para. 12(b); CERD Serbia, above n. 63, at para. 23; CERD Finland, above n. 78, at para. 13.

• 195 S.E. Berry, ‘Aligning Interculturalism with International Human Rights Law: “Living Together” without Assimilation’, 18 Human Rights Law Review 441, at 459 (2018).

• 196 AC-FCNM Austria, above n. 75, at para. 34; AC-FCNM Spain, above n. 45, at para. 50; CERD Poland, above n. 78, at para. 22; CERD General recommendation No. 27 (2000), above n. 187, at para. 9.

• 197 Haslam and Loughnan, above n. 7, at 416.

• 198 G. Bouchard, ‘What is Interculturalism?’, 56 McGill Law Journal 435, at 445 (2011).

• 199 Haslam and Loughnan, above n. 7, at 416.

• 200 AC-FCNM Moldova, above n. 91, at 1.

• 201 R. Zapata-Barrero, ‘Theorising Intercultural Citizenship’, in N. Meer, T. Modood & R. Zapata-Barrero (eds.), Interculturalism and Multiculturalism (2016), at 64.

• 202 CERD General recommendation No. 27, above n. 187, at para. 11.

• 203 AC-FCNM Moldova, above n. 91, at para. 38; AC-FCNM Slovak Republic, above n. 16, at para. 37.

• 204 AC-FCNM, Third Opinion on Poland adopted on 28 November 2013 ACFC/OP/III(2013)004, at 2; AC-FCNM, Fourth Opinion on Bosnia and Herzegovina adopted on 9 November 2017 ACFC/OP/IV(2017)007, at para. 73.

• 205 AC-FCNM Italy, above n. 28, at para. 95; AC-FCNM Malta, above n. 172, at para. 25.

• 206 AC-FCNM Austria, above n. 75, at para. 34.