Criminalising Hate Crime and Hate Speech at EU Level: Extending the List of Eurocrimes Under Article 83(1) TFEU

Abstract

In September 2020, President von der Leyen announced the Commission’s intention to propose to extend the list of EU crimes or Eurocrimes to all forms of hate crime and hate speech, as later reflected in the Commission Work Programme 2021. The article first examines the need for such action at EU level, highlighting also certain shortcomings with existing regulation of hate crime and hate speech across the EU. Next, it raises a few additional challenges, relating to the regulation of the freedom of expression in view of the historical experience in some EU countries, the criminal offence element of public order violation or endangerment, and the harm-offence distinction. It then inspects whether specific conditions for EU action under Article 83(1) TFEU can be satisfied by the proposed area(s) of crime, not only from the perspective of a textual legal analysis but also drawing where appropriate on relevant empirical data. While acknowledging the somewhat controversial topic of regulation in the already sensitive field of EU criminal law as well as certain clear political hurdles linked with the procedural aspects of enlarging the EU crimes’ list, it is submitted that the mentioned crimes and their regulation raise legitimate concerns that warrant the proposed legislative action at EU level.

Introduction

In the State of the Union Address on 16 September 2020, President von der Leyen announced the Commission’s intention to “propose to extend the list of EU crimes to all forms of hate crime and hate speech – whether because of race, religion, gender or sexuality”. This initiative was later mentioned in the Commission President’s letter of intent, included in the Commission Work Programme 2021 and adopted on 9 December 2021.¹ It reflects a recognition that hate crime and hate speech are considered by many as EU-wide problems. The objective of the initiative on having hate speech and hate crime identified as “other areas of crime” for the purposes of Article 83(1) of the Treaty on the Functioning of the European Union (“TFEU”) is officially justified by the claim that both are particularly serious crimes, capable (“can”) of spreading across borders and that developments in crime also justify their inclusion on this list.²

This initiative is set within a wider framework of Commission’s activities, which it aims to complement. The EU Gender Equality Strategy 2020–2025 thus already referred to the inclusion of hate speech and hate crime on grounds of sex on the list of “Eurocrimes”, seeing them as specific forms of gender-based violence and recognising, inter alia, that “[t]oo many people still violate the principle of gender equality through sexist hate speech and by blocking action against gender-based violence and gender stereotypes”.³ The LGBTIQ Equality Strategy 2020–2025 further referred to the extension of the list of the list of EU crimes under Article 83(1) TFEU “to cover hate crime and hate speech, including when targeted at LGBTIQ people”, thereby adding sexual orientation to the list of prohibited grounds.⁴ Moreover, said initiative is meant to complement the parallel Commission initiative on preventing and combatting gender-based violence against women and domestic violence by providing an additional legal basis for addressing those specific forms of serious gender-based violence that can also be defined as hate speech or hate crime on grounds of gender. The Roadmap explains that the extension of the list of EU crimes to include hate speech and hate crime is also “part of the EU’s response to extremist ideologies online and more specifically to the proliferation of racist and xenophobic hate speech on the internet”.⁵ It is thus complementary to the Counter-Terrorism Agenda for the EU and the proposal for a Digital Services Act,⁶ while it also fortifies the Commission’s 2020–2025 anti-racism action plan and the strategy on combating anti-Semitism that has been announced in the Commission Work Programme 2021.⁷

The Framework Decision 2008/913/JHA on racism and xenophobia⁸ (“Framework Decision”) is the only existing EU criminal law instrument that harmonises the definition of and criminal penalties for some specific forms of hate speech and hate crime, namely race, colour, religion, descent or national or ethnic origin. Aiming to ensure that serious manifestations of racism and xenophobia are punishable by effective, proportionate and dissuasive criminal penalties throughout the EU, it requires Member States to take the necessary measures to criminalise the public incitement to violence or hatred on above stated grounds.⁹ Member States were obliged to transpose the obligations imposed on them under the Framework Decision into their national law by 28 November 2010. As a result, they already have or should have national laws in place criminalising hate speech and hate crime on the grounds of race, colour, religion, descent or national or ethnic origin – even though implementation varies greatly (see below). In addition, victims of hate speech and hate crime are required to be provided with an effective remedy and rights, support and protection, as mandated under the Victims’ Rights Directive (2012/29/EU),¹⁰ which recognised victims of hate crime as especially vulnerable victims requiring particular attention (Article 22(3)) and at high risk of experiencing secondary and repeat victimisation, intimidation and retaliation (Recital 57). The Audiovisual Media Services Directive (2010/13/EU) obliges EU Member States to ensure by appropriate means that audio-visual media services provided by media service providers under their jurisdiction do not contain any incitement to hatred based on race, religion, sex or nationality (Article 6). Another complementary action is the EU Code of Conduct on countering illegal hate speech online (2016), aimed at IT companies who are encouraged by the Commission to sign it in order to ensure the effective enforcement of legislation by preventing and countering the illegal hate speech online.¹¹

With respect to hate speech and hate crime based on protected characteristics or prohibited grounds other than those laid down in Framework Decision 2008/913/JHA, there is no harmonisation of criminal offences and sanctions at EU level. It thus depends entirely on the Member States to criminalise other types of hate speech and hate crime, e.g., on grounds of sex and sexual orientation, leading to significant variation in legal protection across the EU. Currently, 21 Member States explicitly include sexual orientation in hate speech and hate crime legislation, or both, as an aggravating factor, while 12 among them also include gender identity and two cover sex characteristics.¹² Additionally, “some Member States are looking to criminalise misogynous hate speech and crime, following its perceived increase both globally and in Europe.”¹³

From 23 February to 20 April 2021, a public and targeted consultations on this initiative were open. The targeted questionnaire was aimed at national and international public authorities, key civil society organisations and networks, European networks of experts and research institutions, which have expertise on the national legal frameworks criminalising hate speech and hate crime and/or collect relevant data. The questions aimed to collect information and views on the current national legal frameworks and the landscape of hate speech and hate crime across Europe. The public consultation yielded 1.486 valid responses, and included contributions from important equality bodies and NGOs.¹⁴

While EU criminalisation powers are not entirely uncontroversial among Member States, which justifies careful and incremental Union steps in this area, its competence post-Lisbon has been entrenched in the Treaty, specifically, Article 83 TFEU (ex 31 TEU), reflecting the approach taken by the Court of Justice (“CJEU”).¹⁵ Article 83(1) TFEU lays down a list of areas of crimes (dubbed EU crimes or Eurocrimes/Euro-crimes) in which the harmonisation of the definition of criminal offences and sanctions by the EU is achievable under the ordinary legislative procedure. EU crimes refer to “areas of particularly serious crimes with a cross-border dimension, resulting from the nature or impact of such offences or from a special need to combat them on a common basis”. The present list of such areas of crime includes: terrorism, trafficking in human beings and sexual exploitation of women and children, illicit drug trafficking, illicit arms trafficking, money laundering, corruption, counterfeiting of means of payment, computer crime and organised crime. These areas largely correspond to the areas that were subject of third-pillar legislation,¹⁶ and some have observed that “that omission of any reference to racism and xenophobia is notable and seems to call into question the future of the provisions of the framework decision”.¹⁷

While the existing list of EU crimes is exhaustive, it can be expanded through a procedure stipulated in the third subparagraph of Article 83(1). On the basis of “developments in crime”, the Council, acting unanimously after obtaining the consent of the European Parliament, may adopt a decision identifying other areas of crime that meet the criteria specified in the first subparagraph. The Commission therefore planned to present an initiative (within the meaning of Article 17(1) TFEU) in the form of a Communication to the European Parliament and the Council with the aim of triggering such a Council decision that would identify hate speech and hate crime as other areas of crime that meet the criteria specified in Article 83(1), first subparagraph, TFEU. Should such a decision be adopted, the Commission would then have competence to propose substantive legislation (a directive) harmonising the definition of and penalties for hate speech and hate crime.

The article first examines, in “The Need to Regulate at EU Level” section, the need for such an action at EU level, highlighting also certain shortcomings with existing regulation of hate crime and hate speech across the EU. In “Additional Challenges Related to the Criminalisation of Hate Speech” section, it addresses a few additional challenges, relating to the regulation of hate speech. Specifically, it looks, first, at the challenge related to the regulation of the freedom of expression in view of the historical experience in some EU countries – which can militate against criminalisation or at least shed some light on the fears and hesitancy to tackle the problem head on. It also addresses the link with public order violation or endangerment, and the harm-offence distinction which impacts on the legitimacy of criminalisation. “Article 83(1) Conditions for Inclusion of New EU Crimes on the List” section then inspects whether specific elements or conditions of EU action under Article 83(1) of TFEU, namely, particular seriousness, cross-border dimension, and developments in crime, can be satisfied by the proposed area(s) of crime, drawing where appropriate on relevant empirical sources, such as victimological data required for an evidence-guided criminal policy and criminalisation. While acknowledging the somewhat controversial topic of regulation in the already sensitive area of EU law, i.e., EU criminal law, as well as appreciating certain political hurdles linked with the procedural aspects of enlarging the EU crimes’ list, it is submitted that the proposed crimes legitimately raise EU concerns and would, moreover, qualify for data-informed EU criminal law intervention.

Lastly, a note on terminology. “Hate” in hate speech and hate crime is a bit of a misnomer, a catch-all word that rather poorly describes what is, in reality, a variety of prejudice-based motivations, some arising out of quite different emotions (not hatred as such), some not directly prompted by emotions at all. This is why hostility, prejudice, bias or intolerance of “the Other” is more accurate and more commonly used in scientific literature – particularly the one that draws on psychological data – and progressively so in official definitions.¹⁸ Hate crime would therefore be more appropriately defined as prejudice crime,¹⁹ prejudice-motivated crime (also bias-motivated crime)²⁰ or intolerance-motivated crime.²¹ Nevertheless, the Commission opted for a more common or everyday term. The notions of hate speech and hate crime are thus used in all of the strategies mentioned and the Code of Conduct. The Recital 9 of the preamble to the Framework Decision speaks of “hatred” and stipulates that hatred for the purposes of this act should be understood as referring to hatred based on race, colour, religion, descent or national or ethnic origin, but it does not engage in providing a definition for the concept as such.²² As these are also the terms used in relation to the discussed initiative and forthcoming COM proposals, it is these nominally hate-based terms that will be used in this article as well, notwithstanding the caveat stated above.

The Need to Regulate at EU Level

Incidence, Prevalence and Increase in Hate Crime and Hate Speech

In view of the incidence of hate crime and hate speech and its increase in the last few years (driven also by populist politicians, illiberal tendencies and the ease of spreading hate speech online), the concern about hate speech and hate crime expressed by the European Commission is certainly warranted. Germany, for example, recorded in 2021 a 6% rise in far-right crimes from the year before. They ranged from displaying Nazi symbols, anti-Semitic remarks to physical attacks and murder, and targeted mainly immigrants, refugees and black Germans, although they also included a rise in anti-Asian violence, linked to the pandemic.²³ Similar phenomena have been observed across Europe. An overview by the Fundamental Rights Agency (“FRA”), inspecting anti-Semitic incidents recorded in the EU between 1 January 2009 and 31 December 2019, has revealed increasing trend lines in anti-Semitic incidents, offences or threats in many countries, even though the numbers behind them cannot necessarily be attributed to the rise of crime (alone) but may (also) be the result of the changes in the reporting system, increased efficiency of the recording system or state efforts to render such conduct more visible.²⁴ The same document reveals that respondents would feel even less comfortable with having a member of the following groups as a neighbour or married to a close relative: a gay, lesbian, bisexual person; a transgender or transsexual person; a Muslim; and asylum seeker/refugee; and lastly (most uncomfortable with) a Roma/gypsy person.²⁵

The Council of Europe’s European Commission against Racism and Intolerance (“ECRI”) noted in its 2020 report that racist and xenophobic incidents connected to the outbreak have been widespread, acknowledging the rise in anti-Semitic, anti-Asian, anti-LGBTI hate speech during these times, in addition to the continuing “alarming levels” of anti-Semitic and anti-Muslim violence, which are “often incited and aggravated by hate speech online”.²⁶ OSCE’s Office for Democratic Institutions and Human Rights (“ODIHR”) latest Hate Crime Report (i.e., for 2019), which is the largest hate crime dataset of its kind, reported “a record number of 3,207 statistical and 3,757 descriptive hate crime incidents reported by civil society, which translates into a minimum of 4,621 hate crime victims”²⁷ in 39 OSCE member states which reported official hate crime data that year. Victims of “sexual orientation or gender identity” were the third largest group of victims – after victims of racism/xenophobia and Roma/Sinti but just before victims of bias against Muslims. Regarding the latter, FRA’s 2017 report demonstrates that about one in four Muslim respondents (27%) reported experiencing hatred-motivated harassment due to their ethnic or immigrant background at least once (although 45% of them have experienced six or more incidents) in the 12 months before the survey.²⁸

Hate and prejudice, however, do not motivate only harassment, threats, hateful expression but also outright attacks. FRA’s 2014 report, based on interviews with 42,000 women across the 28 Member States of the European Union (EU) on violence against women, reveals extensive, yet systematically underreported violence that affects many women’s lives: one in 10 women has experienced some type of sexual violence since the age of 15, and one in 20 has been raped. Just over one in five women has experienced physical or sexual violence, or both, from a current or previous partner. Nevertheless, only 14% of women reported their most serious incident of intimate partner violence (and 13% in the case of most serious incident of non-partner violence) to the police.²⁹ The 2019 ODIHR Hate Crime Report revealed that out of all biased-against groups, victims of bias against disability experienced the largest percentage of violent attacks against the person (compared to “mere” threats and attacks against property): violent personal attacks represented a staggering 67% of all three types of crime mentioned.

AGE Platform Europe reports witnessing an unprecedented amount of ageism and hate speech since the outbreak of the COVID-19,³⁰ which has been condemned also by United Nations’ Independent Expert on the Enjoyment of all Human Rights by Older Persons.³¹ FRA’s 2020 report, aptly titled “A long way to go for LGBTI Equality”, reveals also the high numbers of violence and harassment to which LGBTI communities are exposed, and among them, in particular, trans and intersex people. Trans and intersex people were more likely to experience physical or sexual violence (17% and 22%, respectively) or harassment (48% and 42%, respectively) in the past five years than LGBTI people in general (11% and 38%).³² The prevalence of such crime adds to the seriousness of hate crime against such communities. While in 2018, sexual orientation was the second most commonly reported ground of hate speech by the civil society organisations monitoring the implementation of the EU Code of Conduct on countering illegal hate speech online (15.6%), in 2019, hate speech on the grounds of sexual orientation rose to the first place (33.1%).³³

Hateful, prejudiced or discriminatory expression committed online can spread even more easily. The ease of access, perceived anonymity lowering inhibitions, and more recently also the increase of time spent online during confinement stimulate it further. Based on its surveys, FRA grimly declared that online hatred “has firmly taken root in European societies.”³⁴ Social media, specifically Facebook, Instagram and YouTube, report a disheartening number of posts and comments that they removed due to hate speech, and still insufficiently.³⁵

Lack of Coherent Data Collection and Gaps

The so-called dark figure of crime is, however, likely to be significant in this area and therefore the actual numbers of such crime much bigger than officially reported, partly owing to underreporting³⁶ but partly also owing to the fact that the data collection on such crime is severely inadequate. In its 2013 report, FRA observed that few EU Member States have mechanisms in place to record hate crime comprehensively: only four recorded a range of bias motivations, types of crimes and characteristics of incidents as well as always published the data. Ten Member States with “good” mechanisms recorded a range of bias motivations and generally published the data, while the largest group, composed of 13 Member States, operated with limited mechanisms of data collection, i.e., record few incidents, a narrow range of bias motivations and often do not publish these data. The latter were considered “failing in their duty to tackle hate crime”.³⁷ Its latest, 2018 report reveals that there has been a bit of improvement, although not much. “Of the 19 EU Member States that publish data on recorded hate crime, only 15 disaggregate these data by different bias motivations. Some states publish specific reports on hate crime, providing information on the circumstances of the offenses, which population groups are most at risk of suffering violent offenses, and levels of satisfaction with the police’s response.”³⁸ Despite FRA’s repeated emphasis on how detailed collection and disaggregation of data on hate crime (at least by bias motivation and by type of crime) is necessary in order to monitor the effectiveness of the police response and to prepare effective and targeted policies as well as how access to, publication and dissemination of the data is important to assure victims and communities that hate crime is taken seriously, to send a message to the public that hate crime is addressed and not tolerated, and to improve transparency,³⁹ the availability and quality of data therefore remain lacking.

Other Prohibited Grounds

Considering that the Framework Decision 2008/913/JHA harmonises the definition of, and criminal penalties for, only some specific forms of hate speech and hate crime (i.e., on the grounds of race, colour, religion, descent and national or ethnic origin), leaving aside other personal and widely protected characteristics such as sex, gender, sexual orientation, age, disability and socio-economic status, the new Commission initiative is a welcome step towards filling this gap in the protection of the individual’s rights.

Since hate speech and hate crime are the logical criminal-law derivation of the prohibition of discrimination, the most extreme, most wrongful and harmful manifestations of bias and discrimination, it follows that the grounds on which discrimination and those on which hate speech and hate crime are based should overlap. For these grounds, one should therefore first turn to existing EU anti-discrimination legislation, from which the following grounds can be sourced. Primary legislation expressly mentions nationality (Article 18 TFEU), sex, racial or ethnic origin, religion or belief, disability, age or sexual orientation (Article 19 TFEU) as grounds for prohibited discrimination. Article 21 of the Charter of Fundamental Rights of the EU (“EUCFR”), which became legally binding with the entry into force of the Lisbon Treaty, prohibits “any discrimination based on any ground such as sex, race, colour, ethnic or social origin, genetic features, language, religion or belief, political or any other opinion, membership of a national minority, property, birth, disability, age or sexual orientation” as well as (in the second paragraph) nationality – within the scope of application of the Treaties and without prejudice to any of their specific provisions. There is no legal or logical reason, therefore, to exclude or omit any of those grounds from being prohibited also through the EU hate speech and hate crime legislation.

The Article 21 EUCFR list is similar to that in Article 14 of the European Convention on Human Rights (“ECHR”), which prohibits discrimination on “any ground such as sex, race, colour, language, religion, political or other opinion, national or social origin, association with a national minority, property, birth or other status”. The explicitly mentioned circumstances are in the jurisprudence of the European Court of Human Rights (“ECtHR”) considered “suspect categories” and discrimination on their basis attracts particularly strict scrutiny. Although the ECHR provision is open-ended in terms of grounds, in contrast to Article 21 of EUCFR, while the latter explicitly indicates a few more grounds, the large majority of mentioned grounds are similar. Since all EU Member States are parties to ECHR and ECtHR’s case law is and must be observed by the EU institutions, including CJEU,⁴⁰ it stands to reason that EU anti-discrimination protection, including its prohibited grounds that are directly relevant to the definition of hate crime and hate speech, should not introduce major discrepancies between the two systems.

Even though the Commission’s proposed initiative explicitly mentioned only race, religion, gender or sexuality,⁴¹ it would therefore be legally sound to include all grounds mentioned in the EUCFR – some of which have already been included in the Framework Directive and those that have not been. The public consultation also reveals support for such a wider approach from victim organisations and equality bodies.⁴² Furthermore, in line with societal developments and the EU objective of social inclusiveness or fighting social exclusion (Article 3 TFEU), the inclusion of gender or gender identity – already employed, for example, by ECRI⁴³ – in addition to (the more biological category of) sex would be appropriate.

Disparity in Approaches

Currently, there is a myriad of models tackling hate crime at national level. The main three approaches of incorporating hate crime reforms into the criminal law have been categorised as: (a) the penalty enhancement model (prescribing stricter sanctions for criminal offences carried out with discriminatory motive); (b) the sentence aggravation model (imposing stricter sanctions on perpetrators of crimes carried out with discriminatory motive at sentencing level); and (c) the substantive offence model (stipulating distinct, stand-alone criminal offences that include prejudice or bias as a distinct element of legal definition of the offence).⁴⁴ Some countries have more than one model of hate crime legislation in force, combining, for example, substantive offences and general or specific, or both, aggravating circumstances.⁴⁵

Furthermore, in the absence of specific provision criminalising hate crime as such or treating bias motivation as an aggravating circumstance, some national courts rely on a criminal law provision of general application (e.g., provisions on sentencing powers) to take into consideration the bias motivation when determining the criminal penalties. In Germany, for example, judges may under the general sentencing principles take into account racist, xenophobic and other discriminatory motives “evidencing contempt for humanity” when determining the amount of sentence to apply against a given offender (Section 46(2) of the Criminal Code (“Strafgesetzbuch – StGB”)).⁴⁶ In Slovenia, a similar general provision exists (Article 49 of the Criminal Code) with the exception that the specific bias motivations are not explicitly mentioned. Motivation as such is stated as a factor in the deliberation on sentencing, which leaves the inclusion of the hateful motive into relevant judicial considerations to the discretion of the individual state prosecutor and the judge.⁴⁷ This also implies, however, that where there is reluctance to recognise hate speech and use the relevant incrimination provision, said discretion is likely to reflect it.

Differences in prohibited grounds result in significant variation of legal protection across the EU. At mentioned above, presently, 21 Member States expressly include sexual orientation in hate speech and/or hate crime legislation as an aggravating factor, while 12 amongst them also include gender identity and two cover sex characteristics.⁴⁸ With regard to hate speech, rules vary in scope and severity; there are differences in the definition of hate speech, in the prohibited grounds, in the conditions for the fulfilment of hate speech incrimination, as well as in the actual implementation and execution of the provision on the ground.⁴⁹ The 2014 Report on the implementation of the Framework Decision 2008/913/JHA has already revealed certain issues involving implementation, such as that a number of Member States have not transposed fully or correctly all the provisions of the Framework Decision, specifically provisions in relation to the offences of denying, condoning and grossly trivialising certain crimes. While the majority of Member States have provisions on incitement to racist and xenophobic violence and hatred, these do not appear to always fully transpose the offences covered by the Framework Decision. The report also observes some gaps in relation to the racist and xenophobic motivation of crimes, the liability of legal persons and jurisdiction.⁵⁰ The lack of standardised definitions and terminology pertaining to hate-based crime among EU states has been flagged as a particular concern, as this is reflected in the formulation of national legislation, leading to general imbalances across European legal frameworks.⁵¹

While Article 83 envisages the use of directives – which means that the manner of implementation is left to Member States (although clearly criminal law penalties must be used)⁵² – nevertheless, further EU involvement in this area is likely to reduce certain discrepancies in Member States’ regulation of such crime, for example, in the protected personal characteristics. The Commission’s initiative can be seen as a practical path to aligning definitions and sanctions for hate speech and hate crime in EU Member States. Why is it necessary to ensure that the same behaviour constitutes an offence in all Member States? This question is of course particularly relevant in view of the principle of subsidiarity, which must be given special attention in the area of criminal law. Apart from the usual arguments about differences in the Member States’ criminal laws providing criminals with advantage by enabling them to select the most lenient jurisdiction (forum shopping) or by making the investigation and prosecution of crime more complicated, and constituting barriers to international judicial cooperation,⁵³ it could be argued that certain crimes, such as those violating the most basic EU values (Article 2 TEU), involve particular (and commonly agreed) wrongfulness,⁵⁴ which justifies a common EU action – serving also as a message of EU Member States’ unity in fighting such crime throughout the EU with equal conviction and determination. This is a normative consideration or justification for a common approach that supersedes arguments from effectiveness, and those of trust.⁵⁵ The legislative proposal following upon the Commission initiative – if approved by the Council – could, moreover, correctively address the issues of the lack of accurate data on different types of hate crime and hate speech by requiring from Member States to gather data systematically, consistently and coherently, as well as join all prohibited bias grounds in one place, contributing to the coherence of EU law.

Additional Challenges Related to the Criminalisation of Hate Speech

Freedom of Expression and Historical Experience

Apart from existing gaps and challenges in the hate crime/speech criminalisation across EU, there is – with respect to hate speech – also the issue of a seemingly uneasy relationship between hate speech prohibition and freedom of expression, which is considered a cornerstone of democracies.⁵⁶ However, as Article 10 ECHR emphasises, the exercise of this freedom “carries with it duties and responsibilities” – a reminder that is unique to this freedom – and it may be limited (subject to formalities, conditions, restrictions or penalties as are prescribed by law and are necessary in a democratic society) on a number of grounds, namely, in the interests of national security, territorial integrity or public safety, for the prevention of disorder or crime, for the protection of health or morals, for the protection of the reputation or rights of others, for preventing the disclosure of information received in confidence, or for maintaining the authority and impartiality of the judiciary.⁵⁷ ECtHR tends to approach hate speech in two main ways;⁵⁸ either through Article 17 ECHR (prohibition of abuse of rights), declaring the applications involving certain types of hateful expression (e.g., negationism, revisionism, expression presenting a threat to democracy) to be manifestly unfounded and therefore inadmissible, or through the use of Article 10 ECHR (freedom of expression), examining contextually whether the expression may have been lawfully and legitimately limited in view of the aims pursued as well as necessary in a democratic society, i.e., corresponding to the pressing social need and proportionate to the legitimate aims pursued.

Although freedom of expression is not an absolute right in the European context (in contrast to the US legal system and the First Amendment to the US Constitution, where free speech is the closest to an absolute right, with exceptions thereto being extremely limited) as well as in many others,⁵⁹ the argument from freedom of expression is a powerful tool used to counter limitations on speech. This seems particularly so in post-socialist and post-communist systems, where those often engaging in hate speech are quick to point out that any limitation on their speech would be undemocratic and reminiscent of former regimes (in former Yugoslavia, for example, the crime of “verbal offence” (verbalni delikt) has been misused by the state to deal with political opponents).⁶⁰ This explains, for example, also a certain hesitancy on the part of prosecution and courts in those countries to use the existing hate speech incrimination or, when using it, to interpret it very narrowly. While such concern may still be valid, should political orientation or opinion be included as a grounds for prohibition, it is worth bearing in mind that voices criticising politicians for their views, and especially criticising the government (i.e., the powerful party in this equation), would not qualify under the hate speech criminalisation as currently tentatively suggested by the Commission, since political opinion is not one of the prohibited grounds that are presently being considered for inclusion, and would therefore fail to fulfil one of the necessary elements of the offence. More importantly, such speech is likely to lack the prejudicial, hateful or discriminatory intent most such incriminations require.⁶¹ This, however, also raises the question of the extent to which definitions of hate speech should be asymmetric, taking into account the difference between speech directed against the powerless and that used by the powerless against the powerful. Hate speech regularly involves members of powerful groups speaking against members of vulnerable, often minority groups, and such expression is – from the position of power – particularly harmful, even though this dimension is seldom explicitly considered a necessary element of hate speech. However, such contextual variables are important to consider, at least at the sentencing stage, to accommodate particular vulnerabilities of certain groups and prevent abuse.

This resistance or reluctance to prosecute hate speech in certain parts of European Union, to the extent that this is based on historical grounds, may be lessened if EU is to take action, as proposed, clearly sending a message that criminalising and prosecuting hate speech and hate crime is very much part and parcel of liberal democratic regimes. Furthermore, arguments have been made that hate speech regulation can promote, rather than limit free speech by “lower[ing] the voices of some in order to hear the voices of others”,⁶² thereby balancing out, or compensating for, inequalities in access to platforms in the exercise of the freedom of expression and in participation in public communication.

Endangerment or Violation of Public Order

Another important issue is linked to tying the elements of hate speech to public order violations or endangerment. Some existing criminalisations or interpretations of criminalisations include a violation (or at minimum endangerment) of public order as an element of the offence, a necessary condition to be fulfilled in order to meet the criteria of hate speech.⁶³ Should such an element not be present in the actual manifestation of impeached conduct, the act in question would not be acknowledged as a criminal offence of hate speech, and consequently not prosecuted.

This view is controversial if not problematic in view of today’s knowledge on the impact, harms and victims of hate speech. The Framework Decision stipulates that Member States may choose to punish only conduct which is either carried out in a manner likely to disturb public order or which is threatening, abusive or insulting (Article 1, para. 2). While it therefore allows different options to Member States, for example, to punish the relevant conduct executed only in these two ways⁶⁴ or to choose between the two mentioned ways (e.g., criminalise conduct only if it is carried out in a manner likely to disturb public order or, alternatively, only if it is executed in the second mentioned way), the reports by the UN’s Committee on the Elimination of Racial Discrimination (CERD) criticise states linking criminalisation of such conduct to public order.⁶⁵

There has been an observable shift in the perception and understanding of harms of hate speech, which is undoubtedly connected also to the societal sensibilisation to long-invisible or less visible victimhoods, to voices of minorities and to the increased general knowledge, based on research on the dynamics and impact of such conduct. The primary legal good (or legal interest, Rechstgut) violated by hate speech is human dignity, i.e., dignity of the victimised person, and the primary harm therefore the harm arising out of violation of this legal good. Consequently, this needs to be appropriately recognised by criminal law and become (where this is not already the case) the main object of protection and the main reason that prima facie legitimises criminal law intervention – and not any public order disturbance.⁶⁶

Harm or Offence?

An additional consideration for legitimate criminalisation in criminalisation theory is whether the conduct to be prohibited entails harm or offence, that is, is it harmful or (merely) offensive? While harm requires a wrongful set-back to an interest (i.e., resource over which one has a legitimate claim), offence, i.e., offensive conduct, wrongfully causes offended mental state or psychological discomfort in others – a state that is mostly temporary and does not involve any set-back of an interest.⁶⁷ The EU law, however, refers only to the “seriousness” of conduct, e.g., “areas of particularly serious crime” (Article 83 TFEU). It does not seem to require that seriousness be linked to criminal “harm”, which implies that “serious offence” may be sufficient. Harm as such is not mentioned in the Commission’s 2011 Communication that lays out its views on a “coherent and consistent EU Criminal Policy by setting out how the EU should use criminal law to ensure the effective implementation of EU policies”⁶⁸ either, although it can perhaps be read into it through the requirement to obtain “clear factual evidence about the nature or effects of the crime in question” prior to criminalisation. The policy document on model criminal-law provisions produced by the Council, however, clearly refers to “harm”,⁶⁹ while the European Parliament opted for “damage”.⁷⁰

Despite the primary legislation not invoking harm directly, thereby allowing the possibility of criminalising offence as well, the latter is more controversial in the modern criminal law systems. In the Enlightenment-led model of liberal democracy, the norm so far has been to criminalise only the most morally reprehensible and seriously harmful conduct, while leaving the merely unpleasant, offensive, uncivil behaviour to be censured through informal social control. Hate crime criminalisation may thus, at first blush, be considered easier to justify than hate speech criminalisation, considering that with hate crime the predicate conduct is already, unequivocally recognised as a crime – criminal harm, while some might argue that hate speech is nothing more than “offence”, e.g., a type of insult. Nevertheless, extant research, as will be shown, demonstrates that hate speech includes proper harm-to-others,⁷¹ therefore any objections to its criminalisation that may be raised on the basis of its “mere offensiveness” – which, one may argue, should be tolerated in a pluralist society (cf Handyside v. UK)⁷² – do not hold much weight.

Article 83(1) Conditions for Inclusion of New EU Crimes on the List

We shall now turn to assessing whether hate crime and hate speech meet the requirements set out in Article 83(1) TFEU, namely: areas of crime that are particularly serious and entail cross-border dimension, and whose criminalisation is justified on the basis of developments in crime.

Area of Crime

Article 83(1) TFEU speaks of “areas of crime”, and the Commission’s initiative aims at adding hate crime and hate speech to the list as “other areas of crime”.⁷³ While hate crime can be considered an “area”, encompassing a myriad of criminal offences (predicate crimes) that have been committed with a hate motive, one may wonder whether hate speech is as such a crime “area”. It seems to be a specific criminal offence, rather than an “area of crime”. Although this should not present a problem, objections that may be raised in this regard could be addressed in two ways.

One, more systemic, way would be to suggest to include hate speech as a single offence under a more general crime category or area of “hate crime”. While, in literature, they are often assessed separately⁷⁴ – mostly, it seems, in order to avoid freedom of expression issues in justifying the criminalisation – an argument could be made that hate speech is part of hate crime. Similar to the case of murder which when committed out of hateful motives or prejudice towards a member of the group with protected attributes becomes a hate crime, hate speech could be considered as seriously “harmfully offensive” or harmful speech (often already criminalised in some way in many jurisdictions, e.g., as an insult or defamation) committed with the relevant hate-crime-related elements.

Alternatively, one could argue that the description of EU crimes listed in Article 83(1) as “areas of crime” is merely there to allow the criminalisation of wider harmful conduct that is in some substantive and relevant way linked to the mentioned crimes, but that it does not exclude or prevent any of the therein listed crimes (crime areas) from constituting a criminal offence in itself. Terrorism, for example, is one of the listed crime areas, allowing the criminalisation (or setting of minimum rules…) of terrorism-related activities, such as terrorism financing and so forth, yet it is also a criminal offence in its own right. Similar applies, for example, to money laundering⁷⁵ and many, if not all, other Article 83(1)-listed crime areas.

In the initiative that was later adopted in the form of Communication, however, the Commission decided to argue that hate speech and hate crime are not “areas of crime”, as suggested initially, but one single area of crime, as hate speech and hate crime share an intrinsic feature, i.e., the element of hatred, targeting persons or groups of persons sharing or perceived as sharing the same protected characteristic(s).⁷⁶ While this categorisation of crime based on the underlying emotion (hatred, in this case) stimulating offences, rather than on the targeted object, protected legal value/good or conduct itself, is somewhat unusual and potentially raises some questions (one may wonder, for example, whether terrorism would not then qualify under this area as well), it is certainly an interesting way of organising crime into crime areas.

Particular Seriousness (of Harm)

Any criminalisation exercise, be it national or European, should start with the question: is this conduct seriously harmful to others? In criminal law theory, this is known as the harm principle, originally espoused by Mill (as a “liberty principle”) and phrased by Feinberg as stating that: “[i]t is always a good reason in support of penal legislation that it would be effective in preventing (eliminating, reducing) harm to persons other than the actor (the one prohibited from acting) and there is no other means that is equally effective at no greater cost to other values”.⁷⁷

The prevailing approach in Article 83(1) – in contrast to Article 83(2)⁷⁸ – can be said to be based on the harm principle, as harm (or “particularly serious crime with cross-border dimension”) evidently transpires through the listed EU crimes and clearly seems to be the main guiding principle here. Similar conduct has already been recognised as harmful and criminalisation-worthy in EU Member States, which have in some way or another already criminalised the listed conduct (or conducts related to the listed crime areas) in their criminal codes.⁷⁹

While large-scale empirical studies on the impact of hate crime and hate speech are still relatively scarce, the existing studies show that the detrimental effects of such crime and speech on the individual as well as on members of his or her community (who share the protected personal attribute for which the victim was singled out) are seriously harmful and long-term, which demonstrates their particular seriousness. Apart from the harm caused to the victim and their community, said crime also impacts on society more generally. The relevant harms could therefore be broadly categorised into three groups: (i) harm to the individual victim (the most inner circle of harm), (ii) harms to indirect victims (i.e., to the individual victim’s community with whom they share the relevant personal characteristic), and (iii) wider society.

Hate crime has been described as “message crime”, devised to convey a message from the offender to the victim and their wider community that they “don’t belong”.⁸⁰ It is the victim’s identity that is being directly targeted by hate crime, which affects the impact and perseverance of the emotional and psychological consequences. Sexual minorities are, for example, exposed to higher health risks for anxiety and depression, obesity, higher rates of some forms of cancer, risky drinking patterns, especially among sexual minority women and transgender individuals, which only increases their vulnerability.⁸¹ Prejudice events can be more harmful for their physical health than general life stressors that are not motivated by prejudice.⁸² Harmful consequences, moreover, extend to the victim’s group or community that shares the same personal characteristics as the immediate victim. Some describe these as “in terrorem effects” of hate crime, noting that members of such groups display strikingly similar patterns of emotional and behavioural responses to those experienced by the proximal or immediate victim.⁸³ FRA emphasises that “[h]ate-motivated harassment and violence chip away at people’s trust in public institutions and undermine feelings of attachment to their country of residence”, which impedes social integration.⁸⁴ Wider social costs include the spreading of nationalism, racism, sexism, xenophobia and leading to other hate crime, violence and neo-fascism.⁸⁵ Young people, ethnic minorities, persons with limitations in their usual activities (due to a health problem or disability), lesbian, gay or bisexual people or those who identify in another way, experience physical violence at higher rates.⁸⁶ Hate crime also undermines social cohesion and social stability by creating civil strife between groups.⁸⁷ It undercuts fundamental values of a liberal democratic system, e.g., tolerance, pluralism and the “public good of inclusiveness”.⁸⁸

On the one hand, the EU should be careful not to fall into the trap of casting too wide a net of social control when contemplating criminalisation. While hate crime and hate speech are complex phenomena, rendered particularly salient or exacerbated by today’s social phenomena such as fake news, populism, rule of law and democracy crises, extremism and radicalisation, the focus should remain primarily on the first (and potentially the second) group of victims who can be considered immediate victims, i.e., the ones experiencing immediate harm. The immediate, direct harm must be the one justifying any criminalisation and consequently EU action under Article 83 TFEU. Further away, the issues of legitimacy of criminalisation with regard to “remote” harms (including problems of imputation, causation and intervening acts) come into play.⁸⁹ While such more distant harm is not negligible and can support action against socially undesirable conduct, it alone would not justify punitive state or supranational response.

On the other hand, a particular consideration should be given to political hate speech. Whereas the Commission’s Roadmap on the initiative to extend the list of EU crimes to hate speech and hate crime does not pay particular attention to political hate speech as such, the latter is particularly insidious and detrimental due to its impact, e.g., for normalising such discourse, de-sensitising the public, and inspiring it to speak and act in the same way. Even though, in principle, it may be difficult to legislatively address such hate speech dissimilarly to, and particularly more harshly than, other types of hate speech – as political speech is privileged in the ECtHR case law⁹⁰ – concerns regarding this type of hate speech are valid, and increasingly raised.⁹¹

Cross-Border Dimension

According to Article 83(1) TFEU, the cross-border dimension of a crime can result either from the nature or impact of such criminal offences, or from a special need to combat them on a common basis.

The cross-border requirement may seem, at first sight, the weakest link here. Neither the nature nor the impact of all hate crime and all hate speech is inherently or intrinsically of transnational character; however, it can be and to a significant extent is, particularly its impact. Examples of cross-border nature and impact of hate crime and hate speech include prejudice-motivated offences directed towards immigrants or non-nationals who enter a Member State of hate speech perpetrators. The non-national victim is here a recipient of criminal services – to borrow from the freedom-of-services terminology – that adds the cross-border element to hate crime committed against them. In the case of online hate speech, the cross-border dimension is obvious. Considering the large proportion of such crime is performed in cyberspace that knows no borders, its nature and impact are clearly or typically transnational. When xenophobic hate speech is not migrating across borders,⁹² it is often targeting those across borders and those within them who have foreign roots or ancestors of foreign origin. Such hate speech has impact not only on affected communities inside but also those outside the country, through communication between the two. Furthermore, it may inspire or encourage similar crimes in other Member States.

Increased hate crime and hate speech could also have the effect of discouraging free movement, one of the four freedoms of EU law, to those EU Member States. The noticeable increase in racist and nationalistic hate speech and hate crime around the time of the Brexit referendum (while the UK was still a member of the EU) reduced the attractiveness of UK for EU nationals who already lived and worked there, affecting their motivation to stay.⁹³ Hate speech against LGBTQI+, even more so if emerging from official channels, has also a chilling effect on members of these communities who are considering moving to such a state. It thus impedes free movement of such individuals to EU Member States that are perceived hostile to a particular gender, gender identity or sex orientation.⁹⁴ The free movement of persons entails that citizens of different nationalities travel and work in countries that they may not be nationals of, which increases opportunities for offenders to act upon their prejudice, e.g., xenophobia, and engage in hateful expression. Moreover, increasing social, national, religious and ethnic diversity is likely to provide further opportunities for confronting and dealing (positively and negatively) with prejudices of all sorts.

The existing reports on a notable difference in Member States’ legal regimes with regard to hate crime and hate speech implementation highlight the fact that there is a special need to combat such crimes on a common basis, particularly as they touch upon core EU values. The Framework Directive implementation report as well as various FRA surveys, in addition to reports by other international organisations mentioned above, demonstrate that such crime is not effectively tackled by Member States individually. The issue of prosecutorial or judicial reluctance to process such crime also substantiates the claim that there is a need to combat hate crime and hate speech on a common basis, including requiring accurate, consistent and transparent data on the prevalence of specific types (subcategories) of hate crime and hate speech, broken down by exact bias motives in all Member States. Moreover, in view of the Framework Decision’s limited prohibited grounds and discrepancy in the criminalisation of hate crime and hate speech in Member States, there is also a need to rectify the picture of equality protection by filling the existing legal gaps as regards the protection of victims, i.e., individuals and groups that are traditionally disfavoured and discriminated against – which is supported by the above-presented empirical data provided by victim support organisations.

Developments in Crime

The extension of the Eurocrimes list is possible on the basis of developments of crime, providing that conditions of Article 83(1), subparagraph 1, relating to the substance and type of crime are met. Statistical data on various types of hate crime and hate speech are sometimes difficult to obtain, as the data is often unsegregated by specific bias motivation. FRA has noted in its 2018 report that only 15 of then 28 Member States break down their hate crime data on the various biases behind hate crime, despite the European Court of Human Rights consistently holding that Article 14 ECHR imposes a positive obligation on state authorities to render visible the bias motivation of a crime.⁹⁵

Even though, as a general rule, one has to be careful about drawing conclusions about the extent of crime merely on the basis of the recorded increases of that crime, as the latter may also result from improvements in police recording or be “indicative of a commitment among Member States to combat hate crimes, including through enhanced data collection systems”,⁹⁶ available data, presented above (in “Incidence, Prevalence and Increase in Hate Crime and Hate Speech” section), reveal not only the prevalence but also increases in certain hate crime and hate speech.

It is also likely, however, that what really or even more significantly “developed” is the societal awareness of, and knowledge on, harm generated by hate crime and hate speech, not (only) developments in the criminal phenomena as such. However, developments in the societal realisation of the harmfulness of such crime – which is an essential characteristic of such crime – also warrant legislative changes, as the laws should reflect societal changes, including changes in societal sensitivities. In his theory of the civilising process, Norbert Elias posited a long-term trend toward declining acceptance of violence, brutality, and public suffering,⁹⁷ and this may also partly explain the criminalisation of hate speech and hate crime today.

Conclusion

The current EU crimes list, largely corresponding to the areas that were subject of third-pillar legislation, conspicuously left out racism and xenophobia. The present Commission initiative plans can be seen as the opportunity to correct this omission. While there are some procedural hurdles to overcome⁹⁸ and while some Member States may be hesitant to embrace the proposal wholeheartedly (in general, and particularly on certain grounds),⁹⁹ it has been argued that the Commission’s decision to carry on with the proposed initiative is justified in view of the particular seriousness of hate crime and hate speech harms, their cross-border impact as well as developments in crime, amounting to the pressing need and appropriate timing for EU action.

The Commission’s ambitious attempt to launch this initiative is thus to be welcomed. Many EU countries face increased hate speech, including political hate speech that normalises and stimulates further discriminatory policies and actions.¹⁰⁰ The same may be true of hate crime, in view of the current data and even more so, as ODIHR warns, in view of the fact that the lack of hate crime recording means victims and their needs too often remain invisible.¹⁰¹ The EU is legally committed to tackling discrimination – combatting social exclusion and discrimination being among EU objectives (Article 3 TFEU) – and this entails, as a logical consequence, the prohibition of most egregious forms of attacks against the human dignity of members of groups who are discriminated against on the basis of certain protected characteristics.

Seen from a wider perspective in which this crime occurs or is particularly boosted – rule of law concerns, populist governments, illiberalism, health and economic crises – the need to tackle the problem of hate crime and hate speech on a common basis seems rather difficult to dispute. Such crime goes to the heart of European values and EU public goods,¹⁰² attacking the identity and essence of Europe or, in a sense, the true “European way of life”. Several EU values, as stipulated in Article 2 of the TEU, specifically, the respect for human dignity, freedom, equality, the rule of law and respect of human rights, including the rights of persons belonging to minorities, are intimately connected with the prohibition of hate crime and hate speech. Hate crime and speech legislation – with its clear linkages to discrimination – would provide an additional arsenal to the prohibition of discrimination in other areas of social life. In a society where “pluralism, non-discrimination, tolerance, justice, solidarity and equality between women and men prevail” (Article 2 TEU), there really ought to be no place for immunity against such crime. Commitment to these values requires from the Union and its Member States to conduct an active struggle against hate crime and hate speech. It would of course be preferable, as Rosenfeld argues, “if hate could be defeated by reason. But since unfortunately that has failed all too often, there seems no alternative but to combat hate speech through regulation in order to secure a minimum of civility in the public arena”.¹⁰³

This does not mean, however, that criminalisation is always the answer, the best answer or the only answer. In addition to common legal definitions of hate crime and hate speech, and minimum rules on the definition of offences and penalties, support action should include public awareness raising, incentivising informal social control, robust and detailed data collection, support with administrative law¹⁰⁴ (for minor infractions), relevant staff training, and regular monitoring of implementation and actual execution on the ground. Yet, with hate speech now being able to spread throughout the world in a split second and with nations becoming ever more socially, ethnically, religiously and culturally diverse, the need for regulation becomes increasingly urgent.¹⁰⁵ Moreover, the symbolic or expressive¹⁰⁶ dimension of criminal law is not to be neglected. While in itself, this characteristic of criminal law is insufficient to justify criminalisation, as the latter should primarily be based on sufficiently serious and wrongful harmfulness of the conduct in question,¹⁰⁷ it adds to the gravity of official condemnation of such conduct, which is an important message to the victims of such crimes that they and their victimisation are accorded sufficient weight and due consideration.