The article is here; the Introduction:
In Germany, an artist and activist was arrested multiple times, once for wearing a Palestinian flag and again for holding a sign that read "from the river to the sea, we demand equality." In France, a mayor was held criminally liable, not for what he said, but for failing to remove Islamophobic comments posted by others beneath one of his Facebook posts. In Finland, a member of Parliament faced criminal charges for criticizing her church's support for gay pride events, citing a Bible verse.
Such cases are not exceptional among European Union member states. In many places, they are standard practice.
Yet despite already possessing some of the democratic world's most expansive and actively enforced hate‑speech laws, the European Commission has proposed adding hate speech to the EU's list of "EU crimes" under Article 83(1) of the Treaty on the Functioning of the European Union. In 2020, the European Commission (Commission) President Ursula von der Leyen proposed making hate speech and hate crime European Union (EU)-level offenses. The proposal is framed as part of the Union's broader efforts to promote equality, inclusion, and non-discrimination. Yet it also forms part of a wider trend within EU institutions and Member States, namely a growing belief that expansive restrictions on expression are necessary to protect democratic values, counter extremism, and address social harms.
Since 2021, the Commission has reaffirmed the initiative in formal communications and public statements, with strong support from the European Parliament and other EU bodies, although unanimity in the Council of the European Union has not yet been secured. In its 2026–2030 Anti-Racism Strategy, the Commission signaled that, given the lack of progress, it is considering a legislative initiative to harmonize definitions of online hate offences by drawing on existing areas of crime covered by Article 83(1) of the Treaty on the Functioning of the European Union (TFEU). The shift from treaty expansion to reliance on the existing legal authority of Article 83 reflects not a retreat from harmonization, but a change in legal technique. Where unanimity failed, pursuing similar objectives through digital or computer-crime competences warrants scrutiny as a potential competence-stretching bypassing Member State resistance.
At the time of this report, the proposal has stalled due to a lack of unanimity in the European Council. But the Commission has signaled a narrower, "online‑only" route: a legislative initiative under existing Article 83(1) crime categories (e.g., computer crime) to harmonize definitions of online hate offences. In practice, this would still embed criminal prohibitions more deeply into national criminal law and EU digital enforcement frameworks. It also raises questions about whether speech that might be considered legal when said online suddenly becomes subject to criminal penalties if posted online.
The Commission seeks to justify the new initiative by pointing to an increase in hate speech and hate crimes across the EU. It also highlights divergent national definitions and enforcement practices, arguing that fragmentation undermines victim protection, creates enforcement gaps, and sends "mixed messages" about the seriousness of hate speech. It thus frames the proposal not as a novel expansion but as the logical next step in building a coherent European response that would enable the EU to address serious forms of hate speech more effectively, particularly online.
While it may be unclear how a EU-wide criminalization of hate speech will ultimately take shape, the Commission's proposal fails to meet the foundational requirements of legality, necessity, and proportionality under both European and international human rights law. It further finds that the empirical assumptions underpinning the initiative are unsubstantiated and, in some respects, contradicted by available evidence.
Legality: The principle of legality requires that criminal offenses be formulated with sufficient clarity and precision to enable individuals to regulate their conduct and to foresee, to a reasonable degree, the legal consequences of their actions. This requirement is especially stringent in the case of speech, because uncertainty itself chills expression.
The European Commission's proposal offers no conceptualization of "hate speech." Instead, it relies on broad, vague, and varying moral, identity-based, and sociological categories. The European Parliament has further encouraged the adoption of an open-ended list of protected characteristics, creating a system in which virtually any political, social, or ideological group could invoke criminal law to shield itself from alleged hate speech. If such definitional vagueness is adopted in any subsequent directive, then it will be incompatible with legal certainty, invite arbitrary enforcement, and risk chilling democratic debate on topics of public interest across all 27 Member States.
Necessity: Necessity requires showing that criminalization at the EU level is indispensable, and that less intrusive measures are insufficient to address the identified harm.
The Commission has not demonstrated that EU-level criminalization or harmonization of online hate offences is necessary. Member States already rely on extensive criminal and regulatory tools, including national speech offences, the 2008 Framework Decision on combating racism and xenophobia through criminal law, and the Digital Services Act.
Moreover, the political rationales underpinning the initiative have shifted over time, from the COVID-19 pandemic to post-October 7 tensions, suggesting a sense of urgency driven by the prevailing political climate rather than by a demonstrable legislative need. And empirical research provides little evidence that the criminalization of hate speech reduces hatred or violence. The empirical literature does not establish that broader criminalization reduces hatred or violence in stable democracies, and it identifies plausible counterproductive dynamics (polarization, grievance narratives, displacement). Conversely, stronger expressive freedoms are associated with higher tolerance and lower social conflict in democratic settings.
Proportionality: Proportionality requires that restrictions on expression pursue a legitimate aim using the least intrusive means available. Criminal sanctions represent the most severe form of state interference with speech and therefore demand especially compelling justification. Even under the European Court of Human Rights' comparatively deferential approach to national hate speech laws, criminal sanctions must remain proportionate and narrowly tailored.
While the Commission's Communications briefly acknowledge the importance of freedom of expression, they omit core safeguards required under International Human Rights Law, including intent, imminence, and likelihood of harm. The proposal thus falls short of the strict incitement standard set out in Article 20(2) of the International Covenant on Civil and Political Rights, which must conform with the strict requirements for restricting freedom of expression under Article 19. Article 20(2) mandates only the prohibition of any advocacy of national, racial, or religious hatred that constitutes incitement to discrimination, hostility, or violence, as interpreted by the Rabat Plan of Action.
As a result, the proposal risks criminalizing even "shocking, offensive, or disturbing" speech that is protected under Article 10 of the European Convention on Human Rights (the right to freedom of expression). In practice, this could expose journalists, academics, artists, politicians, activists, and others to criminal liability for expression that is currently lawful in many Member States. And because the Digital Services Act defines "illegal content" by reference to EU and national law, an EU-wide regime harmonizing hate speech or online hate offences would expand platform removal and risk mitigation duties at scale, strengthening incentives to over remove lawful but controversial speech—especially where automated systems struggle with context (satire, minority speech, political critique).
An online-only offence would also risk creating a two-tier criminal regime: Speech that is lawful when delivered offline could trigger criminal liability once posted online. Because many contemporary prosecutions concern online comments, Member States would face strong legal and political pressure to align their general criminal codes with the online definition, effectively achieving the criminal law harmonization that some States have resisted. If they do not, the result is unequal treatment and "digital disappearance." For instance, a lecture could be lawful in a hall and reportable in the press yet be removed from social media (and even expose the speaker or person posting the lecture on social media to prosecution) when uploaded.
International Effects: The proposal also risks reshaping global norms. European approaches to online speech and hate-speech regulation already exert significant influence on countries with weak rule-of‑law protections. Laws in Russia, Turkey, Venezuela, and other jurisdictions have explicitly cited European models when criminalizing or suppressing dissent—even when the effect of implementing the European models in the political systems of those countries is much more speech-restrictive than it would be in Europe. Through the so‑called "Brussels Effect," an EU‑level regime harmonizing online hate offences would further entrench this dynamic, irrespective of the European Union's pro‑democratic intentions.
A More Effective and Rights-Compatible Alternative: Considering the broad range of legal tools already available to Member States to address harmful speech, the European Union should prioritize non-punitive interventions such as counterspeech. This method can foster social resilience without compromising the right to freedom of expression, which underpins democratic life and serves as a safeguard against government overreach, even when pursued with well-intentioned aims.
At a time of heightened political polarization and democratic backsliding, the European Union must reaffirm its commitment to open debate, pluralism, and restraint as a cornerstone of democratic society. The Commission's proposal risks normalizing a model of speech governance that suppresses democratic debate rather than strengthening it. The European Union should regard freedom of expression as an essential safeguard of freedom and democracy—not a "risk" to be managed.
* * *
Rather than strengthening democratic resilience, EU-level criminalization of hate speech or harmonization of online hate offences risks entrenching overbroad enforcement, chilling lawful expression, and amplifying the very social tensions it purports to address. Accordingly, the EU should focus on strengthening rather than weakening freedom of expression and ensure that the criminalization of hate speech is used only as a last resort and in ways that complies strictly with the requirements of legality, necessity and proportionality. The EU should also refocus its crucial efforts to combat hatred and discrimination with an increased emphasis on using non-restrictive means such as counterspeech and education.
Our analysis proceeds from the basic premise that all member states of the EU criminalize certain forms of hate speech and that such criminalization, subject to balancing against the human right to freedom of expression, is both required and compatible with EU law and International Human Rights Law. Accordingly, we assume for our purposes that the repeal of European hate speech laws is both legally and politically unfeasible, even if there are compelling normative and empirical arguments in favor of such a maximalist position. Instead, we examine the proposal through the foundational principles governing restrictions on freedom of expression in European and International Human Rights Law (IHRL): legality, necessity, and proportionality. We assess whether EU-level criminalization is compatible with these standards, whether the need for such a proposal is supported by evidence, and the proposal's implications for democratic debate, online expression, and Europe's broader global influence.
In the following Parts, we situate the initiative's institutional trajectory (Part I); assess its legality, necessity, and proportionality under European and international human rights standards (Part II); analyze Europe's existing hate-speech landscape and escalation dynamics, including the DSA multiplier (Part III); test the proposal's empirical premises (Part IV); examine external diffusion through the "Brussels Effect" (Part V); and develop rights-compatible alternatives (Part VI).
