Presenters Jacob Mchangama
Junko Kotani
Jeffrey Howard
Joan Barata

Oscar Pérez de la Fuente/Tom Herrenberg (chair) This workshop examines historical and comparative aspects of the regulation of hate speech. Controversies surrounding provocative expression date back over millennia, even if the phrase “hate speech” entered the legal lexicon only in recent decades. Today, the rise of online media, along with global surges in populist political and social movements, have turned hate speech into one of the world’s most bitterly disputed social problems. Electronic platforms are now dominated by powerful yet opaque multinational corporations, meaning that questions about the extent of free expression no longer arise primarily between individual citizens and their national governments. These rapid changes have le our current regimes of transnational legal regulation looking murky and unpredictable. Experts overwhelmingly agree about the immorality of hate speech, yet disputes persist about its content, about the political and ethical legitimacy of legal limits, and about the practical efficacy of regulating speech. In our world of instant and incessant global communications, are bans on hate speech politically or ethically desirable, and, even if they are, do bans work in practice? Can they end up fanning the flames they were designed to extinguish? Our Special Workshops devoted to discussing (draft) chapters of the Oxford Handbook of Hate Speech, an interdisciplinary project that examines the most pressing issues surrounding hate speech, forthcoming with Oxford University Press in 2025.

Jacob Mchangama Global historical precedents
In the West, hate speech laws are frequently identified with the post-World War Two order as enshrined in both national and international human rights law. Typically, such laws aim to prevent atrocities as witnessed during World War Two by protecting the equality and dignity of minorities from dehumanization and bigotry. Yet the origin and roots of laws against hate speech run deeper than this ethos of “never again.” Various forms of laws against hate speech, as broadly defined, can be traced back to at least Pope Gregory IX´s decree from 1231, which affirmed that heretics were to incur “The Debt of Hatred.” Moreover, from the early modern era and through the 20th century various forms and guises of hate speech laws have been adopted to protect a variety of interests such as government power, religious doctrines, political pares, and the established social order, as well as vulnerable groups and minorities. This chapter will trace the history of hate speech laws and explore differences and commonalities between such laws ancient and new.

Junko Kotani Hate speech regulation in Japan
I discuss Japan's unique legal approach to addressing the harms of hate speech without criminalizing it. It focuses on Japan's reservations to Article 4 of the ICERD and analyzes the Hate Speech Elimination Act, which declares hate speech intolerable but does not criminalize it. The paper also examines the administrative system designed to prevent hate speech and the local ordinances that aim to fill the gaps in these systems.

Jeffrey Howard Moderation by machine: The ethics of mechanized hate speech removal
In this presentation it is argued that if social media platforms have a duty to remove hate speech—itself a matter of substantial controversy—then the use of AI is indispensable given the platforms’ scale. It nevertheless confronts the problem that current AI cannot distinguish hateful from non-hateful speech with great accuracy. In the face of this challenge, should AI be programmed to under-remove content (thereby permtting some hate speech to stand) or to over-remove content (thereby removing some legitimate speech)? The presentation outlines several ways in which theorists of free speech might respond. The most promising model would be a flexible one: AI would moderate overinclusively in emergency contexts where risks of violent conflict are high, as some collateral censorship can be justified as a side-effect of measures preventing harm. However, in normal or non-emergency circumstances, AI should be adjusted to moderate under-inclusively.

Joan Barata The future of regulating hate speech online
Debates about freedom of expression, its limits and the power of both states and private companies to frame its exercise have surpassed the boundaries of academia and the legal community. This chapter analyzes the challenges and risks associated with the protection of the freedom of expression within the context of the traditional relationship between the individual and state authorities within the context of a global crisis affecting the universal understanding of freedom of expression as a human right. In addition to this, and since the Internet is probably the environment where the expression and dissemination of ideas, opinions and information is subject to a growing number of standards, rules and regulations established both by the state and by powerful private subjects, the text also considers the nature and implications of private speech governance, and the extent to which the new regulatory approaches consisting of introducing procedural safeguards and human rights obligations constitute a viable and appropriate solution to frame and preserve the exercise of freedom of expression under the intermediation of dominant online platforms.