Free speech principles and standards in academia: The case of racism

Kovács, Kriszta; Tóth, Gábor Attila Article — Published Version Free speech principles and standards in academia: The case of racism International Journal of Educational Research Provided in Cooperation with: WZB Berlin Social Science Center Suggested Citation: Kovács, Kriszta; Tóth, Gábor Attila (2025) : Free speech principles and standards in academia: The case of racism, International Journal of Educational Research, ISSN 1873-538X, Elsevier, Amsterdam, Vol. 134, pp. 1-10, https://doi.org/10.1016/j.ijer.2025.102828 This Version is available at: https://hdl.handle.net/10419/331213 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. http://creativecommons.org/licenses/by-nc-nd/4.0/ Free speech principles and standards in academia: The case of racism Kriszta Kov´ acs a,* , G´ abor Attila T´ oth b a Senior research fellow, WZB Berlin Social Science Center, Reichpietschufer 50, D-10785 Berlin, Germany, Associate professor, ELTE University Faculty of Social Sciences, P´ azm´ any P. st 1/a, 1117 Budapest, Hungary b Professor, University of Debrecen, Faculty of Law, 4028-Debrecen, Kassai út 26, Hungary ARTICLE INFO Keywords: Academic speech Research Free speech Racism European Convention US jurisprudence ABSTRACT Hostile and untruthful expressions are notably prevalent across the world. This phenomenon also affects academia. Our legally centred article asks whether academic speech and research are protected even if a professor’s statement, research outcome or research agenda could potentially offend people or even cause moral distress to members of socially disadvantaged groups. The focus of the article is on two regions that have long been widely regarded as offering robust free speech protections: the United States and Europe. The article demonstrates that under both regulatory models, courts inevitably face scientific and moral questions when addressing racism-related legal cases in academia. Courts have traditionally shown deference to science, recognising that addressing scientific matters falls outside the judicial purview and instead belongs to the realm of academia. However, in both models, judges tackle moral dilemmas through the language of law. Notwithstanding their differences, both regulatory models have developed legal principles and standards (the ‘clear and present danger’ standard on the one hand, and the balancing of competing constitutional values on the other) that, when interpreted properly, could help avert the two perils that threaten academic freedom: the limitless dissemination of racist ideas in the name of ‘absolute freedom’ and censorship. 1. Introduction Hostile and untruthful expressions are prevalent throughout the world. Systematic manipulations, including ‘fake news’ and ‘alternative’ facts, are a feature of all of our lives. We encounter hate-filled messages, which are not just disseminated and amplified by hostile individuals, who have always been present, but also by contemporary autocrats and even advanced computational systems like AI. For members of socially disadvantaged groups, such forms of manipulation can have particularly profound effects. They often find themselves the targets of racist, xenophobic and homophobic speech and propaganda, including facing cyber harassment by haters or trolls. This is a situation that once again calls for difficult legal choices to be made in order to avoid two equally undesirable alternatives – as has occurred at many times in history. The first option is to expand the boundaries of constitutionally protected speech, which could lead to greater harm, including bullying or even violence, rather than increased liberty. An alternative option is to impose more * Corresponding author. E-mail addresses: [email protected], [email protected] (K. Kov´ acs). Contents lists available at ScienceDirect International Journal of Educational Research journal homepage: www.elsevier.com/locate/ijedures https://doi.org/10.1016/j.ijer.2025.102828 Received 30 April 2025; Received in revised form 1 September 2025; Accepted 3 October 2025 International Journal of Educational Research 134 (2025) 102828 Available online 24 October 2025 0883-0355/© 2025 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license ( http://creativecommons.org/licenses/by-nc-nd/4.0/ ). limitations on free speech, which might result in censorship, reflecting autocratic tendencies that undermine democracy. Is the dilemma the same in the case of academic freedom? The question arises because when contemporary autocrats endorse unconstrained speech, prejudices and misconceptions might surface in academia, thereby giving rise to pseudoscience (G¨ arditz, 2018, 28). Conversely, imposing strict limitations on speech and academic freedom might result in regulations that autocrats utilise to threaten and defund universities or punish scholars for their allegedly racist speech and engagement in campus protests (Volokh et al., 2025; Garber, 2025; Pedota et al., 2025). There are numerous dimensions and issues associated with the intersection of free speech and academic freedom, and many of these have been the subject of recent analytical articles. One such article analyses the various justifications for free speech and assesses their applicability, among others, to (online) racist speech (Bonotti & Seglow, 2021). Other articles have discussed the relationship between academic freedom and free speech on campus and argued that campuses should not be governed by general free speech principles (Simpson & Srinivasan, 2018; Simpson, 2020). A more recent article has proposed moving beyond the debate about speech restrictions by centring campus speech policies based on their practical effects on the virtues essential to the university’s epistemic mission, such as tolerance and open-mindedness (Marcoci & Oprea, 2024). This article examines the legal dimensions and centres exclusively on a particular issue: whether academic freedom provides any level of protection for statements that might qualify as racist speech articulated by scholars and for research projects that might be considered racist. To sharpen its analytical focus, the article limits its scope to constitutional democracies with a certain degree of the rule of law, diverse media and an independent judiciary. It concentrates on the two rival regulatory models that have long been thought to provide robust free speech protections: one model emerges from the safeguards established by courts that interpret the United States Constitution and another model from the European Convention on Human Rights. The article proceeds as follows. After briefly outlining two rival free speech regulatory models and their supporting principles, it addresses the dilemma of whether reconciling competing values in academic freedom cases is the same as in free speech cases and whether the right to academic speech is distinct from the right to free speech. It discusses a European and a US case involving public statements by professors that socially disadvantaged individuals found harmful and uses these statements to highlight the similarities and differences between the two models and determine whether the legal standards applied by courts are appropriate for academic freedom issues. Considering that scholarly work often involves empirical research and not just the public dissemination of the findings, the article also addresses whether the freedom to conduct research applies to research agendas that might be perceived as racist. The article argues that, when properly interpreted, the balancing act supported by both regulatory frameworks is capable of ensuring ample protection for academic freedom, and it could help avert the two perils that threaten academic freedom: the dissemination of racist ideas in the name of ‘absolute freedom’ and outright censorship. 2. Supporting principles Several foundational principles lend support to the concept of freedom of speech. Theoretical perspectives, spanning from Mill’s consequentialism (Mill, 1962) to Kantian deontology (e.g., Baker, 1989), offer principled reasons why different varieties of expression should or should not be limited. Consequentialists contend that as long as people do not inflict harm on anyone else, they should enjoy a degree of personal autonomy to openly express their thoughts. They also add that while pursuing the truth, various false or erroneous ideas may emerge in the marketplace of ideas, including hateful and false ones, but the market can correct itself. In a marketplace of ideas, sound concepts can be separated from those that are fundamentally flawed. Deontologists, following the Kantian approach, address the intrinsic value of speech. They assert that the right to free speech is anchored in dignity and autonomy that is equally shared by all human beings. Free speech is essential, not because it may contribute to the discovery of truth, but because it enables individuals to develop their own views and express them, no matter what these views entail. Following in the footsteps of these great thinkers, theoreticians have reviewed the principles and analysed the hard cases of free speech, including racist speech. They have proposed rather different solutions. Jeremy Waldron has identified the harm in the fact that disadvantaged minorities face threats, humiliation and discrimination. Hence, he has proposed a new argument supporting hate speech restrictions: An attack on the human dignity of disadvantaged groups undermines their inclusion in society, so, Waldron has argued, it should be legally restricted (Waldron, 2012). Likewise, academics who have recently addressed the modern-day relevance of free speech justifications, where disinformation and online racist speech are common, have contended that restrictions on speech might be justified when speech causes harm to others’ autonomy, as occurs in cases of racist speech (Bonotti & Seglow, 2021, 4). The harm caused by certain racist utterances occurs through norm enactment, meaning that the utterances shape social realities that can result in or perpetuate discriminatory outcomes (McGowan, 2021). These scholars support their arguments by making claims about the consequences of racist speech for the targets of this speech. Another influential argument, pointing to the dangers of paternalism and moralism and the fact that there is no correlation between the existence of racist speech laws and ‘lower levels of abusively expressed prejudice about human difference’ (Ash 2016, 220), directs more attention to the imperilled freedom of speech. For example, Timothy Garton Ash, following theories by Ronald Dworkin (Dworkin, 1977) and Edwin Baker (Baker, 1989), argues that no one has a right not to be offended and recommends principles that foster more and also better speech by focusing on the circumstances of the speech when drafting hate speech laws. Ash suggests refraining from legally enforcing civility in our connected but, at the same time, extremely diverse world (Ash, 2016, 224–229). Two regulatory models have been developed based on these theoretical ideas. One regulatory model takes context-based speech regulation as its starting point. The other does not give free speech a privileged status per se but allows content-based restrictions under specific circumstances. In what follows, we briefly sketch out the two rival models for regulating speech that exist in today’s democratic world. K. Kov´ acs and G.A. T´ oth International Journal of Educational Research 134 (2025) 102828 2 3. Limits on free speech 3.1. Context-based speech regulations In certain states, freedom of speech has a privileged status. For example, the First Amendment to the US Constitution prescribes that ‘Congress shall make no law (…) abridging the freedom of speech’ and the court’s task is to interpret its meaning. At first sight, the constitutional text seems to establish an absolute ban, and indeed, Supreme Court justices argued that it is of an absolute nature, referring to the seminal work of the Anglo-American free speech tradition (Blasi, 2018), Milton’s Areopagitica (Milton, 1644). Yet the majority of the US Supreme Court rejected this view by asserting that ‘the prohibition of encroachment of the First Amendment is not absolute. Restraints are permitted for appropriate reasons’ (L Elrod v. Burns (1976) 427 U.S. 347, 360). What are those appropriate reasons? The underlying idea of this regulatory model is that even extremist messages should not be banned merely based on their content. The US courts have long allowed only context-based speech restrictions, where the manner, place and/or time of expression might justify legal intervention. After the end of WWI, two Supreme Court justices elaborated on the clear and present danger test, according to which hate speech can only be legally prohibited if it poses a direct threat by inciting hatred or violence (Abrams v. United States, 250 U.S. 616 (1919) Justice Oliver Wendell Holmes and Justice Louis Brandeis believed the state is a greater threat to individual freedom than vice versa.). The test was refined later in the case Brandenburg v. Ohio, 395 U.S. 444 (1969), where the Supreme Court held that speech can only be restricted if it is likely to incite or produce imminent lawless action (Lewis, 1991). Therefore, to restrict speech, the court has long demanded lawmakers to demonstrate the likelihood that the action will occur, the speaker’s intention to incite that action, and the imminence and lawlessness of the action (Sunstein, 2024). In applying this test, the judges of the Supreme Court found that the constitutional right to free speech covers, among other actions, the right to use swastikas in a march (National Socialist Party of America v. Village of Skokie, 432 U.S. 43 (1977), to curse gay people in public (Snyder v. Phelps, 131 S.Ct. 1207 (2011), to deny the Holocaust (Schauer, 2012) or to burn the American flag in a demonstrative manner (United States v. Eichman, 496 U.S. 310, 319 (1990). Certain categories of speech are not protected because of the circumstances of the speech. For example, cross-burning that symbolises hatred of Black people is not constitutionally protected speech when it happens in a Black family’s backyard because it is intimidating (Virginia v. Black, 538 U.S. 343, 367 (2003). Moreover, ‘fighting words’, that is, when a speaker uses words in an intolerable way to incite violence, are also not protected because, in this case, there is in effect no speech; the speaker is just emitting sound vibrations that disturb others (R.A.V. v. City of St. Paul, 505 U.S. 377, 393 (1992) and see Meiklejohn, 1948). Even with the profound and rapid changes in the direction of the current US Supreme Court’s jurisprudence, the Court has, to date, only determined whether an expression has protected status after evaluating the circumstances surrounding that expression. For instance, the Court did not outlaw cross burning as such, but it found that if it had the potential to intimidate specific persons, this symbolic expression no longer received constitutional protection. 3.2. Content-based speech regulations The rival regulatory model, which prevails in certain international treaties and in specific countries in Europe, does not just require the context-based restrictions mentioned above but also permits content-based restrictions in the name of democratic values and human dignity, thereby permitting speech to be limited based on its content, notwithstanding the circumstances. Article 4 of the International Convention on the Elimination of All Forms of Racial Discrimination expressly requires states to make it an offense punishable by law to disseminate ‘ideas based on racial superiority or hatred’. There is no such provision in the European Convention on Human Rights. Yet the case law of the European Court of Human Rights associated with Article 10 reveals that the protection of this free speech provision does not cover particular statements. For instance, statements that publicly advocate national socialism aiming at impairing the basic order of freedom and democracy (Kühnen v. Federal Republic of Germany, 12194/86, Decision of 12 May 1988), deny the existence of gas chambers in concentration camps under the National Socialist regime (Honsik v. Austria, 25062/94, Decision of 27 February 1997), agitate against members of the LGBTQ community (Vejdeland and Others v. Sweden, 1813/ 07, Judgment of 9 February 2012), or incite religious hatred (Norwood v. United Kingdom, 23131/03, Decision of 16 November 2004) are not protected under Article 10 because they are incompatible with the values guaranteed by the Convention, notably tolerance, social peace and non-discrimination (Article 17 of the Convention). Nevertheless, these restrictions do not constitute comprehensive prohibitions like the aforementioned Article 4 of the International Convention; they emerge after an examination of the potential impacts. Judges of the European Court of Human Rights balance free speech against competing values such as other people’s human dignity and privacy, public security or enforcement of conventional morality. They examine whether these interests can be considered legitimate grounds for restricting free speech. And as the principle of proportionality requires, courts demand valid justifications to clarify whether a given restriction of free speech is suitable, effective and necessary to achieve a legitimate aim. The fact that the circumstances of the speech matter is even more relevant in the context of EU law. The EU Charter of Fundamental Rights, through its Article 11, grants significant protection for the right to freedom of expression. A Council Framework Decision requires the EU Member States to punish instances of ‘publicly inciting to violence or hatred directed against a group of persons or a member of such a group defined by reference to race, colour, religion, descent or national or ethnic origin’, it allows them to punish ‘only conduct which is either carried out in a manner likely to disturb public order or which is threatening, abusive or insulting’ (The Council of the European Union, 2008). So, the Framework Decision requires the Members States to punish expressions that publicly incite to violance and hatred and it does so by allowing them to implement it as a context-based restriction, focusing on the K. Kov´ acs and G.A. T´ oth International Journal of Educational Research 134 (2025) 102828 3 circumstances of the speech. Although there is a conceptual difference between context-based and content-based restrictions of free speech, in practice, the judicial application of the two regulatory frameworks is very similar. This is because a purely content-based restriction either would not work or would lead to absurd results. Legal restrictions cannot per se apply to communicating specific content but rather its dissemination to the broader public. Holocaust denial is a criminal offence in certain European countries, such as Germany (§130 Die ‘Auschwitzlüge’ Strafgesetzbuch) and according to the European Convention, the negation or revision of ‘clearly established historical facts – such as the Holocaust’ is not protected by Article 10 because of Article 17 (Chauvy and Others v France, Appl no 64915/01, Judgment of 29 June 2004, para 69). But even in this case, individuals who deny the existence of the gas chambers in their local pub will not be prosecuted. Furthermore, educators can address the issue of Holocaust denial as part of their curriculum. The issue can also be present at academic events for historians and literary scholars and may also be reflected on in artistic forms of communication. This approach requires European courts to decide in specific cases whether the speaker intended to inform the public on a matter of public interest or whether the speaker sought to disturb the public order. Hence, the balancing act performed by European courts between free speech and other democratic values is not far from the one US judges have applied when deciding if racist speech, under certain circumstances, poses a clear and present danger and can thus be restricted. Can a similar balancing exercise be applied to academic speech? 4. Limits on academic freedom 4.1. Academic speech The academic world is not immune to prejudices and misconceptions that are inconsistent with the foundational values and principles of constitutional democracy, most importantly the human dignity of free and equal people. Academic publications and statements may, at times, reflect exclusionary attitudes or convey hateful messages that may harm those in disadvantaged social groups. Does academic freedom provide protection for these messages? What is distinctive about academic freedom as a right compared to the right to freedom of speech? Both in the US and Europe, the definition and scope of academic freedom can be inferred from the judicial case law rather than from positive law itself. This is because historical human rights documents, including the US Constitution (Kumm, 2024) and the European Convention on Human Rights, do not make explicit reference to academic freedom (Kov´ acs, 2025). Although Article 13 of the 2009 EU Charter of Fundamental Rights explicitly protects academic freedom, individuals do not have the right to turn directly to the Court of Justice of the European Union if they feel their Charter rights are violated. Therefore, the focus of the next section is on the jurisprudence of the European Court of Human Rights, where academic freedom takes the form of an individual claim. This is what is comparable to the relevant US jurisprudence. The free speech case law developed by both the European and US judges acknowledges that academic freedom is a distinct branch of freedom of speech. Academic speech may take the form of intramural speech, that is, speech inside the walls of the university, including classroom speech, and extramural speech, that is, outside-the-university speech, typically statements and publications made by professors. The meaning of the ‘wall’ metaphor is perceived somewhat differently in the European and US contexts, given the metaphor’s inherent fluidity (Butler, 2025). 4.1.1. The European court of human rights jurisprudence protects academic freedom under article 10 of the convention (Beiter et al., 2023). This protection includes the academic’s intramural expression: the freedom to criticise the academic system in which they work (Sorguç v. Turkey, Appl no 17089/03, Judgment of 23 June 2009) and classroom speech. Although there are no cases that directly address classroom speech, in the case of Vogt v. Germany (Appl no 17851/91, Judgment of 26 September 1995), the Court invoked the principle that disciplinary sanctions for teachers’ expressive activities need to be carefully balanced against their speech-related rights. The protection of academic freedom also extends to the academics’ extramural expression. According to the Strasbourg case law, it is within the academic’s extramural freedom ‘to express freely their views and opinions, even if controversial or unpopular, in the areas of their research, professional expertise and competence’ (Mustafa Erdogan and Others v. Turkey, Appl no 346/04 and 39779/04, Judgment of 27 May 2014, para 40). How much protection is granted to an academic who wishes to freely speak outside of academia depends on whether the subject matter is a public concern and whether the statement falls within the speaker’s expertise. The Aksu case illustrates the extent to which Article 10 protects an academic’s extramural expression (Aksu v. Turkey, Appl no 4149/04, 41029/04, Judgment of 15 March 2012). The book titled Gypsies of Turkey (Türkiye Çingeneleri) written by Associate Professor Ali Rafet ¨ Ozkan, was at the heart of the court proceeding. In the introduction to the book, partly funded by the government, the professor explained the aim of the study and the methodology used during the research in the following way: ‘This study – which I present without any pretensions, but merely in a bid to fill a significant gap (it being the first study of its kind), and to provide guidance to others working on the Gypsies in the future – was prepared using descriptive, comparative and phenomenological methods, in addition to participant observation and interview techniques. (…) This research is intended to present the identity of the Gypsies, these people who have lived among us for centuries and have become an integral part of contemporary Turkish culture, but about whom no comprehensive scientific study has as yet been conducted because their cultural identity has been largely ignored as a result of the difficulties in identifying and defining them. This study will give an account of their socio-cultural characteristics, beliefs, mythologies, festivals and celebrations in all their aspects. (…) In our opinion these people, who suffer from humiliation and rejection everywhere, could be transformed into citizens who are an asset to our State and our nation once their educational, social, cultural and medical problems are addressed. This simply entails focusing on this issue with patience and determination’ (Aksu v. Turkey, paras 10–13). K. Kov´ acs and G.A. T´ oth International Journal of Educational Research 134 (2025) 102828 4 The complainant, Mr Aksu, who is of Roma origin, initiated a court proceeding because he felt that the book included expressions that offended him in his Roma identity. In particular, he referred to that part of the book which provided information about the Roma people’s alleged involvement in illegal activities as ‘thieves, pickpockets, swindlers, robbers, usurers, beggars, drug dealers, prostitutes and brothel keepers’ (Aksu v. Turkey, para 14). He also contended that the book contained several other generalisations that ‘humiliated and debased Gypsies’, for instance, when it stated that they were ‘polygamist and aggressive’ (Aksu v. Turkey, para 14). The European Court of Human Rights acknowledged that negative stereotyping interferes with the person’s capacity to determine her identity and life course. Hence, it is unacceptable under Article 8 of the Convention, which protects the right to private life. It also acknowledged that certain passages of the book, especially the ones cited by Mr. Aksu, ‘read on their own, appeared to be discriminatory and insulting’ (Aksu v. Turkey, para 57). Nevertheless, the Court held that, as a whole, the book was the outcome of an ‘academic study based on scientific research’ (Aksu v. Turkey, para 69) focused on the history and socio-economic living conditions of the Roma people. The Court stressed the importance of such work and referred to its established case law under which ‘any restrictions on the freedom of academics to carry out research and to publish their findings’ must be subject to careful judicial scrutiny (Aksu v. Turkey, para 71). In the given case, the Court asserted that the author of the study did not make negative remarks about the Roma community in Turkey; on the contrary, the author contended that Roma had been ostracised and targeted by vilifying remarks based mainly on prejudice (Aksu v. Turkey, paras 69–70). Therefore, the Court found no violation of Mr Aksu’s right to private life. The relevant Convention case law does not offer a clear and consistent methodology for determining when the moral distress of individuals from disadvantaged groups can justify imposing limits on free speech. While an ordinary person’s political expression can be restricted to protect members of racial minorities (see, for instance the European cases mentioned in section 3.2 on agitation and incitement to hatred), academic freedom seems to trump these concerns if the work in question is a serious work of scholarship. Under such circumstances, Article 10 protects the dissemination of the research outcome even if the otherwise sincere research findings may be offensive to certain disadvantaged group members. Hence, according to the European Court of Human Rights, the moral distress experienced by an individual from a socially disadvantaged group cannot justify censoring an expression if it can be presented as academic work. 4.1.2. Likewise, US law does not allow censorship of the views professors publicly express, as the Levin case suggests. Tenured professor Michael Levin at CUNY’s City College (City College), who held a doctorate in philosophy, expressed the view that ‘the average black is significantly less intelligent than the average white’ in his writings (Levin v. Harleston, 770 F. Supp. 895, 1991 U.S. Dist., 902). More specifically, the American Philosophical Association Proceedings 62(5) gave survey data on numbers of blacks in philosophy and showed that the proportion of blacks in the discipline was considerable below their proportion of the population. Professor Levin published a letter in the next issue of the journal, in which he argued as follows: ‘The black mean IQ is slightly more than one standard deviation below the white mean. In more familiar terms, that amounts to a difference of more than 15 points of IQ as measured by such standard tests as the Wechsler Adult Intelligent Scale. (…) In a recent survey of the psychometric literature, the National Academy of Science concluded that “in the technically precise meaning of the term, (mental) ability tests have not been proved to be biased against blacks; that is, they predict criterion performance as well for blacks as for whites”. (…) Doctors as engineers are recruited from an IQ range of 114 of above. (…) Only 3 % of the black population (as opposed to 16 % of the white population) has an IQ in this range. (…) Making the most optimistic assumptions, given that blacks constitute 12 % of the population, only 2 % (not 12 %) of the profession will be black. That is close enough to current figures for all philosophers to regard themselves free of any discriminatory guilt.’ (Levin v. Harleston, 902–903). A group of university students felt harmed by these public utterances, and as a reaction to these statements, they conducted loud demonstrations outside of Professor Levin’s classes. During the demonstrations, students carrying banners and shouting denounced the professor as a racist. After several such incidents, the City College publicly condemned the views expressed by Professor Levin in his writings. Furthermore, it set up an ad hoc committee to determine whether the professor’s public statements went beyond the bounds of academic freedom and sent a letter to the students, stating that Professor Levin had ‘expressed controversial views’ and offering them an alternative class should they might want to transfer out of the professor’s required introductory philosophy class (Levin v. Harleston, 907). Professor Levin perceived these steps as encroachments on his academic freedom and turned to the courts (Levin v. Harleston, 907). Professor Levin identified several rights infringements, yet our attention will be directed only towards the claim related to academic freedom. The court of first instance, the United States District Court for the Southern District of New York stressed the importance of the professor’s right to speak externally on issues of public importance. It held that Professor Levin’s public ‘statements about his views on the relationship between race and test scores’ were ‘quintessentially “issues of public importance”’ (Levin v. Harleston, 921). Therefore, these statements were constitutionally protected expressions, and the university did not have the right to hinder Professor Levin’s exercise of his right to speak because it felt that exposure to his ideas might be harmful to certain students. Furthermore, pointing to the chilling effect of the proceeding against the professor, the Court condemned the secret deliberations of the ad hoc committee, which, according to the Court, led to Professor Levin’s being ‘forced to “stay as far away as possible from utterances or acts which might jeopardize his living” and therefore declined at least twenty invitations to speak or to write about his views’ (Levin v. Harleston, 920). The appellate court, the United States Court of Appeals for the Second Circuit agreed to this judgment but specified that the explicit or implicit threat of disciplinary proceedings against the professor for his protected extramural speech was itself a violation of his free speech right (Levin v. Harleston, 966 F.2d 85 1992 U.S. App.). Both the first and second instance courts based their arguments on the First Amendment right to free speech. Why? Because in the US case law, the Supreme Court has recognised academic freedom as ‘a special concern of the First Amendment’, highlighting its significance as a specific area of concern related to this constitutional right (Keyishian v. Board of Regents, 385 U.S. 589 (1967) 590). Academic freedom protects both the professors’ intramural expression, including their university classroom speech, though the scope K. Kov´ acs and G.A. T´ oth International Journal of Educational Research 134 (2025) 102828 5 of this protection is debated by scholars (Tribe et al., 2024; Cole, 2025; Whittington, 2025) and the professors’ extramural expression, that is, their speech outside of class (See the recent cases of Suri v. Trump, 1:25-cv-00,480, (E.D. Va.) Order of 20 March 2025 and American Association of University Professors v. Rubio (2025) No. 25-10685-WGY (D. Mass.)). Extramural speech encompasses a professor’s remarks ‘uttered as a citizen in a non-academic space and for a more general public purpose’ (Butler, 2025). In one case, the US Supreme Court distinguished between statements made by a professor within their area of expertise and outside of it, asserting that the former falls under ‘academic freedom’ while the latter is covered by ‘political expression’, with both being distinctive liberties protected by the US Constitution (Sweezy v. New Hampshire, 354 U.S. 234 (1957) 250). Nevertheless, the Court has not yet differentiated between a distinctive First Amendment theory of academic freedom and the general First Amendment rights of free speech, and it has not yet elaborated a distinct theory of academic freedom (Rabban, 2024). 4.1.3. The European and the US legal frameworks, developed over recent decades, share many similarities regarding the nature and the core meaning of academic freedom. In both jurisdictions, academic freedom protection is based on free speech, yet it is linked to the freedom to pursuit knowledge. The European Court of Human Rights has emphasised the importance of the academics’ freedom to distribute knowledge and truth (Sorguç v. Turkey, Appl no 17089/03, Judgment of 23 June 2009). Furthermore, it has specifically stressed the importance of the search for historical truth by providing wide protection to expert speech that discusses historical movements such as the French Resistance during WWII (Chauvy and Others v. France, Appl no 64915/01, Judgment of 29 June 2004, para 69) or the operation of communist secret service regimes in post-Soviet countries (Ungv´ ary and Irodalom Kft v. Hungary, Appl no 64,520/10, Judgment of 3 December 2013, para 63). Besides, the Court often cites the recommendation of the Parliamentary Assembly of the Council of Europe, which explicitly states that academic freedom should guarantee freedom to distribute knowledge and truth (Parliamentary Assembly of the Council of Europe (2006). The Court hence acknowledges the university’s epistemic mission, the pursuit of knowledge. The same point can be made about protecting academic freedom in the US. Academic freedom is within the ambit of the free speech clause of the US Constitution; the US Supreme Court has recognised academic freedom as a First Amendment right. Although, in the early 1950s, justices only argued strongly for the importance of critical inquiry by professors in dissenting opinions (see, e.g., the arguments made by Justice William O. Douglas and Justice Felix Frankfurter in the case Adler v. Board of Education of City of New York, 342 U.S. 485 (1952) 498), the majority later accepted that professors must remain free to inquire because imposing ‘any strait jacket upon the intellectual leaders’ in universities would ‘imperil the future of the nation’ (Sweezy v. New Hampshire, 354 U.S. 234 (1957) 250). Furthermore, in both jurisdictions, academic freedom extends to the professors’ intramural and extramural speech. So, both in the US and Europe, professors are given robust protection for their speech inside and outside the university walls. This is because they may enrich public discourse with their profound expertise and, most importantly, play an important role in advancing the pursuit of knowledge. Hence, they have to be ‘immune’ from the power of others who wish to use their authority to limit academic freedom without justified reasons (Van Alstyne, 1972). Accordingly, both regulatory frameworks have the potential to robustly protect publications and statements by academics, even if certain people who identify as part of a socially disadvantaged group might feel offended, as seen in the Aksu case and in the case of Professor Levin. As the relevant judgments suggest, this robust protection vanishes if circumstances are present that could justify speech being restricted. One example of this would arise if evidence emerged that a given academic has discriminated against students or others based on their ethnicity or race or that their views have had an effect on their teaching. Moreover, scholars are granted this robust protection as long as their speech complies with established academic standards (Rabban, 2024, 9). Therefore, scholars cannot be punished for the views they express publicly as long as these views meet academic standards. When does an expression meet academic standards? When the courts refer to academic standards, they mean the internal standards and mechanisms of academic self-regulation developed by academics themselves. The most common method known today is peer review. All high-quality publications and research funds employ peer review, that is, the rigorous, usually double-blind evaluation of scientific work by other academics working in the same field. The task of those who engage in peer review is to carefully assess whether the claim or approach genuinely qualifies as academic. The key aspects of such assessments include whether the main point of the research is open to critical debate and capable of being challenged and potentially proven or disproven through objective reasoning. If the claim is based on empirical findings, the assessment includes an evaluation of the methods used and an examination of whether the claims are supported by objective, verifiable evidence. Professional standards explicitly forbid academics from acting contrary to those standards, for instance, by misrepresenting and manipulating facts, selectively using data, or plagiarising (Kov´ acs & Spannagel, 2025). Certainly, just as with other human-managed processes, peer review is not without its imperfections. The process does not always screen out non-scientific content, and its integrity can be compromised by many factors. Despite these challenges, no alternative to academic self-control methods has emerged. And in the event that a case involving plagiarism or any other type of academic misconduct were to reach the court, the court would almost certainly consider expert opinion when making its decision. However, academics do not exclusively write peer-reviewed scholarly articles and book chapters; they also contribute to newspapers and may write letters to the editors, as seen in the case of Professor Levin. More and more frequently, they share their opinions on various social media channels (as it happened in the case of Professor Sadurski, see Morijn, 2020). As Strasbourg judges emphasised, the specific format of a statement does not determine whether it is considered academic and thus eligible for academic freedom protection (see the minority opinion to the Mustafa Erdogan v. Turkey case). Scholars are expected to present their findings publicly, and social media channels can be appropriate for this purpose. A social media post can be a judgment that is grounded in extensive, previously conducted and published research or a summary for the public sphere of a conclusion of ongoing and sustained research. It is up to the court to decide if this is indeed the situation. As seen above, several commonalities exist in how the European and the US jurisdictions protect academic freedom. Yet, there is an K. Kov´ acs and G.A. T´ oth International Journal of Educational Research 134 (2025) 102828 6 important difference between the two jurisdictions when it comes to the scope of protection of extramural speech. In Europe, this protection covers only the cases when the academic speaks about matters of public concern within their expertise. Thus, academic freedom extends only to expert expression of the academic outside of the university walls. The external political speech of academics is not related to the expertise of the academic; it is not protected by academic freedom but by freedom of speech. The scope of protection has historically been broader in the US, where, according to established judicial case law, the academic freedom of the faculty generally extends to all political expression by a professor, even if no relationship exists between the professor’s public utterance and their own expertise. The argument behind this wide coverage was that without this protection, trustees of the universities would use the ‘unprotected speech of professors outside their expertise as a pretext for dismissing them for their academic views’ (Rabban, 2024, 24.). According to these standards, Professor Levin’s statement is protected speech. The usual free speech test applies to political speech by professors, meaning that the statement should survive the clear and present danger test mentioned in section 3.2. Professor Levin’s statement probably offended people, but since nothing in it represented a threat or produced a clear and present danger of an ‘imminent lawless action’, it is protected by the First Amendment. In Europe, Levin’s speech is also protected but the reason is different: academic freedom provides immunity for academics to express themselves in a way that is in accordance with academic standards within their own field. The statement cited above is within the expertise of Professor Levin, given that a philosophy professor has claimed that the lower IQ scores of Blacks explain their low percentage in US philosophy departments. Thus, the speech should not face any restrictions, it is protected by academic freedom. In this section, we have examined the long-established legal standards applicable to the extramural speech of academics in the US and Europe. We considered two examples involving speech potentially deemed racist. We found that in the area of extramural academic speech, the standards differ somewhat. In the US, the ‘clear and present danger’ free speech standard is applied to a professor’s public statement, regardless of whether it falls within or outside their area of expertise. In Europe, however, academic freedom protects professors if their external remarks fall within their area of expertise, whereas free speech protects them when making comments outside their field. In spite of these doctrinal differences, the level of protection afforded to academics’ extramural speech does not vary greatly, because, again, both regulatory regimes employ a balancing act as a standard practice. 4.2. Academic research It is not just the academic speech that may posit the existence of superior and inferior ethnic groups but also the topic and focus of the research itself. Is it acceptable to impose limits on research agendas informed by or directed towards such presuppositions, or do such limitations violate academic freedom? More generally, what happens when perspectives emerge under the guise of scientific inquiry that either explicitly or implicitly reject the equal value of human beings? An example of such an agenda may be research on the relevance of biological differences related to dimensions of race or ethnicity. ‘Genetic determinism’ or ‘genetic essentialism’ claims that human traits, abilities and conditions are determined dominantly by genetic factors rather than environmental and social factors; hence, racial differences are genetic (Kowal & Frederic, 2012). There were and are professors whose research centres on racial essentialism and who claim that there are IQ differences between different ethnic groups, and it relates to genetic differences. Arthur Jensen, a well-known US educational psychologist claimed that IQ was 80 % a product of our genes rather than our environments and that the differences between black and white IQs were largely rooted in genetics (Jensen, 1998). Besides, the book Bell Curve, representing Social Darwinism was published (Herrnstein & Murray, 1994), and many ‘scientific racists’ generally followed it. An example is the UK professor Chris Brand, whose book discussed the correlation between race and intelligence and, in the end, concluded that there is a meaningful correlation between certain markers that are connected to identifying someone as being of a certain race and human intelligence (Brandt, 1996; Billig, 1997). Such projects about race and IQ have a history of persistence within academia and remain present even today (Barrow, 2009; Winston, 2020). So, we ask whether such a research agenda is protected by academic freedom. In the US, the boundaries of this freedom are not clearly established by law. Instead of being rooted in a statutory right (Barendt, 2010, 262), academic freedom has been interpreted by the Supreme Court as originating from the First Amendment’s free speech provision (Rabban, 2024, 52). The First Amendment case law prohibits content-based restrictions, including viewpoint-based restrictions; therefore, even intense opposition to the point of view does not justify a prohibition on researching ‘genetic determinism’. Some scholars have explained this by saying, following Condorcet’s ‘zeal for truth’, that the academic task is to question existing knowledge rather than adhere to already established truths (Fish, 2014). Others have come to the same conclusion, though they have justified it by arguing in a Millian way that ‘unregulated markets are self-correcting’ (Bromwich, 2015). As for Europe, the European Court of Human Rights has not established so far that conducting research on certain areas, such as ‘genetic determinism’ would be in violation of the Convention, or that the Convention would allow for the suppression of scientific views based on their content. The rationale for this might be that the Convention is based on liberal democratic values such as equal human rights and obliges the member states to conform to democracy, with pluralism being a key component. The European Court of Human Rights has stressed the importance of pluralism in education (Kjeldsen, Busk, Madsen and Pedersen v. Denmark, Appl nos 5095/ 71, 5920/72, 5926/72, Judgement of 7 December 1976, para 50) and the plurality of opinions in public debates (Handyside v. the UK, Appl no 5493/72, Judgement of 7 December 1976, para 49). It is this pluralism that is required in scientific research, meaning, among other things, that academics can even conduct research on a research topic that may not align with the values of democracy, such as ‘genetic determinism’. The proviso here is that the research does not rely on a range of flawed methodologies and interpretations of evolutionary theory to justify racist beliefs and ideologies, and its outcome is produced in a manner that corresponds to the professional standards of academic discipline. The established case law of both regulatory regimes seems to subscribe to the central principle of academic freedom; that is, the K. Kov´ acs and G.A. T´ oth International Journal of Educational Research 134 (2025) 102828 7 commitment to not silence academic dissent and allow all forms of serious research to proceed. Academia, including the peer reviewers responsible for writing expert reviews of the research proposal, has the duty to confirm if the research lacks legitimate scientific intent or fails to meet the minimal criteria of intellectual rigor. In such an event, the research agenda is not subject to academic freedom protections. And when a scientific proposal is put forward, it is subject to being challenged because in the academic world, there are no taboos and no assertions immune to questioning. It is the academic community’s responsibility to challenge and rebut a scholarly claim of a research proposal. Hence, academics are both ‘beneficiaries and guardians of the conditions under which they work’ (Hermanowicz, 2021, 17), and the courts have limited room for manoeuvre. Consequently, if there is scientific consensus on a particular scientific question, courts acknowledge this consensus when resolving legal cases. A deviation from this is the recent decision by the US Supreme Court, which decided contrary to the existing scientific consensus articulated by major medical associations as it upheld a Tennessee state law that bans hormone therapy for gender dysphoria (United States v. Skrmetti, 605 U.S. __ (2025). How all these considerations apply to conducting research on ‘genetic determinism’, which perceives individuals as automatons shaped mainly by their genes? Although those who pursue research projects in the field tend to sidestep serious and careful scientific criticism, and they often rely on essentialised racial categories, the claims of ‘genetic determinism’ can be rebutted by the academic community (Winston, 2020). But even if genetic determinist scholars were able to demonstrate a correlation between race and intelligence, it would not logically imply that individuals from specific racial or ethnic backgrounds are morally superior or inferior based on those biological differences. A biological reality does not dictate normative moral conclusions. The 1978 UNESCO declaration condemns any theory that proclaims that there are inherently superior or inferior racial or ethnic groups and denies such theories any scientific foundations (UNESCO, 1978; Lerner, 1981). According to Article 2(1), ‘any theory which involves the claim that racial or ethnic groups are inherently superior or inferior, (…) or which bases value judgments on racial differentiation, has no scientific foundation and is contrary to the moral and ethical principles of humanity.’ What the UNESCO declaration truly prohibits, when understood properly, is not conducting research on ‘genetic determinism’ as such, but rather the questioning of the equal moral value of each person’s human dignity. Whether the research on ‘genetic determinism’ qualifies for public funding is a separate issue, though, that warrants exploration in another scholarly article. Nevertheless, we can note here that on the one hand, public funding must not be allocated arbitrarily (such as its withdrawal as a punitive measure Cole, 2025) and must not be distributed discriminatorily (just as happened in National Institutes of Health, et al. v. American Public Health Association, et al. 606 U.S. __ (2025); Garisto and Kozlov, 2025). On the other hand, receiving public funding is not a fundamental right, either under US or European law, but an option for those who conduct serious research and respect the foundational values of constitutional democracy and whose work is backed by the standards of scientific inquiry. This condition does not mean that those who do not share these values are automatically excluded from public funding. For instance, in principle, an academic with expertise in the field can do research on ‘genetic determinism’ and even defend their position in a scholarly manner while complying with the requirement of equal human dignity. Nonetheless, a research proposal advocating for racial differences in intelligence and morality could justifiably be ruled out from receiving public financing. This is because connecting research funding to the requirement of complying with the foundational values of constitutional democracy does not violate academic freedom (Kov´ acs, 2025). 5. Concluding remarks Hostile and untruthful expressions are notably prevalent across the world. This phenomenon also affects academia. Therefore, the article asked whether academic speech and research are protected even if a professor’s statement, research outcome or research agenda could potentially offend individuals or even cause moral distress to members of socially disadvantaged groups. The focus of the discussion was on two regions that have long been widely regarded as offering robust free speech protections: the United States and Europe. In the former context, the US Constitution guarantees the right to free speech, while in the latter context, the European Convention on Human Rights protects the individual’s freedom of expression. The article revealed that both Europe and the US have long provided free speech protections that cover certain instances of speech that might be regarded as racist. Yet in both regulatory regimes, speech can be limited legally as a result of a balancing act, either because of its potential consequences (e.g., if it poses a clear and present danger of an ‘imminent lawless action’, for instance, inciting hatred) or in order to protect other constitutional values (such as equal human dignity). The article then asked whether a similar balancing exercise can be applied to speech in academia that might be considered racist. In order to tackle this question, the article analysed what is distinctive in academic freedom when compared to free speech and demonstrated that the distinction between speech in general and academic speech in particular lies in the rationale of academic freedom, the pursuit of knowledge, which is meant to enhance our comprehension of the world and human affairs in a way that complies with academic norms and standards. An academic has the right to be wrong and to come to conclusions that may even be regarded as offensive by certain socially disadvantaged groups, but this must be accepted in both regulatory models. Academic freedom permits scholars to draw their own conclusions so long as they act according to professional standards. While it allows for the possibility of error, it does not allow a scholar to act in ways that contradict academic standards, for instance, by falsifying and manipulating facts. Both academic speech and research require a certain structured and reasoned form of engagement; therefore, the members of the academic community are those who are qualified to determine whether research agendas and publications are pursuing scientific aims, whether the academics themselves have credible expertise in the field, and whether they are adhering to academic standards. Courts are constrained in their capacity to assess the scientific quality of an academic work. In the process of balancing, judges can decide whether an academic statement or a published research outcome violates legitimate interests, including others’ fundamental rights, or disturbs public order. K. Kov´ acs and G.A. T´ oth International Journal of Educational Research 134 (2025) 102828 8