Academic freedom attracts little interest in ordinary times: academics tend to take it for granted and go about their routine business. But the rise of attacks on academic freedom in recent years, months, and weeks has changed this state of affairs. While sociologists know that quantitative indicators should not be fetishised, it is noteworthy that the Academic Freedom Index (AFI) lists only eight countries with levels of academic freedom higher in 2024 than in 2014 and identifies a staggering 34 cases of decline (Kinzelbach et al., 2025). The fact that this decline takes sometimes spectacular forms even in countries implicitly assumed to be immune to such dynamics has also acted as a wake-up call. The number of academic efforts dedicated to the subject has risen accordingly, with the topic also getting a fair amount of media attention. Drawing from historical models such as the New School University in Exile, set up in 1933 to host German exile scholars, various national and international programs and structures have also been set up to provide help to academics threatened in different ways. The international network Scholars at Risk (SAR, founded in 1999) is an early example. More recent initiatives include the Philipp Schwartz initiative of the Humboldt Foundation (launched in 2015 to allow scholars threatened in their country to continue their work in exile), the German Academy in exile (set up in Dortmund in 2017), the French PAUSE programme (an emergency scheme for hosting scientists in exile also launched in 2017), New School’s New University in Exile Consortium (established in 2018), the AFI (a worldwide assessment of the de facto levels of academic freedom around the globe), the Clemens Heller Institute (an organisation created in 2024 with the aim of offering temporary refuge to academics threatened in their own country). Efforts are also currently made to strengthen the protection of academic freedom within the European Union.

While instruments and organisations like SAR and AFI have contributed to clarifying what we mean by “academic freedom” and empirically documenting its highs and lows, the fact remains that the notion has a flexible meaning and is more often the subject of opinion pieces and political discourses than theoretical reflections and empirical objectifications. There are good reasons for this. One is linked to a sense of urgency and responsibility. Faced with the dire situations of colleagues losing their jobs, their freedom of movement, their funding, or even their lives, the obvious priority is to protest and take countermeasures, not write carefully crafted research. Another reason is that writing about academic freedom in 2025 puts authors in the uneasy position of reflecting on their own vulnerability. When writing about academic freedom, scientists do so as intellectuals, but also as players in a game they are currently losing, albeit to different extents depending on their position. Any output can then be seen as navigating between intellectual effort, political intervention, position-taking in the scientific sphere, and introspection.

Like any other, the notion of academic freedom is a social construct, and this construct is currently subjected to multiple attempts to de- or re-construct it. This symposium cannot claim to be immune to this context. The authors of the papers assembled here have their own circumstances, and those circumstances shape their views on the subject. The objective we have collectively given ourselves is to still try and document academic freedom at the theoretical and empirical level, sometimes by using our own experiences as material, and sometimes by decentring and taking other contexts as objects for study.

Thus Thibaud Boncourt, Gil Eyal, and Eric Fassin all highlight through theoretical and reflective observations that academic freedom is a contested notion, which can be as useful as it is detrimental to academics. Boncourt’s diagnosis (2025) is that the term “academic freedom” (especially with its connotations of “freedom of speech”) has become too slippery to be fully useful as a concept and as a political tool for defending academics and academic activities. He argues, instead, for the concept of academic autonomy both as an analytic tool and as a banner for mobilising efforts. Eyal (2025) relies on personal experience to argue that academics are ambiguous in the ways they practice academic freedom, as one enactment of academic freedom may well be sacrificed to save another. Fassin (2025) digs into the history of the notion of academic freedom and reminds the reader that, contrary to what is sometimes argued, it is not synonymous with freedom of speech and that, therefore, one cannot invoke academic freedom to justify stating nonsense. This is precisely why, in his eyes, academic freedom has come under attack from political forces that use nonsense as a rhetorical tool.

An example of such a controversial use of the notion of academic freedom is provided by Pierre-Nicolas Baudot’s empirical study (2025) of its mobilisation by a group of French scholars and essayists to combat critical approaches that they denounced as “Islamo-leftist” ideology, following and relying on attacks stemming from the political field. Although the “Islamo-leftism” parlance is a French peculiarity originating in French politics, attacks on critical thought, especially gender studies and race and ethnicity cultural studies, are a common denominator to illiberal tendencies. While conservatives denounce these critical approaches as ideology, their prohibition in illiberal regimes is regarded by many scholars as a restriction of academic freedom.

Gisèle Sapiro’s paper (2025) analyses these illiberal tendencies and the various ways in which they threaten what is commonly understood as academic freedom, that is, the freedom of research, teaching, and expression (Beaud, 2010). However, she argues — after returning to the foundations of academic freedom, which she regards as the principle underpinning the autonomy of the academic field — that, in this matter, the apparently clear differences between authoritarian and illiberal regimes, on the one hand, and liberal democracies, on the other, are becoming increasingly blurred for both exogenous and endogenous reasons. The brutal restriction of academic freedom in the US by the Trump administration, described in Eyal’s paper (2025), is an example of the fragility of the academic field’s autonomy even in liberal democracies.

Elif Can and Sümbül Kaya’s paper (2025) provides a very concrete case study of how the illiberal restrictions on academic freedom operate. They focus on a Turkish case, which is paradigmatic of the increasing control exerted by the political field on the academic field. They show how the restrictions of academic freedom since 2016 have operated at both the institutional and individual level, reshaping the academic field in Turkey. Turkish academia is now constrained by strict institutional control, the exclusion of scholars, and the banning of certain topics. In turn, scientific activity expands outside formal academic institutions, into the “academies of solidarity” that have recently emerged.

The complex relationship between institutional autonomy, professional autonomy, scientific autonomy, and academic freedom is one of the questions addressed in different ways — conceptually, theoretically, empirically, practically — by the authors in this symposium. Some of them connect it to Bourdieu’s field theory and others to the sociology of professions, autonomy having different — albeit not incompatible — meanings within these two sociological perspectives (Sapiro, 2019). Field theory concentrates on autonomy vis-a-vis ideological and economic demands, and sees it as guaranteed by self-regulation through peer assessments and specific authorities (Bourdieu, 1988). The sociology of professions focuses on control over access to a group, socialisation through training, professional ethics, and internal discipline, as a right and privilege granted by the state (Abbott, 1988; Freidson, 2001). A conceptual clarification is thus required, and the different angles, whether theoretical or empirical, but also the practical questions and difficulties encountered by organisations defending academic freedom when dealing with concrete cases, are productive to enrich the discussion and help de-construct/re-construct these concepts. We can illustrate these lines of tension with a few examples.

Institutional autonomy means that political authorities delegate to university administrators and/or research institutions the autonomy to decide their budget, recruit their staff and students, and develop their own curricula and research programs, within certain frameworks and rules that can be more or less constraining. Comparing the forms and degrees of institutional autonomy in different countries and within a given country at different times (as Can & Kaya [2025] do in the Turkish case) would help better understand the modalities of control and intrusion of the government (nominations, rather than elections, of rectors or university presidents, for instance). China, for example, is an extreme case of institutional heteronomy, as university presidents are under the control of a representative of the Communist Party (Veg, 2024).

But the notion of autonomy can also be used by governments to impose budgetary reforms. In France, the 2013 Law on Higher Education and Research understood “autonomy” in budgetary and financial terms, with the State disengaging from the payment of university staff salaries and pensions. Moreover, the law substantially changed the governance model of universities, notably by altering the composition of administrative boards. On the one hand, the number of board members was reduced; on the other, more external personalities from the economic world were required, while the presence of representatives from scientific associations and trade unions was no longer mandatory. Thus, as often occurs in neoliberal reforms, the word “autonomy” was used as a kind of antiphrasis, to in fact limit the institutional autonomy (which also depends upon funding) and self-governance of universities and submit them to more external economic constraints. Some papers in this symposium also show that neoliberalism and authoritarian trends are increasingly intertwined.

While institutional autonomy is generally and rightly considered as a condition for the academic freedom of each and all members of the academic community (or individual academic freedom), it has been observed that, depending on their area of specialisation, some scholars can enjoy a certain degree of collective professional autonomy (meaning relying on their expertise and skills to perform certain tasks as recruitment, evaluation, management of research and teaching) and individual academic freedom, even under high institutional control (Veg, 2024). Conversely, as the attacks on “Islamo-leftism” in France illustrate (see Baudot [2025], in this issue), researchers working on topics such as race and gender studies may be threatened even in liberal democracies granting institutional autonomy. This means that limitations on academic freedom affect scholars in an uneven manner. However, the implementation of threats requires a certain level of institutional control (as happened in Florida, where gender studies and Critical Race Theory were banned from college majors and minors programmes in 2023; see Sapiro [2025], in this issue). Inequalities between scholars regarding individual academic freedom also depend on their respective positions: junior researchers and those belonging to minorities are more vulnerable. The recent censorship reforms, which ban more than 350 words in US federal funding schemes, especially those of the National Science Foundation, entailed the cancellation of many research projects and the related salaries of postdoctoral and junior researchers, while most tenured professors are not personally affected at the time of writing. In countries where they are civil servants, professors also benefit from functional protection by their institution in case of attacks, and this can be extended to all personnel hired by universities, but usually not to non-funded affiliated researchers or students. Attacks on academic freedom raise more broadly the question of who is included in the academic community protected by academic freedom. While some consider — explicitly or implicitly — that this freedom applies only to university professors and equivalent staff (Beaud, 2010; Fernandes, 2023), the German tradition of “Lernfreiheit”, underscored by Eric Fassin (2025) in his contribution to this symposium, should include the students. Referring to article 5.3 of Germany’s Basic Law, Christian Calliess (2025) argues that academic freedom protects even more staff, albeit with limitations:

Academic freedom applies to anyone who is independently engaged in scientific work or who intends to pursue such work. It is not just professors who are protected, but also advanced lecturers, postdoctoral researchers, and students who are conducting research for bachelor’s or master’s theses. Protected places and spaces are not only public universities but also non-university research institutions, as well as private universities.

Attacks against academic freedom have also shown that the three dimensions commonly encompassed under the notion of academic freedom — freedom of research, of teaching, and freedom of expression — are not systematically linked in practice. Under institutional control, a professor may self-censor in class (fearing the risk of denunciation) but feel freer in their research when it does not address sensitive topics. By the same token, the freedom to disseminate research results can be limited by ethical considerations as well as by ideological control, but it will also depend on the topics and on the venues: academic circles, public lectures, academic journals, or the media, inside or outside of a country.

This leads to the final, but significant, line of tension explored in several papers: the complex relationship between the notions of academic freedom and freedom of expression, and their respective interpretations. While freedom of expression concerns the communication of opinions and value judgements, academic freedom involves the right to disseminate scientific knowledge. However, as Fassin (2025) contends, scientific knowledge cannot be regarded as neutral and reduced to mere facts. Thus, there is a zone of overlap between the two notions. The freedom of expression of academics may be limited not only by legal restrictions enforced by the state — such as laws on hate speech or protection of privacy in liberal democracies, or prohibitions relating to criticism of the government, religion or certain topics in authoritarian and illiberal regimes — but also by constraints imposed by academic institutions themselves, particularly in light of ethical standards (see Boncourt’s contribution [2025] to this symposium).

Regarding the first type of limitation, discussing the Chinese case, Sebastian Veg (2024) argues that academic freedom is interpreted there by the government in a narrow manner, restricted to academia and academic work within academia. However, even this so-called “freedom” is constrained by banned topics and strong pressure to produce state and partisan propaganda, such as research focused on “Xi Jinping Thought”. By contrast, in liberal democracies, academic freedom is typically more broadly defined and closely related to freedom of expression, to which it is connected without being part of it (while freedom of expression is part of academic freedom). Nonetheless, the boundaries between the two have recently been debated, particularly in the context of pro-Palestinian student mobilisations in the past two years. As a consequence, in the United States, even before the second Trump administration, public universities had already imposed restrictions on professors’ freedom of expression. In addition, freedom of expression for professors can be limited in countries where they are civil servants — as in France, where they benefit from a special exemption from the professional discretion (“obligation de réserve”) required of other civil servants. In France, the National Centre for Scientific Research, whose researchers are also civil servants, undertook an internal consultation on the conditions for researchers to engage in the public space. Its ethics committee considered that there was no incompatibility, in principle, between the norms of scientific research and the public engagement of researchers, so long as “scientific integrity” and “scientific rigour” were respected, and that a balance between freedom and responsibility was observed (COMETS, 2023), and a guidebook was established after a large consultation of all CNRS researchers.

While academic freedom is either approached, in international law and frameworks, at a theoretical and definitional level, or, by activist organisations, as a practice articulated to different national legal frameworks, this symposium is an invitation to initiate a transnational and comparative reflection based on both theoretical issues and empirical research and including experience and practical questions.¹