Writing about academic freedom in the United States right now is dangerous business. Not for the obvious reasons one may think of — I doubt that anybody with political influence cares about what an obscure sociologist writes in an online journal read by other sociologists. It is dangerous because the pace of academic writing and publishing has little chance to keep up with the accelerated pace of political developments. By the time this article goes to print, it is a safe bet that at least some of the analyses below will strike the reader as quaint in comparison with the horrors that transpired since these lines were being written. Forecasting even a few months ahead is like shooting at moving targets, and as my mentor Ivan Szelenyi used to say, sociologists are poor shots. In the summer of 2024, I had the pleasure of being hosted at Bielefeld. At the farewell dinner, my German hosts cautiously and politely asked me if I was worried about the prospect that Donald Trump would win the elections in November. I do not remember my exact glib words, but they were to the effect that this may well happen, but “the institutions will hold”. A little more than a year has passed, and it has revealed the “institutions” to be a lot shakier than I bargained. So, dear reader, read the following with the requisite measure of skepticism. Do not treat me, the author, as an authority, but as one more player caught in the gears of a much larger process, of which I am only able to see my own little corner.

My starting point, with Michel Foucault (1997) and my colleague Peter Bearman (2025), is that academic freedom is a practice, not a thing. This means that academic freedom cannot be possessed as an entitlement, whatever legal guarantees you think you may have. You can never be confident that you have it as a secure possession. If you don’t practice academic freedom, you will lose it, because it is hedged around by a lot of countervailing interests. So, the practitioners of academic freedom would do well, at all times, to cultivate a mild degree of paranoia (Foucault, 1982). The ways in which freedom can be curbed and neutered are manifold and hidden and often self-imposed. It is not always easy to detect the threats to freedom. One has to remain vigilant. But these are not normal times. You do not need to be mildly paranoid to recognize that academic freedom is under attack in the US at the moment.

The major US regulatory science agencies were taken over, downsized, and neutered in early 2025, as part of a larger campaign to dismantle “the administrative state” and bring it under direct control by the head of the executive branch (the so-called “unified executive theory”).¹ The leadership of the agencies, as well as advisory committees, has been fired and is increasingly being recruited from among science sceptics and critics, whose theories and approaches are becoming official government policy. Basic science, as well, was not exempt from this attack. The term “basic science”, dating from the end of WWII in the US (Bush, 2020), encodes a particular social contract, according to which scientists enjoy relative autonomy to determine how public funds are allocated for scientific research. This social contract is now being renegotiated. The relative autonomy of basic science is being undone by major cuts to the budgets of granting agencies; by specific cancellations or freezing of government funding for disfavored research topics and institutions; and by paralyzing the work of review panels and officers at the granting agencies. A related development is a severing or rearrangement of the transnational ties of scientific exchange and collaboration, which gave American scientists relative autonomy to recruit and train promising foreign students. This system is now being rearranged. Other countries rush in to try to benefit from the US retrenchment or to forge alternative transnational networks. It remains to be seen what happens to the relative autonomy of basic science in the process. Finally, and most relevant to the topic of this special issue, the relative autonomy of American universities and other educational institutions to foster a “culture of critical discourse” (Gouldner, 1979) is also being challenged and circumscribed. Certain phrases, concepts, or theories have become a pretext for conservative politicians and activists to push for state supervision of admissions, curriculum, and speech in universities. The pressure is exerted via subpoenas, legal threats, investigations, but most of all by cutting Federal financial support for academic research.

While the focus in discussions of academic freedom is on the third plank in the attack — the freedom to teach and research certain topics — I have also included the attacks on basic and regulatory science. I think they are all of a piece. I think they also demonstrate that academic freedom is a complex set of practices, not always in concord with one another. We may want to distinguish, for example, between formal academic freedom and substantive academic freedom and then ask how they are related. Formal freedom relies on being able to teach, research, and write without interference, dictate, or fear of being fired. Formal freedom is clearly under attack by Republican state legislatures that control state university systems, as happened recently in Texas (Patel & Goodman, 2025). It is less clear whether and to what extent formal academic freedom will be curbed in private universities or in democratic states. But substantive freedom relies on financial and other support from authorities and clients. They can effectively say, “fine, say whatever you want. We will not limit your freedom of speech. But we will also not support you with public funds”. Put differently, academic freedom is not just a negative “freedom from”, it is also a positive “freedom to”. As we have been made painfully aware, the Federal Government possesses awesome powers to curb the latter through the defunding of research. This means, and we have already seen this happen, that threats to substantive freedom can be mobilized in order to win agreements that curtail formal freedom.

What is most striking to observers and participants alike is indeed how weak and vulnerable American universities — even the richest and most storied institutions — appear to be. Presidents of Ivy League institutions resigned or were fired by their own boards of trustees. The very same universities — and not just state universities under the thumb of Republican state legislators — disciplined students and enacted significant curbs on protests in response to outside pressure (Pickering & Davis, 2025). My own institution, Columbia University, capitulated to the pressure and signed an agreement with the Trump Administration that accepts restrictions on protests, requires disciplining of students, interferes in internal administrative processes, imposes scrutiny of admissions, and creates mechanisms that could interfere also in the academic freedom to determine course content, lecture and research in accordance with disciplinary standards — though what will happen in actuality remains to be seen. Why were universities so weak? If academic freedom is a practice, should we explain it by the failure of nerve of the practitioners?

This would be too simplistic. Practices do get congealed in institutions, and institutions are embedded in networks and fields of relations (Bourdieu, 1984). If academic freedom were only a practice, I think we would have lost it long ago. But relations of force can imperceptibly change, and institutions have a way of getting routinized, diverted, self-referential, stuck (Weber, 1978). They are quickly overrun in a war of maneuver (Gramsci, 1971). They have been overrun. One reason why academic institutions proved so vulnerable is precisely that academic freedom is not one thing. It is composed of multiple freedoms, multiple practices, and institutions. These could sometimes work at cross-purposes, and more importantly, they could be pitted against one another and made to work at cross-purposes. One enactment of academic freedom can be mobilized to undermine another. This is definitely part of the story. When they scrambled to get research funds back, universities were not only after money (Columbia University paid $200 million to get back $400 million), but were seeking to reconstitute the social contract of basic science, i.e., they were trying to restore one of the institutions underlying a particular type of substantive academic freedom — the relative autonomy of scientists to determine what lines of research, however esoteric and seemingly divorced from present societal needs, are to be funded. They sacrificed certain formal academic freedoms — let’s say, the freedom to determine how the topic of the Israeli-Palestinian conflict will be taught — for the sake of this substantive freedom. Predictably, natural scientists and scientists who rely on federal grants often cheered the agreement, not simply because they got the money back, but also because they could fund their students and post-docs, and because the money confirmed their self-concept as autonomous natural scientists free to make decisions about what to research and how. Predictably, more social scientists and humanities scholars, who do not enjoy federal largesse as regularly, whose practice of academic freedom does not depend on this financial infrastructure, cried foul.

The people who say that freedom is a practice mean that if you just assume it to exist because it is codified in certain institutionalized forms, you’ll discover one day that you lost it, and that you no longer know how to practice it. It has been hedged about by institutional roadblocks and cul-de-sacs. You may find that one form of institutionalized academic freedom — basic science — is used as a wedge to curb the practice of another academic freedom. But they also mean something else when they say that academic freedom is a practice: we don’t quite know, from day to day, what it means to be free. Freedom is an experiment; what it means now and here remains to be discovered. Academic freedom is a positive “freedom to” do what? It needs to be discovered (Dewey, 2002).

The attack on basic science and the way it was leveraged to curtail other academic freedoms presents an opportunity for reflection on the social contract underlying it, what it means, and whether we want to maintain it. Basic science, as noted earlier, is a term introduced by Vannevar Bush in 1945 (Bush, 2020). Bush was the Director of the wartime Office of Scientific Research and Development (OSRD) and wanted to make sure that the extraordinary federal support for scientific research would continue after the war’s end. The term “basic science” meant that it was worthwhile for the state and the taxpayer to invest large sums in scientists and to leave the allocation of these funds to the scientists themselves. Bush argued that basic scientific research, which is not geared to respond to immediate needs, is precisely what will ultimately yield the most important innovations. “Basic science” was thus part of a strategy of “serendipity”, justifying the funding of esoteric science.² Not everybody agreed with Bush. Senator Harry Kilgore, an influential figure at the time, held a completely opposite view, namely that the Administration or Congress should begin by defining the most important and urgent social needs, and then allocate funds directly to applied science projects. The setup was thus a classic Weberian clash between substantive and formal rationality. Kilgore blocked Bush’s initiative and ultimately led him to resign in 1948, but over time, there was a compromise, and most of Bush’s original ideas were incorporated into the legislation that created the National Science Foundation (NSF). There was an additional layer to this debate. Part of Bush’s reasoning was that by limiting the state to funding research that is not immediately useful, it would prevent it from competing in the market against private firms that were producing “applied science”.³ Bush was a scientist, but he was also one of the founders and owners of the Raytheon company before World War II, and Chairman of the Board of Merck afterwards. Kilgore, on the other hand, believed that the profits from public funds invested in applied science should go back to the state and taxpayers, and that the state should keep the intellectual property and patent rights. You can see that even though this debate took place in the 1940s, it is still very much with us today, and in a perverted way, the Trump Administration is taking Kilgore’s side.

Indeed, this debate is even built into the very structure of the NIH. The pressure to create new institutes within the NIH comes from Kilgorites (like patient lobbies), while the leadership of the NIH prefers general basic science funding that will then “trickle down” to various disease areas. But the leadership also recognized that it could get more funds by starting institutes. The current structure of the NIH reflects this compromise. Finally, the 1980 Bayeh-Dole Act codified the arrangement in which universities get public money for research, keep the patent rights, and lease them to pharmaceutical companies through exclusive licenses. The tensions in this arrangement were evident during the COVID-19 pandemic. The government was indeed able to organize the pharmaceutical industry to respond to an urgent social need, but only to the extent that their exclusive licenses and profits were protected, providing fodder for critics and anti-vaccine activists now leading the charge against basic and regulatory science. Kilgore’s revenge? Yes, but it also exposes the dangers of the Kilgorite position, dangers noted long ago by Max Weber (1978). Substantive rationality, which would define a hierarchy of social needs and socialize investment decisions accordingly, has an elective affinity with an authoritarian administration that claims to represent “the people” and to know how they would rank social needs. Inevitably, such an administration seeks to curb the freedom of truth-producing practices (scientific and legal demonstrations [Rosental, 2013]).

It seems unlikely that the Trump Administration will remain on Kilgore’s side for very long. They have identified a lever that allows them to split the camp of academic freedom into warring factions, weaken the universities, and extort concessions. The strategy depends on the lever — scientists’ relative autonomy to control research budgets — continuing to function. Putting it with Pierre Bourdieu (1984), we could say that one tendency in scientific fields is for the dominant group of experts to seek to control not only the supply of esoteric knowledge and of the producers of this knowledge (through gatekeeping), but also control and monopolize the creation of effective demand for their products, by, as Bourdieu puts it, producing the taste required to consume scientific products. This can lead to a situation where the only legitimate consumers are the producers themselves, organized in tightly knit “core-sets” (Collins, 1985) of similarly trained and similarly endowed scientists. The practitioners of “basic science” have enjoyed some approximation of this situation for 80 years, and they seem willing to pay a high price to keep it so.

The practitioners of basic science and the defenders of accommodation with the Trump Administration may, of course, argue in response that they are not sacrificing anything. They are staunchly committed to defending freedom of expression, they would say, which is a bedrock condition for academic freedom. It is a necessary, but not sufficient, condition of academic freedom. Without freedom of expression, there is, of course, no academic freedom, but the two are not identical because speech from a position of authority as a professor should entail certain responsibilities that lay public speech does not. One should not seek to speak with authority unless one actually has expertise in the subject matter, and even where one has expertise, it is one’s duty to leave politics outside the lecture hall. They would quote Max Weber (1946):

When speaking in a political meeting about democracy, one does not hide one’s personal standpoint; indeed to come out clearly and take a stand is one’s damned duty. The words one uses in such a meeting are not means of scientific analysis but means of canvassing voters and winning over others. They are not plowshares to loosen the soil of contemplative thought; they are swords against the enemies; such words are weapons. It would be an outrage, however, to use words in such fashion in a lecture or in the lecture-room (p. 145).

All we need to do, they would say, is to eliminate such excesses — which they would add are common not among basic scientists but among their social science and humanities colleagues (and who would deny some such “outrages” were and are indeed being perpetrated?) — while defending freedom of expression to the hilt, and Voila! We will maintain our grasp on academic freedom. Then, under their breath, they will add that the right to free expression, as in protests, for example, can also be limited if it is “directly incompatible with the functioning of the university” (University of Chicago, 2015, p. 2). Hence, certain time, location, and manner of expression stipulations are justified and do not threaten academic freedom.

Such an approach is codified in “The Chicago Principles”, formulated during an earlier moment of tumult on university campuses. In July 2014, the President of the University of Chicago, Robert J. Zimmer, appointed a Committee on Freedom of Expression “in light of recent events nationwide that have tested institutional commitments to free and open discourse” (University of Chicago, 2015, p. 1). The Committee was composed of 7 distinguished Professors from varied disciplines, including Law, Psychology, Medicine, Astrophysics, Economics, and English. They were tasked with drafting a statement “articulating the University’s overarching commitment to free, robust, and uninhibited debate and deliberation among all members of the University’s community” (University of Chicago, 2015, p. 1). As titled, the Chicago Principles were not about academic freedom, but about freedom of expression. But even a superficial reading of the text reveals that the boundary between the two is blurred. The strategy of the text is to anchor itself within “the long-standing and distinctive values of the University of Chicago” (University of Chicago, 2015, p. 1); hence, it proceeds by quoting addresses given by successive Chicago Presidents. The quotes refer to “freedom of expression”, and most often to “freedom to discuss”, but they also speak of “freedom of inquiry” and “free and open inquiry in all matters”, which can only take place in “an environment of the greatest freedom” (University of Chicago, 2015, pp. 1, 2). The slippage from “speech” to “inquiry” may simply underline the point that freedom of expression is a necessary condition for academic freedom, and that “inquiry” is an old-fashioned way of referring to Socratic dialogue. But I think the slippage also indicates that the opposition set up by Weber, between the lecture hall and the public square (or the quad), does not exhaust all the contexts within which academic discourse takes place, hence the need for another word like “inquiry”.What is the status of an exhibit on campus that is meant to work its effect precisely by “provoking”? Or an op-ed by a professor in the campus newspaper that seeks to stimulate discussion by “bending the stick in the other way”? Or a professor interviewed on a particular topic — should they be free to say what they wish, use their words as “weapons”, because this is public speech? Or should they avoid “using words in such fashion” because the premise of the interview is their greater authority as experts? For that matter, what about survey questions that could offend the sensibilities of a good part of the public? Freedom of expression is no doubt a condition for academic freedom, but I think the Venn diagram between them has a significant shaded area of overlap, and insisting on a strong boundary between the two ultimately defines the scope of academic freedom far too narrowly. I think it is more accurate to say that the authors of the Chicago Principles (University of Chicago, 2015) perceived constraints on freedom of expression to be the most immediate threat to academic freedom in 2014. They were practicing academic freedom by asserting certain principles that should govern public speech in the university. This is the direction to which their own mild form of paranoia led them.

But the principles they formulated were already outmaneuvered in 2014. There is an uncanny similarity between the situation to which the authors of the Chicago Principles were responding and the present moment. What is common to both moments is a strategy that demands to censor speech by describing it as a form of harassment that creates a hostile environment for a certain protected category of people. Many have already commented on the similarity, noting the irony that the lefty “woke” strategy is now being repurposed by the right, sometimes by the very same people who decried it previously. There is indeed a certain grotesque combination right now of immense prickliness to what the other side can say, coupled with leveraging free speech almost as a cudgel. Now, let us return to Weber (1946) one last time:

The primary task of a useful teacher is to teach his [sic] students to recognize inconvenient facts — I mean facts that are inconvenient for their party opinions. And for every party opinion there are facts that are extremely inconvenient, for my own opinion no less than for others (p. 147).

I strongly believe that Weber is right in characterizing this stance as the duty of the professor and potentially the “moral achievement” of teaching. I try to practice it myself. But it is not hard to see how it will run headlong into the strategy that brands inconvenient facts as a form of harassment. After all, it is precisely the inconvenience, the discomfort in being exposed to arguments that threaten one’s identity, that has been identified as an illegitimate form of “discrimination on the basis of national identity”. There are ways of preempting this clash, of framing the discussion in a way that demonstrates that the teacher is willing to take the body blow first, or that everybody equally undertakes to be exposed to facts inconvenient to their own party’s view. We know how to do it. We, academics, have done this for many years in lectures and seminars. But in a situation where every student is equipped with a smartphone and with the ability to upload videos immediately to social media — another way in which Weber’s opposition of the lecture hall to the public square is, if not obsolete, far less convincing — as happened recently in Texas, leading to the dismissal of the professor (Patel & Goodman, 2025), it is extremely likely that we will not have the chance to practice this art, which takes time and requires inhabiting provocative positions one after the other in quick succession.

It goes without saying that basic scientists are unlikely to find themselves in this situation and are not well-equipped to appreciate it. Astrophysicists who expose their students to inconvenient facts or demonstrate how a good scientist should be willing to look squarely at facts that are inconvenient for their theory are unlikely to experience pushback from their students. A video of a professor instructing a student about the hard facts that a “flat earther” needs to acknowledge may make for viral social media videos but is unlikely to get anybody fired. Natural scientists are relatively secure in a decades-old monopoly on shaping the taste needed to consume their products — their concepts, theories, demonstrations, and assertions of facts, however inconvenient. Social scientists and the humanities, the policy sciences, we are not in this position. Our students and our audience have decided opinions about our matters of expertise, and we do not possess such a monopoly on producing the taste necessary to consume our intellectual products. In between these two poles is medicine. Often funded as a basic science, it can never enjoy a full monopoly over the production of demand since the consumers’ very own bodies are the object of medical expertise, and especially in this age of patient advocacy and lay expertise, they contest medical definitions of normality and illness (Epstein, 2023; Akrich & Rabeharisoa, 2023).

There is a more general point contained in the contestations around medical expertise. When academic freedom is distinguished from freedom of speech, the argument is often made that academic freedom entails the responsibility not only to confront inconvenient facts, but perhaps first of all to confront the inconvenient fact of the limits of one’s expertise. Academic freedom is not the same as freedom of speech. It is not the freedom to say whatever you want, but the freedom to research, write, speak, and teach about what you can claim expertise about — a claim vetted and confirmed by peers, members of the discipline. Speaking as an authority differs from speaking freely as a citizen and therefore requires self-censorship — do not seek to speak authoritatively, especially not in the classroom, about matters that lie beyond your expertise. If we could only hold voluble and self-important academics to this standard, the reasoning goes, it would be very easy to tell what constitutes real attacks on academic freedom and what doesn’t. Alas, things are not so simple (Collins & Evans, 2007; Eyal, 2019). Who is an expert is not always obvious. It may be obvious in astrophysics, and few people would dream about teaching an astrophysics course without belonging to the “core-set” of astrophysicists. But things become murkier quickly as you move away from basic science. Even with something seemingly “hard” as the life sciences, it is not so clear. Who had expertise about COVID? If a physicist spoke about COVID policy choices, would they be out of their jurisdiction? Not so in Israel, where they claimed this jurisdiction in the name of knowing how to model uncertainty and took over the taskforce advising the Prime-Minister (Harel, 2020). And what about engineers? Before the pandemic, we did not know that aerosol engineers had a crucial piece of the puzzle. We probably did not even know that aerosol engineering existed. The point is that when you leave basic science and enter the realm of regulatory science, you encounter “wicked problems” about which nobody is quite an expert (Eyal, 2022). Our experts were like the proverbial seven blind men feeling up an elephant, each knowing only one — potentially misleading — piece of the puzzle. And then there is, of course, the question of the clients. This version of academic freedom seems to assume an ideal situation in which only practitioners can judge the relevance and excellence of other practitioners, but this is increasingly not the case in medicine with the rise of lay experts, and it was always a conundrum in psychology and psychiatry. So yes, we do not want a climate scientist to teach a course about the Israeli-Palestinian conflict and alienate one or the other group of students by trying to pass off political views as facts. “To ‘let the facts speak for themselves’”, as Weber (1946) said, “is the most unfair way of putting over a political position to the student” (p. 146). But what if this climate scientist is teaching a course about the interrelations between climate change and societal conflicts over resources? And they spend several weeks on the struggle over water resources in the West Bank? I think all who are interested in the well-being of academic freedom would say that they should be allowed to do so, and that analysis of Israeli policies that have significantly curbed the water resources available to Palestinians (B’tselem, 2023) does not mean that they are straying from their expertise, nor does it constitute creating a hostile environment in the classroom. But is there a clear, bright line that we can articulate that would tell us when straying somewhat beyond one’s ken does or does not constitute the exercise of academic freedom? After all, “stay in your lane” is precisely what authoritarian governments and risk-averse university administrations would say. Should we leave it, therefore, to the judgment of their peers? But who are the relevant peers? Deciding who to consult in the matter — which is interdisciplinary, emergent, a wicked problem — often prejudges the answer (Jasanoff, 1997; Eyal, 2019). And who should we consult about whom to consult? A university committee?

Come to think of it — don’t we already have such a committee? Isn’t this what tenure decisions are all about? Tenure is a mechanism for transferring risk from an individual to the organization.⁴ The organization socializes the risk and allows individual professors freedom to pursue risky lines of research and teaching, where errors and false starts are likely, and where straying from one’s ken is also likely because one is at the cutting edge of research. Tenure is an organizational practice of academic freedom, and by the same token, it grants the individual freedom to make mistakes, which is, as we know, how science advances. At Columbia, tenure cases go through multiple layers of review, first by the candidate’s department, then by soliciting letters from disciplinary leaders, then through two additional reviews by members of other disciplines and academic leaders. The practice thus includes safeguards against narrow disciplinary groupthink and, by the same token, allows tenured faculty members the freedom to determine by themselves the scope of their research and teaching. But universities have been steadily chipping away at this organizational practice of academic freedom because they increasingly hire adjuncts, thus not just saving money but also no longer being committed to this socialization of risk. Universities have probably made themselves weaker and an easier target as a result. As they ceased to exercise the muscle, the organizational practice that secures academic freedom, it atrophied.

To end on a personal note: in the midst of writing this piece, I was asked by the Provost of my university to serve on a Committee on Academic Freedom advisory to the Provost. Some quipped that this is like passengers on the Titanic forming a “flotation committee”. The implication is that the need is obvious, but that a committee is an inappropriate response where urgent action is necessary. I agree. Academic freedom is a practice, and no committee will save us if members of the university, as well as the organization itself, do not practice academic freedom and do not resist attempts to curb it. The irony is not lost on me that I was asked to join a committee just as I concluded that, first, a university committee is not the proper mechanism to deal with this issue, and second, to the extent that it is, we already have such a committee, but its impact has been diluted. At the same time, once such a committee is constituted, it becomes one of the arenas where the struggle over academic freedom will take place. University committees typically work out some compromise, and where the compromise lands in between the opposing sides depends on their composition and on persuading the wavering middle. Writing this piece allowed me to try out the arguments, hone them, and sharpen them for this purpose. Full disclosure: I joined the committee.