What, precisely, are we defending when we invoke “academic freedom”? A cousin of free speech but not its twin, it promises to shield scholars from unwarranted interference. But should its protections be limited to work within one’s area of expertise, or should it extend more broadly? And is it an individual right belonging to scholars, or does it also encompass the autonomy of universities and departments? 

Efforts to answer these questions have often produced proposals that either are too limited or too broadly construed, in both cases obscuring the essential purpose of academic freedom, which is to create the best conditions for the pursuit of knowledge.

The history of academic freedom has long been marked by these tensions. Its foundations date back to medieval European universities, which carved out enclaves of intellectual autonomy. In 19th-century Germany, the Humboldtian model gave academic freedom clearer form, enshrining the twin principles of Lehrfreiheit (freedom to teach) and Lernfreiheit (freedom to learn) under the rubric of Akademische Freiheit.

In the United States, the forced resignation in 1900 of Stanford economist Edward Ross helped bring the issue to the fore, prompting the creation of the American Association of University Professors (AAUP) and its 1915 Declaration of Principles, a landmark document that set out three pillars of academic freedom: freedom to pursue inquiry and research, freedom to teach, and freedom to speak publicly as a citizen. This document has been updated several times—most notably in AAUP’s 1940 Statement of Principles on Academic Freedom and Tenure, which is still widely cited.

The U.S. Supreme Court began giving academic freedom constitutional shape in the 1950s, recognizing in Sweezy v. New Hampshire (1957) that it is a distinctive First Amendment right. In Keyishian v. Board of Regents (1967), the court described academic freedom as “a special concern of the First Amendment.” Like the AAUP Declaration, these opinions framed academic freedom as belonging to the individual professor. Justice Felix Frankfurter’s concurrence in Sweezy, however, attempted to broaden the concept, quoting a South African statement identifying four freedoms of a university: “to determine for itself on academic grounds who may teach, what may be taught, how it should be taught, and who may be admitted to study.” Beginning in the late 1970s, the court drew upon this to extend academic freedom rights to universities as institutions.

Recently, however, these protections have at least indirectly been called into question by the high court. For example, the constitutional right to freedom of speech for U.S. public employees has been constrained by an employer’s interest in an efficient, disruption-free workplace, with the court’s decision in Connick v. Myers (1983) limiting protection to speech on matters of public concern. This was further whittled down in Garcetti v. Ceballos (2006) to exclude speech made pursuant to an employee’s official duties—though the court left open whether “speech related to scholarship or teaching” might warrant a more protective standard, an exception that some circuits have since recognized.

Outside the constitutional context, definitions of academic freedom differ on the balance between individual and institutional rights. The University of Chicago’s 1972 Shils Report defines it as “the freedom of the individual to investigate, publish, and teach in accordance with his intellectual convictions.” The 1997 UNESCO Recommendation on the issue adopts a wider frame, encompassing both individual freedoms for faculty and institutional autonomy.

Beyond the individual-collective tension, a second fault line runs between professionally grounded speech and broader expression. Should the special prerogatives granted to academics—privileges that go beyond the general free-speech rights granted to most non-academics in their workplaces—be limited to their credentialed areas of academic expertise, or should they apply more broadly? European Court of Human Rights case law conditions enhanced protection on whether the expression flows from professional expertise and meets scholarly standards. However, Britain’s Higher Education (Freedom of Speech) Act, by contrast, does not confine free speech protection to one’s area of expertise. Such a caveat was included in an earlier version of the bill, but was explicitly removed by Parliament during its passage in 2023.

A third fault line concerns the responsibilities that, according to many statements of academic freedom, accompany the rights. The 1940 AAUP Statement counsels “appropriate restraint” when academics speak or write as private citizens. And UNESCO’s 1997 Recommendation includes an obligation of “due respect for evidence, impartial reasoning and honesty in reporting.” Are these responsibilities mandatory or merely hortatory? Is academic freedom forfeited if they are not met?

Keeping the Funnel Open

To make sense of these fault lines, let’s revisit an earlier question: What is academic freedom for? The strongest justification, I believe, is epistemic: Academic freedom is essential to the university’s mission of “truth-seeking and the advancement and dissemination of human knowledge.” Jonathan Rauch introduced the concept of an epistemic funnel to explain the collective process of knowledge creation. At its big end, the funnel must admit a broad range of ideas and propositions. Only those that withstand scrutiny by the “reality-based community”—a social network governed by key rules, norms, and institutions—emerge from the smaller opening at the other end.

Should the special prerogatives granted to academics—privileges that go beyond the general free-speech rights granted to most nonacademics in their workplaces—be limited to their credentialed areas of academic expertise, or should they apply more broadly?

The funnel’s ends operate by antithetical principles. The big end must be as open as possible: Excluding ideas because they offend preconceptions violates the fallibilist rule, which acknowledges the provisional nature of knowledge. Yet once ideas enter the funnel, they are subject to the empirical rule, which rejects personal authority and demands evidence-based validation of propositions. The institutional filters of “organized skepticism”—critical exchange, peer review, rule-based social checking—winnow relentlessly, aiming to ensure that what emerges is guided by merit.

Collapsing the funnel’s zones into a single rubric invites conceptual drift and dilutes protection. If academic freedom exists to create the conditions for truth-seeking, its task is precise: Protect the big end of the funnel. Historically, academic freedom developed to shield universities from church and state; scholars’ intellectual freedom traveled alongside institutional autonomy. But in liberal democracies today, many attempts to censor originate within academia. Autonomy was once a bulwark against infringements of academic freedom; today, it too often enables them.

Former University of Virginia law professor Frederick Schauer argued that the “ability to explore unpopular ideas and challenge the unchallengeable” that we associate with the university's mission is best served by protecting the academic institution—not individual scholars—from political or bureaucratic interference. That may shield broadly mainstream academic views from external pressure, but it does little for dissenters who challenge internal orthodoxy—the very voices most in need of protection.

The better approach, then, is to understand academic freedom as an individual right, securing scholars’ ability to contribute to the funnel against interference from inside or outside the academy. Institutional autonomy is best seen not as a limb of academic freedom but as a separate instrumental good. Autonomous universities are generally better positioned to resist external political pressures on scholarship. Yet when universities come to prize consumer-oriented competition for students, risk-management, and ideological conformity over truth-seeking, autonomy can corrode the epistemic funnel.

How Far Is Far Enough? 

But if academic freedom is only an individual right, how far should it reach? If the right’s primary rationale is epistemic, why protect speech that does not draw on expert knowledge? After all, academics seem no likelier than anyone else to be reliable outside their fields. Equally tempting is to deny protection to work that fails academic standards, on grounds that it does not aid knowledge production.

Yet who decides what counts as “within expertise”? Alfred Wegener’s theory of continental drift was ignored for decades because he was not a geologist. Indeed, it is often cross-disciplinary contributions that prove most consequential (and perhaps also most prone to suppression). The work of the sociologist Michael Biggs on puberty blockers—initially dismissed by the medical establishment—was later cited in the influential Cass Review, demonstrating how scholars coming “from outside” can challenge a discipline’s orthodoxy.

And how do we distinguish “academic standards” from dominant paradigms that need shaking up? Collective judgments of standards shift over time. Economists David Card and Alan Krueger’s contrarian 1995 work on the minimum wage received an excoriating reception from their fellow economists, who likened it to “repealing the law of gravity.” Twenty-six years later, Card won a Nobel Prize for the very same work. Aristotelian professors accused Galileo of trespassing on disciplinary boundaries, while expert theologians declared his heliocentric view “foolish and absurd in philosophy.” We all know how that turned out. Unfortunately, radical challenges to orthodoxy on highly sensitive topics often draw similar treatment from mainstream experts today.

To keep the funnel’s intake genuinely open, academic freedom of investigation and dissemination should not be limited by perceived expertise or made contingent on satisfying current scholarly standards—save for egregious breaches such as falsification, fabrication, or gross plagiarism. The responsibilities commonly linked with academic freedom must be understood as hortatory, not mandatory. This will protect much work that won’t survive the narrowing; it is precisely the safeguard today’s Galileos need. As Biggs remarks:

“Academic freedom comes at a price, of course. It enables scholars to pursue research which might seem frivolous. It protects weird cranks as well as truth seekers. Only in retrospect, however, can they be distinguished.”

Academic freedom of investigation and dissemination should not be limited by perceived expertise or made contingent on satisfying current scholarly standards.

A common claim is that academic freedom excludes pseudoscience or denial of settled knowledge. Yet apart from the fact that “settled” facts are sometimes overturned, ideological pressure and self-censorship can make contested issues look settled when they are anything but. An important recent study surveyed 470 leading psychology professors on 10 “taboo conclusions” about gender, race, and group differences; contrary to the prevailing narrative, it found no scientific consensus against those claims. There was, however, a strong correlation between holding a taboo view and self-censorship. This can lead to a “spiral of silence”: Those who perceive their position to be unsupported will fall silent, reinforcing the illusion of consensus for the opposing view.

University of Texas law professor David Rabban argues that academic freedom protections should not extend to speech outside of the professional realm. But if academic freedom is not confined to supposed expertise, there is no principled boundary excluding such extramural speech. Leaving such speech unprotected, moreover, gives universities a readily exploitable pretext to punish scholars for controversial academic work. The production of knowledge relies on expertise; the rules that best protect that production do not. AAUP co-founder Arthur Lovejoy understood this: “to protect the investigator within his special province, you must protect him outside of it also.” 

At the same time, academic freedom does not entitle a scholar to a speaking invitation, a journal acceptance, or grant funding; these are matters of merit to be judged by the reality-based community. Accordingly, academic freedom does not shield professors from the indirect employment consequences of adverse merit assessment—such as missing out on prizes or promotion—so long as those consequences are not a pretext for viewpoint discrimination.

Finding the Right Balance

We can now see the defect in many frameworks of academic freedom: They are at once over-inclusive and under-inclusive. They overreach when they fold collective rights, institutional autonomy, or peer review into academic freedom. They underreach when they confine protection to speech within expertise, focus on meeting professional standards, and exclude extramural speech.

Properly understood, academic freedom is a broad individual right vested in scholars. It grants them the prerogative to set their research agendas and to express their opinions on any topic. A university may regulate the time, place, and manner of that expression where necessary to protect its essential functions, but it must do so in a viewpoint-neutral way.

Academic freedom also applies to teaching, with some limits. Universities may require coverage of the specified syllabus, disciplinary competence, and basic classroom civility. Beyond that, academics should have wide latitude over teaching methods and course content, and should not be pressured to endorse or reject particular value judgments.

Academic freedom is a necessary (though not the only) condition for the creation of knowledge—humanity's greatest achievement. The touchstone for the academy is whether it keeps the funnel's entrance open while letting merit—not ideology—determine what endures.

Acknowledgements: I am grateful to Alan Sokal for helpful comments on a draft of this article.