The controversy surrounding Nathan Cofnas is usually presented as a simple story: a Cambridge academic expressed racist ideas and universities responded accordingly. But beneath that description lies a more interesting question about academia itself: what happens when an academic challenges one of the assumptions underlying contemporary thinking about race, inequality and DEI?
Is the proper response to demonstrate that he is wrong, or to declare that the argument itself falls outside the boundaries of acceptable academic discussion? Cofnas is an unusual test case precisely because his views are genuinely controversial and challenge the DEI framework consensus.
This article is not an attempt to defend hereditarianism or to argue that its critics are wrong. Its focus is narrower: the limits of academic freedom. Should universities protect the right of academics to advance controversial, unpopular or even deeply offensive ideas, provided they remain within the law and the standards of scholarly debate? Or should some viewpoints be treated as so objectionable that holding or expressing them can justify professional sanctions?
The Cofnas affair matters because it forces academia to confront that boundary directly.
A common interpretation behind DEI policies begins with observed demographic disparities. If a group represents a substantial proportion of the population but is dramatically underrepresented among professors, executives or other prestigious occupations, possible explanations include historical discrimination, unequal opportunities, socioeconomic disadvantage, prejudice and institutional barriers.
This can be in terms of race, gender, sexual orientation, or disabilities. From this perspective, persistent disparities can indicate that apparently neutral systems are not producing genuinely equal opportunities, and DEI policies therefore seek to identify and reduce those barriers.
Cofnas challenges this interpretation at a more fundamental level. In his 2024 essay “A Guide for the Hereditarian Revolution”, he defends hereditarianism: the controversial hypothesis that genetic differences may contribute to some average psychological differences between populations, including differences in measured intelligence.
His political argument follows from this: if demographic groups do not necessarily have identical distributions of every relevant trait, unequal outcomes cannot automatically be attributed to discrimination. This does not mean that he attributes every disparity to genetics or dismisses environmental influences; rather, he argues that biological explanations should not be excluded in advance.
His critics have serious objections. Among them is the difficulty of treating racial categories as biological populations. The US National Academies concluded in 2023 that race is not a scientifically valid measure of the structure of human genetic variation and recommended that researchers should not use race as a proxy for genetic variation. Cofnas also goes much further than merely suggesting that the issue deserves study: he describes “race realism” as scientifically correct and argues that widespread acceptance of hereditarianism would undermine what he considers the intellectual foundations of contemporary progressive DEI politics.
There are therefore perfectly legitimate reasons to think Cofnas is wrong. But that is precisely where the academic-freedom question begins.
In 2024, Emmanuel College terminated Cofnas’s affiliation as a College Research Associate following his controversial writings. According to the letter reported by Daily Nous, the committee concluded that his blog “amounted to, or could reasonably be construed as amounting to, a rejection” of Diversity, Equality and Inclusion policies and that his ideas represented a challenge to the College’s “core values and mission”.
Cofnas was not dismissed from his main Cambridge research fellowship at that point; his College affiliation was separate from his Leverhulme-funded position in the Faculty of Philosophy.
Cambridge University subsequently conducted its own investigation and reached a notably different conclusion. It stated that Cofnas’s published views were regarded by many as offensive but “did not breach the law and did not contravene University regulations designed to uphold free speech”. Cambridge had already made clear that the academics it consulted “vehemently disagree with and challenge the academic validity” of his claims while simultaneously defending his freedom to express them.
That distinction matters. “Your hypothesis is wrong” invites an argument; “your hypothesis is incompatible with our institutional values” potentially closes one. A university should be able to say that an academic’s evidence is weak, his reasoning defective or his conclusions offensive, while still allowing the argument to be tested in public.
Cofnas is not the only academic to have faced serious personal or professional pressure after challenging prevailing views. In 2016, Harvard economist Roland Fryer published a major study of racial differences in police use of force. He found substantial racial disparities in non-lethal force, but, contrary to his own expectations, found no racial differences in officer-involved shootings after examining the available data. The finding was particularly controversial in the context of the Black Lives Matter debate.
Fryer later said that colleagues had warned him not to publish the result and that threats following its publication became serious enough for him and his family to require police protection for around a month. Fryer’s original NBER study Fryer discussing the threats and police protection
A second example is philosopher Kathleen Stock at the University of Sussex. Stock defended gender-critical views, including the position that biological sex remains relevant in law and public policy. Her views prompted sustained protests, posters demanding her dismissal and accusations of transphobia; she was advised by police to install CCTV at her home and ultimately resigned from Sussex in 2021.
Importantly, the university itself defended her academic freedom and stated that there had been no substantive allegations of wrongdoing against her, while acknowledging that she had not been able to work free from bullying and harassment. University of Sussex statement on Stock’s resignation Contemporary account of the protests and pressure on Stock
If Cofnas’s suspension was, even in part, a response to his public exposure of inconsistencies in Jason Arday’s biography, alleged plagiarism and concerns about the quality of his research, that raises a serious question of principle. It would risk looking less like scrutiny of misconduct and more like punishment of the person who exposed it. Nor was public disclosure apparently the first resort.
Concerns about Arday’s work had already been raised with Cambridge as early as 2023, and Times Higher Education later spent months investigating plagiarism allegations, assembling a 63-page dossier and consulting external academics. According to Retraction Watch, the investigation was ultimately abandoned after Arday instructed Carter-Ruck, one of Britain’s best-known defamation law firms, to write to the publication (Retraction Watch). If internal complaints had already failed to resolve the issue and a major news organisation had withdrawn its investigation after legal intervention, taking the evidence directly to the public could reasonably be presented as a last resort.
The concern would be less significant if there were no evidence of a wider chilling effect in academia, but several surveys suggest that political self-censorship is a real phenomenon. A major Policy Exchange survey of UK academics found that 53% of respondents identified with the political left, 35% with the centre and 9% with the right (p. 53). More importantly, 32% of right-leaning academics reported refraining from airing views in teaching or research, compared with around 15% among academics on the centre and left (pp. 55–56). Among currently working right-leaning academics in the social sciences and humanities, the figure reached 50%.
The same report estimated, using an anonymous list experiment, that around 32% of academics would discriminate in a job appointment against a candidate known to have supported Brexit by voting Leave in the 2016 referendum, while noting that political discrimination can occur on both sides.
American evidence points in a similar direction. The FIRE Academic Mind survey of 1,491 US faculty found that 52% worried that something they said could be misunderstood and damage their job or reputation; the figure rose to 72% among conservatives, compared with 56% of moderates and 40% of liberals.
Faculty were also divided exactly 50–50 over mandatory DEI statements: half considered them legitimate employment requirements, while half regarded them as ideological litmus tests that violate academic freedom. Among conservative academics, 90% took the latter view.
A more recent University of Wisconsin–Madison faculty survey found that 70% of faculty identified as liberal and 9% as conservative; in the humanities, the figures were 74% liberal or extremely liberal and 6% conservative. Researchers also conducted randomized hiring experiments and found that, depending on the political issue, faculty were between 11 and 38 percentage points less inclined to hire a candidate who had expressed a conservative position than one expressing the corresponding liberal position.
None of these surveys proves that universities systematically silence every dissenter. Indeed, both Policy Exchange and FIRE found substantial support for academic freedom among faculty. But they do show that fear of professional consequences, ideological imbalance and political self-censorship are not imaginary problems.
None of this depends on hereditarianism being correct. Cofnas may be wrong, his critics may have the stronger scientific arguments, and some of his statements are deliberately provocative. Universities also have legitimate obligations concerning discrimination and harassment. But none of those points removes the need for debate; if anything, they make it more important.
Universities exist partly because difficult questions cannot be settled by deciding which conclusion is most morally comfortable. An academic proposition should ultimately be defeated because its evidence is weak, its methodology is defective or its reasoning is unsound. Calling someone racist may sometimes be an accurate moral criticism, but it is not, by itself, an answer to an empirical argument.
The deeper danger is not simply that many academics share similar political opinions. People are entitled to whatever political beliefs they reach. The problem arises when ideological homogeneity combines with institutional power: when academics believe expressing certain views could damage their careers, when political opinions influence hiring decisions, or when challenging an institutional doctrine is treated as evidence that someone is incompatible with the institution itself.
DEI is intended to stand for diversity, equality and inclusion. Supporting those goals is perfectly legitimate. What is much harder to reconcile with them is a culture in which disagreement is quickly converted into an accusation of racism and then followed by exclusion, suspension or other professional sanctions simply because an academic advances a different interpretation. If diversity means only diversity among people who already agree, it is a very limited form of diversity; if equality disappears when someone holds an unpopular view, it is not much of an equality; and if dissenters are excluded rather than debated, inclusion has also lost much of its meaning.
That is perhaps the central contradiction exposed by the Cofnas affair. DEI can legitimately argue for greater fairness and representation, but it weakens its own principles when disagreement is treated as moral disqualification rather than something to be answered with evidence. An academic culture in which “I think you are wrong” becomes “you are a racist and therefore should not be here” impoverishes scholarship and makes genuine dialogue almost impossible.
Universities should be capable of defending DEI, criticising hereditarianism and rejecting Cofnas’s conclusions without requiring his removal from the conversation. Otherwise, diversity, equality and inclusion risk becoming principles applied to identity but not to ideas.