Essay · Race, universities and the public sphere

The uses of Jason Arday

A national press campaign about scholarly rigour is resting on a self-published, machine-generated blog post. The campaign defending him is resting on a claim the evidence will not carry. Both have made one man stand in for an argument he never agreed to have.

Professor Tahir Abbas  ·  First published 5 August 2026  ·  Updated 6 August 2026

Update Jason Arday resigned from Cambridge and from Jesus College on 5 August, hours after the university opened an investigation into his qualifications and honorary appointments. The piece below is unchanged from publication; a postscript follows at the end.

In September 2025, Times Higher Education had a story ready about Professor Jason Arday of Cambridge. It rested on a sixty-three page dossier of text comparisons between his 2015 doctoral thesis and another academic’s dissertation. The magazine did not run it. Arday’s solicitors, Carter-Ruck, had written to the editor. On 3 August this year, with the story out anyway and running everywhere, the editor Chris Havergal published a note declining to explain that decision but accepting responsibility for it, and adding that the inquiries “reflected matters of legitimate public interest.”

In the eleven months between, the case passed out of the hands of higher education journalism and into the hands of people with quite different interests in it. The philosopher Nathan Cofnas, now at Ghent and until last year at Cambridge, published his own comparison on Substack. Guido Fawkes went after the visiting professorships. The Spectator, spiked and UnHerd took up the theme in near-unison, and Arday is only incidentally its subject. The subject is diversity policy, and what diversity policy is said to produce.

The number doing most of the work in that coverage is 188: sentences in Arday’s thesis described as identical or near-identical to sentences in Paula Zwozdiak-Myers’s 2009 Brunel thesis, with the odds of coincidence reported by the Telegraph at one in a hundred billion. It is worth knowing where the figure comes from. The Telegraph names no statistician. The analysis it draws on was published on 24 July on a personal blog by Alexey Titorenko, and carries a disclosure that its content was generated by an AI model. The post’s own central estimate is nearer one in twelve billion. The 188 is nine sentence pairs at the near-identical threshold, plus a hundred and seventy-nine that share twelve or more consecutive words.

None of which makes the underlying overlap imaginary. Some of it is real and has been examined by people with names and standing. But a self-published, machine-generated analysis has become the evidentiary spine of a national press campaign about scholarly rigour, and nobody on that side of the argument appears to have found this awkward.

I want to start there rather than with the findings, because the findings are not the only thing on the table and they are not, sociologically, the most interesting thing.

I Who gets read twice

Almost no British professor’s doctoral thesis is read again after the viva. A 2015 thesis from a post-1992 university sits in a repository and is consulted, if at all, by two or three people over a decade. What has happened to Arday’s – a sixty-three page dossier, a probability estimate to eleven orders of magnitude, sentence-level comparison against three other theses – I cannot think of more than a handful of British parallels. It is worth being exact about what that establishes and what it does not. It does not establish that the thesis is clean. It establishes that a particular kind of forensic attention was directed at this scholar, and that the question of who attracts such attention is a different question from what it turns up. Both are answerable. They have different answers.

That distinction is the one thing almost nobody in this argument has been willing to make.

The right collapses it in one direction: because the scrutiny found something, the scrutiny must have been disinterested, and anyone raising the disproportion is running interference. The campaign around Arday collapses it in the other. The Good Law Project petition, at nearly fifteen thousand signatures, states that “entirely false allegations of plagiarism were made” and that “investigations have found no evidence whatsoever of wrongdoing.” The open letter, signed by people I respect enormously – Lord Boateng, Lord Woolley, Dame Ijeoma Uchegbu, Kehinde Andrews, Daniel Kebede – says that both institutions “conducted thorough investigations and found no evidence whatsoever of wrongdoing.”

II Both institutions did not

Cambridge conducted none, and says so: its position is that investigations belong to the university where the research was done. Liverpool John Moores ran a confidential procedure and has said publicly only that Arday’s PhD still stands. What is reported of the panel’s reasoning – that the citation issues were honest and reasonable error, and fell well within the accepted standards of the time – has never been published by the university in its own voice. So we have a finding nobody can read, quoted selectively by everybody, against which David Sanders, a biochemist at Purdue who reviewed the THE dossier, says there is “no question that there is extensive plagiarism.”

That is not exoneration and it is not proof. It is an absence, and the absence is doing damage in both directions. My worry about the letter and the petition has nothing to do with their instinct, which is sound. It is that they did not need the additional claim. The case for disproportionate scrutiny stands perfectly well on its own feet. Having attached it to a claim about the evidence that the evidence will not carry, the campaign has staked the credibility of a real and documented argument about racism in British academia on the contents of one man’s footnotes. That was the trade the other side wanted, and it was accepted without a fight.

Our own field ought to have seen this coming. The sociology of race and education is now half a century old, out of the United States in the 1970s, into Britain through the 1980s, into continental Europe more recently, and its central concerns are race, power, identity and labelling. The last of those is doing the work here. A label, once it attaches, becomes the lens through which everything else is read. Arday has now carried two. The first was “Cambridge’s youngest Black professor,” which arrived with the appointment in March 2023 and was repeated in every profile, broadcast and press release. The second is “fantasist.”

You cannot be unmasked unless you have been masked, and the mask was not, in the main, of his own making.

III The Cambridge problem

Which brings me to Cambridge, whose conduct concerns me rather more than anything Arday has done.

The university was approached in 2023, not about the thesis but about a 2018 article in Social Sciences. Its Faculty of Education agreed with him that the article would be taken down while the concerns were addressed, offered him “full support,” and declined the complaint on the grounds that the paper was not within its remit. A correction appeared that December, and another followed in a second journal the year after. The procedural position is defensible on its own terms. Where it has ended up is not. Cambridge now says the allegations have been thoroughly investigated and that Arday “has been cleared by those investigations of any wrongdoing,” and has described the campaign against him as vile. An institution cannot decline jurisdiction and deliver a verdict. Doing both at once is not a defence of a colleague; it is a defence of a decision.

Consider whose interests it serves. Any serious examination of Arday’s record is unavoidably an examination of the appointment – of what was checked, by whom, against what criteria. “There is no case to answer” and “we will not be looking at our own process” are the same sentence in different clothes. Cambridge has attached its institutional dignity to a man it did not protect when protection was cheap, and the consequence is that he is now absorbing damage that properly belongs to a committee.

Paul Sagar, writing in UnHerd this week, put the appointment question bluntly: the record is “so obviously paper-thin that most academics couldn’t have got that position simply based on the CV.” I think that is closer to right than wrong, and I want to be clear that it is a criticism of an institution rather than of a person. Universities have worked out that a professor who is good on television is worth a great deal to them, and the incentives have quietly rearranged themselves to reflect it. Grants, doctoral completions, monographs, the slow accumulation of a record: these still sit on the promotion criteria, and they are still what the rest of us are held to. But an institution that wants a public face will find a way to appoint one. That is a governance failure with a specific address, and the address is not the appointee’s.

IV What the numbers have not done

It is also the characteristic failure mode of equality work done badly. The numbers have not moved.

Thirty years of policy, on and off, and that is where we are. Shifting it means changing structures: recruitment, progression, workload, the distribution of invisible labour that falls on the very few. All of that is slow, expensive and unphotographable. Elevating one person is fast, cheap and photographs beautifully. Symbolic inclusion is what institutions reach for when structural inclusion costs too much, and its defining feature is that the whole weight of an argument comes to rest on an individual who never agreed to carry it.

V The wrong case

So my sympathies fall roughly as follows. With the thousands who signed in good faith and were handed a claim about the evidence that will not survive contact with it. With the colleagues who appointed him and are watching their institution decline to defend its own decision on the merits. Above all with early-career academics of colour, watching this and drawing the obvious lesson, which is that visibility is a liability and the safe career is the quiet one: decline the invitation, refuse the podium, publish and say nothing. That is the chilling effect, and it will outlast every detail of this case.

As for Arday, he says he is not a liar; that his autism led him to rely on mimicry in making sense of information; that his doctoral supervision was inadequate and that he could not afford a proofreader when he started out. Some of that may be true. Some of it may be true and still not be enough. He is entitled to a proper process conducted by people with no stake in its outcome, and he has not had one, because the only body that examined the thesis did so in confidence and has never published its reasoning.

That reasoning should be published. Not because the Telegraph wants blood, though it does, and it will keep pushing until it gets some. Because the alternative is what we have now: a man turned into an argument, two institutions sheltering behind him, a national newspaper campaign resting on an AI-generated blog post, and a field of scholarship that deserves better than to have its legitimacy staked on the outcome.