On free speech, universities need jaw-jaw not see-saw

OfS lost on eight grounds, which my colleague David Kernohan and I have rattled through elsewhere on the site.

The regulator’s own witness evidence – the line about wanting to “send a signal” to the sector – tells you what Sussex really was. A contested case being used to model an enforcement architecture.

But the verdict raises a question that gloating will obscure. How did a regulator come to be applying a free-standing version of academic freedom in the first place?

For me, the answer is that by the time the case reached the regulator, free speech had already been isolated from the surrounding academic settlement and turned into a standalone regulatory object.

Equality, dignity, participation, scholarly standards, and the conditions of academic work had been left elsewhere. Sussex is what happens when that isolation meets a contested case.

And it will keep happening – on different facts, with different protagonists – until universities can do the integration work that isolation made impossible.

That integration isn’t an academic indulgence. It underwrites the claim to institutional autonomy itself. Autonomy is earned by showing a working capacity to reconcile competing views, benefits, risks and harms within a single account of academic life, not by getting any particular decision right under the glare of media coverage.

That capacity – the working version, not the published one – is what government, the public and regulators no longer believe the sector possesses.

Plank statements

I’ve argued before here that much of the debate surrounding the campus culture wars resembles a see-saw – derived from Kant, articulated by Erich Fromm in his 1941 work Escape from freedom, and made famous by Isaiah Berlin’s 1958 essay, Two concepts of liberty.

There’s freedom from harm, most often invoked by EDI advocates. And freedom to speak, most often championed by free speech advocates. At any one time, on any given issue, in any particular political context, a dirty great sandbag is plonked on one end of the see-saw.

It’s why there’s a free speech “champion”. It’s why universities must “root out antisemitism”. It’s why, when students enrol, they’re treated to two of those tickbox online induction modules – one on EDI and harassment, the other on freedom of speech.

Both sides also borrow the other’s arguments – free speech campaigners argue that those with controversial views suffer oppression and harm (i.e. cancellation), while EDI campaigners argue that those with (or who advocate for those with) protected characteristics have their freedom of speech harmed by the oppression they face.

Politicians rarely reconcile the two. They tend to react to a case, a phrase, a protest, or a headline, and even when both ends are mentioned, one is rhetorically maximised while the other is minimised.

So “we stand for absolute freedom of speech within the law but would never tolerate Holocaust denial” places maximal definitional weight on one end, with careful minimalism on the other. “The London hate marches and use of the phrase ‘globalise the intifada’ are encouraging antisemitism, but of course we respect the right to protest” does the same thing in reverse.

The duty on universities should be to integrate the two ends of the see-saw, not to selectively dump sandbags on one end over issue X or issue Y. The fatal flaw in where the legislation and OfS regulation has ended up is that it still treats the two largely as separate issues – you do one with regard to the other, or the other with regard to the one.

Pulled apart

The Higher Education (Freedom of Speech) Act (HEFoSA) is asymmetric, but the structural issue beneath it is isolation. HEFoSA treats free speech as a thing that can be detached from the academic settlement around it and enforced as a single value, by a single function of the regulator, with its own director, against a single duty, through a single complaints scheme.

It then asks the institution to “have particular regard” to that value when balancing it against everything else.

We only have to look at where everything else lives. Equality might sit with an EDI office and an Equality Act compliance team, dignity with HR and the dignity-at-work policy, academic freedom with HR, the academic registrar, or in nobody’s hands at all, and scholarly standards with quality or a validation panel.

On the ground, each of these has a constituency. EDI staff have an office, committees, and colleagues to think with. Free speech advocates have networks – academics with strong views, sometimes a campus group, often a national affiliation. Students have the SU, and within it the clubs and societies organised around exactly the value claims at issue.

Each silo is a community as much as a compliance regime – somewhere to go, someone to talk to, a tribe. Each has its own staff, policies, regulator or quality body, vocabulary and risk register. None owns the question of how the values relate to each other.

But that granular layer is where integration has to happen – where those people meet, where the values are actually held by someone, and where a contested case first arrives. Strategy documents above them don’t reach it.

Once a value is isolated in this way, the regulator that enforces it has no institutional grammar for connecting it back to anything. Arif Ahmed’s brief is to maximise free speech, with a footnote about the law. The C6 harassment compliance brief is to deliver an inclusive environment, with a footnote about lawful expression. Neither is asked to weigh one value against another.

Each enforcement track demands positive engagement, not bare minimum compliance. So each maximises within its own silo, assumes someone else is handling the relationship between values, and leaves the institution to improvise when a contested case arrives.

The result is a parallel set of compliance regimes, each speaking the language of its own remit, with the gaps between them filled by executive judgement and expensive legal advice.

That executive judgement is mostly organised around reputation – the proxy that fills in when an integrated academic account is missing, calibrated to headline risk, funder anxiety, and regulator attention. Whichever side looks most exposed in any given week tends to win the concession, and the underlying value question goes unasked.

That is the architecture that produced Sussex. The regulator overreached on a single duty, but the deeper problem is the sector’s.

Beyond the whistle

The temptation, then, is to read the judgment and wider regime as a balancing problem. Universities have free speech on one side and EDI on the other – the job is to weigh competing claims and reach proportionate outcomes. The see-saw metaphor invites the reading. But, with apologies in advance, I also think it gets the analysis wrong.

A university’s task is to explain how both ends of the see-saw belong within academic life. Treating them as external claims that happen to land on the desk misreads the job.

Speech is constitutive of the academy. Universities depend on critique, disagreement, the pursuit of truth, and the exposure of received positions to challenge. Without protected speech, the institution stops being a place where claims can be tested. It becomes something else.

Dignity and equal participation are constitutive in the same way. Academic life is a collaborative practice. People who are excluded from it, degraded within it, or made unable to participate are being prevented from doing the same activity that the speech protections exist to make possible. Excluding people from inquiry damages inquiry.

Academic freedom is the third element. It’s a professional and institutional condition – security of employment, time to research, stable funding, status, and participation in the governance of the conditions under which work is done – not lawful expression with a protection halo. Without those material conditions, the speech protections protect a hollowed activity.

Integration means an institution can articulate this connection and act on it – an intellectual claim about what universities are for, and a working account of how each value depends on, and is constrained by, the others within a single conception of academic life. Refereeing rival claims and procedural fixes don’t reach it.

Nature abhors

When universities don’t articulate an integrated account, somebody else does it for them.

Parliament does it through HEFoSA, OfS through enforcement priorities, the courts through judicial review, and campaign groups by selecting individual cases for media attention. Executive management does it through risk-managed compromises drafted in the gap between conflicting policies, lawyers through caveats, and consultants through training packages.

None of those external definers can produce an integrated account, because none of them is structurally suited to it. They each see a fragment, and they each maximise their own bit. The cumulative effect, inside the university, is a patchwork of borrowed external framings, none of which the institution actually owns.

It connects to the long-running hollowing of academic governance – the migration of decision-making from Senates and faculty boards to executive teams, the conversion of SU governance roles from membership of the academic community to consumer-voice consultation, and the dominance of Council membership by external members from corporate backgrounds.

None of this is nostalgia for Senates. Articulating integrated values requires shared deliberation among people who actually do the academic work, and “managed” institutions don’t have the structures for that. Published “values” tend to be marketing copy. Operational values are the ones embedded in the silos.

That makes the obvious defensive card – invoking “institutional autonomy” against further regulation – tactically attractive to play, but intellectually empty in its current form.

Institutional autonomy as currently defended means executive autonomy. An executive that hasn’t articulated an integrated account of academic values can’t credibly claim the autonomy to balance them. It’s just substituting managerial discretion for regulatory discretion.

Both produce policy without grammar, both leave the integration question unanswered, and both will lose the next contested case to whoever has the clearest single-value framework – which is never the institution itself.

That is also why external bodies have moved in. Government, regulators, and “the public” (the press) read the executive’s evasions as evidence that the sector can’t hold these values together at all. HEFoSA, the C6 conditions, and the proliferation of single-issue compliance schemes are each drafted on the working assumption that universities can’t be trusted to do the reconciling work themselves. Harsh as the assumption is, it tracks what the sector has visibly stopped doing in public.

Polish lessons

Models are available. As I’ve noted before here, Poland’s 2018 Law on Higher Education and Science requires partial democratic governance at state universities. Within that, Jagiellonian University created the Advocate for Academic Rights and Values in 2020.

The title is the argument – not Free Speech Champion, not Equality Lead, with press-releasable projects for each. An advocate for academic rights and values, plural and integrated. The role exists because the institution refused to split the terrain into rival compliance regimes. Speech, dignity, scholarly standards, and student participation are treated as elements of one academic order, and the office is responsible for all of them.

The first holder, Beata Kowalska, faced a textbook contested case in 2021. Magdalena Grzyb, a criminologist, was due to lecture on the criminological aspects of femicide as part of the 16 Days Against Gender-Based Violence campaign. The co-hosting student society tried to cancel her, citing her past public criticism of certain understandings of non-binary and queer identity. The dean reinstated the lecture on academic freedom grounds.

A trans-activist student, Ernest Figiel, then publicly advocated for “TERFs” to be “thrown into a sack and into a lake”, “disposed of in lime pits”, and praised Stalin’s methods of dealing with “enemies of the people”. Disciplinary proceedings followed, accompanied by counter-campaigns. The university was in serious trouble.

Kowalska did not pick a side. She articulated a single standard. Hate speech allegations are investigated carefully regardless of who makes them. Dehumanising or extermination-metaphor language transgresses freedom-of-speech limits. Grzyb’s lecture proceeded under academic freedom protection. Figiel’s calls for the murder of his ideological opponents fell outside that protection. Same threshold, same proportionality, same forum.

One office working from one framework, protecting both academic freedom and dignity by reference to the same set of academic values.

The model is structural rather than a matter of compromise or mediation. The institution refused, at the level of its job titles and statutory obligations, to let its academic values be split into rival enforcement missions.

The wider European conversation is moving the same way. In April 2023, the League of European Research Universities (LERU) framed academic freedom as a fundamental right with three dimensions, including a positive institutional obligation that can’t be reduced to non-interference.

The Guild of European Research-Intensive Universities has called for a European Ombudsperson – an arbiter forum operating alongside the European Court of Human Rights, not a champion role for one value within a national regulator.

John Heathershaw’s working group has separately argued that severe academic freedom cases require reporting to “a genuinely independent ombudsperson, not to a free speech champion”.

The framing across these interventions is pointed. Europe’s research-intensive universities are asking for institutional arrangements that hold values together, not stronger single-value enforcement. The UK’s director for freedom of speech and academic freedom sits awkwardly with that direction, because it locates one value inside a regulatory enforcement function rather than creating an institutionally trusted forum for academic values as a whole.

Three-line whip

So three tests for the sector.

First, can the university state where its integrated account of academic values is formally owned – one that treats speech, dignity, equality, student participation, academic freedom, and scholarly standards as elements of a single account of academic life, rather than as silo-specific compliance objects with separate vocabularies?

Second, can it show how that account shapes actual machinery – HR, EDI, complaints, academic registry, governance, student participation, and the free speech code – rather than leaving those systems to operate in parallel?

Third, is there a trusted internal mechanism – an ombuds, a Senate process, an academic rights office, or some equivalent – with enough independence and standing to handle contested cases before they are exported to the regulator or the courts?

Not every dispute can be resolved internally. Courts and regulators will continue to exist, and some cases will legitimately reach them. The point is that universities should be able to show an internal academic grammar for these questions, so that external bodies aren’t the first place the values are brought into relation with each other.

What Sussex actually does is postpone the autonomy question. The judgment buys the sector time, but it doesn’t restore the credibility that autonomy actually rests on. That credibility comes from a visible, working capacity to reconcile competing claims internally, and the existence of the dedicated regulatory function is itself evidence that this capacity isn’t currently believed to be there.

The integration work has to do more than defend against the next contested case. It’s how the autonomy claim gets rebuilt – substantively, not rhetorically. Either universities can show that the capacity to hold their academic value framework together lives somewhere inside them as working practice, or they’ve outsourced it.

That government, the public and regulators don’t believe universities to be capable of doing so is the real response to get to work on.