Dismantle the infrastructure of censoriousness | Freddie Attenborough

At least 12 UK universities, including eight Russell Group institutions, have paid a private intelligence firm more than £440,000 since 2022 to monitor staff and student social media, track protest-related speech, and conduct background checks on academics and invited speakers, with a number of reported cases centring on pro-Palestine activism and events linked to the Gaza conflict.

The development raises troubling questions about how universities are beginning to make decisions about lawful expression, as the Government encourages a shift away from the relatively narrow statutory test of whether individuals are being drawn into terrorism towards a broader concern with whether speech may contribute to a “permissive environment” for radicalising influences.

According to a joint investigation by Al Jazeera and Liberty Investigates, Horus Security Consultancy Limited — a firm which describes itself as a “leading intelligence” provider — was commissioned by universities to produce “open-source intelligence” reports on staff, students and external speakers, trawling social media feeds and conducting secret counter-terror “threat assessments”. The company was established in 2006 as a project within the University of Oxford’s security team by former Lieutenant-Colonel Jonathan Whiteley, who, according to the company’s website, has a “23-year career running security, intelligence and counter-intelligence operations all over the world”.

Its higher education clients include Manchester Metropolitan University [MMU], the London School of Economics, the University of Bristol, the University of Oxford, Imperial College London, University College London, King’s College London, the University of Sheffield, the University of Leicester, the University of Nottingham and Cardiff Metropolitan University. There is no suggestion that this activity, which came to light after more than 150 FOI requests to UK universities, was unlawful.

Much of the reporting centres on the surveillance of student protest activity, which is troubling enough. But some of the most striking findings concern the scrutiny of invited speakers and academic events, and the extent to which lawful but contentious scholarship is now being drawn into a securitised framework in which the line between speech and action — long central to free speech protections — is eroded, and ideas are approached less as arguments to be contested than as possible predictors of future violence.

One of the clearest examples concerns MMU. Emails between Horus and university staff, together with a copy of the threat assessment itself, show that in 2023 the university asked the firm to conduct a secret counter-terror “threat assessment” of Rabab Ibrahim Abdulhadi, the 70-year-old Palestine studies scholar at San Francisco State University, ahead of a memorial lecture she had been invited to deliver in honour of British student Tom Hurndall, who was killed by an Israeli sniper in Gaza in 2003.

The resulting report examined Abdulhadi’s social media activity and revisited historic allegations made against her, including claims previously dismissed by her home institution and by a US federal judge. MMU ultimately allowed the event to proceed, concluding that there was no evidence of involvement in proscribed organisations and that any protest was unlikely to be violent.

Separately, the University of Manchester confirmed that it has commissioned similar reports on two guest speakers invited to discuss “geopolitical issues”, although it declined to identify them.

In explaining its decision to seek a threat assessment of Abdulhadi, MMU invoked the Counter-Terrorism and Security Act 2015. Section 26 of that Act placed the core Prevent duty, which had previously existed in policy form, on a statutory footing, requiring universities, as specified authorities, to have “due regard to the need to prevent people from being drawn into terrorism” — in essence, to monitor the exposition of non-violent ideas and ideologies on campus in light of their potential association with radicalisation. 

However, section 31(2) of the same Act makes equally clear that this duty is not freestanding. In carrying it out, universities must “have particular regard to the duty to ensure freedom of speech” and to “the importance of academic freedom”. Section 31(3) further provides that, in issuing guidance to universities, the Secretary of State must also have “particular regard to the duty to ensure freedom of speech”. The rationale is straightforward. Various ideologies identified by government as being associated, in certain contexts, with pathways into radicalisation — for example, Islamism and strands of right-wing thought such as “cultural nationalism” — are also ideas that, in a university setting, may be the object of entirely legitimate academic inquiry, critique or debate.

That statutory balance is, at least in principle, reflected in the Office for Students’ approach to freedom of speech. Regulatory Advice 24 states that providers should maintain a “high tolerance for all kinds of lawful speech” and that there should be a “very strong presumption in favour of permitting lawful speech”. On that account, the question for universities is, or should be, a relatively narrow one: not whether speech is politically contentious or perceived to be troubling, but whether it is lawful and, if so, whether there are reasonably practicable steps that would allow it to proceed, or to be conducted, within the law.

In practice, however, the position is more complicated. Alongside this high-level commitment to lawful speech, the Prevent framework requires universities to “assess the risks associated with Prevent and draw up a plan to mitigate these”, to “have systems for assessing and mitigating risks around external speakers and events on campus”, and to embed those requirements across institutional policy — including IT usage and research governance — while ensuring that students, students’ unions and societies are aware of, and operate within, those policies.

The result is a dual structure in which a strong presumption in favour of lawful speech sits alongside an increasingly elaborate apparatus for identifying and managing risk. As the handling of Abdulhadi’s lecture illustrates, it’s a combination that in practice can pull in different directions. Nor is that dynamic confined to a single case. Over the past decade, a series of incidents has shown how readily lawful academic work can be drawn into a security-oriented frame: students warned to act with caution when engaging with an essay by a leading left-wing academic on the ethics of a socialist revolution; discussions at Glasgow School of Art about referring a student’s artwork on the geopolitics of the Middle East to Prevent teams; a criminology lecturer reporting that she had run her course reading list past the police because of concerns about its content; and a Muslim PhD student cancelling a conference on counterterrorism legislation citing fears of Prevent.

It is not that the legal framework has changed, but that the anticipatory logic of section 26 — and the need to identify susceptibility to radicalising ideologies — has become more prominent in practice, while the section 31 duty has, at times, receded into the background.

The 2023 Independent Review of Prevent, for instance, argued that the programme had, in some respects, drifted towards a safeguarding model, with too much focus on vulnerability and insufficient attention to the power of ideas. In response, it called for a renewed focus on the role of ideology in radicalisation, including the “permissive environments” from which individuals may radicalise and be recruited, and in particular on forms of non-violent extremism that fall below the criminal threshold but may nevertheless create an “environment conducive to terrorism”.

Inevitably, that shift brought universities more directly into view not only as “specified authorities” but as sites of knowledge production and intellectual exchange in which such environments might plausibly be thought to emerge.

Prior to the Review, and perhaps in recognition of the difficulty of reconciling the term “permissive environments” with section 31, the sector-specific guidance for higher education did not use the term at all. Yet in the unified guidance that replaced it in December 2023, the phrase appears repeatedly – more than twenty times, in fact – as part of a broader shift towards identifying and managing the environments in which radicalisation may occur. In that guidance, there is a renewed focus on tackling the “ideological causes of terrorism”, reducing the influence of radicalisers, and addressing “permissive environments… across academia, civil society, communities, government and industry”. At least at the level of framing, universities were thereby drawn into a common conceptual scheme in which they appear, on the surface, much like any other specified authority.

In 2024, the phrase “permissive environment” appears again, this time in the Government’s new non-statutory definition of extremism, carrying the same conceptual weight into a further layer of this increasingly baroque framework. Whereas the earlier 2011 definition focused on “vocal or active opposition to fundamental British values”, the revised definition retains that broad orientation but adds an additional limb: the “intentional” creation of a “permissive environment” for others to negate fundamental rights or undermine liberal democracy.

That same shift in emphasis is now being brought more squarely into the higher education context. In Protecting What Matters, the Government indicates that the Department for Education will issue new non-statutory guidance on external speakers and events under the Prevent duty. According to that policy document, universities will be provided with “principles” to “help… assess when speech… is lawful but could enable a ‘permissive environment’ for radicalising influences”.

There are, of course, safeguards built into the 2023 statutory guidance to protect academic freedom, just as there are in the new definition of extremism — most notably explicit protections for lawful expression, and requirements to base judgments on intention, context and demonstrable behaviour rather than inference — and the forthcoming non-statutory guidance will, presumably, adopt a similar form.

But once translated into institutional policies and training modules, a phrase like “permissive environment” can all too quickly take on a life of its own, coming to be rendered less as a contextual, evidence-based concept linked to a statutory definition, and more as a simple descriptor for a space in which problematic activity is presumed likely to develop if certain forms of speech are permitted. 

That might sound far-fetched, but consider just how many universities over the years have developed “harassment” training that purports to convey the meaning of section 26 of the Equality Act, yet has all too often been reduced to a perception-led standard, detached from the statutory requirements of context and an assessment of whether it is reasonable for the conduct to have that effect.

The risk is that the new non-statutory guidance will effectively lower the practical threshold from a relatively tight statutory concern with people being drawn into terrorism to a broader and more speculative concern with the atmospheres, pathways and ideological climates within which lawful speech takes place — i.e., precisely the kind of expansion that section 31 was intended to guard against.

In a university setting, that creates obvious pressure towards precautionary restrictions on lawful debate, platforming and association. It also helps explain the appeal of firms like Horus and others, including Mitie Intelligence Hub and Global Situational Awareness, operating in a similar surveillance space.

Seen in that light, the Horus material is troubling not only for what it reveals in the here and now, but for how digital technology, private intelligence and a lowered threshold for intervention could combine on campus. The FOI disclosures show how readily this shift in emphasis can be translated into practice, with outsourced data-gathering used to build a wider picture of speakers, associations and public statements to assess whether an event should be permitted to go ahead, often with limited visibility as to how that information is gathered, interpreted or acted upon. So while the statutory protections for academic freedom remain, the question is how far they will continue to shape practice once the language of “permissive environments” is absorbed into training and day-to-day decision-making.