Depoliticising the academy

The final guidance was issued by the Office for Students on 19 June 2025 for implementation of the Higher Education (Freedom of Speech) Act 2023. This guidance is effective from 1 August 2025. In another article, Abhishek Saha and I distilled the primary issues at stake (a longer version is here, including references to the appropriate paragraphs of the guidance).

This is not the end of the story for the Act, by any means. Of the significant parts of the Act, Statutory Instrument 2025 No. 528 (C. 23) passed Sections 1 and 2, dealing with duties of registered higher education providers and constituent institutions, and the guidance relates to these. On 26 June 2025, the Department for Education issued a new document laying out their future plans for the Act, which will involve further legislation. As previously indicated, Sections 3 and 7, dealing with duties of students’ unions and their regulations, will be repealed entirely, as will Section 4, dealing with civil claims (the so-called “statutory tort”). 

New legislation will revise the earlier Section 6, on regulation of duties of registered HE providers, Section 7, the complaints scheme (the DfE make clear that the OfS will be granted a “power” rather than “duty” to consider complaints), and the highly complex situation relating to Section 9, on overseas funding. 

In a special event hosted by the Westminster Higher Education Forum, Director for Freedom of Speech and Academic Freedom at the OfS, Arif Ahmed, in response to a question from me, made clear that he and the OfS appreciate the narrow time scale between the issuing of the guidance on 19 June and its becoming operative on 1 August, and the regular qualifier used throughout the guidance of what is “reasonably practicable” applied to this.

But what can and should providers do now, and in the near future to ensure compliance as soon as possible? I believe it is realistic that they could implement most things within the course of the 2025-26 academic year, and should do so. Here I am laying out a proposed three-step plan which could provide a model, all aspects of which are based upon detailed reading of the guidance. 

The first stage can begin with plain removal of various things in order to preserve academic freedom. This includes all requirements for pledges of allegiance to contested aspects of EDI and other ideologies from processes involving admissions, appointments, promotions, probations and training (as specified in the guidance (hereafter R24), paragraphs 135-153, Exx. 27, 32, 34). This does not of course mean abandoning wider legal EDI duties, and the guidance is emphatic on freedom of speech within the law. No academic should any longer be forced to issue a statement of compliance with the idea of “decolonising the curriculum”, gender identity theory, and so on.

Then highly politicised job titles (Ex. 34) — whether Reader in Education and Social Justice, Professor of Postcolonial and Decolonial Studies, but equally anything which might require commitment to positions on free enterprise, market deregulation, and so on — should be immediately stopped and replaced with new ones which are more politically neutral if they have already been advertised, so that appointments will not involve discrimination (para. 138). No new appointments should have such titles, and I would suggest that existing positions are renamed, to at least lessen the stigma.

All ongoing processes for appointments, promotions and so on need to be fully documented (including, I would argue, shortlisting), for transparency in the event of allegations that some individuals have been penalised for exercise of their academic freedom (R24, paras 135-153). There are already often processes in place for auditing such things, especially with appointments, so it should be a reasonably straightforward step to extend these, or at least bring in a member of administrative staff to keep full records. Ideally the individual who does this should be appointed independently of the department or school in question. 

There remains other “soft” means by which free speech and academic freedom can be curtailed

Institutions would then do best to simply remove all existing statements and pledges of their own on wider subjects (i.e. not relating solely to higher education) such as might limit full support for academics under wider attack if they dispute these (para 207, Exx. 51-52). Ahmed indicated in a speech at a recent event at Oxford University that universities might consider such “institutional neutrality” as part of their duty to promote academic freedom. Provisional free speech codes need to be drawn up in line with the regulations (paras 167-177) and distributed as early as possible in the autumn (for when teaching begins), making clear how these break with previous practice (for example on “misgendering” (Ex. 35), rules against offending religions (Exx. 15, 20), “reputational” concerns (paras 62, 123) and so on), then final versions drawn up by the spring.

The second stage, which I believe can be undertaken during the autumn, involves appointment of an institutional director of free speech/academic freedom. This is not mandatory according to the OfS guidance, but much to be recommended, not least because of the amount of work which will need coordinating to ensure compliance. I would suggest this should be of a comparable level of seniority to the most senior academic tasked with EDI, as the demands are equally important. There could also be associated free speech/academic freedom “leads” in different schools and faculties. Then a time-limited internal process for dealing with staff or student complaints should be drawn up (paras 162-166), as well as guidance for dismissing trivial complaints threatening free speech (para. 166).

The third stage, perhaps the most labour-intensive, is the institution of training for all staff involved in any decision-making for which there are dimensions of freedom of speech and academic freedom (paras 208-212). This is likely to include the majority of academics and professional staff. The nature of the training will need planning (to be meaningful it should be more than just an online questionnaire). This could be done in the Spring Term so that staff undertake such training in the Summer Term. The other exacting duty relates to the filtering of freedom of speech and academic freedom into all committees and governance, as well as audit of professional accreditation, with representations made where appropriate (paras 115-119, 207, Exx. 51-52). This should be able to be completed through the Spring and Summer Terms, so that new practices can be in place by the beginning of the 2026-27 academic year. They can also during this period be guidelines issued on freedom of teaching within approved specifications, whilst considering boundaries to avoid such teaching from being hijacked for political preaching void of critical engagement or for extreme iconoclasm which ignores established methods.

Compliance with the OfS guidance is no small matter, but institutions will gain considerably not only from the very fact of doing so, but also in terms of reputation in terms of their relevant commitment. Issuing a “road map” for implementation by many, then seeing it through, would be an excellent step right now. 

Notwithstanding the detail of this guidance, there do however remain other “soft” means by which free speech and academic freedom can be curtailed. These include discrimination by external organisations to which academics are beholden, or by wider internal processes which can be exclusionary when an individual is ostracised for political reasons: internal assessment of outputs by colleagues (which can affect probation and promotion), exclusion from collaborative projects (participation of which can be very important in some disciplines), delegation of teaching, and so on. The OfS guidance on implementing free speech/academic freedom into governance (paras 189-193) does cover some of this, but more detail and forensic examination of the processes is likely necessary. 

Then there are other institutions to which academics are beholden — research councils and other funding bodies, journals, and of course the Research Excellence Framework (REF). Various principles of HEFOSA, especially relating to non-discrimination on political/ideological grounds, could be extended to apply to funding bodies in some way linked to government. The same process could apply to other institutions which constitute part of “academic citizenship” — conferences, events, positions within scholarly/professional associations, to avoid political exclusion. There are conditions currently placed upon journals in terms of allowing open access, if articles published in them are to be allowed to be submitted to the REF, so it should not be impossible to also insist that they abide by the principles of HEFOSA, also with respect to appointments therein. The problems with REF 2029 are deeper, as I have written about earlier in this journal. Since this article, and an opaque consultation process, the confirmed guidelines strongly imply a shift away from academic excellence towards ideological conformity. All research-active academics are required to participate in the REF, and post-2029 REF needs serious reevaluation in light of this regressive step, at odds with other developments in higher education. 

Then there is the bigger question of heavy politicisation of teaching. This is not strictly excluded according to the OfS guidance, though discrimination against students on political grounds, or requirements for academics to conform to a political agenda, are. Nonetheless, seeing some teaching which makes no attempt to convey any type of balanced perspective and neither incorporates views which differ from those of particular academics, is not an edifying spectacle, and the dissolution of scholarship into politics may be a reason for diminishing wider public confidence in universities as institutions. At the very least, further thought is needed to ensure universities remain politically plural environments, and how to ensure wider academic requirements do not disadvantage those of heterodox but scholarly views.