In the mid-1980s, the Thatcher government introduced a new approach to the allocation of research funding in higher education, entitled the Research Assessment Exercise (RAE). This required a series of the traditional universities to complete a four-part questionnaire on research income, expenditure, planning priorities and output. Assessment was divided between roughly 70 subject categories known as Units of Assessment (UoAs). The RAE ran first in 1986, then again in 1989, 1992, 1996, 2001 and 2008. There were significant criticisms from the outset, on grounds of differing standards between subjects, unclear assessment criteria, and lack of transparency of assessors and an appeals mechanism. Reforms were introduced for the second and then various subsequent exercises. Outputs came to be assessed by a process of peer review from 1989 onwards, two publications were required per member of staff submitted, and wider information was required on research students, external income and future planning. Concomitant with such changes, the proportion of research funding allocated to institutions increased.

Following the 1992 Further and Higher Education Act, allowing polytechnics and other institutions to apply for university status, “post-1992” institutions could submit to the RAE, creating a wider playing field in which they could compete with more traditional institutions. In 1992 itself 192 institutions participated, encompassing over 43,000 full-time equivalent researchers. Nonetheless, the older universities ended up receiving 91 per cent of the available funding. In subsequent rounds, modified systems for ranking outputs were used, research-active staff were required to submit four publications produced during the period since the last exercise, larger “sub-panels” were created, non-UK-based experts were brought in to review work assigned top grades, and other measures were brought in to acknowledge the situation of ‘early career researchers’ not long into their academic careers, those on career breaks, staff transferring between institutions, and so on. Furthermore, the types of research outputs permitted were expanded beyond traditional journal articles and monographs to encompass those in the form of “practice” (especially from 2001 onwards), especially important for those working in the arts, allowing the submission of art works, films and videos, web resources, musical compositions, recorded performances, literary outputs, all of which have stimulated an extensive scholarly literature on the concept of “practice as research”.

After 2008, the system was renamed the Research Excellence Framework (REF), which to date has run twice, in 2014 and 2021. Now institutions were ranked on the basis of three components: “output quality”, weighted at 65 per cent, “impact” at 20 per cent, and “environment” at 15 per cent. “Environment” was assessed on the basis of research income, number of postgraduate research students, and their completion rates. “Impact” was a new category, introduced at the tail end of the last Labour government by the then Business Secretary Lord Mandelson, as an incentive for universities to become more externally engaged. It was defined as “an effect on, change or benefit to the economy, society, culture, public policy or services, health, the environment or quality of life, beyond academia”. Each department had to submit a general statement on “impact”, and between 2 and 7 (depending on the number of research-active staff) impact “case studies”, documenting demonstrable impact, outside of universities, of academics’ work during the cycle dating from the previous exercise. This had to be shown to relate to “underpinning” research of a certain level (a minimum of “2*” quality, from a general REF star system going from 0 up to 4* as the highest) undertaken over a longer period. 

REF 2021 involved wider reforms. Previously, whether an individual academic had four outputs which their institution deemed of sufficiently high quality for REF submission could affect whether they could continue on a research contract or be moved over to teaching-only, whether some on temporary contracts would have these renewed, or whether others would retain their job when a department was cutting staff. This changed as the number of outputs required for a department was modified to 2.5 times the number of full-time equivalent staff submitted (which was all of those whose jobs involved significant responsibility for research). For any single member of staff, anything between 1 and 5 outputs could be submitted, so long as the total was achieved. The weightings of outputs, impact and environment were revised to 60 per cent, 25 per cent and 15 per cent respectively, thus increasing the focus on impact.

REF-bashing is something of a national pastime amongst academics

REF-bashing is something of a national pastime amongst academics. The system, which differs from mechanisms for allocating research funding in almost all other developed countries, has become a central concern for all academics engaged in research, as the primary measure of their achievement in this respect. REF potential is often a major factor in assessing job applications, probation periods, promotion applications and more. It places a good deal of power in the hands of those in departments charged with organising and coordinating a REF submission (often known as “REF leads”), often involving extensive internal processes of assessment prior to this. These can be daunting for individual academics and inevitably vulnerable to charges of unfairness, favouritism, personalisation, and so on. The processes for selecting who acts as REF assessors are also far from wholly transparent, and could be argued in some cases to mirror wider politics, hierarchies and power bases within individual disciplines. Nonetheless, the RAE/REF did enable some academics to flourish to some extent relatively independent of the wider reputation and prestige of their institutions, or the need to curry favour with some senior figures there. Some continental European academics who have come to work in the UK have commented on how the REF gave them opportunities which would not have been possible in their home countries, where advancement was largely dependent upon who one’s PhD supervisor was, and where various scholars stop researching after receiving a permanent job. Furthermore, while the universities of the “Russell Group” (a group set up in 1994 to represent research-intensive universities) still receive the bulk of the available funding, a range of other institutions, including a few post-92 universities, are nonetheless able to access some reasonable research funding through the process. 

From the first RAE onwards, it was inevitable that institutions would do what was required in order to “game” the system to produce results most favourable to them. Major concerns have included such phenomena as “REF jobs”, those temporary positions created for a very short period (sometimes just a few months) simply in order that the institutions can submit the academic’s work to the REF, the possibility of institutions profiting from academic’s work through the REF even after making them redundant (a new type of concern for REF 2021), and a near-exclusive focus on the REF that pushes other research considerations to a distant second place. Various reforms have attempted to respond to these and other concerns and limit the potential for gaming, but it is unrealistic to imagine it could ever be avoided entirely.

But now there are other wider proposed reforms, to affect the next exercise taking place in 2028, which threaten to undermine the very concept of research “excellence” in favour of other ideological considerations. The new London Universities’ Council for Academic Freedom (LUCAF) has undertaken research into these and submitted recommendations. Amongst the initial decisions are a significant shift in weighting, so that a new category of “People and culture” (replacing environment) will be weighted at 25 per cent, up from 15 per cent in 2021, “Engagement and Impact” (replacing impact) will be weighted at 25 per cent, and ‘Contribution to knowledge and understanding’ (replacing outputs) at 50 per cent. Furthermore, the requirement that underpinning research for impact be ranked as minimum 2* is proposed to be dropped. 

In preparation for these proposed changes, UK Research and Innovation, who administer the REF commissioned a report entitled “Harnessing the Metric Tide: indicators, infrastructures & priorities for UK responsible research assessment”, by Stephen Curry, Elizabeth Gad and James Wilsdon, which was published in December 2022. This follows an earlier report, “The metric tide: review of metrics in research assessment”, published in July 2015. This earlier report recommended that where institutions were making use of league tables and ranking measures “alternative indicators that support equality and diversity should be identified and included”, and urged “due attention to the equality and diversity implications of research assessment choices”. Some recommendations, for example factoring in childbearing and caring responsibilities that affect many female researchers, and considering how other inequalities come about through an observable greater tendency on the part of male scholars to self-cite, were meaningful and welcome. But others were much more ideological and open to serious challenge. The report called upon institutions to “engage with external experts such as those at the Equality Challenge Unit” (now part of the organisation Advance HE) which established the Athena SWAN (Scientific Women’s Academic Network), which has since 2020 shifted its focus from categories relating to biological sex to those of gender identity, and effectively recommended that universities eliminate single-sex toilets, changing facilities and the like. As a result, institutions stand to benefit from outsourcing these issues to such external bodies (also to the Stonewall Workplace Equality Index, who make similar ideological recommendations), adhering to their highly contestable wishes, and marginalising or allowing personalised attacks on “gender-critical” academics (as has occurred with philosopher Kathleen Stock, criminologist Jo Phoenix and historian Selina Todd), who continue to assert the fundamental difference between biological sex and gender and insist upon maintaining the collection of data on the former. 

The 2022 report went further in this direction, reflecting the wider priorities of its authors. Curry, Assistant Provost for Equality, Diversity and Inclusion at Imperial London, wrote in 2020 to Liz Truss, then Minister for Women and Equalities, in opposition to restrictions on public provision of gender neutral toilets, and the government dropping wider reforms to the Gender Recognition Act to include self-identification, and had supported the removal of the name and bust of nineteenth-century naturalist Thomas Henry Huxley from the mathematics and computing department building at Imperial on grounds of his racist views, despite describing him as “one of my scientific heroes”.  Gadd had written extensively (for example here) about the use of research metrics and the relationship to equality, diversity and inclusivity (EDI). Wilsdon was co-author of an article welcoming a new network for EDI in science. Referencing the 2015 report, the authors sought to further develop EDI indicators “as a counterweight” to “problematic impacts of assessment”. They also thought existing REF concepts of “excellence” to be “too narrow”, and cited approvingly work on “the biases inherent in the concept of excellence, which sustains ‘epistemic injustice’, by feeding through to unequal acknowledgements of the contributions of women or indeed anyone who isn’t a white, able-bodied, heterosexual man.” The term “epistemic injustice” is well-known to those familiar with the field known as “Critical Social Justice” (CSJ), an outgrowth of postmodern thought which decentres many concepts of knowledge, truth, merit, rendering these as expressions of power dynamics, and can attack modern science as “racist”, “patriarchal”, or “colonial”. There is evidence of censorship and suppression of scientific research which contravenes the narrative of CSJ, and of lack of political diversity creating harmful ideological bias in social science research. 

To see such a concept employed alarmed LUCAF researchers, concerned that these recommendations might lead to the politicisation of science, creating new incentives for academics and universities to endorse without critique what are highly contested renditions of EDI. Furthermore, a requirement to adopt, possibly in a performative manner, particular EDI positions was thought to harm the possibility of academic freedom and viewpoint diversity, at odds with the requirements of the 2010 Equality Act, and displace academic rigour. 

The new “people, culture and environment” element of the new proposals makes clear reference to the value of collecting EDI data, but no reference to academic freedom; nor does such a term appear in the 2022 report. Research culture and environments are not, it appears, to be assessed on the basis of their openness to challenging orthodoxies where appropriate, promoting open, robust but civil debate, or harnessing regular critical engagement such as is a vital part of academic self-regulation. The initial decisions report indicated a “significant minority” of respondents desiring “a positive research culture” as a “core purpose of the REF”, but this was based upon a small sample, and almost as many (21 per cent) thought that research process should be weighted less heavily or not assessed as thought it should be heavily weighted (26 per cent). As such LUCAF, have recommended the inclusion of active promotion of academic freedom as part of the new “people, culture and environment” assessment criteria, and that its weighting should be lowered. 

Already the authors of the 2022 report have published an article as a strong pushback against criticisms, indicating that they think one critique published by two founding members of LUCAF, calling for REF to promote academic freedom, amount to just an attempt “to drag the REF into the culture wars”, and otherwise making suggestions that some “hope that surreptitious lobbying or scaremongering will throttle the proposals”.

These types of sentiments should be viewed in the context of wider developments in academic life and institutions, some of which have come into sharp focus in recent months, many of which stand in stark opposition to established ideals of universities and the generation of knowledge. In 2020 a significant minority of universities had adopted “decolonising the curriculum” as an official policy . This by no means necessarily refers to the introduction of proper historical teaching of global colonialism into the curriculum, but the term “decolonisation” has explicitly been argued by some academics to entail commitment to and support for violent movements such as Hamas, whose actions have been argued by historian Simon Sebag Montefiore to have been legitimised to some by a narrative of Israeli “settler-colonialism”. Others have analysed how, even without this, decolonisation can entail overtly racialised approaches to learning, coding knowledge as “white”, “black” or otherwise and judging it accordingly, described by Doug Stokes, Professor of International Security at the University of Exeter, as “setting up evil Western colonisers against virtuous non-Western victims”. In various areas of maths and sciences, decolonisation objectives can require academics to prioritise the identity of the authors of such work over the scientific cogency of the results. 

Other ideas which have taken hold in academia include “Critical Race Theory”, which in line with other postmodern developments maintains, according to political scientist Yascha Mounk, that “universal ideals and neutral laws are just fig leaves that try to perpetuate racial domination and need to be rejected root and branch”. Crude renditions of ideas originally developed by scholars such as Derrick Bell and Kimberlé Crenshaw present a world in which racial antagonism and hostility are inevitable and unresolvable, and the only legitimate response is the enactment of measures which treat all perspectives, knowledge, research from some groups as by definition racist and therefore automatically suspect. This is part of what Mounk calls the “identity trap”. 

Furthermore, there has over an extended period been a blurring of the distinction between scholarship and activism

Furthermore, there has over an extended period been a blurring of the distinction between scholarship and activism. The two activities are far from incompatible, but are certainly distinct, and should not be confused. No scholar investigates a subject free from assumptions and biases, but a range of strategies have been developed to combat these, involving rigorous self-critique and submission to critique from others, in order to interrogate reasoning, working, theories and arguments for weaknesses and holes. It is then incumbent that scholars modify their work accordingly. An activist by definition is advocating for a particular point of view, and as such is often much less open to question of fundamental assumptions and biases relating to this. It is far from unknown for some ideologically-motivated scholars to cherry pick data, hide banal or unsubstantiated claims behind empty jargon, use loaded language to silence criticism, or engage in ad hominem attacks or present themselves as a victim in the face of critics. Australian law professor Katy Barnett has linked this to increased polarisation of civil society, with mobbing behaviour enacted on social media by groups on the political left and right, and argues that when enacted in academia, it risks turning the academy into an echo chamber. There is a real risk that the loss of the 2* requirement for underpinning research towards impact will allow some activist work of dubious scholarly merits to fill this role.

These developments were not brought about by the REF, but there is a very real chance that proposed changes may exacerbate these serious threats to the free pursuit of knowledge. It is incumbent on all those who believe in the latter, a defining aspect of the role of a university, to oppose attempts to sacrifice merit and excellence to ideologically loaded concepts of “culture” and “environment”, and for UKRI to think again about the proposed changes, to avoid further decline of universities into closed circles policed by ideological gatekeepers.