Research on Extremism in the U.K. Hobbled by Skewed Research Environment

The skew towards generalism appears problematic in that policy applications for research on general phenomena (for example, radicalization) may be unclear in the absence of research on how those phenomena are manifest ‘on the ground’ (for example, in the recruitment practices of specific organizations). The skew towards studies of rightwing extremism, and away from studies of Islamism (especially in a contemporary UK context), could also perhaps be seen as problematic, given that Islamist extremism proportionally represents a far greater terror threat in the UK (see, for example, the Independent Review of Prevent), although that is an inherently political matter.

This report further suggests the existence of multiple social and ideological pressures which could potentially distort the research field:

·  The skew towards general studies may be partly explained by an academic tendency to devalue ‘descriptive’ studies of specific phenomena: ambitious researchers are incentivized to target high prestige journals which tend to favor general and theoretical research, and it may be that such research is also favored by peer reviewers of grant applications

·  It may also be partly explained by the risks run in researching specific organizations which may respond with legal, physical, and other threats (see above)

·  The skew towards studies of the far right may be partly explained by a tendency to focus on short-term government priorities, with one ‘hot topic’ at a time attracting funding (right-wing extremism having been the most recent ‘hot topic’)

·  There is evidence of possibly justified concern that studying Islamist extremism might lead to a researcher’s being labelled as racist

·  There is also evidence of possibly justified concern that being perceived to be critical of actions supporting ‘progressive’ or left-wing causes might lead to negative professional consequences (‘cancellation’)

·  Given another study’s finding that a significant minority of academics may discriminate against funding applications on ideological grounds (whether from a left- or right-wing perspective), there appears to be a plausible risk of silencing and exclusion through peer review, especially as the typical approach to peer review of grant proposals is one that appears particularly vulnerable to bias

·  Beyond this, it seems likely that fear of ideological discrimination may have a general chilling effect, potentially leading researchers to avoid engaging in projects that they suspect could lead to controversy that might be harmful to their careers

The problem of possible ideological bias in peer-review could be mitigated through adoption of open peer review (increasing reviewer accountability) or through a transparent process of reasoned adjudication between contradictory reviews on the part of an identifiable individual (increasing funder accountability). Awarding small grants through open competition with quotas for particular kinds of extremism might help to ensure that study of and expertise in key areas continues to be supported despite cyclical shifts. Universities and HMPPS could assist by ensuring that their approval processes do not obstruct public interest research on extremism.

A governmental commitment to notifying researchers when their work is cited in briefing documents or makes any other contribution would also help to ensure the continued supply of useful research, especially where this is not funded by the government. Value could be added to government-procured research through engagement of entities from across the research ecosystem in the design, analysis, interpretation, and dissemination stages of research projects commissioned from generic research providers, as well as through requirements for data sharing.

Lastly, and perhaps most urgently, the government and the Solicitors Regulatory Authority (SRA) could publicly acknowledge the threat which SLAPPs present to the sharing of information about extremist groups, their associates, and their supporters. Potential remedies might include legislation, as well as guidance issued and action taken by the SRA.

1. Introduction
1.1 Background
Although there are many fields of enquiry in which ‘pure’ research is necessarily dominant, the study of extremism is an applied social science. Extremism researchers aim to influence and inform policymakers, practitioners, and the public. To take just one area in which research on extremism may serve to inform policy and practice, the National College of Policing advocates for the use of ‘the best available evidence … to inform and challenge policing policies, practices and decisions’, and notes that research can help stakeholders not only to ‘develop a better understanding of an issue — by describing the nature, extent and possible causes of a problem or [by] looking at how a change was implemented’ but also to ‘assess the effect of a[n] … intervention — by testing the impact of a new initiative in a specific context or exploring the possible consequences of a change’ (College of Policing, n.d.). Extremism is also a frequent topic of interest for the mass media, with researchers based in universities and think tanks often contributing interviews and op-eds, or simply providing quotes as subject-matter experts. The research community thus provides a vital component in a society’s attempts to respond to the threat of extremism, however it is to be conceived. As a result, the question of how that community carries out the task of researching extremism — of the ways in which individual researchers choose directions for study, and of the ways in which they are supported or frustrated in their efforts to pursue knowledge in those directions and then communicate their findings to stakeholders and the public — is of more than academic interest. In adapting to structures of incentives and disincentives (opportunities for funding or exposure, for example) and navigating obstacles (difficulties with access to research subjects, for example) researchers jointly construct the evidence base on which policy, practice, and public opinion rely when it comes to this key area. Thus, the institutional conditions for research — the research environment, in this study’s terminology — can be seen to set the parameters within which knowledge of extremism can emerge. At a further remove, what is known about extremism is one of the principal determinants of what can be done about extremism. Understanding the national research environment for the study of extremism should thus be understood as a necessary first step in the development of an extremism research policy designed to produce an evidence base adequate to support a democratic, whole-society approach to extremism.

1.2 Definitions
This report adopts the UK government’s official definition of ‘extremism’ as ‘vocal or active opposition to fundamental British values, including democracy, the rule of law, individual liberty, and mutual respect [for] and tolerance of different faiths and beliefs’ (Secretary of State for the Home Department 2011, 107). In common with UK Research and Innovation (UKRI), this report defines ‘research’ as ‘as any form of disciplined inquiry that aims to contribute to a body of knowledge or theory’ (UKRI 2023).

….

4. Conclusions
The two studies presented in this report jointly suggest that there may be substantial structural issues regarding the study of extremism in the UK. In particular, the author of this report wishes to draw attention to the quantitative finding of an apparent lack of grants awarded via open competition for the study of forms of extremism manifest in the UK since 1980 other than those associated with Incels and the far right. It may be that extremism research stakeholders do not consider this to be a problem. However, if they do, then the qualitative part of this report has been able to suggest several factors which might be acting jointly to introduce distortions into the knowledge base: distortions perhaps reflected in or exacerbated by funding patterns as previously mentioned.

Politicization of extremism research emerged as a key theme in those interviews. Interviewees did not appear to be confident that other researchers in the field would always be even-handed and concerns were repeatedly raised (including by a representative of an organization widely seen as being ‘on the left’) regarding the credibility of individuals carrying out ‘activist’ research on the far right. Several interviewees expressed the opinion that studying Islamism could lead to accusations of racism, while at least one — by no means the only interviewee to suggest that Islamism was over-researched — appeared to take the view that such accusations might not be entirely wrong. Some interviewees took the view that ‘left-wing’ or ‘progressive’ forms of extremism (for example relating to environmentalist demands) were under-researched for political reasons, and even that negative professional consequences might follow from trying to study them, while one expressed alarm at the prospect of official pressure to associate such causes with violent extremism. Recent events taking place at the UK’s leading think tank focused on the far right, i.e. CARR, indicate that the stakes of such ideological conflicts are very real. This report does not take a view on who is right or wrong in relation to these events, but notes that scholarly detachment may be unusually difficult to find in a field centrally concerned with political issues and openly conflicted along political lines.

Given the widespread employment of an approach to peer review that seems especially vulnerable to bias (i.e. single-blind review with decisions based on a numerical average of reviewer scores), the political climate in the field may help to explain why the balance of funded projects appeared to lean so heavily firstly towards general studies of extremism (perhaps seen as politically neutral, and therefore ‘safe’) and secondly towards studies of the far right, which, given the different arguments made in both studies presented here, might be expected to receive less hostile reviews than, say, studies of Islamist extremism. This pattern appeared to be funder-dependent: the dominance of studies of the far-right was less pronounced among projects funded by the CCE (which supported a relatively large number of small projects) and by the ESRC (which supported by far the most projects overall, in some cases with very large grants). However, only the CCE appears to have funded studies of recent (i.e. post-1980) Islamist extremism in the UK. Political tendencies might also explain why the dominance of studies of the far right was reduced among publicly funded projects on extremism in countries outside the UK and in the fairly distant past, as these are at a greater remove from political conflicts in which members of the UK extremism research community are likely to feel invested. However, many other problems have been identified, including various difficulties around access to or collection of data, the low academic prestige accorded to non-theoretical research on highly specific phenomena, and the possibility of legal and physical threats to researchers. All of these may have acted to discourage the study of particular extremist movements or groups in the contemporary UK.

Taking all these factors into consideration, it seems plausible that, where researchers are given the opportunity to apply for funds, they might feel incentivized to keep their proposed projects at the highest possible level of generality that can still satisfy the terms of the call. Furthermore, where some degree of specificity is needed, researchers might plausibly feel that it was wisest to take what would appear to be least controversial approach from the point of view of likely peer reviewers —- that is, to focus on the extreme right. This study has highlighted the very high generality of topics specified in CREST calls compared to those in the only CCE call to date: a level of generality which could be argued to render the former of these two approaches particularly attractive. Again, however, it must be emphasized that there is no objective standpoint from which to judge that the balance of funded projects is or is not as it should be: there is no scientific law which states that research topics must be prioritized according to the threat level that they present, and the question of whether there are ‘too many’ or ‘too few’ studies of the extreme right, the Islamist movement, or any other form of extremism is a political matter, as is the question of whether there are ‘too many’ or ‘too few’ studies of extremism in general (as opposed to, say, road safety).

Interviewees characterized the extremism research field as dependent upon a rapidly shifting funding landscape within which long-term research projects are hard to develop. Several researchers suggested that the current configuration of this landscape has recently favored study of the far right, leading to widespread failure to keep track of recent developments in the Islamist sphere. However, it seems likely that, given such a situation, every form of extremism will sooner or later find itself under-researched. This view was corroborated by the Home Office interviewee, who described extremism research as poorly funded and subject to short-term political imperatives. The intellectual entrepreneurship of academic researchers, who operate by identifying an area that has been insufficiently worked over and then ploughing that furrow for a period spanning many years, may mitigate these problems. However, the institutions which make such entrepreneurship possible, i.e. universities, are not well-equipped to support their staff through the type of fast-turnaround projects which may suddenly become possible once policy interest shifts to a previously neglected area, and may see these as financially unviable when supported through the government’s usual contracting system, although they regularly accept smaller sums when these are structured in the form of research grants rather than fees to be paid on a work-for-contract basis. Think tanks are better placed to take on such contracts, but even their representatives suggested that they would prefer to develop and maintain expertise over a longer term than current government research contracting practices can support: the cost of employing experts over many years may be impossible to sustain when the demand is for fast-turnaround research on a rapidly-changing set of policy priorities. Moreover, the practical difficulties of research procurement ensure that much government-commissioned research will continue to be carried out by generic research providers contracted to give insight on multiple topics but lacking in specialized expertise for any one specific project.

Large public research grants provide give researchers and research organizations with a potential way out of the cycle of short-term government, police, and military funding, but these are extremely time-consuming to apply for and have a low success rate. Moreover, the projects which they support appear to have a tendency towards a very high level of generality, which would seem unlikely to aid the development of a wide pool of expertise focused on a diversity of specific extremist threats. In addition, where such projects buck this apparent trend, there seems to be a current bias towards study of one specific particular extremist threat, i.e. the far right. One interviewee also complained that private donors are often reluctant to be associated with extremism research, which suggests that philanthropy may be unlikely to fill any gaps that exist.

Long-term maintenance of a diverse pool of extremism-related research expertise on which stakeholders may draw appears to depend on university staff and independent researchers voluntarily pursuing interests in a wide range of directions despite often uncertain incentives. Independent researchers will follow their own agendas, especially where financial backing is available, although both they and their backers may need to hide their identity to avoid legal and physical reprisals. As for researchers working within a framework in which promotion and even continued employment may depend both upon publication and upon funding acquisition, extremism research —- especially on specific forms whose study their peers might consider to be ideologically suspect —– may come to seem highly unattractive.

In this context, CREST clearly plays a vital role in providing government funding for extremism research, helpfully structuring this funding in grant form (which removes the need for universities to recover full costs). Under the terms of their arrangement with CREST, government agencies can set priorities for CREST open calls, and also participate in the peer-review process by assessing bids for policy impact, but they do not have to issue contracts to individual teams of researchers, and can delegate responsibility for quality control to CREST’s community of peer reviewers. However, this puts members of that community in the position of gatekeepers, which is important to remember, given the impression which this report has gathered of a highly politicized field which may be distorted by several various forms of bias and self-censorship. In this context, it should be recalled that CREST’s funded project pattern of funded projects was particularly skewed towards research on the extreme right. Whether this arose because applications for grants to support other projects were rejected or because they were not made in the first place, it can be argued that more specifically targeted topics such as those employed in the CCE’s sole call could be used to alter the overall pattern, if that were to be felt desirable.

Issues around accessing data and potential research participants are very important. Experienced researchers will not attempt to carry out projects which they know will be made unfeasible by lack of such access, with the possible result that questions of interest to research stakeholders are not addressed. ‘Naturally occurring’ public online data are important, but may not be representative of anything other than themselves, and would ideally supplement data collected by conventional means, rather than being used as a substitute. Obstacles to data collection highlighted by interviewees include risk-averse ethical reviewers, uncooperative gatekeeper institutions, and the lack of a culture of data sharing, both with regard to research projects commissioned using public funds and with regard to key state institutions such as HMPPS.

Difficulties in publishing research on specific extremist groups — which might help to explain why generalized extremism studies dominate publicly funded projects — should also be acknowledged as serious. Researchers may of course present unpublished or unpublishable research findings when called on for advice. However, this raises several problems. Firstly, this kind of research has neither been peer-reviewed nor subjected to public criticism, which means that its value may be hard to gauge. Secondly, it is not discoverable, which means that individuals, organizations and sub-organizational units which might benefit from it may be unaware of its existence. Thirdly, there is little reputational benefit to individuals and institutions carrying out such research, diminishing incentives greatly. When these issues are compounded by legal challenges and threats to researcher safety, researchers and their employers (both in the university sector and in the faster-paced think tank world) may simply avoid this kind of work as risky and unrewarded. As for independent researchers, venturing into such treacherous waters without institutional protection —- as some evidently do —- must clearly require exceptional courage. It is thus to be regretted that Parliament is at present only debating legislation on SLAPPs which relate to economic crime (HM Government 2023). However, it is noted that the SRA has indicated that it is potentially able to take disciplinary action against solicitors who are judged to have issued SLAPPs even where they have not broken the law (SRA 2022).

In summary, this study has found several apparent issues which could be argued to impede or distort extremism studies in the UK. The field seems deeply ideologically conflicted, with the suggestion that expertise is muffled or even excluded due to an alleged requirement in some quarters to conform to emergent political orthodoxy; indeed, the possibility that researchers might be ‘cancelled’ for stepping out of line appears real, and it is emphasized that anticipation of discrimination may have a chilling effect extending beyond any specific implicit or explicit censure. There also seem to be major problems with funding, data collection, and even the dissemination of findings; in some cases, extremists have used legal or physical threats, as well as online harassment, to silence researchers. Considering these issues, it appears likely that key aspects of extremism may be under-represented in the knowledge base. All considered, it would seem unwise to assume either that a sufficient understanding of the current threat environment may be obtained from reading the published literature or that a range of expertise matching the full range of extremist threats will always exist in the UK.

5. Recommendations
Some of the problems identified in this report could be addressed through small changes to stakeholder practices. For example, government agencies could probably do more to leverage the expertise of external researchers if it were recognized, firstly that they often see access to data and stakeholders as a form of compensation, and secondly that their employers may not always expect to recoup full costs of activities which help them to fulfil their institutional mission. For example, universities generally charge full economic costs on research services which their staff provide for clients on a work-for-contract basis, but may charge considerably less where the funder is understood instead to be providing a grant which will support university staff in carrying out public-interest research that will lead to publication in a scholarly venue.

Substantial value could also be added to conventionally-procured research through promotion of greater cooperation between entities across the research ecosystem. For example, engagement of university-based researchers as consultants in assisting in the design of research projects or in analyzing the resulting datasets, or engagement of think tanks in the identification of policy implications and the communication of findings, could bring an infusion of subject-specific expertise to work carried out by generic research providers without compromising on the flexibility which the latter can bring.

A simple commitment to informing individual researchers or research teams that their work has been useful (for example, when it is cited in briefing documents) would also help greatly, both because it gives researchers feedback by indicating that they have been working along beneficial lines, and because it gives them a documented way of claiming credit. This would in turn help those researchers to make a case for continued support from their employers or funders — which is also in stakeholder interests.

However, to solve some of the underlying problems discussed in this report —- or at least, to mitigate their potentially distorting impact on the knowledge base —- it is suggested that an extremism research policy may need to be developed and communicated, identifying research goals which are meaningful to researchers and stakeholders, acknowledging the likely difficulties to be faced in achieving these goals, and setting out concrete measures which might plausibly help these difficulties to be overcome.

Such a policy could, for example, require that, where government agencies issue data collection and analysis contracts, they include clauses requiring both raw and processed data to be deposited in a curated repository, with access by invitation or application where full openness is considered impossible or inadvisable. To give a further example, to be eligible for public funding, British universities could reasonably be expected to demonstrate that their ethical review processes were not neglecting public interest in favor of managing their reputations or avoiding exposure to legal threats. A defined pathway could also be created for accessing data and potential research participants via state agencies, including HMPPS, as it would be of obvious direct benefit to public-interest research.

As an explicitly political act, the setting of quotas for funded projects might also have the advantage of foregrounding the (probably inescapable) politicization of the field itself, and thereby stimulating debate about which forms of extremism research are actually required. A move away from single-blind peer review on the part of the major research funders might also serve to increase confidence in the fairness of the peer review process, and, while it seems unlikely that UKRI will would shift to open peer review for the sake of a field of research field accounting for only a small fraction of a percentage point of its distributed funds, it might perhaps be considered by a specialist funder such as CREST. Alternatively, simply giving an identified individual responsibility for adjudicating between contradictory reviews might reproduce some of the advantages of open peer review, as compared to the ‘computer says no’ approach to decision-making wherein a numerical average is taken across multiple anonymous reviews, potentially enabling a single biased reviewer to torpedo a proposal. Additional measures could be employed as desired: for example, empirical studies of specific extremist groups could receive extra weighting in the approvals process, if stakeholders indeed consider such studies to be valuable, thus ensuring that such studies continued to be supported and incentivized in the face of what this report suggests may be contrary incentives towards generalism.

Joined-up thinking will be essential, however. There is little point in funding empirical studies of specific extremist groups if data on those groups is impossible to collect, or if the groups themselves put researchers and their families at such risk of physical harm as to make it impossible for them to continue in their work, or use spurious legal threats to block the publication of whatever research outputs are eventually produced. With regard to physical threats, one might hope to see employers providing training for those facing a credible threat, [perhaps along with financial support for installing security systems where appropriate, but the misuse of legal process appears to be a more widespread problem. It is therefore welcome that the SRA has issued solicitors with a statement emphasizing that they are ‘expect[ed] … to take reasonable steps to satisfy [themselves] that a claim is properly arguable before putting it forward’ (SRA 2022, n.p.), but a voluntary approach to abandoning SLAPPs would seem unlikely to succeed, given the suggestion (both in interviewee testimony and in parliamentary debate) that certain legal firms have come to specialize in such actions. The SRA may wish to consider its potential role both as a stakeholder and as an instrument of extremism research policy, proactively seeking out or publicly asking to be notified of cases where SLAPPs have been used to suppress the dissemination of information about extremists. However, it seems likely that a legislative solution will be required, just as it has been found to be required in relation to SLAPPs issued in relation to allegations of economic crime. For this reason, a consultation on the use of legal and other silencing tactics by extremist groups and their associates and supporters would seem particularly urgent.