Notes on counter-extremism

Paul Bowman
4 min readOct 8, 2023
Photo by Nikita Pavlov on Unsplash

The following are some notes I originally made in response to an interview by the hosts of Decoding the Gurus podcast with Julia Ebner, a researcher on the far right with the Institute of Strategic Dialogue, and an Oxford U graduate student sharing a supervisor with one of the DtG hosts. As all of the original intended audience are academically engaged with social psychology on one level or another, the first section of what follows may seem a little “inside baseball”. As a disclaimer, I am not myself an academic or professional in social psychology, by any means.

Second and final preliminary note, is that I’m addressing only the theoretical problems of the differences between the counter-extremism approach to researching the far right (FR) compared to that of militant anti-fascists. The political gulfs and mutual hostilities between the establishment “anti-terrorist industrial complex” and militant anti-fascism are already well known to all concerned and are not rehearsed here.

The rather narrow (and somewhat circular) confines of how the counter-extremism (CX) milieu defines “radicalisation” as restricted to ideological tendencies that produce politically-motivated violence as outcomes. In general there are a number of issues with the CX outlook that I think act as restraints or blockages on the development of research approaches to far right (and other) ideological movements. I limit myself to the FR here, because that’s my main field of experience, but obviously the questions generalise to other ideological quarters. I find the limitations of the CX framework (inevitably I’m caricaturing here, for brevity) in 3 areas: foundational, diagnostic and remedial.

On the foundational limitations, I follow Michael Billig’s critique against using cognitive psychology as the default starting point for social psychology approaches to ideological movements (cf his essays in “Ideology and Opinions” and other texts from that era). Billig started as Henri Tajfel’s doctoral student and took part in the design and execution of the original minimal group experiments that Social Identity Theory came out of, so he knows whereof he speaks. After his doctorate, Billig moved onto more concrete research of prejudice in interviewing members of the National Front in the mid-70s, so his views are grounded in practical experience of researching the FR, and well worth a look.

The second issue with the CX framework — diagnostics — is really a result of the attempt to have a “non-political” approach to ideology. This can lead to a “symptomatic” approach to reading rather than a semantic or hermeneutic approach to engaging with the actual narrative and actant-mythologising content. The result of trying to short-circuit the interpretative question by referring solely to violent outcomes, makes it very difficult to answer the “pipeline” question that frequently comes up regarding the exact relationship between the far right and the intermediate layers of right-leaning conspiratorial “heterodox” influencers, like Jordan Peterson, Russell Brand, Maajid Nawaz, the Weinsteins, etc. How can you actually determine if this IDW layer that lies between the cultural mainstream and more extreme radical right ideological territory is acting more as a pipeline from the former to the latter, or some kind of ideological “wetland” that soaks up its audience and acts as a buffer between the shifting sands and the devil’s deep blue sea? With a merely “associative” analysis of the ideological themes of the two spheres it’s impossible to settle the question.

A further illustration of the limitations of a “non-political” associative or symptomatic approach to ideology is the persistent appearance of “horseshoe theory” narratives in CX discourse. As I’ve noted elsewhere, there’s a historical irony in this common liberal centrist maxim, given the Nazi origins of the concept. But aside from simple ignorance, the trope reveals a limitation of the “non-political” approach, which aims to pose as ideologically neutral to state and NGO funders hostile to anti-racist or anti-fascist commitments. Both-sides-ing fascism and anti-fascism is only possible on making crudely generalised models that abstract from concrete ideological particulars to such an extent as to become unfit for any practical use in locating FR radicalism.

Finally, the most practical question, in some ways, the remedial one. The anti-fascist (AF) movement has always significantly relied on the role of conversion — that convinced fascists could experience a damascene conversion and start to work for the other side. In the public eye, repentant converts from Ray Hill, to Martin Collins (UK context here) and many others, have played a role beyond more than simple intelligence, but actively a part of sabotaging and breaking up FR formations. In the necessarily less public role of militant anti-fascist activity, anonymous ex-fascists have made reparations by taking an active part in combatting their former confederates. In both cases, conversion is a crucial question of interest to militant AF practice. By contrast, CX doesn’t really have a model for conversion. Its centrist presuppositions promote a purely negative “ex-extremist” picture. The CX goal is simply to get former extremists to “cease and desist” from any and all activism, rather than the possibility of converting anti-social radicalisation into pro-social activism. In practice a number of the people claiming “ex-extremist” status, for the sake of a stipend from the CX industry, are less than honest. Obvious spoofers like Matt Heimbach fool virtually no one. But historically the CX milieu has a much worse track record than the AF one of telling the spoofers apart from the genuine article. Majiid’s career and the Quilliam debacle is a case in point [context: this is a reference to Ebner’s negative experiences with Quilliam and Nawaz related in the original podcast].

The above is simply a brief outline in note form, rather than an essay-length argument or thesis, and is shared purely for the purposes of discussion for people with an interest in the topic.

--

--