Enactors of the State: The Everyday Coproduction of Security in the Prevention of Radicalisation

Abstract

Since 2001, studies of (counter)terrorism and (counter)radicalisation have burgeoned. However, at times, these literatures have reduced the agency of ordinary citizens, imagining them as ‘actors of the state’ and ‘petty sovereigns’. By integrating vernacular security approaches with allied research in Education, we develop a novel ontological conceptualisation of ordinary citizens as ‘enactors of the state’, engaged in the everyday coproduction of security. The article presents the findings of a survey, semi-structured interviews, focus groups, and participant observations, conducted in Manchester’s Further Education sector, following the 2017 Arena attack. We capture the security enactments of 95 elites, professionals, teachers, and students. Our data reveal the complex interactions, ensembles, and assemblages at the heart of security’s everyday coproduction. In contrast to existing approaches, we show how this creative dynamic operates through citizens’ variegated imbrications with state policy, complex relationalities, and subtle nuances in everyday enactment.

Introduction

Two decades of research on the construction of security, with a focus on terrorist threat, have accompanied the era of the global War on Terror and the United Kingdom’s introduction of counter-terrorism legislation in response to the events of 9/11 (e.g. Croft, 2006; Heath-Kelly, 2013; Jackson, 2005). The focus of much of this important body of work has been state-led constructions of security, appropriate to an era in which political elites articulated a new, omnipresent threat, requiring unprecedented policies and heightened vigilance to root out terrorism (e.g. Holland, 2012; Jarvis, 2009). Broadly in the past decade, however, two important shifts have taken place. The first is theoretical, with growing interest in International Relations on everyday security and the international politics of ordinary life (e.g. Crawford and Hutchinson, 2016; Nyman, 2021). This burgeoning interest in the banal (terroristic) experiences (Katz, 2013) and vernacular security (e.g. Croft and Vaughan-Williams, 2017; Jarvis, 2019; Jarvis and Lister, 2012; Vaughan-Williams and Stevens, 2016) of ordinary citizens has accompanied a second, empirical, shift. From 2015 (to 2023), the United Kingdom altered its domestic approach to counterterrorism, with counter-radicalisation efforts brought under the rubric of a ‘Prevent Duty’ which reimagines the identification of terrorist threat as a broad societal safeguarding responsibility (e.g. Heath-Kelly, 2017). ¹ In this article, we bring these concomitant developments into dialogue, exploring the everyday coproduction of security by ordinary citizens ² involved with and affected by the United Kingdom’s Prevent Duty.

Despite taking ordinary citizens’ lived experiences seriously, research on counter radicalisation, including critical analyses of Prevent, has tended to downplay or overlook the agency of non-elite subjects. In studies of Prevent, education professionals, for example, have been noted for the important role they play as actors of the state, implementing government policy on the ground (e.g. Younis, 2021: 50). As agents for policy’s implementation in everyday spaces, ordinary citizens are conceptualised as petty sovereigns: an ideal-type figure, structured by the logic of governmentality, faithfully reproducing policy preferences in the micro-level capillaries of state power. Faure Walker (2019), drawing on Fairclough (2000), has theorised this hollowing out of agency through a process of nominalisation, with the ordinary citizens implementing Prevent – Judith Butler’s ‘petty sovereigns’ – reduced to vectors of security policy and, simply, actors of the state.

Where critical research has afforded agency to ordinary citizens on Prevent’s frontlines it is often reduced to oppositionality – a reflexive critical resistance (e.g. Dudenhoefer, 2018; Moffat and Gerard, 2020; Sjøen and Jore, 2019). A similar, parallel dynamic is also evident in broader studies of everyday and vernacular security, where a focus on the disruption of state policy can partially displace the possibility of more innovative or hybrid creativities (e.g. Vaughan-Williams and Stevens, 2016). In other instances, theoretical and metaphorical innovations – for example, Heath-Kelly’s (2017) focus on the inductive method of National Health Service (NHS) professionals – can homogenise behaviour, veiling the nuances of creative interactions in specific contexts. Otherwise powerful critical innovations can have the effect of reducing those intended to be safeguarded to the status of input data, ‘perpetuating the exclusion of non-elite meanings and experiences of (in)security’ (Croft and Vaughan-Williams, 2017: 21). Where richer depictions of ordinary citizens’ agency are found, this research remains principally empirical in focus, demonstrating how everyday enactments go beyond the reproduction of, or resistance to, elite policy (e.g. Busher et al., 2019).

Building on these empirical findings and overcoming the elite focus of much critical terrorism and counter-radicalisation research, this article makes contributions to two allied literatures. First, contributing to critical studies of security – with a focus on the everyday, vernacular security literatures – we develop a theorisation of ordinary citizens as coproducers of security. This coproduction works in several ways, including in dialogue with the state and through interaction with other groups of ordinary citizens. We heuristically conceptualise this productive state-public-public relationship as trialogical. Second, contributing to (critical) studies of terrorism and counter radicalisation, we develop an analysis of contemporary counter-radicalisation practices at the level of the everyday which captures the rich nuances and dynamic interactions of groups of citizens, conceptualised as coproducers of security, rather than passive vectors, mere disruptors, or uncritical opponents of the state. Together, the article reconceptualises citizens not as actors of the state, but as enactors of the state, enabling a new understanding of ordinary citizens as everyday security coproducers.

To develop these twin contributions, the article presents the findings of a 3-year study, comprising a survey, semi-structured interviews, focus groups, and participant observation, designed to uncover the full, everyday life of Prevent, by capturing the interactions and enactments of 95 directly engaged respondents, including policy elites, institutional professionals, Further Education (FE) staff, and FE students. As only the fourth study to date capturing the agency of students/pupils, this funded dataset is highly novel; its revealing of the security interactions between key groups underpins the article’s original theoretical contribution. Moreover, conducted following the Manchester Arena attack, the research compiles a significant, original dataset on the Duty’s implementation. This enables our analysis to move studies of Prevent beyond normative evaluations to instead reveal the complex assemblages, ensembles, and interactions at the heart of security’s everyday coproduction. ³

The article proceeds, broadly, in four principal parts. First, we outline the research puzzle and conceptual gap, through a review of Prevent and its associated critical literatures. Second, we develop our theoretical contribution, reconceptualising ordinary citizens as enactors of the state and security coproducers. Third, we outline our four-part methodology. Fourth, we present the findings of our analysis with a thematic focus on security’s trialogical coproduction. We conclude with reflections on the article’s limitations and implications for critical security and counter-radicalisation literatures as well as future studies of UK counter-terrorism policy, setting out an emerging research agenda based on vernacular methods.

Preventing Radicalisation and Its Critique – What Role for Ordinary Citizens’ Agency?

The United Kingdom’s counter-terrorism strategy (CONTEST) was introduced in 2003, in the context of the early War on Terror, dominated by the fallout of the events of the 11th September 2001, and the wars in Afghanistan and Iraq (e.g. Qurashi, 2018). Until 2005 and the London terrorist attacks of the 7th July that year, the Prevent stream of CONTEST was relatively marginalised. Following 7/7, efforts to win the War on Terror through a battle for ‘hearts and minds’ increased (Qurashi, 2018). In the late 2000s, under the Labour government, Prevent was administered through the Department for Communities and Local Government, with a focus on community engagement (Heath-Kelly, 2017: 298; see also Thomas, 2017). After 2010, the coalition government’s review of Prevent, in the context of impending austerity cuts, led to its centralisation under the Office for Security and Counterterrorism, decreasing the highly targeted, localised, and problematic approach of its first era. While the second era would have well-documented problems, what would emerge in the years which followed was a UK counter-radicalisation policy that was effectively nationalised in implementation, with radicalisation inserted into pre-existing national structures for safeguarding, under the new rubric of the ‘Prevent Duty’ (Thomas, 2017).

Introduced in 2015 and hailed as the United Kingdom’s flagship counter-terrorism policy, the Prevent Duty requires public sector workers – teachers, NHS staff, social workers, housing officials, among others – to show ‘due regard’ for potential vulnerabilities to radicalisation, among those with whom they come into contact (HM Government, 2015). Through what Thomas (2017) coins ‘responsibilisation’, the duty extended the responsibility for spotting the signs of radicalisation from counter-terrorism agencies to public sector workers, suggesting that those closest to potential extremists/terrorists could help orchestrate early intervention. This pre-crime approach was facilitated by a shift in language: terrorist and extremist threat would henceforth be understood as a form of harm that could be safeguarded against (James, 2020). For the education system, this was simply to be read as an extension of existing safeguarding practices (James, 2020) and not, as critical readings had suggested, an intrusion on the safe space of education (Dudenhoefer, 2018; Ramsay, 2017). The introduction of the Prevent Duty therefore marked a dramatic transformation of the way in which terrorist and extremist threat was understood and managed. Counter-radicalisation efforts were now an extraordinary but everyday activity that extended across the whole of British society.

A ‘conveyor belt’ approach, conceptualising radicalisation as a teleological process in lieu of timely intervention (Powell, 2016), marks only one of the ways in which Prevent has been criticised as a problematic and dangerous policy by academics and NGOs (e.g. Ecclestone, 2017; Rights Watch UK, 2016; Stanley and Guru, 2015). Among others, the Duty has faced staunch criticism for (1) leaving those responsibilised for implementation uncertain of the repercussions should they fail to spot key indicators (Revell, 2019), (2) silencing classrooms by creating atmospheres of fear and an intolerance of difference (Lundie, 2019; Sjøen and Jore, 2019), (3) securitising vulnerability and criminalising those referred (Durodie, 2016; Heath-Kelly, 2016; Wolton, 2017), and (4) promoting ethno-nationalism through the requirement to integrate fundamental British Values – as antithetical to extremism – in the curriculum (James, 2022).

As Jerome et al (2020: 159) reflect on 20 years of Prevent research, they identify two important trends: first, the normalisation of banal counter-radicalisation efforts and second, the conceptualisation of those responsibilised as either unproblematically reproducing, or wholly opposing the Duty. In the words of Younis and Jadhav (2019: 404), ‘the “good” position is to accept the Prevent Duty, and the “bad” position is to reject it’. Where resistance is encountered, this can be limited by the silencing impact of fear (Younis and Jadhav, 2019). As Younis and Jadhav (2019) note, there is a powerful structural impediment to resistance; with counter radicalisation reimagined as an issue of safeguarding, ‘dissent connotes an additional moral dimension of being lax or supportive of abuse’. Such fears are even evident among those delivering Prevent training, with justifications offered for delivering problematic guidance that absolve the trainer of responsibility through the deliberate reduction of their agency under the label of being a messenger, not the creator of the policy (Younis and Jadhav, 2019). Often, even when officials articulate reluctance to act, they are ultimately reported to still act, carrying out policies with which they do not wholeheartedly agree:

When a trainer told John ‘he’s just here to do the training’, the message was clear: the trainer is not responsible for PREVENT, and so the critique of training is futile. Above all, the greatest worry for participants arose when trainers themselves admitted PREVENT was racist- but delivered the training anyway. The reality there are trainers who reluctantly actualise a policy they acknowledge to be racist raises significant ethical concerns within the NHS. Butler (2004: 56) relates to such trainers as ‘petty sovereigns’; bureaucratized figures mobilized to enact the aims of the institution, without any real power to think or act for themselves. [. . .] The occurrence of reluctant trainers/petty sovereigns raises significant ethical concerns over how perceived unethical policies may be ‘banalized’ within bureaucratic systems in which staff are tasked to ‘play’ their parts (Younis and Jadhav, 2019).

This finding from Younis and Jadhav (2019) is important and reflective of broader studies of Prevent, as well as being in keeping with earlier critical studies of terrorism and security after 9/11. Indeed, we argue that analyses of Prevent have tended to continue the research trends evident in critical scholarship during the early War on Terror, holding the state to account for processes and policies that can do harm in the name of counterterrorism. This effort has, quite understandably, focused on the reach of the state into the lives of citizens and (its internal) Others (Croft, 2012). At the simplest level, this top-down, structural focus has minimised the scope for agency on the part of those affected, or implicated, by state policy.

Teachers, Pupils, and Enactment

We structure this section around three needs for International Relations research, which we outline here and develop below. First, it is vital to conceptualise a richer understanding of agency on the part of educators and policy officials, who offer far more than complicity or opposition. Second, datasets exclusively focused on the responsibilised, in lieu of information on those to be safeguarded, veil the vital interactions at the heart of the Prevent Duty’s enactment as safeguarding mechanism. Third, it is necessary to define ‘enact’ in ways that the critical literature on Prevent has thus far failed to do despite an apparent ubiquity of use, since current uses adopt a variety of synonyms (implement, practice, etc) devoid of productivity and interactivity. After addressing these three needs, we thus set out to develop a novel ontological conceptualisation of ordinary citizens’ agency, defining enact in a way that moves the discipline beyond a focus on petty sovereigns and towards security’s complex everyday coproduction.

The Responsibilised: More Than Petty Sovereigns and Actors of the State

Critical studies of Prevent have set out to defend suspect communities (Breen-Smyth, 2014) with the good intention of limiting and resisting the damaging interventions of the state. Correspondingly, there has been a tendency to portray those involved in Prevent’s implementation as the petty sovereigns and eyes of the state, engaged in enforcement and surveillance on the state’s behalf, even when scepticism is voiced. Ordinary citizens become actors of the state – petty sovereigns ‘mobilised by aims and tactics of power they do not inaugurate or fully control’ (Butler, 2004: 56). Consider, for example, that due to the Prevent Duty:

If a racialised Muslim child appears reclusive, this is not just a natural reaction to distress to be addressed with pastoral support by responsible adults (parents and teachers). Nor is it a behavioural response in need of empowering interventions by psychologists. Now, it is distinctly a risk factor for potential pre-criminality vis-à-vis the nation-state, which frames the responsibility of adults as well as psychologists, who now treat the child’s suffering on this pretence (Younis, 2021: 56).

Such critique is vital and powerful, but within it, studies have tended to the reduction or wholesale evisceration of volition and, with it, the nuances and productivity of ordinary citizens’ agency. Within such a rendering, teachers, education staff, and Designated Safeguarding Leads (DSLs) are reduced to ‘petty sovereigns, unknowing to a degree, about what work they do but performing their acts unilaterally and with enormous consequence’ (Butler, 2004: 65). It is easy to imagine how such a tendency developed in literatures concerned to defend suspect communities from state overreach in the context of a dangerous and damaging War on Terror.

Educationalists working in a non-Prevent context, however, have developed more nuanced takes, suited to the complex interactions of the classroom: ‘policy, even when it is centrally mandated, is translated, adjusted and worked on differently by diverse sets of policy actors’ (Maguire et al., 2010). Policy enactment is characterised by ‘variability and distinctiveness’, across ‘different levels of practice’, including ‘within and between’ institutions and individuals, even when circumstances appear similar (Maguire et al., 2010). Such nuance is also true for Prevent, especially following the reforms of 2015:

As Prevent becomes something banal, direct opposition might diminish, but so too might the accompanying sense of intimidation, anxiety and insecurity among professionals that is likely to distort their professional judgement and foster discriminatory practices. The banalisation of Prevent might also result in education and childcare professionals feeling increasingly emboldened to develop their own take on the Duty and to mould it around their own existing professional and institutional cultures and ethos, in the context of their wider relationships with families and communities. As this happens in early years provision, schools and colleges across the country, it is possible that grassroots policy enactment by education and childcare professionals could not only significantly reconfigure what the Prevent Duty looks like in educational settings, but might also give rise to important ‘bottom-up’ policy innovations that have wider implications for how societies seek to respond to issues such as polarisation, terrorism and political violence (Jerome et al., 2020: 159).

Our analysis, developed in the second half of the article, shows this to be the case.

The Safeguarded: Security’s Coproducers and Their Erasure from Data and Analysis

IR’s empirical focus has tended to remain on the supposed petty sovereigns of Prevent – the responsibilised – rather than those perceived to be at risk of extremism and thus in need of safeguarding. In part, this is easily explained by the methodological and ethical difficulty of accessing pupils and patients (see online Supplemental Appendix). Only four studies to date have attempted to capture the data of students aged under 18 (e.g., Beighton and Revell, 2018; Higton et al., 2018; Moffat and Gerard, 2020). Most, however, focus their data collection at the level of education and health practitioners, or policy leads, rendering those safeguarded invisible, homogenised and lacking volition.

Faure Walker (2019) has traced how Prevent’s logics de-agentify those at risk of being seduced by violent extremism through a process of nominalisation as extremism becomes a thing rather than a process. We argue that a similar logic is evident in studies on the homogenising impact of the Prevent Duty’s national rollout from 2015. Heath-Kelly and Strausz (2019) have noted continuity in Prevent’s pre- and post-2015 eras, with staff ‘on the whole, accepting of their new responsibilities under the Prevent Duty and comfortable with the training provided’. The key move facilitating a minimalisation of responsibilised citizens’ agency and their reduction to petty sovereigns – actors of the state – hinges on the Duty’s new (nationalised) geography. Reconceptualised as a societally homogenised process, Prevent is reimagined, through the development of new, powerful analogies. These analogies – inductive method, algorithm and autoimmune response (Heath-Kelly, 2017; Heath-Kelly and Strausz, 2019) – reduce the agency of ordinary, responsibilised citizens, as well as those citizens to be safeguarded.

Within the confines of the first (inductive) metaphor, ordinary citizens are there to be observed and read, as data, by responsibilised citizens – now positivist researchers – who passively and accurately report that data up the chain of command for processing and intervention where necessary. The second (algorithmic) metaphor performs a similarly structuralising function, removing contingency, volition, creativity, and dialogue from the everyday security politics of counter-radicalisation interactions. In such an imagining, safeguarded and safeguarder are simply input and vector in a top-down state process of data management. The third metaphor (autoimmunity) plays up the biopolitical and epidemiological imagining of the UK as body politic, infected with the threat of radical extremist violence. Again, the metaphor’s impact is to reduce citizens’ agency, with the country now biologically pre-programmed to identify threat and respond appropriately towards its eradication. This evolved homeostasis occurs unconsciously; myopic response replaces agency and interaction. In such an imagining, security politics resides at the state level, rather than in the micro-interactions of everyday life. The ‘act’ of ‘enactment’ is reduced to the automated, predictable and uniform, at the expense of contingency, creativity and heterogeneity.

Enactment – Ubiquitous, Slippery and Imprecise

Enactment is a vital term for counter-radicalisation research. Busher and Jerome’s (2020) The Prevent duty in education: Impact, enactment and implications contains 180 mentions of ‘enact’, including in the title and four chapter titles. It is remarkable that the term’s use in critical studies of Prevent is seemingly ubiquitous despite its definitional vagary, imprecise application and inconsistent use. This is even the case where good empirical work is developed. For example, as da Silva et al. (2020) argue, ‘the enactment of the Prevent Duty in primary schools is dynamic and influenced by the wider socio-political context’. We agree and suggest that there are occasions where the critical literature on Prevent has begun to capture this dynamic empirically but is limited in unpacking its full theoretical implications.

When critical Prevent literatures reflect upon ‘enactment’, it is most commonly drawn from Ball et al.’s (2012) research in education: ‘policy is not simply implemented, but rather it is interpreted by different actors, in different contexts and potentially as part of different policy ensembles’ (Busher and Jerome, 2020: 6). Ball et al.’s (2012) emphasis is on ‘creativity’, ‘innovation’ and ‘transformation’, in moments of opportunity and through ‘processes of interpretation, translation and reconstruction’, as they develop their ‘own take’ on policy (Busher and Jerome, 2020: 6). Busher and Jerome (2020) highlight the situatedness and agency of practitioners as they may adopt or adapt policy, sometimes feeling emboldened to create ‘new forms’ of knowledge and behaviour. They also note that ‘relationships’ to policies and others is key.

Yet, while excellent empirical findings are set out, two modest paragraphs on conceptualisation at a theoretical level feel dwarfed by the term’s importance for the research to follow. In places, the term is juxtaposed to ‘mediate’, at others it is ‘knowledge’ (or interpretation) that is placed in contradistinction, suggesting that enactment can slip back from Ball et al.’s (2012) richer framing to be reduced to replication, implementation or merely practice. Elwick et al. (2020) acknowledge that a ‘complete picture of policy enactment’ requires study of how the ‘policy is experienced by the young people’ it focuses on, but, again, the passivity of the verb ‘experience’ is stark in its theoretical and de-agentifying implications. Two pages later, it is clear that ‘teachers enact Prevent’. While the chapter is admirable for ‘listening to students’, we argue that this empirical insight must be matched by a more empowering ontological conceptualisation of ordinary citizens, which includes those being safeguarded.

In sum, good empirical research exists on enactment (Busher et al., 2017; Lakhani and James, 2021). It requires, however: (1) a fuller dataset, capturing student/pupil agency and (2) a richer theorisation of enactment, cognisant of the agency of ordinary citizens and the political productivity of their interaction. As Taylor and Soni (2017) highlight, there remains ‘a space for important future research into the enactment of government Prevent policy into real-world educational settings’.

Enactors of the State: Security’s Everyday Coproduction

One of the peculiar features of the Prevent literature is its location in a relatively constrained space, centred on the principal outlets for critical studies on terrorism. There is a dearth of Prevent research that crosses over into the domain of critical security studies, even though everyday and vernacular security approaches have burgeoned in the same period as the introduction of the Prevent Duty. Here, therefore, we map the terrain of everyday security studies, bringing this research to bear on Prevent and vice versa in order to conceptualise security’s everyday coproductions in the (post-2015) banal interactions of UK counter-radicalisation. This research contributes to and, in doing so, makes a case for the burgeoning field of everyday security.

There is an ‘enormous diversity of research that seeks – in different ways and for different purposes – to reconsider the politics of security away from the strictures of a “top down,” elitist approach’ (Jarvis, 2019). International Relations, in its historiographies, has tended to ignore the long legacy of critical feminist research which takes seriously the international politics of everyday life (e.g. Cohn, 1987; Enloe, 2014 [1990]). Instead, and in a process of invisibilisation, intellectual genealogies of everyday security often look to the formative influence of Bubandt’s (2005) prescient article. For Bubandt (2005), security should not be understood as a purely analytical term, but rather, a socially situated and discursively produced practice (see also Croft and Vaughan-Williams, 2017). Bubandt’s focus was on the politics of feeling secure in specific contexts; a research agenda that has inspired significant work on everyday security, across a variety of domains, including vernacular security approaches.

Vernacular security approaches are:

characterized by: a curiosity toward variability in the work done by ‘security’ discourses, practices, and technologies in diverse contexts [. . .] and a desire to investigate how ‘elite’ security discourses and technologies are understood, responded to, and (re-)shaped in diverse ways (Jarvis, 2019).

For Jarvis (2019: 5), the term’s conceptual emptiness is a strength, enabling an inductive approach that aims to speak with, rather than for, ordinary citizens (Jarvis and Lister, 2012: 158). The approach is influenced by research in ontological and human security, as well as postcolonial research and Peace Studies (Jarvis and Lister, 2012). Perhaps, most obviously, vernacular security studies are marked out from allied everyday security research through its focus on language – the security articulations of ordinary citizens.

What unites everyday security literatures is a focus on the ordinary and banal – the mundane every day. This might be ordinary feelings (ontological security), language (vernacular), or (everyday) practices. This emphasis on the political productivity of ordinariness is central and juxtaposes this canon of work with structural, top-down approaches frequently at the discipline’s core, historically. As Nyman (2021: 4) notes:

scholarship on the everyday life of security shares a desire to take what we assume to be ordinary, non-important, and pre-political, and to demonstrate that it is in fact political; whether it be ordinary spaces, routine practices and habits, or lived experiences.

At their heart, everyday security approaches acknowledge that most people are not political or security elites, but their lives are nonetheless vibrantly political and fully interwoven with security.

This article, therefore, continues the work of key proponents of the field – Jarvis and Lister (2012, 2013, 2016) in examining how participants talk about threat, Vaughan-Williams and Stevens’ (2016) interrogations of engagement with government counter-terrorism policy, Jackson and Hall’s (2016) analysis of the public consumption of elite counter-terrorism discourses, and Stanley and Jackson’s (2016: 230) analysis of everyday spaces as ‘site[s] of practice’ for security. We, too, study ‘socially specific articulations of security that are contextually and historically situated’ (Jarvis and Lister, 2012: 159), as well as focusing on how security is experienced by ordinary people, ‘to ask not only what security means, but, also, what it does when articulated’, precisely because ‘security assemblages are negotiated, accepted and contested in the spaces and practices of everyday life’ (Jarvis and Lister, 2013: 161 emphasis in original; 758; see also Jarvis and Holland, 2014).

In line with Nyman (2021), we argue that this focus should not reduce analysis to seeing banal, everyday moments as existing in silos, unaltered, or unengaging with the realms of macro-politics (drawing on Davies, 2016b). Instead, the micro and macro should be seen as ‘co-constituted’, existing within a ‘horizontal understanding of relations, recognising that concepts like security “only exist as they are enacted in daily practices, relations and entanglements”’ (Nyman, 2021: 5, quoting Guillaume and Huysmans, 2019: 283). Moreover, adding to this, we argue that these instances should be rethought through their interactive and iterative dimensions. Nyman (2021), building on the arguments of Crawford and Hutchinson (2016), urges analysis to crosscut everyday spaces, practices, and affects, interrogating the synergies between and across these spheres.

Our analysis, therefore, moves beyond normative evaluations of Prevent, or the identification of petty sovereigns, or revealing moments of contestation or pure resistance. We show how ordinary citizens are actively embroiled within multifaceted processes of enactment, through which security is coproduced. We answer Jarvis’ (2019: 19) call:

to engage in conversation with those we might view as security’s subjects in order to begin exploring fundamental questions around: what security means, how security is articulated or constructed in specific (research) environments, how security feels, what conditions or relationships create security and insecurity, with which values security is associated (for instance, order, freedom, equality or justice), and other first order questions.

Enacting the State Through Security’s Trialogical Coproduction

Here, we (1) plug the gap in counter-radicalisation research concerning enactment’s definition, (2) set out to capture and theorise ordinary agency at the heart of everyday security practices, and (3) conceptualise security’s coproduction through routinised, trialogical interactions.

On the first and second, in direct contrast with the existing literature’s focus on those responsibilised by the Prevent Duty – teachers, health professionals, safeguarding leads, trainers and so on – as petty sovereigns and actors of the state, we argue that this diverse group is better conceptualised as enactors of the state. This reformulation of enactment better captures the messy nuances and coexisting tensions within a diverse group across distinct contexts. It repudiates the structuralism and linearity of the dominant theoretical conceptualisation evident in current research. Our approach instead suggests a greater role for volition and productivity. Rather than reducing the responsibilised to a vector of state policy – a conduit forged through co-opted acquiescence and resigned to reluctant implementation – we set out to recognise an agency that goes beyond binaries of support or oppose. Enactment is more than implementation or denial; it is pregnant with the messy possibilities of political productivity.

Third, we argue that the role of the safeguarded is missing from the picture in existing analyses of Prevent. Again, we argue that this group’s agency is vital to security’s coproduction. They too are enactors of the state, in complex and iterative interaction with the responsibilised, in what can usefully be thought of as the trialogical coproduction of security. Beyond the dialogue of policy elites and petty sovereigns or policy elites and suspect communities, we argue that theorising the safeguarded – pupils, students, patients and so on – as political actors enable us to capture their proactive role in producing security. We argue that security is coproduced, trialogically, through iterative interactions between state policy (the Prevent Duty), the responsibilised (teachers and education professionals), and the safeguarded (pupils). Teachers and pupils, together and in productive trialogue with policy, coproduce security in everyday contexts, enacting the state.

Methodology

The article analyses data collected from a 3-year study into the implementation of the Prevent Duty, conducted between 2017 and 2019. In addition to extensive participant observations, it draws on 95 directly engaged participant experiences (through a survey, interviews, and focus groups), comprising 42 staff ⁴ and 45 students, from five different FE institutions in Greater Manchester, ⁵ as well as eight regional policy officials. ⁶ This is the first study to have engaged (trialogically) across policy, the responsibilised and the safeguarded. It is one of only four studies to engage directly with the FE sector (Beighton and Revell, 2018; Higton et al., 2018; Moffat and Gerard, 2020), one of four to speak with students below the age of 18 (Habib, 2018; Higton et al., 2018; Jerome and Elwick, 2019), and one of nine if that is extended to Higher Education students (Abbas et al., 2021; Jarvis et al., 2024; Kyriacou et al., 2017; McGlynn and McDaid, 2019; Zempi and Tripli, 2022).

Three factors influenced the context of our data capture. First, legislation was introduced in 2015 to mandate 16–18-year-olds remain in education or training. As a result, the FE sector experienced a large influx of students. The sector, while also catering to those outside of the typical 16–18-year-old bracket, engages predominantly with the adolescent age range thought of as particularly vulnerable. This becoming-of-age period is one which is thought to be both a potential risk factor to radicalisation in the thrill-seeking of almost adulthood and potentially exploited in just emerging from childhood (e.g. Jackson, 2021). Second, the challenge of FE’s expansion coincided with the introduction of the Prevent duty (see above). Past studies have analysed the education sector broadly (Busher et al., 2017; Miah, 2016; Revell and Bryan, 2018), or paid particular attention to schools (Elwick and Jerome, 2019; Habib, 2018; Quartermaine, 2016; Vincent, 2019) or universities (Brown and Saeed, 2015; McGlynn and McDaid, 2019). This article was one of only two to look solely at the FE sector, as it underwent a seismic transformation (see Moffat and Gerard, 2020). It was also the only study to look across both traditional academic and private, vocational academic providers, which had not previously been subject to the same levels of funding, student numbers, or policy scrutiny. Third, the Manchester Arena bombing – causing 23 deaths and over 1000 injuries – occurred 3 months before the fieldwork period. This had a unique impact on data collection processes and the data collected. Given the proximity of the attack – in terms of both its local geography and targeting of young people – some institutions specifically agreed to participate following the bombing. These participants spoke of their role in efforts to better understand and implement the Duty, precisely because terrorism had just occurred in their city.

Methods – Survey, Interviews, Focus Groups, and Participant Observation

In line with recent research on vernacular security and cognisant of the potential to leave avenues of enquiry unexplored (e.g. Jarvis and Lister, 2013: 287, 2016), we undertook a bottom-up, multi-method approach to capture the empirics of everyday security workings. We prioritised speaking with participants, rather than for them, creating ‘as much space as possible for these voices to be heard’ to ‘contribute towards recent efforts at addressing the historical lack of engagement with everyday (security) politics’ (Jarvis and Lister, 2013). Our approach to data collection was influenced by feminist (Enloe, 2011; Sjoberg, 2009), critical security and terrorism (Heath-Kelly and Strausz, 2019; James, 2022; Moffat and Gerard, 2020), and vernacular security (e.g. Stump, 2017) approaches. Specifically, the article made use of the following four methods: (1) an online survey, (2) semi-structured interviews, (3) focus groups, and (4) participant observations:

1. Survey: 49 of the 83 participants who started the survey completed it. Participants were recruited through targeted emails and snowballing, as well as the Schools and Colleges Network in Greater Manchester. Respondents included a range of education professionals (leadership, DSLs, support and teaching staff) across a variety of (private and state) institutions. Closed questions (see Supplemental Appendix) were analysed using basic and inferential statistics. Open questions enabled respondents to share contextual experiences and reflective comments, in a storytelling mode, with set questions designed to establish subject knowledge, confidence levels, and training frequency. The survey’s anonymity and distance encouraged detailed personal responses in addition to aggregated findings.

2. Interviews: 21 semi-structured interviews were conducted (following Sprague, 2005: 126) – 8 with policy elites, 4 with DSLs and 9 with education professionals. Three of these interviews followed directly from the survey. All but two were conducted face-to-face, in the participant’s workplace, with an average duration of 1 hour, enabling breadth and depth of discussion. A sensitive dialogical approach (Madill, 2011; Whiting, 2008) allowed rapport development and space for interviewees who had the potential to feel ‘uncomfortable or unsure about their responses’ (Revell, 2019: 26–27). Respondents were targeted and contacted in the same way as the survey.

3. Focus groups: To hear the experiences of Prevent among the safeguarded, nine 45–60-minute focus groups were conducted, with 45 students/pupils participating. Access to students was facilitated by DSLs within the participating institutions. Students were largely taken from their tutorial groups, ensuring a range of subjects were studied, with a plurality of classroom experience, and yet familiarity with one another (Wilkinson, 1998: 115–120). Warm-up activities, clear directions relating to task and the safety of the space, as well as efforts to counter dominant/hierarchical voices (Michell, 1999) ensured open, respectful dialogue with all participants involved (with insignificant divergences between groups and genders). As with the interviews, and following vernacular security approaches, participants led the discussions where possible, with enquiries following up on their insights and experiences. The use of focus groups enabled the project to generate data not usually accessible to researchers, revealing the security coproduction of the safeguarded – voices usually unheard (Farquhar, 1999: 62; Stanley, 2016: 2; Wilkinson, 1998: 114).

Data from focus groups (3) and interviews (2) were transcribed manually, before importing into NVIvo for coding. An initial coding framework was created on broader ‘subjects’ of discussion which were dominant across the methods such as ‘safeguarding’, ‘securitisation’, ‘training’, ‘referrals’ and ‘British Values’, with a secondary sub-coding (parent–child nodes) enabling more nuanced analysis. Content and discourse analysis (see Burck, 2005; Milliken, 1999), with a focus on narration (see Stanley, 2016), enabled the identification of aggregated thematics and the ‘storying’ of identities in the relational and temporal contexts of Prevent.

1. Participant Observations: Finally, 9 participant observations were undertaken across two educational institutions, observing three types of ‘classes’: (1) ‘normal’ curriculum delivery classes, (2) classes delivering Prevent-related tutorial sessions, and (3) training provider workshops for trainee teachers. Classes engaged students and education staff, with providers delivering radicalisation awareness lessons through arts-based activities, such as role-play, games, and theatre. During all observations, consent was obtained, with the lead researcher introduced, and anonymity ensured. Written notes were thus used rather than attributable transcription, with sessions enabling the visual assessment of engagement levels, methods for classroom delivery, and responsibilised-safeguarded interactions. In short, participant observation provided real-time-and-space contextual data – ‘unprovoked talk and conduct’ within everyday spaces and interactions (Stump, 2017: 213).

The combination of these methods enabled the triangulation of findings and a holistic analysis of individual moments of enactment, the socio-temporal context of these enactments, and a comparison of different enactments within these different contexts. This ‘methodological vitality’ enabled us to capture ‘multifaceted’ security experiences (Lister, 2019: 22, 26), shedding light on security’s coproduction at a variety of everyday levels.

Enacting Prevent

Analysis of the four-part dataset underpinned the development of our novel ontological conceptualisation of ordinary citizens as enactors of the state, engaged in security’s everyday trialogical coproduction. Here, we present our data in a thematic three-part structure, moving studies of counter radicalisation away from an understanding of the responsibilised as petty sovereigns and the safeguarded as merely acted upon. Both, we show, produce security, in dialogue with each other and the state. To develop that analysis, first, we highlight the messiness of policy reproduction and resistance which co-exist in practitioner and pupil behaviour, emphasising the agency of both groups to chart their own course through the landscape of Prevent, in nuanced everyday contexts. Second, we show how the agency of the responsibilised is iterative, interactive and creative, being shaped in recurrent dialogue between and across those safeguarded and state policy. We focus on how resilience building, and the rewriting of British values, result in novel, trialogically co-produced security assemblages. Third, we demonstrate this complex and contingent three-way coproduction of security with reference to the creative reimagining of Prevent’s referral process.

Beyond Binaries? Reproduction/Resistance, Responsibilised/Safeguarded

Confirming the findings of previous research (e.g. Busher et al., 2017), teachers understood Prevent as safeguarding and something that fitted within existing practitioner efforts around student care (e.g. TI2 DSL; David Wells, CTPNW; AI3 Senior DSL; Trade Union Official – see Supplemental Appendix a). ⁷ Risk, harm/suffering, and exploitation were recurrent themes in interviews. Within an overarching reproduction of the radicalisation ‘journey’, staff confirmed the need for early intervention to interrupt the ‘trajectory’ of ‘grooming’ efforts and ensure ‘wellbeing’ (David Wells, CTPNW; Nigel Lund, North-West DfE Prevent Coordinator; GM Council WRAP Trainer; TI2 DSL; DSL, FE/TI – see Supplemental Appendix b). However, pushing back against existing research, participants simultaneously understood and enacted Prevent as more than merely safeguarding as usual. Our triangulated findings repeatedly confirmed that Prevent was being enacted not simply as part of the usual process of spotting vulnerability, but rather as a complex, negotiated process, requiring a form of care that works both with and against the state and counter-radicalisation policy. Two moves were important here. First, Prevent was understood not as ‘reporting’ students but ‘referring’ them for help. And second, Prevent’s referent object was relocated from the state to cover myriad and variable individuals – the pupil themself, the community, or even ‘everyone’ (Rachael, AI1, Girls; contra Gearon, 2017; Miah, 2017 – see Supplemental Appendix c). These moves to rethink Prevent as ‘care’ were broadly shared by pupils on the basis that the responsibilised ‘are supposed to look after us’ (Aimee, AI3, Mixed Group 1).

Reimagined as part of a duty of care, Prevent was enacted both with and against the United Kingdom’s counter-radicalisation policy. Shifting security’s referent object to the individual to be safeguarded, within a potentially vulnerable community, alongside the re-thinking of intervention as a ‘referral for help’, should have made it easier for the responsibilised to (passively) implement the policies they had been trained to act out. Instead, the responsibilised and safeguarded alike were clear that Prevent went beyond safeguarding. Despite the language of referral, pupils noted that the process required them ‘to basically snitch’ on their peers, given the consequences that flagging concerns could engender (Sara, AI2, Mixed Group 2). Referral was understood as helping some but punitive for others – or, simply, ‘getting into trouble’ (Sara; Maisie, AI2 Group 1; Jake, AI3, Mixed Group2). Prevent, here, was seen as offering security and insecurity simultaneously (e.g. Revell, 2019; Sjøen and Jore, 2019). This translated directly onto concerns from one educationalist, who noted, ‘I don’t feel like I’m supporting students, I feel like I’m shopping them, which is very different with the rest of safeguarding’ (Fine Art Educationalist, AI1). This sense of ‘policing’ students eroded trust (Supplemental Appendix e). For both teachers and pupils, vernacular understandings of care and protection coexisted uneasily alongside fear of the policy, system and potential reprisals (Aimee AI3, Mixed Group1 – Supplemental Appendix f, g). And neither group was empowered to fully opt out of Prevent, if they wanted to remain in the FE sector (Supplemental Appendix h).

Our findings, then, do not dispute that evidence exists for the reproduction of, and resistance to, Prevent. Rather, our data show, first, that reproduction and resistance co-exist in complex ways and, second, that complex mixes of reproduction and resistance are shared by both the responsibilised and safeguarded, with the latter also experiencing a significant degree of responsibilisation. To be clear, our data suggest that it is not just teachers who have been responsibilised by Prevent, but also those to be safeguarded. This is particularly evident in the well-worn terrain of Prevent’s potentially racist undertones and implications (e.g. Breen and Meer, 2019). In a logic similar to Heath-Kelly’s (2017) analogies, some DSLs noted that Prevent’s ubiquity should eliminate the risk of racism (see also Supplemental Appendix i) – ‘it’s got nothing to do with the Prevent agenda’, ‘[it’s] complete and utter naivety, ignorance’, the policy ‘is safeguarding’. For most educationalists and students, however, the inherent risk of racism heightened the importance of their own agency in Prevent’s interpretation and implementation (Supplemental Appendix j, k, l). Teachers noted that Prevent’s intentions were irrelevant; it was the context that mattered, and implementation was complex on the ground, layered by a variety of interlocking engagements. Students noted their own experiences of stigmatisation around Prevent, which was heightened after the Manchester Arena attack (e.g. Supplemental Appendix n). Again, though, this did not lead to a simple rejection of Prevent but, rather, a nuanced evaluation and orientation towards it (Supplemental Appendix m). The point was not that Prevent is good and supported, or bad and opposed; Prevent’s logics were contextual, and agency had to be exercised with knowledge of complex and nuanced everyday heterogeneities in mind. In sum, Prevent inspires creative agency on the part of teachers and pupils, beyond the binaries that can structure extant analyses.

‘So We Kind of Create Our Own Values’

Prevent Duty guidance stipulates that education providers should embed prevention within their classrooms. Here, we focus on resilience building and/through the promotion of ‘British values’ as two key mechanisms through which this is achieved. Both examples help to demonstrate the interactive, iterative, and creative agency of teachers and pupils as security enactors, engaged in security’s trialogical coproduction.

Resilience building is aimed at empowering students to identify and reject extremist ideologies. Such efforts are controversial. Resilience has received significant critique in a neoliberal context generally (e.g. Chandler and Reid, 2016; Joseph, 2013) and regarding Prevent specifically, due to its perceived ‘transfer of responsibility’ to citizens (Stephens and Sieckelinck, 2020: 145, citing Anholt, 2017). The demand to equip pupils with the skills necessary to shield themselves from external harm has placed teachers in a particularly difficult situation (Supplemental Appendix o). Again, participants noted the importance of context, highlighting the variability of knowledge and training, as well as the constraints of time and space, when facing other workplace targets. Repeatedly, teachers reported that interactions with pupils changed based on their own and student knowledge levels. This was especially true for teachers outside of social science and humanities subjects (a focus on critical thinking skills was seen to lend itself more naturally to resilience building). For science, arts, and particularly vocational subjects, participants felt there were few, if any, natural opportunities for Prevent’s inclusion (Supplemental Appendix p). In short, Prevent looks very different from one interaction to another; it is not the homogeneous application of state policy by petty sovereigns. Rather, Prevent’s enactment is contingent upon the knowledge of the responsibilised, in dialogue with those to be safeguarded who possess varying ability levels themselves, in contexts with diverse imperatives for skillset development.

Context, heterogeneity, and creativity shaped these iterative dialogues and structured the nature of trialogical interactions. Two examples can illustrate this. First, one way in which teachers attempt to overcome concerns about knowledge limitations is to draw more heavily upon trusted sources and materials, usually with state approval. Teachers reported making use of materials from trusted outlets, (such as the Education and Training Foundation or Educate Against Hate) to overcome knowledge limitations in Prevent’s delivery (Supplemental Appendix q), counter perceived inadequacies in Prevent training delivery, and avoid racialised discourses (DSL, AI3). A second way in which the state is drawn extensively into security’s trialogue is through the governance role played by reporting mechanisms across areas of policy nexus. Even where support for resilience building was expressed, scepticism abounded over its governance (Supplemental Appendix r; see also Stephens and Sieckelinck, 2020). Teachers repeatedly flagged the importance of compiling an evidence basis for Ofsted, who were tasked with monitoring Prevent’s successful implementation – ‘it’s one more thing that I can say if Ofsted come and go “how does X, Y, and Z know that” it’s just one more thing to back me up’ (Business Administration and Customer Service Educationalist TI2). To meet Ofsted requirements, all institutions found ways of reimagining their delivery. Tutorial programmes and designated workbooks were two examples of ensuring time for delivery and evidence of impact. The selective use of security artefacts can also be interpreted as moments of negotiation and adaptation, showing teachers embroiled in processes of enactment, finding ways to overcome challenges and implement their version of the policy, while simultaneously meeting the governance expectations placed upon them.

Our dataset revealed the trialogical nature of security’s coproduction, as complex assemblages emerged at the intersection of state policy, the responsibilised, and the safeguarded. This was especially true in the reimagining of British values, which Prevent located at the heart of resilience building efforts; the theory being that instilling the right sort of (British) liberal values would inoculate students from those of extremists. Again, our data evidenced creative, nuanced agency, beyond pure reproduction or resistance. Prima facie expressions of scepticism and rejection were elaborated to reveal that it was not the values themselves which were viewed as problematic, but rather, their labelling as British (Supplemental Appendix s) – ‘they’re everyday values’, ‘classroom values’, of which non-British students sought co-ownership. ‘So we kind of create “our values”’ (Counselling, Psychology and Mental Health Educationalist, TI2). Rejecting the values’ association with a muscular liberal agenda, long promoted in the education system (e.g. Davies, 2016a; McGhee and Zhang, 2017; Mythen et al., 2017) and bringing Prevent into line with long-standing equality and diversity work, staff were proud that, with students, they had developed ‘their own values that are relevant to the school, that capture all those British values and that really mean something in their context’ (Trade Union Official).

This creativity was effective. Staff–student interactive agency and iterative dialogue helped to overcome the risk that British values act as a Trojan horse for ethno-nationalist discourses which were seen to work against Prevent’s aims by ‘creat[ing] barriers’ (Fine Art Educationalist AI3) through feelings of ‘isolation’ and ‘marginalisation’ (Counselling, Psychology and Mental Health Educationalist TI2; see also, James, 2022). By reframing the agenda around institutional values, focus was instead placed on promoting the values themselves, rather than allowing them to be used as a vehicle for a political agenda of assimilation (e.g. Mythen et al., 2013; Winter and Mills, 2020). Staff and students, together, were actively working to reimagine Prevent in a way that would not risk reproducing ‘us versus them’ binaries, reminiscent of the War on Terror (e.g. Kundnani, 2014; O’Toole et al., 2016). Students were engaged and vocal in these efforts, noting how a language of Britishness became just another way of ‘protecting ourselves (rather than everyone) and keeping people out’ (James, AI3 Boys). These cooperative efforts to reimagine Prevent were based on the identification of the policy’s creation of insecurity for many, especially in the specific contexts of urban Manchester, following the Arena attack (Supplemental Appendix t). Despite concerns regarding Ofsted, whose inspection framework required explicit engagement with ‘British’ Values and whose inspectors would ‘want to know why’ the language was not being utilised (Andrew Cooke, North-West Head of Ofsted), teachers felt the gain outweighed the risks involved in governance mechanisms. These negotiations provide a useful illustration of the process of enactment, where educators with students simultaneously resisted and reworked state policy, reimagining the framework through which the agenda became embedded in their institution and reclaiming the values as their own, coproducing security on their terms.

Reimagining Prevent’s Referrals Process After the Manchester Arena Attack

Building on the rewriting of Prevent’s ‘values’, our dataset reveals a reimagination of Prevent’s referrals process after the Manchester Arena attack, based on concerns about the insecurities the Duty engenders and the potential to contribute to them. We show how internal, informal referral mechanisms were introduced in response to the attack, as part of the trialogical coproduction of security. This involves state policy in complex and nuanced ways, as it is worked both with and around. Legislation, events, nexus areas of policy governance, and the safeguarded themselves were all involved and intertwined, in iterative interactions, creating complex security assemblages.

Interviews with DSLs revealed misalignment between government discourse and the practice of referrals ‘on the ground’. The notion that the responsibilised simply spot and refer received extensive critique. This critique underpinned a reimagination of Prevent’s referral mechanism. By focusing on collecting information internally and informally, a new security assemblage was constructed. Referrals now remained in-house, rather than pushing concerns – and individuals – directly onto Channel, the multi-agency safeguarding hub tasked with assessing and supporting potentially radicalised individuals. Reimagining the referral process in this way enabled DSLs to, first, minimise potential ‘knee-jerk reactions’ (Supplemental Appendix z) and, second, shift responsibility for ‘official’ referrals away from the teacher. This provided imperative reassurance for several teachers, enabling them to draw on the guidance of colleagues with greater expertise. DSLs were then able to enter the referrals process, in dialogue with the responsibilised, state policy, and the concerning behaviour of a student, as a second level of responsibilised actor. One educator described this process as creating a significantly more ‘positive experience’ of the system, where they had previously been fearful of Prevent (Supplemental Appendix u, v).

In vernacular language that mirrored student concerns, teachers spoke about their reluctance to refer. A desire to see more referrals was in the minority (GM Council Prevent Trainer) and for most DSLs, more referrals should be discouraged. Respondents reported ‘feeling like these are really grey areas’ and wondering ‘are you going to really affect someone’s life if you refer?’ Such concerns highlighted ‘the difference between the Prevent thing and the safeguarding thing’: ‘if I was concerned about a student in a different way, hurting themselves, I wouldn’t have thought twice . . . it’s because it’s such a loaded topic’ (Fine Art Educationalist, AI1). The perceived benefits of a reimagined referrals process were myriad: collective decision making through trusted relationships, greater distance to Channel and its potential implications, and time and space to ‘build up a picture’ (DSL, FE/TI1). Student discussions of referrals likewise highlighted the benefits of a sequential and staggered process, maximising safeguarding and minimising their own responsibilisation (Supplemental Appendix w).

These informal and internal mechanisms were, in significant part, a logical response to accusations of past failings. For example, on one hand, respondents reported having been accused of providing insufficient expertise to justify Channel referrals – ‘that email from the anti-terrorist unit a while ago where they more or less said to me “we want more”’ (DSL, TI2 – see Supplemental Appendix x for full quote). On the other hand, one institution revealed how this reimagination was a direct response to the firsthand experience of the system failing to intervene successfully, with three students losing their lives as terrorists. In the first instance, a good student, with zero warning, went to Syria, where he was killed (Supplemental Appendix aa). Left ‘devastated’ by the events, staff at the institution were in ‘absolute shock’ that contrary to their training, there had been no ‘signs’ that this student had been engaged in any form of radicalisation, nor had he been vulnerable at any point, with no indication from any other service that he may be engaged in these activities. The instance challenged what these educationalists had been told about the process of radicalisation and in turn, they questioned their own enactment of the Duty. A more informal process lowered the bar for internal referrals, increasing the likelihood cases would be picked up with a smaller evidence base.

In the second instance, another student was killed in Syria, after having been referred to Channel, but with officers informing teachers the referral was a mistake (Supplemental Appendix bb). Staff in the institution expressed feeling significant emotional turmoil following the failure to safeguard this individual from harm. It accompanied a feeling of being let down by the authorities, having identified concerns for this individual and not being offered the support they sought. It also created self-doubt among staff, concerning their capacity to implement Prevent. These doubts arose having had concerns downplayed in the first instance, and worries in the second instance that efforts to implement the legislation may have exacerbated insecurity: ‘did we push him over the edge?’ (Supplemental Appendix cc).

In the third instance, a third pupil’s death (as a terrorist) was deemed unpreventable, with staff noting a lack of warning signs (Supplemental Appendix dd). An investigation into the failure to stop this individual from committing his attack assured staff that they could have done no more. Nonetheless, the strength of the vulnerability narrative across FE – and the wider public sector – created a sense of blame from the wider community. The DSL recalled that, to this day, they still receive ‘backlash’: ‘you never told us anything!’ For this DSL and their institution, internal, informal referral mechanisms provided a response to the insecurity produced by having felt – or made to feel – responsibility for these students’ actions (Supplemental Appendix y). The result was a preference to ‘work with’ students, to ‘hopefully get to the bottom of it all’, internally.

In doing so, we posit that it is in these responses to their experiences and circumstances, that DSLs demonstrate enactment. Creating this robust internal system mitigated the challenges posed by limited external support. In resisting a system that did not work, supposed ‘petty sovereigns’ reimagined Prevent, creating a format and scope – at the heart of UK counter radicalisation efforts – that does deliver security, for the state, the community, and those to be safeguarded. Where official security policies have engendered insecurity, creative agency has enabled the formulation of better security processes, malleable in implementation, and responsive to the diverse needs of a heterogeneous citizenry (teachers and students, alike). Reimagining the duty’s implementation and core processes, this creative enactment was vital and rendered staff far more than merely the ‘eyes of the state’ or a passive vector for policy.

Conclusion

The article began by making the case for re-thinking how we understand the production of security. It grounded theoretical and empirical contributions by combining two recent, concurrent trends: a concern in IR with everyday security and the politics of ordinary life; and a reimagination of terrorist threat as a broad societal safeguarding responsibility within UK counter-terrorism policy. We brought these concomitant developments into dialogue through an examination of the everyday experiences of those implicated by Prevent, within the United Kingdom’s FE sector. Mobilising vernacular methods enabled us to analyse a novel dataset and advance the aims of critical, everyday research agendas, hearing the voices of ordinary citizens as a direct challenge to the state-centrism of IR and counter-terrorism research. Challenging the dominant view of responsibilised actors as ‘petty sovereigns’, we reconceptualised security’s coproduction by ordinary publics. Our core theoretical contribution is to have developed a novel ontological conceptualisation of ordinary publics, not as actors of the state, as portrayed in existing critical literatures, but as enactors of state, embroiled in nuanced coproductions of security. This novel understanding enabled the article to demonstrate in its analysis how security is co-produced through iterative trialogical interactions between (government) policy, (responsibilised) teachers, and (safeguarded) students.

As Jerome et al. (2020: 159) speculated, ‘grassroots policy enactment by education and childcare professionals’ could ‘significantly reconfigure’ Prevent and ‘give rise to important “bottom-up” policy innovations’. Our data have shown that this is the case: nuanced and creative dynamics operate through citizens’ variegated imbrications with state policy, complex relationalities, and subtle nuances in everyday enactment. In contrast to existing research, we argue that responsibilised education professionals and safeguarded students, retain creative, iterative, and interactive agency. Teachers are not Butler’s ‘petty sovereigns’, ‘unknowing’ and ‘without any real power to think or act for themselves’ such that staff merely ‘“play” their parts’, ‘unilaterally and with enormous consequence’ (Butler, 2004: 65; Younis and Jadhav, 2019). Rather, they demonstrate considerable knowledge and creative agency, in trialogue with the state and the safeguarded, in complex ‘policy ensembles’ (Busher and Jerome, 2020: 6). Our findings challenge the extant literature’s de-agentification of its subjects and nuance the binaries that structure its analyses.

To further these findings, we suggest that our article contributes to still relatively early calls for a vernacular security studies research agenda (e.g. Jarvis et al., 2024). Two limitations of our article highlight the work still do be done. First, we have developed the novel and potentially useful concept of ‘enactors of the state’. However, our data underpinning this is limited to a detailed case within a specific liberal democratic country. Future research might explore contrasts and parallels beyond the UK and in different types of state. Second, our ontological reconceptualisation of security’s coproduction is derived from a close and detailed analysis of counterterrorism and counter-radicalisation specifically. Future research exploring other sites and sectors of security would bolster this finding and further develop vernacular security research and analyses of the everyday. In sum, we have extrapolated from data on UK counterterrorism to develop a general theoretical finding at the level of security theory. Security is a contested and encompassing concept; it can mean different things in different contexts, shaping politics and normal life in myriad ways. A greater variety of cases from a wide geographical range would bolster the findings of this article. We suggest that the emerging research agenda in vernacular security studies is well set up to address this need.

Supplemental Material

sj-docx-1-psx-10.1177_00323217241236074 – Supplemental material for Enactors of the State: The Everyday Coproduction of Security in the Prevention of Radicalisation

Supplemental material, sj-docx-1-psx-10.1177_00323217241236074 for Enactors of the State: The Everyday Coproduction of Security in the Prevention of Radicalisation by Jack Holland and Natalie Higham-James in Political Studies