One of the key features of the global war on terror has been the development of countering violent extremism (CVE) programmes to meet the threat of ‘home-grown’ terrorists. The UK’s counter terrorism strategy, CONTEST, is made up of four policy strands—Pursue, Protect, Prepare and Prevent. Pursue uses conventional counter terror practices and places suspected terrorists (both home and abroad) under military, policing, intelligence and judicial measures. Protect is based on ensuring that through greater protective security measures, the UK is equipped with the appropriate infrastructure to mitigate the effects of a terror attack. Prepare is focused on having effective procedures in place so that in the event of a terror attack, the country is resilient. Prevent is aimed at ‘countering the ideology’ thought to be propagated by terrorists and seeks to deal with the so-called driving factors of terrorism (Sabir, 2017; p 203). This has led to an ever-expanding army of ‘experts’ dedicated to the task of empirically determining an extremist profile and predicting future terrorists (Wight, 2015, p 210; Monaghan and Molnar, 2016, pp 393–413). As a consequence, extensive lists and classifications have been generated to uncover the ‘signs’ of extremism.

Since its inception, the Prevent strategy has been increasingly challenged for reinforcing Islamophobic prejudices through racialized practices of surveillance on primarily Muslim communities (Sian, 2015a, pp 183–201). For established liberal democracies, such programmes present a significant challenge as they involve a curtailment of civil rights over a sustained period of time. Despite these concerns, there is a general agreement that the identification of violent extremism is key to the success of counter terrorism campaigns. Consequently, the British CVE programme has become the model for many other Western countries and its implementation has had an international significance (Sabir, 2016, pp 1–15). It is against this backdrop of security and regulation that this paper sets out to explore the conceptual overlap between the discourse on counter extremism and its nineteenth century antecedents in the development of positivist criminology.

This paper is interested in developing a critique of Prevent by arguing that the discourse (including various classifications, indicators, and categories) that has been used and applied to ‘detect’ extremism shares a conceptual familiarity with 19th century race-thinking and positivist criminology, and is therefore likely to suffer the same methodological flaws. Through an analytical engagement with the works of Cesare Lombroso (1876) and his Criminal Man thesis, this paper will build an account to demonstrate the political function of (race) science and criminalisation in the disciplining of racialized populations. The comparison being made between Lombroso’s thesis and the Prevent programme is of particular importance for this paper. In the same way that Lombroso employed the ‘scientific method’ to develop typologies to draw conclusions around criminality, the Prevent programme has similarly attempted to use a scientifically based framework to draw evaluations around extremism. This has been done through the formulation of various indicators used by Prevent to determine extremism. Such indicators are based on the Extremism Risk Guidance 22+ (ERG 22+) framework developed by psychologists, Christopher Dean and Monica Lloyd (Qureshi, 2016, pp 12–13).

In a similar way to Lombroso’s model of identifying criminals, ERG22+ is based on trying to identify extremism by establishing a series of classifications that claim to have a ‘psychology evidence base’ (ibid). The ERG22+ framework has as such been given credence to ‘scientifically’ predict future terrorists, as well as influence referrals to the governments de-radicalisation programme, Channel (ibid). Prevent’s ‘Channel Vulnerability Assessment Framework 2012’ contains a total of 22 indicators of extremism, all of which have been derived from ERG22+ (ibid). The paper to follow will examine these indicators alongside broader strategies around policing, as a way to tease out the complex and textured intersections between Lombroso and current counter terror practice. By exploring the limits of such positivist approaches, the paper will argue that these frameworks appear to be structured more so by the logics of racism/Islamophobia rather than fact/truth.

Lombroso and positivism

Positivism broadly refers to a philosophical outlook that maintains the idea that all phenomena, from physics to sociology, can be understood through the application of scientific method. It can be understood as an epistemological stance that rejects social constructionism and genealogical approaches, as such it represents, “one way of thinking about the basis on which knowledge can claim to be scientific” (Walklate, 1998, p 3). Positivists subscribe to the notion that scientific knowledge and reason can control and manage not only nature, but also social life. This position represents a key characteristic of modern society and influenced not only the natural sciences, but also the social sciences including criminology. The desire to produce scientific, objective and value-free knowledge in the field of criminology has been particularly strong, and according to Sandra Walklate (1998) positivist criminology is underpinned by, “a commitment to objectivity; understanding the determined nature of criminal behaviour; and by its ability to measure criminal behaviour” (ibid, p 4). She goes on to suggest that through an understanding of social, psychological and biological factors, it is claimed that the actions and behaviours of criminals can be measured objectively, thus maintaining a commitment to the scientific project and the production of universal explanations (ibid).

For Walklate, the impulse to develop an analysis of crime that would apply to all contexts has formed a key feature of criminology and is linked to the underlying desire to devise policy (ibid). This point is particularly important for this paper, which goes on to show how such articulations of positivist criminology (both past and present) shape reductionist ideas, policies and programmes around criminality. In the case of Prevent, positivism provides a necessary scientific veneer to various constructions of Muslims and extremism that make the policy possible. In order to understand the influence of earlier positivist thinking on contemporary counter terror policy and practice, we might want to begin with Cesare Lombroso who is among the most widely cited early positivist criminologists in standard European criminology textbooks.

Lombroso is regarded as the founder of positivist criminology (Agozino, 2003, p 4; Walklate, 1998, p 14; Young, 2011, p 71). He drew upon racial science, eugenics, psychiatry, biology, and social Darwinism to compile his works on the Criminal Man, first published in 1876. In his most infamous statement, Lombroso claimed that individuals were ‘born criminal’ proposing that criminality was inherent, and not a product of the social environment. Lombroso invested much of his energy into identifying and classifying criminal types, and in Criminal Man his examinations of the skulls of criminals led him to propose that they could be categorised by their physical traits or ‘defects’ including, deformed skulls, being taller than a ‘healthy person’, having misshaped noses, slanted foreheads, big jaws, and having dark eyes, dark hair, and dark skin (ibid). What emerged from Lombroso’s account was both the continuation and the expansion of racial science thinking, whereby non-whites (and women who did not conform to traditional gender roles) were constructed as essentially criminal. Lombroso built upon earlier ideas of race-thinking which began to emerge in Europe during the long sixteenth century as part of the new world inaugurated by colonial expansion.Footnote 1

Race-thinking was largely rooted within anthro-scientific theories and gained prominence particularly during the period of European Enlightenment (Lee, 2015, p 26). Notions of race were mobilised to arrange humans according to phenotypical differences and as systems of classifications continued to advance, a schema governed by white superiority and non-white inferiority was established throughout Europe (ibid, p 33). This hierarchy served as an important ideological tool to legitimise and justify European practices of slavery, colonialism and genocide. The appeal of racial science came in its claims of scientific objectivity, reliability, and universal truths, however as Sandra Soo-Jin Lee (2015) reminds us, such a discourse arose from a social landscape of a particular set of beliefs (ibid, p 32), thus racial science was adopted less for its claims for being ‘value free’ and more for its capacity to justify racist policies through the creation of a seemingly ‘natural’ division of races (ibid, p 32).

As a consequence, earlier ethnological approaches (which were guided by general standards of degrees of difference between the races) were replaced with a new framework that sought to establish an absolute difference between whites and non-whites (Young, 1994, p 160). As Robert Young (1994) suggests, in this new criterion, “non-whites were classified in terms of how much they deviated from white standard. This constitutive difference of the species became the central focus of racial theory” (ibid). In other words, by declaring non-whites a different species, the constitution of the US (which proclaimed that ‘all men were born equal’) could be preserved, as such a classification meant that equality did not apply to non-whites who were categorised as an entirely ‘other’ breed (ibid). Slavery was given the green light, and from the 1840s onwards, ideas of polygenesis continued to spread throughout Europe (ibid). Such thinking went on to inform a number of academic disciplines-in particular the field of criminology, which was keen to incorporate such ideas.

The work of Lombroso was long one of the most influential in the application of racial science to address social problems. Lombroso’s most recognised assertion was that criminals were ‘born with evil inclinations,’ and through his examination of cranial anatomy, biology, genetics, psychiatry, and anthropology, read through a positivist epistemology, he was able to compile a comprehensive catalogue to mark out specific criminal types. Lombroso’s research fundamentally turned the study of crime into a science. Through the use of Darwin’s theory of natural selection Lombroso was able to create hierarchies of evolution and make claims that the skulls of criminals were deformed, resembling the skulls of rodents (Agozino, 2003, p 25). As such, Lombroso’s theory of born criminals could be applied by identifying a range of defects to verify that criminals were simply savages (ibid). Biko Agozino (2003) rightly points out that the main problem for Lombroso lay in the fact that his catalogue of defects paralleled the features of dark skinned Sicilians, who had been racially constructed in both criminal justice discourse and popular culture as representing the largest source of criminality in Italy (ibid, Young, 2011, p 71). Although Lombroso went on to revise his crude biological determinism, Agozino suggests that his ideas remained influential as they confirmed common sense notions of pure blood and aristocracy, and more importantly, he was the first to have executed a scientific experiment in the field of criminology (ibid).

The violent extremist man: racialization and prevent

The current British counter terrorism Prevent programme shares with Lombroso’s work a number of features. To make clear the degree of overlap between Lombroso’s positivist criminology and the development of Prevent, it is useful to outline the British state’s attempt to deal with the home front of the war on terror. Prevent is a key strand of the governments wider counter-terrorism strategy CONTEST that was officially introduced in 2006 by the Labour government following the London bombings (7/7). The focus of the policy is to deal with so-called ‘extremist ideology’ propagated by groups and individuals loosely defined as ‘Islamic extremists.’ Consecutive governments and policy documents on Prevent have all made claim to the idea that extremist ideology is at the very centre of driving terrorism (HM Government, 2011, p 1).

As such, the key goal of Prevent is to identify and contain suspected extremists. It does this by targeting those deemed vulnerable to extremist ideologues and placing them through a ‘de-radicalisation’ process (ibid). This approach is therefore based on the assumption that recognising the early signs of extremism is a prerequisite in effectively counteracting extremism and thus terrorism (Qureshi, 2016, pp 4–50, Monaghan and Molnar, 2016, pp 393–413). Once a suspected extremist is identified, the individual is then referred for de-radicalisation under the Channel programme. Channel utilises a ‘scientifically’ based vulnerability assessment framework (ERG22+) in order to evaluate the most appropriate ‘de-radicalisation’ package that will be applied to the individual (Qureshi, 2016, pp 18–20). This framework consists of 22 empirically based indicators that aid the task of spotting signs of extremism (HM Government, 2012, pp 1–18). These include: Grievance/Injustice; Threat; Identity; Meaning and belonging; Status; Excitement, comradeship or adventure; Dominance and control; Susceptibility to indoctrination; Political/moral motivation; Opportunisitc involvement; Family and/or friends support extremist offending; Transitional periods; Group influence and control; Mental health; Over-identification with a group, cause or ideology; Them and Us thinking; Dehumanisation of the enemy; Attitudes that justify offending; Harmful means to an end; Harmful objectives; Individual knowledge, skills; Access to networks, funding or equipment for terrorism; Criminal capability (ibid).

Although such indicators appear broad and general and could seemingly apply to anybody, in the context of the war on terror, the underlying focus is exclusively upon Muslim bodies, who through a particular set of markers, have been subjected to practices of racialization (Sian, 2015a, pp 183–201). The racialization of groups is not only restricted to those biological markers which previously dominated race-thinking,Footnote 2 rather in the war on terror, racialization has come to signify a radical ‘otherness’ which distinguishes individuals and groups by a stronger emphasis on origin and heritage, as well as physical appearance (Meer and Modood, 2010, p 77). As such religious, cultural, and phenotypical identifications infuse together to mark out, or racialize Muslims (and those perceived to be Muslim based on these characteristics) (Meer and Modood, 2010, p 77).

On further inspection of the indicators, it becomes clear to see that such ‘signs’ are underpinned by racialized understandings of Muslims, which are dependent upon culturally and religiously deterministic frameworks (Sian, 2015a, pp 187–189). For example, under the heading, Us and Them Thinking the literature states: “An example is the Al Qa’ida influenced ‘single narrative’ explanation of world events in which Muslims are portrayed across history as the victims of an evil and decadent West” (HM Government, 2012, p 14). In relation to Dehumanisation of the enemy, the literature asserts: “Al Qa’ida influenced offenders may justify targeting members of their own Muslim community by moral disgust at their hypocrisy and perceived betrayal of their fellow Muslims” (ibid, p 15). Al Qa’ida activity is over-represented in the documentation, it is only under the Political/moral motivation category that we see ‘other’ forms of extremism briefly mentioned (e.g., IRA, far right extremists, anti-abortionists and the Basque separatist movement). In terms of Identity, Meaning and Belonging it is suggested that:

This may be particularly attractive to those who have been disrupted by moving from one country to another and who may be confused by conflicting cultural demands…individuals may have experienced an aimless lifestyle or have indulged in behaviour contrary to a particular code of faith and be attracted to extremist values to displace this lifestyle and assuage guilt (ibid, p 9).

This example shows that even when not explicitly referring to Muslims or Islam, the explanations strongly evoke ideas of Muslimness, in that the potential extremist is linked to religion and culture.

Prevent practitioners however are keen to insist that they are targeting all forms extremism rather than focusing on particular communities (ibid, 2015, p 3), yet there are various documented cases which can be seen to trouble these claims. For example, in 2015, Mohammed Umar Farooq, a University of Staffordshire postgraduate student, was interrogated by a university official for reading a textbook on terrorism relating to his MA programme on terrorism, crime and global security, in his university library.Footnote 3 Needless to say the student was identified as Muslim. Similarly in 2016, teachers referred Muslim schoolboy, Rahman Mohammadi to the police under Prevent for wearing a free Palestine badge.Footnote 4 In another case, teachers at a school in London questioned a Muslim pupil under the Prevent policy about ISIS after he used the word “eco-terrorist” in a French lesson.Footnote 5 These cases draw connections with the disproportionate targeting of Muslim bodies in which, “the broad array of indicators posited as drivers of radicalisation have furthered discriminatory practices targeting the Muslim community as ‘usual suspects’” (Monaghan and Molnar, 2016, p 394).

Prevent policy claims that there is no single profile of an extremist, though such an assertion is not really possible given the nature of the dependent variable “terrorist.” In Channel awareness training packages, which have been rolled out across civil society organisations on ‘how to spot an extremist,’ the profiling of Muslims is explicit. Another warning sign often presented in Prevent discourse is, “significant changes to appearance/behaviour” (HM Government, 2008, p 18). Channel awareness training seeks to examine this issue by providing a scenario concerning a Muslim school girl who replaces her Western dress and socialising activities, with wearing a hijab and taking a keen interest in Islam. This change in appearance/behaviour is framed as an early warning that she could go on to potentially engage within extremist activity (see Fig. 1).

Fig. 1
figure 1

Channel Awareness Training. This image is not covered by a CC BY license

Alongside this, it has also been suggested that Muslim men face potential criminalisation under Prevent for growing a beard.Footnote 6 In their research on counter terrorism policing practices in Canada, Monaghan and Molnar (2016) found similar patterns at work. For instance, in a training slide designed for the RMPC the same assertion that, “there is no identifiable profile but common indicators” was presented. However despite this claim the discourse was unable to disassociate Muslims from radicalisation, as such Islam/Muslims inevitably dominated the discussion on (de)radicalisation:

As these training slides demonstrate, the efforts to convey complexities in academic accounts of radicalisation are absent in the constructions of difference that define police operations. In this application of indicators, the nuance of radicalisation studies is lost in the essentialising and racialised imagery… Under the banner of a ‘bias-free’ policy, the search for indicators tends towards problematic conjecture that draws upon racialised inferences in pre-emptive operational contexts (Monaghan and Molnar, 2016, p 408).

It therefore seems apparent that the discourse of de-radicalisation has disproportionately profiled Muslim communities despite claims of neutrality, objectivity and non-discrimination. Official statistics go on to support this argument and reveal how Prevent overwhelmingly targets Muslims. Data released by the Association of Chief Police Officers under the Freedom of Information Act, for example, shows that from 2007/08 until 2010/11, 67% of all people referred to Channel were Muslim (Traquair, 2014). For unknown reasons, data was not collected on religion between January 2011 and March 2012. When data was re-collected from 2012 until 2014, it revealed that 57% of all referrals made to Channel were Muslim. Then from 2014 until 2016, it shows that 87% of Channel referrals were Muslim (ibid, 2016). Undoubtedly the ‘central fixation’ for law enforcement officials has been upon the profiling of Muslims (Monaghan and Molnar, 2016, p 402). As Monaghan and Molnar go on to suggest, “it is precisely in the reification of particular attributes as part of the governance of future risks that radicalisation knowledge are twinned with the identification of generic characteristics of Islam that could be indicators of future violence” (Monaghan and Molnar, 2016, p 401).

The indicators around ‘extremist types’ are therefore not bias-free largely due to the fact that the very signifier of ‘radical’ comes with a set of distinct essentializing features (ibid). As a result, the indicators are guided by an Islamophobic logic; in very much the same way that Lombroso’s observations of criminal types were also guided by a racist logic. In Criminal Man, for example, Lombroso writes:

Those who have read this far should now be persuaded that criminals resemble savages and the coloured races. These groups have many characteristics in common, including, thinness of body hair, low degrees of strength and below-average weight, small cranial capacities, sloping foreheads, and swollen sinuses. Members of both groups frequently have sutures of the central brow ridge, precocious synostes or disarticulation of the frontal bones, upwardly arching temporal bones, sutural simplicity, thick skulls, overdeveloped jaws and cheekbones, oblique eyes, dark skin, thick and curly hair, and jug ears (Lombroso, [1876] 2006, p 91).

He then goes on to suggest:

Among habitual criminals as among savages, we find less sexual differentiation between normal men and women. In addition, in both we find insensitivity to pain, lack of moral sense, revulsion for work, absence of remorse, lack of foresight, vanity, superstitiousness, self-importance, and finally, an underdeveloped concept of divinity and morality (ibid).

In the former explanation provided by Lombroso we see overt support for scientific/biological accounts of race. Such a reading of race as described earlier, is characteristic of the times in which Lombroso was writing whereby assumptions that particular ethnic groups have biologically inheritable characteristics became institutionally embedded. In the latter explanation, the attributes listed by Lombroso are broader and move beyond physical traits to cultural traits, which appear on the surface to be more subjective. However, as we have seen, Lombroso is guided by a particular understanding of race and crime. In the same way that the 22 indicators of extremism are wide-ranging, it might be useful to consider that they too are guided by a particular cultural understanding of Muslims and extremism which assumes a causal link between the two (Monaghan and Molnar, 2016, p 401).

The Prevent programme appears to share with Lombroso an attempt to ‘objectively’ link criminality with race, and extremism with Muslims, respectively. This of course is the main weakness of positivist criminology. Such a linking, whether through science, biology or culture, does not emerge naturally. Rather it is established to operate politically as a way of disciplining and punishing racially marked bodies; historically through systems of slavery, colonialism, the holocaust, and apartheid, and in more recent times, through systems of racial profiling, stop and search, and de-radicalisation programmes. What we are seeing in the Prevent policy then is the continuation of 19th century ideas that sought to regulate racialized communities. Perhaps the only twist is that the seemingly biological has been replaced by a stronger focus on the cultural (Goldberg, 2002, p 26). In the knowledge formations around contemporary Islamic extremism the idea of race appears to serve as a central feature to essentialize Muslims in a misguided and irresponsible quest to correct, reverse or stop the radicalisation process.

For Lombroso the prevention of crime could be best achieved through ‘scientific policing.’ He argued that:

Hitherto policing was conducted much as wars used to be waged: randomly and on the basis of hunches. Successful investigations depended on the astuteness and dedication of a few individuals. What we need now is to apply the scientific method to the identification of criminals (Lombroso, [1876] 2006, p 331).

Since 9/11 exceptional measures have been implemented to gather intelligence on citizens. This has included the monitoring of telephone patterns without permission, the snooping of Internet browsing histories, and unlawful wiretapping (Lustick, 2006, p 32; Bauman and Lyon, 2013, p 107). Advanced surveillance technologies continue to seep into everyday life-worlds shaping the experiences of Muslims and racially marked populations more broadly. The mapping of the ‘electronic footprint,’ the collection of biometric data, and the use of complex algorithmic data in government watchlists have all been normalised in the West’s war against terrorism (Amoore and Goede, 2008, p 182).

In the informal space, Lombroso proposed that crime prevention measures should operate in schools:

Born criminals should not be allowed to infiltrate elementary schools because education would harm both them and society… Advances in criminal anthropology have now made possible the preventative isolation of criminals- the most important measure of social defence. Teachers are now able to identify in children the incurable signs of inborn criminality and use these signs to distinguish between innate criminality and the temporary criminality of all youth. These signs include physiological and craniological anomalies in children, coupled with tendencies toward wrongdoing (Lombroso, [1876] 2006, pp 334–335).

Lombroso would perhaps take solace in the workings of the current Prevent strategy. Under the Counter-Terrorism and Security Act (Sept 2015) public sector bodies (e.g., education, health, local councils, amongst others) across the UK have a legal duty to refer potential ‘extremists’ to an appropriate body if they show signs of becoming ‘radicalised’ (Sian, The Guardian, 21 July 2015b). As we have already seen in the above examples, risk averse administrators have used this obligation to criminalise a whole range of legitimate activities, including book-reading, talking about eco-activism etc. The argument that British universities are crucibles of radicalisation continues to animate government thinking on this topic and strongly echoes Lombroso’s thesis around schools playing a central role in detecting criminality in children. The legal obligation of teachers to essentially spy on students via training on extremism has proven a farce at best (Sian, 2015a, pp 183–201). In the same way that it is absurd to suggest that teachers should be able to detect ‘physiological and craniological anomalies in children, coupled with tendencies toward wrongdoing’ as proposed by Lombroso, it is just as nonsensical to place a legal obligation on teachers to (impossibly) spot the signs of extremism in children. It would therefore seem to be the case that these measures serve a specific political function and sit within a broader historical narrative of racial repression.

Contemporary ‘scientific’ surveillance measures have undoubtedly targeted Muslim populations, however the social control of racially marked groups is by no means a new phenomenon. Practices designed to socially control racialized populations gathered pace in the 19th century prompted largely by the modern eugenics movement, which was spearheaded by Sir Francis Galton in 1865 (Lee, 2015, p 33). Such thinking inspired forced sterilisation programmes of blacks, Native Americans and Latinos across the USA, and of indigenous communities in Australia and Canada. The uptake of eugenicist approaches also paved the way for genocide in Nazi Germany, the exercise of miscegenation laws in the USA, and strict immigration policies across Western nations (Garner, 2010, pp 71–75). Although the eugenics movement lost its popularity following the death of millions of Jews in Nazi Germany, the logics of population control through what can be described as racialized governmentality stayed put (Hesse and Sayyid, 2006, p 21), particularly within the sphere of law enforcement. That is, positivist methods have remained paramount in the restricting, management, and surveillance of racially marked groups who have increasingly come to represent criminals, deviants, threats, suspects and undesirables in the Western psyche (Amin, 2014, p 102).

In his pioneering work Policing the Crisis, Stuart Hall et al. (2013) demonstrated that from the early 1970s onwards, British law enforcement officials, political discourse and sensational media coverage were all central in the production of an (empty) association between black youth and ‘mugging,’ to the extent by which, “mugging and black crime are now virtually synonymous” (Hall, et al. 2013, p 321). Through a series of moral panics the theme of ‘black crime’ generated public hysteria around social order and racial ‘others.’ By ‘objectively’ equating mugging to black youth, official responses in the form of oppressive street policing in inner city areas i.e., ‘trouble spots,’ were legitimised (ibid, p 326). The disproportionate use of stop and search and racial profiling of black youth in both the UK and USA has been documented widely. As Angela Davis (2003) argues, “that it is possible to be targeted by the police for no other reason than the colour of one’s skin is not mere speculation” (Davis, 2003, pp 30–31). Davis goes on to point out that in the US, “police departments in major urban areas have admitted the existence of formal procedures designed to maximise the numbers of African-Americans and Latinos arrested- even in the absence of probable cause” (ibid, p 31).

Lombroso’s application of the ‘scientific method’ to the identification of criminals, in place of ‘policing based on hunches,’ is now a dominant method of fighting crime. In other words, positivist assumptions tend to underlie modern policing practices which are increasingly organised around the idea that a crime is caused by easily identifiable indicators. Not only in popular culture have we seen this shift being documented in TV series with savvy CSI style programmes replacing the traditional Columbo genre of crime entertainment, but also in practice societies have witnessed an expediential growth and sophistication of DNA databases, retina recognition technologies, and forensic biometrics, which continue to advance in the pursuit of social control. The use of such scientific methods in law enforcement agencies reproduces race-thinking logics. As Robert Carter (2007) argues, police DNA databases reflect racial (gender and class) biases of policing, by developing specific correlations between the genetic profiles of different types of criminals (Carter, 2007, p 552). In terms of contemporary migration and border control regulation, Katja Franko Aas (2011) documents extensively the ways in which racially marked migrants in particular have become criminalized and managed through a plethora of systems monitoring their moves. From the Schengen Information System which serves as the largest operational database related to police, judicial assistance and external border control in Europe, to the Eurodac database which registers and compares the fingerprints of asylum seekers, alongside the Visa Information System, the Automatic Border Crossing system and the Entry/Exists System (to name but a few) the ‘crimmigrant body’ is subjected to hi-tech modes of surveillance and governance (Aas, 2011, pp 334–336).

These examples all seem to point to the way in which the political construction of the racialized body has served as a tool, both historically and in the present, to ensure the hegemony of whiteness, Western superiority, and positivism. Prevent might be best understood as yet another chapter in the dusty catalogue of European positivist criminology- that is a field as well as a practice which has always been dependent upon the ‘scientific’ category of ‘race.’

The limits of positivism

In his book, Counter-Colonial Criminology, Agozino offers critical insights into positivist criminology and the work it has done, and continues to do, in reproducing colonial knowledge formations. His ideas are thus significant for the development of the critique of positivist forms of thinking and its relationship with the colonial. One way to understand the war on terror is as a colonial campaign of counter-insurgency being waged in a postcolonial context.Footnote 7 The intersection between colonialism and positivism is not purely accidental, in that, the colonial enterprise was imagined and implemented through the positivist appropriation of the colonised, through surveys, censuses and various classificatory schemes.Footnote 8 Agozino makes the explicit link between Lombroso’s positivism and the racial science upon which it was built:

Positivism may assume that sense-impressions are the only basis of knowledge, but Lombroso’s theory is not ‘positivistic’ simply because it adopted operational definitions of causal attributes. Lombroso’s claim to valid knowledge was not merely based on the appeal to immediate apprehensible measurements. It was based on ideological ‘truths’ derived from race-class-gender stereotypes and imperial dogma (Agozino, 2003, p 82).

In his reading of Stephen Pfohl and Gordon’s (1986) essay on ‘Criminological Displacements: A Sociological Deconstruction,’ Agozino builds a convincing set of arguments around how we might challenge the enterprise of positivist criminology, which as this paper has mapped out has been central to the governing of racial/Muslim populations. Agozino examines Pfohl and Gordon’s analysis to demonstrate the limits of positivist criminology, which is argued to be rooted within three interconnected ‘pleasures’ as opposed to ‘truths.’ They include, firstly the pleasures of sadism: mastering the facts of crime as ‘things.’ This refers to the hierarchy between the criminologist as ‘he’ and the criminal as ‘she.’ In this positivist framing, the criminal (like the woman) represents a constitutive ‘Other,’ which is to be ‘penetrated’ as a way to ‘master’ the causes of criminal behaviour (Agozino, 2003, p 96). This mirrors the way in which the dominant male gaze functions to subordinate women, i.e., “master them, penetrate them, lay them bare and control them” (ibid). In this way, the sadism is less about rationality and more about the pleasure of exerting power over others (ibid). In other words, the pleasure comes from the ability to inflict pain on other bodies.

The second aspect is, the pleasure of surveillance: the eye upon her is the gaze that categorises the Other. Here it is argued that positivism is viewed as ‘the end of ideology’ as it is based upon the assumption that the Other can be measured, controlled, and differentiated through close observation. It assumes that those who adopt positivist methodologies will generate the same set of results allowing for claims to be checked for reliability (ibid).

Thirdly is, the pleasure of truth: the normal subject and his Other. This pleasure is centred around the notion of ‘truth,’ whereby the assumption of criminologists, police officials, judges, psychiatrists and so on, is that both victims and offenders are telling the truth, and if they are not this can be unmasked through skilful questioning. In this way criminal justice actors are able to make claims for the superiority of their truth, “because it is not the truth of an individual but the balanced, objective truth gleaned from all the individual truths told by interviewees” (ibid). As Agozino goes on to suggest, in his critique of reason Foucault (1961) proposes that it is precisely this ‘pleasure of truth’ underlying positivism which must be critiqued, not for the purpose of revealing a higher truth, but rather, “for the purpose of exposing the will to power inherent in what seems to be a disinterested objective view” (ibid, pp 96–97).

In the Prevent strategy, it would appear that the mastering of the (Muslim) ‘Other’ has been a focal point in organising lengthy classifications around ‘extremist types,’ (similar to Lombroso’s categories around criminals types). By utilising positivist thinking based on the ERG22+, Prevent practitioners (as well as law enforcement officials and teachers etc.) seek to dominate and discipline Muslim bodies.Footnote 9 We have also seen how surveillance operates as the principle feature in Prevent discourse, whereby civil society and public sector bodies, have been drawn in to observe, spy upon and control the Muslim body. The rolling out of the Prevent policy in such a way follows suit with the positivist tradition to ensure that its results can be measured, compared and classified. This has given credence to ideas that the various indicators and profile analysis have a ‘definite truth’ and operate objectively (Monaghan and Molnar, 2016, p 400).

Finally we can observe in the Prevent strategy the way in which the idea of a superior ‘truth’ guides the regulation of Muslim bodies. As Monaghan and Molnar (2016) note, surveillance practices around counter-terrorism can be seen as Orientalist knowledge constructions. In this way, Western knowledge operates to legitimise representations of Muslims, “through inferior, illiberal and irrational essentialisations” (ibid). The ‘truth’ that appears to be driving Prevent is often presented as being evidence based, however, despite its grand claims for scientific objectivity, we have seen a degree of uncertainty in the Prevent strategy, which seems to be guided more by historic policies and practices associated with racism and contemporary forms of structured Islamophobia, as opposed to logic.

Instead of the obsessive prodding and poking of Muslims, it might be more useful to look beyond positivist inspired de-radicalisation programmes like Prevent, both conceptually and empirically, because the issue is perhaps not simply about ‘correcting’ ‘defected’ behaviours and mindsets, but rather about profoundly changing structural conditions. Following Weber’s (1970) seminal works on the limits of science, Stuart Hall, (1992) reminds us that scientific rationality, “leads not to the ‘emancipation’ which the Enlightenment hoped for, but to ‘a senseless hustle in the service of worthless,…self contradictory,… antagonistic ends’” (Hall, 1992, p 256). Perhaps this best sums up positivist thinking and practice.

The value of Prevent

A decade of Prevent has seen its expansion both in terms of the various sections of society brought under its preview as well as its export to other countries. The decade has also brought an entrenchment of Prevent, as criticisms of the programme are dismissed as matters of empirical detail: faulty delivery, over-zealous implementation etc. This paper demonstrates that the problems with Prevent are conceptually deeper. By reading Prevent in comparison to Lombroso’s studies in criminology, it is possible to place Prevent where it belongs—in a history of racial science. What we see in Prevent is the persistence of positivist ideas used to aimlessly classify and categorise particular individuals and groups in the pursuit of countering extremism. As this paper has illustrated, there appears to be an overwhelming lack of consistency to such classifications and categorisations, which seem to confirm the parameters of positivism. The common feature grouping together Lombroso’s account of criminal types; eugenics and Nazi racial theory; muggings and stop and search; and extremism and the Prevent strategy, is the idea of a fixed racial/religious/cultural essence.

In the same way that the race-thinking behind Lombroso’s criminal types ultimately failed, those trying to get to the root cause of extremism through Islamophobic, positivist frameworks, are also destined to fail, because perhaps the answer to this puzzle is that there is no ‘essence’ to be found. These approaches are guilty of reproducing and reinforcing biological, scientific, and cultural forms of racism, while, at the same time, dismissing and failing to identify the contingent and political nature of the process. In its current form, Prevent can only offer us an ever-lengthening index of key features on how to spot an extremist (and not much more). It might therefore be more useful to instead dismantle these typologies since there is no radical ‘gene’ to be discovered and eliminated. Such a lens limits us to seeing crime, deviance, and political violence as merely a pathology to be eradicated (Young, 2011, p 199). By understanding radicalisation in these frameworks all that can be done is to furnish more descriptions to a never-ending list, which appears to be spiralling out of control.

Lombroso was widely discredited because people began to realise that his positivist measurements didn’t quite work for understanding crime. Perhaps then, Western societies might also realise that his measurements do not work for understanding radicalisation either. Ultimately, when political and social actors grasp that these issues cannot be measured according to skull sizes, nose shapes, beards, veils, and culture clashes, we might begin thinking sensibly about how we are to change the system as opposed to the individual. That is, understanding crime cannot simply be reduced to uncovering quantifiable patterns of criminal activity. Its analysis requires us to engage with historical and social contexts, and it is precisely these contexts that positivism denies in its quest for a universal understanding of criminals and criminality unencumbered (Walklate, 1998, p 3). If we are serious about challenging discourses on radicalisation, it would be more useful to critique those societies which are at the forefront of waging illegal wars, abusing human rights and breaching the rule of law (Wight, 2015, p 212). In other words, it is perhaps less about culture, biology or science, and more about the political.

Data availability

All data generated or analysed during this study are included in this published article.