Over the course of the past decade, “radicalization” has become prevalent as an analytical paradigm to interpret and explain phenomena of political violence, notably in research on jihadist terrorism and Western “foreign fighters” in Syria and Iraq. The concept rose to prominence in the wake of so-called “home-grown” terrorism in Western countries [Neumann Reference Neumann2008: 3; Sedwick 2010: 480; Kudnani Reference Kudnani2012: 4-7; Schmid Reference Schmid2013: 1; Crone and Harrow Reference Crone and Harrow2011: 522-524; Crone Reference Crone2016: 589]. Researchers, policy-makers, and the public had to come to terms with the fact that the perpetrators of these atrocities were young men who had been born and brought up in Europe, which meant that jihadist terrorism could no longer be conceived of as an “external” threat. Thus, the puzzle of how these attacks could have occurred was increasingly re-framed in the form of the question: “how did seemingly ordinary young men become radicalized?”
Thereby, as some have argued, the notion of “radicalization” provided a welcome opportunity to address the “roots of terrorism,” which had become difficult in the political climate after 9/11 [Neumann Reference Neumann2008: 4; Schmid Reference Schmid2013: 2]. Yet, it also significantly re-shaped the way in which phenomena of political violence were analyzed and explained. In contrast to the term’s earlier use in research on political violence and social movements, which had emphasized relational dynamics in processes of escalation at the collective level, “radicalization” came to be understood predominantly as the gradual adoption of “extremist” ideas that promote and eventually lead to acts of terrorism, thus focusing attention on processes of cognitive and ideological transformation, mainly at the individual level.
This shift in perspective did not go unchallenged. In recent years, a number of authors have begun to chart and assess the emerging field of “radicalization-studies,” questioning the notion of radicalization as primarily a cognitive and ideological process [Horgan Reference Horgan2008a; Bjorgo and Horgan Reference Bjorgo and Horgan2009: 3-5; Borum Reference Borum2011b: 8], criticizing its excessive focus on the individual level of analysis and its tendency to de-contextualize the phenomenon, and calling for greater attention to be given to the meso-level of radical movements and milieus and the role of the wider societal and political environment [Horgan 2008: 81; Sedgwick Reference Sedgwick2010: 480; Kudnani Reference Kudnani2012: 5; Schmid Reference Schmid2013: 3-4; Malthaner and Waldmann Reference Malthaner and Waldmann2014; Crone Reference Crone2016].
The purpose of this article
The purpose of this article is to examine some of the main strands of development in research on radicalization, focusing, in particular, on the evolution of the concept as an analytical paradigm and the way it has been applied in empirical studies, with reference to and within the context of broader sociological research on political violence. After discussing, in the remainder of this introduction, the main conceptual fault-lines, I review, in the second section, the way in which the term “radicalization” has been introduced in earlier studies on social movements and political violence and further developed in a recently emerging literature at the intersection of these fields. The third section, then, presents some of the main strands of research on jihadist radicalization in more detail before returning to the critical debates mentioned above and reviewing recently emerging sub-fields of research, such as works on places of radicalization and radical milieus, the radicalization of terrorist lone actors, and pathways of jihadist “foreign fighters.”
Two things need to be made clear at the outset. Firstly, the focus of this paper rests on violent radicalization, that is, processes of radicalization connected to phenomena of political violence, rather than mere political radicalization in the sense of shifts towards more “radical” political beliefs or demands. Secondly, while seeking to provide an overview of the major lines of research and theoretical perspectives, I should emphasize that my aim is not to provide a comprehensive account of the literature on radicalization, which, given the size of the field, would be beyond the scope of this article. Footnote 1
Conceptual fault-lines
“Radicalization,” as various reviews of the literature have noted, is a contested concept that has been conceived of in very different ways and is often used in a vague and ill-defined manner [see inter alia Schmid Reference Schmid2013: 5-6]. Moreover, who or what is defined as “radical” necessarily depends on an itself problematic notion of what is “normal,” “moderate,” or “mainstream” as a point of reference and has changed considerably over time [Sedgwick Reference Sedgwick2010]. While specific definitions used in the various lines of research will be presented below, the aim of this section is to identify some of the main conceptual fault-lines that lie beneath differences in analytical perspectives as well as theoretical debates in the field.
What seems clear is that radicalization refers to a process of “becoming more radical.” While semantically the term can also refer to situations, in research on political violence it is mainly used with reference to actors and/or forms of action (movements or groups “become radicalized,” or we observe a “radicalization of forms of protest”), whereas the term “escalation” is used to refer to situations and patterns of interactions (i.e. confrontations “escalate”). Thus, conceptual fault-lines concern, firstly, the question: what are the nature and end-point processes of radicalization? Thereby, radicalization as a social process—a sequence of happenings unfolding over time, bound together either by a particular type of change produced or via a particular class of events [see Bosi, Demetriou and Malthaner 2014: 4]—can be conceived of in terms of two kinds of change: (1) the transformation of aims, attitudes, and perceptions, or (2) changes in forms of activism and action, or both. In other words, radicalization can be understood of as the radicalization of beliefs (i.e. the adoption of an extremist ideology) and/or as the radicalization of behavior; that is, shifts towards violence [Della Porta and LaFree Reference Della Porta and LaFree2012]. Secondly, the concept may refer to different kinds of actors, individual or collective (groups, movements), as well as to one or to both sides in a conflict (oppositional groups and state actors). And, finally, radicalization can be conceived of as a process within different kinds of conflicts or social situations, such as, for example, radicalization in the context of inter-group conflict, or radicalization in the context of social movements and escalating protests. These questions, as I will try to show, are not merely concerned with delimiting the concept’s empirical referent, but touch upon the epistemological foundations of the way processes of radicalization and the emergence of violence are analyzed and explained.
The escalation of conflict and individual pathways towards high-risk activism: the concept of radicalization in research on social movements and political violence
One of the most notable developments in research on political violence during the past decade was the increasing influence of theoretical approaches from social movement studies, which, paralleled by a growing interest in political violence among social movement scholars, expanded into a distinct strand of literature at the intersection of both fields [see for example Alimi Reference Alimi2011; Alimi, Demetriou and Bosi 2012; 2015; Bosi, Demetriou and Malthaner Reference Bosi, Demetriou and Malthaner2014; Della Porta 1992; 1995; 2013; Fillieule Reference Fillieule2005; Reference Fillieule2010; Reference Fillieule, Klandermans and van Stralen2015; Goodwin Reference Goodwin1997; Gunning Reference Gunning2007; Reference Gunning, Jackson, Smyth and Gunning2009; Hafez Reference Hafez2004; Hegghammer Reference Hegghammer2010; Malthaner Reference Malthaner2011; Tilly Reference Tilly2003; Reference Tilly2004; Wiktorowicz Reference Wiktorowicz2004; Reference Wiktorowicz2005; Wood Reference Wood2003]. Its influence on “mainstream”-research on radicalization has been considerable, discernible, for example, in theoretical elements such as the role of pre-existing personal ties in processes of mobilization that have been selectively adopted into models of jihadist radicalization [Sageman Reference Sageman2004; Reference Sageman2008; Wiktorowicz Reference Wiktorowicz2005; see below]. Yet, this literature also represents a distinct approach to radicalization in its own right. In fact, the term “radicalization” emerged, as Della Porta and LaFree note [2012: 6], in earlier research on violence in the context of social movements of the 1960s and 70s to emphasize the dynamic and interactive nature of these processes [Della Porta and Tarrow Reference Della Porta and Tarrow1986; Della Porta Reference Della Porta1995]. Thereby, radicalization is conceived of as a shift towards more violent forms of action, Footnote 2 analyzed in particular at the level of groups or movements but also with respect to individual trajectories towards militant activism, and embedded in its social and political context as well as broader processes of contention. While this line of research has its origin in the field of social movement studies, it is important to note similar developments in earlier research on terrorism and political violence, with respect to an emphasis on processes and relational dynamics [Neidhardt Reference Neidhardt, Von Alemann and Thurn1981; Reference Neidhardt, von Baeyer-Katte, Claessens, Feger and Neidhardt1982], social context [Waldmann Reference Waldmann1992; Reference Waldmann1998; Crenshaw Reference Crenshaw and Crenshaw1995], as well as explanatory frameworks that integrate different levels of analysis [Crenshaw Reference Crenshaw1981].
Radicalization at the collective level: relational dynamics and processes of escalation
In her seminal work on political violence in the context of social movements, Della Porta dismisses explanations of violence as an effect of economic, social, or political structural conditions as well as explanations focusing on particular ideological characteristics of political organizations. While systemic explanations are unable to account for the behavior of small political organizations, ideological ones rely on simplistic assumptions about the relationship between aims and chosen means [Della Porta Reference Della Porta1995: 5-7]. Instead, what she found in her comparative study of the leftist movements in Italy and Germany was that violence emerged as the outcome of a process of interaction between social movements and their opponents, and in reaction to particular patterns of protest policing and repression [1995: 7, 57, 81-82]. Footnote 3 What makes her work particularly interesting for the study of radicalization, then, is that she links these dynamics of escalation with patterns of organizational radicalization, pointing at the ways in which “environmental conditions triggered organizational processes that in turn favored the diffusion of violence” [Della Porta Reference Della Porta1995: 83]. In the case of the Red Army Faction, for example, the radicalization of small networks concurred with the overall decline of mobilization and was reinforced by the transformation of available resources that favored militant forms of action and processes of isolation with their social environment [Della Porta Reference Della Porta1995: 89-110]. With reference to Neidhardt [Reference Neidhardt, Von Alemann and Thurn1981], she emphasizes the self-reinforcing nature of these processes of escalation and radicalization, which often become “vicious circles,” in which “spirals of negative feedback” in sequences of actions and reactions produce unintended effects and replace the actors’ original objectives [Della Porta Reference Della Porta1995: 111; Neidhardt Reference Neidhardt, Von Alemann and Thurn1981: 244-245]. What is crucial for developing a more precise understanding of radicalization, thereby, is that this process also entails—and is reinforced by—a shift in frames of interpretations, expectations, and perceptions of the “enemy”; that is, a cognitive dynamic of radicalization [Neidhardt Reference Neidhardt, Von Alemann and Thurn1981: 248; Della Porta Reference Della Porta1995: 136].
This relational approach to collective radicalization and escalation has been further developed, with comparative research covering more diverse sets of cases (including militant Islamist groups) and, in particular, with the growing influence of the contentious politics paradigm pioneered by McAdam, Tarrow, and Tilly [Reference McAdam, Tarrow and Tilly2001], from which a more systematic understanding of processes and mechanisms was adopted [see i.a Alimi Reference Alimi2011; Alimi, Bosi and Demetriou 2012; 2015; Bosi, Demetriou and Malthaner Reference Bosi, Demetriou and Malthaner2014; Della Porta 2008; 2013; 2014; Zwerman, Steinhoff and Della Porta Reference Zwerman, Steinhoff and Della Porta2000; Zwerman and Steinhoff Reference Zwerman, Steinhoff, Davenport, Johnston and Mueller2005]. The basic idea behind this line of research was to identify recurring causal mechanisms which, in varying combinations and with context-specific outcomes, shape processes of radicalization and escalation; an ambition that was underlined by the broader scope of comparative research. In her own more recent work on clandestine political violence, Della Porta distinguishes and examines in detail several mechanisms of radicalization at the collective level, including Escalating Policing, which refers to the reciprocal adaptation of forms of policing and repertoires of protest, and Competitive Escalation, as a mechanism that emerges from a dynamic of competition between different groups within the same movement. To that she adds mechanisms that account for the transformation of radicalizing groups, such as Organizational Compartmentalization, which refers to patterns of increasing social isolation and detachment, and Ideological Encapsulation, as a cognitive dynamic that triggers a shift towards more exclusive ideological frameworks [Della Porta Reference Della Porta2013: 67-69, 74-76, 146-152, 176-178, 206-209]. Similar efforts have been undertaken by Alimi, Bosi, and Demetriou, who specify a set of recurring mechanisms as well as three corresponding arenas of interaction as the specific relational contexts from which these mechanisms emerge (2012; 2015); Footnote 4 as well as by Bosi, Demetriou and Malthaner [Reference Bosi, Demetriou and Malthaner2014], who propose a framework of four types of dynamics that shape processes of radicalization and escalation (dynamics of interaction between oppositional movements and the state, dynamics of intra-movement competition, dynamics of meaning formation and transformation, and dynamics of diffusion).
In sum, what characterizes the perspective on (collective) radicalization in this line of research at the intersection of social movement studies and political violence studies is, firstly, its emphasis on relational dynamics in explaining violence, combined with an understanding of violence as emergent. What this implies is, secondly, the analytical embedding of radical movements and militant groups within a broader relational field of actors involved in political conflict. Thirdly, radicalization is understood also to imply a process of cognitive transformation, without, however, considering both to be identical. Patterns of interaction with opponents and rivals are seen as producing—and as being reinforced by—shifts in perceptions and beliefs. In fact, processes of radicalization are understood to result from the interaction of environmental, cognitive, and relational mechanisms [McAdam, Tarrow and Tilly Reference McAdam, Tarrow and Tilly2001: 25-28; Tilly Reference Tilly2003: 7-9; Alimi, Demetriou and Bosi Reference Alimi, Demetriou and Bosi2015: 15], although priority is certainly given to the way in which relational dynamics shape and put into effect cognitive mechanisms. Thereby, cognitive processes are often analyzed on the basis of the concept of (interpretative) frames and collective action frames as well as processes of frame-alignment between a movement and its potential followers (and frame resonance) [Della Porta Reference Della Porta2013: 18; Bosi, Demetriou and Malthaner Reference Bosi, Demetriou and Malthaner2014: 11-15; Snow et al. 1986; Snow and Byrd Reference Snow and Byrd2007]. Another concept used to capture the cognitive aspect of radicalization is identity—as a transformation of notions of collective identity and as processes of identification [Della Porta Reference Della Porta1995: 179-181; Cross and Snow Reference Cross and Snow2011: 118-119], which, for example, van Stekelenburg and Klandermans place at the center of their conceptualization of radicalization in the context of inter-group conflict [van Stekelenburg and Klandermans Reference Van Stekelenburg and Klandermans2010/Reference Van Stekelenburg, Klandermans, Azzi, Chryssochoou, Klandermans and Simon2011]. Finally, beyond the broader relational fields and arenas of interaction mentioned above, the works in this line of research also pay particular attention to social environments and spatial settings, such as micro-mobilization settings in the form of countercultural milieus, radical milieus, or radical networks emerging at the fringes of social movements as well as movement safe spaces [Della Porta Reference Della Porta1995: 136-137; 149-151; Malthaner Reference Malthaner2014; Cross and Snow Reference Cross and Snow2011; Waldmann and Malthaner 2012; 2014]. Free spaces or safe spaces denote small-scale settings within a community or movement that are to some extent removed from the control of authorities or opponents, and play a crucial role in allowing for movement activities that generate the cultural challenge that precedes or accompanies political mobilization and facilitate the formation of mobilizing networks [see i.a. Polletta Reference Polletta1999]. For radical activists, as Cross and Snow argue, free spaces are particularly important:
Free spaces, particularly those embedded in other activists’ spaces, where they are welcome, or at the very least tolerated, give radicals places where they can engage in radical identity work, meet like-minded activists, and even do some limited planning of radical actions [Cross and Snow Reference Cross and Snow2011: 119].
Individual pathways of radicalization
Individual pathways towards militant activism have been a particular concern of scholars studying social movements and political violence. The starting point is their shared skepticism towards approaches that explain individual participation in protest and political violence as a result of psychological or social “pathologies” [McAdam Reference McAdam1986: 65; Snow, Zurcher and Ekland-Olson Reference Snow, Zurcher and Ekland-Olson1980: 789; Della Porta Reference Della Porta and Della Porta1992: 6-7; Diani Reference Diani, Snow, Della Porta, Klandermans and McAdam2013]. Instead, this line of research emphasizes the role of social ties and interpersonal processes—in other words, the relational dimension of radicalization—as well as the way they shape and interact with cognitive processes. Moreover, individual trajectories are linked to social context, in the form of radical networks and milieus as “micromobilization-settings” as well as by examining the effects of broader processes of escalation on individual motivations and pathways. Thus, while presented here as a separate topic of research, the analysis of individual pathways is often closely linked to research on collective radicalization and, in some cases, integrated in multi-level analytical frameworks [Della Porta 1995; 2013].
The most well-known theoretical element of this literature is probably the notion of mobilization via pre-existing social ties. It refers to the consistent finding that participation in (or recruitment into) movements is often initiated via personal (friendship or kinship) ties to activists that precede involvement [inter alia Snow et al. 1980; McAdam Reference McAdam1986; Della Porta Reference Della Porta and Della Porta1992: 8; McAdam and Paulsen Reference McAdam and Paulsen1993; Passy Reference Passy, Diani and McAdam2003; Diani Reference Diani, Snow, Della Porta, Klandermans and McAdam2013]. Beyond facilitating initial contact to movements, interpersonal networks are also important at later stages of the process. Particularly with respect to joining armed groups, the more relevant networks often are not pre-existing ones but those formed in action, that is, militant networks [Della Porta Reference Della Porta2013; Viterna Reference Viterna2006/Reference Viterna2013]. Close personal bonds formed in subcultural milieus and radical networks are a powerful inducement to participation in militant forms of protest and generate personal trust and loyalty that sustain commitment under pressure [Snow, Zurcher and Eckland-Olson Reference Snow, Zurcher and Ekland-Olson1980; McAdam Reference McAdam1986; Della Porta Reference Della Porta and Della Porta1992]. “Low-risk” activism in broader social movement thus can contribute to “paving the way” towards “high-risk” forms of activism as it provides tentative, safe forays into new roles, facilitates connections to other activists, and entails processes of socialization that contribute to the adoption of perceptions, attitudes, values, and identities [McAdam Reference McAdam1986: 69-70]. In other words, cognitive radicalization, from this perspective, is intimately linked to social processes of dense interaction in radical networks and groups. In fact, while ideological affinity is not irrelevant, attitudes and motivations for participation are seen not as pre-existing but as formed in the process, or, as Snow, Zurcher and Eckland-Olson put it: “the ‘whys’ or ‘reasons’ for joining arise out of the recruitment itself” [1980: 799]. Comparative studies found social ties and personal networks to be crucial in very different forms of militant activism, including left-wing as well as ethno-nationalist and religious movements, although radical networks and milieus can take very different forms and consequently shape individual pathways in very different ways [Bosi and Della Porta Reference Bosi and Della Porta2012; Della Porta Reference Della Porta2013; Malthaner and Waldmann Reference Malthaner and Waldmann2012; 2014; Waldmann 1993]. A number of works have also identified varieties of individual motivations (i.e. ideological, instrumental, and solidaristic) and pathways among the members of particular groups [Bosi and Della Porta Reference Bosi and Della Porta2012; see also Dorronsoro and Grojean Reference Dorronsoro and Grojean2004], as well as particular motivations and trajectories of female activists [i.e. Passerini Reference Passerini and Della Porta1992; Viterna Reference Viterna2013].
Yet, as Della Porta makes clear, personal networks alone cannot explain radicalization: “Focusing on affective ties provides only a partial explanation of individual motivations, since they cannot account for the specific form that social networks take” [Della Porta Reference Della Porta and Della Porta1992: 10-11]. In addition, she argues, we need to examine “the environmental conditions that make an individual receptive to the use of political violence” [Della Porta Reference Della Porta and Della Porta1992: 11], in particular the effects of repression and confrontations with countermovements on individuals and their experience of activism [Della Porta Reference Della Porta1995: 161-162]. Confrontation with the police and experiences of persecution and arrests increase solidarity among activists and create powerful motivations in the form of injustice frames and the legitimization of violence as revenge. Thus, rather than reducing radicalization to abstract ideological processes, it is seen as unfolding via the lived experience of activism, which is inseparably connected to broader processes of escalation, as episodes of collective action entail experiences that re-shape perceptions and frames of interpretation.
Research on jihadist terrorism and the evolution of the field of “radicalization studies”
The field of “radicalization studies” that emerged after 9/11 and expanded, in particular, after 2005 was to some extent influenced by research on social movements and political violence but developed a very distinct analytical perspective on radicalization. Its focus rested predominantly on jihadist radicalization and on individual dispositions and trajectories, to some extent embedded in radical networks and group-dynamics, while paying little attention to processes of collective radicalization and their broader political context. The background to this shift was, as mentioned in the introduction, the particular pattern of jihadist militancy in the West. Radicalization, in this case, evolved in local milieus and networks connected to a dispersed transnational movement, and was driven less by spirals of repression and radicalization in direct interaction with Western governments than by outrage over military interventions and the plight of Muslim “brothers and sisters” in countries in the Middle East and Asia. This meant that individual pathways seemed to be linked only indirectly to collective processes of radicalization, and the latter were thus less relevant.
In the following, I will try to chart this field of “radicalization studies” by discussing, firstly, what could be called the “master-narrative” of jihadist radicalization that in one form or another shapes the theoretical foundation on which many empirical studies of the past decade are based. Secondly, I present approaches that have developed subsequently or in parallel, placing greater emphasis on psychological and/or socio-structural and cultural explanations. The following sections, then, address theoretical advancements and conceptual debates before discussing research that focuses on particular aspects such as the spatial and social context of radicalization, lone actor radicalization, and the phenomenon of Western foreign fighters in Syria and Iraq.
Pathways towards home-grown terrorism: the emergence of the “master narrative” of jihadist radicalization
Between 2005 and 2009, a central strand of research on jihadist radicalization emerged in the form of a series of empirical studies that not only were similar in their methodology but also converged in their basic theoretical perspective around what I call the “master-narrative” of radicalization [Bakker Reference Bakker2006, Bokhari, Hegghammer, Lia, Nesser and Tonnessen Reference Bokhari, Hegghammer, Lia, Nesser and Toennessen2006; Gartenstein-Ross and Grossman Reference Gartenstein-Ross and Grossman2009; Nesser Reference Nesser2006; Neumann and Rogers Reference Neumann and Rogers2008; Precht Reference Precht2007; Silber and Bhatt Reference Silber and Bhatt2007; Taarnby Reference Taarnby2005]. Two works, in particular, influenced this line of research: that of Marc Sageman [2004; 2008] and Quintan Wiktorowicz [Reference Wiktorowicz2005]. The impact of Marc Sageman’s seminal work on jihadist networks and processes of radicalization [2004; 2008], thereby, was methodological as much as theoretical. By compiling a large dataset of individuals involved in jihadist terrorist attacks based on news reports and restricted sources, from which he could draw descriptive statistics as well as qualitative insights on specific individuals and groups, Sageman introduced a method that in the following years became a kind of standard-approach to the study of radicalization. In his analytical approach, he particularly emphasizes the role of social ties, small-group-dynamics, and networks, drawing on elements of social movement theory, network analysis, as well as research on new religious movements. In terms of social background, Sageman finds some commonalities within certain “generations” of militant jihadists. Whereas jihadists radicalized during the 1990s, such as the cell around Mohammad Atta, often were from religious upper- or middle-class families, well educated, and often married, and had come to European countries to study, the generation of “homegrown” terrorists was made up of second or third generation migrants who were younger, less well educated, and more often had a history of petty crime or gang-membership [Sageman Reference Sageman2004: 69-97; 2008: 48-50, 58-63]. Yet, he argues that because of the problem of specificity, “profiles based on such personal characteristics […] and socioeconomic background are of very little value in identifying true terrorists” [Sageman Reference Sageman2004: 99]. Rather, what is particular about those who eventually join a terrorist group is “that they have made a link to the jihad” [2004: 99]. Therefore, he argues, pre-existing friendship or kinship-ties are crucial in connecting individuals to radical networks, which (combined) he found to be of relevance in 75 % of his sample [2004: 111]. The process of radicalization itself, then, takes place as a group-process, within cliques of friends (“bunches of guys”) who form strong bonds and generate small-group dynamics that transform individual perceptions and values [2004: 115, 154-155; 2008: 69, 86-87, 116-117]. Yet, Sageman also specifies several cognitive and ideological “elements” of radicalization that precede or evolve in parallel to an individual’s integration in radical networks: moral outrage at the discrimination or suffering of Muslims, the belief that a war is being waged on Islam, and resonance with personal experiences [2008: 71].
Whereas Sageman refers to it only in passing, Wiktorowicz explicitly builds his qualitative analysis of the al-Muhajiroun movement in the UK on social movement theory. Similar to Sageman he emphasizes social networks and personal relationships as “the social pathways for joining”, arguing that: “attitudinal affinity may predispose an individual to join a movement, but social ties are critical for transforming interest and availability into actual activism” [Wiktorowicz Reference Wiktorowicz2005: 15]. In addition, he points to frame alignment between individuals and movements as a necessary precondition for recruitment and, in particular, stresses processes of socialization within movements [Wiktorowicz Reference Wiktorowicz2005: 16]. With reference to Melucci [1989] he emphasizes the fact that radical movements and milieus create “networks of shared meaning” that shape individuals’ identity, perceptions, and motivations. Thereby, Wiktorowicz identifies characteristic steps in pathways of individual radicalization. At the outset, a personal crisis of some form (failure in educational or professional careers, loss of a family member, experiences of victimization, etc.) produces a cognitive opening that shakes previously held beliefs and can lead to a process of religious seeking. The latter refers to a quest for meaning that is channeled towards religiosity by prior socialization or personal networks [Wiktorowicz Reference Wiktorowicz2005: 20-24; 85-86; 92-94]. Thereby, radical movements can themselves influence this process via outreach activities and “moral shock” tactics, deliberately creating cognitive openings and opportunities for encounters. Interaction with movement activists is then gradually intensified, shifting from lectures and open teaching sessions to closed study groups where socialization towards more radical and political beliefs then takes place, involving strong emotional appeals and religious notions of individual obligation that help to overcome obstacles to high-risk activism [Wiktorowicz Reference Wiktorowicz2005: 24-25, 176-182, 208; see also Dalgaard-Nielsen Reference Dalgaard-Nielsen2010: 801-803].
Subsequent empirical research on jihadist radicalization along these lines was typically based on datasets of jihadist militants or clusters of activists in Europe and the U.S., with very few in-depth studies on specific milieus or networks. Largely confirming Sageman’s findings with respect to the general pattern in socio-economic backgrounds of the most recent “generation” of jihadist activists, many studies identify personal crises, identity crises, and frustrations at the outset of a process of radicalization that is then shaped, in particular, by social bonds and group-dynamics in small cliques of friends [i.e. Bakker Reference Bakker2006; Bokhari et al. 2006; Nesser Reference Nesser2006; Neumann and Rogers Reference Neumann and Rogers2008; Precht Reference Precht2007; Silber and Bhatt Reference Silber and Bhatt2007]. The ideological and behavioral transformation within this process was then captured more systematically in various “phase models”, for example by Silber and Bhatt who identified subsequent stages of (1) “Pre-Radicalization” (vulnerability), (2) “Self-Identification” (cognitive opening, religious seeking, contact to like-minded individuals), (3) “Indoctrination” (progressive intensification of beliefs within group/milieu), and (4) “Jihadization” (progression to violent action) [2007: 6-7; also Neumann and Rogers 2007: 43; Precht Reference Precht2007: 32-37]. Another way in which authors sought to systematize causal analysis was to specify different types of factors shaping the process of radicalization, as, for example Precht, who distinguished between background factors, such as identity crisis and personal trauma; trigger factors, such as political events; and opportunity factors, such as certain environments and meeting-places [Precht Reference Precht2007: 6]. Further developing this line of research, a number of studies contributed important insights on settings and places of radicalization, the role of senior activists, mentor-figures, and other “recruitment-agents,” as well as on variations in motivations and pathways of radicalization [Bokhari et al. 2006; Neumann and Rogers 2007: 19-25, 36-8; Precht Reference Precht2007: 60-65; Slootman and Tillie Reference Slootman and Tillie2006: 90]. Nesser, in particular, shows that members of jihadist groups play different roles, which he specified as the “entrepreneur,” his “protégé,” “misfits,” and “drifters,” which correspond to different patterns of radicalization Nesser Reference Nesser2006]: “The entrepreneur and the protégé are often religiously devout idealists who appear to join through intellectual processes and appear to be driven mainly by political grievances and a call for social justice. Misfits appear to join cells mainly to deal with personal problems or out of loyalty to other cell members, whereas the drifters join a cell more unconsciously, through their social networks” [Nesser in Bokhari et al. Reference Bokhari, Hegghammer, Lia, Nesser and Toennessen2006: 11-12].
In sum, a core strand of empirical research on jihadist radicalization that emerged after 2005 to some extent converged around an analytical perspective that conceived of radicalization as a process triggered by a personal crisis, facilitated by (pre-existing) personal ties, and driven by dynamics within small groups of friends. It is important to note that, notwithstanding its emphasis on personal ties and group-dynamics, the analytical focus of this perspective rests squarely on the individual level, and on the way in which individuals are transformed as a result of their “exposure” to certain radical environments. Thereby it has, as Crone put it, an “intellectualist” (cognitive) bias [Crone Reference Crone2016: 604], meaning that the process of radicalization is conceived of primarily as a process of cognitive and ideological transformation with the implicit or explicit assumption that radical beliefs (at least in some cases) somehow lead to violent action. Footnote 5 The explanatory logic of this perspective with respect to the emergence of violence thus revolves around a notion of “propensity” to engage in violence, as a particular cognitive-ideological state of an individual (the state of “having become radicalized”) [see inter alia Moghaddam Reference Moghaddam2005: 161; Bouhana and Wikström Reference Bouhana and Wikström2011].
Socio-structural conditions and individual vulnerabilities to radicalization
Studies on radicalization and terrorism from a psychological perspective have, again and again, pointed out the lack of empirical evidence for any notion that perpetrators of political violence are psychologically abnormal, or that any particular type of “terrorist personality” exists, focusing, instead on risk- and background-factors that can under certain circumstances make individuals vulnerable to radicalization [inter alia Borum Reference Borum2014; Horgan 2005; Reference Horgan2008a; Silke Reference Silke2008; Taylor and Horgan Reference Taylor and John2006]. These factors, however, are often either rather general social characteristics (such as age and gender), or contingent upon an individual’s personal situation and prior patterns of political beliefs, quite similar to notions of personal crisis mentioned in the studies discussed above [Sageman Reference Sageman2004: 95-98; Silber and Bhatt Reference Silber and Bhatt2007: 6-7; Wiktorowicz Reference Wiktorowicz2005: 20-21]. Horgan, for example, lists “the presence of some emotional vulnerability, in terms of feelings of anger, alienation […], and disenfranchisement”, dissatisfaction with one’s current activity (including political), identification with victims (personal or vicarious), the belief that engaging in violence against the state is not immoral, and a sense of reward about participating in a movement as predisposing risk-factors [2008a: 84-85; see also Silke Reference Silke2008]. In a more recent article, Borum to some extent returns to the idea of predisposing personalities, focusing on types of “mindsets”, referring also to Adorno and Frenkel-Brunswick’s notion of the “authoritarian personality,” as well as psychological vulnerabilities (need for meaning/identity, need for belonging) and maladaptive cognitive and emotional patterns that can increase the likelihood of involvement with violent extremism [Borum Reference Borum2014]. Of particular relevance, thereby, is the work of Gambetta and Hertog on “engineers of jihad” [Reference Gambetta and Hertog2009; Reference Gambetta and Hertog2016]. Based on a meticulous empirical study on the educational background of militant Islamists and attitudinal profiles connected to certain career-choices, they develop the notion of an “extremist mindset,” characterized by “the tendency to experience disgust,” the need for cognitive closure, and polarizing in-group and out-group distinctions, to explain the overrepresentation of engineering-students among jihadists (and right-wing extremists) across very different societal contexts [2016: 128-134].
Another approach to explaining predisposition to radicalization links individual vulnerability to socio-structural conditions. While economic deprivation and collective grievances are frequently mentioned rather vaguely as relevant “background-factors” for processes of radicalization (for a discussion see Schmid Reference Schmid2013: 2-3, 20-21; Egerton Reference Egerton2011: 36-43), a rather prominent strand of literature has emphasized the particular conditions of second and third generation Muslim migrants in Western countries, and the Muslim “diaspora”, that make them susceptible to violent and “globalized” interpretations of Islam [Khosrokhavar Reference Khosrokhavar2005; Reference Khosrokhavar2009; Reference Khosrokhavar2015; Roy Reference Roy2004; Kepel Reference Kepel2004; see also Sageman Reference Sageman2008; Waldmann Reference Waldmann2009; Dalgaard-Nielsen Reference Dalgaard-Nielsen2010: 799-801]. “Homegrown” jihadists from Europe tend to originate from socio-economically more disadvantaged strata of society, with lower levels of education and higher levels of unemployment and criminality. Yet, it is in particular experiences of discrimination and marginalization in combination with the specific crisis of identity of a young generation of Muslims caught in a situation of cultural disconnection, between the traditionalist perspective of their parents and their Westernized lifestyle, that explains their particular receptivity to powerful notions of belonging and meaning inherent in radical Islamism: “A second and third generation born of Muslim migrants may recast their feeling of being excluded […] Islam is cast as the ‘otherness’ of Europe and thus may be recast as an alternative identity for youngsters in search of a reactive identity” [Roy Reference Roy2004: 45, 118-141]. While many of these studies acknowledge the need to consider the role of social ties and other factors in pathways of radicalization, they insist that contemporary radicalization of young Muslims in Europe cannot be understood without its broader social, cultural and historic context. According to Khosrokhavar, a sociological approach to radicalization needs “to raise the question of the forms of activism within a broader perspective and to analyze the underlying motivations of extremists by inquiring, in particular, into the long-term effects of stigmatization, humiliation, and insidious forms of rejection or exclusion of which disadvantaged populations are the objects in society” [2015: 14].
Theoretical advances and conceptual debates
A number of studies and debates provided, at different points in time, important theoretical impulses towards further developing the field of “radicalization research.” One among these was the work by Horgan and Taylor [Horgan 2005; 2008a; 2009; Taylor and Horgan 2006], who under the programmatic heading “from profiles to pathways” promoted a more explicitly processual understanding of radicalization. While it is true that the idea that “radicalization is a progression which plays out over a period of time and involves different factors and dynamics” had been, in some form or another, widely accepted among scholars on radicalization from relatively early on [Neumann Reference Neumann2013: 874], Horgan and Taylor’s approach contributed to a more nuanced and precise understanding of (individual) trajectories towards terrorist violence. Conceiving of developmental pathways as sequences of transitions, they emphasize the fact that different dynamics but also motivations and perceptions shape different phases of the process, which implies that “answering questions about why people may wish to initially become involved in terrorism may have little bearing on what they do (or are permitted to do) as terrorists or how they actually become engaged in specific terrorist operations” [Horgan Reference Horgan2008a: 81; see Taylor and Horgan 2006: 589-590]. Moreover, pathways of radicalization also involve migration between different roles within a movement or group [Horgan Reference Horgan2008a: 81, 86]. In other words, this perspective, which was further developed in particular by Fillieule, emphasizes that radicalization must not be reduced to a linear development with a specific point of culmination (the violent act), but rather as part of an “activist career”, understood as “a long-lasting social activity articulated by phases of joining, commitment, and defection”; a perspective that was of influence particularly in research on de-radicalization and disengagement [Bjorgo and Horgan Reference Bjorgo and Horgan2009; Horgan Reference Horgan2009; Fillieule 2005; 2010; 2015; Sommier Reference Sommier2012]. Beyond the complexity and non-linearity of trajectories of radicalization, Horgan and Taylor also emphasize their quality as a fundamentally social, context-dependent process, which evolves through experiences of social learning within “communities of practice” [Taylor and Horgan 2006: 589-590; see also Fillieule Reference Fillieule2010; Sommier Reference Sommier2012] and is shaped by the appreciation and legitimacy of radical groups and their violent campaigns within broader (ethnic or religious) communities [Horgan Reference Horgan2008a: 87-88]. Radicalization, thus, is understood as a multidimensional phenomenon, emerging from the interaction of ideological and social dynamics and shaped by individual, meso-organisational, and macro-contextual factors [see Taylor and Horgan 2006: 590-593].
Another approach that contributed to further developing a processual analysis of radicalization was introduced by McCauley and Moskalenko, who, from a background of social-psychology and based on a broadly comparative analysis, identified recurring mechanisms at different levels of analysis that can combine in various ways to shape trajectories towards violence [McCauley and Moskalenko Reference McAuley and Moskalenko2008; 2011]. Distinguishing individual radicalization from group radicalization and “mass radicalization”, they specify, at the individual level, mechanisms such as personal victimization or grievances, loyalty and effective ties (“the power of love”) or dynamics of incrementally intensifying engagement in radical activism (“slippery slope”), as well as the search for risk and status [McCauley and Moskalenko 2008: 418-421; 2011: 31-33, 41-47, 53-54, 62-64]. Another important mechanism of radicalization at the individual level is “unfreezing,” meaning the weakening or loss of social connection, which “can open an individual to new ideas and [a] new identity that may include political radicalization” [McCauley and Moskalenko 2011: 75]. What is remarkable about McCauley and Moskalenko’s work is that they not only embed individual pathways within the context of networks or milieus but also consider collective processes of radicalization, in particular at the group and inter-group level, addressing mechanisms of group-polarization, competition between different groups, and group isolation [2011: 95-148]. Their work converges with experimental socio-psychological research on small-group dynamics of radicalization that studies shifts in attitudes and opinions as a result of in-group deliberation and in interaction with other groups [see i.e. Hogg Reference Hogg, Hogg and Blaylock2012]. Thereby, dynamics of polarization within particular types of groups called “enclaves”––confined groups in which members share ideological beliefs––are of particular relevance and have been studied also with respect to online-interactions [Wojciezak Reference Wojciezak2010].
Some time after but closely connected to this literature, an important conceptual debate emerged over (as its critics saw it) the prevalent notion of radicalization as primarily a cognitive-ideological process and its implicit assumptions about the emergence of violence [see Borum 2011a; 2011b; McCauley and Moskalenko 2014; Neumann Reference Neumann2013]. Warning that this perspective “risks implying that radical beliefs are a proxy––or at least a necessary precursor––for terrorism,” Borum, among others, pointed out that “[m]ost people who hold radical ideas do not engage in terrorism, and many terrorists––even those who claim to a ‘cause’––are not deeply ideological and may not ‘radicalize’ in any traditional sense” [Borum Reference Borum2011b: 8]. Instead, he and other scholars argued in favor of analytically separating––and studying the interaction of––cognitive/ideological and behavioral (violent action) dimensions of radicalization [Horgan 2005; 2008a; 2009; Borum 2011a; 2011b; 2011c; McCauley and Moskalenko 2014]. Borum, for example, distinguished radicalization, as the process of adopting extremist ideologies and beliefs, from action pathways (or pathways of terrorism involvement), as the process of engaging in violent actions [2011a: 2-3, 2011b: 8-9], while McCauley and Moskalenko separated radicalization of opinion from radicalization of action [2014: 70-73]. It is important to emphasize that this conceptual debate is not about different end-points of radicalization (radicalization of means or radicalization of ends) [see Neumann Reference Neumann2013: 874-875], or about the difference between pathways towards (non-militant) political activism versus towards militancy [Barlett and Miller 2012; Moskalenko and McCauley 2009]. Rather, it is about the causal relation between cognitive processes and violence within pathways of violent radicalization, which is central to our understanding of radicalization as an analytical paradigm to explain the emergence of violence. It is certainly true that some of the criticism in that debate is overstated, as hardly any of the more prominent models or theories of radicalization suggests a “‘unidirectional relationship’ between extremist beliefs and terrorism” [Neumann Reference Neumann2013: 880]. Yet, the fact remains that the way they are assumed to causally interact is often not made explicit and has rarely been theorized precisely, and that a tendency exists to consider cognitive radicalization (via the notion of propensity) as a precondition and as somehow “leading to” violence. Further theoretical progress on the causal relationship between radical beliefs and violence, however, has been limited. To some extent, relational approaches offer an alternative perspective by emphasizing that attachment to a political ideology does not necessarily precede joining a militant movement, but is acquired gradually as a result of processes of secondary socialization, and by pointing to the role of small-group dynamics, personal loyalty, and peer pressure in shaping decisions to participate in violent action [Horgan Reference Horgan2008a; Fillieule 2010; 2015; Sommier Reference Sommier2012]. Moreover, Crone proposed that, rather than the other way round, radical beliefs may also be formed as a result of violent actions, and that “violence can, conversely, be a precondition for engaging with extremist ideology” [Crone Reference Crone2016: 592]. This perspective to some extent echoes insights from social movement studies on the transformative effects of activism and the impact of participation in collective action as a lived experience [Della Porta 1992; 1995].
Places and settings of radicalization, radical milieus and radical networks
A recurrent issue of critical debates within recent research on radicalization was the need to embed individual pathways within their social context. While many of the studies mentioned above acknowledged the role of certain social and spatial environments, these remained weakly conceptualized and, due to the methodological focus on individuals, empirically under-researched. Silber and Bhatt, for example, mention “radicalization incubators”, whereby they refer to “venues that provide the extremist fodder or fuel for radicalization”, such as certain mosques [2007: 20]. Neumann and Rogers developed a more nuanced understanding of “recruitment grounds”, that is, the places and settings where recruitment into jihadist networks takes place, distinguishing two different types of settings according to their characteristics and function in processes of radicalization: firstly, places where Muslims meet and congregate, such as mosques or Islamic book shops; and, secondly, places where individuals are particularly vulnerable and potentially receptive to the message of radical movements such as prisons or refugee centres [Neumann and Rogers Reference Neumann and Rogers2008: 19; see also Rabasa and Benard Reference Rabasa and Benard2015]. In some cases, mosques can turn into “recruitment magnets” widely known to harbour radical activists, which therefore attract individuals seeking to connect to these networks, and where targeted recruitment by these radical activists take place. But in other cases, mosques may simply provide the setting for cliques to form, often at the margins of the wider congregation [Neumann and Rogers Reference Neumann and Rogers2008: 20]. Thereby, it is important to note that the role of certain environments such as mosques changes over time and in interaction with the wider social and political context. As a result of increasing surveillance by security services, radical activities around mosques have become less visible and have partly shifted away from mosques towards more private spaces [Neumann and Rogers Reference Neumann and Rogers2008: 22). Radicalization in prisons, thereby, has developed into an important topic of research in its own right [inter alia Khosrokhavar Reference Khosrokhavar2004; Reference Khosrokhavar2013; Beckford and Khosrokhavar Reference Beckford and Khosrokhavar2005; Hannah, Clutterbuck and Rubin Reference Hannah, Clutterbuck and Rubin2008; Trujillo et al. Reference Trujillo, Jordán, Gutiérrez and González-Cabrera2009; Brandon Reference Brandon2009; Neumann Reference Neumann2010]. As Neumann and Rogers point out, “prisons are […] a highly––some would say uniquely––conducive environment for radicalization and recruitment” [2008: 23], due to the fact that they are particularly unsettling environments and contain a population of individuals who are often at a point of crisis or crossroads in their lives, creating “cognitive openings” and the need for support and belonging [Brandon Reference Brandon2009; Neumann and Rogers Reference Neumann and Rogers2008; Hannah, Clutterbuck and Rubin 2008; Rabasa and Benard Reference Rabasa and Benard2015: 112-116].
Whereas a number of empirical case-studies provided valuable empirical insights into the makeup and dynamics of jihadist networks [i.e. Jordan, Manas and Horsburgh Reference Jordan, Manas and Horsburgh2008; Kirby Reference Kirby2007; Schuurman, Eijkman and Bakker Reference Schuurman, Eijkman and Bakker2014], a more focused approach to conceptualizing and analysing the social environment of violent groups and pathways of radicalization was developed by Malthaner and Waldmann based on the concept of radical milieu, which refers to the immediate (formative and supportive) social environment of clandestine groups and can comprise a number of very different settings and places (and spaces) [Malthaner and Waldmann 2012; 2014; Malthaner Reference Malthaner2014]. Drawing inter alia on Della Porta’s work on micro-mobilization settings [2013: 117] they analyse the formation and makeup of radical milieus in the context of social movements as well as at the fringe of religious or ethnic communities, differentiating between different types of milieus (radical subcultures, radical communities, and radical networks) as well as patterns of resulting relations with broader movements and communities. For individual activists, the radical milieu is the formative social environment in which they are socialized and adopt frameworks of interpretation, values and symbols, and in which they share experiences of persecution and violent confrontations. And it includes the social networks and friendship ties that facilitate subsequent pathways into clandestine groups and which emotionally reinforce their commitment and solidarity. Yet, while, on the one hand, representing a milieu that may contribute to processes of radicalization, the radical milieu, on the other hand, also constitutes an environment which, for various reasons, may constrain these pathways by offering, for example, alternative (non-militant) forms of activisms as well as a place to go––a viable “exit”-option––for activists who disengage from violent clandestine groups. Complementary works have further extended and adapted the concept to address, for example, the issue of radical online-milieus [Conway Reference Conway2012].
Recently emerging sub-fields of research: lone actor radicalization and jihadist “foreign fighters”
Two more specialized lines of research have emerged during the past years in reaction to recent phenomena of political violence: research on so-called terrorist “lone actors”, that is, individuals preparing and carrying out attacks on their own, and studies examining pathways of (particularly western) “foreign fighters” in Syria and Iraq.
Research on lone actors (or “lone wolfs”) is quite often comparative, examining jihadist and right-wing extremist (as well as “single-issue”-terrorists). Most studies, thereby, rely on relatively small sets of case-studies based on open-source information to analyse biographical backgrounds and pathways, with some authors developing typologies of lone actors or models of radicalization [i.a. Bakker and de Graaf 2014; McCauley and Mosalenko 2014, Pantucci Reference Pantucci2011, Spaaij Reference Spaaij2010; Reference Spaaij2012; Gartenstein-Ross Reference Gartenstein-Ross2014, Van Buuren and De Graaf Reference Van Buuren and De Graaf2014]. Among the few studies based on larger-n datasets, the work of Gill and colleagues is the most extensive to date [Gill, Horgan, and Deckert 2012; 2014]. While there is no general socio-demographic profile of lone actor terrorists, there are several marked differences with individuals engaged in other forms of political violence. Gill et al. found them to be on average slightly older, more often socially unattached (nor married and without children), and un- or under-employed relative to their level of education [Gill et al. Reference Gill, Horgan and Deckert2014: 428]. Moreover, a slightly higher-than-average proportion of individuals in their sample had a history of mental illness or personality disorder (31.9 %) [Gill et al. 2014: 428; see also de Roy van Zuijdewijn and Bakker Reference De Roy van Zuijdewijn and Bakker2016: 44], which in many studies has been identified as the most notable common particular characteristic of lone actor terrorists [see also Nesser Reference Nesser2012: 66; Spaaij Reference Spaaij2010: 862]. The phenomenon of lone actors, thus, to some extent exacerbated existing tendencies in the analysis of radicalization, with an even greater emphasis on cognitive-psychological processes at the individual level. Yet, Gill et al.’s research also shows that the notion of lone actors as isolated and mentally disturbed loners is misleading. Firstly, the distribution of personality disorders is very uneven, and far more prominent among right-wing extremists and single-issue terrorists than among jihadist lone actors. Secondly, lone actors are not all that “alone,” but frequently are connected to political movements and radical groups, radical mentors, or virtual communities. Gill et al. found that in almost half of the cases in their sample, lone actors had interacted face-to-face with members of a wider network of political activists, and even more claim to belong to and act as part of a wider movement [2014: 430; see also Nesser Reference Nesser2012: 67-69; Borum, Fein, and Vossekuil Reference Borum, Fein and Vossekuil2012; Berntzen and Sandberg Reference Berntzen and Sandberg2014; Gartenstein-Ross Reference Gartenstein-Ross2014]. Rather than “lone” actors, these individuals appear to be peripheral or former members of political movements and milieus [Malthaner and Lindekilde Reference Malthaner, Lindekilde, Stohl, Burchill and Englund2016; Pitcavage Reference Pitcavage2015]. In a detailed analysis of a medium-N dataset and several in-depth case-studies, Lindekilde, Malthaner and O’Connor found significant variation with respect to the ways in which lone actors interact with other militant activists and radical milieus during the process of radicalization, distinguishing, accordingly several types of lone actors and corresponding pathways of radicalization: peripheral lone actors remain at the margins of radical movements or milieus as a result of the interaction of particular personal traits (withdrawn/indecisive, anti-social, or volatile/unstable personalities) with social dynamics within these settings; whereas (formerly/partially) embedded lone actors become disconnected from radical groups due to external (police persecution) or internal (struggles over leadership) reasons or decide to carry out violent attacks alone because of strategic considerations [Lindekilde, Malthaner and O’Connor forthcoming]. One factor that has received particular consideration in connection with lone actor radicalization is the Internet and the emergence of strategies of propaganda by terrorist organizations aimed at inspiring and instigating terrorist lone actors as part of a strategy of “leaderless resistance” [see Kaplan Reference Kaplan1997, Pantucci Reference Pantucci2011, Weimann Reference Weimann2012; see also Sageman Reference Sageman2008: 122].
The rising numbers of Western foreign fighters joining jihadist groups in Syria and Iraq became a major concern in 2013 and even more so after the rapid expansion of the so-called “Islamic State” in summer 2014 [see inter alia ICSR 2013; Hegghammer Reference Hegghammer2013; Weggemans, Bakker, and Grol Reference Weggemans, Bakker and Grol2014; Lindekilde, Bertelsen and Stohl Reference Lindekilde, Bertelsen and Stohl2016]. Research based on large-N datasets on foreign fighters as well as several in-depth case-studies confirm that basic patterns of radicalization correspond to those identified in jihadist militancy more generally. A study on a total of 677 individuals who left Germany for Syria or Iraq between 2012 and June 2015, for example, found the largest group of foreign fighters to be men in their early 20s, often from migrant families, but emphasized broad variance in terms of personal, ethnic, and educational background. In terms of factors influencing trajectories of radicalization, personal relationships and radical milieus stand out as paramount. 69 % of those who later went to Syria were active members of Salafist milieus, and the combined influence of milieus and family and friendship ties is assessed as relevant in 96 % of all cases. While the Internet plays a role in 42 % of the cases, it does so mainly in combination with social settings and only very rarely on its own [GTAZ 2015]. In other words, similar to processes of radicalization in general, trajectories of foreign fighters are embedded in radical milieus and cliques of friends who radicalize together [see also Weggemans, Bakker and Grol Reference Weggemans, Bakker and Grol2014: 108; Lindekilde, Bertelsen and Stohl Reference Lindekilde, Bertelsen and Stohl2016: 864-865]. Qualitative studies emphasize, among other things, a dynamic of integration in radical networks reinforced by increasing isolation from prior social environments, either in the form of actively cutting prior ties or in the form of the social environment expelling radical activists. They also identify a number of factors that may create vulnerability or predisposition to radical groups, including a strong sense of frustration or traumatic experiences or, as Lindekilde et al. argue from a perspective of life psychology, “experiences threatening life embeddedness” [Lindekilde, Bertelsen and Stohl Reference Lindekilde, Bertelsen and Stohl2016: 861-863; see also Coolsaet Reference Coolsaet2015: 17]. Particular questions addressed in research on foreign fighters include explaining differences in pathways and causal factors between those who leave to become foreign fighters, those who stay behind, and those who carry out attacks in Europe upon their return; and the relatively high rate of individuals with a history in petty crime or gangs has drawn attention to the role of criminal milieus, skills, and personal histories in processes of radicalization [Hegghammer Reference Hegghammer2013; Basra and Neumann Reference Basra and Neumann2016]. Moreover, a number of studies have examined the motives and trajectories of young women travelling to Syria and Iraq, who in the German sample mentioned above represented 15 % of those leaving before June 2014 and 38 % of those leaving between June 2014 and June 2015 [GTAZ 2015]. While women often seem to aspire to roles other than those of active fighters, these studies emphasize their agency and the strength of their ideological beliefs [see i.e. Bakker and de Leede 2015; Kneip Reference Kneip2016; Navest, de Koning and Moors Reference Navest, De Koning and Moors2016; Peresin and Cervone Reference Peresin and Cervone2015].
Conclusions
“Radicalization” has become an analytical paradigm that crucially shapes the way we think about and seek to explain certain forms of political violence. It rose to prominence in response to a very particular phenomenon––that of “homegrown” jihadist terrorism––with a focus on individual pathways of “becoming an extremist” as a process of cognitive-ideological transformation, facilitated and driven by personal networks and small-group dynamics. A look at earlier research on social movements and political violence is instructive in better understanding the shift in analytical focus and explanatory logic that this perspective entails. Social movement studies integrate the analysis of individual pathways and collective processes of escalation with larger processes of contentious politics, conceiving of violence as, in principle, a deliberate choice by militant groups in response to perceived opportunities and constraints and in reaction to their opponents. Footnote 6 In contrast, the notion of radicalization as creating a personal, cognitive-ideological propensity for violence, combined with the idea of susceptibility to radicalization resulting from frustration or identity-crises, tends towards an understanding of violence as individual behavior triggered by certain beliefs and environmental stimuli, rather than deliberate action in the context of a broader political conflict. Yet, insights from episodes of mobilization and repression in, for example, the left-wing movements of the 1970s or Northern Ireland, are not readily transferable to the current phenomenon of transnational jihadist radicalization. And with all its merits, the field of social movement research had little to say about how to study relational dynamics of repression and escalation, and how to integrate the analysis of individual pathways with the broader political conflict, in cases in which radicalization takes place in dispersed clusters of activists and local milieus, loosely connected to a transnational movement, and oriented towards and enraged by armed conflicts in foreign countries which they observe from a distance.
During the past decade, significant theoretical advances have been made with respect to the processual dynamic of radicalization as well as the relation of cognitive radicalization and violent action, as a result of debates and approaches from psychology and social psychology as well as sociological works at the intersection of social movement studies and research on political violence. Thereby, two emerging lines of analysis seem particularly important to further developing the field. The first is concerned with contextualization and embedding individual pathways within micro-mobilization settings as well as broader processes of political conflict, which has been a recurrent concern of authors on all sides of recent debates [Crone Reference Crone2016; Della Porta and LaFree 2012; Horgan 2008; Neumann Reference Neumann2013]. The challenge, here, is to move beyond a static understanding of social environments of radicalization and towards analyzing (and conceptualizing more precisely) the co-evolution of individual pathways and social (i.e. Salafist) movements in interaction with their broader societal and political environment, and to examine the formation and transformation of radical cliques and networks as specific environments of radical micro-mobilization [see inter alia de Koning 2012; Malthaner Reference Malthaner2014]. Secondly, the notion of experiences (in particular experiences of violence and collective action) seems to emerge as a promising alternative way of developing a more precise understanding of the relation between cognitive radicalization, social interactions, and violence [Crone Reference Crone2016]. While analyzing the internalization of radical beliefs as a process of socialization within certain milieus or groups is certainly enlightening, it only partially captures the way perceptions, emotions, and interpretative frameworks shape and are at the same time transformed by violent interactions and the lived experience of militant activism. This involves questioning the idea of cognitive radicalization as the adoption of abstract “teachings”, and focusing on the way in which violence relies upon and activates basic notions of identity as well as perceptions of opponents and audiences, and the transformation of subjectivities and “embodied capacities” [Crone Reference Crone2016: 600].
