(The Role of) Religion in
Radicalisation Research:
Problems, Pitfalls and Possibilities
Written by: Nora Kindermann Student number: 10532005 First reader: Dr. Ulrike Popp-Baier Second reader: Dr. Carolina Ivanescu
MA thesis Religious Studies – Religion and Identity in the Modern World
March, 29th 2020 Word count: 24455
Table of Content
TABLE OF CONTENT ... 1
ABBREVIATIONS ... 3
ABSTRACT ... 4
CHAPTER 1 – INTRODUCTION ... 5
1.1 DEVELOPMENT OF THE RESEARCH ON RADICALISATION ... 5
1.2MODELLING THE PROCESS OF RADICALISATION ... 6
1.3 DEFINITIONAL STRUGGLES AND RELATED PROBLEMS. ... 7
Unclear definitions and their consequences ... 7
Backlash from radicalisation research ... 8
Relation of radical belief and radical action ... 9
1.4 RESEARCH QUESTIONS ... 10
CHAPTER 2 – METHODOLOGY ... 12
THE SAMPLING OF THE LITERATURE ... 12
EXTRACTING THE DATA ... 14
CHAPTER 3 – PSYCHOLOGICAL MODELS DEVELOPED BETWEEN 2005-2019 ... 18
3.1 STAIRCASE MODELS ... 18
Type (a) staircase models ... 19
Type (b) staircase models ... 19
3.2 PYRAMID MODELS ... 20
3.3 SYSTEMATIC MAPPING OF MECHANISMS ... 21
Type (a) systematic mapping of mechanisms ... 22
Type (b) systematic mapping of mechanisms ... 23
3.4 DEFINITIONS OF RADICALISATION ... 23
3.5 CONCLUSION ... 25
CHAPTER 4 – THE ROLE OF RELIGION AND IDEOLOGY IN THE RADICALISATION PROCESS ... 27
4.1 THE ROLE OF IDEOLOGY, WORLDVIEW AND RELIGION IN THE RADICALISATION MODELS ... 27
Type (a) staircase model ... 27
Type (b) staircase models ... 30
Pyramid models ... 32
Type (a) systematic mapping of mechanisms ... 34
Type (b) systematic mapping of mechanisms ... 37
4.2 THE ROLE(S) OF IDEOLOGY, WORLDVIEWS AND RELIGION - AN OVERVIEW ... 40
The role of ideology ... 40
The role of worldviews ... 42
The role of religion ... 42
4.3 CONCLUSION ... 46
CHAPTER 5 - ASSUMPTIONS AND BIASES IN THE MODELS ... 48
5.1 THE USE OF EMPIRICAL EVIDENCE IN MODELS THAT DRAW ON FINDINGS FROM OTHER (EMPIRICAL) STUDIES ... 48
5.2 THE USE OF EMPIRICAL EVIDENCE IN MODELS THAT WERE TESTED BY APPLICATION TO CASE STUDIES ... 49
5.3THE USE OF EMPIRICAL EVIDENCE IN MODELS BASED ON THE OWN COLLECTION AND/OR ANALYSIS OF EMPIRICAL EVIDENCE ... 50
5.4 ASSUMPTIONS CONCERNING THE LINK BETWEEN COGNITIVE AND BEHAVIOURAL RADICALISATION ... 50
5.4 CONCLUSION ... 51
CHAPTER 6 – CONCLUSION ... 53
6.1 DISCUSSION ... 53
Definitional issues ... 53
Empirical issues ... 53
The transgression from cognitive and behavioural radicalisation ... 55
6.2 ANSWERING THE RESEARCH QUESTION ... 55
6.3 LIMITATIONS ... 56
BIBLIOGRAPHY ... 57
Abbreviations
ETA - Euskadi Ta Askatasuna
LTTE - Liberation Tigers of Tamil Eelam PIRA - Provisional Irish Republican Army TMT – Terror Management Theory
Abstract
It has been argued that radicalisation research in the twenty-first century has been plagued by various conceptual and empirical problems that allowed for implicit and potentially unfounded assumptions and biases to take hold in the field of radicalisation studies. Conceptual problems concern the definition of the terms ‘radicalisation,’ ‘extremism,’ and ‘terrorism,’ as well as the conflation of cognitive and behavioural radicalisation. Empirical problems concern the overrepresentation of Islamic radicalisation and of violent radicalisation in contemporary radicalisation research.
The lack of conceptual clarity enables and is enforced by the politicisation of the term ‘radicalisation.’ This, in turn, might allow for stigmatisation and discrimination of Muslims. The conflation of radical belief with radical action hinders research on the role that belief and ideology (and religious belief and Salafi-jihadi ideology more specifically) play in the process of radicalisation. A better understanding of the role of (religious) belief in the process of radicalisation is needed as there is no academic consensus on this topic, and as outside academia religion (and especially Islam) is often ascribed a central role in the process of radicalisation.
This thesis discussed the question “What role is ascribed to religion in psychological models of radicalisation that were developed between 2005-2019?” by conducting a systematic literature review of articles developing psychological models of the radicalisation process. When analysing the role of religion in these models, specific attention is paid to definitional concerns, the relation between religion and ideology, between cognitive and behavioural radicalisation, and to the use of (and possible biases in) empirical evidence on radicalisation processes.
Chapter 1 – Introduction
Since the beginning of the twenty-first century political interest in and academic research on radicalisation has exploded. This interest has been prompted by various religiously motivated terrorist attacks in the West in 2001 and 2005. In justifying their attacks, the perpetrators referred to Islam and the dichotomy between the West and Islam (see for example Horgan, 2005; McCauley & Moskalenko, 2017; Wilner & Dubouloz, 2011). This seemingly intimate connection between Islam and terrorism prompted an academic and policy-related debate in the West and on the role of Islam and (religious) belief in the process of radicalisation. As will become clear in the pages that follow, a large amount of assumptions concerning concepts such as ‘radicalisation’ or ‘extremism,’ the role of religious belief and Islam, as well as the connection between cognition and behaviour influence (and arguably hinder) contemporary research on radicalisation.
This thesis systematically reviews several models of the process of radicalisation, paying attention to the assumptions concerning the role of religion in radicalisation and related assumptions. In the rest of this first chapter, an overview of the radicalisation research since the beginning of the twenty-first century is given, including the critical engagement with that research. The first chapter concludes with the formulation of a research question and sub-questions that guided the research and structure the subsequent chapters. Chapter two discusses the methodology of the systematic literature review that was conducted. Chapter three, four and five present the findings of the systematic literature review, thereby providing an overview of the models under review, their dealing with religion, and the presence of assumptions and biases in the models. Chapter six discusses these findings and scrutinises their implications for future of radicalisation research. This thesis concludes by answering the main research question.
1.1 Development of the research on radicalisation
Since 9/11, and especially since 2005, academic interest in the phenomenon of radicalisation and in the use of the term ‘radicalisation’ has grown (Borum, 2017; Githens-Mazer & Lambert, 2010; Pisoiu, 2015; Schmid & Price, 2011; Trip et al., 2019). Since then, radicalisation has been researched in various disciplines, such as criminology, sociology, (social) psychology and psychology of religion (Borum, 2011; 2017; Schuurman & Taylor, 2018).
Due to the amount of publications it is difficult to get an overview of the state-of-the-art in radicalisation research. Several attempts to classify the literature on radicalisation have been made. Pisoiu (2015), for example, classifies the literature according to underlying paradigms: social movement theory; socio-psychological theories focusing on the individuals; and theories focusing on the macro level, for example on diasporas, or cultural and religious specificities. Illustrating the large variety of research on radicalisation from 2005-2011, Schmid and Price (2011), classify the literature according to thematic foci, such as the examination of the vulnerable individuals,1 adventure seeking as a driver of radicalisation, places of radicalisation, triggers of radicalisation, and the roles of ideology and religion.
Malthaner (2017) classifies research on radicalisation into four thematic groups. The first group contains studies from the perspective of Social Movement Theory (SMT). The second group addresses radicalisation at a collective level, highlighting the relational dynamics in the process of radicalisation, as well as the process of escalation between a social movement and its adversary.2 The third group focuses on individual radicalisation pathways, emphasising the role of interpersonal processes, social ties and their interaction with cognitive processes.
1 On the individual level, psychological explanations of radicalisation have first focussed on so-called ‘clinical’
The fourth group focuses on “jihadist radicalization and on individual dispositions and trajectories” (p. 378). These studies pay attention to the embedding of these processes “in radical networks and group-dynamics, while paying little attention to processes of collective radicalization and their broader political context” (p. 378).
According to Malthaner (2017), this last category of studies demarks the field of ‘radicalisation studies’ which has emerged in the aftermath of 9/11 and the 2005 terrorist attacks. This strand of research is characterised by a master narrative which is defined by 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” (p. 382).3 Many of the authors associated with this strand of research present psychological radicalisation models or models with a substantial psychological component.
The observation that there is a consensus to understand radicalisation as process occurring on various levels is confirmed by Neuman (2013) and reflected in Borum (2017)’s classification of radicalisation literature. Borum (2017) classifies research on radicalisation according to the levels on which the radicalisation process has been studied: macro-level, group-level, and individual level. On the macro-level, SMT has tried to account for processes of radicalisation. On a group-level, findings from social psychology have been applied to aid the understanding of the mechanisms underlying radicalisation. Research on the individual-level “explore[s] how otherwise normal mental states and processes - built on characteristic attitudes, dispositions, inclinations, and intentions - might affect a person's propensity for involvement with violent extremist groups and actions” (p. 20).
1.2 Modelling the process of radicalisation
Neuman (2013), Borum (2017) and Malthaner (2017) point out that the radicalisation process can be studied on various levels, and that recently radicalisation scholars have focused on the interaction of processes on various levels. Borum (2017) makes this explicit by stating that a full understanding of radicalisation “should focus on how people come to think what they think, and, ultimately, how they progress – or not – from thinking to action” (Borum, 2017, p. 18). As factors on the micro-, meso- and macro-levels can play a role in that process, a multi-level and multidisciplinary view is needed (Borum, 2017; see also Malthaner, 2017; Horgan & Taylor, 2006 for similar points.) It has been argued that models can provide a multidisciplinary view by integrating processes occurring on various levels during radicalisation (Borum, 2017).
Various integrated models have been developed in the recent years (see below).4 Despite the promising approach of multidisciplinary and integrative modelling of the radicalisation process, the following criticism has been voiced: Borum (2017) and King and Taylor (2011) notice that most models are based on thin empirical foundations because it is difficulty to develop methods for testing models in their entirety. It is more realistic to develop methods that can test individual mechanisms (King & Taylor, 2011; Taylor & Horgan, 2006; see also
3 As will become clear below, many authors argue that the radicalisation process is set in motion by a so-called
trigger event or a sever personal crisis. This can include the experience of individual or group discrimination of the confrontation with death.
4 Borum (2017) points to the existence of common elements in different integrated models. He argues that the
presence of common elements suggests that “researchers may fundamentally agree on the key factors in the radicalization process, but conceptually parse those factors in different ways” (Borum, 2017, p. 24, see also King & Taylor, 2011). King and Taylor (2011), who reviewed 5 influential models of the radicalisation, also point out following commonalities between the models:
1. “[T]he assumption that radicalization is a transformation based on social-psychological processes" (p. 609). 2. The fact that issues such as relative deprivation and identity crisis can lead to cognitive openings.
3. The fact that relative deprivation plays a role in the process.
4. The fact that identity-related issues (such as discrimination or the management of a dual identity) might contribute to the process of radicalisation.
Dawson, 2018). This makes it challenging to empirically test the interaction between the separate mechanisms.
Additionally, there are several problems within the field of radicalisation research that have implication for the construction of models. These problems and their relevance modelling the process of radicalisation are discussed in the next section.
1.3 Definitional struggles and related problems. Unclear definitions and their consequences
Neumann (2013) pointed out that “[a]t the most basic level, radicalization can be defined as the process whereby people become extremists” (p. 874). To date, however, there is no consensus on a definition of radicalisation. As pointed out by Striegher (2015), Borum (2017) and Bötticher (2017), the terms ‘radicalisation,’ ‘violent extremism,’ and ‘terrorism’ lack a clear demarcation and are often used interchangeably. Sedgwick (2010) or Malthaner (2017) also point out that the concept ‘radicalisation’ is often vague, ill-defined, or has a variety of meanings in different contexts. The lack of clear definitions of radicalisation has several negative consequences.
Academic: The lack of a clear conceptual distinction between the term ‘radicalisation’
and related terms such as ‘extremism’ or ‘terrorism’ has profound negative consequences for radicalisation research. Lacking or weak conceptual foundations may impede objective and systematic research (Ackerman & Burnham, 2019). Aly and Striegher (2012), for example, state that “some definitions are so broad as to categorize all opinions that differ from societal norms as radical or extreme. Other definitions conflate radicalisation with a tendency toward or support for the use of violence as a legitimate avenue for achieving stated objectives” (p. 850). Moreover, a lack of solid conceptual foundations hinders the dialogue of researchers from different fields and incites the conceptual lumping together of potentially very different forms of radicalisation. Borum (2017) and Pisoiu (2015) point out the problematic conflation of cognitive and behavioural radicalisation which conceals the distinction between violent and non-violent radicalisation. Malthaner (2017) identifies the following conceptual fault-line in the (ill-defined) concept of ‘radicalisation:’ “what are the nature and end-point processes of radicalization? Thereby, radicalization as a social process (…) 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” (pp. 371-372). Additionally, stronger and more systematic definitions would improve consistency across studies when it comes to operationalising the terms at hand (e.g. for coding) (Ackerman & Burnham, 2019).
Social/political: In addition to having negative academic consequences, the lack of
conceptual clarity has social and political ramifications, as the definition of certain problems “has profound implications for how we understand and address them” (Borum, 2017, p. 17). Regarding the question of what kind of radicalisation is meant, Neumann (2013) shows how understanding radicalisation as cognitive or behavioural radicalisation shapes counter-radicalisation policies. Reflecting on the potentially negative implication of insufficiently distinguishing between the concepts ‘radicalisation’ and ‘terrorism,’ Bötticher (2017) notes: “the danger is that all forms of radical rebellion - even legitimate resistance against corrupt and violent authoritarian regimes - are disqualified as illegitimate extremism. This has the dangerous potential effect of driving pro-democratic radicals into the arms of anti-democratic extremists” (p. 76).
Moreover, the concept ‘radicalisation’ has been criticised for importing political assumptions. Githens-Mazer and Lambert (2010), state that “[r]adicalization is a research topic plagued by assumption and intuition, unhappily dominated by ‘conventional wisdom’ rather than systematic scientific and empirically based research” (Githens-Mazer & Lambert, 2010,
[w]ithout a clear definition of terms and ideas of what we are trying to observe and understand, a ‘we know it when we see it’ approach to understanding radicalization becomes lackadaisical and promotes stereotyping. It justifies a policy-making and media approach to radicalization that promotes emotional or politically driven feelings about who poses a security threat over a scientific, empirically derived from of knowledge and understanding about what this threat actually is or is not. (p. 901)
Githens-Mazer (2012) articulates the above point even more forcefully, arguing that the lack of conceptual clarity turns the concept into a “political machination” (p. 562), and can lead to the furthering of “the social construction of risk attributed to Islam and Muslims in the European Context” (p. 561). The author furthermore warns that any understanding of the concept must be explicitly separated from Islam, and definitions that understand the challenge of the status quo as an essential part of radicalisation must be eyed with caution: “We must consider the collectively defined though individually enacted phenomenon of challenging an existing status quo as an act which has no scientific basis for moral claims as to what constitutes ‘good’ or ‘bad’ or risky or constructive forms of radicalization” (p. 562).
Kundnani (2012) also observes the presence of ideological assumptions in the concept ‘radicalisation,’ a concept which “led to the construction of Muslim populations as ‘suspect communities’” (p. 3). He argues that this is enforced by the fact that research on radicalisation and the field of policy making are heavily intertwined. The author critically notes that some academic work on radicalisation takes over the “ideological assumptions that determine what counts as legitimate and illegitimate within the terms of this discourse” (p. 8) of the consumers of their knowledge. Revisiting Kundnani (2012)’s article, Derek Silva (2018) rearticulate Kundnani (2012)’s criticism. Stating that since 2012 most studies were cultural-psychological and theological-psychological in nature, he argues that
such studies (…) begin by looking at a predetermined problematic group or social location (i.e., Muslim communities or attacks perpetrated by Muslims), and extend their findings to broader populations. Indicative of government (and perhaps social) anxieties surrounding Islam as the primary site of homegrown terrorism in the post-Iraq and Afghanistan ‘war on terror’, research beginning with Muslim communities is at risk of simple confirmation bias. (p. 40; see also Jensen, Atwell Seate & James, 2018, p. 2)
To summarise the above: on the one hand, it has been argued that the intertwining of policy and research, which led to the import of ideological assumptions into radicalisation research, hindered conceptual clarity of the term ‘radicalisation’ in the 21st century. On the other hand, it is precisely that lack of conceptual clarity that allows the import of assumptions. In other words, while the lack of conceptual clarity allows for imported assumptions and confirmation bias, at the same time, these assumptions are partly responsible for the lack of conceptual clarity. It has been argued that the resulting vicious circle promoted stereotyping, especially of Islam.
Backlash from radicalisation research
Stereotyping, besides being problematic in itself, might have an effect on processes of radicalisation: Githens-Mazer and Lambert (2010) point out the stigmatisation of certain Muslim groups which is related to the ‘conventional wisdom’ of radicalisation. Many authors point out that discrimination of Muslims and the alleged overreaction of (Western) governments to (Islamic) terrorist attacks might enhance the process of and conditions for radicalisation (Della Porta, 2013; Hafez & Mullins, 2015; McCauley & Moskalenko, 2011; 2017; Pyszczynski, Motyl & Abdollahi, 2009). As discussed above, Githens-Mazer (2012) points to
the association of risk with Muslims/Islam. It is likely that such associations contribute to islamophobia and negative/biased attitudes towards Muslims. These social conditions and grievances, it has been argued, contribute to/influence the process of radicalisation.
Relation of radical belief and radical action
One assumption in radicalisation research concerns the role of Salafi-jihadist ideology in radicalisation (e.g. Bhatt & Silber, 2007; Wiktorowicz, 2005). This assumption is connected to the assumption that (radical) ideology and (religious) belief precede and are somehow causally connected to (radical) action, an assumption readily made by some researchers (Borum, 2017). It is not clear what role ideology and (religious) belief play in the process of radicalisation into violent extremism/radical action. As noted by McCauley and Moskalenko (2017), research on the relation between thought and action does not support the assumption that radical belief necessarily follows radical action. With regard to the process of radicalisation, it is widely accepted that ideology and (religious) belief is neither necessary nor a sufficient condition for radical action. McCauley and Moskalenko (2017) state that “[r]esearch in social psychology has long established that attitudes do not easily translate into actions” (p. 19). Borum (2017) also notes: “Empirically (…) it is unclear whether systematic and significant differences may exist between the believers and the doers. If such differences do exist, they may involve differing social psychological mechanisms, pertain to individual characteristics (e.g., vulnerabilities and propensities), and/or relate to contrasting life experiences, situations, or group dynamics” (p. 25). Therefore, studying radical belief is important because “many – perhaps most – people with radical ideas and violent justifications do not engage in terrorism” (p. 18). Besides, “many individuals move to jihadist action without jihadist ideas – for personal revenge, status, escape, or love” (McCauley & Moskalenko, 2017, p. 21).
The overrepresentation of violent cases in radicalisation research has hindered our knowledge on the connection between radical belief and radical action. Jensen, Atwell Seate and James (2018) argue that “studies that only examine violent extremists as a sample tend to downplay the complex relationship between belief and behaviour by assuming that one naturally produces the other” (p. 2). This lack of insight in the relation between belief and action is connected to the so-called ‘problem of specificity.’ There are two versions of the problem of specificity: (a) the question why, while so many share the same circumstances, only some radicalise (cognitively and/or behaviourally); and (b) why do only few radicalise behaviourally compared to a large(r) amount of people that are cognitively radicalised (Cassam, 2020, quoted in Sardoc, 2020). According to Dawson (2018 a,b), to solve the problem of specificity concerning the fact that most of the people with radical ideas do not restore to radical action, we must “identify the necessary and sufficient conditions prompting individuals to undertake each type of action, especially violence” (p. 145).
The distinction between radical belief and radical action is important in order to define the phenomenon under research, to assess the role of ideology in general, and Salafi-jihadi ideology specifically (see e.g. Dawson, 2018 a, b; Neumann 2013), and can help to address the problem of specificity. Moreover, this distinction as well as a clarification of the role of ideology and religious belief in the process of radicalisation into violent extremism is important in order to circumvent government’s alleged overreactions to in their counter-radicalisation programmes. As McCauley and Moskalenko (2017) argue, “seeing ideas as the threat leads to a war on ideas. Targeting ideas is likely to produce a backlash” (p. 18).
Many authors eviscerate the role of ideology and (religious) belief in the process to radical action, prioritising other mechanisms instead. Aly and Striegher (2012) as well as Schuurman and Taylor (2018), Sparke (2019) and Verkuyten (2018) argue that the role of religion in radicalisation is less important than is often assumed in policy and research (e.g. in
Mazer and Lambert (2010) explicitly connect the conflation of racial belief and action to the stigmatisation of Islam. They pointed out that, instead of focusing on role of Salafi-Jihadi ideology, one should focus on difference between doers and sayers: “If we truly seek to understand the difference between an individual who will and one who will not commit terrorism, might it in fact be more apposite to start to address the difference between 'doers' and 'sayers', rather than engage in normative debates about the existential threats of Islamic ideas?” (p. 896). Schuurman and Taylor (2018) make a similar point, arguing that to ascribe an important role to religion in the process of radicalisation contributes to the stigmatisation of Muslims and does not help to solve the problem of specificity. McCauley and Moskalenko (2017) add that politics and policy-related definitions of radicalisation often understand terrorist violence as being motivated by ideology. They argue that “[t]he language that focuses on Islamic Extremism as the enemy is a major impediment to the theoretical separation between radicalization in opinion and action” (p. 24).
Criticising the, what he calls, “curious erasure of religion” (and ideology), Dawson (2018 a,b) challenges the above stances. He argues that
[t]here is strong prima facie evidence that religiosity plays a larger role in the radicalization of some jihadists than many researchers think. Challenging the logical and empirical adequacy of some key assumptions of those seeking to downplay the role of ideology and, hence, religion, I argue that we will acquire a better grasp on ‘religious terrorism’ if we give more credence to what jihadists say about their motivations. (Dawson, 2018a, p. 99) Understanding Salafi jihadism as political ideology, Neuman (2013) argues in line with Dawson: “simply put, what makes some individuals restore to political violence while others do not is, in many cases, impossible to understand without looking at the ideological assumptions which they have come to accept and believe in” (p. 881). The arguments made by Crone (2016) are in line with Dawson and Neuman. These authors argue that a better understanding of the role of religion and/or ideology5 might help understand the development of certain behaviours and thereby possibly help to address the problem of specificity.
1.4 Research questions
To summarise the above: in recent years, a processual understanding of radicalisation has become salient, requiring the development of models of radicalisation processes that include mechanisms on the individual-, group-, and societal level. The models that have been developed in the aftermath of 9/11, and especially since 2005, are psychological in nature or include significant psychological components (Malthaner, 2017). This thesis focuses exclusively on psychological models of the process of radicalisation. This focus has been chosen because, according to King and Taylor (2011), insights into the psychology of radicalisation can help to address the problem of specificity concerning the link between cognitive and behavioural radicalisation.
There are several interconnected problems with regard to radicalisation research that affect the development of models of radicalisation. First of all, there are problems concerning the definition and conceptualisation of the term ‘radicalisation.’ The lack of conceptual clarity enables and is enforced by the politicisation of the term. This, in turn, might allow for stigmatisation and discrimination of Muslims, which might contribute to the radicalisation process. Secondly, and connected to the above problems, radical belief is often conflated with radical action. On the one hand, this conflation contributes to the lack of conceptual clarity. On
5 As will be shown below part of the problem lies in the lack of clarity regarding the definition of the terms
the other hand, the conflation of belief and action hinders research on the role that belief and ideology (and religious belief and Salafi-jihadi ideology more specifically) play in the process of radicalisation. A better understanding on the role of (religious) belief in the process of radicalisation is needed as there is no academic consensus on this topic, and as outside academia religion (and especially Islam) is often ascribed a central role in the process of radicalisation.
Taking for granted that, in order to get a full grasp of the process of radicalisation, conceptual models are needed. Addressing the two sets of problems described above, we must ask what role is ascribed to religious belief specifically, and to ideology more generally, in the process of radicalisation, and via which mechanisms religious belief and ideology impact the process of radicalisation. Insights into and an evaluation of the understanding of the relationship between religion/ religious ideology and radicalisation might help to diminish stereotyping of religious groups (especially of Muslims); to sharpen the definition of ‘radicalisation,’ thereby circumventing that academia unintentionally contributes to stigmatisation of and biases towards certain religious groups; and to countering the conflation of radical belief and action. Based on the above discussion, this thesis addresses the following research question:
What role is ascribed to religion in psychological models of radicalisation that were developed between 2005-2019?
1. What psychological models of the process of radicalisation have been developed between 2005-2019?
2. How, if so, do these models address the role of religion in the process of radicalisation? 3. How do (implicit) assumptions and biases feature in the psychological models of the
process of radicalisation developed between 2005-2019?
4. What are the implications of the above findings with regard to the academic research on radicalisation (and religion) developed between 2005-2019?
Chapter 2 – Methodology
In order to answer the main and sub-questions, a systematic literature review (see Xiao & Watson, 2019) has been conducted. The following method was applied in order to make the data acquisition and analysis transparent, minimise bias in the selection of the articles, and ensure repeatability.
The sampling of the literature
In order to find and select relevant articles to include in the literature review, the databases Google Scholar, Web of Science and Scopus were searched. Google Scholar was selected as it shows academic publications from the entire web.6 Web of Science was chosen as it contains publications from various fields of science, such as social sciences and life-science.7 Also Scopus was selected because of its broad database and its advanced search tools.8 Thereby, a broad search was made possible. Only articles published within the timeframe 01-01-2005 until 31-05-2019 (end of data collection) were selected. This timeframe was set because, as stated above, the use of the term ‘radicalisation’ rapidly increased after the terrorist attacks in 2005. The endpoint of data collection was chosen pragmatically. After 31-05-2019 the selected articles were analysed and reviewed.
The following combinations of search terms were used:
(1) Religion AND radicalisation OR radicalization AND “violent extremism” OR terrorism;
(2) Political AND radicalisation OR radicalization AND “violent extremism” OR terrorism AND model;
(3) Radicalisation OR radicalization AND “violent extremism” OR terrorist AND model; (4) Political AND radicalisation OR radicalization AND “violent extremism” OR terrorism. The selection of the above search terms is based on the fact that radicalisation is often defined in terms of violent extremism and terrorism (see chapter one). Moreover, the terms ‘political’ and ‘religion’ were included because this research is interested in radicalisation models in general, and not only models of religious radicalisation.
With regard to the data-bases Web of Science and Scopus, all results were screened to select the articles to be reviewed. The number of search results in Google Scholar exceeded 9000 publications. As it was impossible to scan such a large amount of publications, I skimmed through and made a selection based on the first 750 results per search. This threshold was set as the relevance of the articles rapidly decreased, and the repetition of articles increased after the first 750 results.
In order to determine which articles are included in the review, the following inclusion and exclusion criteria were established.
Inclusion criteria:
- Conceptual (systematic review) articles concerning psychological models of the process of radicalisation. Models were understood in line with Norman (1983): “A conceptual
model is invented to provide an appropriate representation of the target system,
appropriate in the sense of being accurate, consistent, and complete” (p. 7);
- Articles with the goal to develop new psychological models of the process of radicalisation;
- Articles with the goal to further develop existing psychological models of the process of radicalisation;
6https://www.lib.umn.edu/faq/5342
7https://clarivate.libguides.com/webofscienceplatform/coverage 8https://www.elsevier.com/solutions/scopus/how-scopus-works
- Articles published since 2005; - Articles published in English. Exclusion criteria:
- Articles focusing on the topic of countering violent extremism; - Articles focusing on the topic pf preventing violent extremism;
- Articles focusing on deradicalisation (the reason for this is that deradicalisation is often understood as a different process than radicalisation, rather than the reversed process of radicalisation);
- Articles that apply certain models of the process of radicalisation in order to understand specific cases of radicalisation (i.e. articles where the focus is on understanding a specific empirical phenomenon instead of developing or enhancing the understanding of the process of radicalisation);
- Case studies that aim to understand a certain context/specific group/phenomenon of radicalisation specific to a geographic location;
- Articles addressing only specific mechanisms as laid out in models of the process of radicalisation. That is, articles that do not address entire models or aim to understand a certain mechanism within a model;
- Studies focussing solely on empirical evidence and not on the development of models; - So-called grey literature, i.e. articles that have not been peer-reviewed;
- Articles that have been published before 2005;
- Articles that have been published in a language other than English.
Step 1: Based on the above inclusion and exclusion criteria, all titles that were found via the
combinations of search terms were screened. Articles whose relevance could not be determined based on their title, were included in the first selection. These articles were then screened on the basis of their abstracts (and, if necessary, the conclusion section).
Step 2 – backward search: Relevant articles that were cited in the articles selected in step one
were included.
Step 3 – forward search: Relevant articles that cited the articles selected in step one and two
were included (making use of various citation functions in the databases).9 This procedure also helped to include articles that are not published digitally. Backward and forward searches were also based on authors’ names. The search for literature was cyclical as a better understanding of the field of radicalisation studies provided new authors or topics to include in the search. Based on the above search strategy, forty-one articles were selected.
Step 4: These forty-one articles were scan-read again in order to determine whether they
properly satisfy all the above named inclusion criteria. Several of the articles that were included into the literature review at first, were excluded after the first round of analysis. Some of these articles which were excluded they focused on the description of mechanisms instead of the conceptualisation of a consistent and complete model. Furthermore, the initial selection included several articles by the same author (namely Kruglanski) were included. Since the first round of analysis showed that the differences between these articles was small, several of these articles were excluded in order to avoid repetition. Moreover, one article, by Trip et al. (2019), was excluded on the basis of quality concerns.10
The following seventeen articles were included in the final selection and reviewed:
1. Ali, R. B. M., Moss, S. A., Barrelle, K., & Lentini, P. (2017). Does the pursuit of meaning explain the initiation, escalation, and disengagement of violent extremists?. Aggression and violent behavior, 34, pp. 185-192.
2. Borum, R. (2014). Psychological vulnerabilities and propensities for involvement in violent extremism. Behavioral sciences & the law, 32(3), pp. 286-305.
3. Decety, J., Pape, R., & Workman, C. I. (2018). A multilevel social neuroscience perspective on radicalization and terrorism. Social neuroscience, 13(5), pp. 511-529. 4. Doosje, B., Moghaddam, F. M., Kruglanski, A. W., De Wolf, A., Mann, L., & Feddes, A.
R. (2016). Terrorism, radicalization and de-radicalization. Current Opinion in Psychology, 11, pp. 79-84.
5. Ferguson, N., & Binks, E. (2015). Understanding radicalization and engagement in terrorism through religious conversion motifs. Journal of Strategic Security, 8(1-2), pp. 16-26.
6. Horgan, J. (2005). The psychology of terrorism. Routledge.
7. Jensen, M. A., Atwell Seate, A., & James, P. A. (2018). Radicalization to violence: A pathway approach to studying extremism. Terrorism and Political Violence, pp. 1-24. 8. Kruglanski, A. W., & Fishman, S. (2009). Psychological factors in terrorism and
counterterrorism: Individual, group, and organizational levels of analysis. Social Issues and Policy Review, 3(1), pp. 1-44.
9. Kruglanski, A. W., Gelfand, M. J., Bélanger, J. J., Sheveland, A., Hetiarachchi, M., & Gunaratna, R. (2014). The psychology of radicalization and deradicalization: How significance quest impacts violent extremism. Political Psychology, 35, pp. 69-93. 10. McCauley, C., & Moskalenko, S. (2011). Friction: How radicalization happens to them
and us. Oxford: Oxford University Press.
11. McCauley, C., & Moskalenko, S. (2017). Understanding political radicalization: The two-pyramids model. American Psychologist, 72(3), pp. 1-34.
12. Moghaddam, F. M. (2005). The staircase to terrorism: A psychological exploration. American psychologist, 60(2), pp. 161-169.
13. Moghaddam, F. M. (2006). From the terrorists' point of view: What they experience and
why they come to destroy. Praeger Security International.
14. Pyszczynski, T., Motyl, M., & Abdollahi, A. (2009). Righteous violence: killing for God, country, freedom and justice. Behavioral Sciences of Terrorism and Political Aggression, 1(1), pp. 12-39.
15. Taylor, M., & Horgan, J. (2006). A conceptual framework for addressing psychological process in the development of the terrorist. Terrorism and political violence, 18(4), pp. 585-601.
16. Wilner, A. S., & Dubouloz, C. J. (2011). Transformative radicalization: Applying learning theory to Islamist radicalization. Studies in Conflict & Terrorism, 34(5), 418-438. 17. Winter, D. A. & Feixas, G. (2019). Towards a constructivist model of radicalization and
deradicalization: A conceptual and methodological proposal. Frontiers in
psychology, 10(412), pp. 1-11.
Importantly, the two publications by Moghaddam present the same model and were analysed together. The same counts for the two publications of McCauley and Moskalenko. Therefore, a total of fifteen models was analysed.
Extracting the data
As noted by Xiao and Watson (2019), the type as well as the goal of a systematic literature review determines the method and data extraction process. The goal of the present review is
descriptive and analytic, i.e. to describe and analyse the role ascribed to religion in psychological models of radicalisation that were developed between 2005-2019. The aim is to extract different conceptions of the role of religion in the process of radicalisation. The type of systematic literature review most fit for this research is a textual-narrative synthesis. This type of literature review is characterised by having a standard data-extraction format and
often requires studies to be organised into more homogenous subgroups. The synthesis will then compare similarities and differences across studies based on the data that was extracted (Lucas et al. 2007). Because of the standardized coding format, the review may include a quantitative count of studies that has each characteristic (…) and a commentary on the strength of evidence available on the research question. (Lucas et al., 2007 in Xiao & Watson, 2019, p. 95)
The articles selected for review were analysed qualitatively; the technique of ‘thematic analysis’ was applied (see Lucas et al., 2007). The coding was conducted in line with the below themes (a)-(h).
Reference
Ad 1 (a) Summary of model and its mechanisms/components
[Codes: mechanisms, components of the model]
Ad 1 (b) Level of analysis (i.e. are the processes and mechanism located on the level of the individual, the group and/or society?)
[Code: level of analysis of mechanism or component]
Ad 1 (c) Definition of the term ‘radicalisation’
[Code: definition radicalisation, i.e. specification of the conceptual components of the term ‘radicalisation’]
Ad 2 (d) Explicit definition of religion/ideology/worldview or implicit understanding of religion/ideology
[Codes: definition ideology, definition religion, definition worldview]
Ad 2 (e) Role of religion/ideology/worldviews in the model
[Codes: role of religion; role of ideology; role of worldview]
A 2 (f) Addressing the problem of specificity with regard to the relation of cognitive and behavioural radicalisation
[Code: relation belief/action]
Ad 3 (g) Assumptions in the model, such as the assumption that radical belief precedes radical action
[Code: unfounded assumptions; relation belief/action]
Ad 3 (h) Biases reflected in the model, such as the occurrence of confirmation bias
[Codes: data-generation, model-testing, reference to empirical evidence, reference to empirical evidence by others]
After the first round of coding, (a) a short textual description of the article under review was given and the radicalisation model extracted. (b) The levels of analysis of the models were determined (e.g. a psychological model will prioritise mechanisms on an individual level, whereas a social-psychological model might pay more attention to contextual influences of group dynamics). Moreover, (c) definitions of the term ‘radicalisation’ were extracted, in order to determine whether all models focus on the same empirical phenomenon. Insight into (a)-(c) helped to categorise the articles under review. This categorisation can be understood as a first/preliminary synthesis (see Lucas et al. 2007) on the basis of which the first sub-question, - What psychological models of the process of radicalisation have been developed between 2005-2019? – was answered.
In order to answer sub-question two – “How, if so, do these models address the role of religion in the process of radicalisation?” –, (d) the explicit definition of religion/ideology/worldview or the implicit understanding of these terms was extracted, as well as (e) the role that religion/ideology/worldview play in the process of radicalisation. Despite the research question’s focus on the role of religion, the definition/understanding of ideology and worldview was also extracted. This was done due to the close connection between, and sometimes conflation of religion (religious ideology), ideology and worldview within the models.
As will become clear in chapter four, the terms ‘religion’ or ‘religious’ are frequently used but never defined in the reviewed articles. Instead, the authors mostly refer to known religion, such as Islam or Christianity. Moreover, no specific role is ascribed to religion per se. Religions are mostly referred to in order to provide examples when citing empirical evidence. To analyse and compare the roles that religions (can) have in the models under review, Smart (1998)’s understanding of religion in terms of seven dimension was used as an analytical lens. Smart (1998)’s understanding was chosen because it “help[s] to characterize religions as they exist in the world” (p. 21). This approach to religion matches the references to and understandings of religion in the articles under review. Moreover, Smart’s approach in terms of seven dimensions also include secular worldviews.11 Smart claims that there is no fundamental difference between world’s religions and secular/political worldviews and understand both in terms of the seven dimensions. That means that in secular world views as well as religions all seven dimensions are manifested, even though in various degrees. As will become clear below, the term ‘worldview’ is used by several authors under review and can be included in the analysis by applying Smart’s seven dimensions.
The seven dimensions are: (1) the practical and ritual dimension, referring to prayers, preaching, etc.; (2) the experiential and emotional dimension referring to the feelings generate by religious practices and rituals; (3) the narrative and mythic dimension referring to the stories about founders, evil, the here-after, etc.; (4) the doctrinal and philosophical dimension underlying the narrative and mythic dimension; (5) the ethical and legal dimension concerning religious and political values in society, as well as the individual’s moral life; (6) the social and institutional dimension referring to the fact that religion is practices by groups and relies to a certain degree of institutionalisation including formal religious officials; and (7) the material
11 Smart (1998) states: “It is important for us to recognize secular ideologies as part of the story of human
worldviews. It is artificial to divide them too sharply from religions, partly because they sometimes function in society like religions, and partly because the distinction between religious and secular beliefs and practices is a modern Western one and does not represent the way in which other cultures categorize human values” (p. 10).
dimension concerning religious buildings, artworks, etc. Based on the analysis of the component parts of the model, and of the role of religion, ideology and/or worldview in the model, (f) the model’s capacity to address the problem of specificity regarding the relation between cognitive and behavioural radicalisation was addressed.
In order to answer sub-question three – ‘How do (implicit) assumptions and biases feature in the psychological models of the process of radicalisation developed between 2005-2019?’ –, (g) assumptions in the model were identified, for example propositions or causal relations that were accepted as true without proof or argument (think of the assumption that radical belief precedes radial action). Besides, (h) biases regarding (certain) religions were identified. As discussed above, such biases are often reflected in the empirical foundation of radicalisation research. Therefore, the evidence base of the models was reviewed (e.g. are models based on own data collection? Has the model been tested?).
Chapter 3 – Psychological models developed between 2005-2019
This chapter answers sub-question one, “What psychological models of the process of radicalisation have been developed between 2005-2019?” Answering this question, the models are classified into three categories: staircase models chapter 3.1), pyramid models (sub-chapter 3.2) and models systematically mapping interactive mechanisms of the radicalisation process (sub-chapter 3.3). In addition, the definitions of the term ‘radicalisation’ are discussed (sub-chapter 3.4). This chapter functions as a background against which sub-questions two and three will be answered in the subsequent chapter.3.1 Staircase models
Six of the models present a so-called staircase towards radicalisation. Moghaddam (2005), Horgan (2005), Winter and Dubouloz (2011), Ferguson and Brinks (2015), Doosje et al. (2016) and Winter and Feixas (2019) present concrete phases or steps which those who radicalise pass though in a certain order. These models can be sub-categorized based on their main level of analysis: (a) staircase models prioritising processes on the individual level, and (b) staircase models prioritising the relation between external processes and processes on the individual level. Staircase models (a ) prio riti sin g p ro ce ss es o n th e in div id ua l le ve l Moghaddam (2005) Interpretation of material conditions Displacement of
aggression Moral engagement Solidification Terrorist attack
Wilner and Dubouloz (2011)
Trigger phase (effect on interpretation)
Change of meaning
perspective (reinterpretation)
Outcome phase (novel behaviour)
Ferguson and
Brinks (2015) Context Crisis Quest
Contact with group
Meaningful
interaction Commitment Consequence
Winter and Feixas (2019) History of invalidation of construction of meaning Transformative trigger Collapse of meaning and experience of uncertainty Reconstructing of meaning via radical views
(including the reconstruction of violence as acceptable) (b ) re la tio n ex te rn al/in div id ua l le ve l p ro ce sse s Doosje et al.
(2016) Sensitivity (uncertainty) Group membership Action
Horgan (2005) Predisposing events (becoming a terrorist) Catalyst event (becoming a terrorist) Identification (becoming a terrorist) Sustained involvement (being a terrorist) Terrorist attack (being a terrorist)
Type (a) staircase models
According to Moghaddam (2005), there are five floors that are transgressed during the process of radicalisation. On the ground floor, material conditions are interpreted in a certain way. On the first floor “perceived options to fight unfair treatment” (p. 163) are assessed. The second floor is characterised by the displacement of aggression towards out-groups. The third floor denotes the phase of moral engagement and the fourth floor the “solidification of categorical thinking and the perceived legitimacy of the terrorist organisation” (p. 163). Within tightly controlled terrorist groups, the individual passes on to the fifth floor, the terrorist act. Importantly, Moghaddam (2005) prioritises processes on the individual level: “As individuals climb the staircase, they see fewer and fewer choices, until the only possible outcome is the destruction of others, or oneself, or both” (p. 161.
Wilner and Dubouloz (2011) apply the transformative leaning theory to the process of radicalisation. They differentiate between three stages: the trigger phase following a sever crisis (such as socio-political alienation or foreign policy events) which undermines existing ways of interpretation; the process of change phase in which reflection leads to a transformation of meaning perspective (such as religious ideals and practices leading to a reinterpretation of the environment); and the outcome phase in which “the individual establishes a competence and self-confidence in his or her new role, which leads to novel behaviour that is reflective of the individual’s changed personal worldview” (p. 421). The authors focus on “individual psycho-cognitive processes associated with radicalization” (p. 418), specifically on the individual’s transformation of meaning perspective.
Ferguson and Brinks (2015) utilise Rambo’s (1993 quoted in Ferguson & Brinks, 2015) seven stages religious conversion model to explain cognitive radicalisation, the stages being context, crisis, quest, contact with new movement, interaction, commitment, and consequence. The authors focus on the individual as well, specifically on the “transformational process whereby the individual goes from believing in and adhering to and/or practicing one set of values/teaching, to believing in, adhering to, and practicing another set of values” (p. 21).
Winter and Feixas (2019) build a radicalisation model based on the personal construct theory which focusses on the way people construct meaning and anticipate the world. The first stage of their model refers to the “history of invalidation of [an individual’s] construing” (p. 3) of meaning. Via so-called transformative triggers, the individual experiences a collapse of his or her meaning construct which leads to uncertainty. Radical views, often structured and presented as certain, help the individual to overcome this uncertainty. Within this new meaning construct, a negative view of the other is construed, and, in close contact with other group members, violence is reconstructed as being acceptable. The authors “aim to understand the radicalized person’s ‘own internal “map” of reality’” (p. 2), thereby prioritising processes on the individual level.
Type (b) staircase models
According to Horgan (2005), two broad phases can be distinguished – the phase of becoming a terrorist, and the phase of being a terrorist – both consisting of several steps. The becoming of a terrorist is triggered by what Horgan calls ‘catalyst events’ that prompt identification with a terrorist movement. The phase ‘being a terrorist’ is characterised by psychological processes (such as the acquisition and the articulation of a group’s language, a growing sense of empowerment, or “a lowering in inhibition in relation to the expression of violent behaviour” p. 138) that sustain involvement and facilitate the engagement in terrorist offences. As will become clear in the subsequent chapters, Horgan (2005)’s model prioritises the interaction of processes on the individual level and external processes.
According to Doosje et al. (2016), the first phase is called ‘sensitivity’ and refers to feelings of uncertainty or insignificance that make a person sensitive to radicalisation. The second phase is called ‘group-membership’ and denotes the joining of radical groups after a cognitive opening. The third phase, called ‘action,’ describes the turn towards violence. During all three phases various processes on the micro (i.e. individual), meso (i.e. group) and macro (i.e. societal) level occur. By analysing processes on these three levels, Doosje et al. (2016) clearly analyse external processes as well as processes on the individual level.
3.2 Pyramid models
While staircase models presume that all stages/phases/floors must be transgressed in due order, pyramid models, while also presenting a process ordered in distinguishable stages, do not require the transgression of the distinct stage in due order. Both pyramid models prioritise the relation between external processes and processes on the individual level.
Pyramid models re la tio n ex te rn al/in div id ua l le ve l p ro ce sse s McCauley & Moskalenko (2011; 2017) Opinion pyramid:
4. Personal moral obligation to engage in violence 3. Justifier 2. Sympathiser 1. Neutral Action pyramid: 4. Terrorist 3. Radical 2. Activist 1. Inert Kruglanski et al. (2014) Motivational element: quest for personal significance
Ideology: identification of means to the goal of personal significance. Delegitimising functions.
Social process: bringing individuals in touch with ideology.
McCauley and Moskalenko (2017) are the only authors who present two pyramid models of radicalisation: one for cognitive and one for behavioural radicalisation. While they distinguish between different phases of involvement and also order these phases hierarchically, the authors emphasise that neither models are staircase models: “individuals can skip levels in moving up and down the opinion [and the action] pyramid” (pp. 19-20). Regarding the opinion pyramid, McCauley and Moskalenko (2017) distinguish between individuals who are neutral regarding political causes; sympathisers; and justifiers of violent action; and “those who feel a personal
moral obligation to take up violence in defence of the cause” (p. 19). Regarding the action
pyramid, inerts who are not doing anything for a particular cause, are distinguished from activists who engage in legal action for a certain cause, and radicals who engage in illegal action. “[A]t the apex of the pyramid [are] those engaged in illegal action that targets civilians” (p. 20). McCauley and Moskalenko (2011; 2017) clearly analyse the process of radicalisation on an individual level in relation to group-level processes.
Kruglanski et al. (2014) focus on processes of the individual (regarding motivation and ideology) and the social level. They view radicalisation as a matter of degree: a person holding a radicalised ideology is less radicalised than a person who also joins a terrorist group, whereas a terrorist group’s member who is not directly involved in terrorist attacks is less radicalised than a suicide bomber. As the authors themselves state, their understanding of the process of radicalisation mirrors other pyramid models of radicalisation. Their model, however, is conceptualised not in terms of stages of radicalisation but in terms of processes on different
levels. Despite this model being classified as pyramid model, it has significant overlap with the next category. Kruglanski et al. (2014) analyse motivational and ideological elements on an individual level in relation to “the social process (networking, group dynamics) serving as the vehicle whereby the individual comes in contact with the ideology” (p. 80), thereby classifying as type (b).
3.3 Systematic mapping of mechanisms
Instead of distinguishing clear phases in the process of radicalisation, the remaining models map various interactive mechanisms on either one or various levels. Again, (a) systematic mappings of mechanisms that prioritise processes on the individual level can be distinguished from (b) systematic mappings of mechanisms that prioritise the relation between external processes and processes on the individual level.
Only Jensen, Atwell Seate and James (2018) fall outside that categorisation. These authors have developed models based on mechanisms previously identified by scholars. These mechanisms encompass the structural, group-based and individual level. The different models are characterised by the presence of different mechanisms in pathways to radicalisation. As the authors equally draw from “individual-level psychological models, group-level social identity models, recruitment models, social movement models, and cost/benefit models” (p. 3), it is not possible to identify one particular focus (regarding the individual level or the relation with external processes) in the pathways developed.
Systematic mapping of mechanisms
pr ior iti sin g p ro ce ss es o n th e in div id ua l le ve l Ali et al. (2011) Constellation of
circumstances that ignite radicalisation: perceived injustice, polarising ideologies, pursuit of lasting achievements, pursuit of novelty and adventure Conditions that foster meaning: e.g. just and supportive community, unambiguous standards and duties, stable values over time Biases that reinstate meaning: e.g. need for closure, belief that society is just, inflated stability of values,
Mechanisms that transfer biases into violent extremism: e.g. pursuit of revenge, contempt towards other communities, aggressive defence of worldviews Borum (2014) Worldviews: authoritarianism, apocalypticism, dogmatism, fundamentalism Vulnerabilities: need for meaning,
identity and
belonging, perceived injustice
Propensities: motivational, attributional, attitudinal, volitional
Pyszczynski, Motyl and Abdollahi (2009) Individual dimension: Fear of death.
Worldviews as buffer against existential death anxiety and providing us with self-esteem.
Threat to worldview leads to death anxiety.
Group dimension:
Worldviews as social construct, contributing to in-group coherence.
Threat to worldview due to humiliation and grievances. Constructing other as evil.
Desire to restore justice, potentially by violent action. Glorification of terrorist, helping to overcome social sanctions.
Taylor and Horgan (2006)
Setting events, e.g. family, life experience peers, social and political events
Social/political/ organisational context
Personal factors:
E.g. actional or marginal violence.
re la tio n ex te rn al/in div id ua l le ve l p ro ce sse s Decety, Pape and Workman (2017)
Group dynamics: formation and deepening of social bonds, group-isolation Interpersonal processes: e.g. obedience, conformity, dehumanisation Micro-sociology: e.g. extremist ideologies, values and beliefs, extremist and victimisation narratives
Kruglanski and Fishman (2009)
Individual level:
Significance loss, threat thereof, or opportunity for significance gain.
Quest for personal significance Group level: socialisation into terrorist ideologies, recruitment function, institutional component. Organisational level: decision-making structures, rationality of terrorism, production of terrorism operatives. Jensen, Atwell Seate and James (2018) Pathway 1:
personal crisis, community crisis, psychological vulnerability,
psychological reward,
physiological vulnerability, recruitment, group norms, group biases, cognitive frame
Pathway 2: community crisis, cognitive frame Pathway 3; personal crisis, psychological vulnerability, physiological vulnerability, material reward, cognitive frame Pathway 4: community crisis, recruitment, group norms, group biases, cognitive frame
Type (a) systematic mapping of mechanisms
Ali et al. (2017) apply the so-called ‘meaning maintenance model’ to the process of radicalisation. This model argues that individuals, in order to foster and maintain meaning and coherence in life, assume that their social environment is just, wish to “feel certain of which duties or standards they must fulfil” (p. 187), expect core societal values to remain steady, and have confidence in their capabilities. In case these conditions are not fulfilled, the individual can enter the pathway of radicalisation. The mechanisms along this path are several cognitive biases: the inflation of justice in the world which stimulates revenge, the need for closure which stimulates disapproval of out-groups, an aggressive defence of values that are perceived as stable, and a wish to inflate power and capabilities which might encourage the risk-taking involved in violent extremism. The authors prioritise individual level cognitive processes that contribute to the fostering and maintenance of meaning and coherence in life.
Borum (2014) presents a model in which worldviews (namely authoritarianism, apocalypticism, dogmatism and fundamentalism) interact with psychological vulnerabilities (namely the need for meaning and identity, the need for belonging, and perceptions of injustice and humiliation) and with motivational (such as status- or revenge-related motivations), attributional (such as confirmation bias or hostile attribution bias), attitudinal (such as proviolence attitudes or sensation-seeking) and volitional (referring to the “nature and degree of control that persons have over their emotions, motivations/needs, thoughts, impulses and behaviours” (p. 297)) propensities. Borum (2014) focusses on the individual’s mental states and processes. Despite there being some attention to external components such as an external threat, the emphasise lies on the perception thereof. Borum (2014)’s aim is to “explore how otherwise normal mental states and processes - built on characteristic attitudes, dispositions, inclinations, and intentions - might affect a person's propensity for involvement with violent extremist groups and actions” (p. 286).
Pyszczynski, Motyl, Abdollahi (2009) draw on the terror management theory (TMT) to model radicalisation. Core to this model is the interaction of existential threat, worldviews as buffer of this threat, and the need to defend one’s worldview in order to be protected from
existential threat. The authors pay attention to external factors such as group-dynamics or micro-sociological processes. However, these authors prioritise the individual level by using TMT “to explore the psychological, social and cultural forces that lead diverse groups and individuals to endorse, promote and enact violence against innocent civilians” (p. 12).
Understanding the individual as rational actor, Taylor and Horgan (2006)’s model focusses on unravelling the “relationship between political context (and especially its ideological qualities), organisation framework, and the individual by breaking down the process into parts, and examining the relationship between them” (p. 587). To do so, they identify three process variables – setting events, personal factors, and social/political/organisational contexts –, that interact to produce the increasing involvement with terrorism. Taylor and Horgan (2006) mainly focus on individual level processes, understanding the individual as rational actor. Despite the fact that these authors discuss individual level processes in relation to external factors, so-called setting events and the social/political/organisational context, they are subsumed under group (a). The reason for this is that they do not include external processes in their model.
Type (b) systematic mapping of mechanisms
Decety, Pape and Workman (2018) analyse and provide neurological evidence for radicalisation processes on the group level (social bonds with a fictive kin), interpersonal level (processes such as group-conformity, obedience, dehumanisation and deindividuation), and micro-sociological level (such as a stricter adherence to extremist ideologies or narratives), and the interaction between these levels. The authors pay attention to external factors such as group-dynamics or micro-sociological processes. Despite focussing on the hardwiring/neurological underpinnings of such processes, processes on group and micro-sociological levels are captured in the model.
Kruglanski and Fishman (2009)’s model emphasises the complex interaction between the interdependence of processes on the individual, group, and organisational level:
To pursue its objectives, the organization relies on group-level processes which in turn operate on individual psychologies. Organizational functioning presupposes individuals' readiness to participate and contribute to tasks defined by the organization; such readiness derives from potential members' individual propensities and inclinations, and the amplification of these via group-level processes of socialisation and indoctrination. In other words, individuals contribute their motivation and dedication to the advancement of terrorist causes, groups affect the socialization into terrorist ideologies, and organizations determine the structure, mechanisms, and rewards needed for group survival. (p. 5)
Kruglanski and Fishman (2009), therefore, integrate group-level and organisational processes in their model, and analyse individual level processes in connection with group-level and organisational processes.
3.4 Definitions of radicalisation
Regarding different understandings of radicalisation, four trends can be noted. For an overview of the different definitions of radicalisation, please see annex 1.
Definitions in terms of terrorist violence: Moghaddam (2005), Horgan (2005), Taylor
and Horgan (2006), Kruglanski and Fishman (2009), Borum (2014), McCauley and Moskalenko (2017)’s models all address the processes involved in or pathway leading towards terrorist violence. While, Moghaddam (2005), Kruglanski and Fishman (2009), and Borum (2014) explicitly define the term ‘radicalisation,’ Horgan (2005) and Taylor and Horgan (2006) do not mention that term. Most of these authors define terrorism as the committing of violent
