Authors: Rik Peels
Categories: Social movements and Radicalisation in Europe, Behavioural radicalisation, cognitive radicalisation, conceptualization, extremism, racism, Research Article
Source: Journal of Contemporary European Studies
The term ‘radicalisation’ is relatively new. It is mostly the result of the political climate since 2005, but now widely used in work on extremism, fundamentalism, conspiracism, fanaticism, terrorism, and counter-terrorism. But exactly what is radicalisation and can we still properly use the term in the face of the many objections that have been levelled against it? I defend a conception of radicalisation that combines the fourmain approaches in the literature, the so-called monist and pluralist, as well as the absolutist and relativist ones. It does so on the basis of conceptual analysis, reflective equilibrium and particular case studies. Since the term will not be going away, it is wiser to be as lucid on how one defines it as possible. Such a definition matters for three there is much confusion in the public debate about radicalisation, e.g. about Islamism, increasing right-wing radicalisation in Europe and North America, and the views of conspiracy thinkers and anti-vaxxers, the term ‘radicalisation’ is also widely used in the academic literature on terrorism and counter-terrorism, but there is much unclarity about its relation to violence, to phenomena like fundamentalism, extremism, terrorism, and, finally, in order to be fruitful in research we need a definition that can be operationalized.
KEYWORDS: Behavioural radicalisation, cognitive radicalisation, conceptualization, extremism, racism
The concept of radicalisation is ubiquitous in academic work on extreme belief and extreme behaviour. We find the notion in empirical, historical, and theoretical studies on fanaticism, extremism, fundamentalism, terrorism and counter-terrorism, conspiracy theorizing, insurrectionism, radicalism, radicalisation and de-radicalisation, prevention, and societal resilience. The term ‘radicalisation’ is relatively new, especially in comparison with notions like ‘fundamentalism’, ‘extremism’, ‘terrorism’ and even ‘radicalism’. It is mostly the result of the political climate after 9/11 and particularly home-grown terrorism since 2005, but now widely used in academic work, prevention policy, and the public debate. To say that it is used is an there are more than 80,000 referenced works on ‘radicalisation’ (Kaya 2020). Particularly influential in the field of radicalisation research has been Walter Laqueur’s seminal work (2004). He argued that in explaining the turn to extremist and terrorist violence, researchers should move away from political and structural conditions to individual psychological character traits, particularly those related to theological ideology, and thus to the temporal process of radicalisation of particular subjects.
But exactly what is radicalisation? Various analyses of radicalisation that are often used can easily be disqualified. Peter Neumann defines ‘radicalisation’ as ‘the process whereby people become extremists’ (Neumann 2003). This may well be true, but it tells us little about the nature of the process of radicalisation. Societal understandings of radicalisation as ‘what goes on before the bomb goes off’ are even more problematic, as it is widely agreed in the literature that extremist violence need not be suicidal violence and even that one can radicalise without actually employing violence at any point. As we shall see in more detail below, it is widely acknowledged that one can cognitively radicalise – radicalize in one’s beliefs, ideas, thoughts, and ways of thinking – without that manifesting itself in any kind of behaviour.
As various scholars have pointed out, the term ‘radicalisation’ is often ill-defined (e.g. Schmid 2013, 5–6). Etymologically, the word is, of course, derived from the Latin ‘radix’ (root), so one might think that radicalisation has to do with going to the roots of an alleged problem (Calhoun 2011; see Bötticher 2017 for the historical use of ‘radicalism’). However, this is not only somewhat uninformative, it is also insufficiently activists also try to get at the roots of problems. Etymology may be interesting, then, but will not all by itself answer the question of how we should conceptualize radicalisation.
A rigorous conceptualization of radicalisation matters for three reasons. First, it can bring clarity and precision to the public debate. The term was traditionally reserved for Islamism and right-wing extremism in Europe and North America, but nowadays it is also used for conspiracy thinkers fighting the Great Reset, anti-vaxxers, or radical BLM supporters. A proper conceptualization can help identify the boundaries of the concept. Second, the term is widely used in the academic literature, but there is much unclarity about its exact relation to violence, and to conceptually closely related phenomena – several of which are also covered in this special issue – like fundamentalism, extremism, terrorism, fanaticism, populism, and non-conventional forms of political mobilization. Third, in order to be fruitful in qualitative and quantitative research, we need an analysis of radicalisation that can be operationalized. Many existing conceptualizations of radicalisation do not meet this criterion. Fourth and finally, the concept of radicalisation has been criticized for various reasons, such as its being a tool to police certain minorities (Kundnani 2012) and its being Western centric (Ilyas 2021, 5). A rigorous conceptualization might help to meet such worries.
Let us be explicit how this article fits into this special issue. Various authors address drivers and frames of radicalisation, so what explains radicalisation (Daniele Albertazzi and Donatella Bonansinga, Dimitrios Anagnostakis, Cristiano Bee, Stavroula Chrona in this issue). Others explore the relation of radicalisation to phenomena such as conventional and non-conventional forms of political mobilization (Martin Mejstřík, Cristiano Bee, Stavroula Chrona in this issue), and reactionism and populism (Ayhan Kaya, Max-Valentin Robert in this issue). Still others explore various empirical aspects of radicalisation, such as gender roles (Ayşenur Benevento, Martin Mejstřík in this issue), global and local properties (Tahir Abbas, Ayhan Kaya, Max-Valentin Robert, Metin Koca in this issue), or specific varieties of radicalisation, such as right-wing radicalisation (Valeria Bello, Metin Koca, Dimitrios Anagnostakis in this issue). Yet others explore responses to radicalisation on a state level or civil society (Cristiano Bee, Stavroula Chrona, Martin Mejstřík in this issue). This article addresses the preliminary issue of whether there is such a single thing as radicalisation at all and whether it is a concept that can usefully be employed. Whereas most other contributions are primarily empirical in nature, this article is theoretical and even specifically largely philosophical in nature, even though it of course takes the empirical work on radicalisation into account. Naturally, if the concept were deeply flawed, that would spell trouble for these other contributions, but it will be argued that the notion of radicalisation can actually be fruitfully developed if we pay careful attention to certain pitfalls.
In this article the issue is addressed head-on by developing and defending a conception of radicalisation that combines the main approaches in the literature, namely the so-called monist and pluralist ones as well as the absolutist and relativist ones. After formulating various desiderata (§2), several objections that might be levelled against using the notion of radicalisation are addressed (§3). It is then shown what the four main approaches to radicalisation in the literature, namely monism and pluralism as well as absolutism and relativism, amount to (§4). Subsequently, it is shown how the insights of the four approaches can be combined in such a way that their strengths are taken on board and their weaknesses circumvented (§5). Finally, the article discusses in detail various worries that might be raised for this account (§6).
It seems that any fruitful concept of radicalisation should be explicative in the sense that it fine-tunes our current use of the concept. It should, therefore, stick relatively close not so much to societal usaget (which can vary wildly) but to academic usage. It cannot deviate too much from how the word is commonly used in the literature on radicalization. After all, that would result in a new concept being introduced rather than an existing concept being ameliorated. In this section, then, five desiderata will be formulated that can be derived from the current academic debate on radicalisation.
First, it is widely agreed there are two kinds of cognitive and behavioural (e.g. Malthaner 2017, 371; Sageman 2004, 2008). Others have made a similar distinction but used different terms, such as, respectively, ‘belief radicalisation’ and ‘methods radicalisation’ (Cassam 2021; Cassam in Sardoč 2020).^1^ Cognitive radicalisation occurs when one develops radical ideas and comes to embrace extremist ideologies. It can come without the disposition to act on those beliefs. Behavioural radicalisation occurs when one engages in extreme, particularly extremist behaviour, such as using violence. One can hold various extreme beliefs without engaging in extreme behaviour, for instance, because one prefers to keep it to oneself (this holds for various highly-educated neo-fascists, as Sterkenburg 2021a, 2021b has shown). One can engage in radical behaviour even if one does not have the corresponding beliefs, e.g. because one does it out of admiration for a group leader or merely because of various so-called grievances. Possibly, there are further ways to radicalise than just cognitively or behaviourally – an issue to which we return below – but it can at least be said that an account of radicalisation should do justice to these two faces of radicalisation.
Second, violence is of course one important expression of behavioural radicalisation, as said above. Yet, it is not the only one. One can (increasingly) disadvantage sexual, gender, ethnic, or religious minorities. One can estrange from democratic processes, for example by no longer voting (therefore, increasingly not doing things can count as radicalisation) (see Benevento in this SI). One can exclude any opposing voices from one’s group by no longer allowing them to enter one’s home, or by excluding them from the church or mosque. And so on. (See Cassam 2021, Anagnostakis in this SI; Koca in this SI). Any plausible concept of radicalisation should be able to take this on board.
Third, it should leave room for the so-called ‘problem of the few’ (e.g. Cassam in Sardoč 2020, 167). In fact, there are two problems of the only a few who are exposed to extremist ideology actually cognitively radicalise and only a few of those who cognitively radicalise also behaviourally radicalise. A conceptualization of radicalisation should be compatible with this.
Fourth, two main lines of thought in analysing radicalisation are so-called monism and pluralism (or universalism and contextualism, or generalism and particularism) (see Cassam in Sardoč 2020). Monism says that there is a single core that is common to all processes of radicalisation. In other words, particular instances of radicalization across radically different extremist movements, such as Salafi Jihadism, right-wing extremism, or the radical left, all have something in common. Pluralism, however, says that there are many different processes of radicalisation and that they often have little or nothing in common, there is no core that they all share. Another important distinction here is that between absolutism and relativism. Absolutism says that there are absolute truths about whether someone has radicalised. Relativism on the other hand says that whether someone has radicalised is always relative to a particular set of values or principles, there is no absolute or objective or mind-independent truth about that. It always depends on the perspective of a particular audience, such as the majority of a society at a particular time. It would clearly be preferable if an account of radicalisation should do justice to the main intuitions underlying each of these four positions. That might sound impossible, because these positions seem to involve diametrically opposed pairs. As we shall see, though, there are plausible ways to understand them such that they turn out to be perfectly compatible.
Fifth, work on radicalisation is primarily empirical, both qualitative and quantitative. It is only secondarily historical or theoretical (conceptual, normative, etc.). A proper concept of radicalisation, then, should lend itself for fruitful use in empirical research and that means that it should be operationalizable. To say that means that the somewhat abstract concept of radicalisation should be analysed in such a way that it can be translated into and measured by concrete observations, e.g. in questionnaires or interviews.
Before we try to develop a fruitful concept of radicalisation, there are some challenges to this very project that we first need to consider. As we shall see, they will teach us important lessons that ought to be taken into account in conceptualizing radicalisation. Here, we will consider what seem to be the three most important ones and explain what they mean for the project at hand.
First, Mark Sedgwick (2010) argues that the concept is used in three different contexts that all have their own agenda and purposes. These are the security context, the integration context, and the foreign-policy context. Suggesting that there is a common core to all three of them that we can denote by the term ‘radicalisation’ is misleading. In the security context, the idea is that anarchists and terrorists may pose a threat to public security, that is why it uses the notion of radicalization. The term ‘radicalisation’ is then used to denote certain people gradually becoming a threat to public security. In the integration context, those are dubbed radicalised who are thought to oppose things like republican principles, fundamental democratic values, active citizenship, co-citizenship, or acceptance of homosexuality. The reason for the use of the concept of radicalization here is it is thought to be an obstacle to the proper integration of newcomers to that particular society. In foreign policy, one can label certain kinds of opposition in one’s own country as protests or violence by radicalised people, in order to justify, in the face of the international community, taking repressing measures against them (for a more detailed description of these contexts with their own agendas, see Sedgwick 2010, 485–487). Here, the reason for the use of the concept of radicalization is the justification of one’s own policy towards a specific part of one’s population. Since there is no common core to what counts as ‘radicalised’ in these three different contexts, we should either give up the notion or highly contextualize it. The lesson is fairly straightforward any plausible concept of radicalisation should be able to make sense of crucial differences, but, in case Sedgwick’s suggestion misses out on relevant things, also important commonalities across these three contexts.
Arun Kundnani (2012, 2014) has shown in detail how the term ‘radicalisation’ has been used and in fact hijacked for numerous political purposes. The process of radicalisation has often been applied only to Muslims embracing extremist ideologies and Kundnani considers this a clear expression of deep-seated racism (for more on the issues of terrorism and counter-terrorism and race, see Groothuis 2020). Among other things, it has wrongly led to mass surveillance among Muslim minorities in the U.S. and other countries. Kundnani rightly points out that funding (money flows) and ideological assumptions have had a seriously distorting effect on the use of the concept of radicalisation and he may well be right that Walter Laqueur’s approach (e.g. in Laqueur 2004) to radicalisation has been particularly influential here. The lesson, it seems, is that any proper concept of radicalisation should be entirely neutral on ethnic and religious issues. Any concept that implies the slightest amount of racism ought to be firmly rejected.
One may worry that such connotations are too strongly attached to the term ‘radicalisation’ and that we should, therefore, abandon it altogether. This seems unwise for two reasons. First, as pointed out above, the term is widespread in numerous debates and despite many objections levelled against it, it is more popular than ever. In all likelihood, it will not go away. It is pragmatically entirely justified, therefore, to seek elucidation and clarification of the concept in the sense of seeking necessary and sufficient conditions for radicalization, as well as identifying constitutive elements of cognitive and behavioural radicalization. That will be useful to the debate, for undefined notions lead to much conceptual confusion. Second, it is also epistemically justified. In other words, there is something to be gained intellectually. As we shall see below, using the notion of ‘radicalisation’ across radically different paths of radicalisation will allow one to study important psychological similarities (e.g. the role of cognitive vices), political and social similarities (e.g. hostility), and content-similarities (e.g. in narratives) that one would miss out on if one lacked a more generic notion for all these different phenomena.
A third and final challenge is the idea that radicalisation may well be a so-called essentially contested concept, a notion first introduced by Gallie (1956). Something is an essentially contested concept if actual use of it comes with endless disputes about what he proper use is and if such disputes cannot be settled by empirical evidence, linguistic usage, or logic. Examples are art, religion, power, and philanthropy. Surely, the perennial debates on what radicalisation is and whether we should use the term at all suggest something along these lines. In what follows, we’ll explore whether this worry can be assuaged.
On monism, there is such a thing as radicalisation. In other words, there is a common core to all processes of radicalisation, whether or not it plays out differently across fundamentalist, extremist, and terrorist movements, such as left-wing extremism, Jihadism, neo-Nazism, and RSS Hindu extremism. The alternative is pluralism. Quassim Cassam (in Sardoč 2020) has argued that there probably is no such a thing as ‘the radicalisation process’. In all likelihood, there are just many different kinds of radicalisation processes. If one studies each case, one will note remarkable differences in pretty much any regard in which subjects socio-economic background, political circumstances, theological and other beliefs, religiosity, grievances, experienced and/or perceived injustice, relation to the group, etc. The lesson we can take from this is that in conceptualizing radicalisation, one should bear in mind two important is it indeed true that particular instances of radicalisation differ in pretty much any theological, cultural, economic, social, political, and ideological regard, and if this is so, what does that mean regarding the issue of whether there is such a thing as the process of radicalisation or only numerous distinct processes of radicalisation?
The other main distinction is that between absolutism and relativism. On the absolutist approach, whether someone is radicalised does not depend on the beliefs or values of anyone else. Sophia Moskalenko and Clark McCauley, for instance, contrast radicalisation with activism. They define ‘activism’ as ‘readiness to engage in legal and non-violent political action’ (Moskalenko and McCauley 2009, 240), whereas ‘radicalism’ – as something crucially different from radicalisation – is defined as ‘readiness to engage in illegal and violent political action’ (Malthaner 2017, 240). The idea here, then, is that radicalism is engaging in illegal and violent action; it can, therefore, be defined without reference to anybody’s beliefs or values – at least not anybody beyond the person or group who is thought to be radical.
Their particular variety of absolutism faces several worries. Among other things, we saw that one can radicalise without engaging in violent action. Also, there can clearly be justified resistance, even if it was illegal at the time, such as the German theologian Dietrich Bonhoeffer’s involvement in an attempt to assassinate Hitler in 1944 (Mengus 1992). We would not say that he was radicalised—that would be to confer a new meaning upon the term. Maybe somewhat confusingly, we would say that he was radical, or maybe even a radical, and that he displayed radicalism. But we would not say that he was radicalised. Independently of the particular worries one might phrase about their specific account of radicalisation, the point should be various authors think that whether or not one is radicalised is an absolute matter.
On a relativist understanding, radicalisation is a matter of representing or supporting an extreme section in society and it is always relative to a particular belief-system or value-system of a particular group or audience what should be considered moderate and what extreme. Relativism raises worries of its own, though. For instance, is it simply a matter of perspective whether someone is radicalised? The phrase ‘one person’s terrorist is another person’s freedom fighter’ is well-known but many find it problematic. It seems to undermine any non-question begging justification for preventing and fighting radicalisation.
In the following section, we will explore whether we can do justice to the main intuitions underlying both positions, while simultaneously avoiding the pitfalls that plague these positions. Again, these positions may seem entirely opposed, so that one cannot combine them. We shall see, though, that there are plausible ways to understand them such that they are perfectly compatible. We will see that there are various traits shared by all processes of radicalisation, yet when it comes to fleshing them out, it turns out that what makes them cases of radicalisation is heavily context-dependent.
It seems that there are various properties of the process of radicalisation that are shared by all processes of radicalisation. In other words, a sober inspection of the data about what are widely considered to be processes of radicalisation all exemplify these properties. First, trivially, if there is a process of radicalisation, then there is a subject that radicalises. Slightly less trivially, that subject might be an individual, a group of people, or a social entity (a student body, an institute, an organization, a government). Some scholars may well be right that the literature and the public debate have wrongly focused on the radicalisation of individuals at the cost of those of groups (Malthaner 2017, 370). This may also hold for social entities. This matters because social epistemology has shown in detail that, for, say, group belief, it is not required that all members that form the group individually hold the belief in question. What is relevant is what the operative members of the group – those persons who have group authority to take decisions – think of the issue and what the group dynamics are. Thus, governments can radicalise, as, according to some, the Bush administration did after 9/11. Entire peoples can radicalise, as happened to Iran after 1979. Institutes and organizations can radicalise, as happened with the Rote Armee Fraktion which was non-violent at first (Aust and Bell 2009). And that is perfectly compatible with disagreement and even opposition by parts of the government, population, or organization.
The second uncontroversial thing – the thing shared by all processes of radicalisation – is that radicalisation is a temporal process (Kruglanski, Bélanger, and Gunaratna 2019; Malthaner 2017, 371). More specifically, it is a temporal process that takes place on a continuum or multiple continua. One person can be more radicalised than another or one can now be more radicalised than last year around this time. Note that it does not follow that radicalisation is a straightforward linear process. That does not need to be the case at all. One can move back for a while at some point, or be stationary for some time. The process can be somewhat capricious and unpredictable. All that matters is that in the course of time one moves up the scale. Note also that this does not exclude an extremely short process of radicalisation, in which, say, a religious experience or a moment of revelatory experience plays a central role (see De Graaf 2021). Some people may even be extremists without ever maybe they have been extremists all along (for instance, due to being raised in an extremist community) or maybe they more or less instantly turned extremist due to a traumatic experience. Straight away, then, we see also elements that make pluralism how quickly people radicalise and whether extremists have radicalised at all varies from subject to subject.
Third, radicalisation is a diachronic movement from the moderate to the extreme. Of course, ‘moderate’ and ‘extreme’ need to be fleshed out and it is up for debate whether that can be done in absolutist terms (we return to that below). Before we do so, though, let us note that both for behavioural and cognitive radicalisation, there are various constitutive elements that we can all agree on count as elements of radicalisation. For behavioural
For cognitive radicalisation we find similar but at some points crucially different constitutive
What is often overlooked, it seems, is that there is radicalisation in other domains as well, not just in those of belief and behaviour. One can affectively radicalise, for instance, that is, radicalize with regard to one’s affections, emotions, and passions. As Githens-Mazer and Lambert (2010, 894) argue, ‘(t)his notion of passion is lacking in theoretical approaches to violent radicalisation’. It seems one can affectively radicalise, for instance, by developing stronger emotions (say, from frustration to anger to hatred, or from deception to resentment and vengefulness), emotions that are more strongly embedded, emotions that have a wider scope of objects (one fears more things or hates more things), and so on.
Similarly, one can conatively radicalise, that is, radicalise with regard to one’s wishes, desires, and goals (one’s conative states). One’s desires and goals can broaden, deepen, become more central, or have a wider scope. Where those desires and goals have to do with extremist goals and purposes, such as racial purification or a preoccupation with one’s national identity, change in such conative states can amount to conative radicalization.
Again, we see what lends supports to monism – all subjects who radicalise behaviourally, cognitive, affectively, or conatively, and often it is a combination of these – but also what lends support to one person can radicalise more affectively, whereas another person radicalises behaviourally, and so on. This brings us to the issue of absolutism versus relativism. Most scholars of radicalisation and extremism would say that a particular kind of increasing devotion to activism and reformation should not as such count as radicalisation. Mahatma Gandhi, Martin Luther King, a liberal waging war against royalists, and others may have gone through a process as just captured, yet without radicalising. And, as we noted above, the purpose here is to fine-tune and clarify existing use of ‘radicalisation’ rather than to develop an entirely new concept. How, then, to distinguish these cases from cases of true radicalisation?
It seems that there are two ways to do so. First, one can build in various normative – pejorative in particular – elements in one’s definition. Here are a few options of elements that one could make constitutive of the process of
Of course, one may flesh out these normative notions relativistically. In other words, what is moral or immoral, rational or irrational, and so on, depends on a person’s values and beliefs. Yet, one may equally well suggest that there are objective moral and epistemic truths about these things (no matter how hard they are to know sometimes). In fact, moral realism is the majority position in ethics nowadays. Such normative components in an analysis of radicalisation, then, would not rule out absolutism.
However, a well-known problem with building normative constituents into one’s definitions is that they are notoriously hard or impossible to operationalize. We can measure what people feel is immoral or think is irrational, but how are we supposed to measure actual immorality and irrationality? Fortunately, as we shall see below, there have recently been attempts to meet such worries.
The second way to distinguish benign radicalism from radicalisation would be to suggest that only radicalised individuals moved not just on a continuum from moderate to extreme, but all the way to extremism.^3^ Here, ‘extremism’ would denote movements such as neo-fascism, violent Trump-supporting American nationalism, the Rote Armee Fraktion (RAF), the Irish Republican Army (IRA), RSS Hindu ethno-nationalism, and Salafi Jihadism. One could rely on a definition of ‘extremism’ here, or just provide a list and thus give an exemplary definition. This would clearly leave room for absolutism, as it is compatible with the idea that there are facts about what counts as an extremist movement.
So far, the absolutist part. Can we also do justice to the intuitions underlying relativism? It seems we can, in at least two ways. First, people can radicalise across different scales. We already saw that people can radicalise with respect to their beliefs, actions, emotions, and goals and desires. Perhaps there are further scales, like one’s preoccupations, such as an obsession with purity, virtue, humiliation, or revenge. In fact, many of these scales can have multiple dimensions all by in addition to the dimensions distinguished for behavioural and cognitive radicalisation above, one can think of a scale of decreasing rationality for beliefs (even more in conflict with public evidence, for instance) or another scale that concerns the moral status of those beliefs. The point is that people can score differently on these different scales and on the various dimensions of these scales.
Moreover, these scales and dimensions may well be incommensurable: they cannot properly be compared. And if they can be compared at all, there may be no theory-neutral way to do so. Contextual values and epistemic and moral paradigms will come in. Hence, the degree of radicalisation for a particular individual, group, or social entity will inevitably be a contextual matter.
Second, one might suggest that if radicalisation is a process from the moderate to the extreme or even to extremism, something is extreme or an instance of extremism always from a particular perspective and that such as perspective entails a set of values. This is also the case historically, as Ayhan Kaya points ‘Ideas that are radical at some point could be liberal or even conservative at another. Liberals and democrats of the 19th century were then the radicals’ (Kaya 2020). What Kaya rightly draws our attention to is that what is considered to be radical, radicalised, extreme, and extremist is to some extent time- and place-dependent, because background values and conditions can differ from time to time and from place to place. This is of course true and a conceptualization of radicalisation should do justice to this.
This line of reasoning faces two problems, though. On the one hand, there is an argumentative gap from the idea that what is perceived as radicalised differs from time to time to the idea that what it actually is to be radicalised differs from time to time. On the other hand, if it is all a matter of one’s background values, then it would follow that Martin Luther King was as radicalised as a contemporary neo-Nazi. After all, neo-Nazis are regarded as radicalized by the majority of contemporary Western societies, but so was Martin Luther King by the majority of U.S. citizens in the sixties. It seems the former are rightly so regarded whereas the latter was wrongly so regarded. To say that what counts as radicalized is determined by the majority position at some point seems to miss out on various important differences between various people and groups, such as that King chose non-violent means, whereas many neo-Nazis do not, that King sought harmony and peace rather than division and polarization.
Unless one is willing to adopt a highly relativist position that has counter-intuitive consequences to the effect that Martin Luther King was not just a radical, but in fact also radicalised and an extremist, then, one should reject this second line of reasoning. This is fully compatible with the realization that what is considered or thought of or perceived as radicalised and as extremist often differs from time to time and place to place and that such perceptions depend on background values. This is surely something to pay close attention to, especially since the debate on radicalisation is usually carried out from a rather Western (North-American and European) perspective, sometimes even with drastic practical ramifications for minorities, such as mass surveillance for Muslims.
Finally, let us return to the five desiderata that we formulated in Section 2 (1) The account is perfectly compatible with the idea that people can cognitively as well as behaviourally radicalise. In fact, it entails it, even though it is quick to add that there is more than just cognitive and behavioural radicalisation. (2) It clearly follows from the account that radicalisation can but need not come with the increasing use or support of violence. (3) The conceptualization is compatible with the fact that out of those exposed to extremist ideology, only a few cognitively radicalise and that out of those only a few behaviourally radicalise. (4) The conceptualization does justice to the main intuitions underlying monism and pluralism, as well as those of absolutism and relativism, namely respectively that there is a common core to all paths of radicalisation but that those paths may be radically different, and there may be a plausible absolutist approach to radicalisation (if moral realism holds), but that the fact that multiple and possibly incommensurable scales are involved in processes of radicalisation suggests the truth of relativism about some aspects of radicalisation. (5) Finally, the concept can relatively easily be operationalized and we can use various existing measurement tools and scales in doing so. For instance, one can use something along the lines of Altemeyer’s and Hunsberger’s (2004) Fundamentalism Scale in order to measure whether people hold extreme beliefs of various kinds and how strongly they hold them. Some of the normative elements explored above have even been operationalized as well, for instance, in Meyer, Alfano, De Bruin (2021).
Let me summarize the findings so far in the two tables , the one displaying the monist, pluralist, absolutist, and relativist elements of radicalization, where the monist and pluralist elements mentioned are not that controversial and can easily be combined, where the absolutist and relativist elements are clearly more controversial (Table 1), and the other one displaying the main kinds and most important varieties of radicalization (Table 2):
Now that we have conceptualized radicalisation in a way that synthesizes monist and pluralist as well as absolutist and relativist approaches, let us return to the objections that we explored in the section on desiderata and some further objections – five in total – and see whether said conceptualization holds water.
Philosophy of language also supports this point. When words have multiple meanings, there are two options. Either the word is ambiguous: it then means completely different things, like the word ‘bank’ is ambiguous between, say, the bank of a river or the bank where one can deposit one’s money. Or it is polysemous: it has then different, but overlapping meanings with a common core, e.g. ‘to get’, which can mean to procure, to become, or to understand (all related). Clearly, ‘radicalisation’ is much closer to polysemy than to the three contexts in which the term is used lead to overlapping meanings.
Another objection of Segdwick (2010) is that it is arbitrary where we draw the line between when someone is radicalised and when they are not. For any delineation point, one can easily justifiably pick another cut-off point that functions equally well for the purposes at hand. This seems right. As said, radicalisation is a matter of degree and that means that any cutting-off point will be arbitrary to some extent. Someone who rejects all democratic values and is willing to use violence to overturn democracy is clearly radicalised, someone else can be less radicalised because she is not willing to use violence, or because she rejects only some democratic values. And so on. This is exactly the result we want. For practical purposes, such as when to intervene, one can draw the line here or there. For example, one can draw the line where someone uses violence or where particular groups are excluded from education or where religious sermons encourage not partaking in democratic processes. Measures here concern closing of schools, taking down websites, and in rare cases even incarcerating people. But all such practical matters for delineating radicalisation from its absence should be differentiated from the conceptual radicalisation remains a matter of degree.^4^
Arun Kundnani (2012 and 2014) worries that the concept of radicalisation only serves for governments to intrude into the lives of its citizens, Muslims in particular. He thinks this is especially true for theological explanations of ‘The problem is that, if there is no real reason to think that these radical religious beliefs are associated with terrorist violence, then the theological radicalisation model is merely legitimizing unwarranted state intrusion into the private religious lives of large numbers of citizens’. In his wake, Derek Silva has argued that ‘the concept (of radicalisation; RP) itself contributes to discriminatory government policies targeting diverse Islamic communities’ (Silva 2018). This has to do with a number of factors, one of which is that much radicalisation research is directly funded by the government. According to Silva, this has led to a lack of funding of research that is critical of contemporary and former counter-radicalisation government practices, of status-quo surveillance, and of various intelligence and policing strategies. It has led to a focus on research aiming to identify the theological, cultural, psychological, and social characteristics of those who radicalise. This, in turn, has led to a pre-occupation with indicator-based models of radicalisation in mass-media.
In reply, it should be pointed out that the concept of radicalisation laid out in the previous section neither says nor implies anything about specific extremisms, or the reasons for which people radicalise, or when or how it is legitimate to intervene, or what power relations hold between the radicalized and various factors and persons that contribute to the radicalization process. The concept does not entail anything about policing tools and strategies. Kundnani is right that various concepts of radicalisation have served at times as a vehicle for racism. We should, therefore, always remain vigilant as to how and for what purposes the term is used. This, however, should not blind us to the fact that arguably there is a process of radicalisation in some people. The words ‘truth’, ‘democracy’, ‘freedom’, and ‘knowledge’ are each also frequently abused for numerous purposes, yet few people question whether these things exist or matter. In fact, the same holds for various other personal (partly psychological) processes, such as conversion and intellectual growth. These terms can be abused, but it seems hard to deny that they denote real phenomena.
Not only should we be aware of the purposes for which the concept or radicalization is used. We should also realize that processes of radicalization and treating something as a process of radicalization are often related. The subtitle of Clark McCauley and Sophia Moskalenko’s influential 2011 study Friction already was How Radicalization Happens to Them and Us (in the revised 2017 How Conflict Radicalizes Them and Us). Frequently, radicalization is a mutual phenomenon. Particular media representations, for instance, easily lead to support for radical measures in fighting radicalization (Stürmer et al. 2019).
It seems that Sedgwick is (rightly) criticizing some particular understandings of radicalisation here, not the very concept of radicalisation. The concept of radicalisation can or should be descriptive in the sense that it tells us what happens in individuals and groups – it does not as such provide an explanation of that. One can perfectly well appeal to various root causes of a sociological, economic, cultural, or political nature at the macro-level. Or one could explain it in terms of ideology, or beliefs, or religiosity, or a combination of these things at the micro-level. Or one could go for an explanation at the meso-level, such as one in terms of group dynamics. In fact, one can combine elements of these explanations. All of this is up for debate. Nothing about this follows from the very concept of radicalisation laid out in this article.
The same point applies to many critical comments on radicalisation that we find in the literature. According to Githens-Mazer and Lambert (2010, 889), for instance, ‘(r)adicalisation is a research topic plagued by assumption and intuition, unhappily dominated by conventional wisdom rather than systematic scientific and empirically based research’. Yet, a closer look at their article shows that what they really argue is not so much that the very concept of radicalisation is problematic, but that what they consider to be the ‘conventional wisdom’ view on causes of radicalisation is untenable. On this view, individuals radicalise because of identity issues, such as marginalization and alienation, and exposure to extremist ideas, such as ‘Salafi-Jihadi’ ideology. They take various cases, such as Rhaman Adam’s involvement in the 7/7 bombing, to show that such explanations of radicalisation fail. They may well be right – it should be obvious, though, that their objection concerns particular alleged causes of radicalisation rather than the very concept of radicalisation as such.
In fact, Derek Silva (2018), who himself is rather critical of the concept of radicalisation, admits this much in his observations about the radicalisation ‘Despite often using methodological techniques and approaches that Kundnani had already appropriately critiqued, the data here suggest that academic publishers continue to disseminate research attempting to find the social psychological and theological causes associated with radicalisation, whilst using pre-selected case studies without reference to alternative cases’. The problem, if he is right, is with specific accounts of alleged causes of radicalisation – because the selection of case-studies on which they are based is biased, for instance – not with the very concept of radicalisation.
All this also means that radicalisation is not an essentially contested concept, only parts of the concept are contested. As long as one makes explicit what the monist part and possibly the absolutist part of the concept is, one may well turn out to agree with other people with whom one initially seems to disagree.
The concept of radicalisation as developed in this article concerns a shorter or longer temporal process in which certain developments take an individual, a group, or a social entity moves from the moderate to the extreme – cognitively, behaviourally, or both – and, one could argue, all the way to extremism. We saw that we might even want to broaden the concept of radicalisation (an important focus of this special issue) by including affective and conative radicalisation.
This account, therefore, does not as such commit one to any explanations (referring to causes, reasons, enabling conditions, and the like) for radicalisation. They can be psychological, sociological, political, entirely person-dependent, a matter of moral outrage, friendship, personal experiences, and kinship as Sageman (2004, 2008) has argued, or whatever. It may be theology and doctrine, as Laqueur (2004) has argued. Nor does it as such identify indicators that have predictive power or risk factors as they might better be called (Cassam 2020). It does not identify any group or religious view or ideology as particularly likely to radicalise. The concept of radicalisation that has been defended in this article is perfectly compatible with seeking clues and indicators on micro-, meso-, and macro-levels (as e.g. Gartenstein-Ross and Grossman 2009 do) to find out what contributes to radicalisation and perhaps in particular to what Wiktorowicz (2005) calls the ‘cognitive opening’ for the use of violence. And it can be applied to both native social groups and migrant-origin social groups, as this special issue seeks to do.
As noted, some have suggested we should rather talk about radicalisation processes than the radicalisation process (Cassam 2020). It is probably wisest to do both, that is, to keep an eye both on what makes these many radicalisation processes unique in different times and places and across radically different extremist movements, ranging from right-wing neo-fascism to Salafi Jihadism, and on what they have in common with one another.^5^
This article has been limited to the study of radicalization into extremism. But as noted earlier (in footnote 3) there are parallel and to some extent similar processes of radicalization into other kinds of extreme action, extreme belief, and extreme affection. Here, we can think of radicalization into religious nationalism (Kinnvall 2004) or radicalization into conspiracy theorizing (e.g. Daniel and Douglas, 2014; Erisen 2022). This matters because they interact with one another in various ways. Research shows, for instance, that extremism is a good predictor for conspiracy theorizing (Van Prooijen, Krouwel, and Pollet 2015). It would be worthwhile to explore to what extent the concept of radicalization as developed in this article (e.g. the distinction between cognitive, behavioral, affective, and conative radicalization) also applies to these closely related phenomena. Hopefully, the present article provides a good starting point for doing so.
No potential conflict of interest was reported by the author.