Deradicalization or Disengagement?

Deradicalization or Disengagement?

A Process in Need of Clarity and a Counterterrorism Initiative in Need of Evaluation

By John Horgan


As a result of the overwhelming preoccupation with uncovering the process of radicalization into terrorist activity, little attention has been paid to the related, yet distinct processes of disengagement and deradicalization from terrorism. This continuing neglect is ironic because it may be in the analysis of disengagement that practical initiatives for counterterrorism may become more apparent in their development and feasible in their execution. While a variety of deradicalization initiatives worldwide are currently receiving enormous interest from afar, it is inevitable they will be subjected to intense scrutiny regarding their alleged outcomes and claimed successes. A more immediate challenge, however, is to assert some conceptual and terminological clarity. While deradicalization has become the latest buzzword in counterterrorism, it is critical that we distinguish it from disengagement and stress that not only are they different, but that just because one leaves terrorism behind; it rarely implies (or even necessitates) that one become 'deradicalized'. One of several implications arising from this distinction may be a more realistic appraisal of how our knowledge and understanding of the disengagement processes (and not undefined and poorly conceptualised deradicalization efforts) may be put to effective use in the short-medium term development of research agendas.


In a book from several years ago,[1] I outlined the complexity of understanding what involvement in terrorism entails. In particular, I suggested that one useful way of appreciating the complexity of involvement is by viewing it as a process. This psychological and behavioural process comprises at least three discrete phases for the individual terrorist: becoming' a terrorist, 'being' a terrorist [understood as both a) remaining involved and b) engaging in actual terrorist offences] and 'disengaging' from terrorism.

One of many challenges arising from making these distinctions relates to identifying its implications for counterterrorism. First, a critical conceptual point for informing response strategies (at whatever stage they may be focused) lies in the recognition that the factors that impinge upon the individual at each of these phases may (1) not be necessarily related to each other and (b) may not necessarily reflect upon each other. In other words, answering the call of one of these phases of the process may not reveal anything useful or insightful about the other.

Taking this a little further, we realize that thinking about involvement in terrorism in such a way is essentially a recognition that answering questions about why people may wish to become involved in terrorism then may have little bearing on the answers that explain what they do (or are allowed to do) as terrorists (or something else). In other words, the factors that determine how someone becomes involved are rarely related to the kinds of factors that determine how, when, and to what extent that person will engage in specific terrorist operations. Similarly, answering questions about what keeps people involved with a terrorist movement may have surprisingly little, if any, bearing on what subsequently causes them to disengage from terrorist operations or from the organization (and/or broader network or movement) altogether.

This kind of thinking is not especially new. Criminologists are familiar with these kinds of distinctions, but as far as terrorism is concerned, the implications are rarely appreciated at the level of policy initiatives. This is even more apparent when it came to recognizing the enormous potential in studying the process of disengagement. Despite some important contributions in the 1980s and early 1990s, little is known even today about what happens for the individual terrorist to leave terrorism behind. Consequently, there is no available evidence that may serve to inform policy that is in any way related to thinking critically about what could be developed to facilitate or promote disengagement at any level.

In the terrorism research community, we have mainly focused on issues of becoming involved, thereby leaving a significant gap in our knowledge. One reason for this has been an ambivalent perception by researchers towards issues concerning and arising from, disengagement. For example, there is the assumption that terrorists and their movements are somehow no longer relevant or deserving of serious, urgent study once their involvement in terrorism has ceased (or the movement has entered a cease-fire or peace process). Yet it is precisely at this phase that systematic analyses of these issues represent a realistic objective for researchers. Despite perennial difficulties in gathering reliable data, former terrorists are most likely at this phase to be willing to speak to researchers and provide valuable documentation. Since early 2006, I have been conducting interviews with former members of over a dozen radical and extremist movements around the world. Though the interviews often arose in a practical sense from extensive planning and meticulous preparations, the important point is that former terrorists are willing to speak about their experiences if one asks the correct questions.

In thinking about disengagement from terrorism, there are various questions we need to answer in relation to why people leave terrorist movements, what happens to people who leave terrorism, what influences them to leave, and how leaving can influence response strategies. Indeed, a broader issue here, and one that is especially relevant given the complexity of what analyses of 'becoming involved' seems to suggest, concerns what we mean by 'disengaging' or 'leaving'. Disengaging might suggest critical cognitive and social changes, in terms of leaving behind the shared social norms, values, attitudes and aspirations so carefully forged while the individual was still a member of a terrorist network. On the other hand it might indicate some continued adherence to these values and attitudes, and engaging in some other socially relevant 'support' behavior but no longer engaging in actual terrorist operations.

Perspectives from psychology would appear relevant at this point. An interesting, though brief, report by Renee Garfinkel [2] outlines significant factors involved in the psychological transformation inherent in moving from involvement in violence to non-violent activity. Garfinkel conducted a series of telephone interviews with former members of militant groups, all of whom were described as "now working for peaceful change" (p.3). Garfinkel describes this transformation as taking place against a backdrop of vulnerability, catalysed often by stress, crisis and trauma. Garfinkel describes how some of the interviewees (religious extremists in particular) underwent what might be conceptualised as a form of de-radicalization – "a reorientation in outlook and direction". A key factor, Garfinkel notes, in the transformation from violence to peace was the role played by personal relationships: "change often hinges on a relationship with a mentor or friend who supports and affirms peaceful behavior" (p.1).

A broader consideration of the process of 'exiting' roles comes from Helen Ebaugh. [3] Her study incorporated a systematic examination of the accounts of 185 people, each of whom underwent significant role change. Her sample comprises ex-convicts, ex-alcoholics, ex-doctors, ex-nuns, and others who experienced voluntary "role exit processes". Key commonalities included a sense of disillusionment with the individual's current persona or identity, an attempt to identify and locate an alternative, more satisfactory role, particular triggering factors that facilitate final decisions to leave the role, to finally, the creation of a new identity as an "ex-".

Some of these processes can be found in accounts of involvement in terrorist activity. [4] Two immediate sets of disengagement factors that may be described both as factors leading to, and expressions of, disengagement include psychological issues and physical issues.

Psychological, or emotional issues leading to disengagement from terrorism may include the following:

  1. The development of negative sentiments as a result of experiencing negative qualities associated with sustained, focused membership (e.g. pressure, anxiety, the gradual dismantling of the fantasy or illusion that served to lure the recruit in the first place etc.) and as a result;
  2. A change in priorities (e.g. the longing for a social/psychological state which (real or imaginary) to regain something that the member feels is lacking, or existed before membership, often a result of self-questioning, but mostly following prolonged social/psychological investment as a member from which little return appears evident);
  3. A sense of growing disillusionment with the avenues being pursued, or some quality of them (e.g. with the political aims or with operational tactics and the attitudes underpinning them).

Physical disengagement from terrorism is slightly different in that while there is change, it is externally identifiable. Involvement in even the smallest of terror networks is not discrete or static – it is a constantly changing process of engagement in and occupation with an array of different activities that might seem unconnected, and often may entail fulfilment of more than one role (both at any one time, or over time).

What could be said to constitute physical disengagement from terrorism might involve any of the following, none of which should be considered exclusive:

  • Apprehension by the security services, perhaps with subsequent imprisonment (or if not, forced movement by the terrorist leadership of the member into a role whereby he or she is less likely to risk arrest);
  • Forced movement into another role, for example, as a result of disobeying orders: at the very least ostracism may occur, if not outright execution, but if there is some mitigating circumstance the member may instead be pushed into another functional role;
  • An increase in 'other role' activity whereby the original role becomes displaced (e.g. an area of specialization that relates directly to the commission of terrorist offences such as exploiting one's technical acumen by assisting in the preparation of equipment), or increased involvement in political activity (often as a result of imprisonment, which, ironically for some represents a final consolidation of communal identity);
  • Being ejected from the movement (e.g. for improper use of arms, money, etc. or some disrespectful behaviour that warrants dismissal but not execution);
  • As with psychological disengagement, a change in priorities.

The crucial difference between physical and psychological disengagement is that in the case of the former; the terrorist may continue playing a part in the network, but may move into another role/function in order to facilitate new circumstances (e.g. personal, such as getting married or having children, and moving into a support or ancillary role as a result or organizational if there is a deficit in numbers in a particular area of specialization). The person may still continue to engage in 'terrorism'-related behaviours, but not in a direct or obvious way with respect to the immediate behaviors associated with terrorist operations per se, i.e. digging a hole several hours before a roadside bomb is put there by someone else. The other direction from which this role change might emerge is from the leadership, who may place a heavier emphasis on political activity in the months approaching an election. In simple practical terms, this might involve an active terrorist engaging in distributing posters, storing weapons or equipment or helping to organize political rallies. The significance of knowledge about this sense of 'migration' between roles cannot be underestimated in the context of counterterrorism initiatives

To summarise then, individual disengagement from terrorism may be, broadly speaking, the result of an individual or collective process (or some combination of both). It may stem from experiences as diverse as role change, attitudinal shift, and may or may not result in the kind of 'reorientation in outlook' described by Garfinkel. We can identify both psychological and physical dimensions of disengagement. Further complicating matters, we might think of disengagement as broadly voluntary in origin (e.g. when an individual has made a decision that continued membership of the group is no longer as important as some overriding personal issue) or involuntary (e.g. an individual is forced to leave in the face of some external issue such as the reality of decommissioning, or some new legislative initiative, and the implications this has for organizational dissipation), or a combination, for instance in the form of an outright rejection of the group's ideals as a result of a political shift in the group's stance.


Even a cursory examination of accounts suggests that the pathway to disengagement may not necessarily be the same for everyone, nor are the qualities of that disengagement process as experienced by the individual the same for each person. Additionally, of course, the disengaged terrorist may not necessarily be repentant or 'de-radicalised' at all. [5] Often there can be physical disengagement from terrorist activity, but no concomitant change or reduction in ideological support or indeed, the social and psychological control that the particular ideology exerts on the individual.

This issue is very significant, and exploring its implications may well represent the start of a more realistic and comprehensive policy-based exploration of what kinds of challenges disengagement programmes may well need to face. It also raises some serious issues in relation to how risk and danger are conceptualized in this context. In fact, in the sample of individuals I interviewed from 2006-2008, while almost all of the interviewees could be described as disengaged, not a single one of them could be said to be 'de-radicalized'. [6] In fact, even the process of disengagement was highly idiosyncratic for those interviewed. For some, leaving the movement was temporary, with members opting to come back to the movement at some later stage. Sometimes, this was to a different role, otherwise it was a return to the same role or function held before the initial departure.

Given the lack of conceptual development around not only disengagement issues, but from involvement in terrorism more generally, it might appear both naïve and premature to draw generalized conclusions from a discussion about disengagement. However, we do have some very interesting starting points. It is clear that there is significant promise in thinking about the development of some kind of disengagement initiatives. At its heart (and although military responses to terrorism will continue) terrorism will continue to be a social problem, and initiatives informed by psychological principles perhaps not previously considered in a serious way will ultimately warrant much greater consideration in the future.

This, however, is not entirely the same as the rapidly growing phenomenon that is collectively described as 'the de-radicalization programmes' currently developing throughout countries as diverse as the challenges they face.

At present, there is a flurry of activity worldwide aimed at what on the surface appears to be a pursuit of similar objective, but upon closer inspection reveal such diverse conceptual underpinnings that they can only realistically be unified in terms of promoting some kind of move away from terrorism. Yemen, Saudi Arabia, Northern Ireland, Colombia, Indonesia, Malaysia, and Singapore – each of these countries (and many more) has seen the development of its own particular approach to promoting disengagement of some kind from terrorism. Delving deeper, however, they seem to be less about actual deradicalization and more about attempts to promote disengagement and desistance from terrorist activity in some limited sense.

An immediately interesting feature of these programmes is the range of terminology used in the development and implementation. "Desertion", "demobilisation", "defection", "de-escalation", "rehabilitation" and others [7]; while essentially implying a move away from involvement in terrorism, carry within them different, sometimes quite nuanced, assumptions about the way in which disengagement initiatives ought to develop (or be implemented) in different settings, at whatever stage in the terrorist campaign (or level of the moment) they may be directed. The Saudi Arabian government uses the Arabic equivalent of 'rehabilitation'. One of the interviewees in my own sample, a radical Sunni preacher based in Tripoli in North Lebanon, prefers the term 'reforming'. To paraphrase one of my interviews with him, if there is nothing wrong with 'radicalization', then it is offensive and misleading to speak of 'de-radicalization'.

Each of these different programmes has developed in different ways and has attempted to express and be guided by different strategies. According to Tore Bjørgo, [8] some focus on imprisoned terrorists, while others target terrorists that are currently involved in illegal activity. Some try to reduce the size of the terrorist movement by reducing the number of active participants (e.g. the Child Combatant Programme as part of the Ministry of Interior and Justice's Reincorporation Programme in Colombia), and may require only an apparent shift in attitude and behavior away from violent extremism as the precondition for an individual radical to be released from prison.

As Bjørgo describes, other approaches provide amnesty only in exchange for testimony against former colleagues in the terrorist group. The historical precedent to this is well known. In Italy, members of the Red Brigades, as well as the Mafia, were offered reduced sentencing or even amnesty if they agreed to testify against their comrades. This system was considered an important tool in helping disrupt both the Red Brigades and the Mafia in the 1970s, 1980s, and early 1990s, but abuses including false testimonies and conviction of innocent persons undermined the system, and the system, as with the Supergrass trials in Northern Ireland, become generally seen as largely unfavorable.

One of the first disengagement programmes to emerge was the "Committee for Dialogue" initiative in Yemen. The scheme, now discontinued, remains infamous for reasons that bear little resemblance to the internal logic that underpinned the design of that programme. First described in detail by Michael Taarnby, [9] the scheme involved a simple process whereby Islamic clerics enter into dialogue with imprisoned jihadists in an effort to challenge the basis of the ideologies to which they have been exposed in an effort to undo the process of identification with that ideology.

In some ways, the Yemen initiative was the spark that lit the fire. Elements of that programme were exported to other countries, with little regard for its limitations or context-bound nature. Though recriminations and accusations of bad faith and mistrust now surround much of the external commentary on the Yemen initiative, the programme raised some challenging questions – can a programme be designed to 'change' terrorists or their behavior? On the surface, the Yemen (and now in particular, the Saudi Arabian rehabilitation programme) initiative perhaps most closely resemble what one would expect of 'deradicalization': a softening of views, an acceptance that the individual's pursuit of his objectives using terrorism were illegitimate, immoral and unjustifiable.

Part of the acrimony that began with the Yemen initiative, and now increasingly surrounding the Saudi Arabian effort arose from a failure to answer the obvious question – do these programmes actually work? And if so, how? And why?

There is an old joke that psychology professors tell their students – how many psychologists does it take to change a light bulb? To which the answer is: only one, but that bulb really has to want to change. Though a silly illustration, it highlights at once the enormously complex issue at the heart of what is implicitly conveyed in these programmes – the idea of a quick fix, 'de-radicalizing', undoing the gradual, incrementally experienced process of social and behavioral learning that has culminated in this individual person engaging in terrorist activities. But can this actually be done in the ways that such initiatives suggest? The answer to this is, very simply, no – at least not for the reasons assumed in some of the programmes. We can certainly develop mechanisms for facilitating or promoting disengagement, and often using a variety of innovative and imaginative interventions at specific phases – using former terrorists to highlight negative aspects of the reality of 'being involved', disrupting networks by providing incentives that result in substituted and displaced activity on the part of individual network members and so on, but we should take a long hard look at claims to change or control terrorist behavior in any context.

Right now, we should at least agree on one thing - it is too difficult to draw anything more than generalizations from these initiatives. Access to data is difficult (even when data exists), and there has been little by way of dialogue about the assumptions that underpin the criteria for success (or failure) in these programs. The point here is not to be negative but to highlight the fact that we have to now begin to critically examine the basis of what we think 'deradicalization' is, and what efforts to promote this can actually deliver in practice as a purportedly effective means of changing terrorist behavior.

Obviously, there can be no ideal template for what might be assumed of any of these programmes. If the development of terrorism is a product of its own time and place, it follows that issues of disengagement (and all that that implies) will also be context-specific and necessarily nuanced both in terms of how the programmes are constructed, implemented, and promoted in different countries, and for different reasons. Different perceptions about broader issues of disengagement will exist not only between and within the individual terrorist group, but also how the issue may be differentially perceived within the constituent population that is represented (or otherwise) by the terrorist movement.

Moving beyond the confusion brought about by the potentially misleading term deradicalization, there are more promising avenues we must consider. The promotion of disengagement (in whatever way, and at whatever level) will necessarily have to be tailor-made to not only the specific movement in question, but will need to be carefully positioned within the context of the specific socio political or other issues experienced by the non-state movement and opposing regime at any particular moment in time. As above, disengagement from terrorism, because of the broad factors that may facilitate it from an individual perspective, may not necessarily result in the emergence of a "repentant community" of terrorists becoming integrated into the community it previously claimed to represent. While some minor efforts have developed in Northern Ireland to initiate restorative justice programmes between terrorists and families of their victims, these have been largely unsuccessful and, in some cases, have led to the development of significant tensions.

In Colombia, an array of legal and judicial issues put in place in the context of 'demobilization' have attempted to facilitate the disengagement not just existing prisoners, but active members of FARC also. To facilitate exit from FARC, the Colombian judiciary essentially 'suspends' the militant's trial in an attempt to encourage and sustain their 'demobilization'. If (as happened extensively in Northern Ireland), the militant engages in illegal activity of any kind, they are subsequently forced to face trial for their original offences. However, as with the Northern Irish case, there is no 'blanket' effect – the extent to which the 'amnesty' is used is usually a function of the illegality of the offence. Therefore, a person who has been directly involved in unambiguous terrorist activity (i.e. shooting or bombing operations) will necessarily be treated more harshly than a person who is involved in fringe activity (that, if continued, may or may not eventually lead to increased involvement in the former).

The continued success of the Colombian initiative, however, is in part due to the integrated systemic approach taken by a variety of agencies. In the very formulation of the scheme, as well as its obvious implementation, the programme for reintegration invokes comprehensive expert legal advice, relevant input and resources from psychosocial support input (to reflect re-insertion difficulties), police monitoring (for the reasons given above, as well as to ensure that the 'demobilization' from the AUC, or FARC etc. is actually meaningful), and finally ongoing research evaluation into the progress of the scheme and its adherents will be vital. While the Colombian initiative is unique in many respects, it is this systematic approach and ongoing evaluative effort that may mark its lasting contribution with respect to inspiring other similar initiatives to develop elsewhere.


The basis of this article began with a simple premise – there is no evidence to suggest that disengagement from terrorism may result in deradicalization. For this and other reasons, particularly the growth of these diverse programmes, it may be worthwhile to seek greater conceptual clarity between the two. While there are a number of ways in which we can do this, one immediate objective is to engage in research to evaluate, through multiple means, the increasing prevalence of the deradicalization programmes worldwide. The otherwise unverified and often spurious claims to success by the proponents of some of these programmes does not mean that we should dismiss their claims, rather we should engage in efforts to ask one of the most basic questions in all counterterrorism: do they actually work (and why?). The need for multi-attribute evaluation studies of each and every disengagement and deradicalization initiative would be a welcome start to greater clarity around these issues. Critically examining claims for success in each case, as well as an examination of the ways in which the criteria for success are conceptualized and measured in each setting would be a useful step. It is useful not only to help identify policy-relevant lessons for the development of counterterrorism initiatives of this kind in regions where there are none, but more immediately would lead to greater clarity around the basic social and behavioral processes involved, that have, it would seem, escaped scrutiny in many cases to date.

About the Author: John Horgan is Director of the International Center for the Study of Terrorism at Penn State. His forthcoming book Walking Away from Terrorism: Accounts of Disengagement from Radical and Extremist Movements will be published by Routledge at the end of 2008.


[1] Horgan, John, The Psychology of Terrorism. London: Routledge, 2005.

[2] Renee Garfinkel (2007), 'Personal Transformations: Moving From Violence To Peace', United States Institute of Peace Special Report 186, April.

[3] Helen Rose Fuchs Ebaugh (1988) Becoming an Ex: The Process of Role Exit (Chicago: University of Chicago Press).

[4] See Horgan, note 1.

[5] Most of the explicitly disengaged terrorists interviewed by Horgan (across ethno-nationalist, Jihadist, and other religious terrorist movements in Europe, the Middle East and South-East Asia) cannot be said to be 'de-radicalized' per se, even in some cases despite a clear, and acrimonious departure from their movement – see John Horgan (forthcoming) Walking Away From Terrorism: Accounts of Disengagement from Radical and Extremist Movements, London: Routledge.

[6] Horgan, Walking Away from Terrorism.

[7] Ibid.

[8] Bjorgo, T. (2005). 'Reducing Recruitment and Promoting Disengagement from Extremist Groups: The Case of Racist Sub-Cultures', in C. Benard (Ed.) A Future For The Young: Options for Helping Middle Eastern Youth Escape the Trap of Radicalization. A RAND Working Paper WR-354. Washington, D.C.: RAND National Security Research Division.

[9] Michael Taarnby, Yemen's Committee for Dialogue: the relativity of a counter terrorism success, in C. Benard (Ed.) (2005). A Future For The Young: Options for Helping Middle Eastern Youth Escape the Trap of Radicalization. A RAND Working Paper WR-354. Washington, D.C.: RAND National Security Research Division.


Perspectives on Terrorism is  a journal of the Terrorism Research Initiative and the Center for Terrorism and Security Studies

ISSN  2334-3745 (Online)

Disclaimer, Terms and Conditions