Prisons and Terrorism
Radicalisation and
De-radicalisation in 15 Countries

A policy report published by
the International Centre for
the Study of Radicalisation
and Political Violence (ICSR)
In partnership with the
National Consortium for
the Study of Terrorism and
Responses to Terrorism
(START)

ABOUT THIS REPORT

The author of this report is Peter R. Neumann.
Its empirical basis are 15 country reports that
were written by the following experts: Omar
Ashour (Algeria and Egypt); Laila Bokhari
(Pakistan); Chris Boucek (Saudi-Arabia and
Yemen); Andrew Coyle (United Kingdom); Boaz
Ganor and Ophir Falk (Israel); Rohan Gunaratna
(Singapore); Bob de Graaff and Eelco Kessels
(Netherlands); Arie Kruglanski and Michelle
Gelfand (Philippines); Jean-Luc Marret (France);
Sidney Jones (Indonesia); Marisa Porges
(Afghanistan); Manuel Torres (Spain); and
Bert Useem (United States).
The project could not have been undertaken
without the kind and very generous support of:
•	 The Department of Foreign Affairs
and Trade, Australia
•	 The National Coordinator for
Counterterrorism, The Netherlands
•	 The Office for Security and Counterterrorism,
Home Office, United Kingdom
The views expressed in this report do not
necessarily reflect the opinions of any of the
above mentioned experts, partners or sponsors.

CONTACT DETAILS

For questions, queries and additional copies
of this report, please contact:
ICSR
King’s College London
138 – 142 Strand
London WC2R 1HH
United Kingdom
T. + 44 20 7848 2065
F. + 44 20 7848 2748
E. mail@icsr.info
Like all other ICSR publications, this report can
be downloaded free of charge from the ICSR
website at www.icsr.info.
© ICSR 2010

Summary
Introduction
•	 This report offers a wide-ranging analysis of the role prisons can
play in radicalising people – and in reforming them. In doing so, it
examines the policies and approaches of 15 countries, identifying
trade-offs and dilemmas but also principles and best practices
that can help governments and policymakers spot new ideas and
avoid costly and counterproductive mistakes.
•	 Prisons matter. They have played an enormous role in the
narratives of every radical and militant movement in the modern
period. They are ‘places of vulnerability’ in which radicalisation takes
place. Yet they have also served as incubators for peaceful change
and transformation.
•	 Much of the current discourse about prisons and radicalisation is
negative. But prisons are not just a threat – they can play a positive
role in tackling problems of radicalisation and terrorism in society
as a whole. Many of the examples in this report demonstrate how
prisons can become net contributors to the fight against terrorism.

Prison Regimes for Terrorists
•	 Terrorists are not ‘ordinary’ offenders. They often use their time in
prison to mobilise outside support, radicalise other prisoners, and
– when given the opportunity – will attempt to recreate operational
command structures.
•	 There are no hard and fast rules about whether terrorist prisoners
should be concentrated together and/or separated from the rest of
the prisoner population. Most of the countries that were included in
the sample practice a policy of dispersal and (partial) concentration,
which distributes terrorists among a small number of high security
prisons. Even within such mixed regimes, however, it rarely seems
to be a good idea to bring together leaders with followers and mix
ideologues with hangers-on.
•	 The ‘security first’ approach of most countries results in missed
opportunities to promote reform. Many prison services seem to
believe that the imperatives of security and reform are incompatible.
In many cases, however, demands for security and reform are more
likely to complement than contradict each other.
•	 Prison services should be more ambitious in promoting positive
influences inside prison, and develop more innovative approaches
to facilitate prisoners’ transition back into mainstream society.

1

Prison Radicalisation
•	 Prisons are often said to have become breeding grounds for
radicalisation. This should come as no surprise. Prisons are ‘places
of vulnerability’, which produce ‘identity seekers’, ‘protection
seekers’ and ‘rebels’ in greater numbers than other environments.
They provide near-perfect conditions in which radical, religiously
framed ideologies can flourish.
•	 Over-crowding and under-staffing amplify the conditions that lend
themselves to radicalisation. Badly run prisons make the detection
of radicalisation difficult, and they also create the physical and
ideological space in which extremist recruiters can operate at
free will and monopolise the discourse about religion and politics.
•	 Religious conversion is not the same as radicalisation. Good
counter-radicalisation policies – whether in or outside prison
– never fail to distinguish between legitimate expression of faith
and extremist ideologies. Prison services should invest in staff
training, and consider sharing specialised resources (for
example, translators).
•	 In the case of Islamist militant radicalisation, prison imams have
an important role to play, but they are not a panacea. Their
independence and credibility need to be protected. It is neither
reasonable nor realistic to expect them to be spiritual advisers,
welfare officers and terrorism experts all at the same time.

Collective Disengagement and De-Radicalisation
•	 In several cases, the (imprisoned) leaders of terrorist groups
have used their influence to pursue collective processes of
de-radicalisation and disengagement that have resulted in the
standing down of entire armed groups. However, such instances
are rare, and the conditions in which they are likely to happen
are limited.
•	 Collective disengagement and de-radicalisation is predicated
upon the existence of strong and authoritative leaderships; the
existence of hierarchical command and control structures; and,
not least, a conducive environment in which the leadership
perceives that the armed campaign has faltered or is seen to
have become less useful than other means of contention.
•	 Where such processes are possible, governments should
support them. This may involve: opening appropriate lines of
communication; providing the right mix of ‘carrots and sticks’
to keep the process on track; and – most importantly – enabling
prisoners’ release and facilitating their integration back into
mainstream society.
•	 When political concessions form part of the ‘deal’ between
the government and the terrorists, collective disengagement
and de-radicalisation can form the basis for a fully-fledged
peace process, and requires the necessary skills, resources,
and – above all – patience.

2

Individual Disengagement and De-radicalisation
•	 Individual de-radicalisation and disengagement programmes
aim to reduce the number of active terrorists in a given society
by helping individual terrorists abandon terrorism and easing their
re-integration into mainstream society.
•	 One of the biggest difficulties lies in measuring their success, as
different local contexts, rules on eligibility and the short period for
which they have run, have produced data that is difficult to judge
and nearly impossible to compare.
•	 As long as the political momentum is no longer with the insurgents
and other external conditions are conducive, their impact can
be significant. Even in the best circumstances, however, they
complement rather than substitute other instruments in the fight
against terrorism.
•	 The report identifies the key underlying drivers and principles of
individual disengagement and de-radicalisation programmes, which
may help policymakers understand the phenomenon and identify
elements of best practice. They are:
•	 A mix of different kinds of programming, typically combining
ideological and/or religious re-education with vocational training.
•	 Credible interlocutors, who can relate to prisoners’ personal and
psychological needs.
•	 Emphasis on prisoners’ transition back into mainstream society,
typically by providing them with the means for a new beginning
and by establishing social networks away from extremism.
•	 Sophisticated methods for locking prisoners into multiple
commitments and obligations towards family, community, and
the state.
•	 Material inducements, which – while useful – do not seem to be
decisive on their own.
•	 The positive and outward-looking approach of individual
disengagement and de-radicalisation programmes is to be
commended and should serve as an inspiration for policymakers
and prison authorities all over the world.

3

3

4

Contents
1 Introduction	
1.1 Argument	
1.2 Methodology	
Scope
Methods
Structure

7
8
8

2 Prison Regimes for Terrorists	
2.1 Issue	
Terrorists in prison
Differences between terrorists
Challenges

13
13

2.2 Practice	
Concentration, separation, and isolation
Containment versus reform
After-care

17

2.3 Recommendations	

21

3 Prison Radicalisation	
3.1 Issue	
How big a problem?
What does it look like?
Challenges

25
25

3.2 Practice	
General conditions
Monitoring and Training
Prison imams
Other external influences

29

3.3 Recommendations	

35

4 Collective De-radicalisation and Disengagement	
4.1 Issue	
Pre-conditions
Challenges

39
39

4.2 Practice	
Leadership and environment
Repression
Inducements

41

4.3 Recommendations	

45

5

5

5 Individual De-radicalisation and Disengagement	
5.1 Issue	
Debates
Challenges

47
47

5.2 Practice	
Programming
Interlocutors
Inducements
External stakeholders
After-care
Environment

50

6 Conclusion	

59

Acknowledgements 	
Further Reading	

61
63

Tables and Boxes	
Table 1: Country Samples 	
Box 1: Key Terms and Definitions 	

11
12

6

Prisons and Terrorism Radicalisation and De-radicalisation in 15 Countries

1 Introduction

I

n the debate about radicalisation and violent extremism,
prisons are often described as ‘hotbeds’ of terrorism. The
American criminologist Harvey Kushner, for example, argues
that Western prisons are one of the main recruitment grounds for
Al Qaeda. 1 So did the late Ian Cuthbertson, who was certain that
Al Qaeda had ‘taken full advantage’ of Western prisons and their
‘relatively lax practices’. 2 An American policy report, published
in 2006, which drew on the insights and opinions of a 15-strong
expert panel, arrived at similar conclusions, stating that prisoner
radicalisation posed ‘a threat of unknown magnitude to the
security of the [United States]’. 3
How true are such claims, and what should be done? This report is
the first to examine government policies on prison radicalisation and
de-radicalisation in 15 countries across the globe. Drawing on the
research carried out by more than a dozen experts, it conceptualises,
examines and compares the policies and practices of these countries
and their prison systems. Avoiding sensationalism, it aims to offer
commonsense principles and practical recommendations which may
help to reduce radicalisation and, more importantly, turn prisons into
net contributors in the struggle against terrorism.
Whether or not one believes that prisons have become Al Qaeda’s
‘universities’ or ‘finishing schools’, 4 there can be no question that
prisons matter:
•	 Prisons have played an enormous role in the narratives of every
radical and militant movement in the modern period. No matter
how different their causes or backgrounds, Egyptian Islamists,
German Marxists and Irish Republicans have all regarded their
comrades’ imprisonment as traumatic turning points in the histories
of their movements. The prisoners and the ways they were treated
came to be focal points for their groups’ campaigns, and they
significantly influenced their supporters’ attitude towards violence
and the state.
•	 Prisons are highly unsettling environments in which individuals are
more likely than elsewhere to explore new beliefs and associations.
Confronted with existential questions and deprived of their
existing social networks, prisoners with no previous involvement
in politically motivated violence are vulnerable to being radicalised
and recruited into terrorism. Prisons, therefore, are ‘places of
vulnerability’ in which radicalisation can take place.5

1	
2	
3	
4	
5	

Harvey Kushner (with Bart Davis), Holy War on the Home Front (New York: Sentinel, 2004).
Ian M. Cuthbertson, ‘Prisons and the Education of Terrorists’, World Policy Journal, 22 September 2004,
p. 16.
‘Out of the Shadows: Getting Ahead of Prisoner Radicalization’, A Special Report by The George
Washington University Homeland Security Policy Institute and the University of Virginia Critical Incident
Analysis Group, September 2006, p. i.
See ‘Al Qaeda’s British Propagandists on Trial’, Jihadwatch, April 2007; available at
http://www.jihadwatch.org/2007/04/al-qaedas-british-propagandists-on-trial.html.
Peter Neumann and Brooke Rogers, ‘Recruitment and Mobilisation for the Islamist Militant Movement in
Europe’, European Commission, Directorate General for Justice, October 2007. Also Peter R. Neumann,
‘Joining Al Qaeda: Islamist Militant Recruitment in Europe’, Adelphi Paper 399, International Institute for
Strategic Studies, January 2009, Chapter 3.

7

Prisons and Terrorism Radicalisation and De-radicalisation in 15 Countries

•	 At the same time, prisons have on many occasions been
incubators for peaceful change and transformation. Whether
through individual de-radicalisation and disengagement or as
basis for entire peace processes, prisons have made significant
contributions towards reversing the process of radicalisation
and undermining terrorist campaigns on the outside. Prisons,
in other words, have served as engines for positive change
whose impact has been felt far beyond the prison walls.
The main part of this report will investigate each of these issues,
looking – in turn – at prison regimes for terrorists; policies to prevent
prisoner radicalisation; (prison-centred) collective de-radicalisation;
and (prison-centred) individual de-radicalisation and disengagement
programmes. The intention is to understand key dynamics and tradeoffs; compare different government policies and programmes; and
make practical suggestions for how prisons can play a positive role
in countering terrorism.

1.1 Argument
If there is an overall thread – or underlying message – that
connects the different parts of the report and the many different
national experiences it reflects, it is the idea that prisons can present
opportunities for combating radicalisation and terrorism, and that their
potential for doing so needs to be fully understood and acted upon.
Much of the current discourse about prisons and radicalisation is
negative. For policymakers, prisons are of little interest or concern
unless prisoners escape or cause unrest, while prison authorities are
mostly concerned about making sure that their institutions do not
attract negative attention. As a consequence, many governments’
efforts revolve around overly defensive notions of ‘reducing’ or
‘preventing’ radicalisation in prisons, but rarely go beyond the basic
aim of containing the threat within the prison walls.
None of the de-radicalisation or disengagement programmes in
the Middle East or South East Asia is perfect. Still, some of their
underlying principles may be relevant and applicable elsewhere.
More fundamentally, what they demonstrate is that prisons are not
just a threat but can make a positive contribution to tackling problems
of radicalisation and terrorism in society as a whole: the positive
and outward-looking approach that is exhibited in several of these
programmes should serve as an inspiration for policymakers and
prison authorities all over the world.

1.2 Methodology
Before delving into the analysis, it is important to explain what this
study has looked at and how the research was carried out.

Scope
The report is based on 15 country reports that were researched
and written by external contributors (see below). The countries were
selected because of their significance in relation to the subject, but
also to ensure a good representation of different regions, continents

8

Prisons and Terrorism Radicalisation and De-radicalisation in 15 Countries

and types of prison system. Not every country with a history of
prison radicalisation could be included in the ‘sample’, but most of
the relevant dynamics and national experiences have been captured.
While much of the recent experience with prison radicalisation and
de-radicalisation relates to Islamist militants, the investigation has not
been limited to this particular expression of the phenomenon. On the
contrary, the report will show that valuable lessons can be learned
from comparing different kinds of terrorism and how they have
played out in the prison environment. In fact, several of the countries
contained that were part of the sample – France and Spain, for
example – have had to cope with strikingly different types of terrorist
prison populations at the same time.
It should also be pointed out that the recommendations in this
report are focused on prison policy, not the micro-management of
prisons. The author is in no position to tell prison staff how many
hours a day to keep prisoners locked up in their cells, or how best
to prevent mobile phones from being smuggled in. As a result,
most of the attention has been directed at the principles that are
underpinning different countries’ policies and programmes and how
they are implemented. While the report is by no means an operational
handbook, it may nevertheless prove valuable for policymakers and
prison officials who are hoping to understand the key dynamics and
trade-offs that inform operational procedures.

Methods
The report is part of a project that was carried out between May
2009 and May 2010 by the International Centre for the Study of
Radicalisation and Political Violence (ICSR) in partnership with the
National Consortium for the Study of Terrorism and Responses
to Terrorism (START), based at the University of Maryland. Its
empirical basis are 15 country case studies that were written by
country experts who were chosen after careful deliberation.
They are:
•	
•	
•	
•	
•	
•	
•	
•	
•	
•	
•	
•	
•	

Omar Ashour (Algeria and Egypt);
Laila Bokhari (Pakistan);
Chris Boucek (Saudi-Arabia and Yemen);
Andrew Coyle (United Kingdom);
Boaz Ganor and Ophir Falk (Israel);
Rohan Gunaratna (Singapore);
Bob de Graaff and Eelco Kessels (Netherlands);
Arie Kruglanski and Michelle Gelfand (Philippines);
Jean-Luc Marret (France);
Sidney Jones (Indonesia);
Marisa Porges (Afghanistan);
Manuel Torres (Spain); and
Bert Useem (United States).

Not all 15 countries were relevant in relation to each area of research
(see Table 1), but each author was asked to structure their case study
in a similar way, thus making sure that findings from different case
studies could be compared.
The countries that were looked at for their prison regimes for
terrorists were France, the Netherlands, Spain, the United Kingdom,

9

Prisons and Terrorism Radicalisation and De-radicalisation in 15 Countries

and the United States. Policies and approaches to deal with prison
radicalisation were analysed in Afghanistan, France, the Netherlands,
Pakistan, the Philippines, Spain, the United Kingdom, and the United
States. The analysis of collective disengagement and de-radicalisation
is based on the experiences of Algeria, Egypt, and Israel. For
individual disengagement and de-radicalisation, the report looked at
Afghanistan, Indonesia, the Philippines, Saudi-Arabia, Singapore,
and Yemen.
While the analysis and recommendations expressed in this report do
not necessarily reflect the individual views or any consensus among
the contributing experts, their collective insight into the phenomenon
– often based on years, if not decades, of study of the countries
in question – is its unique strength. Unlike other studies of prison
radicalisation, the report achieves depth as well as breadth, drawing
on detailed and highly informed assessments of past and current
policies in countries as different as Indonesia and Israel.
In the empirical part of the paper, where specific countries are
referred to, readers should assume that the information was gleaned
from the relevant expert’s country report unless a footnote and/
or reference points to a different source. (All expert papers will be
published in full length as part of an edited volume in 2011.)
In addition to the 15 case studies, the author – assisted by ICSR
staff – has carried out an extensive review of the literature on prison
radicalisation and de-radicalisation, which has made it possible to
contextualise some of the overarching themes and close gaps in
the understanding of concepts and theories. The report also draws
on many – often informal – conversations with other researchers,
representatives from prison services, and policymakers and
their staff.
Not least, the author has benefited from attending several
international conferences on the topic, including a workshop on
prison radicalisation and de-radicalisation at King’s College London
in November 2009, which ICSR and START organised in order to
provide expert contributors with an opportunity to present the first
drafts of their papers.

Structure
The structure of the report follows the questions and issues that
were set out earlier:
Chapter 2 deals with prison regimes for offenders who are held for, or
have been convicted of, terrorism related charges, concentrating on
those countries in which specific policies exist or have been outlined.
Chapter 3 looks at the issue of prisoner radicalisation, exploring the
conditions under which non-terrorism related (that is, ‘ordinary’)
prisoners may be exposed to radicalisation and what has been done
to counter this.
Chapter 4 explores the potential and limits of collective
de-radicalisation and disengagement.

10

Prisons and Terrorism Radicalisation and De-radicalisation in 15 Countries

Chapter 5 examines past and ongoing efforts at promoting
individual de-radicalisation and disengagement within prison,
identifying similarities and differences between programmes
as well as establishing factors that contribute to
their effectiveness.
Chapter 6 sums up the findings and highlights key
recommendations that were identified in the course
of the research.

TABLE 1 COUNTRY SAMPLES
PRISON
RADICALISATION

COLLECTIVE DERADICALISATION

INDIVIDUAL DERADICALISATION

France

Afghanistan

Algeria

Afghanistan

Netherlands

France

Egypt

Indonesia

Spain

Netherlands

Israel

Philippines

United Kingdom

Pakistan

Saudi-Arabia

United States

Philippines

Singapore

Spain

Yemen

PRISON REGIMES
FOR TERRORISTS

United Kingdom
United States

11

Prisons and Terrorism Radicalisation and De-radicalisation in 15 Countries

BOX 1 Key terms & Definitions
RADICALISATION Most of the definitions currently in circulation describe
radicalisation as the process (or processes) whereby individuals or groups
come to approve of and (ultimately) participate in the use of violence for
political aims. Some authors refer to ‘violent radicalisation’ in order to
emphasise the violent outcome and distinguish the process from non-violent
forms of ‘radical’ thinking.
DE-RADICALISATION AND DISENGAGEMENT The terms de-radicalisation
and disengagement describe processes whereby individuals (or groups)
cease their involvement in organised violence and/or terrorism. However,
while de-radicalisation aims for substantive changes in individuals’ (or
groups’) ideology and attitudes, disengagement concentrates on facilitating
behavioural change, that is, the rejection of violent means. According to
John Horgan, ‘the disengaged terrorist may not be “deradicalized” or
repentant at all. Often physical disengagement may not result in any
concomitant change or reduction in ideological support’. 6 Additionally, many
authors distinguish between collective and individual de-radicalisation and/or
disengagement, depending on whether the process is led by, or aimed
at, individuals or entire groups. 7
POLITICALLY MOTIVATED OFFENDERS The principal difference between
politically motivated offenders and ‘ordinary’ criminals lies in their intention.
While ‘ordinary’ criminals commit crimes in pursuit of selfish and/or personal
goals, politically motivated offenders believe that they are acting on behalf of
a certain group, society or humanity as a whole. Politically motivated offenders
commonly distinguish between ‘legality’ and ‘legitimacy’, arguing that breaking
the law is justified when a particular policy or the entire political or legal system
are illegitimate. 8 Not all politically motivated offenders are terrorists, but all
terrorists are politically motivated offenders.
TERRORISM The definitional problem, which has haunted terrorism
research for decades, is unlikely to be resolved by this report. As a tactic,
terrorism typically consists of symbolic acts of violence which are intended to
influence the political behaviour of a target group via the deliberate creation of
fear. 9 The formula which many governments and international organisations
have chosen to adopt describes terrorism as politically motivated violence that
intentionally targets civilians and/or non-combatants. 10 For the purposes of
this report, the term terrorist is used for all individuals who have been charged
with, or convicted of, offences that are proscribed under their countries’
anti-terrorism laws.
EXTREMISM The term can be used to refer to political ideologies that
oppose a society’s core values and principles. In the context of liberal
democracies, this could be applied to any ideology that advocates racial or
religious supremacy and/or opposes the core principles of democracy and
human rights. However, the term can also be used to describe the methods
through which political actors attempt to realise their aims, that is, by using
means that ‘show disregard for the life, liberty, and human rights of others’. 11
Many governments refer to terrorists as ‘violent extremists’.
6	

John Horgan, ‘Individual disengagement: a psychological analysis’ in Tore Bjørgo and John Horgan (eds.),
Leaving Terrorism Behind (London and New York: Routledge 2009), p. 27.
7	 See, for example, Omar Ashour, The Deradicalization of Jihadists (London: Routledge, 2009).
8	 See Nikos Passas, ‘Political Crime and Political Offender: Theory and Practice’, Liverpool Law Review, 8(1) (1986).
9	 See Peter Neumann, Old and New Terrorism (Cambridge: Polity Press, 2009), Chapter 1.
10	 Ibid.
11	 Roger Scruton, The Palgrave Macmillan Dictionary of Political Thought, 3rd ed. (Basingstoke: Palgrave Macmillan 2007).

12

Prisons and Terrorism Radicalisation and De-radicalisation in 15 Countries

2 Prison Regimes for Terrorists

F

or Gudrun Ensslin, one of the founders of the West German
Red Army Faction, life in prison was a microcosm of the
injustices, inequalities and oppression which she believed
characterised the capitalist system as a whole. Responding to a
letter from the father of her child, she wrote that ‘nowhere else do
the workings of the system become more obvious [than in prison]:
the two classes – those with keys, and those without’. 12 Her
attitude – especially the acute sense that the prison system had
to be fought and struggled against – typifies one of the ways in
which terrorists and other politically motivated offenders respond
to being incarcerated.
Constructing prison regimes for people like Ensslin is not a new
challenge, nor is it confined to particular expressions of terrorism.
Governments everywhere have had to ask themselves how
terrorists are different from ‘ordinary’ prisoners and what this means
for how they should be treated in prison. Their experiences have
resulted in prison regimes that mirror countries’ particular histories
and experiences with extremism, but also bring out many common
and recurring problems.
This chapter argues that prison regimes for terrorists need to be
informed by a sophisticated understanding of the motivations and
behaviours of politically motivated offenders, who – unlike ‘ordinary
prisoners’ – may want to mobilise outside support, radicalise
other prisoners, and (in the case of terrorists) recreate operational
command structures. With some notable exceptions, the five
countries that were examined – France, the Netherlands, Spain,
the United Kingdom, and the United States 13 – have addressed
these challenges in similar ways. Most practice a policy of dispersal
and (partial) concentration, which distributes terrorists among a
small number of high security prisons. The overall approach can
be characterised as ‘security first’, with little attention or effort
expended on attempts at rehabilitation and reform. Furthermore,
with just one exception, no particular provisions have been made
for ‘after-care’ – sometimes for entirely understandable reasons.
Even so, the chapter argues that the emphasis on containment and
security – however justified and appropriate in many cases – means
that opportunities to pursue reform and rehabilitation are missed.
The analysis shows that it would be mistaken to see a contradiction
between the dual demands for security and reform, and prison
authorities should be more ambitious in pursuing both and at the
same time.

2.1 Issue
In an effort to deny terrorists legitimacy and popular support, one
response of governments all over the world has been to refuse to
12	 ‘Gudrun Ensslin and Bernward Vesper, 11/1/69’ in Gudrun Ensslin and Bernward Vesper, ‘Notstandsgesetze
von Deiner Hand’: Briefe 1968/1969 (Frankfurt: Suhrkamp, 2009), p. 199.
13	 The authors of the country case studies were Andrew Coyle (United Kingdom), Bob de Graaff and Eelco
Kessels (Netherlands), Jean-Luc Marret (France), Manuel Torres (Spain), and Bert Useem (United States).

13

Prisons and Terrorism Radicalisation and De-radicalisation in 15 Countries

acknowledge the political motivation behind their actions. In public
messages, terrorists are labelled ‘criminals’ and their actions are
portrayed as ‘crimes’. The same principle applies to prisons, where
governments are quick to point out that terrorists are dealt with like
all other criminals. Terrorists, in the words of a senior prison official
from Britain, are ‘regard[ed]... as we regard all prisoners’. 14
At first glance, this approach appears to make good sense,
especially when considering negative experiences like the conflict
in Northern Ireland where prisoners were granted what, in essence,
amounted to ‘political status’ and could run their own prison
wings with little interference from the authorities. What it fails to
acknowledge is that, however much governments would like to
deny it, terrorists’ political motivation does make a difference: it
alters their self-perception, behaviours, and – for good reasons and
throughout history – has produced different kinds of prison regimes.
Beginning with the Republican Jacobins in the early 19th century,
most European countries have made ‘exceptions’ for politically
motivated offenders, 15 often based on the assumption that they
were of ‘good character’ and that their crimes – however misguided
– had resulted from an ‘altruistic motivation’. 16 Politically motivated
prisoners were the first ones to be given amnesties or released
once hostilities had ended or particular rights had been granted.
At the same time, during their incarceration, it was common for
such offenders to be separated from ‘non-political’ prisoners,
knowing that the former could exert a ‘subversive’ influence on
the latter – a phenomenon which, in today’s circumstances,
would probably be described as radicalisation. 17
Even today, most criminal justice systems recognise – sometimes
explicitly – that individuals charged with or convicted of terrorism
related offences are different from ‘ordinary’ criminals. They are
investigated by special sections of the police forces and brought to
court by special prosecutors. The laws under which they are charged
– anti-terrorism laws – were created specifically for them, and the
courts at which they stand trial are special (or specialised) courts.
This – governments argue – reflects the particularly serious nature
of their crimes, which are directed against the state and society as
a whole. 18 Yet it also demonstrates that the idea of ‘criminalising’
and ‘de-politicising’ terrorist offences – if taken seriously and fully
internalised – is contradictory at best, and may in fact prevent
prison authorities from developing a sophisticated understanding
of the particular challenges posed by politically motivated offenders,
including terrorists, in prison.

Terrorists in prison
In the prison context, most difficulties in dealing with terrorists
are caused precisely by the fact that these offenders do not see
themselves as criminals. Rather than quietly serving their sentence,
14	 Phil Wheatley, Director General of the Prison Service for England and Wales, cited in Josh Lefkowitz,
‘Terrorists Behind Bars’, NEFA Report, 5 May 2008, p. 12.
15	 See Christine van den Wijngaert, The Political Offence: Exception to Extradition (Deventer: Kluwer, 1980),
pp. 29-30. For the case of the Jacobins, see Michael von Tangen Page, Prisons, Peace and Terrorism
(Houndmills: Macmillan, 1998), p. 25.
16	 Von Tangen Page, Prisons, pp. 5, 31-3, 46.
17	 Ibid., p. 39.
18	 See ibid., p. 46.

14

Prisons and Terrorism Radicalisation and De-radicalisation in 15 Countries

many regard their time in prison as an opportunity to continue the
‘struggle’, and involve themselves in activities ranging from passive
resistance to turning prison into a ‘battleground’ from which to
support the wider campaign. Typical behaviours include: 19
•	 Refusing to cooperate with the prison authorities. Politically
motivated offenders believe that they have been incarcerated for
their beliefs rather than their actions, and that the process that led
to their incarceration is illegitimate. As a result, they may choose
not to cooperate with prison staff, or even attempt to establish
parallel structures.
•	 Developing the movement’s strategy and ideology. Unable to
participate in operations and planning, imprisoned terrorists may
want to use their time in prison to help develop their movements’
strategy or ideology. From Gerry Adams’ Brownie columns to
Mohammed Al Maqdisi’s theories of ‘jihadist’ warfare, some of
the most influential tracts and articles in the history of militant
and/or terrorist movements were written while their authors were
in prison.
•	 Mobilising supporters. Prisons can provide an environment
from which to continue the confrontation with the state, especially
by highlighting and/or exaggerating the kinds of grievances
and injustices that are also fuelling the movements’ campaigns
on the outside. This can involve allegations of discrimination,
mistreatment and torture, and result in negative media coverage,
protests and even dramatic confrontations such as hunger strikes.
•	 Participating in violent campaigns. Terrorists will consider it
their duty to attempt to contribute to their movements’ violent
campaigns. The most obvious way is to try to escape from prison
and rejoin their groups. 20 Where this is impossible, they may seek
to become involved in strategic and operational planning, or issue
operational guidance or authorisation.
Implicit in this description is a further difference between politically
motivated offenders and ‘ordinary’ prisoners, namely that political
offenders are more likely to have a constituency in the form of vocal
supporters on the outside, who – like the offenders themselves –
refuse to accept the legitimacy of their incarceration and regard
them not as perpetrators of crimes but as victims of injustices and
discrimination.
As a consequence, prisons will be subject to a higher level of
public scrutiny;21 that is, the number of people who take an interest
in politically motivated prisoners is certain to go well beyond their
families and immediate associates: it may include thousands of
political supporters – often organised in transnational support
groups – who will observe, respond to and publicise every action
undertaken by the prison authorities.

19	 The typology draws on Andrew Coyle’s case study report on the UK as well as a recently published
RAND Corporation study. See Greg Hannah, Lindsay Clutterbuck and Jennifer Rubin, Radicalization or
Rehabilitation: Understanding the Challenge of Extremist and Radicalized Prisoners (Cambridge: RAND,
2008), p. 46.
20	 Von Tangen, Prisons, p. 32.
21	 Ibid., p. 29.

15

Prisons and Terrorism Radicalisation and De-radicalisation in 15 Countries

Differences between terrorists
The political motivation is what makes terrorists different from
the rest of the prison population. But not all terrorists are the same.
In constructing prison regimes, it is vital to account for the unique
characteristics and behaviours of particular terrorist groups
and networks. 22
For example, terrorist groups differ in relation to their internal
structure and cohesion. Terrorist groups are no longer always
coherent, firmly structured entities, but – like Al Qaeda – they may
constitute loose networks which revolve around personal relationships
rather than military hierarchies. Some of these groups are said to
have implemented the idea of ‘leaderless resistance’ whereby a
movement’s leadership and its followers have no direct contact but
are expected to act on their own initiative and without support from
other parts of their ‘organisation’. 23 When imprisoned, different
members of such networks are unlikely to have met – in many
cases, they will not even have heard of each other. By contrast,
members of more cohesive and hierarchically structured groups like
the Irish Republican Army (IRA) will not only know each other, but
seek to maintain or recreate operational command structures within
the prison walls.
Likewise, not all terrorist groups have the same attitude towards
radicalisation and recruitment inside prison. The IRA, for example,
wanted nothing to do with ‘ordinary’ criminals, who were seen as
unreliable, ill-disciplined and potentially harmful to the group’s
image of a liberation army seeking to attain political objectives. As
a result, its emphasis was on gaining autonomy within prison, with
no deliberate efforts to radicalise and/or recruit among the general
prison population. Many Al Qaeda affiliated prisoners, on the other
hand, see it as their duty to propagate their faith and political
ideology (dawa). They realise that prison constitutes a potentially
fruitful place for conversion and radicalisation (see Chapter 3), and
will consequently exploit whatever opportunities they are offered to
approach other offenders and turn them into followers of the group.

Challenges
The points raised in this section demonstrate that the prison
behaviours of terrorists and other politically motivated offenders are
different from the rest of the prison population. Understanding those
differences helps in conceptualising the particular challenges in
dealing with terrorist inmates. They are:
•	 Preventing the radicalisation of (non-terrorist) inmates;
•	 Preventing the maintenance and/or recreation of operational
command structures; 24
•	 Preventing the exploitation of the prison environment for the
purpose of mobilising outside support.
22	 As in the previous section, this typology draws on Andrew Coyle’s case study report on the UK.
23	 See Marc Sageman, Leaderless Jihad (Philadelphia: University of Pennsylvania Press, 2008); also
Neumann, Old and New, Chapter 3.
24	 The one exception is if the leadership has turned against violence and wants to pursue a process
of collective de-radicalisation and/or disengagement (see Chapter 3).

16

Prisons and Terrorism Radicalisation and De-radicalisation in 15 Countries

In addition, prison regimes for terrorists should also aspire to:
•	 Provide opportunities for de-radicalisation and
disengagement; and
•	 Make a positive contribution to reducing terrorism and
radicalisation on the outside.
If and how governments have managed to reconcile these (sometimes
competing) demands will be explored in the following section.

2.2 Practice
This section uses data from five countries that have no formal
disengagement or de-radicalisation programmes but established
practices, policies or approaches towards dealing with prisoners who
have been charged with, or convicted of, terrorism related offences.
They are: France, the Netherlands, Spain, 25 the United Kingdom
(UK),26 and the United States of America (US). 27
While similar in many respects, it should be noted that comparisons
between the countries in question are not as straightforward as
it seems. Their respective terrorism laws and justice systems are
very different, and – as a result – produce different terrorist prisoner
populations. Numbers differ too: whereas French and Spanish
prisons currently deal with several hundred suspected and convicted
terrorists from various backgrounds (Basque, Islamist, and – in the
French case - Corsican), the Netherlands hold exactly five. All these
factors are likely to influence the nature of countries’ prison regimes
and make them more difficult to compare.
At the same time, there are several dilemmas that are similar to all
the countries in the sample, no matter what background or size their
terrorist prisoner population. They follow from the challenges and
dilemmas that were highlighted in the previous section, and will be
explored – one after the other – in this section. They are: the question
of whether to concentrate, separate and/or isolate terrorist prisoners;
whether the emphasis should be on containment or reform; and to
what extent provisions have been made for after-care.

Concentration, separation and isolation
All prison services that deal with terrorists need to decide how to
distribute this prisoner population around their systems. In most
cases, this boils down to three related questions, namely whether
they should all be held in one place (concentration); whether
they should they be separated from the general prison population
(separation); and if they should be isolated from each other (isolation).
Concentration, separation and isolation may be good ways of
stopping efforts to radicalise and recruit the general prisoner
population, and they will prevent any emerging links between
25	 All Spanish data excludes figures for Catalonia.
26	 All UK data relates to the Prison Service for England and Wales, which comprises well over 90 per cent of
the prisoner population of the United Kingdom.
27	 Our study considers US domestic prisons only. It excludes US detainees abroad, such as those held at the
facilities in Guantanamo or Bagram airbase in Afghanistan.

17

Prisons and Terrorism Radicalisation and De-radicalisation in 15 Countries

terrorists and ‘ordinary’ criminals. At the same time, concentration,
separation and isolation provide a focus for public attention and
protest; 28 they emphasise the different (that is, political) nature of
their crimes; and – if prisoners are concentrated and separated
but not isolated – they enable prisoners to re-create operational
structures and confront the prison authorities as a united force.
Of the five countries in the sample, only one – the Netherlands –
has implemented the principle of concentration fully – most likely
because of the small numbers involved: all five are located in the
‘terrorism wing’ of the high security prison in Vught. In the US’ federal
prison system, almost all terrorists are held in just three prisons, yet
– in contrast to the Netherlands – there is no dedicated ‘terrorism
wing’ within those prisons. 29 One may argue, therefore, that the US
government has opted for a mix of dispersal and concentration,
allowing the authorities to establish clusters of terrorist prisoners
within a small number of (mostly high security) prisons. This policy –
dispersal and (partial) concentration – is even more pronounced in the
remaining countries: the UK, France, and Spain all practice dispersal
policies, meaning that most terrorists are classified as ‘high security’
upon conviction and then dispersed across their respective countries’
high security prisons. Doing so enables prison services to avoid
clear focal points for public protests and, simultaneously, makes
it possible for governments to maintain the idea that terrorists
are treated just like any other prisoner.
None of the countries have implemented regimes that involve the
permanent isolation of terrorist prisoners, which would be illegal under
various international and European human rights conventions. This
means that, in the Dutch case, terrorist prisoners can communicate
and interact with other terrorists, and – in all other countries – that
they are also exposed (albeit to a limited degree) to the general
prisoner population. The extent to which this is true varies, however.
Whereas Spanish practices differ between prisons and depend on
the nature of the terrorist prisoner population (see below), French and
American high security prison regimes can be very harsh. The French
practice of moving the most hardened terrorists from one prison to
another, together with its notorious DPS regime, 30 has been strongly
criticised by human rights organisations. The UK practice seems less
‘isolationist’, although – like the French DPS regime – it also provides
for a special regime in the form of so-called specialist units, which law
enforcement staff have described as ‘prisons within prison’. 31
The Spanish approach is particularly interesting. Whereas the
country’s 75 charged or convicted terrorists with an Islamist
background are concentrated in a small number of high security
prisons, Basque separatists, which mostly belong to the group ETA,
have been dispersed across the country. The reasons are, first, that
the Basque group ETA is a tightly structured organisation which, if
permitted to concentrate their prisoners, would attempt to re-create
its operational command structures, make desertions more difficult,
and present the prison authorities with a united front. Second, and
even more importantly, prison policy in relation to ETA is considered
28	 Examples include Stammheim prison in Germany, where all members of the Red Army Faction (RAF) were
held; and, more recently, the American detention facilities at Guantanamo Bay.
29	 Prisoners with terrorist ties may nevertheless be subject to different rules, especially regarding contact with
the outside world.
30	 DPS stands for Détenu Particulièrement Signalé, which translates as Specially Designated Detainees.
31	 Conversation with British police officer, 20 January 2010.

18

Prisons and Terrorism Radicalisation and De-radicalisation in 15 Countries

a potential bargaining chip, which may be used in future peace
negotiations. In fact, Spain is perhaps the best illustration of how
terrorist prison regimes have been adapted to reflect the needs and
risks of particular terrorist prisoner populations while being part of
a comprehensive political strategy.

Containment versus reform
Prison everywhere serves the triple aims of punishment,
incapacitation (preventing individuals from committing further
crimes), and reform (making sure that offenders will not return to
crime after being released). 32 Prison regimes need to achieve a
sensible balance between those aims, especially when it comes to
‘reform’ which public opinion and the popular press often portray
as a ‘luxury’ that can be dispensed of when costly, inappropriate,
or otherwise inconvenient. How have the countries in the sample
resolved the dilemma?
No doubt, incapacitation and containment are the overwhelmingly
dominant considerations for the prison services in all the countries.
With very few exceptions, terrorists are considered ‘high risk’ and are,
therefore, treated in similar or identical ways to non-terrorist offenders
in this category. This not only means that they are sent to ‘high
security’ prisons but also that – compared to the general prisoner
population – they are sometimes locked up in their cells for longer;
they are monitored more closely; and their communications are
subject to greater scrutiny. Indeed, some of these regimes are fairly
harsh and have prompted judicial challenges.
Even the most restrictive regimes, however, do not provide total
security. For instance, even under the United States’ high security
system, and despite being subject to ‘Special Administrative
Measures’, the so-called Blind Sheikh, who served as the inspiration
behind the 1993 World Trade Center attacks, managed to send
inflammatory messages from his prison cell to his followers in Egypt. 33
Similar examples can be found in all the countries in the sample,
including Britain, where an imprisoned supporter of Al Qaeda used
a mobile phone to direct the construction of an extremist website, 34
or Spain, where Algerian terrorists planned a terrorist attack for the
first anniversary of the 2004 Madrid train bombings from their prison
cells.35 Even in the best run prisons, such incidents will occur (see
Chapter 3). That, however, does not make them acceptable.
In all the countries observed, high security regimes have left little
space for activities geared towards reforming prisoners. Three of the
countries in the sample – the UK, France, and Spain – have provided
incentives for separatist prisoners who are willing to turn against
their comrades, but this rarely involved any conscious efforts or
programmes on behalf of the prison authorities. 36 As far as Islamist
militants are concerned, France’s and the United States’ legal,
political and even constitutional traditions make it difficult for prisons
32	
33	
34	
	
35	
36	

Additional aims may include deterrence and retribution. See Tangen, Prisons, p. 145.
See Lefkowitz, ‘Terrorists behind’, p. 23.
Ibid., p. 25. For other examples from the UK, see James Brandon, Unlocking Al-Qaeda: Islamist
Extremism in British Prisons (London: Quilliam Foundation, 2009), pp. 26-32.
Julio Lazaro, ‘Absueltos 14 de los 20 condenados en la Operación Nova’, El Pais, 7 October 2008.
In addition, there are several instances of ‘collective disengagement’ – initiated by the group’s outside
leadership and often embedded in wider peace processes (see Chapter 4) – which led to prisoners being
released on licence (Northern Ireland), ‘re-inserted’ into society (Spain), or full-scale amnesties (France).

19

Prisons and Terrorism Radicalisation and De-radicalisation in 15 Countries

to engage in attempts to ‘de-radicalise’ prisoners, confront offenders
with political and/or theological challenges, or even recognise the
concept of ‘radicalisation’. In other countries – especially Spain
and the UK – sporadic efforts have been made, but there exists no
systematic programme.
When it comes to reform, the lack of ambition is obvious and – in
many cases – perfectly understandable. Constitutional constraints
aside, expending money and effort on reforming terrorist prisoners
is not just a political minefield but widely regarded as futile. In Spain
and France, for example, most of the Islamist inmates are foreign
nationals, who will be sent back to their home countries once they
have served out their sentences. In the United States, no terrorist
in any domestic prison will be released for several years; and in the
Netherlands, the numbers are so small that a major investment
would not be justified. In the absence of compelling reasons for
change, the ‘security first’ approach is likely, therefore, to continue.

After-care
One lesson that emerges from the de-radicalisation and
disengagement programmes that are discussed in Chapters 4 and 5
is the importance of preparing prisoners for their release, and – once
they have been released – providing ‘after-care’ to facilitate their
transition into stable environments and make them less vulnerable
to being re-recruited into extremist activities. To what extent do
such arrangements exist in countries that have no formalised
de-radicalisation programmes?
In all the countries in the sample, prisoners are entitled (or obliged) to
participate in the same probationary arrangements that are available
to other prisoners. This means that, prior to an offender’s release,
risks and needs will be assessed, resulting in a probationary regime
which may include a list of conditions (ranging from prohibitions to
see or communicate with certain people to the requirement to enrol in
full-time education or training) and will be supervised by a probation
officer whom prisoners are likely to have met prior to their release.
While, in most countries, probationary arrangements are made
and supervised by a dedicated Probation Service (which, in turn,
collaborates with other state bodies, including the police and local
authorities), Spain has no agency for this purpose, and in France,
many of the functions are entrusted to a group of non-governmental
organisations (Associations), which ‘specialise’ in particular needs or
types of prisoner (for example, foreign inmates; drug users, etc.).
The absence of special arrangements or programmes for terrorist
prisoners reflects some of the issues and circumstances that were
mentioned in the previous section. In the US, for example, no
terrorists are likely to be released from domestic prisons any time
soon, which means that probationary or after-care arrangements are
not high on the list of priorities. The same is true for Spain and (to a
lesser extent) France, where – although prison sentences are shorter
than in the US – a high percentage of Islamist militants are foreign
nationals, who will be deported to their countries of origin
after serving their sentences.
As far as separatists are concerned, the prevailing wisdom seems
to be that the provision of sophisticated, reform oriented post-release

20

Prisons and Terrorism Radicalisation and De-radicalisation in 15 Countries

programmes makes little sense when prisoners are released back
into communities – be it in Corsica or the Basque Country – which
are overwhelmingly supportive of the cause that led to the offender’s
conviction. The result here, as in other countries, is to have a
‘security first’ approach towards ‘after-care’, which is to ensure
that prisoners are under tight supervision and monitoring by the
police and intelligence services.
The one exception is Britain, where the prison authorities are
beginning to put in place more ambitious structures. With several
dozen ‘home-grown’ terrorists up for release in the coming years,
relationships have been forged with several community groups
who were given active roles in the implementation of prisoners’
probationary regimes. In addition to providing vocational training
and, in some cases, academic education, these groups take an
active interest in shaping former prisoners’ social environments and
attitudes, including – where necessary – psychological counselling
and the discussion of theological and political issues. Similar to
the de-radicalisation and disengagement programmes in Middle
Eastern and Southeast Asian countries, the aim is to provide former
prisoners with stable and healthy environments in which a return
to extremism becomes undesirable and high cost. Arguably, the
emerging British approach towards after-care comes as close as
anything that has been observed in the countries in the sample to
drawing on the principles and experiences of de-radicalisation or
disengagement programmes (see Chapters 4 and 5).

2.3 Recommendations
This chapter has shown that – despite the ‘criminalisation’
doctrine – most prison authorities understand that terrorists and
other politically motivated offenders represent a distinct prison
population which presents particular risks and, consequently,
requires different kinds of prison regimes.
There are no hard and fast rules about whether terrorist prisoners
should be concentrated or separated and isolated. It all depends on
how a policy is implemented and, more importantly, on the nature and
dynamics of the particular group in question. If, like Al Qaeda in the
West, the group is dispersed and largely leaderless, bringing together
its followers in prison would help to create structures that had not
existed before, and should be avoided where possible. If, on the other
hand, concentration serves the purpose of bringing together terrorists
who have turned against their former comrades and enable them to
support each other, it could have a positive impact.
The current mix of dispersal and (partial) concentration, which is
practiced in most of the countries in thesample, balances the high
risks inherent in this prisoner population with the desire not to
create focal points for public protest or human rights concerns. But
it may not be appropriate in all cases: for instance, where a group
undergoes a process of collective de-radicalisation (see Chapter 4)
or where governments are engaged in peace processes, bringing
together the prisoners in one place may contribute to the wider aim
of furthering conflict resolution.
The lessons are both obvious and simple:

21

Prisons and Terrorism Radicalisation and De-radicalisation in 15 Countries

•	 The importance of developing a sophisticated understanding of the
terrorist prisoner population cannot be overstated. This includes
a thorough understanding of the groups and movements to which
terrorist prisoners belong, as well as regular assessments of each
individual in question, ensuring, for example, that ‘fellow travellers’
and marginally involved foot soldiers are not mixed with hardcore
terrorists and charismatic leaders.
•	 Training and educating frontline staff is a priority. Prison staff
need to ensure that terrorists – like other prisoners – are treated
correctly and fairly at all times, which will make it more difficult
for terrorists and their outside supporters to exploit the prison
environment for propaganda purposes, including, for example,
allegations of discrimination and mistreatment. Furthermore,
where terrorist and non-terrorist prisoners have opportunities to
mix, it is important for prison staff to be able to detect changes
in behaviour, including attempts to influence others or assume
leadership roles, but also signs of disillusionment with the group
and its ideology (see Chapter 3).
The ‘security first’ approach, which dominates many prison services’
thinking in relation to terrorists, is understandable and not entirely
without justification or merit. For example, prisons need to make
sure that terrorist prisoners are closely monitored. This is particularly
important when it comes to their communications – after all, it is
in this respect that such prisoners, who are otherwise unable to
contribute to their movements’ campaigns, are potentially most
problematic. Even the best prisons cannot be hermetically sealed, but
prison staff must know and understand how central the monitoring
of communications is to dealing with convicted terrorists and other
politically motivated offenders.
‘Security first’, however, should not be mistaken for ‘security only’.
Whether or not a terrorist can be rehabilitated or de-radicalised
depends on many factors, and it may well be true that – in certain
cases – such attempts are futile. In others, however, signs of
disenchantment with the movement, its methods or ideas should be
taken seriously, and – where possible and appropriate – they should
be encouraged, nurtured, and capitalised upon.
Reform does not need to come at the expense of security. In
many situations, in fact, the two imperatives go hand in hand. For
example, making sure that the prison is not flooded with extremist
or inflammatory literature is a security issue, but it also represents a
pre-condition for reform and rehabilitation. Going one step further,
if ‘negative’ influences ought to be kept out of prison, ‘positive’
influences need to be more actively promoted.
The fostering of positive influences does not necessarily require
a formalised de-radicalisation programme or even a fully-fledged
de-radicalisation centre, but – again, depending on the nature
of a particular movement and the needs of individuals prisoners
– may consist of ‘soft’ interventions, such as providing ideological
challenges to the extremist ideology, bringing in speakers who
can debate relevant issues and present alternative viewpoints, or
making more educational or vocational opportunities more available.
There may be perfectly good political, constitutional, and practical
reasons for not wanting a formal de-radicalisation programme. That

22

Prisons and Terrorism Radicalisation and De-radicalisation in 15 Countries

does not mean, however, that the principles and positive practices
that have been gleaned from such schemes should not be adapted to
local circumstances, including the above-mentioned attempts to offer
prison programming and enable challenges to extremist beliefs and
justifications of violence, but also – and importantly – their emphasis
on ‘after-care’.
Again, the demands of security and rehabilitation are more likely to
complement than contradict each other: the more former prisoners
are engaged with support structures and programmes after their
release from prison, the easier it becomes to monitor them. At the
same time, for ‘after-care’ to succeed in facilitating the transition from
prison to society and minimising the risk of re-recruitment, it needs
to begin while offenders are still in prison. For that reason, it may
well turn out that establishing effective after-care will lead to a more
systematic approach towards reform and rehabilitation overall.

23

Prisons and Terrorism Radicalisation and De-radicalisation in 15 Countries

24

Prisons and Terrorism Radicalisation and De-radicalisation in 15 Countries

3 Prison Radicalisation

I

n a recently released statement, Abu Qatada – often referred to
as Osama bin Laden’s ‘Ambassador to Europe’ 37 – talked about
‘see[ing] the signs of Allah’ within British prisons, where – in his
own words – ‘young men enter into Islam and then... learn Arabic and
the Sharia in a short number of months’. 38 Similar observations have
been made by extremists all over the world, and it is no exaggeration
to say that prisons – for all kinds of movements and in all periods of
history – have been fruitful places for radicalisation and recruitment. 39
This chapter focuses on Islamist militants, because they represent
the most widespread manifestation of the phenomenon globally.
Examining eight countries – Afghanistan, France, the Netherlands,
Pakistan, the Philippines, Spain, the United Kingdom, and the United
States40 – the chapter looks at underlying drivers as well as the
policies and approaches that have been adopted by governments
to counter prison radicalisation. It demonstrates that prison
radicalisation is driven by behaviours and conditions that are typical
of the prison environment – especially religious seeking, defiance,
and the need for protection – and that efforts aimed at preventing
radicalisation should seek to address and mitigate these factors.
The difficulties are considerable. Distinguishing between legitimate
expression of faith and extremist ideologies can be difficult, and
training and detection in this area is highly uneven. The desire to deny
extremists the ‘religious space’ from which to radicalise and recruit
has led to renewed interest in the institution of the prison imam, but
its role and influence are at risk of being overrated. Furthermore,
the gap between the vision of creating ‘stable prison societies’ and
the reality of disorderly, overcrowded and under-staffed prisons is
sometimes quite stark, especially in developing countries.
The chapter suggests that more needs to be done to confront the
challenges of detecting and preventing prison radicalisation, but
that prescriptions differ depending on country and circumstances.
In fact, many of the issues that play out in the prison environment
are not actually ‘prison issues’ at all, but touch on wider and more
fundamental question of counter-radicalisation policy.

3.1 Issue
When experts and policymakers speak about the dangers of prison
radicalisation,41 they typically raise two kinds of concern. One is
37	 See, for example, ‘Britain “sheltering al-Qaeda leader”, BBC News, 8 July 2002;
see http://news.bbc.co.uk/1/hi/uk/2115371.stm.
38	 Cited in ‘An Address to the Muslims from Abu Qatada’, NEFA Foundation, 23 March 2009; available at
http://www.nefafoundation.org/miscellaneous/FeaturedDocs/nefaabuqatada0309.pdf. The quotation was
originally found in Brandon’s Unlocking Al Qaeda, p. 1.
39	 See, for example, Anti-Defamation League, ‘Dangerous Convictions: Extremist Activities in Prisons’,
Topics in Extremism, 2(2002).
40	 The authors of the country case studies were Laila Bokhari (Pakistan), Andrew Coyle (United Kingdom),
Bob de Graaff and Eelco Kessels (Netherlands), Jean-Luc Marret (France), Arie Kruglanski and Michelle
Gelfand (Philippines), Marisa Porges (Afghanistan), Manuel Torres (Spain), and Bert Useem (United States).
41	 Among the policymakers who have warned of the potential threat from prison radicalisation are US
	
Attorney General Alberto Gonzales and former UK Home Secretary Jacqui Smith. See Gonzales, quoted
in Mark S. Hamm, ‘Terrorist Recruitment in American Correctional Institutions: An Exploratory Study of
Non-Traditional Faith Groups – Final Report’, US Department of Justice Commissioned Report, December
2007, p. 10; Smith, quoted in Raffaello Pantucci, ‘Britain’s Prison Dilemma: Issues and Concerns in Islamic
Radicalization’, Jamestown Terrorism Monitor, 24 March 2008.

25

Prisons and Terrorism Radicalisation and De-radicalisation in 15 Countries

that prison brings together politically motivated offenders, including
terrorists, with ‘ordinary’ criminals, creating the potential for an
‘unholy alliance’ between the two. Instead of reducing the risk of
terrorism, prison may thus help to produce an even more serious
threat by combining the terrorists’ ideological fervour with ‘ordinary’
offenders’ criminal energy and skills. It is this concern – among others
– which underlies the dilemma between ‘concentrating’ imprisoned
terrorists or allowing them to mix with ‘ordinary’ criminals (see
Chapter 2). 42
The second point is that prison represents a ‘place of vulnerability’
in which individuals are more likely to be radicalised and go on
to become involved in extremist and/or terrorist activities. The
term ‘place of vulnerability’ depicts locations in which individuals
experience social isolation or personal crises, both of which are
widely believed to be factors that increase a person’s responsiveness
to extremist messages and/or approaches. 43 Prison matches this
description: offenders are removed from their friends and family; they
are confronted with their past; and they need to find their way in a
new (and often hostile) social universe.
Accordingly, the two motivations that are most likely to make inmates
susceptible to new faith and belief systems – including extremist
and militant interpretations of Islam – are the search for meaning and
identity (which new prisoners experience in particularly intensive ways)
and the need for (physical) protection, which is offered by traditional
gangs and Islamic groups alike. 44 In addition, prison also seems to
breed a desire to defy the authorities, which means that ‘identities of
defiance’ – enabling prisoners to articulate their grievances and rebel
against the system – are in high demand. 45

How big a problem?
Anecdotal evidence of prison radicalisation is easy to find. The most
spectacular (and, by extension, most frequently cited) cases are
those in which ‘free world’ terrorist plots had their origins in the prison
environment. They include:
•	 Khalid Kelkal, who was radicalised in a French prison in the early
1990s. Recruited by radical Algerians, he went on to become
involved in the murder of a moderate imam in Paris and the
attempted bombing of the high-speed rail link between Paris
and Lyon.
•	 Operation Nova in Spain, which involved supporters of the Algerian
Armed Islamic Group (GIA), who recruited new followers during
their stay at a prison in Salamanca. By 2004, they were planning
a terrorist attack against the Spanish Central Court.

42	 See, for example, FBI Director Robert Mueller’s speech in Ohio in 2006; quoted in Hamm,
‘Terrorist Recruitment’, p. 24.
43	 Neumann, ‘Recruitment and Mobilisation’.
44	 Hamm, ‘Terrorist Recruitment’, Chapter 3. See also Gabriele Marranci, ‘Faith, Ideology and Fear: The Case
of Current and Former Muslim Prisoners’, IQRA Annual Lecture – House of Lords, 26 June 2007.
45	 See Bert Useem’s case study report on the United States; Hamm, p. 5. In Farhad Khosrokhavar’s words,
radical Islam it is fast becoming prisoners’ preferred ideology of resistance – ‘the religion of the oppressed’
– beginning to play the role that once belonged to Marxism. See Khosrokhavar, cited in Craig S. Smith,
‘French Prisons – Radicalizing Large Muslim Populations’, New York Times, 20 December 2004.

26

Prisons and Terrorism Radicalisation and De-radicalisation in 15 Countries

•	 The Authentic Assembly of God (JIS) – an Islamist group which
formed in a California state prison in 1997. Seven years later, its
leader was recruiting inmates who were instructed to attack Army
recruitment centres, synagogues and other targets. 46
•	 Kevin Gardner – later known as Abbas Shafiq – who turned
towards extremism during his stay at a Young Offenders’ Institution
in England in 2006-07. He became obsessed with the British Army
and plotted an attack on a military base from within his cell. 47
Instances of prison radicalisation in non-Western countries are less
well-documented, but they are not hard to find. To take the most
dramatic example, numerous attacks by Al Qaeda and the Taliban are
said to have originated from within Afghan prisons, where terrorists
and insurgents mix freely with ‘ordinary’ criminals and instances of
radicalisation are said to be extremely frequent. As the Commander of
U.S. Forces in Afghanistan, General Stanley McChrystal, pointed out,
there are ‘more insurgents per square foot in corrections facilities than
anywhere else in Afghanistan’. 48
However, it is difficult to generalise based on anecdotal evidence
alone, and neither governments nor researchers know exactly how
much their respective prison systems are affected. Some of the
problems in quantifying the phenomenon relate to different ideas
about the concept as such and how it should be measured. A number
of commentators and even government agencies, for example,
confuse ‘conversion’ with ‘radicalisation’, implying that prisoners
who take up purist strains of Islam – often described as Salafi – are
all at risk of becoming terrorists. 49 In reality, the two are not always
connected. While prison conversions to Islam are frequent, and
significant percentages of converts and ‘born again Muslims’ are
attracted by the purity and rigour of the Salafist tradition, no serious
researcher claims that this automatically translates into support for
terrorism. More importantly, freedom of religious expression is not
something that prison authorities can (or should) criminalise.
Furthermore, it is not always clear whether an individual’s
radicalisation took place in prison or happened before. Take, for
example, the case of Richard Reid, the ‘shoe bomber’ who attempted
to blow himself on board a transatlantic flight in 2001. He converted
to Islam while imprisoned at a Young Offenders Institution in England,
but many believe that his interest in Al Qaeda only started when he
became involved with a group of extremists at Brixton Mosque. 50 The
same can be said of Jose Padilla, the American terrorist who was
accused of plotting to detonate a ‘dirty bomb’. He converted to Islam
in prison, but followed a moderate interpretation of that faith until his
release. His radicalisation is said to have been prompted by a friend
whom he had met at a local mosque. 51 Both cases are frequently held
up as evidence of prison radicalisation, 52 yet – under closer scrutiny –
turn out to have little to do with prison. (If anything these cases are
a good illustration of why the systematic provision of after-care is
so critical.)
46	 For the most extensive analysis of the JIS case, see Hamm, ‘Terrorist Recruitment’, Chapter 2.
47	 Brandon, Unlocking Al-Qaeda, p. 18; also Ben Goldby, ‘Al Qaeda recruiting at Midlands prison’, Sunday
Mercury, 10 January 2010.
48	 Gen. McChrystal, cited in Melisa Porges’ case study report on Afghanistan.
49	 See Jean-Luc Marret’s case study report on France.
50	 See Andrew Coyle’s case study report on the UK.
51	 Deborah Sontag, ‘Terror Suspect’s Path From Streets to Brig’, New York Times, 25 April 2004.
52	 See, for example, Alex Wilner, ‘We must make sure that prison isn’t terrorism school’, Globe and Mail,
19 October 2009.

27

Prisons and Terrorism Radicalisation and De-radicalisation in 15 Countries

Equally, radicalised prisoners may not necessarily go on to become
involved in extremism once they are released. This is particularly
true for those who became part of a radical group – or prison gang
(see below) – for the purpose of seeking protection. They are not
always ‘authentic’ conversions, which are likely to be followed up or
acted upon in the ‘free world’. 53 This is not to downplay the problem,
but – rather – to emphasise the particular conditions of the prison
environment which increase the potential for radicalisation, but may
also produce behaviours that are no longer relevant or appropriate
when offenders have been released.

What does it look like?
Prison radicalisation comes in many guises. In categorising its most
frequent manifestations, it is useful to distinguish between ‘channels’
of radicalisation that emanate from outside the prison environment
and those that come from within.
Among the external drivers of radicalisation are extremist books,
videos or websites, which are sent to the prison or made available to
inmates. External ‘channels’ of radicalisation may also include outside
visitors, who promote extremist messages. Most prisons will seek to
prevent prisoners from having the opportunity to interact with known
extremists, but such individuals are not always easy to recognise,
especially when they claim to look after prisoners’ religious needs and
the prison authorities have no way of distinguishing ‘moderates’ from
extremists. As a result, there have been a (small) number of cases in
which ‘radical imams’ have gained access to prisons and spread their
message among the inmates.
Internal sources of radicalisation include terrorist inmates who
– unless they are completely separated from the general prison
population – may approach ‘ordinary’ inmates, befriend them and
‘convert’ them to their beliefs and ideology. A typical behaviour is
for convicted terrorists to attempt to assume leadership roles
among the wider prison population, for example by leading Friday
prayers or taking the lead in negotiating concessions on behalf of
particular prisoners or ‘Muslim prisoners’ as a whole.
A different phenomenon are so-called Muslim prison gangs, which
are more typical of Western prisons (where being Muslim is a marker
of identity) than prisons in Muslim majority countries. Like traditional
prison gangs, Muslim prison gangs are based on religious (sometimes
also ethnic) affiliation and provide members with a strong sense
of identity and in-group loyalty. Moreover, they allow members to
articulate their grievances and protect them against other gangs or
groups of prisoners. Muslim prison gangs rarely include convicted
terrorists, nor do they generally seem to have a good grasp of the
Islamic faith or extremist ideology, and it is questionable to what
extent their members have internalised, or genuinely committed
themselves to, either. Nevertheless, ‘cut and paste’ versions of radical
Islam frequently form the basis of their ‘ideology’, and particularly
notorious terrorists are frequently glorified and held up as heroes. 54

53	 See Hamm, ‘Terrorist Recruitment’, Chapter 1.
54	 See ‘Out of the Shadows’, Special Report, p. iv.

28

Prisons and Terrorism Radicalisation and De-radicalisation in 15 Countries

Challenges
Prison radicalisation results from – and is related to – the conditions
inherent in the prison environment, including: the presence of terrorist
inmates; the need for protection; the search for meaning and identity;
and the desire to defy a system which may be perceived as unjust.
The challenge for policymakers and prison authorities is to mitigate
those conditions, so they are less likely to feed into radicalisation.
It follows that prison policies should be geared towards
•	
•	
•	
•	

Recognising radicalisation;
Monitoring the activities of terrorist inmates
Controlling external influences;
Encouraging prisoners to embrace ‘inmate identities’ rather than
defy the authorities; 55
•	 Mitigating conflicts between ethnic, racial and religious groups;
•	 Channelling prisoners’ search for meaning and identity into
productive directions.
To what extent current policies have addressed these objectives
will be shown in the next section.

3.2 Practice
This section uses information from eight case studies in order to
examine how different countries and their respective prison systems
have dealt with the issue of prison radicalisation. The countries are:
Afghanistan, France, the Netherlands, Pakistan, the Philippines,
Spain, the UK, and the US.
As in the previous chapter, it is important to be conscious of the
differences between these countries and their prison systems.
For example, two of the countries in the sample – Pakistan and
Afghanistan – are Muslim majority countries in which ‘Muslim prison
gangs’ are unlikely to be relevant and issues such as anti-Muslim
prejudice are insignificant or non-existent. It also needs to be kept
in mind that prison radicalisation in Afghanistan – and, to a lesser
extent, Pakistan and the Philippines – takes place in the context
of ongoing insurgencies, with wider political conflicts spilling over
into the prisons regardless of what kinds of policies the prison
authorities have decided to adopt.
The first two parts of this section look at internal drivers (general
conditions, and monitoring and training), while the following two
examine external drivers (the role of prison imams and external
influences).

General conditions
Many experts believe that there is a correlation between potential
for radicalisation and the degree to which prisons are safe and
orderly. In Useem’s view, for example, ‘stable prison societies’
– which require sufficient space and staffing but also work and
educational opportunities – minimise the space for sub-cultures
55	 See Bert Useem’s case study report on the US.

29

Prisons and Terrorism Radicalisation and De-radicalisation in 15 Countries

and conflicts between inmates, and they facilitate the emergence
of ‘inmate identities’, with prisoners being ‘mere prisoners, not rebels
with a cause’. 56 Importantly, safe and orderly prisons also make
it easier for the authorities to collect intelligence and pick up on
emerging signs of radicalisation. 57
With the exception of the Netherlands (which has recently started
leasing spare capacity to neighbouring Belgium), all the prison
systems in the sample suffer from over-crowding and, in most cases,
a shortage of staff as well as work and educational opportunities. This
does not always result in chaotic conditions, but – following Useem’s
analysis – it increases the potential for conflict and reduces the
authorities’ ability to pick up on extremist activities. Indeed, anecdotal
evidence from France, Spain and the UK suggests that ‘radicalisers’
do take advantage of poorly run and overcrowded prisons, and that
Muslim prison gangs tend to form in environments in which resources
are scarce, ethnic and religious conflicts are rife and the prison
management can no longer ensure the safety of inmates. 58
The contrast between the well-resourced and well-staffed prison
system in the Netherlands, where the authorities were able to stop
instances of radicalisation from spreading at an early stage, and
those in Afghanistan, the Philippines and Pakistan could not be
starker. In all three countries, prisons are not only understaffed and
over-crowded, there are no provisions for monitoring terrorist inmates
who – with the possible exception of the Philippines – are free to
mix with ‘ordinary’ inmates. Bribery and corruption are common,
meaning that militant groups can frequently count on the complicity
of prison staff when targeting inmates for radicalisation and
recruitment. Furthermore, instead of preventing the use of violence
and ensuring the safety of all inmates, it is the prison staff who
initiate many violent attacks.
Arguably, such conditions not only provide the ‘breeding ground’
for radicalisation but may represent one of its causes. Take Pakistan,
for example, where 92,000 prisoners share 41,000 prison places,
often mixing juvenile with adult offenders and making no distinction
between minor offenders, hardened criminals and politically motivated
militants. Staff are in short supply and badly trained, with the result
that reports of mistreatment and even torture are frequent. Moreover,
the government’s recent successes in fighting domestic militancy
have produced 4,000 additional inmates, of which 3,700 are
thought to be affiliated with Al Qaeda or the Taliban. Together with
detainees from Guantanamo, which have recently been sent back to
Pakistan, militants of all kinds now represent the fastest growing part
of the Pakistani prison population, making the overcrowding worse
and exposing ‘ordinary’ inmates to increasing numbers of battlehardened ‘jihadists’.
The Pakistani case shows that it would be neither fair nor
appropriate to look at prison radicalisation as a problem of the
prison system alone. In Pakistan and the Philippines, the ‘root cause’
of the overcrowding are highly inefficient judicial systems, which
often produce verdicts only years after the accused have been
56	 Ibid.
57	 Ibid.
58	 Similarly, Useem believes it may be no accident that the one terrorist plot that emerged from an
American prison (see previous section), occurred in California, whose state prison system is notorious
for overcrowding and its general lack of resources, staffing, and programming. See ibid.

30

Prisons and Terrorism Radicalisation and De-radicalisation in 15 Countries

thrown into prison. 59 Likewise, corruption and mistreatment in
prisons are indicative of political systems in which access to justice is
limited and the rule of law is weak. In cases such as these, it seems
clear that dealing with prison radicalisation requires governments to
consider (and tackle) wider structural problems rather than attempting
to ‘fix’ the prisons alone.

Monitoring and training
Islamist militant prison radicalisation is a relatively new phenomenon
which prison authorities the world over have only recently started to
grapple with. Everyone agrees that improving the capacity to spot
radicalisation is vital, but that doing so represents one of the most
difficult aspects of dealing with the phenomenon.
As far as the five Western countries in the sample are concerned,
three trends could be identified. First, all five have recognised the
issue and have determined that raising awareness among prison
staff should be a priority. While some (such as the UK) have
developed high-level strategies and created new, often centralised
structures, others (such as the Netherlands) have relied on less
formal mechanisms and permitted local stakeholders – individual
prison governors, for example – to take ownership of the issue
and lead nationwide efforts.
All these initiatives emphasise the need to train prison staff,
recognising that prison guards and other frontline staff who interact
with prisoners on a daily basis are critical in making early detection
work. In Andrew Coyle’s words, ‘[Only frontline staff] can sense when
there is a change of atmosphere in a prison wing. They can “smell
trouble” before it begins’. 60It is not always clear, however, to what
extent such announcements have been translated into practice. In the
United States, for example, where 50 state prison systems and the
federal Bureau of Prisons all enjoy autonomy, implementation is very
uneven. In France and the UK, radicalisation training programmes
have only recently been embarked upon, and – in some cases (such
as Spain) – they are still at a conceptual stage.
(Relevant training does not necessarily need to address the
radicalisation issue directly. Afghanistan, Pakistan and the Philippines,
for example, have intensified their efforts – often with Western
help – to educate their staff about basic practices of good prison
management, helping to avoid the worst excesses and instances
of mistreatment. Likewise, Muslim minority countries such as Britain
have come to understand that instances of discrimination, or the
perception thereof, feed into the extremist narrative of victimisation,
and have adopted racial and/or religious equality training as part of
the counter-radicalisation agenda.) 61
A second trend among the Western countries in the sample is the
development of indicators for radicalisation. France/the European
Union (EU), 62 the UK and the Netherlands have all produced leaflets
59	 In Pakistan, an astonishing 80 per cent of the prison population are held on remand. See Laila Bokhari’s
case study report on Afghanistan.
60	 See Andrew Coyle’s case study report on the UK.
61	 James A. Beckford, Daniele Joly and Farhad Khosrokhavar, Muslims in Prison: Challenge and Change
in Britain and France (Basingstoke: Palgrave Macmillan, 2005), Chapter 5.
62	 The French manual is the result of a joint initiative by France, Germany and Austria during the French
Presidency of the European Union in 2008. See Jean-Luc Marret’s case study report on France.

31

Prisons and Terrorism Radicalisation and De-radicalisation in 15 Countries

or manuals with instructions on what prison staff should look out
for. The Dutch booklet stresses changes in behaviour and beliefs,
including ‘(sudden) interest in religion’, ‘an exaggerated distinction
between what is “pure” and “impure”’, and ‘a desire to convert fellow
inmates’, together with expressions of defiance such as ‘challenging
the words of an(other) imam’ and ‘displaying Messiah-like behaviour’.
By contrast, the British, while covering much of the same ground,
are more careful to downplay the religious and/or Islamic aspect,
emphasising not only that there are other kinds of extremists
(dissident Irish Republicans, far right activists, etc.), but also that –
rather than looking for potentially benign behaviours, such as sudden
interest in religion – prison staff should act on expressions of ‘support
for radical extremist causes or leaders’, look for ‘materials... that
express sympathies to extremist behaviours/actions’, and ‘contacts
with people involved or suspected of involvement with extremist
activity’. The French/EU handbook falls somewhere in the middle,
with a mixture of indicators that address religious practice, physical
appearance and political expression. It also provides a ‘traffic
light’ system, with ‘yellow’ marked by the appearance of Islamist
propaganda, graffiti, and changes in prisoners’ appearance, and ‘red’
consisting of acts of outright defiance, including the destruction of
public property and attacks against prison staff.
Well-intentioned and worthwhile in principle, such leaflets may
have unintended consequences. The Dutch, for example, with their
emphasis on religious practice and changes in outward appearance,
may encourage an over-reporting of religious conversions, which are
not necessarily indicative of radicalisation. Indeed, when conversions
to Islam are met with harsh and overly repressive responses by prison
staff, this not only violates prisoners’ freedom of religious expression,
it also creates the impression that the prison authorities – and, by
extension, the state – are against Muslims in general, feeding into the
extremist narrative of the West at war with Islam. On the other hand,
the British leaflet – in trying to avoid this ‘trap’ – has become almost
circular in its logic. There are numerous references to ‘extremist
behaviours/actions’ and ‘extremist expressions’ without actually
explaining what such behaviour consists of: British prison guards,
in other words, are asked to detect extremist behaviours by looking
for extremist behaviours. The only useful and truly reliable indicators,
which are contained in all three booklets, relate to acts of open
defiance, including attempts to replace and marginalise the prison
imam. At that point, however, it may already be too late for
early interventions.
A third trend among Western prison authorities is for more
centralised and professional intelligence systems to be put in place.
All countries have recognised that it is vital for information and
resources to be pooled, because individual prisons on their own are
unlikely to possess the ability to monitor communications in foreign
languages, assess the content of suspicious books and videos, or
evaluate the credentials of external visitors (see below). In Spain,
France and the Netherlands, it is the security services which have
taken the lead and work closely with prisons’ security departments
(France, Netherlands) or ‘observation and control groups’ (Spain).
In Britain, in addition to the Prison Service’s high security directorate,
the National Offender Management Service now houses an
extremism unit. The United States’ federal prison service is the most
advanced system, having established a close relationship with the
Federal Bureau of Investigation (FBI) in the form of the Correctional

32

Prisons and Terrorism Radicalisation and De-radicalisation in 15 Countries

Intelligence Initiative. This partnership enables the collection and
sharing of information about prisoners and prison imams, but
also deploys centrally located resources such as translators and
information on extremist literature.

Prison imams
One of the most interesting recent developments is the upsurge of
interest in the institution of the prison imam. In all the Muslim minority
countries in the sample, Muslims in prison are over-represented
relative to the country’s overall population, sometimes significantly so:
in the US by a factor of 3; in Spain, the Netherlands and the UK by a
factor of 4; and in France by an estimated factor of 10. 63 Yet, until the
mid-2000s, little attention had been paid to the provision of religious
services for this segment of the prison population. Only recently
have prison services begun to embrace the institution of the prison
imam as a counter-radicalisation tool, realising that prison imams
can help to minimise, if not deny, the space available to extremists
who – rightly or wrongly – have used religious language and motifs to
establish their credibility and recruit followers, for example by leading
Friday prayers or providing spiritual care.
Before their recent renaissance, prison imams were regarded with
indifference or suspicion in most of the Muslim minority countries in
the sample. Numbers were small, and little was done to support their
work. The element of suspicion related to a small number of cases in
which radical imams had gained access to prisons and established
themselves as legitimate providers of religious services, with the
result that – following the terrorist attacks in September 2001 – many
governments’ initial reaction was to ‘crack down’ on prison imams,
reducing their numbers further and making life more difficult for those
who remained within the system.
With the possible exception of the Netherlands, whose government
had invested in the provision of prison imams throughout the 1990s,
the realisation that prison imams could be helpful in combating
radicalisation and that their role should be strengthened occurred only
in the mid-2000s. Today, even countries such as France – which had
long been reluctant to support prison imams based on the suspicion
that doing so would violate the state doctrine of laicité – are fully
committed to the institution. Indeed, only Spain seems not to have
changed its (indifferent) attitude, with all the other Western countries
in the sample being enthusiastic ‘converts’.
The sudden enthusiasm for the prison imam has gone hand in
hand with an expansion of their role. Not only are today’s ‘new’
prison imams supposed to be competent in the provision of religious
services and spiritual care, they are also expected to be counsellors,
social workers, experts in radicalisation and extremism, and – more
generally – act as interlocutors between the prison authorities
63	 The US number is based on the number of individuals seeking Islamic religious services in the federal
prison system; state figures are not available, but are thought to vary considerably. See Hamm, ‘Terrorist
Recruitment’, Chapter 1. In Spain, the number relates to prisoners who are citizens of Muslim majority
countries, as the number of Spanish citizens of Muslim faith is believed to be very small. See Humberto
Trujillo et al, ‘Radicalization in Prisons? Field Research in 25 Spanish Prisons’, Terrorism and Political
Violence, 21(2009), p. 559. In France, no official statistics on religious affiliations are kept, but
Khosrokhavar’s figure of 70 – 80% is widely quoted and thought to be a reliable estimate. See Beckford,
Muslims in Prison, Chapter 1. The figures in the UK and the Netherlands are based on official statistics,
quoted in Andrew Coyle’s case study report on the UK and Bob de Graaff and Eelco Kessels’ case
study report on the Netherlands.

33

Prisons and Terrorism Radicalisation and De-radicalisation in 15 Countries

and Muslim prisoners for all kinds of everyday concerns. To make
sure that all those services are delivered effectively, France, the
Netherlands, the UK and – to some extent – the US have started to
institutionalise and professionalise the position as well as formalise
the processes through which prison imams are appointed, with the
result that imams now have to fulfil a language requirement, provide
evidence of religious competence, and commit to ongoing training
exercises. In France and the Netherlands, there are state-funded
Islamic Prison Chaplaincies, while the UK has created the position
of Muslim Advisor. In the US, the federal prison service has forged
closer relationships with a number of religious service providers.
Despite governments’ best intentions, the institutionalisation of the
prison imam has prompted the concern that Western authorities
are introducing a state-sanctioned form of Islam, which can easily
be undermined by extremists and fails to reflect the diversity of the
Islamic faith, especially when prison services’ faith advisors are
all drawn from the same religious tradition. 64 Another concern is
that the prison imam’s ever increasing mandate will lead to Muslim
prisoners having to define themselves as ‘Muslim’ in order to gain
access to welfare services and pastoral care for which non-Muslim
prisoners would rely on the prison management. Overall, however,
there can be no question that Western governments’ recent attempts
to improve the provision of Islamic services compares favourably
with the situation in the remaining countries in the sample, where no
systematic efforts have been made to offer religious services and it is
– more often than not – left to the prisoners themselves to lead Friday
prayers and perform other spiritual duties. In places like Pakistan and
Afghanistan, which have significant numbers of extremists in their
prison systems and allow ‘ordinary’ prisoners and extremists to mix,
the consequence is that extremists are free to lead Friday prayers
and use their supposed religious authority to radicalise and recruit.

Other external influences
In his extensive study of Islam in American prisons, Mark Hamm
found that prison radicalisation is an essentially social phenomenon,
drawing its strength primarily from the personal interaction between
inmates.65 Attention, however, must also be paid to external
influences, which can inspire, reinforce, direct and substantiate
the interactions in which prisoners participate.
All the countries in the sample claim to have strict policies on
prisoners’ communication with the outside world: mail deliveries
are said to be screened; phone calls and the use of the internet are
meant to be restricted. In reality, however, individual prisons often
lack the capacity to perform these functions effectively, and instances
of external infiltration could be found in each of the eight countries.
The explanation for such failures typically lies in insufficient staff
numbers (in Spain, for example, the staff/prisoner ratio is 1:160),
or prison guards who have no idea how to distinguish between
‘legitimate’ religious and political information and extremist materials.
For prisons in Muslim minority countries, whose Islamic inmates
are mostly foreign nationals or the descendants of immigrants, an
64	 See Brandon, Unlocking Al-Qaeda, p. 113.
65	 See Mark Hamm, ‘Prisoner Radicalization: Assessing the Threat in US Correctional Institutions’,
National Institute of Justice Journal, October 2008; available at
http://www.ojp.usdoj.gov/nij/journals/261/prisoner-radicalization.htm#note8.

34

Prisons and Terrorism Radicalisation and De-radicalisation in 15 Countries

additional obstacle lies in the fact that communication often takes
place in foreign languages.
While countries such as Britain have started training prison staff
in basic Islamic concepts and expressions, the US federal prison
system’s Correctional Intelligence Initiative will enable prisons to
access federal resources (for example, translators) and share and
access information on external visitors and literature. Despite such
activities, the wider question of what constitutes ‘extremist’
material and to what extent prisoners’ freedom of expression and
access to information should be restricted in the name of counterradicalisation remains unaddressed. In one country in the sample,
prison governors prison governors have been advised to ‘reject
“gifts” from foreign embassies’, which probably refers to Saudisponsored translations of the Quran that are frequently criticised
for being overly militant and sectarian. Would other countries be
prepared to go as far as sidelining particular interpretations of
the Quran?

3.3 Recommendations
In dealing with radicalisation, prison services are confronted with
two challenges: preventing radicalisation, and detecting radicalisation
where it has occurred. The analysis of the problem and the policies
adopted by the eight countries in the sample showed that running
safe and orderly prisons is critical to addressing both. In fact, doing
so is the pre-condition for all other measures that may be taken to
address one or the other:
•	 Overcrowded, chaotic prisons make it difficult for prison staff
to understand what is happening ‘on the ground’ and, as a
result, will allow extremists to operate at free will. Under such
conditions, effective monitoring and detection of radicalisation
becomes impossible.
•	 Unsafe and disorderly prisons aggravate the conditions that make
prisoners vulnerable to radicalisation. Extremists will find it easier
to fill the (spiritual and material) vacuum created by prisons that
fail to provide prisoners with a perspective and a set of meaningful
activities towards which their energies can be directed. They also
create a ‘security dilemma’ in which inmates feel compelled to turn
to extremists – or organise themselves as prison gangs
– for protection.
This highlights the need for sufficient numbers of well-trained staff,
and it reinforces the fundamental importance of prison programming
in creating ‘inmate identities’ that will reduce the occurrence of
defiant, hopeless and vulnerable prison populations. This may seem
like an obvious priority for developing countries, where prison systems
are often in a desperate state, but it also applies to Western countries
in which problems with disorder have arisen too.
Another important element in combating prison radicalisation is to
deny religious space to extremists. The recent upsurge of interest
in the institution of the prison imam shows that prison services
have understood the significance of preventing extremists from
monopolising the Islamic narrative. Even so, the professionalisation
of the position and its ever increasing role have led to a number of

35

Prisons and Terrorism Radicalisation and De-radicalisation in 15 Countries

concerns, which should be taken into account when developing
their role further:
•	 As much as their revival is welcome, prison imams are not a
panacea. It is neither reasonable nor possible to expect imams to
be spiritual advisers, welfare officers and terrorism experts all at
the same time. Their primary role and responsibility is to look after
prisoners’ spiritual needs and ensure the provision of adequate
religious services – they are not a substitute for services which
can and should be delivered by the prison service.
•	 Prison imams’ independence and credibility needs to be
protected. This requires a difficult balancing act: while wanting to
professionalise and ‘control’ the institution, prison services have
to make sure that prison imams reflect the diversity of the Islamic
faith and are not seen as corrupted by the authorities.
More generally, the analysis demonstrated that the prison environment
poses challenges and problems that are similar – if not the same – as
those encountered in society at large:
•	 Discrimination and anti-Muslim prejudice (sometimes referred to
as ‘Islamophobia’) are often said to provide the pretext for – and
one of the drivers behind – radicalisation and extremist recruitment
in Muslim minority countries. This is also true for prisons, except
that in prisons, the confrontation between Muslims and the state
(represented by the prison service) is more direct, frequent and
immediate than outside. Needless to say, there should be zero
tolerance for any form of racism and anti-Muslim prejudice among
prison staff.
•	 Conversion is not the same as radicalisation, and good counterradicalisation policies – whether in or outside prison – never fail
to distinguish between the two. Changes in religious practice,
especially the adoption of strict and literalist varieties of the Islamic
faith, need not pose a security problem, and prisoners’ freedom of
religious expression should always be respected. In fact, efforts to
‘clamp down’ on Salafi Islam in the name of counter-radicalisation
may have the opposite effect, namely to create the impression
that prisoners are singled out for negative attention by the prison
authorities on account of their (Islamic) faith alone. 66
•	 Given the nature of Islamist extremism, which combines political
and religious motifs and consciously uses the language and
symbols of faith, it can be difficult to distinguish between legitimate
expressions of faith and politically inspired extremism. To take the
best known example, the term ‘jihad’ can have vastly different 	
meanings depending on who it is used by and for what purpose. 67
Training for prison staff can go some way towards making basic
differences and subtleties understood, but it would be unrealistic
to expect prison guards to become experts in Islamic theology
and/or ‘jihadist’ ideology. Making shared resources available to
individual prisons, enabling them to seek advice and access
66	 It should be added that religion per se can be of great benefit for a prisoner’s psychological well-being,
especially when being deprived of freedom, friends, family and other things to hold on to. See Hamm,
‘Terrorist Recruitment’, Chapter 1. The author also wishes to thank Marc Linning for emphasising
this point.
67	 See, for example, Jared M. Brachman, Global Jihadism: Theory and Practice (London and New York:
Routledge, 2009), p. 5; John Esposito, Unholy War: Terror in the Name of Islam (Oxford: Oxford University
Press, 2002), Chapter 1.

36

Prisons and Terrorism Radicalisation and De-radicalisation in 15 Countries

specialised services (for example, translators and experts), clearly
has an important role to play too.
However, many of the issues that play out in the prison environment
are not actually ‘prison issues’ at all. They touch on fundamental
questions of counter-radicalisation policy, such as: what constitutes
radicalism and/or extremism? Is any discussion of foreign conflicts in
Islamic terms problematic, or should it be encouraged? Where should
authorities draw the line between acceptable yet radical political views
and political extremism? Can non-violent Islamists ever be partners in
combating radicalisation? Should prison authorities try to limit, ignore
or even encourage the spread of ‘quietist’ Salafi Islam?
The wider significance of these issues should be understood and
acknowledged by prison services. It is not up to prison governors or
even the heads of prison services to decide matters of national policy,
but they have every right to convey their views and experiences, and
expect their governments to issue clear guidance and direction where
national policy is not clear or consistent.

37

Prisons and Terrorism Radicalisation and De-radicalisation in 15 Countries

38

Prisons and Terrorism Radicalisation and De-radicalisation in 15 Countries

4 Collective De-radicalisation

and Disengagement

W

hen Bobby Sands and other IRA prisoners went on
hunger strike in 1981, this galvanised Irish nationalist
opinion across the island. Many Catholics saw them not
as murderers but as decent men – patriots even – who had been
locked up for acting on their convictions. In the words of Mairead
Corrigan, who founded the Northern Ireland peace movement and
had a strong record of opposing IRA violence: ‘They were men
from our community. We know how they have come to be there.
And above all, we don’t want them suffering within the prisons’. 68
Governments often fail to grasp the powerful position of terrorist
prisoners vis-à-vis their supporters and constituencies. In several
instances, the leaders of terrorist groups have used their standing
and influence as ‘living martyrs’ in order to pursue collective
processes of de-radicalisation and disengagement (for definitions
of these terms, see Box 1), which have resulted in the declaration
of permanent ceasefires and the standing down of armed groups.
However, this chapter will show that such instances are rare, and
the conditions in which they are likely to happen are limited.
Using data from Egypt, Algeria and Israel, 69 the analysis provided
in the following demonstrates that collective disengagement
and de-radicalisation is predicated upon the existence of strong
and authoritative leadership; the existence of hierarchical command
and control structures; and – not least – a conducive environment
in which the leadership perceives that the armed campaign has
faltered or is at risk of inflicting unacceptable losses to the group
or its wider constituency.
The chapter argues that – if those conditions are in place –
governments can (and should) play an active role in facilitating the
process, for example by supporting the dialogue between leaderships
and their grassroots; easing repressive measures for those willing
to join the process; and, importantly, making available early release,
financial support and other measures of socio-economic reintegration.
Depending on the nature and character of the process, even
measures of political reform and reconciliation may be appropriate.

4.1 Issue
Collective de-radicalisation and disengagement shares a number
of characteristics with the process of disarmament, demobilisation
and reintegration (DDR), which international institutions – especially
the United Nations (UN) – have long practiced in former conflict
zones. In essence, DDR is about facilitating the process whereby
armed insurgent groups cease to engage in organised violence
and their members are (re-)integrated into (post-conflict) civilian
68	 Mairead Corrigan, quoted in Padraig O’Malley, The Uncivil Wars: Ireland Today (Belfast: Beacon Press,
1983), p. 268.
69	 The authors of the country case studies were Omar Ashour (Algeria and Egypt), and Boaz Ganor and
Ophir Falk (Israel).

39

Prisons and Terrorism Radicalisation and De-radicalisation in 15 Countries

life, thus creating the foundations for economic recovery
and development.
The literature on DDR is extensive, 70 and some of its concepts
and practices are relevant to collective de-radicalisation and
disengagement. It is equally important, however, to recognise
the differences:
•	 DDR happens in the context of transitions from war to peace,
and its instruments are aimed at assisting the implementation
of existing peace agreements. This is not necessarily the case
for collective de-radicalisation and disengagement, where
ceasefires may have been declared but wider conflicts often
remain unresolved and no formal political negotiations have
taken place.
•	 DDR deals with issues that result from the standing down
of an armed group, whereas collective de-radicalisation and
disengagement focuses on facilitating the process whereby
armed groups decide to stand down in the first place. In other
words, collective disengagement and de-radicalisation happens
before DDR has even started.
What DDR and collective disengagement and de-radicalisation
share is their concern for the successful reintegration of insurgents
and/or terrorists, which both regard as an essential benchmark for
success. Unsurprisingly, then, it is in the areas of reinsertion and
reintegration that lessons from DDR are most relevant to collective
disengagement and de-radicalisation.

Pre-conditions
Collective disengagement or de-radicalisation attempts to bring
about a process which leads to large numbers of terrorists –
typically, the vast majority, if not all, of the members of an armed
group – changing their attitudes towards the use of violent means
or the pursuit of extremist aims. Two conditions are critical:
•	 The armed groups in question need to have functioning
leaderships which support and facilitate the process. To some
extent, therefore, the process of collective disengagement and
de-radicalisation is ‘top down’, because it typically emerges
from, and is actively pursued by, groups’ leaderships.
•	 Armed groups have to be hierarchically structured, so that any
decision to abandon violence – taken by the leadership and
negotiated in partnership with the grassroots – will be adhered
to by the organisation as a whole. Collective de-radicalisation
and disengagement is less likely to succeed in networked or
‘leaderless’ organisations where leaderships find it harder to
influence, or even reach, their membership.

70	 See, for example, Mark Knight, ‘Guns, Camps and Cash: Disarmament, Demobilization and Reinsertion
of Former Combatants in Transitions from War to Peace’, Journal of Peace Research, 41(4) (2004), pp.
499 – 516; Robert Muggah, ‘No Magic Bullet: A Critical Perspective on Disarmament, Demobilization and
Reintegration (DDR) and Weapons Reduction in Post-conflict Contexts’, The Round Table, 94(379) (2005),
pp. 239-52; Peter Swarbrick, ‘Avoiding Disarmament Failure: The Critical Link in DDR’, Small Arms Survey,
February 2007; Sigrid Willibald, ‘Does money work? Cash transfers to ex-combatants in disarmament,
demobilization and reintegration processes’, Disasters, 30(3) (2006), pp. 316-39.

40

Prisons and Terrorism Radicalisation and De-radicalisation in 15 Countries

Ideally, therefore, collective disengagement and de-radicalisation
requires most of the group – including its leadership – to be
incarcerated and their command and control structures to have
remained largely intact.

Challenges
Governments rarely provide the initial impetus for processes of
collective de-radicalisation and disengagement, but they can
be critical in assisting the process and leading it to a successful
conclusion. In doing so, governments need to:
•	 Recognise changes in attitudes and beliefs among armed
groups’ leaderships;
•	 Open appropriate channels of communication through which
collective de-radicalisation and disengagement efforts can
be supported;
•	 Provide the right mix of sanctions and inducements (the
proverbial ‘carrots and sticks’) that will keep the process
on track and move it towards a successful conclusion;
•	 Enable the release and re-integration of terrorists and/or
insurgents, ensuring that any return to violence
remains unlikely.
How a number of selected countries have addressed these
challenges will be examined in the next section.

4.2 Practice
This section looks at three countries which meet the pre-conditions
for collective de-radicalisation and disengagement: they have all
been confronted with armed groups whose structures have been
hierarchical and whose imprisoned members are concentrated in
so-called ‘security prisons’, where the groups maintain command
and control structures.
Even so, only two of the countries – Algeria and Egypt – have seen
actual disengagement and de-radicalisation processes succeed. The
third case – Israel – was included in the sample to demonstrate some
of the difficulties and obstacles in making collective disengagement
and de-radicalisation processes happen. Indeed, the analysis will
show that the advantages of dealing with leader-led, hierarchically
structured organisations can easily turn into stumbling blocks when
the wider environment is negative and/or leaderships are reluctant
to end their campaigns, making both collective and individual
disengagement and de-radicalisation efforts impossible.
The section is divided into three parts, dealing with leadership
and environment, repression, and inducements respectively.

41

Prisons and Terrorism Radicalisation and De-radicalisation in 15 Countries

Leadership and environment
The case studies in the sample show that authoritative leadership
and a conducive environment are essential to making collective
processes of disengagement and de-radicalisation work. Equally
important, however, is governments’ ability to recognise these
conditions and help facilitate the implementation of programmes.
The Egyptian Islamic Group (IG) is the best known example of
prison-based collective disengagement and de-radicalisation.
Faced with defeat, the loss of popular support and the prospect of
an ‘Algerian-type’ civil war, the IG’s leadership decided to embark
on a comprehensive re-interpretation of the group’s doctrine on the
legitimacy of using armed force, which resulted in the publication
of 25 volumes. In 2002, the leaders – who had themselves been
imprisoned – toured prison facilities across the country, holding
discussions (first) with middle-ranking commanders and (then) the
group’s foot soldiers. By the end, nearly 15,000 militants had agreed
to follow the new course. The process led to no significant splits,
and no terrorist attacks or armed operations have been attributed
to the IG since 1999.
The Egyptian government’s initial response to the process, however,
was marked by a lack of interest, if not suspicion. For four years after
the group first declared a ceasefire in 1997, there was little support.
Only after the September 11 attacks in 2001 did the government –
now under intense pressure from the United States – begin to help
the IG’s leadership in promoting the new course, mostly by organising
the leadership’s ten month tour of the country’s security prisons.
Even so, the government’s initial lack of response and even tacit
support had not only slowed down the IG’s collective process, it
also (unnecessarily) prolonged the group’s campaign of violence.
The disengagement of the Islamic Salvation Army (AIS) in Algeria
took place in the second half of the 1990s and concluded with the
standing down of the group in 2000. In contrast to Egypt, it was
the government which first approached the group in 1993. Three
attempts to negotiate a settlement with the imprisoned leadership
of the Islamic Salvation Front (FIS; the AIS’ political parent
organisation) led to nothing, partly because the FIS’ leadership still
felt confident about its chances of toppling the government. By
1997, the situation had changed dramatically, with Algeria having
descended into outright civil war and the various insurgent groups
rapidly losing support among the population (see below). This time,
the decision to embark on a process of collective disengagement
was taken by the AIS’ military leaders, who had hidden in the
group’s mountain strongholds. In the end, the political leadership,
though reluctant, had little choice but to accept the outcome of a
process which most of their imprisoned comrades (and their
families) supported.
In Israel, neither the political nor the military leaderships of the most
important Palestinian groups – Fatah and Hamas – have showed any
interest in pursuing collective de-radicalisation or disengagement
outside the context of a negotiated political settlement. Even when
they were under military and political pressure, the groups always
insisted on being treated as equals, negotiating on behalf of ‘their
people’ and achieving the settlement of the conflict as a whole. As
a consequence, there have been no known attempts by the Israeli

42

Prisons and Terrorism Radicalisation and De-radicalisation in 15 Countries

government to facilitate prison-based collective disengagement
or de-radicalisation efforts.

Repression
The question of whether government repression helps or hinders
disengagement and de-radicalisation efforts touches on wider
debates about the relative benefits of ‘hard’ and ‘soft’ counterterrorism approaches, which this report will not be able to resolve. 71
In the three cases that form part of the sample, government
repression has triggered conflicting responses, which – depending
on context and environment – have led to further radicalisation or
intensified efforts at promoting disengagement and de-radicalisation.
In the Egyptian case, government repression in the 1990s –
which involved the mass arrests of Islamist sympathisers and their
systematic mistreatment in prison – further radicalised sections of
the Egyptian population as well as the prisoners themselves. At
the same time, it seems clear that – over time – the government’s
excesses contributed to the IG leadership’s decision to re-think and,
eventually, change their mind on the use of violence. Violent jihad, the
group’s leadership argued, could be justified only if it furthered the
movement’s aims, yet the group’s violent campaign had achieved the
opposite: mosques had been closed, preachers banned, the families
of activists were suffering, and most of the movement had ended
up behind bars. According to one of the group’s leaders, if God
was on their side, none of this would have happened, which meant
that something had to be ‘theologically wrong’ with the decision to
confront the regime.
In Algeria, AIS’ response was similar, even if the context was entirely
different. The country had slid into civil war, with massacres of
civilians being committed by all sides, principally the Armed Islamic
Group (GIA) and the government. The massacres often took place in
AIS strongholds while most of the group’s active members had been
arrested and deported to detention centres in the Algerian desert,
leaving their families defenceless. The public, on the other hand,
often failed to distinguish between different Islamist groups, with the
(paradoxical) result that the AIS, whose supporters had been among
the principal victims of the GIA and government repression, felt it
needed to disengage from violence in order to stop the massacres
and prevent the loss of public support. In the leadership’s view, the
situation had become untenable – the ‘jihad’, in the words of an AIS
leader, ‘was about to be buried by its own sons’.
In Israel, despite the ongoing conflict, the situation never reached
a point at which repression alone would have ‘broken’ any of the
Palestinian groups. Palestinian security prisoners in Israeli prisons
have not been exposed to the kind of systematic mistreatment that
used to be common in Egypt. Unlike Algeria, the situation in the
Palestinian territories has never quite reached the point of sliding
into civil war, even if the conflict has imposed significant strains on
the civilian population for several decades. Moreover, the Israeli
government’s practice of releasing prisoners in return for kidnapped
Israeli soldiers or as ‘goodwill gestures’ in the context of peace
negotiations has not encouraged armed groups to change their
71	 See Ronald Crelinsten, Counterterrorism (Cambridge: Polity, 2008), Chapters 2 and 4.

43

Prisons and Terrorism Radicalisation and De-radicalisation in 15 Countries

behaviour but – rather – allowed them to ‘sit tight’ and wait
until it is their turn to be set free. 72

Inducements
Inducements and aftercare are important elements of most
disengagement and de-radicalisation programmes, because they
provide incentives for engaging with the process and then sustain the
transition from prison back into mainstream society. In certain cases,
collective programmes may go as far as offering political concessions
as part of a negotiated ‘package’ that, more typically, includes
promises of early release and cash payments.
Once the Egyptian government started embracing the IG’s
disengagement initiative, it quickly moved from suspicion towards
active facilitation. Not only did the repression stop and prisoners
were offered relaxed prison regimes (including, for example, better
meals, more visits, better opportunities to mix with other inmates,
etc.), the government organised the leadership’s tour of the country’s
security prisons and even helped with the publication of its books.
Also, for the first time, the group’s efforts were covered, and praised,
on state television.
Critical was the promise of early release for those who had joined the
programme. Of the 15,000 IG prisoners in 2002, 13,000 had been
released by 2006. In 2007, only a few hundred were left. Furthermore,
those who have been released now receive ‘pensions’, which –
though administered by the IG – are widely thought to be funded with
the assistance or, indeed, direct support from the government.
In Algeria, the government went even further. Not only were the
AIS’ prisoners released back into society, they received promises
of employment (which, however, were not always kept). Families of
FIS activists who had been victimised by the government were paid
compensation, and some former members were given permission
to wear guns in order to protect themselves against revenge attacks.
Political concessions formed an essential part of the ‘package’ that
had been negotiated with the Algerian government. It committed to
reducing the military’s influence over public life, and gave assurances
that former members of the AIS would be allowed to establish political
parties and participate in political life. Again, not all these promises
have been kept, but they illustrate the scale of the government’s
commitment and the very wide-ranging nature of the process, which
– in many respects – resembled a more traditional peace process.
Once more, the Israeli case demonstrates why it is impossible to
separate disengagement and de-radicalisation efforts – whether
collective or individual (see Chapter 5) – from the wider political
environment. Even if inducements were to be provided to prisoners
belonging to certain armed groups, or factions within those groups,
72	 That said, the Israeli government’s rapprochement with Fatah, and the consequent relaxation of prison
regimes for Fatah prisoners, especially when compared to supporters of Hamas and Islamic Jihad, has
made a significant contribution to stabilising the relations between Israel and the (Fatah led) Palestinian
Authority. It provides a good illustration of how prison policy can serve wider political imperatives, and
may – at some point – provide an important element of peace negotiations. The author wishes to thank
Marc Linning for this insight and idea.

44

Prisons and Terrorism Radicalisation and De-radicalisation in 15 Countries

it is far from clear whether they would convince them to
engage in collective disengagement or de-radicalisation. Short
of a comprehensive settlement, participating prisoners would be
released back into hostile communities, with participants likely to
be shunned by their families and neighbours. For such programmes
to work, all participants would have to receive resettlement aid and
move to entirely different locations, which neither the prisoners nor
their families would find acceptable.

4.3 Recommendations
Processes of collective disengagement and de-radicalisation have
proved powerful and effective in facilitating insurgents’ and terrorists’
transition from violence to non-violent activism. However, it should
be remembered that the circumstances in which they are likely to
succeed are quite limited:
•	 They seem to work best when the group’s military leadership and
most of their followers are behind bars.
•	 They also presuppose that groups have strong and authoritative
leaderships, which are able to exercise a degree of hierarchical
control.
•	 Most importantly perhaps, collective disengagement and
de-radicalisation tends to happen when the feasibility and/or
viability of the armed campaign has come into question. The
process, therefore, requires a conducive environment, that is,
a situation in which the leadership perceives that the armed
campaign has faltered or is at risk of inflicting unacceptable
losses to the group or its wider constituency.
Even when those conditions are met, the local and historical context
may dictate different kinds of approaches. For instance, the Israeli
case shows that collective disengagement or de-radicalisation are
unlikely unless they are accompanied by a full settlement of the
conflict. In fact, it may at times be difficult to distinguish the two,
especially when – as in the Algerian case – political concessions form
part of the ‘package’ that is offered to the terrorists and/or insurgents.
When success is judged to be likely, governments should help
facilitate the process:
•	 They can support the process of reform and dialogue, for example
by allowing reforming leaderships to meet with their followers; relax
prison regimes for those who have signed up to the programme;
and disseminate their writings and publications.
•	 Repression is a double-edged sword. While it may be useful
in ‘breaking’ an armed group’s will to carry on and shift their
perception from ‘possible victory’ to ‘certain defeat’ and
‘unacceptable losses’, it may also have radicalised many
prisoners in the first place.

45

Prisons and Terrorism Radicalisation and De-radicalisation in 15 Countries

•	 Whatever role repression plays prior to the beginning of a process
of collective disengagement and de-radicalisation, the cases that
have been looked at seem to show that – once a process has
begun – the easing of repressive measures provides an immediate
reward for those who have signed up to the programme.
•	 Inducements and aftercare can make an important contribution
to sustaining the success of collective disengagement and
de-radicalisation. Early release is by far the most important
inducement for those still in prison, while financial support and/or
employment appear to be critical in ensuring that armed groups
and their members remain peaceful.
The case of Algeria showed that political reform and reconciliation
may form part of a ‘deal’ that brings about collective de-radicalisation
and disengagement. This raises important questions about the nature
and identity of the process: if a full ‘package’ is negotiated that
includes the standing down of the armed groups in return for political
concessions, collective disengagement and de-radicalisation may turn
into a fully-fledged peace process and benefit from drawing on the full
set of instruments provided by conflict resolution and peace-making.

46

Prisons and Terrorism Radicalisation and De-radicalisation in 15 Countries

5 Individual De-radicalisation

and Disengagement

A

small sign at the Bagram detention facility in Afghanistan
sums up the approach that underpins individual
disengagement and de-radicalisation. It reads: ‘Turning
Taliban into productive citizens of Afghanistan one detainee at a
time’.73 Afghanistan has only recently started ‘rolling out’ individual
disengagement and de-radicalisation programmes across the
country’s Afghan-run prisons, but programmes in other countries
– aimed primarily at Islamist militants – have been in existence for
several years. Nearly a dozen countries are believed to have carried
out such programmes, yet – despite the considerable publicity
they have attracted – systematic evaluations of their content and
underlying principles are rare.
Examining programmes in Afghanistan, Indonesia, the Philippines,
Saudi-Arabia, Singapore, and Yemen, 74 this chapter demonstrates
that programmes are too different and too dependent on local
context and conditions to measure success and compare their
results across the board. Nevertheless, some of the key ‘ingredients’
that seem to be underlying the more effective and sophisticated
programmes have been identified. They include: a mix of prison
programming, consisting primarily of religious re-education and
vocational training; credible interlocutors who can relate to prisoners’
personal and psychological needs; consistent efforts to facilitate
prisoners’ transition into social networks away from extremism; and
the systematic fostering of long-term commitments towards family,
community, and the state, which aim to reduce opportunities for
re-offending and increase the social and material cost of doing so.
The chapter stresses that no one size fits all. What works in one case
may be counter-productive in another, which means that programmes
always need to be adapted to the specific context in which they are
run. Equally important, programmes cannot be judged in isolation
from the wider social and political context in which they take place.
Whether they can make a substantial, strategic contribution to ending
a campaign of terrorism often depends on whether the external
conditions are conducive. Even if they are not, their positive and
outward-looking approach can send a powerful message, which will
prevent extremists from exploiting prison as a grievance.

5.1 Issue
The programmes that are discussed in this chapter aim to
provide structures through which individual de-radicalisation and
disengagement can be facilitated. These structures can be fairly
loose and ad hoc, but they all share the aim of reducing the number
of active terrorists in a given society by helping individual terrorists

73	 Cited in Marisa Porges’ case study report on Afghanistan.
74	 The authors of the country case studies were Chris Boucek (Saudi-Arabia and Yemen), Rohan Gunaratna
(Singapore), Arie Kruglanski and Michelle Gelfand (Philippines), and Marisa Porges (Afghanistan).

47

Prisons and Terrorism Radicalisation and De-radicalisation in 15 Countries

abandon terrorism and easing their re-integration into
mainstream society. 75
As Bjørgo and Horgan have pointed out, all individual
disengagement and de-radicalisation programmes consist of
measures that deal with ‘push’ and/or ‘pull’ factors. Push factors
are the reasons why individuals may have started having doubts
about their involvement in terrorist groups, including, for example,
disillusionment with the group’s goals, its methods, leaders, social
relations, their own status within the group and the pressures
of being underground. Pull factors, on the other hand, are
comprised of the ‘incentives’ that may attract individuals to life
outside terrorism, such as the availability of amnesties or reduced
sentences, education and training, financial inducements, as well
as the possibility of establishing new social networks and a family. 76
Where programmes differ is in emphasis, implementation and in
how various measures are combined.
The difference between individual and collective disengagement
and de-radicalisation lies in the former’s focus on dealing with
individual members rather than insurgent or terrorist groups
as a whole. This explains why individual disengagement and
de-radicalisation programmes are all based in prisons, where
individuals can be assumed to be unhappy about their situation
(that is, being in captivity), and it may be easier to isolate them
from group discipline and other external influences.

Debates
Individual disengagement and de-radicalisation programmes have
caused much debate and controversy in recent years. 77 The most
obvious point of contention is the idea behind such programmes.
Not everyone agrees that terrorists, who have been found guilty
(or are strongly suspected) of the most heinous crimes, should be
released back into society and receive ‘rewards’ for ceasing an
activity they should not have started in the first place. By contrast,
the programmes’ supporters maintain that individual disengagement
and de-radicalisation is effective in reforming individuals, that it saves
lives and prevents future atrocities from taking place. There is no
simple solution for the ethical trade-offs involved in this debate, nor
can one or the other view be dismissed easily. Indeed, as will be
shown, the limits of what is ethically and politically acceptable in a
given society plays a significant role in shaping the programme and
may determine its sustainability and effectiveness.
A more practical debate revolves around measuring the
success of programmes. Policymakers – understandably – want to
understand how programmes compare and have researchers tell
them which ones are most effective. Such comparisons, however,
are not as straightforward as they seem. As Chris Boucek points
75	 Tore Bjørgo, remarks at the ‘Countering Violent Extremism’ conference in Amman, Jordan,
16 March 2010.
76	 John Horgan, ‘Individual disengagement’, pp. 21-2; Tore Bjørgo, ‘Processes of disengagement from
violent groups of the extreme right’ in Bjørgo, Leaving Terrorism, pp. 36-40. Also Michael Jacobson,
‘Terrorist Dropouts: Learning From Those Who Have Left, Policy Focus #101, The Washington Institute
for Near East Policy, January 2010, Chapter 2.
77	 See, for example, Shiraz Maher, ‘A Betty Ford Clinic for Jihadis’, The Sunday Times, 6 July 2008;
Peter Neumann, ‘Table Tennis for the Taliban’, Spiegel Online, 28 January 2010; Marisa Porges and
	
Jessica Stern, ‘Getting Deradicalization Right’, Foreign Affairs, May/June 2010.

48

Prisons and Terrorism Radicalisation and De-radicalisation in 15 Countries

out, recidivism rates ‘may not be the best metric with which to
measure [relative] success’ 78, especially when many programmes
have started only very recently and many years may be needed
to gauge whether or not an individual has been fully ‘rehabilitated’.
Moreover, programmes differ in relation to who is eligible, with
many programmes excluding hardcore militants in favour of those
who were only marginally involved. This skews the results and
makes programmes which are open to all offenders look less
successful than those which concentrate on the ‘easy’ cases.
The most significant obstacle to measuring success is
undoubtedly the (often overlooked) fact that individual disengagement
and de-radicalisation programmes do not occur in a political and
societal vacuum, and that – consequently – the effectiveness
of such programmes cannot be judged without considering the
political circumstances in which they take place. For example,
the disengagement programme in Iraq, which is widely seen as
a success, may have produced entirely different results had it been
introduced in 2006, when Iraq was on the brink of a civil war and
the Sunni insurgency was gaining ground, instead of 2008, when
the Sunni population had turned against the insurgents. In fact,
General Douglas Stone, who commanded the U.S. detention facilities
in the country at the time, went as far as conceding that ‘improved
security’ – not the programme itself – had been critical in reducing
recidivism rates. 79 Programmes’ effectiveness, therefore, may at
times have little to do with any of their individual components but
– rather – be related to the broader environment in which they are
carried out, making comparisons that are based on the programmes’
content alone methodologically questionable.
A related debate concerns the transferability of individual
disengagement and de-radicalisation programmes. Many countries,
including several Western states, have been playing with the idea of
running their own ‘de-rad’ programmes. However, there is no one
‘template’ or ‘blueprint’ that could be copied and pasted. Not only
do programmes depend on the context and (political) environment
in which they are implemented, they often rely on uniquely local
dynamics and structures – tribal customs in the Middle East, for
example – that may be difficult to recreate elsewhere. Indeed, there
seems to be an emerging consensus that no one size fits all: while
certain general principles may apply everywhere, even the most
appealing programmes have to be adapted to go with the grain
of the societies in which they are set.
Perhaps the most thorny question is what role individual
disengagement and de-radicalisation programmes can (and should)
play in relation to countries’ wider counter-terrorism strategies. Are
they a substitute for ‘hard’ counter-terrorism measures, or should
they be treated as complementary? Can the difference they make be
strategic, or are they more like ‘clean-up’ operations which should
only be used once a particular terrorist campaign or insurgency is
winding down anyway? In other words, are they worth all the money,
effort, and attention they have recently received?

78	 See Chris Boucek’s case study report on Saudi-Arabia.
79	 Anne Speckhard, ‘Prison and Community Based Disengagement and De-Radicalization Programs
for Extremists Involved in Militant Jihadi Terrorism Ideologies and Activities’, Studies in Conflict and
Terrorism, forthcoming.

49

Prisons and Terrorism Radicalisation and De-radicalisation in 15 Countries

Challenges
Needless to say, this report will be unable to resolve all the dilemmas
and debates that were outlined above. Nevertheless, in examining
the programmes that formed part of the sample, a number of specific
questions and challenges may help to inform the discussion:
•	 How have programmes addressed – and been conditioned
– by the wider environment in which they were implemented?
•	 What kinds of incentives and instruments are deployed and in
what combination?
•	 Who leads the programmes and how?
•	 How do programmes ensure that prisoners’ disengagement
and/or de-radicalisation is sustained after their release?
These questions will guide the analysis of six individual
disengagement and de-radicalisation programmes in the next section.

5.2 Practice
As Anne Speckhard and others have pointed out, prison-based
disengagement and de-radicalisation has become ‘fashionable’ in
recent years. 80 For reasons of time and scope, not all the existing
programmes have been included in the sample – missing, for
example, are Jordan, Iraq, Malaysia, and Qatar. In selecting the case
studies, the intention was to ensure a good range of older and newly
emerging programmes, and combine programmes that seemed
highly sophisticated with those that appeared looser and more ad
hoc. Hence, the sample includes the two most developed and best
resourced programmes (Saudi-Arabia and Singapore) and two which
relied mostly on individual initiative and had only limited resources
at their disposal (Yemen and Indonesia). In addition, it comprises
two programmes which – at the time of writing – have only been
piloted and are yet to be rolled out across their respective countries
(Afghanistan and the Philippines).
As always, it is important to keep in mind that this report can only
provide the briefest insight, and that – for a full understanding – they
need to be considered in greater detail and on their own merit. The
first three variables that have been examined are often thought to
be the core, prison-based elements of individual disengagement
and de-radicalisation programmes, namely the actual programming
(vocational training and ideological/religious ‘re-education’); the role
of interlocutors; and the availability of inducements and material
incentives. The remaining variables look beyond the prison walls,
considering the involvement of external stakeholders; after-care
provisions; and the likely impact of the wider political and
social environment.

Programming
One of the most contentious debates among students of
radicalisation is about the role of (religiously inspired) ideology in the
process that leads to violent extremism. Does ideology matter, or
has its role and significance been overrated at the expense of more
80	 Ibid.

50

Prisons and Terrorism Radicalisation and De-radicalisation in 15 Countries

important factors, such as group relations and/or grievances? 81
It should come as no surprise, then, that different approaches
towards finding the right combination of religious and ideological
‘re-education’ and other instruments of prison programming,
especially education and professional training, can be found
among the programmes in the sample.
Four of the programmes contain a substantive element of religious
and ideological re-education – partly, one may argue, because they
were inspired by each other. All of them share the notion of terrorists
as ‘victims’ whose religious ideas are based on misinformation and
a lack of proper knowledge about Islam, making re-education and
reform both necessary and possible. The programme in Yemen,
for example, which ran from 2002-05, consisted of a series of
religious dialogues between prisoners and religious clerics, which
aimed at convincing the prisoners that their justification for taking
up violent ‘jihad’ had no basis in scripture. The Saudi programme,
which emerged in 2004, initially adopted Yemen’s strong emphasis
on religious ‘re-education’, with prisoners having to attend courses
in which contentious religious concepts are discussed. Formally
launched in 2003, the Singaporean Religious Rehabilitation
Group has been led by religious clerics and understood religious
re-education to be its core mission. The programme in the
Philippines, which is about to be rolled out, will be modelled
on the Singaporean experience and have a similar approach.
Despite the similar emphasis, the scope and content of the religious
re-education efforts differs significantly. In Saudi-Arabia and Yemen,
the authorities’ ambitions have been fairly modest. In both cases,
their focus has been on re-affirming the Islamic legitimacy of their
respective regimes, avoiding more complicated questions such as
the permissibility of ‘jihad’ abroad and/or against un-Islamic regimes.
In Muslim minority countries, where the case for ‘Islamic legitimacy’
is more difficult to make, religious re-education has been more farreaching and comprehensive. In Singapore and, by extension, the
Philippines, the aim is to challenge and de-construct the full range of
concepts that underlie and inform the Islamist militant movement. 82
With the possible exception of Yemen, however, none of the
programmes consist of religious or ideological re-education alone.
As the Saudi programme has evolved, more emphasis has been put
on fostering ‘social responsibility’ among prisoners, resulting in the
creation of job training and education schemes, which are meant
to provide the basis for a meaningful and socially secure existence.
Furthermore both Saudi-Arabia and Singapore have recently added
artistic classes to their programming.
By contrast, neither Indonesia nor Afghanistan pay much
attention to religious re-education. Indonesia’s first full programme,
which ran from 2005 to 2007, had former terrorists (rather than
religious clerics) as interlocutors (see below), and their central
argument revolved around the strategic utility of armed force,
not questions of principle or religious legitimacy. In Afghanistan’s
emerging programme, the assumption is that (Taliban) foot soldiers
carry little ideological baggage, and that they are best served by
81	 See, for example, Tore Bjørgo (ed.), Root Causes of Terrorism (New York and London: Routledge, 2005);
Sageman, Leaderless Jihad.
82	 For an overview of these concepts, see Brachman, Global jihadism, Chapter 2.

51

Prisons and Terrorism Radicalisation and De-radicalisation in 15 Countries

– in the first instance – separating them from the ‘ideologues’ and
– then – providing them with useful skills (beginning, in many cases,
with reading and writing) which will allow them to become valued,
respected and socially responsible members of their communities
upon their return.
Given the complexity of the programmes, it is difficult to say
whether religious re-education or skills training, or any combination
thereof, works best. The Yemeni programme, for example, is widely
considered a failure, but – as will be shown – this cannot be explained
by its reliance on religious dialogues alone. Nevertheless, it is worth
noting that, as the Saudi programme has evolved, it has progressively
adopted additional instruments, indicating that the strong initial
emphasis on religious re-education was no longer seen as sufficient.
Furthermore, the sample illustrates that the type of programming
offered is informed, at least to some extent, by the needs and
nature of the prisoner population. The creators of the Singaporean
programme, for instance, assumed they were dealing with
‘ideologues’, who had been indoctrinated in the religious schools of
the Jemaah Islamiyah (JI), whereas Taliban foot soldiers in Afghanistan
are thought to be mostly untouched by ideology. If those assumptions
are correct, both approaches – though different – may be correct.

Interlocutors
Almost all of the programmes in the sample have come to recognise
the critical role played by those who deliver the programme. The
principal interlocutors in most of the programmes are religious clerics,
who have been confronted with varying degrees of suspicion because
of their association with the government.
In Yemen, the solution to this dilemma was for the clerics to challenge
the prisoners to enter into a ‘dialogue of equals’, with both sides
pledging to accept the other side’s arguments if they were seen as
more plausible and convincing. (Needless to say, the government’s
clerics won all rounds of this ‘competition’.) In Saudi-Arabia,
Singapore and the Philippines, the approach has been to compensate
for the clerics’ lack of perceived legitimacy with added social and
psychological skills. In Singapore, for example, as well as being
competent scholars of Islam, clerics need to train as counsellors and
demonstrate their ability to relate to prisoners’ psychological needs.
In Saudi-Arabia, clerics who are not seen as ‘brotherly’ are ruthlessly
screened out, regardless of their religious qualifications or knowledge.
The two notable exceptions to having clerics as the principal
interlocutors are Indonesia and Afghanistan. In Indonesia, the
programme was run by former terrorists, who may not have
possessed much religious knowledge but were nevertheless
regarded as ‘credible’, if not ‘charismatic’, by many of the prisoners,
in particular those they had personally recruited or trained. 83 The
fledgling programme in Afghanistan, on the other hand, does not
seem to have given much thought yet to identifying any principal
interlocutor. Given how important their role in establishing credibility
and social bonds, this may turn out to be a major shortcoming.
83	 However, they were much less successful with prisoners who belonged to other groups or had no prior
relationship to them.

52

Prisons and Terrorism Radicalisation and De-radicalisation in 15 Countries

Inducements
The inducements offered to prisoners are the most public,
and in many ways the most controversial, aspect of individual
disengagement and de-radicalisation programmes. Typically,
inducements come in the form of more comforts during participants’
time in prison; early release or amnesty; and financial assistance
during and after their stay in incarceration. In the public’s eye, these
incentives are often perceived not as essential ingredients in the
process of rehabilitation but, rather, as ‘rewards for terrorism’.
Such perceptions exist not only in Western countries, where
de-radicalisation and disengagement programmes are frequently
regarded with suspicion, but also in the (non-Western) countries in
which the programmes are carried out. Indeed, the governments that
are currently running programmes have always had to gauge what
level of ‘generosity’ towards terrorist prisoners would be acceptable,
and this has undoubtedly played a role in preventing certain kinds of
prisoners from taking part. In Yemen, for example, terrorists who were
suspected to be guilty of ‘serious offences’ were excluded from the
religious dialogues. In Saudi, individuals with ‘blood on their hands’
may participate in the programme but will not be released early.
Furthermore, there have been significant variations in the way in
which material inducements have been delivered. At one end of the
spectrum is the Indonesian approach, which saw former terrorists
going into prisons with cash in their hands, handing out money to
everyone who seemed willing to cooperate. By contrast, the Saudi
programme provides very significant material incentives too, but
it calibrates them more carefully, offering financial assistance to
prisoners’ families during their stay in prison as well as providing
prisoners with the means for a new beginning (including cars,
houses, and flat screen TVs) after their release. The Singaporean
programme is more modest, but – like the Saudi programme – it
combines support for prisoners’ families during their incarceration
with ‘re-settlement’ aid after the prisoners have been released. It
is safe to assume that the emerging programme in the Philippines
will follow a similar model.
The only programme in which material incentives do not seem to
play a major role is that in Afghanistan. Here, the thinking seems to
be that early release, combined with the long-term effects of
equipping prisoners with practical skills, education and the means
to earn an independent livelihood, will more than outweigh any
amount of (short-term) financial assistance. If this is seen to work,
it could represent an interesting counter-point to the Saudi and
Indonesian approaches. In either case, what clearly does not work
is the Yemeni tactic of promising financial assistance and then not
paying up, which is cited by many observers as one of the reasons
why many participants in the Yemeni dialogues returned to violence.

External stakeholders
All the individual disengagement and de-radicalisation
programmes in the sample acknowledge that, for offenders to
desist from deviant behaviour, it is essential to facilitate their
re-integration into mainstream social networks, encouraging them
to establish stable commitments outside extremist circles. Simply

53

Prisons and Terrorism Radicalisation and De-radicalisation in 15 Countries

put, the more committed a former terrorist is to social networks
away from extremism, the fewer opportunities there will be to
return to violence and the higher the (social) cost of re-offending.
The nature of the social networks which are being leveraged by
individual disengagement and de-radicalisation programmes differ
from country to country. Even so, all the programmes have identified
the family as an important focal point. The most basic way of
involving prisoners’ families in the process of disengagement and
de-radicalisation is by providing economic assistance or employment
opportunities for wives and relatives, which – for instance – a more
recent iteration of the Indonesian programme has incorporated. In
Yemen, families were meant to vouch for the good behaviour of a
former terrorist by signing a ‘contract’ with the government. In both
cases, the involvement of families has been rather crude and rested
on questionable assumptions, such as the idea that prisoners’ wives
and families are necessarily in need of economic assistance, or
that families are always pro-government and will honour their
commitment to ensuring ‘good behaviour’.
More sophisticated mechanisms exist in Saudi-Arabia and Singapore.
In Saudi, families’ needs are individually assessed and tailored to their
particular needs. Throughout the programme, ‘family reconciliation’
is a priority, and their active participation in aspects of the programme
is actively encouraged. (Where prisoners are unmarried, the
authorities are known to have helped finding wives and paid for
weddings.) Similarly, Singapore seeks to include prisoners’ families
in religious re-education and psychological counselling sessions and,
thereby, make them active stakeholders in the process from day
one. The underlying approach and emphasis, however, is markedly
different: whereas the Saudi programme assumes prisoners have
been abandoned by their families and are in need of ‘reconciliation’,
Singapore believes that prisoners’ families are radical too and should
be ‘reformed’. Though different, both approaches may be effective
in their respective environments, reflecting the peculiar dynamics of
violent extremism in the two countries.
Going beyond families, many programmes have aimed to mobilise
larger social networks. In Saudi and Yemen, for example, prisoners’
tribes are expected to vouch for a prisoners’ good behaviour.
Singapore has involved community and professional bodies in its
rehabilitation programme, and a similar approach is likely to be
adopted in the Philippines. The case of Afghanistan, on the other
hand, demonstrates the very substantial obstacles to reaching out to
external stakeholders when a country is effectively in a state of civil
war. Despite attempts to involve village elders and organise symbolic
‘welcome ceremonies’, the programme struggles to overcome tribes’
hostility and the difficulties involved in travelling around the country.

After-care
In the context of disengagement and de-radicalisation, after-care is
designed to help prisoners ease back into society and sustain their
newfound commitment to refraining from terrorism. Accordingly, the
more sophisticated programmes invest considerable resources in
making after-care provisions effective.

54

Prisons and Terrorism Radicalisation and De-radicalisation in 15 Countries

The Saudi programme is undoubtedly the most extensive. It
creates so many obligations and personal commitments on behalf
of the former prisoners that it would seem almost irrational – if not
physically impossible – for anyone to return to violence. In addition to
periodical ‘check-ups’ with members of the security services, Saudistyle after-care involves the above-mentioned social and economic
assistance for former prisoners and their families, help with finding
wives, jobs and even start-up funds for budding entrepreneurs.
Equally important are regular meetings with their interlocutors, the
religious clerics (see above), whose continued association with the
former prisoners eases their transition from prison to society and
allows them to carry on a social relationship which is likely to have
been critical in their transformation.
Singapore uses a similar approach, even if the ‘golden handcuffs’
on offer are less generous than the Saudis’. In Indonesia, the
approach towards after-care seems to have evolved. Initially, little
systematic after-care was provided, except for former prisoners’
relationships with the former terrorists who had convinced them
to abandon their commitment to violence. More recently, the
Indonesian authorities have put into place extensive ‘livelihood’
programmes, which make former prisoners and their wives participate
in co-operative enterprises. Whether this will deliver the expected
benefits remains to be seen: while providing former prisoners with a
stable income, the scheme does not replace their social networks.
On the contrary, it cements their relationships with other former
extremists and may expose them to the danger of ‘re-radicalisation’.
Yemen is the best example of how not to do it. The government
showed little interest in after-care and made few resources available.
Rather than receiving the support and assistance which they had
been promised, former extremists often ended up being the victims
of police harassment. It became obvious that the government’s
commitment to the programme was low, and this – as well as
many other factors – explains why many of the programme’s
participants returned to violence. In Afghanistan, on the other hand,
the government and the Coalition forces would very much like to
provide after-care, but are prevented from doing so by the turbulent
conditions in the country. As even the authorities admit, systematic
follow-up will be difficult, if not impossible, and this may well turn
out to be the programme’s Achilles heel.

Environment
As noted above, individual disengagement and de-radicalisation
programmes cannot be looked at in isolation from the wider social
and political environment in which they are carried out. In judging their
success, it is vital, therefore, to consider external conditions and how
they are likely to have influenced the programmes’ outcomes.
Afghanistan, for example, launches its programme in the worst
possible conditions. Not only does the ongoing insurgency prevent
the implementation of after-care and the building of relationships with
external stakeholders (see above), former prisoners will be exposed
to all kinds of radicalising influences when they return to their villages.
Unlike Iraq, where the US introduced a similar programme in 2007-08,
the communities from which Taliban are recruited have not yet turned
against the insurgents. On the contrary, as long as the Taliban seem

55

Prisons and Terrorism Radicalisation and De-radicalisation in 15 Countries

to be winning the wider conflict, it is doubtful whether any strategy
that relies on ‘turning Taliban into productive citizens... one detainee
at a time’ will make a meaningful contribution to turning the tide.
Contrasting the programmes in Indonesia and Yemen provides further
compelling insight. Both programmes were highly unstructured,
and they both relied on the personal initiative of a small number of
individuals. In many ways, in fact, they were both deeply flawed, yet
the Indonesian programme and its successors are widely considered
successes, whereas the programme in Yemen is seen as a failure.
The difference lies in the external conditions under which they were
taking place. In Indonesia, local conflicts in Poso and Maluku were
winding down, with the programmes serving as ‘clean-up operations’,
dealing with men who were unlikely to return to violence now that
their respective causes had gone away. In Yemen, by contrast,
local conflicts were starting to intensify when the programme was
launched, and global developments – especially the invasion of Iraq
– appeared to confirm rather than contradict the former prisoners’
earlier beliefs.
Even on the Saudi programme, which has attracted most of the
attention in recent years, the jury is still out. Hailed as a success by
the Saudi government itself, it is likely to have benefited substantially
from the Saudi population’s increasing rejection of Al Qaeda and its
methods. Whether its overall success can be sustained, and how
this will affect recidivism rates, remains to be seen.

5.3 Recommendations
The emergence of individual disengagement and de-radicalisation
programmes has added a new – and hitherto unknown – instrument
in countering terrorism. Their novelty means that experts and
policymakers still do not know enough about how these programmes
work, and what contribution they can make to the fight
against terrorism.
As noted at the beginning of this chapter, one of the biggest
difficulties lies in measuring success, as different rules on eligibility,
changing external conditions and local contexts, and the short
period for which most of these programmes have been running, have
produced data that is difficult to judge and nearly impossible
to compare.
This report has found no solution to this problem – in fact, if anything,
it argues that a solution is unlikely to be found. Nevertheless, based
on the sample, the following observations can be made:
•	 The more developed programmes contain a mix of different kinds
of programming, typically combining ideological and/or religious
re-education with vocational training.
•	 Credible interlocutors are key. Also important is their ability to
establish ‘brotherly’ relationships and relate to the prisoners’
personal and psychological needs.

56

Prisons and Terrorism Radicalisation and De-radicalisation in 15 Countries

•	 Good programmes pay attention to facilitating prisoners’ transition
from prison back into mainstream society, typically by providing
them with the means for a new beginning and by establishing (or
re-establishing) social networks away from extremism.
•	 One of the overriding aims is to reduce opportunities for
re-offending and increase the social and material cost of doing
so. Much of the activities in sophisticated programmes are
consequently geared towards locking prisoners into commitments
and obligations towards family, community, and the state.
•	 Inducements play an important role in every programme, but
material incentives do not seem to be decisive on their own.
These principles can help guide country’s efforts in bringing elements
of disengagement and de-radicalisation to their own prison systems,
even – and especially – when they are reluctant, for political or
other reasons, to establish a full and formalised disengagement or
de-radicalisation programme of their own.
Another important finding is that seemingly effective programmes
cannot be simply ‘copied and pasted’. One size does not fit all.
What works in one case can be counter-productive in another. For
programmes to be effective, their scope, structure and instruments
must reflect local contexts and conditions, in particular:
•	 the prisoner population, and their individual and collective needs
and motivations;
•	 the nature and ideology of the groups to which prisoners used
to belong;
•	 the society from which they originate, its structure and customs;
•	 the dynamics of the wider conflict and other external conditions,
which may affect the programme’s outcome.
Furthermore, in order for programmes to be sustainable, they need
to consider what the societies in which they take place find politically
and ethically acceptable.
Finally, what contribution can such programmes make? How
important are they as instruments in the fight against terrorism? The
analysis in this chapter shows that they can be significant when both
external conditions and the wider political environment are conducive.
In particular, they can play an important role in facilitating the
transition from conflict to peace when the political momentum is no
longer with the terrorists and/or conflicts are winding down.
Whether they can make a substantial, strategic contribution on their
own remains to be proven. The difficulties in Afghanistan show that
an over-reliance on individual disengagement and de-radicalisation
programmes as the primary means of conflict resolution or counterterrorism would be disastrous as long as the overall situation
continues to deteriorate. More generally, it should always be
remembered that disengagement and de-radicalisation programmes
can never be judged in isolation from the wider political and social
context in which they are carried out.

57

Prisons and Terrorism Radicalisation and De-radicalisation in 15 Countries

That, of course, does not mean that they are pointless:
•	 Even when external conditions are adverse and programmes are
unlikely, therefore, to make a substantial contribution to reducing
the number of people engaged in terrorism, they may prevent
extremists from exploiting prison as a grievance.
•	 Examples of individuals who have abandoned terrorist groups are
a powerful tool in discouraging others from joining, especially by
offering a credible counter-balance to the radical narrative. 84
•	 De-radicalised or disengaged individuals are more likely to
cooperate with the authorities in a meaningful way. They may aid
on-going investigations, serve as government witnesses or work
with prison authorities to encourage other prisoners to abandon
violent associations.85
•	 Like terrorism and radicalisation, successful disengagement and
de-radicalisation can be contagious. Only recently, the example
set by the Saudi programme has prompted Islamist militant
prisoners in neighbouring Jordan to ask for religious counselling
and opportunities to ‘de-radicalise’ of their own. 86 The impact of
disengagement and de-radicalisation programmes, in other words,
can be truly global.
Most importantly, individual de-radicalisation and disengagement
programmes show that prisons are not just about locking people
away, but that they can make a real and positive contribution to
tackling problems of radicalisation and terrorism in society
as a whole.

84	 The author’s report is grateful to Vanessa Haas Hood for this idea.
85	 Ibid.
86	 Hamed El Said, remarks at the ‘Countering Violent Extremism’ conference in Amman, Jordan,
16 March 2010.

58

Prisons and Terrorism Radicalisation and De-radicalisation in 15 Countries

Conclusion
Prisons matter. Often ignored by the public and policymakers,
they are important vectors in the process of radicalisation, and
they can be leveraged in the fight against it.
Much of the public debate about prisons and terrorism is about
locking people away. This report has aimed to develop a more
sophisticated understanding of the role prisons can play in
radicalising people – and in reforming them. In doing so, it has
analysed the policies and approaches of 15 countries, identifying
trade-offs and dilemmas but also principles and best practices that
will help governments and policymakers spot new ideas and avoid
costly and counterproductive mistakes.
Take, for example, prison regimes for terrorists. Governments
everywhere have had to address a trade-off between wanting
to treat terrorists ‘just like other prisoners’ and preventing them
from mobilising outside support, recreating operational command
structures, and radicalising others. The report showed that there
are no hard and fast rules about whether terrorist prisoners
should be concentrated, separated and/or isolated. In fact, most
of the countries that have been looked at practice a policy of
dispersal and (partial) concentration, which distributes terrorists
among a small number of high security prisons. Even within such
mixed regimes, however, it rarely seems to be a good idea to
bring together leaders with followers and mix ideologues
with hangers-on.
The wider, and perhaps even more important, problem is that – in
most of the countries that have been looked at – prison regimes for
terrorists are informed by the demand for security before everything
else. While understandable, the ‘security first’ approach has resulted
in missed opportunities to promote reform. Many prison services
seem to believe that the imperatives of security and reform are
incompatible. In reality, though, reform does not need to come at
the expense of security. Prison services should be more ambitious
in promoting positive influences inside prison, and develop more
innovative approaches in facilitating prisoners’ transition back into
mainstream society.
Another issue which this report has devoted much attention to is
that of prison-based radicalisation. Prisons are often said to have
become breeding grounds for radicalisation. This should come as no
surprise. Prisons are ‘places of vulnerability’, which produce ‘identity
seekers’, ‘protection seekers’ and ‘rebels’ in greater numbers than
other environments. They provide near-perfect conditions in which
radical, religiously framed ideologies can flourish. While the extent
of the problem remains unclear, the potential for prison radicalisation
is significant, and the issue clearly needs to be addressed.
Based on the research, it seems obvious that over-crowding
and under-staffing amplify the conditions that lend themselves
to radicalisation. Badly run prisons also create the physical and
ideological space in which extremist recruiters can operate at
free will and monopolise the discourse about religion and politics.

59

Prisons and Terrorism Radicalisation and De-radicalisation in 15 Countries

For many prison systems, therefore, the first and most important
recommendation is to improve general conditions, avoid overcrowding, train staff, and provide meaningful programming that
allows prisoners to develop stable inmate identities. Prison imams
are important in denying religious space to extremists, but they
are not a panacea.
Not all the findings in this report are negative. The report shows
that, while certain countries fall short of even the most basic good
practices, others have recognised the enormous potential for prisons
to become net contributors in the fight against terrorism, and have
encouraged – sometimes sponsored – initiatives that seek to
promote disengagement and de-radicalisation.
For example, where groups are hierarchical and have strong,
authoritative leaderships, collective disengagement and
de-radicalisation becomes possible. Such processes have to be
carefully managed. When political concessions form part of the ‘deal’
between the government and the terrorists, collective disengagement
and de-radicalisation may, in fact, assume the character of a fullyfledged peace process, and requires the necessary skills, resources,
and – above all – patience.
Individual disengagement and de-radicalisation, on the other hand,
remains understudied and is often misunderstood – despite all the
publicity that programmes have attracted in recent years. Looking
at six individual disengagement and de-radicalisation programmes
in the Middle East and South East Asia, the report has identified key
principles, which will help policymakers understand the phenomenon
and identify elements of best practice that can be adopted in their
own prison systems.
Whether individual disengagement and de-radicalisation programmes
can make a strategic contribution to bringing a terrorist campaign
to an end remains to be proven. The programmes that have been
examined suggest that they can – as long as the political momentum
is no longer with the insurgents and other external conditions are
conducive. An over-reliance on individual disengagement and
de-radicalisation programmes as the primary means of fighting
terrorism should be avoided, therefore – they complement rather
than substitute other instruments in the fight against terrorism.
In bringing together the experiences of 15 countries, the report
has attempted to show the diversity of policy and practice across
the world. Not every lesson may be relevant or applicable in every
context, but – taken together – they demonstrate the enormous
possibilities for prisons to make a positive and significant
contribution to countering terrorism.

60

Prisons and Terrorism Radicalisation and De-radicalisation in 15 Countries

Acknowledgements
Many individuals were involved in making this report possible. First
among them are the authors of the 15 case studies, whose insights
and expertise underpin the entire analysis: Dr Omar Ashour, Laila
Bokhari, Dr Chris Boucek, Andrew Coyle, Ophir Falk, Dr Boaz Ganor,
Dr Michelle Gelfand, Dr Rohan Gunaratna, Dr Bob de Graaff, Eelco
Kessels, Dr Arie Kruglanski, Dr Jean-Luc Marret, Sidney Jones,
Marisa Porges, Dr Manuel Torres, and Dr Bert Useem.
We wish to thank the governments whose support – financial and
intellectual – allowed us to turn our idea into reality. Particularly
important were Julie Dowdle and Ambassador Bill Paterson at the
Australian Department of Foreign Affairs and Trade; Dr Hans van Miert
at the Dutch Counterterrorism Coordinator; and Prof Paul Grasby at
the British Home Office.
Also important were the reviewers, who took the time to read the
report, provide feedback and correct mistakes. We are grateful to
Gary Ackerman, Dr John Bew, Ryan Evans, Vanessa Haas Hood,
Marc Linning, and Dr Michael von Tangen Page.
At ICSR, numerous people helped in keeping the project on track,
including Andreas Berg, Steve Rickatson, and Katie Rothman. An
essential role was played by Alexandra Matine, who coordinated the
entire process and organised the November 2009 workshop, which
brought everyone together. We should also mention ICSR’s Trustees
– Henry Sweetbaum, Rt Hon Kim Campbell, and John Sacher – who
kept us focused.
ICSR’s partners at START – especially Dr Gary LaFree, Kathleen
Smarick, and Danielle Hawkins – were always helpful and willing to
put their network’s impressive intellectual resources at our disposal.
Needless to say, any and all mistakes in this report are the
author’s alone.

61

Prisons and Terrorism Radicalisation and De-radicalisation in 15 Countries

62

Prisons and Terrorism Radicalisation and De-radicalisation in 15 Countries

Further Reading
Books, articles and policy reports
•	 Anti-Defamation League, ‘Dangerous Convictions: Extremist
Activities in Prisons’, Topics in Extremism, 2(2002).
•	 Omar Ashour, The Deradicalization of Jihadists (London:
Routledge, 2009)
•	 James A. Beckford, Daniele Joly and Farhad Khosrokhavar,
Muslims in Prison: Challenge and Change in Britain and France
(Basingstoke: Palgrave Macmillan, 2005).
•	 Tore Bjørgo and John Horgan (eds.), Leaving Terrorism Behind
(London and New York: Routledge 2009).
•	 Jared M. Brachman, Global Jihadism: Theory and Practice
(London and New York: Routledge, 2009).
•	 James Brandon, Unlocking Al-Qaeda: Islamist Extremism in British
Prisons (London: Quilliam Foundation, 2009).
•	 Ian M. Cuthbertson, ‘Prisons and the Education of Terrorists’,
World Policy Journal, 22 September 2004.
•	 Mark S. Hamm, ‘Terrorist Recruitment in American Correctional
Institutions: An Exploratory Study of Non-Traditional Faith Groups
– Final Report’, US Department of Justice Commissioned Report,
December 2007.
•	 Greg Hannah, Lindsay Clutterbuck and Jennifer Rubin,
Radicalization or Rehabilitation: Understanding the Challenge of
Extremist and Radicalized Prisoners (Cambridge: RAND, 2008).
•	 Ellie Hearne and Naureen Chowdhury Fink, ‘Beyond Terrorism:
Deradicalization and Disengagement from Violent Extremism’,
International Peace Institute report, October 2008.
•	 John Horgan, Walking Away From Terrorism: Accounts of
Disengagement from Radical and Extremist Movements (New York
and London: Routledge, 2009).
•	 Michael Jacobson, ‘Terrorist Dropouts: Learning From Those Who
Have Left, Policy Focus #101, The Washington Institute for Near
East Policy, January 2010.
•	 Mark Knight, ‘Guns, Camps and Cash: Disarmament,
Demobilization and Reinsertion of Former Combatants in
Transitions from War to Peace’, Journal of Peace Research, 41(4)
(2004), pp. 499-516.
•	 Josh Lefkowitz, ‘Terrorists Behind Bars’, NEFA Report,
5 May 2008.
•	 Gabriele Marranci, ‘Faith, Ideology and Fear: The Case of Current
and Former Muslim Prisoners’, IQRA Annual Lecture – House of
Lords, 26 June 2007.
•	 Peter R. Neumann, ‘Joining Al Qaeda: Islamist Militant Recruitment
in Europe’, Adelphi Paper 399, International Institute for Strategic
Studies, January 2009.
•	 Peter R. Neumann, Old and New Terrorism (Cambridge: Polity
Press, 2009).
•	 ‘Out of the Shadows: Getting Ahead of Prisoner Radicalization’,
A Special Report by The George Washington University Homeland
Security Policy Institute and the University of Virginia Critical Incident
Analysis Group, September 2006.
•	 Raffaello Pantucci, ‘Britain’s Prison Dilemma: Issues and
Concerns in Islamic Radicalization’, Jamestown Terrorism Monitor,
24 March 2008.

63

Prisons and Terrorism Radicalisation and De-radicalisation in 15 Countries

•	 Marisa Porges and Jessica Stern, ‘Getting Deradicalization Right’,
Foreign Affairs, May/June 2010.
•	 Anne Speckhard, ‘Prison and Community Based Disengagement
and De-Radicalization Programs for Extremists Involved in Militant
Jihadi Terrorism, Ideologies and Activities’, Studies in Conflict and
Terrorism, forthcoming.
•	 Michael von Tangen Page, Prisons, Peace and Terrorism
(Houndmills: Macmillan, 1998).
•	 Humberto Trujillo et al, ‘Radicalization in Prisons? Field Research
in 25 Spanish Prisons’, Terrorism and Political Violence, 21(2009).
	

64

ABOUT ICSR

The International Centre for the Study of
Radicalisation and Political Violence (ICSR)
is a unique partnership of six great academic
institutions: King’s College London; the
University of Pennsylvania; the Jordan Institute
of Diplomacy; the Interdisciplinary Center
Herzliya; the Centre for Policy Research,
New Delhi; and the Pakistan Institute for
Peace Studies.
The aim and mission of ICSR is to bring
together knowledge and leadership to address
the challenge of radicalisation and political
violence. For more information, please visit
www.icsr.info

www.icsr.info