The Routledge Handbook on Radicalisation and Countering Radicalisation [1 ed.] 9780367476847, 9781032573809, 9781003035848

Citation preview

“This is a super book, worthy of its important subject matter. The line-up of contributors is stellar, and the range of perspectives is deeply impressive. It is destined to become the standard reference text on radicalisation.” Andrew Silke, Professor of Criminology, Royal Holloway, University of London, UK “Brilliant! This handbook should be on every student, researcher, practitioner and policymaker’s shelf. A much-needed comprehensive, diverse, multidisciplinary and, most importantly, critical approach to understanding and preventing violent radicalisation.” Ghayda Hassan, Professor of Clinical Psychology, Université du Québec à Montréal, Canada; Director of the Canadian Practitioners Network for the Prevention of Radicalization and Extremist Violence “With definitive contributions from so many of the top scholars in radicalisation studies, this comprehensive handbook is an invaluable resource for academics, practitioners, and students alike and guaranteed to remain so for the foreseeable future.” Noémie Bouhana, Professor of Crime Science and Counter Extremism, University College London, UK “A key strength of the The Routledge Handbook on Radicalisation and Countering Radicalisation is its multidisciplinary approach to analysing the concept of radicalisation and its various cognitive and behavioural expressions. Contributions raise critical issues relevant to the policy and practice of counter radicalisation and it will be of interest to both practitioners and scholars.” Adrian Cherney, Professor, School of Social Science, University of Queensland, Australia “This book is a significant contribution to our understanding of radicalisation. It offers a series of coherent conceptual analyses addressing some of the most challenging areas, and also develops practical insights and strategies to address radicalisation in communities and with individuals at both practical and policy levels.” Max Taylor, Visiting Professor, Department of Security and Crime Sciences, University College London, UK

THE ROUTLEDGE HANDBOOK ON RADICALISATION AND COUNTERING RADICALISATION

This handbook provides a theoretical and methodological exploration of the research on radicalisation and counter-radicalisation, one of the most influential concepts in Security Studies, International Relations, and Peace and Conflict Studies. Sitting at the heart of high-profile research and policy agendas on preventing and countering violent extremism (P/CVE), radicalisation as a concept has transformed the way researchers, policymakers, and societies think about how to counter terrorism and political violence. Deliberations about radicalisation and countering radicalisation have become further embedded as efforts to prevent and counter violent extremism have been ‘mainstreamed’ into other areas of public policy and practice, such as education, gender relations, health, peacebuilding, aid, and development. Theoretically and methodologically pluralistic, this handbook addresses radicalisation and countering radicalisation as they relate to a wide range of groups and milieus, articulating diverse ideological positions, drawing together insight and experience from multiple geographic and institutional settings, integrating global perspectives, and including scholarship focused on a range of policy fields. This book will be an essential reference point for anybody working on radicalisation, countering radicalisation, or terrorism and political violence more broadly. The insight that it provides will be relevant for both academics and members of relevant policy and practitioner communities. Joel Busher is Professor of Political Sociology at the Institute for Peace and Security, Coventry University. He is co-editor of Researching the Far Right: Theory, Method and Practice (Routledge, 2021) and author of The Making of Anti-Muslim Protest (Routledge, 2016), which was awarded the British Sociological Association’s Philip Abrams Memorial Prize. Leena Malkki is Director of the Centre for European Studies at the University of Helsinki. She is the chair of the ECPR Standing Group on Political Violence and a member of the editorial board of Terrorism and Political Violence. Sarah Marsden is Senior Lecturer in the School of International Relations at the University of St. Andrews. She is the author of Reintegrating Extremists: Deradicalisation and Disengagement.

THE ROUTLEDGE HANDBOOK ON RADICALISATION AND COUNTERING RADICALISATION

Edited by Joel Busher, Leena Malkki, and Sarah Marsden

Designed cover image: Getty Images First published 2024 by Routledge 4 Park Square, Milton Park, Abingdon, Oxon OX14 4RN and by Routledge 605 Third Avenue, New York, NY 10158 Routledge is an imprint of the Taylor & Francis Group, an informa business © 2024 selection and editorial matter, Joel Busher, Leena Malkki and Sarah Marsden; individual chapters, the contributors The right of Joel Busher, Leena Malkki and Sarah Marsden to be identified as the authors of the editorial material, and of the authors for their individual chapters, has been asserted in accordance with sections 77 and 78 of the Copyright, Designs and Patents Act 1988. All rights reserved. No part of this book may be reprinted or reproduced or utilised in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers. Trademark notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation without intent to infringe. British Library Cataloguing-in-Publication Data A catalogue record for this book is available from the British Library Library of Congress Cataloging-in-Publication Data Names: Busher, Joel, editor. | Malkki, Leena, editor. | Marsden, Sarah V., editor. Title: The Routledge handbook on radicalisation and countering radicalisation / Edited by Joel Busher, Leena Malkki and Sarah Marsden. Description: Abingdon, Oxon ; New York, NY : Routledge, 2024. | Includes bibliographical references and index. Identifiers: LCCN 2023023664 (print) | LCCN 2023023665 (ebook) | ISBN 9780367476847 (hbk) | ISBN 9781032573809 (pbk) | ISBN 9781003035848 (ebk) Subjects: LCSH: Radicalization. Classification: LCC HN49.R33 R69 2024 (print) | LCC HN49.R33 (ebook) | DDC 303.48/4—dc23/eng/20230802 LC record available at https://lccn.loc.gov/2023023664 LC ebook record available at https://lccn.loc.gov/2023023665 ISBN: 978-0-367-47684-7 (hbk) ISBN: 978-1-032-57380-9 (pbk) ISBN: 978-1-003-03584-8 (ebk) DOI: 10.4324/9781003035848 Typeset in Times New Roman by Apex CoVantage, LLC

CONTENTS

List of contributors

xi

 1 Introduction Sarah Marsden, Leena Malkki, and Joel Busher

1

PART I

The history of research on radicalisation

17

  2 Before ‘radicalisation’: explaining individual involvement in terrorism before the popularisation of the radicalisation concept Leena Malkki

19

  3 ‘Radicalisation’ and ‘countering radicalisation’: the emergence and expansion of a contentious concept Rik Coolsaet

34

  4 Analysing ‘radicalisation’ in historical cases Tim Wilson PART II

53

Analytical approaches to radicalisation

69

  5 The role of ideology in radicalisation Kumar Ramakrishna

71

vii

Contents

  6 Identity and extremism: sorting out the causal pathways to radicalisation and violent self-sacrifice Julia Ebner and Harvey Whitehouse   7 Social movement theory and research on radicalisation Stefan Malthaner   8 Criminological perspectives on extremist radicalisation and terrorist acts Gary LaFree and Yesenia Yanez   9 Insights from the study of new religious movements into the process of radicalisation Lorne L. Dawson

85 99 113

132

10 The pen and the sword: cognition, emotion, communication, and violent radicalisation150 Kurt Braddock 11 Gender perspectives on radicalisation Elizabeth Pearson

164

12 Radicalisation and psychopathology Emily Corner

180

13 Belonging is just a click away: extremism, radicalisation, and the role of online communities Amarnath Amarasingam

196

14 Radicalisation of ‘lone actors’ Francis O’Connor, Lasse Lindekilde, and Stefan Malthaner

213

15 Radicalisation of ‘foreign fighters’ Jeanine de Roy van Zuijdewijn

231

PART III

Countering radicalisation: key debates

245

16 Countering violence or ideas? The politics of counter-radicalisation Kodili Chukwuma and Lee Jarvis

247

viii

Contents

17 Responding to radicalisation with different ideological roots: how similar is the ‘problem’? How similar are the ‘solutions’? Daniel Koehler

262

18 Who should be involved with counter-radicalisation policy and practice? Francesco Ragazzi and Josh Walmsley

276

19 Working with communities to counter radicalisation Paul Thomas and Michele Grossman

292

20 The role of research and researchers in counter-radicalisation policy and practice Alice Martini and Laura Fernández de Mosteyrín

306

21 How can we meaningfully evaluate the effects and effectiveness of programmes to prevent or counter radicalisation? Tom Fisher and Joel Busher

320

22 Re-colonising the field of evaluation of prevention of violent radicalisation: a critical, cross-regional perspective Pablo Madriaza, David Morin, and Vivek Venkatesh

338

PART IV

Countering radicalisation: strategies and challenges

353

23 Rethinking CVE and public health prevention: towards health promotion Keiran Hardy

355

24 Multi-agency approaches to countering radicalisation Robin Andersson Malmros and Jennie Sivenbring

369

25 Naturalisation through mainstreaming: counter-terrorism and counter-radicalisation in UN and EU discourse Rita Augestad Knudsen

384

26 Countering radicalisation while expanding the criminal law Tufyal Choudhury

399

27 Responding to radicalisation through education Stijn Sieckelinck and William Stephens

415

ix

Contents

28 Counter narratives and strategic communications, offline and online Benjamin Lee

431

29 Deradicalisation and disengagement: lessons from the Indonesian experience446 Julie Chernov Hwang 30 Interventions with ‘at risk’ individuals Raquel da Silva

461

31 How can general violence risk assessment and management inform that of violent extremist risk? Paul Gill, Zoe Marchment, Amber Seaward, Philip Doherty, and Kirsty Goodman

476

32 Content moderation: social media and countering online radicalisation Bharath Ganesh

498

33 Radicalisation, counter-radicalisation, and governance in fragile states Oscar Gakuo Mwangi

514

34 Local contours and global discourses in countering violent radicalisation and extremism: a perspective from the global south Fathima Azmiya Badurdeen

528

Index541

x

CONTRIBUTORS

Amarnath Amarasingam is Assistant Professor in the School of Religion and is cross-appointed to the Department of Political Studies at Queen’s University in Ontario, Canada. He is also Senior Fellow with the International Centre for the Study of Radicalisation. He is the author of Pain, Pride, and Politics: Sri Lankan Tamil Activism in Canada (2015), the co-editor of Stress Tested: The COVID-19 Pandemic and Canadian National Security (2021), and has published numerous peer-reviewed articles and book chapters. Fathima Azmiya Badurdeen is Lecturer at the Department of Social Sciences, Technical University of Mombasa, Kenya. Since 2012, she has worked in the East African region as an academic in the fields of countering terrorism, violent extremism, forced migration, human trafficking, and gender. She has increasingly used ethnographic methods in her research work. Prior to her work in Kenya, she has worked as a researcher and practitioner in Asia, mainly in Sri Lanka and India, in the fields of forced migration, post-conflict development, conflict transformation, and peacebuilding. Kurt Braddock is Assistant Professor of Public Communication at American University. His research centres on the persuasive strategies used by extremist groups to draw vulnerable audiences to their cause. He is the author of Weaponized Words: The Strategic Role of Persuasion in Violent Radicalization and Counter-Radicalization, 2020. Dr. Braddock has also advised several national and international organisations, including the US Department of Homeland Security, the US Agency for International Development, and the UN Office of Counter-Terrorism. Joel Busher is Professor of Political Sociology at the Centre for Trust, Peace and Social Relations, Coventry University. His research addresses far-right and exclusivist activism; the escalation and inhibition of violence associated with contentious politics; and the evolution and enactment of efforts to prevent and counter ‘violent extremism’. He is the co-editor of Researching the Far Right: Theory, Method and Practice (Routledge) and The Prevent Duty in Education: Impact, Enactment and Implications. His book, The Making of Anti-Muslim Protest (Routledge), was awarded the British Sociological Association’s Philip Abrams Memorial Prize.

xi

Contributors

Tufyal Choudhury is Associate Professor at Durham Law School, Senior Research Fellow on Security and the Rule of Law at the Bingham Centre for the Rule of Law, and Senior Legal and Policy Advisor at Rights and Security International. His teaching and research focus on issues of counter-terrorism law and policy, racial and religious discrimination, and integration. Recent publications include Suspicion, Discrimination and Surveillance: The Impact of Counter-terrorism Law and Policy on Racialised Groups at Risk of Racism in Europe (2021). Kodili Chukwuma is Assistant Professor of International Security in the School of Government and International Affairs at Durham University. PhD was awarded to Kodili by the University of East Anglia in 2022, in which he explored Nigeria’s counter-terrorism strategy. Kodili’s research focuses on the politics of security, terrorism, and counter-terrorism and straddles the fields of International Relations, Critical Security Studies, Critical Terrorism Studies, African Politics, and Postcolonial studies. His current projects include work on terrorism trials in Nigeria, on postcolonial space(s) and critical terrorism studies, on the proscription of terrorist organisations in Nigeria, and on counter-terrorism strategies in the Lake Chad Basin. His work has been published or is forthcoming in leading journals such as Security Dialogue, International Political Sociology, Critical Studies on Terrorism, and African Security. Rik Coolsaet is Professor Emeritus of International Relations at Ghent University (Belgium) and Senior Associate Fellow at Egmont, Royal Institute for International Relations (Brussels). His fields of interest include the dynamics of radicalisation and the history of terrorism, international relations, and the interaction between domestic and foreign policy. In the 1980s and 1990s, he served as deputy chief of the Cabinet of the Belgian Minister of Defence and of the Minister of Foreign Affairs. Emily Corner is Senior Lecturer in Criminology at the Centre for Social Research and Methods at the Australian National University. Emily’s research focuses on lone and group-based terrorism, radicalisation, mass murderers, fixated individuals, and grievance-fuelled violence. Dr. Corner’s doctoral research focused on examining mental disorders and terrorist behaviour and won the Terrorism Research Initiatives Thesis Award in 2016. Dr. Corner has significant experience as a researcher on multiple international projects and Centres of Excellence funded by the Defence Science and Technology Laboratory, the European Union, the National Institute of Justice, the U.S. Department of Defence, the Australian Department of Home Affairs, and the Australian Institute of Criminology. Lorne L. Dawson is Professor Emeritus in the Departments of Religious Studies, and Sociology and Legal Studies at the University of Waterloo (Canada). He has published three books, five edited books, and 88 academic articles and book chapters. He is Co-Founder and Co-Director of the Canadian Network for Research on Terrorism, Security and Society (2012–23; www.tsas.ca), and his recent research has focused on such topics as foreign fighters, the role of religion in motivating religious terrorism, and the social ecology of radicalisation. Raquel da Silva is Assistant Professor of International Relations at the School of Economics, University of Coimbra, and Integrated Researcher at CEI-Iscte. Her research explores processes of engagement in and disengagement from violent organisations through the narrative study of the life stories of former militants. She is also interested in the intersection between violent and non-violent ideologically motivated activism and the impact of Preventing/Countering Violent xii

Contributors

Extremism (P/CVE) interventions in the lives of individuals, groups, and communities. Raquel values creative public engagement and the translation of research into practice and policy. She has published a monograph entitled Narratives of Political Violence: Life Stories of Former Militants (Routledge, 2019), as well as different articles in national and international journals. Her work has been funded by the Foundation for Science and Technology, the British Academy, and the European Union, among others. She tweets at @RaquelBPSilva. Jeanine de Roy van Zuijdewijn is Assistant Professor of terrorism studies at Leiden University. Her research interests include dealing with the impact of terrorism and foreign fighters. She aims to make academic knowledge relevant for wider audiences, ranging from the general public to policymakers and politicians. Philip Doherty is currently a PhD student at University College London (UCL) in the Department of Security and Crime Science. He holds a BA in Criminology and Psychology, an MSc in Countering Organised Crime and Terrorism, and MRes in Security and Crime Science. Philip’s area of research lies in terrorist risk assessment criteria and process validity, hostile reconnaissance, and terrorist decision-making. Some of his work has looked at simulation-based deterrence methods and exercise processes, sentiment analyses of prior terrorist writings and language styles, and linguistic trends between writing types. Philip has a background in Strategic Intelligence, having previously worked as an intelligence analyst identifying patterns and trends of malicious and nonmalicious cyber activity globally, in particular in the APAC region. Julia Ebner is Researcher specialising in radicalisation, cumulative extremism, and terrorism prevention. She is a DPhil candidate in Anthropology at the Centre for the Study of Social Cohesion, University of Oxford, and holds a senior research fellowship at the Institute for Strategic Dialogue. She is the author of The Rage (2017) and Going Dark (2020). Laura Fernández de Mosteyrín is Lecturer at Universidad Nacional de Educación a Distancia, UNED. She is a fellow researcher at the Group for the Study of Society and Politics (UCMUNED). Her most recent publication is Meaning and context in analysing extremism: the banalisation of the far-right in Spanish public controversies, Critical Studies on Terrorism, 15:1, 38–60. Her upcoming book is Los significados de la seguridad: una aproximación a los discursos y las prácticas de la ciudadanía en España. Tom Fisher is Evaluation Researcher, Social Policy Analyst, and Lecturer based at the Centre for Trust, Peace and Social Relations at Coventry University. Over the past 12 years, Tom has evaluated social interventions across many policy fields, from intercultural dialogue and migrant integration programmes to international peacebuilding and food poverty. More recently, Tom has specialised in the evaluation of preventing and countering violent extremism, leading a range of high-profile programme evaluations in the United Kingdom, Europe, and East Africa. Bharath Ganesh is Assistant Professor in the department of Media Studies and the Institute for Logic, Language and Computation at the University of Amsterdam. Bharath’s research focuses on racism and social media, platform governance, and transnational far right digital cultures. Paul Gill is Professor of Security and Crime Science at University College London. He has over 1,000 publications on the topic of terrorist behaviour. He has conducted research funded xiii

Contributors

by the European Research Council, the Office for Naval Research, the Department of Homeland Security, DSTL, the European Union, the National Institute of Justice, CREST, Public Safety Canada, and MINERVA, among others. Collectively, these grants have been worth over €10 million. These projects focused on various aspects of terrorist behaviour, including IED development, creativity, terrorist network structures, lone-actor terrorism, and counter-terrorism evaluation. Kirsty Goodman holds an MRes from University College London’s Department of Security and Crime Science and acted as a research assistant on various funded projects during that time. Michele Grossman is Professor and Research Chair in Diversity and Community Resilience at the Alfred Deakin Institute for Citizenship and Globalisation, where she is also Director of the Centre for Resilient and Inclusive Societies (CRIS) and Convenor of the AVERT (Addressing Violent Extremism and Radicalisation to Terrorism) Research Network. Her research focuses on engaging communities in countering and addressing modes of violent extremism across the ideological spectrum, on which she has published widely. Her work has been funded by a range of national and international research schemes in Australia, Canada, the United States, and Europe. She is a past Robert Schuman (Distinguished Scholar) Fellow at the European University Institute in Florence (2018–22) and is Visiting Professor at the School of Education and Professional Development at the University of Huddersfield. Keiran Hardy is Senior Lecturer in the Griffith Criminology Institute and School of Criminology and Criminal Justice, Griffith University, Australia. He has published widely on counter-terrorism law and policy, countering violent extremism, national security whistleblowing, and the accountability of intelligence agencies. He contributes regularly to parliamentary inquiries and the Australian media. He is the author of Law in Australian Society: An Introduction to Principles and Process (Routledge, 2019). Julie Chernov Hwang is Associate Professor of political science and international relations at Goucher College. She is the author of Why Terrorists Quit: The Disengagement of Indonesian Jihadists (2018), Peaceful Islamist Mobilization in the Muslim World: What Went Right (2009), and the co-editor of Islamist Parties and Political Normalization in the Muslim World (2014). Her articles have been published in Political Psychology, Studies in Conflict and Terrorism, Terrorism and Political Violence, Asian Survey, Asian Security, Contemporary Southeast Asia, Asia-Pacific Issues, Southeast Asia Research, Nationalism and Ethnic Politics, The Washington Post, CTC Sentinel, and Lawfare. Her new book manuscript, Becoming Jihadis: Radicalization and Commitment in Southeast Asia, forthcoming in 2023, explores the social and emotional reasons why Indonesian and Filipino Muslims join, commit to, and take on high-risk roles in Islamist extremist groups. Lee Jarvis is Professor of International Politics at the University of East Anglia in the United Kingdom. He is the author or editor of 14 books and over 70 articles/chapters on the politics of security, including Times of Terror: Discourse, Temporality and the War on Terror; Anti-Terrorism, Citizenship and Security (with Michael Lister); and Banning Them, Securing Us? Terrorism, Parliament and the Ritual of Proscription (with Tim Legrand). Lee’s work has been funded by bodies including the ESRC, the AHRC, the Australian Research Council, and NATO.

xiv

Contributors

Daniel Koehler is Founding Editor and Director of the Journal for Deradicalization and the German Institute on Radicalization and De-Radicalization Studies (GIRDS). He is a member of the Editorial Board of the International Centre for Counter-Terrorism in The Hague and Research Fellow at the Polarization and Extremism Research and Innovation Lab (PERIL) of the American University in Washington, DC, as well as the Royal United Service Institute (RUSI). His work can be followed on Twitter at @GIRD_S. Rita Augestad Knudsen is Senior Researcher at the Norwegian Institute of International Affairs (NUPI), Group for Security and Defence, and Managing Director of the Consortium for Research on Terrorism and International Crime. Her work covers counter-terrorism and counter-radicalisation, especially in the United Kingdom, Scandinavia, and the rest of Europe, and in particular issues around risk assessment and mental health. An intellectual historian/international historian by training, she has also published on other issues of international security, especially ones at the intersection between ideas, politics, security, and law. Gary LaFree is Professor in the Department of Criminology and Criminal Justice and the Founding Director of the National Consortium for the Study of Terrorism and Responses to Terrorism (START) at the University of Maryland. His research focuses on the causes and consequences of violent crime and terrorism. His most recent books are Putting Terrorism in Context (with Laura Dugan and Erin Miller), The Handbook of the Criminology of Terrorism (with Josh Freilich), and Countering Terrorism (with Martha Crenshaw). Benjamin Lee is Senior Research Associate at the Centre for the Study of Terrorism and Political Violence at St. Andrews University, where his work is financed by the Centre for Research and Evidence on Security Threats (CREST). Benjamin’s main research focuses on the extreme right and right-wing terrorism. Lasse Lindekilde is Professor at the Department of Political Science, Aarhus University.  He holds a PhD from the European University Institute. His research is focused on political mobilisation, violent radicalisation, and the implementation of countering violent extremism policies. His most recent research has focused on online political hostility and bystander reactions. In the fall of 2020, he received an ERC Consolidator Grant for the project ‘Standing by: Pro-social Bystander Reactions to Online Political Hostility’. Pablo Madriaza is Professor of Social Work at the Université du Québec en Outaouais in Canada. He holds a degree in psychology and a master’s degree in anthropology, both obtained in Chile, as well as a master’s degree and a doctorate in sociology, obtained in France. He has participated in numerous research projects and publications on violence and social conflict, including school violence, delinquency, and social movements. The last period of his career has been devoted to the study of the prevention of violent extremism and hate-motivated acts, particularly from the perspective of intervention practices and programme evaluation. He is particularly interested in analysing such socio-political violence from a critical, decolonial, and postcolonial approach. Leena Malkki is Director of the Centre for European Studies at the University of Helsinki. Her research focuses on political violence in post-war Europe, and her current research interests

xv

Contributors

include the transnational dynamics of political violence, societal resilience to violent extremism, school shootings, and lone-actor violence. She is the chair of the ECPR Standing Group on Political Violence and a member of the editorial board of Terrorism and Political Violence. Robin Andersson Malmros is Deputy Director of The Segerstedt Institute, a Swedish national centre for studies on violent extremism and P/CVE, at the University of Gothenburg. He holds a PhD degree in Public Administration, and his work primarily focuses on organisational aspects of local P/CVE work. He has published articles and chapters on topics such as local translations of radicalisation and P/CVE, multi-agency work, and coordination and management of P/CVE. Stefan Malthaner is Senior Researcher at the Hamburg Institute for Social Research and Visiting Professor at Leuphana University, Lüneburg. His research focuses on militant movements, political violence and collective urban violence, processes of radicalisation, and civil wars, with a particular focus on the relationship between armed groups and their social environment. Zoe Marchment is Post-doctoral Research Fellow at University College London. She completed her PhD in UCL’s Department of Security and Crime Science in 2019. Her PhD research examined the spatial decision-making of terrorist target selection, with a focus on lone actors and violent dissident Republican activity. Her post-doctoral research has involved evaluating counter-terrorism risk assessment and management, developing risk assessment instruments, and leading the deterrence strand of work for CREST. She also holds a BSc in Psychology and an MSc in Countering Organised Crime and Terrorism. Zoe has been a Co-Principal Investigator for projects funded by the Centre for Research and Evidence on Security Threats (CREST), the UK Home Office and CPNI, including the CREST funded project ‘Further Development of Risk Assessment Schemes for Channel’. She has also worked on projects for the European Research Council (Grievance), the Defence Science and Technology Laboratory (Dstl); FP7 Preventing, Interdicting and Mitigating Extremism (PRIME); and the VOX-Pol Network of Excellence. Sarah Marsden is Senior Lecturer in the School of International Relations at the University of St. Andrews. Sarah’s research looks at the process and impact of different kinds of political contention, from transnational social movements to violent militant networks. Her current research includes a body of work on political violence through the Centre for Research and Evidence on Security Threats (CREST), with a focus on counter-extremism. She is the author of Reintegrating Extremists: Deradicalisation and Disengagement. Alice Martini is Lecturer in International Relations at the Universidad Complutense de Madrid, Spain. Her work focuses on global discourses and practices of terrorism, and counter-terrorism and international security. She has been a board member of the EISA Early Career Development Group and the BISA Critical Studies on Terrorism Working Group. Among other works. She is the author of UN and Counter-terrorism. Global hegemonies, Power, and Identities (Routledge, 2021) and a co-editor of Encountering Extremism. Theoretical Issues and Global Challenges (2020). David Morin is Full Professor in the School of Applied Politics in the Faculty of Arts and Humanities at the Université de Sherbrooke in Canada. His areas of expertise and professional experience include national and international security issues, particularly violent extremism, radicalisation, disinformation, and conspiracy. He is a co-holder of the UNESCO Chair in the Prevention of xvi

Contributors

Violent Radicalization and Extremism (UNESCO-PREV Chair). He served on the Expert Advisory Panel on Online Safety to provide advice to the Government of Canada on a legislative and regulatory framework to best address harmful content online. He also serves as a designated expert on the steering committee of the Francophone Network for the Prevention of Violent Radicalization and Extremism that could lead to Terrorism (FrancoPREV) supported by the International Organization of La Francophonie. Oscar Gakuo Mwangi is Associate Professor of Political Science, Department of Political and Administrative Studies, National University of Lesotho. He teaches various courses in Comparative Politics, International Relations, and Political Theory. His research interests are governance, conflict, security, and development in eastern and southern Africa, with a focus on state fragility, democratisation, radicalisation, counterterrorism, and environmental politics. He has authored several refereed publications. He is also Adjunct Associate Professor, Department of Social Sciences at Pwani University, Kilifi, Kenya. Francis O’Connor is Marie Curie Skłodowska Post-doctoral Fellow in Rural Sociology at Wageningen University in the Netherlands. His research addresses the grey area between violent and non-violent radicalisation and mobilisation. He particularly focuses on the relationship between insurgent movements and their supporters; his current project, Routinised Insurgent Space, looks at the spatial dynamics of insurgent support in the cases of the PKK in Turkey and the M-19 in Colombia. Elizabeth Pearson is Lecturer in Criminology with the Conflict, Violence and Terrorism Research Centre at Royal Holloway, University of London, and Associate Fellow with RUSI, the Royal United Services Institute, and with the International Centre for Counter-Terrorism, ICCT, in The Hague. Her research interests are in gender, extremism, and counter-extremism. Francesco Ragazzi is Associate Professor of International Relations at Leiden University (the Netherlands) and Associate Scholar at the Centre d’Etude sur les Conflits, Liberté et Sécurité (France). He is also Co-director of ReCNTR (Leiden University’s Center on Multimodal and Audio-visual Methods). His research interests include counter-radicalisation, counter-terrorism, and digital surveillance. His current research project explores the security uses of computer vision in areas such as biometric surveillance, social media content moderation, and border control. Kumar Ramakrishna is Professor of National Security Studies in the S. Rajaratnam School of International Studies at Nanyang Technological University, Singapore. Described by the leading journal Perspectives on Terrorism as ‘one of Southeast Asia’s leading counterterrorism experts’, Ramakrishna has published numerous articles and books on the subject. His recent research has focused on understanding, preventing, and countering violent extremism in Southeast Asia. His latest book is Extremist Islam: Recognition and Response in Southeast Asia (2022). Amber Seaward is a PhD candidate at University College London (UCL), where her areas of interest are radicalisation, fixated threats, and threat assessment. Currently, her research involves evaluating the human and operational sides of threat assessment processes and identifying risk factors that distinguish violent from non-violent involvement in extremism. She holds a BA in Philosophy, Politics, and Economics from Oxford University and an MSc in Countering Organised Crime and Terrorism from UCL. xvii

Contributors

Stijn Sieckelinck is concerned with issues regarding social education. As a ‘thinker in a world of doing’, he tries to contribute to better insights and approaches in science, practice, and policy, wherever our relationship with young people is at stake. Since March 2021, he has been a professor at the University of Applied Sciences of Amsterdam, the Netherlands, where he coordinates research on youth work. He holds a PhD in Social Educational Theory. Jennie Sivenbring is Research Fellow at The Segerstedt Institute at the University of Gothenburg. She holds a PhD degree in Child and Youth Studies, and her research primarily focuses on policy on prevention against violent extremism on a Nordic level and the management of children and youth within the area of violent extremism and terrorism. William Stephens is a postdoctoral researcher at Utrecht University. His research addresses social and pedagogical responses to polarisation and radicalisation, and resilient trajectories of development in youth. Alongside his research, William spends much of his time working alongside groups of youth engaged in efforts for constructive social change in neighbourhoods in Europe. The insights he gains from hundreds of young people in these different communities continuously inspire and give purpose to his research. Paul Thomas is Professor of Youth and Policy and Associate Dean (Research) in the School of Education and Professional Development at the University of Huddersfield, the United Kingdom. Paul’s research focuses on how state policies such as Community Cohesion and the Prevent counter-terrorism strategy have been understood and enacted by practitioners, particularly educationalists, and experienced by communities. It has led to books such as Race, Space and Place in Northern England: The (M62) Corridor of Uncertainty (with Miah and Sanderson, 2020), as well as articles in many leading journals. Paul’s research collaborations around ‘community reporting thresholds’ with colleagues in the United Kingdom, the United States, and Canada have focused on barriers to community members reporting concerns about an ‘intimate’ becoming involved in violent extremism, and the UK study has directly informed the ‘Act Early’ campaign there. Paul is a qualified Youth and Community Worker. Vivek Venkatesh is Filmmaker, Musician, Curator, and Applied Learning Scientist focused on building community resilience and tolerance against hate through a resolutely public pedagogical approach. He holds the UNESCO Chair in Prevention of Radicalisation and Violent Extremism (UNESCO-PREV) and is Codirector of the Centre for the Study of Learning and Performance (CSLP) at Concordia University, where he is Full Professor of Inclusive Practices in Visual Arts and Chair of the Department of Art Education. Josh Walmsley is a PhD researcher in International Relations at King’s College London’s Department of War Studies. His current research explores the transnational politics of counter-radicalisation, inquiring, in particular, into struggles over expert authority and the role of technology. In recent years, Josh has contributed to reports on counter-terrorism for the European Parliament and the Council of Europe and has participated in expert panels for a number of organisations, including the Academy of European Law and the International Commission of Jurists. Harvey Whitehouse is Professor of Social Anthropology and Director of the Centre for the Study of Social Cohesion at the University of Oxford. His research involves surveys, experiments,

xviii

Contributors

and immersive fieldwork in many societies around the world. He is also a founding director of Seshat: Global History Databank. His books include Inside the Cult (1995), Arguments and Icons (2000), Modes of Religiosity (2004), and The Ritual Animal (2021). Tim Wilson assumed the Directorship of Handa CSTPV in 2016. As the Director of Europe’s oldest research centre devoted solely to the academic study of terrorism, he is frequently invited to comment on terrorism in the media. His last book, Killing Strangers: How Political Violence Became Modern, appeared in September 2020. Yesenia Yanez is a PhD student and McNair Scholar in the Department of Criminology and Criminal Justice at the University of Maryland. Her research focuses on the predictors of violent crime and extremism. Yesenia’s current work analyses the impact of situational factors on mass shooting fatalities.

xix

1 INTRODUCTION Sarah Marsden, Leena Malkki, and Joel Busher

Radicalisation has been one of the most influential concepts in security studies over the last two decades and has also had a significant impact in cognate areas, such as peace and conflict studies and research on social movements and contentious politics. Sitting at the heart of globally prominent research and policy agendas on preventing and countering violent extremism (P/CVE), it has transformed the way researchers, policymakers, and societies more broadly think about terrorism and political violence and how to counter it (Richards, 2015). The concepts of radicalisation and countering radicalisation have become more embedded as efforts to prevent and counter violent extremism have been integrated or ‘mainstreamed’ into other areas of public policy and practice, such as education, gender relations, health, peacebuilding, aid, and development (Ambrozik, 2019; Knudsen, this volume; Shanaah & Heath-Kelly, 2022). Not surprisingly, policy and public interest in radicalisation and countering radicalisation have translated into an explosion of research in this area. Studies on radicalisation have been one of the main beneficiaries of the growth of investment and interest in terrorism studies that took place in the first two decades of the 21st century. As more researchers have been drawn into the field, radicalisation research has become increasingly multidisciplinary, with contributions from scholars rooted in a wide range of disciplinary and theoretical traditions (Morrison, 2022). As attested to by many of the contributions to this volume, this has driven significant innovation in the field and led to the generation of important empirical and conceptual insights into how, why, when, and under what conditions people engage with radical worldviews and move towards, and sometimes away from, political violence. This explosion of research has not been without its challenges, however. There has been extensive criticism of the lack of conceptual clarity and consistency in the field and its implications for research, policy, and practice (Peels, 2023; Sedgwick, 2010). While the growing multidisciplinarity of the research has undoubtedly helped to advance theoretical and practical understanding of radicalisation and countering radicalisation, it has been argued that the field has often not maximised effectively the insights from prior research of theoretical and empirical relevance (Youngman, 2020). The pressure to respond to emerging and often urgent policy and practical requirements has created a context that does not always prioritise methodological, empirical, and theoretical rigour (Schuurman, 2020b). More fundamentally, questions have been raised about whether the discourse of radicalisation and concomitant P/CVE agendas are contributing, whether by accident or by 1 DOI: 10.4324/9781003035848-1

Sarah Marsden, Leena Malkki, and Joel Busher

design, to processes of societal securitisation, polarisation, and the erosion of fundamental human rights (Baker-Beall et al., 2015; Kundnani & Hayes, 2018). Recognising the advances that have been made over the last two decades, and the challenges and critiques that this research has attracted, the purpose of this Handbook is to map the state of the art with respect to research on radicalisation and countering radicalisation, and to consider some of the most important priorities, opportunities, and challenges in the next phase of these research agendas. The volume is organised into four parts. Part I focuses on the history of radicalisation, with chapters that trace how the field evolved and the extent to which it has intersected with wider scholarship on the causes and dynamics of terrorism and/or violent extremism. Part II explores prominent analytical approaches to radicalisation, how they have developed over the last two decades, and their various strengths and limitations. Part III turns to countering radicalisation, and some of the key debates that have dominated this field. Part IV continues the focus on countering radicalisation, examining some of the strategies that have been developed in the field, how they have been put into practice, and the way these have played out on the ground. Throughout the volume, the chapters aim to provide an overview of existing research on radicalisation and countering radicalisation. We intentionally sought to bring together and bring into dialogue contributions from scholars from a wide range of disciplinary, theoretical, and methodological traditions, including from more critical and policy-facing perspectives, working on diverse groups and milieus across a range of geographic, social, and institutional settings. This was achieved through both contributor selection and the review process, whereby contributors also acted as reviewers for other chapters within the volume. One of our hopes for this volume is that it will provide an important stepping stone towards a more concerted and constructive conversation between these different strands of research on radicalisation and countering radicalisation than has sometimes been the case. Given the scale and scope of this volume, we do not attempt to summarise all of the chapters in this introduction. What we do instead is draw on the contributions to this volume and the wider literature on radicalisation and countering radicalisation to briefly reflect on three central aspects of the research: conceptual issues in radicalisation research, the evolving evidence base, and research practice.

Conceptual issues in radicalisation research As has been well documented (Coolsaet, 2016, and this volume; Malkki, this volume), the current concept of radicalisation rose to prominence in the wake of the mass casualty attacks in New York in 2001, Madrid in 2004, and London in 2005, as policy planners and practitioners sought ways to understand and address such threats. ‘Radicalisation’ was initially used to conceptualise the process whereby otherwise ‘ordinary’ individuals or citizens come to engage in or actively support terrorist violence (Neumann, 2013). It soon gained popularity, particularly among policy and practitioner communities, for whom it provided a way of talking about the problem of ideologically motivated political violence that felt fairly intuitive and effectively located it as a form of deviance and a social problem that could be managed through appropriate interventions. Radicalisation also became a focus for researchers seeking to understand this process and how it might be inhibited. As it gained popularity, however, it became something of a catch-all term for, to use Peter Neumann’s (2008, p. 4) much-cited formulation, ‘what goes on before the bomb goes off’. The ensuing conceptual fuzziness and concomitant debates have been a feature of much of the academic literature ever since. It is therefore useful to outline here some of the main focal points of these debates. 2

Introduction

Radicalisation at different scales Radicalisation can be studied at different scales. At the individual level, studies of radicalisation tend to explore how and why individuals adopt increasingly radical ideological views and/or engage in extremist violence. Radicalisation can also be studied as a collective process. Studies of group radicalisation examine how and why a certain group comes to adopt an extremist ideology, justify the use of violence, commit violent acts, and recruit new members (e.g. della Porta, 2018) while studies of mass radicalisation (McCauley & Moskalenko, 2017) examine the processes through which mass publics radicalise in the context of inter-group conflict. Of the three, while group and mass radicalisation have received some attention – particularly in the wider literature on social movements, civil war, and insurgency (della Porta & LaFree, 2012; Malthaner, 2017) – in terrorism studies, the individual level has undoubtedly received the greatest attention to date. This coincides with the individual-level focus of most P/CVE programming, a focus that has been reinforced in recent years by interest in and growing concern about so-called lone actors or self-radicalisers, particularly in the Global North (Schuurman et al., 2018). It also reflects a more general tendency towards methodological individualism within research on crime and deviance, and social problems more broadly (Ewert et al., 2021). Mass radicalisation in particular has rarely been integrated under the banner of radicalisation studies, even though it is actively studied, for example, in social psychology (McCauley & Moskalenko, 2017) and political sociology (Malthaner, 2017). An important question going forward is how these different scales of analysis can be more effectively integrated. There is growing recognition of the benefits of multi-level explanations and responses to radicalisation (Alimi et al., 2015; Bouhana, 2019; Dawson, 2017). Interventions that take account of these multiple levels – of the role of families, peers, community members, and wider socio-cultural and political dynamics in radicalisation processes – are also now emerging (Ellis et al., 2022). Such approaches offer a more holistic approach to interpreting and responding to radicalisation, and provide alternative sites for intervention (Lewis & Marsden, 2021a), such as programmes that build resilience at the community level (Grossman, Hadfield et al., 2020), or provide routes for otherwise hard-to-reach groups to access support on the basis of partnership working between statutory agencies and communities (Puigvert et al., 2020). It is important that the research base keeps pace with these innovations.

Cognitive versus behavioural radicalisation One of the most prominent conceptual debates has concerned the extent to which radicalisation should be conceived of as a cognitive process, a behavioural process, or a combination of the two, and if it is the latter, how cognitive and behavioural aspects relate to one another (Ramakrishna, this volume). Put another way, to what extent is radicalisation about changing ideas, beliefs, opinions, or worldviews, and to what extent is it about the adoption of increasingly radical actions, either in terms of participation in or support for violent acts or in terms of participation in groups or milieus that use or support the use of political violence? One of the main criticisms of radicalisation research, particularly in its earlier iterations, was that it overplayed and overgeneralised about the role of ideas and ideology in moving people towards engagement (Crone, 2016; Sageman, 2004). This criticism gained traction amid mounting evidence that ideological radicalisation is not always a prerequisite for participation in or support for political violence (Horgan & Taylor, 2011). As a result, more recent scholarship has sought to conceptualise the relationship between ideas and behaviours in ways that better accommodate the 3

Sarah Marsden, Leena Malkki, and Joel Busher

complexity of this relationship (e.g. Holbrook & Horgan, 2019; Khalil et al., 2022). In doing so, scholars have also drawn insight from other fields. For example, Lorne Dawson (2009 and this volume) has demonstrated the value of existing empirical and theoretical research on pathways into new religious movements for understanding radicalisation into violent extremism, while Daniela Pisoiu (2015) has brought subcultural approaches into dialogue with work on radicalisation. Alongside this, scholarship is increasingly moving beyond the cognitive versus behavioural binary. For example, contemporary research draws attention to the role of emotion and affect in interpreting radicalisation (Crone, 2016; della Porta, 2020; McDonald, 2018; Malthaner, 2017, and this volume; Pilkington, 2016; van Stekelenburg, 2017) and the importance of embodied practices and experiences in shaping counter-radicalisation processes (Duriesmith & Huda Ismail, 2022; Mykkanen, 2022). These debates have significant implications for understanding what it means to counter radicalisation, the parameters of ‘legitimate’ state intervention, and what ‘success’ for such programmes looks like (Berger, 2016; Chukwuma & Jarvis, this volume; Elshimi, 2017; Silke, 2011). In the case of efforts to help people leave terrorism or violent extremism behind through reintegration programming, such debates have provoked extensive discussion about the appropriate emphasis on deradicalisation, commonly understood as efforts to support attitudinal change, and disengagement, typically reflecting behavioural change (Marsden, 2017). Given the importance of these issues, it is likely that they will remain live areas of discussion for the foreseeable future.

Radicalisation endpoints and the relationship with ‘terrorism’ and ‘violent extremism’ As discussed earlier, contemporary interest in radicalisation initially arose as part of efforts to explain and understand the pathways into violent extremism or terrorism. As such, ‘radicalisation’ has often been at least implicitly defined by its outcomes. The challenges with this have been wellrehearsed in the literature. Most notably, it has been pointed out that in many cases, the adoption of more radical ideas or behaviours does not result in something that would, by most standard definitions, be understood as terrorism or violent extremism (Expert Group, 2008; Githens-Mazer, 2012). For example, in the context of civil resistance movements, some activists might adopt more radical ideological or tactical positions, often as a result of encountering escalated repression by state security actors, but not engage in inter-personal violence due to adherence to principles of non-violence (Schock, 2015). In recent years, there have been some attempts within the radicalisation literature to disrupt and critically interrogate the association between ‘radicalisation’ and ‘violent extremism’ or ‘terrorism’. For example, Graham Macklin and Joel Busher (2015) draw attention to the often ‘truncated’ nature of processes of ‘reciprocal radicalisation’. Research on individual radicalisation pathways also highlights that the most common outcome of such processes is not, in fact, participation in violence (Schuurman, 2020a). Nonetheless, both in policy language and in popular usage, the terms ‘radicalisation’, ‘terrorism’, and ‘violent extremism’ remain tightly intertwined, at least in part due to the dominance of the radicalisation (into violence) concept in much of the policy and practice that falls under the P/CVE banner. The complex and contested relationships between radicalisation, violent extremism, and terrorism have implications for the relationship between countering radicalisation and counterterrorism. One of the main differences between counter-radicalisation and ‘traditional’ counterterrorism is the focus on ‘upstream’ intervention in the processes believed to lead to violence rather than interdiction or intelligence gathering when individuals are already substantively involved in the 4

Introduction

preparation of terrorism. Counter-radicalisation involves a range of activities that go beyond the ‘harder’ interventions deployed by the police and security personnel, using everything from mentoring and social work to education, sports, and the arts in efforts to prevent or counter radicalisation (Harris-Hogan et al., 2016; Koehler, 2017). Similarly, while both counterterrorism and counter-radicalisation involve actors from multiple agencies and sectors, counter-radicalisation is, at least in theory, less reliant on the methods and resources of the police and the security and intelligence services and typically involves a wider group of stakeholders, from community members, statutory actors that do not have a traditional security role, such as teachers or healthcare professionals, to families, peers, and even former extremists (Irwin, 2015; Lewis & Marsden, 2021a). In practice, however, the relationship between these areas has been the subject of extensive debate. While in some national, institutional, and policy contexts, counter-radicalisation has been seen as an independent, though related, area of policy and practice, in others, it has been seen as one pillar of counterterrorism (Martins & Ziegler, 2018). One area where these debates are particularly important is when we look at the evolving legal landscape. As Tufyal Choudhury (in this volume) argues, the rapid development of counterterrorism legislation in some jurisdictions means that the distinction between the ‘pre-crime’ space, within which counter-radicalisation interventions operate, and the ‘criminal’ space, which was once the exclusive domain of the criminal justice system, has become blurred. Monitoring and understanding the implications of this should be a priority over the coming years.

Contested models and metaphors A further conceptual debate has revolved around the models and metaphors used to talk about radicalisation. One function of models and metaphors is to facilitate communication, often with wider or new audiences. As such, they tend towards simplification. This was certainly the case with early models and metaphors used to talk about radicalisation, such as ‘conveyer belts’ (U.S. Department of State, 2006), ‘staircases’ (Moghaddam, 2005), or ‘pyramids’ (Leuprecht et al., 2010; McCauley & Moskalenko, 2017). Such models and metaphors have, however, faced criticism, and more recent academic literature has instead tended to emphasise the open-ended, contingent, and nonlinear nature of radicalisation (Feddes et al., 2020), and the importance of understanding the complex interplay between individual- and group-level processes and the wider context (Clemmow et al., 2023). More recent scholarship is turning to models and metaphors that better accommodate this complexity and dynamism. Noémie Bouhana (2019), for example, has drawn on ecological models in her S5 framework; Ken Reidy (2018) has used the metaphor of radicalisation as a ‘vector’, while John Horgan (2020) has recently explored how tornados might provide a useful metaphor for talking about radicalisation. In a similar vein, the 3N framework put forward by Arie Kruglanski and colleagues (e.g. Webber & Kruglanski, 2017) conceptualises radicalisation as an interplay of three factors – needs, narratives, and networks – which vary in order and importance depending on the individual in question. Turning to countering radicalisation, one of the most influential models in P/CVE in recent years has been the so-called public health model. This describes the different tiers of interventions in counter-radicalisation work in line with the logic of preventing negative public health outcomes (Harris-Hogan et al., 2016). Primary prevention takes a broad approach focused on the community level, aiming to address the root causes of terrorism and raise awareness about violent extremism, analogous to public health campaigns highlighting the dangers of smoking. Secondary prevention targets those considered at risk of radicalisation with more 5

Sarah Marsden, Leena Malkki, and Joel Busher

focused efforts to try and disrupt the move towards violent extremism, similar to targeted public health interventions in high-risk areas or with high-risk populations. Tertiary interventions are targeted at those already involved in violent extremism, or convicted of a terrorism offence, seeking to support them to disengage and/or deradicalise, and could be compared to smoking cessation interventions. This public health metaphor has gained significant traction in part because it provides a heuristic that helps to capture the vast array of existing intervention approaches and aims. It is also an approach that is legible to policymakers and fits the logic of contemporary social policy and its emphasis on nurturing resilience to different social harms (Stephens et al., 2021). However, the public health approach has faced criticism. Some argue that it risks conceptualising violent extremism as a disease rather than a complex socio-political phenomenon (Aggarwal, 2018) and that the broad reach of, in particular, primary prevention efforts significantly expands the scope of those subject to security policies (Richards, 2015). Others have argued that the broad reach and focus of public health-oriented approaches make it difficult to understand what is being countered, and hence assess whether interventions are ethical, appropriate, or effective (Fisher & Busher, this volume; Harris-Hogan et al., 2016; Richards, 2015). Engaging with these debates, Keiran Hardy’s chapter in this volume argues that it might be preferable to adopt a ‘health promotion’ approach, concerned with nurturing healthy societies able to reduce the likelihood of radicalisation, rather than a ‘disease prevention’ approach.

Towards intersectional approaches to radicalisation One of the ways in which the research on radicalisation and countering radicalisation has advanced, particularly during the last decade, has been through a move towards an increasingly precise description of the mechanisms, processes, and outcomes under consideration, how they are being measured, and in which populations. This has enabled the emergence of a growing body of systematic reviews that illustrate the strengths and limitations of research on, for example, risk and protective factors for radicalisation (Wolfowicz et al., 2020; Lösel et al., 2018); the processes that inform radicalisation to violence (Vergani et al., 2020); lone-actor terrorism (Kenyon et al., 2021); the effectiveness of counter-narrative interventions (Carthy et al., 2020); and multi-agency approaches to counter radicalisation (Mazerolle et al., 2021). One of the things to come from this is the benefit of adopting a more intersectional approach to interpreting trajectories towards and away from violence, looking across as well as within particular types of explanations to offer a more holistic approach to understanding and explaining radicalisation. Some work has already begun to develop a more intersectional reading of radicalisation. Of particular note here is the growing literature on the relationships between gender – including masculinities – ethnicity, religion, and class within processes of radicalisation (Pearson et al., 2020; Jensen & Larsen, 2021). There is, however, much more that can be done to develop more sophisticated and less reductive conceptual frameworks. How, for example, might theories of radicalisation and countering radicalisation more effectively take account of global inequalities, forced migration, and new information technologies alongside considerations about gender and class (e.g. Mansour-Ille, 2019)? And what roles do intergenerational processes of trauma, adversity, conflict, and mental health play (e.g. Lewis & Marsden, 2021b)? Such approaches would appear well-suited to accommodating and providing insight about the multiple, dynamic influences on radicalisation processes and enabling the development of counter-radicalisation approaches that grapple effectively with these complexities.

6

Introduction

Radicalisation as a social problem versus radicalisation as a social construct A more fundamental critique of radicalisation-related research and policy has been that it represents an agenda driven forward less by evidence than by a particular configuration of assumptions and interests, including the expansion of state power. As Charlotte Heath-Kelly (2013, p. 394) argues, by making ‘terrorism pre-emptively governable and knowable’, the concept of radicalisation identifies new sites for intervention. Reflecting such critiques, considerable attention has been paid to the implications of the radicalisation concept for policy and practice, with a particular focus on how counter-radicalisation policies and programmes focus, either by design or by accident, on specific practices, often associated with specific populations (Busher et al., 2023; Mythen et al., 2017). In effect, then, the literature reflects two distinct approaches to interpreting the radicalisation construct: on the one hand, as a process ‘out there in the world’, the understanding of which can help scholars, practitioners, and policymakers to better understand and more effectively interdict political violence; and on the other, as a social construct that is being used to shape policy and practice and transform understanding of the ‘problem’ of political violence and how it might be addressed. These two approaches are not mutually exclusive, however. It is possible to acknowledge that concepts are the product of particular historical moments, reflecting their various interests, blind spots, and prejudices, without having to abandon the idea that they still refer, albeit imperfectly, to some kind of real process. Similarly, radicalisation can be understood as a significant social problem while still recognising the importance of understanding where the concept comes from, how it is being used, and to what ends and effects. One of our hopes for this volume is that it creates space for dialogue between, and encourages the effective leveraging of the most important advances within and across the literatures that engage with these two approaches.

Towards a stronger empirical base During the early wave of interest in radicalisation, detailed empirical studies were conspicuous by their relative absence (Borum, 2011), although the situation did seem to be better than in other areas of terrorism research (Neumann & Kleinmann, 2013). One of the main reasons for this was the challenge associated with deploying the sort of data sources and methods otherwise commonly used in social and political research (Freilich & LaFree, 2016). Where it is possible at all, gaining access to and carrying out research with radical milieus can be a difficult, dangerous, and timeconsuming process, and many of the datasets that might shed new light on both radicalisation and countering radicalisation are deemed sensitive by national intelligence or security communities and are held out of reach of most researchers. This has meant that most research has relied rather heavily on open-source data, often coming from court cases and media reporting, which have their own challenges and blind spots (LaFree & Yanez, this volume). In the last decade or so, there has, however, been a very significant expansion in the empirical literature on radicalisation and countering radicalisation (Schuurman, 2019, 2020b). Researchers are proving effective at finding strategies to circumvent many of the issues that initially hindered significant empirical research, and research designs have become more sophisticated. Within the field today, one can find the widespread use of a range of standard social and political science approaches, such as interview-based methods, ethnographic research, and surveys, while online research is also creating new opportunities to study and better understand radical milieus (MacDonald et al., 2023). Nonetheless, there are a number of important limitations and challenges to be addressed.

7

Sarah Marsden, Leena Malkki, and Joel Busher

Some of these limitations are associated with the scope of current research in terms of its geographic, linguistic, temporal, and ideological focus. Reflecting the primacy of the Global North in research on terrorism and radicalisation, there is an overwhelming focus on research and practice emerging from the Global North and, to a lesser extent, those parts of the Global South that have been identified by actors in the Global North as strategic priority areas (Jackson et al., 2020; Mohammed, 2022). Other regions, including Russia and Latin America, are largely neglected in the evidence base. Even research on transnational phenomena, such as foreign fighters, is still largely focused on the experiences of countries in the Global North (De Roy van Zuijdewijn, this volume). This is problematic not only in terms of gaps in knowledge but also in terms of how this might be distorting local, national, and international research and policy agendas (Azmiya, this volume; Kundnani & Hayes, 2018; Mwangi, this volume), and how such dynamics, which take as their starting point the problems facing actors in the Global North rather than the needs and interests of communities in the Global South, reflect colonial patterns of knowledge production and diffusion (Madriaza, Morin, & Venkatesh, this volume). Similarly, in terms of languages, although research on the online manifestations of violent extremism has included Arabic as well as English language content (Amarasingam, this volume), the evidence base more broadly reflects an Anglophone bias both with respect to the language research is published in (Mohammed, 2022) and the data that is collected to inform the study of terrorism (LaFree, 2019). With respect to ideological focus, one of the more commented-upon features of P/CVE practice and research has been its heavy concentration on Islamist terrorism and radicalisation (Schuurman, 2019). More recently, the balance is beginning to shift, and a modest but growing body of research is examining the extreme right (Ahmed & Lynch, 2021). Contemporary research is in fact focusing more on far-right online milieus and established platforms than jihadist online milieus and newer platforms (Amarasingam, this volume). Nevertheless, other forms of radicalisation, particularly outside the Global North, still receive only scant, if any, scholarly attention (Azmiya, this volume; Ragazzi & Walmsey, this volume). Addressing these biases or gaps in the literature would not only address empirical limitations but also enable more comparative analyses, thereby helping to further develop and refine theoretical insights about radicalisation and countering radicalisation. As Tim Wilson argues in this volume, another way to leverage such comparisons is by exploiting more effectively the insights that can be gained from looking into historic cases of radicalisation. Furthermore, a more historically oriented approach provides a route to understanding dynamics that unfold over centuries rather than years and decades, making it possible to grasp more fully how macro-historical processes shape the more specific forms of radicalisation that emerge in certain socio-historical moments (Sánchez-Cuenca, 2019). A second set of challenges relates to persistent issues around counterfactuals and meaningful comparison groups. A common criticism of radicalisation research has been that much of the time it in effect selects on the dependent variable, making it susceptible to a range of biases and potentially false inferences and to the risk of overpredicting the likelihood of radicalisation (Clemmow et al., 2020). Some studies have sought to address this through the construction of comparison groups (Bartlett & Miller, 2012; Knight et al., 2017), but here researchers have had to confront difficult questions about how they identify the shared group-ness of those who do and do not engage in violence. More recently, there have been a number of efforts to advance an understanding of radicalisation by studying the relative ‘lack’ of political violence within particular settings, milieus, or life histories (Malkki, 2020), such as Ignacio Sánchez-Cuenca’s (2019) comparative study of the New 8

Introduction

Left terrorism, Nick Brooke’s (2018) study on nationalism and terrorism in the United Kingdom, and Bart Schuurman and colleagues’ (Schuurman, 2020a) analysis of individual radicalisation journeys that do not end in terrorist violence. In a similar vein, a number of studies have started to take more seriously questions about how we can understand and explain processes of restraint within violent milieus and what this can tell us about radicalisation pathways more broadly (e.g. Busher & Bjørgo, 2020). On a parallel but related track, research has also been undertaken seeking to identify base rates of particular characteristics to understand their prevalence in the general population, thereby enabling a more effective and meaningful comparison with populations that engage in political violence (Clemmow et al., 2020). Research and evaluation studying the outcomes and effectiveness of P/CVE programmes, in the primary, secondary, or tertiary space, have faced similar problems. These have been particularly acute due to concerns around the ethical implications of using control or comparison groups in such a high-risk field. However, while this problem at one time seemed almost insurmountable, the last decade has seen research and evaluation teams identify and put into practice a number of evaluation designs that go a long way towards mitigating these issues (Fisher & Busher, this volume). While challenges persist, particularly around scale-ability, there are now an increasing number of P/CVE interventions that have been studied using experimental and quasi-experimental designs (e.g. Parker & Lindekilde, 2020; Aldrich, 2014; Webber et al., 2018). This, however, brings us to a third area, which concerns the persistence of important blindspots around counter-radicalisation programmes. Many of the contributions to this volume show how the empirical research base around counter-radicalisation work has become stronger in recent years, highlighting more effectively the strengths, weaknesses, and problems with a range of P/ CVE approaches, such as work with communities (Thomas & Grossman, this volume); public health and whole society approaches (Hardy, this volume; Ragazzi & Walmsey, this volume); and multi-agency approaches (Andersson Malmros & Sivenbring, this volume). There are, however, still a number of major limitations in the research. Some of these are about basic descriptive gaps. Notwithstanding the increase in research, and the willingness of some stakeholders to provide researchers access to P/CVE programmes, there is still not enough information about how these programmes actually operate in practice. As Ragazzi and Walmsey (this volume) argue, if knowledge and understanding about how counter-radicalisation initiatives do and do not work are really to advance, it is crucial that more is done to open the black box of P/CVE interventions, particularly when it comes to understanding the experiences of those who are the subject/target group of those initiatives. There is still strikingly scant empirical research on or with those referred into P/ CVE programmes (Thomas & Grossman, this volume) or on the experiences and effects (rather than effectiveness) of participation in intervention programmes (Da Silva, this volume). The other major area where further research effort is required concerns questions about what effective counter-radicalisation programming looks like and how it can be achieved in practice. Again, there have been a number of important advances in this area in recent years. There is now a broad consensus that there are no one-size-fits-all solutions but instead that P/CVE initiatives should be tailored to their context (Koehler, this volume). Alongside this, as researchers have developed and embraced more sophisticated research designs (Lewis et al., 2020), there has emerged a growing and increasingly high-quality evidence base underpinning our understanding and assessment of, to use the realist evaluation formula, ‘what works, for whom, under what circumstances, and why’ (Gielen, 2019). Key questions now include how to formulate transferable practical learnings about programme and evaluation design, how different approaches and initiatives are ‘translated’ when moving from one context to another (Andersson Malmros & Sivenbring, this volume), how to move beyond having a series of examples of good evaluation 9

Sarah Marsden, Leena Malkki, and Joel Busher

practice to fostering the emergence of ‘an overall system of constructive P/CVE evaluation practice’ (Baykal et al., 2021, p. 7), and the implications of this for policy and practice. Looking forward, a further challenge that is likely to remain is how the literature on radicalisation and countering radicalisation keeps pace with changing societal, technological, and political contexts. The world in which radicalisation and countering radicalisation happen is changing, and while certain fundamental insights might continue to hold, it is reasonable to expect that these changes will also impact the processes of radicalisation and countering radicalisation. Keeping pace with these changes will in part be about understanding, at speed and as they arise, specific opportunities, developments, policies and approaches, and their wider implications. In this volume, for example, Bharath Ganesh discusses the theoretical and practical issues around online content moderation – on the one hand, a specific P/CVE tool, but one that also has important implications for much wider debates around fundamental rights and civil liberties that are playing out against a backdrop of increased polarisation and state intervention in the online space. Research will also need to grapple with how and to what extent these societal changes require new conceptual and theoretical advances. For example, as Amarnath Amarasingam (this volume) points out, while numerous articles refer to the ‘online community’ or ‘virtual community’, very few studies interrogate this idea to consider how developing a sense of belonging to a community – virtual or otherwise – works in a digital age and the relevance of these processes to radicalisation. Understanding better how young people who have grown up with the Internet as a major source of community and information interpret, consume, and engage in politics will be important to help make more sense of what is nowadays often referred to as hybridisation, or ‘salad bar extremism’, where individuals pick and choose aspects of different ideologies or subcultures to generate new, syncretic forms of extremism (e.g. Koch, 2021). Geo-political changes also tend to demand adaptation and innovation on the part of researchers working in the social and political sciences. How will research on radicalisation and countering radicalisation be affected by, for example, the apparent decline in the cultural and political hegemony of the United States in the field of international relations or by the advance of what some scholars refer to as ‘post-politics’? Whatever the answer, it is likely that effective adaptation will require engagement with other fields, such as social movement studies, or peace and conflict studies, that have frameworks to help interpret new forms of mobilisation and contestation and their relevance to the dynamics of political violence (Malthaner, this volume).

The practice of radicalisation research It is also worth saying something briefly about research practice. As discussed earlier, this is a relatively young field of research, at least in its current configuration, that draws across multiple disciplines and research traditions and where the stakes have often felt high. It is perhaps unsurprising, therefore, that there have been extensive and often heated debates about how to do research in this area. A full summary of these debates goes well beyond the scope of this Introduction, but we believe it is helpful to draw attention to what appear to us to be some of the most important aspects of these debates within the literature. A first point to make is simply that, while contributors to these debates might often disagree with one another, there can be little doubt about the level of interest in ethical questions among scholars working in this field, whether they come from more ‘critical’ or ‘policy-focused’ traditions. Particularly as empirical research has become more common and as researchers have found themselves having to negotiate often highly risk-averse institutional ethics committees, there have been a number of attempts to draw out ethical principles for use in research on terrorism and 10

Introduction

(violent) extremism (e.g. Conway, 2021; Mills et al., 2020; Morrison et al., 2021). Among the most prominent issues of concern have been questions about how to apply the principle of informed consent without compromising the respondents’ safety, the implications of studying vulnerable populations, and what kind of legal ramifications collecting terrorist content may have for the researcher and research participants. Another ethical issue that has received a great deal of attention concerns the relationship between researchers and counterterrorism policy planners and practitioners (Martini & Fernández de Mosteyrín, this volume; Toros, 2016). These debates have a long history, having been discussed in terrorism studies for many years prior to the emergence of interest in radicalisation (Brannan et al., 2001; Gorriti, 1991; Zulaika & Douglass, 1996). They are particularly salient in the field of radicalisation and countering radicalisation because so much of the radicalisation research to date has responded fairly directly to emerging policy agendas and problems, sometimes enabled through direct government funding, access to data, or broader efforts at research-practice knowledge exchange (Phillips, 2023). In contrast to Marc Sageman’s (2014) assessment of the state of terrorism research and what he described as the ‘unbridgeable gaps’ between practitioners and researchers, the issue raised by some is how researchers remain alert to the influence of practitioner priorities on knowledge production (Schuurman, 2019). Discussion of these issues is increasingly visible in research on terrorism and, by extension, the study of radicalisation. Scholars are now reflecting on how to navigate these relationships, how to carry out research, and how to form and maintain the relationships required to do so while remaining alert to the potential negative consequences of too close a collaboration between policy planners, practitioners, and academics (Martini & Fernández de Mosteyrín, this volume; Morrison, 2022; Ragazzi & Walmsey, this volume). Part of the answer lies in the importance of providing realistic appraisals to policy and practitioner communities about what can and cannot be achieved – what can and cannot be concluded with confidence from the research (Gill et al., this volume). This is easier said than done, however, particularly when researchers often find themselves under considerable pressure to find ‘solutions’ and compete for attention and investment. This requires researchers to find ways to articulate the nuance in their findings while communicating effectively and holding the attention of policy and practitioner audiences. A further topic that has received growing attention has been researcher safety and well-being. Arguably long overdue, it is likely that one of the drivers of interest in the issue has been wider societal concerns about mental health and well-being, particularly in the workplace. It has also been argued that while studying violent extremism and counterterrorism has always come with a psychological toll, these issues have become more acute today with the increasing availability of online, and often highly graphic, research data. Several strategies to mitigate the negative psychological impacts of such research have been discussed, and academic institutions and research groups are increasingly alert to the need to provide support for researchers to help deal with the emotional cost and possible trauma of this work (Pearson et al., 2023; Winter, 2019). There are also a growing number of publications that share learning and recommendations about conducting terrorism-related fieldwork (Dolnik ed., 2013; Sluka, 2020). There is undoubtedly more work to be done on working through and reflecting on research practice, and how this is changing amid societal and technological developments. Nonetheless, the fact that these discussions are taking place and that questions about ethics, integrity, safety, and wellbeing are being taken seriously is, we believe, indicative of an increasingly mature research field.

*** 11

Sarah Marsden, Leena Malkki, and Joel Busher

The overall impression we were left with after putting this Handbook together was that while there is still much to be done to expand and enhance the research on radicalisation and countering radicalisation, scientific understanding has advanced greatly since the early 2000s. What has been particularly notable is the considerable advancement of research methods, the growth in empirical research, and the increasingly rigorous use of theoretical approaches drawn from across various disciplines. We hope that future research on radicalisation and countering radicalisation will follow this track and that this Handbook will provide inspiration and direction for new research programmes as well as serve as an introduction to students and researchers who are new to the area. As the editors of the Handbook, we also wish to express our gratitude to all the contributors to this volume. Their commitment to producing insightful and comprehensive chapters, and providing feedback to other contributors, is testimony to the vigour of the research field today. Our warmest thanks also go to Valpuri Alanen (University of Helsinki), who provided crucial help in the final stages of the editing process.

References Aggarwal, N. K. (2018). Questioning the current public health approach to countering violent extremism. Global Public Health, 14(2), 309–317. Ahmed, Y., & Lynch, O. (2021). Terrorism studies and the far right – the state of play. Studies in Conflict & Terrorism, 1–21. Aldrich, D. P. (2014). First steps towards hearts and minds? USAID’s countering violent extremism policies in Africa. Terrorism and Political Violence, 26(3), 523–546. Alimi, E., Demetriou, C., & Bosi, L. (2015). Dynamics of radicalisation: Relational and comparative perspective. Oxford University Press. Ambrozik, C. (2019). Countering violent extremism globally. Perspectives on Terrorism, 13(5), 102–111. Baker-Beall, C., Heath-Kelly, C., & Jarvis, L. (2015). Counter-radicalisation in Europe: Critical perspectives. Routledge. Bartlett, J., & Miller, C. (2012). The edge of violence: Towards telling the difference between violent and non-violent radicalization. Terrorism and Political Violence, 24(1), 1–21. Baykal, A., Bressan, S., Friedrich, J., Pasquali, G., Rotmann, R., & Wagner, M. (2021). Evaluating P/CVE: Institutional structures in international comparison. Global Public Policy Institute. www.gppi.net/media/ Friedrich-et-al_PrEval_ENG_UPDATED.pdf Berger, J. M. (2016). Making CVE work: A focused approach based on process disruption. International Centre for Counter-Terrorism. Borum, R. (2011). Radicalization into violent extremism II: A review of conceptual models and empirical research. Journal of Strategic Security, 4(4), 37–62. Bouhana, N. (2019). The moral ecology of extremism: A systemic perspective. Paper prepared for the UK Commission for Countering Extremism. Brannan, D. W., Esler, P. F., & Strindberg, N. T. A. (2001). Talking to “terrorists”: Towards an independent analytical framework for the study of violent substate activism. Studies in Conflict & Terrorism, 24(1), 3–24. Brooke, N. (2018). Terrorism and nationalism in the United Kingdom: The absence of noise. Palgrave. Busher, J., & Bjørgo, T. (2020). Restraint in terrorist groups and radical milieus. Perspectives on Terrorism, 14(6), 2–13. Busher, J., Choudhury, T., & Thomas, P. (2023). Surveillance and preventing violent extremism: The evidence from schools and further education Colleges in England. In M. Kwet (Ed.), The Cambridge handbook of race and surveillance (pp. 267–287). Cambridge University Press. Carthy, S. L., Doody, C. B., Cox, K., O’Hora, D., & Sarma, K. M. (2020). Counter‐narratives for the prevention of violent radicalisation: A systematic review of targeted interventions. Campbell Systematic Reviews, 16(3), e1106. Clemmow, C., Rottweiler, B., Wolfowicz, M., Bouhana, N., Marchment, Z., & Gill, P. (2023). The whole is greater than the sum of its parts: Risk and protective profiles for vulnerability to radicalization. Justice Quarterly, 1–27.

12

Introduction Clemmow, C., Schumann, S., Salman, N. L., & Gill, P. (2020). The base rate study: Developing base rates for risk factors and indicators for engagement in violent extremism. Journal of Forensic Sciences, 65(3), 865–881. Conway, M. (2021). Online extremism and terrorism research ethics: Researcher safety, informed consent, and the need for tailored guidelines. Terrorism and Political Violence, 33(2), 367–380. Coolsaet, R. (2016). ‘All radicalisation is local’: The genesis and drawbacks of an elusive concept (Vol. 84). Egmont-Royal Institute for International Relations. Crone, M. (2016). Radicalization revisited: Violence, politics and the skills of the body. International Affairs, 92(3), 587–604. Dawson, L. L. (2009). The study of new religious movements and the radicalization of home-grown terrorists: Opening a dialogue. Terrorism and Political Violence, 22(1), 1–21. Dawson, L. L. (2017). Sketch of a social ecology model for explaining homegrown terrorist radicalisation. ICCT research note. The International Centre for Counter-Terrorism (ICCT). Della Porta, D. (2018). Radicalization: A relational perspective. Annual Review of Political Science,  21, 461–474. Della Porta, D. (2020). Protests as critical junctures: Some reflections towards a momentous approach to social movements. Social Movement Studies, 19(5–6), 556–575. Della Porta, D., & LaFree, G. (2012). Guest editorial: Processes of radicalization and de-radicalization. International Journal of Conflict and Violence, 6(1), 4–10. Dolnik, A. (Ed.). (2013). Conducting terrorism field research: A guide. Routledge. Duriesmith, D., & Ismail, N. H. (2022). Embodied militarism and the process of disengagement from foreign fighter networks. Critical Military Studies, 8(1), 22–38. Ellis, B. H., Miller, A. B., Schouten, R., Agalab, N. Y., & Abdi, S. M. (2022). The challenge and promise of a multidisciplinary team response to the problem of violent radicalization. Terrorism and Political Violence, 34(7), 1321–1338. Elshimi, M. S. (2017). De-radicalisation in the UK prevent strategy: Security, identity, and religion. Routledge. Ewert, B., Loer, K., & Thomann, E. (2021). Beyond nudge: Advancing the state-of-the-art of behavioural public policy and administration. Policy & Politics, 49(1), 3–23. Expert Group. (2008). Radicalisation processes leading to acts of terrorism. Report prepared by the European Commission’s Expert Group on Violent Radicalisation. The list of members and their affiliation, as well as the text of the report is available at. www.clingendael.org/sites/default/files/pdfs/20080500_cscp_report_ vries.pdf Feddes, A. R., Nickolson, L., Mann, L., & Doosje, B. (2020). Psychological perspectives on radicalization. Routledge. Freilich, J. D., & LaFree, G. (2016). Measurement issues in the study of terrorism: Introducing the special issue. Studies in Conflict & Terrorism, 39(7–8), 569–579. Gielen, A. J. (2019). Countering violent extremism: A realist review for assessing what works, for whom, in what circumstances, and how? Terrorism and Political Violence, 31(6), 1149–1167. Githens-Mazer, J. (2012). The rhetoric and reality: Radicalization and political discourse. International Political Science Review, 33(5), 556–567. Gorriti, G. (1991). Terrorism, research and public policy: An experience, some thoughts. Terrorism and Political Violence, 3(1), 103–116. Grossman, G., Hadfield, K., Jefferies, P., Gerrand, V., & Ungar, M. (2020). Youth resilience to violent extremism: Development and validation of the BRAVE measure. Terrorism and Political Violence, 34(3), 468–488. Harris-Hogan, S., Barrelle, K., & Zammit, A. (2016). What is countering violent extremism? Exploring CVE policy and practice in Australia. Behavioral Sciences of Terrorism and Political Aggression, 8(1), 6–24. Heath-Kelly, C. (20130). Counter-terrorism and the counterfactual: Producing the radicalisation discourse and the UK PREVENT strategy. The British Journal of Politics and International Relations, 15(3), 394–415. Holbrook, D., & Horgan, J. (2019). Terrorism and ideology. Perspectives on Terrorism, 13(6), 2–15. Horgan, J. (2020, December 2). Terrorism and tornadogenesis: A multi-level, multi-role, multi-mechanism socio-psychological model of terrorism and its participants. Paper presentation, ECPR Standing Group on Political Violence Seminar Series. Horgan, J., & Taylor, M. (2011). Disengagement, de-radicalization and the arc of terrorism: Future directions for research. In R. Coolsaet (Ed.), Jihadi terrorism and the radicalisation challenge: European and American experiences (2nd ed., pp. 173–186). Ashgate.

13

Sarah Marsden, Leena Malkki, and Joel Busher Irwin, N. (2015). The complexity of responding to home-grown terrorism: Radicalisation, de-radicalisation and disengagement. Journal of Policing, Intelligence and Counter Terrorism, 10(2), 166–175. Jackson, R., Toros, H., Jarvis, L., & Heath-Kelly, C. (Eds.). (2020). Critical terrorism studies at ten: Contributions, cases and future challenges. Routledge. Jensen, S. Q., & Larsen, J. F. (2021). Sociological perspectives on Islamist radicalization – bridging the micro/macro gap. European Journal of Criminology, 18(3), 426–443. Kenyon, J., Baker-Beall, C., & Binder, J. (2021). Lone-actor terrorism – a systematic literature review. Studies in Conflict and Terrorism. https://doi.org/10.1080/1057610X.2021.1892635 Khalil, J., Horgan, J., & Zeuthen, M. (2022). The attitudes-behaviors corrective (ABC) model of violent extremism. Terrorism and Political Violence, 34(3), 425–450. Knight, S., Woodward, K., & Lancaster, G. (2017). Violent versus non-violent actors: An empirical study of different types of extremism. Journal of Threat Assessment and Management, 4(4), 230–248. Koch, A. (2021). The ONA network and the transnationalization of neo-nazi-satanism. Studies in Conflict and Terrorism. https://doi.org/10.1080/1057610X.2021.2024944 Koehler, D. (2017). A typology of ‘de-radicalisation’ programmes. In L. Colaert (Ed.), Deradicalisation: Scientific insights for policy (pp. 63–81). Flemish Peace Institute. Kundnani, A., & Hayes, B. (2018). The globalisation of countering violent extremism policies. Undermining human rights, instrumentalising civil society. Transnational Institute. LaFree, G. (2019). The evolution of terrorism event databases. In A. Gofas, E. Chenoweth, E. English, & S. N. Kalyvas (Eds.), The Oxford handbook of terrorism (pp. 50–68). Oxford University Press. Leuprecht, C., Hataley, T., Moskalenko, S., & McCauley, C. (2010). Containing the narrative: Strategy and tactics in countering the storyline of global Jihad. Journal of Policing, Intelligence, and Counter Terrorism, 5, 42–57. Lewis, J., & Marsden, S. V. (2021a). Countering violent extremism interventions: Contemporary research. Centre for Research and Evidence on Security Threats. https://crestresearch.ac.uk/resources/ evaluating-programmes-to-prevent-and-counter-extremism/ Lewis, J., & Marsden, S. V. (2021b). Trauma, adversity and violent extremism. Centre for Research and Evidence on Security Threats. https://crestresearch.ac.uk/resources/trauma-adversity-and-violent-extremism/ Lewis, J., Marsden, S. V., & Copeland, S. (2020). Evaluating programmes to prevent and counter extremism. Centre for Research and Evidence on Security Threats. https://crestresearch.ac.uk/resources/ countering-violent-extremism-interventions/ Lösel, F., King, S., Bender, D., & Jugl, I. (2018). Protective factors against extremism and violent radicalization: A systematic review of research. International Journal of Developmental Science, 12(1–2), 89–102. Macdonald, S., Pearson, E., Scrivens, R., & Whittaker, J. (2023). Using online data in terrorism research. In A research agenda for terrorism studies (pp. 145–158). Edward Elgar Publishing. Macklin, G., & Busher, J. (2015). The missing spirals of violence: Four waves of movement – countermovement contest in post-war Britain. Behavioral Sciences of Terrorism and Political Aggression, 7(1), 53–68. Malkki, L. (2020). Learning from the lack of political violence: Conceptual issues and research design. Perspectives on Terrorism, 14(6), 27–36. Malthaner, S. (2017). Radicalization: The evolution of an analytical paradigm. European Journal of Sociology/Archives Européennes de Sociologie, 58(3), 369–401. Mansour‐Ille, D. (2019). Social media and the dynamics of radicalization and violent extremism among female migrant workers. Studies in Ethnicity and Nationalism, 19(3), 248–268. Marsden, S. V. (2017). Reintegrating extremists: Deradicalisation and desistance. Springer. Martins, B. O., & Ziegler, M. (2018). Counter-radicalization as counter-terrorism: The European Union case. In Expressions of radicalization: Global politics, processes and practices (pp. 321–352). Mazerolle, L., Cherney, A., Eggins, E., Hine, L., & Higginson, A. (2021). Multiagency programs with police as a partner for reducing radicalisation to violence. Campbell Systematic Reviews, 17(2), e1162. McCauley, C., & Moskalenko, S. (2017). Understanding political radicalization: The two-pyramids model. American Psychologist, 72(3), 205. McDonald, K. (2018). Radicalization. Polity Press. Mills, T., Massoumi, N., & Miller, D. (2020). The ethics of researching “terrorism” and political violence: A sociological approach. Contemporary Social Science, 15(2), 119–133. Moghaddam, F. M. (2005). The staircase to terrorism: A psychological exploration. American Psychologist, 60(2), 161.

14

Introduction Mohammed, I. (2022). Decolonialisation and the terrorism industry. Critical Studies on Terrorism,  15(2), 417–440. Morrison, J. F. (2022). Talking stagnation: Thematic analysis of terrorism experts’ perception of the health of terrorism research. Terrorism and Political Violence, 34(8), 1509–1529. Morrison, J. F., Silke, A., & Bont, E. (2021). The development of the framework for research ethics in terrorism studies (FRETS). Terrorism and Political Violence, 33(2), 271–289. Mykkanen, T. (2022). Intersubjective body mapping for reintegration: Assessing an art-based methodology to promote reintegration of foreign terrorist fighters. Critical Studies on Terrorism, 15(4), 988–1022. Mythen, G., Walklate, S., & Peatfield, E. J. (2017). Assembling and deconstructing radicalisation in PREVENT: A case of policy-based evidence making? Critical Social Policy, 37(2), 180–201. Neumann, P. R. (2008). Introduction. In P. R. Neumann, J. Stoil, & D. Esfandiary (Eds.), Perspectives on radicalisation and political violence: Papers from the first international conference on radicalisation and political violence. ICSR. Neumann, P. R. (2013). The trouble with radicalization. International Affairs, 89(4), 873–893. Neumann, P. P., & Kleinmann, S. (2013). How rigorous is radicalization research? Democracy and Security, 9(4), 360–382. Parker, D., & Lindekilde, L. (2020). Preventing extremism with extremists: A double-edged sword? An analysis of the impact of using former extremists in Danish schools. Education Sciences, 10(4), 111. Pearson, E., Whittaker, J., Baaken, T., Zeiger, S., Atamuradova, F., & Conway, M. (2023). Online extremism and terrorism researchers’ security, safety, and resilience: Findings from the field. VOX-Pol. www.voxpol. eu/download/report/Online-Extremism-and-Terrorism-Researchers-Security-Safety-Resilience.pdf Pearson, E., Winterbotham, E., & Brown, K. E. (2020). Countering violent extremism: Making gender matter. Palgrave Macmillan. Peels, R. (2023). Towards a fruitful concept of radicalisation: A synthesis. Journal of Contemporary European Studies, https://doi.org/10.1080/14782804.2023.2185594 Phillips, B. J. (2023). How did 9/11 affect terrorism research? Examining articles and authors, 1970–2019. Terrorism and Political Violence, 35(2), 409–432. Pilkington, H. (2016). Loud and proud: Passion and politics in the English defence league (p. 328). Manchester University Press. Pisoiu, D. (2015). Subcultural theory applied to Jihadi and right-wing radicalization in Germany. Terrorism and Political Violence, 27(1), 9–28. Puigvert, L., Aiello, E., Oliver, E., & Ramis-Salas, M. (2020). Grassroots community actors leading the way in the prevention of youth violent radicalization. PLoS One, 15(10), e0239897. Reidy, K. (2018). Radicalization as a vector: Exploring non-violent and benevolent processes of radicalization. Journal for Deradicalization (14), 249–294. Richards, A. (2015). From terrorism to ‘radicalization’ to ‘extremism’: Counterterrorism imperative or loss of focus? International Affairs, 91(2), 371–380. Sageman, M. (2004). Understanding terror networks. University of Pennsylvania Press. Sageman, M. (2014). The stagnation in terrorism research. Terrorism and Political Violence, 26(4), 565–580. Sánchez-Cuenca, I. (2019). Historical roots of political violence: Revolutionary terrorism in affluent countries. Cambridge University Press. Schock, K. (2015). Civil resistance today. Polity Press. Schuurman, B. (2019). Topics in terrorism research: Reviewing trends and gaps, 2007–2016. Critical Studies on Terrorism, 12(3), 463–480. Schuurman, B. (2020a). Non-involvement in terrorist violence. Perspectives on Terrorism, 14(6), 14–26. Schuurman, B. (2020b). Research on terrorism, 2007–2016: A review of data, methods, and authorship. Terrorism and Political Violence, 32(5), 1011–1026. Schuurman, B., Bakker, E., Gill, P., & Bouhana, N. (2018). Lone actor terrorist attack planning and preparation: A data‐driven analysis. Journal of Forensic Sciences, 63(4), 1191–1200. Sedgwick, M. (2010). The concept of radicalization as a source of confusion. Terrorism and Political Violence, 22(4), 479–494. Shanaah, S., & Heath-Kelly, C. (2022). What drives counter-extremism? The extent of P/CVE policies in the west and their structural correlates. Terrorism and Political Violence, 1–29. Silke, A. (2011). Disengagement or deradicalization: A look at prison programs for jailed terrorists. CTC Sentinel, 4(1), 18–21.

15

Sarah Marsden, Leena Malkki, and Joel Busher Sluka, J. (2020). Too dangerous for fieldwork? The challenge of institutional risk-management in primary research on conflict, violence and ‘terrorism’. Contemporary Social Science, 15(2), 241–257. Stephens, W., Sieckelinck, S., & Boutellier, H. (2021). Preventing violent extremism: A review of the literature. Studies in Conflict & Terrorism, 44(4), 346–361. Toros, H. (2016). Dialogue, praxis and the state: A response to Richard Jackson. Critical studies on terrorism (pp. 126–130). Routledge. U.S. Department of State Country Reports on Terrorism. (2006). Chapter 1: Strategic assessment. www. globalsecurity.org/security/library/report/2007/c-rprt-terrorism_2006-01.htm van Stekelenburg, J. (2017). Radicalization and violent emotions. Political Science & Politics, 50(4), 936–939. Vergani, M., Iqbal, M., Ilbahar, E., & Barton, G. (2020). The three Ps of radicalization: Push, pull and personal. A  systematic scoping review of the scientific evidence about radicalization into violent extremism. Studies in Conflict and Terrorism, 43(10), 854–854. Webber, D., Chernikova, M., Kruglanski, A. W., Gelfand, M. J., Hettiarachchi, M., Gunaratna, R., Lafreniere, M.-A., & Belanger, J. J. (2018). Deradicalizing detained terrorists. Political Psychology, 39, 539–556. https://doi.org/10.1111/pops.12428 Webber, D., & Kruglanski, A. (2017). Psychological factors in radicalisation: A “3N” approach. In G. LaFree & J. Freilich (Eds.), The handbook of the criminology of terrorism (pp. 33–46). Wiley Blackwell. Winter, C. (2019). Researching Jihadist propaganda: Access, interpretation & trauma. RESOLVE Network. https://doi.org/10.37805/rve2019.1 Wolfowicz, M., Litmanovitz, Y., Weisburd, D., & Hasisi, B. (2020). A field-wide systematic review and metaanalysis of putative risk and protective factors for radicalization outcomes. Journal of Quantitative Criminology, 36(3), 407–447. Youngman, M. (2020). Building “terrorism studies” as an interdisciplinary space: Addressing recurring issues in the study of terrorism. Terrorism and Political Violence, 32(5), 1091–1105. Zulaika, J., & Douglass, W. A. (1996). Terror and taboo: The follies, fables, and faces of terrorism. Routledge.

16

PART I

The history of research on radicalisation

2 BEFORE ‘RADICALISATION’ Explaining individual involvement in terrorism before the popularisation of the radicalisation concept Leena Malkki Introduction Why terrorism happens has been a key question within terrorism studies ever since it emerged as a distinct research area in the early 1970s. However, using the term ‘radicalisation’ to describe the process leading to terrorism is of much more recent vintage. As Rik Coolsaet explains in the next chapter, the concept became popular in the policymaking arena and, consequently, in academic research during the first decade of the 21st century. This chapter maps the different approaches and conceptualisations that were used to study what is now called radicalisation during the first three decades of terrorism studies. Studies from that era have their missteps and methodological weaknesses, and some of the presented explanatory approaches have since been abandoned. At the same time, these decades saw the publication of several key studies that already include many current assertions about radicalisation that have been put forward in more recent literature. This chapter begins with a short reflection on why the term ‘radicalisation’ was not used before the 2000s. It then introduces the most notable lines of research on individual involvement in terrorism during that period and discusses how the legacy of these studies is visible in contemporary studies of radicalisation. The chapter concludes with a short discussion of the pros and cons of the term ‘radicalisation’ vis-à-vis earlier conceptualisations.

Why was ‘radicalisation’ framed differently before the 2000s? Radicalisation has become such an important organising concept in contemporary research and policymaking that it may be difficult to imagine talking about the emergence of terrorism and violent extremism without it. The term is an integral part of the current explanatory approaches. How the emergence of terrorism has been interpreted and approached has changed over time, and changing terminology is one reflection of that. A simple explanation for the absence of the term ‘radicalisation’ in earlier research is that preventing radicalisation had not yet become a policy priority. Approaches and research questions that have dominated terrorism studies have traditionally been strongly influenced by contemporary events and debates. As Martha Crenshaw (2011), one of the most influential early scholars 19

DOI: 10.4324/9781003035848-3

Leena Malkki

of terrorism, notes in her retrospective account, the study of terrorism ‘has always been driven by events and policy interests’ and has taken ‘place in a mixed and messy world of academics, think tanks, government analysts, journalists, and pundits’ (p. 1). A rather significant part of the research has always sought to answer questions that seem to support contemporary counterterrorism efforts. Consequently, the trends in counterterrorism have had a strong impact on what kinds of questions have been asked in the research field and partly on what kinds of analytical approaches have been preferred. In previous decades, counterterrorism had a different orientation and, thereby, different knowledge needs. The underlying puzzle was the same – explaining why terrorism happens and what can be done about it – but the strategies adopted to tackle that question were different. What initially gave the impetus for the emergence of terrorism studies was the ‘rise of modern terrorism’ in the late 1960s and 1970s (Stampnitzky, 2013, pp. 21–27; Crenshaw, 2019). This referred to the almost simultaneous emergence of plane hijackings by Palestinian organisations, far-left terrorism in West Germany, Italy, the United States, and other countries, and the rising levels of violence in the ongoing ethnic conflicts in Northern Ireland and the Basque country. The ensuing research on terrorism was often geared towards analysing these and other ongoing terrorist campaigns and how to deal with them. The main interest was to learn more about those already involved in terrorism. Knowledge about terrorist profiles, mindsets, and decision-making was expected to produce insights on ‘what kind of a being the terrorist really is’ (Cooper, 1977, p. 20) and what can be expected from them. Counterterrorism did include a predictive element, but it was geared towards predicting what those already involved in terrorist groups might do next. Early prevention of terrorism, in the form of trying to stop individuals from joining terrorist groups or adopting extremist worldviews, was not yet really an objective of counterterrorism. This was also not what the researchers preoccupied themselves with. In those few instances in which researchers discussed the possibility of ‘forecasting’ who becomes involved in terrorism, they were pessimistic about whether the current theories allowed for that. As Franco Ferracuti (1982) warned, this would come with a substantial risk of ‘overgeneralization and overprediction’ (p. 134). One difference between research on terrorism today and in previous decades is the relative importance attached to understanding individual participation in terrorism. Before the 2000s, terrorism was treated most of all as a group activity, and explaining why individuals joined those movements was not considered a priority. When explaining the emergence of terrorism, most academic researchers were more inclined to prioritise research on the group level and the larger socio-political context while also acknowledging that there are individual-level factors that should be considered. This approach is well summarised by Martha Crenshaw in her seminal article ‘Causes of Terrorism’ (Crenshaw, 1981), which continues to be one of the most cited academic articles on terrorism (Silke & Schmidt-Petersen, 2017) and included in several readers (e.g. Neumann [ed.], 2015). Crenshaw (1981) argued that a comprehensive analysis of the causes of terrorism should pay attention to three levels: the setting, the decision-making in terrorist groups, and individual participants. This emphasis is also visible in one of the earliest contributions discussing the prevention of radicalisation (Sprinzak, 1991, pp. 67–68): effective anti-terrorist policy may be initiated long before the actual appearance of terrorism. If terrorism is the highest stage of a long process of delegitimation, then the prudent policy-maker does not have to wait until all the conditions for the emergence of terrorism are present. We may act much earlier to either stop the radicalization that produces terrorism or be prepared for its upsurge. The indeterminism of the process of delegitimation makes rebel terrorism extremely vulnerable to prudent intervention. It implies the existence of 20

Before ‘radicalisation’

weak spots in the pre-terroristic stages at which wise governmental action can either prevent the evolution of terrorism or reduce its effectiveness a great deal. It is quite conceivable that research in this direction could yield ‘radicalization indicators’ and be developed into a sophisticated ‘anti-terrorist early warning system’. With ‘radicalisation indicators’, Sprinzak (1991) clearly thinks about something other than the individual-level risk assessment tools we have nowadays. What he has in mind are group- (or societal-) level indicators. Instead of explaining terrorists, he finds it more important to work towards a ‘comprehensive understanding of larger social processes that produce terrorism’ (Sprinzak, 1991, p. 68). Still, studies on individual involvement in terrorism have been conducted since the emergence of terrorism studies. Such studies initially emerged partly in response to popular stereotypes that characterised terrorists as crazy or abnormal. Later, they developed into a more versatile study of the individual motivations behind participation in terrorism and the personality traits that may make an individual inclined to get involved. Towards the end of the 1980s, the focus started to shift from the study of individual qualities to the processes of socialisation and recruitment. What is important to note is that instead of radicalisation, these studies used several different terms to describe and conceptualise the object of study, depending on the exact aspect being studied and the approach adopted. Among the terms used were, for example, terrorist personality, motivations for terrorism, political socialisation of terrorism, mobilisation or recruitment to terrorist groups, participation in underground organisations, and politicisation of violence. We will now take a closer look at these studies and the ways in which issues related to radicalisation were analysed and explained before the 2000s.

Studies on psychopathology and terrorist profile When terrorism studies started to take shape in the early 1970s, the first studies considered terrorism mainly as one type of insurgency and analysed it as such (see, e.g., Wilkinson, 1974). This approach was quickly foreshadowed as terrorism became stigmatised and politicised. In public debate, terrorists were increasingly presented as irrational and abnormal (Stampnitzky, 2013, pp. 49–82). As terrorist attacks were extreme and rare events that often involved ruthless or indiscriminate killings of people, it seemed logical to think that those involved must somehow be abnormal, too. During the 1970s, the question about why terrorism happens translated regularly into a question about what is wrong with those involved in it (for an overview, see Gill & Corner, 2017; Victoroff, 2005; Silke, 1998; Horgan, 2014, pp. 47–76). It was suggested that participation in terrorism is driven by the psychological needs and qualities of those involved rather than by their socio-political objectives and rational thinking. While acknowledging that individuals participating in terrorism were far from similar to each other, some researchers still presented bold sweeping generalisations about their mental states and motivations. One particularly prevalent proposition put forward repeatedly around this time was that at least some terrorists might suffer from psychopathy or sociopathy, something that in today’s terminology would be called antisocial personality disorder (Victoroff & Adelman, 2012). Psychopathy seemed like a possible explanation because it typically involved such remorselessness and a lack of empathy that it would make committing violent acts easier. While not claiming that all terrorists would be psychopaths, some contributors found parallels between descriptions of psychopathy and what we know about individual terrorists (e.g. Cooper, 1978) and sometimes other mental illnesses (Pearce, 1977). 21

Leena Malkki

These early psychopathological explanations were soon met with criticism, especially because they were built on very little (or no) empirical data and did not follow the methodological standards demanded by academic psychological or psychiatric research (Corrado, 1981). There were also other, often better, empirically grounded studies that produced contrasting results, pointing towards the relative absence of mental disorders among terrorists (Rasch, 1979). Another way researchers sought to shed more light on who terrorists were was by collecting information on the socio-economic and demographic backgrounds of those involved in terrorist groups. These studies were, in a way, predecessors of similar studies on individuals involved in radical Islamist terrorism (Sageman, 2004; Bakker, 2006, 2011) and later studies on Western foreign fighters in Syria and Iraq (e.g. Gustafsson & Ranstorp, 2017; Marone & Vidino, 2019; Bakker & de Bont, 2016). One of the earliest studies on the socio-economic profiles of terrorists was published in the very first issue of the academic journal Terrorism in 1977. This study claimed that terrorists were typically young single males from middle- or upper-class family backgrounds who had at least some university education (Russell & Miller, 1977; for a critique, see Merari, 1991, pp. 92–95). Later, more studies on socio-economic profiles were published, for example, on American terrorists in the 1960s, 1970s (Handler, 1990), and 1980s (Smith & Morgan, 1993), members of the Basque separatist group Euskadi Ta Askatasuna (ETA) (Clarke, 1986), and German left-wing and rightwing extremists (Schmidtchen, 1981). These studies indicated that while many terrorists were young and male, their backgrounds were far from similar and seemed to differ from one context to another. This meant that it was likely not possible to present any generalisable socio-economic or demographic terrorist profile.

Psychological studies on terrorist minds During the 1980s, psychopathological explanations for terrorism became largely sidelined. This does not mean that researchers would have thought that psychological research did not have any role in explaining terrorism. Instead of psychopathology, answers were sought by studying how psychological needs, experiences, and personality traits might contribute to individual participation in terrorism. Even though most studies started from the declared assumption that terrorists were ‘normal’, the idea of terrorists as somehow different from others did not disappear altogether (Silke, 1998; Horgan, 2014, p. 48). As we will see, studies continued to make claims that came very close to suggesting psychological abnormality. Like earlier psychopathy studies, many of these contributions have been built on very limited empirical evidence. Therefore, they should be read (and were sometimes explicitly meant to be read) as hypotheses and frameworks for further research. Some of the contributions mainly discussed how research on other types of violence and political activity could help in understanding terrorism (e.g. McCauley & Segal, 1987). The late 1970s and early 1980s, however, also saw the publication of the first major analytically oriented empirical studies on individual terrorists. For example, the German Ministry of the Interior financed a large project in which social scientists studied 250 imprisoned political extremists. The results of this project were published in a four-volume book series, Analysen zum Terrorismus (1981–4). Jeanne Knutson, for her part, interviewed convicted terrorists in US prisons (Knutson, 1981), while Franco Ferracuti and Francesco Bruno studied left-wing and right-wing terrorists in Italy (Ferracuti & Bruno, 1981). While this was surely a step forward, the number of empirical studies was still low, and the research projects conducted had significant methodological weaknesses and limitations (Crenshaw, 1986, pp. 381–382; Horgan, 2005, pp. 54–55). The choice of case studies was also notably Eurocentric. 22

Before ‘radicalisation’

One topic of discussion in the 1980s literature was that even if there was no terrorist profile, perhaps it would still be possible to identify some characteristics that were exceptionally common among terrorists. Such characteristics were sought not only from the still scant empirical evidence but also by reflecting on what involvement in terrorism requires from an individual and what kind of person would therefore be particularly suitable for or drawn to it. Among the personality traits most often discussed were a potential predisposition to violence and aggression and a positive attitude towards violence more generally. Based on available empirical evidence, many reached the conclusion that ‘attraction to violence does not appear to be the dominant aspect of their personalities’ (Crenshaw, 1986, p. 386; see also, e.g., Knutson, 1981). Interviews with terrorists showed that many of them expressed ambivalent attitudes towards violence. There were also other kinds of observations and views. For example, Peter Merkl, drawing especially from his research on interwar far-right activism in Germany, suggested that violent dispositions may even play a central role and come first, while participation in terrorism comes later. In other words, what we call radicalisation may actually be best understood as the politicisation of violence (Merkl, 1986; see also Kellen, 1982). As terrorism involved danger and stressful situations, as well as intensive activities that exceeded everyday life, it was thought to be attractive to those who enjoyed stressful situations. Therefore, stress seeking, being action-oriented, and being ready for adventure were seen as possible qualities of terrorists (e.g. Crenshaw, 1986, p. 388; Post, 1990, p. 27). Perhaps the most common theme in these studies was, however, the role of an individual’s psychological needs and deficits in early socialisation. One version of this argument was presented by Jerrold Post, who claimed that ‘political terrorists are driven to commit acts of violence as a consequence of psychological forces, and that their special psycho-logic is constructed to rationalize acts they are psychologically compelled to commit’ (Post, 1990, p. 25). By joining a terrorist group, an individual attempts to consolidate a fragmented identity and rebuild their relationship with society. A major attraction of terrorism, according to him, is that it answers one’s need to belong and provides a framework within which to blame others for one’s problems. He claims that those who are attracted to terrorism often come from troubled family backgrounds or have experienced failures in their lives (Post, 1984, 1990). Post was not alone in drawing lines between terrorist behaviour and early childhood experiences. The studies that highlighted the role of early socialisation often built upon psychodynamic approaches, which argue that human behaviour is greatly influenced by ‘a range of latent, unconscious desires, the origins of which are argued to have developed as a result of real or imagined unresolved childhood conflicts’ (Horgan, 2005, p. 60). One popular theoretical anchor for such discussions was Erik Erikson’s identity theory (see, e.g., Crenshaw, 1986). According to this theory, an individual faces several developmental stages and crises and, by successfully solving them, develops a mature, integrated identity. This may also fail, leading to identity confusion and psychological problems in later life. Following this line of explanation, what makes an individual a candidate for participation in terrorism is the failure to meet developmental tasks when growing up, resulting in identity crises to which participation in terrorist groups seems to provide a solution. Identity theory was used in a number of studies and discussions on individual terrorists during the 1980s, including many empirically grounded studies of this time (Böllinger, 1981; Knutson, 1981). Another recurring topic that also draws a line between early socialisation and terrorism is the relationship between terrorism and narcissism or narcissistic injury – damage to self-esteem in early life. Some researchers involved in the aforementioned West German state-funded project found narcissistic traits among the individuals studied. The topic comes up implicitly or explicitly 23

Leena Malkki

in several studies mentioned in this section (e.g. Shaw, 1986; Post, 1987). The narcissism explanation was later elaborated on by Richard Pearlstein (1991). A common and significant weakness of most studies in this period is, once again, the lack of empirical evidence. While not all of these contributions rest on speculation alone, they still draw from a small number of cases examined as part of studies that have not involved control groups and have suffered from other significant methodological challenges. Several contributions also have internal inconsistencies (Horgan, 2014, pp. 61–64). One of these inconsistencies was that several studies claimed they considered terrorists psychologically normal but still attached to them characteristics and personality traits that were strikingly similar to those typical of various psychiatric conditions (Silke, 1998, p. 64). Not all researchers have found research on individual personalities to be a particularly promising or worthwhile endeavour. Maxwell Taylor and Ethel Quayle (1994, p. 33) argued that ‘[e]xplanations of terrorism in terms of “personality” serve to push the explanations inside the individual – they offer accounts in terms of essentially unobservable and unmeasurable qualities of people’. A similar critique has been presented against psychodynamic explanations in general; such explanations are virtually impossible to confirm or falsify (Victoroff, 2005, p. 26). Echoes of early psychological studies are still visible in the post-9/11 research on the radicalisation of terrorism. Difficult family situations, childhood traumas, and failures that damage self-esteem, for example, come up in several accounts of risk factors for radicalisation. What is typical, as Daniela Pisoiu and Sandra Hain (2018, pp. 53–57) note, is that the familiar arguments now often appear in a ‘diluted’ form without making direct references to the concepts and theories upon which these studies were originally built. A common feature of the psychological literature on terrorism in the 1980s is that it deals primarily with what terrorists are like and much less with how individuals become terrorists. While it was broadly understood that becoming a terrorist is a process, few studies attempted to produce any models of that process. A significant exception is the personal pathway model presented by Erik D. Shaw (1986), which draws from a number of previously mentioned studies. The personal developmental path to terrorism included, according to him, four components: early socialisation processes (to values and views supported by parents), narcissistic injuries (events significantly affecting an individual’s self-esteem and self-image), escalatory events (e.g. confrontations with the police), and personal connections to terrorist group members. It is also one of the few publications from this time that explicitly outlines a potential at-risk population. According to Shaw (1986, p. 365): The personal pathway model suggests that terrorists come from a selected, at-risk population, who have suffered from early damage to their self-esteem: Their subsequent political activities may be consistent with the liberal social philosophies of their families, but may go beyond their perception of their contradiction in their family’s beliefs and lack of social action. Family political philosophies may also serve to sensitize these persons to the economic and political tensions inherent throughout modern society. As a group, they appear to have been unsuccessful in obtaining a desired traditional place in society, which has contributed to their frustration. Shaw also proposes that personal needs and group membership provide a more powerful motivation for joining terrorist groups than ideology. He also points out that, in addition to early socialisation, ideas and norms that support social action may be learned in later life. While psychological explanations dominated individual-level studies on terrorism until this point, it is also important to recognise that they were not the only avenues adopted for explaining 24

Before ‘radicalisation’

terrorism. Approaches that link terrorism to political and social grievances have been part of terrorism studies virtually from the beginning, for example, in the form of various analyses influenced by the relative deprivation theory of Ted Robert Gurr1 (e.g. Friedland, 1992; Corrado, 1981; Heskin, 1985; Wilkinson, 1974) and studies that link terrorism to the state’s actions and lack of political opportunities within the political system for new contenders (e.g. Bonanate, 1979). These analyses typically looked at terrorism mainly at the group or societal levels. Explanations for individual participation were usually sought from other directions, as it seemed difficult to explain it (solely) in these terms.

Studies on terrorist belief systems Towards the end of the 1980s, studies on terrorist personalities started to fade into the background, and other approaches for explaining (individual) participation in terrorism started to gain ground. What the emerging approaches in the 1980s and 1990s had in common was that rather than asking ‘what terrorists are like’, it was more productive to search for answers for ‘how individuals get involved in terrorism’. Not only was this question easier to study, but the most significant explanation for how individuals become terrorists may lie in the process of joining rather than in the qualities of an individual. The idea of terrorists as somehow special was largely abandoned, and studies operated with the assumption that terrorist behaviour can be explained with the same approaches and theories as any human behaviour (see, e.g., Taylor & Quayle, 1994). Involvement in terrorism became to be seen, to an important degree, as a product of social relations and the wider socio-political context. With the shifting research focus, researchers acquired a new kind of interest in documents and testimonies from those involved in terrorism. While such sources were examined earlier mainly to find clues about their personalities or psychological states, they were now analysed to understand how they viewed the world. The need for such research was underlined, for example, by Bonnie Cordes (1987). She called for paying more attention to how terrorists perceived themselves and their motivations and argued that researchers should listen carefully to what they said because ‘[t]errorist statements give us our best, and sometimes, only inside view of terrorist life and ­thinking within the group’ (Cordes, 1987, p. 150). In order to understand how individual terrorist groups functioned and what motivated individuals to remain involved, it was necessary to understand their ‘subjective reality’, in other words, how they interpreted the world around them (Della Porta, 1992; Crenshaw, 1992; Taylor & Quayle, 1994). This kind of research was facilitated by changes in public attitudes towards terrorism. The late 1980s and 1990s were contradictory periods in terrorism studies. The number of publications was in decline (Reid, 1997), but at the same time, many influential and academically ambitious studies were conducted. The declining number of publications most likely reflected the fact that, towards the end of the 1980s, public interest in terrorism started to wane. On the other hand, as an Italian scholar, Donatella della Porta, remarked in the early 1990s, ‘the end of the terrorism emergency’ after the peak years in the 1970s had ‘provided a more congenial atmosphere for research projects focusing precisely on the nature of the individual motivations in the underground’ (Della Porta, 1992, p. 4; Crenshaw, 2000, p. 407). This seems to be the experience of European scholars in particular. Terrorism was also a controversial topic in European countries, but there seemed to be somewhat more leeway (allegedly already before the late 1980s; see Kellen, 1979, p. 4) to analyse terrorism from different perspectives, including the motivations of those involved and sociological perspectives that studied the impact of various societal developments on terrorism. 25

Leena Malkki

In this light, it is hardly surprising that many of the studies that analysed terrorist groups in the wider societal context focused on groups that operated on European soil. Several contemporaneous studies on left-wing and national-separatist campaigns in Europe dealt with the role of wider political and social conditions in the emergence of terrorism (see, e.g., articles in Merkl [ed.], 1986). The studies discussed in this section mainly (though not solely) draw from the study of farleft terrorist groups in the 1970s and 1980s, especially those active in Italy, West Germany, and, to some degree, the United States. One question that was explored more elaborately than before was the nature and development of the belief systems of terrorist groups. Researchers pointed out that while terrorist groups had different objectives and beliefs, they appeared to share certain characteristics at a more general level. Martha Crenshaw (1988) noted that the belief systems of terrorists tend to be excessively abstract, highly moralistic, and characterised by black-and-white thinking. Terrorists present themselves as an elite of right-minded people, and the enemy is often dehumanised as thoroughly immoral and evil. There is no possibility of reconciliation and no middle ground (Della Porta, 1992, pp. 20–21). Terrorist belief systems have a tendency to become self-referential, inward-looking, and resistant to conflicting information. What contributes to the development of the belief system in this direction is that terrorist groups, as underground organisations, tend to be relatively isolated from their environment and require a high level of commitment from their members. As Della Porta (1992) notes, ‘the more isolated the groups are, the more abstract, ritualistic and inaccessible to factual arguments their ideologies become’ (p. 20). A crucial part of the belief system is the justification of violence. Besides the characterisation of the enemy as evil and inhumane, this is often accomplished by presenting the prevailing situation as a war in which the group is fighting an adversary guilty of injustice (Crenshaw, 1988; see also Ferracuti, 1982). Their argumentation often shows elements of the mechanisms of moral disengagement identified by Albert Bandura (1990a, 1990b). These psychosocial mechanisms help an individual overcome internalised moral standards that prohibit violent behaviour by reconstructing the act itself, its effects, or the victim in ways that make violence justifiable. While the belief systems of terrorist groups may have their own peculiarities, these studies pointed out that they do not develop in a vacuum. Instead, the researchers highlighted how these beliefs drew from the symbols and narratives available in their political and cultural contexts. They often share ideas and symbols with broader social movements from which they have developed. Moreover, Khachig Tololyan (1987, 1989), in particular, has underlined that terrorism should not be seen only as a political phenomenon because it also draws strongly from the surrounding cultural milieu. This is particularly the case with terrorist groups that have a clear connection to a wider (ethnic or religious) community. The terrorist group can tap into the project narrative of this community, which is a shared story that ‘not only tells a story of the past, but also maps out future actions that can imbue the time of individual lives with transcendent collective values’ (Tololyan, 1989, p. 101). This allows it to connect its own struggle to historical examples and thereby produce ‘new heroes for old stories’ (Tololyan, 1989, p. 117). What is typical for the aforementioned and several other studies in this period is that what is now called ‘cognitive radicalisation’ is not conceptualised as strongly through political ideology (or religion) as it is often nowadays. Instead, researchers more typically talk about belief systems, narratives, subjective realities, and identities. Political ideologies appear in these contributions as one element from which terrorist groups draw to justify their actions and construct their narratives. Moreover, rather than a process of adopting an extremist ideology, the cognitive change is conceptualised more typically as a moral one – an adoption of another value system that justifies violence.

26

Before ‘radicalisation’

A good example of a study that approaches cognitive change from this perspective is the process of delegitimation theory presented by Ehud Sprinzak (1991, 1995). It describes the lengthy psychopolitical process during which the moral views and identity of a group undergo changes, culminating in terrorist acts. Sprinzak calls this process ‘radicalisation’, but this is not his primary conceptualisation. Sprinzak sees terrorism as a ‘direct behavioural extension of non-terroristic opposition politics’ (Sprinzak, 1991, p. 50) and the process towards terrorism essentially as ‘a slowly evolving legitimacy crisis between an insurgent movement and the government’ (Sprinzak, 1995, p. 18). It develops through three stages (crisis of confidence, conflict of legitimacy, and crisis of legitimacy). How these stages manifest themselves is different when terrorism emerges within a new social or political movement that opposes the regime (transformational delegitimation) than when it happens in the context of an existing oppositional nationalist movement that struggles for independence (extensional delegitimation). Another version of the process can be found in ethnically, nationally, or religiously divided societies, where the radicalising group is ‘in conflict with two opposing entities: a community (or class) of people who are considered illegitimate, and a regime’ (Sprinzak, 1991, p. 52), which does not always become an object of delegitimation (split delegitimation). This type of delegitimation process is typical of far-right groups in particular (Sprinzak, 1995). Another analysis that resonates strongly with Sprinzak’s theory was presented by Mark Juergensmeyer. On the basis of his research on religiously motivated terrorism, he presented his model of satanisation about how the worldviews of those involved in religious terrorism develop towards acceptance of terrorism. This analysis is presented in the first edition of his highly regarded book, Terror in the Mind of God (Juergensmeyer, 2000). Collective belief systems of terrorist groups or larger radical subcultures have a significant impact on individual participants and their beliefs. However, the research from this period also recognised that not all individual participants develop similar belief systems (e.g. Taylor & Quayle, 1994, p. 38). Individual members can play different roles in the group. More specifically, the belief systems of those in leadership positions may be more complex, and their authority may be based on putting the implicit beliefs held by their followers into words (Crenshaw, 1988, p. 19). Moreover, the process of becoming involved in terrorism, including the development of one’s belief system, may be significantly different for the first-generation members who were part of the group when it moved to terrorism than for the later generations who joined a group already involved in it (e.g. Moyano, 1992; Wasmund, 1983; Wasmund, 1986).

Social movement studies on individual involvement An integral and important part of the shifting focus in terrorism studies at the turn of the 1980s and 1990s was the emergence of social movement studies-based approaches and theories on terrorism. This line of research was exemplified especially by Donatella della Porta’s publications (1992, 1995) and presents one of the most academically grounded and theoretically developed lines of research on the emergence of terrorism before the 2000s. These studies analyse the emergence of terrorism as a relational process, which means that it takes place through complex interactions between various actors (the social movement, its rival movements, the political system, and security authorities). The main focus is usually on the group level, but the studies build strongly on the notion that understanding the emergence, continuation, and decline of terrorism requires a multilevel analysis that also takes into account societal and individual-level developments.2 Analysing complex interactions provides a way to uncover how these different levels are linked to each other.

27

Leena Malkki

The most important contribution of social movement studies research to the understanding of individual participation in terrorism is its emphasis on social ties: the majority of those who get involved in terrorist groups do so together with their friends or relatives, or they know someone who joined earlier. This means that ‘to understand differential recruitment and escalation of commitment in terrorist groups as well as in social movement organizations, we must look not at the psychological characteristics of the activists but at their relational positions’ (Della Porta, 1992, p. 22). This observation was not entirely surprising. Studies on participation in social movements had already shown that social movements usually gain new members from the social networks of those already involved (Della Porta, 1992, p. 8). The significant role of social relations had already come up in studies on German terrorism published in the early 1980s (e.g. Jäger, 1981; Böllinger, 1981; Wasmund, 1983). This line of research also highlighted the role of interaction within subcultural communities (e.g. the West Berlin counterculture in the case of the Red Army Faction; see Wasmund, 1986) and more generally in peer groups constructed during adolescence (Della Porta, 1992). These were typically the contexts in which crucial social relations were formed and political beliefs were shaped. According to Della Porta, affective ties formed with other people in these contexts were an important motivation for continuing and deepening the commitment to political activities and eventually joining a terrorist group. Furthermore, social relations also shaped cognitive radicalisation (the transformation of belief systems) as it took place in social groups within a larger political counterculture (Della Porta, 1992). Involvement in countercultural networks is not the only environmental precondition that influences the decision to engage in violent action. Confrontations with the police and experiences of arrests and repression were also significant (Della Porta, 1992, pp. 15–16, 1995, pp. 55–82). The researchers also pointed out that there was significant variation in the social context in which terrorist groups developed. There were differences between the contexts of political socialisation as well as the relationship between the terrorist group, the radical milieu, the wider community, and society (Waldmann, 1992; Neidhardt, 1992). It is worth paying attention to the language used in these studies to discuss the process of individual involvement in terrorism. The term ‘radicalisation’ occasionally pops up as a generic term to denote the change of beliefs or behaviour towards a more radical direction, but it has no defined specific meaning. The ‘pre-terrorist’ phase of the process towards involvement in terrorism is conceptualised as political socialisation, while the actual joining of a terrorist group is typically referred to as recruitment or mobilisation. Many of the ideas included in early social movement studies on terrorism have been influential in how radicalisation has been discussed later in the research literature.3 This goes especially for the importance of social ties, which became widely acknowledged and highlighted with the publication of Marc Sageman’s book Understanding Terror Networks (2004), which provided further support for this finding. Through Sageman’s work, the important role that social ties and socialisation in small groups and networks play in the radicalisation process became known as the ‘bunch of guys’ theory. More generally, researchers have continued to build upon the work of Donatella della Porta and others and developed what has now become known as the relational approach to radicalisation (e.g. Alimi et al., 2015; Bosi et al., 2014; Della Porta, 2013; Malthaner, 2017). It has become one of the most important approaches to explaining individual radicalisation and the emergence of terrorism. While much of the early social movement research on terrorism is still relevant, there are also limitations. One significant limitation derives from the type of organisational structures that these studies dealt with. Like many other studies on terrorism before the 2000s, these studies focused on 28

Before ‘radicalisation’

organised groups that operated clandestinely and against the government of their country of operation. Since then, the development of communications technology, together with a number of other societal changes, has changed the way radical milieus emerge and function, as well as how those involved in terrorism organise. As Stefan Malthaner has rightly noted, social movement studiesbased approaches have significantly less to say about ‘cases in which radicalization takes place in dispersed clusters of activists and local milieus, loosely connected to a transnational movement, and oriented towards and enraged by armed conflicts in foreign countries which they observe from distance’ (Malthaner, 2017, p. 393).

Conclusion When research under the flag of ‘radicalisation’ emerged in the mid-2000s, it did not start with an empty plate. As we have seen in this chapter, many notions that have become common in radicalisation research have their roots in the terrorism studies literature in the earlier decades (see also Neumann  & Kleinmann, 2013). The role of mental health problems was one of the first issues discussed in relation to individual participation. The search for a terrorist profile or personality also started in the 1970s. That was also when scepticism towards that goal was voiced for the first time. The discussion about the link between terrorism and early socialisation dates back to the early 1980s. The significant role of social ties has also been well recognised, at least from the early 1990s onwards. These early studies are often not directly cited in the radicalisation literature. Instead, it appears that their conclusions are familiar to current researchers, mostly through a number of widely read studies and literature reviews (e.g. Victoroff, 2005; Sageman, 2004; Horgan, 2005). What cannot be found in the earlier literature is the concept of radicalisation. To be more precise, the word radicalisation can be found. It made occasional appearances in terrorism studies research, especially in the 1990s, but it did not have the same role and meaning as it has nowadays. It most typically appears as a general word to denote a change in ideas or behaviour in a more radical direction. It was more often used at the collective level than at the individual level, and it was hardly ever used to conceptualise the phenomenon under study or any of its dimensions.4 Most importantly, there was no overarching concept that would have been used to bring several dimensions of individual or collective involvement in terrorism together in the way that the term ‘radicalisation’ is used nowadays. Instead, studies used different terms to refer to various processes that now fall under the concept of radicalisation. How people get recruited into terrorist groups, how socialisation into terrorism happens, what motivates individuals to take part, how violence is legitimised, how terrorist belief systems develop, and how terrorist groups end up adopting violent methods – all these questions were treated as conceptually separate (although empirically related) issues. How exactly these different issues were conceptualised depended on the approach and focus of the study. When set against earlier conceptualisations, the introduction of the term ‘radicalisation’ has not only its benefits but also its downsides. The fact that the term has been so widely adopted shows that it is a useful general label for a phenomenon that has become the subject of broad public attention and policymaking. It is a go-to term to refer to everything that happens ‘left of bang’. The flexibility and fluidity of the term are partly what have made it such a useful term in the policymaking context. When it comes to academic research, it is less clear whether the adoption of the term can be seen as a conceptual advancement. Rather, it seems to have increased conceptual confusion and sometimes even paved the way for conceptual regression. The situation is not entirely dissimilar to what Paul Gill and Emily Corner have observed with regard to research on terrorism and mental 29

Leena Malkki

health. They claim that more recent research on this topic has tended to ignore that individuals have different roles in terrorist groups and that there may be psychological differences between individuals in these different roles – something that was better acknowledged and conceptualised in earlier studies on terrorism and mental health (Gill & Corner, 2017). In the same vein, earlier research on the causes of terrorism and individual involvement was more precise in its conceptualisations. With the policy relevance of explaining ‘radicalisation’, it appears that the appetite to respond to the need for sweeping and straightforward explanations has also further contributed to conceptual confusion.

Notes 1 This theory is introduced in Gurr (1970). In a survey among terrorism researchers conducted in the early 1980s, terrorism researchers were asked which theory they found helpful for understanding terrorism. Gurr’s work was mentioned most often (Schmid & Jongman, 1988, pp. 62–63). As Alex P. Schmid, who conducted the survey, notes, Gurr did not himself think his model could be used to explain terrorism in the 1970s (Schmid & Jongman, 1988, p. 68). For Gurr’s own take on analysing terrorism, see for example, Gurr (1988) and Gurr (1990). 2 There are also other contributions from this period that present multi-level models and frameworks for analysing terrorism. For a social-psychology-based framework for the study of terrorism, see Rabbie (1991). For a multi-level psychological-structural model for explaining the causes of terrorism, see Ross (1999). 3 Included in this body of literature is also one of the first studies on individual support for terrorism that involved a control group. Robert W. White (1992) studied why some residents of the Irish Republic supported the political violence of the IRA while others did not. What he found out was that supporting it required one to think that Ireland should be united. That was not, however, enough on its own. Supporters also typically had personal connections that made this objective and the related ideology relevant and familiar for them. In order for that to develop into support for the IRA, one also had to believe that political violence would be effective. 4 One of the very few instances in terrorism studies in which the term ‘radicalisation’ has been used in a more substantial manner before the 2000s is by Ted Robert Gurr. He claims that there are two main ways groups become willing to accept extreme means. He calls these radicalisation and reaction. ‘Radicalization refer to a process in which the group has been mobilised in pursuit of a social or political objective but has failed to make enough progress toward the objective to satisfy all activists’ (Gurr, 1990, p. 87).

References Alimi, E. Y., Demetriou, C., & Bosi, L. (2015). The dynamics of radicalization: A relational and comparative perspective. Oxford University Press. Bakker, E. (2006). Jihadi terrorists in Europe: Their characteristics and the circumstances in which they joined the jihad: An exploratory study. Netherlands Institute of International Relations Clingendael. Bakker, E. (2011). Characteristics of Jihadi terrorists in Europe. In R. Coolsaet (Ed.), Jihadi terrorism and the radicalisation challenge: European and American experiences (2nd ed., pp. 131–144). Ashgate. Bakker, E., & de Bont, R. (2016). Belgian and Dutch Jihadist foreign fighters (2012–2015): Characteristics, motivations, and roles in the War in Syria and Iraq. Small Wars & Insurgencies, 27(5), 837–857. Bandura, A. (1990a). Selective activation and disengagement of moral control. Journal of Social Issues, 46(1), 27–46. Bandura, A. (1990b). Mechanisms of moral disengagement. In W. Reich (Ed.), Origins of terrorism: Psychologies, ideologies, theologies, states of mind (pp. 161–191). Cambridge University Press. Böllinger, L. (1981). Die Entwicklung zu terroristischem Handeln als psychosozialer Prozess. In H. Jäger, G. Schmidtschen, & L. Süllwold (Eds.), Analysen zum Terrorismus 2: Lebenslausanalysen (pp. 175–231). Springer Fachmedien Wiesbaden. Bonanate, L. (1979). Some unanticipated consequences of terrorism. Journal of Peace Research, 16(3), 197–211.

30

Before ‘radicalisation’ Bosi, L., Demetriou, C., & Malthaner, S. (Eds.). (2014). Dynamics of political violence: A process-oriented perspective on radicalization and the escalation of political conflict. Routledge. Clarke, R. P. (1986). Patterns in the lives of ETA members. In P. H. Merkl (Ed.), Political violence and terror: Motifs and motivations (pp. 283–310). University of California Press. Cooper, H. H. A. (1977). What is a terrorist: A psychological perspective. Legal Medical Quarterly, 1(1), 16–32. Cooper, H. H. A. (1978). Psychopath as terrorists. Legal Medical Quarterly, 2(4), 253–262. Cordes, B. (1987). When terrorists do the talking: Reflections on terrorist literature. The Journal of Strategic Studies, 10(4), 150–171. Corrado, R. R. (1981). A critique of the mental disorder perspective of political terrorism. International Journal of Law and Psychiatry, 4, 293–309. Crenshaw, M. (1981). The causes of terrorism. Comparative Politics, 13(4), 379–399. Crenshaw, M. (1986). The psychology of terrorism. In M. G. Hermann (Ed.), Political psychology: Contemporary problems and issues (pp. 379–413). Jossey-Bass. Crenshaw, M. (1988). The subjective reality of the terrorists: Ideological and psychological factors in terrorism. In R. O. Slater & M. Stohl (Eds.), Current perspectives on international terrorism (pp. 12–46). MacMillan Press. Crenshaw, M. (1992). Decisions to use terrorism: Psychological constraints on instrumental reasoning. In D. della Porta (Ed.), International social movement research. Vol.4: Social movements and violence: Participation in underground organizations (pp. 29–42). JAI Press. Crenshaw, M. (2000). The psychology of terrorism: An agenda for the 21st century. Political Psychology, 21(2), 405–420. Crenshaw, M. (2011). Explaining terrorism: Causes, processes and consequences. Routledge. Crenshaw, M. (2019). Constructing the field of terrorism. In E. Chenoweth, R. English, A. Gofas, & S. N. Kalyvas (Eds.), The Oxford handbook of terrorism. Oxford University Press. Della Porta, D. (1992). Introduction: On individual motivations in underground political organizations. In D. della Porta (Ed.), International social movement research. Vol.4: Social movements and violence: Participation in underground organizations (pp. 3–28). JAI Press. Della Porta, D. (1995). Social movements, political violence, and the state: A comparative analysis of Italy and Germany. Cambridge University Press. Della Porta, D. (2013). Clandestine political violence. Cambridge University Press. Ferracuti, F. (1982). A sociopsychiatric interpretation of terrorism. Annals of the American Academy of Political and Social Science, 463, 129–140. Ferracuti, F., & Bruno, F. (1981). Psychiatric aspects of terrorism in Italy. In I. L. Barak Glantz & C. R. Huff (Eds.), The mad, the bad, the different: Essays in honor of Simon Dinitz (pp. 199–213). Lexington Books. Friedland, N. (1992). Becoming a terrorist: Social and individual antecedents. In L. Howard (Ed.), Terrorism: Roots, impacts, responses (pp. 81–93). Praeger. Gill, P., & Corner, E. (2017). There and back again: The study of mental disorder and terrorist involvement. American Psychologist, 72(3), 231–241. Gurr, T. D. (1970). Why men rebel. Princeton University Press. Gurr, T. D. (1988). Empirical research on political terrorism: The state of the art and how it might be improved. In R. O. Slater & M. Stohl (Eds.), Current perspectives on international terrorism (pp. 115–154). MacMillan Press. Gurr, T. D. (1990). Terrorism in democracies: Its social and political bases. In W. Reich (Ed.), Origins of terrorism: Psychologies, ideologies, theologies, states of mind (pp. 86–102). Cambridge University Press. Gustafsson, L., & Ranstorp, M. (2017). Swedish foreign fighters in Syria and Iraq: An analysis of opensource intelligence and statistical data. Swedish Defence University. www.diva-portal.org/smash/get/ diva2:1110355/FULLTEXT01.pdf Handler, J. S. (1990). Socioeconomic profile of an American terrorists: 1960s and 1970s. Terrorism, 13(3), 195–213. Heskin, K. (1985). Political violence in Northern Ireland. The Journal of Psychology, 119(5), 481–494. Horgan, J. (2005). The psychology of terrorism. Routledge. Horgan, J. (2014). The psychology of terrorism: Revised and updated second edition. Routledge. Jäger, H. (1981). Die individuelle Dimension terroristischen Handelns: Annäherungen an Einzelfälle. In H. Jäger, G. Schmidtschen, & L. Süllwold (Eds.), Analysen zum Terrorismus 2: Lebenslausanalysen (pp. 120–174). Springer Fachmedien Wiesbaden.

31

Leena Malkki Juergensmeyer, M. (2000). Terror in the mind of God: The global rise of religious violence: Updated editions with a new preface. University of California Press. Kellen, K. (1979). Terrorists – what are they like? How some terrorists describe their world and actions. A Rand note. RAND. Kellen, K. (1982). On terrorists and terrorism. A Rand note. RAND. Knutson, J. N. (1981). Political and psychodynamic pressures toward a negative identity: The case of an American revolutionary terrorist. In Y. Alexander & J. M. Gleason (Eds.), Behavioral and quantitative perspectives on terrorism (pp. 105–150). Pergamon Press. Malthaner, S. (2017). Radicalization: The evolution of an analytical paradigm. European Journal of Sociology/Archives Européennes de Sociologie, 58(3), 369–401. Marone, F., & Vidino, L. (2019). Destination Jihad: Italy’s foreign fighters (p. 10). The International Centre for Counter-Terrorism. https://icct.nl/app/uploads/2019/03/Marone-Vidino-Italys-Foreign-FightersMarch2019.pdf McCauley, C. R., & Segal, M. E. (1987). Social psychology of terrorist groups. In C. Hendrick (Ed.), Group processes and intergroup relations. Review of personality and social psychology 9 (pp. 231–256). SAGE Publications. Merari, A. (1991). Academic research and government policy on terrorism. Terrorism and Political Violence, 3(1), 88–102. Merkl, P. H. (Ed.). (1986). Political violence and terror: Motifs and motivations. University of California Press. Merkl, P. H. (1986). Approaches to the study of political violence. In P. H. Merkl (Ed.), Political violence and terror: Motifs and motivations (pp. 19–59). University of California Press. Moyano, M. J. (1992). Going underground in Argentina. A look at the founders of a guerrilla movement. In D. della Porta (Ed.), International social movement research. Vol.4: Social movements and violence: Participation in underground organizations (pp. 105–129). JAI Press. Neidhardt, F. (1992). Left-wing and right-wing terrorist groups: A comparison for the German case. In D. della Porta (Ed.), International social movement research. Vol.4: Social movements and violence: Participation in underground organizations (pp. 215–235). JAI Press. Neumann, P. R. (Ed.). (2015). Radicalization. Routledge. Neumann, P. R., & Kleinmann, S. (2013). How rigorous is radicalization research? Democracy and Security, 9(4), 360–382. Pearce, K. I. (1977). Police negotiations. A new role for the community psychiatrist. Canadian Psychiatric Association Journal, 22(4), 171–175. Pearlstein, R. (1991). The mind of the political terrorist. Scholarly Resources. Pisoiu, D., & Hain, S. (2018). Theories of terrorism: An introduction. Routledge. Post, J. M. (1984). Notes on a psychodynamic theory of terrorist behavior. Terrorism: An International Journal, 7(3), 241–256. Post, J. M. (1987). “It is us against them”: The group dynamics of political terrorism. Terrorism, 10(1), 23–35. Post, J. M. (1990). Terrorist psycho-logic: Terrorist behavior as a product of psychological forces. In W. Reich (Ed.), Origins of terrorism: Psychologies, ideologies, theologies, states of mind (pp. 25–40). Cambridge University Press. Rabbie, J. M. (1991). A behavioral interaction model: Toward a social-psychological framework for studying terrorism. Terrorism and Political Violence, 3(4), 134–163. Rasch, W. (1979). Psychological dimensions of political terrorism in the Federal Republic of Germany. International Journal of Law and Psychiatry, 2, 79–85. Reid, E. O. F. (1997). Evolution of a body of knowledge: An analysis of terrorism research. Information Processing & Management, 33(1), 91–106. Ross, J. I. (1999). Beyond the conceptualization of terrorism: A psychological-structural model of the causes of this activity. In C. Summers & E. Markusen (Eds.), Collective violence: Harmful behavior in groups and government (pp. 169–192). Rowman & Littlefield Publishers. Russell, C. A., & Miller, B. W. (1977). Profile of a terrorist. Terrorism: An International Journal, 1(1), 17–34. Sageman, M. (2004). Understanding terror networks. University of Pennsylvania Press. Schmid, A. P., & Jongman, A. J. (1988). Political terrorism: A new guide to actors, authors, concepts, data bases, theories & literature. Transaction Publishers.

32

Before ‘radicalisation’ Schmidtchen, G. (1981). Terroristische Karrieren. Soziologische Analyse anhand von Fahndungsunterlagen und Prozeβakten. In H. Jäger, G. Schmidtschen, & L. Süllwold (Eds.), Analysen zum Terrorismus 2: Lebenslausanalysen (pp. 14–77). Springer Fachmedien Wiesbaden. Shaw, E. D. (1986). Political terrorists: Dangers of diagnosis and an alternative to the psychopathological model. International Journal of Law and Psychiatry, 8, 359–368. Silke, A. (1998). Cheshire-cat logic: The recurring theme of terrorist abnormality in psychological research. Psychology, Crime, and Law, 4(1), 51–69. Silke, A., & Schmidt-Petersen, J. (2017). The golden age? What the 100 most cited articles in terrorism studies tell us. Terrorism and Political Violence, 29(4), 692–712. Smith, B. L., & Morgan, K. D. (1993). Terrorists right and left: Empirical issues in profiling American terrorists. Studies in Conflict & Terrorism, 17, 39–57. Sprinzak, E. (1991). The process of delegitimation: Towards a linkage theory of political terrorism. Terrorism and Political Violence, 3(1), 50–68. Sprinzak, E. (1995). Right-wing terrorism in a comparative perspective: The case of split delegitimization. Terrorism and Political Violence, 7(1). Stampnitzky, L. (2013). Disciplining terror: How experts invented “terrorism”. Cambridge University Press. Taylor, M., & Quayle, E. (1994). Terrorist lives. Brassey’s. Tololyan, K. (1987). Cultural narrative and the motivation of the terrorist. The Journal of Strategic Studies, 10(4), 217–233. Toloyan, K. (1989). Narrative culture and the motivation of the terrorist. In J. Shotter & K. J. Gergen (Eds.), Text of identity (pp. 99–118). Sage Publications. Victoroff, J. (2005). The mind of the terrorist. Journal of Conflict Resolution, 49(1), 3–42. Victoroff, J., & Adelman, J. (2012). Why do individuals resort to political violence? Approaches to the psychology of terrorism. In M. Breen-Smyth (Ed.), The Ashgate research companion to political violence (pp. 148–178). Routledge. Waldmann, P. (1992). Ethnic and sociorevolutionary terrorism: A comparison of structures. In D. della Porta (Ed.), International social movement research. Vol.4: Social movements and violence: Participation in underground organizations (pp. 237–257). JAI Press. Wasmund, K. (1983, Fall). The political socialization of terrorist groups in West Germany. Journal of Political and Military Sociology, 11, 223–239. Wasmund, K. (1986). The political socialization of West German terrorists. In P. H. Merkl (Ed.), Political violence and terror: Motifs and motivations (pp. 191–228). University of California Press. White, R. W. (1992). Political violence by the nonaggrieved: Explaining the political participation by those with no apparent grievances. In D. della Porta (Ed.), International social movement research. Vol.4: Social movements and violence: Participation in underground organizations (pp. 79–103). JAI Press. Wilkinson, P. (1974). Political terrorism. Palgrave.

33

3 ‘RADICALISATION’ AND ‘COUNTERING RADICALISATION’ The emergence and expansion of a contentious concept Rik Coolsaet Introduction ‘Radicalisation’ has become the primary prism for understanding and countering terrorism. To many, terrorism as the result of a ‘radicalisation process’ is an unquestioned axiom and considered self-evident. But the concept does not have a long pedigree, neither in academia nor in counterterrorism policies. In relation to terrorism, it only emerged after the 9/11 attacks and, in particular, after the string of attacks that struck Europe in 2004 and 2005. Contrary to the concept itself, however, the questions it seeks to answer are as old as terrorism: Why and how does one become a terrorist? What makes a terrorist? What motivates people to join terrorist groups? When do they embrace violence to advance political objectives? What is the trigger that makes an individual cross the threshold between holding extremist ideas and actually using violence? Has the new concept succeeded in giving better answers to the old questions? This will be the common thread throughout the chapter. The chapter will proceed as follows. First, it will reconstruct the emergence of the concept as a political construct and describe how and why academia and think tanks subsequently embraced the concept. The role of the European Union has been crucial throughout this process. Countering radicalisation rapidly became the holy grail of European counterterrorism. From the outset, however, scholars have been warning about its contentious nature as well as its potential for politicisation. The chapter will then turn to the key debates in radicalisation studies, which echo similar debates in the era of terrorism studies in the 1970s and 1980s. Inevitably, these debates had direct relevance for the elaboration of policies devised to counter radicalisation. Conceived in Europe, the radicalisation paradigm went global, in particular after the United States launched, in 2011, its own parallel concept, ‘CVE’, for ‘Countering Violent Extremism’, followed by the UN and its ‘PVE’, for ‘Preventing Violent Extremism’. Many countries adopted the new idioms, but not always with the same goal of addressing the environment that allowed radicalisation to emerge. Finally, the chapter will return to the initial question of whether the new concept of radicalisation has resulted in a better understanding terrorism and, in particular, individual motivations. And have the numerous de-radicalisation and CVE initiatives contributed to higher-performing counterterrorism policies? With hindsight, the new idioms have been less of a conceptual breakthrough DOI: 10.4324/9781003035848-4

34

‘Radicalisation’ and ‘countering radicalisation’

than initially anticipated, but progress has nevertheless been made in understanding how individuals get involved in terrorism. Countering radicalisation, on the other hand, still remains very much mired in ambiguity and work in progress.

A quick march through the European institutions ‘Radicalisation’ is commonly understood as the process through which individuals and groups get involved in terrorism. For more than a century, scholars and observers have been trying to apprehend what makes a terrorist. But ‘radicalisation’ was not part of their vocabulary. Nor did it belong to the toolkit of counterterrorism policies: No one talked of the IRA being radicalised, or Shining Path, or Black September or the Red Brigades. Though all of these older groups certainly were by our modern understanding. (Silke & Brown, 2016) If the word ‘radicalisation’ was used at all, it was generally in its generic sense, describing individuals or groups moving away from an assumed mainstream position in the pursuit of their goals. It was applied to ideas as well as to action. It was somehow synonymous with activism, protesting, and extremism. It was in that loose sense that some European intelligence services and police officers at the start of the 2000s used it to describe growing expressions of anger and frustration among young people in diasporic and Muslim communities in Europe (Europol, 2004, p. 30). The Dutch domestic intelligence agency Algemene Inlichtingen- en Veiligheidsdienst (AIVD, General Intelligence and Security Agency) was among the first to attempt to identify the root cause of this anger. In a public assessment in December 2002, the agency opined that many young men of Moroccan origin, aged between 18 and 31, born in the Netherlands or who had grown up there from early childhood, were vulnerable to indoctrination and manipulation by foreign extremists. Unsure about their identity and blaming Dutch society for not respecting who they were, they were said to be easy prey to recruiters. This report followed an incident whereby two Dutch young men of Moroccan origin travelled to Kashmir, where they were soon killed in combat. An investigation by the AIVD discovered that they had been recruited in the Netherlands by ‘radical Muslims’, who ‘spiritually prepared [them] to participate in the violent jihad’ (AIVD, 2002, pp. 5, 9). In the immediate post-9/11 period, Western intelligence services commonly depicted Al-Qaeda as a worldwide organised recruitment ring targeting vulnerable young people to join foreign war zones (AIVD, 2002; Europol, 2004; Sageman, 2014, p. 567). But the US invasion of Iraq in 2003 led to an unintended but dramatic change in the terrorism scene, transforming it from an external security challenge into a domestic threat. According to the official American rationale, the invasion of Iraq was about fighting the terrorists in Iraq and across the world so that Western countries did not have to face them at home. Instead, the Iraq war enhanced the bourgeoning anger among youngsters, as intelligence services and police officers in the field observed. It generated and mobilised a new generation of radicals, outraged by the invasion. Worldwide, thousands of volunteers travelled to Iraq (about 100 came from Europe). By adding a new layer of frustration within Muslim communities worldwide, it also pushed jihadi terrorism into a new dimension: it was rapidly becoming a decentralised, largely home-grown patchwork of scattered jihadi groups, linked by a common world view and opportunistic links, and recruitment was a rather spontaneous local process, developing through pre-existing kinship and friendship bonds (Pillar, 2004; COREU, 2004a, 2004c; Sageman, 2004, pp. 99–135; Coolsaet, 2007, p. 9; AIVD, 2007, p. 18; Malthaner, 2017). 35

Rik Coolsaet

The need to understand what was going on became all the more urgent when a string of terrorist attacks struck Europe. Unlike the 9/11 attacks, the Madrid train bombings of March 2004, the murder of the Dutch movie director Theo van Gogh in November, and the 7 July 2005 London bombings were not perpetrated by foreign terrorists or organised by Al-Qaeda. Most of the attackers originated from immigrant communities who grew up in Europe and were often born there. The attacks triggered a scramble for answers to the century-old question of why and how individuals came to resort to terrorism and turn against their own countrymen. Some European officials were convinced that this changing face of terrorism would put Europe’s social fabric at risk since such attacks boosted radical-right political parties, xenophobia, and nativism. As a result, European societies would be polarised between natives and migrant communities and between Muslims and non-Muslims. Officials at the European Commission and in the European Council Secretariat started actively pushing member states to address the ‘root causes’ of this anger among young people. Upstream prevention, through the identification of the underlying factors that could lead to (recruitment to) terrorism, became their leitmotiv. Two months after the Madrid attacks, the two standing counterterrorism committees of the European Union issued their first ever joint (confidential) assessment of the potential causes of this ‘anger’. They were wide-ranging (and not put in any specific order of priority, nor really operationalised): regional conflicts and failed or failing states (and the perception of Western double standards), globalisation and socioeconomic factors, alienation, propagation of an extremist worldview, and systems of education (madrasas). Addressing specifically the alienation of youngsters in Europe, the report stated: Within Europe, young Muslims may often feel themselves to be subject to discrimination and participate relatively little in mainstream politics and public life. In this context of sometimes real grievances, a lack of any real opportunities to effect change or vent frustration and a consequent sense of anger and helplessness, the unambiguous messages of extremist propaganda can become very attractive, particularly to the youth population. (COREU, 2004a, para 7) It was the very first time an official (albeit confidential) EU document mentioned ‘radicalisation’ in relation to terrorism. The report warned that the ‘processes by which individuals are drawn into terrorism [were] very complex. . . . Accordingly, generalisations must be viewed with great caution’: Recruitment to terrorist organisations is preceded, in many cases, by a process of radicalisation, where an individual may become attracted to extremist ideologies. There are a number of factors that may contribute to this, although the specific process of radicalisation can vary from individual to individual. (COREU, 2004a, para 4) In October 2004, echoing this assessment, the European Commission released its first public paper introducing the concept of ‘radicalisation’, accompanied by the qualification ‘violent’ (European Commission, 2004). Speaking of ‘disrupting the conditions facilitating the recruitment of terrorists’, it announced its intention to draw upon existing academic expertise to help sort out ‘where European policies and instruments can play a preventive role against violent radicalisation’. The two EU counterterrorism committees drew up a second joint report in November 2004 (COREU, 2004b). This one zoomed in on the personal trajectories of individuals and privileged 36

‘Radicalisation’ and ‘countering radicalisation’

ideological factors as the sources of eventual anger over the contextual factors that had been prominent in their first report. Focusing on the process an individual undergoes from his or her original ‘normal’ status to becoming a terrorist, the November report insisted that it tended to take place via loosely connected networks and individuals rather than through specific organisations dedicated to the task. The original emphasis on the role of an external recruiter actively taking the lead in the process of radicalisation had now given way to ‘endogenous factors (that) have an increasing influence on radicalisation’. While contextual factors were not entirely neglected – underlying factors such as social deprivation or lack of integration were indeed briefly mentioned – the emphasis was clearly on ideological factors as the key drivers of (self)radicalisation (COREU, 2004b, para A). The July 2005 bombings in London further boosted counterterrorism work and thinking at EU level. Radicalisation as a prism for understanding and countering terrorism was now rapidly gaining traction within EU counterterrorism thinking. But officials were aware that it was an extremely complex issue. The endeavour was deemed worth the effort, though: if one could understand the process individuals undergo in their slide towards terrorism, it was assumed, timely interventions could be devised to prevent this from happening. The rapidly expanding use of the concept of ‘violent radicalisation’ did, however, not imply that everybody thought along the same lines. Still, no univocally agreed definition or its causes were agreed upon. The first EU Counter-Terrorism Coordinator Gijs de Vries recalls that ‘at the time the EU Strategy was adopted [2005, RC] Member States did not agree about what constitutes radicalisation, or what causes it’: Faced with the lack of consensus [on radicalisation] among prevention analysts and practitioners, the EU has opted for open-ended formulations. The latest Commission document on radicalisation, for example, does not provide a definition of radicalisation. As to drivers, it notes that these ‘may include’ a strong sense of personal or cultural alienation, perceived injustice or humiliation reinforced by social marginalisation, xenophobia and discrimination, limited education or employment possibilities, criminality, political factors as well as an ideological and religious dimension, unstructured family ties, personal trauma and other psychological problems. The Commission is right: radicalisation may indeed include any or all of these factors. The trouble is that policy-makers and practitioners are left none the wiser. (De Vries, 2021, pp. 236–237) That bewildering array of topics under the umbrella of ‘radicalisation’ was aptly summarised by Peter Neumann, founder of the London-based International Centre for the Study of Radicalisation, as ‘what goes on before the bomb goes off’ (ICSR, 2008). In 2004–5, the European discussions on radicalisation thus went in all directions. Some Member States subscribed to the UK view that considered ideology the major driving force of radicalisation. This allowed authorities to set aside possible domestic drivers: ‘For politicians and many policy makers, ideology is a magical, formless, mostly-Islamic force that bends vulnerable minds toward violence’ (Berger, 2017). Most governments were adamant that the emphasis should be on preparedness and repression. But a small number of Member States, together with the Commission and the Council Secretariat, were adamant about addressing contextual factors as part of a long-term strategy for countering radicalisation and recruitment into terrorism, according to one of the participants in these discussions (Rietjens, 2007). In December 2005, the EU heads of state and government officially adopted the ‘EU CounterTerrorism Strategy’ (European Council, 2005a). It closely mirrored the structure of the UK’s previously adopted counter-terrorism strategy. It had taken some discussion among the Member States, 37

Rik Coolsaet

but by identifying ‘Prevent’ (‘preventing people turning to terrorism by tackling the factors or root causes which can lead to radicalisation and recruitment, in Europe and internationally’) as the primary strategic objective, the EU clearly stressed the preventive work that needed to be undertaken to successfully combat terrorism in the long term. As part and parcel of this new overall strategy, the European Council simultaneously adopted the ‘Strategy for Combating Radicalisation and Recruitment to Terrorism’, thus confirming that radicalisation had become one of the central threads in Europe’s ‘root cause’ approach to counterterrorism. This document called for a better understanding of the ‘motives behind such a decision [i.e. to become involved in terrorism]’ and for a way to ‘identify and counter the ways, propaganda and conditions through which people are drawn into terrorism and consider it a legitimate course of action’ (European Council, 2005b). The strategy acknowledged that within the Union, too, structural factors could exist that might create disaffection and susceptibility to the overtures of extremists, such as social and economic inequalities among relevant minority groups. By now, radicalisation had become the holy grail of European counterterrorism. It was commonly referred to in official documents and speeches. It was picked up in media reporting. It was taken for granted and considered self-evident. But it was not. With hindsight, one can indeed see how competing paradigms about the drivers of radicalisation were pulling the discussions in different directions without the participants really being aware of it. What was not clear at that time either was the divergence in policies that would flow from these competing paradigms. If contextual factors were of the essence, then upstream prevention through a whole-of-government effort had to be developed, encompassing complex societal issues such as integration, multiculturalism, the place of religion in Europe, and social cohesion, and stitching it all together in a broadened security agenda. If ideology – or rather an extremist interpretation of religion – was the crucial driver, countering radicalisation was essentially a war of ideas through the promotion of ‘moderate Islam’ (whatever that may mean) to combat the negative influence of ‘radical Islam’. If recruitment by foreigners was seen as the core issue, then counterradicalisation essentially rested with the security services. And if foreign policies were involved in the radicalisation process, then the effect of domestic policy changes would be marginal. The early debates on radicalisation had been largely limited to officials from the Commission, the Secretariat, and the EU member states, with little to no involvement of scholars. The results were not judged satisfactory. Commission officials realised that, contrary to the popular and public appeal of the new concept, not much was known about what happened in the black box labelled ‘radicalisation’ or what exactly made an environment conducive to radicalisation. They also acknowledged that the relationship between ideas and action, between radicalisation and terrorism, was poorly understood. Past studies on why, how, and when individuals became involved in terrorism had indeed never produced definite answers or operational policy options. In 2006, the Commission reached out to the scholarly community.

Academia joins in In April  2006, the European Commission set up the very first ‘Expert Group on Violent Radicalisation’. Its officials hoped to plug into the existing academic knowledge on radicalisation to bring some order to the chaotic discussions and to suggest policy advice for countering violent radicalisation. They defined violent radicalisation as ‘the phenomenon of people embracing opinions, views and ideas which could lead to acts of terrorism’. They were fully aware that this was ‘a very complex question with no simple answers and which requires a cautious, modest and wellthought [out] approach’ (European Commission, 2005). The Expert Group was given the mandate 38

‘Radicalisation’ and ‘countering radicalisation’

to establish a preliminary contribution on the state of knowledge on violent radicalisation. Originally planned for June 2006, it was only by mid-May 2008 that the final report was submitted to the Commission (Expert Group, 2008). The report confirmed the obvious: the concepts of ‘radicalisation’ and ‘violent radicalisation’ had originated in EU policy circles after the 2004 Madrid bombing, and had not been widely used in social science. Moreover, there was no uniform usage of the terms ‘radicalisation’ and ‘violent radicalisation’ in the academic literature. The experts also voiced some misgivings about the very concept. They considered the term ‘violent radicalisation’ potentially ‘misleading’ because ‘the socialisation process itself [leading to the use of violence] does not have to be ‘violent’ and even ‘socialisation into violence is not necessarily co-terminous with socialisation into terrorism’. ‘Radicalisation’, in turn, is problematic, they argued, ‘in that its relationship to ‘radicalism’ as an expression of legitimate political thought, still reflected in the titles of some political parties in Europe, is confusing’. The Expert Group also concluded that there ‘are . . . remarkable parallels between radicalisation to current jihadist terrorism and radicalisation to left-wing, right-wing or nationalist separatist terrorism’. Warning against its apparent simplicity and political exploitation, they offered an alternative wording: ‘socialisation to extremism which manifests itself in terrorism’. By the time the Expert Group’s report was submitted to the Commission, the first academic contributions on the topic had begun to appear. Scholars had been studying involvement in terrorism for decades, but the search for trajectories, individual traits, and specific profiles had been inconclusive (Kellen, 1979; Taylor, 1988; Horgan, 2008). While not referring to the novel concept of radicalisation that was being debated in the EU, the American forensic psychologist Randy Borum was the first to present a simplified model of a four-stage ‘process of ideological development’ that led to the emergence of a terrorist mindset. His model was intended to enable investigators to ‘better identify persons who represent desirable candidates or recruitment, . . . possible sites of indoctrination, . . . and extremists or groups that may use violent tactics’. Echoing similar warnings from the pioneers of terrorism studies in the 1970s and 1980s, he argued that ‘ideology may be a factor, but not necessarily the factor in determining motive’ (Borum, 2003). A more elaborate model was offered by the American psychologist Fathali M. Moghaddam in 2005. It became one of the most popular and influential representations of a staged process of ‘radicalisation’. Moghaddam offered the metaphor of an ever-narrowing staircase leading to the top of a building – terrorism. On the ground floor, perceptions of fairness and feelings of relative deprivation dominate. Step-by-step, an individual, ‘influenced by leaders’, eventually gets recruited to terrorist organisations and might reach the top floor, where they are ‘trained to sidestep inhibitory mechanisms that could prevent them from injuring and killing both others and themselves, and those selected are equipped and sent to carry out terrorist acts’ (Moghaddam, 2005). While not exactly a model, the Turkish-American scholar Zeyno Baran offered an appealing metaphor of non-violent Islamist organisations, whose ‘radical ideology’ acts as a ‘conveyor belt’ that encourages, seemingly irresistibly, ‘frustrated youth who have lost faith in the systems of the countries to which they or their parents came’ to become terrorists (Baran, 2005). In these and other early models and approaches, committed organisations played a key role in a gradual process of sequential steps through which individuals were lured into the use of violence. But the invasion of Iraq in 2003 and the subsequent mobilisation of a new generation of angry young men, willing to seek revenge for the invasion of yet another Muslim country, showed the limits of the existing models since no professional recruitment ring was involved. Radicalisation models grew in complexity, involving multi-factor dynamics with a considerable number of variables and mechanisms. 39

Rik Coolsaet

American scholar and former CIA case officer Marc Sageman confirmed earlier findings from the 1980s that individuals typically joined the global jihad through pre-existing social bonds that often preceded ideological commitment to the cause of jihadism (Sageman, 2004). He coined the term ‘bunch of guys’ to that effect. He defined radicalisation as the dynamics of four ‘prongs’, that interact in a non-linear way: moral outrage about the way Muslims worldwide are being treated; interpretation that this is part of a war on Islam, constituting a mobilising narrative; resonance of this perception with one’s personal experience of discrimination; and, finally, mobilisation of a small number of those who share this perception through networks of kin and friends (Sageman, 2008a). The American social psychologists Clark McCauley and Sophia Moskalenko in turn tried to solve a long-standing and tricky issue in the study of terrorism: how do radical ideas and radical action relate. Many share radical ideas, but only a handful cross the threshold to action. They suggested distinguishing between two kinds of radicalisation: radicalisation of opinion and radicalisation of action. The relationship between these two is weak since, according to them, radical ideology is not a ‘conveyor belt’ that mechanically moves an individual on to violent action. They came up with a complex set of mechanisms of radicalisation at the individual, group, and mass levels that could generate a wide variety of combinations, pushing individuals into a radicalisation pathway (McCauley & Moskalenko, 2008, 2011, 2017). The successive mobilisations of foreign volunteers following the invasion of Iraq and even more so in the Syrian civil war (with some 5,500 Europeans involved) offered a dramatic reallife test case for the radicalisation models. Research into the phenomenon of foreign fighters was revived (Malet, 2015, p. 2; Hegghammer, 2013; Duyvesteyn & Peeters, 2015). Simultaneous jihadi plots by small hubs and so-called lone-actor attacks alike, in particular in 2015–6, further triggered a huge surge in scholarly endeavours trying to come to grips with the dynamics of radicalisation. Radicalisation studies multiplied, widened, and deepened. Earlier assumptions that radicalisation always needed time to mature into action were challenged. A number of Syrian travellers and jihadi plotters indeed jumped literally from living a normal life or from drug trafficking and petty criminality into the jihadi scene without a protracted process of radicalisation. Other studies identified local environments as a key variable in radicalisation, a neglected dimension in terrorism and radicalisation studies (Rosenblatt, 2016; Dodwell et al., 2016). New research also strongly suggested that radicalisation was not necessarily a straightforward linear process. Personal trajectories or pathways were increasingly depicted as murky processes with ins and outs, just as in criminal gangs, neither fixed nor pre-determined but highly ‘individualised and non-linear, with a number of common ‘push’ and ‘pull’ factors but no single determining feature’ (Emmerson, 2016; see also Horgan, 2014). Explicitly rejecting models based upon an orderly sequence of steps that produce an output, the American scholars Mohammed Hafez and Creighton Mullins put forward an alternative metaphor of a ‘puzzle’, consisting of four pieces or factors that come together to produce violent radicalisation: personal and collective grievances; networks and interpersonal ties; political and religious ideologies; and enabling environments and support structures (Hafez & Mullins, 2015). Most leading scholars and practitioners in the field today view radicalisation broadly in the same way scholars in the 1980s viewed terrorism: as a complex interaction between personal characteristics, group dynamics, belief systems, and contextual factors. But beyond this broad but real consensus on the key variables that play a crucial role in radicalisation leading to terrorist violence, substantial conceptual disagreements exist, sometimes leading to febrile scholarly debates.

40

‘Radicalisation’ and ‘countering radicalisation’

Questioning radicalisation When advancing the concept of radicalisation in 2004–5, European officials readily acknowledged that the notion of ‘radicalisation’ was an oversimplification of an extremely complex phenomenon. The main reason for going along with this concept was that it allowed us to speak about ‘root causes’ in politically less contentious terms. ‘Root causes’ was indeed rapidly becoming a controversial idiom. After 9/11 (and until a reassessment of counterterrorism policies by the Obama administration in 2011), the United States in particular considered that speaking of ‘root causes’ implied condoning certain terrorist acts. ‘Roots was a taboo in the Bush administration, . . . with “evil” the only acceptable explanation for the attacks of Sept. 11’ (Gellman, 2002). Within the EU too, some officials were reluctant to use the words ‘root causes’. Not only did they notice that within the United Nations, this was indeed used to condone cases of political violence. They also feared that it could be hijacked by radicals in the Basque region or Northern Ireland to justify their terrorist tactics. They therefore chose to rather speak about ‘conditions conducive to the spread of terrorism’ and about ‘radicalisation’. The latter was gradually embraced as a pithy shorthand for the former and, in particular, as a less politically charged word to discuss ‘root causes’ of terrorism. But ‘radicalisation’ was a source of ambiguity and confusion from the start, not unlike the word ‘terrorism’ itself in the 1970s and 1980s, when a lot of scholarly energy was spent on how to define ‘terrorism’. While the scope of radicalisation studies rapidly expanded, involving a widening array of academic disciplines and a new generation of terrorism scholars, warnings about the ill-defined, ambiguous, and controversial nature of the very notion of radicalisation also persisted. In 2006, the Commission’s own Expert Group had already cautioned against the ambiguity of the concept. It soon became apparent that academic definitions of radicalisation indeed varied as much as official ones and that not all were up to academic standards (Schmid, 2013a). Because of its intrinsic ambiguities, no obvious metrics are available that objectively gauge radicalisation, notwithstanding numerous endeavours in academia, police, and policy circles. A number of scholars, including the Irish-born foremost expert on the psychology of terrorism, John Horgan, also highlighted early on the politicised nature of the term (Schmid, 2013a, 17). According to some, ‘radicalisation’ was instrumentalised from the start. Reflecting on the UK’s counter-radicalisation programme, called ‘Prevent’, launched by the Blair government in the wake of the July 2005 bombings, Arun Kundnani, former editor of Race and Class, equated it with ‘mass surveillance of Muslim populations’ (Kundnani, 2012, p. 19; see also Thomas, 2010; Lambert, 2011). Some scholars, including John Horgan and Marc Sageman, even suggested skipping the concept of radicalisation altogether (Schmid, 2013b). Most scholars do not go that far. Calling radicalisation ‘one of the great buzzwords of our time’ (Neumann & Kleinmann, 2013), terrorism scholar Peter Neumann nevertheless maintains that something called radicalisation does exist, but that ‘its meaning is ambiguous, and all the major controversies and debates that have sprung from it are linked to the same inherent ambiguity’ (Neumann, 2013). The scholarly debates indeed reveal several interrelated conceptual fault lines, not unlike the competing paradigms that were at play in the early debates among EU officials, and for the same reasons: the absence of a consensus definition of radicalisation and diverging hypotheses about the nature of its drivers.

Debating radicalisation The emergence of the radicalisation paradigm has led to a tremendous increase in terrorism-related research. Scholarly energy devoted to the issue far exceeds that of the earlier heyday of terrorism

41

Rik Coolsaet

studies in the 1970s and 1980s. But opinions are divided on the result of these efforts. Some scholars are sceptical that much progress has been made at all and argue that terrorism research has basically stagnated (Sageman, 2014). Others dispute this assessment as overly pessimistic and are more positive ‘because we feel we have ourselves learned a number of things about terrorism since 9/11’ (McCauley & Moskalenko, 2014). But even those who claim that real progress has been made in understanding how individuals get involved in terrorism readily acknowledge that gaps continue to exist and that a number of issues remain subject to debate and sometimes even frenzy among academics. Several major conceptual ‘fault-lines’ (Neumann, 2013, p. 873) can be identified. Context versus ideology as the main driver of radicalisation is the first major fault line in radicalisation studies. Many standard definitions of radicalisation, official as well as academic, define it as a process of gradually embracing radical and extremist ideas that act as a conveyor belt and predictor of terrorist violence. As a corollary, research has often emphasised the micro level of the individual to the detriment of the macro level of contextual factors (Sedgwick, 2010, pp. 480–481). In a sense, this de-contextualisation has been a step backwards compared to the broad consensus among the leading terrorism scholars of the 1980s that individual involvement in terrorism never happens in a vacuum. While many scholars of radicalisation do acknowledge that contextual factors are important, the sheer magnitude of the potential causes that have been proposed and their intrinsic political nature have prevented a consensus on this issue. Some scholars have pointed to ‘robust findings that group-based feelings of injustice reliably predict collective action’ (King & Taylor, 2011, p. 610) and suggested applying these findings from research on collective action to the study of terrorism. Still other scholars point at macro-structures, such as globalisation or resistance to modernisation, failed states, or the absence of freedom and the rule of law. Some link the emergence of radicalisation and terrorism to Western colonialism (Burgat, 2019), population growth, and the existence of a youth bulge (Urdal, 2006). Other scholars, however, dispute such sweeping assessments and insist that they cannot account for the limited number of individuals involved in acts of terrorism. The same question was already raised in the 1980s (Crenshaw, 1981, p. 389). Fraught with methodological difficulties and confronted with a seemingly endless amount of contextual conditions and factors to be taken into consideration, scholars at the time pointed out that the ‘root cause’ approach was promising on a theoretical level but questioned its feasibility on a practical level (Van de Voorde, 2011, p. 49; Wilkinson, 1974, p. 126). Many therefore preferred to approach terrorism in more pragmatic ways and opted for empirical, policy-oriented studies or crisis management analyses, resulting in direct and more useful, applied knowledge (Jenkins, 1999, pp. viii, xi; Jenkins, 2001). This observation remains valid today. It helps to understand why the study of individual pathways and trajectories has become prevalent in radicalisation studies. This offered a seemingly more practical and promising approach to understanding involvement in terrorism, as EU officials too had concluded (Bjørgo, 2005, pp. 5–6). Most scholars subscribe to the idea that radicalisation is the result of many variables. But weighing the relative importance of these variables represents the second fault line in scholarly debates. Two variables vie for pre-eminence as the main driver of radicalisation, and both have received a lot of academic attention: the role of ideology and the influence of kinship and friendship bonds. Few scholars will dispute that ideology is an important variable. It can cover ideas, belief systems, a fixed set of axioms, and narratives (hence, some ambiguity also exists as to what exactly is meant by ideology). But exactly how important these are is a matter of debate. To some, it 42

‘Radicalisation’ and ‘countering radicalisation’

represents the key driver of radicalisation (Kepel, 2015; Dawson, 2018, p. 99), while others argue that it serves a variety of different purposes, depending very much on the individuals involved: Ideology is relevant, though poorly understood. Some individuals embrace an ideology before joining the group while others adopt ideology afterwards. Some espouse the group’s ideology when preparing for an actual attack or because they need to justify their actions to themselves or to others after the fact. Others seek out an ideology they’ve decided they want to do something. For those who become involved in violent extremist groups, . . ., acquiring the ideology, the language, and the mindset happens further along the path – they are just some of the acquired qualities of engagement as opposed to prerequisites for initial involvement in the first place. (Horgan, 2019, p. 3) There also exists a strong body of scholarship – reaching back to research on involvement in neoNazi groups in the 1990s (Bjørgo, 2002) – that suggests mobilisation via pre-existing social bonds is a much better gauge to understand how individuals get involved in terrorist acts than ideology (Bjørgo & Horgan, 2009; Roy, 2015, p. 8). It earned its credentials through the earlier social movement theories (Malthaner, 2017), but it was definitely introduced in the field of radicalisation studies by Marc Sageman (2004, 2008b). These academic discussions on the respective importance of ideology and social bonds are part of a broader – an older – debate on the exact nature of the relationship between idea and action and, as a corollary, on the observation that many carry radical ideas but few act on them. In radicalisation studies, the former is labelled ‘cognitive radicalisation’ while the latter goes by the concept of ‘behavioral radicalisation’. While many official de-radicalisation strategies still use some form of conveyor belt model whereby ideas lead to action, most scholars will reject this crude metaphor and distinguish between developing extremist ideas and engaging in terrorism (Borum, 2011, p. 30; McCauley & Moskalenko, 2017). A third conceptual fault line in radicalisation studies is closely related to both previous ones. Is Islam a key factor in radicalisation or – more broadly – is involvement in religious terrorism inherently different from involvement in non-religious, political terrorism? Before 9/11, terrorism in the name of a religion was largely absent from terrorism research. In one of the few studies on the subject, RAND scholar Bruce Hoffman explicitly stated that religious terrorism was driven by value systems and worldviews that were radically different from those of secular terrorists. As a result, violence was more unrestrained and indiscriminate (Hoffman, 1993, 1998, pp. 94–95). Following 9/11 and at least until the 2010s, when new forms of violence-prone extremism started to emerge, terrorism and radicalisation research had a strong Islam bias. To some, this implied that only scholars with a solid knowledge of Islam could authoritatively address this form of terrorism (Kaplan, 2016, p. 9). The least one can say is that most western scholars had to adapt (and still do) to unfamiliar concepts, more or less related to Islam. This has reinforced the already ambiguous and contentious nature of the concept of radicalisation: When applied to Islam and Muslims, the term radical is often being used interchangeably and opaquely with terms such as fundamentalist, Islamist, Jihadist and neo-Salafist or Wahabbist with little regard for what these terms actually mean, and instead indicate signals about political Islam that these members of the media and politicians wish to transmit. (Githens-Mazer, 2010, p. 8) 43

Rik Coolsaet

The conceptual debates on the role of Islam in radicalisation coincide with the ‘old’ versus ‘new’ terrorism debate. At one end of the spectrum, there are scholars who subscribe to Hoffman’s claim that religion and thus jihadi terrorism represent a significant difference compared to politically motivated terrorism (Juergensmeyer, 2000a, 2000b; Simon & Benjamin, 2000; Dawson, 2017). On the other side of the spectrum, scholars dismiss the novelty thesis, pointing out essential continuities with previous expressions of terrorist violence (Duyvesteyn, 2004; Crenshaw, 2011) or arguing that ‘there is no important practical difference between terrorism on behalf of political ideology and that on behalf of religion’ (Fettweis, 2009, p. 277; Stern, 2003; Expert Group, 2008). When the issue of foreign fighters became a top research concern in 2012–3, the debate between the proponents of both schools sharpened. To some, religion was a key factor in the mobilisation of thousands of volunteers, while others concluded that religion only played a minor role compared to other variables, such as frustration, a lack of perspectives, or the characteristics of the local environment (Aly & Striegher, 2012; Weggemans et al., 2014). The same fault line is at play in the discussion about whether these volunteers originated from a youth subculture at the fringes of Muslim communities and organisations or, on the contrary, constituted the vanguard of a radicalising Umma. According to the latter, the world is facing a ‘global Islamist insurgency’ (Kilcullen, 2005; Lia, 2016), in which non-violent Islamist movements are pursuing the same goals as jihadists, that is, the creation of an Islamic state, and should be countered with the same determination. A less-dramatic portrayal of jihadism positions that rather than an irresistible global uprising, it represents a movement (the French word ‘mouvance’ is a more adequate phrasing) that is riven by internal disputes on ideology, tactics, and leadership and very much dependent upon local circumstances that allow it to prosper – or not. In most countries where it manifests itself, it can hardly count on the support of more than a tiny slice of the population (Arab Opinion, 2015, 2020; Molenbeek, 2017). In the wake of the Paris terrorist attacks in 2015, these divergent scholarly opinions reached a climax in a heated and still ongoing debate between two of the leading French scholars of Islam and jihadism, Gilles Kepel and Olivier Roy. The former considers Salafism the cradle of jihadism and jihadi terrorists the result of Islamist-dominated enclaves in France where individuals are groomed to become terrorists. The latter, on the contrary, sees jihadism as a generational revolt, not unlike the Italian Red Brigades or the German RAF of the 1970s and 1980s. According to Roy, the current generation of jihadi terrorists did not go through the Salafist funnel nor did they originate from Salafist enclaves, so no link of causality can be established between Salafism and jihadism. The former thus sees the radicalisation of Islam as the core ingredient, warning of a civil war that threatens France and the West. The latter invokes an Islamisation of radicalism, thus relegating religion to the fringes of jihadi terrorism (Roy, 2016, p. 15). Looking back on the evolution of radicalisation studies since 2004–5, we see a thriving field of research with an increasing number of scholarly disciplines involved and a widening research agenda. Themes keep adding up: gender, mental health and psychopathology, the Internet and social media, culture, the crime-terror nexus, migration, and discourse analyses. The concept has gone global. With all its ambiguities and caveats, it has been embraced by scholars worldwide, moving beyond the original Western-dominated pool of experts (English, 2016, pp. 145–146; Alava et al., 2017; Jones, 2018; Rink & Sharma, 2018; Timbuktu Institute, 2021; Bencherif et al., 2022; Smith et al., 2022). Since the late 2010s, scholars have started to address the applicability of the existing knowledge on radicalisation to other forms of radicalisation (RAN, 2016, p. 5). Confronted with the rise of a ‘melting pot’ of potentially violent extremisms (Coolsaet & Renard, 2022), how relevant are the findings of radicalisation studies, construed against the background of jihadi terrorism? This 44

‘Radicalisation’ and ‘countering radicalisation’

issue represents both a challenge to the validity of existing knowledge on radicalisation and an opportunity to strengthen its academic moorings.

Countering radicalisation In 2004–5, EU officials added ‘radicalisation’ to the toolkit of counterterrorism. Upstream prevention, so it was envisaged, would tackle the factors underlying the (self)recruitment of individuals into terrorism and complement the traditional counterterrorism instruments of law enforcement and intelligence gathering. Counterterrorism would thus become a long-term strategy aimed at obtaining sustainable results, something that could not be achieved by purely reactive and repressive measures. Did the results live up to expectations? Devising effective preventive policies proved to be more difficult than initially thought. The United Kingdom was among the first European member states to adopt this new dimension in 2003 as part of its overall counterterrorism strategy. But ‘Prevent’ was the least developed strand of its CT policy (CONTEST, 2009, p. 82). Moreover, from the start, competing counter-radicalisation models clashed. While tackling ideology was privileged as the main objective of Prevent, deep disagreement existed as to whether authorities could partner with Salafis and Islamists to tackle Al-Qaeda influence in the United Kingdom (Lambert, 2011, p. 216). In addition, criticisms of stigmatising and spying on Muslim communities made Prevent a toxic programme from the outset. Instead of fostering community cohesion, Prevent was said to further alienate Muslims from the wider society and from authorities (Kundnani, 2009; Mandaville, 2017, p. 5). What happened to the UK’s Prevent was a paradigmatic example of the difficulty for policymakers worldwide to come up with policies that were commensurate with the complexity of the issue. The United States long considered radicalisation to be a quintessential European problem, opposing the so-called European marginalisation of their Muslim communities with the successful American integration of Muslims, often living in affluent suburbs rather than poor ethnic enclaves as they did in Europe (Vidino, 2010, p. 2). But in 2009, young Americans of Somali descent living in Minneapolis went on a suicide mission in Somalia. They were soon followed by a group of American Muslims from Virginia who travelled to Pakistan, supposedly to join the Taliban. This came as a shock to the American counterterrorism community and led, in June 2010, to the organisation of the very first EU-US high-level encounter on radicalisation on the premises of the European Commission. In order to distinguish itself from George Bush’s Global War on Terror, the Obama Administration promoted the concepts of ‘Violent Extremism’ and ‘Countering Violent Extremism’ (CVE). ‘CVE’ was launched in August 2011 (White House, 2011) and became a landmark signature of Obama’s counterterrorism approach, soon adopted by governments worldwide. It essentially covered the same ground as Europe’s counter-radicalisation: ‘The shift in aim was to prevention and resilience primarily at home, shaping individual attitudes to resist the ideological and emotional attractions of appeals to violence rather than crushing the actors behind it’ (Crenshaw & LaFree, 2017, p. 175). Making environments non-permissive for terrorists seeking to exploit them was a long-term challenge that could not be met by even the best intelligence operations and law enforcement efforts, as the American administration now also acknowledges. In December 2015, the United Nations introduced its own acronym ‘PVE’, Preventing Violent Extremism, which basically covers the same preventive, non-coercive corollary to counterterrorism as the American CVE and the European counter-radicalisation (U.N., 2015; Stephens & Sieckelinck, 2019). By then, the unprecedented mobilisation of foreign fighters for the conflict in Syria had given a new impulse to prevention programmes worldwide (Renard, 2022, p. 24). 45

Rik Coolsaet

In Europe, many countries have set up a wide array of national and local de-radicalisation programmes. These programmes often aim at creating, sometimes uneasy, partnerships between law enforcement, police, social services, and, in some countries, civil society organisations. Best practices are being shared between practitioners, and dedicated networks have been set up, with the EU-sponsored Radicalisation Awareness Network (RAN) being the best-known example. New risk assessment tools have been developed for use within the police, prisons, and social services that accompany individuals ‘at risk’ of radicalisation. Reflecting the ambiguity and multifaceted nature of the concept of radicalisation, countering radicalisation in some European countries accordingly resembles a potpourri of objectives and programmes of all kinds, for which it is not easy to assess what works and what doesn’t. In other countries, such as the United Kingdom and France, while not totally excluding contextual factors, ideology is the primary objective in counter-radicalisation strategies (Bonelli & Ragazzi, 2019, p. 121). Still other European countries have not invested significantly in countering radicalisation and violent extremism (see Ragazzi & Walmsey in this volume). At the height of the ISIS mobilisation, the American CVE programme was an ambitious, expansive ‘whole-of-government’ programme, covering many departments and agencies, and establishing local pilot programmes for community outreach. Homeland Security had to establish an interagency CVE task force to coordinate the multitude of counter-radicalisation initiatives (Crenshaw & LaFree, 2017, pp. 29, 136). But confronted with the complexity of the issue and growing political resistance, national agencies whittled down their CVE workload to a bare minimum, and only sporadic local programmes remained in place, concentrating on one-on-one interventions with individuals already on the radar of law enforcement (Clifford & Hughes, 2022). In the Global South too, numerous de-radicalisation programmes have been set up, such as in Saudi Arabia, Malaysia, and Singapore. Here, they place a heavy emphasis on the religious dimension since authorities consider radicalisation to be a deviation from righteous religious beliefs. In the case of the Saudi programme, its effectiveness is questioned by many possibly due to its links to the military (LaFree & Freilich, 2019, p. 11). In some countries, civil society organisations have long played a key role in bridge-building efforts at the local level and are pursuing this effort under the new idioms. In still other cases, authorities have adopted the new concepts, but not with the expressed intention of addressing the environment that allows radicalisation to emerge (Renard, 2019; Pinfari, 2020; Trauthig, 2021). Also, some did so with the hope of receiving Washington’s support in their policy of criminalising their opposition (Hawkins, 2015; Emmerson, 2015, p. 4). Countering radicalisation is still very much a very young area of research and policymaking, and thus work is in the making. Just like the literature on the effectiveness of counterterrorism itself, research on countering radicalisation remains significantly under-theoretised (Renard, 2022, p. 135). By their very nature, all prevention policies are hard to evaluate through evidence-based assessments (see also Busher & Fisher in this volume). The jury is thus still out as to whether counterradicalisation programmes actually work. As the German scholar Daniel Koehler, editor of the only dedicated academic journal on the subject, Journal for Deradicalization, explains, there are different types of intervention programmes with different characteristics and performance expectations (Koehler, 2017). As will become evident in Sections 3 and 4 of this handbook, programmes vary according to who is eligible to participate, who the instructors and other staff are, the curriculum’s content, the tools employed, and the goals and desired outcomes (LaFree & Freilich, 2019, p. 11). As a result, they are hard to compare.

46

‘Radicalisation’ and ‘countering radicalisation’

But significant efforts are being undertaken to systematically assess the effectiveness of P/ CVE programmes (Colaert, 2017; Pistone et al., 2019; Nickolson et al., 2021; CPN-PREV). Some scholars, however, remain sceptical: That we do not know precisely what we are preventing, let alone knowing how or whether we might have prevented it, does not make for a bright future. (Horgan, 2014, p. 10)

In the end, what have we learnt? The emergence of a new wave of terrorism and the introduction of the concept of radicalisation in the toolkit of counterterrorism policies and research have set in motion an avalanche of public money, many times superior to the amounts available in the earlier terrorism era of the 1970s and 1980s. The irresistible success of the new concept had less to do with its intrinsic value than with the social and political environment in which it was embraced. It made it possible to put a label on the unfamiliar phenomenon of political Islam. Reflecting on the status the concept of radicalisation had gained over the years, Alex Schmid, former Officer-in-Charge of the Vienna-based Terrorism Prevention Branch of the United Nations and a lifelong scholar on terrorism, concluded: [W]e have to admit that in the final analysis, ‘radicalisation’ is not just a socio-psychological scientific concept but also a political construct, introduced into the public and academic debate mainly by national security establishments faced with political Islam in general and Salafist Jihadism in particular. The concept was ‘pushed’ to highlight a relatively narrow, micro-level set of problems related to the causes of terrorism that Western governments faced in their efforts to counter predominantly ‘home-grown’ terrorism from second and third generation members of Muslim diasporas. (Schmid, 2013a, p. 19) The absence of an agreed-upon understanding of what radicalisation turned it into a catch-all concept. The post-9/11 political and social environment ultimately allowed for the new concept of radicalisation to take centre stage in counterterrorism thinking, first in Europe and then worldwide. It got embroiled in the concerns over immigration, integration, and multiculturalism, as well as the unease over Islam and Muslims, that had all been growing steadily since the 1980s, particularly in Europe. Right-wing pundits and movements in particular had been eager to exploit this environment to cultivate divisive identity politics (UNDP, 2004, pp. 1–2, 76). Radicalisation made it possible to speak about these issues in a seemingly sanitised way that differed from the anti-Islam rhetoric of the far right. Embraced by the polity, the media, and academia, it became a self-sustaining concept. New departments and think tanks have been created. New jobs and new career opportunities have been made available. If so much money and effort are being spent on it and so many researchers are working on it, then radicalisation has to be a topic worth scholarly and political attention. Has this resulted in a corresponding breakthrough in our understanding of terrorism and political violence, going beyond the answers provided by the earlier generation of terrorism scholars? Most scholars and first-line practitioners acknowledge that radicalisation is a complex and multifaceted phenomenon that defies easy answers and recipes. This is not unlike the broad consensus

47

Rik Coolsaet

on terrorism among scholars in the 1970s and 1980s. Moreover, radicalisation and violent extremism are as contentious concepts as terrorism and as easily politicised. Finally, it is important to keep in mind that the way radicalisation is perceived has direct relevance for the elaboration of policies intended to counter it. Today’s debates on radicalisation are not so different from yesterday’s debates on terrorism: the difficulty of operationalising root causes; the correlation between beliefs and violence; the nature of the threat. One might mention other issues that resurface with every terrorism wave, such as the ever-returning question of mental health. As a result, predicting who will become a terrorist today remains as elusive a goal as it was back then. With hindsight, the concept of radicalisation has not been a conceptual breakthrough. It has been less helpful at explaining terrorism and, in particular, individual motivations than its European advocates anticipated in 2004, in part because of its over-reliance on micro-level research. Nevertheless, scholarly knowledge on why and how individuals engage in terrorism has undoubtedly improved. The myriad radicalisation studies produced since 2004 have confirmed that involvement in terrorism in most cases indeed results from socialisation and mobilisation dynamics (Schmid, 2013a), in which interpersonal ties are often more important than ideology (even if heated discussions are ongoing on the exact role ideology plays in this process). Radicalisation is thus habitually tied up with local conditions. Contrary to the earlier mentioned pessimistic assessments in 2014, our understanding of what happens to individuals once they get involved in such a dynamic has undeniably deepened. This can result in terrorism in a way that is quite similar to other forms of socialisation into deviant behaviour, like criminal gangs, religious sects, and youth street gangs. Moreover, scholars and first-line practitioners have confidently assessed that not all individuals follow the same trajectory into violence. It does not necessarily resemble a logical and staged process of successive steps. There are many individual trajectories involving different sets of factors, dissimilar timelines, and diverse triggers. As a result, countering radicalisation turned out to be trial and error with scant evidence of the effectiveness of specific interventions and policies – beyond the acknowledgement that a single one-size-fits-all programme to stem an individual’s slide into extremism and violence will not work. National characteristics, local conditions, and personal trajectories are simply too diverse. A final word on the issue of contextual and structural drivers – the weak spot in both terrorism and radicalisation studies. It will always remain a complex issue to relate individual motivations and behaviour to context. But this shouldn’t discourage scholars from investing in the exploration of why and when an environment becomes conducive to radicalisation. Terrorism is never an automatic reaction to conditions (Crenshaw, 1981, p. 389). Neither is radicalisation. The accumulated knowledge on terrorism and radicalisation over the decades gives credence to the assessment that the decision to perpetrate a terrorist attack or to join a terrorist group is as much an individual decision as the compound effect of a conducive environment, group dynamics in which personal trajectories run their course, and a mobilising narrative. But for terrorism to emerge and take hold, one additional dimension is required, something more intangible: the feeling that an historical moment has arrived (Crenshaw, 1981, p. 388; Coolsaet & Renard, 2022). When terrorists and terrorist organisations are able to proclaim that history is on their side and that their tactics present a unique opportunity to radically change existing conditions, then terrorism becomes a credible and enticing offer for action. The interplay of all these dimensions will always be hard to quantify or mould into a model. But all of them have nevertheless to be present for terrorism to emerge, and for explaining why and how individuals become terrorists. 48

‘Radicalisation’ and ‘countering radicalisation’

References AIVD. (2002, December). Recruitment for the Jihad in the Netherlands. AIVD. AIVD. (2007, October). The radical dawa in transition. AIVD. Alava, S., Frau-Meigs, D., & Hassan, G. (2017). Youth and violent extremism on social media. Mapping the research. UNESCO. https://unesdoc.unesco.org/ark:/48223/pf0000260382 Aly, A., & Striegher, J. (2012). Examining the role of religion in radicalisation to violent extremism. Studies in Conflict and Terrorism, 35(12), 849–962. Arab Opinion. (2015). Arab opinion index. Arab Research & Policy Studies. www.dohainstitute.org/en/lists/ ACRPS-PDFDocumentLibrary/Arab_Opinion_Index_2015_Results_in_Brief_REVISED_JUNE_2016.pdf Arab Opinion. (2020, June). The 2019–2020 Arab opinion index. Arab Center. https://arabcenterdc.org/ resource/the-2019-2020-arab-opinion-index-main-results-in-brief/#section7 Baran, Z. (2005). Fighting the war of ideas. Foreign Affairs, 84(6), 68–78. Bencherif, A., Belporo, L. C., & Morin, D. (2022, January). Étude internationale sur les dispositifs de prévention de la radicalisation et de l’extrémisme violents dans l’espace francophone. UNESCO. https://chaireunescoprev.ca/wp-content/uploads/2022/03/UNESCO-PREV_EspaceFranco_RapportFinal.pdf Berger, J. M. (2017, June 6). “Defeating IS ideology” sounds good, but what does it really mean? ICCT. https://icct.nl/publication/defeating-is-ideology-sounds-good-but-what-does-it-really-mean/ Bjørgo, T. (2002, June). Exit Neo-Nazism. Reducing recruitment and promoting disengagement from racist groups. NUPI. www.files.ethz.ch/isn/27380/627.pdf Bjørgo, T. (2005). Root causes of terrorism. Myths, reality and ways forward. Routledge. Bjørgo, T., & Horgan, J. (2009). Leaving terrorism behind. Individual and collective disengagement. Routledge. Bonelli, L., & Ragazzi, F. (2019, January). La lutte contre la “radicalisation”. Genèse et expansion d’un nouveau lieu commun administratif en France et dans l’Union européenne. Archives de politique criminelle, 41. Borum, R. (2003, July). Understanding the terrorist mind-set. FBI Law Enforcement Bulletin, 72(7), 7–10. Borum, R. (2011). Radicalization into violent extremism I: A review of social science theories. Journal of Strategic Security, 4(4), 7–36. Burgat, F. (2019). Understanding political Islam. Manchester University Press. Clifford, B., & Hughes, S. (2022, July 17). The history of countering violent extremism tends to repeat. It shouldn’t. Lawfare. Colaert, L. (2017, September). ‘De-radicalisation’. Scientific insights for policy. Flemish Peace Institute. CONTEST. (2009, March). Pursue, prevent, protect, prepare: The United Kingdom’s strategy for countering international terrorism. HMG. Coolsaet, R. (2007, September). Belgium and counterterrorism policy in the Jihadi era. Egmont Institute, Egmont Paper 15. Coolsaet, R., & Renard, T. (2022, June 9). What the zeitgeist can tell us about the future of terrorism. ICCT. https://icct.nl/publication/zeitgeist-future-of-terrorism/ COREU. (2004a, May 26). COTER thematic assessment on recruitment to terrorism. DUB COREU, CFSP/ PRES/DUB/1167/04. COREU. (2004b, November 18). Joint COTER/TWG Report on Recruitment. HAG COREU, CFSP/PRES/ HAG/1565/04. COREU. (2004c, December  13). Presentation on decentralisation of Islamist terrorism. HAG COREU, CFSP/PRES/HAG/1812/04. CPN-Prev, Montreal. https://cpnprev.ca/ Crenshaw, M. (1981). The causes of terrorism. Comparative Politics, 3(4); Same author (1991), How terrorism declines. Terrorism and Political Violence, 3(1), 69–87. Crenshaw, M. (2011). The debate over ‘old’ vs ‘new’ terrorism. In R. Coolsaet (Ed.), Jihadi terrorism and the radicalisation challenge. European and American experiences (pp. 57–67). Ashgate. Crenshaw, M., & LaFree, G. (2017). Countering terrorism. Brookings. Dawson, L. (2017). Discounting religion in the explanation of homegrown terrorism: A critique. In J. R. Lewis (Ed.), Cambridge companion to religion and terrorism (pp. 32–45). Cambridge University Press. Dawson, L. (2018). Debating the role of religion in the motivation of religious terrorism. Nordic Journal of Religion and Society, 31(2), 98–117. De Vries, G. (2021). EU counterterrorism policy 2005–2020. In C. Höhn, I. Saavedra, & A. Weyembergh (Eds.), The fight against terrorism: Achievements and challenges. Liber Amicorum Gilles de Kerchove (pp. 217–251). Bruylant.

49

Rik Coolsaet Dodwell, B., Milton, D., & Rassler, D. (2016, April). The Caliphate’s global workforce: An inside look at the Islamic state’s foreign fighters paper trail. CTC. Duyvesteyn, I. (2004). How new is the new terrorism? Studies in Conflict & Terrorism, 27(5), 439–454. Duyvesteyn, I., & Peeters, B. (2015, October). Fickle foreign fighters? A cross-case analysis of seven Muslim foreign fighter mobilisations (1980–2015). ICCT, Research Paper. Emmerson, B. (2015, September 18). Report of the special rapporteur on the promotion and protection of human rights and fundamental freedoms while countering terrorism. Human Rights Council, A/70/371. Emmerson, B. (2016, February 22). Report of the special rapporteur on the promotion and protection of human rights and fundamental freedoms while countering terrorism. Human Rights Council, A/HRC/31/95. English, R. (2016). The future study of terrorism. European Journal of International Security, 1(2), 135–149. European Commission. (2004, October 20). Prevention, preparedness and response to terrorist attacks. Communication from the Commission to the Council and the European Parliament. COM(2004) 698. European Commission. (2005, September 21). Communication from the commission to the council and the European parliament entitled. Terrorist Recruitment: Addressing the Factors Contributing to Violent Radicalisation. COM(2005) 313. European Council. (2005a, November 30). The European union counter-terrorism strategy. 14469/4/05 REV 4. Brussels. European Council. (2005b, November 24). The European union strategy for combating radicalisation and recruitment to terrorism. 14781/1/05 REV 1. Brussels. Europol. (2004, December 2). Terrorist activity in the European Union: Situation and trends report (TE-SAT). The Hague, Europol. Expert Group. (2008). Radicalisation processes leading to acts of terrorism. Report prepared by the European Commission’s Expert Group on Violent Radicalisation. The list of members and their affiliation, as well as the text of the report is available at. www.clingendael.org/sites/default/files/pdfs/20080500_cscp_report_vries.pdf Fettweis, C. J. (2009). Freedom fighters and zealots: Al Qaeda in historical perspective. Political Science Quarterly, 124(2), 269–296. Gellman, B. (2002, December 24). In U.S., terrorism’s peril undiminished. Washington Post. Githens-Mazer, J. (2010, January). Rethinking the causal concept of Islamic radicalisation. Committee on Concepts and Methods. www.concepts-methods.org/Files/WorkingPaper/PC%2042%20Githens-Mazer.pdf Hafez, M., & Mullins, C. (2015). The radicalization puzzle: A theoretical synthesis of empirical approaches to homegrown extremism. Studies in Conflict & Terrorism, 38(11), 958–975. Hawkins, S. W. (2015, February 19). Obama’s anti-extremism plan lacks human rights safeguards. Al-Jazeera. Hegghammer, T. (2013). Should I stay or should I go? Explaining variation in western Jihadists’ choice between domestic and foreign fighting. American Political Science Review, 107(1), 1–15. Hoffman, B. (1993). Holy Terror: The implications of terrorism motivated by a religious imperative. RAND. Hoffman, B. (1998). Inside terrorism. Columbia University Press. Horgan, J. (2008). From profiles to pathways and root to routes: Perspectives from psychology on radicalization into terrorism. The Annals of the American Academy of Political and Social Science, 618, 80–94. Horgan, J. (2014). Peace arena. Insight. USIP, Spring. Horgan, J. (2019). Psychological approaches to the study of terrorism. In E. Chenoweth et al. (Eds.), The Oxford handbook of terrorism. Oxford University Press. ICSR. (2008). Perspectives on radicalisation and political violence: Papers from the first international conference on radicalisation and political violence. https://icsr.info/wp-content/uploads/2008/03/Perspectives-on-Radicalisation-Political-Violence.pdf Jenkins, B. (1999). Foreword. In B. Hoffman (Ed.), Countering the new terrorism. RAND. Jenkins, B. (2001). Terrorism and beyond: A 21st century perspective. Studies in Conflict and Terrorism, 24(5), 321–337. Jones, S. (2018). Radicalisation in the Philippines: The Cotabato cell of the “East Asia Wilayah”. Terrorism and Political Violence, 30(6), 933–943. Juergensmeyer, M. (2000a). Understanding the new terrorism. Current History, 99(636), 158–163 Juergensmeyer, M. (2000b). Terror in the mind of god. University of California Press. Kaplan, J. (2016, November). Waves of political terrorism. Oxford Research Encyclopedias. https://oxfordre. com/politics/view/10.1093/acrefore/9780190228637.001.0001/acrefore-9780190228637-e-24/version/0 Kellen, K. (1979). Terrorists – what are they like? How some terrorists describe their world and actions. RAND.

50

‘Radicalisation’ and ‘countering radicalisation’ Kepel, G. (2015). Terreur dans l’Hexagone. Genèse du djihad français. Gallimard. Kilcullen, D. (2005). Countering global insurgency. Journal of Strategic Studies, 28(4), 597–617. King, M., & Taylor, D. (2011). The radicalization of homegrown Jihadists: A review of theoretical models and social psychological evidence. Terrorism and Political Violence, 23(4), 602–622. Koehler, D. (2017). Understanding deradicalization. Methods, tools and programs for countering violent extremism. Routledge. Kundnani, A. (2009, October). Spooked: How not to prevent violent extremism. Institute of Race Relations. Kundnani, A. (2012). Radicalisation: The journey of a concept. Race & Class, 54(2), 3–25. LaFree, G., & Freilich, J. D. (2019). Government policies for counteracting violent extremism. Annual Review of Criminology, 2(13), 1–22. Lambert, R. (2011). Competing counter-radicalisation models in the UK. In R. Coolsaet (Ed.), Jihadi terrorism and the radicalisation challenge. European and American experiences (pp. 215–225). Ashgate. Lia, B. (2016, December). Jihadism in the Arab world after 2011: Explaining its expansion. Middle East Policy, 23(4), 74–91. Malet, D. (2015). Foreign fighter mobilization and persistence in a global context. Terrorism and Political Violence, 27(3), 454–473. Malthaner, S. (2017). Radicalization: The evolution of an analytical paradigm. European Journal of Sociology, 58(3), 369–400. Mandaville, P. (2017, May 8). Designating Muslims. Islam in the western policy imagination. Cambridge Institute on Religion and International Studies. McCauley, C., & Moskalenko, S. (2008). Mechanisms of political radicalization: Pathways toward terrorism. Terrorism and Political Violence, 20(3), 415–433. McCauley, C., & Moskalenko, S. (2011). Friction: How radicalization happens to them and us. Oxford University Press. McCauley, C., & Moskalenko, S. (2014). Some things we think we’ve learned since 9/11: A commentary on Marc Sageman’s “the stagnation in terrorism research”. Terrorism and Political Violence, 26(4), 601–606; Same authors. McCauley, C., & Moskalenko, S. (2017). Understanding political radicalization: The two-pyramids model. American Psychologist, 72(3), 205–216. Moghaddam, F. A. (2005). The staircase to terrorism. A psychological exploration. American Psychologist, 60(2), 161–169. Molenbeek. (2017, May). Molenbeek and violent radicalisation: A ‘social mapping’. European Institute of Peace. https://view.publitas.com/eip/eip-molenbeek-report-16-06/page/1 Neumann, P. (2013). The trouble with radicalization. International Affairs, 89(4), 873–893. Neumann, P., & Kleinmann, S. (2013). How rigorous is radicalization research? Democracy and Security, 9(4), 360–382. Nickolson, L., van Bergen, N., Feddes, A., Mann, L., & Doosje, B. (2021). Extremist thinking and doing. A systematic study of empirical findings on the radicalisation process. WODC. Pillar, R. (2004). Counterterrorism after al-Qaeda. Washington Quarterly, 27(3), 101–113. Pinfari, M. (2020, December). Country reports on national approaches to extremism. Framing Violent Extremism in the MENA region and the Balkans – Egypt. European Institute of the Mediterranean. https:// h2020connekt.eu/wp-content/uploads/2021/01/Egypt_CONNEKT_Approaches_to_extremism.pdf Pistone, I., Eriksson, E., Beckman, U., Mattsson, C., & Sager, M. (2019, Summer). A scoping review of interventions for preventing and countering violent extremism: Current status and implications for future research. Journal for Deradicalization, 19. RAN. (2016, December). Radicalisation research – gap analysis. https://radical.hypotheses.org/ files/2018/06/201612_radicalisation_research_gap_analysis_en1.pdf Renard, T. (2019, December). Returnees in the Maghreb. A European perspective. Egmont, Security Policy Brief. Renard, T. (2022). The evolution of counter-terrorism since 9/11. Understanding the paradigm shift in liberal democracies. Routledge. Rietjens, P. (2007). België, de EU en het Jihadi-terrorisme. Studia Diplomatica, LX, Special Issue. Rink, A., & Sharma, K. (2018). The determinants of religious radicalization: Evidence from Kenya. Journal of Conflict Resolution, 62(6), 1229–1261. Rosenblatt, N. (2016, July). All Jihad is local. What ISIS’ files tell us about its fighters. New America.

51

Rik Coolsaet Roy, O. (2015). What is the driving force behind Jihadist terrorism? A scientific perspective on the causes/circumstances of joining the scene. Bundeskriminalamt, Autumn Conference. file:///C:/Users/rikco/Downloads/ herbsttagung2015Roy%20(1).pdf Roy, O. (2016). Le djihad et la mort. Seuil. Sageman, M. (2004). Understanding terror networks. University of Pennsylvania Press. Sageman, M. (2008a). Leaderless Jihad. Terror networks in the twenty-first century. University of Pennsylvania Press. Sageman, M. (2008b, July/August). Does Osama still call the shots? Debating the containment of al Qaeda’s leadership. Foreign Affairs, 87(4), 163–166. Sageman, M. (2014). The stagnation in terrorism research. Terrorism and Political Violence, 26(4). Schmid, A. (2013a, March). Radicalisation, de-radicalisation, counter-radicalisation: A conceptual discussion and literature review. ICCT Research Paper. www.icct.nl/download/file/ICCT-Schmid-Radicalisation-De-Radicalisation-Counter-Radicalisation-March-2013.pdf Schmid, A. (2013b, July). The end of radicalisation? ICCT. https://icct.nl/publication/the-end-of-radicalisation/ Sedgwick, M. (2010). The concept of radicalization as a source of confusion. Terrorism and Political Violence, 22(4), 479–494. Silke, A., & Brown, K. (2016). ‘Radicalisation’: The transformation of modern understanding of terrorist origins, psychology and motivation. In S. Jayakumar (Ed.), State, society, and national security: Challenges and opportunities in the 21st century. World Scientific Publishing. Simon, S., & Benjamin, D. (2000). America and the new terrorism. Survival, 42(1), 59–75. Smith, C., Resnyansky, L., & Taylor, C. (2022, March 13). Radicalization in the Asia-pacific region: Themes and concepts. Studies in Conflict and Terrorism. www.tandfonline.com/doi/full/10.1080/1057610X.2022. 2034232?scroll=top&needAccess=true Stephens, W., & Sieckelinck, S. (2019). Working across boundaries in preventing violent extremism: Towards a typology for collaborative arrangements in PVE policy. Journal of Deradicalization, 20, 272–312. Stern, J. (2003). Terror in the name of god: Why religious militants kill. Ecco. Taylor, M. (1988). The terrorist. Brassey’s. Thomas, P. (2010). Failed and friendless: The UK’s ‘preventing violent extremism’ programme. British Journal of Politics & International Relations, 12(3), 442–458. Timbuktu Institute. (2021). Les pays côtiers face à la menace terroriste: Comment éviter les ‘erreurs’ du Sahel ? https://timbuktu-institute.org/index.php/toutes-l-actualites/item/531-les-pays-cotiers-face-a-lamenace-terroriste-comment-eviter-les-erreurs-du-sahel. See also the other studies by the Institute. Trauthig, I. K. (2021). Counterterrorism in North Africa: From police state to militia rule and the quagmire of “CVE”. ICSR. U.N. (2015, December 24). Plan of action to prevent violent extremism. Report of the Secretary-General. United Nations: Secretary General. www.un.org/en/ga/search/view_doc.asp?symbol=A/70/674 UNDP. (2004). Human development report 2004. Cultural liberty in today’s diverse world. UNDP. Urdal, H. (2006). A clash of generations? Youth bulges and political violence. International Studies Quarterly, 50(3), 607–629. Van de Voorde, T. (2011). ‘Terrorism studies’: A critical appraisal. In R. Coolsaet (Ed.), Jihadi terrorism and the radicalisation challenge. European and American experiences (pp. 45–54). Ashgate. Vidino, L. (2010). Countering radicalization in America. Washington, USIP. www.usip.org/sites/default/files/ resources/SR262%20-%20Countering_Radicalization_in_America.pdf Weggemans, D., Bakker, E., & Grol, P. (2014). Who are they and why do they go? The radicalisation and preparatory processes of Dutch Jihadist foreign fighters. Perspectives on Terrorism, 8(4), 100–110. White House. (2011, August). Empowering local partners to prevent violent extremism in the United States. www.dhs.gov/sites/default/files/publications/empowering_local_partners.pdf Wilkinson, P. (1974). Political terrorism. MacMillan Press.

52

4 ANALYSING ‘RADICALISATION’ IN HISTORICAL CASES Tim Wilson

Introduction Does contemporary radicalisation have a pre-history? If it does, what might we learn from it? This chapter focuses exclusively on the ‘micro-level processes’ by which individuals or very small groups attempt system-changing violence (Marsden, 2020). I concentrate on projects of violence that have been launched by very few people and with very limited resources, a focus that leads me away from considering larger movements such as the Sicarii or Nizari Isma’ilis (i.e. assassins). Such violence is shocking in any age, if only for its sheer ambition. What I call ‘micro-radicalisation’ therefore offers a recognisably recurrent phenomenon that it makes sense to examine across different ages. I keep the focus on Europe and the Western world since this is where the paradigm of radicalisation first arose (Malthaner, 2017, p. 371). In making this attempt, this chapter does three things. First, it offers a very brief chronological account of some of the better-recorded instances of radicalisation since their appearance in Europe in the late 16th century. This survey is not comprehensive, but indicative. It is merely a ‘trial trenching’ exercise: the first investigation to guide where others might dig more systematically in the future. Second, it tries to allow some discussion of how contemporaries made sense of such radicalisation. Lastly, and with necessary brevity, it tries to stand back and discern the significance of the fluctuating incidence of such violence over the long term.

The emergence of radicalisation ‘In western Europe between the middle of the sixteenth century and the 1630s’, Colin Kidd has observed (2012, p. 269) ‘the threat of assassination came to loom very large indeed’. King Henry IV of France was to survive (at least) 23 assassination attempts before his luck finally ran out in the middle of a Paris traffic jam (Greengrass, 1987, pp. 251–252; Mousnier, 1973, pp. 21–26, 105). Moreover, this was a new pattern of threat: ‘in the West at least the solitary political assassin is a phenomenon of rather recent origin, dating only from the sixteenth-century wars of religion’ (Van Kley, 353). Why, then, did Western Europe function in the late 16th century as a laboratory of radicalisation? Since 1517, the Reformation has shattered the cultural unity of Western Europe as various reform – or ‘Protestant’ – movements have broken away to challenge the moral authority

53

DOI: 10.4324/9781003035848-5

Tim Wilson

of the Catholic Church. At a popular level, the resulting conflicts to establish the ‘correct’ version of Christian belief and practice were turbo-charged by the emergent technology of printing (Pettegree, 2015, 2016). Although this remained an age of vertiginous social hierarchy, new cultures of publicity created new sub-cultures of religious militancy. Such conditions revived old debates among Catholic intellectuals over tyrannicide: the belief that it might sometimes be just to kill unjust rulers (Bell, 1979/2005, p. 13). In the words of the firebrand Catholic preacher, Jean Boucher, it was ‘a most meritorious act to kill a king who was a heretic or supporter of heretics’ (in Knecht, 2014, p. 307). It is abundantly clear that there was some degree of popular sympathy, or at least ambivalent tolerance, for such beliefs (Knecht, 2014, pp.  303, 307–308). Given the complex and unstable interplay of religious and political forces, this was unsurprising. Henry IV’s conversion to Catholicism looked tactical, apparently taken to remove the last obstacle to his assuming the French crown. His own private sympathies seemed to sceptics to be better reflected in the 1598 Edict of Nantes (that extended limited civic rights to French Protestants). Given his own ongoing international disputes with the Papacy, he remained deeply distrusted by militant Catholics. Hence the problem of Catholic radicalisation – what the historian Roland Mousnier has called the existence of a ‘large body of possible murderers’ (1973, p. 48) – went to the very heart of maintaining stable government in the early modern period. Can we be more precise about the identity of these Catholic ‘possible murderers’ of early modern Europe? Nearly all were male (though on two separate occasions, Portuguese women did try to stab Philip II of Spain) (Parker, 2015, p. 294). In an age where illiteracy remained the norm, these men seem to have been – relatively – literate and well-travelled. Beyond this, it is hard to generalise. Certainly, they display no one dominant social profile. Some were highly educated. Dr. William Parry had graduated with a doctorate of law from Paris and holds the dubious honour of being the only English MP executed for treason (on 2 March 1585) (Alford, 2013, pp. 188–192). John Somerville, the Warwickshire gentleman who in October 1583 unwisely announced his intention of killing Queen Elizabeth I of England to his fellow drinkers in a pub, had previously studied at Oxford (Wizeman, 2008; Alford, 2013, p. 134). The 19-year-old Jean Chastel, who stabbed Henry IV of France in the mouth on 27 December 1594, was a student of law at Paris and came from a respectable family of merchant drapers. He had previously studied classics and philosophy at a Jesuit College (Greengrass, 1987, p. 252; Descimon, 2009, p. 86). Similarly, it was as a student at the Catholic university at Dôle in the early 1580s that Balthasar Gérard first announced his intention of killing the Count of Nassau (and international Protestant icon), William of Orange (Jardine, 2005, p. 55). Some belonged to the gentry class, especially in England. That was true not only of Somerville but also of most of the Gunpowder Plotters (who audaciously plotted to blow up both the king and parliament in 1605). In general, the Catholic gentry of the English midlands seem to have been a seedbed of militancy (Fraser, 1996, pp. 46–49, 148). Although not great aristocrats, these remained socially privileged figures despite the intermittent persecution and heavy fines inflicted upon them by the government. Only such figures had the resources to plan as ambitious a conspiracy as the Gunpowder Plot of 1605. Others, though, were truly marginal figures: killers from the social fringe. Both of the successful French regicides were nobodies. On 1 August 1589, a young Dominican friar named Jacques Clément stabbed King Henry III to death (on his commode, of all places). Since Clément was instantly killed in turn by the king’s guard, we know relatively little about him. But it is clear he was a figure whom no one had ever taken very seriously: his fellow friars recalled afterwards that when he spoke, people tended to laugh at him (Knecht, 2014, p. 298). We know far more about his 54

Analysing ‘radicalisation’ in historical cases

successor in regicide, François Ravaillac. On 14 May 1610, Ravaillac simply leant into the royal carriage in the congested Rue de la Ferronerie and killed King Henry IV with two blows (Mousnier, 1973, pp. 25–26). Subsequent interrogations allow us to reconstruct his biography in some detail. At 31, the unmarried Ravaillac had been imprisoned for debt: his past also included stints as a ‘rejected monk and failed schoolteacher’ (Horne, 2002, p. 100). By Christmas 1609, he was in Paris. He went to see Father Daubigny, a Jesuit priest, for confession. Ravaillac waved a knife around and spoke wildly of visions. He was convinced that he must persuade King Henry IV to convert the French Protestants by force. And he looked rough. Daubigny therefore advised him ‘that he ought to eat good soup, go to his own country [i.e. home region], tell his [rosary] beads, and pray to God’ (Goldsmid, 1885, pp. 31–33). This was excellent advice: in effect, it attempted de-radicalisation through social re-integration. But it was not sufficient. Ravaillac seems to have spent the next few weeks shuttling indecisively back and forth between Paris and his hometown of Angoulême in south-western France, a journey of eight days on foot each way. He tried to speak with the king near St. Innocents in Paris but was brushed aside. By Easter Sunday 1610, over a month before his attack and now back in Angoulême, he could not bring himself to take Holy Communion because the temptation to kill the king was troubling him too much. He stole a knife on the road, broke the tip to try to render it useless, and then had it re-sharpened again. Finally, and fatally, there was a chance conversation in a pub. Some soldiers he met expressed their concern that Henry IV was about to declare war against the Pope. Ravaillac’s outrage at this prospect consolidated his own sense of purpose: ‘for this cause he yielded to the temptation which prompted him to kill the king, because he made war against God, inasmuch as the pope is God, and God the pope’ (Goldsmid, 1885, p. 16: Mousnier, 1973, pp. 27–37). Over four centuries later, Ravaillac’s own account of becoming an ideological killer remains compelling for its arresting immediacy (Goldsmid, 1885, pp. 5–52). It offers a vivid portrait of an inner battle fought, and lost, against a relentless obsession. As such, it also invites wider reflection on the making of early modern assassins. How typical was this journey when compared to other ‘potential Ravaillacs’ (Mousnier, 1973, p. 48)? Answering that question is not easy: then as now, there were a variety of pathways into ‘godly killing’ (Kidd, 2012). Humans were no simpler in the 17th century than they are today, nor were they any more predictable. Still, in practice, several distinct pathways are identifiable. One was ‘status-seeking’ (Moskalenko & McCauley, 2020, p. 100). Both serious attempts against the life of William of Orange were galvanised, in part, by the huge price set on his head by King Philip II of Spain: 25,000 gold coins (Ford, 1985, p. 161; Parker, 2015, pp. 250, 279, 293). By contrast, other assassins conceived of their mission in entirely self-sacrificial terms. Neither Jean Chastel nor François Ravaillac made any attempt to flee the scene of their attacks. More theatrical was the ex-soldier John Felton, who knifed the Duke of Buckingham in a crowded Portsmouth pub on 23 August 1628. Buckingham’s disastrous leadership of an expedition to support the French Protestants at La Rochelle had made him hugely unpopular. Amidst the chaos that followed his deed, Felton simply announced: ‘I am the one’. Statements stitched into his hatband further elaborated ‘that, as a solider and a gentleman, he had chosen to hazard this life for the honour of his God, king, and country’ (Bellany, 2008; also Aikin, 1833, pp. 222–224). Assassins were also motivated by personal crises. Four of the Gunpowder Plotters underwent a sudden ‘rejection of youthful misdemeanours, a ricochet towards ardent piety’ (Fraser, 1996, p. 50). Some turning points stand out sharply. Jean Chastel ‘had hit his mother and his reading of a book by the theologian Poncet persuaded him that he would never obtain forgiveness for the act’ 55

Tim Wilson

(Descimon, 2009, p. 90). François Ravaillac became severely distressed in telling his interrogators about his expulsion from the Cistercians for having visions: ‘this he uttered with tears’ (Goldsmid, 1885, pp. 8–9). John Felton came back from the slaughter at La Rochelle in late 1627 a changed man: according to his family, ‘of melancholy disposition’ and troubled ‘by dreams of fighting’ (quoted in Bellany, 2008). What, then, shifted personal crisis towards political killing? Some sense of external affirmation was needed. Some self-starters seem to have found such ‘validating charters’ more or less by chance (Thomas, 1991, p.  503). Jean Chastel ‘was apparently inspired to kill the king after reading a devotional treatise’ (Greengrass, 1987, p. 252). John Felton was already privately bitter at the Duke of Buckingham, who, he believed, had thwarted his promotion to captain and withheld £80 in backpay. But he only shifted decisively towards killing him when he discovered an anti-Buckingham ‘remonstrance’ drawn up by the House of Commons. This document convinced ‘him that his personal suffering at Buckingham’s hands was part of a much larger story, that of the duke’s wicked and treacherous rule over the whole nation’ (Bellany, 2008). In the more elaborate conspiracies, personal relationships remained key. Edmund Neville was recruited into a plot to assassinate Queen Elizabeth I by his cousin, William Parry (Alford, 2013, p. 187). Juan Jáuregui agreed to shoot William of Orange ‘for the love of his master and for the honour of his name’ (quoted in Jardine, 2005, p. 70). The Gunpowder Plot was similarly welded together by ‘strongly felt male bond[s]’ (Haynes, 2010, p. 9). Here, the driving personality was the deeply charismatic Robin Catesby. We can trace the half-life of his influence through interrogation records. When the group’s explosives expert, Guy Fawkes, declared to King James that a ‘dangerous disease required a desperate remedy’, he was recycling a ‘comforting catchphrase that had been in general use among the conspirators’ (Fraser, 1996, pp. 90–93, 98, 174). Radicalising others was often hard work. In 1585, an English spy in Dunkirk reported that a friar had attempted to convince him to kill Queen Elizabeth I with the aid of a picture of Balthasar Gérard shooting William of Orange: Look how this Burgundian did kill this prince. In such manner and sort, there will not want [be forthcoming] such another Burgundian to kill that wicked woman and that before it be long, for the common wealth of all Christendom. (quoted in Alford, 2013, p. 135) It is also worth noting in passing that William of Orange must have been dead for only a few months at this point, but pictures of his killing were clearly circulating widely. ‘Assassins’, remarks Colin Kidd (2012, p. 269), ‘were household names in early modern Europe’. If assassins were such a talking point, it is worth briefly considering what informed contemporaries made of them. How do they think they were created? What did they think drove them? In general, political assassination was seen as an extension of spiritual battlefields. To his interrogators, ‘the Devil had inspired Chastel to do the deed’ (Descimon, 2009, p. 91). Ravaillac faced an identical line of questioning (Mousnier, 1973, pp. 28–38). To those it would have detonated, the Gunpowder Plot was simply ‘Satan’s policy’ (Fraser, 1996, p. 279). Common sense in the period also reflected patrician condescension. As a student, Balthasar Gérard had explicitly been told by an older companion that it was not for the likes of him to think of killing great princes (Wedgewood, 1945, p. 214). All the great lords who interrogated François Ravaillac thought ‘he was too mean and inconsiderable a person to have conceived such a design, and that he must have been advised and supported by others’ (Goldsmid, 1885, pp. 25, 51).

56

Analysing ‘radicalisation’ in historical cases

Where such assumptions led was towards a relentless search for hidden conspiracy. The phenomenon of the lone assassin was barely acknowledged. When Henry III of France was killed, ‘it was assumed that Clément had not acted alone’ (Knecht, 2014, p. 308). Right until the very end, when he was being pulled apart by wild horses, Ravaillac was urged to reveal his accomplices (Goldsmid, 1885, p. 48). It took those who investigated John Felton three months to accept, with the greatest reluctance, that he was what he said he was: a lone killer (Bellany, 2008). Given their formidable reputation for intrigue, the Jesuits were frequently assumed to be the true puppet masters of conspiracy (Mousnier, 1973, pp.  53–60, 213–228). In 1594, the Jesuits were briefly expelled from the Paris region following Jean Chastel’s attack on Henry IV (Greengrass, 1987, p. 252). In March 1606, the Attorney General of England, Sir Edward Coke, brazenly declared that he would name the Gunpowder Plot ‘the Jesuits’ treason’ (in Fraser, 1996, p. 244). Generally, the making of an individual assassin was assumed to be a rather mechanical process. According to Robert Burton (1620/2001, p. 345), the Jesuits would bring a future assassin into a melancholy dark chamber, where he shall see no light for many days together, no company, little meat, ghastly pictures of devils all about him, and leave him as he will himself, on the bare floor, in this chamber of meditation as they call it, on his back, side, belly till by this strange usage they make him quite mad and beside himself. And then after some ten days, as they find him animated and resolved, they make use of him. All in all, a general indifference to the mind of the individual assassin remains a notable feature of the early modern period.

Radicalisation: the gap Understood as a micro-process, the history of radicalisation in Western Europe becomes harder to trace after c. 1630. The reason is simple: the supply of assassinations begins to dwindle in the later 17th century and all but disappears for most of the 18th century. In effect, we are confronted with a puzzling absence: ‘a near-moratorium on political murder’ (Ford, 1976, p. 213). Why? That is a bigger question than can be explored here: but three tentative suggestions are worth offering. First, kings stopped sponsoring the assassination of their peers. The late 16th century was an age where, as seen, Philip II could advertise for a contract killer to remove William of Orange and order the bells of Madrid to be rung in celebration at the St. Bartholomew’s Day Massacre of Protestants in Paris. By the late 17th century, those who planned to kidnap William III of England on behalf of the exiled James II were notably coy about admitting they planned to kill him. They knew that would be vetoed (Garrett, 1980, p. 76). Elite expectations seem to have shifted. Second, the (relative) stabilisation of sectarian relations that followed the Peace of Westphalia in 1648 seems to have dried up the supply of religious killers from below (Te Brake, 2017). Finally, after the mid-17th century civil wars across Britain and France, a culture of social deference was re-established and the idea of killing kings began to fade. Contrasts with the previous century could hardly have been starker: the French ‘kings at Versailles, almost unguarded, lived in a perpetual crowd, and yet, in a hundred years there was only one half-hearted attempt at assassination’ (Mitford, 1969, p. 88). That attempt still repays attention. On 5 January 1757, Louis XV was stabbed with a penknife between the fourth and fifth ribs. His assailant was François Damiens, a 42-year-old domestic servant currently on the run for stealing 240 louis [gold coins] from a previous master

57

Tim Wilson

(De Kley, 1984, pp. 2–5, 13). Damiens’s assault thus represented a form of ‘forward escape’ from an intense personal crisis. But it was also a calculated intervention in a long-simmering political crisis. At its heart, that crisis was a demarcation dispute between the legal and ecclesiastical spheres (De Kley, 1984, p. 10). Damiens identified with the magistrates against the clergy (who he saw as over-taxing the poor). He was enraged by the tactic introduced by the Archbishop of Paris of denying sacraments to opponents (De Kley, 1984, p. 39). All along, Damiens maintained his intention had been only to ‘touch’ the king and, by so doing, to pressurise him into backing the magistracy [parlement] decisively. He said he had never heard of any books that argued that it was permissible to kill a king. Damiens represented an enigma to his interrogators. It was absolutely inconceivable to them that any person of humble social extraction such as Damiens could have independently conceived the idea of killing or hurting the king, because the necessary motivation would be lacking. In order to possess this motivation, one would have had to be personally offended by or otherwise in direct conflict with the king and therefore move in exalted social circles. (De Kley, 1984, p. 66) In short, they embarked on a futile hunt to find ‘accomplices’ because they could not accept that Damiens could have developed any informed political opinions of his own. Yet Damiens had seen first-hand how much employment depended on the legal profession. Between 1752 and 1754, he had been a personal servant to one of the leading magistrates in the Paris region. He had spent hours listening to the lawyers discussing ‘very loudly’ the gathering crisis and how ‘the king risked much by not preventing the evil conduct of the archbishop of Paris’. All of this, in his words, ‘heated up his head’ (in De Kley, 1984, p. 45). A key turning point for him seems to have been the mass resignation of Paris magistrates on 13–14 December– 1756: a move calculated, like his own attack three weeks later, to pressurise the king into siding against the archbishop (De Kley, 1984, pp. 44, 51). It seems clear that Damiens never had any accomplices. But he did have potential imitators (despite his grisly execution). Over the next few years, rumours of another attempt on the life of the king repeatedly surfaced; loose talk briefly swelled to a floodtide (De Kley, 1984, pp. 240–245, 264, 345). Even if these did not come to anything, the Damiens affair reminds us how mass publicity can foster individual ambitions whose maverick trajectories are impossible to predict.

Radicalisation and revolution The late 18th century’s twin revolutions in British America (1775–83) and France (1789–99) are often taken as marking the dawn of modern Western politics. Strikingly, this watershed period did not spawn hosts of lone assassins or micro-conspiracies as the late 16th century had done. Revolutions do not seem to have mass produced radicalised individuals willing to strike alone (or almost so). Still, two notable exceptions are worth briefly discussing here: James Aitken and Charlotte Corday. Born into poverty in Edinburgh in 1752, James Aitken was well educated at an orphanage school. He was later to declare that it had been his early reading about ancient Greece and Rome that kindled in him ‘the desire of accomplishing some great achievement’ (in Warner, 2005, p. 10). His early life was chaotic; he became variously a thief, highwayman, deserter from the British Army, and runaway servant in the American colonies. What he did develop were strong proAmerican sympathies in the gathering struggle against British rule. 58

Analysing ‘radicalisation’ in historical cases

By mid-1775, he was back in England. A chance encounter in an Oxford pub proved fateful. Aitken’s fellow drinkers all agreed: the Royal Navy depended on its dockyards. Without them, the war in America would be lost. For Aitken, it was a Eureka moment: ‘It is amazing with what force this conversation kept possession of my mind’ (in Warner, 2005, p. 95). Visits to the royal dockyard revealed to him that they were protected against external attack but not sabotage. In Paris, he introduced himself to the American Congress’s representative in France and secured his (rather formulaic) support (Warner, 2005, pp. 105–116). James Aitken began his four-month arson campaign in late 1776. He set several small fires in the dockyards at Portsmouth and Bristol. Ultimately, he managed to cause only limited destruction with the resources available to him. He struggled to develop an incendiary device that was portable, concealable, and also efficiently combustible. He was also indiscreet: boasting too often to strangers of his future notoriety. Such a solo run could not be sustained indefinitely; exhausted, he was eventually run to the ground in a Hampshire village (Warner, 2005, pp. 117–128, 154–168, 171, 225). Aitken came close to achieving massive devastation. He decisively succeeded in generating widespread alarm that an organised conspiracy of arsonists was at work (Warner, 2005, p. 167). Commentators ‘with classical educations compared him to Herostratus, “who to become immortal in history, set fire to the famous temple of Diana at Ephesus” ’ (Warner, 2005, p. 248). That was as close as anyone came to understanding the ‘quest for significance’ that drove, in part, Aitken’s behaviour. He was soon forgotten. Charlotte Corday (born 1768) came from an impoverished aristocratic family in Normandy. In 1784, her mother died, and her father sent her to the Abbaye-aux-Dames convent in Caen to complete her education. In 1789, she decided to become a nun. But the French Revolution broke out that summer (Loomis, 1964, pp. 21, 30, 36, 39–40). By 1791, the convent itself had been shut down. Corday found herself boarding with an elderly aunt in Caen, circumstances rather conducive to unhealthy brooding (Loomis, 1964, pp. 32–33, 39–40, 49–51). She conceived a hatred of Jean-Paul Marat, a rising demagogue on the left of the National Assembly who she saw as dragging France into civil war. She took to writing down the same question: ‘Shall I or shall I not?’ (Loomis, 1964, p. 9). By June 1793, she was committed to self-sacrifice to save France. She gave away her pencil box to a child: ‘Always be good, and kiss me, because you will never see me again’ (in Loomis, 1964, pp. 102, 106). On 9 July  1793, Corday took the stagecoach to Paris. She had never been before and was amazed that the hack drivers did not know where Marat lived. But she persevered. On Saturday 13 July, she bought a 6-inch kitchen knife (Loomis, 1964, pp. 113–114). That afternoon, she gained direct access to Marat at home at his bathtub-cum-writing desk (where he habitually sought relief from a skin complaint). She had promised him information on his enemies in Normandy. And then she butchered him with one decisive downward blow into his chest: ‘a student of anatomy could not have done better’ (Loomis, 1964, p. 122). Inside her bodice was found ‘An Address to the French People’ (giving her motives), and her baptismal certificate (as identity) (Loomis, 1964, p. 112). However, her interrogators pursued the spectre of a wider conspiracy since they assumed ‘so atrocious a deed could not have been committed by a woman your age’. Her answer is worth quoting for its self-awareness: ‘ “You do not understand the human heart”, she replied. “It is far easier to execute a plan such as this when one is inspired by one’s own hatred than when one is inspired by the hatred of others” ’ (in Loomis, 1964, pp. 131–133). Officialdom remained obsessed with trying to explain her act through her gender, or, rather, her sexuality. A formal report of her autopsy solemnly reported that she had died a virgin (Loomis, 1964, pp. 34, 138–139). 59

Tim Wilson

A modest drip-feed of broadly comparable instances continues throughout the first half of the 19th century (Ford, 1985, pp. 212–216). One case gained particular notoriety. On 23 March 1819, Carl Ludwig Sand called upon the 57-year-old playwright August Von Kotzebue at his home in Mannheim and killed him with a flurry of knife blows before, ineptly, stabbing himself in the stomach. Von Kotzebue’s offence in Sand’s eyes was as a traitor to the German nation, both for his close links to leading figures in the Russian government, and also for writing ‘immoral’ plays (Williamson, 2000, pp. 890–892). Sand was a 24-year-old student of theology at Jena University. He was the son of a legal official from Franconia. His childhood had been upended by the French occupation that began in 1806. Like so many others of his generation, he was attracted to pan-German nationalism and its vision of building a strong and unitary German state. Sand had personal torments to manage, too. Rumours dogged him that he had failed to save a friend from drowning through ‘indecision and lack of courage’. His diary suggests a personality tormented by a desire to achieve moral purity. As early as 1816, he was reflecting on Christ’s death on the cross. Self-sacrifice appeared to offer a way out: ‘Couldn’t I myself in that moment subdue the snakes of natural instinct and laziness?’ (in Williamson, 2000, pp. 921–922). Sand identified closely with the Burschenschaft movement of student fraternities. Extravagant nationalist posturing was expected here (‘you are a weakling if you can’t peacefully cut bread for a meal with the same knife that you have just used to kill a noble man’) (in Williamson, 2000, pp. 933–934). Von Kotzebue was already unpopular in these circles, but a series of court battles in which he sued publications close to Burschenschaften circles only ratcheted up their hatred for him. As early as May 1818, Sand was writing in his journal: ‘I often think that someone should bravely take it upon himself to stick a sword in the guts of Kotzebue or another such traitor to the fatherland’ (in Williamson, 2000, p. 929). But it still took the next ten months for this thought to evolve fully from a daydream into a life mission. He carefully prepared a manifesto entitled ‘Death Blow to August Von Kotzebue’, which he intended to pin to the latter’s front door with his dagger. And when it finally came to the key moment, he feinted first to draw Von Kotzebue’s arms upwards and then stabbed hard into the unprotected chest. Sand planned this move after studying his friends’ reactions to being startled with a stick (Williamson, 2000, pp. 932, 935). By contrast, his own suicide attempt was botched (Ford, 1985, p. 215). ‘The immediate aftermath of the Kotzebue murder was dominated by a determined search for evidence of a wider conspiracy’ (Ford, 1985, p. 215). None were found. Sand’s own fate was to be beheaded at Mannheim on 20 May 1819. The execution scaffold was recycled to provide the hallowed frame of a small hut where Heidelberg Burschenschaftler would meet in secret (Williamson, 2000, p. 940; Ford, 1985, p. 215). Less secretively, Sand’s picture sold well: ‘some students wished that they, rather than Sand, had enjoyed the honor of martyrdom for Germany’. Sand’s more lasting monument was the reactionary Carlsbad Decrees. Drawn up by Prince Metternich of Austria (and ratified in September 1819 by the Federal Diet of German states at Frankfurt), these erected an integrated system of counter-radicalisation across Central Europe that clamped down on all freedom of inquiry. Liberal freedoms, Prince Metternich was firmly convinced, would create a ‘deluge of assassins’ (Zamoyski, 2014, pp. 233, 437).

Radicalisation and the rise of mass politics If there was no deluge of assassins, there would – eventually – be a trickle. Between April 1865 and May 1866, assassination attempts were made against: President Abraham Lincoln (of the United States); Tsar Alexander II (of Russia); and Chancellor Otto Von Bismarck of Prussia (Steers, 2014; 60

Analysing ‘radicalisation’ in historical cases

Verhoeven, 2009; Schoeps, 1984). Great national controversies such as the ‘Irish question’ in British politics now began to generate a stream of self-appointed attackers, often with pre-prepared manifestos (Wilson, 2020a, 2020b, p. 125). Lone assassins came of age. We can illustrate this dynamic by zeroing in on the German Empire. Here, the Kulturkampf (‘culture struggle’) of the 1870s represented an attempt by the new German state to subordinate the Catholic Church to its project of nation-building. Control of education and ecclesiastical appointments emerged as the key points of attack. Since almost 40% of all Germans were Catholics, the Vatican could organise effective, and non-violent, civil disobedience (Abrams, 1996, pp. 22–24; Hoyer, 2021, pp. 80–86). As the sympathetic Cologne Gazette observed: ‘fanaticism amounting to assassination affects in Germany only isolated and perverse persons’ (in Times, 16 July 1874). Yet between 1872 and 1877, five assassination plots against Chancellor Otto Von Bismarck emerged (Carlson, 1972, pp. 132–133; Ross, 1998, pp. 22, 146–148). By far, the most dangerous occurred on 13 July 1874 in Kissingen, northern Bavaria. Seated in his carriage, Bismarck was just raising his hat to greet an acquaintance when a bullet grazed his rising hand (Mühlnikel, 2014, pp. 15–17). The gunman turned out to be a 20-year-old journeyman cooper named Eduard Kullmann from Magdeburg. His alcoholic father had been a seller of dried fish; and his mother was in a lunatic asylum (Times, 16 July 1874; The Argus, 26 December 1874). Upon arrest, Kullmann had dramatically declared that it was ‘better to die than that all religion go to ruin’. And yet, until recently, he himself ‘seldom or never went to church’ (Morning Post, 31 October 1874). Kullmann attributed his own radicalisation to the taunts of his fellow apprentices, who had taunted him for being Catholic. But the micro-environment mattered, too. Attracted by cheap cigars and beer, he had joined the Catholic Men’s Society in Salzwedel. According to his defence counsel, the lectures here ‘made him a fanatic’. He began to follow the course of the Kulturkampf through the Catholic newspapers at the club; the news of the arrest of the Archbishop of Posen hit hard. His behaviour changed. He started fights: and talked of civil war. By now he was attending Mass again – but at Easter 1874, he abruptly stopped going to confession. Although his attempt was still 14 weeks off, he already knew he was going to shoot Bismarck (Morning Post, 30, 31 October 1874; The Mail, 30 October 1874). At his trial, Kullmann’s radicalisation was treated primarily as a medical problem (though some efforts were made to find accomplices). Professor Reinecke did not consider Kullmann to be a religious or political fanatic; neither had he a natural disposition to crime; nor was he a common murderer. His powers of comprehension were unimpaired, but on the other hand, Kullmann was hereditarily afflicted with a moral defect. All this had ‘produced in Kullmann a certain want of independence in forming opinions, a susceptibility to external impressions, coldness of heart, and a tendency to violence and vanity’. To the police doctor, Dr. Vogt, he ‘was not a fanatic, but a “fanaticised” man, and of violent disposition’. Kullmann got 14 years imprisonment (Argus, 26 December 1874). Such a sentence would soon look distinctly lenient (Carr, 2007, pp. 48–50).

Anarchist radicalisation surges: c. 1880–1920 Once more, the tide of individual assassination attempts and micro-conspiracies rose abruptly across Europe, and this time it carried over to the New World as well. Between 1880 and 1912, anarchists killed six heads of state and helped pioneer mass casualty attacks (Carr, 2007, p. 39). 61

Tim Wilson

Several macro-forces helped set the stage for this drama. State power had become impregnable to spontaneous uprisings by the late 19th century. But class tensions had risen with industrialisation: and so, too, had the impact of mass publicity on politics. Some anarchist radicals therefore discerned an opportunity to jump-start social revolution through ‘propaganda of the deed’ attacks. In short, these were propitious conditions for the mass production of lone assassins (Wilson, 2020b). Who were these individuals? One study of the Italian gunmen and bombers concludes that they were men in their mid-twenties: ‘overall they could be described as quite ordinary members of the Italian working class’ (Pernicone & Ottantelli, 2018, p. 175). Most were of humble origins – and this also seems true of French anarchists such as François Ravachol (who, with the aid of four companions, launched a dynamite bombing campaign in the spring of 1892) and Auguste Vaillant (who bombed the Chamber of Deputies on 9 December 1893) (Carr, 2007, pp. 47–48; Merriman, 2009, pp. 137–138; Vizetelly, 1911, p. 118). Still, there were a few examples of downward social mobility as well. Alexander Berkman (who shot the Carnegie Vice President in Pittsburgh in 1892) had grown up in ‘a prosperous Jewish bourgeois family’ with servants in Vilnius (Nic Dháibhéid, 2018, p. 16). Likewise, Dr. Carl Nobiling (who blasted Kaiser Wilhelm I with a shotgun in 1878) ‘was the son of a landowner and a lady of title’ (Vizetelly, 1911, p. 48). In summary, there was no single social profile. Unsurprisingly, their experience of formal education was also highly uneven – Giovanni Passanante (who attacked King Umberto I of Italy on 17 November 1878) had once been sacked as a cook because his employer caught him reading (Pernicone & Ottanelli, 2018, p. 27). But all do seem to have been able to read and, indeed, were often guided to their targets by press announcements of forthcoming events. Anarchist killers belonged to the new age of mass literacy. What helped predispose them towards anarchism? Early humiliations seem significant. FrançoisRavachol begged as a young child, was mocked for his shabby clothes, endured dead-end jobs, and was murdered for money – all before becoming an anarchist bomber (Merriman, 2009, p. 69). Max Hödel (who shot at Kaiser Wilhelm I on 11 May 1878) ran away from home at age 12 and was publicly flogged as a thief (Carlson, 1972, p. 116). Emma Goldman (19), who assisted Berkman, had endured a tyrannical home life and been raped as a teenage shop assistant. Anarchists were often strangers in strange lands: ‘Europe at that time knew no passports, and frontiers hardly existed’ (Serge, 1951, p. 22). It was no accident that members of the Italian and Jewish diasporas featured so prominently among the anarchist killers (Jensen, 2015, p. 114). They were products of a world on the move; their social marginality was further exacerbated by the centrifuge of global migration. Berkman arrived in New York knowing no one and with no English (Nic Dháibhéid, 2018, p. 17). How, though, did they become anarchists? For individuals such as Émile Henry (Merriman, 2009, p. 40) or Victor Serge (1951, pp. 4–12), radicalisation seems to have been helped along by the involvement of other close family members in radical sub-cultures. And yet, among the Italian anarchists studied by Pernicone and Ottanelli (2018, p. 175), ‘none had been exposed to radical ideas or activities within their immediate families’. Indeed, a traumatic breakdown of relations with immediate family seems a recurrent leitmotif in trajectories of anarchist radicalisation (Goldman, 1931; Pernicone & Ottanelli, 2018, p. 121; Nic Dháibhéid, 2018, p. 16). ‘The conclusion which begins to emerge’ summarises Kedward (1971, p. 32) ‘is that anarchism had a number of social origins but that in many cases personal and intellectual reasons were predominant’. In other words, the key drivers were subjective. Sometimes it is possible to pinpoint the personal crises that accelerated the radicalisation process. For Émile Henry, it was the experience of working for his uncle, a civil engineer in Italy, or, rather, his uncle’s attempt to recruit him into 62

Analysing ‘radicalisation’ in historical cases

undercover surveillance of workers (Merriman, 2009, pp. 33–34). For Leon Czolgosz, the future assassin of President McKinley, the turning point came in 1893. During a bitter strike over wages, he and his brother ‘prayed very hard’ in keeping with their Catholic upbringing. But the strike was lost (Jensen, 2014, p. 240). What is clear is that radicalisation lent purpose to troubled lives. According to an anarchist who knew Émile Henry, ‘the only thing he thought of was to strike a blow and die’. Czolgosz was clear-eyed as to the consequences of his choices: ‘I realized I was sacrificing my life’ (in Jensen, 2014, pp. 41, 245). Victor Serge (1951, p. 23) recalled his time on the fringes of the Bonnot Gang: ‘Anarchism swept us away completely because it both demanded everything of us and offered us everything’. Such an ordering of priorities seems eloquent: cause before rewards. How did contemporaries understand anarchist radicalisation? Some commentators saw the radicalised as obedient recruits to tightly organised (but also fluidly transnational) conspiracies. Such assumptions constituted a kind of official common sense during the early years. When Passanante attacked King Umberto I in 1878, the possibility that he might have acted alone was one ‘that the authorities refused to concede’ (Pernicone & Ottanelli, 2018, p. 27). What is more striking is how hard such thinking died. Thirty years later, the psychiatrist Dr. Edward Spitzka discerned ‘an international army organized to wage war upon society, directed by skilful generals’. He hinted at a ‘directing mind in London’ presiding over ‘an international society of murder’ (New York Times, 6th June 1909). And yet violent anarchist conspiracies remained small. Perceptive observers noticed that Everything at present rests with the ‘group’, which is, at the same time, very small and of an extremely fluctuating character. Five, seven, or at most a dozen men unite in a group according to occupation, personal relationships, propinquity of dwelling, or other causes; only after a certain time to separate again. (Zenker, 1897, p. 181) Alexander Berkman’s attack on Frick fits this pattern exactly: he struck alone, but had been crucially assisted by four accomplices (Nic Dháibhéid, 2018, p. 20). Even more puzzling was the appearance of lone actors (or solitaires in French police slang) (Coolsaet, 2022, p. 176). Standard interpretations of the solitaires emphasise that ‘morbid thirst for notoriety which is the bane and curse of our modern civilization’ (Lord Salisbury in Jensen, 2014, p. 225). Such thinking led inexorably away from the political motivations of the assassins and towards an exclusive focus on their mental health. Indeed, the most striking feature of the contemporary study of the violent anarchists is that it was so medicalised. Dr. Christison held that such a monstrous conception and impulse as the wanton murder of the President of the United States [by Czolgosz], arising in the mind of so insignificant a citizen, without his being either insane or a degenerate, could be nothing short of a miracle. (in Lydston, 1904, pp. 255–256) Cesare Lombroso, the distinguished Italian criminologist, subscribed to biological determinism: anarchists were ‘always epileptics [and] moral madmen’. Despite its obvious novelty, violent anarchism was frequently interpreted as a pre-modern survival. Lombroso thus ‘asserted that anarchist ideology . . . ideas harkened back to prehistoric society before the rise of institutions of authority’ (in Pernicone & Ottanelli, 2018, pp. 2–3). Similarly, 63

Tim Wilson

George Frank Lydston (1904, pp. 255–256), Professor of both Genito-Urinary Surgery and Criminal Anthropology, concluded that ‘the Italian murderer is often the result of non-progression of the race from the days of barbarism’. Medical investigation generated elaborate agnosticism as well. After examining Czolgosz, Carlos Macdonald (1902), Professor of Mental Diseases and Medical Jurisprudence, recorded: there was no insanity in his family, and . . . he had not suffered any serious illness or injury during his life-time; . . . he had never been subject to fits, spasms or vertigo; . . . he usually ate and slept well, and . . . his bowels were always regular. He admitted having had sexual intercourse with women, but denied masturbation or other unnatural practices. Dr. Spitzka (1902), in performing an autopsy on Czolgosz’s brain, found nothing ‘that could have been at the bottom of any mental derangement’. He concluded that Czolgosz had been ‘socially diseased and perverted, but not mentally diseased’. By contrast, those well placed on the left to dissect the dynamics of anarchist radicalisation received less attention. In 1884, the Marxist theoretician Karl Kautsky asserted that state repression was a key driver in ‘breeding’ [züchtet] anarchists (Gabriel, 2014, pp. 97, 244). In 1897, the Austrian journalist Ernst Zenker foreshadowed ‘relative deprivation theory’: the average man does not much mind his rich fellow-man riding in his carriage while he himself cannot even pay his tram-fare; but that he should be abandoned by society to every chance official of justice, as a prey that has no rights . . . that makes his blood boil. (in Coolsaet, 2022, pp. 176, 180) In short, how anarchist attackers were created remained largely mysterious to their contemporaries. And as anarchist terror itself began to fade out in the 1920s so, inevitably, did public and professional curiosity. Once more, lone attackers became a sporadic phenomenon. No further transnational wave of micro-radicalisations on this spectacular scale was to occur again until the early 21st century.

Radicalisation in the 20th century Mass atrocities in the 20th century did provoke one new transnational phenomenon: radicalised avengers. An early example was Shalom Schwartzbard. Described by his psychiatrist as ‘an idealist, pursuing wholeheartedly the course of justice’, Schwartzbard shot dead Symon Petliura in a Paris restaurant on 25 May 1926. As ruler of Ukraine, Petliura organised the anti-Jewish pogroms of 1919 (Nedava, 1982, pp. 30, 33). Similar killings followed. On 4 February 1936, David Frankfurter shot dead Wilhelm Gustloff, the leader of the Nazi Party in Switzerland. Seeing his Jewish uncle abused on a visit to Berlin had affected him deeply (Fuhrer, 2010, pp. 54–59). On 7 November 1938, Herschel Grynszpan shot dead Ernst vom Rath in Paris in protest at the deportation of 18,000 Polish Jews, including his parents, from Germany. On 13 March 1940, in London, Udham Singh gunned down Sir Michael O’Dwyer – the former Lieutenant Governor of the Punjab during the Amritsar massacre of 1919. On 4 June 1968, Sirhan Bishara Sirhan, a Palestinian, killed Senator Robert Kennedy for supporting Israel (Bell, 1979, pp. 71). As sporadic gate-crashers of international politics, such lone attackers provoked little official curiosity. One notable exception was Georg Elser, whose bomb came close to killing Adolf Hitler 64

Analysing ‘radicalisation’ in historical cases

at the Bürgerbräukeller in Munich on 8 November 1939 (Kershaw, 2000, pp. 271–274). Given the professionalism of the attempt, the Nazi elite’s first instinct was to suspect a British plot. But evidence of any wider conspiracy was non-existent, forcing attention back onto Elser. Even Hitler asked: ‘What type are we dealing with, from the point of view of criminal psychology?’ (in Hochhuth, 1984, p. 139). At 36, Elser remained unmarried; ‘had little education to speak of, did not read books, and scarcely bothered with newspapers’. He had little interest in politics, beyond leftist sympathies. But he was clearly also a meticulous craftsman. He had spent 30 nights in the Bürgerbräukeller hollowing out a pillar in which to conceal his bomb (Kershaw, 2000, pp. 271–273). The Head of the Gestapo, Heinrich Müller, who personally interrogated Elser, probably got closest to uncovering the origins of his radicalisation: ‘he hates Hitler because one of his brothers who had been a Communist sympathiser was arrested and put in a concentration camp’ (in Hochhuth, 184, 136). A hypnotist’s report concluded Elser ‘had psychotic compulsions, related especially to technical matters, which sprang from an urge to achieve something really noteworthy . . . reinforced by a thirst for vengeance for the alleged injustice which had been done to his brother’ (in Hochhuth, 1984, p. 145). In summary, Nazi torture and pseudo-science pieced together an account of Elser’s radicalisation that was both dynamic and multi-dimensional.

Conclusion What can this brief survey of historical cases contribute to our understanding of micro-radicalisation processes? Such a survey certainly expands the universe of possible case studies. One basic observation is the sheer range of possible causes to kill for. Radicalisation under very different ideological banners seems, however, to have worked in broadly similar ways, though ideology did mould end-stage action. Religious zealots often seemed to feel self-justification was superfluous. Secular killers tended to invest more effort in public persuasion. In 1628, John Felton, setting out to murder the Duke of Buckingham with his pre-prepared manifestos, was ahead of his time. Many of the trends uncovered by social psychology turn up here: group grievance, personal grievance, risk and status-seeking, ‘disconnection’, and escape from personal problems (Moskalenko & McCauley, 2020, p. 100). But then, as now, such observations only take us so far. Any attempt to move beyond subjective motivations hits the classic problem of over-determination. The hardships that feature so prominently in the biographical sketches of the radicalised offered here have been the fate of millions all down recorded history. And yet ideological killers remain rare. Ideological killers of this self-motivated type always remain rare, but not absolutely or consistently so. A scan of long-term horizons therefore reveals some pronounced bunching effects. Before the late 16th century, such actors simply did not exist. Thereafter, they appeared only sporadically. And yet such micro-radicalisations have also surged spectacularly in just three periods over the last 450 years: between c. 1580 and 1630; from c. 1880 to 1920; and again since c. 2005. Why? Ideology mattered because it gave violent action grandiose meaning. Western Christendom collapsing into schism; the ‘social question’ during the Industrial Revolution; the erosion of traditional authority across the Islamic world – all these episodes of ideological turmoil sparked flurries of micro-radicalisation. Such crises were remarkable for both their transnational reach and social penetration. That matters because ideology becomes most potent as a catalyst for violence when it is embedded ‘in the pouches of micro-level solidarities’ (Malesevic, 2022, p. 91). ‘Shall we always, gentlemen, talk and never do anything?’ Thomas Percy asked his friends, the future Gunpowder Plotters, in May 1604 (Fraser, 1996, p. 98). 65

Tim Wilson

Technology mattered here, too. Each of the ideological conflicts earlier was (partially) driven by a communications revolution: the printing press in the 16th century; the mass circulation daily press of the late 19th century; and the rise of the Internet age in the early 21st century. Each generated ferocious wars of words. Each made new worlds seem within reach. And each disseminated example of violence might just help achieve them (Vizetelly, 1911, p. 86). Against the longer history of micro-radicalisation, two other macro-changes seem worth noting. First, the rise of the coercive infrastructure of the modern Western state has helped squeeze out opportunities for large-scale violence: ‘Time and again, it is only society’s dreamers and deranged who have dared to mount any kind of sustained violent challenge to the state’ (Wilson, 2020b, p. 21). Second, micro-radicalisation has been accelerated by democratisation. Mass religious passions first inspired a handful of lowly-born assassins in the early modern period. But in such a hierarchically structured society, their appearance as political actors constituted a mysterious phenomenon. In 1610, the French aristocrats who interrogated Ravaillac did not know what to make of him. Only in the later 19th century, as individual political choices became validated through the expansion of electoral franchises, did micro-radicalisation increase in tempo. In summary, ideological lone killers belong to the age of mass politics. What, finally, of future radicalisation? A political phenomenon with deep historical roots in Western societies will not disappear soon. And since neither communication revolutions nor ideological upheavals – the twin forces that drove previous peaks of micro-radicalisation – look set to disappear, we may confidently expect more trouble ahead. Such trouble is likely to appear under increasingly diverse banners. Radicalisation into idiosyncratic causes is not itself new, as François Damiens demonstrated. Even in 1757, his attack inspired fears of repetition. But in an age of social media, such violent repetition is not just far more likely. It is all but certain. One point more. Looking back on the attempts by our most distinguished predecessors to understand radicalisation, it is hard not to be impressed by what they did not see, and this despite their contemporary access to detailed information now lost to time. Yet we cannot assume these experts are our intellectual inferiors. Why do we think we are wiser about the radicalisation of our own age?

*** References Abrams, L. (1996). Bismarck and the German empire, 1871–1918. Routledge. Aikin, L. (1833). Memoirs of the court of King Charles the first (Vol. 1). Longman. Alford, S. (2013). The watchers. Penguin. Bell, J. B. (1979/2005). Assassin. Transaction. Bellany, A. (2008). John Felton. DNB. Burton, R. (1620/2001). The anatomy of melancholy. New York Review Books. Carr, M. (2007). The infernal machine. The New Press. Carlson, A. R. (1972). Anarchism in Germany (Vol. I). The Scarecrow Press. Coolsaet, R. (2022). When do individuals radicalize? In D. Muro & T. Wilson (Eds.), Contemporary terrorism studies. Oxford University Press. De Kley, D. K. (1984). The Damiens affair and the unravelling of the Ancient Regime, 1750–1770. Princeton University Press. Descimon, R. (2009). Chastel’s attempted regicide (27 December 1594) and its subsequent transformation into an ‘affair’. In A. Forrestal et al. (Eds.), Politics and religion in early bourbon France. Palgrave MacMillan. Ford, F. L. (1976). Assassination in the eighteenth century: The dog that did not bark in the night. Proceedings of the American Philosophical Society, 120(3), 211–215.

66

Analysing ‘radicalisation’ in historical cases Ford, F. L. (1985). Political murder. Harvard University Press. Fraser, A. (1996). The gunpowder plot: Terror and faith in 1605. Weidenfeld & Nicolson. Fuhrer, A. (2010). Tod in Davos. Metropol. Gabriel, E. T. (2014). Assassins and conspirators. NIU Press. Garrett, J. (1980). The triumphs of providence. Cambridge University Press. Goldman, E. (1931, 2018). Living my life. Knopf. Goldsmid, E. (Ed.). (1885). The trial of Francis Ravaillac. Forgotten Books. Greengrass, M. (1987). France in the age of Henri IV. Longman. Haynes, A. (1994/2010). The gunpowder plot. History Press. Hochhuth, R. (1984). Tell’ 38. Little, Brown and Co. Horne, A. (2002). Seven ages of Paris. Pan. Hoyer, K. (2021). Blood and Iron: The rise and fall of the German empire 1871–1918. The History Press. Jardine, L. (2005). The awful end of prince William the silent. Harper Perennial. Jensen, R. B. (2014). The Battle against anarchist terrorism. Cambridge University Press. Jensen, R. B. (2015). Anarchist terrorism and counter-terrorism in Europe and the world, 1878–1934. In R. Law (Ed.), The Routledge history of terrorism. Routledge. Kedward, R. (1971). The anarchists. Macdonald. Kershaw, I. (2000). Hitler: 1936–45. Allen Lane. Kidd, C. (2012). Assassination principles in Scottish political culture: Buchanan to Hogg. In C. Erskine & R. Mason (Eds.), George Buchanan. Ashgate. Knecht, R. (2014). Hero or tyrant? Henry III, king of France, 1574–89. Ashgate. Loomis, S. (1964/1970). Paris in the terror. Penguin. Lydston, G. F. (1904). The diseases of society. Lippincott. MacDonald, C. F. (1902). The trial, execution, autopsy and mental status of Leon F. Czolgosz, Alias Fred Nieman, the assassin of president McKinley. Medical News, 80(1). Malesevic, S. (2022). Why humans fight. Cambridge University Press. Malthaner, S. (2017). Radicalization: The evolution of an analytical paradigm. European Journal of Sociology, 58(3), 369–401. Marsden, S. (2020). Radicalization: Trajectories in research, policy and practice. International Journal of Conflict and Violence, 14(2), 1–6. Merriman, J. (2009). The dynamite club. R Books. Mitford, N. (1969). The sun king. George Rainbird. Moskalenko, S., & McCauley, C. (2020). Radicalization to terrorism. Oxford University Press. Mousnier, R. (1973). The assassination of henry IV. Faber. Mühlnikel, M. (2014). Fürst, sind Sie unverletzt? Schöningh. Nedava, Y. (1982/2015). Some aspects of individual terrorism: A case study of the Schwartzbard affair. In Y. Alexander & K. Myers (Eds.), Terrorism in Europe. Helm. Nic Dháibhéid, C. (2018). Terrorist histories: Individuals and political violence since the 19th Century. Routledge. Parker, G. (2015). Imprudent king. Yale University Press. Pernicone, N., & Ottanelli, M. (2018). Assassins against the old order: Italian anarchist violence in Fin de Siècle Europe. University of Illinois Press. Pettegree, A. (2015). The invention of news. Yale University Press. Pettegree, A. (2016). Brand Luther: 1517, printing, and the making of the reformation. Penguin. Ross, R. (1998). The failure of Bismarck’s Kulturkampf. Catholic University of America Press. Schoeps, J. (1984). Bismarck und sein Attentäter. Ullstein. Serge, V. (1951/2012). Memoirs of a revolutionary. New York Review of Books. Spitzka, E. A. (1902). The postmortem examination of Leon F. Czolgosz. Medical News, 80(1). Steers, E. (2014). Lincoln’s assassination. Southern Illinois University. Te Brake, W. (2017). Religious war and religious peace in early modern Europe. Cambridge University Press. Thomas, K. (1991). Religion and the decline of magic: Studies in popular beliefs in sixteenth- and seventeenthcentury England. Penguin. Verhoeven, C. (2009). The odd man Karakozov: Imperial Russia, modernity and the birth of terrorism. Cornell University Press. Vizetelly, E. (1911). The anarchists. Bodley Head. Warner, J. (2005). John the painter. Profile.

67

Tim Wilson Wedgewood, C. (1945). William the silent. Cape. Williamson, G. (2000). What killed August Von Kotzebue? Journal of Modern History, 72(4), 890–943. Wilson, T. K. (2020a). Rightist violence. ICCT Research Papers. Wilson, T. K. (2020b). Killing strangers. Oxford University Press. Wizeman, W. (2008). John Somerville. DNB. Zamoyski, A. (2014). Phantom terror. Collins. Zenker, E. (1897/2014). Anarchism. Outlook.

*** Newspapers Argus, 26 December 1874 The Mail, 30 October 1874 Morning Post, 31st October 1874 New York Times, 6th June 1909 Times, 16th July 1874

68

PART II

Analytical approaches to radicalisation

5 THE ROLE OF IDEOLOGY IN RADICALISATION Kumar Ramakrishna

In February 2011, the former British Prime Minister David Cameron argued that ‘the ideology of extremism is the problem’ and that ‘terrorism’s root lies in the existence of this extremist ideology’ (Forsyth, 2011). Cameron was reflecting an oft-articulated academic and policy notion on the role of ideology in radicalisation into violent extremism.1 To be sure, Cameron’s view of ‘terrorism’s root’ did not go uncontested (Weaver, 2015). The academic literature has engaged extensively with the notion of the ‘root causes’ of violent extremism and tends to eschew single-factor explanations. As one study puts it, terrorists may be ‘deprived, uneducated, affluent and from both sexes’, while at the same time, terrorism can occur ‘in developed and undeveloped countries, in a variety of regimes’ and ‘encompasses ideology and religion’ (Lia & Skjølberg, 2004). Terrorism scholar Louise Richardson (2006), rather than suggesting that ideology is the root per se, argues that ‘legitimizing ideology’ in any case is very much a key ingredient of the ‘lethal cocktail’ of factors giving rise to terrorism. In this respect, there has been much written about the ways in which ‘Salafi-Jihadi’ ideology in particular is to blame for the emergence of Al-Qaeda, its affiliates, and offshoots like ISIS worldwide (Maher, 2016; Wagemakers, 2016). Such assessments appear to also dominate the policy and scholarly discourse on violent Islamist extremism in Southeast Asia, where the current author is from, incidentally. A report in the early 2010s on Indonesian prison radicalisation concurred that at the ‘heart of the current wave of global Islamist terrorism is a legitimizing extremist narrative that resonates with individuals from Morocco to the Philippines’ (ASPI/RSIS, 2011). A contemporaneous Australian Strategic Policy Institute study iterated that ‘despite tactical counterterrorism successes, a changed geopolitical landscape and the strategic weaknesses of the al-Qaeda movement, the unchecked ideology of violent jihadism ensures that homegrown radicalization’ will continue to be a focus of policy concern’ (Khalil, 2012). Similarly, in recent years, academic and policy concern has become increasingly focused on the role of far-right ideology, incorporating various ideas animating ‘a bewildering amalgam of White nationalists, some White Christian evangelicals, racists, anti-government militias, misogynists, anti-globalisers, and anti-vaxxers, amongst others’, seeking to exploit the ‘global social and political upheaval’ – including the current pandemic – to promote intolerance ‘and at times inflict violence’ (Ramakrishna, 2021a). But what is the role of ideology in the radicalisation process? Is it really the root, or does it play some other role? This chapter addresses the issue in the following way. The next section unpacks 71

DOI: 10.4324/9781003035848-7

Kumar Ramakrishna

the concept of ideology, importantly, showing how it is inextricably linked and with, and emerges from, the wider cultural context. The third section then examines how ideology is viewed by various scholars to function at the individual level, while the fourth examines how it is said to do so at the group level. The fifth and final section then examines relatively recent and novel discussions about how ideology is said to function within globally dispersed violent extremist networks, viewed as complex, self-organising adaptive systems. The chapter concludes by offering one possible, albeit rather more philosophical, answer to the question of ‘terrorism’s root’.

Ideology unpacked To better grasp what ‘ideology’ means, it is helpful to take a step back and examine the wider theoretical construct of ‘culture’. In the seminal research of Dutch social psychologist Geert Hofstede (Hofstede & Hofstede, 2005), culture consists of the ‘unwritten rules of the social game’. Culture, understood as ‘patterns of thinking, feeling and potential acting’, is ‘learned throughout’ an individual’s lifetime, starting with childhood. More than that, Hofstede notes that such social learning within a community can lead to a ‘collective programming of the mind that distinguishes the members of one group or category of people from others’. Hofstede’s view of culture was shaped by the French sociologist Pierre Bourdieu’s concept of habitus: ‘a socially constituted system of cognitive and motivating structures’ (Juergensmeyer, 2000; also Hofstede & Hofstede, 2005). Other cross-cultural psychologists concur that when it comes to the concept of culture, rather than emphasising any particular individual’s ‘values, beliefs, or other attributes’, it is more useful to study the extent to which larger social collectives such as ‘nations can be characterized by sets of shared beliefs, meanings or mentalities’ (Smith et al., 2006). Mark Juergensmeyer (2000) concurs, adding that the notion of culture draws attention to how ‘ideas and social groupings’ coalesce strategically to produce ‘terrorist acts’. An important aspect of culture is highlighted by Ervin Staub (1989), who observes that culture is shared largely without awareness. That is, culture profoundly, if tacitly, shapes the beliefs and behaviour of individual members in ways that they are generally unaware of and to such an extent that they believe that ‘their preferences are obvious, natural and taken-for-granted’ (Nasir et al., 2010). Academic research since the mid-1980s suggests that culture should not be regarded as a coherent, integrated system of meanings that directly motivates individual behaviour. Rather, cultural meanings exist mainly to enable individuals to rationalise or make sense of their behaviour after the fact. In other words, culture is ‘best viewed as a loose repertoire of justifications that rationalize or make sense of the choices that individuals make in their lives’ (Vaisey, 2009; DiMaggio, 1997). The argument is that culture functions basically as a ‘tool-kit’ of ‘symbols, stories, rituals, and world-views, which people use in varying configurations to solve different kinds of problems’ (Swidler, 1986). This unsystematic, pragmatic tool-kit approach occurs because human beings are rather ‘cognitively impoverished’ and unable to preserve coherently in their minds highly elaborated cultural systems (Lizardo & Strand, 2010). In other words, humans are in general ‘remarkably bad at giving consistent reasons for their behavior’ (Vaisey, 2009) and quite capable of holding on to contradictory and inconsistent cultural schemata – or interpretive and representational mental structures – at the same time (DiMaggio, 1997). As Vaisey (2009) explains, while individuals function at ‘two primary levels of consciousness – deliberative and automatic’, in effect ‘most of our cognitions occur below the level of conscious awareness’; that is, the ‘unconscious schemas that people develop through life experience’ actively shape emotions and intuitions, with individuals reflexively drawing upon ‘cultural-cognitive structures’ that are ‘built up out of experience’ to respond to situations. These mental structures function as ‘deep, largely unconscious networks of neural associations that facilitate perception, interpretation, and action’. 72

The role of ideology in radicalisation

In the context of the current discussion, therefore, culture can be understood as the largely taken-for-granted, unconscious impact of ‘collective programming’ that permeates what Mark Juergensmeyer (2000) calls the ‘culture of violence’ and psychoanalyst Willard Gaylin (2003) terms the ‘culture of hatred’. Understanding ‘these cultures of violence’, James W. Jones (2008) adds, ‘is clearly a crucial part of understanding religious terrorism’. Within a culture of violence, out-group prejudice and even hate develop through a process of what Matthew Williams (2022) calls ‘cultural feeding’ of ‘negative stereotypes’ of outgroups, that begins ‘early in life’ and remains ‘unchecked and unnoticed’ – meaning individuals ‘remain unaware of its effect’ unless an effort is later made to recognise it. What should be recognised here is that free-floating, inchoate, outgroup biases, prejudices, and anecdotes must be deliberately and systematically weaponized in the form of ‘calculated propaganda’ so as to ‘mobilize the population’ towards out-group violence (Gaylin, 2003). It is this notion of calculated propaganda that moves us beyond the broader realm of culture into the more specific, action-oriented domain of ideology. While culture, as noted, is unwritten and shared largely without mass conscious awareness, ideology is primarily a consciously held and relatively well-articulated set of beliefs and ideas (Waller, 2005). Ann Swidler (1986) adds that while cultural traditions as seen are ‘taken for granted’ and seemingly ‘inevitable parts of life’, ideology, in contrast, is ‘a highly articulated, self-conscious belief and ritual system, aspiring to offer a unified answer to problems of social action’. Terrorism scholar Alex Schmid (2013) offers a similar interpretation of ideology as those ‘patterns of beliefs and expressions that people use to interpret and evaluate the world in a way designed to shape, mobilise, direct, organize and justify certain modes and courses of action’. For his part, John Wilson (1973) similarly refers to ideology as ‘both a cognitive map of sets of expectations and a scale of values in which standards and imperatives are proclaimed’, and that ideology ‘thus serves both as a clue to understanding and as a guide to action, developing in the mind of its adherents an image of the process by which desired changes can best be achieved’. It is worth noting here that a somewhat related concept to that of ideology in the extant literature is what social movement theorists would call frames. Frames offer ‘interpretive schemata that provide a cognitive structure for comprehending the surrounding environment’, explaining ‘causation, evaluating situations, and offering prescriptive remedies’ (Wiktorowicz, 2005). That said, some scholars argue that the concepts of ‘frame’ and ‘ideology’ are distinct and should not be used interchangeably (Oliver & Johnston, 2000). Reiterating the close nexus between ideology and the wider culture on which it tends to draw, Wiktorowicz (2005) emphasises that to effectively mobilise action, an ideology must demonstrate ‘fidelity with cultural narratives, symbols, and identities’. Agreeing, Glazzard (2017) argues that: ‘Culture shapes ideology, and in looking for explanations for terrorist behaviour’, there is a need to ‘pay attention to cultural heritage’. After all, ‘even from a narrow study of terrorist statements’, it is apparent that ‘themes repeat themselves, but so do figures of speech and even particular verbal phrases’. At the same time, while it derives its content and authenticity from a wider culture, ideology’s role in radicalisation is to in a sense weaponize the random and unstructured cultural biases, prejudices, and sentiments of in-group members by concretely diagnosing the problem they are purportedly facing, identifying clearly the ‘enemies’ putatively responsible for their collective predicament, and finally suggesting and legitimising courses of action, including possibly an armed struggle to ‘defend’ the in-group against those ‘evil’ enemies (Ramakrishna, 2005). Swidler (1986) adds that deliberately constructed ideological formulations become prominent, especially during times of social and political upheaval, ‘when competing ways of organizing action are developing or contending for dominance’; thus, ideology is in a sense ‘contested culture’. A key aspect of ideological formulation involves integrating new episodes into the master ideological narrative as coherently as possible so that while specific stories or episodes ‘come and 73

Kumar Ramakrishna

go, change and morph’, the overall master narrative that has ‘endured the test of time’ remains resilient (Halverson, 2011). In sum, what seems to emerge from an analysis of the wider literature at this juncture is the notion that the basic role of ideology in the radicalisation process is to selfconsciously explain and legitimise, in culturally familiar ways, violence perpetrated by in-group members against putatively evil out-groups. In the following sections, we shall see how such ‘bad ideology’ (McCauley & Moskalenko, 2011), if not necessarily the ‘root cause’, is certainly part of the ‘lethal cocktail’ that generates terrorist action (Richardson, 2006). We shall be exploring this issue at three levels of analysis: the individual, the group, and the system.

The individual level In their well-known ‘two-pyramids model of radicalizations’, the social psychologists Clark McCauley and Sophia Moskalenko (2017) argue that ‘radicalization to extremist opinions is psychologically a different phenomenon from radicalization to extremist action’. They add elsewhere that individual ‘beliefs alone are a weak predictor of action’ and that there are ‘many paths to radicalization that do not involve ideology’, with some individuals joining a ‘radical group for thrill and status, some for love, some for connection and camaraderie’, while personal and group-based grievances ‘can move individuals toward violence’ (McCauley & Moskalenko, 2011). Agreeing, Andrew Glazzard (2017) opines that ideological explanations for violent radicalisation are ‘at best a gross over-simplification’ that ignore important alternative factors ranging ‘from socioeconomic grievances to the lure of adventure to the primary human need for survival’, as well as ‘identification with a group, socialization, and the effect of civil conflicts’. In a media interview, terrorism scholar Andrew Silke noted that ‘at the early stages those that become involved in terrorism have a very limited understanding of the ideology – they are not scholars’ (Weaver, 2015). This point is driven home by reports that certain individuals preparing to fly to Syria to fight for ISIS had ordered copies of The Koran for Dummies and Islam for Dummies online prior to their departure. This appeared to suggest that any systematic ideological basis for their desire to join the fighting overseas was seriously deficient (Batrawy & Hinnant, 2016). A similar situation exists on the far right. Julia Ebner (2020) has shown that rather than playing a leading role, far-right ideology works best when individuals are facing ‘moments of weakness’, as the vulnerable, lost and impressionable women she observed on an extremist online counselling forum were ‘gradually indoctrinated’ into the ‘norms and ideologies of their coaches and peers’ who adroitly linked ‘the group ideology to personal questions’. This analysis fits well with Petter Nesser’s (2015) description of ‘misfits’ and ‘drifters’ falling prey to skilled ‘identity entrepreneurs’ and their ‘proteges’. This is not to suggest, however, that ideology has an unimportant role in the individual radicalisation process. The literature suggests that rather than being a root cause, the impact of ideology at the individual level is as a ‘source of justification’ or a ‘rationalization for violence’ (McCauley & Moskalenko, 2011). Louise Richardson (2006) alludes to a similar idea in her notion of a ‘legitimizing ideology’. Recent research adds that the intensity with which such violence-legitimising rationalisations are held also matters: ‘stronger attitudes’ shaped by such rationalisations are more predictive of violence (Khalil et al., 2019). More precisely, it has been suggested that ideology, understood as a ‘violence-justifying narrative’, is tied to an individual’s quest for significance. That is, it ‘arouses the exposed individuals’ quest for significance by highlighting and dramatizing the humiliation one’s group may have suffered’, as well as forging ‘the critical link between violence and significance, suggesting that engagement in the former will bring about the latter’ (Kruglanski et al., 2019). This fundamental human quest for significance plays out in both the Islamist and farright extremist domains. For instance, Islamist extremist ideologues argue that engaging in jihad 74

The role of ideology in radicalisation

against the Jews and Crusaders said to be oppressing the ummah worldwide is a worthy and noble enterprise that would ‘earn an individual the group’s everlasting gratitude and accord her or him a coveted hero and/or martyr status’ (Kruglanski et al., 2019). Meanwhile, individuals motivated by far-right, white supremacist ideologies of hate towards non-white minorities are likely compensating for deep-rooted ‘feelings of inadequacy’ due to troubled family backgrounds (Williams, 2022). Anti-Semitic white supremacist groups in the United States, for instance, tend to target ‘emotionally disturbed youths between the ages of 10 and 18’, who are regarded as particularly susceptible to such ideas (Lavin, 2020). Joan Smith (2020) noted that the terrorists she studied, be they Islamist or far-right, tended to display ‘symptoms of insecurity, resentment and rage before they turned to terrorism, suggesting that the extremist ideology they discovered was a ‘justification’ for acting on pre-existing violent impulses’. Smith points out, for instance, that Thomas Mair, who murdered the Labour Party MP Jo Cox in June 2016, was not just a ‘committed extremist’ with links to the far right; he was also someone who had had a history of violent misogynistic impulses and mental illness (although it was assessed that the latter was not a factor when he attacked the MP). Other scholars observe that mental illness, often as a result of childhood trauma, has likely rendered some individuals more vulnerable to violence-justifying ideologies later in life (Williams, 2022). It is equally important to underscore that it is not particularly crucial for radicalised individuals to demonstrate a strong grasp of relevant ideological themes with a high degree of comprehension and accuracy. As noted earlier, ‘they are not scholars’ (Weaver, 2015). In a related vein, Lorne Dawson (2015) in this respect argues that ‘faulty theology is not a reliable indicator of the degree or primacy of religion in someone’s motivations’ – what matters is the ‘perpetrator’s conception of the actions’. It has been widely observed that terrorists often ‘develop an instrumentalized cut-and-paste interpretation of a given ideology in order to justify their recourse to violence’ (Holbrook & Horgan, 2019). As is the case for broader cultural tropes, ideological beliefs need not be fully logical or entirely consistent. As Holbrook and Horgan (2019) argue, it is perhaps the emotional impact rather than precise theological/ideological content that plays the decisive role in the radicalisation process at the individual level: A nasheed [Islamic song] celebrating the rewards of the martyrs once they ascend to the highest stages of heaven may evoke a powerful emotional response, irrespective of the recipient’s cognitive capacity to dissect any theological underpinnings that might be associated with it. Indeed, we know that they are popular among individuals who do not even understand the lyrics, simply because they strike an emotional cord [sic] or reflect ‘membership’ of a social collective deemed desirable. (my emphasis) The notion that ideology reflects membership in a ‘social collective deemed desirable’ is pertinent. Holbrook and Horgan (2019) posit that ‘the role of ideology can best be explained through highlighting its social components’, not just, as argued, ‘the collective maps and shared perspectives that help us make sense of the world’, but also importantly, defining ‘who is or is not part of our community’. In other words, ideology also has a role to play in radicalisation at the group level. This is a key issue to which we now turn.

The group level It is important to understand what is meant by ‘group’. A group in this context refers to ‘a number of people in contact with each other’; a group is thus not the same as a ‘category’, which 75

Kumar Ramakrishna

refers to people ‘who, without necessarily having contact, have something in common’, such as, for instance, ‘all women managers, or all people born before 1940’ (Hofstede & Hofstede, 2005). More importantly, what defines a group in the current discussion is ‘a specific set of connections between people in the group’ (Christakis & Fowler, 2011). In an evolutionary sense, a group represents ‘the maximum number of humans in which every member knows every other member, knows whether they are friendly or hostile, and knows the relationships among them’. As Christakis and Fowler (2011) argue, in a face-to-face group, relationships among members are stable and can be ‘picked up again on meeting after an absence without any need to re-establish where you stand’. In this regard, what is the role of ideology in radicalisation at the group level? As mentioned, some scholars suggest that a key contribution is delimiting the boundaries of the in-group in relation to an out-group by making explicit ‘who is or is not part of the community’ (Holbrook & Horgan, 2019). Scholars such as Sara Savage (2011) concur, adding that most ‘studies of radicalization leading to violence show there is a period of immersion, at some point, in an extremist group’, and that individuals ‘are rarely radicalized in a vacuum’. The extant literature sheds additional useful light on how ideology operates within the group context. It does so by highlighting the so called echo chamber effect (Sageman, 2008) where outgroup prejudice is intensified and potentially transformed – via what Cass Sunstein (2009) calls ‘group polarization’ – into hate-filled violence. This requires elaboration. Initially, group members, on balance, while highly prejudiced against the out-group, may nevertheless remain cautious about engaging in actual physical violence against out-group members. However, intense group discussion within physically and psychologically insulated ‘free spaces’ set apart from the wider societal mainstream – such that no alternative ideas or views can easily circulate within the confines of the group – tends to facilitate group polarisation in the form of a ‘risky shift’ from latent prejudice to outright violent action (Sunstein, 2009; also Fink, 2014). McCauley and Moskalenko (2011) add that ‘isolated groups – terrorist groups, youth gangs, religious cults, soldiers in combat – have unchecked power to determine value and meaning’, and hence ‘group isolation opens groups to mutual influence in group dynamics’ in radicalisation towards out-group violence. The impact of group isolation on fostering group polarisation and radicalisation is well documented. For instance, the Al-Qaeda terrorist Mohammed Mansour Jabarah, a Canadian of Kuwaiti extraction, trained at the isolated Al Farooq camp in Afghanistan – a location chosen not merely to ‘conceal the camp from attackers’ but also to help ‘the instructors mould their young pupils into indoctrinated radicals by keeping them free from outside distractions and focused on the task at hand’ (Bell, 2005). The far right also appears to grasp the importance of isolated spaces for the purposes of ideological indoctrination. White supremacists have set up ‘summer camps for children in Germany and Ukraine’, while an annual summer camp for ‘far right Identitarians in France’ seeks to train future ‘militants’, focusing on ‘those twenty to twenty-five years old and combining physical-fitness training with history classes and simulated street riots’ (Miller-Idriss, 2020). The echo chamber effect thus intensifies and focuses the impact of ideology at the group level, transforming a small group of believers into something akin to a religious cult. The clinical professor of psychiatry Arthur Deikman (2003) identified four features of such cults: dependence on a leader; utter compliance with the group; suppression of dissent; and finally, devaluation of outsiders. The devaluation of outsiders is shaped by ideological indoctrination and, if successful, generates out-group hate. Hate, it should be recognised, is much more than anger. While anger is fleeting and ephemeral, hatred is ‘a sustained emotion of rage’ that is obsessive and preoccupies an individual (Gaylin, 2003). The British psychologist Russell Razzaque (2008) paints a vivid

76

The role of ideology in radicalisation

picture of the isolated small-group echo chamber in the throes of ideological polarisation towards out-group hatred and potentially, violence: There are few places where [the men] can really speak their mind, unleash the ferocity of their passion and anger and feel understood, supported, even encouraged. This is why their hideaway becomes so important – it becomes like a tunnel, transporting them to another dimension. . . . The initiates spend more and more time there, neglecting work, obligations, friends and family. . . . They would hold their hands in prayer . . . have tiffs and arguments and then make up with one another, and move in a pack, unable to perform even the smallest tasks without being together. . . . Spending more and more time in each other’s company leads ultimately to de-individuation, in which each identity begins to merge until there is very little left of the individual outside the identity of the group as a whole. They feed off each other’s notions, pushing one another’s worldviews inexorably into further and further extremes. (my emphasis) Razzaque’s description of the ideological indoctrination process within the group reaching a point where ‘each identity begins to merge until there is very little left of the individual outside the identity of the group as a whole’ suggests another critical element of group-level radicalisation worth noting: identity fusion (Swann & Buhrmester, 2015; Reese & Whitehouse, 2021). Identity fusion is a ‘visceral sense of “oneness” with a group and its individual members that motivates personally costly, pro-group behaviors’ (Swann & Buhrmester, 2015). It describes an ‘extreme’ situation where the ‘individual and group believe they share an “essence”, where one cannot imagine existing without the other’ and where ‘an attack on the group is experienced as a direct attack on the individual’ (Williams, 2022). In essence, what the foregoing suggests is that at the group level, as Williams (2022) argues, it is ‘strong identity fusion, and not so much extremist beliefs’ per se that ‘may tell us who will become the next suicide bomber and who will not’. Hence, as is the case for the individual level, the role of ideology at the group level – rather than being a root cause of radicalisation – is much more nuanced; ideology plays an important but nevertheless supporting role in other more critical social psychological processes, as described.

The system level In recent years, some scholars, influenced by systems and complexity studies, have attempted to conceptualise terrorist networks as complex, adaptive and self-organising systems (for instance, Lichtblau et al., 2006; Marion & Uhl-Bien, 2003; Mesjasz, 2015; Campedelli et al., 2019). The system level of analysis is useful when one thinks of transnational violent extremist networks – like Al-Qaeda and ISIS for instance – as systems comprising large numbers of individuals and groups that are geographically dispersed and cross-border in their operations, and hence far larger than the smaller, face-to-face groups described in the previous section. In this connection, the professor of psychiatry Marc Galanter (1999) has coined the term charismatic group to describe an entity that comprises not merely ‘a dozen or more members’, but ‘even hundreds or thousands’, whose members are characterised by four core ‘psychological elements’: a ‘high level of social cohesiveness’, a strong adherence to group ‘behavioral norms’, belief in the ‘charismatic or (sometimes divine) power’ of the ‘group or its leadership’ – and most significantly for our purposes – ‘a shared belief system’ – or ideology. The charismatic group is scalable, not only

77

Kumar Ramakrishna

including small face-to-face social groups, as described in the previous section, like ‘families and villages’ (Christakis & Fowler, 2011), but also comprising much larger social collectives dispersed geographically and existing online as well. Mark Juergensmeyer (2000) calls such large transnational groups ‘e-mail ethnicities’ or ‘transnational networks of people tied culturally despite the diversity of their places of residence and the limitations of national borders’ – and ‘united by Web sites and the Internet’. Cass Sunstein (2009) in this regard insists that ‘virtual online groups’ such as chat-rooms and discussion forums have as much power as ‘face-to-face groups’ in influencing individual attitudes and behaviours. A large, transnational charismatic group of ‘hundreds and thousands’ of members who may not necessarily be in direct, personal contact with one another can arguably be seen, from the vantage point of complexity theory, as a complex, adaptive, and self-organising system composed of multiple individuals and cells interacting in numerous, unpredictable ways, but which in essence represent a ‘group of people with a common ideology who try to achieve certain general goals’ (Jackson, 2006, p. 248). A relatively young discipline associated with the well-known Santa Fe Institute (www.santafe.edu) in New Mexico in the United States, complexity theory seeks to encompass both the hard and soft sciences in a quest for a unifying, inter-disciplinary synthesis (Waldrop, 1992). Complexity theorists aver that there is at present no single theory, let alone science, of complexity (Mitchell, 2009; Mesjasz, 2015). Nevertheless, there has been much interest in the past decade in the insights and potential real-world applications of complexity perspectives (Snowden & Boone, 2007; Ramo, 2009). At the core of complexity theory is the idea of a complex, self-organising and adaptive system. M. Mitchell Waldrop (1992) puts it succinctly: [P]eople trying to satisfy their material needs unconsciously organize themselves into an economy through myriad individual acts of buying and selling; it happens without anyone being in charge or consciously planning it. The genes in a developing embryo organize themselves in one way to make a living cell and in another way to make a muscle cell. Flying birds adapt to the actions of their neighbors, unconsciously organizing themselves into a flock. Organisms constantly adapt to each other through evolution, thereby organizing themselves into an exquisitely tuned ecosystem. (my emphasis) The philosopher Howard Bloom (2000) argues that the human body itself could be thought of as a complex adaptive system that – like a ‘collective learning machine’ – constantly reacts to changing environmental stimuli so as to ensure its survival. Importantly, he adds that within the human ‘superorganism’ cognition and information processing are distributed or shared. Complexity theorists D. Snowden and C.F. Kurtz (2003) likewise assert that in complex systems, ‘no director or designer is in control’, and what is known in complexity theory parlance as emergent order is ‘selforganizing’ and arising from ‘the interaction of many entities’. Biologists have long argued that complex self-organising systems in nature survive by seeking homeostasis or equilibrium within their respective ecological niches. Homeostasis is thus the aim of such systems, as it has allowed organisms, aggregated in groups, to adapt effectively to potentially hostile environments. Precisely because complex, self-organising, adaptive systems tend towards homeostasis, they mitigate the impact of conflict and competition among individual units (Harman, 2011). Ultimately, a new, self-organised emergent order arises as an adaptive, homeostatic response to environmental perturbations (Bloom, 2000; Christakis & Fowler, 2011). Applying the above complexity perspective to the type of transnational charismatic group of interest for our purposes – the transnational violent extremist network – it could be said that the 78

The role of ideology in radicalisation

latter resembles a protean, complex, self-organising system seeking homeostasis with its environment as well. Thus, ‘terrorist networks’, far from being ‘stagnant organisms’, are ‘dynamic and in a constant state of flux in an attempt to adjust to their surrounding environment’ (Fink, 2014). To further elaborate: two individuals sharing the same extremist ideology can be said to be the smallest ‘building block’ of a larger complex adaptive system ever mutating and self-organising to keep on top of environmental stresses (Waldrop, 1992). Such building blocks grow larger, continually taking cues from one another, combining and re-combining in countless non-linear permutations, forming more complex – and unpredictable – ‘global’ structures in response to unstable environmental conditions based on simple, local rules (Johnson, 2001). In the case of members of transnational Islamist extremist networks, for instance, deriving from the well-known central Salafi Jihadist concept of al-wala’ wa-al-bara, two such simple basic local rules of conduct would be: shun unbelievers and apostates; and interact only with fellow believers (Weismann, 2017; Maher, 2016). From such simple micro-level axioms, an entire universe of macro-behaviours running the gamut from non-violent if adversarial social and political activism, as on the supposedly non-violent Hizb ut-Tahrir model (Ayoob, 2008), to violent behaviours, including ISIS-motivated terrorism, can conceivably evolve as a homeostatic response to rapidly evolving and diverse operational and political contexts. Likewise in the case of transnational far-right extremist networks, simple rules such as shun Jews and non-whites; and interact only with fellow whites, can spawn a similarly complex range of macro-behaviours spanning social ostracisation and discrimination of Jews and non-whites at one end, to outright violence against them at the other extreme. Observers in this respect have argued that the ‘ideologies and language that underpin’ the violent extremist far-right call to action ‘have long ago reached the mainstream’, with supposedly non-violent far-right ‘populists who sit in parliaments or head governments’ playing a ‘crucial role in normalising, legitimising and amplifying the ideologies that drove the Christchurch and Poway attackers’ (Ebner, 2020).2 ‘Complex behavior need not have complex roots’, Waldrop (1992) thus argues, quite simply because ‘local interactions between large numbers of agents, governed by simple rules of mutual feedback’, can do the job (see also Johnson, 2001). The constituent actors and cells within transnational violent extremist networks therefore can display considerable autonomy and diversity of action and yet – through interactions shaped by a common ideology – retain a paradoxical overarching separate-yet-interconnected quality. As applied to global Islamist terrorist networks like Al-Qaeda, for instance, this notion of how a transnational, complex, self-organising adaptive system operates enables us to better understand the role and strategic importance of the ‘jihadisphere’ – encompassing a wide cross-section of producers and consumers, from al-Qaeda Central to media arms of various al-Qaeda franchise organisations to the globally dispersed array of ‘jihobbyists’ with no formal links to any violent jihadist organisation, all contributing to the everyday making and remaking of the violent jihadi narrative. (Conway, 2012) The far right similarly possesses a global, if fragmented and far from ideationally monolithic, ideological ecosystem, as violent far-right actors around the world are inspired by one another, cite one another’s manifestos, and draw on similar conspiracy theories around the great replacement, the idea of an organized group of global (Jewish) elites orchestrating multicultural societies, and the need to accelerate violence toward the end times. (Miller-Idriss, 2020)3 79

Kumar Ramakrishna

As Miller-Idriss notes, ‘white-supremacist extremism and far-right movements are decidedly global in nature’. Tania Lavin (2020) adds that ‘white supremacists in Eastern Europe, Western Europe, Australia, and the United States commingle, sympathize with one another, and spread propaganda’, while also facilitating the ‘cross-border spread of money, tactical coordination and even personnel’ – as has been the case in Ukraine, where the far-right Azov Battalion has trained white supremacist militants from the United States (Soufan Center, 2019). What holds these transnational violent Islamist and far-right systems together, however imprecisely, is a common overarching ideology incorporating a stock master narrative linking diverse ‘systems of stories’ (Halverson, 2011). This ideological master narrative, by providing ‘common and accepted “stories” or “narratives” ’, works to ‘knit the members’ scattered throughout the network together and ‘shape their actions’ via strategic influence rather than formal, top-down strategic control per se (Jackson, 2006). This notion of strategic influence is akin to strategic theorist Lawrence Freedman’s (2006) concept of a strategic narrative in helping widely dispersed irregular groups connect emotionally to an identity and mission and thus cohere to a general, overall strategic aim. Likewise, social network theorists have spoken of how the right ‘story’ can help keep members of terrorist groups connected in a loose network and prevent defection (Ronfeldt & Arquilla, 2001). What holds the whole system together – to reiterate – is a shared ideology. It is in therefore this sense that at the transnational system level, ideology, functioning as a strategic narrative, sustains a radicalised transnational terror network, however imperfectly. Marion and Uhl-Bien (2003) thus describe the transnational Al-Qaeda system this way: Its structure, which resulted from bottom-up coordination of individuals who voluntarily came together based on common need rather than from top-down hierarchical control, clearly demonstrates the power of a networked system based on relationships and shared vision and mission. (my emphasis) Ideology here acts as what the 19th-century Prussian philosopher of war Carl von Clausewitz (1780–1831) would have called the ‘centre of gravity’ of the transnational violent extremist network or system, or ‘the hub of all power and movement, on which everything depends’ (Von Clausewitz, 1832/1976). Antulio J. Echevarria’s (2002) interpretation of the Clausewitzian centre of gravity concept seems especially germane for our purposes. That is, to ‘find the CoG in any particular situation’, Echevarria explains, ‘we must look for the thing that is providing a certain centripetal, or center-seeking, force (as opposed to centrifugal, which is outward-seeking) for the enemy’. Put another way, the centre of gravity is ‘the one element within a combatant’s entire structure or system that has the necessary centripetal force to hold that structure together’. In other words, at the transnational system level, ideology – as a strategic narrative – provides some degree of coherence among the myriad and often uncoordinated activities of a large number of otherwise disparate militant cells, lone actors, and followers of an amorphous transnational violent extremist network. Interestingly, this ties in with Islamist extremist conceptions of a ‘global Islamic resistance’ (Lia, 2009) and far-right notions of an equally transnational ‘leaderless resistance’ (Musharbash, 2021). Without such a common ideology providing overarching strategic influence, the latter could lose not merely its systemic coherence but also its resilience and longevity in the face of international and local counter-terrorism pressures. In short, without a relatively strongly shared common ideology, it may be that the balance between centrifugal and centripetal forces – what complexity theorists call the ‘edge of chaos’ – would be disrupted, and the transnational terrorist charismatic group or network likely ‘dissolve into turbulence’ (Waldrop, 1992). At 80

The role of ideology in radicalisation

the transnational system level, therefore, to reiterate, the role of ideology – rather than being a root cause – is to sustain a radicalised transnational terror network.

Conclusion In this chapter, we have attempted to unpack the role of ideology in radicalisation into violent extremism. We have seen that ideology draws upon the inchoate, free-floating, and nebulous outgroup prejudices embedded in the wider culture – be it Islamist or far right – and weaponises them into a relatively systematically articulated and action-oriented programme of action that legitimises violence against relevant out-groups. We examined the role of ideology at the level of the individual, group, and system, unpacking how ‘bad ideology’ is part of the ‘lethal cocktail’ that generates terrorist action. We saw that at the individual level, terrorists, rather than having a systematic understanding of the extremist ideology supposedly motivating them, often possess a rough and ready, cut-and-paste interpretation in order to justify out-group violence in ways that may not be logical or consistent. What matters at the individual level is rather the emotional impact and legitimising role of ideology on socioeconomic, political and even idiosyncratic grievances, as well as the desire for out-group violence, that often already pre-exist. At the group level, we saw that ideology plays a key role in demarcating the boundaries of the in-group and out-group. In addition, it was noted that intense group discussion within physically and psychologically isolated enclaves set apart from the wider societal mainstream ensures that no alternative ideas can easily circulate within the confines of the group, and this in turn tends to facilitate group polarisation and a ‘risky shift’ from latent prejudice to outright violent action. It is within the echo chambers of such socially isolated enclaves that identity fusion between the individual and the group occurs, and an assault on the group is then experienced as a direct assault on the individual – social psychological processes often resulting in violent radicalisation. At the group level, therefore, ideology plays an important but supporting, violence-justifying role. Finally, at the level of the system, we saw that a complexity perspective is useful when one thinks of transnational violent extremist networks – like ISIS and the relatively more fragmented global far right – as complex, self-organising, and adaptive systems comprising large numbers of individuals and groups that are geographically dispersed and yet, in their myriad interactions, are guided somewhat by a shared belief system. That is, at the transnational system level, ideology, functioning as a strategic narrative, sustains a radicalised transnational violent extremist network, however imperfectly. The upshot of the earlier discussion suggests that rather than ideology being the ‘root’ of radicalisation, it appears more accurate to think of it as a supporting enabler of the process at the individual, group, and system levels of analysis. To end off, what then could be the root? Clausewitz helps shed provisional light on this rather philosophical issue. His classic tome, On War, remains an influential treatise on the nature of human conflict, after all, with many scholars asserting that it is ‘not simply the greatest but truly the only great book on war’ (Paret & Howard, 1976). In essence, Clausewitz famously argued that what animates human conflict is the ‘blind natural force’ of ‘primordial violence, hatred, and enmity’ (Von Clausewitz, 1832/1976). In short, one possible answer is that while the precise causes, grievances – and ideologies – that fuel violent extremism are myriad, geographical context-driven and issue-specific, at root, in a Clausewitzian sense, ‘primordial violence, hatred, and enmity, which are to be regarded as a blind natural force’, is what fuels violent extremist individuals, groups, and terrorist networks or systems – Islamist and far right – worldwide (Bassford & Villacres, 1995). As mentioned, this is of course only a provisional perspective, one that is certainly wide open to further contestation. This exercise, however, is well 81

Kumar Ramakrishna

beyond the scope of this chapter, which, for reasons of space, focuses only on the role of ideology in the Islamist and white supremacist far-right spheres. Future research should ideally examine the role of ideology within other types of far-right extremism as well – such as the Hindu and Buddhist varieties in Asia (Ramakrishna, 2021a, 2021b). Only through studying such additional varieties of violent extremism can the role of ideology in radicalisation be even better understood – and the intriguing Clausewitzian perspective on its real ‘root’ receive perhaps more extensive and critical interrogation.

Notes 1 Radicalisation is understood here to mean the process by which ‘an individual adopts ideas that are severely at odds with those of the mainstream, refutes the legitimacy of the existing social order, and seeks to replace it with a new structure based on a completely different belief system’. See Vidino (2010). 2 On 15 March 2019, Brenton Tarrant killed 51 people at two different mosques in Christchurch, live-streaming the first attack on Facebook (The Soufan Center, 2019). Just over a month after the Christchurch attack, on 27 April 2019, a copycat attack took place on the other side of the Pacific in Poway, California. John T. Earnest opened fire at a synagogue, killing one woman (Spagat, 2019). 3 The essential argument of the ‘great replacement’ idea popularised by white supremacists is that white, Christian nations are being overrun by masses of black and brown Muslim immigrants from North and sub-Saharan Africa (McAuley, 2019).

References Australian Strategic Policy Institute/S. Rajaratnam School of International Studies. (2011). Jihadists in Jail: Radicalisation and the Indonesian Prison Experience. ASPI-RSIS Joint Report, Issue 40. Ayoob, M. (2008). The many faces of political Islam: Religion and politics in the Muslim world. NUS Press. Bassford, C., & Villacres, E. J. (1995). Reclaiming the Clausewitzian trinity. Parameters, 25(1). www.clausewitz. com/readings/Bassford/Trinity/TRININTR.htm Batrawy, A., & Hinnant, L. (2016, September 26). Leaked ISIS documents show recruits have poor grasp of Islamic faith. The Independent. www.independent.co.uk/news/world/middle-east/isis-documents-leakrecruits-islam-sharia-religion-faith-syria-iraq-a7193086.html Bell, S. (2005). The Martyr’s Oath: The apprenticeship of a homegrown terrorist. John Wiley and Sons Canada Ltd. Bloom, H. (2000). Global brain: The evolution of mass mind from the big bang to the 21st century. John Wiley and Sons. Campedelli, G. M., Cruickshank, I., & Carley, K. M. (2019). A complex networks approach to find latent clusters of terrorist groups. Applied Network Science, 4(59), https://doi.org/10.1007/s41109-019-0184-6 Christakis, N., & Fowler, J. (2011). Connected: The amazing power of social networks and how they shape our lives. Harper Press. Conway, M. (2012). From al-Zarqawi to al-Awlaki: The emergence and development of an online radical milieu. Combating Terrorism Exchange, 2(4), 12–22. Dawson, L. (2015). Challenging the curious erasure of religion from the study of religious terrorism. Numen, 63, 141–164. Deikman, A. J. (2003). Them and us: Cult thinking and the terrorist threat. Bay Tree Publishing. DiMaggio, P. (1997). Culture and cognition. Annual Review of Sociology, 23, 263–287. Ebner, J. (2020). Going dark: The secret social lives of extremists. Bloomsbury. Echevarria II, A. J. (2002). Clausewitz’s center of gravity: Changing our warfighting doctrine-again! Strategic Studies Institute. Fink, L. (2014). Understanding radicalisation and dynamics of terrorist networks through political-psychology. International Institute for Counter-Terrorism. Forsyth, J. (2011, March 3). Clegg collides with Cameron over extremism. The Spectator. www.spectator. co.uk/article/clegg-collides-with-cameron-over-extremism Freedman, L. (2006). The transformation of strategic affairs. Adelphi Paper 379. Routledge. Galanter, M. (1999). Cults: Faith, healing, and coercion (2nd ed.). Oxford University Press. Gaylin, W. (2003). Hatred: The psychological descent into violence. Public Affairs.

82

The role of ideology in radicalisation Glazzard, A. (2017). Losing the plot: Narrative, counter‐narrative and violent extremism. International Centre for Counter Terrorism. Halverson, J. R. (2011, December 8). Why story is not narrative. Center for Strategic Communication, Arizona State University. http://csc.asu.edu/2011/12/08/why-story-is-not-narrative/ Harman, O. (2011). The price of altruism: George price and the search for the origins of human kindness. Vintage. Hofstede, G., & Hofstede, G. J. (2005). Cultures and organizations: Software of the mind: Intercultural cooperation and its importance for survival (2nd ed.). McGraw-Hill. Holbrook, D., & Horgan, J. (2019). Terrorism and ideology: Cracking the nut. Perspectives on Terrorism, 13(6), 2–15. Jackson, B. A. (2006). Groups, networks, or movements: A command-and-control-driven approach to classifying terrorist organizations and its application to Al Qaeda. Studies in Conflict and Terrorism, 29(3), 241–262. Johnson, S. (2001). Emergence: The connected lives of ants, brains, cities and software. Penguin. Jones, J. W. (2008). Blood that cries out from the earth: The psychology of religious terrorism. Oxford University Press. Juergensmeyer, M. (2000). Terror in the mind of god: The global rise of religious violence, updated edition with a new preface. University of California Press. Khalil, J., Horgan, J., & Zeuthen, M. (2019). The attitudes- behaviors corrective (ABC) model of violent extremism. Terrorism and Political Violence. doi:10.1080/09546553.2019.1699793 Khalil, L. (2012). US counter-radicalisation strategy: The ideological challenge. Australian Strategic Policy Institute. Kruglanski, A. W., Belanger, J. J., & Gunaratna, R. (2019). The three pillars of radicalization: Needs, narratives, and networks. Oxford University Press. Kurtz, C. F., & Snowden, D. J. (2003). The new dynamics of strategy: Sensemaking in a complex and complicated world. IBM Systems Journal, 42(3), 462–483. Lavin, T. (2020). Culture warlords: My journey into the dark web of white supremacy. Hachette Books. Lia, B. (2009). Architect of global Jihad: The life of Al-Qaeda strategist Abu Mus’ab Al-Suri. Hurst. Lia, B., & Skjølberg, K. (2004). Causes of terrorism: An expanded and updated review of the literature? FFI/ RAPPORT-2004/04307. Norwegian Defence Research Establishment, Kjeller. Lichtblau, T. E., Haugh, B. A., Larsen, G. N., & Mayfield, T. (2006). Analyzing adversaries as complex adaptive systems. IDA Paper P-3868. Institute for Defense Analyses. Lizardo, O., & Strand, M. (2010). Skills, toolkits, contexts and institutions: Clarifying the relationship between different approaches to cognition in cultural sociology. Poetics, 38(2), 204–227. Maher, S. (2016). Salafi-Jihadism: The history of an idea. Penguin Books. Marion, R., & Uhl-Bien, M. (2003). Complexity theory and Al Qaeda: Examining complex leadership. Emergence, 5(1), 54–76. McAuley, J. (2019, June 17). How gay icon Renaud Camus became the ideologue of white supremacy. Nation. www.thenation.com/article/archive/renaud-camus-great-replacement-brenton-tarrant/ McCauley, C., & Moskalenko, M. (2017). Understanding political radicalization: The two-pyramids model. American Psychologist, 72(3), 205–216. McCauley, C., & Moskalenko, S. (2011). Friction: How radicalization happens to them and us. Oxford University Press. Mesjasz, C. (2015). Complex systems studies and terrorism. In P. V. Fellman, Y. Bar-Yam, & A. Minai (Eds.), Conflict and complexity (pp. 35–71). Springer. Miller-Idriss, C. (2020). Hate in the homeland: The new global far right. Princeton University Press. Mitchell, M. (2009). Complexity: A guided tour. Oxford University Press. Musharbash, Y. (2021). The globalization of far-right extremism: An investigative report. CTC Sentinel, 14(6). https://ctc.westpoint.edu/the-globalization-of-far-right-extremism-an-investigative-report/ Nasir, K. A., Pereira, A. A., & Turner, B. S. (2010). Muslims in Singapore: Piety, politics and policies. Routledge. Nesser, P. (2015). Islamist terrorism in Europe. A history. Oxford University Press. Oliver, P. E., & Johnston, H. (2000). What a good idea! Ideologies and frames in social movement research. Mobilization: An International Journal, 4(1), 37–54. Paret, P., & Howard, M. (Eds.). (1976). Carl von Clausewitz, on war. Princeton University Press. Ramakrishna, K. (2005). It’s the story, stupid: Developing a counter-strategy for neutralizing radical Islamism in Southeast Asia. CSRC Discussion Paper 05/48. http://scholar.google.com.sg/scholar?cluster=1672 7912978685463890&hl=en&as_sdt=0,5&sciodt=0,5

83

Kumar Ramakrishna Ramakrishna, K. (2021a, January 20). The growing challenge of the extreme right. RSIS Commentary. www. rsis.edu.sg/rsis-publication/icpvtr/the-growing-challenge-of-the-extreme-right/#.YSdlpdMza8U Ramakrishna, K. (2021b). Deconstructing Buddhist extremism: Lessons from Sri Lanka. Religions, 12, 970. https://doi.org/10.3390/rel12110970 Ramo, J. C. (2009). The age of the unthinkable: Why the new world disorder constantly surprises us and what we can do about it. Little, Brown and Company. Razzaque, R. (2008). Human being to human bomb: Inside the mind of a terrorist. Icon Books. Reese, E., & Whitehouse, H. (2021). The development of identity fusion. Perspectives on Psychological Science, 16(6), 1398–1411. Richardson, L. (2006). What terrorists want: Understanding the terrorist threat. John Murray. Ronfeldt, D., & Arquilla, J. (2001). What next for networks and netwars? In J. Arquilla & D. Ronfeldt (Eds.), Networks and netwars: The future of terror, crime and militancy. RAND. Sageman, M. (2008). Leaderless Jihad. University of Pennsylvania Press. Savage, S. (2011). Four lessons from the study of fundamentalism and the psychology of religion. Journal of Strategic Security, 4(4), 131–150. Schmid, A. P. (2013). Radicalisation, de-radicalisation, counter-radicalisation: A conceptual discussion and literature review. International Center for Counterterrorism (ICCT) Research Paper. Smith, J. (2020). Home grown: How domestic violence turns men into terrorists. Riverrun. Smith, P. B., Bond, M. H., & Kagitcibasi, C. (2006). Understanding social psychology across cultures: Living and working in a changing world. Sage. Snowden, D. J., & Boone, M. E. (2007). A leader’s framework for decision making. Harvard Business Review, 1–8. Soufan Center. (2019). White supremacy extremism: The transnational rise of the violent white supremacist movement. The Soufan Center. Spagat, E. (2019, September 20). Poway killer told 911 he was ‘defending our nation against the Jewish people’. The Times of Israel. www.timesofisrael.com/poway-killer-told-911-he-was-defending-our-nationagainst-the-jewish-people/ Staub, E. (1989). The roots of evil: The origins of genocide and other group violence. Cambridge University Press. Sunstein, C. R. (2009). Going to extremes: How like minds unite and divide. Oxford University Press. Swann, Jr., W. B., & Buhrmester, M. D. (2015). Identity fusion. Current Directions in Psychological Science, 24(1), 52–57. Swidler, A. (1986). Culture in action: Symbols and strategies. American Sociological Review, 51(2), 273–286. Vaisey, S. (2009). Motivation and justification: A dual-process model of culture in action. American Journal of Sociology, 14(6), 1675–1715. Vidino, L. (2010). Countering radicalization in America: Lessons from Europe. United States Institute of Peace. Special Report 262. Von Clausewitz, C. (1832/1976). On war (M. Howard & P. Paret, Ed. & Trans.). Princeton University Press. Wagemakers, J. (2016, August). Salafism. Oxford Research Encyclopaedia: Religion. https://oxfordre.com/ religion/view/10.1093/acrefore/9780199340378.001.0001/acrefore-9780199340378-e-255 Waldrop, M. M. (1992). Complexity: The emerging science at the edge of order and chaos. With a New Afterword by the Author. Simon and Schuster. Waller, J. (2005). Becoming evil: How ordinary people commit genocide and mass killing. Oxford University Press. Weaver, M. (2015, July 20). Cameron’s anti-terror strategy is ‘barking up wrong tree’, says expert. The Guardian. www.theguardian.com/uk-news/2015/jul/20/david-cameron-anti-terror-strategy-wrong-expert-says Weismann, I. (2017). A perverted balance: Modern salafism between reform and Jihad. Die Welt Des Islams, 57, 33–66. Wiktorowicz, Q. (2005). Radical Islam rising: Muslim extremism in the west. Rowman & Littlefield. Williams, M. (2022). The science of hate: How prejudice becomes hate and what we can do to stop it. Faber & Faber. Wilson, J. (1973). Introduction to social movements. Basic Books.

84

6 IDENTITY AND EXTREMISM Sorting out the causal pathways to radicalisation and violent self-sacrifice Julia Ebner and Harvey Whitehouse

What are the distinctive cognitive and behavioural patterns associated with pathways towards violence? How do personal and group identities contribute to the process of radicalisation? What kinds of social environments and shared experiences may be conducive to violent extremism? How are the identity formation processes involved in radicalisation affected by globalisation and digitalisation? This chapter addresses these questions in the broader context of psychological perspectives on radicalisation. Our focus is on the role of identity fusion – an extreme form of group cohesion that has been consistently linked to extreme self-sacrificial violence and terrorism.

Psychological explanations of radicalisation In the aftermath of 9/11, researchers and policy experts were divided over the question as to whether violent extremism was motivated primarily by ideologies or by unresolved grievances. These two possibilities are commonly referred to as ‘Kepel vs Roy’, named after the two French scholars and former friends who came to disagree about the underlying causes of radicalisation (Nossiter, 2016). Gilles Kepel, a political scientist and Arabist who teaches at Sciences Po in Paris, championed the idea that ideology and collective narratives play a key role in the process of radicalisation towards violent extremism (2003). Oliver Roy, another leading French scholar and professor at the European University Institute in Florence, stressed the psychological state of suffering as well as the desire for recognition that lead individuals to embrace revolutionary youth movements that promise heroism and brotherhood (Roy, 2015). While Kepel viewed ideology as the main driver of violent extremism, Roy saw ideological frameworks as merely outlets for personal struggles. In the context of jihadism, Kepel spoke of a ‘radicalisation of Islam’ (Kepel, 2015), while Roy argued that we are witnessing an ‘Islamisation of radicalism’ (Roy, 2017). Although both Kepel and Roy focused their work predominantly on jihadist violence, their lines of argumentation have been applied to other forms of violent extremism and reflect a wider debate in radicalisation research. The first two decades of the 21st century were marked by a strong focus on radicalisation towards violent jihadist causes. Research frequently highlighted the role of identity but tended to emphasise its ideological underpinnings, for example, Muslim identity politics and religious identification – as key elements of identity and driving forces in the radicalisation process (Choudhury, 2007). Importantly, this research pointed to underlying similarities in the 85

DOI: 10.4324/9781003035848-8

Julia Ebner and Harvey Whitehouse

radicalisation process across a diversity of groups (Lee & Knott, 2020; Bailey & Edwards, 2017), ranging from jihadists to far-right extremists (Ebner, 2017). This research brought to light various ways in which the narratives of violent extremists connected personal grievances to group-based struggles (Fielitz et al., 2018). In line with this, many researchers are increasingly viewing radicalisation as an identity transformation process that incorporates both personally transformative and ideologically shared components (Whitehouse H., 2018). High-profile political engagement and media reporting on the topic of radicalisation have made it difficult to distinguish between the drivers and dynamics that are well-established in scientific research and those that are merely gaining traction in the public discourse but have little empirical basis (e.g. explanations based exclusively on personality profiles or psychopathological considerations (Kruglanski & Fishman, 2009)). Researchers at the Royal United Services Institute (RUSI) conducted an extensive review of post-2000 literature and academic articles about radicalisation to classify common hypotheses about pathways to extremism according to their scientific credentials (Allan et al., 2015). They found little or no support in existing studies for the arguments that masculinity and honour play significant roles in radicalisation. Likewise, the idea that low levels of literacy and limited availability of information drive extremism is not empirically well supported. While the role of education, employment, poverty, and blocked political participation may play a role, the evidence is mixed. Likewise, efforts to establish distinctive terrorist and extremist profiles have proven inconclusive (Gill et al., 2014; Neumann P. R., 2017; Gruenewald et al., 2013). Researchers studying radicalisation have increasingly shifted their focus towards more evidence-based approaches (Smith et al., 2016; Gøtzsche-Astrup, 2018; Ahearn et al., 2020). On the basis of a systematic review of scientific literature on radicalisation towards violent extremism across different ideologies and geographies, Vergani et al. proposed the ‘3Ps of Radicalisation’ model – differentiating ‘push’, ‘pull’, and ‘personal’ factors in the radicalisation process. They concluded that pull factors – which include cognitive drivers, social mechanisms, and group processes as well as emotional and material incentives – were the most regularly emphasised in the literature (Vergani et al., 2018). There is substantial credible evidence that identity is a key element in radicalisation (Allan et al., 2015). In particular, research has identified a wide range of factors associated with group psychology, and especially the dynamic interplay between personal and group identities in motivating violent extremism.

Psychological theories of radicalisation Social psychologists have developed a range of models that seek to explain radicalisation, using a combination of individual and social attitudinal and behavioural factors. Proposed psychological drivers include ‘Frame Alignment’ (Wiktorowicz, 2005), ‘Uncertainty-Identity Theory’ (Hogg & Adelman, 2013), the ‘Quest for Significance Model’ (Kruglanski et al., 2018), the ‘Staircase to Radicalisation Model’ (Moghaddam, 2005), the ‘Four-Stage Process’ (Sageman, 2008), the ‘Duplex Theory of Hate’ (Sternberg, 2003), the ‘Moral Disengagement Theory’ (Bandura A., 2004, S. 121), the ABC Model (Khalil et al., 2019), the Two Pyramids Model (McCauley & Moskalenko, 2017), and the ‘Fusion-Plus-Threat Model’ (Whitehouse H., 2018). Psychological approaches suggest that the perspectives of the two French theorists, Kepel and Roy, are not as incompatible as they and their followers have tended to assume. While Roy was right to point to the importance of shared suffering as a major source of motivation to engage in extreme forms of pro-group action, Kepel’s collective narratives also play a role in motivating ideological extremism. Collective narratives can contribute to perceptions of shared suffering and thereby contribute to the deep psychological motivations that lead to violence. On the basis of 86

Identity and extremism

extensive research into the British Islamist extremist organisation Al-Muhajiroun, Wiktorowicz called the matching of individual factors and group factors ‘frame alignment’, that is, when an individual’s ‘cognitive opening’ coincides with the narrative offered by an extremist group (Wiktorowicz, 2005). But there is growing evidence that neither demography nor ideology is capable of motivating extreme self-sacrifice on their own. Research on radicalisation is increasingly moving away from viewing radicalisation as a one-dimensional, linear process towards a more holistic, multi-level approach to framing extremism (McCauley & Moskalensko, 2008; Pisou & Ponesh, 2016; Ranstorp, 2010; Allan et al., 2015). In 2005, the psychologist Fathali M. Moghaddham introduced his famous ‘staircase to radicalisation’ model. According to this model, most people inhabit the ground floor, which stands for perceptions of injustice, but some then move on to higher floors, trying to find options to resolve the perceived state of unfair treatment (first floor) or the channelling of aggression towards an enemy figure or group (second floor). In the next steps, radicalised individuals would become morally engaged to justify hostility towards a demonised out-group (third floor) and increasingly endorse a terrorist organisation as their categorical thinking solidifies (fourth floor). Few people reach the uppermost floors, actually committing acts of terrorism and violence against out-groups (fifth floor) (Moghaddam, 2005). A 2011 study found empirical evidence for Moghaddam’s ‘staircase’ model (Lygre et al., 2011). But further research is needed to clarify all the predictions of the model, including the key question of causality – in particular, what drives individuals to ascend the staircase that leads to violent extremism? Another influential model, developed by Marc Sageman, conceptualises radicalisation as a four-stage process: ‘a sense of moral outrage, interpreted in a specific way, which resonates with one’s personal experiences, and is channelled through group dynamics, both face-to-face and online’ (Sageman, 2008, S. 223). While the model offers a useful framework to understand the connection between personal grievances and group narratives, it lacks an evidence-based explanation of the transitions between the four stages. Social Identity Theory helps to explain how social identification and group membership can motivate and influence individual behaviour (Tajfel & Turner, 1979). Many studies have explored how strong group alignment (based on shared attributes such as class, party, gender, ethnicity, and religion) is linked to radicalisation (Beelmann, 2020; Smith et al., 2016; Harris et al., 2014). Strindberg argued that ‘group identity draws the battlelines’ and that social identification processes are central to pathways towards extremism (Strindberg, 2020). Uncertainty-identity theory is based on the premise that uncertainty about one’s self is a strong motivator to join extremist groups that offer high levels of group entitativity and identification (Hogg  & Adelman, 2013). There is substantial evidence in support of the causal link between identity uncertainty and group identification (Gøtzsche-Astrup, 2018). It is widely recognised by social identity theorists and terrorism researchers that extreme attitudes and behaviours entail ‘us and them’ thinking, which expresses favourable attitudes towards the in-group and negative attitudes towards the outgroup (Harris et al., 2014). J.M. Berger adopts a framework based on social identity theory for his assessment of ideological shifts towards extremism and the construction of in-group/out-group dichotomies (Berger J., 2017). Drawing on Haroro J. Ingram’s linkage concept (Ingram, 2016), Berger argues that extremists link out-groups to a threat and the in-group to a vulnerability, resulting in a crisis construct that may sit on a spectrum from mild to apocalyptic. Kruglanski et al. (2018) introduced the so-called Quest for Significance Model, which combines three different factors to explain radicalisation: psychological need, ideological narrative, and social network – the 3Ns. The authors argue that an individual’s need for personal 87

Julia Ebner and Harvey Whitehouse

significance or desire ‘to be someone’ is the key factor underlying radicalisation, which is then expressed through a narrative that promises a pathway to achieving significance through violent action in support of a collective cause and a network of people who similarly endorse the narrative (Kruglanski et al., 2018). A number of interview-based and observational studies found evidence for the significance quest approach (Neumann P., 2015; Speckhard et al., 2017). However, both the causality and directionality of the triangular relationship have not been tested systematically. The Attitudes-Behaviour Corrective (ABC) model problematises the concept of radicalisation, emphasising that violent behaviour cannot be predicted on the basis merely of attitudes or ideology (Khalil et al., 2019, 2020). Like the ABC model, the Two Pyramids model by ­McCauley and Moskalenko assumes two separate forms of radicalisation: radicalisation in opinion and r­ adicalisation in action, with only the latter giving rise to terrorist activity (McCauley & Moskalenko, 2017). While many of the models reviewed earlier provide useful frameworks and data for exploring the different psychological factors that have reappeared as patterns in radicalisation, they also leave many questions unanswered. Most importantly, they do not satisfactorily distinguish between psychological causes of violent extremism and mere correlates or consequences of the radicalisation process. This constitutes a major problem for prevention and the detection of criminal intent. Many individuals may exhibit beliefs and behaviours associated with radicalisation that are not in themselves illegal and may never lead to acts of violent extremism. Is there one underlying factor that drives criminal manifestations of radicalisation as distinct from merely extreme beliefs and attitudes? Is there some special concatenation of features that, once combined, leads to acts of terrorism? Or are all the features currently associated with radicalisation mere epiphenomena that are caused by the same processes that drive not only violent extremists but also many harmless ideologists and bigots? If so, how can we get at causation and what can we do about it? To address these questions, we argue it is necessary to focus on an extreme form of group alignment known as ‘identity fusion’, a visceral sense of oneness that has been strongly linked to violent self-sacrifice (Swann W. B. et al., 2009).

Identity fusion: an emerging psychological explanation of violent extremism As explained in the previous section, Social Identity Theory fruitfully distinguishes two kinds of identity: the personal self (comprising traits that make us unique as individuals) and the social group (comprising traits that mark off a collective of some kind, whether it is a small relational one like a family or a football team or a large categorical one such as a nation, ethnic group, or world religion). Social Identity Theory is centred around group identification, which commonly presents personal and group identities as quite separate and unconnected. However, an alternative form of group alignment is presented by identity fusion, which entails a porous boundary between personal and group identities (Swann et al., 2012). In this section, we will draw on a growing body of evidence to argue that these two forms of group cohesion tend to lead to different outcomes in extremism: cognitive versus violent extremism. Identification with a group is depersonalising in the sense that making the group salient will cause the personal identity to become less accessible, and vice versa (Tajfel & Turner, 1997). For example, when you think about your identity as a Muslim (e.g. the beliefs and practices associated with that group), your unique personal identity will become less relevant. And the same applies the other way around – if you think about your unique self as an individual, your

88

Identity and extremism

sense of yourself as just another member of a big religious organisation will be less salient and relevant. With identity fusion, however, it is a very different story. When people fuse with a group, their personal and group identities are activated synergistically rather than hydraulically (Swann & Buhrmeister, 2015). If you make the person’s identity as a Muslim salient, it activates their sense of personal agency at the same time. And conversely, if you make their personal identity salient, that taps directly into their conception of what it means to be a Muslim. This fusion of personal and group identities motivates very strong forms of progroup action. Most of the time, identity fusion is a very positive thing. For example, many people are fused with their families and are therefore willing to make great sacrifices for their relatives, whether by giving them help financially or in more extreme cases taking care of them during prolonged illnesses or donating organs when transplants are required (Swann et al., 2014). But there is also a dark side to fusion. Because the self and the group are so tightly bound together, when the group is threatened, fused individuals take it personally. As a result, they will stop at nothing to defend the group, displaying far greater willingness to fight and die for each other than the most highly identified individuals. While strong forms of identification with a group can motivate commitment to extreme ideologies, this form of group alignment does not appear to be sufficient to motivate extreme forms of self-sacrifice (Whitehouse H., 2018; Swann & Buhrmeister, 2015; Swann et al., 2012). Selfpreservation is a strong motivating force in humans and other animals, overriding coalitional commitments when they imperil survival. By contrast, since fused individuals integrate the group into their sense of self, they are more likely to throw themselves on a grenade to save their fellows (Gómez A. et al., 2011; Swann W. et al., 2010). Evidence for this comes from a variety of studies, including fieldwork, surveys, and interviews with armed groups that regularly risk life and limb to protect their fellow group members. For example, fusion has been linked to extreme pro-group violence in Libyan revolutionary battalions (Whitehouse et al., 2014), Cameroonian herders and farmers (Buhrmeister et al., 2020), Indonesian religious fundamentalists (Kavanagh et al., 2020), British and Brazilian football hooligans (Newson, 2019), as well as captured fighters from Islamic State and Kurds on the battle frontlines (Atran, 2016). The fusion-plus-threat model (see Figure 6.1) integrates previous inter-disciplinary evidence about the drivers and consequences of fusion and identification into a broader theory of ‘ritual modes’, in which differing memory systems give rise to contrasting forms of group alignment (Whitehouse H., 2018). The theory proposes that an emotional event stored in episodic memory and usually accompanied by the individual’s reflection on its meaning can contribute to personal identity transformation. When such a transformative experience is perceived to be shared with a group, this helps to create a feeling of shared group essence that can produce fusion. A second pathway to fusion is based on phenotypic matching, hence perceived biological kinship (Whitehouse et al., 2017). Both pathways – the sharing of personally transformative experiences and the sense of biological kinship – lead to a form of psychological kinship and fusion with other group members. When faced with an external threat to the in-group, identity fusion can significantly heighten an individual’s willingness to fight and die for the group. Fusion therefore offers an explanation for acts of self-sacrifice such as suicide terrorism and martyrdom operations, as well as other forms of extremely risky pro-group behaviours (Whitehouse H., 2018). On the basis of the open commentary in response to Whitehouse’s fusion-plus-threat model, several variables that might mediate or moderate pathways from fusion to violence have been identified (see Whitehouse, 2021 for a fuller account of this framework). As outlined by Ebner

89

Julia Ebner and Harvey Whitehouse

90 Figure 6.1  Pathways to fusion and self-sacrifice Source: Whitehouse (2018)

Identity and extremism

et al. (Ebner et al., 2022), the following factors have been frequently cited in literature as contributing to an escalation towards violence: (1) Perceived out-group entitativity (Choi et al., 2018; Lickel et al., 2006), which was traced via the use of ‘us versus them’ narratives (Berger J., 2017), in particular the use of language that denounces, demonises, or dehumanises an entire out-group (Giner-Sorolla et al., 2011); (2) Perceived out-group threat (Whitehouse H., 2018), which may manifest itself in narratives of an existential threat posed to the in-group (Hirschberger et al., 2016), the belief in a conspiracy of the out-group (Mashuri & Zaduqisti, 2015), or the belief in an inevitable war between the in- and out-group (Ebner, 2017); and (3) Violence condoning norms (Louis et al., 2018), which may include the justification of violence, the glorification of violence via martyrdom narratives or the so-called warrior mentality (Louis et al., 2018), the identification with a violent role model (Davey & Ebner, 2019), and perceived hopelessness of alternative solutions (Thomas & Louis, 2014). Figure 6.2 illustrates the relevant variables on the fusion to violence pathway. Armed with a robust theory of the causal pathways to identity fusion and the mediators between fusion and violent extremism, it should now be possible, at least in theory, to tackle the problem of winnowing out from a large population of ideological extremists which ones are most likely to commit hate crimes or acts of terrorism. While many ideological extremists are highly identified with their groups, those who are also strongly fused present the most significant threat – and so the challenge is to find those individuals before it is too late. Thus, the race is on to find a set of diagnostic tools to pick out highly fused would-be terrorists from a much larger population of ideologues. Likewise, the fusion-plus-threat model could offer a useful starting point for future policy and practice in the field of radicalisation prevention. For example, de-fusion approaches could be used to reverse extreme forms of alignment with violent organisations by demonstrating that group-defining experiences are not as widely shared as their members suppose.

Policy and practice: from ideology-focused to multi-level approaches Prevention and deradicalisation approaches have evolved in parallel with emerging evidencebased research. Policymakers and frontline practitioners are now increasingly aware of psychological drivers that have emerged as consistent patterns in radicalisation pathways. In the post-9/11 era, the idea that ideology is the starting point from which individuals mechanically move towards violent extremism became known in policy circles as the ‘conveyor belt’ theory. This doctrine

Figure 6.2  Relevant variables on the fusion to violence pathway Source: Ebner et al. (2022)

91

Julia Ebner and Harvey Whitehouse

gained prominence in the United Kingdom when former Prime Minister David Cameron used the term in a 2011 speech at the House of Commons to call for stronger action to tackle non-violent Islamist extremism (Cameron D., 2011). In recent years, radicalisation prevention policies have undergone notable changes, mirroring both the theoretical shifts towards multi-dimensional explanation models and integrating new evidence emerging from psychological research on radicalisation (for a more detailed account, see Koehler, 2016). A number of transnational organisations, including the European Parliament (2014), the United Nations (UNDP, 2016), the OSCE (Neumann P. R., 2017), and the Human Rights Council (Emmerson, 2016), have made efforts to do justice to the complexity and diversity of pathways to radicalisation. For example, an OSCE report by Peter Neumann of the International Centre for the Study of Radicalisation and Political Violence (ICSR) at Kings College named five recurring factors: grievance, needs, ideas, people, and violence (Neumann P. R., 2017, S. 17–18). Magnus Ranstorp, who helped to establish the European Union’s flagship radicalisation research programme RAN (Radicalisation Awareness Network), described radicalisation as a ‘multifaceted combination of push-pull factors involving a combination of psychological factors, political grievance, religious motivation and discourse, identity politics and triggering mechanisms’ (Ranstorp, 2010). The US national terrorism risk assessment guide highlights motivational factors as indicators of a threat to violent mobilisation. While it cites ideology, relationships, and intent as important motivation indicators, it fails to mention the role of identity and group cohesion in the radicalisation process (NCTC et al., 2021). The New York State Police Department treats radicalisation leading to jihadist terrorism as a process composed of four distinct phases that echo Sageman’s model: preradicalisation, self-identification, indoctrination, and jihadisation (Silber & Bhatt, 2007). Sageman’s credentials as a former CIA case officer and the first scholar in residence at the New York Police Department have no doubt contributed to the influence of this model on US policy and security circles. The UK government highlighted the role of identity crises in bridging psychological and ideological contributions to radicalisation. The Home Office’s framework, which informs all top-down prevention and intervention programmes in the United Kingdom, depicts radicalisation as a cocktail of personal grievances, identity crises, exposure to charismatic recruiters and their extremist narratives, and ideologies (HM Government, 2010). While the British Prevent programme initially had a strong focus on Islamist ideologies, attracting much criticism for its disproportional focus on religious and cultural minority communities, it has increasingly emphasised the role of overarching patterns in social processes and identity transformations in recent years. Building on existing research on radicalisation pathways, today’s Channel programme – the radicalisation referral programme within Prevent – integrates a multitude of psychological factors identified as relevant in the radicalisation process (HM Government, 2015). Following an initial confidential screening of referred radicalisation cases by the police, referrals may then be forwarded to the multi-agency ‘Channel panel’, which consists of local authority representatives, education and health practitioners, and radicalisation experts. The Channel committee’s decision on whether referred radicalisation cases should receive support via Channel’s one-on-one deradicalisation programme is informed by the Vulnerability Assessment Framework, which covers a mix of psychological and environmental risk factors on an individual level: (a) engagement with a group, cause, or ideology (e.g. feelings of grievance and injustice, perceived threat, need for identity or belonging, desire for status, and desire for political or moral change); (b) intent (e.g. ‘us versus them’ thinking, dehumanisation of the enemy, and violence justification); and (c) capability to cause harm (e.g. individual knowledge, skills and competencies, access to networks, and funding or equipment) (HM Government, 2012). 92

Identity and extremism

Despite recent developments in policy approaches to radicalisation, there is potential to further enhance prevention and intervention models based on evidence from psychological research. While there are many ways to achieve this, the following section focuses on the future of early warning systems that can assist frontline workers and security services in identifying recurring risk factors for violent extremism.

Future research and prevention: spotting would-be violent extremists in the crowd Preventing acts of terrorism requires multi-dimensional risk assessment and prediction models that take into account both environmental macro-factors and individual micro-factors. A growing number of theoretically informed, data-driven models to predict the risk of terrorism have been designed in recent years (Ding et al., 2017; Python et al., 2021). Many studies have focused on designing statistical models that seek to identify and quantify the effects of terrorism drivers in space and time (Nemeth et al., 2014; Python et al., 2018). Previous studies have used a combination of social factors (e.g. ethnic diversity, major drug regions, and population density) and natural factors (e.g. night-time lights, population density, and topography) to simulate future conflict (Ball, 2012; Taylor et al., 2004). For example, Python et al. introduced an agent-based model using machine learning techniques, which made it possible to predict the location of terrorist events on a fine spatial and temporal scale (Python et al., 2021). Models grounded in human drivers and psychological risk factors have not yet reached similarly high predictive power. Neither academics nor policymakers have agreed on a universal definition of terrorism and violent extremism on a global scale. Against this background, it is not surprising that few measures to assess individual and group-based proneness to extreme violence exist to date. A reliable method of identifying would-be terrorists before they carry out attacks has not yet been found. Even if extreme ideology is frequently associated with terrorist violence, it is not a sufficient condition, and its causal role has not been well established. Moreover, in recent years, a number of researchers have warned that the role of ideology in radicalisation has been exaggerated in counterextremism approaches (Kundnani A., 2015; Powell, 2016). As noted earlier, the effort to establish terrorist and extremist profiles has been inconclusive (Gill et al., 2014; Neumann P. R., 2017). Several think tanks, tech firms, and government bodies have made progress on measuring violent extremism and hateful language deemed conducive to violence in virtual environments. For example, the Institute for Strategic Dialogue and The Center for the Analysis of Social Media jointly created the ‘Hate Mapper’, a tool that determines the concentration of anti-minority hatred on a borough level in different geographies, including the City of London and the State of Victoria in Australia. The Hate Mapper is based on lists of keywords that are seen as markers for antiSemitic and anti-Muslim extremism (ISD, 2019). Though not developed for the purpose of examining terrorism only, it is worth mentioning Swann et al.’s ‘fight and die’ scale as an alternative approach to measuring willingness to engage in violent pro-group behaviour (Swann W. et al., 2014). Their original study that introduced the measure combined Swann et al.’s measure of endorsement of extreme pro-group behaviours (2009) with two items that indicate willingness to die for their country (e.g. ‘I would sacrifice my life if it saved another country member’s life’) to form a more comprehensive ‘fight and die’ index. Drawing on existing research from the fields of social psychology, anthropology, and linguistics, our latest study identified and measured recurring psychological patterns in the manifestos of convicted terrorists and compared them to the patterns observed in non-violent political manifestos. We designed a new language-based framework grounded in fusion theory that can help assess 93

Julia Ebner and Harvey Whitehouse

psychological risk factors that make members of virtual groups more prone to committing acts of extreme violence (Ebner et al., 2022). Our comparative manifesto analysis examined 15 manifestos – ranging from terrorist manifestos published prior to high-profile attacks in 2011–2022 (including, e.g. the attacks in Oslo & Utøya 2011; Isla Vista 2014; Charleston 2015; Christchurch 2019; Poway 2019; Halle 2019; and El Paso 2019) to ideologically extreme and moderate manifestos that were not followed by violent attacks committed by their authors. Based on a mix of qualitative text analysis and quantitative natural language processing (NLP), narrative patterns and related linguistic markers were distilled and statistically assessed (Ebner et al., 2022; Ebner et al., 2023). The results showed that identity fusion, out-group dehumanisation and violence-condoning norms were significantly higher in violent as opposed to non-violent manifestos. While the study highlights the validity of the fusion-plus-threat model, it also supports compatible alternative models such as Sternberg’s ‘Duplex Theory of Hate’ (Sternberg, 2003, S. 303), Bandura’s ‘Moral Disengagement Theory’ (Bandura A., 2004, S. 121), Kruglanski et al.’s ‘Quest for Significance Model’ (Kruglanski et al., 2018), and Louis et al.’s research on the power of group norms (Louis et al., 2018).

Conclusions We have reviewed a wide range of psychological research on radicalisation and violent extremism, arguing that it has had limited success in establishing the underlying causes of these phenomena, making it difficult to establish effective methods of preventing or managing them. Despite recent shifts away from simplistic theories towards multi-dimensional models that recognise the importance of psychological factors, there is potential to learn from evidence-based research on the role of identity and shared history in violent extremism. Our model predicts that fusion drives violent extremism via three mediators: perceived outgroup threat, violence-condoning norms, and perceived out-group entitativity. We argue, however, that extreme ideology is a product of strong forms of identification rather than fusion and cannot in itself explain violent extremism. It is therefore fruitless to use extreme ideology as a diagnostic tool for singling out would-be terrorists. Since we now have a good understanding of the pathways to fusion, it should be possible to use linguistic markers to detect highly fused actors at risk of carrying out acts of violent self-sacrifice without their being aware of it. The research presented here therefore holds out the prospect for early detection and intervention to prevent terrorist violence that promises to be much more effective than existing programmes focusing primarily on ideological deradicalisation.

Acknowledgements The authors would like to thank Dr. Daniel Köhler and the editors for instructive feedback and literature recommendations. Ebner’s work on this chapter was funded by the UK’s Economic and Social Research Council (ESRC) and St. John’s College at Oxford University under the ESRC Grand Union Doctoral Training Partnership (DTP) Studentship, and Whitehouse was supported by a European Research Council (ERC) Advanced Grant under the European Union’s Horizon 2020 Research and Innovation Programme (Advanced Grant #694986).

References Ahearn, E.-R., Bhui, K., & Jones, E. (2020). What factors are truly associated with risk of radicalisation? A secondary data analysis within a UK sample. Transcultural Psychiatry, 58(5).

94

Identity and extremism Allan, H., Glazzard, A., Jesperson, S., Reddy-Tumu, S., & Winterbotham, E. (2015). Drivers of violent extremism: Hypotheses and literature review. Royal United Services Institute (RUSI). Atran, S. (2016). The devoted actor: Unconditional commitment and intractable conflict across cultures. Current Anthropology, 57, 192–201. Bailey, G., & Edwards, P. (2017). Rethinking ‘radicalisation’: Microradicalisations and reciprocal radicalisation as an intertwined process. Journal for Deradicalization, 10(Spring Issue), 255–281. Ball, P. (2012). The transformation of war: Modelling modern conflict. Springer. Bandura, A. (2004). The origins and consequences of moral disengagement: A social learning perspective. In F. Moghaddam & A. J. Marsella (Eds.), Understanding terrorism: Psychosocial roots, consequences, and interventions. American Psychological Association. Beelmann, A. (2020). A social-developmental model of radicalization: A systematic integration of existing theories and empirical research. International Journal of Conflict and Violence, 14(1). Berger, J. (2017). Extremist construction of identity: How escalating demands for legitimacy shape and define in-group and out-group dynamics. ICCT. Buhrmeister, M., Zeitlyn, D., & Whitehouse, H. (2020). Ritual, fusion and conflict: The roots of agro-pastoral violence in rural Cameroon. Group Processes & Intergroup Relations, 25(1), 298–311. Cameron, D. (2011, May 3). Statement to house of commons HC Deb. cc 461, 473. Choi, V., Gelfand, M., & Jackson, J. (2018). The role of entitativity in perpetuating cycles of violence. Behavioral and Brain Sciences, 41. Choudhury, T. (2007). The role of Muslim identity politics in radicalisation. HM Government Report Commissioned by the Preventing Extremism Unit. Davey, J., & Ebner, J. (2019). ‘The great replacement’: The violent consequences of mainstreamed extremism. ISD. Ding, F., Ge, Q., Jiang, D., Fu, J., & Hao, M. (2017). Understanding the dynamics of terrorism events with multiple-discipline datasets and machine learning approach. PLoS One, 12(6). Ebner, J. (2017). The rage: The vicious circle of Islamist and far-right extremism. I. B. Tauris. Ebner, J., Kavanagh, C., & Whitehouse, H. (2022). Is there a language of terrorists? A comparative manifesto analysis. Studies in Conflict and Terrorism, 1–28. Ebner, J., Kavanagh, C., & Whitehouse, H. (2023). Measuring Socio-Psychological Drivers of Extreme ­Violence in Online Terrorist Manifestos: An Alternative Linguistic Risk Assessment Model. Journal of Policing, Intelligence and Counter Terrorism. Emmerson, B. (2016). Report of the special rapporteur on the promotion of human rights and fundamental freedoms while countering terrorism. Human Rights Council. Fielitz, M., Ebner, J., Guhl, J., & Quent, M. (2018). Loving hate: Anti-Muslim extremism, radical Islamism and the spiral of polarization. IDZ/ISD. Gøtzsche-Astrup, O. (2018). The time for causal designs: Review and evaluation of empirical support for mechanisms of political radicalisation. Aggression and Violent Behavior, 39, 90–99. Gill, P., Horgan, J., & Deckert, P. (2014). Bombing alone: Tracing the motivations and antecedent behaviors of lone-actor terrorists. Journal of Forensic Science, 59(2). Giner-Sorolla, R., Leidner, B., & Castano, E. (2011). Dehumanization, demonization, and morality shifting. In M. Hogg & D. Blaylock (Eds.), Extremism and the psychology of uncertainty (pp. 165–182). Wiley-Blackwell. Gómez, A., Brooks, M. L., Buhrmeister, M. D., Vázquez, A., Jetten, J., & Swann, W. B. (2011). On the nature of identity fusion: Insights into the construct and a new measure. Journal of Personality and Social Psychology, 100, 918–933. Gruenewald, J., Chermak, S., & Freilich, J. D. (2013). Distinguishing “loner” attacks from other domestic extremist violence: A comparison of far-right homicide incident and offender characteristics. Criminology and Public Policy, 12(1), 65–91. Harris, K., Gringart, E., & Drake, D. (2014). Understanding the role of social groups in radicalisation. Edith Cowan University Australian: Security and Intelligence Conference. Retrieved November 4, 2019, from http://ro.ecu.edu.au/cgi/viewcontent.cgi?article=1034&context=asi Hirschberger, G., Ein-Dor, T., Leidner, B., & Saguy, T. (2016). How is existential threat related to intergroup conflict? Introducing the multidimensional existential threat (MET) model. Frontiers in Psychology, 7(1877). HM Government. (2010). Channel: Supporting individuals vulnerable to recruitment by violent extremists (p. 10). HM Government.

95

Julia Ebner and Harvey Whitehouse HM Government. (2012). Channel: Vulnerability assessment framework. HM Government. HM Government. (2015). Counter-terrorism and security act 2015. HM Government. www.legislation.gov. uk/ukpga/2015/6/contents/enacted Hogg, M., & Adelman, J. (2013). Uncertainty-identity theory: Extreme groups, radical behavior, and authoritarian leadership. Journal of Social Issues, 69(3), 436–454. Ingram, H. J. (2016). A “linkage-based” approach to combating militant Islamist propaganda: A two-tiered framework for practitioners. ICCT. ISD. (2019). Strong cities network. Retrieved November 24, 2019, from https://strongcitiesnetwork.org/en/ incorporating-new-approaches-our-approach-to-evaluating-p-cve-interventions/ Kavanagh, C., Kapitány, R., Putra, I. E., & Whitehouse, H. (2020). Exploring the pathways between transformative group experiences and identity. Frontiers in Psychology, 11(1172), 1–16. Kepel, G. (2003). Jihad: The trail of political Islam. Harvard University Press. Kepel, G. (2015). Terreur dans l’Hexagone: Genèse du djihad français. Gallimar. Khalil, J., Horgan, J., & Zeuthen, M. (2019). The attitudes-behaviors corrective (ABC) model of violent extremism. Terrorism and Political Violence, 1–26. Khalil, J., Horgan, J., & Zeuthen, M. (2020). The ABC model: Clarifications and elaborations. Terrorism and Political Violence, 1–8. Koehler, D. (2016). Understanding deradicalization. Methods, tools and programs for countering violent extremism. Routledge. Kruglanski, A., & Fishman, S. (2009). Psychological factors in terrorism and counterterrorism: Individual, group, and organizational levels of analysis. Social Issues and Policy Review, 3(1), 1–44. Kruglanski, A., Jasko, K., & Webber, D. (2018). The making of violent extremists. Review of General Psychology, 22(1). Kundnani, A. (2015). A decade lost: Rethinking radicalisation and extremism. Claystone. Lee, B., & Knott, K. (2020). More grist to the mill? Reciprocal radicalisation and reactions to terrorism in the far-right digital milieu. Perspectives on Terrorism, 14(3). Lickel, B., Miller, N., Stenstrom, D., Denson, T., & Schmader, T. (2006). Vicarious retribution: The role of collective blaming in intergroup aggression. Personality and Social Psychology Review, 10(4), 372–390. Louis, W. R., McGarty, C., Thomas, E. F., Amiot, C. E., & Moghaddam, F. M. (2018). The power of norms to sway fused group members. Behavioral and Brain Sciences, 41, 1–62. Lygre, R., Eid, J., Larsson, G., & Ranstorp, M. (2011). Terrorism as a process: A critical review of Moghaddam’s “staircase to terrorism”. Scandinavian Journal of Psychology, 52(6), 609–616. Mashuri, A., & Zaduqisti, E. (2015). The effect of intergroup threat and social identity salience on the belief in conspiracy theories over terrorism in Indonesia: Collective angst as a mediator. International Journal of Psychological Research, 8(1), 24–35. McCauley, C., & Moskalensko, S. (2008). Mechanisms of political radicalization: Pathways to terrorism. Terrorism and Political Violence, 20(3), 415–433. McCauley, C., & Moskalenko, S. (2017). Understanding political radicalization: The two-pyramids model. American Psychologist, 72(3), 205. Moghaddam, F. M. (2005). The staircase to terrorism: A psychological exploration. American Psychologist, 60(2), 161–169. NCTC, FBI, & DHS. (2021). U.S. violent extremist indicators, edition 2021. National Counterterrorism Center (NCTC). Nemeth, S. C., Mauslein, J. A., & Stapley, C. (2014). The primacy of the local: Identifying terrorist hot spots using geographic information systems. The Journal of Politics, 76(2), 304–317. Neumann, P. R. (2015). Victim, perpetrators, assets: The narratives of Islamic state defectors. The International Centre for the Study of Radicalisation and Political Violence. Neumann, P. R. (2017). Countering violent extremism and radicalisation that lead to terrorism: Ideas, recommendations, and good practices from the OSCE region. OSCE. Newson, M. (2019). Football, fan violence, and identity fusion. International Review for the Sociology of Sport, 54(4), 431–444. Nossiter, A. (2016, July 12). ‘That ignoramus’: 2 French scholars of radical Islam turn bitter rivals. New York Times. Retrieved November 16, 2019, from www.nytimes.com/2016/07/13/world/europe/france-radicalislam.html Parliament, E. (2014). Preventing and countering youth radicalisation in the EU. EU Directorate General for Internal Policies.

96

Identity and extremism Pisou, D., & Ponesh, S. (2016). Youth, families and communities. Radicalisation Awareness Network (RAN). Powell, L. (2016). Counter-productive counter-terrorism. How is the dysfunctional discourse of prevent failing to restrain radicalisation? Journal for Deradicalisation, 6. Python, A., Bender, A., Nandi, A. K., Hancock, P. A., Arambepola, R., Brandsch, J., & Lucas, T. C. (2021). Predicting non-state terrorism worldwide. Science Advances, 7. Python, A., Illian, J., Jones-Todd, C., & Blangiardo, M. (2018). A Bayesian approach to modelling subnational spatial dynamics of worldwide non-state terrorism, 2010–2016. Journal of the Royal Statistical Society: Series A (Statistics in Society), 182(1), 323–344. Ranstorp, M. (2010). Understanding violent radicalisation: Terrorist and Jihadist movements in Europe. Routledge. Roy, O. (2015, November 19). What is the driving force behind Jihadist terrorism? – A scientific perspective on the causes/circumstances of joining the scene. Speech at BKA Autumn Conference: International Terrorism: How can prevention and repression keep pace? Bundeskriminalamt. BKA Autumn Conference. Retrieved November 19, 2019, from https://life.eui.eu/wp-content/uploads/sites/7/2015/11/OLIVIER-ROYwhat-is-a-radical-islamist.pdf Roy, O. (2017). Jihad and death: The global appeal of Islamic state. Hurst. Sageman, M. (2008). A  strategy for fighting international Islamist terrorists. The Annals of the American Academy of Political and Social Science, 618, 223–231. Silber, M. D., & Bhatt, A. (2007). Radicalization in the west: The homegrown threat. New York Division: NYPD Intelligence. Smith, B. L., Snow, D. A., Fitzpatrick, K., Damphousse, K. R., Roberts, P., Tan, A., . . . Klein, B. (2016). Identity and framing theory, precursor activity, and the radicalization process. U.S. Department of Justice. Speckhard, A., Shajkovci, A., & Yayla, A. (2017). Following a military defeat of ISIS in Syria and Iraq: What happens next after the military victory and the return of foreign fighters? Journal of Terrorism Research, 8(1), 81–89. Sternberg, R. (2003). A duplex theory of hate: Development and application to terrorism, massacres, and genocide. Review of General Psychology, 7. Strindberg, A. (2020). Social identity theory and the study of terrorism and violent extremism. Swedish Defence Research Agency. Swann, W., Buhrmeister, M., Gómez, A., Jetten, J., Bastian, B., Vázquez, A., . . . Zhang, A. (2014). What makes a group worth dying for? Identity fusion fosters perception of familial ties, promoting self-sacrifice. Journal of Personality and Social Psychology, 106, 912–926. Swann, W., Gómez, A., Huici, C., Morales, F., & Hixon, J. (2010). Identity fusion and self-sacrifice: Arousal as catalyst of pro-group fighting, dying and helping behavior. Journal of Personality and Social Psychology, 99, 824–841. Swann, W., Jetten, J., Gómez, A., Whitehouse, H., & Bastian, B. (2012). When group membership gets personal: A theory of identity fusion. Psychological Review, 119, 441–456. Swann, W. B., & Buhrmeister, M. D. (2015). Identity fusion. Current Directions in Psychological Science, 24(1), 52–57. Swann, W. B., Gómez, A., Seyle, D., Morales, J. F., & Huici, C. (2009). Identity fusion: The interplay of personal and social identities in extreme group behavior. Journal of Personality and Social Psychology, 96, 995–1011. Tajfel, H., & Turner, J. (1979). The social identity theory of intergroup behavior. In S. Worchel & W. G. Austin (Eds.), The social psychology of intergroup relations. Nelson-Hall Publishers. Tajfel, H., & Turner, J. (1997). An integrative theory of intergroup conflict. In W. Austin & S. Worchel (Eds.), The social psychology of intergroup relations (pp. 33–47). Brooks Cole Publishing. Taylor, G., Frederiksen, R., Vane, R., & Waltz, E. (2004). Agent-based simulation of geo-political conflict. Innovative Applications of Artificial Intelligence, 884–891. Thomas, E. F., & Louis, W. R. (2014). When will collective action be effective? Violent and non-violent protests differentially influence perceptions of legitimacy and efficacy amongst supporters. Personality and Social Psychology Bulletin, 40, 263–276. UNDP. (2016). Preventing violent extremism through promoting inclusive development, tolerance and respect for diversity. United Nations Development Programme (UNDP). Vergani, M., Iqbal, M., Ilbahar, E., & Barton, G. (2018). The three Ps of radicalization: Push, Pull and Personal. A systematic scoping review of the scientific evidence about radicalization into violent extremism. Studies in Conflict & Terrorism, 43(10), 854–886.

97

Julia Ebner and Harvey Whitehouse Whitehouse, H. (2018). Dying for the group: Towards a general theory of extreme self-sacrifice. Behavioral and Brain Sciences, 41, 1–62. Whitehouse, H. (2021). The ritual animal: Imitation and cohesion in the evolution of social complexity. Oxford University Press. Whitehouse, H., Jong, J., Buhrmeister, M. D., Gómez, A. B., Bastian, C. M., Newson, M., . . . Gavrilets, S. (2017). The evolution of extreme cooperation via shared dysphoric experiences. Scientific Reports, 7, 1–10. Whitehouse, H., McQuinn, B., Buhrmeister, M., & Swann, W. (2014). Brothers in arms: Libyan revolutionaries bond like family. Proceedings of the National Academy of Sciences USA, 111, 17783–17785. Wiktorowicz, Q. (2005). Radical Islam rising: Muslim extremism in the west. Rowman & Littlefield Publishers.

98

7 SOCIAL MOVEMENT THEORY AND RESEARCH ON RADICALISATION Stefan Malthaner

Introduction The relationship between research on radicalisation and political violence, on the one hand, and social movement studies, on the other, has been not only close but also complicated. They are close because, for one thing, the phenomena they examine are often part of the same political processes: militant underground organisations and terrorist cells frequently emerge in the context of protest movements, as precursors or as ‘fallout’ of broader waves of mobilisation; and terrorist attacks at times result from the escalation of other, initially non-violent, repertoires of protest (see Bosi et al., 2019; della Porta, 2013; Gunning, 2009, p. 156). Moreover, there is an overlap in some of the aspects of these processes that the two fields are interested in, such as interpretative and cognitive processes (whether as ‘ideology’ or ‘framing’) or the question why individuals end up participating in movements or militant groups. Yet, for quite some time, both pursued their research interests largely separately – as social movement scholars had, by and large, ‘little to say about terrorism’ (Goodwin, 2004, p. 259), and terrorism research, in turn, tended to isolate and de-contextualise violent groups from broader movements (Tilly, 2004; della Porta, 2013) – and within rather different analytical paradigms. This chapter takes a closer look at the evolution of social movement theory and, in particular, a more recent literature that has emerged at the intersection of social movement studies and research on radicalisation, political violence, and civil wars. It argues that the interaction between these fields has been very productive, drawing attention to the relevance of networks, socialisation, framing, or spaces/milieus in research on radicalisation, and contributing to an analytical shift towards context, relations, and processual approaches more generally. Recent research on the role of emotions as well as performative and experiential aspects of protest and micro-mobilisation might provide further, so far underexploited analytical resources for the study of radicalisation. Yet, a critical evaluation of the use of social movement theory in research on radicalisation also shows that underlying differences have not disappeared, and that the epistemological implications of this perspective have not yet been fully appreciated. At the same time, we have to ask whether some of the (often unacknowledged) premises of social movement studies – in particular the notion of protest as by and large organised and politically rational collective claim-making

99

DOI: 10.4324/9781003035848-9

Stefan Malthaner

addressing the state – may limit its applicability and usefulness in the face of a transforming landscape of (semi-)political violence.

The evolution of social movement studies and research on political violence: context and process The famously turbulent 1960s not only catapulted social movements to the top of the research agenda across the social sciences but – on the background of a more reflective and critical academic debate in sociology and other disciplines – also induced a paradigmatic shift in how protest was conceptualised and analysed. Its antipole, in opposition to which the emerging field of social movement studies defined itself, was hitherto prevalent collective behaviour-, relative deprivation-, or status inconsistency-theories, which, for all their differences, were seen as sharing a notion of protest as an irrational, quasi-psychological reaction to societal strain or breakdown, entirely distinct from ‘normal’, institutional politics (see Jenkins, 1983, pp. 528–530; McAdam, 1982, pp. 5–18). A new generation of social movement scholars, in contrast, conceived of protest as ‘politics by other means’ – as political challenges raised by organised collective actors who respond rationally to opportunities and constraints in their environment, thus, as Buechler put it, ‘recasting the study of collective action from an instance of deviance and social disorganization to a topic of political and organizational sociology’ (Buechler, 2011, pp. 111). Within that perspective, various theoretical approaches were developed, one of the most prominent being Resource Mobilisation Theory (McCarthy & Zald, 1977), which, inspired by economic approaches, focused on the availability of financial and organisational resources and the ability of social movement organisations (SMOs) to act as ‘movement entrepreneurs’ who strategically deploy these resources to produce incentives for participants and external consensus. A parallel strand that later became known as the Political Process approach, in contrast, put stronger emphasis on political conflict, interests, political opportunities, and the creation or appropriation of organisational networks (Tilly, 1978; McAdam, 1982). Arguing that protest responds to – and requires – openings in the institutional political system as well as informal, cultural forms of political opportunities, political process scholars examined the ways regime shifts, periods of political instability, or changes in the composition of elites facilitate mobilisation and shape the repertoires of action adopted by social movements. Equally important, movements were understood to respond to political context in the form of threats and closing political opportunities (see also Bosi et al., 2019). Thereby, as McAdam made clear, processes of mobilisation were not seen as determined by the given structure of political opportunities alone but as dynamically shaped by the social control response of other groups and, in particular, the state, which means that with ‘the outbreak of the insurgency, . . . the movement itself introduces a new set of causal dynamics into the study of collective protest activity’ (McAdam, 1982, p. 53). Thus, resource mobilisation and political process approaches introduced, among other things, two aspects that later also became central to research on political violence: first, an emphasis on context, in the sense of situating protest in its political environment; and, second, an at least rudimentary idea of process, more specifically dynamics resulting from sequences of interaction between movements and the state (and other parties), which retroact on the conditions and motivations that gave rise to a protest movement. Yet, for quite some time, the relationship between the emerging field of social movement studies and the study of violence was an ambivalent one, despite the extensive work on revolutions and violent insurgencies done by some of its main proponents (e.g. Tilly, 1978). Social movement scholars often wrote from a perspective of sympathy with the movements they studied, and, as Piven and Cloward have pointed out, the fact that they emphasised continuities with conventional, 100

Social movement theory and research on radicalisation

institutional politics, in order to normalise protest as ‘politics by other means’, came at the price of losing sight of the violence that frequently accompanies protest movements (Piven & Cloward, 1992; Piven, 2012, p. 19), and many social movement scholars shied away from addressing the apparent connection between the two. Nevertheless, in the late-1980s, a number of researchers started to study the formation of militant underground groups in the context of protest movements and the escalation of violence during particular phases of cycles of protest (e.g. Della Porta & Tarrow, 1986; Della Porta, 1988, 1995; White, 1993; Zwerman et al., 2000). In this line of research, the term ‘radicalization’ emerged to denote shifts towards violent repertoires of action as well as changes in frameworks of interpretation, analysed in particular not only at the level of groups or movements but also with respect to individual pathways towards participation in militant activism and embedded in its social and political context as well as broader processes of contentious interactions with the state (Della Porta & LaFree, 2012, p. 6; see also Malthaner, 2017, pp. 372–378). It thus combined insights from resource mobilisation, political opportunities, and political process approaches with (at that time) more recent work on individual pathways towards participating in political activism (‘micro-mobilization’) (McAdam, 1986; Snow et al., 1980) and framing and frame alignment (Snow et al., 1986; Snow, 2013), mirroring what by then was regarded as the dominant theoretical ‘synthesis’ in social movement studies (McAdam et al., 1996). Research on micro-mobilisation was linked to the Political Process perspective and later became particularly relevant to research on radicalisation, as it sought to explain how people ended up participating in (‘high-risk’) political activism. Sceptical of notions of protestors as deviants or misfits, these works emphasised the interpersonal processes that shaped recruitment – in particular the role of pre-existing social ties and activist networks. The former refers to the finding that recruitment into activist groups is often initiated via personal relationships and that sustained commitment and progression to more demanding or high-risk forms of activism depend on activist or militant networks that emerge during the process of mobilisation (i.e., Snow et al., 1980; McAdam, 1986; Della Porta, 1992, p. 8; McAdam  & Paulsen, 1993; Passy, 2003; Diani, 2013). As this line of research pointed out, close personal bonds formed in subcultural milieus and radical networks are a powerful inducement to participation in militant forms of protest and generate personal trust and loyalty that sustain commitment under pressure (Snow et al., 1980; McAdam, 1986; Della Porta, 1992). Thus, ‘low-risk’ activism in a broader social movement can contribute to ‘paving the way’ towards high-risk forms of activism as it provides tentative, safe forays into new roles, facilitates connections to other activists, and entails processes of socialisation that contribute to the adoption of perceptions, attitudes, values, and identities (McAdam, 1986, pp. 69–70). Yet, work at the intersection of research on protest and political violence, in particular the work of Donatella della Porta, thereby also went beyond the state of social movement studies at the time in that she closely interlinked various levels of analysis, in particular individual processes of micro-mobilisation, organisational transformation, and interactions between movements and their adversaries. In her comparative study of the leftist movements in Italy and Germany, for example, she showed not only that violence emerged as the outcome of a process of adaptation and escalation between social movements and their opponents, but also that the latter produced – and was, in turn, shaped by – a process of organisational and individual radicalisation, which entailed the isolation of clandestine cells as well as a shift in frames of interpretations and perceptions of the ‘enemy’, producing ‘spirals of negative feedback’ that drove the escalation of violence and replaced the actors original political objectives (Della Porta, 1995, pp. 111, 136). The analytical perspective that emerged from these works thus remained, in a way, faithful to the idea of continuity with ‘normal’ politics by refusing to exceptionalise political violence and instead tracing the emergence of violence as one of several forms of confrontation 101

Stefan Malthaner

within a wider repertoire of protest actions and strategies: movements, but also violent groups, adopt protest tactics from available repertoires in response to changing environments and to the actions of their opponents. This perspective, thus, contextualises violence, first, by situating militant groups within the broader field of actors involved in the conflict, including movements, state authorities, police, and counter-movements. Second, it places violent radicalisation within the causal dynamics of broader processes of contention, as the dynamic interactions of the various parties contribute to cycles of protest in which violence may emerge as a product of escalation, a trigger of broader mobilisation, or a symptom of decline. Violent forms of action, as della Porta put it, ‘emerge, and are transformed, in the course of physical and symbolic interactions among social movements and not only their opponents but also their potential allies’ (Della Porta, 2013, p. 19). This emphasis on context and process in research on social movements1 resonated with a parallel development in the study of political violence and civil wars (Crenshaw, 1981, 1995; Neidhardt, 1981, 1982; Waldmann, 1992, 1998), giving rise to debates and collaborations, and with the turn of the millennium, a vibrant literature emerged at the intersection of these fields (see, e.g., Alimi, 2011; Della Porta, 1995, 2008, 2013; Fillieule, 2010, 2015; Goodwin, 1997, 2006, 2012; Gunning, 2007, 2009; Hafez, 2004; Hegghammer, 2006, 2010; Malthaner, 2011; Viterna, 2013; Wiktorowicz, 2004, 2005; Wood, 2003; Zwerman et al., 2000; Zwerman & Steinhoff, 2005). Thereby, building on the Contentious Politics Paradigm (McAdam et al., 2001; Tilly & Tarrow, 2007), a number of scholars sought to develop a more systematic understanding of processes of radicalisation based on the notion of mechanisms (Alimi et al., 2012, 2015; Bosi et al., 2014; Della Porta, 2013). The basic idea behind this line of research is to identify recurring causal patterns – mechanisms – that, in varying combinations and with context-specific outcomes, shape processes of radicalisation and escalation. Della Porta, for example, examined a set of characteristic mechanisms shaping processes of radicalisation, which include escalating policing, that is, the reciprocal adaptation of forms of policing and repertoires of protest, often in a dynamic of mutually reinforced escalation (Della Porta, 2013, pp. 67–69, 74–76). Another mechanism, competitive escalation, results from the dynamics of interaction between similar activist groups within a movement that compete for the same constituencies and resources, which may lead to political outbidding. Moreover, several organisational and cognitive mechanisms can be identified that over the course of processes of escalation, reshape militant groups and their perceptions and frameworks of interpretation, resulting in increasing organisational isolation and ideological radicalisation (organisational compartmentalisation and ideological encapsulation) (Della Porta, 2013, pp. 146–152, 176–178, 206–209).2 Social movement scholars have also contributed to research on radicalisation and political violence in other ways. In addition to works that further develop and apply a framing perspective to the study of militant movements and terrorism (see Alimi & Johnston, 2014; Snow & Byrd, 2007), the concept of identity has been used, for example, to analyse processes of escalation and radicalisation by focusing on the transformation of notions of collective identity and processes of identification (van Stekelenburg & Klandermans, 2010, 2011; see also Della Porta, 1995, pp. 179–181; Cross & Snow, 2011, pp. 118–119). Finally, studies on the social environments and spatial settings of radicalisation have referred to the notion of ‘safe spaces’ (Cross & Snow, 2011; Polletta, 1999; see also Ancelovici, 2021). Free spaces or safe spaces denote small-scale settings within a community or movement that are to some extent removed from the control of authorities or opponents and play a crucial role in allowing for movement activities that generate the cultural challenge that precedes or accompanies political mobilisation and facilitate the formation of mobilising networks (see, e.g., Polletta, 1999). For radical activists, as Cross and Snow argue, free spaces are 102

Social movement theory and research on radicalisation

particularly important, as they ‘give radicals places where they can engage in radical identity work, meet like-minded activists, and even do some limited planning of radical actions’ (Cross & Snow, 2011, p. 119).

Social movement theory and research on radicalisation: networks, socialisation, and radical milieus In research on terrorism and political violence, radicalisation became prominent as a concept and developed into a field of research of its own, particularly in the historical context after 9/11 and in response to the phenomenon of ‘home grown’ jihadist terrorism, that is, violent attacks perpetrated by young men (rarely women) who had been born or raised in Europe or North America (Neumann, 2008, p. 3; Sedgwick, 2010, p. 480; Kundnani, 2012, pp. 4–7; Crone, 2016, p. 589; see also Malthaner, 2017). The particular configuration of this form of political violence as local manifestations of a transnational movement had far-reaching implications for the analytical approaches applied to studying it. Terrorist cells and individual perpetrators responsible for attacks in Europe or the United States considered themselves to be part of a worldwide jihadist movement, but to Western observers, the latter not only seemed more abstract and elusive than the protest movements of the 1960s and more of an extremist religious movement than an actual political protest movement, but was also understood as something alien to Western societies. As a result, the research focus shifted to individual trajectories of radicalisation and the way personal and sociocultural conditions produced individual vulnerabilities and predispositions to radicalisation – for example, how socio-economic inequalities, marginalisation, and experiences of racism and discrimination combined with the fragile and conflicted identities of second-generation migrants – as well as the particular factors and settings that facilitated and drove individual pathways towards jihadist terrorism (for a discussion, see Crone, 2016; Malthaner, 2017). Social movement theory, thereby, became a reference for critical, self-reflective debates in terrorism and radicalisation research (Beck, 2008; Della Porta, 2008, 2013; Crone, 2016; Gunning, 2007); but selected elements of social movement theory were also adapted and applied to the explanation of individual trajectories of jihadist radicalisation, in particular works on network dynamics in micro-mobilisation. Notably, the influential works of Marc Sageman (2004, 2008) and Quintan Wiktorowicz (2004, 2005) brought social movement theory into the debate on jihadist radicalisation. Sageman, thereby, emphasised that radicalisation was typically facilitated by personal ties and shaped by small-group dynamics (Sageman, 2004, pp. 99, 111, 2008, pp. 86–87, 116–117), thus echoing insights from research on individual participation in protest movements. Wiktorowicz, who built his analysis of the al-Muhajiroun movement in the United Kingdom even more explicitly on social movement theory, highlighted the role of pre-existing personal relations, arguing that ‘attitudinal affinity may predispose an individual to join a movement, but social ties are critical for transforming interest and availability into actual activism’ (Wiktorowicz, 2005, p. 15). Moreover, he drew on framing approaches, emphasising that recruitment via networks depends on frame alignment between individuals and movements and, in particular, stressed processes of socialisation within movements (Wiktorowicz, 2005, p. 16). In other words, instead of focusing exclusively on individuals, Wiktorowicz examined how perceptions, motivations, and frameworks of interpretation are adopted as a result of interactions in the context of movements and milieus, which he conceived of as ‘networks of shared meanings’ – a concept developed by Italian social movement scholar Alberto Melucci (1989). He observed that the trajectory of many members of al-Muhajiroun started with a quest for meaning, often triggered by personal crises and moments of cognitive opening, which was then channelled towards religiosity by personal networks or, in 103

Stefan Malthaner

some cases, the movement’s active outreach activities (Wiktorowicz, 2005, pp. 20–24, 85–86, 92–94). Once an individual started to participate in movement activities, interactions with other activists and more experienced mentors – as well as teaching sessions in study groups, for example – contributed to an (also emotional) intensification of commitment and a process of socialisation that resulted in the adoption of more radical religious and political beliefs as well as more demanding and ultimately militant forms of activism (Wiktorowicz, 2005, pp. 24–25, 176–182; see also Borum, 2011a/2011b; Dalgaard-Nielsen, 2010, pp. 801–803). Building on social movement theory as well as Wiktorowicz’ work, Malthaner and Waldmann developed a contextualising approach to analyse radicalisation and political violence based on the concept of radical milieu, which refers to the immediate (formative and supportive) social environment of clandestine groups that can comprise a number of very different settings and places (and spaces) (Malthaner & Waldmann, 2012, 2014; Malthaner, 2014). Radical milieus contain activist networks that enable and shape pathways towards joining armed groups, and they are the formative social environment in which individuals adopt frameworks of interpretation, values, and symbols and in which they share experiences of persecution and violent confrontations that emotionally reinforce identities and commitment. Thereby, radical milieus are, in turn, situated in relation to broader movements or religious or ethnic communities. They form, as Della Porta has shown in her similar work (using the concept of ‘micro-mobilization settings’; Della Porta, 2013, p. 117), in ambiguous and sometimes conflictual processes, in some instances by separating and isolating themselves from broader movements. And at times, virtual spaces can act as online radical milieus or mediate and complement face-to-face interactions among activists (Conway, 2012). The notion of radicalisation as a pathway facilitated and shaped by networks has found its way into numerous studies on jihadist terrorism and other forms of political violence (see, e.g., Bokhari et al., 2006; Neumann & Rogers, 2008). Yet, an important insight from social movement studies that has not always been taken into consideration in these studies is the fact that pre-existing personal ties and activist networks are not simply a conveyor belt towards radicalisation, but can have complex, ambivalent effects. In particular, family ties as well as relations to non-militant activists may constrain or prevent participation in militant activism, which is why Snow, Zurcher, and Ekland-Olson have identified ‘availability’, that is, the absence of constraining ties to other groups or individuals, as a precondition of participation (Snow et al., 1980, pp. 789–790). Among the radicalisation scholars who have taken up this insight from a social-psychological perspective are McCauley and Moskalenko, who list ‘unfreezing’ – the sudden weakening of constraining social ties – as a mechanism of radicalisation (McCauley & Moskalenko, 2008).3 Similarly, at the level of milieus and movements, Malthaner and Waldmann emphasise that while, on the one hand, activism in non-militant parts of a movement may contribute to radicalisation by facilitating connections and socialisation into new identities and frameworks of interpretation, they may, on the other hand, also inhibit pathways towards violence by offering, for example, alternative (non-militant) forms of activism as well as a viable ‘exit’ option for activists who disengage from violent clandestine groups (Malthaner & Waldmann, 2014).

Explanatory perspectives Thus, in sum, during the past two decades, on the background of a growth of research and theoretical exchange at the intersection of social movement studies with research on political violence, analysts of radicalisation have adopted various concepts and elements of social movement theory. Yet, as mentioned earlier, in social movement theory – of the Political Process brand and even more in the more recent Contentious Politics paradigm – these concepts and elements are 104

Social movement theory and research on radicalisation

embedded in a broader theoretical perspective and a logic of explanation that emphasises political conflict, interactions (and continuities) between protest and ‘normal’ institutional politics, as well as the autonomous dynamics of processes of mobilisation; in other words, a concern for context and processual dynamics within a fundamentally relational perspective (McAdam et al., 2001; Della Porta, 2013; Bosi & Malthaner, 2022; Bosi et al., 2014; Malthaner et al., 2022). These more elementary aspects of social movement theory have exerted a certain influence upon – and resonated with independent, parallel developments within – research on radicalisation, too, but did so within certain limits. The processual perspective developed in social movement theory corresponded with notions of processuality that had emerged, in parallel, in research on radicalisation, for example in John Horgan’s famous essay ‘From Profiles to Pathways’ (Horgan, 2008; see also Horgan, 2005, 2009; Taylor & Horgan, 2006), and it inspired others in this field to develop more systematic mechanismprocess explanatory approaches (e.g., McCauley & Moskalenko, 2008, 2011). While it is true, as Peter Neumann pointed out, that the idea that ‘radicalization is a progression which plays out over a period of time and involves different factors and dynamics’ has been, in some form or another, widely accepted among scholars on radicalisation from relatively early on (Neumann, 2013, p. 874), what makes the understanding of the process developed in these works non-trivial is the notion of emergence, autonomous causal dynamics, and non-linearity. This means that factors preceding radicalisation, be they socio-structural preconditions or individual predispositions, cannot fully account for the outcome because turning points can reshape the trajectory at various junctures, and attitudes and motives are not pre-given but may be produced and transformed in the process of radicalisation itself. As earlier research on micro-mobilisation had already made clear, ‘the “whys” or “reasons” for joining arise out of the recruitment itself’ (Snow et al., 1980, p. 799). Quite similarly, Horgan and Taylor argued that ‘answering questions about why people may wish to initially become involved in terrorism may have little bearing on what they do . . . as terrorists or how they actually become engaged in specific terrorist operations’ (Horgan, 2008, p. 81; see Taylor & Horgan, 2006, pp. 589–590). This perspective, which was further developed, among others, by social movement scholar Olivier Fillieule, emphasises that trajectories of radicalisation are not linear but complex and context-dependent and must not be reduced to a linear path leading to the violent act as its point of culmination (Horgan, 2009; Fillieule, 2005, 2010, 2015). To capture the dynamics that shape these processes, McCauley and Moskalenko specified a set of recurring mechanisms at different levels of analysis, not only individual trajectories but also collective processes of radicalisation at the group and inter-group level (McCauley & Moskalenko, 2008, 2011). Closely related to process, context – and the relational quality of processes of radicalisation – have been emphasised in various studies and explanatory approaches, for example, by drawing attention to settings of radicalisation (Neumann & Rogers, 2008), subcultures (Pisoiu, 2014; see also the chapter by Pisoiu in this volume), radical milieus (Malthaner & Waldmann, 2014), and multi-level ecological perspectives. Yet, despite the growing emphasis on process, relations, and context – and frequent references to social movement theory – there remains a quite fundamental difference with respect to the underlying logic of explanation adopted by many studies on radicalisation. Not only are relations, milieus, and movements in many cases studied mainly as ‘context’ of pathways of individual radicalisation, rather than as part of a dynamic relational field that is itself part of the phenomenon to be understood and explained. Moreover, relational dynamics are often, implicitly or explicitly, seen as ancillary to another process, which is seen as the one that actually ‘explains’ political violence: the gradual transformation of attitudes and ideological beliefs and the emergence of a readiness or propensity for violence. Charles Tilly has pointed out this tendency of terrorism 105

Stefan Malthaner

research to rely on what he calls dispositional explanations that seek to explain the actions of terrorists ‘by means of their orientations just before the point of action’ (Tilly, 2005, p. 19). In other words, there is a tendency to reduce the question of how we can explain terrorism to the question of how someone reaches a state (of mind, perceptions, motivation, and emotion) that predisposes them to violence: the state of having become ‘radicalized’. Relational factors and context, insofar as they are taken into account, are treated ‘not as direct causes but as factors that promote or inhibit the central cause of terrorism: formation of a consciousness dividing the world starkly into us and them’ (Tilly, 2005, p. 20). While Tilly did not argue that the transformation of perceptions, attitudes, or orientations was irrelevant to understanding terrorism, he questions the tendency to regard dispositions as some kind of causal node (that ultimately explains violence), suggesting instead to put relations and relational dynamics at the centre of the analysis.

Challenges ahead and ways forward Social movement studies and research on radicalisation are, one could say, close relatives, even if at times estranged ones. The phenomena they study are connected, and they share an interest in some of the same analytical aspects, such as individual pathways towards participation, interpretative processes, spaces and milieus, or the impact of state policies. During the past two decades, empirical research and theoretical insights from social movement studies have informed approaches to study and theorise radicalisation as well as critical debates reflecting upon the limitations and political implications of radicalisation as an explanatory paradigm (see, e.g., Crone, 2016; Gunning, 2009; Malthaner, 2017; for an overview of theoretical approaches, see also Pisoiu & Hain, 2018). Moreover, a number of more specific elements of social movement theory, such as framing theory, the notion of safe spaces, networks, competition (intergroup radicalisation), and political opportunity structures, have been applied very fruitfully in recent empirical and conceptual work. Marsden (2016), for example, developed a typology of militant organisations based on social movement theory, focusing on organisational characteristics and capacities as well as characteristics of the political context in which these groups operate. To some extent, building on the work of Snow and Byrd, who applied framing theory to the study of radical ideology in the case of Al-Qaeda (Snow & Byrd, 2007), Andersen and Sandberg (2020) drew on the frame concept in their analysis of ‘Islamic State’ (IS) propaganda, as had Berntzen and Sandberg (2014) in an earlier study, which used framing theory in order to situate the attacks of Anders Behring Breivik within the debates and meaning-making processes of the broader Anti-Islamic Movement. Similarly, if with a somewhat different focus, Papale applied the framing concept to examine the imagery used by Al-Shabaab in their recruitment and mobilisation strategy in Kenya (Papale, 2021). In his research on the Irish Republican Movement, Clubb (2016) used social movement theory to study de-radicalisation and disengagement processes, focusing on framing and the structural transformation of movements. Whereas Lygren and Ravndal, for example, applied social movement theory to develop a mechanisms approach to explain different processes of ‘reciprocal intergroup radicalization’ (RIR) emerging from interactions between Islamists and Anti-Islamists in Britain and Norway (Lygren & Ravndal, 2021; on movement-countermovement interactions, see also De Fazio, 2017). Yet, as productive as this interaction has been, there also seem to be limits to the applicability of social movement theory in radicalisation studies. Frictions result not only from the incompatibility of explanatory logics but also from different normative positions. Moreover, and maybe most significantly, research on radicalisation is currently facing a transforming landscape of political violence, for which ‘classic’ social movement theories such as Resource Mobilization or Political 106

Social movement theory and research on radicalisation

Process approaches might not be particularly well-equipped. It seems obvious that attacks by lone actors and isolated micro-cells, dispersed attacks (hate crimes) against marginalised groups, as well as phenomena that challenge the conceptual differentiation between political and non-political violence (and between organised crime and terrorism), cannot be readily understood in terms of organised political claim-making directed at the state. Yet, rather than rendering the intellectual exchange between radicalisation research and social movement studies obsolete, I would argue that it might, on the contrary, be even more important today to develop new ways of theorising transforming phenomena of political violence instead of relegating them to the realms of deviant behaviour and individual psychology. At the most general level, reflections by social movement scholars on new forms of political mobilisation and contestation in an era of post-politics (see, e.g., Blühdorn & Deflorian, 2021) can provide a point of departure for reflecting on changing contexts of violent radicalisation. Changing societal realities and understandings of what is ‘political’ and what is protest – as well as the transformation of practices, repertoires, and spaces of protest – impact the shape of radical politics and individual pathways towards militant activism. Moreover, various bodies of work might offer ways to theorise more specific aspects of these phenomena. For example, by exploring social movement research on the diffusion of practices and repertoires of action, or on ‘signaling spirals’, radicalisation scholars might be able to conceptualise the dynamics of more emergent, de-centralised, and weakly coordinated violent collective action, including clusters of lone-actor attacks. Similarly, more recent work at the intersection of social movement studies and radicalisation research on the eventfulness of activism and its impact on identities and relationships among participants, emotions, as well as experiences – including the affective and emotional effects of experiences and their aesthetic and visceral, embodied aspects – has proven productive in charting new ways of understanding radicalisation as well as radical activism (and violence) as a lived reality (see Pilkington, 2016; McDonald, 2018; Crone, 2016; Malthaner, 2017; Stekelenburg, 2017; see also della Porta, 2011, 2020). By reconsidering the relation between ‘intellectual’ radicalisation and violence, these perspectives seem much more suited to analysing new phenomena of radicalisation, without de-politicising them and to capture different and more fluent patterns of social embeddedness and connectivity. Drawing on elements of social movement theory, Kevin McDonald, for example, develops a perspective that analyses the experiential, sensual, and affective dimensions of radicalisation; that is, patterns of embodied meaning-making constructed around relationships with ‘us’ (notions of community), ‘you’ (conflict and competition with a group constructed as ‘other’), and ‘self’ (notions of self-transformation and self-discovery) (McDonald, 2018, pp. 20-21). This understanding allows us not only to develop a new and more nuanced understanding of what is ‘political’ about the experience of radicalisation but also to explore radicalisation as ‘something produced by active participants, attempting to make sense of themselves and their world’ (McDonald, 2018, p. 15). Moreover, one sub-field in which social movement theory has been adopted to great benefit is research on the radical right (Blee, 2003, 2017, 2018; Castelli Gattinara & Pirro, 2018; Pilkington, 2016), which seems among the currently most relevant not only because right-wing violence is on the rise but also because some of the forms (individual or spontaneous group attacks) and targets (powerless, marginalised groups) of violence challenge the notion of political violence as emerging from organised contentious politics directed at powerful elites. In addition to work that explores the dynamics of more dispersed waves of violent attacks, Hilary Pilkington, for example, develops a perspective that offers new ways of understanding the experience of participating in radical movements, emphasising the interrelation of emotional and politically expressive dimensions of political activism (Pilkington, 2016). Thus, we are seeing a transformation of forms, actors, and contexts of radicalisation, no doubt. But the fact that research at the intersection of social movement studies and 107

Stefan Malthaner

radicalisation studies seems to face changing and, in a certain way, liminal phenomena can also be understood as a valuable analytical opportunity, which gives theoretical developments at this intersection the potential to contribute something important to both fields.

Notes 1 For a more general discussion of interaction, process, and temporality in social movements, see Jasper and Duyvendak (2015), Jasper (2021), and Gillian and Edwards (2020); for a discussion of context in social movement studies, see Ancelovici (2021). 2 Similar efforts have been undertaken by Alimi, Bosi, and Demetriou, who specify a set of recurring mechanisms as well as three corresponding arenas of interaction as the specific relational contexts from which these mechanisms emerge (2012/2015); as well as by Bosi et al. (2014), who propose a framework of four types of dynamics that shape processes of radicalisation and escalation (dynamics of interaction between oppositional movements and the state, dynamics of intra-movement competition, dynamics of meaning formation and transformation, and dynamics of diffusion). 3 For an application of this approach to the study of lone-actor terrorism see, Lindekilde et al. (2019), Malthaner and Lindekilde (2017), as well as the chapter by O’Connor, Lindekilde, and Malthaner in this volume.

References Alimi, E. Y. (2011). Relational dynamics in factional adoption of terrorist tactics: A comparative perspective. Theory and Society, 40(1), 95–119. Alimi, E. Y., Bosi, L., & Demetriou, C. (2012). Relational dynamics and processes of radicalization: A comparative framework. Mobilization, 17(1), 7–26. Alimi, E. Y., Demetriou, C., & Bosi, L. (2015). The dynamics of radicalization: A relational perspective. Oxford University Press. Alimi, E. Y., & Johnston, H. (2014). Contentious interactions, dynamics of interpretations, and radicalization: The Islamization of Palestinian nationalism. In L. Bosi, C. Demetriou, & S. Malthaner (Eds.), Dynamics of political violence: A process-oriented perspective on radicalization and the escalation of political conflict (pp. 169–188). Ashgate. Ancelovici, M. (2021). Conceptualizing the context of collective action. Social Movement Studies, 20(2), 125–138. Andersen, J. C., & Sandberg, S. (2020). Islamic state propaganda: Between social movement framing and subcultural provocation. Terrorism and Political Violence, 32(7), 1506–1526. Beck, C. J. (2008). The contribution of social movement theory to understanding terrorism. Sociology Compass, 2(5), 1565–1581. Berntzen, L. E., & Sandberg, S. (2014). The collective nature of lone wolf terrorism: Anders Behring Breivik and the anti-Islamic social movement. Terrorism and Political Violence, 26(5), 759–779. Blee, K. M. (2003). Inside organized racism: Women in the hate movement. University of California Press. Blee, K. M. (2017). How the study of white supremacism is helped and hindered by social movement research. Mobilization: An International Quarterly, 22(1), 1–15. Blee, K. M. (2018). Understanding racist activism: Theory, methods, and research. Routledge. Blühdorn, I., & Deflorian, M. (2021). Politicisation beyond post-politics: New social activism and the reconfiguration of political discourse. Social Movement Studies, 20(3), 259–275. Bokhari, L., Hegghammer, T., Lia, B., Nesser, P., & Toennessen, T. H. (2006). Paths to global Jihad: Radicalisation and recruitment to terror networks, Proceedings from a FFI Seminar. Norwegian Defence Research Establishment (FFI). Retrieved June 8, 2009, from www.mil.no/multimedia/archive/00077/ Paths_to_global_jiha_77735a.pdf Borum, R. (2011a). Rethinking radicalization. Journal of Strategic Security, 4(4), 1–6. Borum, R. (2011b). Radicalization into violent extremism I: A review of social science theories. Journal of Strategic Security, 4(4), 7–36. Bosi, L., della Porta, D., & Malthaner, S. (2019). Organizational and institutional approaches: Social movement studies perspectives on political violence. In E. Chenoweth, R. English, A. Gofas, & S. N. Kalyvas (Eds.), The Oxford handbook of terrorism. Oxford University Press.

108

Social movement theory and research on radicalisation Bosi, L., Demetriou, C., & Malthaner, S. (Eds.). (2014). Dynamics of political violence: A process-oriented perspective on radicalization and the escalation of political conflict. Ashgate. Bosi, L., & Malthaner, S. (2022). Processual perspectives. In D. A. Snow, D. della Porta, B. Klandermans, & D. McAdam (Eds.), The Wiley-Blackwell encyclopedia of social and political movements. Wiley-Blackwell. Buechler, S. M. (2011). Understanding social movements: Theories from the classical era to the present. Routledge. Castelli Gattinara, P., & Pirro, A. L. P. (2018). The far right as social movement. European Societies, 21(4), 447–462. Clubb, G. (2016). Social movement de-radicalization and the decline of terrorism: The morphogenesis of the Irish republican movement. Routledge. Conway, M. (2012). From al-Zarqawi to al-Awlaki: The emergence and development of an online radical milieu. CTX: Combating Terrorism Exchange, 2(4), 12–22. Crenshaw, M. (1981). The causes of terrorism. Comparative Politics, 13(4), 379–399. Crenshaw, M. (1995). Terrorism in context. The Pennsylvania State University Press. Crone, M. (2016). Radicalization revisited: Violence, politics and the skills of the body. International Affairs, 92(3), 587–604. Cross, R., & Snow, D. A. (2011). Radicalization within the context of social movements: Processes and types. Journal of Strategic Security, 4(4), 115–129. Dalgaard-Nielsen, A. (2010). Violent radicalization in Europe: What we know and what we do not know. Studies in Conflict and Terrorism, 33, 797–814. De Fazio, G. (2017). Hostile countermobilization and political violence: Loyalist contention and radicalization in Northern Ireland, 1968–1969. Research in Social Movements, Conflicts and Change, 41, 3–31. Della Porta, D. (1988). Recruitment processes in clandestine political organizations: Italian left-wing terrorism. In B. Klandermans, H. Kriesi, & S. Tarrow (Eds.), New social movements in Western Europe and the United States (pp. 155–169). JAI Press. Della Porta, D. (1992). Introduction: On individual motivations in underground political organizations. In D. Della Porta (Ed.), Social movements and violence: Participation in underground organizations (pp. 3–28). Jai Press Inc. Della Porta, D. (1995). Social movements, political violence and the state. Cambridge University Press. Della Porta, D. (2008). Research on social movements and political violence. Qualitative Sociology, 31, 221–230. Della Porta, D. (2011). Eventful protest, global conflicts. Distinction Journal of Social Theory, 9(2), 27–56. Della Porta, D. (2013). Clandestine political violence. Cambridge University Press. Della Porta, D. (2020). Protests as critical junctures: Some reflections towards a momentous approach to social movements. Social Movement Studies, 19(5–6), 556–575. Della Porta, D., & LaFree, G. (Eds.). (2012). (De-)Radicalization. Special-Focus of the International Journal of Conflict and Violence, 6(1). Della Porta, D., & Tarrow, S. (1986). ‘Unwanted children’: Political violence and the cycle of protest in Italy, 1966–1973. European Journal of Political Research, 14, 607–632. Diani, M. (2013). Networks and social movements. In D. A. Snow, D. Della Porta, B. Klandermans, & D. McAdam (Eds.), The Wiley-Blackwell encyclopedia of social movements. Blackwell Publishers. Fillieule, O. (2005). Le Désengagement militant. Belin. Fillieule, O. (2010). Some elements of an interactionist approach to political disengagement. Social Movement Studies, 9(1), 1–15. Fillieule, O. (2015). Disengagement from radical organizations: A process and multi-level model of analysis. In B. Klandermans & C. van Stralen (Eds.), Movements in times of transition. University of Minnesota Press. Gillian, K., & Edwards, G. (2020). Time for change. Social Movement Studies, 19(5–6), 501–515. Goodwin, J. (1997). No other way out. Cambridge University Press. Goodwin, J. (2004). Review essays: What must we explain to explain terrorism? Social Movement Studies, 3, 259–265. Goodwin, J. (2006). A theory of categorical terrorism. Social Forces, 84(4), 2027–2046. Goodwin, J. (2012). Introduction to a special issue on political violence and terrorism: Political violence as contentious politics. Mobilization: An International Journal, 17(1), 1–5. Gunning, J. (2007). Hamas in politics. Hurst &Company.

109

Stefan Malthaner Gunning, J. (2009). Social movement theory and the study of terrorism. In R. Jackson, M. Breen Smyth, & J. Gunning (Eds.), Critical terrorism study: A new research agenda (pp. 156–177). Routledge. Hafez, M. (2004). Why Muslims rebel: Repression ad resistance in the Islamic world. Lynne Rienner Publishers. Hegghammer, T. (2006). Terrorist recruitment and radicalization in Saudi Arabia. Middle East Policy, 13(4), 39–60. Hegghammer, T. (2010). Jihadism in Saudi Arabia. Cambridge University Press. Horgan, J. (2005). The psychology of terrorism. Routledge. Horgan, J. (2008). From profiles to pathways and roots to routes: Perspectives from psychology on radicalization into terrorism. Annals of the American Academy of Political and Social Science (618), 80–94. Horgan, J. (2009). Walking away from terrorism. Routledge. Jasper, J. M. (2021). Linking arenas: Structuring concepts in the study of politics and protest. Social Movement Studies, 20(2), 243–257. Jasper, J. M., & Duyvendak, J. W. (Eds.). (2015). Players and arenas: The interactive dynamics of protest. Amsterdam University Press. Jenkins, C. J. (1983). Resource mobilization theory and the study of social movements. Annual Review of Sociology, 9, 527–553. Kundnani, A. (2012). Radicalization: The journey of a concept. Race & Class, 54(2), 3–25. Lindekilde, L., Malthaner, S., & O’Connor, F. (2019). Peripheral and embedded: Relational patterns of loneactor terrorist radicalization. Dynamics of Asymmetric Conflict, 12(1), 20–41. Lygren, S., & Ravndal, J. A. (2021). Why reciprocal intergroup radicalisation happened between Islamists and Anti-Islamists in Britain but not in Norway. Terrorism and Political Violence, 35(2), 390–408. Malthaner, S. (2011). Mobilizing the faithful: The relationship between militant Islamist groups and their constituencies. Campus. Malthaner, S. (2014). Contextualizing radicalization: The emergence of the ‘Sauerland-group’ from radical networks and the Salafist movement. Studies in Conflict and Terrorism, 37(8), 638–653. Malthaner, S. (2017). Radicalization. The evolution of an analytical paradigm. European Journal of Sociology, 58(3), 369–401. Malthaner, S., della Porta, D., & Bosi, L. (2022). The social science of political violence: Processual approaches. In D. Muro & T. Wilson (Eds.), Contemporary terrorism studies. Oxford University Press. Malthaner, S., & Lindekilde, L. (2017). Analyzing pathways of lone-actor radicalization: A relational approach. In M. Stohl, R. Burchill, & S. Englund (Eds.), Constructions of terrorism. University of California Press. Malthaner, S., & Waldmann, P. (Eds.). (2012). Radikale Milieus: Das soziale Umfeld terroristischer Gruppen. Campus. Malthaner, S., & Waldmann, P. (2014). The radical Milieu: Conceptualizing the supportive social environment of terrorist groups. Studies in Conflict and Terrorism, 37(12), 979–998. Marsden, S. V. (2016). A social movement theory typology of militant organizations: Contextualizing terrorism. Terrorism and Political Violence, 28, 750–773. McAdam, D. (1982). Political process and the development of Black insurgency, 1930–1970. The University of Chicago Press. McAdam, D. (1986). Recruitment to high-risk activism: The case of freedom summer. American Journal of Sociology, 92(1), 64–90. McAdam, D., McCarthy, J. D., & Zald, M. N. (1996). Introduction: Opportunities, mobilizing structures, and framing processes – toward a synthetic, comparative perspective on social movements. In D. McAdam, J. McCarthy, & M. N. Zald (Eds.), Comparative perspectives on social movements: Political opportunities, mobilizing structures, and cultural framings (pp. 1–20). Cambridge University Press. McAdam, D., & Paulsen, R. (1993). Specifying the relationship between social ties and activism. American Journal of Sociology, 99, 640–667. McAdam, D., Tarrow, S., & Tilly, C. (2001). Dynamics of contention. Cambridge University Press. McCarthy, J. D., & Zald, M. N. (1977). Resource mobilization and social movements: A partial theory. American Journal of Sociology, 82(6), 1212–1241. McCauley, C., & Moskalenko, S. (2008). Mechanisms of political radicalization: Pathways towards terrorism. Terrorism and Political Violence, 20(3), 415–433. McCauley, C., & Moskalenko, S. (2011). Friction: How radicalization happens to them and us. Oxford University Press.

110

Social movement theory and research on radicalisation McDonald, K. (2018). Radicalization. Polity Press. Melucci, A. (1989). Nomads of the present: Social movements and individual needs in contemporary society. Temple University Press. Neidhardt, F. (1981). Über Zufall, Eigendynamik und Institutionalisierbarkeit absurder Prozesse: Notizen am Beispiel einer terroristischen Gruppe. In H. Von Alemann & H. P. Thurn (Eds.), Soziologie in weltbürgerlicher Absicht: Festschrift für Rene König zum 75. Geburtstag (pp. 243–257). Westdeutscher Verlag. Neidhardt, F. (1982). Soziale Bedingungen terroristischen Handelns. Das Beispiel der ‘Baader-Meinhof-Gruppe (RAF).’ In W. von Baeyer-Katte, D. Claessens, H. Feger, & F. Neidhardt (Eds.), Analysen zum Terrorismus 3: Gruppenprozesse (pp. 318–393). Westdeutscher Verlag. Neumann, P. (2008, January 17–18). Introduction. In Perspectives on radicalization and political violence. Papers from the First International Conference on Radicalization and Political Violence, The International Center for the Study of Radicalization and Political Violence. http://icsr.info/wp-content/uploads/2012/10 /1234516938ICSRPerspectivesonRadicalisation.pdf Neumann, P. (2013). The trouble with radicalization. International Affairs, 89(4), 873–893. Neumann, P., & Rogers, B. (2008). Recruitment and Mobilisation for the Islamist Militant Movement in Europe, ICSR report. http://icsr.info/wp-content/uploads/2012/10/1234516791ICSREUResearchRep ort_Proof1.pdf Papale, S. (2021). Framing frictions: Frame analysis and Al-Shabaab’s mobilization strategies in Kenya. Critical Studies on Terrorism, 14(1), 1–23. Passy, F. (2003). Social networks matter. But how? In M. Diani & D. McAdam (Eds.), Social movements and networks. Relational approaches to collective action (pp. 21–48). Oxford University Press. Pilkington, H. (2016). Loud and proud: Passion and politics in the English defence league. Manchester University Press. Pisoiu, D. (2014). Subcultural theory applied to Jihadi and right-wing radicalization in Germany. Terrorism and Political Violence, 27(1), 9–28. Pisoiu, D., & Hain, S. (2018). Theories of terrorism: An introduction. Routledge. Piven, F. F. (2012). Protest movements and violence. In S. Seferiades & H. Johnston (Eds.), Violent protest, contentious politics, and the neoliberal state (pp. 19–28). Routledge. Piven, F. F., & Cloward, R. A. (1992). Normalizing collective protest. In A. Morris & C. McClurg Mueller (Eds.), Frontiers in social movement theory (pp. 301–325). Yale University Press. Polletta, F. (1999). ‘Free spaces’ in collective action. Theory and Society, 28, 1–38. Sageman, M. (2004). Understanding terror networks. University of Pennsylvania Press. Sageman, M. (2008). Leaderless Jihad: Terror networks in twenty-first century. University of Pennsylvania Press. Sedgwick, M. (2010). The concept of radicalization as a source of confusion. Terrorism and Political Violence, 22(4), 479–494. Snow, D. A. (2013). Framing and social movements. In D. Snow, D. Della Porta, B. Klandermans, & D. McAdam (Eds.), The Blackwell encyclopedia of social and political movements. Wiley-Blackwell. Snow, D. A., & Byrd, S. C. (2007). Ideology, framing processes, and Islamic terrorist movements. Mobilization: An International Quarterly Review, 12(1), 119–136. Snow, D. A., Rochford, B., Worden, S. K., & Benford, R. D. (1986). Frame alignment processes, micromobilization, and movement participation. American Sociological Review, 51, 464–481. Snow, D. A., Zurcher, L. A., & Ekland-Olson, S. (1980). Social networks and social movements: A microstructural approach to differential recruitment. American Sociological Review, 45(5), 787–801. Taylor, M., & Horgan, J. (2006). A conceptual framework for addressing psychological process in the development of the terrorist. Terrorism and Political Violence, 18(4), 585–601. Tilly, C. (1978). From mobilization to revolution. Addison-Wesley. Tilly, C. (2004). Terror, terrorism, terrorists. Sociological Theory, 22(1), 5–13. Tilly, C. (2005). Terror as strategy and relational process. International Journal of Comparative Sociology, 46(1–2), 11–32. Tilly, C., & Tarrow, S. (2007). Contentious politics. Paradigm Publishers. Van Stekelenburg, J. (2017). Radicalization and violent emotions. Political Science & Politics, 50(4), 936–939. Van Stekelenburg, J., & Klandermans, B. (2010). The social psychology of protest. Sociopedia. ISA (e-journal), 1–13. Van Stekelenburg, J., & Klandermans, B. (2011). Radicalization. In A. E. Azzi, X. Chryssochoou, B. Klandermans, & B. Simon (Eds.), Identity and participation in culturally diverse societies: A multidisciplinary perspective. Wiley-Blackwell.

111

Stefan Malthaner Viterna, J. (2013). Women in war: The micro-processes of mobilization in El Salvador. Oxford University Press. Waldmann, P. (1992). Ethnic and sociorevolutionary terrorism: A comparison of structures. International Social Movement Research, 4, 237–257. Waldmann, P. (1998). Terrorismus: Provokation der Macht. Gerling Akademie Verlag. White, R. W. (1993). Provisional Irish republicans: An oral and interpretive history. Greenwood Press. Wiktorowicz, Q. (2004). Islamic activism: A social movement theory approach. Indiana University Press. Wiktorowicz, Q. (2005). Radical Islam rising: Muslim extremism in the west. Rowman & Littlefield. Wood, E. J. (2003). Insurgent collective action and civil war in El Salvador. Cambridge University Press. Zwerman, G., & Steinhoff, P. G. (2005). When activists ask for trouble: State-dissident interactions and the new left cycle of resistance in the United States and Japan. In C. Davenport, H. Johnston, & C. Mueller (Eds.), Repression and mobilization (pp. 85–107). University of Minnesota Press. Zwerman, G., Steinhoff, P. G., & Della Porta, D. (2000). Disappearing social movements: Clandestinity in the cycle of new left protest in the U.S., Japan, Germany, and Italy. Mobilization, 5(1), 85–104.

112

8 CRIMINOLOGICAL PERSPECTIVES ON EXTREMIST RADICALISATION AND TERRORIST ACTS Gary LaFree and Yesenia Yanez

While there were important criminology contributions to the study of violent political extremism1 prior to 2000 (e.g. Hamm, 1993; Kittrie, 1978; Smith, 1994; Turk, 1982), they were relatively uncommon, and the study of violent extremism was not widely recognised as a major criminological specialisation. This situation has changed dramatically in the past two decades. Criminology research on political extremism has expanded enormously since 2000 and is now arguably the fastest-growing specialisation in the field (for reviews, see Forst et al., 2011; Freilich & LaFree, 2015; LaFree & Ackerman, 2009; Lum & Kennedy, 2012; Sageman, 2014). The reasons for this explosive growth are beyond the scope of this chapter, but the development was no doubt accelerated in the United States by the coordinated 11 September 2001 attacks and in Europe by highprofile attacks like the 2004 Madrid train bombings, the 2005 London bombings, the series of attacks carried out in 2015 in Paris, and the 2016 Berlin Christmas market attacks. There are also other important signs that the study of radicalisation, political extremism, and terrorism is becoming part of the essential fabric of mainstream criminology. Research on political extremism now appears in nearly all criminology journals; college courses on political extremism and terrorism have been added to the curricula of many criminology and criminal justice departments; and a growing number of criminology students are choosing political extremism as a suitable topic for class papers, research topics, theses, and dissertations. The American Society of Criminology (ASC) has added a Division on Terrorism and Bias Crime that now has as many members as some long-standing specialisations, such as organised crime and juvenile delinquency. ASC, the European Society of Criminology, and the Academy of Criminal Justice Studies routinely feature dozens of papers and panels on radicalisation, political extremism, and terrorism each year at their annual meetings. While the scientific rigour of this rapidly expanding research literature has been uneven (LaFree & Ackerman, 2009; Silke, 2001, 2009), it is clear that research on political extremism and terrorism represents one of the major growth areas in social science scholarship over the past two decades, and furthermore criminology has figured prominently in these developments.



113

DOI: 10.4324/9781003035848-10

Gary LaFree and Yesenia Yanez

Despite the growing literature, criminology research on political extremism and terrorism is frequently criticised for being under-theorised (Freilich & LaFree, 2015). For example, Borum (2004, p. 4) points out that even though massive resources throughout the government and private sectors have been allocated . . . to the task of preventing terrorism. These efforts . . . often lack a conceptual . . . foundation for understanding terrorists and their acts of violence. The purpose of this chapter is to examine how criminological theories have contributed to the aetiology of radicalisation and political extremism during the past two decades. How often have criminological perspectives been applied to the study of political extremism in leading criminology journals? What does this criminology literature tell us about the structures and processes that lead individuals to radicalise2 to the point of engaging in violence to achieve their political ends? Are criminology theories playing a larger role in explaining radicalisation and political extremism now than they were in 2000? What are some of the most important insights we can gain by examining criminological perspectives on radicalisation and political extremism?

Criminological perspectives on extremism To better understand how criminological perspectives have contributed to the study of radicalisation and extremism, we identified a set of 12 refereed journals that publish high-quality empirically based research.3 It is important to note at the outset that a substantial number of terrorism-related papers fall outside the scope of these journals. For example, we exclude articles by criminologists in non-criminology journals, including those that specialise in terrorism research. We thus use these parameters not to represent the contributions of criminology to the entirety of the terrorism literature but rather to highlight how radicalisation and extremism are analysed and understood within many of the most influential outlets of criminological research. In these 12 journals, we looked for all articles from 2000 to 2020 that contained the terms ‘radicalization’, ‘extremism’, or ‘terrorism’. We excluded articles that focused only on hate crimes, non-ideologically motivated mass shootings, or genocide. The journals we reviewed are shown in Table 8.1. We identified a total of 103 articles in these criminology journals that dealt with violent political extremism for the 21 years included. Table 8.1 shows a wide variation among the journals in terms of how many extremism-related articles they published over the time period covered. All 12 journals had at least one relevant article. The Journal of Criminology and Public Policy had the largest number of applicable articles – over a quarter of the total. The Journal of Quantitative Criminology and the European Journal of Criminology were the next most frequent in terms of extremism-related articles. Other journals that published fewer than five articles on extremism and terrorism over the 21-year period included several of the most prestigious journals in criminology (e.g. Criminology, Journal of Research in Crime and Delinquency). The journals that had the fewest articles (one each) were the Journal of Crime and Justice and Law and Human Behavior. It is interesting to note that the journal that published the largest number of articles on extremism (Criminology and Public Policy) is known for not requiring extensive theoretical justifications for publication. We next reviewed the 103 articles and identified their primary theoretical focus (full references in Appendix A). Note that four articles used multiple theoretical perspectives, increasing the total number of theoretical applications to 107. In Table 8.2, we show the proportion of articles under major theoretical headings and also indicate those that we classified as atheoretical. We also 114

Criminological perspectives on extremist radicalisation Table 8.1  Articles Used for Analysis Journal

Total Articles Used

Criminology Journal of Quantitative Criminology Journal of Research in Crime and Delinquency Justice Quarterly Annual Review of Criminology British Journal of Criminology Journal of Criminal Law and Criminology Journal of Crime and Justice Criminology and Public Policy European Journal of Criminology Journal of Experimental Criminology Law and Human Behavior Total

4 17 3

4% 17% 3%

5 2 12 11 1 28 17 2 1 103

5% 2% 12% 11% 1% 27% 17% 2% 1% 100%

Source: Created by the authors Table 8.2  Criminological Theories Used for Research on Terrorism in 12 Journals, 2000–20 Theoretical Perspective

N

%N

N since 2010

% since 2010

Situational/routine activities Rational choice/deterrence Anomie/strain Criminal subcultures Life course Differential association/learning Social control Social disorganisation Collective action theory Conflict/radical Psychological Situational action Social construction Symbolic interaction Atheoretical analysis and reviews Total

20 10 4 4 4 3 2 2 1 1 1 1 1 1 52 107

19% 9% 4% 4% 4% 3% 2% 2% 1% 1% 1% 1% 1% 1% 49% 100%

19 8 4 2 4 3 2 2 1 1 1 1 1 1 35 85

95% 80% 100% 50% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 67% 79%

Source: Created by the authors

distinguish the proportion of perspectives that have been published in articles since 2010. One of the striking features of Table 8.2 is that nearly half of the total articles (49%) were atheoretical. Atheoretical contributions included articles on anti-radicalisation and counter-terrorism policies, empirical studies of specific aspects of terrorism, literature reviews, and studies of the criminal processing of terrorist perpetrators. 115

Gary LaFree and Yesenia Yanez

According to Table 8.2, of the remaining articles, 55 (51%) applied at least one major theory. The most common theoretical perspective was situational/routine activity theory, accounting for nearly one-fifth of the total (or one-third of the theoretical articles). The next most common were rational choice/deterrence perspectives, accounting for one-fifth of the theoretical articles. Anomie, criminal subcultures, and life course theories were next with four articles each. Differential association followed with three articles. Social control and social disorganisation were the major theoretical justifications for two articles, and six theories (collective action, conflict, psychological, situational action, social construction, and symbolic interaction) appeared in only one article. To examine how the use of criminological perspectives in the study of extremism changed over time within this sample, we split the sample in half, comparing the full sample to just those studies that were published after 2010. The results are fairly striking. According to Table 8.2, 85 of the 107 theory references (79.4%) were published after 2010. In other words, there is strong evidence that research on radicalisation, extremism, and terrorism in major criminology journals is becoming far more common over time. Moreover, of the 14 different theories identified in the articles, 11 have appeared exclusively since 2010. Strong majorities of two of the other theories have also appeared in articles since 2010: 95% of situational/routine activities and 80% of rational choice applications. Finally, the fact that only 67% of atheoretical analyses and reviews were found in articles since 2010 suggests that the proportion of extremism-related articles based on specific criminology theories is increasing over time. In the sections that follow, we consider some of the specific contributions made by the articles under each of the theoretical perspectives.

Situational criminology and the routine activities theory While some criminological theories focus on individual-level variables that contribute to radicalisation, a situational approach shifts attention away from the offender and onto the crime itself (Clarke, 1980). Situational or routine activity approaches focus on the contextual variables that make mobilisation for political extremism more likely. According to Table 8.1, the situational perspective was the most frequently employed in the studies reviewed here. Among the articles that relied on situational or routine activity perspectives, the most common themes dealt with crime concentration or hot spots (e.g. Hasisi et al., 2019),4 soft and hard target characteristics (e.g. Sturup et al., 2020), and situational crime prevention (e.g. Perry et al., 2017). An early application of situational crime theory by Sherman et al. (1989) established that a large proportion of the total number of crimes in a given community are concentrated in a small number of micro geographic places. Perry (2020) confirms this pattern among terrorist attacks in the city of Jerusalem. Specifically, he finds that one-quarter of all attacks occurred in just four hot spots in Jerusalem, or 1.6% of all attack locations. This finding has important policy implications as it suggests that terrorism, like more ordinary crimes, does not occur randomly but rather is highly concentrated in relatively small areas. This insight raises the possibility that societies can deploy police and other preventive measures to reduce the number of successful terrorist attacks. Behlendorf et al. (2012) incorporate this reasoning into their concept of ‘microcycles of violence’, or localised bursts of activity, and show how this thinking can help policy makers understand where and when future attacks are likely to occur. Closely linked to the situational approach, Cohen and Felson’s (1979) routine activity theory suggests that crime occurs when suitable targets, motivated offenders, and the absence of capable guardians converge in time and space. Several of the terrorism studies articles reviewed have focused on soft and hard target characteristics along with security-related variables. For example, Fahey 116

Criminological perspectives on extremist radicalisation

et al. (2012) examined whether terrorist hijackings situationally differed from non-ideologically motivated aerial hijackings. By examining contextual variables, including the number of hijackers, day of the week, and departure country, the authors found that measures of organisational resources (e.g. the number of hijackings and weapon type) differed significantly between terrorist and non-terrorist perpetrators. By focusing on characteristics like these, the researchers were able to offer situation-specific policy guidance. Clarke (1980) proposed situational crime prevention (SCP) as a method for reducing crime through managerial and environmental changes. SCP rests on the assumption that offences, whether they are carefully planned or not, are heavily influenced by situational opportunities (Clarke, 1997, 2016). Thus, the goal of SCP is to raise the costs and reduce the benefits of crimes to make them less attractive to individuals. Clarke and Homel (1997) offer 16 opportunity-reducing techniques to help reduce the risk of crime (e.g. target hardening, natural surveillance, reducing temptation, and facilitating compliance). These strategies can also be applied to the prevention of terrorism. For example, based on a situational perspective, Perry et al. (2017) concluded that the ‘west bank barrier’ in Jerusalem was effective in preventing suicide bombings.

Rational choice and deterrence The second most common theoretical perspective in our review of criminology journals is rational choice/deterrence. Clarke and Cornish’s (1985) rational choice theory borrows from economics, psychology, and sociology to model offenders’ decision-making processes. In criminology, this perspective is used primarily to predict an individual’s initial involvement in crime. Because the rational choice perspective assumes that all individuals are decision makers who take incentives and risks into account, deterrence is often the focus of rational choice studies. Among the articles we reviewed, rational choice was most commonly used to examine the deterrent effects of antiradicalisation programmes and counterterrorist strategies. The rational choice perspective is particularly interesting in the study of political extremism, as terrorist perpetrators may perceive the benefits of their crimes to go beyond their personal gains in support of more general political aims. Radicalised individuals may thus be more likely to engage in political extremism if they perceive the rewards of their actions as outweighing the risks or costs of a particular social movement. In response to these expected calculations, governments may attempt to deter potential terrorists by increasing the risk of apprehension. The aerial hijackings of 11 September 2001 undoubtedly encouraged US efforts to enforce protective measures at airports. On the basis of the rational choice approach, Dugan et al. (2005) modelled aerial hijackings and found that some measures of target hardening, like metal detectors and law enforcement at passenger checkpoints significantly reduced the rate of new hijacking attempts (although the effects were limited to non-terrorist perpetrators). Likewise, Carson et al. (2020) apply rational choice to study the impact of government actions on radical eco-movement attacks and find that when government policies increase the costs of committing this type of extremist behaviour, incidents decline. While the deterrence perspective suggests that criminal behaviour can be modified by increasing the threat or imposition of punishment, several of the studies reviewed here have noted a backlash effect where government actions may unintentionally increase the likelihood of future attacks. For example, LaFree et al. (2009) compare deterrence and backlash outcomes by examining rates of terrorist attacks in Northern Ireland between 1969 and 1992 as the British imposed various types of counter terrorism interventions. In support of backlash interpretations, the authors discovered that three of the six counterterrorist interventions used by the British during this period 117

Gary LaFree and Yesenia Yanez

(internment, criminalisation/Ulsterisation, and the Gibraltar incident) significantly increased the hazard of future terrorist strikes, two interventions (the Falls curfew and the Loughgall incident) had no significant effect in either direction, and only one (Operation Motorman) found a shortlived deterrence effect.5 Similarly, Hsu and McDowall (2020) found that repressive counterterrorism actions increased violence in Israel and that backlash effects were dependent on the magnitude of government repression, the target, and the lethality of terrorist attacks.

Anomie/strain Criminologists studying radicalisation and extremism have also applied deprivation theories such as anomie and strain. In his original formulation of anomie theory, Merton (1938) proposed that crime is a result of limited or blocked access to normative goals. Thus, crime rates may be higher in poor communities because they share many of the aspirations of rich communities but have fewer legitimate means of fulfilling these aspirations. Similarly, in his version of strain theory, Agnew (2007) argues that failure to achieve important goals can lead individuals to experience strain, which in turn encourages criminal behaviour to reduce the strain. In general, researchers applying strain theories to terrorism claim that strain increases the likelihood of terrorism by leading to negative emotional states such as frustration or humiliation and reducing the effectiveness of social control through legal channels. However, support for this theory with regard to terrorist perpetrators is weak, with one article in our review finding that collective strain does not have a direct effect on violent extremist attitudes (Nivette et al., 2017); one finding an opposite effect (i.e. Freilich et al. [2015] found that economically deprived counties were less likely to have far-right perpetrators); and another finding effects of collective deprivation only for far-left terrorism (Varaine, 2019). In contrast to more ordinary forms of crime, it does not appear that terrorism and political violence are consistently associated with poverty and economic deprivation.

Criminal subcultures Supporters of criminal subculture perspectives in criminology argue that there are social groups in which it is normal to behave in ways that are deemed criminal by members of mainstream society. For example, Wolfgang and Ferracuti (1967) hypothesise that a subculture of violence exists in many low-socio-economic communities in the United States, and in these communities, violence is accepted as a normal part of life. While this type of subcultural application is uncommon in criminology research on radicalisation, Cottee’s (2020) description of a western jihadi subculture provides an example. He argues that this jihadi subculture is composed of three distinct features that heighten the likelihood of extremist violence: (1) violence and machismo, (2) death and martyrdom, and (3) disdain for the temporal world and its earthly concerns and possessions (dunya). The other three articles that apply a criminal subculture approach all focus on subcultures within prison and how they are connected to radicalisation. From a subcultural perspective, research on radicalisation within prison suggests that individuals assimilate to the prison subculture by adopting violent or antisocial attitudes, leading them to engage in more extremist beliefs and actions. While some researchers conclude that there are low or modest levels of radicalisation in prison (Useem & Clayton, 2009), others find that spending time in prison and being exposed to radicalisation in prison increases the likelihood of violent political extremism after release (Hamm, 2009; LaFree et al., 2020).

118

Criminological perspectives on extremist radicalisation

Life-course perspectives Unlike perspectives that explain crime patterns between individuals or groups, the life-course perspective focuses on how criminality develops within an individual’s lifetime. Studies that apply life-course perspectives (Sampson & Laub, 2009) overlap to some extent with other theoretical perspectives, especially social control and deterrence theories. In our sample, four studies investigated the process of radicalisation throughout the life course. Simi et al. (2016) used in-depth life-history interviews with former violent extremists to determine how childhood- and adolescent-related variables influence later patterns of radicalisation and violent political extremism. They argue for a three-stage risk-factor model such that the individual (1) experiences different types of adversity in childhood, (2) subsequently has conduct problems in adolescence, and (3) is motivated to seek out circumstances that lead to participation in political extremist groups. While Simi et al. focus on psychological processes leading to radicalisation, Carlsson et al. (2020) instead emphasise the social processes leading to terrorism and political extremism. For example, the authors argue that the process of radicalisation occurs in three stages: (1) a weakening of informal social controls; (2) association with individuals close to extremist groups; and (3) a process of ‘meaning-making’ in relation to the group and one’s identity. Life-course perspectives not only are used to explain the process of getting into crime but often help explain recidivism and desistance. Hasisi et al. (2020) study the offending patterns of imprisoned terrorists and find that terrorist recidivism rates match or exceed recidivism rates for other types of criminal offences. They illustrate similarities between terrorism-related crimes and other crimes by showing that many of the risk factors for recidivism are the same (e.g. age and prior criminal record). While the life-course perspective typically follows individuals, one of the studies in our review focuses on the emergence, offending, and decline of terrorist organisations. Miller (2012) examines the terrorist activities of 557 organisations and finds that terrorist organisations that arise more rapidly and attack more frequently also survive longer. Given that applications of life course perspectives are especially demanding in that they require longitudinal data, it is interesting that we have found four examples since 2010.

Differential association/learning Originally formulated by Sutherland (1939, 1947), learning perspectives begin with the premise that criminal behaviour, like other behaviour, is learned. Specifically, individuals interact in small, intimate groups to learn and adopt values that favour breaking the law. By this logic, social learning sometimes overlaps with criminal subcultures. LaFree et al. (2020) argue that imprisonment and prison radicalisation increase the probability of post-prison violent extremism because inmates, through associations with peers in prison, learn and adopt radical values. This argument is also supported outside prison, as individuals with radical peers are more likely to engage in violent political acts (LaFree et al., 2018). Further, some scholars argue that learning does not necessarily come from associations with deviant peers but can be facilitated by previous high-profile examples. In other words, crime can be contagious through more distant learning. In our sample, Miller and Hayward (2019) apply learning theory to understand the contagion-like nature of highly publicised vehicle-ramming terrorist attacks.

119

Gary LaFree and Yesenia Yanez

Social control Proponents of the social control perspective argue that crime occurs when an individual’s bonds with prosocial communities or institutions are weakened (Hirschi, 1969). Common prosocial bonds include employment, education, marriage, and military experience. LaFree et al. (2018) apply this perspective to a sample of American political extremists and find that one variable associated with social control theory (lack of stable employment) is a significant predictor of violent political extremism. By contrast, marital status, educational attainment, and military experience had no effect. In a model that connects social control and strain theories to political aggression, Pauwels et al. (2020) find that social integration, or an accumulation of social bonds, increases perceived respect for others and commitment to procedural justice.

Social disorganisation In their classic formulation of social disorganisation theory, Shaw and McKay (1942) argue that when communities are unable to facilitate strong social bonds, individuals in those communities will be more likely to engage in a wide variety of antisocial behaviours, including crime. This perspective suggests that crime and deviant behaviour, more generally, will be higher in areas that lack shared norms and have high population heterogeneity, residential mobility, and concentrated disadvantage (Bursik, 1988). To apply this theory to political extremism, LaFree and Bersani (2014) investigate terrorist attacks in the United States and find that counties with high population heterogeneity and residential instability experience high rates of terrorist attacks. The authors also find that counties with high concentrations of poverty are associated with fewer attacks. In a more general attempt to explore the applicability of various macro-level theories for understanding terrorist attacks, Freilich et al. (2015) find little support for measures commonly associated with social disorganisation theory.

Less common perspectives In our sample of articles, six looked at theoretical perspectives that appeared only once in our review. Four of these articles (Ruggiero, 2010; Ilan & Sandberg, 2019; Mythen & Walklate, 2006; Arrigo, 2010) take a critical look at mainstream criminology theories and argue that it would be useful for researchers to look beyond traditional criminology theories for understanding radicalisation and political violence. Three of these articles appeared in the British Journal of Criminology, and the other appeared in the European Journal of Criminology. All four articles recommend some version of a conflict/constructionist perspective as an alternative to a more traditional criminological framework. We classify the fifth and sixth articles with only a single theoretical mention as situational action and psychological perspectives. Clemmow et al. (2020) draw on situational action theory to devise a typology of person-exposure patterns among lone terrorists. They use cluster analysis to identify relationships across three components: propensity, situation, and network. Propensity measures include low self-control and difficulties with anger management; situation measures include whether others are aware of the perpetrators’ grievances and their ideology and whether they have been a target of prejudice; and network measures include face-to-face interactions with members of a wider network and claiming to be a part of a broader movement. The authors argue that lone-actor terrorism is extremely heterogeneous, and developing a more nuanced understanding of the forms it can take is vital to countering threats. Corner and Gill (2020) analyse 90 case studies to examine the onset of psychological distress across 120

Criminological perspectives on extremist radicalisation

three phases of terrorist involvement (engagement, disengagement, and post-disengagement). The authors demonstrate that the relationship between mental disorders and terrorist engagement is extremely complex and that psychological distress is mediated by several variables and combinations of variables.

Atheoretical analyses and reviews We classified a total of 52 articles as atheoretical (see Appendix A). A quick look at these articles shows that they represent a diverse number of themes and perspectives. We find it useful to group the atheoretical articles into four broad categories: (1) policy arguments; (2) literature reviews; (3) empirical examinations of specific aspects of extremism or terrorism; and (4) criminal justice processing of extremist/terrorist cases. We classified 20 (38.5%) of the articles as dealing mostly with different aspects of terrorism policy. For example, in this category are Banks’ (2009) article on ‘smart’ counterterrorism, Forst’s (2017) article on targeted assassinations, and Anderson’s (2004) article on citizen-detainees’ right to due process. Empirical examinations of specific aspects of extremism or terrorism were next most common, accounting for 14 (26.9%) of the articles. Atheoretical empirical articles are generally descriptive studies that report on specific characteristics of terrorism without providing an overarching theoretical perspective to interpret them. The characteristics reported on are diverse, including detailed analyses of lone wolf attacks, right-wing extremists, Islamist extremists, targeted killings, Somali freedom fighters, the impact of the Internet on terrorism, terrorist target selection, how members of criminal gangs compare to political extremists, and behaviours terrorists engage in before their attacks. Examples of this diverse group include Pyrooz et al.’s (2017) analysis of gang members and political extremists; Gruenewald et al.’s (2013) comparison of lone wolves and other extremists; and Morris’s (2015) study of terrorist target selection. We classified 13 (25%) of the atheoretical articles as literature reviews. Examples of literature reviews include Lum et al.’s (2006) systematic review of counter terrorism policies; Wolfowicz et al.’s (2020) systematic review and meta-analysis of radicalisation risk factors; and Ahmad and Monaghan’s (2019) review of the criminology literature on radicalisation. The final category and the least common (five studies, or 9.6%) are studies that examine how extremist/terrorist cases are processed by the criminal justice system. Examples include Amirault et al.’s (2016) study of sentencing outcomes for terrorist perpetrators in Canada, Norris and GrolProkopczyk’s (2015) study of entrapment in terrorism cases, and Rushchenko’s (2019) assessment of ‘separation centres’ for terrorist perpetrators in the United Kingdom.

Discussion and conclusions Our review of 12 major criminology journals over the past 21 years yielded 103 terrorism/extremismrelated articles, suggesting that research on radicalisation, political extremism, and terrorism now represents a major sub-field within criminology. Moreover, the pace of change seems to be accelerating, with nearly four-fifths of the 103 articles we found appearing after 2010. Our review also provides support for the argument that criminology research on terrorism has been under-theorised, as about half of the articles we identified were atheoretical. For those articles that were grounded in a specific theory, the most common perspectives were situational/routine activities and rational choice/deterrence, followed at some distance by criminal subcultures, life course approaches, anomie, and differential association. Many influential criminology theories were uncommon (e.g. social control and symbolic interaction) or altogether absent (e.g. labelling 121

Gary LaFree and Yesenia Yanez

and biology). Among the atheoretical articles, the most common topics (in order) were policy arguments, empirical examinations of specific aspects of radicalisation, extremism, or terrorism (e.g. lone wolf attacks and female perpetrators), literature reviews, and articles about the criminal justice processing of extremist/terrorist cases. While the 12 journals we chose to review are among the most important in the field of criminology, we concede that our sample of journals is arbitrary. In particular, our list leaves off articles by criminologists in general social science journals (e.g. American Sociological Review and Social Forces) or journals that specialise in the study of terrorism (e.g. Terrorism and Political Violence, Studies in Conflict and Terrorism). We also exclude more specialised criminology journals (e.g. Homicide Studies and the Journal of Interpersonal Violence) and most journals connected to specific countries (e.g. the Australian and New Zealand Journal of Criminology and the Canadian Journal of Criminology and Criminal Justice). In addition, our focus is only on articles published in refereed journals. Our review excludes contributions from monographs and book chapters. We noted earlier that only about half of the articles reviewed included a specific theoretical framework. However, we also pointed out that the proportion of articles that were theoretical increased a good deal after 2010. This pattern is consistent with the idea that criminological research on terrorism and extremism is becoming more theoretical over time. Researchers generally agree that among the purposes of science, description is the most basic. For example, fields like biology began by simply describing flora and fauna. As a field matures, its practitioners are likely to develop increasingly complex scientific methods that move beyond description to explanation, prediction, and control. Many of the atheoretical criminology articles reviewed here were efforts to describe specific aspects of extremism, terrorism, or the radicalisation process. More recent articles, which more often incorporate theoretical perspectives, generally move beyond simple description. The mix of theories represented in the articles reviewed is also of interest. The most common theoretical perspectives we identified were situational/routine activities and rational choice/ deterrence perspectives. This may be attributed in part to our journal selection or the fact that our analysis includes articles that focus on both the process of radicalisation and terrorist acts.6 By contrast, we find relatively few articles applying criminological staples like social control, strain, and learning theories. These patterns may tell us something about the current state of data on terrorism and political extremism, in particular the strong reliance on open-source data. Criminologists traditionally rely on three types of data for research: ‘official’ data collected by legal agents (most often the police); ‘victimization’ data collected from the general population of victims and nonvictims; and ‘self-report’ data collected from offenders. None of these sources is very useful for the study of terrorism and political extremism. Victimisation surveys are of little use because, even with extremely large samples, few individuals in most countries have been directly victimised by terrorists. Moreover, because terrorist victimisation is often random – victims simply happen to be in the wrong place at the wrong time – victims are unlikely to encounter perpetrators, much less know them. And unfortunately, in many cases, victims of terrorism are killed by their attackers, making it impossible for them to relate their experiences. Self-report data, where researchers collect information on terrorist acts from those who committed those acts, have been more fruitful than victim data. For example, the article in our review by Simi et al. (2016) is based on self-reported narratives from terrorist perpetrators. However, self-reported survey data in the study of terrorism are relatively rare. Most active terrorists are unwilling to participate in interviews, and even when they are willing, getting access to known terrorists for research purposes raises obvious logistical challenges. 122

Criminological perspectives on extremist radicalisation

At present, most countries in the world do not make public official police data on terrorist perpetrators. In fact, the majority of offenders suspected of terrorism in the United States are not legally processed for terrorism-specific charges but rather for other related offences such as weapons violations and money laundering. Moreover, while a good deal of information on individual terrorist perpetrators is collected by clandestine organisations, these data are usually unavailable to researchers working in an unclassified environment. As a result of these limitations, most of the research in the criminology journals reviewed here has been based on open-source data. Open-source databases on radicalisation, terrorism, and political extremism generally rely on court cases and news reports from the print and electronic media to collect detailed information on the characteristics of terrorist attacks. In general, open-source data on terrorism can be classified by whether it focuses on terrorist attacks, terrorist groups, or terrorist perpetrators (for a review, see LaFree, forthcoming, 2022). The fact that nearly all of the theoretical studies we report in this chapter are based on open-source terrorism/extremism data has important implications for the theories being tested and the timing of those tests. In general, open-source terrorism data likely lend themself to situational and deterrence perspectives. Given that much of the open-source data on terrorism examine the characteristics of terrorist incidents, these data usually include quite a bit of information about the situations under which attacks occur (e.g. location, degree of urbanisation, and time of day). This makes situational perspectives an obvious theoretical choice. Likewise, because much of the open-source data being examined focus on terrorist attacks, it is relatively straightforward to test rational choice/deterrence models by looking at what happens to the rate of terrorist attacks before and after some specific intervention. By contrast, open-source event data are often less useful for understanding individual-level decision-making. Many of the mainstream criminology theories that are underrepresented in our survey of terrorism-related articles (e.g. social control, differential association, and situational action theory) require detailed information on the characteristics of perpetrators. These considerations may also help explain the timing of theoretical contributions to the literature on political extremism and terrorism. The only three theories that were present before 2010 in the articles reviewed were situational, rational choice, and criminal subcultures. Because some open-source databases have been available for decades (Mickolus, 1976; LaFree et al., 2014), situational and rational choice perspectives may have been easier to apply before 2010. Further, the two studies based on criminal subcultures published before 2010 (Hamm, 2009; Useem & Clayton, 2009) were based on separate data collection projects on political extremists in prison. The open-source data situation has changed considerably since 2010, as new individual-level databases on political extremism have started to appear. This makes it increasingly possible to test criminological theories at the individual level, including examples of life course perspectives (Hasisi et al., 2020), strain (Varaine, 2019), differential association (LaFree et al., 2020), and social control (LaFree et al., 2018). As more individual-level databases become available in the future, we are likely to see a growing number of articles based on criminological theories that have, to this point in time, been relatively uncommon.7 Criminological interest in radicalisation and terrorism has grown dramatically since the turn of the 21st century. While there are indications that this specialisation within criminology has been under-theorised, we also find evidence that applications of criminological theories to understand terrorism are becoming more common over time. A major challenge in terms of applying criminological theories to radicalisation and political extremism is that the nature of terrorism makes it difficult to track individual perpetrators through traditional criminological data sources such as victimisation or self-report surveys, or police files. In response to this challenge, terrorism researchers have turned to open-source data, which most often count terrorist attacks rather than 123

Gary LaFree and Yesenia Yanez

individual pathways to terrorism. As a result, popular criminology theories that focus on predicting individual offender behaviour have been less useful. Scholars will likely find it easier to do theoretically driven research in the future as more individual-level data become available.

Notes 1 We adopt the FBI’s (2017) definition of violent extremism as ‘encouraging, condoning, justifying, or supporting the commission of a violent act to achieve political, ideological, religious, social, or economic goals.’ 2 We adopt the FBI’s definition of radicalisation as ‘the process by which individuals come to believe their engagement in or facilitation of non-state violence to achieve social and political change is necessary and justified’ (Hunter & Heinke, 2011). 3 For purposes of this exercise, we adopted the 12 journals that the Department of Criminology and Criminal Justice at the University of Maryland currently ranks as the most important in the field. 4 We include the 103 references that we identified from our review of research articles in Appendix A. Other references are included in the reference section. 5 Briefly, internment was a policy that relaxed legal requirements for imprisoning suspects; criminalisation/ Ulsterisation was a strategy to treat perpetrators as ordinary criminals rather than political prisoners; the Gibraltar and Loughall incidents were targeted assassinations; and Operation Motorman was a massive military surge. 6 The emphasis on rational choice/deterrence perspectives supports Ahmad and Monaghan’s (2019) claim that criminological work on radicalisation has privileged individualistic theories. However, the emphasis on situational/routine activities is more complex in that the perspective emphasises both individual (motivated offenders) and situational/sociological elements (suitable targets, capable guardians). 7 The growing interest in individual-level data on terrorism itself likely represents a shift in research away from explaining crime events (i.e., terrorist attacks) towards understanding criminal development (i.e., radicalisation). As 9/11 recedes, there appears to be growing interest in understanding radicalisation. In general, rational choice and situational perspectives are likely more important for understanding why attacks happen, while theories like social control and situational action theory require individual-level data on the characteristics and behaviours of perpetrators.

References (see Appendix A for criminology theory references) Agnew, R. (2007). Pressured into crime: An overview of general strain theory. Oxford University Press. Borum, R. (2004). Psychology of terrorism. University of South Florida. Bursik, R. J. (1988). Social disorganization and theories of crime and delinquency: Problems and prospects. Criminology, 26(4), 519–552. Clarke, R. V. (1980). Situational crime prevention: Theory and practice. British Journal of Criminology, 20(2), 136–147. Clarke, R. V. (Ed.). (1997). Situational crime prevention: Successful case studies (2nd ed.). Harrow and Heston. Clarke, R. V. (2016). Situational crime prevention. In R. Wortley & M. Townsley (Eds.), Environmental criminology and crime analysis (2nd ed., pp. 286–303). Routledge. Clarke, R. V., & Cornish, D. B. (1985). Modeling offenders’ decisions: A framework for research and policy. Crime and Justice, 6, 147–185. Clarke, R. V., & Homel, R. (1997). A revised classification of situational crime prevention techniques. In S. P. Lab (Ed.), Crime prevention at a crossroads. Academy of Criminal Justice Sciences. Cohen, L. E., & Felson, M. (1979). Social change and crime rate trends: A routine activity approach. American Sociological Review, 44(4), 588–608. Federal Bureau of Investigation. (2017). What is violent extremism? Department of Justice. Fisher, D., & Becker, M. H. (2021). The heterogeneous repercussions of killing Osama bin Laden on global terrorism patterns. European journal of criminology, 18(3), 301–324. Forst, B., Greene, J. R., & Lynch, J. P. (2011). Criminologists on terrorism and homeland security. Cambridge University Press.

124

Criminological perspectives on extremist radicalisation Freilich, J., & LaFree, G. (2015). Criminology theory and terrorism. Terrorism and Political Violence, 27(1), 1–8. Hamm, M. S. (1993). American skinheads: The criminology and control of hate crime. Santa ABC-CLIO. Hirschi, T. (1969). Causes of delinquency. University of California Press. Hunter, R., & Heinke, D. (2011). Radicalization of Islamic terrorists in the western world. FBI Law Enforcement Bulletin, 80. Jensen, S. Q., & Larsen, J. F. (2021). Sociological perspectives on Islamist radicalization–bridging the micro/ macro gap. European journal of criminology, 18(3), 426–443. Kittrie, N. N. (1978). A new look at political offenses and terrorism. In M. H. Livingston, L. B. Kress, & M. G. Wanek (Eds.), International terrorism in the contemporary world (pp. 354–375). Greenwood. Kupatadze, A., & Argomaniz, J. (2019). Introduction to Special Issue–Understanding and conceptualizing European jihadists: Criminals, extremists or both?. European Journal of Criminology, 16(3), 261–277. LaFree, G. (forthcoming). Terrorism open-source databases. In D. Muro & T. Wilson (Eds.), Contemporary terrorism studies. Oxford University Press. LaFree, G. (2022). Terrorism open source databases. Contemporary Terrorism Studies, 113–135. LaFree, G., & Ackerman, G. (2009). The empirical study of terrorism: Social and legal research. Annual Review of Law and Social Science, 5, 347–374. LaFree, G., Dugan, L., & Miller, E. (2014). Putting terrorism in context: Lessons from the global terrorism database. Routledge. Lum, C., & Kennedy, L. W. (Eds.). (2012). Evidence-based counterterrorism policy. Springer. Merton, R. K. (1938). Science and the social order. Philosophy of Science, 5(3), 321–337. Mickolus, E. F. (1976). International terrorism: Attributes of terrorist events: ITERATE. Inter-university Consortium for Political and Social Research. Pyrooz, D. C., LaFree, G., Decker, S. H., & James, P. A. (2018). Cut from the same cloth? A comparative study of domestic extremists and gang members in the United States. Justice Quarterly, 35(1), 1–32. Sageman, M. (2014). The stagnation in terrorism research. Terrorism and Political Violence, 26(4), 565–580. Sampson, R. J., & Laub, J. (2009). Shared beginnings, divergent lives. Harvard University Press. Shaw, C. R., & McKay, H. D. (1942). Juvenile delinquency and urban areas. University of Chicago Press. Sherman, L. W., Gartin, P. R., & Buerger, M. E. (1989). Hot spots of predatory crime: Routine activities and the criminology of place. Criminology, 27(1), 27–56. Silke, A. (2001). The devil you know: Continuing problems with research on terrorism. Terrorism and Political Violence, 13(4), 1–14. Silke, A. (2009). Contemporary terrorism studies: Issues in research. In R. Jackson, M. B. Smyth, & J. Gunning (Eds.), Critical terrorism studies: A new research agenda. Routledge. Smith, B. L. (1994). Terrorism in America: Pipe bombs and pipe dreams. SUNY Press. Sutherland, E. H. (1939). Principles of criminology. Lippincott. Sutherland, E. H. (1947). Principles of criminology. Lippincott. Turk, A. (1982). Political criminality: The defiance and defense of authority. Sage. Wolfgang, M., & Ferracuti, F. (1967). The subculture of violence. Tavistock. Zedner, L., & Ashworth, A. (2019). The rise and restraint of the preventive state. Annual Review of Criminology, 2, 429–450.

125

APPENDIX A. REFERENCES BY CRIMINOLOGICAL THEORY

Situational/routine activities   1. Argomaniz, J., & Bermejo, R. (2019). Jihadism and crime in Spain: A convergence settings approach. European Journal of Criminology, 16(3), 351–368.  2. Behlendorf, B., LaFree, G., & Legault, R. (2012). Microcycles of violence: Evidence from terrorist attacks by ETA and the FMLN. Journal of Quantitative Criminology, 28(1), 49–75.   3. Braithwaite, A., & Johnson, S. D. (2012). Space – time modeling of insurgency and counterinsurgency in Iraq. Journal of Quantitative Criminology, 28(1), 31–48.   4. Fahey, S., LaFree, G., Dugan, L., & Piquero, A. R. (2012). A situational model for distinguishing terrorist and non-terrorist aerial hijackings, 1948–2007. Justice Quarterly, 29(4), 573–595.   5. Freilich, J. D., Chermak, S. M., & Klein, B. R. (2020). Investigating the applicability of situational crime prevention to the public mass violence context. Criminology & Public Policy, 19(1), 271–293.   6. Hasisi, B., Perry, S., Ilan, Y., & Wolfowicz, M. (2019). Concentrated and close to home: The spatial clustering and distance decay of lone terrorist vehicular attacks. Journal of Quantitative Criminology, 1–39.   7. Hsu, H. Y., & McDowall, D. (2017). Does target-hardening result in deadlier terrorist attacks against protected targets? An examination of unintended harmful consequences. Journal of Research in Crime and Delinquency, 54(6), 930–957.   8. Hsu, H. Y., Vásquez, B. E., & McDowall, D. (2018). A time-series analysis of terrorism: Intervention, displacement, and diffusion of benefits. Justice Quarterly, 35(4), 557–583.   9. Hsu, H. Y., Vásquez, B. E., & McDowall, D. (2020). A deadlier post-9/11 terrorism landscape for the USA abroad: A quasi-experimental study of backlash effects of terrorism prevention. Journal of Experimental Criminology, 16(4), 607–623. 10. LaFree, G., Dugan, L., Xie, M., & Singh, P. (2012). Spatial and temporal patterns of terrorist attacks by ETA 1970 to 2007. Journal of Quantitative Criminology, 28(1), 7–29. 11. LaFree, G., Morris, N. A., & Dugan, L. (2010). Cross-national patterns of terrorism: Comparing trajectories for total, attributed and fatal attacks, 1970–2006. The British Journal of Criminology, 50(4), 622–649. 12. Legault, R. L., & Hendrickson, J. C. (2009). Weapon choice and American political violence: A comparison of terrorists and other felons in federal custody. Criminology & Public Policy, 8(3), 531–559. 13. Marchment, Z., Gill, P., & Morrison, J. (2020). Risk factors for violent dissident republican incidents in Belfast: A comparison of bombings and bomb hoaxes. Journal of Quantitative Criminology, 36(3), 647–666. 14. Perry, S. (2020). The application of the “law of crime concentration” to terrorism: The Jerusalem case study. Journal of Quantitative Criminology, 36(3), 583–605. 15. Perry, S., Apel, R., Newman, G. R., & Clarke, R. V. (2017). The situational prevention of terrorism: An evaluation of the Israeli West Bank barrier. Journal of Quantitative Criminology, 33(4), 727–751.

126

Criminological perspectives on extremist radicalisation 16. Perry, S., Hasisi, B., & Perry, G. (2019). Lone terrorists: A study of run-over attacks in Israel. European Journal of Criminology, 16(1), 102–123. 17. Rossmo, D. K., & Harries, K. (2011). The geospatial structure of terrorist cells. Justice Quarterly, 28(2), 221–248. 18. Sturup, J., Gerell, M., & Rostami, A. (2020). Explosive violence: A near-repeat study of hand grenade detonations and shootings in urban Sweden. European Journal of Criminology, 17(5), 661–677. 19. Wilson, M. A., Scholes, A., & Brocklehurst, E. (2010). A behavioural analysis of terrorist action: The assassination and bombing campaigns of ETA between 1980 and 2007. The British Journal of Criminology, 50(4), 690–707. 20. Yang, S. M., & Jen, I. C. (2018). An evaluation of displacement and diffusion effects on eco-terrorist activities after police interventions. Journal of Quantitative Criminology, 34(4), 1103–1123.

Rational choice/deterrence   1. Carson, J. V. (2014). Counterterrorism and radical eco-groups: A context for exploring the series hazard model. Journal of Quantitative Criminology, 30(3), 485–504.   2. Carson, J. V. (2017). Assessing the effectiveness of high‐profile targeted killings in the “war on terror” a quasi‐experiment. Criminology & Public Policy, 16(1), 191–220.   3. Carson, J. V., Dugan, L., & Yang, S. M. (2020). A comprehensive application of rational choice theory: How costs imposed by, and benefits derived from, the US federal government affect incidents perpetrated by the radical eco-movement. Journal of Quantitative Criminology, 36(3), 701–724.  4. Dugan, L., LaFree, G., & Piquero, A. R. (2005). Testing a rational choice model of airline hijackings. Criminology, 43(4), 1031–1065.   5. Fisher, D., & Becker, M. H. (2019). The heterogeneous repercussions of killing Osama bin Laden on global terrorism patterns. European Journal of Criminology, 18(3), 301–324.   6. Fisher, D., & Dugan, L. (2019). The importance of governments’ response to natural disasters to reduce terrorist risk. Justice Quarterly, 1–32.   7. Freilich, J. D., Adamczyk, A., Chermak, S. M., Boyd, K. A., & Parkin, W. S. (2015). Investigating the applicability of macro-level criminology theory to terrorism: A county-level analysis. Journal of Quantitative Criminology, 31(3), 383–411.   8. Hsu, H. Y.,  & McDowall, D. (2020). Examining the state repression‐terrorism nexus: Dynamic relationships among repressive counterterrorism actions, terrorist targets, and deadly terrorist violence in Israel. Criminology & Public Policy, 19(2), 483–514.   9. LaFree, G., Dugan, L., & Korte, R. (2009). The impact of British counterterrorist strategies on political violence in Northern Ireland: Comparing deterrence and backlash models. Criminology, 47(1), 17–45. 10. Mills, C. E., Schmuhl, M., & Capellan, J. A. (2020). Far-right violence as backlash against gender equality: A county-level analysis of structural and ideological gender inequality and homicides committed by far-right extremists. Journal of Crime and Justice, 43(5), 568–584.

Criminal subcultures 1. Cottee, S. (2020). The Western Jihadi subculture and subterranean values. The British Journal of Criminology, 60(3), 762–781. 2. Hamm, M. S. (2009). Prison Islam in the age of sacred terror. The British Journal of Criminology, 49(5), 667–685. 3. LaFree, G., Jiang, B., & Porter, L. C. (2020). Prison and violent political extremism in the United States. Journal of Quantitative Criminology, 36(3), 473–498. 4. Useem, B., & Clayton, O. (2009). Radicalization of US prisoners. Criminology & Public Policy, 8(3), 561–592.

Life course 1. Carlsson, C., Rostami, A., Mondani, H., Sturup, J., Sarnecki, J., & Edling, C. (2020). A life-course analysis of engagement in violent extremist groups. The British Journal of Criminology, 60(1), 74–92.

127

Gary LaFree and Yesenia Yanez 2. Hasisi, B., Carmel, T., Weisburd, D., & Wolfowicz, M. (2020). Crime and terror: Examining criminal risk factors for terrorist recidivism. Journal of Quantitative Criminology, 36(3), 449–472. 3. Miller, E. (2012). Patterns of onset and decline among terrorist organizations. Journal of Quantitative Criminology, 28(1), 77–101. 4. Simi, P., Sporer, K., & Bubolz, B. F. (2016). Narratives of childhood adversity and adolescent misconduct as precursors to violent extremism: A life-course criminological approach. Journal of Research in Crime and Delinquency, 53(4), 536–563.

Anomie/strain 1. Freilich, J. D., Adamczyk, A., Chermak, S. M., Boyd, K. A., & Parkin, W. S. (2015). Investigating the applicability of macro-level criminology theory to terrorism: A county-level analysis. Journal of Quantitative Criminology, 31(3), 383–411. 2. Nivette, A., Eisner, M., & Ribeaud, D. (2017). Developmental predictors of violent extremist attitudes: A test of general strain theory. Journal of Research in Crime and Delinquency, 54(6), 755–790. 3. Varaine, S. (2019). Revisiting the economics and terrorism nexus: Collective deprivation, ideology and domestic radicalization in the US (1948–2016). Journal of Quantitative Criminology. doi:10.1007/ s10940-019-09422-z 4. Pauwels, L. J., Ljujic, V., & De Buck, A. (2020). Individual differences in political aggression: The role of social integration, perceived grievances and low self-control. European Journal of Criminology, 17(5), 603–627.

Differential association/learning 1. LaFree, G., Jensen, M. A., James, P. A., & Safer‐Lichtenstein, A. (2018). Correlates of violent political extremism in the United States. Criminology, 56(2), 233–268. 2. LaFree, G., Jiang, B., & Porter, L. C. (2020). Prison and violent political extremism in the United States. Journal of Quantitative Criminology, 36(3), 473–498. 3. Miller, V., & Hayward, K. J. (2019). ‘I did my bit’: Terrorism, Tarde and the vehicle ramming attack as an imitative event. The British Journal of Criminology, 59(1), 1–23.

Social control 1. LaFree, G., Jensen, M. A., James, P. A., & Safer‐Lichtenstein, A. (2018). Correlates of violent political extremism in the United States. Criminology, 56(2), 233–268. 2. Pauwels, L. J., Ljujic, V., & De Buck, A. (2020). Individual differences in political aggression: The role of social integration, perceived grievances and low self-control. European Journal of Criminology, 17(5), 603–627.

Social disorganisation 1. Freilich, J. D., Adamczyk, A., Chermak, S. M., Boyd, K. A., & Parkin, W. S. (2015). Investigating the applicability of macro-level criminology theory to terrorism: A county-level analysis. Journal of Quantitative Criminology, 31(3), 383–411. 2. LaFree, G., & Bersani, B. E. (2014). County‐level correlates of terrorist attacks in the United States. Criminology & Public Policy, 13(3), 455–481.

Collective action theory 1. Ruggiero, V. (2010). Armed struggle in Italy: The limits to criminology in the analysis of political violence. The British Journal of Criminology, 50(4), 708–724.

128

Criminological perspectives on extremist radicalisation

Conflict/radical 1. Ilan, J., & Sandberg, S. (2019). How ‘gangsters’ become Jihadists: Bourdieu, criminology and the crime – terrorism nexus. European Journal of Criminology, 16(3), 278–294.

Psychological 1. Corner, E., & Gill, P. (2020). Psychological distress, terrorist involvement and disengagement from terrorism: A sequence analysis approach. Journal of Quantitative Criminology, 36(3), 499–526.

Situational action 1. Clemmow, C., Bouhana, N., & Gill, P. (2020). Analyzing person‐exposure patterns in lone‐actor terrorism: Implications for threat assessment and intelligence gathering. Criminology & Public Policy, 19(2), 451–482.

Social construction 1. Mythen, G., & Walklate, S. (2006). Criminology and terrorism: Which thesis? Risk society or governmentality? British Journal of Criminology, 46(3), 379–398.

Symbolic interaction 1. Arrigo, B. A. (2010). Identity, international terrorism and negotiating peace: Hamas and ethics-based considerations from critical restorative justice. The British Journal of Criminology, 50(4), 772–790.

Atheoretical   1. Ahmad, F., & Monaghan, J. (2019). Mapping criminological engagements within radicalization studies. The British Journal of Criminology, 59(6), 1288–1308.   2. Ajil, A. (2020). Politico-ideological violence: Zooming in on grievances. European Journal of Criminology, 1477370819896223.   3. Amirault, J., Bouchard, M., Farrell, G., & Andresen, M. A. (2016). Criminalizing terrorism in Canada: Investigating the sentencing outcomes of terrorist offenders from 1963 to 2010. The Journal of Criminal Law and Criminology (1973–2010), 769–810.   4. Anderson, J. B. (2004). Hamdi v. Rumsfeld: Judicious balancing at the intersection of the executive’s power to detain and the citizen-detainee’s right to due process. The Journal of Criminal Law and Criminology, 95, 689.   5. Banks, W. C. (2009). Smart counterterrorism. Criminology & Public Policy, 8, 593.   6. Barberet, R. (2005). Country survey: Spain. European Journal of Criminology, 2(3), 341–368.   7. Bassiouni, M. C. (2007). The new wars and the crisis of compliance with the law of armed conflict by non-state actors. The Journal of Criminal Law and Criminology, 98, 711.   8. Boers, K., Walburg, C., & Kanz, K. (2017). Crime, crime control and criminology in Germany. European Journal of Criminology, 14(6), 654–678.   9. Borum, R. (2013). Informing lone-offender investigations. Criminology & Public Policy, 12, 103. 10. Brown, I., & Korff, D. (2009). Terrorism and the proportionality of internet surveillance. European Journal of Criminology, 6(2), 119–134. 11. Donovan, J., & Coupe, R. T. (2013). Animal rights extremism: Victimization, investigation and detection of a campaign of criminal intimidation. European Journal of Criminology, 10(1), 113–132. 12. Forst, B. (2009). Criminologists and terrorism-finding firearms under lampposts. Criminology & Public Policy, 8, 655.

129

Gary LaFree and Yesenia Yanez 13. Forst, B. (2017). Targeted killings: How should we assess them. Criminology & Public Policy, 16, 221. 14. Freilich, J. D., Chermak, S. M., & Caspi, D. (2009). Critical events in the life trajectories of domestic extremist white supremacist groups: A case study analysis of four violent organizations. Criminology & Public Policy, 8(3), 497–530. 15. Garduno, L. S., & Bales, W. D. (2014). Preventing terrorism: Complex social, scientific, and political tasks. Criminology & Public Policy, 13, 451. 16. Gill, P.,  & Corner, E. (2013). Disaggregating terrorist offenders: Implications for research and practice. Criminology & Public Policy, 12, 93. 17. Gruenewald, J. (2017). Do targeted killings increase or decrease terrorism. Criminology & Public Policy, 16, 187. 18. Gruenewald, J., Chermak, S., & Freilich, J. D. (2013). Distinguishing “loner” attacks from other domestic extremist violence: A comparison of far‐right homicide incident and offender characteristics. Criminology & Public Policy, 12(1), 65–91. 19. Jacques, K., & Taylor, P. J. (2013). Myths and realities of female-perpetrated terrorism. Law and Human Behavior, 37(1), 35. 20. Jensen, S. Q., & Larsen, J. F. (2019). Sociological perspectives on Islamist radicalization – bridging the micro/macro gap. European Journal of Criminology, 18(3), 426–443. 21. Joosse, P., Bucerius, S. M., & Thompson, S. K. (2015). Narratives and counternarratives: Somali-Canadians on recruitment as foreign fighters to Al-Shabaab. British Journal of Criminology, 55(4), 811–832. 22. Kleemans, E. R. (2008). Introduction to special issue: Organized crime, terrorism and European criminology. European Journal of Criminology, 5(1), 5–12. 23. Kupatadze, A.,  & Argomaniz, J. (2019). Introduction to special issue – understanding and conceptualizing European Jihadists: Criminals, extremists or both? European Journal of Criminology, 16(3), 261–277. 24. LaFree, G. (2007). Expanding criminology’s domain: The American society of criminology 2006 presidential address. Criminology, 45(1), 1–31. 25. LaFree, G. (2009). Criminology’s third war: Special issue on terrorism and responses to terrorism. Criminology & Public Policy, 8, 431. 26. LaFree, G. (2017). Terrorism and the internet. Criminology & Public Policy, 16, 93. 27. LaFree, G., & Freilich, J. D. (2019). Government policies for counteracting violent extremism. Annual Review of Criminology, 2, 383–404. 28. LaFree, G., Yang, S. M., & Crenshaw, M. (2009). Trajectories of terrorism: Attack patterns of foreign groups that have targeted the United States, 1970–2004. Criminology & Public Policy, 8(3), 445–473. 29. Lonky, H. (2017). Revisiting the public safety exception to Miranda for suspected terrorists. The Journal of Criminal Law and Criminology, 107(3), 393–420. 30. Lum, C., Kennedy, L. W., & Sherley, A. (2006). Are counter-terrorism strategies effective? The results of the Campbell systematic review on counter-terrorism evaluation research. Journal of Experimental Criminology, 2(4), 489–516. 31. Marguiles, J. (2011). Deviance, risk, and law: Reflections on the demand for the preventive detention of suspected terrorists. The Journal of Criminal Law and Criminology, 101, 729. 32. McNeal, G. S. (2011). The status quo bias and counterterrorism detention. The Journal of Criminal Law and Criminology, 101, 855. 33. Mohapatra, M. (2004). Learning lessons from India: The recent history of antiterrorist legislation on the subcontinent. The Journal of Criminal Law and Criminology, 95, 315. 34. Morris, N. A. (2015). Target suitability and terrorism events at places. Criminology & Public Policy, 14, 417. 35. Mythen, G., & Walklate, S. (2016). Counterterrorism and the reconstruction of (in) security: Divisions, dualisms, duplicities. British Journal of Criminology, 56(6), 1107–1124. 36. Norris, J. J., & Grol-Prokopczyk, H. (2015). Estimating the prevalence of entrapment in post-9/11 terrorism cases. The Journal of Criminal Law and Criminology, 105, 609. 37. Ortblad, V. (2008). Criminal prosecution in sheep’s clothing: The punitive effects of OFAC freezing sanctions. The Journal of Criminal Law and Criminology, 1439–1466. 38. Pelfrey Jr, W. V. (2014). Policing in an omnicultural environment: Population heterogeneity and terrorism prevention. Criminology & Public Policy, 13, 483. 39. Pyrooz, D. C., LaFree, G., Decker, S. H., & James, P. A. (2017). Cut from the same cloth? A comparative study of domestic extremists and gang members in the United States. Justice Quarterly, 35(1), 1–32.

130

Criminological perspectives on extremist radicalisation 40. Rushchenko, J. (2019). Terrorist recruitment and prison radicalization: Assessing the UK experiment of ‘separation centres’. European Journal of Criminology, 16(3), 295–314. 41. Scott, J. (2011). Confronting foreign intelligence: Crawford roadblocks to domestic terrorism trials. The Journal of Criminal Law and Criminology, 101, 1039. 42. Silke, A. (2008). Holy warriors: Exploring the psychological processes of Jihadi radicalization. European Journal of Criminology, 5(1), 99–123. 43. Smith, B. L., & Damphousse, K. R. (2009). Patterns of precursor behaviors in the life span of a US environmental terrorist group. Criminology & Public Policy, 8(3), 475–496. 44. Stern, J. (2009). Strengths and limits of criminological research on terrorism. Criminology & Public Policy, 8, 661. 45. Szmania, S., & Fincher, P. (2017). Countering violent extremism online and offline. Criminology & Public Policy, 16, 119. 46. Taylor, P. J., Holbrook, D., & Joinson, A. (2017). Same kind of different. Criminology & Public Policy, 16, 127. 47. van der Wilt, H., & Paulussen, C. (2019). The role of international criminal law in responding to the crime – terror nexus. European Journal of Criminology, 16(3), 315–331. 48. Walen, A. (2011). Criminalizing statements of terrorist intent: How to understand the law governing terrorist threats, and why it should be used instead of long-term preventive detention. The Journal of Criminal Law and Criminology, 101, 803. 49. Wolfowicz, M., Litmanovitz, Y., Weisburd, D., & Hasisi, B. (2020). A field-wide systematic review and meta-analysis of putative risk and protective factors for radicalization outcomes. Journal of Quantitative Criminology, 36(3), 407–447. 50. Wormeli, P. (2014). Developing policies for countering terrorism. Criminology & Public Policy, 13, 493. 51. Young, J. K. (2017). Morality, efficacy, and targeted assassination as a policy tool. Criminology & Public Policy, 16, 225. 52. Zedner, L., & Ashworth, A. (2019). The rise and restraint of the preventive state. Annual Review of Criminology, 22, 429–450.

131

9 INSIGHTS FROM THE STUDY OF NEW RELIGIOUS MOVEMENTS INTO THE PROCESS OF RADICALISATION Lorne L. Dawson ***

In situations where data are scarce or fragmentary, it can be beneficial to supplement the direct study of any social phenomenon with insights gleaned from the study of analogous ones. Care must be taken in making trans-situational comparisons to consider unique aspects of each context, but the challenges of studying extreme social phenomena often compel researchers to seek explanatory insights by identifying the fundamental psychological and social processes they share. Such is the case with efforts to understand the process of radicalisation leading to violence. The data available from interviews with radicalised individuals, or those close to them, is limited, in terms of the amount of data and its quality. The data available is often fragmentary, and when samples of radicalised individuals are acquired, they are not random or representative. The information available on specific issues (e.g. the educational attainments of terrorists) is usually incomplete, inconsistent, and sometimes non-comparable. The problem of specificity, a recurrent shortcoming of studies reliant on sampling the dependent variable, looms large. Insights into the process of radicalisation have come almost exclusively from studies of the radicalised, with little meaningful comparison with others, such as those exposed to similar conditions who did not radicalise or control groups. Consequently, the factors contributing to radicalisation that such studies isolate are insufficiently specific. They apply as well, it appears, to individuals who did not become violent extremists (Silke, 2008; Hodwitz, 2019; Dawson, 2019a; Schuurman, 2020; Derfoufi, 2022). In these circumstances, it is surprising that more researchers have not turned to the study of similar phenomena to help ameliorate the data and specificity problems. Some studies have called attention to the ways in which theories from criminology may be pertinent to understanding violent extremists (e.g. Wilström & Bouhana, 2017; LaFree et al., 2018; Sarnecki, 2018; Ahmad & Monaghan, 2019; Carlsson et al., 2020), and some scholars have begun to compare radicalisation with the process of joining violent youth gangs (e.g. Decker & Pyrooz, 2011, 2015; Benavides et al., 2016; Pyrooz et al., 2018; Becker et al., 2022). In general, a link has been presumed between criminal activity and terrorism (Mullins, 2009; Basra & Neuman, 2016; Agromaniz & Bermejo, 2019). Yet while many jihadists in Europe are criminals before they radicalise, most are not (Stuart, 2017; Dawson, 2021a, pp.  23–27; Jensen et al., 2020), and there is little evidence of prior DOI: 10.4324/9781003035848-11

132

Insights from the study of new religious movements

criminality for North American jihadists (Simcox & Dyer, 2013; Vidino & Hughes, 2015; START, 2018; Wilner, 2019). Another strong parallel exists between radicalisation and involvement in new religious movements (NRMs). This comparison, however, has received far less attention (e.g. Post, 1984; McCauley & Segal, 1987, pp. 241–247; Mayer, 2001; Introvigne, 2009; Dawson, 2010), despite the prominence of religious terrorism since 9/11. For jihadists, the most consistent and conspicuous mark of radicalisation is a surge in religiosity. The vast majority of ‘home-grown’ Western jihadists, whether they commit offences at home or leave to fight abroad, experience a religious conversion (or reversion) before engaging in terrorism. It is their involvement in an Islamic NRM, Salafi Jihadism (Meijer, 2013; Maher, 2016), that distinguishes them, and a disproportionate number of them are converts to Islam (e.g. Mullins, 2015; Schuurman et al., 2016; Jones & Dawson, 2023). Yet the copious research on conversion, based largely on the study of NRMs, has received little serious attention in radicalisation studies (e.g. Jones, 2008; Ferguson & Binks, 2015; Van den Elzen, 2018). Reticence to explore the comparison with NRMs may stem, as some have argued, from a failure to see jihadists as genuinely religious (e.g. Bale, 2013; Dawson, 2017a, 2018, 2021b, 2021c; Larsen, 2020; de Graaf & van den Bos, 2021). There are several natural points of convergence, however, between the empirical findings of research on NRMs and terrorism. One is with regard to who joins, how, and why. Another is with regard to explanations of why some NRMs become violent (see Dawson, 2010). Insights can be gleaned as well from the study of how NRMs respond to the failure of prophecy (see Dawson, 2017b). Within the limited bounds of this chapter, I will only sketch some of the parallels in who joins NRMs and jihadist terrorist groups, how, and why. This leads to consideration of some insights from the study of the process of conversion in NRMs and radicalisation. In conclusion, I reflect briefly on some deeper parallels in the explanatory contexts of studying NRMs and the radicalisation of religious terrorism. Each field of study faces challenges stemming from the normative nature of the phenomena under study and the ways they defy easy explanation in terms of rational choice modelling. Sometimes, the study of NRMs has something new to offer terrorism scholars. More often, the comparison simply involves reinforcing insights gained, with limited evidence, into the process of radicalisation, as well as highlighting the need for greater methodological reflexivity in explaining both types of extreme behaviour. The parallels between the research questions pursued, and the answers found, are real and strong, but messier than can be conveyed here. Research on both topics is often less than systematic because much of the work is impacted by a ‘hermeneutic of crisis management’ (Brannan et al., 2001), as researchers have responded to the transient fears and demands of governments and the public (Dawson, 2021b, pp. 9–11).

Similarities in who joins, how, and why? Discussions of who joins NRMs or becomes radicalised have followed a similar trajectory, one that can be summarised in terms of a movement through a series of explanatory conjectures and partial refutations. Popular conceptions of who joins NRMs are riddled with stereotypes. Sometimes, converts are thought to be young, idealistic, and gullible people duped by cunning cult recruiters. Sometimes they are maladjusted and marginal losers who have found a safe haven in the controlled life of a cult. Other times, it is suggested that everyone is susceptible to being recruited. In the media and the propaganda of the anti-cult movement (Melton, 1999), the three views are often blended together using anecdotal evidence (Dawson, 2006, p. 71). Much the same is true of popular conceptions of who becomes a terrorist. They are either crazy, duped, evil, or somehow all three. In the face of the challenges that the actions of cultists and terrorists pose to our 133

Lorne L. Dawson

conventional conceptions of the good life, such stereotypical thinking is common, and it persists long after the stereotypes are called into question by research. In seeking to explain these similar forms of deviance, researchers have commonly assumed that some form of deprivation is instrumental to understanding what is happening and why, because they cannot otherwise understand why individuals would turn their backs on the conventional norms and aspirations of society. The deprivations evoked tend to have three conceptual foci: socio-economic, psychological, and identity issues. In the midst of the ‘great American cult scare’ (Bromley & Shupe, 1981) in the 1970s and 1980s, many assumed those who joined these marginal, seemingly deviant, and perhaps even anti-social groups were driven by a desire to compensate for some deprivation. The assumption stemmed, in part, from the limited knowledge of such successful NRMs of the late 19th and early 20th centuries as the Mormons, Pentecostalists, and Jehovah’s Witnesses. These groups drew their followers disproportionately from the under-privileged segments of society, and it was hypothesised that converts were seeking to offset their economic distress with the intense emotional and social rewards of participating in highly cohesive groups that declared their members to be morally superior (Niebuhr, 1957). The ideology of these sects reversed the existing social order; they fervently proclaimed the apocalyptic ideal that on judgement day, the first shall be last and the last first. In the case of terrorist groups, it seems common sense recommends a similar view. Suicide bombers in the Palestinian-Israeli conflict and elsewhere, for example, were driven by desperation in the face of their dim economic, political, and personal prospects, while domestic jihadists in the West were seeking revenge for the social and economic marginalisation suffered by generations of Muslim immigrants. The thinking of sociologists of religion shifted with the recognition that sectarian groups were becoming more middle-class and thriving while mainstream religions were going into decline (Roof & McKinney, 1987; Finke & Stark, 2005). Likewise, it was becoming apparent that those joining NRMs were actually the children of privilege. In addition to being young, they had aboveaverage educations and came almost exclusively from middle- to upper-class families (see Dawson, 2006, pp. 84–85, for an overview of the data). The study of religious terrorism followed a similar pattern. Initially, many assumed that poverty played an important role in motivating terrorism, until several studies cast significant doubt on that claim (e.g. Sageman, 2004; Krueger, 2007). As Krueger concluded (2007, p. 89): neither the available micro-level data nor the more aggregated data show much of a correlation between income and the origins of terrorism. Likewise, educational attainment and involvement in terrorism, at both the individual level and the country level are either uncorrelated or positively correlated. For some years, many researchers assumed that economic factors were a poor predictor of radicalisation. The research in hand, however, was inconsistent, and some scholars began to question that view (e.g. Piazza, 2011; Hegghammer, 2016). Many European scholars examining the surge in jihadist foreign fighters travelling to fight in Syria and Iraq (from 2012 to 2016), for example, gravitated back to the view that the mobilisation of jihadists was heavily influenced by their social and economic marginalisation (e.g. Weggemanns et al., 2014; Bakker & Grol, 2015; Weenink, 2015; Coolsaet, 2016; Gustafsson & Ranstorp, 2017). Some have argued that contemporary jihadism is more a manifestation of a rebellious youth movement among the second generation of Muslim immigrants, born of their social and economic struggles, than an outgrowth of religious extremism (Roy, 2015, 2017; Coolsaet, 2016). The findings undergirding this low-prospect 134

Insights from the study of new religious movements

interpretation, however, are fragmentary, often statistically weak, and methodologically problematic (Dawson, 2021a, pp. 13–22). Bergema and van San (2019), for example, present findings on the class background, education, and occupations of the 217 Dutch foreign fighters. Overall, they conclude, ‘the majority . . . came from the lower levels of society, lacking (tertiary) school or vocational qualifications, and are oftentimes unemployed or stuck in unskilled labour’ (2019, p. 656). Yet, they admit, assessing the socioeconomic status of the fighters involves some ‘crude judgement’ (2019, p. 645), and the small size of the data samples available for each category within their already small sampling of Western foreign fighters limits the generalisability of their findings. The youthfulness of the sample, plus the high percentage of students in many cases, poses problems as well, and no baseline information is presented for comparable youth, whether immigrants or otherwise. From this data alone, it is difficult to draw any conclusions about whether these fighters were actually reacting to their low prospects. How many of these young people, for example, would have finished their education if they had not radicalised and left for Syria and Iraq? Most of the research on European jihadists is subject to similar doubts. Certainly, the low prospects theory does not appear to apply anywhere near as uniformly to foreign fighters from the United Kingdom, the United States, and Canada. Commenting on the situation, Meleagrou-Hitchens et al. (2018) note: ‘The sample of American IS supporters cut across economic boundaries, and American Muslims as a population tend to experience greater levels of economic success and integration than their counterparts in other Western countries’. While in Canada, 57% of domestic jihadist terrorists either completed an undergraduate degree or were at least enrolled at university, and if college is included, 63% undertook some education beyond high school (Harris-Hogan et al., 2020, p. 88). In a large transnational study, Benmelech and Klor (2020) compare data on the numbers and origins of foreign fighters in Syria and Iraq with data on the social, political, and economic conditions of the countries from which they came. Using measures of political freedom, social fragmentation, economic development, inequality, and unemployment, they discovered there is little correlation between poor economic conditions and the recruitment of foreign fighters. Contrary to expectation, they conclude (2020, p. 1459): the number of ISIS foreign fighters is positively correlated with a country’s GDP per capita and its Human Development Index (HDI). In fact, many foreign fighters originate from countries with high levels of economic development, low income inequality, and highly developed political institutions. In fact, they go on to stress, ‘controlling for other socioeconomic variables, income inequality is associated with fewer – not more – ISIS foreign fighters’ (2020, p. 1475). What then accounts for the flow of foreign fighters from Western European countries? The answer lies in two other positive correlations they detected: the size of a country’s Muslim population and the degree of its linguistic and cultural homogeneity (2020, p. 1459). Highly homogeneous countries with larger Muslim minorities produce more fighters. They postulate that this is because Muslim immigrants experience more difficulties assimilating into these societies, which leaves more individuals open to radicalisation (2020, p. 1460). The perceived lack of acceptance, the social isolation, matters most, and ISIS creatively exploited this with its ideology and propaganda (2020, p. 1476). The latter finding suggests that it is not so much real deprivation that matters as perceived or ‘relative deprivation’. When people think there is a discrepancy between the social rewards they 135

Lorne L. Dawson

feel entitled to and the rewards they are getting, or they believe others are getting, and they do not accept some rational explanation for their deprivation, there is an incentive to join a movement that promises change or compensation. As an explanation for involvement in extremism, the idea of relative deprivation seems plausible, and this line of thinking has influenced the study of both NRMs and terrorism (e.g. Glock, 1964; Wilson, 1978; Gurr, 1970; Gupta, 2005; Spalek, 2007; Franz, 2007). The relevant perception of deprivation, however, need not only be economic. As speculated long ago (Glock, 1964, pp. 27–28; Aberle, 1962), other forms of relative deprivation may help to explain NRMs. Forms of ‘social deprivation’, associated with the unequal distribution of social status or sense of belonging, may be relevant. As might forms of ‘psychic deprivation’ related to the unequal fulfilment of various kinds of psychological needs, such as affection or self-worth. Alternatively, ‘ethical deprivation’, stemming from the sense that the dominant values of society will not support living a meaningful life, may be crucial. No matter how we conceive the options, it is easy to see how many types of relative deprivation may also play a role in becoming a terrorist. Whatever the merits of this admittedly rather vague way of conceiving the impetus for joining an NRM or an extremist group, the role of relative psychic deprivation has received considerable attention. In each case, the research followed a trajectory similar to that of studies examining real economic deprivations. For a prolonged period, it was widely assumed that people must be ‘crazy’ to engage in such strange and socially unacceptable activities. In the case of the jihadists, the deviation from normality involved participation in the indiscriminate killing of others and the sacrifice of one’s own life (through death or long imprisonment). For those joining NRMs, such as the Unification Church, people were perplexed that middle-class students, with promising futures, would abandon school to proselytise ten hours a day for a self-declared messiah from Korea who spoke no English. Recruits lived ascetic lives in groups dedicated to studying the new revelations and seeking funds for creating the Kingdom of God on earth (Barker, 1984). In both cases, the actions suggested a ‘psychological anomaly’ (Silke, 1998, p. 52), and many assumed they must be rooted in diagnosable psychopathologies or personality disorders. In the context of NRMs, the quest for an underlying mental disorder was part of the belief that NRMs brainwashed their recruits (Conway & Seigelman, 1978; Singer, 1995). By medicalising the issue, the anti-cult movement hoped to find a legally defensible rationale for interfering with such groups (Robbins & Anthony, 1982; Richardson, 1991; Reichert et al., 2015). The traits and behaviours proposed, however, were too vague, ambiguous, or insufficiently present to justify identifying cultists or terrorists as psychologically deprived, in a real or relative sense. Summary analyses of the research on NRMs are provided by Rochford et al. (1989), Saliba (1993, 2004), Richardson (1995), and Aronoff et al. (2000); for terrorists, summary analyses are provided by Silke (1998), Horgan (2005), and Victoroff (2005). In the end, Silke concludes, it is the ‘normality’ of the terrorists that stands out (1998, p. 62), while Levine observes that, after clinical interviews with hundreds of converts to NRMs, they ‘showed no more sign of pathology . . . than is found in any youthful population’ (1984, p. 4). While few scholars any longer expect to find a clear relationship with one, or a few psychological disorders, some have begun looking for more refined, complex, and heterogeneous correlations between an array of psychological issues and different relevant behaviours. Using bigger datasets that disaggregate the information on terrorists and a wider array of behavioural indicators, they use more sophisticated statistical analyses to discern correlations (Gill et al., 2014; Corner & Gill, 2015; Corner et al., 2016; Corner & Gill, 2017; Gill et al., 2021). Mental health issues play a larger, but still less than anticipated, role in instances of lone-actor terrorism but not in groupbased terrorism. They are also more prevalent with single-issue lone actors (e.g. animal rights and anti-abortion) than jihadist or far-right terrorists (Corner & Gill, 2015, pp. 30–31). The prevalence 136

Insights from the study of new religious movements

of mental disorders among group-based terrorists, however, is actually lower than in the general population. For a more detailed discussion of the extensive new research available on the variable presence, prevalence, and relevance of a wide array of psychological variables and personality traits, readers should consult Emily Corner’s chapter on ‘Radicalisation and Mental Health’ in this volume. However, psychopathology remains only one possible factor in the process of radicalisation for a minority of the individuals who become extremists, and even then, in diverse ways, we are still figuring out how. Similar studies of coverts and NRMs have gone beyond the simple psychopathology versus normality debate. They investigate aspects of the cognitive and emotional well-being of members of some NRMs in comparison with the general population and other religious converts (Buxant et al., 2007; Namini & Murken, 2009). While there are some reliability issues because of the reliance on self-reports, the findings are consonant with the earlier studies. There are no strong links between mental health issues and conversion to NRMs. In this regard, converts to NRMs do not differ from converts to other religions, and psychological tests of converts overall indicate that, after converting, they have higher levels of psychological well-being than the general population (Paragament, 1997, pp. 295–297). As interest in socio-economic deprivation and psychopathology waned, researchers turned more to the vaguer notion that the actions of converts and extremists were in response to identity crises and deficits. The youthfulness of most converts and terrorists pointed to this possibility, given the vicissitudes of identity so characteristic of adolescence and early adulthood. The hypothesised link is conceived in many ways (Tipton, 1982; Roof, 1993; Dawson, 2006, pp. 39–60, 2017c; Roy, 2004; Taylor & Louis, 2004; Schwartz et al., 2009; Berger, 2018). These studies, however, are more speculative than systematic, relative to the vast body of empirical and theoretical literature available (e.g. Giddens, 1991; Baumeister, 1997; Sen, 2007; Hornsey, 2008). Consequently, here I am limited to addressing the link indirectly in terms of the analysis of conversions. Changes in identity remain the implicit substrate of conversion studies, and some of the key developments in this field have implications for the study of the process of identity change in radicalisation.

Religious conversions and the process of radicalisation In its simplest sense, the term conversion describes a switch in someone’s religious beliefs and affiliations. The change can be from having no religion to having one, between religions, or sometimes towards a more intense practice of one’s religion (e.g. becoming a ‘born-again’ Christian). Conversion research, however, tends to focus on dramatic instances of such change, ones going beyond a mere change in affiliation or adhesion. A conversion in this sense involves a real rupture with one’s previous life and a significant change in one’s sense of self and worldview (Heirich, 1977; Bankston et al., 1981; Snow & Machalek, 1984). It does not involve, however, changes in basic personality traits and styles (Paloutzian et al., 1999). Decades of research on conversions, prompted primarily by the study of NRMs, demonstrate that conversion is a complex and multifaceted process. As with radicalisation, the experience is unique to each individual, yet there are similarities indicative of a generic pattern. Attempts to capture this generic process are too numerous to summarise here (e.g. Lofland & Stark, 1965; Lofland & Skonovd, 1981; Snow & Machalek, 1984; Richardson, 1985; Gartrell & Shannon, 1985; Rambo, 1993; Gooren, 2010), and in this chapter, I can only indicate the relevant parallels with radicalisation by explicating six basic insights from the study of conversion in NRMs. First, while much of the study of conversion involved case studies or the application of specific theories to a handful of groups, by the 1990s there was increasing recognition of the need 137

Lorne L. Dawson

to integrate a wider array of factors, encompassing micro (personal), meso (group), and macro (societal) levels of analysis. Scholars of radicalisation are now seeking to do the same and devising models that capture the diverse interactions of individuals with multiple environments (e.g. Dawson, 2017c, 2023; Bouhana, 2019; Kruglanski et al., 2018). In this regard, as Ferguson and Binks (2015) suggest, there may be value in emulating Lewis Rambo’s comprehensive and multi-level model of conversion (Rambo, 1993). Rambo’s work provides a synthesis of conversion research, delineating what is known about the ‘Context’ of conversions, the experiences of ‘Crisis’ and ‘Quest’ preceding conversions, the ‘Encounter’, ‘Interaction’, and ‘Commitment’ phases of conversions, and finally the ‘Consequences’ of conversions. These categories may need to be adapted for the new context and made more consistently processual, but they are suggestive of the full range of data that should be taken into consideration in delineating the interactions generating identity change in the instance of radicalisation. Unlike early efforts to model radicalisation (e.g. Moghaddam, 2005; Silber & Bhatt, 2007), this approach is not about charting a series of stages through which everyone passes. Rambo stresses that these categories are analytical constructs for organising data on the process that may be pertinent in different cases to differing degrees in somewhat different ways and sequences (Rambo & Bauman, 2012). Second, scholars of conversion have demonstrated that conversions are heavily conditioned by the normative expectations and language of different traditions (Beckford, 1978; Taylor, 1978; Stromberg, 1993). Conversion is a social construct that shapes what is happening as it happens, not just after the fact. This insight has two important implications for the study of conversions and radicalisations. First, it points to the active role played by ideology in the processes of change experienced, and not just as a post-hoc rationalisation of the change. Second, it underlines, nevertheless, why we must be cautious in using the accounts of the change provided by participants. Very little of the experience of conversion or radicalisation is uninterpreted (i.e. raw data). Yet, as outsiders to the processes, researchers must strive to become familiar with the participants’ ‘definition of the situation’ to minimise introducing other socio-cultural conceptions of what is happening into scholarly interpretations (Dawson, 2019b). As scholars of conversion seem to realise far more than those studying radicalisation, the interpretive challenge is far more complex than many seem to appreciate (Dawson, 2017a, 2021c). Third, in the context of conversion to NRMs, much had been made of the seeming suddenness of the conversions of young people, and this claim was used to bolster the argument that the conversions were not genuine but rather the result of psychological manipulation by NRM recruiters. A closer study revealed, however, that the choice to join a group usually involved a more gradual and private process of self-scrutiny than realised (Barker, 1984; Levine, 1984; Richardson, 1985; Jones, 2008). The conversion process tends to be more prolonged and tentative than commonly appreciated. In their seminal theory of conversion to a deviant perspective, Lofland and Stark (1965) stress that converts only pass from being ‘verbal converts’ to ‘true converts’ after they have traversed multiple stages and become fully integrated into their new religion. Studies also reveal that it is common for converts to lay ‘side-bets’ about their future, should things not work out as planned (Becker, 1961; Richardson, 1985). Little or no comparable research exists on the precise nature of the struggles of individuals radicalising. The limited data available, largely from case studies, suggests a variety of scenarios, but as Klausen et al. (2016, p. 69) found, the process should be ‘measured in months and years rather than days and weeks’. Jones and Dawson (2023) discovered that the radicalisation of converts to Islam is probably nowhere near as swift as most of the studies seeking to explain the disproportionate presence of converts among Western jihadists presumed. On average, the gestation period of radicalisation for many jihadists, converts or otherwise, may be more gradual than commonly thought. 138

Insights from the study of new religious movements

The process may be more tentative as well. Sampling a diverse array of NRMs, Bird and Reimer (1982) found that two-thirds of the most highly involved members eventually left, while Levine’s (1984, p. 15) interviews with hundreds of participants in NRMs revealed that over 90% left within two years. Carefully tracing those who attended a Unification Church workshop in London in 1979 (N = 1,017), Barker found that only 0.5% were ‘associated with the movement two years later, and by no means all of these [were] full-time members’ (1984, pp. 147, 259). These findings led Dawson (2010, p. 9) to ask if the same pattern might apply to political extremists, as some anecdotal evidence suggests. At this point, we do not know. These findings suggest that joining an NRM may be only a phase of identity experimentation and development, a view reinforced by the relative psychological health of converts and (voluntary) ex-converts. They also indicate, however, that some people, admittedly very few, are satisfied with their involvement in an NRM. Can the same be said about extremists and terrorists? Fourth, as Sageman (2004) established for Al-Qaeda and others have confirmed for other terrorists (e.g. Della Porta, 1988; Malthaner, 2014; Reinares et al., 2017; Saal, 2021; Kanol, 2022), recruitment is largely through pre-existing social networks, and much depends on the interpersonal bonds developed. Sageman (2004, pp.  125–130) calls on analogous findings from the study of NRMs to support his argument (e.g. Lofland & Stark, 1965; Barker, 1984; Galanter, 1989). This research, he notes (Sageman, 2004, p. 128), shows that ‘[m]embership spreads through social networks, and faith constitutes conformity to the religious outlook of one’s intimates (Stark & Bainbridge, 1980; Bainbridge, 1997, p. 177)’. In a comprehensive study of Sōka Gakkai in the United Kingdom, Wilson and Dobbelaere (1994, p. 50) found: Only 6 per cent of those in our sample had encountered [the group] through the impersonal agencies of the media – through exhibitions, concerts, the movement’s own publicity, or various media accounts of the organization. . . . Ninety-four per cent met the movement through social interaction. Friends represented the largest category of people who introduced members, amounting to some 42 per cent; 23 per cent were brought into contact with it through their partners or family members. The remainder were first presented with information by acquaintances, work or student colleagues most particularly, but 14 per cent owed the encounter to casual acquaintances. The advent of the Internet has certainly significantly augmented the networking capacities of all movements (Sardarnia  & Safizadeh, 2019), but it seems that in-person network relations remain fundamentally important (Nesser, 2015; Gill et al., 2017; Saal, 2021). Fifth, under the influence of social movement theory, scholars studying NRMs and terrorism first stressed the role of social influences and socialisation in the relevant processes of identity change, or what should perhaps be more accurately identified as processes of resocialisation and deconditioning (Wilson, 1984). This view fits well with the tight-knit NRMs at the heart of the cult scare, such as the Unification Church, Hare Krishna, and Scientology, and it seems to apply to such traditional terrorist organisations as the IRA and Red Brigades or such groups as al-Muhajiroun (Della Porta, 1988; Wiktorowicz, 2005; Sageman, 2011). The fit is less obvious, however, in the more amorphous circumstances of homegrown terrorism. Most homegrown jihadists are not exposed to systematic programmes of resocialisation and deconditioning. Their new identity involves bonding with local co-conspirators as well as more abstractly identifying with the global ummah. Unlike recruits to the classic NRMs, with their totalistic demands, recruits to homegrown terrorism have many more opportunities to desist. In this regard, homegrown terrorists have more in common with converts to the more loosely organised new age, neo-pagan, and 139

Lorne L. Dawson

alternative spirituality movements that have multiplied in Western societies (Hanegraaff, 1996; Luhrmann, 1989; Pike, 2000). In both instances, scholars have found it harder to explain why the conversion happens. Tanya Luhrmann developed a sophisticated theory of ‘cognitive drift’ to explain the path of identity transformation taken by those drawn to these alternative spiritualities. It seeks to explain how even educated professionals learn to live comfortably within a magical sub-culture in a scientific age. Lessons might be learned from such research for understanding how religious extremists also emerge from similar personal, familial, and social backgrounds. The research could augment the multifaceted models of radicalisation emerging in response to the explanatory challenges of homegrown terrorism (Dawson, 2017c, 2023; Kruglanski et al., 2018; Bouhana, 2019). Sixth, the surprising success of movements of alternative spirituality in the late modern Western world highlighted the need to recognise the agency of converts to NRMs. The popular debate about the social problem of NRMs focused on the accusation of brainwashing levelled by the anti-cult movement (Dawson, 2006, pp. 95–124). This happened in the context of scholarly conceptions of conversion that conformed too closely to traditional religious understandings of conversion, which cast converts as passive recipients of conversion. Like St. Paul on the road to Damascus, blinded by a flash of light from which Jesus spoke, conversions were typically thought to be sudden and dramatic, emotional, or even irrational events. They involve the intrusion of some external force into people’s lives (whether natural, supernatural, or social). The empirical study of NRMs, however, suggested another view. Most conversions appeared to result from a combination of prolonged tensions, consideration of options, and interactions between individuals and groups. Coverts are not converted; they convert. They are active participants in a process marked by three features that previous research tended to overlook. First, most conversions are preceded by periods of ‘seekership’, during which the potential converts search for answers to their spiritual questions and needs (Strauss, 1976; Sutcliffe, 2017). Second, conversions often involve some negotiation between the potential convert and the recruiting group, whereby the group seeks to win the convert and the convert learns about the benefits of converting. Third, the conversions commonly involve role-play. Converts don the convert role, mimicking expected behaviours, before eventually committing to the new worldviews (Strauss, 1976; Bromley & Shupe, 1979; Balch, 1980; for an overview, see Dawson, 2006, pp. 117–119). The discovery of ‘side-bets’ also points to the agency of converts. Even individuals in leadership roles in NRMs often reveal they have considered their options, in case life in the group proves unsatisfactory (Kilbourne & Richardson, 1988, pp. 10–11). This relative activism resonates with much that we know about many (if not most) cases of homegrown and lone-actor jihadists (Bakker & Grol, 2015; Marone, 2016; McDonald, 2018; Sieckelinck et al., 2019; Alonso & Delgado, 2023). But we still lack a sufficient understanding of the specific cognitive and affective struggles experienced by ‘converts’ to violent extremism. We lack the psychological insights available from the study of converts to NRMs (Barker, 1984; Levine, 1984). If the parallelism of radical identity change in conversions and radicalisation holds, then many aspects of the active conception of religious conversions may have implications for understanding radicalisation. Faced with the specificity problem, especially as manifest in the gap between radical talk and action, many researchers have sought to differentiate ‘cognitive radicalisation’ and ‘behavioral radicalisation’. It is the latter, they argue, that really matters when it comes to countering terrorism, and researchers should stop prioritising the role of ideology in favour of tracking the behavioural indicators of radicalisation (Borum, 2011; Neumann, 2013; Khalil, 2014; McCauley & Moskalenko, 2014, 2017). As the study of conversion to NRMs indicates, however, drawing a sharp distinction between the cognitive and behavioural aspects of radicalisation may 140

Insights from the study of new religious movements

be misleading. Beliefs alone do not lead to violent extremism, as some have naively implied, and the clearest signs of radicalisation may often be behavioural. Yet, as conversion studies also reveal, behavioural changes alone may not be a reliable indicator of actual commitment to a cause. It may only indicate that someone is what Lofland and Stark (1965) call a ‘verbal convert’, and not a ‘total convert’. Ninety per cent of converts to NRMs leave in less than two years. Surely, role-playing may factor into the process of resocialisation and deconditioning called radicalisation. Learning new roles is a common aspect of assuming new identities, and people often imitate what they aspire to be as part of the process of becoming them. There is, moreover, a growing recognition that radicalisation often involves some fantasy and imaginative role-play. This is part of the gradual process of engaging with a sub-culture like those created by NRMs, including Salafi Jihadism (Pisoiu, 2015; McDonald, 2018; Larsen & Jensen, 2019; Cottee, 2020; Dawson & Amarasingam, 2021). Reversing the onus of the arguments of Borum (2011), McCauley and Moskalenko (2014, 2017), and others, behavioural changes may not be the most reliable indicators of radicalisation. In fact, Pisoiu’s (2015) application of two dominant schools of sub-cultural analysis to the lives of four jihadists and three right-wing extremists in Germany corroborates three insights from the study of NRMs. First, she found no evidence that socio-economic marginalisation and strain played a determinant role in the radicalisation of these extremists. Second, she argues, it is the more cultural aspects of a process of identity struggle and its sub-cultural resolution that are pivotal. Third, calling on aspects of sub-cultural theory, she stresses that it is important to realise that the implicit determinism of most theories of radicalisation is misleading. Those radicalising, like those creatively engaging in sub-cultures, display significant agency. These considerations need to be kept in mind when talking about the ‘vulnerability’ and resilience of youth to recruitment by extremists, a phrasing commonly encountered in discussions of preventing violent extremism. The emphasis placed on detecting and diverting ’vulnerable individuals’ can lend itself to an overly passive reading of the process of radicalisation, which is curiously at odds with the agency implicitly invoked in the partnerships, and mentoring relationships, so often at the core of CVE efforts. Like converts, those who are radicalised are cast as victims, to curry public support for efforts to ‘rescue’ them. Yet, like programmes designed to reform alcoholics and drug addicts, it is widely recognised that a change for the better depends significantly on the will of the person being assisted. Their agency matters a great deal, though appropriate social and structural supports are needed to facilitate and guide this agency to achieve a successful result. Even the religious revivalists, waiting for the saving grace of God, know that salvation is only possible when the sinner decides to be saved and recommits daily to the way of the Lord. Overall, then, we should resist dichotomist approaches to the cognitive and behavioural aspects of radicalisation (Holbrook & Horgan, 2019; Dawson, 2021c, pp. 14–15). We need to focus on the ways in which the cognitive (ideological) and behavioural aspects of radicalisation co-evolve throughout the process. The behavioural changes are more obvious, especially in the early stages, but as research on NRMs indicates, they are likely the external manifestations of the more cognitive process of seeking. As people interact with the jihadist sub-culture the nature of their involvement and the reasons for it change. Accordingly, Khalil et al. (2022) recommend tracing the shifts in the balance of attitudinal (cognitive) and behavioural aspects of the radical identity of violent extremists over time.

Concluding remarks: on the need for greater methodological reflexivity At a deeper and less obvious level, researchers studying why people join NRMs and the process of radicalisation share a set of similar methodological tensions. In both fields, they feel the torque of 141

Lorne L. Dawson

competing agendas and assumptions. In line with the strong societal urge to condemn both activities, there is a marked tendency to treat the process of involvement and those getting involved as not only morally suspect but also non-rational or irrational, or at best passive victims of manipulation. Yet that assumption hinders understanding, and as partially indicated here, it does not hold up to empirical scrutiny. In most respects, both converts to NRMs and terrorists are remarkably ordinary, and as such, their voluntary decisions and actions remain relatively open to understanding (i.e. rational explanation). Yet many of the beliefs held and actions undertaken by these individuals are extra-ordinary. They deviate significantly from societal norms. The conundrum of the ordinary doing the extraordinary makes the processes involved in identity transformation perplexing. In this situation, there is a recurrent tension between the desire to understand what is happening and to reject it morally and socially. Deterrence and moral condemnation go hand in hand, and in principle so should understanding and deterrence, but the quest for understanding often seems to be curtailed, in subtle ways, by the fear that humanising the processes involved implies condoning or legitimating them. In the end, converts to NRMs and the radicalised are ‘othered’ in problematic ways by the limits of our capacity to understand them, both morally and factually. This tension is aggravated when the processes under study are religious. Contemporary, and largely secular, scholars tend to have an ambiguous relationship with religious phenomena. Since they doubt the veracity of religious beliefs, they are suspicious of claims about the primacy and/ or authenticity of religious commitments. Religion and false consciousness are closely associated, and researchers are strongly inclined to search for other reasons why these people are engaging in these processes of change. There is a compunction to do so even more when the beliefs and practices deviate so much from the accepted conventions of society and other religions. Yet these same scholars live in societies that continue to see religion as a social good, and place an absolute value on the freedom of religious expression. This right is integral to sustaining the liberal-democratic social order to which most are committed, morally and rationally. This situation sets up another implicit tension in the explanation of conversions to NRMs and radicalisation. Many researchers were prompted to study NRMs by their commitment to the liberal-democratic order. They were seeking the information required to protect NRMs from the prejudicial and specious accusations of brainwashing levelled by the anti-cult movement. They were intent on helping these minority religions escape legal interference and popular persecution. Yet few of the scholars in question would have been pleased to see their own sons and daughters join an NRM. In the case of those studying radicalisation, the tensions play out somewhat differently. Even recognising the religiousness of the process, in the case of religious terrorism, poses problems. Since secular scholars are inclined to view religious commitments as non-rational, recognising the religiousness of the process seems to diminish the capacity to explain it, and hence strategically counter it. This conundrum supports the search for the latent (i.e. non-religious) reasons or functions of the process – the psychological and social ones that they think are more subject to rational explanation and assessment. Yet this reductive approach entails some ambivalence because they too live in societies that valorised religion and being religious, and they wish to avoid implying, in the case of jihadism, that adherence to Islam is implicated in setting the conditions for radicalisation. In each case, in the study of NRMs and religious terrorism, the tension is ameliorated, in practice, by harkening to the deviant (i.e. unorthodox and unconventional) nature of the religious beliefs and practices in question. That means, however, that a normative judgement, and a religious one at that, is conditioning the research, and that is contradictory to the social scientific aspirations of the research. It is the ambivalence implicitly felt about these crosscutting explanatory tensions, I suspect, that accounts for the reluctance of researchers studying the radicalisation of religious 142

Insights from the study of new religious movements

extremists to investigate the insights available from the study of NRMs, and more particularly the process of religious conversion. Alternatively, with greater methodological reflexivity, we can grant greater possible explanatory significance to religiosity in some instances of radicalisation without either legitimating the objectionable goals of the extremists or undermining our capacity to understand what is happening to them. We must prioritise understanding further and learn how to better bracket our moral qualms about the phenomenon we are studying.

References Aberle, D. (1962). Millennial dreams in action. In S. L. Thrupp (Ed.), Comparative studies in society and history, Supplement II (pp. 209–214). Mouton. Agromaniz, J., & Bermejo, R. (2019). Jihadism and crime in Spain: A convergence setting approach. European Journal of Criminology, 16(3), 351–368. Ahmad, F., & Monaghan, J. (2019). Mapping criminological engagements within radicalization studies. British Journal of Criminology, 59(6), 1288–1309. Alonso, R.,  & Delgado, P. (2023). The radicalization of young Jihadi convicted of membership of terrorist organization in Spain: Analyzing the pieces of the puzzle. Studies in Conflict and Terrorism, 46(5), 594–617. Aronoff, J., Lynn, S. J., & Malinowski, P. (2000). Are cultic environments psychologically harmful? Clinical Psychology Review, 20(1), 91–111. Bakker, E., & Grol, P. (2015, July). Motives and considerations of potential foreign fighters from the Netherlands. International Centre for Counter-Terrorism, Policy Brief. www.icct.nl/wp-content/uploads/2015/07/ ICCT-Bakker-Grol-Motives-and-Considerations-of-Potential-Foreign-Fighters-from-the-NetherlandsJuly2015.pdf Bainbridge, W. S. (1997). The sociology of religious movements. Routledge. Balch, R. (1980). Looking behind the scenes in a religious cult: Implications for the study of conversion. Sociological Analysis, 41, 137–143. Bale, J. M. (2013). Denying the link between Islamist ideology and Jihadist terrorism: ‘Political correctness’ and the understanding of counterterrorism. Perspectives on Terrorism, 7(5), 5–46. Bankston, W. B., Forsyth, C. J., & Floyd, H. H. (1981). Toward a general model of the process of radical conversion: An interactionist perspective on the transformation of self-identity. Qualitative Sociology, 4(4), 279–297. Barker, E. (1984). The making of a Moonie: Choice or brainwashing? Blackwell. Basra, R., & Neumann, P. R. (2016). Criminal pasts, terrorist futures: European Jihadists and the new crimeterror nexus. Perspectives on Terrorism, 10(6), 25–40. Baumeister, R. F. (1997). Identity, self-concept, and self-esteem: The self lost and found. In R. Horgan, J. Johnson, & S. Briggs (Eds.), Handbook of personality psychology (pp. 681–710). Elsevier. Becker, H. S. (1961). Outsiders: Studies in the sociology of deviance. The Free Press. Becker, M. H., Decker, S. H., LaFree, G., Pyrooz, D. C., Ernest, K., & James, P. A. (2022). A comparative study of initial involvement in gangs and political extremism. Terrorism and Political Violence, 34(8), 1647–1664. Beckford, J. (1978). Accounting for conversion. British Journal of Sociology, 29(2), 249–262. Benavides, A. D., Keyes, L. M., & Pulley, B. (2016). Understanding the recruitment methods and socialization techniques of terror networks by comparing them to youth gangs: Similarities and divergences. In S. Ekici et al. (Eds.), Countering terrorist recruitment in the context of armed counter-terrorism operations (pp. 40–54). IOP Press. Benmelech, E., & Klor, E. F. (2020). What explains the flow of foreign fighters to ISIS? Terrorism and Political Violence, 32(7), 1458–1481. Bergema, R., & van San, M. (2019). Waves of the Black banner: An exploratory study on the Dutch Jihadist foreign fighter continent in Syria and Iraq. Studies in Conflict and Terrorism, 42(7), 636–661. Berger, J. M. (2018). Extremism. MIT Press. Bird, F., & Reiner, B. (1982). Participation rates in new religious movements. Journal for the Scientific Study of Religion, 21(1), 1–14. Borum, R. (2011). Radicalization into violent extremism I: A review of social science theories. Journal of Strategic Security, 4(4), 7–36.

143

Lorne L. Dawson Bouhana, N. (2019). The moral ecology of extremism: A systematic perspective. U.K. Commission for Countering Extremism. https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/834354/Bouhana-The-moral-ecology-of-extremism.pdf Brannan, D. W., Esler, P. F., & Strindberg, N. T. A. (2001). Talking to terrorists: Towards an independent analytical framework for the study of violent substate activism. Studies in Conflict and Terrorism, 24(1), 3–24. Bromley, D. G., & Shupe, A. D. (1979). Just a few minutes seem like a lifetime: A role theory approach to participation in religious movements. In L. Kriesberg (Ed.), Research in social movements, conflict and change (Vol. 2, pp. 159–186). JAI Press. Bromley, D. G., & Shupe, A. D. (1981). Strange gods: The great American cult scare. Beacon Press. Buxant, C., Saroglou, V., Casalfiore, S., & Christians, L-L. (2007). Cognitive and emotional characteristics of new religious movement members: New questions and data on the mental health issue. Mental Health, Religion and Culture, 10(3), 219–238. Carlsson, C., Rostami, A., Mondani, H., Sturup, J., Sarnecki, J., & Edling, C. (2020). A life-course analysis of engagement in violent extremist groups. British Journal of Criminology, 60(1), 74–92. Conway, F., & Seigelman, J. (1978). Snapping: America’s epidemic of sudden personality change. J. B. Lippincott. Coolsaet, R. (2016, March). Facing the fourth foreign fighter wave: What drives Europeans to Syria, and to Islamic state? Insights from the Belgian Case, Egmont – The Royal Institute for International Relations. www.egmontinstitute.be/facing-the-fourth-foreign-fighters-wave/ Corner, E., & Gill, P. (2015). A false dichotomy? Mental illness and lone-actor terrorism. Law and Human Behavior, 39(1), 23–34. Corner, E., & Gill, P. (2017). Is there a nexus between terrorist involvement and mental health in the age of the Islamic state? CTC Sentinel, 10(1), 1–10. Corner, E., Gill, P., & Mason, O. (2016). Mental health disorders and the terrorist: A research note probing selection effects and disorder prevalence. Studies in Conflict and Terrorism, 39(6), 560–568. Cottee, S. (2020). The Western Jihadi subculture and subterranean values. The British Journal of Criminology, 60(3), 762–781. Dawson, L. L. (2006). Comprehending cults: The sociology of new religious movements (2nd ed.). Oxford University Press. Dawson, L. L. (2010). The study of new religious movements and the radicalization of home-grown terrorists: Opening a dialogue. Terrorism and Political Violence, 22, 1–21. Dawson, L. L. (2017a). Discounting religion in the explanation of homegrown terrorism: A critique. In J. R. Lewis (Ed.), Cambridge companion to religion and terrorism (pp. 32–45). Cambridge University Press. Dawson, L. L. (2017b). The failure of prophecy and the future of IS. The International Centre for Counter-Terrorism. https://icct.nl/wp-content/uploads/2017/09/ICCT-Dawson-The-Failure-of-Prophecy-and-The-Future-ofISIS-Sept-2017.pdf Dawson, L. L. (2017c). Sketch of a social ecology model for explaining homegrown terrorist radicalisation. International Centre for Counter-Terrorism. https://icct.nl/publication/sketch-of-a-socialecology-model-for-explaining-homegrown-terrorist-radicalisation/ Dawson, L. L. (2018). Challenging the curious erasure of religion from the study of religious terrorism. Numen, 65(2), 141–164. Dawson, L. L. (2019a, Spring). Clarifying the explanatory context for developing theories of radicalization: Five basic considerations. Journal for Deradicalization, 18, 50–87. Dawson, L. L. (2019b). Taking terrorist accounts of their motivations seriously: An exploration of the hermeneutics of suspicion. Perspectives on Terrorism, 13(5), 65–80. Dawson, L. L. (2021a, February 10). A comparative analysis of data on western foreign fighters in Syria and Iraq: Who went and why? International Centre for Counter-Terrorism. https://icct.nl/app/uploads/2021/02/ Dawson-Comparative-Analysis-FINAL-1.pdf Dawson, L. L. (2021b). Bringing religiosity back in: Critical reflection on the explanation of western homegrown religious terrorism (part I). Perspectives on Terrorism, 15(1), 2–16. Dawson, L. L. (2021c). Bringing religiosity back in: Critical reflection on the explanation of western homegrown religious terrorism (part II). Perspectives on Terrorism, 15(2), 1–21. Dawson, L. L. (2023). The social ecology model of homegrown Jihadist radicalisation. In A. N. Awan & J. R. Lewis (Eds.), Radicalisation in comparative perspective. Hurst & Oxford University Press.

144

Insights from the study of new religious movements Dawson, L. L., & Amarasingam, A. (2021). Homegrown terrorist radicalization: The Toronto 18 in comparative perspective. In M. Nesbitt, K. Roach, & D. Hofmann (Eds.), Canadian terrorism: Multidisciplinary perspectives on the Toronto 18 terrorism trials. Manitoba Law Journal (Special Issue). de Graaf, B. A., & van den Bos, K. (2021). Religious radicalization: Social appraisals and finding radical redemption in extreme beliefs. Current Opinions in Psychology, 40, 56–60. Decker, S., & Pyrooz, D. (2011). Gangs, terrorism, and radicalization. Journal of Strategic Security, 4(4), 151–166. Decker, S., & Pyrooz, D. (2015). ‘I’m down for a Jihad’: How 100 years of gang research can inform the study of terrorism, radicalization and extremism. Perspectives on Terrorism, 9(1), 104–112. Della Porta, D. (1988). Recruitment process in clandestine political organizations: Italian left-wing terrorism. In B. Klandermans, H. Kriesi, & S. Tarrow (Eds.), International social movement research (Vol. 1, pp. 155–169). JAI Press. Derfoufi, Z. (2022). Radicalization’s core. Terrorism and Political Violence, 34(6), 1185–1206. Ferguson, N., & Binks, E. (2015). Understanding radicalization and engagement in terrorism through religious conversion motifs. Journal of Strategic Security, 8(1), 16–26. Finke, R., & Stark, R. (2005). The churching of America, 1776–2005: Winners and losers in our religious economy. Rutgers University Press. Franz, B. (2007). Europe’s Muslim youth: An inquiry into the politics of discrimination, relative deprivation, and identity formation. Mediterranean Quarterly, 18(1), 89–112. Galanter, M. (1989). Cults: Faith, healing, and coercion. Oxford University Press. Gartrell, C. D., & Shannon, Z. K. (1985). Contacts, cognitions, conversion: A rational choice approach. Review of Religious Research, 27(1), 32–48. Giddens, A. (1991). Modernity and self-identity. Polity Press. Gill, P., Clemmow, C., Hetzel, F., Rottweiler, B., Salman, N., Van Der Vegt, I., Marhcment, Z., Schumann, S., Zolghadriha, S., Schulten, N., Taylor, H., & Corner, E. (2021). Systematic review of mental health problems and violent extremism. The Journal of Forensic Psychiatry, 32(1), 51–78. Gill, P., Corner, E., Conway, M., Thornton, A., Bloom, M., & Horgan, J. (2017). Terrorist use of the Internet by the numbers: Quantifying behaviors, patterns, and processes. Criminology and Public Policy, 16, 99–117. Gill, P., Horgan, J., & Deckert, P. (2014). Bombing alone: Tracing the motivations and antecedent behaviors of lone-actor terrorists. Journal of Forensic Sciences, 59(2), 425–435. Glock, C. Y. (1964). The role of deprivation in the origin and evolution of religious groups. In R. Lee & M. Marty (Eds.), Religion and social conflict (pp. 24–36). Oxford University Press. Gooren, H. (2010). Religious conversion and disaffiliation: Tracing patterns of change in faith practices. Palgrave Macmillan. Gupta, D. K. (2005). Exploring roots of terrorism. In T. Bjørgo (Ed.), Root causes of terrorism (pp. 16–32). Routledge. Gurr, T. R. (1970). Why men rebel. Princeton University Press. Gustafsson, L., & Ranstorp, M. (2017). Swedish foreign fighters in Syria and Iraq: An analysis of open-source intelligence and statistical data. Center for Asymmetric Threat Studies, Swedish Defence University. www.diva-portal.org/smash/record.jsf?pid=diva2%3A1110355&dswid=-450 Hanegraaff, W. J. (1996). New age religion and western culture. E. J. Brill. Harris-Hogan, S., Dawson, L. L., & Amarasingam, A. (2020). A comparative analysis of the nature and evolution of the domestic Jihadist threat to Australia and Canada (2000–2020). Perspectives on Terrorism, 14(5), 77–102. Hegghammer, T. (2016, November 8). Revisiting the poverty-terrorism link in European Jihadism. Paper presented to the Society for Terrorism Research. http://hegghammer.com/_files/Hegghammer_-_poverty. pdf.Hegghammer Heirich, M. (1977). Change of heart: A test of some widely held theories about religious conversion. American Journal of Sociology, 83(3), 653–680. Hodwitz, O. (2019, October). Recognizing and resolving issues in terrorism research, data collection, and analysis. Researching Violent Extremism Series, RESOLVE Network. www.resolvenet.org/research/ recognizing-resolving-issues-terrorism-research-data-collection-analysis Holbrook, D., & Horgan, J. (2019). Terrorism and ideology: Cracking the nut. Perspectives on Terrorism, 13(6), 2–15.

145

Lorne L. Dawson Horgan, J. (2005). The psychology of terrorism. Routledge. Hornsey, M. (2008). Social identity theory and self-categorization: A historical review. Social and Personality Psychology Compass, 2(1), 204–222. Introvigne, M. (2009). Of ‘cultists’ and ‘martyrs’: The study of new religious movements and suicide terrorism in conversation. In M. Al-Rasheed & M. Shterin (Eds.), Dying for the faith: Religiously motivated violence in the contemporary world (pp. 43–48). I. B. Tauris. Jensen, M. A., Safer-Lichtenstein, A., James, P. A., & LaFree, G. (2020). The link between prior criminal record and violent political extremism in the United States. In D. Weisburd, E. U. Savona, B. Hesis, & F. Calderoni (Eds.), Understanding recruitment to organized crime and terrorism (pp. 121–146). Springer. Jones, D. A.,  & Dawson, L. L. (2023). Re-examining the explanations of convert radicalization in SalafiJihadist terrorism with evidence from Canada. Behavioral Sciences of Terrorism and Political Aggression, 15(2), 246–273. Jones, J. W. (2008). Converting to terrorism: What psychology of religion tells us about religiously motivated terrorism. Paper presented at the American Psychological Association annual meeting, www. researchgate.net/profile/James-Jones-12/publication/237498525_CONVERTING_TO_TERRORISM_ WHAT_THE_PSYCHOLOGY_OF_RELIGION_TELLS_US_ABOUT_RELIGIOUSLY_MOTIVATED_TERRORISM/links/56cf3ab708ae059e375965a6/CONVERTING-TO-TERRORISM-WHATTHE-PSYCHOLOGY-OF-RELIGION-TELLS-US-ABOUT-RELIGIOUSLY-MOTIVATED-TERRORISM.pdf Kanol, E. (2022). Contexts of radicalization of Jihadi foreign fighters from Europe. Perspectives on Terrorism, 16(3), 45–62. Khalil, J. (2014). Radical beliefs and violent actions are not synonymous: How to place the key disjuncture between attitudes and behaviors at the heart of our research into political violence. Studies in Conflict and Terrorism, 37, 198–211. Khalil, J., Horgan, J., & Zeuthen, M. (2022). The attitude-behaviors corrective (ABC) model of violent extremism. Terrorism and Political Violence, 34(3), 225–250. Kilbourne, B., & Richardson, J. T. (1988). Paradigm conflict, types of conversion, and conversion theories. Sociological Analysis, 51(1), 1–21. Klausen, J., Campion, S., Needle, N., Nguyen, G., & Libretti, R. (2016). Toward a behavioral model of “homegrown’ radicalization trajectories. Studies in Conflict and Terrorism, 39(1), 67–83. Krueger, A. B. (2007). What makes a terrorist? Economics and the roots of terrorism. Princeton University Press. Kruglanski, A., Webber, D., Jasko, K., Chernikova, M., & Molinario, E. (2018). The making of violent extremists. Review of General Psychology, 22(1), 107–120. LaFree, G., Jensen, M. A., James, P. A., & Safer-Lichtenstein, A. (2018). Correlates of violent political extremism in the United States. Criminology, 56(2), 233–268. Larsen, J. F. (2020). The role of religion in Islamist radicalisation processes. Critical Studies on Terrorism, 13(2), 396–417. Larsen, J. F., & Jensen, S. Q. (2019). Jihadism from a subcultural perspective. Critical Criminology, 27, 421–436. Levine, S. V. (1984). Radical departures: Desperate detours to growing up. Harcourt Brace Jovanovich. Lofland, J., & Stark, R. (1965). Becoming a world-saver: A theory of conversion to a deviant perspective. The American Sociological Review, 30(6), 863–874. Lofland, J., & Skonovd, N. (1981). Conversion motifs. Journal for the Scientific study of Religion, 29(4), 373–385. Luhrmann, T. (1989). Persuasions of the witch’s craft. Harvard University Press. Maher, S. (2016). Salafi-Jihadism: The history of an idea. Hurst. Malthaner, S. (2014). Contextualizing radicalization: The emergence of the ‘Sauerland-group’ from radical networks and the Salafist movement. Studies in Conflict and Terrorism, 37(8), 638–653. Marone, F. (2016, August). Italy’s Jihadists in the Syrian civil war. International Centre for Counter-Terrorism. https://icct.nl/publication/italys-jihadists-in-the-syrian-civil-war/ Mayer, J-F. (2001). Cults, violence, and religious terrorism: An international perspective. Studies in Conflict and Terrorism, 24(5), 361–376. McCauley, C. R., & Moskalenko, S. (2014). Toward a profile of lone wolf terrorists: What moves an individual from radical opinion to radical action. Terrorism and Political Violence, 26(1), 69–85.

146

Insights from the study of new religious movements McCauley, C. R., & Moskalenko, S. (2017). Understanding political radicalization: The two-pyramids model. American Psychologist, 72(3), 205–216. McCauley, C. R., & Segal, M. E. (1987). Social psychology of terrorist groups. In C. Hendrick (Ed.), Review of personality and social psychology, vol. 9, Group processes and intergroup relations (pp. 231–256). Sage. McDonald, K. (2018). Radicalization. Polity Press. Meijer, R. (2013). Global Salafism: Islam’s new religious movement. Oxford University Press. Meleagrou-Hitchens, A., Hughes, S., & Clifford, B. (2018). The travellers: American Jihadist in Syria and Iraq, Program on Extremism. George Washington University. https://extremism.gwu.edu/sites/g/files/ zaxdzs2191/f/TravelersAmerican-JihadistsinSyriaandIraq.pdf Melton, G. J. (1999). Anti-cultists in the United States: An historical perspective. In B. Wilson & J. Cresswell (Eds.), New religious movements: Challenge and response (pp. 213–233). Routledge. Moghaddam, F. M. (2005, February–March). The staircase to terrorism: A psychological exploration. American Psychologist, 161–169. Mullins, S. (2009). Parallels between crime and terrorism: A social psychology perspective. Studies in Conflict and Terrorism, 32, 811–830. Mullins, S. (2015). Re-examining the involvement of converts in Islamist terrorism: A comparison of the U.S. and U.K. Perspectives on Terrorism, 9(6), 72–84. Namini, S., & Murken, S. (2009). Self-chosen involvement in new religious movements (NRMs): Well-being and mental health from a longitudinal perspective. Mental Health, Religion and Culture, 12(6), 561–585. Nesser, P. (2015). Islamist terrorism in Europe: A history. Oxford University Press. Neumann, P. R. (2013). The trouble with radicalization. International Affairs, 89(4), 873–893. Niebuhr, H. R. (1957 [1929]). The social sources of denominationalism. World Pub Co. Paloutzian, R. F., Richardson, J. T., & Rambo, L. (1999). Religious conversion and personality change. Journal of Personality, 67, 1047–1079. Paragament, K. I. (1997). The psychology of religious coping. Guilford Press. Piazza, J. A. (2011). Poverty, minority economic discrimination, and domestic terrorism. Journal of Peace Research, 48(3), 339–353. Pike, S. (2000). Earthly bodies, magical selves. University of California Press. Pisoiu, D. (2015). Subcultural theory applied to Jihadi and right-wing radicalization in Germany. Terrorism and Political Violence, 27(1), 9–28. Post, J. M. (1984). Notes on a psychodynamic theory of terrorist behavior. Terrorism: An International Journal, 7, 241–256. Pyrooz, D. C., LaFree, G., Decker, S. H., & James, P. A. (2018). Cut from the same cloth? A comparative study of domestic extremists and gang members in the United States. Justice Quarterly, 35(1), 1–32. Rambo, L. R. (1993). Understanding religious conversion. Yale University Press. Rambo, L. R., & Bauman, S. C. (2012). Psychology of conversion and spiritual transformation. Pastoral Psychology, 61, 879–894. Reichert, J., Richardson, J. T., & Thomas, R. (2015). ‘Brainwashing’: Diffusion of a questionable concept in legal systems. International Journal for the Study of New Religions, 6(1), 3–26. Reinares, F., Garcia-Calco, C., & Vicente, A. (2017, July). Differential association explaining Jihadi radicalization in Spain: A quantitative study. CTC Sentinel, 29–34. Richardson, J. T. (1985). The active vs. Passive convert: Paradigm conflict in conversion/recruitment research. Journal for the Scientific Study of Religion, 24(2), 163–179. Richardson, J. T. (1991). Cult/brainwashing cases and freedom of religion. Journal of Church and State, 33(1), 55–74. Richardson, J. T. (1995). Clinical and personality assessment of participants in new religions. International Journal for the Psychology of Religion, 5(3), 145–170. Robbins, T., & Anthony, D. (1982). Deprogramming, brainwashing and the medicalization of deviant religious groups. Social Problems, 29(3), 283–297. Rochford, E. B., Purvis, S., & Eastman, N. (1989). New religions, mental health, and social control. In M. Lynn & D. Moberg (Eds.), Research in the social scientific study of religion (Vol. 1, pp. 57–82). JAI Press. Roof, W. C. (1993). A generation of seekers: The spiritual journeys of the baby boom generation. Harper-Collins. Roof, W. C., & McKinney, W. (1987). American mainline religion: Its changing shape and future. Rutgers University Press.

147

Lorne L. Dawson Roy, O. (2004). Globalized Islam: The search for a new ummah. Columbia University Press. Roy, O. (2015, November 24). Le djhadisme est une révolte générationnelle et nihillisté. Le Monde. www.lemonde.fr/idees/article/2015/11/24/le-djihadisme-une-revolte-generationnelle-et-nihiliste_4815992_3232.html Roy, O. (2017). Jihad and death: The global appeal of Islamic state. Hurst. Saal, J. (2021). The dark social capital of religious radicals: Jihadi networks and mobilization in Germany, Austria and Switzerland, 1998–2018. Springer. Sageman, M. (2004). Understanding terror networks. University of Pennsylvania Press. Sageman, M. (2011). The turn to political violence in the west. In R. Coolsaet (Ed.), Jihadi terrorism and the radicalisation challenge (pp. 117–129). Ashgate. Saliba, J. A. (1993). The new religions and mental health. In D. G. Bromley & J. K. Hadden (Eds.), Religion and the social order, vol. 3 – the handbook on cults and sects in America, part B (pp. 99–113). JAI Press. Saliba, J. A. (2004). Psychology and the new religious movements. In J. R. Lewis (Ed.), The Oxford handbook of new religious movements (pp. 317–332). Oxford University Press. Sardarnia, K., & Safizadeh, R. (2019). The internet and its potentials for networking and identity seeking: A study on ISIS. Terrorism and Political Violence, 31(6), 1266–1283. Sarnecki, J. (2018). A criminological perspective on recruitment of men and women to Daesh. International Annals of Criminology, 56, 122–144. Schuurman, B. (2020). Research on terrorism, 2007–2016: A review of data, methods, and authorship. Terrorism and Political Violence, 32(5), 1011–1026. Schuurman, B., Grol, P., & Flower, S. (2016, June). Converts and Islamist terrorism: An introduction. International Centre for Counter-Terrorism. http://icct.nl/publication/converts-and-islamist-terrorism-an-introduction/ Schwartz, S. J., Dunkel, C. S., & Waterman, A. S. (2009). Terrorism: An identity theory perspective. Studies in Conflict and Terrorism, 32(6), 537–559. Sen, A. (2007). Identity and violence: The illusion of destiny. Penguin. Sieckelinck, S., Sikkens, E., van San, M., Kotnis, S., & de Winter, M. (2019). Transitional journeys into and out of extremism: A biographical approach. Studies in Conflict and Terrorism, 42(7), 662–682. Silber, M. D., & Bhatt, A. (2007). Radicalization in the West: The Homegrown Threat. New York City Police Department. https://sethgodin.typepad.com/seths_blog/files/NYPD_Report_radicalization_in_the_West. pdf Silke, A. (1998). Cheshire-cat logic: The recurring theme of terrorist abnormality in psychological research. Psychology, Crime and Law, 4(1), 51–69. Silke, A. (2008). Research on terrorism: A review of the impact of 9/11 on the global war on terrorism. In H. Chen, E. Reid, J. Sinai, A. Silke, & B. Ganor (Eds.), Terrorism informatics knowledge management and data mining for homeland security (pp. 27–50). Springer. Simcox, R., & Dyer, E. (2013). Al-Qaeda in the United States: A complete analysis of terrorism offences. Henry Jackson Society. https://henryjacksonsociety.org/wp-content/uploads/2013/02/Al-Qaeda-in-theUSAbridged-version-LOWRES-Final.pdf Singer, M. T. (1995). Cults in our midst: The hidden menace in our everyday lives. Jossey-Bass. Snow, D., & Machalek, R. (1984). The sociology of conversion. In R. H. Turner & J. P. Short Jr. (Eds.), Annual review of sociology (pp. 167–190). Annual Review Inc. Spalek, B. (2007). Disconnection and exclusion: Pathways to radicalisation? In T. Abbas (Ed.), Islamic political radicalism: A European perspective (pp. 192–206). University of Edinburgh Press. Stark, R., & Bainbridge, W. S. (1980). Networks of faith: Interpersonal bonds and recruitment to cults and sects. American Journal of Sociology, 80(6), 1376–1395. START. (2018). Pre-radicalization criminal activity of United States extremists. Research Brief, National Consortium for the study of Terrorism and Responses to Terrorism, University of Maryland. https://start. umd.edu/publication/pre-radicalization-criminal-activity-united-states-extremists Strauss, R. (1976). Changing oneself: Seekers and the creative transformation of life experience. In J. Lofland (Ed.), Doing social life (pp. 252–273). Wiley & Sons. Stromberg, P. G. (1993). Language and self-transformation: A study of the Christian conversion narrative. Cambridge University Press. Stuart, H. (2017). Islamist terrorism: Analysis of offences and attacks in the UK (1998–2015). Henry Jackson Society. https://henryjacksonsociety.org/wp-content/uploads/2017/03/Islamist-Terrorism-preview-1.pdf Sutcliffe, S. J. (2017). Seekership revisited: Explaining traffic in and out of new religions. In E. V. Gallagher (Ed.), Visioning new and minority religions (pp. 33–46). Routledge.

148

Insights from the study of new religious movements Taylor, B. (1978). Recollection and membership: Converts’ talk and the ratiocination of commonality. Sociology, 12, 315–324. Taylor, D. M., & Louis, W. (2004). Terrorism and the quest for identity. In F. M. Moghaddam & A. J. Marsella (Eds.), Understanding terrorism: Psychological roots, consequences, and interventions (pp. 169–185). American Psychological Association. Tipton, S. M. (1982). Getting saved from the sixties. University of California Press. Van den Elzen, J. (2018). Radicalisation: A subtype of religious conversion? Perspectives on Terrorism, 12(1), 69–80. Victoroff, J. (2005). The mind of the terrorist: A review and critique of psychological approaches. Journal of Conflict Resolution, 49(1), 3–42. Vidino, L., & Hughes, S. (2015). ISIS in America: From retweets to Raqqa. Program on Extremism, George Washington University. https://extremism.gwu.edu/isis-america Weenink, A. (2015). Behavioral problems and disorders among radicals in police files. Perspectives on Terrorism, 9(2), 17–33. Weggemanns, D., Bakker, E., & Grol, P. (2014). Who are they and why do they go? The radicalisation and preparatory processes of Dutch Jihadist foreign fighters. Perspectives on Terrorism, 8(4), 100–110. Wiktorowicz, Q. (2005). Radical Islam rising: Muslim extremism in the west. Rowman & Littlefield. Wilner, A. (2019). Canadian terrorists by the numbers: An assessment of Canadians joining and supporting terrorist groups. MacDonald-Laurier Institute. https://macdonaldlaurier.ca/files/pdf/20190205_MLI_ Canadian_Terrorists_Wilner_PAPER_WebFinal.pdf Wilson, B. R. (1978). Becoming a sectarian: Motivation and commitment. In D. Baker (Ed.), Religious motivation: Biographical and sociological problems for the church historian (pp. 481–507). Blackwell. Wilson, B. R., & Dobbelaere, K. (1994). A time to chant: The Sōka Gakkai Buddhists in Britain. Clarendon Press. Wilson, S. R. (1984). Becoming a yogi: Resocialization and deconditioning as conversion processes. Sociology of Religion, 45(4), 301–314. Wilström, P-O. H., & Bouhana, N. (2017). Analyzing radicalization and terrorism: A situational action theory. In G. LaFree & J. D. Freilich (Eds.), The handbook of the criminology of terrorism (pp. 175–186). Wiley-Blackwell.

149

10 THE PEN AND THE SWORD Cognition, emotion, communication, and violent radicalisation Kurt Braddock

Since the empirical study of terrorism and other forms of political violence began in earnest in the late 20th century, the field has been dominated by several academic disciplines. Scholars from political science, psychology, sociology, and criminology comprise the majority of academics exploring terrorism. Radicalisation typically refers to a social and psychological process by which an individual experiences greater commitment to a terrorist ideology, the extension of which is violent radicalisation, which theoretically puts individuals at greater risk for engaging in terrorist violence. Because radicalisation often occurs as a result of an individual’s interaction with terrorist propaganda, other people, or some combination thereof, communication studies can provide some value in explaining the process by which some individuals undergo the process. In this chapter, I explore this perspective, with a particular focus on two models of social influence that can explain how communication processes can motivate terrorist activity. This chapter will also offer insight on how communication studies can inform the development of strategic efforts to challenge terrorist propaganda with the goal of preventing violent radicalisation on the part of vulnerable audiences. Rather than engage with issues related to counter-narratives (as is skilfully done by Benjamin Lee in this volume), this chapter will discuss other counter-persuasion perspectives with promise for challenging terrorist messaging. These include reasoned action theory (a cognitive-based perspective) and discrete emotion theory (an affective-based perspective). Although these two perspectives are alluded to in counter-radicalisation efforts, they are rarely (if ever) explicated to the point that their contributions to knowledge and practice are fully leveraged. It is my goal to provide such an explanation to facilitate the use of these perspectives in developing messages that undermine violent radicalisation. Of course, these are not the only theories related to communication that can be used to understand social and psychological processes related to ideological violence. There exist a number of communication theories related to social influence, entertainment, education, and a host of other foci. However, the reasoned action and discrete emotions perspectives were specifically chosen for this chapter for their association with behavioural motivation. Both theories are not only convenient for, but indeed designed for, explaining all kinds of human behaviour – from joining a book club to setting off a bomb strapped to one’s chest. The reasoned action and discrete emotions perspectives were also chosen for this chapter because of their complementary foci. Reasoned action theory is exclusively concerned with DOI: 10.4324/9781003035848-12 150

The pen and the sword

perception and cognition – assuming that individuals make decisions about behaviours as a result of careful deliberation related to behaviour utility, norms, and perceived control over one’s own behaviour. The discrete emotions perspective takes the opposite approach, explaining behaviour as a function of our visceral responses to environmental stimuli. The truth about human behaviour likely lies somewhere between these two extremes, so it benefits the reader to understand how both cognition and emotion might influence one’s decision to engage in politically motivated violence. However, before turning to the theories themselves and how they inform our understanding of radicalisation to violence and how to prevent it, it is necessary to place radicalisation in the proper context as a process of social influence. The next section seeks to address this goal by defining radicalisation to violence as a process that is inextricably linked to communication generally and persuasion specifically.

Radicalisation to violence as a process of social influence Despite conceptual disagreements about how radicalisation to violence occurs, there are themes that cut across different conceptualisations that allow us to boil it down to its fundamental defining characteristics. In comparing different perspectives on radicalisation, there are three factors that cut across all descriptions. First, radicalisation is almost always described as being phasic in nature. Although there is disagreement about what constitutes a ‘phase’, radicalisation is largely understood to be a cumulative process that occurs over time and across multiple personal, political, and social dimensions. Second, radicalisation is often described as being characterised by increased commitment to beliefs and attitudes consistent with an extremist ideology, but not necessarily a parallel commitment to engage in violence on behalf of that ideology. Finally, radicalisation is driven, at least in part, by exposure to content that advocates beliefs and attitudes consistent with an extremist ideology. Each of these factors overlaps with definitions and conceptualisations of persuasion and social influence in the communication literature. In this vein, social influence theorists commonly equate persuasion with behavioural modification. We generally consider someone ‘persuaded’ when they either begin doing something we want them to do or stop doing something we don’t want them to do. Although we tend to equate persuasion to behavioural conversion, behaviours do not change in and of themselves. For someone to be persuaded to engage in a behaviour, they also need to develop certain beliefs, attitudes, and intentions with respect to that behaviour. So, despite the fact that we may be ultimately interested in changing how someone acts, persuasion also involves attempts to modify a target’s beliefs, attitudes, or intentions. It is also important to note that persuasion is not only an active process; that is, it is not always something that occurs such that ‘Person A’ simply communicates with ‘Person B’ in a way that gets ‘Person B’ to do something they want. For instance, Person B might engage with material that Person A created and ­published on the Internet years ago. Regardless of the distinction as to whether persuasion occurs actively or passively, the target of persuasive messages often experiences a change in their ideological position. Given this, we can define persuasion as a phenomenon whereby beliefs, attitudes, intentions, or behaviours (or any combination thereof) are modified by messages that appeal to the reason and/or emotions of the message recipient(s). It therefore follows that if someone is exposed to persuasive messages that induce changes in their beliefs or attitudes such that they become more closely aligned with those advocated by an extremist ideology, then radicalisation is an inherently persuasive process. And if radicalisation involves a shift in beliefs and attitudes such that they are consistent with an extremist ideology, radicalisation to violence involves a shift in intentions and behaviours such that they further extremist ideologies or goals through physical violence. 151

Kurt Braddock

Moreover, if radicalisation to violence is a process that is inherently related to persuasion and social influence, it follows that we can explain radicalisation (and counter-radicalisation) using theories and perspectives salient to these processes. In the next section of this chapter, I outline two of these perspectives: reasoned action theory and discrete emotion theory. These theories were chosen because they respectively exemplify cognitive- and affective-focused perspectives. However, there exist a wide range of other communication-based theories and perspectives that could also be useful in explaining radicalisation through violence and counter-radicalisation. Some of these can be found in other chapters in this volume.

Using cognitive and affective models of social influence to understand radicalisation and counter-radicalisation processes This section of the chapter explores the utility of two theories for understanding radicalisation to violence and how counter-radicalisation (i.e. the process of preventing the adoption of beliefs, attitudes, intentions, or behaviours that support the use of extremist violence) might be effectively implemented. These theories are reasoned action theory and discrete emotion theory.

Reasoned action theory: cognition, radicalisation, and counter-radicalisation Reasoned action theory dictates that an individual’s beliefs about a given behaviour – which are themselves predicated on several background factors – are the most critical predictors of whether the individual will engage in that behaviour. The initial developers of reasoned action theory, Martin Fishbein and Icek Ajzen, further argued that although a person’s beliefs about a behaviour form the foundation of the behaviour’s performance, the relationship between beliefs and action is not direct. Instead, someone’s ultimate decision to engage in a behaviour is a function of several relationships linking beliefs to several other outcomes, which themselves are linked to behavioural intentions, which in turn predict manifest behaviour (Fishbein & Ajzen, 2010). The multifaceted nature of this system of relationships is represented in Figure 10.1. Following the arrows that illustrate the connections between different elements of the theory shows that although different beliefs about a behaviour are the primary determinants of that behaviour, the link between beliefs and behaviour is not direct. Instead, a person’s beliefs about a behaviour affect their attitudes, perceived norms, and perceptions of control they have over performing that behaviour. In turn, these factors affect the person’s intention to perform the behaviour, which then has a direct effect on whether the behaviour is engaged in. Although one’s beliefs predict their behaviour, they are only the starting point. To fully capture how the series of relationships that comprise reasoned action theory influence one’s behaviour, it is important to consider each of the theory’s component relationships. The following sections explicate these respective relationships.

Elements of reasoned action theory BACKGROUND AND CONTEXTUAL FACTORS

Although beliefs about a behaviour represent the first (indirect) cognitive determinants of that behaviour’s performance, beliefs about a behaviour have their own predictors as well. In the framework of reasoned action theory, factors that predict an individual’s beliefs about a behaviour are called background and contextual factors. 152

The pen and the sword

Figure 10.1  Reasoned action theory Source: Braddock (2020), Chp.5

Background and contextual factors are environmental and experiential elements that form the basis for a person’s understanding of the world around them. This can include their religion, their education, their job, their friends, their extent belief systems, or any manner of other factors that might influence how they perceive certain behaviours. Although many of these background factors are related to relatively static characteristics (e.g. religion and education level), there are others that are influenced by dynamic communicative processes. Specifically, Fishbein and Ajzen indicate that the messages to which one is exposed can affect that person’s beliefs, attitudes, perceived norms, and perceptions of behavioural control in relation to a behaviour. The next section explicates these critical elements of the theory. PREDICTORS OF BEHAVIOURAL INTENTIONS

Because the terms are regularly used interchangeably, there is often confusion about the difference between beliefs and attitudes. The difference between the two concepts is predicated on valence. That is, whereas beliefs are ideas about the nature of the world in which a person exists, attitudes incorporate value judgements into the characteristics of that world. As I have written elsewhere, ‘beliefs are not good, bad, favorable, unfavorable, pleasant, unpleasant, happy, [or] sad’ (Braddock, 2020, p. 144). Instead, they are merely one’s perceptions about the environment around them. For instance, if, after watching a local weather forecast, a person states, ‘I think it will rain tomorrow’, that person would be stating a belief. There is no value appended to their thought; it simply represents their conception of the weather in the coming 24 hours. However, if, after watching that same local weather forecast, a different person states, ‘the weather will be unpleasant tomorrow’, they would be expressing an attitude about the weather. By attaching value (unpleasantness) to the idea that it will rain in the coming day, the latter person has developed a specific attitude about the rain. 153

Kurt Braddock

We will return to attitudes later in this section, but for now, it is important to consider the different kinds of beliefs that are influenced by background factors. Specifically, reasoned action theory contends that background factors directly influence one’s beliefs about (a) the nature of the behaviour itself, (b) norms associated with the behaviour, and (c) their ability to perform the behaviour (owing to personal capacity and environmental restraints). Behavioural beliefs and attitudes about a behaviour  Behavioural beliefs are an individual’s unvalenced ideas about engaging in a behaviour and the consequences that will result from doing so. Over time and experiences engaging in a behaviour, a person will develop a substantial set of beliefs about that behaviour. Eventually, the beliefs that the individual has cultivated through their experiences will form the basis for their sense of how favourable it might be to partake in the behaviour. Unlike beliefs, which do not have value judgements attached to them, attitudes represent judgements about a behaviour that include perceptions of the behaviour’s favourability. This favourability is determined by the individual’s beliefs of (a) how useful a behaviour may be for achieving a specific goal and (b) how pleasurable the behaviour will be to engage in. Normative beliefs and perceived norms  Unlike behavioural beliefs, which relate to an individual’s perceptions about a behaviour, normative beliefs relate to an individual’s notions about others’ (particularly others whom the individual values) stances on the behaviour under consideration. Research on norms and normative pressure has shown that there are two different kinds of norms that place pressure on an individual’s intention to engage in a behaviour. These norms are respectively called injunctive norms and descriptive norms. Injunctive norms represent an individual’s beliefs about whether others would approve (or disapprove) of the individual engaging in a specific behaviour. For instance, if an individual is considering marching in support of a white-power militia, that individual will consider what his family, friends, teachers, co-workers, and others would think of his doing so. Descriptive norms are an individual’s beliefs about a given behaviour given the degree to which he sees others engaging (or not engaging) in that behaviour. Using the aforementioned example, an individual may develop beliefs about the favourability of joining a white-power militia if he observes his friends doing so. Taken together, injunctive and descriptive norms form perceived norms, which are characterised by the pressure one feels to engage (or not engage) in a behaviour given his perceptions of others’ experiences and beliefs about that behaviour. Control beliefs and perceived behavioural control  Finally, control beliefs represent a person’s understanding of how likely it is that she will have the time, resources, opportunities, and capability to engage in a specific behaviour. Control beliefs essentially indicate whether a person believes that her performance of a behaviour is likely or not. As in the case of attitudes and perceived norms, perceived behavioural control is also comprised of two kinds of control beliefs. First, the degree to which an individual will be able to perform a behaviour is determined by their capacity to do so. One’s behavioural capacity is their possession of the skills, resources, and opportunity to perform a behaviour. For example, a would-be terrorist’s capacity to carry out a bombing attack is contingent (in part) on their technical ability to build a bomb or their access to those who can do so. The second element of perceived behavioural control is autonomy. Autonomy represents the degree to which an individual believes that the performance of a behaviour is up to them. 154

The pen and the sword BEHAVIOURAL INTENTIONS AND MANIFEST BEHAVIOUR

One’s behavioural intentions indicate their readiness to engage in a behaviour. This readiness is dictated by the three key factors outlined earlier: behavioural attitude, perceived behavioural norms, and perceived behavioural control. Reasoned action theory dictates that the most direct predictor of whether someone will engage in a behaviour is their intention to do so. As such, if an individual (a) has a positive attitude about a behaviour, (b) feels pressure to engage in the behaviour as a function of their perceived norms, and (c) perceives that they have control over engaging in the behaviour, then they are likely to engage in that behaviour. Of course, a person’s intention to engage in a behaviour will not necessarily result in that behaviour being performed. Several environmental factors can preclude a person from engaging in a behaviour they intended to perform. Referring to our earlier example, consider a person who has a positive attitude about joining a white-power militia, perceives normative pressure to join that militia, and understands themselves as having the control to join the militia. Taken together, these three outcomes would lead to an individual’s intention to join the white-power group. However, security forces may have been monitoring the individual’s online activity, leading to his arrest. Despite having all the ingredients necessary for the individual to join the white-power militia group, a key environmental factor (e.g. security intervention) prevented the transformation of the individual’s intentions into manifest behaviour.

Leveraging knowledge of reasoned action processes to inform counter-radicalisation efforts Reasoned action theory dictates that if we can identify message targets’ attitudes, perceived norms, and perceived control over performing a behaviour, we can predict their attitudes, intentions, and performance of that behaviour. It follows that if we can influence the beliefs that underpin those attitudes, perceived norms, and perceptions of control, we can similarly influence these outcomes, ultimately shaping the behaviour of message targets. In this way, reasoned action theory guides us to target different kinds of beliefs that can, in theory, have different implications for whether message targets eventually engage in (or avoid) a desired (or undesired) behaviour. But before a message designer can even consider developing an intervention intended to affect message targets’ beliefs, attitudes, intentions, or behaviours surrounding an extremist ideology, several other steps must be taken to improve the chances of the intervention’s success. The first of these seems obvious but is often overlooked – we must identify the specific behaviour that we hope to promote or suppress. For example, we may be interested in changing an individual’s beliefs about specific violent activities, like the use of weapons to achieve political goals. Note that message designers might also attempt to change beliefs related to more general behavioural categories. These outcomes have less defined boundary conditions and can encapsulate any number of specific behaviours. For instance, whereas messages intended to undermine support for specific behaviours may target beliefs related to building a bomb, more general behaviours can also be targeted – like illegal activity in general. Although research has provided support for targeting both kinds of behaviours with persuasive messaging in the context of reasoned action processes, Fishbein found that interventions based on the first category, changing specific behaviours, are the most likely to be effective. A second step in leveraging reasoned action processes to undermine extremist messaging is to distinguish message targets in terms of their vulnerability to persuasion given the factors associated with their intentions. The theory of reasoned action indicates that a person’s decision to engage in a given behaviour hinges on two factors: (1) they have the intention to perform the behaviour and (2) they have the skills and opportunity to perform the behaviour. This is not to say 155

Kurt Braddock

that an individual will be successful in their endeavour; just that they make a decision to attempt the behaviour. However, in the absence of these factors – intention, skills, and opportunity – the behaviour cannot be performed. So, among those who do not engage in a behaviour, there are two types of message targets – those with the intention but not the skills/opportunity, and those with the skill/opportunity but not the intention. Given that there are two subsets of the target audience, message designers will need to develop different interventions to influence those with intention but not ability and those with ability but not intention. To determine the degree to which the population being targeted is primarily characterised by the former or latter category, message designers should obtain measures of the targets’ beliefs, attitudes, norms, self-efficacy, and intentions in relation to the behaviour in question. Not only will this help us to segment our intended audience, but it will also help us to understand which of the three primary predictors of violent intention is the most potent – attitudes, perceived norms, or perceived behavioural control. In addition to the elements of reasoned action theory that can be used to guide which kinds of ideas to target (i.e. attitudes, norms, or perceived control), some research has identified other concepts that message designers should consider when determining how to persuade message targets to engage in behaviours. First, the targeted belief should be strongly correlated with the behaviour that the message designer wishes to change. Though the message designer should have already chosen to develop a communicative intervention based on the strongest predictor of message targets’ intentions, reasoned action theory indicates that attitudes, perceived norms, and self-efficacy are only indirectly related to behaviour via mediation through intention. To ensure the salience of the belief being targeted, message designers can evaluate the degree to which the belief and behaviour are related. For instance, if a message designer wished to leverage reasoned action processes to motivate nonviolent political action, there should be a salient relationship between engagement in non-violent political action and the achievement of the message targets’ goals. Second, for an intervention to be successful, there need to be enough people in the target audience that do not hold the belief that the intervention promotes. That is, message designers must consider whether the development of a communicative intervention intended to change audience beliefs will affect enough people to make it worth the time, effort, and, in some cases, resources. This is particularly important in the domain of violent extremism, where monetary resources are often restricted to efforts that yield substantial effects. As a final consideration in choosing which belief to target, it is critical that message designers know whether targeted beliefs are even capable of changing. As succinctly argued by Fishbein and Yzer (2003), ‘not all beliefs are equally amenable to change, and relatively little will be accomplished by attacking a belief that is very difficult, if not impossible, to change’. This is largely a subjective evaluation on the part of the message designer, but beliefs based on experience tend to be more resistant to influence than beliefs based on information received via communication from others. In the context of preventing radicalisation to violence, intervention designers should recognise whether message targets’ attitudes, perceived norms, and perceived behavioural control are a function of exposure to propaganda or their lived experiences. If any one of these factors is affected by the latter, it may be more effective to target factors that have been affected by the former.

Discrete emotion theory: affect, radicalisation, and counter-radicalisation Despite a lengthy history of research that has approached decision-making as a purely cognitive and rational process, human experience tells us that some decision-making processes are 156

The pen and the sword

not resolved with cognition alone. Human beings are not passionless computers; they are individuals characterised and moulded by subjective experiences, memories, and biochemistries that can shape how we respond to different kinds of communicative stimuli. In short, human beings can experience emotions, and those emotions can have a substantial impact on how we make decisions. Though not a theory of communication per se, the discrete emotions perspective allows us to understand how the elicitation of certain emotions via communication might motivate message targets to engage (or not engage) in specific behaviours. Just as reasoned action theory shows us how communication might affect cognitions that lead to behavioural decisions, discrete emotions theory demonstrates how communication can lead to similar outcomes via emotion-motivated responses. As defined by the discrete emotions perspective, emotions are physical and psychological reactions to cognitive appraisals. That is, emotions represent mental and biochemical processes that are prompted by environmental factors that help or hinder us in achieving our goals. Although there exist various models for understanding emotions, work by key emotion theorists like Richard Lazarus, Bernice Lazarus, Nico Frijda, Peter Kuipers, and Carroll Izard contends that different emotions can be conceptualised on the basis of different criteria. First, emotions have different qualitative ‘feel’. That is, the experiences associated with different emotions feel subjectively different from each other. Happiness feels slightly different from surprise; anger feels slightly different from jealousy; sadness feels slightly different from fear; and so on. Second, different emotions are associated with different physiological changes in the body. Some stimulating emotions, for example, can trigger an increase in adrenaline output or spike blood pressure. In contrast, more sedated emotions can induce reductions in these physiological metrics, reducing our energy levels. Similar to the physiological changes associated with emotional experience, different emotions also prompt neurological changes. For instance, experiencing anger or fear can increase neurological activity in the brain. Different emotions can also trigger changes in an individual’s cognition by narrowing her perceptual field or focusing attention on the stimulus that prompted the emotional response. Finally, different emotions trigger specific appearances in terms of both facial expressions and body posture. For instance, disgust can often induce a person to ‘scrunch’ up their face and tighten their lips; fear is often associated with a cowering position; and sadness is associated with frowning. Taken together, these five outcomes – subjective feeling, physiological changes, neurological and cognitive responses, and manifest expressions – comprise the experience of that which is called an ‘emotion’. But perhaps the most important element of the discrete emotions perspective is not the collection of cognitive, psychological, and physiological outcomes associated with the experience of an emotion but the behavioural outcome that corresponds with it. Specifically, some researchers (including Lazarus and Izard) have distinguished emotions in terms of the different behavioural motivations they trigger. These motivations, dubbed action tendencies, induce the individual to either approach or avoid the stimulus in question. Table 10.1 summarises common emotions included in the discrete emotions model (Dillard & Peck, 2001, p. 41; Lazarus, 1991; Izard, 1977; Frijda et al., 1989), including both the stimulus that prompts the emotional experience (i.e. appraisal) and the action tendency associated with that experience. With conceptualisation of emotions (per the discrete emotions perspective) complete, it is possible to consider some of the different emotions that are part of the discrete emotions models. 157

Kurt Braddock Table 10.1  Discrete Emotions, Appraisals, and Action Tendencies Emotion

Appraisal

Action Tendency

Anger

Unwarranted obstruction of a goal

Fear

Probability of harm to one’s body

Disgust

Probability of harm to one’s health

Guilt Sadness

Violation of personally held moral Irrevocable failure to achieve a salient goal

Envy Happiness

Recognition that one’s goal (performance or possession of an object) has been achieved by another Acute movement towards a goal

Hope

Change in probability of goal achievement

Pride

Recognition of credit for an achievement by oneself or a group with which one identifies

Approach: Attack, remove, or reject the source of the obstruction Avoidance: Retreat from or acquiesce to a threat Avoidance: Abstain from interacting with or consuming material that can make oneself ill Approach: Redress the moral violation Avoidance: Review plan for continued pursuit of goal; regain strength and resources Approach: Seek to obtain that which rival possesses; dispossess rival of that which he/ she has Approach: Bask in continued success towards a valued goal Approach: Renew and strengthen efforts towards achieving a valued goal Approach: Bask in celebration of completion of goal

Note: Adapted from Dillard and Peck’s Table 1, as well as different works by Lazarus, Izard, and Frijda and Kuipers.

Anger, fear, and radicalisation to violence A consideration of all the emotions outlined in Table 10.1 is beyond the scope of this chapter. However, there are three emotions that are ever-present in extremist messaging and are associated with motivating extremist violence: anger, fear, and pride. Following, this section will turn to how these emotions can be used in counter-radicalisation efforts to undermine attempts to motivate radicalisation to violence. ANGER

Anger is prompted by the recognition that one’s goals have been threatened, prevented, or undermined (Lazarus, 1991; Lazarus & Lazarus, 1994; Reiser, 1999). Upon making this recognition, the experience of anger triggers a physiological response whereby the individual’s blood pressure and muscle capacity increase, thereby rendering the individual more capable of removing barriers that block the path to achieving their goals (Turner, 2007; Izard, 1991; Estés, 1992). Given the physiological and neurological responses associated with anger, it is often associated with negative behavioural tendencies, including aggression and antisocial anger. In addition, the anger response can motivate an individual to attack or otherwise punish the stimulus that caused the anger (van Doorn et al., 2017). In this way, anger has been linked to ‘lashing out’ at those perceived as blocking the pursuit of valued goals. For instance, some research has shown that when a person perceives that another individual is being treated unfairly or victimised, those perceptions can lead to feelings of anger, motivating the individual to attack the source of the unfairness or victimisation (Van Doorn et al., 2017; Raihani & McAuliffe, 2012). 158

The pen and the sword

Given that anger is associated with potent behavioural motivations, some researchers have explored its role in persuasive processes. This work has indicated that the efficacy of persuasive messages designed to elicit anger on the part of message recipients is contingent on three key factors. First, the effectiveness of an anger appeal to motivate action depends on message recipients’ perceptions of their own efficacy. If a message intended to induce anger is to motivate actual action on the part of audiences, those audiences must believe that they are able to engage in a behaviour that will alleviate their anger and that the performance of that behaviour will be effective in doing so. Second, there exists a positive relationship between the intensity of one’s anger and the likelihood that they will react to it. Work by Fein, for example, has shown that when a message evokes a substantial amount of anger, the message recipient is likely to aggress against the object that the individual perceives as responsible for their anger (Fein, 1993; Pfau et al., 2001). If, however, the object evokes only a moderate anger response, the individual is more likely to engage in analytical thinking and problem solving to alleviate their anger rather than attack the object of it (Thomas, 1993; Averill, 1983). Finally, whether a message target will respond favourably to the behaviour advocated for in the anger appeal is dependent (in part) on the target’s prior attitudes about the behaviour in question. Anger appeals will be most effective for those with a pre-existing disposition to agree with the message but who have not yet acted in accordance with it. In this way, when a message elicits anger towards an object and advocates a specific behaviour to resolve that anger, the message can assist in giving someone a ‘nudge’ towards engaging in a behaviour that had previously been deemed too difficult to perform (Turner, 2007, p. 116). Given these characteristics, it is clear how extremists may use anger appeals in their messaging to influence vulnerable message targets. Specifically, extremist groups take advantage of anger’s action tendency to approach the groups’ enemies and engage in actions against them. This is a common propaganda tactic among terrorist groups, which often depict the victimisation of those they claim to represent. On its website, for example, Hamas maintains a repository of news stories relating to Israeli aggression against Palestinian civilians. In one of these stories, Hamas describes a story in which an Israeli soldier shoots a Palestinian and leaves him to bleed in the street. The story goes on to describe how other Israeli soldiers were standing around ignoring the pain-filled cries of the wounded Palestinian until one soldier turned to shoot him in the head, executing him. To further enflame anger that might result from reading this story, Hamas included a picture of the killed Palestinian bleeding in the street. Anger can be a particularly dangerous emotion to contend with, should extremist groups be successful in evoking it. In other contexts, anger’s approach-oriented action tendency has been shown to arouse hostility and engagement in activities that people perceive as self-defence against an aggressor. In the context of political violence, this can translate to support for or engagement in violence against those depicted as the source of the anger. This very phenomenon is often referenced in discussions with former terrorists, who claim that their decision to get involved with terrorism was at least partially influenced by the anger they felt against those they perceived as enemies (Horgan, 2009). FEAR

Fear appeals are persuasive messages that highlight dangers to an individual’s safety if they neglect the recommendations intrinsic to the message. These messages are typically designed to warn target audiences of physical harm to induce them to (a) avoid behaviours that increase the 159

Kurt Braddock

chances that they will avoid the physical harm that was warned about or (b) proactively engage in behaviours that reduce this likelihood (Tannenbaum et al., 2015; Dillard et al., 1996; Meczkowski & Dillard, 2017). To promote (or prevent) specific behaviours, fear appeals are typically designed to elicit three responses: the aforementioned fear, perceived threat, and perceived efficacy. First, fear appeals are designed to trigger message recipients in a manner consistent with their recognition that they may experience harm. Second, fear appeals should amplify message targets’ perceptions of the threat posed by the fear-inducing stimulus. In other words, effective fear appeals highlight message recipients’ susceptibility to the harms described in the message and/or the severity of that harm should they fail to follow the recommendations embedded in the message. Third, fear appeals should contain language designed to increase recipients’ perceptions that they have the capacity to engage in (or avoid) behaviours that reduce their chances of experiencing the harms described in the message (Witte & Allen, 2000; Witte, 1998). In a meta-analysis of fear appeals and their persuasive efficacy, Tannenbaum and her colleagues analysed dozens of studies populated by thousands of participants. The researchers concluded that fear appeals are potent tools for influencing audiences’ attitudes, intentions, and behaviours. Moreover, they found that this persuasive effect of fear appeals is largely consistent across studies (Tannenbaum et al., 2015). The efficacy of fear appeals identified by Tannenbaum and colleagues has not been lost on extremist groups attempting to influence their audiences. One of the fundamental objectives of their communication is to promote and spread fear among as many people as possible to facilitate the achievement of political goals. This typically occurs in one of two ways: First, extremist groups can arouse fear in relation to those they identify as their enemies. The avoidance action tendency associated with fear can create the perception that those individuals or organisations (i.e. the ‘enemies’) are dangerous and are worth being attacked (or at least avoided). By arousing fear about their supposed enemies, extremist groups can also cast themselves as the only ones capable of protecting target audiences from these enemies. This characterisation, in turn, justifies the use of violence against these ‘enemies’ by the group. Although terrorist groups often weaponise fear to characterise their enemies in a certain way, they can also use fear to illustrate their willingness to engage in violence (as a form of punishment) if their demands are not met. In doing so, extremist groups seek to activate the avoidance tendency associated with fear by inducing target audiences to avoid potential harm by submitting to the group’s demands (Garrison, 2004). Regardless of the manner in which an extremist group seeks to elicit fear among target audiences, the avoidance-orientation associated with fear’s action tendency exerts a powerful force on the target audience’s responses. Specifically, by inducing fear about carefully selected targets that the group casts as enemies or characterising themselves as a dangerous entity to be respected and obeyed, extremist groups leverage humans’ natural tendency to avoid harm by aligning with their ideology or acquiescing to their demands, respectively. To be sure, extremist groups are adept at manipulating the emotions of their audiences, particularly with respect to anger and fear. However, as outlined earlier, there are other emotions – each with their own action tendencies – that can be elicited among target audiences to undermine the persuasive attempts of extremist groups. Although all the emotions in the discrete emotions perspective can be leveraged in this manner to some degree, two key emotions that can be particularly useful in this regard are hope and pride. In the next section, these two emotions are explored, and their persuasive potential as part of counter-radicalisation efforts is examined. 160

The pen and the sword

Leveraging hope and pride in counter-radicalisation efforts HOPE

Hope is intrinsically connected to one’s perceptions about the likelihood of achieving a valued goal (MacInnis & de Mello, 2005; Chadwick, 2015). Specifically, hope is aroused by one’s recognition that their goals are possible to achieve through specific actions but are not a forgone conclusion. When individuals feel hopeful, they seek to leverage opportunities they perceive as helpful in achieving their valued goals. In this vein, message designers that aim to elicit hope as an emotion meant to promote counter-radicalisation should produce and deliver messages from which audience members derive motivation to achieve goals that run counter to the ideologies and actions of extremist groups. Given that the persuasive value of hope is dependent on audiences’ continued pursuit of valued goals, the development of hope appeals (in any context) first requires that message designers identify target audiences’ valued goals. Once these goals are identified, it is then necessary to identify specific behaviours that represent opportunities for audience members to achieve these goals. By (a) identifying audience goals that run contrary to terrorist ideologies and objectives and (b) recommending behaviours that can help them achieve those goals, it becomes possible to elicit hope such that individuals who desire outcomes that are contrary to terrorist groups’ objectives will be motivated to engage in suggested behaviours to achieve those ends. For instance, consider young, white individuals targeted for recruitment by right-wing extremist organisations. Many individuals targeted for recruitment into these groups have often been found to lack a sense of identity or belonging (Horgan, 2014). Extremist groups exploit these issues by eliciting fear and anger towards other groups on which they can place blame. In this way, right-wing extremists weaponise these individuals’ failure to achieve important social goals (i.e. identity, purpose, and belonging) to draw them towards their groups. To counter the persuasive efficacy of these kinds of messages, it would be useful to offer message recipients hope about achieving these goals without having to join extremist groups to do so. Many counter-radicalisation efforts have sought to do so by providing those targeted by extremist groups with other outlets for their discontent, including more peaceful political expression and the chance to engage with those with whom they have built friendships and a collective identity. By providing these individuals with the impression that they can resolve their problems and achieve their social goals without resorting to membership in an extremist group, hope appeals can exert a powerful persuasive force on those vulnerable individuals who might otherwise be led astray.

PRIDE

Whereas the persuasive power of hope is related to the degree to which an individual perceives that s/he might pursue and ultimately achieve a desired goal, there is another persuasive emotion that motivates action on the basis of whether that goal is completed. This emotion, pride, has an approach orientation such that individuals are motivated to bask in the accomplishment of one’s goals (Leach et al., 2002). The true persuasive power of pride in the context of counter-radicalisation, however, resides in the fact that individuals can experience vicarious pride as well. Specifically, achievement-induced pride does not result only from the completion of goals that one has set for him/herself but also from goals achieved by those with whom individuals consider themselves affiliated. Extremist groups often attempt to elicit pride among target audiences by lionising the group’s actions, particularly if audience members identify with that group in some way 161

Kurt Braddock

(e.g. culturally and ideologically). By stressing (a) the affiliation between the audience member and the group and (b) the goals that the group has achieved, extremist groups seek to elicit helping behaviours on the part of the audience members resulting from the pride those individuals feel. That said, even as extremist groups can exploit feelings of affiliation and the resultant pride to motivate behaviour consistent with the group’s goals, it can also be elicited by counter-radicalisation messages to prevent such behaviours. Similar to how hope and its link to alternative forms of political expression and affiliation can undermine the persuasive power of extremist messaging, highlighting audience affiliations with non-violent entities can be an effective method for challenging extremist messaging through the elicitation of pride as well.

Closing remarks It is impossible to overstate the extent to which communication influences radicalisation and counter-radicalisation processes. Regardless of the nature of how an extremist group or counterextremist organisation seeks to influence their audiences, communication is at the centre of these efforts. Still, just as the magnitude of communication’s importance is impossible to overstate, so too is it impossible for a single volume, let alone a single chapter, to capture all the perspectives that might explain how communication influences persuasive processes in the context of radicalisation and extremism. This chapter provides only small portions of cognitive and emotional frameworks that help explain how and why certain messages resonate with audiences targeted by extremist messaging and counter-radicalisation efforts. A complete understanding of how different forms of communication affect salient audiences psychologically, cognitively, emotionally, and personally requires a full survey of the literature on persuasion and social influence. Nevertheless, it is the author’s hope that this chapter has provided an introduction to these issues for those who wish to use communication as a tool for challenging extremist groups. Specifically, a consideration of the target audience’s goals, the manner in which extremist groups target them, and the mechanisms by which messages persuade them can assist in developing messages that might steer those audiences away from beliefs, attitudes, and behaviours that could lead them to hurt themselves and others.

References Averill, J. R. (1983). Studies on anger and aggression: Implications for theories of emotion. American Psychologist, 38(11), 1145–1160. Braddock, K. (2020). Weaponized words: The strategic role of persuasion in violent radicalization and counterradicalization. Cambridge University Press. Chadwick, A. E. (2015). Toward a theory of persuasive hope: Effects of cognitive appraisals, hope appeals, and hope in the context of climate change. Health Communication, 30, 598–611. Dillard, J. P., & Peck, E. (2001). Persuasion and the structure of affect: Dual systems and discrete emotions as complementary models. Human Communication Research, 27(1), 38–68. https://doi.org/10.1093/ hcr/27.1.38 Dillard, J. P., Plotnick, C. A., Godbold, L. C., Freimuth, V. S., & Edgar, T. (1996). The multiple affective outcomes of AIDS PSAs: Fear appeals do more than scare people. Communication Research, 23(1), 44–72. Estés, C. P. (1992). Women who run with the wolves (1st ed.). Ballantine Books. Fein, M. (1993). I. A. M.: A common sense guide to coping with anger. Praeger. Fishbein, M., & Ajzen, I. (2010). Predicting and changing behavior: The reasoned action approach. Routledge. Fishbein, M., & Yzer, M. C. (2003). Using theory to design effective health behavior interventions. Communication Theory, 13(2), 164–183. https://doi.org/10.1111/j.1468-2885.2003.tb00287.x

162

The pen and the sword Frijda, N. H., Kuipers, P., & ter Schure, E. (1989). Relations among emotion, appraisal, and emotional action readiness. Journal of Personality and Social Psychology, 57(2), 212–228. https://doi.org/10.1037/ 0022-3514.57.2.212 Garrison, A. H. (2004). Defining terrorism: Philosophy of the bomb, propaganda by deed and change through fear and violence. Criminal Justice Studies, 17(3), 259–279. https://doi.org/10.1080/1478601042000281105 Horgan, J. (2009). Walking away from terrorism: Accounts of disengagement from radical and extremist movements. Routledge. Horgan, J. (2014). The psychology of terrorism. Routledge. Izard, C. E. (1977). Human emotions. Springer. Izard, C. E. (1991). The psychology of emotions. Basil Blackwell. MacInnis, D. J., & de Mello, G. E. (2005). The concept of hope and its relevance to product evaluation and choice. Journal of Marketing, 69(1), 1–14. Meczkowski, E. J., & Dillard, J. P. (2017). Fear appeals in strategic communication. In P. Rössler (Ed.), The international encyclopedia of media effects. John Wiley and Sons. Lazarus, R. S. (1991). Emotion and adaptation. Oxford University Press. Lazarus, R. S., & Lazarus, B. N. (1994). Passion and reason: Making sense of our emotions. Oxford University Press. Leach, C. W., Snider, N., & Iyer, A. (2002). “Poisoning the consciences of the fortunate”: The experience of relative advantage and support for social equality. In I. Walker & H. J. Smith (Eds.), Relative deprivation: Specification, development, and integration (pp. 136–163). Cambridge University Press. Pfau, M., Szabo, A., Anderson, J., Morrill, J., Zubric, J., & H-Wan, H.-H. (2001). The role and impact of affect in the process of resistance to persuasion. Human Communication Research, 27, 216–252. Raihani, N. J., & McAuliffe, K. (2012). Human punishment is motivated by inequity aversion, not a desire for reciprocity. Biology Letters, 8(5), 802–804. Reiser, C. (1999). Reflections on anger: Women and men in a changing society. Praeger. Tannenbaum, M. B., Hepler, J., Zimmerman, R. S., Saul, L., Jacobs, S., Wilson, K., & Albarracín, D. (2015). Appealing to fear: A  meta-analysis of fear appeal effectiveness and theories. Psychological Bulletin, 161(6), 1178–1204. Thomas, S. P. (1993). Anger and its manifestation in women. In S. P. Thomas (Ed.), Women and anger, vol. 15: Focus on women (pp. 40–67). Springer. Turner, M. M. (2007). Using emotion in risk communication: The anger activism model. Public Relations Review, 33(2007), 114–119. van Doorn, J., Zeelenberg, M., & Breugelmans, S. M. (2017). The impact of anger on donations to victims. International Review of Victimology, 23(3), 303–312. Witte, K. (1998). Fear as motivator, fear as inhibitor: Using the EPPM to explain fear appeal successes and failures. In P. Andersen & L. Guerrero (Eds.), Handbook of communication and emotion (pp. 423–450). Academic Press. Witte, K., & Allen, M. (2000). A meta-analysis of fear appeals: Implications for effective public health campaigns. Health Education & Behavior, 27(5), 591–615.

163

11 GENDER PERSPECTIVES ON RADICALISATION1 Elizabeth Pearson

Introduction Radicalisation is a contested term and has been widely critiqued by scholars and communities. Understood by the United Kingdom government as the ‘process by which people become involved in terrorism or extremism leading to terrorism’ (Home Office, 2011), it is also widely acknowledged in the literature that there is no one radicalisation process (King & Taylor, 2011; Neumann, 2013), and radicalisation can hardly even be labelled a ‘process’ at all (Mandel, 2009). Radicalisation is not just theoretically problematic. It can be politically harmful. Use of the term expanded following 9/11, and its use has stigmatised and even criminalised Muslim communities in the global north (Awan, 2012; Brown, 2008; Kundnani, 2014; Rashid, 2014), and become an instrument of neo-colonialism in the global south (Brown, 2010; Pearson et al., 2020; Shepherd, 2006). Additionally, radicalisation lacks boundaries of scope. It is used in two ways: cognitive (thought) and behavioural (actions) (Neumann, 2013), often interchangeably. Still, the term ‘radicalisation’ remains the analytic paradigm used within government and terrorism studies to explain adherence to jihadist groups (Malthaner, 2017) and, increasingly, the far right, the two key security concerns of western governments (Pearson et al., 2020). This chapter therefore engages with the term radicalisation as useful in that it can be understood to mean the pathways towards support or activity for politically extreme or violent groups. However, the chapter also contributes to the wider literature contesting the term and the theoretical models underpinning it (Horgan, 2012; Sedgwick, 2010). What about gender? Gender is a way of understanding power and how it is distributed and has been historically neglected in the literature on radicalisation. Gender is socially constructed, describing ‘the socially constituted behavioural expectations, stereotypes and rules that construct masculinity and femininity’ (Sjoberg, Cook and Neal, 2011, p. 6). While mainstream radicalisation theories have made little mention of gender, gender is nonetheless directly relevant. Indeed, feminist critiques have outlined how radicalisation is gendered (Pearson et al., 2020). First, the population of contemporary violent extremists that radicalisation describes is predominantly – but by no means exclusively – made up of men (Gentry, 2022; Sjoberg, 2013). Second, gender is relevant to the effects of radicalisation policy. The term radicalisation has enabled gendered counter-radicalisation practices against the violent Jihad, such as the securitisation DOI: 10.4324/9781003035848-13 164

Gender perspectives on radicalisation

of Muslim men and the co-optation of Muslim women as allies in peace (Brown, 2008; Rashid, 2014). Policy makers understood a particular demographic, Muslim men, as particularly vulnerable to Jihadist radicalisation. As a result, entire communities have been stigmatised, despite the tiny incidence of either extreme or violent actors relative to general populations. This approach had the gendered effect of both securitising peaceful Muslim men and failing to recognise the possibility of a small minority of violent women. The latter have been characterised as lacking agency, with the media, for instance, terming them ‘jihadi brides’ and ‘brainwashed’ into travel to Syria by predatory men (Jackson, 2021). Third, international security norms and legal statutes point to the importance of gendered approaches in understanding and countering terrorism. The United Nations Security Council has produced ten resolutions as part of the Women, Peace and Security (WPS) agenda. In 2015, UNSCR 2242 addressed the importance of a gender perspective in dealing with terrorism. This resolution recommended the inclusion of gender as a ‘cross-cutting issue’ in counter-terrorism activities (see sections 11–3 of UNSCR 2242, 2015). Then in 2019, UNSCR 2467 discussed sexual violence by terrorist groups as ideological and strategic. Finally, gender is relevant because it organises the expression of societal power, producing hierarchies through masculinities and femininities (Connell, 2005). Terrorism and radicalisation also concern power and violence; one definition of terrorism is the use of violence against symbolic targets to send a message to achieve an ideological goal (Richards, 2014). Terrorism therefore aims to change the status quo, whether by states to further cement power or by non-state actors who seek power. Gender analysis is therefore relevant to understanding how power operates within radicalisation to masculinist groups seeking to elevate (particular) men’s status and rights. And yet, to date, gender has either been missing in mainstream theory or apparent as a synonym for ‘women’, as this chapter will outline. It is not that there is no work on gender and radicalisation: there is much. It is that this work has yet to be integrated into mainstream approaches to radicalisation, whether for jihadist or far-right groups or any other emergent terrorist threat.2 Much of the gender work on radicalisation has in fact focused on women, a historic subject of neglect in security studies and relegated to the margins of study and policy (Enloe, 2001, p. 6; OSCE, 2011; Parpart & Zalewski, 2008, p. 1; Tickner, 1997, p. 611). Such was the extent of this neglect that the travel of some 5,000 women from Western Europe to join the so-called Islamic State (IS) took policymakers by surprise (Cook & Vale, 2018). Radicalisation theories had offered little to account for this, given their focus on men’s recruitment journeys to extreme or terrorist groups. Nor did these theories reflect on the reasons for men’s dominance in the numbers joining IS or historic terrorist organisations; theory took men’s power and violence for granted. The outcome has been, as Gentry (2022) states, ‘continual surprise’ within the terrorism studies mainstream that women are ever actively involved in terrorism or that misogyny counts as political ideology. However, as Gentry (2022) warns, ‘surprise is no longer an option and only reveals ignorance’ (p. 221). Ignorance is not harmless. Ignorance creates a policy blind spot with tangible outcomes: women giving birth in warzones to children who have no citizenship; men lacking the resilience to resist violent rhetoric offering them masculine status; an incomplete understanding of how terrorism works. To understand terrorist organisations holistically – their ideology, violence, mobilisation tactics and propaganda – it is necessary to engage not just with gender as the fact of women’s violence but also with gender as an approach. This approach recognises radicalisation to violent organisations as situated within social structures within which gendered interaction provides the ‘interactional scaffolding’ (West & Zimmerman, 1987, p. 147). It is not enough to ‘add’ gender as women’s activity to existing theory; a shift is required in how we conceptualise radicalisation and gender. Gender constructs pathways to extremist activism and, for some, to violence. Using gender reveals 165

Elizabeth Pearson

how radical identities are constituted within power structures (Jensen & Larsen, 2019). This chapter begins with an assessment of the key categories appearing in the dominant models of radicalisation in the terrorism studies literature, emphasising the absence of gender. Gender becomes an aside, considered only in the subfield of literature on women’s radicalisation, a synonym for women. The chapter then goes on to explore the ways in which a gendered approach can produce a deeper understanding of the key features of radicalisation identified, advocating gender as a framework going forward. An intersectional approach is important: race, sexuality, creed, and ethnicity are all invoked in extreme narratives. Intersectional approaches understand that identities are not fixed, nor are power hierarchies (Boesten, 2014; Yuval-Davis, 2016). However, the emphasis in this chapter is on gender, which is at the forefront of understanding contemporary radicalisation. This is because contemporary extreme movements, such as the far-right or violent jihadists, are masculinist and organised according to gendered binaries in which men’s and women’s roles are regarded as opposite and complementary, as the section “Gendering radicalisation: an integrated approach” will explore.

Radicalisation theory and the absence of gender There is no one theory of radicalisation in the mainstream literature. Attempts to understand the processes behind people’s journeys into violent and extreme groups have drawn on a variety of disciplines and methodologies, and a range of empirical data. What is described as theory is in reality a series of models. Researchers have taken different approaches (Dalgaard-Nielsen, 2010; Neumann & Kleinmann, 2013; Sageman, 2014; Silke, 2003). They have analysed the life histories of people convicted of terrorism (Bakker, 2006; Sageman, 2004); they have talked to radicals and extremists (Blee, 2007; Busher, 2015; Kenney, 2018; Kimmel, 2018; Pilkington, 2016; Weeks, 2020; Wiktorowicz, 2005); they have talked to those who know them (Pearson & Winterbotham, 2017); they have developed psychological models (McCauley & Moskalenko, 2008; Moghaddam, 2005). They have created typologies to describe the roles of the radicalised (Nesser, 2006; Pantucci, 2011; Venhaus, 2010). Despite the variety of approaches and different emphasis of the models, reviews of the literature reveal that radicalisation theories share key themes. One is the presence of extreme ideology, although the importance of this in the radicalisation process is disputed (Sageman, 2004, 2008). Sageman believes a social group forms first, later finding a radical ideology; other authors believe ideology comes first. Radical ideology can be understood as a mobilising aspect of a group, ‘pulling in’ recruits (Dawson, 2021). Another shared theme across models, therefore, is the recognition of so-called push and pull factors (Khalil & Zeuthen, 2016). Push factors are understood as situational and external to the individual; they are often structural, such as poverty or discrimination. Pull factors are those that draw an individual into a group – adventure, brotherhood, or the chance of money or a job, for instance. However, the ‘push and pull’ factor model is conceptually underdeveloped, and these factors overlap in ways that mean the distinction between push and pull is not clear (Cherney et al., 2021). Another shared theme across models is the absence of gender (Jensen & Larsen, 2019; Kimmel, 2018). For instance, radicalisation models engage with the physical and social environments within which individual radicalisation occurs without putting emphasis on the importance of gender within them. They have noted the impact of societal factors, geographical spaces, and locations. For those focused on the macro-conditions of radicalisation, what is termed ‘root’ causes, emergent factors can include poverty, or more specifically, relative deprivation: a perception of poverty relative to expected status or the status of others. These causes are read as structural 166

Gender perspectives on radicalisation

‘strains’, against which individual agency is exercised (Cottee, 2011; Smelser, 2010). Societal challenges become ‘grievances’, pushing an individual towards radicalisation. In the aftermath of 9/11, the literature emphasised particular spaces as ‘locations of vulnerability’ to radicalisation in which grievances were exploited by radical influencers, preachers, and ideologues. Among these were spaces that have particular gender dynamics: mosques, gyms, or prisons (Pearson, 2015); alongside were spaces where the gender dynamics are less obvious but still important, such as universities or the Internet (Bermingham et al., 2009; Nesser, 2016; Neumann, 2010; Quilliam, 2010) and necessitating restrictive measures to remove and suppress hateful narratives (HM Government, 2020; Waddell, 2016). Radicalisation models have considered the importance of the people around the radicalised individual, the ‘specific, immediate social environment’, which Malthaner and Waldmann (2014) term the ‘radical milieu’. They have tended not to consider gender within this (Jensen & Larsen, 2019). Social movement theory (SMT) has much to contribute to radicalisation theory (della Porta & Haupt, 2012; Malthaner, 2017; della Porta & Diani, 2005). It has provided key concepts, such as the need for extreme group messaging to resonate with an intended audience and for violent actors to emerge from a wider social milieu. Radicalisation models thus suggest emergence over time and as part of a process, using metaphors such as the pyramid, conveyor belt, or staircase to represent ideological progression alongside a narrowing demographic group (Dalgaard-Nielsen, 2010; Moghaddam, 2005). SMT has provided the theoretical basis for several empirical studies of radicalisation, including Wiktorowicz’s (2005) seminal work on the banned UK Islamist group al-Muhajiroun. In a review of the literature more than ten years ago, Dalgaard-Nielsen noted the social aspect of radicalisation as a shared theme in the various inter-disciplinary models: sociological, social network/movement, and empiricist. Their conclusions cohere in their emphasis on the role of collective and group dynamics (Dalgaard-Nielsen, 2010), yet without discussion of the importance of the gender dynamics of these groups. In other research from this period, McCauley and Moskalenko also emphasised the role of the collective without explicit consideration of gender. They identified 12 distinct mechanisms of radicalisation, ten of which are collective (McCauley & Moskalenko, 2008). Indeed, one emphasis has been on the emergence of radicalisation within discrete clusters centred around friendship and kinship groups (Rogers & Neumann, 2007; Gendron, 2016; Sageman, 2004; Wiktorowicz, 2005). In her analysis of UK jihadist radicalisation, Klausen (2019) observes that radicalisation occurs within discrete locations and among clusters of people already well known to one another. Similarly, in his seminal so-called ‘bunch of guys’ model, Sageman (2004) suggested that men’s friendship group formation was a fundamental precursor to the adoption of radical ideology. Such work has nonetheless tended to neglect relevant feminist theory. Researchers have employed empirical data from almost entirely homosocial datasets of convictions for violent terrorist incidents, for instance (Bakker, 2006; Gartenstein-Ross & Grossman, 2009; Nesser, 2010; Sageman, 2004; Venhaus, 2010). The theory derived from this data is therefore inevitably gendered, given that most convictions for violence considered involve men, yet the authors do not reflect in depth on gender in their findings. Despite recognition of the social and structural drivers of radicalisation, theory repeatedly returns to the individual, the rational actor, subject to the cognitive transformation towards violence that characterises radicalisation (Malthaner, 2017). In the discussion of Jihadist radicalisation this actor is both male and young. Roy notes the prevalence of ‘unmoored youth’, a ‘lost generation’ (Roy, 2007, pp. 55–56). This actor is also understood as vulnerable, enabling counter-radicalisation strategies that focus on safeguarding and the development of ‘thinking skills’, even though critical thinking is what leads some to extremism (Wiktorowicz, 2005). Not simply vulnerable, the individual is also in crisis, a so-called identity crisis, a salient feature of 167

Elizabeth Pearson

several models of radicalisation (King & Taylor, 2011). This precipitates a ‘cognitive opening’, which increases receptivity to the extreme ideology at radicalisation’s heart. Neumann (2008) concludes, therefore, that radicalisation concerns grievance, ideology, and mobilisation. The British government concurs, suggesting the ‘vulnerable individual’ confronted with an ‘extremist ideology’ and a ‘radicalising influencer’ might be radicalised (HM Government, 2015, p. 21). Both acknowledge the individuals discussed are largely men, but this is not further reflected upon. Instead, gender becomes a synonym for ‘women’, and the academic work on women’s radicalisation is separated from that on men’s, as the next section outlines.

Add women and stir As discussed, mainstream radicalisation theories have tended to assume most radicals are men and gender applies only to women. Theories have been slow to consider gender in men’s behaviours, or ideology, and have failed to read misogyny as ideological (Gentry, 2022). They have also exceptionalised women’s extremism and support for violence. This exceptionalism has relegated the study of women to the margins of the literature, separated from and, by implication, irrelevant to the wider field. To add women to the gender-blind radicalisation terrain is to read gender empirically (Peterson, 2005) and has two possible outcomes: the introduction of a new category, biological sex, or a new category in the field, women’s violence (Pearson et al., 2020). Consistent with International Relations (IR) more broadly, men’s violence is not framed as a function of gender, and women’s participation and violence are hardly seen at all (OSCE, 2011; Parpart & Zalewski, 2008; Enloe, 2001; Tickner, 1997). When terrorism studies did engage with prominent women involved in violent groups, for instance in left-wing groups prevalent in the 1970s, such as the Red Army Faction, explanations often relied on stereotypes of deviant emotional women persuaded into terrorism by violent men (Jacques & Taylor, 2009; Gentry & Sjoberg, 2007). The neglect of gender was not about failing to see women but seeing them only in particular ways. It was about the inability of the epistemological framework of mainstream/malestream theory to admit that women’s violence could be political. This same framework was also unable to read men’s terrorist violence as gendered, or misogyny in group ideology as important or political (Gentry, 2022). To look for gender in radicalisation theory and terrorism studies meant, for a long time, therefore, a diversion from the mainstream texts and into – the predominantly – women’s literature on women. This is vitally important work, drawing on feminist IR scholarship to address the lack of attention paid to women in conflict. It draws attention to women’s experiences, voices, violence, and the impacts of war and extremism on their lives and bodies. Yet much of this work – necessary to a complete picture of radicalisation – engages, again, with gender as a synonym for women rather than adapting a gendered approach. Therefore, one key focus in the literature on female terrorism has been women’s agency. Security studies literature has routinely neglected agency where this applies to women, framing them as victims upon whom external forces operate or as appendages to male actors, wives, or girlfriends, ignored and minimised (Brown, 2013, 2020; Clegg, 2006; Cook & Vale, 2018; Mohanty & Mohanty, 1990). In an early review of the literature on women’s terrorism, Jacques and Taylor found a focus on women’s agency (Jacques & Taylor, 2009, p. 505). This is evident in work focused on particular case studies or to make a broader point about the political nature of women’s engagement, which has historically been ignored (d’Estaing, 2017; Margolin, 2018; Poloni-Staudinger & Ortbals, 2012). Much work on women and terrorism has emerged since 2014, and the visibility of young women recruited on a global scale to Islamic State (IS) (see Cook & Vale, 2018 for a sense of the scale of recruitment). The 168

Gender perspectives on radicalisation

recruitment of women to IS revealed an international blind spot in policy on countering jihadist groups (UNSCR 2242). Scholars of women and terrorism have succeeded in drawing attention to what mainstream theory had minimised and ignored: the gendered complexities of women’s activism, their participation in and support for violent Salafi-Jihadi groups or the far right, for instance (Blee, 2008; Bloom, 2011; Brown, 2013; Davis, 2006, 2013; Eggert, 2015; Mattheis, 2018; Parashar, 2011; Pilkington, 2017; Von Knop, 2006). Authors have demolished the stereotypes of passivity and lack of agency. They have demonstrated women’s motivation and mobilisation, their fundraising and propagandising, their importance in ensuring the continuation of conflict and grievance, raising children within Jihadist ideologies, or as mothers committed to the supremacy of the white race (Mattheis & Winter, 2019; Sageman et al., 2010; Von Knop, 2007; Wiktorowicz, 2005). They have detailed the ways in which IS propaganda seeks to mobilise women (Chatterjee, 2016; Cook & Vale, 2018; Ingram, 2017; Klausen, 2015; Lehane et al., 2018; Pearson, 2018b; Winter, 2015; Winter & Margolin, 2017), and how women in turn use online spaces to express their support (Huey et al., 2017; Huey & Witmer, 2016; Klausen, 2015; Scrivens, 2018). Within terrorism studies, such work has remained an aside, not yet mainstream, even when the authors are respected professors consulting with the United Nations. Authors are often still regarded as concerned with ‘women’s things’, a sideshow, not the main event. It is then easy for mainstream authors to regard this marginalised literature as the only necessary location for work on gender; gender remains a synonym for women with no relevance to men. Gender is read as sex, not as a social construct. While this work adds much to an understanding of terrorism, it cannot therefore progress a gendered analysis of radicalisation pathways as it is the epistemological twin of mainstream literature, which only sees men. The next section explores an alternative. It considers a growing body of work beginning to integrate these distinct bodies of understanding through a gendered analysis. This provides a more holistic account of radicalisation, providing answers to the questions of how and why. It also recognises gender as power and the key organising principle of contemporary violent groups and movements: the far right, jihadists, and misogynists.

Gendering radicalisation: an integrated approach The previous section revealed gender as a blind spot in radicalisation theory. Women’s radicalisation has been historically misunderstood. Later, the work of feminist scholars was relegated to a sub-category, with gender as a synonym for sex, further read as women’s sex: an outlier variable in the radicalisation equation (Pearson et al., 2020). Men have been regarded as gender neutral, leaving the relevance of gender in their radicalisation journeys unconsidered. However, as the field has evolved, alternative approaches have emerged, based on feminist and men’s studies in IR and criminology, and theoretically integrating gender and race. Reading gender not as sex but as a ‘meaning system’ (Peterson, 2005, p. 499), they demonstrate the co-construction of group and individual – men and women understood together – through gendered norms, values, and beliefs. This meaning system shapes radical masculinities and femininities, actions, and thoughts, and is in turn shaped by them. Additionally, it is also situated, gender embedded in context, and space (Miller, 2002).

Gender as fundamental organising principle of violent groups Work engaging gender, not sex, does not have to be complex. It involves gathering the same data and analysing the same materials, yet with a different approach to their analysis. Its starting point is an epistemological position that understands gender as socially constructed, regulating social 169

Elizabeth Pearson

organisation. As Scott (1986) writes, gender is ‘a way of referring to the social organization of the relationship between the sexes’ (p. 1053). Criminological and sociological approaches to extremism and terrorism have frequently employed this definition, using a gendered lens. They have built on the work of masculinities scholars, who have explicitly engaged with men’s violence, as men (Jensen & Larsen, 2019; Kimmel, 2007; Messerschmidt & Rohde, 2018). Connell’s work on gender has been important in drawing attention to masculinities within the gendered hierarchies structuring social relations. Furthermore, women are not absent in this approach, as ‘patterns of masculinity are socially defined in contradistinction from some model (whether real or imaginary) of femininity . . . focusing only on the activities of men occludes the practices of women in the construction of gender among men’ (Connell & Messerschmidt, 2005, p. 848). Since Connell’s earlier work, the emphasis is on the plurality of masculinities, the various possibilities for being a man, and the ways in which women can also perform masculine identities (Connell & Messerschmidt, 2005). To think about radicalisation from this gendered perspective provides fresh insights into existing work, contributing the next analytical step in theorising radicalisation, and a solution to the bifurcation of the literature into radicalisation theory (men) and work explicitly on women. Additionally, a gendered approach furnishes the field with a framework within which agent – man or woman – and structure are understood as co-constitutive aspects of one dynamic system (Jensen & Larsen, 2019; Miller, 2002). The first advantage of such an approach is that it renders men visible within violent or extreme organisations as men, responding to gendered structural pressures against them and shaping those structures in turn (Jensen & Larsen, 2019; Miller, 2002; Pearson, 2022). Location matters here – as mainstream theory identified – as gendered power is situated in location. For Connell (1995), masculinities are about ‘the practices through which men and women engage that place in gender, and the effects of these practices in bodily experiences, personality and culture’ (p. 71). How and where men seek and find power matters. A gendered approach means men’s radicalisation as men is no longer taken for granted and unexamined simply because they are over-represented in crimes of violence in contemporary far-right or Jihadist groups (Blee, 2008; Pearson, 2018a; Pilkington, 2017; Sjoberg, 2013). For instance, masculinities authors Kimmel and Messerschmidt have framed radicalisation as an attempt at the creation of a compensatory identity in response to the reduction of status that mainstream theory associates with grievance or deprivation. Using a gendered approach, radicalisation is understood as a series of choices between hierarchies of masculinities that enable a particular status within gendered social structures. Hegemonic masculinities are understood within social systems as most high-status and desirable, but they are only accessible to particular men (Connell & Messerschmidt, 2005). Terrorism is not exceptional; it is a form of crime. A gendered approach understands crime as a resource to do gender (Messerschmidt, 1993). Kimmel and Connell have also explored men’s radicalisation as a form of ‘protest’ masculinity. Radicalisation then constitutes a gendered response to a disruption to social status through perceived emasculation. Emasculation might entail changes to traditional gender relations and men’s power to fulfil breadwinner roles as heads of households. Protest masculinities are a resource enabling men to reassert their influence and find status, even as they believe changes to structural power and long-held gender norms destabilise them (Connell, 2004, 2011; Kimmel, 2018; Kimmel & Ferber, 2000; Kimmel, 2003). They are exploited by and constituted within extremist groups. For instance, IS has pushed back against social gendered change, which has left women ‘not fulfilling their fundamental roles’ in the home because ‘of the rise in the number of emasculated men’ (Winter, 2015, p. 17). Incel groups similarly lament their own – self-perceived – lack of masculinity when 170

Gender perspectives on radicalisation

compared with the hegemonic masculinity they find evident in those men they believe can exercise sexual choice, so-called Chads (Leidig, 2021). Violence, misogyny, and the performance of ‘warrior’ masculinities are understood as a compensatory resource. Existing work, even within mainstream literature, has recognised the significance of the term manhood when analysing contemporary violent groups and movements. Hegghammer (2017) has even discussed the performance of a range of masculinities in Jihadi culture, from the scholarly and the emotional weeping fighter to the warrior. Yet mainstream authors have tended to neglect references to gender literature (see, for instance, Awan, 2014; Sageman, 2004, 2008). Sageman (2008) has referenced the macho ‘rock star’ of jihad and pointed to a ‘masculine conception of heroism among terrorists’ (p. 111). Another key author on radicalisation, Roy (2017), has noted that the macho culture of violent Jihad enables ‘male losers’ to become the ‘avenging hero’ (pp. 45, 51–52). Yet he does not take the next analytical step, using feminist or criminological gender theory to link emasculation to an understanding of masculinities, hierarchies of power, structural systems, misogyny, or existing masculinities. Without reference to the wider gender literature on power and masculinities, such observations dangle, and the role of gender in constituting homosocial socialisation and radicalisation is neglected, unexplored. The second advantage of gendered approaches is that they unify understanding of men’s and women’s participation in political violence and extremism; men and women are now understood as subject to the same macro-forces (Kimmel, 2018). Such an approach sees men and women not as distinct objects of radicalisation, but as situated within shared ‘hierarchies of structural position’ (Miller, 2002, p.  439). Gendered oppositions and binaries constitute an organising principle of the hierarchies of contemporary terrorism currently prioritised by western states – jihadist and far right – the backbone of both belief and action. Gender is therefore vital in understanding the ‘pull’ factors of extreme organisations for both men and women. Both Jihadist and far-right organisations, particularly those advocating violence, tend to position men’s and women’s roles as complementary and oppositional (Mattheis & Winter, 2019; Pearson, 2018b). The founding father of modern Islamism, Qutb, associated strictly separated gender roles with the moral value of any society. According to his schema, women were to be protected by the Islamist collective of men, guided by a strictly fundamentalist reading of Islam (Qutb, 2006). Qutb’s ideas underpin contemporary Jihadist ideology and are evident in its propaganda and actions. In previous work, I (2015, 2018a, 2019a) and my co-author Emily Winterbotham (2017) have suggested IS narratives are structured around a gender binary, valorising particular masculinities. Following Qutb, Jihadist groups such as IS present themselves as the only place for ‘real’ men, who demonstrate a ‘rightful masculinity’ when contrasted with ‘emasculated’ westerners (Winter, 2015). For nationalist and far-right movements too, manhood becomes associated with nationhood and the protection of the white race. Motherhood is exalted and symbolic; mothers are crucial as the ‘biological reproducers of “the nation” ’ (Yuval-Davis, 1996, p. 22). Sexuality also matters here. Far-right and jihadist groups have persecuted men and women understood as homosexual or trans, casting sexualities that are not understood to contribute to reproduction as deviant. Islamic State for instance disseminated videos depicting the execution of LGBTI individuals as part of its propaganda campaign (United States Department of State, 2017). Furthermore, gendered binaries and gender-based violence are not simply an aspect of contemporary violent groups arguably focused on issues of religion or race; rather, gender is the fundamental organising principle through which they are constructed. The incel (involuntary celibate) movement is a case in point; it is composed of men who believe high-status women are denying them sex because they are low-status men, and feminism is the enemy (Gentry, 2022; Williams, 2018). It is the gendered order of a movement or group that enables violence, constituting men’s 171

Elizabeth Pearson

violence as legitimate/desirable and women’s as transgressive. A focus on structures here both better explains how activism happens and provides a theoretical framework that sees women’s terrorism not as distinct or exceptional but as part of the same gendered systems within which men’s violence is created.

What gender tells us about groups A gendered approach is not simply about better understanding men’s violence; it is also about reframing women’s violence through consideration of their negotiation of patriarchal structural norms in extreme groups (Blee & Deutsch, 2012; Pearson, 2018b). My previous work (2017) has explored women’s engagement with gendered role definition in Jihadist groups, outlining how IS women online have enabled and enforced strict gendered norms prohibiting particular forms of female behaviour. For instance, some IS women have collectively censured those women not deemed jihadist enough, who for example use online photographs showing hair in profiles, or they have publicly chastised men for failing to travel to Islamic State territory to fight Jihad. There are various accounts demonstrating women’s enthusiastic participation in the all-female IS AlKhansaa Brigade, a form of police that became notorious for brutality against women infringing Jihadi norms (Bhutia, 2015; Eleftheriou-Smith, n.d.; Winter, 2015). A gendered approach considering individual responses to structural inequalities and their effects in further constituting these enables scholars to outline the mechanisms by which, particularly young women experiencing Islamophobia and marginalisation in the West, can be drawn towards ideologies in which hypermasculinities of warrior-dom and violence have high status, and women’s support of these is read as empowerment (see also Lahoud, 2018). In her work on the alt-right, Mattheis (2018) has also employed a gendered approach to detail women’s enthusiastic engagement with patriarchal norms and their attempts at self-empowerment through maternalist narratives. Women can ‘do’ criminal masculinities taking on roles that confirm men’s expected behaviour within crime (Miller, 2002) or by supporting masculinities that are apparently misogynist. As gendered approaches are situated, they enable ethnomethodological insights gained from a practical engagement with those affected by radicalisation. Protest masculinities, for instance, are evident in the misogynist incel movement, in Jihadist groups, as well as the far right and others, and authors have produced detailed analyses to evidence this through interviews and case studies (Duriesmith, 2020; Kimmel, 2018; Kimmel & Ferber, 2000; Messerschmidt & Rohde, 2018; Pearson, 2019b). These case studies reveal a multiplicity of situated masculinities, not simply one ‘terrorist masculinity’ (Duriesmith & Ismail, 2022, p. 2). These masculinities are dynamic, destabilised through extreme activism and its status within wider norms and structures, and then reasserted. They are local and regional (Popovski & Aslam, 2012), yet also global (Messerschmidt & Rohde, 2018). Such work challenges assertions that violent Jihad always concerns high-status warrior masculinities (Jensen & Larsen, 2019). This insight is important in countering and understanding jihadist messaging, which narratives resonate with which people, and why. This gendered ethnography is also the kind of first-hand, empirical work emphasised as both lacking and important in the terrorism studies field research literature (Chappuis & Krause, 2019; Dolnik, 2013; Sageman, 2014). It is particularly important to understand that ‘terrorist masculinities’ cannot be universalised. Through field research and interviews, Duriesmith (2018, 2019) and Duriesmith and Ismail (2019), for instance, have charted the multiple layers through which masculinity constitutes radical identities in Indonesia, noting the importance of local dynamics. The authors draw attention to alignments between masculinities of resistance and anti-colonialism and the multiplicity of masculinities that constitute ‘Indonesian’. Warrior masculinities are multiple and do not all share 172

Gender perspectives on radicalisation

the same status; foreign fighters of Asian ethnicities must prove their masculinity, as Asian men are regarded as having low status in the Afghan conflict. With Winterbotham (2017), my work has also discussed how gendered dynamics differ between western countries in catalysing radicalisation. In France and Germany, we found that bans on the wearing of the hijab in some contexts were important for the potential resonance of messaging from Islamist groups with both men and women. In the United Kingdom, government messaging responsibilising certain Muslim communities for radicalisation (mothers who did not speak English well, for instance) divided communities and created gaps in resilience. Focus group data also pointed to the relative lack of resilience of young men to economic hardship when compared to young women and the impact on the appeal of IS messaging. Ethnographic studies taking a gendered approach prioritise narratives articulated by those affected by radicalisation in both understanding and countering it. They furnish the field with practical understandings of the pressures and rewards of the radical life and, thereby, an increased ability to counter them. Such work provides lessons for counter-radicalisation initiatives on the necessity of considering specific local narratives in interventions as well as the complexity of masculinities evident in radicalisation pathways. Yet mainstream theories, as yet, tend not to apply these to existing theories, although they provide lessons both for the field and for practitioners.

Conclusion The past gender blindness of radicalisation studies is no accident. The wider historic IR and security studies field to which it belongs has lacked explicit engagement with feminist theory and gendered narratives. Yet no field can escape gender. The mainstream is itself gendered and racialised, in that it has historically been produced by men for men and through privileges of race and power. It is alert to some things and blind to others, gender among them (Enloe, 2001; Gentry, 2022; Hooper, 2001; Zalewski, 2007). This blindness is an impediment to understanding, especially in the current context, in which masculinist terrorist groups with misogynist ideologies appear dominant. The failure is one of epistemological and ontological position: how we decide to understand the field and what counts as worthy of study within it. The world of conflict and violence is one of masculinism, yet is security studies itself. It has been incapable of recognising the importance of analysis of that which it itself embodies, and that cannot go on. As Gentry (2022) and Leidig (2021) have observed, without a gendered approach, it is not possible for mainstream scholars to fully understand even an explicitly gendered and masculinist movement, such as the incels. A failure to see gender or to read misogyny as political is a failure to understand how gender constructs activism through hierarchies of social power. It is also a missed opportunity to counter it. This chapter has explored both the accumulated knowledge but also the inadequacies of past approaches within radicalisation and terrorism studies and the ways in which the field is adapting and changing. It has presented ideas outlining what a new understanding of gender in this arena could – and does, increasingly – look like and why this matters. Thinking about gender is, of course, not just about thinking about women, although some sections of security studies have been slow to even consider this. Gendered approaches provide a framework for thinking about radicalisation that neither ignores women nor takes men’s violence for granted. It understands these as being linked. The framework produced can engage with the increasingly complex landscape of gendered relations; gendered approaches understand how violent groups mobilise men and women around gendered societal change. Societal debates about gender are in constant evolution. Gender is, of course, about women, and indeed, women scholars continue to struggle to ensure their contributions are heard. But gendered approaches go beyond the study of women by women. 173

Elizabeth Pearson

Gendered approaches are not magical or mysterious; they are about subjecting the data gathered to a different kind of analysis. If there is still ‘stagnation’ in radicalisation studies, as Sageman (2014) has claimed, it might lie in the inexplicable reticence of radicalisation studies to look beyond their own self-imposed parameters. Radicalisation studies can no longer remain, as Jensen and Larsen (2019) observe, ‘strangely uninformed’ by other fields, such as masculinities studies. Nor can they overlook the existing work of women on radicalisation in their own field. The stakes are too high. Genderbased violence, gendered binaries, misogyny: these are not simply peripheral aspects of contemporary political violence, they are organising principles constructing violence, behaviour, recruitment, and activism in the key movements considered security priorities in the west. Gender cannot be ignored without putting people at risk, whether of the ruined lives of radicalisation or of the violence that can result from it. Gender is key. The trouble with radicalisation theory has been the historic neglect of this; the hope is that the growing numbers of scholars working with gender in terrorism studies will change the field in ways that will benefit scholars of this subject and increase the chances of understanding and therefore challenging pathways into political violence.

Notes 1 This work was supported by the Economic and Social Research Council [grant number ES/J500057/1] for a PhD project at King’s College London entitled “To what extent does gender matter in extremism in the UK?” 2 Critical scholars use the term ‘mainstream approaches’ to describe those relied on by policy-makers tasked with tackling radicalisation; such approaches are produced by predominantly male authors ensconced within an academic-security nexus that has tended to eschew critical or reflexive perspectives.

References Awan, A. N. (2014). Spurning this worldly life: Martyrdom amongst British Jihadists. In D. Janes & A. Houen (Eds.), Martyrdom and terrorism: Pre-modern to contemporary perspectives. Oxford University Press. Awan, I. (2012). “I am a Muslim not an extremist”: How the prevent strategy has constructed a “suspect” community. Politics & Policy, 40(6), 1158–1185. https://doi.org/10.1111/j.1747-1346.2012.00397.x Bakker, E. (2006). Jihadi terrorists in Europe: Their characteristics and the circumstances in which they joined the Jihad: An exploratory study [Clingendael CSCP Paper]. Clingendael Institute. www.clingendael.nl/sites/default/files/20061200_cscp_csp_bakker.pdf Bermingham, A., Conway, M., Mcinerney, L., Hare, N. O., & Smeaton, A. F. (2009). Combining social network analysis and sentiment analysis to explore the potential for online radicalisation. IEEE International Conference on Advances in Social Network Analysis and Mining, 231–236. Bhutia, J. (2015, December 28). Isis’s Al-Khansaa Brigade kills woman for breast-feeding baby in public in Raqqa. International Business Times UK. www.ibtimes.co.uk/isiss-al-khansaa-brigade-kills-womanbreast-feeding-baby-public-raqqa-1535110 Blee, K. M. (2007). Ethnographies of the far right. Journal of Contemporary Ethnography, 36(2), 119–128. https://doi.org/10.1177/0891241606298815 Blee, K. M. (2008). Women and organized racial terrorism in the United States. In C. D. Ness (Ed.), Female terrorism and militancy: Agency, utility, and organization (1st ed., pp. 201–216). Routledge. Blee, K. M., & Deutsch, S. M. (2012). Introduction. In Women and the right: Comparisons and interplay across borders (pp. 1–17). Penn State University Press. http://muse.jhu.edu/chapter/491433 Bloom, M. (2011). Bombshell: The many faces of women terrorists. C Hurst & Co Publishers Ltd. Boesten, J. (2014). Sexual violence during war and peace: Gender, power, and post-conflict justice in Peru. Palgrave Macmillan. http://ebookcentral.proquest.com/lib/kcl/detail.action?docID=1699326 Brown, K. E. (2008). The promise and perils of women’s participation in UK mosques: The impact of securitisation agendas on identity, gender and community. The British Journal of Politics & International Relations, 10(3), 472–491. https://doi.org/10.1111/j.1467-856X.2008.00324.x

174

Gender perspectives on radicalisation Brown, K. E. (2010). Gender and counter-terrorism: UK prevent and de-radicalisation strategies. American Political Studies Association, British Politics Group. www.britishpoliticsgroup.org/BPG%202010-Brown.pdf Brown, K. E. (2013). Gender and counter-radicalization: Women and emerging counter-terror measures. In J. Huckerby & M. L. Satterthwaite (Eds.), Gender, national security and counter-terrorism. Routledge. Brown, K. E. (2020). Gender and anti-radicalization. Oxford University Press. Busher, J. (2015, December 2). Understanding the English defence league: Living on the front line of a ‘clash of civilisations’[LSE University Website]. British Politics and Policy at LSE. http://blogs.lse.ac.uk/politicsandpolicy/ understanding-the-english-defence-league-life-on-the-front-line-of-an-imagined-clash-of-civilisations/ Chappuis, F., & Krause, J. (2019). Research dilemmas in dangerous places. In Secrecy and methods in security research (pp. 112–125). Routledge. Chatterjee, D. (2016). Gendering ISIS and mapping the role of women. Contemporary Review of the Middle East, 3(2), 201–218. https://doi.org/10.1177/2347798916638214 Cherney, A., Putra, I. E., Putera, V. S., Erikha, F., & Magrie, M. F. (2021). The push and pull of radicalization and extremist disengagement: The application of criminological theory to Indonesian and Australian cases of radicalization. Journal of Criminology, 54(4), 407–424. https://doi.org/10.1177/26338076211034893 Clegg, S. (2006). The problem of agency in feminism: A critical realist approach. Gender and Education, 18(3), 309–324. https://doi.org/10.1080/09540250600667892 Connell, R. W. (1995). Masculinities. Polity Press. Connell, R. W. (2004). Globalisation, imperialism and masculinities. In M. S. Kimmel, J. R. Hearn, & R. W. Connell (Eds.), Handbook of studies on men and masculinities (pp. 71–89). SAGE Publications, Inc. Connell, R. W. (2005). Masculinities (2nd ed.). Polity Press. Connell, R. W. (2011). Masculinity research and global society. In E. Ngan-Ling Chow, M. Texler Segal, & L. Tan (Eds.), Analyzing gender, intersectionality, and multiple inequalities: Global, transnational and local contexts (Vol. 15, pp. 51–72). Emerald Group Publishing Limited. www.emeraldinsight.com/doi/ pdfplus/10.1108/S1529-2126%282011%290000015008 Connell, R. W., & Messerschmidt, J. W. (2005). Hegemonic masculinity: Rethinking the concept. Gender & Society, 19(6), 829–859. https://doi.org/10.1177/0891243205278639 Cook, J., & Vale, G. (2018). From Daesh to diaspora: Tracing the women and minors of Islamic state/ ICSR. International Centre for the Study of Radicalisation and Political Violence. http://icsr.info/2018/07/ icsr-report-launch-daesh-diaspora-tracing-women-minors-islamic-state/ Cottee, S. (2011). Jihadism as a subcultural response to social strain: Extending Marc Sageman’s “bunch of guys” thesis. Terrorism and Political Violence, 23(5), 730–751. https://doi.org/10.1080/09546553.2011.611840 Dalgaard-Nielsen, A. (2010). Violent radicalization in Europe: What we know and what we do not know. Studies in Conflict & Terrorism, 33(9), 797–814. https://doi.org/10.1080/1057610X.2010.501423 Davis, J. (2006). Women and radical Islamic Terrorism: Planners, perpetrators, patrons? The Canadian Institute of Strategic Studies, 136. Davis, J. (2013). Evolution of the global Jihad: Female suicide bombers in Iraq. Studies in Conflict & Terrorism, 36(4), 279–291. https://doi.org/10.1080/1057610X.2013.763598 Dawson, L. (2021). A comparative analysis of the data on western foreign fighters in Syria and Iraq: Who went and why? ICCT Research Paper. https://icct.nl/app/uploads/2021/02/Dawson-Comparative-Analysis-FINAL-1.pdf della Porta, D., & Diani, M. (2005). Social movements: An introduction (2nd ed.). Wiley-Blackwell. della Porta, D., & Haupt, H.-G. (2012). Patterns of radicalization in political activism: An introduction. Social Science History, 36(3), 311–320. d’Estaing, S. G. (2017). Engaging women in countering violent extremism: Avoiding instrumentalisation and furthering agency. Gender & Development, 25(1), 103–118. https://doi.org/10.1080/13552074.2017.1279823 Dolnik, A. (2013). Conducting terrorism field research: A guide. Routledge. Duriesmith, D. (2018). Hybrid warriors and the formation of new war masculinities: A case study of Indonesian foreign fighters. Stability: International Journal of Security and Development, 7(1). https://doi. org/10.5334/sta.633 Duriesmith, D. (2019). Masculinity and new war: The gendered dynamics of contemporary armed conflict (1st ed.). Routledge. Duriesmith, D. (2020). Adaptation of militarized masculinity and violent extremism in the Southern Philippines. In K. O’Neil (Ed.), Conflicting identities: The nexus between masculinities, femininities and violent extremism in Asia (pp.  11–30). United Nations Development Programme (UNDP) and UN Women. www.unwomen.org/en/digital-library/publications/2020/03/conflicting-identities-the-nexusbetween-masculinities-femininities-and-violent-extremism-in-asia

175

Elizabeth Pearson Duriesmith, D., & Ismail, N. H. (2019). Militarized masculinities beyond methodological nationalism: Charting the multiple masculinities of an Indonesian Jihadi. International Theory, 11(2), 139–159. https://doi. org/10.1017/S1752971919000034 Duriesmith, D., & Ismail, N. H. (2022). Masculinities and disengagement from Jihadi networks: The case of Indonesian militant Islamists. Studies in Conflict  & Terrorism, 1–21. https://doi.org/10.1080/10576 10X.2022.2034220 Eggert, P. (2015). Women fighters in the “Islamic state” and Al-Qaida in Iraq: A comparative analysis. Die Friedens-Warte, 90(3/4), 363–380. http://www.jstor.org/stable/24868700 Eleftheriou-Smith, L.-M. (n.d.). Escaped Isis wives describe life in the punishing all-female Al-Khansaa brigade. The Independent. Retrieved April 25, 2015, from www.independent.co.uk/news/world/middle-east/ escaped-isis-wives-describe-life-in-the-allfemale-alkhansa-brigade-who-punish-women-with-40-lashesfor-wearing-wrong-clothes-10190317.html Enloe, C. (2001). Bananas, beaches and bases: Updated edition with a new preface: Making feminist sense of international politics (2nd Revised ed.). University of California Press. Gartenstein-Ross, D., & Grossman, L. (2009). Homegrown terrorists in the US and UK: An empirical examination of the radicalization process. FDD Center for Terrorism Research. Gendron, A. (2016). The call to Jihad: Charismatic preachers and the internet. Studies in Conflict & Terrorism, 1–18. https://doi.org/10.1080/1057610X.2016.1157406 Gentry, C. E. (2022). Misogynistic terrorism: It has always been here. Critical Studies on Terrorism, 15(1), 209–224. https://doi.org/10.1080/17539153.2022.2031131 Sjoberg, Cooke, and Neal, Introduction in Women, Gender and Terrorism. Eds. Sjoberg, Laura, Gentry, Caron E. Athens: University of Georgia Press, 2011. Hegghammer, E. T. (2017). 8. Non-military practices in Jihadi groups. In Jihadi culture: The art and social practices of militant Islamists (pp. 171–220). Cambridge University Press. HMGovernment.(2015).Counter-extremismstrategy.www.gov.uk/government/publications/counter-extremism-strategy HM Government. (2020). Online harms white paper: Full government response to the consultation. Home Office. www.gov.uk/government/consultations/online-harms-white-paper/outcome/online-harms-whitepaper-full-government-response Home Office. (2011). Prevent strategy. Home Office. www.gov.uk/government/uploads/system/uploads/ attachment_data/file/97976/prevent-strategy-review.pdf Hooper, C. (2001). Manly states: Masculinities, international relations, and gender politics. Columbia University Press. http://ebookcentral.proquest.com/lib/kcl/detail.action?docID=952902 Horgan, J. (2012, September 28). Discussion point: The end of radicalization? | START.umd.edu. www.start. umd.edu/news/discussion-point-end-radicalization Huey, L., Inch, R., & Peladeau, H. (2017). “@ me if you need shoutout”: Exploring women’s roles in Islamic state Twitter networks. Studies in Conflict  & Terrorism, 0(ja), 00–00. https://doi.org/10.1080/10576 10X.2017.1393897 Huey, L., & Witmer, E. (2016). #IS_Fangirl: Exploring a new role for women in terrorism. Journal of Terrorism Research, 7(1), 1. https://doi.org/10.15664/jtr.1211 Ingram, K. (2017). IS’s appeal to western women: Policy recommendations. International Centre for CounterTerrorism. https://icct.nl/publication/iss-appeal-to-western-women-policyrecommendations/ Jackson, L. B. (2021). The monstrous and the vulnerable: Framing British Jihadi brides. Hurst. Jacques, K., & Taylor, P. J. (2009). Female terrorism: A review. Terrorism and Political Violence, 21(3), 499–515. https://doi.org/10.1080/09546550902984042 Jensen, S. Q., & Larsen, J. F. (2019). Sociological perspectives on Islamist radicalization – bridging the micro/macro gap. European Journal of Criminology, 1477370819851356. https://doi.org/10.1177/ 1477370819851356 Kenney, M. (2018). The Islamic state in Britain: Radicalization and resilience in an activist network. Cambridge University Press. https://doi.org/10.1017/9781108557108 Khalil, J., & Zeuthen, M. (2016). Countering violent extremism and risk reduction: A guide to programme design and evaluation (Whitehall Report No. 2; p. 45). RUSI. Kimmel, M. (2007). Racism as adolescent male rite of passage: Ex-Nazis in Scandinavia. Journal of Contemporary Ethnography, 36(2), 202–218. https://doi.org/10.1177/0891241606298825 Kimmel, M. (2018). Healing from hate: How young men get into―and out of―violent extremism (1st ed.). University of California Press.

176

Gender perspectives on radicalisation Kimmel, M., & Ferber, A. L. (2000). ‘White men are this nation’: Right-wing militias and the restoration of rural American masculinity. Rural Sociology, 65(4), 582–604. Kimmel, M. S. (2003). Globalization and its Mal(e)Contents: The gendered moral and political economy of terrorism. International Sociology, 18(3), 603–620. https://doi.org/10.1177/02685809030183008 King, M., & Taylor, D. M. (2011). The radicalization of homegrown Jihadists: A review of theoretical models and social psychological evidence. Terrorism and Political Violence, 23(4), 602–622. https://doi.org/10.1 080/09546553.2011.587064 Klausen, J. (2015). Tweeting the Jihad: Social media networks of western foreign fighters in Syria and Iraq. Studies in Conflict & Terrorism, 38(1), 1–22. https://doi.org/10.1080/1057610X.2014.974948 Klausen, J. (2019). Neighbourhood effects – how Jihadist recruitment really works. In A. Richards, D. Margolin, & N. Scremin (Eds.), Jihadist terror: New threats, new responses. CoJiT-UK. Kundnani, A. (2014). The Muslims are coming: Islamophobia, extremism, and the domestic war on terror. Verso Books. Lahoud, N. (2018). Empowerment or subjugation: An analysis of ISIL’s gendered messaging. UN Women. Lehane, O., Mair, D., Lee, S., & Parker, J. (2018). Brides, black widows and baby-makers; or not: An analysis of the portrayal of women in English-language Jihadi magazine image content. Critical Studies on Terrorism, 1–16. https://doi.org/10.1080/17539153.2018.1471054 Leidig, E. (2021). Why terrorism studies miss the mark when it comes to incels. https://icct.nl/publication/ why-terrorism-studies-miss-the-mark-when-it-comes-to-incels/ Malthaner, S. (2017). Radicalization: The evolution of an analytical paradigm. European Journal of Sociology/Archives Européennes de Sociologie, 58(3), 369–401. https://doi.org/10.1017/S0003975617000182 Malthaner, S., & Waldmann, P. (2014). The radical milieu: Conceptualizing the supportive social environment of terrorist groups. Studies in Conflict & Terrorism, 37(12), 979–998. https://doi.org/10.1080/1057 610X.2014.962441 Mandel, D. R. (2009). Radicalisation – what does it mean? In T. M. Pick, A. Speckhard, & B. Jacuch (Eds.), Home-grown terrorism: Understanding and addressing the root causes of radicalisation among groups with an immigrant heritage in Europe – Volume 60 . . . Series – E: Human and Societal Dynamics. Margolin, D. (2018). Neither feminists nor victims: How women’s agency has shaped Palestinian violence. Institute for Global Change. https://institute.global/policy/neither-feminists-nor-victims-howwomens-agency-has-shaped-palestinian-violence Mattheis, A. A. (2018). Shieldmaidens of whiteness: (Alt) Maternalism and women recruiting for the far/altright. Journal for Deradicalization (17), 128–162. Mattheis, A. A., & Winter, C. (2019). The greatness of her position (p. 32). International Centre for the Study of Radicalisation and Political Violence. McCauley, C., & Moskalenko, S. (2008). Mechanisms of political radicalization: Pathways toward terrorism. Terrorism and Political Violence, 20(3), 415–433. https://doi.org/10.1080/09546550802073367 Messerschmidt, J. W. (1993). Masculinities and crime: Critique and reconceptualization of theory. Rowman & Littlefield. Messerschmidt, J. W., & Rohde, A. (2018). Osama bin laden and his Jihadist global hegemonic masculinity. Gender & Society, 32(5), 663–685. https://doi.org/10.1177/0891243218770358 Miller, J. (2002). The strengths and limits of ‘doing gender’ for understanding street crime. Theoretical Criminology, 6(4), 433–460. https://doi.org/10.1177/136248060200600403 Moghaddam, F. M. (2005). The staircase to terrorism: A psychological exploration. American Psychologist, 60(2), 161–169. https://doi.org/10.1037/0003-066X.60.2.161 Mohanty, C. T., & Mohanty, S. P. (1990). Contradictions of colonialism: Review of Kumkum Sangari and Sudesh Vaid, eds., recasting women: Essays in colonial history. Women’s Review of Books, 19–21. Nesser, P. (2006). Jihadism in Western Europe after the invasion of Iraq: Tracing motivational influences from the Iraq war on Jihadist terrorism in Western Europe. Studies in Conflict & Terrorism, 29(4), 323–342. https://doi.org/10.1080/10576100600641899 Nesser, P. (2010). Joining Jihadi terrorist cells in Europe. In M. Ranstorp (Ed.), Understanding violent radicalisation: Terrorist and Jihadist movements in Europe. Routledge. http://ebookcentral.proquest.com/lib/ kcl/detail.action?docID=481128 Nesser, P. (2016). Islamist terrorism in Europe: A history. C Hurst & Co Publishers Ltd. Neumann, P. R. (2008). Introduction. The Adelphi Papers, 48(399), 5–10. https://doi.org/10.1080/ 05679320802686783 Neumann, P. R. (2010). Chapter two: Recruitment grounds. The Adelphi Papers, February 2013, 37–41.

177

Elizabeth Pearson Neumann, P. R. (2013). The trouble with radicalization. International Affairs, 89(4), 873–893. https://doi. org/10.1111/1468-2346.12049 Neumann, P. R., & Kleinmann, S. (2013). How rigorous is radicalization research? Democracy and Security, 9(4), 360–382. https://doi.org/10.1080/17419166.2013.802984 OSCE. (2011). Women and terrorist radicalization. OSCE. http://polis.osce.org/library/f/4061/3772/ OSCE-AUT-RPT-4061-EN-3772 Pantucci, R. (2011). A typology of lone wolves: Preliminary analysis of lone Islamist terrorists editorial assistants. International Centre for the Study of Radicalisation and Political Violence. Parashar, S. (2011). Gender, Jihad, and jingoism: Women as perpetrators, planners, and patrons of militancy in Kashmir. Studies in Conflict & Terrorism, 34(4), 295–317. https://doi.org/10.1080/1057610X.2011.551719 Parpart, J. L., & Zalewski, M. (2008). Introduction: Rethinking the man question. In J. L. Parpart & M. Zalewski (Eds.), Rethinking the man question: Sex, gender and violence in international relations: 1 (1st ed.). Zed Books. Pearson, E. (2015). The case of Roshonara Choudhry: Implications for theory on online radicalization, ISIS women, and the gendered Jihad. Policy & Internet, n/a-n/a. https://doi.org/10.1002/poi3.101 Pearson, E. (2017). Online as the new frontline: Affect, gender, and ISIS-take-down on social media. Studies in Conflict & Terrorism, 1–25. https://doi.org/10.1080/1057610X.2017.1352280 Pearson, E. (2018a). Why men fight and women don’t: Masculinity and extremist violence (Debunking Myths on Gender and Extremism). Tony Blair Institute for Global Change. http://institute.global/insight/ co-existence/why-men-fight-and-women-dont-masculinity-and-extremist-violence Pearson, E. (2018b). Wilayat Shahidat: Boko Haram, the Islamic state, and the question of the female suicide bomber. In J. Warner & J. Zenn (Eds.), Boko Haram: Behind the headlines. CTC Sentinel. Pearson, E. (2019a). To what extent does gender matter in UK extremism? King’s College London. Pearson, E. (2019b). Extremism and toxic masculinity: The man question re-posed. International Affairs, 95(6), 1251–1270. https://doi.org/10.1093/ia/iiz177 Pearson, E. (2022). Beyond women: Rethinking gender and radicalisation. In A. N. Awan & J. Lewis (Eds.), Radicalisation: A global and comparative perspective. Hurst & Co. www.hurstpublishers.com/book/ radicalisation/ Pearson, E., & Winterbotham, E. (2017). Women, gender and Daesh radicalisation. The RUSI Journal, 1–13. https://doi.org/10.1080/03071847.2017.1353251 Pearson, E., Winterbotham, E., & Brown, K. E. (2020). Countering violent extremism: Making gender matter. Palgrave Macmillan. Peterson, S. V. (2005). How (the meaning of) gender matters in political economy. New Political Economy, 10(4), 499–521. https://doi.org/10.1080/13563460500344468 Pilkington, H. (2016). Loud and proud: Passion and politics in the English Defence League. Manchester University Press. www.oapen.org/search?identifier=607920 Pilkington, H. (2017). ‘EDL angels stand beside their men . . . not behind them’: The politics of gender and sexuality in an anti-Islam(ist) movement. Gender and Education, 29(2), 238–257. https://doi.org/10.108 0/09540253.2016.1237622 Poloni-Staudinger, L., & Ortbals, C. D. (2012). Terrorism and violent conflict: Women’s agency, leadership, and responses (2013 ed.). Springer. Popovski, V., & Aslam, M. (2012). Gender-based explosions: The nexus between Muslim masculinities, Jihadist Islamism and terrorism. United Nations University Press. Quilliam. (2010). Quilliam briefing paper: The threat of radicalisation on British university campuses. Quilliam. Qutb, S. (2006). Milestones. Islamic Book Service. Rashid, N. (2014). Giving the silent majority a stronger voice? Initiatives to empower Muslim women as part of the UK’s ‘War on Terror’. Ethnic and Racial Studies, 37(4), 589–604. https://doi.org/10.1080/014198 70.2013.816759 Richards, A. (2014). Conceptualizing terrorism. Studies in Conflict & Terrorism, 37(3), 213–236. Rogers, Brooke and Neumann, Peter. (2007). Recruitment and Mobilisation for the Islamist Militant Movement in Europe, ICSR. Commissioned by Directorate General for Justice, Freedom and Security of the European Commission. Roy, O. (2007). Islamic terrorist radicalisation in Europe. In S. Amghar, A. Boubekeur, & M. Emerson (Eds.), European Islam: The challenges for society and public policy. Centre for European Policy Studies. http:// archives.cerium.ca/IMG/pdf/Roy-IslamicTerroristRadicalisation.pdf

178

Gender perspectives on radicalisation Roy, O. (2017). Jihad and death: The global appeal of Islamic state. C Hurst & Co. Sageman, M. (2004). Understanding terror networks. University of Pennsylvania Press. Sageman, M. (2008). Leaderless Jihad: Terror networks in the twenty-first century (1st ed.). University of Pennsylvania Press. Sageman, M. (2014, May). The stagnation in terrorism research. Terrorism and Political Violence, 1–16. https://doi.org/10.1080/09546553.2014.895649 Sageman, M., Groen, J., & Kranenberg, A. (2010). Women warriors for Allah: An Islamist network in the Netherlands (R. Naborn, Trans.). University of Pennsylvania Press. Scott, J. W. (1986). Gender: A useful category of historical analysis. The American Historical Review, 91(5), 1053–1075. https://doi.org/10.2307/1864376 Scrivens, R. (2018, August). Stormfront women. VOX-Pol. Sedgwick, M. (2010). The concept of radicalization as a source of confusion. Terrorism and Political Violence, 22(4), 479–494. https://doi.org/10.1080/09546553.2010.491009 Shepherd, L. J. (2006). Veiled references: Constructions of gender in the Bush administration discourse on the attacks on Afghanistan post-9/11. International Feminist Journal of Politics, 8(1), 19–41. https://doi. org/10.1080/14616740500415425 Silke, A. (Ed.). (2003). Research on terrorism: Trends, achievements and failures. Routledge. Sjoberg, L. (2013). Gendering global conflict: Toward a feminist theory of war. Columbia University Press. Sjoberg, Cooke, and Neal, Introduction in Women, Gender and Terrorism. Eds. Sjoberg, Laura, Gentry, Caron E. Athens: University of Georgia Press, 2011. Smelser, N. J. (2010). The faces of terrorism: Social and psychological dimensions. Princeton University Press. Tickner, J. A. (1997). You just don’t understand: Troubled engagements between feminists and IR theorists. International Studies Quarterly, 41(4), 611–632. United States Department of State. (2017). Country reports on human rights practices for 2016 United States department of state. Bureau of Democracy, Human Rights and Labor. https://web.archive.org/ web/20170307090250/www.state.gov/documents/organization/265710.pdf UNSCR 2242. (2015). via. www.securitycouncilreport.org/atf/cf/%7B65BFCF9B-6D27-4E9C-8CD3CF6E4FF96FF9%7D/s_res_2242.pdf Venhaus, C. J. M. ‘Matt’. (2010). Why youth Join al-Qaeda [Special Report]. United States Institute of Peace. Von Knop, K. (2006). Women in religious terror organizations. In L. K. Johnson (Ed.), Strategic intelligence: Volume 1, understanding the hidden side of government: Vol. I (pp. 154–177). Praeger. Von Knop, K. (2007). The female Jihad: Al Qaeda’s women. Studies in Conflict & Terrorism, 30(5), 397–414. https://doi.org/10.1080/10576100701258585 Waddell, K. (2016, February 19). Twitter’s account suspensions are surprisingly effective against ISIS. The Atlantic. www.theatlantic.com/technology/archive/2016/02/twitters-account-suspensions-aresurprisingly-effective-against-the-islamic-state/463440/ Weeks, D. (2020). Al Muhajiroun: A case study in contemporary Islamic activism (1st ed.). Palgrave Macmillan. West, C., & Zimmerman, D. H. (1987). Doing gender. Gender & Society, 1(2), 125–151. https://doi. org/10.1177/0891243287001002002 Wiktorowicz, Q. (2005). Radical Islam rising: Muslim extremism in the west. Rowman & Littlefield Publishers, Inc. Williams, Z. (2018,April 25). ‘Raw hatred’:Why the ‘incel’movement targets and terrorises women.The Guardian. www.theguardian.com/world/2018/apr/25/raw-hatred-why-incel-movement-targets-terrorises-women Winter, C. (2015). Women of the Islamic state. The Quilliam Foundation. www.quilliamfoundation.org/wp/ wp-content/uploads/publications/free/women-of-the-islamic-state3.pdf Winter, C., & Margolin, D. (2017). The Mujahidat Dilemma: Female combatants and the Islamic state. CTC Sentinel, 10(7). https://ctc.usma.edu/v2/wp-content/uploads/2017/08/CTC-Sentinel_Vol10Iss7.pdf Yuval-Davis, N. (1996). Women and the biological reproduction of ‘the nation’. Women’s Studies International Forum, 19(1/2), 17–24. Yuval-Davis, N. (2016). Power, intersectionality and the politics of belonging. In W. Harcourt (Ed.), The Palgrave handbook of gender and development: Critical engagements in feminist theory and practice. Palgrave Macmillan Limited. http://ebookcentral.proquest.com/lib/kcl/detail.action?docID=4720388 Zalewski, M. (2007). Do we understand each other yet? Troubling feminist encounters with (in). 9, 302–312. https://doi.org/10.1111/j.1467-856x.2007.00287.x

179

12 RADICALISATION AND PSYCHOPATHOLOGY Emily Corner

There and back again Psychopathology, defined as both the scientific study of abnormal cognition and accompanying maladaptive behaviours, and the manifestation of behaviours and experiences indicative of mental illness or psychological impairment, has been associated with terrorist behaviour since the birth of study in the area. The application of psychopathology in this space should be anticipated, as such explanations have continuously been asserted in investigations seeking to identify causality across multiple types of non-political violent behaviour (Leistedt, 2013). The terrorist suffering from a mental disorder is initially an attractive concept, as the callous, destructive actions and behaviours involved in terrorism invite a stigmatising label. The academic inquiry of the relationship between becoming a terrorist and mental health has experienced a number of paradigms of conceptual and empirical development. These paradigms have differed from one another with regard to their conceptualisations of terrorist involvement, the specific mental disorders attributed to the acts, and the underlying empirical evidence. The earliest academic research seeking to explain terrorist behaviour looked to the most commonly attributed cause of extreme behaviour in society – psychopathy (Cooper, 1977, 1978; Ferracuti & Bruno, 1981; Pearce, 1977; Strentz, 1988; Tanay, 1987). Characterised by a lack of empirical evidence, the works within this paradigm offered psychopathy as a cause of terrorist involvement. Using post-hoc assessments of known terrorism cases and focusing on individual drives, the studies categorised terrorist involvement as a yes/no dichotomy. The rationale underpinning this paradigm was predominately derived from popular culture and the desire to attribute mental disorders to those committing such heinous, violent acts (Victoroff, 2005). The later emergence of complex clinical procedures for accurately capturing psychopathy likely underpinned the lack of systematic empirical interrogation of its role in radicalisation and terrorist behaviour. Concurrent to the early research asserting psychopathy as a cause of terrorism, a team of German researchers published their investigation, the results of which spurred the second paradigm and remain some of the most influential in the field today. The Analysen zum Terrorismus was a highly comprehensive mixed-methods study published across four volumes, one of which included Schmidtchen’s interviews and subsequent analyses of 250 terrorist careers (Jäger et al., 1981). The results of the interviews highlighted a range of different personality traits across both leaders and DOI: 10.4324/9781003035848-14 180

Radicalisation and psychopathology

followers, focusing particularly on narcissism. The impact of these findings should not be underestimated. Schmidtchen’s findings still resonate in literature seeking to explain terrorist behaviour today (Houssier, 2016; Lloyd & Kleinot, 2017; MacDonald, 2014; Opoku-Agyemang, 2017; Rae, 2012). Despite the impact of the Analysen zum Terrorismus, much of the subsequent literature examining personality1 was conducted following violent events, using methods focused on profiling individuals based on the nature of the attack behaviour and lacking rigorous clinical diagnostic procedures (Akhtar, 1999; Baruch, 2003; Berko, 2007; Billig, 1985; De Cataldo Neuberger & Valentini, 1996; DeMause, 2002; Kellen, 1982; Pearlstein, 1991; Post, 1984; Taylor & Quayle, 1994). A range of seminal reviews conducted within the third paradigm synthesised the existing evidence base, rightly questioning the causative nature and prevalence of psychopathy and personality traits (Borum, 2004; Horgan, 2003, 2005; Silke, 1998, 2003; Victoroff, 2005). These reviews have become extremely influential, amassing over 4,000 citations at the time of writing, and have focused on the ambiguities and seemingly contrasting findings regularly uncovered within various empirical studies in this area. They suggest that the differences may be a by-product of misunderstandings, methodological approaches, sampling, and interpretation. The fields examining personality and mental disorders are vast and conflicted, and the reviews reflected that conflict. Importantly, the range of traits and disorders identified across studies span multiple theoretical models and constructs, which, as the reviews noted, makes it extremely difficult to draw out singular observations that may have credibility and predictive value. Intrinsically related to this, the reviews highlighted that no single mental disorder or personality trait is reliably associated with radicalisation or terrorism. This is true across all forms of violence, so it is unsurprising that it is also reflected here. Despite the intention of the reviews, subsequent misinterpretations contributed to some commonly held, yet false, generalisations within the literature. Many studies citing the reviews not only rejected the explanatory power of personality or mental disorders as concluded within the reviews but also likely downplayed the presence of any mental health problems among terrorists in the first place (Post, 2005, 2007; Sher & Rice, 2015; Spaaij, 2011). Whereas the reviews used highly specific diagnostic language, the citations generalised to a higher level of abstraction, overlooking many nuances. It was not uncommon to see terms such as ‘cognitively normal’ (Abrahms, 2011, p. 591), ‘bounds of normality’ (O’Gorman, 2010, p. 64), ‘not normally crazy’ (Lutz & Lutz, 2013, p. 16) throughout the citing literature. This contributed to the false assumption that there was no relationship between psychopathology and terrorist involvement, which had major implications for both practice and media portrayal of terrorist and mass casualty events. The intention of the reviews was not to halt the study of psychopathology and terrorism. Rather, they intended to highlight the problems with existing studies and argue for less simplistic linear assumptions and more empirically rigorous investigations moving forward.

Debunking the dichotomy Despite the influence of the literature citing the reviews, the study of how individuals become radicalised and enact terrorist offences has developed rapidly in the past 15 years, and conceptual development and empirical analyses in this area have flourished, encapsulating the current paradigm. The research in this paradigm is typified by an expansive range of samples, ranging from non-radical populations through to terrorists. Each sample offers more insight into the complexity of the processes undergone by individuals who radicalise and move to enact terrorist behaviour. There is now a consensus that no single variable can explain these phenomena, and the studies within this paradigm are carving a middle ground between the hardcore causal line adopted by 181

Emily Corner

paradigms one and two and the mistaken assumptions that followed seminal reviews of paradigm three. This paradigm is characterised by the understanding that radicalisation and terrorist involvement are highly complex processes (Gill, 2012, 2015; Horgan, 2005), and the understanding of the role of psychopathology within these processes is still in its infancy. The greater conceptual and empirical sophistication of these works has broadened our understanding of the role of psychopathy in radicalisation and terrorist behaviour. Recent research has measured the relationship between both radicalisation and terrorist involvement and both traits (De Waele & Paulwels, 2016; Egan et al., 2016; Nussio, 2017; Pauwels et al., 2014; Pauwels & De Waele, 2014; Pauwels & Hardyns, 2018; Pauwels & Heylen, 2014; Pauwels et al., 2020; Pauwels & Schils, 2016; Pauwels & Svensson, 2017; Perry et al., 2018; Schils & Pauwels, 2016; Stys et al., 2014) and clinical facets of psychopathy (Bélanger et al., 2014; Jones, 2013). The results of which highlight that, at best, specific traits of psychopathy – including sensation-seeking, poor self-control, low empathy, and impulsivity – in conjunction with a wide range of experiences and behaviours may play an indirect role in an individual’s radicalisation to terrorism. The inquiry into the role of personality in radicalisation has also developed in more recent years, with researchers now employing a range of methods to investigate the causal role of a range of personality traits, including surveys, questionnaires, focus groups, interviews, and closed and open-source data analysis (Corner et al., 2021). Empirically significant relationships2 have been identified between radicalisation and adventure seeking (Bartlett & Miller, 2012), dominance (Dunwoody & McFarland, 2018), goal commitment (Bélanger et al., 2014), greed (Arjona & Kalyvas, 2009), inflexibility (Zmigrod et al., 2019), moral neutralisation (Nivette et al., 2017), passion (Bélanger et al., 2014; Sikkens, 2018), status seeking (Bartlett & Miller, 2012), superiority (De Waele & Pauwels, 2016; Pauwels & Heylen, 2014), uncertainty (Gøtzsche-Astrup, 2019, 2020), and weak morality (Perry et al., 2018). These findings emphasise the inherent complexity of examining personality; each human being is unique, so therefore their personality is. Despite the improvements in understanding, studies examining personality in this space have a long way to go. In their expansive systematic review of personality in terrorism, Corner et al. (2021, p. 398) note: many of the variables identified during this review were drawn from studies including personality traits as variables without an appreciation for the strict methodological procedures required for valid measurement of personality, or in some instances, an appreciation that such variables are personality traits. Other studies included personality traits as variables, with a predominant theoretical and conceptual focus on other, non-personality related variables. Beyond empirical improvements in the focus of earlier paradigms, much of the work in this paradigm has focused on attempting to understand the role of a range of other mental disorders. The predominant interest in this research endeavour has focused on establishing prevalence.3 Since the first study examining ethno-nationalist offenders in Northern Ireland, carried out by Lyons and Harbinson (1986), there have been 24 studies examining the prevalence rates of mental disorders in terrorism (Gill et al., 2021). These studies have investigated a range of samples, including groups (Hewitt, 2003; Gruenewald et al., 2013; Corner & Gill, 2015, 2019) and lone (Hewitt, 2003; Gruenewald et al., 2013; Gill et al., 2014; Capellan, 2015; Capellan & Anisin, 2018; Liem van Buuren et al., 2018; Zeman et al., 2018; Gill et al., 2019) offenders, and jihadist (Bakker, 2006; Leygraf, 2014; Chermak & Gruenewald, 2015; Perry et al., 2017; van Leyenhorst & Andreas, 2017; King et al., 2018; Böckler et al., 2018; Weenink, 2019; Dhumad et al., 2020), far-right (Gruenewald et al., 2013; Bubolz & Simi, 2019; Challacombe & Lucas, 2019), and far-left (Chermak & Gruenewald, 2015) offenders. The prevalence rates vary greatly across studies, ranging from 0% to 182

Radicalisation and psychopathology

57%. These differences are predominately due to the form of data that are used to reach prevalence estimates. In studies using clinical examinations, diagnoses were more prevalent (33.47%) than those using police or judicial sources (16.96%) or those using open sources (9.82%).4 These studies, while useful for helping debunk the misconception that psychopathology does not occur in terrorists, treat mental disorders as dichotomous,5 and fail to acknowledge that clinical diagnostics span an extensive range of disorders, each with different impacts on behaviour. A second stream of studies has moved beyond overall prevalence to investigate specific disorders in terrorist samples, including the identification of instances of attention-deficit/hyperactivity disorder (ADHD) (van Leyenhorst & Andreas, 2017; Weenink, 2015), autism spectrum disorders (Knight et al., 2017; Weenink, 2015), borderline personality disorder (van Leyenhorst & Andreas, 2017), depression (Bakker, 2006), pervasive development disorder (Weenink, 2015), post-traumatic stress disorder (PTSD) (van Leyenhorst & Andreas, 2017; Weenink, 2015), psychotic disorder (Leygraf, 2014; van Leyenhorst & Andreas, 2017; Weenink, 2019), schizophrenia (Knight et al., 2017; Leygraf, 2014; Weenink, 2015), and substance abuse disorders (Weenink, 2015). Furthering these works, two studies have sought to ascertain specific prevalence rates of disorders. Corner et al. (2016) identified extensive prevalence rates across their sample of 153 lone-actor terrorists, and Gill et al. (2019) drew prevalence rates from closed source police files, identifying mood disorders (12.2%), schizophrenia (10.2%), intellectual disabilities (4.1%), and personality disorders (2%). While studies examining diagnostic prevalence rates have value and have undoubtedly contributed to our increased knowledge of the role of mental health in radicalisation and subsequent terrorist behaviour, the use of diagnostics is not without issue. The field of diagnostics is highly dependent on a range of factors, including populations, cultures, and participation, and is met with a range of constraints, including, but not exclusively, the exclusion of populations with high proportions of severe mental disorders, for example the homeless, survey nonresponse due to refusal, and systematic non-reporting following errors or failures. Highlighting differences in diagnostics in their analysis of World Mental Health surveys across 17 countries, Kessler and Üstün (2008) noted that the recorded prevalence for ‘any disorder’ ranged from 12.0% (Nigeria) to 47.4% (United States) with an average prevalence of 27.43%. Further to these issues, according to Stone et al. (2020, p. 797), a ‘good diagnosis provides a valid interpretation of a person’s experience. . . . It should also capture what is known about the causation and course of the illness, and it should be clinically helpful’. In practice, diagnostics should be made in clinical settings with access to individuals within the field of interest. In reality, only one of the above studies achieves this (Leygraf, 2014). Some studies use prison or police case files (Knight et al., 2017; Weenink, 2015, 2019), others use judicial information (van Leyenhorst & Andreas, 2017), and most rely on open-source information to develop diagnostics. So, while these studies are a necessary endeavour in the first steps to examining prevalence, their validity remains unclear. To counter such challenges, some use diagnostic aids specifically designed for research, such as the ICD-10 (Corner & Gill, 2015, 2019) while others have worked alongside trained psychologists who performed diagnostics on cases (Gill et al., 2019). Others still have moved away from diagnostics of specific disorders and sought a more holistic approach, examining the stated lived experience of psychopathology, focusing on identifying prevalence of overall mental state of terrorist offenders (Altier et al., 2017; Bubolz & Simi, 2019; Corner & Gill, 2019, 2021; Horgan et al., 2017). These studies are able to offer deeper insight into the role of psychopathology in radicalisation and terrorist behaviour than the other prevalence studies and are the foundation of the move towards relevance, which will be discussed later in this chapter. Although the above results demonstrate a tentative correlational relationship between the presence of particular mental disorders and specific forms of terrorist behaviour, the answers to the 183

Emily Corner

relevance of psychopathology remain elusive due to research designs and data limitations. Despite the increasing empirical value of terrorism research, particularly in the domain of psychopathology, one lingering problem in both the study of terrorism and the subsequent implementation of prevention protocols is that of base rates. Authors have argued that solely focusing on those who engage in violence on behalf of a political or religious cause unduly narrows our understanding of the relationship between psychopathology and extreme violence (Corner et al., 2018). However, comparison group studies are extremely rare in this research area; indeed, Desmarais et al.’s (2017) systematic review identified just six investigations. Generalising results from research designs that lack suitable comparison groups likely overpredicts the role of factors of terrorism. Within the domain of psychopathology, although the current overall diagnostic prevalence rates are high for specific samples (e.g. lone actors), research is yet to concretely determine whether this prevalence is unique to lone-actor terrorists or whether prevalence rates are less, just as much, or more prevalent in other populations. Studies investigating psychopathology in terrorism that have employed comparison groups include Merari’s (2010) investigation, where a range of clinical interviews and personality tests,6 were administered to failed suicide bombers and a control group of other terrorists and non-political criminals. Suicide bombers received significantly more diagnoses of avoidant and dependent personality disorders, depressive symptoms, and suicidal tendencies. They were also significantly less likely to have psychopathic tendencies and impulsive-unstable tendencies compared to the control group. Horgan et al. (2016) compared lone-actor terrorists to mass casualty offenders, identifying that lone actors present with a higher prevalence of delusional disorders, PTSD, schizophrenia, and traumatic brain injuries, whereas mass casualty offenders have higher recorded prevalence rates of ADHD, anxiety disorders, autism spectrum disorders, drug dependence, depression, OCD, and personality disorders. Corner et al. (2016) compared lone-actor terrorists to both group actor terrorists and the general population, identifying that lone-actor terrorists present with higher rates of autism spectrum disorders, delusional disorder, personality disorders, schizoaffective disorder, and schizophrenia than the general population. Dhumad et al. (2020) compared terrorists to community controls and convicted murderers, identifying that the terrorist sample presented with conduct disorder and anti-social personality disorder more often than community controls but less often than murderers. The results here demonstrate the heterogeneity of psychopathology across offenders and other population samples and are the first major step towards understanding the relevance of psychopathology in this domain. Collectively, as the above works have highlighted,7 there is no one factor in psychopathology that acts as a predominant driver for individuals who engage in terrorism. This should not be a surprise. The fields of psychopathology and personality are vast and conflicted, and they have yet to reach a consensus on a wide range of non-criminal and criminal behaviours. Therefore, it is not surprising that the application of these disciplines to terrorist behaviour mirrors that conflict.

From presence to relevance To date, empirical efforts to understand the role of psychopathology in terrorism have been largely guided by the taxonomic approach. This approach, while a necessary first step towards progress in any scientific field (Bailey, 1994), has some important limitations. While well-designed and validated taxonomic research is able to provide useful classifications and systematic categorisation in an emerging area of research, the purpose remains organisational: a typology describes what a phenomenon is, but it does very little to explain why it is so (Bouhana et al., 2022). The above investigations each have value, as they identify diagnoses and traits that often co-occur with 184

Radicalisation and psychopathology

terrorist behaviour, and the collective conclusions of the works in the current paradigm fundamentally differ from the early paradigms, as they highlight the diversity rather than the homogeneity of terrorism. The current lack of a discrete terrorist profile is not because of a lack of interest or effort. Instead, it is a distinct empirical finding. However, what is now needed, as emphasised in the reviews of paradigm three, is a move from description to explanation and an exploration of the causal processes that account for the outcome under scrutiny. We need to develop a more holistic understanding of the patterning of interactions between factors of radicalisation, a move from identifying presence to identifying relevance. The most recent works within the current paradigm follow the conclusions of this research inquiry, and now largely agree on two core developmental concepts, equifinality and multifinality. With regard to equifinality, there are multiple pathways into radicalisation and subsequent terrorist behaviour. Individuals with very different initial situations will experience a wide range of factors8 that interact to contribute to each individual’s pathway. The principle of multifinality notes that different people with similar initial situations may experience different outcomes, and the impact of experiencing a single factor may impact an individual’s development in very different ways. Translating these concepts to relate to the role of psychopathology in terrorism, it is extremely unlikely that psychopathology alone is the cause of radicalisation or terrorist behaviour. Even for those who, at some point, suffer from mental health problems and ultimately become radicalised and enact terrorism, the role of these disorders or traits will differ across cases, and they will be mediated by a wide range of other experiences and situations. A single specific disorder will likely have a different impact on different people. Sometimes the experience of mental health problems may compound other problems. Sometimes other problems may compound the mental health problems. It is also possible that underlying psychopathology mediates other factors and moves individuals away from radicalisation, or there may be differences in the specific disorders and traits between those who become radicalised and those who act violently. What is needed now is the development of a more holistic understanding of the patterning of interactions between factors of radicalisation, a move from identifying presence to identifying relevance. This can be achieved by shifting our academic focus in three key areas: moving from static to dynamic analyses, including the pre-violence space, and expanding our focus on mental health.

From static to dynamic While the aforementioned studies focused on establishing the presence of psychopathology in offenders, others have aimed to identify behavioural associations to understand the psychopathological mechanisms, thus seeking to identify relevance. In this area, research typically follows the case study/vignette route (e.g. see Bubolz & Simi, 2019; Erlandsson & Meloy, 2018; Faccini & Allely, 2017; Gill, 2015; Hemmingby & Bjørgo, 2018; Holt et al., 2018; Inderberg et al., 2019). Many studies examine the impact of psychopathology on the adoption of radical and terrorist views. However, some studies go into further detail, examining the functional link between specific disorders and symptoms and terrorist involvement. Bubolz and Simi (2019) conducted life history interviews with a sample of former violent US White supremacists. The results that provided insight into the development of psychopathology were linked with involvement in the movement. The authors concluded that ‘there may be a type of synergy that exists between unstable individuals and movements that valorize violence as a preferred mode of expression’ (p. 11). Faccini and Allely (2017) outlined a number of case studies demonstrating the functional links of autism spectrum disorders and terrorist threats and engaging in terrorism. The authors concluded that being diagnosed with an autism spectrum disorder does not, in and of itself, satisfactorily 185

Emily Corner

explain how someone progresses to conduct an act of terrorism. Inderberg et al. (2019) highlighted three vignettes detailing the impact of autism spectrum disorders and psychosis on attraction to extreme and violent ideologies. The authors concluded that the combination of both disorders exacerbates both conditions, which may provoke a collapse of cognitive functioning, impulse control, and the ability to determine the dangerousness/offensiveness of situations. Far less efforts to establish relevance of psychopathology have been made in large-N studies. One exception is Corner et al. (2019), who performed probability-based analyses to quantitatively identify life course directions of those who conducted lone-actor terrorist acts. The results highlighted that psychopathology may be both a catalyst and an inhibitor for moving towards radicalisation. In some instances, following experiencing psychological distress, it is the move towards substance abuse that appears to lead an individual towards violence. This violence then leads to an arrest, which then eventually moves an individual towards a religious conversion and the expression of a radical ideology. A higher probability sequence suggested that psychopathology may also play an inhibitory role in the radicalisation process. It is only following psychiatric treatment that individuals isolate themselves from previous social networks. Either this isolation follows a direct route to expression of a radical ideology or the expression follows unemployment and an outward intensification of ideology and religion. In a further life course study, Corner and Gill (2019) examined whether involvement in terrorism increased the likelihood of developing psychopathology. The results highlighted that terrorists who reported experiencing psychopathology were also significantly more likely to report experiencing a wide range of stressors, and it was the combination of these stressors that had long-term consequences for their well-being. Each of these studies has empirically demonstrated the importance of moving beyond static interpretations of the prevalence of psychopathology and investigating the interactions between the dynamic processes in the lead-up to terrorist violence. However, these studies continue to rely on the measurement and comparison of the prevalence of a range of behaviours and experiences and have tended to focus on those who conducted, or at least attempted to conduct, violence. These studies are therefore not able to present the most holistic picture of the role of psychopathology in the movement towards radicalisation. To answer this inquiry, we can look to those who do not go on to commit violence, specifically individuals radicalised by, or attitudinally affiliated9 with, a violent cause.

The pre-violence space Solely focusing on those who engage in violence on behalf of a political or religious cause narrows our understanding of the relationship between psychopathology and extreme violence. Studies focusing on individuals who hold attitudinal affinity with violent radical causes are an emerging literature field and have huge potential for expanding our understanding of the relevance of psychopathology in radicalisation and terrorist behaviour. In these works, the focus of inquiry rests on individuals who hold radical beliefs and ideals but have not yet espoused any violent ideation in support of such beliefs. These studies further highlight the importance of disaggregating psychopathology to include personality traits, along with several other personal, situational, and attitudinal measures. Recent interrogation of terrorist offenders who are not motivated to commit violence is one stream of research in this area. Knight et al. (2017) compared 24 violent and 16 non-violent terrorist offenders in the United Kingdom. Knight et al. found no significant differences across a range of personality traits (fantasist/narcissism, obsessiveness, paranoia, grandiosity, irrationality, and delusional) and specified mental disorders (autism spectrum disorders, depression, personality 186

Radicalisation and psychopathology

disorders, and schizophrenia). Rates of substance abuse were markedly higher in violent actors, and unspecified psychopathology was noted as more prevalent in non-violent actors, but this did not yield significant values. Other investigations rely on general population samples, measuring the espoused views of individuals across a range of backgrounds. Victoroff et al. (2010) used self-report data to study traits of depression and anxiety and indicators of aggression in supporters and non-supporters of militant groups in Gaza. The results highlighted that 13% of participants self-reported traits of severe depression and 23% of severe anxiety. Further, those who scored highly on the depression and anxiety inventories also scored significantly higher on trait aggression. Those who reported anxiety also scored significantly higher on support for ‘religio-political aggression’.10 Stankov et al. (2010) compared personality traits against the proviolence, vile world, divine power scales of their militant extremist mindset scale in a sample of 2,424 students. Psychoticism was the only personality trait tested11 that held a positive correlation with proviolence. Psychoticism also held the strongest positive correlation with divine power (enhanced religiosity) and the vile world (negative views of the current world). Coid et al. (2016) compared psychiatric morbidity in 3,679 British males espousing pro- or anti-British extremist or neutral sentiments. Those who espoused neutral sentiments were significantly more likely to be noted as depressed, and those who espoused extremist sentiments were significantly more likely to have anti-social personality disorder. Nivette et al. (2017) developed a four-item violent extremist attitudes scale that was deployed to 1,288 adolescents in Switzerland. Personal strain12 was associated with significantly higher support for violent extremism. However, this result explained little of the variance within the sample and largely disappeared following the inclusion of other social and individual variables. Soliman et al. (2016) conducted structural equation modelling on eight self-reported measures examining three categories: social, cognitive, and psychopathological13 in 662 Egyptian nationals. They found that no one category was able to explain radical expression when tested individually; however, when all three categories were combined, the model produced a better fit. Each of these studies highlights the complexity of the formation of radical views, and collectively highlights that they cannot be predicted by one factor alone. In a series of studies examining sympathies for violent protest and terrorism, Bhui and colleagues developed (Bhui, Warfa et al., 2014), refined (Bhui et al., 2014; Bhui et al., 2016), and validated (Bhui et al., 2020) a scale to measure radicalisation. The sympathies for violent protest and extremism scale are made up of 16 questions. Bhui, Warfa et al. (2014) identified that people with self-reported poor health were significantly less likely to show support for violent protest and terrorism and that depression and anxiety scores showed no significant differences. Bhui, Everitt et al. (2014) moved the analysis forward, refining the items (following input from focus groups) within the scale. Bhui, Everitt et al. (2014) found that those most sympathetic to using violence were significantly more likely to also report mild depressive symptoms. There were no significant differences in anxiety scores. Bhui et al. (2020) expanded the analyses to include a wider range of psychopathological diagnoses. Sympathies were positively associated with co-morbid major depression and dysthymia, and with co-morbid PTSD and anxiety symptoms. However, this relationship was not identified in those with autism spectrum or personality disorders. Research in this area is rapidly expanding our understanding of the relevance of psychopathology in radical belief formation. In some instances, some diagnoses or traits appear to be related to radical beliefs and attitudinal affinity. However, each study is characterised by the complexity of outcomes, once again confirming the importance of the holistic combination of factors, with psychopathology being just one. Despite helping develop our understanding of the process of radicalisation across a range of settings, these studies still focus on tangible traits and diagnoses of 187

Emily Corner

psychopathology, which continues to restrict our understanding of the relevance of psychopathology in these processes.

From psychopathology to ‘needs’ Within criminology, research has long shifted from examining the presence (or absence) of factors. Instead, life- course research has identified multiple interrelated and dependent events that influence offending behaviour. Research has consistently demonstrated the importance of the combination and interaction of multiple factors. Examination of different disorders, situations, and demographics, along with unique experiences, has offered more rounded answers regarding the attribution of mental disorders to criminal and violent behaviour. This includes childhood experiences and abuse (Monahan et al., 2001), parenting practices (Monahan et al., 2001), age and employment (Uggen, 2000), marriage and spousal choice (Van Schellen et al., 2012), mental disorder and non-compliance to medication (McFarland et al., 1989), mental disorder and homelessness (Martell, 1991), mental disorder and prior arrests (Shore et al., 1989), and neighbourhood context (Monahan et al., 2001). As the aforementioned studies show, psychopathology is likely to co-occur alongside a range of other psychological, social, and situational factors. Rather than an overt focus on mental disorders, practice is moving to critically examine a range of presenting needs exhibited by individuals prior to radicalisation. Indeed, the National Health Service (NHS) defines a cluster of areas that impact on, and are impacted by, mental well-being: personal lives and relationships; discrimination; money/work/housing; life changes; physical health issues; traumatic events; and smoking/drinking/gambling/drug misuse (Gill et al., 2021). This perspective is starting to creep into research, indeed Gøtzsche-Astrup and Lindekilde (2019, p. 984) implored: Rather than asking if mental illness is or is not present in radicalization and terrorism, we ought to look at the range of functioning from adaptive to maladaptive behavior. Furthermore, the dispositional and social explanations should not be seen as competing perspectives, but rather as parts of a larger whole. In their comprehensive systematic review, Gill et al. (2021) provide insight into a wide range of studies that have examined the areas defined by the NHS as interrelated to mental health. However, much like the majority of the studies examining the presence of psychopathology, these investigations focus on the prevalence of a number of factors within each area, with little attention paid as to the relationship between factors and psychopathology and how these factors interact in the move towards radicalisation and through to terrorist violence.14 In the ongoing endeavour to advance our knowledge of the relationship between psychopathology, radicalisation, and terrorism and understand the relevance of psychopathology, the natural next step is to interrogate this research area.

Conclusion From a position of the false dichotomy and a lack of data almost a decade ago, a wide range of studies now highlight a presence of psychopathology and other co-occurring needs in radical and terrorist populations. The scope of these studies is far broader than this chapter, which has drawn continual focus to psychopathology alone. Overall, although improvements can be made in the inclusion of standardised psychological tests, definitions, and moving the focus to relevance, the outcomes above suggest that improving our knowledge in this space is a worthwhile pursuit. 188

Radicalisation and psychopathology

As a distinct group, terrorists often present with similar behaviours, yet the above research has highlighted how they develop from a wide range of pathways, some of which may involve psychopathology. Treating radicals and terrorists as a homogeneous group because of their presenting behaviours and instituting a small number of interventions will always be insufficient. There are a multitude of reasons for individuals engaging in radical and terrorist behaviours, including curiosity, fixation, identity-seeking, peer pressure, relative deprivation, social isolation, and a range of other factors both related and unrelated to psychopathology. It is not surprising that as interventions become more common, debates proliferate in the mental health professions about the role such practitioners should play in countering terrorism (Augestad Knudsen, 2020; Bhui & Jones, 2017; Dom et al., 2018; Khoshnood, 2017; McGarry, 2016; Weine & Kansal, 2019). Preventing individuals with underlying psychopathology from becoming radicalised or engaging in terrorism will always necessitate unique tailoring rather than broad, generalised interventions and a focus on matching treatment to the needs of the individual under scrutiny.

Notes 1 Personality is defined as the patterns of thinking, perceiving, reacting, and relating. These patterns are relatively stable over time. Personality disorders are defined as rigid, maladaptive traits that impair interpersonal functioning (Zimmerman, 2022). 2 In studies with a methodological quality score that reflects ‘Well conducted case control or cohort studies with a low risk of confounding or bias and a moderate probability that the relationship is causal’ (Corner et al., 2021). 3 Mirroring the initial pursuit of exploring psychopathology in wider criminal and violent behaviours in works carried out in criminology. 4 As Gill et al. (2019) and Gill et al. (2021) note, open-source datasets are all created differently, so to capture the accuracy of data, care must be taken in determining the provenance, depth of sources, and resourcing in coding processes. 5 A practice that has been long disregarded in psychology and criminology. 6 Including the Rorschach Inkblot Test (Rorschach, 1951), the Thematic Apperception Test (TAT; Murray, 1943), the House-Tree-Person Drawing Test (Buck, 1948), and a shortened (300-item) version of the California Personality Inventory (CPI; Gough, 1956). 7 As it should be expected. 8 These factors and their relative causal weight differ widely between individuals. 9 Those who hold affinity with those who espouse radical attitudes or intent. 10 Measured using statements; ‘Religious ends justify any means’, and ‘Harming civilians is a justifiable tool in the Muslim arsenal’. 11 Other traits examined included conscientiousness, honesty/propriety, agreeableness, resiliency, extraversion, and originality/intellect. 12 Including measures of personal stressors, negative life events, and prior admissions to psychiatric facilities 13 Activism-radicalism intention scale (Moskalenko & McCauley, 2009), The short Coolidge axis II inventory (Coolidge et al., 2010), Cognitive complexity instrument (Bagdasarov, 2009), Intolerance of uncertainty scale, short form (Carleton et al., 2007), Rational decision-making style (Scott & Bruce, 1995), Cognitive style index (Allinson & Hayes, 1996), The frustration-discomfort scale (Harrington, 2005), and Need to belong scale (Leary et al., 2013). 14 One exception to this is the aforementioned work of Bubolz and Simi (2019).

References Abrahms, M. (2011). Does terrorism really work? Evolution in the conventional wisdom since 9/11. Defence and Peace Economics, 22(6), 583–594. https://doi.org/10.1080/10242694.2011.635954 Akhtar, S. (1999). The psychodynamic dimension of terrorism. Psychiatric Annals, 29(6), 350–355. https:// doi.org/10.3928/0048-5713-19990601-09

189

Emily Corner Allinson, C. W., & Hayes, J. (1996). The cognitive style index: A measure of intuition‐analysis for organizational research. Journal of Management Studies, 33(1), 119–135. Altier, M. B., Leonard Boyle, E., Shortland, N., & Horgan, J. (2017). Why they leave: An analysis of terrorist disengagement events from eighty-seven autobiographical accounts. Security Studies, 26(2), 305–332. Arjona, A., & Kalyvas, S. N. (2009). Rebelling against rebellion: Comparing insurgent and counterinsurgent recruitment. Paper presented at human security and ethnicity workshop: Mobilisation for political violence: What do we know? Centre for Research on Inequality. Augestad Knudsen, R. (2020). Measuring radicalisation: Risk assessment conceptualisations and practice in England and Wales. Behavioral Sciences of Terrorism and Political Aggression, 12(1), 37–54. https://doi. org/10.1080/19434472.2018.1509105 Bagdasarov, Z. (2009). Developing and validating a measure of cognitive complexity: The role of cognitive complexity in processing of health messages. Rutgers The State University of New Jersey, School of Graduate Studies. Bailey, K. D. (1994). Typologies and taxonomies: An introduction to classification techniques. Sage. Bakker, E. (2006). Jihadi terrorists in Europe: Their characteristics and the circumstances in which they joined the jihad: An exploratory study. Netherlands Institute of International Relations Clingendael. Bartlett, J., & Miller, C. (2012). The edge of violence: Towards telling the difference between violent and non-violent radicalization. Terrorism and Political Violence, 24(1), 1–21. https://doi.org/10.1080/09546 553.2011.594923 Baruch, E. H. (2003). Psychoanalysis and terrorism: The need for a global talking cure. Psychoanalytic Psychology, 20(4), 698–700. https://doi.org/10.1037/0736-9735.20.4.698 Bélanger, J. J., Caouette, J., Sharvit, K., & Dugas, M. (2014). The psychology of martyrdom: Making the ultimate sacrifice in the name of a cause. Journal of Personality and Social Psychology, 107(3), 494–515. https://doi.org/10.1037/a0036855 Berko, A. (2007). The path to paradise: The inner world of suicide bombers and their dispatches. Praeger Publishers. Bhui, K., Everitt, B., & Jones, E. (2014). Might depression, psychosocial adversity, and limited social assets explain vulnerability to and resistance against violent radicalisation? PloS One, 9(9), e105918. https://doi. org/10.1371/journal.pone.0105918 Bhui, K., & Jones, E. (2017). The challenge of radicalisation: A public health approach to understanding and intervention. Psychoanalytic Psychotherapy, 31(4), 401–410. https://doi.org/10.1080/02668734.2017.1354908 Bhui, K., Otis, M., Silva, M. J., Halvorsud, K., Freestone, M., & Jones, E. (2020). Extremism and common mental illness: Cross-sectional community survey of white British and Pakistani men and women living in England. The British Journal of Psychiatry, 217, 547–554. https://doi.org/10.1192/bjp.2019.14 Bhui, K., Silva, M. J., Topciu, R. A., & Jones, E. (2016). Pathways to sympathies for violent protest and terrorism. The British Journal of Psychiatry, 209(6), 483–490. https://doi.org/10.1192/bjp.bp.116.185173 Bhui, K., Warfa, N., & Jones, E. (2014). Is violent radicalisation associated with poverty, migration, poor selfreported health and common mental disorders? PloS One, 9(3), e90718. https://doi.org/10.1371/journal. pone.0090718 Billig, O. (1985). The lawyer terrorist and his comrades. Political Psychology, 6(1), 29–46. https://doi. org/10.2307/3791269 Böckler, N., Leuschner, V., Zick, A., & Scheithauer, H. (2018). Same but different? Developmental pathways to demonstrative targeted attacks – Qualitative case analyses of adolescent and young adult perpetrators of targeted school attacks and Jihadi terrorist attacks in Germany. International Journal of Developmental Science, 174(4), 5–24. https://doi.org/10.3233/DEV-180255 Borum, R. (2004). Psychology of terrorism. University of South Florida. Bouhana, N., Corner, E., & Gill, P. (2022). A risk analysis framework of lone-actor terrorism. In J. C. Holzer, A. J. Dew, P. R. Recupero, & P. Gill (Eds.), Lone-actor terrorism: An integrated framework (pp. 287– 303). Oxford University Press. Bubolz, B. F., & Simi, P. (2019). The problem of overgeneralization: The case of mental health problems and US violent white supremacists. American Behavioral Scientist, 1–17. https://doi.org/10.1177/ 0002764219831746 Buck, J. N. (1948). The H-T-P technique. A qualitative and quantitative scoring manual. Journal of Clinical Psychology, 4(4), 319–396. https://doi.org/10.1002/1097-4679(194810)4:4%3c317:AID-JCLP2270040402% 3e3.0.CO;2-6

190

Radicalisation and psychopathology Capellan, J. A. (2015). Lone wolf terrorist or deranged shooter? A study of ideological active shooter events in the United States, 1970–2014. Studies in Conflict & Terrorism, 38(6), 395–413. https://doi.org/10.108 0/1057610X.2015.1008341 Capellan, J. A., & Anisin, A. (2018). A  distinction without a difference? Examining the causal pathways behind ideologically motivated mass public shootings. Homicide Studies, 22(3), 235–255. https://doi. org/10.1177/1088767918770704 Carleton, R. N., Norton, M. P. J., & Asmundson, G. J. (2007). Fearing the unknown: A short version of the intolerance of uncertainty scale. Journal of Anxiety Disorders, 21(1), 105–117. Challacombe, D. J., & Lucas, P. A. (2019). Postdicting violence with sovereign citizen actors: An exploratory test of the TRAP-18. Journal of Threat Assessment and Management, 6(1), 51. https://doi.org/10.1037/ tam0000105 Chermak, S., & Gruenewald, J. A. (2015). Laying a foundation for the criminological examination of rightwing, left-wing, and Al Qaeda-inspired extremism in the United States. Terrorism and Political Violence, 27(1), 133–159. https://doi.org/10.1080/09546553.2014.975646 Coid, J. W., Bhui, K., MacManus, D., Kallis, C., Bebbington, P., & Ullrich, S. (2016). Extremism, religion and psychiatric morbidity in a population-based sample of young men. The British Journal of Psychiatry, 209(6), 491–497. https://doi.org/10.1192/bjp.bp.116.186510 Coolidge, F. L., Segal, D. L., Cahill, B. S., & Simenson, J. T. (2010). Psychometric properties of a brief inventory for the screening of personality disorders: The SCATI. Psychology and Psychotherapy: Theory, Research and Practice, 83(4), 395–405. Cooper, H. H. A. (1977). Whither now-terrorism on the brink. Chitty’s Law Journal, 25(6), 181. Cooper, H. H. A. (1978). Psychopath as terrorist. Legal Medical Quarterly, 2(4), 253–262. Corner, E., Bouhana, N., & Gill, P. (2019). The multifinality of vulnerability indicators in lone-actor terrorism. Psychology, Crime & Law, 25(2), 111–132. https://doi.org/10.1080/1068316X.2018.1503664 Corner, E., & Gill, P. (2015). A false dichotomy? Mental illness and lone-actor terrorism. Law and Human Behavior, 39(1), 23. https://doi.org/10.1037/lhb0000102 Corner, E., & Gill, P. (2019). Psychological distress, terrorist involvement and disengagement from terrorism: A sequence analysis approach. Journal of Quantitative Criminology, 36, 499–526. https://doi. org/10.1007/s10940-019-09420-1 Corner, E., & Gill, P. (2021). Psychological distress and terrorist engagement: Measuring, correlating, and sequencing its onset with negative life events, social factors, and protective factors. Transcultural Psychiatry, 58(5), 697–711. Corner, E., Gill, P., & Mason, O. (2016). Mental health disorders and the terrorist: A research note probing selection effects and disorder prevalence. Studies in Conflict & Terrorism, 39(6), 560–568. https://doi.org /10.1080/1057610X.2015.1120099 Corner, E., Gill, P., Schouten, R., & Farnham, F. (2018). Mental disorders, personality traits, and grievancefueled targeted violence: The evidence base and implications for research and practice. Journal of Personality Assessment, 100(5), 459–470. Corner, E., Taylor, H., Van Der Vegt, I., Salman, N., Rottweiler, B., Hetzel, F., Clemmow, C., Schulten, N., & Gill, P. (2021). Reviewing the links between violent extremism and personality, personality disorders, and psychopathy. The Journal of Forensic Psychiatry & Psychology, 32(3), 378–407. https://doi.org/10.1080 /14789949.2021.1884736 De Cataldo Neuberger, L., & Valentini, T. (1996). Woman and terrorism. Palgrave Macmillan. De Waele, M., & Pauwels, L. J. R. (2016). Why do Flemish youth participate in right-wing disruptive groups? In C. Maxson & F. Esbensen (Eds.), Gang transitions and transformations in an international context (pp. 173–200). Springer. DeMause, L. (2002). The childhood origins of terrorism. The Journal of Psychohistory, 29(4), 340–348. Desmarais, S., Simons-Rudolph, J., Brugh, C. S., Schilling, E., & Hoggan, C. (2017). The state of scientific knowledge regarding factors associated with terrorism. Journal of Threat Assessment and Management, 4(4), 180–209. Dhumad, S., Candilis, P. J., Cleary, S. D., Dyer, A. R., & Khalifa, N. (2020). Risk factors for terrorism: A comparison of family, childhood, and personality risk factors among Iraqi terrorists, murderers, and controls. Behavioral Sciences of Terrorism and Political Aggression, 12(1), 72–88. https://doi.org/10.1080/19434 472.2019.1591481

191

Emily Corner Dom, G., Schouler-Ocak, M., Bhui, K., Demunter, H., Kuey, L., Raballo, A., Frydecka, D., Misiak, B., Gorwood, P., & Samochowiec, J. (2018). Mass violence, radicalization and terrorism: A role for psychiatric profession? European Psychiatry, 49, 78–80. https://doi.org/10.1016/j.eurpsy.2018.01.001 Dunwoody, P. T., & McFarland, S. G. (2018). Support for anti – Muslim policies: The role of political traits and threat perception. Political Psychology, 39(1), 89–106. https://doi.org/10.1111/pops.12405 Egan, V., Cole, J., Cole, B., Alison, L., Alison, E., Waring, S., & Elntib, S. (2016). Can you identify violent extremists using a screening checklist and open-source intelligence alone? Journal of Threat Assessment and Management, 3(1), 21. https://doi.org/10.1037/tam0000058 Erlandsson, Å., & Meloy, J. R. (2018). The Swedish school attack in Trollhättan. Journal of Forensic Sciences, 63(6), 1917–1927. https://doi.org/10.1111/1556-4029.13800 Faccini, L., & Allely, C. (2017). Rare instances of individuals with autism supporting or engaging in terrorism. Journal of Intellectual Disabilities and Offending Behaviour, 8(2), 70–82. https://doi.org/10.1108/ JIDOB-11-2016-0022 Ferracuti, F., & Bruno, F. (1981). Psychiatric aspects of terrorism in Italy. In I. L. Barak Glantz & C. R. Huff (Eds.), The mad, the bad, and the different: Essays in honor of Simon Dinitz (pp. 199–213). Lexington Books. Gill, P. (2012). Terrorist violence and the contextual, facilitative and causal qualities of group-based behaviors. Aggression and Violent Behavior, 17(6), 565–574. https://doi.org/10.1016/j.avb.2012.08.002 Gill, P. (2015). Lone-actor terrorists: A behavioural analysis. Routledge. Gill, P., Clemmow, C., Hetzel, F., Rottweiler, B., Salman, N., Van Der Vegt, I., Marhcment, Z., Schumann, S., Zolghadriha, S., Schulten, N., Taylor, H., & Corner, E. (2021). Systematic review of mental health problems and violent extremism. The Journal of Forensic Psychiatry, 32(1), 51–78. https://doi.org/10.10 80/14789949.2020.1820067 Gill, P., Corner, E., McKee, A., Hitchen, P., & Betley, P. (2019). What do closed source data tell us about lone actor terrorist behavior? A research note. Terrorism and Political Violence, 1–18. https://doi.org/10.1080 /09546553.2019.1668781 Gill, P., Horgan, J., & Deckert, P. (2014). Bombing alone: Tracing the motivations and antecedent behaviors of lone-actor terrorists. Journal of Forensic Sciences, 59(2), 425–435. https://doi. org/10.1111/1556-4029.12312 Gøtzsche-Astrup, O. (2019). Personality moderates the relationship between uncertainty and political violence: Evidence from two large US samples. Personality and Individual Differences, 139, 102–109. https://doi.org/10.1016/j.paid.2018.11.006 Gøtzsche-Astrup, O. (2020). Pathways to violence: Do uncertainty and dark world perceptions increase intentions to engage in political violence? Behavioral Sciences of Terrorism and Political Aggression, 1–18. https://doi.org/10.1080/19434472.2020.1714693 Gøtzsche-Astrup, O., & Lindekilde, L. (2019). Either or? Reconciling findings on mental health and extremism using a dimensional rather than categorical paradigm. Journal of Forensic Sciences, 64(4), 982–988. https://doi.org/10.1111/1556-4029.14014 Gough, H. G. (1956). California psychological inventory. Consulting Psychologists Press. Gruenewald, J., Chermak, S., & Freilich, J. D. (2013). Distinguishing “loner” attacks from other domestic extremist violence: A comparison of far-right homicide incident and offender characteristics. Criminology & Public Policy, 12(1), 65–91. https://doi.org/10.1111/1745-9133.12008 Harrington, N. (2005). The frustration discomfort scale: Development and psychometric properties. Clinical Psychology & Psychotherapy: An International Journal of Theory & Practice, 12(5), 374–387. Hemmingby, C., & Bjørgo, T. (2018). Terrorism target selection: The case of Anders Behring Breivik. Perspectives on Terrorism, 12(6), 164–176. Hewitt, C. (2003). Understanding terrorism in America: From the Klan to al Qaeda. Psychology Press. Holt, T., Freilich, J. D., Chermak, S. J., & LaFree, G. (2018). Examining the utility of social control and social learning in the radicalization of violent and non-violent extremists. Dynamics of Asymmetric Conflict, 11(3), 125–148. https://doi.org/10.1080/17467586.2018.1470661 Horgan, J. (2003). The search for the terrorist personality. In A. Silke (Ed.), Terrorists, victims and society: Psychological perspectives on terrorism and its consequences (pp. 3–27). Wiley. Horgan, J. (2005). The psychology of terrorism. Routledge. Horgan, J., Alteir, M. B., Shortland, N., & Taylor, M. (2017). Walking away: The disengagement and deradicalization of a violent right-wing extremist. Behavioral Sciences of Terrorism and Political Aggression, 9(2), 63–77.

192

Radicalisation and psychopathology Horgan, J., Gill, P., Bouhana, N., Silver, J., & Corner, E. (2016). Across the universe? A comparative analysis of violent radicalization across three offender types with implications for criminal justice training and education. The Office of Justice Programs, NCJRS. Houssier, F. (2016). Terrorist acting out, narcissism and psychopathology of identifications. Soins Psychiatrie, 37(302), 26–28. https://doi.org/10.1016/j.spsy.2015.11.006 Inderberg, A. M. S., Horndalsveen, K., Elvehaug, A. H., Mehmi, Y., Jørstad, I., & Bakken, T. L. (2019). Autism, intellectual disabilities and additional psychosis, and affiliation to groups with violent ideology.  Journal of Intellectual Disabilities and Offending Behaviour, 10(1), 1–7. https://doi.org/10.1108/ JIDOB-09-2018-0010 Jäger, H., Schmidtchen, G., & Süllwold, L. (1981). Lebenslaufanalysen (Vol. 2). Westdeutscher Verlag. Jones, D. N. (2013). Psychopathy and Machiavellianism predict differences in racially motivated attitudes and their affiliations. Journal of Applied Social Psychology, 43(2), E367–E378. https://doi.org/10.1111/ jasp.12035 Kellen, K. (1982). On terrorists and terrorism: A RAND note. RAND. Kessler, R., & Üstün, T. B. (2008). The WHO world mental health surveys: Global perspectives on the epidemiology of mental disorders. Cambridge University Press. Khoshnood, A. (2017). The correlation between mental disorders and terrorism is weak. BJPsych Bulletin, 41(1), 56. https://doi.org/10.1192/pb.41.1.56 King, S., Endres, J., Schwaβ, M., Stemmler, M., Lauchs, L., Armborst, A., Scheithauer, H., Leuschner, V., Böckler, N., Akhgar, B., & Nitsch, H. (2018). Prisoners with Islamist relations: Are prisoner files a valuable data source for individual assessment and for research? International Journal of Developmental Science, 12(1–2), 129–141. https://doi.org/10.3233/DEV-170243 Knight, S., Woodward, K., & Lancaster, G. L. (2017). Violent versus nonviolent actors: An empirical study of different types of extremism. Journal of Threat Assessment and Management, 4(4), 230. https://doi. org/10.1037/tam0000086 Leary, M. R., Kelly, K. M., Cottrell, C. A., & Schreindorfer, L. S. (2013). Construct validity of the need to belong scale: Mapping the nomological network. Journal of Personality Assessment, 95(6), 610–624. Leistedt, S. J. (2013). Behavioural aspects of terrorism. Forensic Science International, 228(1–3), 21–27. https://doi.org/10.1016/j.forsciint.2013.02.004 Leygraf, N. (2014). On the phenomenology of Islamic terrorist offenders. Forensische Psychiatrie, Psychologie Kriminologie, 8(4), 237–245. https://doi.org/10.1007/s11757-014-0291–2 Liem, M., van Buuren, J., de Roy van Zuijdewijn, J., Schönberger, H., & Bakker, E. (2018). European lone actor terrorists versus “common” homicide offenders: An empirical analysis. Homicide Studies, 22(1), 45–69. https://doi.org/10.1177/1088767917736797 Lloyd, M., & Kleinot, P. (2017). Pathways into terrorism: The good, the bad and the ugly. Psychoanalytic Psychotherapy, 31(4), 367–377. https://doi.org/10.1080/02668734.2017.1360380 Lutz, J. M., & Lutz, B. (2013). Global terrorism. Routledge. Lyons, H. A., & Harbinson, H. J. (1986). A comparison of political and non-political murderers in Northern Ireland, 1974–84. Medicine, Science and the Law, 26(3), 193–198. https://doi. org/10.1177/002580248602600304 MacDonald, P. (2014). Narcissism in the modern world. Psychodynamic Practice, 20(2), 144–153. https:// doi.org/10.1080/14753634.2014.894225 Martell, D. A. (1991). Homeless mentally disordered offenders and violent crimes: Preliminary research findings. Law & Human Behavior, 15(4), 333–347. https://doi.org/10.1007/BF02074075 McFarland, B. H., Faulkner, L. R., Bloom, J. D., Hallaux, R., & Bray, J. D. (1989). Chronic mental illness and the criminal justice system. Hospital and Community Psychiatry, 40(7), 718–723. McGarry, P. J. (2016). Terrorism: It’s not mental illness – It’s politics. BJPsych Bulletin, 40(5), 285. https:// doi.org/10.1192/pb.40.5.285a Merari, A. (2010). Driven to death: Psychological and social aspects of suicide terrorism. Oxford University Press. Monahan, J., Steadman, H. J., Silver, E., Appelbaum, P. S., Clark Robbins, P., Mulvey, E. P., Roth, L. H., Grisso, T., & Banks, S. (2001). Rethinking risk assessment: The Macarthur study of mental disorder and violence. Oxford University Press. Moskalenko, S., & McCauley, C. (2009). Measuring political mobilization: The distinction between activism and radicalism. Terrorism and Political Violence, 21(2), 239–260. Murray, H. A. (1943). Thematic apperception test. Harvard University Press.

193

Emily Corner Nivette, A., Eisner, M., & Ribeaud, D. (2017). Developmental predictors of violent extremist attitudes: A test of general strain theory. Journal of Research in Crime and Delinquency, 54(6), 755–790. https://doi. org/10.1177/0022427817699035 Nussio, E. (2017). The role of sensation seeking in violent armed group participation. Terrorism and Political Violence, 32(1), 1–19. https://doi.org/10.1080/09546553.2017.1342633 O’Gorman, R. (2010). The evolutionary logic of terrorism: Understanding why terrorism is an inevitable human strategy in conflict. In A. Silke (Ed.), The psychology of counter-terrorism. Political violence (pp. 62–75). Routledge. Opoku-Agyemang, K. (2017). Narcissism over ideology: Revealed versus stated terrorist preferences. University of California. Pauwels, L. J., Brion, F., Schils, N., Lafinneur, J., Verhage, A., De Ruyver, B., & Easton, M. (2014). Explaining and understanding the role of exposure to new social media on violent extremism: An integrative quantitative and qualitative approach. Academia Press. Pauwels, L. J., & De Waele, M. (2014). Youth involvement in politically motivated violence: Why do social integration, perceived legitimacy, and perceived discrimination matter? International Journal of Conflict and Violence, 8(1), 135–153. ISSN: 1864–1385. Pauwels, L. J., & Hardyns, W. (2018). Endorsement for extremism, exposure to extremism via social media and self-reported political/religious aggression. International Journal of Developmental Sciences, 12(1– 2), 51–69. Pauwels, L. J., & Heylen, B. (2014). Perceived group threat, perceived injustice, and self-reported right-wing violence: An integrative approach to the explanation of right-wing violence. Journal of Interpersonal Violence, 35(21–22), 1–27. https://doi.org/10.1177/0886260517713711 Pauwels, L. J., Ljujic, V., & De Buck, A. (2020). Individual differences in political aggression: The role of social integration, perceived grievances and low self-control. European Journal of Criminology, 17(5), 603–627. https://doi.org/10.1177/1477370818819216 Pauwels, L. J., & Schils, N. (2016). Differential online exposure to extremist content and political violence: Testing the relative strength of social learning and competing perspectives. Terrorism and Political Violence, 28(1), 1–29. https://doi.org/10.1080/09546553.2013.876414 Pauwels, L. J., & Svensson, R. (2017). How robust is the moderating effect of extremist beliefs on the relationship between self-control and violent extremism? Crime and Delinquency, 63(8), 1000–1016. https:// doi.org/10.1177/0011128716687757 Pearce, K. (1977). Police negotiations. Canadian Psychiatric Association Journal, 22(4), 171–174. https:// doi.org/10.1177/070674377702200405 Pearlstein, R. M. (1991). The mind of the political terrorist. Scholarly Resources. Perry, G., Wikström, P.-O. H., Roman, G. D., Scheithauer, H., Leuschner, V., Böckler, N., Akhgar, B., & Nitsch, H. (2018). Differentiating right-wing extremism from potential for violent extremism: The role of criminogenic exposure. International Journal of Developmental Science, 12(1–2), 103–113. https://doi. org/10.3233/DEV-170240 Perry, S., Hasisi, B., & Perry, G. (2017). Who is the lone terrorist? A study of vehicle-borne attackers in Israel and the West Bank. Studies in Conflict & Terrorism, 41(11), 899–913. https://doi.org/10.1080/105 7610X.2017.1348101 Post, J. M. (1984). Notes on a psychodynamic theory of terrorist behavior. Terrorism, 7(2), 241–256. https:// doi.org/10.1080/10576108408435577 Post, J. M. (2005). Addressing the causes of terrorism: Psychology. Paper presented at the International Summit on Democracy, Terrorism and Security. Post, J. M. (2007). The mind of the terrorist: The psychology of terrorism from the IRA to al-Qaeda. Palgrave Macmillan. Rae, J. (2012). Will it ever be possible to profile the terrorist? Journal of Terrorism Research, 3(2), 64–74. https://doi.org/10.15664/jtr.380 Rorschach, H. (1951). Psychodiagnostics: A diagnostic test based on perception (5th ed.). Verlag Hans Huber. Schils, N., & Pauwels, L. (2016). Political violence and the mediating role of violent extremist propensities. Journal of Strategic Security, 9(2), 70–91. https://doi.org/10.5038/1944-0472.9.2.1491 Scott, S. G., & Bruce, R. A. (1995). Decision-making style: The development and assessment of a new measure. Educational and Psychological Measurement, 55(5), 818–831. Sher, L., & Rice, T. (2015). Prevention of homicidal behaviour in men with psychiatric disorders. The World Journal of Biological Psychiatry, 16(4), 212–229. https://doi.org/10.3109/15622975.2015.1028998

194

Radicalisation and psychopathology Shore, D., Filson, C. R., Johnson, W. E., Rae, D. S., Muehrer, P., Kelley, D. J., Davis, T. S., Waldman, I. N., & Wyatt, R. J. (1989). Murder and assault arrests of White House cases: Clinical and demographic correlates of violence and subsequent to civil commitment. American Journal of Psychiatry, 146(5), 645–651. https://doi.org/10.1176/ajp.146.5.645 Sikkens. (2018). Relating to radicalism: Family and upbringing experiences in radicalization and de-radical ization [PhD thesis]. Utrecht University. Silke, A. (1998). Cheshire-cat logic: The recurring theme of terrorist abnormality in psychological research. Psychology, Crime & Law, 4(1), 51–69. https://doi.org/10.1080/10683169808401747 Silke, A. (2003). Becoming a terrorist. In A. Silke (Ed.), Terrorists, victims and society: Psychological perspectives on terrorism and its consequences (pp. 29–54). Wiley. Soliman, A., Bellaj, T., & Khelifa, M. (2016). An integrative psychological model for radicalism: Evidence from structural equation modeling. Personality and Individual Differences,  95, 127–133. http://doi. org/10.1016/j.paid.2016.02.039 Spaaij, R. (2011). Understanding lone wolf terrorism: Global patterns, motivations and prevention. Springer. Stankov, L., Saucier, G., & Knežević, G. (2010). Militant extremist mind-set: Proviolence, vile world, and divine power. Psychological Assessment, 22(1), 70–86. https://doi.org/10.1037/a0016925 Stone, L., Waldron, E., & Nowak, H. (2020). Making a good mental health diagnosis: Science, art and ethics. Australian Journal of General Practice, 49(12), 797–802. https://doi.org/10.31128/AJGP-08-20-5606 Strentz, T. (1988). A  terrorist psychosocial profile: Past and present. FBI Law Enforcement Bulletin, 57, 13–19. ISSN: 0014 5688. Stys, Y., Gobeil, R., Harris, A. J. R., & Michel, S. (2014). Violent extremists in federal institutions – estimating radicalization and susceptibility to radicalization in the federal offender population. Correctional Service of Canada. Tanay, E. (1987). Pseudo-political terrorism. Journal of Forensic Sciences, 32(1), 192–200. https://doi. org/10.1520/JFS12342J Taylor, M., & Quayle, E. (1994). Terrorist lives. Brassey’s. Uggen, C. (2000). Work as a turning point in the life course of criminals: A duration model of age, employment, and recidivism. American Sociological Review, 65(4), 529–546. https://doi.org/10.2307/2657381 van Leyenhorst, M., & Andreas, A. (2017). Dutch suspects of terrorist activity: A study of their biographical backgrounds based on primary sources. Journal for Deradicalization, 12, 309–344. http://journals.sfu.ca/ jd/index.php/jd/article/view/119 Van Schellen, M., Apel, R., & Nieuwbeerta, P. (2012). “Because you’re mine, I walk the line”? Marriage, spousal criminality, and criminal offending over the life course. Journal of Quantitative Criminology, 28(4), 701–723. https://doi.org/10.1007/s10940-012-9174-x Victoroff, J. (2005). The mind of a terrorist: A review and critique of psychological approaches. Journal of Conflict Resolution, 49(1), 3–42. https://doi.org/10.1177/0022002704272040 Victoroff, J., Quota, S., Adelman, J. R., Celinska, B., Stern, N., Wilcox, R., & Sapolsky, R. M. (2010). Support for religio‐political aggression among teenaged boys in Gaza: Part I: Psychological findings. Aggressive Behavior, 36(4), 219–231. https://doi.org/10.1002/ab.20348 Weenink, A. W. (2015). Behavioral problems and disorders among radicals in police files. Perspectives on Terrorism, 9(2), 17–33. www.jstor.org/stable/26297357 Weenink, A. W. (2019). Adversity, criminality, and mental health problems in Jihadis in Dutch police files. Perspectives on Terrorism, 13(5), 130–142. www.jstor.org/stable/26798583 Weine, S., & Kansal, S. (2019). What should global mental health do about violent extremism? Global Mental Health, 6(14), 1–7. Zeman, T., Břeň, J., & Urban, R. (2018). Profile of a lone wolf terrorist: A crisis management perspective. Journal of Security & Sustainability Issues, 8(1), 1. https://doi.org/10.9770/jssi.2018.8.1(1) Zimmerman, M. (2022, September). Overview of personality disorders. MSD manual: Professional version. https://www.msdmanuals.com/en-au/professional/psychiatric-disorders/personality-disorders/overviewof-personality-disorders Zmigrod, L., Rentfrow, P. J., & Robbins, T. W. (2019). Cognitive inflexibility predicts extremist attitudes. Frontiers in Psychology, 10, 1–13. https://doi.org/10.3389/fpsyg.2019.00989

195

13 BELONGING IS JUST A CLICK AWAY Extremism, radicalisation, and the role of online communities Amarnath Amarasingam Introduction As the field of terrorism studies evolves, there continues to be increasing interest in how the Internet and social media platforms are impacting extremist groups, militant movements, and individual radicalisation to violence. Especially with the rise of the Islamic State, and the migration of tens of thousands of foreign fighters to Syria and Iraq, there was urgent academic and policy concern about how ISIS’s glossy magazines, high-definition videos, and social media presence were contributing to this sudden spike in jihadist mobilisation around the world (Rudner, 2017; Barrett, 2017; Dawson & Amarasingam, 2017). Policy discussions soon followed, with substantial resources going into countering violent extremism (CVE) strategies (Vidino & Hughes, 2015; Berger & Perez, 2016; Winter et al., 2020; Clifford et al., 2020). Focus was also placed on deplatforming extremist content as well as new initiatives by social media companies to work together to ensure that their platforms were not exploited for harmful purposes (Amarasingam, 2021; GIFCT, 2017; Meleagrou-Hitchens & Kaderbhai, 2017). Other researchers looked more closely at the nexus between radicalisation and online extremist content (Koehler, 2014; Hassan et al., 2018; Odag et al., 2019) and how online networks are formed and developed among extremists (Ferrara, 2017; Veilleux-Lepage & Archambault, 2019; Caiani & Wagemann, 2009). This rush to research, though, has produced staggered results: some regions have been ignored, and propaganda in languages other than Arabic and English remains understudied. For instance, based on a rudimentary Google Scholar search, there are over 50 articles on Dabiq and Rumiyah, the Islamic State’s English-language online magazine. Commendable work thus far has analysed visual strategies (Abdelrahim, 2019), the portrayal of particular countries in these magazines (Stempien, 2019), images of children (Kaczkowski, 2019), and calls to violence (Lakomy, 2019), among a host of other topics. Similarly, there have been numerous studies examining extremist online support networks as well as the impacts of deplatforming on these networks (Berger & Perez, 2016; Amarasingam et al., 2021; Pearson, 2018; Chandrasekharan et al., 2017; Conway et al., 2021; Alexander, 2017). What remains relatively understudied, however, is the importance of community for individuals who are radicalising in the online space. This is an area of research that is beginning to receive further attention and is coming to be seen as an important area that warrants further investigation. While numerous articles mention ‘online community’ or ‘virtual DOI: 10.4324/9781003035848-15 196

Belonging is just a click away

community’ in passing, there are only a handful of studies truly unpacking the concept or exploring its significance in the field. There are also different terminologies used in publications attempting to capture the phenomenon: online ecology (Conway et al., 2021), radical milieu (Conway, 2012), collective identity (Futrell & Simi, 2004), as well as cyberculture (Simi & Futrell, 2006). For this chapter, I will be using virtual community and online community interchangeably, and examining, first, the literature on online community that exists outside the field of terrorism studies. Second, I explore some of the ways that extremism researchers have thus far deployed the concept in their work. The chapter concludes with some recommendations for future research. I argue that, going forward, there needs to be much more cross pollination of these bodies of work for two reasons: first, the research on online communities complicates what we know about how extremists use the online space and, second, the research on extremism complicates the literature on online communities. For instance, some work on virtual communities suggests that interpersonal trust is less necessary when communities exist online. For extremist communities, however, this is not true. Similarly, research in online communities suggests that most people go into online communities to experiment with different identities and be creative about who they are and can be. As Turkle (1999, pp. 643–644) notes, ‘For many people, joining online communities means crossing a boundary into highly charged territory. Some feel an uncomfortable sense of fragmentation, some a sense of relief. Some sense the possibilities for self-discovery’. For extremists, as I show below, the opposite is often true. It is in their real life that they are pretending to be who they are not, and it is within transnational communities in the online space that they truly express their beliefs and form trusted relationships (Amarasingam, 2015).

What is a community? Those working in terrorism studies inevitably hear the words ‘community’ and ‘belonging’ quite often. When journalists ask former extremists questions about how they joined jihadist groups or far-right groups, it usually only takes a few minutes before they mention that they were ‘in search of belonging’ or ‘looking for community’ (Kelsey-Sugg, 2019). So, what exactly is community? While the term itself is heavily debated by sociologists, a shorter history of how definitions have evolved should suffice for our purposes. For the ancient Greeks, especially in the work of Aristotle, the notion of community was seen as synonymous with society, pictured in political and contractual terms. As Delanty (2010, p. 2) notes, from ‘the ancient Greeks to the Enlightenment, community expressed the essence of society, not its antithesis’. With the Enlightenment, the community was envisioned as lying outside the state and political realities. In this sense, ‘community expressed a dream impossible to realize: a vision of a pure or pristine social bond that did not need a state’ or government (Delanty, 2010, p. 3). Aspects of modern thought have imagined the state and government as a kind of necessary evil, to control and keep humankind in line through laws and tactics of control. Community, in this thinking, is imagined as something outside the state, more pure, more personal, and exists as a ‘perpetual critique of the state’ (Delanty, 2010, p. 3). While space constraints prevent a complete discussion of the theoretical and definitional debates around the notion of community, there is one conceptual advancement that demands mention, and forms the foundation for our discussion below on virtual community. This is Benedict Anderson’s (1983) idea of ‘imagined communities’. While Anderson’s study is first and foremost about nationalism and arguing against the nation as some sort of ‘natural’ group, his work has been extensively used to help understand everything from identity to social integration, archives, and more. As Anderson (1983, p. 6) notes, nations and communities are imagined because ‘the members of even the smallest nation will never know most of their fellow-members, meet them, 197

Amarnath Amarasingam

or even hear of them, yet in the minds of each lives the image of their communion’. Anderson argues that symbolic attachment to a community does not have to be local or face-to-face and does not have to be based on interdependence. Rather, they can be entirely abstract and global. Muslim youth in Paris and Peshawar can imagine themselves as part of a global religious community, Dalit activists in India and in the Indian diaspora can be part of a transnational anti-caste activist community, and similar ideological communities can form around women’s rights, environmental activism, or human rights. As Delanty (2010, p. 137) notes, ‘No community, not even a traditional, occupationally-based one such as that of coalminers or a rural community, could exist as a community without a capacity to symbolically imagine themselves as a community’. It is by internalising Anderson’s notion of communities and nations as imagined that we get closer to understanding some of the debates around online communities. More broadly, it is also important to note some of the key debates and multidisciplinary developments in the conceptualisation of community. Bruhn (2011) and Hunter (2007) both offer strong overviews of the conceptions of community that are rooted in a variety of geographies, disciplines, and contexts. During the mid-20th century, scholars shifted from a Chicago School conception, which framed urbanisation as having led to a loss of community (Hunter, 2007), to an understanding of how communities exist in inner-city and suburban communities (Bruhn, 2011). Hunter (2007) also notes that technology and globalisation in the 1980s and 1990s ‘liberated’ the community, freeing it from the historical constraints of space. Bruhn makes note of Barry Wellman’s work that characterises communities as networks. He explains that as networks have a high density of ties that are directly tied to each other, communities function in the same way, with most people being members of multiple communities. The community has also changed in the context of rationalisation, as demonstrated by Bakardjieva and Feenberg (2002), who use the term ‘democratic rationalization’. This concept refers to users in virtual communities intervening to challenge unfair power structures and barriers to communication, which can force design changes and greater user collaboration (186–7). Hunter (2007) further notes the importance of rationalisation when looking at the theory of communities of limited liability. This point argues that individuals will rationally invest in their local communities to the limited degree that they perceive they are receiving valued benefits from their engagement (Hunter, 2007, p. 26).

What is an online community? Howard Rheingold’s book The Virtual Community (1993) was the first major study of the notion of community in the online space and largely formed the foundation for many subsequent treatments of the phenomenon. The state of the Internet in 1993 was, of course, entirely different from what it is now, but the work remains an important milestone in helping us rethink belonging, relationships, communication, and space in virtual domains. Since Rheingold, two works on virtual communities of note are Ridings and Gefen (2006) and Feenberg and Bakardjieva (2004). The former broadly characterises virtual communities as ‘ “groups of people with common interests and practices that communicate regularly and for some duration in an organized way over the Internet through a common location or mechanism’ (Ridings & Gefen, 2006, p. 4) and further argues that people join communities for one or more of several reasons: (1) information and knowledge exchange, (2) the exchange of social support, (3) friendship, and (4) recreation. In Feenberg and Bakardijeva’s conceptualisation of virtual communities, they rely on Benedict Anderson’s idea of ‘imagined communities’ (2004, p. 37). As a result, they argue that virtuality is a perfectly normal aspect of communities (Feenberg & Bakardjieva, 2004, p. 37). Other scholars such as Zizi Papacharissi and danah boyd use conceptualisations aside from virtual communities. Papacharissi uses the term 198

Belonging is just a click away

‘affective publics’ to understand how public formations are mobilised, connected, and disconnected through ‘expressions of sentiment’ (2015, p. 125). Boyd similarly refers to ‘networked publics’, which are ‘publics that are restructured by networked technologies’ (2011, p. 39). This results in an imagined collective at the intersection of people, technology, and practice (boyd, 2011, p. 39). In terms of methodologies when studying online communities, Christine Hine’s and Sarah Pink’s respective works have much to offer. Hine (2000) described a virtual ethnography as a site for interaction, while noting that one should not assume that any single online domain would be distinguishable from the ‘real’ world. In contrast to the dominant literature, Postill and Pink (2012) argue that ethnographers of digital worlds should move away from the typical paradigms of networks and communities. Instead, they propose that engaging with concepts of routine, movement, and sociality enables researchers to better understand the ethnography of virtual worlds (123). It is also important to acknowledge that approaches to online ethnography have developed in tandem with the Internet itself (Robinson & Schulz, 2009). Much of the research on the impact of the Internet on human behaviour also cautions us against interpreting some digital activities as entirely different from real-world activities. As Cowan (2005, p. 51) notes, ‘in most respects what is commonly called “virtual reality” is at best a cyber-shadow, an electronic reflection of real life. Shopping online may offer consumers more convenience, for example, but presents them with little more than catalogues accessed electronically’. Similarly, most of the emails and text messages we send are to family, friends, and coworkers, with whom we already ‘share a common lifeworld’ (Delanty, 2010, p. 145). This idea is supported by Schroeder, who questions the idea that online worlds create communities (2011, p. 151). He argues that online worlds would be communities if ‘community’ simply meant a shared interest, but most online worlds largely consist of masses of undifferentiated users and encounters between strangers. One of the ongoing debates among those who study online communities, then, is whether what we are seeing is merely an electronic extension of people’s real lifeworlds, as well as whether online-only communities truly evince a ‘sense of community’ as traditionally understood. It is to these questions that we now turn. In 1974, Sarason first expounded on the psychological benefits for individuals when they feel a ‘sense of community’. In 1986, McMillan and Chavis proposed four criteria for understanding, and perhaps measuring, the ‘sense of community’ that individuals experience: membership, influence, integration and fulfilment of needs, and shared emotional connection. Membership, they argue, is a sense that an individual has invested something of themselves in the community and thus has a right to be part of it. Membership is also closely tied to ideas of boundaries and emotional safety. Boundaries define clearly who is part of the community and who is not, and these boundaries are often mediated by a ‘common symbol system’ (McMillan & Chavis, 1986, p. 10). By influence, McMillan and Chavis mean that ‘members are more attracted to a community in which they feel that they are influential’, and that there is a harmonious relationship between an individual’s feeling that they have some influence in the group and the group’s need for members to conform (1986, p. 12). The third component of ‘sense of community’ discussed by McMillan and Chavis is integration and fulfilment of needs. By this, they mean that ‘for any group to maintain a positive sense of togetherness, the individual-group association must be rewarding for its members’ (1986, p. 12). These rewards could be rooted in status, competence of other members, and shared values, which ‘foster the belief that in joining together they might be better able to satisfy these needs and obtain the reinforcement they seek’ (McMillan & Chavis, 1986, p. 13). The final element, a shared emotional connection, often has to do with shared history, frequency of contact, the quality of these 199

Amarnath Amarasingam

interactions, a spiritual bond, as well as a shared experience of moments of crisis. As McMillan and Chavis (1986, p. 15) conclude, a strong sense of community arises when members are given ‘positive ways to interact, important events to share and ways to resolve them positively, opportunities to honor members, opportunities to invest in the community, and opportunities to experience a spiritual bond among members’. Subsequent research, which we do not need to explore in full here, has operationalised different aspects of the four elements discussed earlier, nuanced certain elements, and added more (Perkins & Long, 2002; Fisher & Sonn, 2007, 2002). For our purposes, though, the important question is whether a true ‘sense of community’ is viable in the online space. The sense of community research has unsurprisingly, since the 2000s, attempted to incorporate the virtual domain as well, in what is often called the ‘sense of virtual community’ (Blanchard & Markus, 2004; Koh & Kim, 2003). The rise of the Internet and the scholarly debate around the viability of online communities coincided with changes that researchers were noticing in how individuals approached real-life community as well. Membership in groups and communities has become, over the last several decades, much more flexible and reflexive. As Memmi (2006, p. 292) has noted, A typical modern behavior has emerged, where group membership is constantly re-evaluated and renegotiated. The modern individual belongs to several groups (professional, cultural, political . . .) at the same time but doesn’t identify too closely with any of them. He or she views the association with any given group as potentially temporary, to be discarded without trepidation when circumstances have changed. The debate about whether virtual communities and networks can provide a replacement for the decline of community in modernity is ongoing. Some argue that online communities are just as rich in human relationships as anything found in real life (Rheingold, 1993; Wellman, 1999; Amarasingam, 2015; Roberts et al., 2002). Others argue that relationships online tend to be fleeting, transactional, and temporary. Memmi (2006, p. 295) outlines some of the salient features and characteristics often found in virtual communities: • • • • • • • • • • •

Participation is often occasional, or a one-off occurrence. Participants are frequently anonymous or use pseudonyms. Groups may be quite large, with hundreds or thousands of participants. There are not only active participants but also many passive readers. Group membership is often temporary. There seems to be little group awareness. Group structure is highly flexible. Contributions to the discussion are often addressed to no one in particular. Many contributions are apparently ignored. There are few personal relationships, and they are unstable. The discussion style is usually cold and unemotional (except for some aggressiveness which serves social control purposes). • Interactions are not between persons, but revolve about a common object, goal or task. • Interactions contribute to the construction of a common workspace. • Contributions are mostly goal-oriented. None of the characteristics outlined by Memmi point to robust relationships, long-term friendships, or a sense of belonging. This is not to say that there cannot be online communities that 200

Belonging is just a click away

evince a high degree of social cohesion or personal interactions. His point is that these characteristics are more frequent in online communities, which tend to more often resemble a group assignment in a classroom as opposed to rewarding communal experiences. As he (2006, p. 295) notes, interventions are posted in a public workspace in order to further some common goals, but the individual origin of interactions is less important than their effect on the state of the common discussion or task. . . . Again, this is the picture of an ideal type. One doesn’t always observe all these characteristics at the same time or to the same degree. But they are typical of a new kind of group where a common task supersedes interpersonal relations. As Memmi goes on to note, online communities – because of the way most online communication is structured – are particularly well-suited for engaging with technical problems, solving tasks, or organising events. As he (2006, p. 296) writes, The factual character of written interactions, the time lag required to respond, the lack of affective overtones are very useful to solve technical problems without undue emotional noise. . . . The narrow bandwidth, slow rate of interaction and (mostly) written exchanges are inadequate for vague, poorly defined and open-ended problems. While much has changed in the online and social media landscape since Memmi was writing in 2006, the notion that communication in the online space is more conducive to tasks and organising is still relevant for understanding how many extremists are engaging with each other in the online world. This also relates to Ridings’ and Gefen’s work (2006, p. 5), who find that information exchange is one of the primary reasons individuals join virtual communities, with knowledge and information functioning as valuable currencies and social resources within them. As we’ll see below, even while more personal and strong bonds are being formed in online communities, most of the conversation and posts on social media platforms, from Facebook to Telegram, is still very much depersonalised and task oriented – whether it is trying to establish a white ethnostate or overturn President Biden’s election victory in 2020. However, some extremist communities also complicate this narrative. Because they operate in secrecy and under the threat of law enforcement intervention, these communities are different from other non-extremist online communities. The task or goal they are working towards requires trust, emotional bonds, and very strict boundaries around who is a member and who is not. As mentioned earlier, the research on online communities complicates much of what we understand about how extremists use the Internet, but research on extremism also complicates much of what we know about how online communities typically function. Given the changing dynamics of how communities are operating, there has also been some thought given to detecting the presence of online communities. Lorne Dawson (2004, p. 83) has proposed six elements that are useful for telling us if we are in the presence of a virtual community: interactivity, stability of membership, stability of identity, netizenship and social control, personal concern, and occurrence in a public space. Interactivity means that conversations online, to some extent, ‘relate to each other in sequence’, are related to earlier messages in the discussion, and display a ‘continuous feedback loop fostering the comprehension of a shared interpretive context, one that facilitates the emergence of new jointly produced meanings’ (Dawson, 2004, pp. 83–84). We know from some studies of online communities that the vast majority of people on these platforms are silent readers and rarely contribute to the discussion. A recent study of QAnon users on Telegram, for instance, found that 74.62% of users had never posted or posted only once (Wildon & Argentino, 2021). 201

Amarnath Amarasingam

The second element, stability of membership, means that members of these communities post fairly regularly and stay part of these groups for extended periods of time. ‘These two indicators of stability tell us we are in the presence of something more than a mere exchange of views of some passing topic of common concern’ (Dawson, 2004, p. 84). Of course, this stability, particularly when we are talking about extremist groups online, could be impacted by outside forces, such as social media platforms shutting down channels or suspending users (Berger & Perez, 2016). Third, according to Dawson (2004, p. 84) and others, because some semblance of trust must undergird community, stable identities in the online space are important. This could be achieved by individuals using consistent usernames, consistent profile pictures, and so on. Anonymity can still be maintained, but there should be some way for community members to feel like they are talking to people they have spoken to before. Fourth, netizenship and social control mean that ‘participants in the computer-mediated communication display a sense of responsibility for sustaining the communication by negotiating and enforcing norms to guide future communications and protect participants from various forms of deviance’ (Dawson, 2004, p. 84). It is unclear how much this element is often found in extremist groups online, as they usually spend more time protecting the group from outside interference, as opposed to guarding against internal disagreements (which are ignored or removed by the administrator) (McNamee et al., 2010, p. 275). Fifth, true community starts to form when at least some individuals in the group show personal concern for the lives of others in the group. As noted below, in the case of the Islamic State’s online supporters, especially in close knit groups, there was constant discussion of the struggles that individuals were undergoing at home, at work, and at school. Individuals who struggled with poverty were sent money by other members in the group and so on. This can be seen in examples of other virtual communities, such as in pro-anorexia forums, where, as Maloney (2012, p. 119) explains, members support, care for, and check on each other. It was not the case, in other words, that ISIS supporters talked about nothing other than ISIS. Personal bonds were formed, which led to them taking increased risks for each other because they found these spaces to be safe, supportive, and rewarding. Finally, and related to the first element, much of the interaction between individuals in online communities should take place ‘in a context where it is being viewed by others’ (Dawson, 2004, p. 85). This does not of course preclude more private conversation between group members, but ‘the notion of community implies a shared experience amongst a reasonably large number of people’ (Dawson, 2004, p. 85). As noted earlier, this may still mean that there are several observers on the sidelines who rarely contribute but nevertheless feel a sense of community from having access to conversations about topics that are important to them.

The online community in extremism research As discussed thus far, the definitional and conceptual debates around community, sense of community, and online community are robust in the academic literature. The cross pollination of this research into studies of extremist groups online has been limited and scattered, apart from some key studies, some of which we will examine later. For the most part, though, many publications mention online community or virtual community in passing but rarely dwell on unpacking what it means. Other times, research adopts a quantitative approach, which takes a deep dive into one or two of Dawson’s elements, but not others – largely due to methodological limitations. Zickmund (1997) and Daniels (2009) were two of the earliest researchers to look at online communities with a nexus to extremist movements and radicalisation. Since then, Maura Conway (2012, p. 1) has deployed the notion of ‘radical milieus’ to argue that the Internet ‘has been shown to facilitate the virtual establishment of strong social and personal bonds in many different contexts’ and ‘there 202

Belonging is just a click away

is no reason to believe that this should not hold true with respect to violent political extremists’ online. Conway’s article is also one of the few that examines the online community of jihadists; much of the research since has focused almost exclusively on the far right. Despite this, there are a handful of notable exceptions (Benigni et al., 2017; Awan, 2017; Piazza & Guler, 2021; Pearson, 2018; Criezis, 2021). Benigni and colleagues (2017), in particular, examined how ISIS sympathisers, propagandists, and fighters interacted with each other on Twitter to form what they call an ‘online extremist community’. Interestingly, they also point to how the Internet and social media have impacted counter-terrorism operations: law enforcement used to use individuals to gain insights about the network, but with social media, they now use networks to gain insight into individuals (Benigni et al., 2017, p. 3). Simi and Futrell (2006) were among the first to notice that a lot of the research literature on virtual communities (Rheingold, 1993; Turkle, 2005) is increasingly relevant to understanding how extremist groups also function online. They make use of Gamson’s (1996) notion of ‘free space’ to argue that the online space provides extremist movements with a safe space where they can ‘be themselves without pressure or self-consciousness about what members of the more powerful group may think or how they will react’ (Gamson quoted in Simi & Futrell, 2006, p. 118). These spaces are important for activists in general, as they give them some sense of autonomy from the mainstream ‘where they can nurture oppositional cultures and organize collective action’ (Simi & Futrell, 2006, p. 118). Simi and Futrell’s interviews with white power movement members about the significance of the online community are quite revealing. One member tells them, for instance, that the online world keeps them ‘aware’ that they are part of this movement, even as they do not have time to attend rallies and festivals: ‘I think the internet just makes it easier to be a racialist when you know what’s out there and how many other people all over the world are fighting for pretty much the same thing you are’ (interview participant quoted in Simi & Futrell, 2006, p. 128). This particular point is supported by Dentice and Bugg’s (2014, p. 120) findings concerning white racial identity, which suggest that the social construction of white racial identities on the online forum Stormfront is a result of unquestioned beliefs and perceived threats to white identity and culture. In this way, the online community also provides a corrective to the potential dilution of movement ideology when individuals are out in the real world, interacting with diverse communities and ideas. The online brotherhood and sisterhood they come home to serves to alleviate any doubts the real world may have deposited in their minds throughout the day. One of the ways this occurs, according to Simi and Futrell (2006, p. 129), is by providing movement members easy access to the white power movement’s ‘cultural items and the Aryan aesthetic enacted daily by many members’. This includes access to photos, children’s clothing, tattoos, flags, music, games, and magazines. As one member tells them: It’s really cool how you can get all this shit off the net now. Ten years ago, there really wasn’t that much stuff you could get . . . but now you’ve got all the music, the clothes. . . . I mean you can get pretty much anything you can think of I bought my daughter a toy figure of Hitler from a movement website. I mean you’re not going to find that at Wal-Mart. These kinds of cultural and aesthetic artefacts help to ‘reduce the psychological distance’ between an individual’s real-life and their involvement in the movement, which prior to the Internet may have been more sporadic (Simi & Futrell, 2006, p. 129). In the contemporary world, these trends have only been more amplified, and the hate aesthetic has come to include, among other things, memes, podcasts, and social media groups. 203

Amarnath Amarasingam

De Koster and Houtman’s 2008 study of online communities on Stormfront was also one of the earliest to point out that while researchers have long analysed far-right and other hate groups online (see Adams & Roscigno, 2005; Duffy, 2003; Schafer, 2002; Whine, 2000), there is still a dearth of attention paid to the social and psychological significance these online communities hold for the individuals involved. Their study provides further confirmation of several of Simi and Futrell’s findings and is also one of the few studies that engage with the research literature on ‘sense of community’. As De Koster and Houtman (2008, p. 1167) point out, part of the value of extremist online communities, a point I reiterate below, is that it allows individuals to feel like they are part of a broader ‘embattled’ sub-group: individuals linked transnationally undergoing the same struggle. As they (2008, p. 1167) write, It is important to stress that the members who experience stigmatization offline are strongly attached to the community function Stormfront fulfils, whether engaged in passing or not. Some of them reflect on this aspect of the forum themselves: ‘We have a cohesive factor: love for our people and fatherland, incomprehension by outsiders, and loyalty. [A community exists] because we are an oppressed species, this creates a bond’. (Herman) In a similar manner, Pearson (2018) finds that Twitter ISIS supporters frame account suspension as an act of war against their community, thus victimising and making a martyr of the sub-group. In many of these communities, there is also, as mentioned earlier, an important process of boundary maintenance and policing of conduct (Roberts, 2017). Both often go hand in hand, as leaders in these groups make clear that there is a code of conduct to be followed, a mutual respect that must be maintained, and a deviant style of life that must be shunned – all of which separate ‘us’ from ‘them’. In other words, the community cannot itself descend into chaos, deviance, and internal strife because that is what characterises the world ‘out there’ – and we are better than ‘them’. In many of these studies, there is also an interesting difference present in the way members of far-right online communities and (based on my own research) jihadist online communities discuss stigmatisation, real-life expression of views, and the value of the online space. For instance, ISIS supporters in Western countries shy away from expressing their true beliefs in real-world interactions not because of ‘stigmatisation’ but because there could also be real legal ramifications. In this sense, how individual members navigate stigma and publicly expressing their views are negotiated through whether particular groups are proscribed entities, the laws around glorifying terrorism in the countries where individuals reside, and a host of other issues. In other words, reallife geography matters for how individuals behave in online spaces. Bowman-Grieve also rightly noted back in 2009 that extremists are using the Internet not simply to consume content but to navigate ideas, form social bonds, and validate their worldviews. As she (2009, p. 990) notes of virtual communities: From a psychological perspective the discourses constructed within them have much to tell, including (but not limited to) the meaning of support, processes of identity negotiation that lead to the acceptance of ideologies that promote the use of terrorism, the specific propaganda disseminated by different movements, or factions within movements. Her study also looks at Stormfront, which, at the time of her study in 2009, had 159,000 members. Bowman-Grieve (2009, p. 997) finds many of Dawson’s elements for online community in Stormfront: community members interact quite frequently through a conversational style, individuals 204

Belonging is just a click away

recognise each other through unique usernames, and members demonstrate that they are aware of the opinions and worldviews of their specific interlocutors when posting. They also place themselves in vulnerable psychological and emotional positions as they recount how they found the far-right movement, openly discuss struggles in their own lives, and talk about how this online community has provided them with a safe space of support. Bliuc et al. (2019) examined how offline events impact collective identity on the far-right forum Stormfront. Downunder is the Australian sub-forum of Stormfront. Continuous communication in the online space allows not only for the solidification of ingroup identity but also for organic transformation in response to developments in the movement, particularly pivotal events in the real world like attacks, riots, and elections. These threshold incidents could be events the group organises themselves or could be occurrences outside the movement that nevertheless serve to crystallise identity, grievances, and objectives. The authors focus on the 2005 Cronulla riots in Sydney, Australia. The riots were preceded by a physical altercation between a group of Middle Eastern youth and three lifeguards the week prior. The incident was widely discussed in right-wing media, and protests were planned. On 11 December 2005, 5,000 people gathered to ‘reclaim the beach’ (Johns et al., 2017, p. 249). Sporadic violence continued throughout the day. What Bliuc and colleagues (2019, p. 1781) found was that not only were the Cronulla riots mentioned numerous times on Stormfront Downunder, but also that they resulted in ‘significant transformations in the collective identity of the online community’ and that there was a change in the group discussion on the sub-forum, from a focus on generic, exclusivist topics which were broadly aligned to (global) White supremacy principles (e.g. race, nationalistic values and out-groups such as Jews), to a more crystallised focus on (local) specific ethnic and religious out-groups. They found, relatedly, that the online discourse became more extreme and anti-Muslim sentiment grew stronger after the riots.

Case study: the Baqiya family What is clear from the discussion earlier is that academic research into extremist online communities has several limitations: the vast majority of the work is focused on the far-right (in particular, Stormfront and related forums); much of the work uses computational and quantitative approaches to try and understand what is at base a largely qualitative question; and there is a scarcity of work addressing newer platforms, such as the Iron March Forum, as well as newer movements and ideologies, including QAnon. In this short section, I draw from my own research and extensive interviews with Islamic State supporters to make the case that a more sustained focus on the importance of online communities is increasingly necessary in the study of extremist movements. Starting in late 2012, tens of thousands of people, young and old, began leaving their countries of origin and travelling to Syria to help a variety of militant and terrorist groups push back against the brutal regime of Bashar al-Assad. While the phenomenon of the foreign fighter was not unique – there have been similar migrations to Afghanistan, Bosnia, Chechnya, and Somalia (Hegghammer, 2020; Byman, 2019 – Syria was a fundamental game changer with respect to the number of people who travelled as well as the number of countries they came from (Barrett, 2017). What was also unique was that many of the Muslim youth who travelled to join the cause were so-called digital natives, who were tech and social media savvy. Many of them landed in Syria and maintained their accounts on Twitter, Facebook, Instagram, Snapchat, Tumblr, and a whole host 205

Amarnath Amarasingam

of other platforms. This created an interesting opportunity for researchers and journalists. We no longer had to travel to war zones to interview militants or rely strictly on their official propaganda to try to piece together an understanding of what was happening on the ground. It became possible to reach out to these individuals – ISIS supporters living in Western countries or frontline fighters in Syria – and request interviews (Dawson & Amarasingam, 2016). Starting in 2013 and continuing to the present day, I have conducted dozens of interviews with militants themselves and supporters all over the world (Dawson & Amarasingam, 2017; Amarasingam & Dawson, 2018). While my interviews with ISIS supporters began by exploring how they came to see the group as legitimate, how they feel about travelling to fight on its behalf, as well as theological and ideological matters, conversations inevitably were tinged with the importance of the online community. Very early on, the online support network of the Islamic State started to become very close with each other and started to refer to themselves as the Baqiya family – they were friends, took risks for each other, helped each other with struggles in their real lives, sent each other money and gifts, and even got married (Amarasingam, 2015; Pearson, 2015). I had been talking to one ISIS supporter from the United Kingdom, whom I will call Abu Abdullah, for several years, and he was a trusted fixture in the Baqiya family. One day he messaged me to say that law enforcement had paid him a visit and told him that it was in his best interest to stop engaging in online activity for the benefit of the Islamic State. His father was unsurprisingly furious and demanded that he shut down all of his social media accounts and focus on his education. Abu Abdullah messaged me to complain about his father. ‘I know he’s worried about me but he is literally going overboard’, he tells me, He was going on and on earlier until my mom got really upset. He wants me to literally stop my whole life and cut all relations with my Muslim friends online. Trust me, I’ve never felt like I belong anywhere until I met the brothers and sisters online. Now he wants me to cut it all off? I can’t do that. There are several interesting aspects to this text message from Abu Abdullah. First, there is the statement that his online community is equivalent to his ‘whole life’ and, second, the important observation that he never felt like he belonged anywhere except in his online community. As mentioned earlier, this complicates some of the existing research on online communities, which tends to suggest that the online space is an arena for identity experimentation to publicly express aspects of yourself behind the anonymity of the username. For members of extremist groups, the opposite is often the case. It is with their families, at their schools, and at their places of work where they are putting on an act, where they are not being themselves – sometimes for the simple reason that they do not want to be ostracised or arrested for being a ‘jihadist’ or a ‘neo-Nazi’. But online, they become part of a likeminded collective, a transnational brotherhood and sisterhood that truly understands them and also understands the ‘act’ they put on in their real lives. Abu Abdullah continues: I want to be with people who are on the same level as me. The Muslim brothers and sisters who have the same aqeeda [creed] and manhaj [methodology] as you are very beloved. Hanging around with coconuts is the worst thing ever. Baqiya Twitter is one place I felt at home. The internet keeps us connected. Keeps us as a family. Sometimes its like the person online is the real you (emphasis added). Another important element of the Baqiya family that became evident throughout my interviews was the ways in which they engaged in boundary maintenance, a process by which groups decided 206

Belonging is just a click away

who made up the ingroup and who was part of the outgroup (Roberts, 2017). Quite unexpectedly for me, one way in which boundary maintenance was practiced by members of the Baqiya family was by navigating Twitter suspensions. From 2013 to 2015, Twitter in particular became the platform of choice for the Islamic State. While Twitter was steadily suspending ISIS accounts throughout these years, the suspensions really ramped up in 2015. Members of the Baqiya family came up with several strategies for surviving Twitter’s mass suspensions, from hacking stale accounts and turning them into pro-ISIS ones to having ‘shoutout’ accounts, which had the sole purpose of tweeting to the rest of the community that someone who got suspended has returned with a new account. One individual, whom I will call Umm Hafida, created several Twitter accounts daily and simply kept them ‘at the ready’ until someone got suspended. She would then give the login details of this account to the individual so they could come back online. This process of surviving Twitter suspensions itself became important for community identity building. By getting suspended, you were communicating to your fellow Baqiya members that you had some skin in the game and that you were taking risks for the cause, as opposed to simply tweeting content that the platforms, and mainstream Western society by extension, would not find threatening. As Umm Hafida tells me, When the wave of mass Twitter suspensions started, those who spread the truth or supported IS got suspended. So basically those who had accounts suspended were considered trustful. Back in that time, everybody wanted to get suspended. Even I thought, ‘ah must be cool to get suspended and receive all that support from friends’. When I got suspended . . . lots of people were saying, ‘welcome to the Baqiya family!’ Umm Hafida continues: It is considered as a kind of shahada [martyrdom]. That’s why we say, my Twitter account just got shahada, Alhamdulillah! Suspensions show that we are in the right path, spreading the truth. Because the West doesn’t want people to know the truth. We know how trustworthy an account is by the number of times it got suspended. So, if an account got suspended ten times, you would know that the bro/sis spread beneficial tweets and trustworthy news. In other words, getting suspended was one of the ways you proved to the community that you were a part of them, that you were also taking real-life risks for the movement, and that you were not simply a free-rider. Asking ISIS supporters about what they were doing online and what it meant to them unearthed many insights about basic human needs that were being fulfilled by helping these communities online.

Conclusion The sense of community that many people feel in the online space is also true for members of extremist movements. Research in terrorism studies has at times touched on the importance of online communities in passing, but, as I argue in this chapter, they deserve more substantive treatment. There has been an overwhelming focus, largely due to the Islamic State’s unusually high propaganda output, on the content itself and its presumed radicalising impact on individuals. This has subsequently led to the development of counter-narratives and deplatforming as the main policy option for dealing with extremist content in the online space. What I have argued in this chapter, and what others (like Conway, Simi and Futrell, and Bowman-Grieve) have also noticed, 207

Amarnath Amarasingam

is that there is something else, perhaps even more important, going on that is receiving relatively less attention from extremism researchers and policy makers. As I have written elsewhere, members of these online communities were not simply consuming content; they also cared for each other, celebrated the birth of children, respected online boundaries of marriage and gender, developed relationships and got married, expressed condolences at the loss of a fighter, shared news, and served as a support group for youth who are undergoing hardship thousands of miles away. Everything we do in our community, on and offline, they did as well. (Amarasingam, 2015) Going ahead, research on online extremist communities should focus on moving the conversation forward and understanding better how these communities impact individual identities and perhaps contribute to radicalisation processes. First, more extremism research needs to incorporate the extensive research that already exists looking at sense of community and sense of online community and test some of these findings in extremist communities to see if extremist communities are somehow unique. Second, research on modern community tends to point out the increasing collapse of groups and communities today (Putnam, 2000). An important research question would be to explore whether extremist communities online are providing individuals with the muchneeded sense of belonging that is being lost in our everyday life. Third, research is clear that the vast majority of members in online communities are passive readers or lurkers. Studying, although methodologically challenging, how passive members of extremist communities online feel a sense of belonging, how this membership may be impacting their offline lives, and so on, is increasingly important. Fourth, and relatedly, it is clear that regardless of the transnational nature of online communities, geography still matters. Future research should also unpack how the urban/rural divide, the counterterrorism legal landscape, and the presence of CVE programmes in a particular locality impact individuals’ online behaviour. Finally, more significant event analysis is also needed. While some researchers have started to fill this gap in the research (Scrivens et al., 2021), obtaining a better sense of how elections, protests, riots, attacks, and other major events impact online behaviour and crystallise movement ideology will be increasingly important going forward. All of these suggestions further apply to policy makers, who are logistically best equipped to aid in the deradicalisation process and improve CVE programmes. In order to do so, they require a more nuanced understanding of the online communities that extremists are members of. This means that research that engages with online communities and extremism will better inform policy makers. Online communities are here to stay and are only proving to be more significant and impactful as the extremism space grows more eclectic, as trust in mainstream news sources and expert systems declines, and political discourse becomes increasingly polarised (Argentino et al., 2022). As one young person interviewed by CBS News’s Adam Yamaguchi stated: ‘My moral compass, and my reference point for what’s right and wrong, my politics, my worldview, my philosophy all comes from the internet’ (CBS News, 2020). When Yamaguchi wondered how this young man might feel knowing that the vast majority of society views what he is saying and the community that he is part of online as racist, he responded: ‘Well, ninety-nine percent of them can go fuck themselves. They’re from another planet. They’re not from our world’. As individuals spend more and more time on this ‘other planet’, it is imperative that researchers better understand the impact that it is having on their lives and on the lives of those around them. 208

Belonging is just a click away

References Abdelrahim, Y. A. A. (2019). Visual analysis of ISIS discourse strategies and types in Dabiq and Rumiyah online magazines. Visual Communication Quarterly, 26(2), 63–78. Adams, J., & Roscigno, V. J. (2005). White supremacists, oppositional culture and the world wide web. Social Forces, 84(2), 759–778. Alexander, A. (2017). Digital decay: Tracing change over time among English-language Islamic state sympathizers on Twitter. George Washington University Program on Extremism. https://extremism.gwu.edu/ sites/g/files/zaxdzs2191/f/DigitalDecayFinal_0.pdf Amarasingam, A. (2015, December 30). What Twitter really means for Islamic state supporters. War on the Rocks. https://warontherocks.com/2015/12/what-twitter-really-means-for-islamic-state-supporters/ Amarasingam, A. (2021, January 12). Does deplatforming work? A quick survey of literature in the wake of the capitol hill attack. Intrepid Blog. www.intrepidpodcast.com/blog/2021/1/12/does-deplatformingwork-a-quick-survey-of-literature-in-the-wake-of-the-capitol-hill-attack Amarasingam, A., & Dawson, L. L. (2018). ‘I left to be closer to Allah’: Learning about foreign fighters from family and friends. ISD Report. www.isdglobal.org/wp-content/uploads/2018/05/Families_Report.pdf Amarasingam, A., Maher, S., & Winter, C. (2021, January 8). How telegram disruption impacts Jihadist platform migration. CREST Report. https://crestresearch.ac.uk/resources/how-telegramdisruption-impacts-jihadist-platform-migration/ Anderson, B. (1983). Imagined communities. Verso. Argentino, M-A., Amarasingam, A., & Conley, E. (2022). ‘One struggle’: Examining narrative syncretism between accelerationists and Salafi-Jihadists. International Centre for the Study of Radicalization (ICSR) Report. https://icsr.info/wp-content/uploads/2022/01/ICSR-Report-One-Struggle-Examining-NarrativeSyncretism-between-Accelerationists-and-Salafi%E2%80%91Jihadists.pdf Awan, I. (2017). Cyber-extremism: Isis and the power of social media. Society, 54(2), 138–149. https://doi. org/10.1007/s12115-017-0114-0 Bakardjieva, M., & Feenberg, A. (2002). Community technology and democratic rationalization. The Information Society, 18(3), 181–192. https://doi.org/10.1080/01972240290074940 Barrett, R. (2017, October 31). Beyond the caliphate: Foreign fighters and the threat of returnees. The Soufan Center. https://thesoufancenter.org/wp-content/uploads/2017/11/Beyond-the-Caliphate-Foreign-Fighters-andthe-Threat-of-Returnees-TSC-Report-October-2017-v3.pdf Benigni, M. C., Joseph, K., & Carley, K. M. (2017). Online extremism and the communities that sustain it: Detecting the ISIS supporting community on Twitter. PLoS One, 12(12), e0181405. https://doi. org/10.1371/journal.pone.0181405 Berger, J. M., & Perez, H. (2016). The Islamic state’s diminishing returns on Twitter: How suspensions are limiting the social networks of English-speaking ISIS supporters. George Washington University’s Program on Extremism, Occasional Paper. https://extremism.gwu.edu/sites/g/files/zaxdzs2191/f/downloads/ JMB%20Diminishing%20Returns.pdf Blanchard, A. L., & Markus, M. L. (2004). The experienced ‘sense’ of a virtual community: Characteristics and processes. Database for Advances in Information Systems, 35(1), 65–79. Bliuc, A-M., Betts, J., Vergani, M., Iqbal, M., & Dunn, K. (2019). Collective identity changes in far-right online communities: The role of offline intergroup conflict. New Media & Society, 21(8), 1770–1786. Bowman-Grieve, L. (2009). Exploring ‘stormfront’: A virtual community of the radical right. Studies in Conflict & Terrorism, 32, 989–1007. boyd, d. (2011). Social network sites as networked publics affordances, dynamics, and implications. In Z. Papacharissi (Ed.), A networked self: Identity, community and culture on social network sites. Routledge. Bruhn, J. G. (2011). Conceptions of community: Past and present. In The sociology of community connections. Springer Netherlands. https://doi.org/10.1007/978-94-007-1633-9 Byman, D. (2019). Road warriors: Foreign fighters in the Armies of Jihad. Oxford University Press. Caiani, M., & Wagemann, C. (2009). Online networks of the Italian and German extreme right. Information, Communication & Society, 12(1), 66–109. https://doi.org/10.1080/13691180802158482 CBS News. (2020, March 12). Extremists next door. CBS News Originals. www.youtube.com/ watch?v=x7xVcsmO8rg Chandrasekharan, E., Pavalanathan, U., Srinivasan, A., Glynn, A., Eisenstein, J., & Gilbert, E. (2017). You can’t stay here: The efficacy of reddit’s 2015 ban examined through hate speech. Proceedings of the ACM on Human-Computer Interaction, 1, No. CSCW, Article 31), 1–22.

209

Amarnath Amarasingam Clifford, B., Hughes, S.,  & Meleagrou-Hitchens, A. (2020, December  29). An abridged history of America’s terrorism prevention programs: Fits and starts. Lawfare. www.lawfareblog.com/ abridged-history-americas-terrorism-prevention-programs-fits-and-starts Conway, M. (2012). From Al-Zarqawi to Al-Awlaki: The emergence and development of an online radical milieu. CTX: Combating Terrorism Exchange, 2(4), 1–10. https://core.ac.uk/download/pdf/14760 3165.pdf Conway, M., Khawaja, M., Lakhani, S., & Reffin, J. (2021). A snapshot of the Syrian Jihadi online ecology: Differential disruption, community strength, and preferred other platforms. Studies in Conflict & Terrorism, 1–19. https://doi.org/10.1080/1057610X.2020.1866736 Cowan, D. E. (2005). Cyberhenge: Modern pagans on the internet. Routledge. Criezis, M. (2021). ‘Many sisters wish they were men’: Gendered discourse and themes in pro-ISIS online communities. Journal for Deradicalization, 25, 35–73. Daniels, J. (2009). Cyber racism: White supremacy online and the new attack on civil rights. Rowman & Littlefield. Dawson, L. L. (2004). Religion and the quest for virtual community. In L. L. Dawson & D. E. Cowan (Eds.), Religion online: Finding faith on the internet. Routledge. Dawson, L. L., & Amarasingam, A. (2016). Trying to talk to terrorists: Ethical and methodological challenges in Canada. TSAS Working Paper Series. No. 16–13. www.tsas.ca/wp-content/uploads/2018/02/ TSASWP16-13_Dawson-Amarasingam.pdf Dawson, L. L., & Amarasingam, A. (2017). Talking to foreign fighters: Insights into the motivations for Hijrah to Syria and Iraq. Studies in Conflict and Terrorism, 40(3), 191–210. De Koster, W., & Houtman, D. (2008). ‘Stormfront is like a second home to me’: On virtual community formation by right-wing extremists. Information, Communication & Society, 11(8), 1155–1176. Delanty, G. (2010). Community (2nd ed.). Routledge. Dentice, D., & Bugg, D. (2014). Fighting for the right to be white: A case study in white racial identity. Journal of Hate Studies, 12, 101–128. Duffy, M. E. (2003). Web of hate: A fantasy theme analysis of the rhetorical vision of hate groups online. Journal of Communication Inquiry, 27(3), 291–312. Feenberg, A., & Bakardjieva, M. (2004). Virtual community: No ‘killer implication.’ New Media & Society, 6(1), 37–43. https://doi.org/10.1177/1461444804039904 Ferrara, E. (2017). Contagion dynamics of extremist propaganda in social networks. Information Sciences, 418–419, 1–12. https://doi.org/10.1016/j.ins.2017.07.030 Fisher, A. T., & Sonn, C. C. (2002). Psychological sense of community in Australia and the challenges of change: One community or a community of communities. Journal of Community Psychology, 30, 579–609. Fisher, A. T., & Sonn, C. C. (2007). Sense of community and dynamics of inclusion-exclusion by receiving communities. The Australian Community Psychologist, 19(2), 26–34. Futrell, R., & Simi, P. (2004). Free spaces, collective identity, and the persistence of U.S. white power activism. Social Problems, 51(1), 16–42. Gamson, W. (1996). Safe spaces and social movements. Perspectives on Social Problems, 8, 27–38. GIFCT. (2017). Story. Global Internet Forum to Counter Terrorism. Formation of GIFCT. https://gifct.org/ about/story/ Hassan, G., Brouillette-Alarie, S., Alava, S., Frau-Meigs, D., Lavoie, L., Fetiu, A., Varela, W., Borokhovski, E., Venkatesh, V., Rousseau, C., & Sieckelinck, S. (2018). Exposure to extremist online content could lead to violent radicalization: A systematic review of empirical evidence. International Journal of Developmental Science, 12(1–2), 71–88. https://doi.org/10.3233/DEV-170233 Hegghammer, T. (2020). The Caravan: Abdallah Azzam and the rise of global Jihad. Cambridge University Press. Hine, C. (2000). Virtual ethnography. SAGE. Hunter, A. (2007). Contemporary conceptions of community. In R. A. Cnaan & C. Milofsky (Eds.), Handbook of community movements and local organizations (pp.  20–33). Springer US. https://doi. org/10.1007/978-0-387-32933-8_2 Johns, A., Noble, G., & Harris, A. (2017). After cronulla: ‘Where the bloody hell are we now?’ Journal of Intercultural Studies, 38(3), 249–254. Kaczkowski, W. (2019). Qualitative content analysis of images of children in Islamic state’s Dabiq and Rumiyah magazines. Contemporary Voices, 1(2), 26–38.

210

Belonging is just a click away Kelsey-Sugg, A. (2019, August 19). Why Christian Picciolini joined a violent Neo-Nazi group as a teen – and what got him out. ABC Radio National. www.abc.net.au/news/2019-08-16/neo-nazi-extremisminside-the-mind-of-christian-picciolini/11405856 Koehler, D. (2014). The radical online: Individual radicalization processes and the role of the internet. Journal for Deradicalization, 1, 116–134. Koh, J., & Kim, Y-G. (2003/2004). Sense of virtual community: A conceptual framework and empirical validation. International Journal of Electronic Commerce, 8(2), 75–93. Lakomy, M. (2019). Recruitment and incitement to violence in the Islamic state’s online propaganda: Comparative analysis of Dabiq and Rumiyah. Studies in Conflict & Terrorism, 44(7), 565–580. Maloney, P. (2012). Online networks and emotional energy: How pro-anorexic websites use interaction ritual chains to (re)form identity. Information, Communication & Society, 16(1), 105–124. https://doi.org/10.1 080/1369118X.2012.659197 McMillan, D., & Chavis, D. (1986). Sense of community: A definition and theory. Journal of Community Psychology, 14, 6–23. McNamee, L. G., Peterson, B. L., & Peña, J. (2010). A call to educate, participate, invoke and indict: Understanding the communication of online hate groups. Communication Monographs, 77(2), 257–280. https:// doi.org/10.1080/03637751003758227 Meleagrou-Hitchens, A., & Kaderbhai, N. (2017). Research perspectives on online radicalisation: A literature review 2006–2016. Vox-Pol. https://icsr.info/wp-content/uploads/2017/05/ICSR-Paper_ResearchPerspectives-on-Online-Radicalisation-A-Literature-Review-2006–2016.pdf Memmi, D. (2006). The nature of virtual communities. AI and Society, 20, 288–300. Odag, Ö., Leiser, A., & Boehnke, K. (2019). Reviewing the role of the internet in radicalization processes. Journal for Deradicalization, 21, 261–300. Papacharissi, Z. (2015). Affective publics: Sentiment, technology, and politics. Oxford University Press. Pearson, E. (2015, November 11). Wilayat Twitter and the battle against Islamic state’s Twitter Jihad. Vox Pol. www.voxpol.eu/wilayat-twitter-and-the-battle-against-islamic-states-twitter-jihad/ Pearson, E. (2018). Online as the new frontline: Affect, gender, and ISIS-take-down on social media. Studies in Conflict & Terrorism, 41(11), 850–874. https://doi.org/10.1080/1057610X.2017.1352280 Perkins, D. D., &Long, D. A. (2002). Neighborhood sense of community and social capital: A multi-level analysis. In A. T. Fisher, C. C. Sonn, & B. J. Bishop (Eds.), Psychological sense of community: Research, applications, and implications (pp. 291–318). Plenum. Piazza, J. A., & Guler, A. (2021). The online caliphate: Internet usage and ISIS support in the Arab world. Terrorism and Political Violence, 33(6), 1256–1275. https://doi.org/10.1080/09546553.2019.1606801 Postill, J., & Pink, S. (2012). Social media ethnography: The digital researcher in a messy web. Media International Australia, 145(1), 123–134. https://doi.org/10.1177/1329878X1214500114 Putnam, R. D. (2000). Bowling alone: The collapse and revival of American community. Simon & Schuster. Rheingold, H. (1993). Virtual communities. MIT Press. Ridings, C. M., & Gefen, D. (2006). Virtual community attraction: Why people hang out online. Journal of Computer-Mediated Communication, 10(1). https://doi.org/10.1111/j.1083-6101.2004.tb00229.x Roberts, D. (2017). Subcultural boundary maintenance in a virtual community for body modification enthusiasts. International Journal of Cultural Studies, 20(4), 361–376. Roberts, L. D., Smith, L. M., & Pollock, C. (2002). Mooing till the cows come home: The search for sense of community in virtual environments. In A. T. Fisher, C. C. Sonn & B. J. Bishop (Eds.), Psychological sense of community: Research, applications, and implications (pp. 223–245). Kluwer Academic/Plenum Publishers. Robinson, L., & Schulz, J. (2009). New avenues for sociological inquiry: Evolving forms of ethnographic practice. Sociology, 43(4), 685–698. https://doi.org/10.1177/0038038509105415 Rudner, M. (2017). “Electronic Jihad”: The internet as Al Qaeda’s catalyst for global terror. Studies in Conflict & Terrorism, 40(1), 10–23. https://doi.org/10.1080/1057610X.2016.1157403 Sarason, S. B. (1974). The psychological sense of community: Prospects for a community psychology. Jossey-Bass. Schafer, J. A. (2002). Spinning the web of hate. Web-based hate propagation by extremist organizations. Journal of Criminal Justice and Popular Culture, 9(2), 69–88. Schroeder, R. (2011). Being there together: Social interaction in shared virtual environments. Oxford University Press.

211

Amarnath Amarasingam Scrivens, R., Burruss, G. W., Holt, T. J., Chermack, S. M., Freilich, J. D., & Frank, R. (2021). Triggered by defeat or victory? Assessing the impact of presidential election results on extreme right-wing mobilization online. Deviant Behavior, 42(5), 630–645. Simi, P., & Futrell, R. (2006). Cyberculture and the endurance of white power activism. Journal of Political and Military Sociology, 34(1), 115–142. Stempien, M. S. (2019). Russia in the eyes of Islamic state: An analysis of the content of Dabiq and Rumiyah magazines and Russia’s involvement in the fight against the Islamic state. Terrorism and Political Violence, 1–23. https://doi.org/10.1080/09546553.2019.1657097 Turkle, S. (1999). Cyberspace and identity. Contemporary Sociology, 28(6), 643–648. Turkle, S. (2005). The second self: Computers and the human spirit (20th Anniversary ed.). MIT Press. Veilleux-Lepage, Y., & Archambault, E. (2019). Mapping transnational extremist networks: An exploratory study of the soldiers of Odin’s Facebook network, using integrated social network analysis. Perspectives on Terrorism, 13(2), 21–38. Vidino, L., & Hughes, S. (2015). Countering violent extremism in America. The George Washington University Program on Extremism. www.lawfareblog.com/abridged-history-americas-terrorism-preventionprograms-fits-and-starts Wellman, B. (Ed.). (1999). Networks in the global village. Westview Press. Whine, M. (2000). Far right extremists on the internet. In D. Thomas & B. D. Loader (Eds.), Cybercrime. Law enforcement, security and surveillance in the information age (pp. 234–250). Routledge. Wildon, J., & Argentino, M-A. (2021, July 28). QAnon is not dead: New research into telegram shows the movement is alive and well. GNET Insights. https://gnet-research.org/2021/07/28/qanon-is-not-deadnew-research-into-telegram-shows-the-movement-is-alive-and-well/ Winter, C., Neumann, P., Meleagrou-Hitchens, A., Ranstorp, M., Vidino, L., & Fürst, J. (2020). Online extremism: Research trends in internet activism, radicalization, and counter-strategies. International Journal of Conflict and Violence (IJCV), 14, 1–20. https://doi.org/10.4119/ijcv-3809 Zickmund, S. (1997). Approaching the radical other. In S. Jones (Ed.), Virtual culture: Identity and communication in cybersociety. SAGE.

212

14 RADICALISATION OF ‘LONE ACTORS’ Francis O’Connor, Lasse Lindekilde, and Stefan Malthaner

Introduction1 In February 2020, a far-right extremist, Tobias R., went on a terrorist shooting spree in the German city of Hanau. He targeted two shisha bars, which were popular with many of the city’s residents with immigrant backgrounds, killing nine and injuring five. The perpetrator had publicly expressed racist views and his violent opposition to immigration but was not known to have had any links to the local far-right extremist scene (see Crawford & Keen, 2020). The attack came in the wake of a series of far-right terror attacks in Germany, including the National Socialist Underground’s decade-long terror campaign, a number of killings and attacks on politicians, and the 2019 attack on a synagogue in Halle. Additionally, potential far-right infiltration of the German police and army had come to public attention, and in 2018, the former head of the Federal Office for the Protection of the Constitution, Hans-Georg Maaßen, was ushered out of his position due to his perceived political sympathies with the arguments of the German far-right. In the aftermath of the attack, migrant associations, left-of-centre parties, and movements feared that the political aspects of the Hanau attack would be downplayed and the attack would be simply categorised as the actions of a ‘lone wolf’ or decontextualised as a ‘lone attack’. This led to a widespread nationwide campaign rejecting the idea that the attack was an Einzelfall (single case) or carried out by an Einzeltäter (single attacker) (Lüdecke, 2020; Sturm, 2020), arguing instead that it was the latest instance of a broader pattern of far-right terrorist attacks. A pattern that has been fuelled by the political stigmatisation of migrants, their legitimisation as targets of violence, state institutions’ complacency regarding the threat of far-right violence as well as police incompetence and, at times, complicity with the far-right milieu. Yet, Tobias R. fulfilled all the academic criteria that distinguish an act of lone-actor terrorism from other forms of collective terrorism. Lone-actor terrorists are commonly differentiated from their group-based counterparts according to three dimensions. First, they must prepare and carry out the attack as a single perpetrator. Second, they must not be formally affiliated with a terrorist organisation/group, and finally, a lone actor does not act on the command of other individuals or groups (Lindekilde et al., 2018, p. 24). Tobias R. met all of these criteria: he conducted the attack on his own; there was no evidence of any substantial involvement in local far-right movements or even in online contexts; and he did not act on the commands of others.

213

DOI: 10.4324/9781003035848-16

Francis O’Connor, Lasse Lindekilde, and Stefan Malthaner

Many people adopt extremist or radical views that endorse the use of violence, but only an infinitesimal minority ever conduct individual (or indeed collective) acts of violence or terrorism. Some form of incremental socialisation is usually necessary for somebody to cross the threshold to carry out acts of terrorism. This radicalisation process evolves when select individuals or groups are vilified and dehumanised, leading violence against them to be not only normalised but also exhorted as a virtue necessary to protect one’s community or people. And if the perpetrator of the Hanau massacre had little contact with far-right extremists, what drove him, in contrast to all of the others with similar political views, to conduct his attack? This is where the ‘kein Einzelfall’ campaign and the academic consensus on lone-actor radicalisation actually coincide, albeit the use of differing terminology has resulted in confusion surrounding the academic concept of lone-actor terrorism and radicalisation. This confusion is perhaps understandable given the popular focus on the ‘loneness’ of lone actors rather than their immediate social environment, their social and political ties, and prevailing public discourse. Research has confirmed that lone actors are in fact not as ‘lone’ as the description suggests (Gill, 2015). Decades of empirical and theoretical work have demonstrated that collective radicalisation is a relational process driven by interactions with other individuals, movements, and institutions and consolidated by emotional bonds (Bosi et al., 2014). Our findings have shown that lone-actor radicalisation is a similarly relational process: the ‘loneness’ of lone actors is always relative and never absolute. Being alone is itself a relational status, albeit one characterised by few rather than multiple social relations (Malthaner & Lindekilde, 2017). ‘Loneness is derived from interactive patterns of relational embedding and disembedding in various social settings. . . . not an inherent quality but a result of social processes shaped by individual lone actors’ personalities and capacities for social interaction’. (Lindekilde et al., 2018, p. 31). Indeed, the more ‘lone’ cases like the ‘Unabomber’ Ted Kaczynski and Anders Breivik are ‘exceptional rather than archetypical’ (Schuurman et al., 2019, p.  772). Most lone actors benefit from the moral support and encouragement of their online and offline ideological milieus, and some even benefit from partial material support in attack planning. Lone-actor radicalisation is driven by differing interactive patterns of social ties in online and offline environments. This leads back to the position that what much of the radicalisation and terrorism literature classifies as lone-actor terrorism actually validates the ‘Kein Einzelfall’ focus on societal, institutional, and broader political responsibility for the context that engenders violence rather than simply individualising the blame and pathologising ‘mad’ individuals. Accordingly, lone-actor terrorism is a reflection of macro-political tendencies and norms, bearing much similarity and continuity with parallel forms of movement-based terrorism while remaining a distinct operational tactic within wider ideological milieus, with a higher importance afforded to individual triggers and personality issues than in movements. This chapter will outline how the field has developed in recent years as the study of lone-actor radicalisation has carved a distinct niche within broader terrorism and radicalisation studies. It will then introduce the Lone Actor Radicalisation and Terrorism (LART) dataset and explain its origins, data sources, and coding protocols. This dataset is used to provide an empirical overview of an empirical overview of recent patterns of lone actor violence. It also outlines some of the challenges inherent in coherently categorising lone-actor terrorism and the justification for the inclusion of interrupted attacks as well as completed attacks. It continues by listing four issues that the authors argue are of specific relevance to better understanding contemporary LART. We conclude the article by pointing to avenues for future research that would further enhance the field.

214

Radicalisation of ‘lone actors’

Lone-actor radicalisation and terrorism – a brief literature review Until 2012, lone-actor radicalisation scarcely existed as a topic of research in the discipline (Schuurman, 2019, p. 7); it is, therefore, a relatively recent sub-field in the area of terrorism studies and radicalisation. The concept of lone-actor terrorism itself remains contested, and several overlapping and diverging definitions are in widespread use (Kenyon et al., 2021, p. 5). Disputes centre on the clarity of the political motive necessary to distinguish it from other forms of spree killings such as school shootings or even domestic violence. As it is a low-frequency phenomenon, data scarcity is a problem, and due to the security risks associated with the phenomenon and the high fatality rate of the attackers, much research has been reliant on secondary rather than primary sources. Pioneering attempts to understand lone-actor radicalisation were based on limited data (Simon, 2016; Spaaij, 2011) and rooted in theoretical frameworks developed for collective forms of violence (McCauley & Moskalenko, 2017). This research tended, with some notable exceptions (Gill, 2015; Gill et al., 2014), to focus on single case studies (inter alia Böckler et al., 2015; GartensteinRoss, 2014) with little comparative analysis (see Lindekilde et al., 2018, p. 24). Much of this early work emphasised lone actors’ isolation and was oriented towards empirically mapping the phenomenon and discovering the ‘profile’ of the types of individuals likely to carry out acts of individual terrorism (Danzell & Montañez, 2016; McCauley & Moskalenko, 2014). However, similarly to group-based terrorism (Horgan, 2008), aside from a clear preponderance of male over female lone actors, no robust profile has been identified. As research continued to develop in the field, it became clear that lone actors tended to report higher rates of mental illness than both the general public and members of violent movements or groups (Corner & Gill, 2015). Yet, arguments proposing mental illness as a causal explanation for lone-actor radicalisation failed to account for the fact that the overwhelming majority of people suffering from mental illness never commit any acts of violence. Accordingly, the explanation for lone-actor terrorism cannot be intrinsic to the actors themselves but rather a combination of how certain external factors interact with forms of mental illness to enhance individual pathways towards violence. More recent research has disaggregated the overarching term of mental illness to analyse the differing roles specific psychiatric conditions can potentially play in inhibiting or accelerating radicalisation (Gill et al., 2020); Gøtzsche‐Astrup & Lindekilde, 2019). As mentioned earlier, most lone actors tend to be much less isolated than anticipated. Consequently, the field started to investigate lone actors’ social and political environment and radical milieu in an effort to disentangle the mechanisms and triggers that lead to acts of violence (Berntzen & Sandberg, 2014; Bright et al., 2020; Malthaner & Lindekilde, 2017; Nesser, 2012). This broader focus also includes the non-ideological or political dimensions of the attackers’ lives. This includes the focus on the so-called crime-terror nexus (Neumann & Basra, 2018) and work drawing links between previous patterns of non-political violent behaviour such as domestic violence and eventual lone-actor terrorism (McCulloch et al., 2019). This shift in emphasis from the individual to the social and environmental level of analysis has resulted in a less pronounced focus on lone-actor terrorism and radicalisation as a universal phenomenon with more importance afforded to local and national contexts (Gattinara et al., 2018). Work on the promotion of leaderless resistance as a strategy of the North American far-right (Joosse, 2017; Kaplan et al., 2014) and a similar approach by international jihadist movements (Ramsay & Marsden, 2015) has highlighted the incorporation of lone-actor terrorism into movement repertoires and strategy. Parallel to this focus on existing national sub-cultures of political violence, research has also highlighted a conspicuously internationally oriented trend in lone-actor

215

Francis O’Connor, Lasse Lindekilde, and Stefan Malthaner

violence, comprising reciprocal symbolic referencing, imitative attacks, and explicit acknowledgement in attackers’ written and oral declarations. Much of this interaction occurs in online forums before producing deadly real-world consequences such as the Christchurch attack and the cluster of attacks it partially inspired (Koehler, 2019; Macklin, 2019a, 2019b).

Empirical overview: introducing the lone-actor radicalisation and terrorism dataset Scope of dataset In the following, we provide an empirical overview of lone-actor terrorism based on the LART. The dataset contains 306 instances of planned and conducted lone-actor terrorism between the years 1990 and 2021. The threshold of violence is set relatively high, focusing on attacks aimed to inflict physical damage (serious injury or death) on persons using weapons. It places terrorism on a continuum of coercion ranging from hate crime (e.g. threats and vandalism), political violence (beatings, property damage), and ultimately terrorism, which has the objective of killing people (see Bjørgo & Ravndal, 2019). The case inclusion is comprehensive for several western European countries (the United Kingdom, Germany, Italy, France, Denmark, Sweden, Belgium, the Netherlands, Spain, Luxembourg, Sweden, Norway, Finland, and Ireland), Canada, and the United States (see Figures 14.1 and 14.4). Due to language issues and the regional expertise of the authors, it does not make any claims of generalisability for other regions in the world. The dataset comprises two distinct phases of data collection. It is based on the initial dataset (N = 119) created by Paul Gill and his colleagues (Gill et al., 2014), with whom the authors collaborated within the PRIME consortium on lone-actor extremism. This dataset covers the years 1990 until 2014. It also includes certain lone-actor terrorists who committed attacks before 1990 but were only captured after that date, most notoriously the so-called Unabomber Ted Kaczynski. In the second phase, the authors expanded the dataset’s geographical scope to include countries like Italy and Germany, which had not been systematically included in Gill et al.’s original dataset. It also included all cases in the regions outlined above that occurred between 2014 and 2020 (N  =  306).2 Cases were identified through regular reviews of online broadsheets and by crosschecking the dataset with other publicly available datasets on terrorism, in particular those focused

Figure 14.1  Geographic distribution * Note: Total number of attacks = 299. Australia and New Zealand are excluded from the plot. Source: Created by the authors

216

Radicalisation of ‘lone actors’

Figure 14.2  Chronological distribution * Note: Total number of attacks = 306. Source: Created by the authors

Figure 14.3  Ideological distribution * Note: N = 306. Source: Created by the authors

217

Francis O’Connor, Lasse Lindekilde, and Stefan Malthaner

on lone-actor terrorism. They included the dataset on North American lone actors gathered by Hamm and Spaaij (2017), right-wing terrorism in Europe (Ravndal, 2016), and annual EUROPOL reports on terrorism in Europe.3

Ideological distribution The dataset is dominated by attacks from two large ideological tendencies: far-right extremism and Islamist extremism. The former includes ‘attacks whose target selection is based on extremeright beliefs and corresponding enemy categories – immigrants, minorities, political opponents, or governments’ (Bjørgo & Ravndal, 2019, p. 5). These patterns of far-right extremism vary according to their socio-political contexts. Attacks in countries like Italy are shaped by the country’s neo-fascist milieu, whereas in the United States, attacks are often affiliated with the so-called patriot militia or white supremacist scenes. The category of Islamist Extremism includes attacks that followed the targeting logic of groups like ISIS or Al-Qaeda, focusing on targets associated with Western governments, as well as following calls to target ‘unbelievers’ in a broad sense. It also includes other attacks focusing on Islamic minorities, such as the killing of Ahmadiyya cleric Asad Shah in Glasgow by a Sunni extremist from Bradford in 2016 (Parveen, 2016). Many of these attackers are beholden to some form of Salafi Jihadism, others to violent interpretations of Barelvi or Deobandi doctrine. The next most prominent group of attackers is motivated by single issues such as anti-abortion or Black Nationalism. The anti-abortion movement in the United States has a strong overlap with the American far-right scene, while Black Nationalists were historically associated with the far left. Nevertheless, they have been both coded as a separate category because their attacks were almost exclusively focused on their specific grievances, for example, directly attacking doctors who perform abortions or shooting policemen seen as the embodiment of the racist state. The dataset includes five cases of terrorism inspired by left-wing extremism. Left-wing extremism serves as an umbrella term, including various strands of left-wing politics, rather than distinguishing between specific tendencies such as anti-fascism or anarchism. In addition to these larger groupings, the dataset also contains a range of attacks or plots ideologically categorised as ‘other’. These attacks contain a myriad of ideological motivations, from non-specific opposition to the state (Henley, 2002), attacks conducted to draw attention to the situation in North Korea (Madhani, 2003), motivated by Irish republicanism (Kearney, 2017), to opposition to the 1990s peace process between Israel and Palestine (Sack, 1997). Nevertheless, these attacks all contained a specific political, albeit often incoherent, motivation. There have been multiple other attacks, notably school shootings, that are similar in modus operandi but lack the sufficient political dimension necessary for them to be considered terrorism and are therefore excluded from the dataset (Malkki, 2014).

Attack type The dataset includes instances of shootings, bombings, stabbings (which include all attacks with bladed weapons), run-over attacks, and arson. Arson attacks on property or buildings that could have been reasonably presumed to be empty and therefore unlikely to injure people are excluded. The attack by an anarchist, Eric King, on Missouri US Democratic Representative Emanuel Cleaver’s office is not included (Reuters Staff, 2016). On the contrary, Roland James Smith, a Black Nationalist activist who seized hostages in a store before setting it alight, killing seven people, was included in the dataset as the attack either deliberately sought to cause human victims harm or was 218

Radicalisation of ‘lone actors’

Figure 14.4  Attack type * Note: N = 306 Source: Created by the authors

recklessly indifferent to the likelihood of their harm (Goldman, 1995). Some attacks incorporated multiple means of violence. In order to facilitate coding, the attacks were coded according to the means that resulted in the most injuries and death. Accordingly, Breivik’s 2011 terrorist attack was coded as a shooting, as many more victims were shot on Utøya Island than injured in the bomb he planted earlier in the day in Oslo (Hemmingby & Bjørgo, 2016).

Interrupted plots Importantly, interrupted plots are also included in the dataset. Forty-nine cases were stopped by the authorities before attacks could actually take place. As difficult as it is to obtain data on attacks, obtaining data on prevented attacks is even more difficult as they attract less media coverage. Plus, it is known that many lone actors back out at the last moment, so it is possible that planned attacks might never have actually taken place. To minimise the risk of including interrupted cases, which amount to no more than online braggadocio, we have included only cases with evidence confirming active steps were taken towards the attack, such as weapons’ procurement or target scoping. The 2019 cases inspired by the Christchurch attack, of James Dylan Grimes, who threatened to blow up a school, and of Thomas Alonzo Bolin, who discussed an unspecified shooting attack with his cousin, were considered for inclusion but later excluded as no concrete steps appear to have been taken (‘Louisville Man Arrested after Threatening “9,000 Kiddies” ’, 2019; Stack, 2019). 219

Francis O’Connor, Lasse Lindekilde, and Stefan Malthaner

However, the 2014 case of Brusthom Ziamani is included, as he was on his way to behead a British soldier in London and had a knife and hammer in his rucksack when he was stopped by the police (‘Brusthom Ziamani “Arrested on Way to Behead Soldier” ’, 2015). Although hoaxes can, on occasion, be prosecuted under terrorism legislation (see Bell, 2020), we have excluded them from the dataset because, although they seek to generate fear, they do not usually lead to physical harm. Accordingly, cases like that of white supremacist David Parnham, who sent fake anthrax to the Queen of England, are not included (Murphy, 2019). LART also does not include cases of people charged with supporting terrorist groups or violence by lone actors affiliated with the state, such as police officers engaging in vigilante attacks. Another challenging factor in the coding of lone-actor plots (as opposed to actual attacks) is assigning a date to them. For reasons of consistency in coding, it was decided to code the date of arrest (when available) in the absence of an attack date. The type of planned attack in interrupted plots was also included in the attack type section.

Limits of the dataset The dataset is limited by its focus on western countries, thus limiting its generalisability to the countries included in the study rather than the broader phenomenon per se. It is reliant largely on secondary sources (Schuurman, 2018), save for a number of cases, which have been further analysed in small and medium-N studies where restricted source material was obtained. Although since 2016, new cases have been contemporaneously added to the dataset, there is likely an underrepresentation of historic cases in the countries not included in Gill and colleagues’ first dataset (Gill et al., 2014). The challenges in the re-construction of a historic dataset in contrast to the contemporaneous development of a current one are self-evident.4 In the early 1990s in particular, when Internet sources were less comprehensive, it is likely that cases that resulted in a limited number of deaths or interrupted plots were less covered in the media and therefore simply missed.

Recent trends in lone-actor terrorism Building on the extensive empirical evidence in the LART dataset, the authors have identified four recent trends influencing and likely shaping the phenomena of LART in the near future. First, politics in recent years has seen an incorporation of extremist discourse of the political margins by institutional political actors across the Western world, leading to a normalisation of the stigmatisation of minorities and/or political adversaries. Second, there has been a simplification of attacks away from complex plots involving explosives to ones using everyday devices and tools such as run-over attacks or mass stabbings, making them to harder to interdict. Third, the tendency to analyse radicalisation as either online or offline remains prevalent, notwithstanding academic research that has demonstrated the blended nature of radicalisation, including online and offline elements. Finally, the changing nature of online platforms has also changed how lone actors hint at their plans, conduct their attacks, and, in many cases, try to broadcast them as they unfold.

Mainstreaming of radical discourse Party politics literature has long debated the ‘pathological normalcy’ or otherwise of far-right politics in western societies (Mudde, 2010). The dataset presented above dates to 1990, and in the last 30 years, there have been several far-right parties in western parliaments or, indeed, on some occasions, in government. Yet, there has been a notable swing in the last decade towards a 220

Radicalisation of ‘lone actors’

greater and ever more vocal presence of the far right in power. This is clear in countries like Italy (Lega) and Austria (Freiheitliche Partei Österreichs, FPÖ), notably during Donald Trump’s reign in the United States, and by a marked increase in nativist hostility towards migrants in parties of the centre right like the Conservatives in the United Kingdom and even parties of the centre like Emmanuel Macron’s La République En Marche in France. This has led to a legitimisation of far-right discourse and a blurring of the boundaries between institutional and extra-parliamentary and potentially violent actors, culminating in ‘the fluent transition between organized right-wing parties, non-institutional anti-immigration movements and right-wing terrorism’ (see Gattinara & Pirro, 2019, p. 3; Koehler, 2018, p. 117). While direct causality between politicians’ statements at the macro-level and specific acts of violence at the micro level is difficult to establish, the open endorsement and promotion of views that identify and stigmatise specific communities as legitimate political targets, and, on occasion, directly call for violence against them is a worrying development. In central Italy in 2018, a far-right extremist shot and wounded six African migrants with the intention of ‘killing them all’ on the streets of Macerata. The previous year, the leader of the Lega party, Matteo Salvini, who later served as Deputy Prime Minister, declared in an electoral campaign that Italy ‘needs a mass cleansing, street by street, neighborhood by neighborhood, doing this the hard way if needed’ (Gattinara & O’Connor, 2018). Prior to the aforementioned attack in Hanau, the AfD ran a national and regional campaign explicitly identifying shisha bars with general criminality and, in one case, gang rape (Kinkartz, 2020). Such utterances by institutionally legitimate politicians and parties whose comments enjoy massive public reach can serve as indirect encouragement cues for lone actors to engage in violence (Lindekilde et al., 2018, p. 13). Even in the wake of the attack, the AfD’s regional party chairperson suggested that the shisha bars somehow brought the attack on themselves because they are disliked by many people (himself included) and that they are a public nuisance (in Hetrodt & Rösmann, 2020). Such pronouncements can be considered forms of ‘motivational frames’ convincing individuals or groups of the need to take action (Benford & Snow, 2000, p. 617). Encouragement cues are particularly important if they are expounded by figures considered in-group authorities, for example, nationally important political parties, and if the action demanded by the cues is likely to incur a potentially high cost, such as conducting a violent attack (Malthaner et al., 2017, p. 28). In the absence of support mechanisms and a sense of reciprocal obligation towards one’s comrades in violent movements, encouragement cues are of particular importance for lone actors. Although conspiracy theories are not a new phenomenon – take, for example, the notorious ‘Protocol of the Elders of Zion’ tract, which is more than 100 years old (Zia-Ebrahimi, 2018) – they are enjoying a recent period of unprecedented modern salience. They can facilitate violence by serving as ‘radicalisation multipliers’ (Emberland, 2020), and conspiratorial thinking tends to recur in violent milieus and subcultures from jihadism to the far-right (Amarasingam, 2019, p. 27). Until recently, conspiracies tended to circulate in existing extremist milieus or subcultures, whereas now they are expounded by so-called mainstream politicians, rendering them ever more dangerous by virtue of their enhanced credibility. Additionally, the promotion of conspiracy theories by authoritative political figures extends their reach to broader, often less-politicised audiences. Erstwhile US President Donald Trump and many of his allies in the Republican Party openly or implicitly endorsed the QAnon conspiracy, validating a belief system that has convinced several lone actors to carry out violent attacks in the United States (Fisher et al., 2016; Watkins, 2019). Conspiracy theorising is of course not confined to the United States; several conspiracy theories such as ‘the Great Replacement’ and ‘White Genocide’ have taken root in international far-right milieus and motivated attacks like the 2019 221

Francis O’Connor, Lasse Lindekilde, and Stefan Malthaner

one in Christchurch (Veilleux-Lepage et al., 2020). Regionally specific conspiracies such as the ‘love-jihad’ expounded by Hindutva extremists, which argues that Muslim men are deliberately seducing Hindu women in order to undermine the Hindu character of India, are a key pillar of violence against Muslims and other non-Hindus in the country (Leidig, 2020). The Corona virus pandemic has given impetus to invigorated conspiracy theorising by a wide array of political actors, including former President Trump, which has led to an increase in violence against people of Asian heritage. The long-term consequences of this in terms of violence are yet to unfold.

Simplification and copycatting of attacks The foreboding image of highly competent, hyper-focused lone actors lurking in the shadows waiting to strike is erroneous (Schuurman et al., 2018). Indeed, the majority of lone actors, in contrast to the meticulous minority, strike with rudimentary weapons against proximate targets and employ little planning or operational security measures (Lindekilde et al., 2019, p. 120). Increasingly, violent Islamist extremists and far-right milieus have made a virtue of lone actors’ inability to launch more ambitious and sophisticated attacks. Notoriously, ISIS’ spokesperson Abu Muhammad Adnani directed its supporters in the West to carry out any form of attack on ‘disbelievers’ using whatever weapons were at hand, resulting in a spate of lone-actor attacks and plots (GartensteinRoss & Barr, 2016). Many of these attacks have been conducted with bladed weapons and have resulted in limited numbers of casualties. The bloodiest attacks with mundane weapons have undoubtedly been run-over attacks, resulting in massive casualties in the 2016 attacks on the Berlin Christmas market and in Nice in 2016. Since then, the tactic has ideologically migrated and has also been used by far-right lone actors, with deadly consequences in London and at the ‘Unite the Right’ rally in Charlottesville in 2017 (Miller & Hayward, 2019). Target hardening has been proven to limit the effectiveness of vehicular attacks (Perry et al., 2018), but they remain a challenge for authorities. On the other hand, the use of bladed weapons in attacks has continued unabated in recent years, as evidenced by the cluster of attacks in France in 2020, including the decapitation of a schoolteacher near Paris and the killing of three people in a Nice church. Given the ubiquity of potential weapons, they are nigh on impossible to prevent, and as evidenced in France, they retain the ability to provoke dramatic government reactions (or indeed overreactions). Such responses can be viewed as successful political outcomes from an extremist perspective, resulting in greater stigmatisation of vulnerable communities and thus potentially inadvertently fuelling further grievances that could be harnessed by online and offline extremist milieus and networks.

Persistent online-offline interaction The fusion of online and offline radicalisation is an increasingly well-researched and nuanced subject of academic research. As McDonald and Whittaker surmise, the ‘use of the term online radicalization perpetuates a sharp distinction between the offline and online realms even though this distinction is problematic and appears to be unsustainable’ (2020, p. 42). Nevertheless, the myth of dangerous individuals ‘self-radicalized in the darkest corners of the internet’ (Der Spiegel, 2019) perseveres in certain academic circles and in public debates. To give one concrete example, the youth who attacked American soldiers at a German airport in 2011: was not a member of a terrorist organization, nor had he visited any of the infamous training camps for terrorists. His entire radicalization, from early attraction to jihadi preaching to the 222

Radicalisation of ‘lone actors’

final deadly mission, was accomplished online. . . . [His] is a typical case of the new trend of terrorists being engaged through the newest online platforms, commonly known as the ‘new media’ or ‘social media’. (Weimann, 2014, p. 1) Unfortunately, this stark conclusion, neatly fitting the self-radicalisation narrative, is based on incomplete data and is incorrect. Research on the same German case based on access to local restricted sources revealed that the attacker had in fact persistently attempted to embed himself in the local Islamist extremist milieu, attending radical preachers’ lectures and prayer circles, but he had simply gone unnoticed by others in the scene, and no efforts were made to involve him in any collective activities (Lindekilde et al., 2018). His case was therefore a textbook example of a combined pattern of online and offline radicalisation. As it is not easy to access restricted sources or have local knowledge of cases, there is arguably a methodological tendency towards confirmative bias overemphasising online dimensions. Information on perpetrators’ online radicalisation is often made available in their trials or through the activities of activists who access their social media histories, whereas offline ties are much harder to verify, and ties to other extremists or violent organisations can simply be denied by involved parties (see Gattinara et al., 2018). By conceptualising radicalisation as being conducted through networks of stronger and weaker ties in both online and offline realms, one can avoid this problematic dichotomy. Strong ties characterised by prolonged interactions and genuine bonds of affection or solidarity occur in everyday settings, such as university and sports clubs; however, they can also occur in online spaces, where many individuals spend hours of their lives. As Conway argues, ‘today’s Internet does not simply allow for the dissemination and consumption of “extremist material” in a one-way broadcast from producer to consumer, but also high levels of online social interaction around this material’ (2017, p. 80). Accordingly, the pendulum of most relevant social ties can in some cases swing more towards online spaces and in other cases more towards offline spaces; it is exceedingly rarely an either/or situation. These social ties also play different roles in different phases of the radicalisation process. Online ties can familiarise prospective radicals with some core ideological tenets or signpost them to places where further ties can be consolidated, for example, on a particular Discord forum or channel on Telegram. However, the motivation to act might be triggered in an offline space through familial contacts before the would-be attacker might once again seek certification in the online world from some authority that would sanction their attack.

New developments in online behaviour Political extremists, and right-wing ones in particular, have long demonstrated their openness to technological innovation and were early adapters of the Internet for political purposes (Conway et al., 2019, p.  3). In recent years, after political pressure on social media companies to shut down extremist accounts on platforms like Twitter and Facebook, there has been a migration to other platforms such as the various ‘Chan’ imageboards, Telegram, and, especially for right-wing extremists, Gab (Conway et al., 2019). In 2019, a series of right-wing lone-actor terrorist attacks in Christchurch, New Zealand, El Paso, Texas, and Poway, California, was announced and promoted, and efforts were even made to live stream these attacks directly on 8Chan. The attack on a synagogue in Halle and a mosque near Oslo was shared on successor platforms to 8Chan, Meguca, and EndChan, respectively (Baele et al., 2020, p. 2). 223

Francis O’Connor, Lasse Lindekilde, and Stefan Malthaner

This cluster of attacks does indeed present a qualitative change in modus operandi regarding lone actors’ usage of social media and potentially its role in the attackers’ radicalisation. Traditionally, attackers have used social media to promote their attacks and ideological or religious justifications to a wider offline public, attempting to intimidate their adversaries, embolden and inspire their supporters, and provoke desired responses from state actors. In short, attackers’ online communication around acts of violence was oriented towards broader audiences. This cluster of attackers’ use of the Chan platforms is noticeably different. First, efforts were made to live stream the attacks, involving the ‘absent audience’ or their online community in the attack (Schattka, 2020). In the case of the Halle attack, the attacker directly addressed the audience, reproaching himself for his failure to break into the synagogue and the poor quality of his weapons. Some of the ‘audience’ in turn criticised his low ‘score’ in comparison to the number of deaths in the Christchurch attack. This has been described as the gamification of terrorism, whereby gaming terminology and discourse are applied to real-world violence (Evans, 2019b). Second, the attackers in this cluster reference one another explicitly as inspirations for their violence. Interestingly, the Halle attacker communicated mostly in English, confirming his audience was international rather than oriented towards the local or national far-right milieu. All of the attackers attempted to frame their attacks as following in the legacy of Anders Breivik’s Utøya massacre in 2011, thus hinting at the self-narrativisation of their attacks. Finally, these platforms are image boards and contain little or often deliberately misleading text. They are characterised by in-jokes and memes with coded meanings only comprehensible to members of these online communities (Conway et al., 2019, p. 11). The use of ‘shitposting’, which can be understood as ‘the act of throwing out huge amounts of content, most of it ironic, low-quality trolling, for the purpose of provoking an emotional reaction in less Internet-savvy viewers’ (Evans, 2019a), is not designed to easily communicate with broader audiences as is the case, for example, in some jihadist martyr videos or attackers like Breivik. They are for internal consumption, aimed at winning the approval of their own anonymous community. The impact this internal orientation has on the types of attacks planned, targets, or likelihood of further copycat attacks remains unclear, as does the reciprocal dynamic between real-world attacks and these online communities (Baele et al., 2020). While these are new developments in the arena of radicalisation and lone-actor terrorism, which demand careful scholarly attention, it is also worth recalling that these are a small cohort of loneactor attacks, albeit bloody ones. The majority of other right-wing attackers in the last few years were radicalised by their socialisation in existing right-wing milieus, through real-world social ties with other violent extremists, and potentially polarising rhetoric from mainstream political actors. This was the case of the veteran member of the Neo-Nazi scene, Frank Steffen, who stabbed Cologne’s mayoral candidate Henriette Reker in 2015 over her support for asylum seekers. In other words, their radicalisation patterns demonstrate significant continuity with other far-right lone actors predating more recent technological developments. One could argue, somewhat provocatively, that there is a degree of confirmative bias in some of the emphasis on the ‘Chan’ milieu and other forms of online radicalisation. Once the attack has been committed, perpetrators are most commonly arrested or killed during the attack. This is clearly a significant judicial basis to prosecute them, combined with often publicly available online behaviour, thus confirming an individual trajectory to violence, backed up by strong evidence. However, it has been established that radicalisation is a social process shaped by one’s immediate social environment and the social ties within it (Lindekilde et al., 2018). To take the case of Stephan Baillet, the 2019 Halle attacker, a leading German media outlet described his self-radicalisation ‘in the darkest corners of the internet’ (Der Spiegel, 2019), an 224

Radicalisation of ‘lone actors’

explanation that summarily dismisses all of the other potential offline radicalising influences. ‘Balliet, from what is known so far, had no previous ties or contacts with the organized right-wing extremist milieu in Germany’ (Koehler, 2019, p. 17). Here, the focus should lie on ‘what is known so far’. His home state of Saxony-Anhalt is home to a thriving far-right scene; Halle, the city of the attack is home to the German headquarters of the far-right extremist association Generation Identity, and the regional branch of the AfD has open contacts with local neo-Nazi actors (Eichler, 2017). Indeed, his mother, with whom he lived until the attack included anti-Semitic and conspiracy theory remarks in a suicide note (Tag24, 2020). But as sufficient evidence for a conviction in such cases is readily available through perpetrators’ online behaviour, guilty pleas, and the physical proof from the attack itself, there seems little to be gained from digging deeper into the societal context that shaped the radicalisation process. Once again, this produces a bias towards looking for an explanation in terms of individual pathologisation rather than the social and political context that promotes and legitimises the stigmatisation of vulnerable minorities. Finally, by focusing on lone actors’ pre-attack broadcasts, manifestos, and online behaviour, much of which is assiduously tailored, there is a risk of prioritising the narrative and symbolic terrain preferred by the attackers themselves, which carries the danger of inadvertently enhancing their objectives. While this is more commonly an issue with media coverage, it can also seep into academic research and have potential policy implications.

Conclusion Like the field of radicalisation itself, research on lone-actor terrorism is extremely westernfocused, with the exception of research on Palestinian lone actors (Perry et al., 2017). It is simply unknown how widespread the phenomenon is in the non-Western world. Is there an overlap between regionally specific, individualised forms of violence like amok attacks common in southeast Asia (Horowitz, 2001, pp. 102–109) and lone-actor violence? Does the encouragement or even the implicit sub-contracting of violence by states to their ‘patriotic citizens’ in countries like Turkey to punish presumed supporters of terrorism (O’Connor & Baser, 2018) or in India to target non-Hindu populations (Leidig, 2020) lead to differing patterns of lone-actor terrorism? The application of existing concepts in non-Western contexts would also serve as an ulterior validity test and enhance a broader understanding of lone-actor radicalisation. So far, research on lone-actor terrorism and radicalisation has been almost completely ‘event driven’ and guided by urgent public and political demand for explanations of what can seem like inexplicable attacks (see Schuurman, 2019). This has in turn shaped the evolution of the field, leading to often fragmented and overlapping research that is often not part of a coherent research agenda. If the field is to consolidate itself, there is a need to embed patterns of lone-actor radicalisation in broader patterns of contentious politics as well as link them to broader political and social developments, not merely the newest form of online communication technology or as a response to another bloody attack. As suggested here, what may appear to be Einzelfall (single case) can often also be seen as part of a broader milieu or wave of attacks. Nevertheless, the field has qualitatively and quantitatively blossomed in recent years, characterised by commendable inter-disciplinarity and robust methodological pluralism. Importantly, the field’s glaring blind spot to right-wing extremism (Schuurman, 2019) is increasingly being rectified with much comparative and empirical research on instances of right-wing extremism and violence. However, it is important not to overcompensate and neglect violent Islamist extremism, which 2021 attacks in Austria (Bennhold et al., 2020) and France (Pantucci, 2020) have shown remains a substantial threat. 225

Francis O’Connor, Lasse Lindekilde, and Stefan Malthaner

Notes 1 The authors would like to thank Brian Kitt for his feedback on early drafts and Mette Jørgensen for assistance with creating the geographic visualisation of the data. 2 Correct as of 18 March 2021, in figure 14.1, attacks conducted in New Zealand and Australia are excluded from the visualisation. 3 The data was initially coded to include 11 variables, generating 3,366 original data points: if the attack occurred or was interrupted, the date of the attack, the country of attack, the city/region of attack, the type of attack (Bombing, Shooting, Stabbing, Run-over, Arson, or Other), and the number of casualties, and the number of injuries. On the perpetrator, it gathered the name, age at the date of the attack, ideology behind the attack, and the outcome of the attack, if the perpetrator was arrested, killed, committed suicide, or escaped. The coding process is an iterative one, contingent on new information being made available, often many years later, regarding attackers’ motivations or connections to other militant groups or networks. 4 This is clear when considering the travails of the reconstruction of 1993 in the GTD database (Acosta & Ramos, 2017).

References Acosta, B., & Ramos, K. (2017). Introducing the 1993 terrorism and political violence dataset. Studies in Conflict & Terrorism, 40(3), 232–247. https://doi.org/10.1080/1057610X.2016.1184061 Amarasingam, A. (2019). The impact of conspiracy theories and how to counter them: Reviewing the literature on conspiracy theories and radicalization to violence. In A. Richards, D. Margolin, & N. Scremin (Eds.), Jihadist terror: New threats, new responses (pp. 27–40). Bloomsbury. Baele, S. J., Brace, L., & Coan, T. G. (2020). The “Tarrant effect”: What impact did far-right attacks have on the 8Chan forum? Behavioral Sciences of Terrorism and Political Aggression, 1–23. https://doi.org/10.1 080/19434472.2020.1862274 Bell, S. (2020, September 25). RCMP arrests “Abu Huzayfah” for allegedly faking his past with ISIS. Global News. https://globalnews.ca/news/7350363/rcmp-arrests-abu-huzayfah-for-faking-past/ Benford, R. D., & Snow, D. A. (2000). Framing processes and social movements: An overview and assessment. Annual Review of Sociology, 26(1), 611–639. https://doi.org/10.1146/annurev.soc.26.1.611 Bennhold, K., Eddy, M., & Schuetze, C. F. (2020, November 3). Vienna reels from a rare terrorist attack. The New York Times. www.nytimes.com/2020/11/03/world/europe/vienna-attack-shooter.html Berntzen, L. E., & Sandberg, S. (2014). The collective nature of lone wolf terrorism: Anders Behring Breivik and the anti-Islamic social movement. Terrorism and Political Violence, 26(5), 759–779. https://doi.org/1 0.1080/09546553.2013.767245 Bjørgo, P. T., & Ravndal, D. J. A. (2019). Extreme-right violence and terrorism: Concepts, patterns, and responses. ICCT. https://doi.org/10.19165/2019.1.08 Böckler, N., Hoffmann, J., & Zick, A. (2015). The Frankfurt airport attack: A case study on the radicalization of a lone-actor terrorist. Journal of Threat Assessment and Management, 2(3–4), 153–163. http://dx.doi. org.ez.statsbiblioteket.dk:2048/10.1037/tam0000045 Bosi, L., Demetriou, C., & Malthaner, S. (Eds.). (2014). Dynamics of political violence: A process-oriented perspective on radicalization and the escalation of political conflict. Ashgate Publishing Limited. Bright, D., Whelan, C., & Harris-Hogan, S. (2020). Exploring the hidden social networks of “lone actor” terrorists. Crime, Law and Social Change. https://doi.org/10.1007/s10611-020-09905-2 Brusthom Ziamani “arrested on way to behead soldier”. (2015, February 9). BBC News. www.bbc.com/news/ uk-31305813 Conway, M. (2017). Determining the role of the Internet in violent extremism and terrorism: Six suggestions for progressing research. Studies in Conflict  & Terrorism, 40(1), 77–98. https://doi.org/10.1080/10576 10X.2016.1157408 Conway, M., Scrivens, R., & Macnair, L. (2019). Right-wing extremists’ persistent online presence: History and contemporary trends. https://icct.nl/publication/right-wing-extremists-persistent-online-presence-historyand-contemporary-trends/ Corner, E., & Gill, P. (2015, February). A false dichotomy? Mental illness and lone-actor terrorism. American Psychological Association. http://search.proquest.com.ez.statsbiblioteket.dk:2048/docview/1553716320/ abstract/FFBA1C8E1F4E4E49PQ/1

226

Radicalisation of ‘lone actors’ Crawford, B., & Keen, F. (2020, March 24). The Hanau terrorist attack: How race hate and conspiracy theories are fueling global far-right violence. Combating Terrorism Center at West Point. https://ctc.usma.edu/ hanau-terrorist-attack-race-hate-conspiracy-theories-fueling-global-far-right-violence/ Danzell, O. E., & Montañez, L. M. M. (2016). Understanding the lone wolf terror phenomena: Assessing current profiles. Behavioral Sciences of Terrorism and Political Aggression, 8(2), 135–159. https://doi.org/1 0.1080/19434472.2015.1070189 Der Spiegel. (2019, December 11). Far-right terrorism in Germany: Shooting exposes lapses in security apparatus. www.spiegel.de/international/germany/far-right-terrorism-in-germany-shooting-exposes-lapsesin-security-apparatus-a-1291075.html Eichler, H. (2017, July 15). Zugang zu sensiblen Daten: Rechter Spitzel im Ausschuss? Mitteldeutsche Zeitung. www.mz-web.de/sachsen-anhalt/landespolitik/zugang-zu-sensiblen-daten-rechter-spitzel-imausschuss-27973754 Emberland, T. (2020, February 24). Why conspiracy theories can act as radicalization multipliers of far-right ideals. C-REX – Center for Research on Extremism. www.sv.uio.no/c-rex/english/news-and-events/rightnow/2020/conspiracy-theories-radicalization-multipliers.html Evans, R. (2019a, March 15). Shitposting, inspirational terrorism, and the Christchurch Mosque Massacre. Bellingcat. www.bellingcat.com/news/rest-of-world/2019/03/15/shitposting-inspirational-terrorism-andthe-christchurch-mosque-massacre/ Evans, R. (2019b, August 4). The El Paso shooting and the gamification of terror. Bellingcat. www.bellingcat. com/news/americas/2019/08/04/the-el-paso-shooting-and-the-gamification-of-terror/ Fisher, M., Cox, J. W., & Hermann, P. (2016, December 6). Pizzagate: From rumor, to hashtag, to gunfire in D.C. Washington Post. www.washingtonpost.com/local/pizzagate-from-rumor-to-hashtag-to-gunfire-indc/2016/12/06/4c7def50-bbd4-11e6-94ac-3d324840106c_story.html Gartenstein-Ross, D. (2014). Lone Wolf Islamic terrorism: Abdulhakim Mujahid Muhammad (Carlos Bledsoe) case study. Terrorism and Political Violence, 26(1), 110–128. https://doi.org/10.1080/09546553.20 14.849921 Gartenstein-Ross, D., & Barr, N. (2016, July 20). Bloody Ramadan: How the Islamic state coordinated a global terrorist campaign. War on the Rocks. https://warontherocks.com/2016/07/bloody-ramadanhow-the-islamic-state-coordinated-a-global-terrorist-campaign/ Gattinara, P. C., & O’Connor, F. (2018, February 9). Analysis | An Italian neo-fascist shot 6 immigrants. So why won’t Italy’s political parties condemn xenophobia? Washington Post. www.washingtonpost. com/news/monkey-cage/wp/2018/02/09/an-italian-neo-fascist-shot-6-migrants-how-does-this-play-intothe-upcoming-elections/ Gattinara, P. C., O’Connor, F., & Lindekilde, L. (2018). Italy, no country for acting alone? Lone actor radicalisation in the neo-fascist Milieu. Perspectives on Terrorism, 12(6), 136–149. Gattinara, P. C., & Pirro, A. L. P. (2019). The far right as social movement. European Societies, 21(4), 447– 462. https://doi.org/10.1080/14616696.2018.1494301 Gill, P. (2015). Lone-actor terrorists: A behavioural analysis. Routledge. Gill, P., Clemmow, C., Hetzel, F., Rottweiler, B., Salman, N., Vegt, I. V. D., Marchment, Z., Schumann, S., Zolghadriha, S., Schulten, N., Taylor, H., & Corner, E. (2020). Systematic review of mental health problems and violent extremism. The Journal of Forensic Psychiatry & Psychology, 32(1), 51–78. https://doi. org/10.1080/14789949.2020.1820067 Gill, P., Horgan, J., & Deckert, P. (2014). Bombing alone: Tracing the motivations and antecedent behaviors of lone-actor terrorists. Journal of Forensic Sciences, 59(2), 425–435. https://doi.org/10.1111/ 1556-4029.12312 Goldman, J. J. (1995, December 9). 8 Die as Gunman sets afire N.Y. store tied to dispute: Violence: Harlem shop had been target of protests over clash between white owner and another business. Los Angeles Times. www.latimes.com/archives/la-xpm-1995-12-09-mn-12115-story.html Hamm, M. S., & Spaaij, R. (2017). The age of lone wolf terrorism. Columbia University Press. Hemmingby, C., & Bjørgo, T. (2016). The dynamics of a terrorist targeting process: Anders B. Breivik and the 22 July attacks in Norway. Springer. https://doi.org/10.1007/978-1-137-57997-3 Henley, J. (2002, March 28). Eight die in council chamber massacre. The Guardian. www.theguardian.com/ world/2002/mar/28/france.jonhenley1 Hetrodt, E., & Rösmann, T. (2020). AfD-Reaktion auf Hanau: “Shisha-Bars sind Orte, die vielen missfallen”. FAZ.NET. www.faz.net/aktuell/rhein-main/hessische-afd-weist-mitschuld-an-mord-in-hanauzurueck-16645531.html

227

Francis O’Connor, Lasse Lindekilde, and Stefan Malthaner Horgan, J. (2008). From profiles to pathways and roots to routes: Perspectives from psychology on radicalization into terrorism. The Annals of the American Academy of Political and Social Science, 618(1), 80–94. https://doi.org/10.1177/0002716208317539 Horowitz, D. L. (2001). The deadly ethnic riot. University of California Press. Joosse, P. (2017). Leaderless resistance and the loneliness of lone wolves: Exploring the rhetorical dynamics of lone actor violence. Terrorism and Political Violence, 29(1), 1–27. https://doi.org/10.1080/09546553. 2014.987866 Kaplan, J., Lööw, H., & Malkki, L. (2014). Introduction to the special issue on lone wolf and autonomous cell terrorism. Terrorism and Political Violence, 26(1), 1–12. https://doi.org/10.1080/09546553.2014.854032 Kearney, V. (2017, June 20). Christine Connor: Swedish model to “lone wolf dissident”. BBC News. www. bbc.com/news/uk-northern-ireland-40330784 Kenyon, J., Baker-Beall, C., & Binder, J. (2021). Lone-actor terrorism – a systematic literature review. Studies in Conflict & Terrorism, 1–24. https://doi.org/10.1080/1057610X.2021.1892635 Kinkartz, S. (2020, February 21). Die völkischen Brandstifter. Deutsche Welle. www.dw.com/de/ die-v%C3%B6lkischen-brandstifter/a-52462739 Koehler, D. (2018). Right-wing terrorism in the 21st century: The “national socialist underground” and the history of terror from the far-right in Germany. Routledge. Koehler, D. (2019). The Halle, Germany, Synagogue attack and the evolution of the far-right terror threat. Combating Terrorism Center at West Point. https://ctc.usma.edu/halle-germany-synagogueattack-evolution-far-right-terror-threat/ Leidig, E. (2020). Hindutva as a variant of right-wing extremism. Patterns of Prejudice, 1–23. https://doi.org /10.1080/0031322X.2020.1759861 Lindekilde, L., Malthaner, S., & O’Connor, F. (2018). Peripheral and embedded: Relational patterns of loneactor terrorist radicalization. Dynamics of Asymmetric Conflict, 12(1), 20–41. https://doi.org/10.1080/17 467586.2018.1551557 Lindekilde, L., O’Connor, F., & Schuurman, B. (2019). Radicalization patterns and modes of attack planning and preparation among lone-actor terrorists: An exploratory analysis. Behavioral Sciences of Terrorism and Political Aggression, 11(2), 113–133. https://doi.org/10.1080/19434472.2017.1407814 Louisville man arrested after threatening “9,000 kiddies”. (2019, March 22). Kentucky Today. www.kentuckytoday.com/stories/louisville-man-arrested-after-threatening-9000-kiddies,18530 Lüdecke, R. (2020). Hanau: Weder Einzelfall noch Einzeltäter. Amadeu Antonio Stiftung. www.amadeu-antonio-stiftung.de/hanau-weder-einzelfall-noch-einzeltaeter-59077/ Macklin, G. (2019a). The Christchurch attacks: Livestream terror in the viral video age. Combating Terrorism Center at West Point. https://ctc.usma.edu/christchurch-attacks-livestream-terror-viral-video-age/ Macklin, G. (2019b). The El Paso terrorist attack: The chain reaction of global right-wing terror. Combating Terrorism Center at West Point, 12(11), 1–10. Madhani, A. (2003, October 15). Activist faces sentence for shots at UN. Chicago Tribune. www.chicagotribune.com/news/ct-xpm-2003-10-15-0310150329-story.html Malkki, L. (2014). Political elements in post-Columbine school shootings in Europe and North America. Terrorism and Political Violence, 26(1), 185–210. https://doi.org/10.1080/09546553.2014.849933 Malthaner, S., & Lindekilde, L. (2017). Analyzing pathways of lone-actor radicalization: A relational approach. In M. Stohl, R. Burchill, & S. Englund (Eds.), Constructions of terrorism: An interdisciplinary approach to research and policy (pp. 163–180). University of California Press. Malthaner, S., Lindekilde, L., O’Connor, F., & Bouhana, N. (2017). D5.4 lone actor radicalisation subscript (PRIME, pp. 1–164). FP7 PRIME Project. McCauley, C., & Moskalenko, S. (2014). Toward a profile of lone wolf terrorists: What moves an individual from radical opinion to radical action. Terrorism and Political Violence, 26(1), 69–85. https://doi.org/10. 1080/09546553.2014.849916 McCauley, C., & Moskalenko, S. (2017). Friction how conflict radicalizes them and us (Revised and Expanded ed.). Oxford University Press. McCulloch, J., Walklate, S., Maher, J., Fitz-Gibbon, K., & McGowan, J. (2019). Lone wolf terrorism through a gendered lens: Men turning violent or violent men behaving violently? Critical Criminology, 27(3), 437–450. https://doi.org/10.1007/s10612-019-09457-5 McDonald, S., & Whittaker, J. (2020). Online radicalization: Contested terms and conceptual clarity. In J. R. Vacca (Ed.), Online terrorist propaganda, recruitment and radicalization (pp. 33–47). CRC Press.

228

Radicalisation of ‘lone actors’ Miller, V., & Hayward, K. J. (2019). “I did my bit”: Terrorism, Tarde and the vehicle ramming attack as an imitative event. The British Journal of Criminology, 59(1), 1–23. https://doi.org/10.1093/bjc/azy017 Mudde, C. (2010). The populist radical right: A pathological normalcy. West European Politics, 33(6), 1167– 1186. https://doi.org/10.1080/01402382.2010.508901 Murphy, S. (2019, September  3). White supremacist who sent fake anthrax to Queen jailed for 12 years. The Guardian. www.theguardian.com/uk-news/2019/sep/03/white-suprematist-who-sent-fakeanthrax-to-queen-jailed-for-12-years-david-parnham Nesser, P. (2012). Single actor terrorism: Scope, characteristics and explanations. Perspectives on Terrorism, 6(6), 61–73. Neumann, D. P. R., & Basra, R. (2018). The crime-terror Nexus in the United Kingdom and Ireland. Crime Terror Nexus. O’Connor, F., & Baser, B. (2018). Communal violence and ethnic polarization before and after the 2015 elections in Turkey: Attacks against the HDP and the Kurdish population. Southeast European and Black Sea Studies, 18(1), 53–72. https://doi.org/10.1080/14683857.2018.1451035 Pantucci, R. (2020, October 23). A murder in Paris: France’s grim reminder of the terrorism threat. RUSI. https://rusi.org/commentary/murder-paris-frances-grim-reminder-terrorism-threat Parveen, N. (2016, August 9). Glasgow murder of Asad Shah spurred by sectarianism in Pakistan. The Guardian. www.theguardian.com/uk-news/2016/aug/09/glasgow-of-asad-shah-spurred-by-sectarianism-in-pakistan Perry, S., Hasisi, B., & Perry, G. (2017). Who is the lone terrorist? A study of vehicle-Borne attackers in Israel and the West Bank. Studies in Conflict & Terrorism, 41(11), 899–913. https://doi.org/10.1080/105 7610X.2017.1348101 Perry, S., Hasisi, B., & Perry, G. (2018). Lone terrorists: A study of run-over attacks in Israel. European Journal of Criminology, 16(1), 102–123. https://doi.org/10.1177/1477370818769257 Ramsay, G. A., & Marsden, S. V. (2015). Leaderless global Jihadism: The paradox of discriminate violence. Journal of Strategic Studies, 38(5), 579–601. https://doi.org/10.1080/01402390.2015.1032408 Ravndal, J. A. (2016). Right-wing terrorism and violence in Western Europe: Introducing the RTV dataset. Perspectives on Terrorism, 10(3). www.terrorismanalysts.com/pt/index.php/pot/article/view/508 Reuters Staff. (2016, March  3). Missouri man admits to attempted fire bombing of congressman’s office. Reuters. www.reuters.com/article/us-kansas-firebomb-idUSKCN0W52Q7 Sack, K. (1997, February 26). Florida man put bomb at Peres speech, police say. The New York Times. www. nytimes.com/1997/02/26/us/florida-man-put-bomb-at-peres-speech-police-say.html Schattka, C. (2020). Halle (Saale), 9. Oktober 2019 Protokoll eines Anschlags. In S. Malthaner & T. Hoebel (Eds.), Von einsamen Wölfen und ihren Rudeln. Zum sozialen Phänomen des Einzeltäters: Mittelweg 36, Heft 4–5 Oktober/November 2020 (1st ed., pp. 45–62). Hamburger Edition. Schuurman, B. (2018). Research on terrorism, 2007–2016: A review of data, methods, and authorship. Terrorism and Political Violence, 32(5), 1011–1026. https://doi.org/10.1080/09546553.2018.1439023 Schuurman, B. (2019). Topics in terrorism research: Reviewing trends and gaps, 2007–2016. Critical Studies on Terrorism, 12(3), 463–480. https://doi.org/10.1080/17539153.2019.1579777 Schuurman, B., Bakker, E., Gill, P., & Bouhana, N. (2018). Lone actor terrorist attack planning and preparation: A data-driven analysis. Journal of Forensic Sciences, 63(4), 1191–1200. https://doi. org/10.1111/1556-4029.13676 Schuurman, B., Lindekilde, L., Malthaner, S., O’Connor, F., Gill, P., & Bouhana, N. (2019). End of the lone wolf: The typology that should not have been. Studies in Conflict & Terrorism, 42(8), 771–778. https://doi. org/10.1080/1057610X.2017.1419554 Simon, J. D. (2016). Lone wolf terrorism: Understanding the growing threat. Prometheus Books. Spaaij, R. (2011). Understanding lone wolf terrorism: Global patterns, motivations and prevention. Springer Science & Business Media. Stack, L. (2019, April 4). Christchurch messages on Facebook lead to charge of lying to F.B.I. (Published 2019). The New York Times. www.nytimes.com/2019/04/04/nyregion/thomas-alonzo-bolin-whitesupremacist.html Sturm, G. (2020, February 20). Grünen-Politikerin über Terror in Hanau: “Rechtsterrorismus ausgeblendet”. Die Tageszeitung: Taz. https://taz.de/!5665289/ Tag24. (2020, July 29). Tag 4 im Halle-Prozess: Mutter des Attentäters wollte sich umbringen. TAG24. www. tag24.de/justiz/vierter-tag-im-halle-prozess-stephan-balliets-familie-und-umfeld-im-fokus-1593976

229

Francis O’Connor, Lasse Lindekilde, and Stefan Malthaner Veilleux-Lepage, Y., Daymon, C., & Amarasingam, A. (2020). The Christchurch attack report: Key takeaways on Tarrant’s radicalization and attack planning. The International Centre for Counter-Terrorism. https://doi.org/10.13140/RG.2.2.15294.31047 Watkins, A. (2019, July 21). He wasn’t seeking to kill a Mob Boss: He was trying to help Trump, his lawyer says. The New York Times. www.nytimes.com/2019/07/21/nyregion/gambino-shooting-anthony-comellofrank-cali.html Weimann, G. (2014). New terrorism and new media. Wilson Centre. Zia-Ebrahimi, R. (2018). When the elders of Zion relocated to Eurabia: Conspiratorial racialization in antisemitism and Islamophobia. Patterns of Prejudice, 52(4), 314–337. https://doi.org/10.1080/00313 22X.2018.1493876

230

15 RADICALISATION OF ‘FOREIGN FIGHTERS’ Jeanine de Roy van Zuijdewijn

The emergence of studies into foreign fighters A cursory search of academic databases using the keyword ‘foreign fighters’ seems to suggest that they are a truly modern phenomenon. The number of studies on this topic prior to 2010 was low. A Google Scholar search displays 2,750 hits for the period 1900–2009, compared to 17,100 hits for the decade that followed (2010–20). The level of academic interest reflected in these figures does not correspond to the level of involvement of foreign fighters in conflicts; they have been present in conflicts for centuries. As a matter of fact, before the emergence of nation-states with large armies that gradually developed between the 17th and early 19th centuries, fighting forces often consisted of large numbers of foreign mercenaries. When nation-states could start to rely on their own national forces, this changed, and the number of mercenaries involved in conflicts slowly decreased (O’Connor & Piketty, 2020, p. 2). Nevertheless, the phenomenon of people taking up arms abroad remained a common feature of conflicts. David Malet studied the involvement of foreign fighters in conflicts between 1816 and 2005, showing that 70 out of 331 civil conflicts attracted such combatants (Malet, 2013, p. 43). In that sense, it is no surprise that conflicts today also attract foreign fighters. What has changed in recent years, however, is the unprecedented surge in both academic and public interest in the topic. This recent increase in interest did not emerge from academic debates in conference halls. The spark that ignited this (academic) fire could be found in Syria around 2011. Thousands of people from across the globe travelled to the country, initially to partake in the fight against the regime of President Bashar al-Assad. The nature of this conflict rapidly transformed, as jihadist groups such as Jabhat alNusra (later Jabhat Fateh al-Sham) and Islamic State (IS) increasingly gained power and eventually controlled large swaths of territory in Syria and Iraq. Estimates of the total number of fighters vary, but the United Nations Security Council calculated in late 2017 that over 40,000 foreign fighters from over 110 countries had travelled to the region (United Nations, 2017). This steep rise in the number of foreign fighters was followed by a wave of around 17,000 publications that referred to the topic. Figuratively speaking, this meant that a new publication emerged for every third foreign fighter who joined the conflict. Among those thousands of foreign fighters were around 6,000 originating from European Union member states (5,684) and North America (430) (Pokalova, 2019, p. 818).

231

DOI: 10.4324/9781003035848-17

Jeanine de Roy van Zuijdewijn

Most scholars were interested in this diverse group of Western jihadist fighters, who dominated the news for many years. The question of why men (and later also women) from relatively well-off backgrounds decided to join an organisation that has been involved in terrible crimes, including genocide, according to the UN (UN News, 2021), intrigued academics. The rise in studies was also a response to the growing concerns among policymakers, practitioners, and the wider public about the potential threat these fighters could pose. This threat was increasingly defined as the one that returnees would pose upon returning to their home countries (Bakker & De Roy van Zuijdewijn, 2015). As a result, much of our understanding today is based on this specific sub-set of jihadist foreign fighters and has been informed by policy-oriented questions or concerns. More than a decade after the start of the Syrian conflict, it is high time to assess where the academic community stands with regard to understanding the foreign fighter phenomenon. This chapter aims to assess the state of the art of research into foreign fighters. As many recent studies have investigated jihadist foreign fighters, this cohort features prominently in this chapter. However, the chapter also adopts a broader view by looking at various studies into other types of foreign fighters, using examples from, among others, the Spanish Civil War and the Waffen-SS. Many studies into the radicalisation process of foreign fighters have tried to answer one or more of the following questions: who are the foreign fighters, what are their motivations, and how are they mobilised or recruited? This chapter will discuss how academics have approached and answered these three questions. It will examine how and why scholars have different interpretations of the role of specific factors in the radicalisation process of foreign fighters. The chapter will end by assessing where we stand today and reflecting on some of the main gaps and questions for future research.

The definition question: contested elements Similar to most terms in the field of terrorism studies, defining foreign fighters has not been straightforward or uncontested. Malet defined them as ‘non-citizens of conflict states who join insurgencies during civil conflicts’ (Malet, 2013, p. 9, originally proposed in 2009). Hegghammer built on Malet’s definition to provide his own description of the term: an agent who (1) has joined, and operates within the confines of, an insurgency, (2) lacks citizenship of the conflict state or kinship links to its warring factions, (3) lacks affiliation to an official military organization, and (4) is unpaid. (Hegghammer, 2011, pp. 57–58) Both authors position foreign fighters within the context of an insurgency where they aid the non-state side, which is generally understood to be the weaker party involved in a conflict. Foreigners fortifying the ranks of an official military organisation, such as state forces, would generally not be considered foreign fighters, nor are soldiers who are sent abroad as part of their national military service. Yet, the question can be raised why foreigners who join or receive support from states should be excluded, such as the thousands of Shia fighters backed by Iran who travelled to Syria after 2012 to keep Assad’s regime in power (Reiff, 2020; Smyth, 2015). The reasoning behind excluding such fighters is tied to the fourth element of Hegghammer’s definition: such state-sponsored fighters are usually paid. Foreign fighters who join non-state factions, however, also often receive some form of payment; examples are found in Syria and with regard to those fighting with separatists in Ukraine (Bakker & De Bont, 2016, p. 849; Mareš, 2017, p. 33; Rácz, 2017, p. 68). 232

Radicalisation of ‘foreign fighters’

The question of payment is therefore not necessarily dichotomous, but it does point to an underlying assumption about who foreign fighters are: they fight because they believe in a cause, not because of a desire to make money or to simply have a form of employment. Some scholars thus chose to approach this issue of monetary reward less rigidly. Moore and Tumelty, for instance, defined foreign fighters as ‘non-indigenous, non-territorialized combatants who, motivated by religion, kinship, and/or ideology rather than pecuniary reward, enter a conflict zone to participate in hostilities’ (Moore & Tumelty, 2008, p. 412). This definition would also include those who join the non-insurgent side of a conflict, but it excludes mercenaries, who often show little to no interest in the reasons behind the conflict itself (UN, 1989). These examples show that the definition of foreign fighters is not only based on what they do but also based on why they do it. Another debated element of the definition is the extent of overlap between the terms foreign fighter and terrorist, as both could travel abroad to take part in hostilities. This issue became more pressing over recent years against the backdrop of the thousands of foreign fighters who joined IS, a designated terrorist organisation according to many countries and international organisations such as the UN and EU. The same overlap was visible in the case of those who joined the infamous Al-Qaeda training camps in the Afghanistan-Pakistan border region while participating in combat activities around the turn of the century (Wright, 2006, pp. 322, 347). Historically, however, such overlap has been rare, and it does not seem to warrant the general and frequent conflation of the two terms. There is usually a difference between those who travel to fight in an insurgency and those who go abroad to attend a terrorist training camp or acquire skills with the primary aim of applying those at home (Mendelsohn, 2011, p.  193; De Roy van Zuijdewijn, 2014b, p. 61). The current terminology of ‘foreign terrorist fighters’ that is widely used in policy circles might increase such conflation (United Nations Security Council, 2014). The use of the term also draws attention away from the fact that at least hundreds of foreign fighters have joined the Syrian-Kurdish People’s Protection Units (YPG) or the Iraqi-Kurdish Peshmerga active in the same conflict and are not listed as terrorist organisations by the UN, EU, or US (Tuck et al., 2016, p. ii). A final element of the definition is linked to the question of when someone can be called a fighter. Historian Nir Arielli preferred to use the term ‘foreign volunteers’, as many fighters preferred to call themselves. Examples are the Freiwilligen, who joined the Waffen-SS during World War II, or the volontaires who assisted France in World War I (Arielli, 2017, p. 5). While this term might sound euphemistic, Arielli explained that the term indicates that people could also take on non-combat roles during a conflict. Performing more supportive, non-combat duties is vital to sustaining a conflict. This question regarding the extent of involvement in combat duties for someone to be considered a foreign fighter became more pressing in recent years, particularly after IS declared its Caliphate in 2014 and attracted a much broader audience of supporters. Thousands of women from across the globe travelled to the Caliphate to support the organisation, albeit rarely in a direct combat role (Cook & Vale, 2018). Nevertheless, these women are widely perceived to be part of the ‘foreign fighter phenomenon’ and are generally included in the statistics of foreign fighter counts. Thus, when referring to foreign fighters today, it implicitly also includes those who play a supporting role in combat activities. It often remains unclear how direct and concrete those efforts need to be. This question also has an important legal dimension. Some public prosecution services try to circumvent the issue by charging women with participation in a terrorist organisation rather than participation in an illegal fighting force. Concluding this section on the definition question, there seems to be consensus that foreign fighters are individuals who travel to another country to join a conflict for ideological reasons. Contested elements include the side they fight on (state or non-state), which is indirectly connected to the issue of payment versus the primacy of ideological motivations; the difference between a 233

Jeanine de Roy van Zuijdewijn

terrorist and a foreign fighter; and lastly, how concrete their contribution to the combat itself needs to be in order to be considered fighters.

The ‘who’ and ‘why’ questions: analytical lenses to understand the profiles Over the past years, many studies into the profiles of foreign fighters and their motivations have been published. The majority of these studies have focused on those who joined jihadist groups in Syria and Iraq. Before discussing the main results and debates regarding these jihadists, it is useful to assess the state of the art when it comes to understanding the profiles and motivations of foreign fighters more broadly. This section will first outline some of the broader and historical perspectives and then discuss studies that focus on case studies or individual fighters.

Broader and historical perspectives versus individual perspectives The earlier-mentioned works of Malet and Arielli are prime examples of such a broader approach to the foreign fighter phenomenon. Malet offered a unique perspective by studying foreign fighter mobilisations in widely different contexts, ranging from the Texas Revolution (1835–6), the Spanish Civil War (1936–9), and the Israeli War of Independence (1947–9) to the mujahideen, or holy warriors, who fought in Afghanistan (1978–92). Despite the wide variety in the backgrounds and ideological commitments of the parties involved, Malet noted a common pattern in recruitment: the recruitment messages aimed to convince potential fighters of the idea that a transnational community was facing an existential threat and hence, needed to be helped (Malet, 2013, p. 211). Thus, foreign fighters join a battle to protect a group, community, identity, or set of values. Examples that demonstrate the emphasis on defensive frames can be found across conflicts with widely different ideological configurations. For instance, those who fought alongside the Republicans in the Spanish Civil War wanted to stop the rise of fascism and were often motivated by communist views. The thousands who fought on the other side, aiding General Franco and his forces, wanted to protect Catholicism from the threat of communism (Keene, 2001; Petram & Kruizinga, 2020, p. 73). Similarly, those who heeded the calls of sheikh Azzam in Afghanistan in the 1980s responded to his claims that the ummah – the Islamic community – was under attack by the Soviet Union – a godless regime – and had to be defended (Azzam, 1987, pp. 15–16; Hegghammer, 2020, p. 303; De Roy van Zuijdewijn, 2014a, pp. 36–37). Some of the ‘volunteers’ who joined the Waffen-SS on the eastern front during World War II explained that they did not primarily fight to support National Socialism, but that they were first and foremost engaged in a protective war of ‘Europe against Bolshevism’ (Smith et al., 1999, p. 76). Foreign fighters on the side of the Ukrainian forces who went there around 2014–6 and again in 2022, wanted to expel the Russians who threatened the integrity of the Ukrainian nation by invading the Crimean Peninsula in 2014 and even larger parts of the country in 2022 whereas many Russian fighters who bolstered separatist groups claimed to fight to stop the rise of fascism that they allegedly witnessed in Ukraine (Rácz, 2017, p. 68; Weijenberg & De Roy van Zuijdewijn, 2021). Despite the large differences between those conflicts, they overlap in the sense that the main mobilisation rhetoric referred to a defensive war against a force that threatened the core identity and values of a community. While Malet focused mostly on the messages of recruitment instead of the personal motivations of fighters, he attempted to briefly link these two in his conclusions. He argued that the potential recruitment power of foreign fighter movements could be limited by decreasing the salience of transnational identities for individuals. Instead, feelings of attachment to the nation and fellow citizens should be strengthened, as they would decrease the chance of someone becoming a foreign 234

Radicalisation of ‘foreign fighters’

fighter (Malet, 2013, p. 213). This observation shifts the focus from propaganda and recruitment towards the role of identity and meaning at the individual level. If someone is strongly tied to his or her own environment, attempts to activate a different, transnational identity might be less successful. This idea was further examined by Arielli, who identified the ‘search for meaning’ as the common denominator for all foreign fighters across conflicts (Arielli, 2017, p. 80). He argued that this search for meaning is both internal and social. Arielli provided several examples where he witnessed this search for meaning. For instance, he discussed Western fighters who joined Kurdish forces in the fight against IS, quoting a Canadian fighter: ‘I have to admit I’m having the time of my life out here. If you ever have the chance, I strongly recommend fighting terrorists as a great way to get yourself out of your various existential funks’ (Arielli, 2017, p. 86). The frame of fighting against terrorists provided the ideological framework that supported someone in dealing with personal issues.

Ideological versus personal factors It is this mix between ideological convictions and personal, more mundane factors motivating fighters that is often found in detailed accounts of individual fighters. This mix, however, also seems to trouble those who want a straightforward answer to the question of which set of factors is most important when understanding the radicalisation of foreign fighters. As will be discussed below in the context of jihadist foreign fighters, scholars do not agree on the relative weight of various factors to explain why someone becomes a foreign fighter. Some of these differences reflect genuine academic or fundamental disagreements about what mobilises people to fight, or more broadly, to become violent extremists. This chapter would argue, however, that these differences are at least also partially due to a difference in focus. Scholars who use broader, historical, or comparative perspectives aim to explain why foreign fighters join particular conflicts on a macro level. They often discuss the ideological configurations and broader trends in mobilisation and recruitment. Such studies rely on the propaganda and recruitment material of insurgent groups, sources that tend to emphasise ideological frames or messages. Other studies adopt a more micro-level focus, diving into the personal (hi)stories of foreign fighters while often paying little to no attention to recruitment messages or the broader historical and cultural contexts. In that case, mundane factors will start to appear, including personal grievances, life-changing events, and social factors. It is virtually inevitable that researchers will find mundane factors when zooming in on someone’s life – regardless of the topic being studied – while it would be complicated to discover them when exclusively looking at broader trends but not at the individuals involved. Similarly, broader ideological trends or historical settings easily get lost when you dive deep into a person’s trajectory. As Dutch football legend Johan Cruijff once said: ‘you’ll only see it when you understand it’. Several scholars have attempted to adopt a broader focus by looking at ideological factors and discussing how these might play out on the individual level. For instance, Rekawek noted that many foreign fighters who joined the conflict in Ukraine during the first wave of mobilisation around 2014–6 had ‘extreme right wing or “nationalist” convictions’ and often came from similar ideological milieus (Rekawek, 2020, p. 3). Yet he also concluded that the decision about which side they were fighting, separatist or pro-Ukrainian, was often not based on ideological motivations but on individual or group connections. He argued that this demonstrates the ‘shallowness of the ideological convictions’ of those involved (Rekawek, 2020, p. 4). Another example is a study of Dutch fighters who joined the International Brigades in the Spanish Civil War. Many of those 235

Jeanine de Roy van Zuijdewijn

fighters had long-standing communist sympathies and were members of the Dutch Communist Party. Their decision to join the fight, however, was often preceded by various forms of personal hardship. Many young Dutch men had become unemployed in the late 1930s and, in some cases, had also lost their sense of purpose in life. They found a renewed sense of identity and meaning in fighting fascism in Spain (Petram & Kruizinga, 2020, pp. 32–36). Similarly, around half of the American volunteers in the Lincoln Brigade said they had been unemployed prior to joining the conflict (Arielli, 2013). Such a conjunction of political convictions and personal circumstances was also found in the case of the volunteers in the Waffen-SS, the military branch of the SS, who fought in World War II. It has been estimated that more than half of its (at least) 600,000 soldiers did not originate from Germany but predominantly came from North Western European countries (Gingerich, 1997, p. 815; Smith et al., 1999, p. 73). Among them were around 6,000 Danish fighters, whose ideological convictions differed widely. For a significant number of fighters, there was no ‘consistent ideological belief that would identify them as convinced National Socialists’ (Smith et al., 1999, p. 88). For the more highly ranked officers, however, ideological motivations might have played a crucial role. Such a stronger focus on ideological motivations was emphasised in another study of Danish, Swedish, and Swiss Waffen-SS volunteers. The author sketched the following profile of these fighters: a man of ‘a middle or upper-class upbringing, well-educated and integrated into his society. Moreover, he would have a cosmopolitan outlook or an openness to the wider world [combined with an] ideological inclination towards fascism with an imbedded pan-Europeanist tension’ (Gutmann, 2013, pp. 591–592). There have been numerous examples of highly educated, successful individuals who decided to take up arms to support Hitler’s National Socialist regime. This description runs counter to what many would expect (or find morally acceptable) when referring to those who voluntarily played a role in such a horrific episode in European history. There is a widespread tendency to label those whose actions we find incomprehensible as one of the following two options: they are either long-time, die-hard ideological fanatics who are no longer susceptible to any form of reasoning, or they are desperate, low-opportunity, or mentally challenged individuals who fell prey to life’s circumstances and hence were too weak to resist the lure of evil. Neither description, despite their instinctive appeal, suffices to understand those who joined the Waffen-SS (Gutmann, 2013, p. 588). Similar simplified descriptions have also surfaced time and again in relation to the thousands of jihadist fighters who joined another genocidal regime: that of IS.

Jihadists foreign fighters since 2011 Various scholars have decided to accept the challenge of studying, almost in real-time, the large numbers of foreign fighters who joined the fight in Syria and Iraq. Many initially tried to make sense of ‘their’ foreign fighter contingent by looking at data and profiles of those who departed from their countries of origin (Bakker & De Bont, 2016; Gustafsson & Ranstorp, 2017; Marone & Vidino, 2019; Reynolds & Hafez, 2019; Van Ostaeyen, 2017). While these studies acknowledge the difficulty of establishing a profile of the fighters, most still attempt to do so. While the list of included characteristics varies per study, there seems to be tacit agreement that certain factors are relevant to include. Commonly found are gender, age, ethnic background, religious beliefs, socio-economic status, mental health issues, and crime history. Dawson analysed the findings of 34 studies on Western foreign fighters published between 2014 and 2019 and the profiles that these studies sketched. Close to 80% of the foreign fighters were male; the average age was about 26. Many were citizens of the countries they departed from, but they had an immigrant 236

Radicalisation of ‘foreign fighters’

background and around 15% might have converted to Islam (Dawson, 2021, p. 2). Results that strongly diverged between countries or were less conclusive were related to socio-economic status, mental health issues, and crime history. Those from continental Europe often came from lower socioeconomic backgrounds, while this was not commonly observed in the North American or British context (Dawson, 2021, p. 2). Data on mental health and a criminal past was often inconclusive (Dawson, 2021, p. 3). While most scholars who sketched such a profile of their national contingent warned their readers not to draw (causal) conclusions from their findings, their decision to include certain characteristics implies that these hold some kind of explanatory value. We do not find any studies that include the musical preferences, dietary requirements, shoe size, or hair colour of foreign fighters in an attempt to educate the reader. Such profiles influence perceptions and could be used to legitimise policies applicable to groups of people who share such characteristics. A well-known issue with profiles, however, is the specificity problem: many individuals share the same characteristics, but only very few of those become foreign fighters (Sageman, 2004, p. 69). This raises doubt about the utility of such profiles, especially when there is no (hypothesised) underlying mechanism that could explain the correlation of some factors with the act of becoming a foreign fighter.

Push and pull factors Moving beyond the profiles, many studies looking into jihadist fighters attempted to answer the ‘why question’ by distinguishing between so-called push and pull factors: what pushes someone away from their own country and what pulls them towards a particular conflict? This discussion shares similarities with the earlier mentioned discussion on ideological versus personal factors for non-jihadist fighters. Ideology, in this case the belief in the necessity to engage in jihad, is often considered to be the most important pull factor (others include the desire to help, adventure, or a sense of comradeship), while personal issues are often discussed as part of the push factors. Oftenmentioned push factors pertain to socio-economic status, especially the issue of relative deprivation and feelings of injustice and discrimination that might be linked to it. It seems common sense to note that an individual is affected by both push and pull factors. Scholars do not agree on the significance of factors, however, and ascribe greater explanatory value to some than to others. Coolsaet, for instance, questioned the alleged dominance of explanations that focused on the role of ideology. He argued that group dynamics and other factors that are part of the socialisation process are more important to understand why someone becomes a foreign fighter or a terrorist. As he states, ‘in this process of political radicalisation into extremism, it is not the narrative (i.e. the ideology) that eventually lures an individual into terrorism’ (Coolsaet, 2015, p. 6). He also pointed at the importance of socio-economic conditions. While he and others have warned that these do not form a ‘direct pathway’ to terrorism, they are nevertheless seen as significant markers of the environment of would-be fighters (Bakker & De Bont, 2016, p. 845; Coolsaet, 2015, p. 16). References to socio-economic factors are widely found and seem to have gained more support in the past decade after having been deemed irrelevant or even unscientific for many years. Examples of studies that aimed to redirect the spotlight on such factors include Weenink’s work (which also focused on mental health, among other factors) and Hegghammer’s plea for a renewed academic investigation of the terrorism-poverty link during the Society for Terrorism Research Conference in 2016 (Hegghammer, 2016; Weenink, 2019). Scholars do not necessarily consider socio-economic factors to be significant in absolute terms. They mostly surface in relative terms, often in conjunction with other factors. For instance, a study into Swedish foreign fighters 237

Jeanine de Roy van Zuijdewijn

concluded that more than 70% came from so-called exposed areas, ‘socially deprived areas hit by high criminality and low socio-economic status’ (Gustafsson & Ranstorp, 2017, p. 104). Such studies draw attention to a wide range of structural factors that compose the environment in which an individual is situated and that might push someone to become violent. Similarly, there has also been renewed attention on a so-called crime-terror nexus (Basra & Neumann, 2016; Ljujic et al., 2017; Marcus, 2021). This move to more thoroughly investigate push factors and increasingly stress their importance as an explanatory factor has been met with criticism. For instance, Dawson questioned the alleged dominance of such factors in terrorism studies, which he argued has resulted in ideology and religion being considered mostly irrelevant (Dawson, 2021, p. 31). He ascribed this (perceived) widespread neglect of the role of ideology and religion to a lack of understanding of religiosity among scholars in Western, secular societies. Scholars might interpret quick conversion processes as being unauthentic and therefore showing their irrelevancy. Conversely, Dawson argued that such quick and intense changes in someone’s ideological or religious outlook, when followed by a decision to engage in violence, actually demonstrate their significance (Dawson, 2021, p. 32). He warned not to overstate the importance of religion or ideology: ‘religiosity alone is not explanatorily adequate, any more than socioeconomic push factors. Both explanations lack the requisite specificity’ (Dawson, 2021, p. 32). This is similar to Coolsaet’s warning that socio-economic factors are not considered to be a direct pathway to terrorism. In another study, Dawson and Amarasingam conducted interviews with foreign fighters from several (predominantly Western) countries and concluded that the relative importance of pull factors was probably greater than that of push factors (Dawson & Amarasingam, 2017, p. 206). As the authors also noted, acknowledging that both types of factors play a role seems sensible. Researchers could try to determine the importance of certain factors in particular circumstances, although that remains an arduous task.

Macro-level approaches A different way to approach the questions of who foreign fighters are and what drives them is to look at country-specific factors. Is there any link between country characteristics and the number of foreign fighters, and could they explain what pushes individuals away from their own countries to take up arms abroad? Scholars have performed various forms of regression analysis on national statistics to see if any correlation with the number of foreign fighters could be found. These studies were often motivated by a desire to test assumptions about structural factors and their link to violence. Most studies found little evidence for the importance of structural factors on the national level, such as those linked to poverty. For instance, Pokalova looked at various political, economic, demographic, and social factors in 190 countries – including 103 from which fighters departed – and performed regression analysis to establish how these correlated to the number of foreign fighters. Interestingly, she found that countries with higher Human Development Index (HDI) levels were positively correlated to higher numbers of foreign fighters, yet so was the unemployment rate, but only in non-Muslim majority countries (Pokalova, 2019, p. 809). Similarly, Benmelech and Klor looked at various economic, political, and social conditions. Among other things, they also found that the number of foreign fighters was positively related to the HDI level (as well as GDP). Their conclusion was that economic or political factors might not explain the foreign fighter phenomenon and that ideology and social issues such as assimilation would be more relevant: ‘the difficulties of Muslim immigrants to integrate into rich, secularized and homogenous societies 238

Radicalisation of ‘foreign fighters’

might lead them to radicalise’, the authors concluded based on their statistical analysis (Benmelech & Klor, 2020, p. 1459). While Pokalova did not include any indicators related to relative deprivation, Benmelech and Klor included the Gini coefficient – a measure of income inequality – in some of their models. They found that, contrary to widespread expectations, the number of foreign fighters was negatively correlated with the level of economic inequality. They also found that countries with less ethnic and linguistic fractionalisation generally saw more individuals departing to become foreign fighters (Benmelech & Klor, 2020, pp. 1468, 1477). Both of these studies thus seem to provide little evidence for socio-economic (and to some extent social) push factors on a structural or country level. It must be acknowledged, however, that such studies cannot be used to explain or assess personal experiences. They cannot look into the minds of those involved to measure the experienced level of deprivation or inequity, regardless of national statistics.

Limits to understanding the ‘who and why’ Similar to the ongoing debates about violent extremism, radicalisation, and terrorism more broadly, scholars continue to discuss why people become foreign fighters. While research can help improve our understanding, there might be limits to what can be scientifically proven. A realistic, or perhaps pessimistic, view would entail that it might be very difficult, if not impossible, to assess the ‘true’ reasons why someone did something. As Morrison recently noted, asking such broad questions in terrorism studies means that ‘we are potentially setting ourselves up for failure’ (Morrison, 2021, p. 4). Researchers have to deal with a plethora of challenges. This becomes even more pertinent when interviewing foreign fighters with the desire to use their testimonies to assess their motivations rather than considering these to be justifications or self-representations, which are much easier to determine. The academic challenges researchers face in such interviews include distorted memory, hindsight bias, path dependency, a tendency to ex-post provide (or even believe in) rationalised ‘accounts’ to explain their decisions, attempts to impress the researcher, and fear of reprisals or prosecution. All of these factors could lead to altered testimonies (for a more thorough discussion of the benefits and challenges of conducting interviews, see, for instance, Dawson, 2019; Horgan, 2012; Khalil, 2019; Morrison, 2021). Further complicating the matter is the simple yet unwelcome fact that people might perform the same action for various different reasons or perform widely different actions for similar reasons. The painfully unsatisfying answer to the question of why people become foreign fighters is that it differs for each person. This does not mean that it is impossible to better understand some larger, commonly found groups of factors or the mechanisms by which these might have affected a particular individual. Concluding, scholars will most likely continue to debate, and disagree, why someone becomes a foreign fighter.

The ‘how’ question: analytical lenses to understand recruitment and mobilisation As the question of why people become foreign fighters remains difficult to answer, some scholars have instead tried to understand how this is taking place. This corresponds to a broader trend in terrorism studies of scholars preferring to discuss (de-)mobilisation or (dis)engagement practices, in other words, behaviour, rather than processes of radicalisation or beliefs, or, in other words, thoughts. The underlying assumption is that many people share similar radical ideas and are confronted with similar structural conditions, but only some of them become foreign fighters. 239

Jeanine de Roy van Zuijdewijn

Processes of recruitment and mobilisation, inextricably tied to social networks, are therefore the sine qua non needed for people to join movements. The importance of social networks in terrorism studies more generally was strongly articulated by Sageman in his landmark work Understanding Terror Networks (Sageman, 2004). He noted that profiling attempts are of little value because of the specificity problem, but he identified a common denominator in the fact that all these individuals somehow ‘made a link to the jihad’ (Sageman, 2004, p. 99). Jihadist groups were mostly composed of young men who found each other through friendship, kinship, and discipleship and, later on, adopted jihadist beliefs and translated these into violent actions. Thus, Sageman considered ideology to be of secondary importance compared to social bonds. The process of recruitment was mostly bottom-up, through existing local networks and ties. Other scholars, such as Hoffmann, have argued that terrorist groups have been mostly involved in top-down recruitment efforts (Hoffman, 2006). What both authors have in common, along with many others, is the emphasis on the necessity of either top-down or bottomup recruitment as having more explanatory value to understand why someone becomes a foreign fighter than focusing on structural (or personal, for that matter) factors only.

The importance of recruitment: historical examples As foreign fighters often travel to distant conflicts, they might not be aware of cultural norms or do not speak the local language, which means that they would probably need some support to make their way into a conflict group. Such groups might introduce methods of vetting to ensure that those who intend to join have some value to the group or, less ambitiously, are not spies trying to infiltrate. The level of organised recruitment varies by conflict and group. There are several examples where organisations indeed played a crucial role in recruitment and mobilisation efforts. For instance, the Communist International (Comintern), controlled by the Soviet Union, helped many prospective fighters reach the battlefield in Spain (Raeburn, 2020, p. 721). Volunteers from various countries often met in Paris, where the Comintern (and other groups) had their offices (Arielli, 2013, p. 227; Petram & Kruizinga, 2020, p. 65). For the Comintern itself, mobilising people to join the fight had significance beyond the conflict in Spain. It would be portrayed as a demonstration of international comradeship and brotherhood, bringing together the working classes from various nations, as strongly desired by the Soviet Union. The recruitment of foreign fighters into the Waffen-SS also seemed to have been organised in a centralised, top-down manner. Heinrich Himmler, the Reich Leader of the SS, aimed to recruit ‘Germanic’ volunteers from countries that were deemed racially similar to Germany, such as Denmark, Norway, the Netherlands, and Belgium. For instance, in June 1940, soon after the occupation of the Netherlands, he issued a document titled ‘Guidelines for the Acquiring of Dutch Volunteers’ that outlined the various ways in which Dutch citizens could be recruited for the Westland regiment (Gingerich, 1997, p. 822). Centralised recruitment efforts in these ‘Germanic lands’ would follow from 1941 onwards. The urgency of such efforts grew when Hitler planned to attack the giant army of the Soviet Union in Operation Barbarossa. High casualty rates among the German armed forces were to be expected, so reinforcements were strongly welcomed. As such, the recruitment of foreign volunteers was deemed significant for the German war efforts, although the actual numbers of foreigners that the Waffen-SS managed to enlist remained low at that particular time. The importance of organised recruitment in explaining the high numbers of foreign fighters was also noticeable in Afghanistan after the Soviet invasion in 1979. There was no significant influx of foreigners during the first years of the war. Abdullah Azzam was disappointed at this lack of 240

Radicalisation of ‘foreign fighters’

engagement by Muslims worldwide and played a crucial role in mobilising fighters. He travelled to various countries to give lectures in order to convince people to join the jihad in Afghanistan. Equally important, he set up the Services Bureau to provide various forms of logistical support to foreign fighters and their families (Hegghammer, 2020, pp. 205, 245; Wright, 2006, p. 119). From 1985 onwards, thousands of fighters joined the conflict, which can at least be partially ascribed to Azzam’s zealous recruitment efforts and the global dissemination of propaganda material (Hegghammer, 2020, p. 244). There are also examples of foreign fighter mobilisations characterised by less organised recruitment. For instance, the recent conflict in Syria and Iraq seemed to have attracted large numbers of fighters who joined the battle independently and without being facilitated. Especially in the early years of the conflict, many foreign fighters departed in small groups to travel to the Syrian border themselves without having any connections to armed groups (Vidino, 2014, p. 221). Several of them turned up unannounced and were sometimes rejected by groups, although not every jihadist movement had a similar approach to scrutinising prospective fighters. For instance, Holman noted that IS initially had less strict procedures than Al-Qaeda-related groups (Holman, 2016, p.  6). The national cohorts of foreign fighters subsequently played an important role in mobilising sympathisers at home to also come and join the conflict, providing them with practical advice on how to do so (Bakker & De Bont, 2016; Coolsaet, 2015; Vidino, 2014). In addition, IS was well-versed in producing slick propaganda material that spread widely via the Internet and showed would-be fighters rosy pictures of life in the Caliphate and tried to convince them to find ways to travel there. The role of the Internet in mobilising and facilitating people’s travel to the battlefields in Syria and Iraq has been widely debated. While there has been ample evidence that the Internet has provided people with more opportunities to access propaganda material, most studies conclude that physical, face-to-face interactions are often indispensable for people to join extremist movements (Behr et al., 2013). There is little support for a scenario in which individuals ‘radicalised’ fully online and then decided to travel to a distant conflict without having prior physical contact with other individuals, although some exceptions do exist (Whittaker, 2018, p. 118).

Dispersed recruitment: what the map can tell us As the conflict in Syria and Iraq attracted thousands of foreign fighters from a long list of countries across the globe, it is easy to forget that recruitment and mobilisation have mostly been clustered at the local level. Especially in the Western European context, where this type of data is available, large regional differences can be observed. Reynolds and Hafez studied the mobilisation in Germany and noted strong clusters: around half of the foreign fighters in their dataset came from the state of North-Rhine Westphalia (Reynolds & Hafez, 2019, p. 682). In the Dutch context, relatively large clusters departed from The Hague, Delft, Gouda, and Arnhem (Ginkel & Entenmann, 2016, p. 36). The Belgian cities of Vilvoorde and Mechelen stood out, while the same applied to Aarhus (Denmark) and Gothenburg (Sweden) (Ginkel & Entenmann, 2016, pp. 25, 29, 48). Similarly, in Austria and Switzerland, facilitation networks for foreign fighters were found in specific places, such as Vienna and Graz (Austria) and Winterthur in Switzerland (Saal & Lippe, 2021, p. 37). Such differences could not be explained by the relative size of the Muslim population or by large differences in, for instance, socio-economic conditions, as these are relatively similar within countries in Western Europe. As such, they are another indication of the importance of local networks and recruitment efforts in explaining foreign fighter mobilisations. 241

Jeanine de Roy van Zuijdewijn

Concluding remarks: where do we stand today? There has been a strong increase in academic interest in the foreign fighter phenomenon after thousands of people travelled to Syria and Iraq to participate in the conflict. Many studies have focused on jihadist fighters, particularly those from Western countries. There is much to learn, however, from studying historical cases or adopting broader approaches that also aim to understand the who, why, and how of the radicalisation of foreign fighters. As with other forms of violent extremism, it has been difficult, if not impossible, to establish concrete profiles of these fighters. Scholars also continue to debate the importance of push and pull factors, for instance, focusing on ideology or socio-economic factors. As these questions have proven difficult to answer, other scholars have instead tackled the how question, looking at recruitment and mobilisation. The ‘who, why and how’ questions have been central to many studies. This means that the ‘with what effects’ question has mostly remained unexplored, particularly beyond the (Western) countries from which fighters departed. Now that the initial wave of interest and strong policydriven need to understand (Western) fighters have subsided, scholars could devote more attention to trying to investigate the effects of foreign fighters on the conflicts they join. Some authors have already tried to study such questions and discuss the tangible effects of foreign fighter involvement in conflicts (Chu & Braithwaite, 2017; Cragin & Stipanovich, 2019; Daymon et al., 2020; Moore & Tumelty, 2008; Rich & Conduit, 2014; De Roy van Zuijdewijn, 2016). In addition, the assumption that returnees to Western countries would pose a significant threat does not seem to have materialised. Rather, there seems to have been a continuation of the relatively small threat posed by returnees in historical cases (De Roy van Zuijdewijn, 2014b). As a result of this Western, threat-focused perspective, other questions have remained underexplored. Hopefully, now that the ‘acute’ phase of the foreign fighter issue from a Western perspective has (at least for now) ended, scholars can dedicate their attention to some of these underexplored questions.

References Arielli, N. (2013). Getting there: Enlistment considerations and the recruitment networks of the international brigades during the Spanish civil war. In N. Arielli & B. Collins (Eds.), Transnational soldiers: Foreign military enlistment in the modern era (pp. 219–232). Palgrave Macmillan. Arielli, N. (2017). From Byron to Bin Laden: A history of foreign war volunteers. Harvard University Press. Azzam, A. (1987). Join the caravan. Azzam Publications. Bakker, E., & Bont, R. De. (2016). Belgian and Dutch Jihadist foreign fighters. Small Wars & Insurgencies, 27(5), 837–857. Bakker, E., & Roy van Zuijdewijn, J. De. (2015). Jihadist foreign fighter phenomenon in Western Europe: A low-probability, high-impact threat (ICCT Research Paper). ICCT. Basra, R., & Neumann, P. R. (2016). Criminal pasts, terrorist futures. Perspectives on Terrorism, 10(6), 25–40. Behr, I. V., Reding, A., Edwards, C., & Gribbon, L. (2013). Radicalisation in the digital era: The use of the Internet in 15 cases of terrorism and extremism. RAND Corporation. Benmelech, E., & Klor, E. F. (2020). What explains the flow of foreign fighters to ISIS? Terrorism and Political Violence, 32(7). Chu, T. S., & Braithwaite, A. (2017). The impact of foreign fighters on civil conflict outcomes. Research and Politics, 4(3). Cook, J., & Vale, G. (2018). From Daesh to “diaspora”: Tracing the women and minors of Islamic state. ICSR. Coolsaet, R. (2015). What drives Europeans to Syria, and to IS? Insights from the Belgian case. (Egmont Paper Issue 75). Egmont Institute. Cragin, R. K., & Stipanovich, S. (2019). Metastases: Exploring the impact of foreign fighters in conflicts abroad. Journal of Strategic Studies, 42(3–4), 395–424. Dawson, L. L. (2019). Taking terrorist accounts of their motivations seriously: An exploration of the hermeneutics of suspicion. Perspectives on Terrorism, 13(5), 74–89. www.jstor.org/stable/26798579?seq=1&cid=pdf

242

Radicalisation of ‘foreign fighters’ Dawson, L. L. (2021). A comparative analysis of the data on Western foreign fighters in Syria and Iraq: Who went and why? International Centre for Counter-Terrorism. Dawson, L. L., & Amarasingam, A. (2017). Talking to foreign fighters: Insights into the motivations for Hijrah to Syria and Iraq. Studies in Conflict and Terrorism, 40(3), 191–210. Daymon, C., Roy van Zuijdewijn, J. De., & Malet, D. (2020). Career foreign fighters: Expertise transmission across insurgencies. RESOLVE Network. Gingerich, M. P. (1997). Waffen SS recruitment in the “Germanic lands,” 1940–1941. The Historian, 59(4), 814–830. Ginkel, B. V., & Entenmann, E. (2016). The foreign fighters phenomenon in the European context: Profiles, threats and policies. Terrorism and Counter-Terrorism Studies, 7. Gustafsson, L., & Ranstorp, M. (2017). Swedish foreign fighters in Syria and Iraq: An analysis of open-source intelligence and statistical data. Swedish Defence University. Gutmann, M. (2013). Debunking the myth of the volunteers: Transnational volunteering in the Nazi WaffenSS officer corps during the second world war. Contemporary European History, 22(4), 585–607. Hegghammer, T. (2011). The rise of Muslim foreign fighters: Islam and the globalization of Jihad. International Security, 35(3), 53–94. Hegghammer, T. (2016). Revisiting the poverty-terrorism link in European Jihadism. Society for Terrorism Research Annual Conference, Leiden. Hegghammer, T. (2020). The caravan: Abdullah Azzam and the rise of global Jihad. Cambridge University Press. Hoffman, B. (2006). Inside terrorism. Columbia University Press. Holman, T. (2016). “Gonna get myself connected”: The role of facilitation in foreign fighter mobilizations. Perspectives on Terrorism, 10(2), 2–23. Horgan, J. (2012). Interviewing the terrorists: Reflections on fieldwork and implications for psychological research. Behavioral Sciences of Terrorism and Political Aggression, 4(3), 195–211. Keene, J. (2001). Fighting for Franco: International volunteers in nationalist Spain during the Spanish civil war. Continuum. Khalil, J. (2019). A guide to interviewing terrorists and violent extremists. Studies in Conflict and Terrorism, 42(4), 429–443. Ljujic, V., van Prooijen, J. W., & Weerman, F. (2017). Beyond the crime-terror nexus: Socio-economic status, violent crimes and terrorism. Journal of Criminological Research, Policy and Practice, 3(3), 158–172. Malet, D. (2013). Foreign fighters: Transnational identity in civil conflicts. Oxford University Press. Marcus, R. D. (2021). ISIS and the crime-terror Nexus in America. Studies in Conflict & Terrorism, 1–26. Mareš, M. (2017). Foreign fighters in Ukraine: Risk analysis from the point of view of NATO. In Not only Syria: The phenomenon of foreign fighters in a comparative perspective (pp. 31–39). IOS Press. Marone, F., & Vidino, L. (2019). Destination Jihad: Italy’s foreign fighters. International Centre for Counter-Terrorism. Mendelsohn, B. (2011). Foreign fighters-recent trends. Orbis, 55(2), 189–202. Moore, C., & Tumelty, P. (2008). Foreign fighters and the case of Chechnya: A critical assessment. Studies in Conflict and Terrorism, 3. Morrison, J. F. (2021). Analyzing interviews with terrorists. RESOLVE Network. O’Connor, S., & Piketty, G. (2020). Introduction – foreign fighters and multinational armies: From civil conflicts to coalition wars, 1848–2015. European Review of History: Revue Européenne d’histoire, 27(1–2). Petram, L., & Kruizinga, S. (2020). De oorlog tegemoet: Nederlanders en de strijd om Spanje, 1936–1939. Atlas Contact. Pokalova, E. (2019). Driving factors behind foreign fighters in Syria and Iraq. Studies in Conflict and Terrorism, 42(9), 798–818. Rácz, A. (2017). The elephant in the room: Russian foreign fighters in Ukraine. In Not only Syria: The phenomenon of foreign fighters in a comparative perspective (pp. 60–73). IOS Press. Raeburn, F. (2020). Politics, networks and community: Recruitment for the international brigades reassessed. Journal of Contemporary History, 55(4), 719–744. Reiff, J. C. (2020). When Ali comes marching home: Shi’a foreign fighters after Syria. Studies in Conflict and Terrorism, 43(11), 989–1010. Rekawek, K. (2020). Career break or a new career? Extremist foreign fighters in Ukraine. Counter Extremism Project. Reynolds, S. C., & Hafez, M. M. (2019). Social network analysis of German foreign fighters in Syria and Iraq. Terrorism and Political Violence, 31(4), 661–686.

243

Jeanine de Roy van Zuijdewijn Rich, B., & Conduit, D. (2014). The impact of Jihadist foreign fighters on indigenous secular-nationalist causes: Contrasting Chechnya and Syria. Studies in Conflict & Terrorism, 38(2), 113–131. Roy van Zuijdewijn, J. De. (2014a). Fearing the Western Muslim foreign fighter: The connection between fighting the defensive Jihad and terrorist activity in the West. Utrecht University. Roy van Zuijdewijn, J. De. (2014b). The foreign fighters’ threat: What history can(not) tell us. Perspectives on Terrorism, 8(5), 59–73. Roy van Zuijdewijn, J. De. (2016). Terrorism and beyond: Exploring the fallout of the European foreign fighter phenomenon in Syria and Iraq. Perspectives on Terrorism, 10(6), 82–96. Saal, J., & Lippe, F. (2021). The network of the November Vienna attacker and the Jihadi threat. CTC Sentinel, 14(2), 33–43. www.ctc.usma.edu/sentinel/ Sageman, M. (2004). Understanding terror networks. University of Pennsylvania Press. Smith, P. S., Poulsen, N. B., & Christensen, C. B. (1999). The Danish volunteers in the Waffen SS and German warfare at the Eastern front. Contemporary European History, 8(1), 73–96. Smyth, P. (2015). The Shiite Jihad in Syria and its regional effects (Issue 11). The Washington Institute for Near East Policy. Tuck, H., Silverman, T., & Smalley, C. (2016). “Shooting in the right direction”: Anti-ISIS foreign fighters in Syria & Iraq horizons series no. 1. Institute for Strategic Dialogue. UN. (1989). International convention against the recruitment, use, financing and training of mercenaries. https://treaties.un.org/doc/Treaties/1989/12/1989120408-54AM/Ch_XVIII_6p.pdf UN News. (2021). ISIL crimes against Yazidis constitute genocide, UN investigation team finds. UN News. https://news.un.org/en/story/2021/05/1091662 United Nations. (2017). United Nations security council – 8116th meeting. United Nations. United Nations Security Council. (2014). Resolution 2178 (2014). www.undocs.org/S/RES/2178(2014) Van Ostaeyen, P. (2017). Belgian foreign fighters in Syria and Iraq – May 2016. In K. Rekawek (Ed.), Not only Syria: The phenomenon of foreign fighters in a comparative perspective (pp. 40–51). IOS Press. Vidino, L. (2014). European foreign fighters in Syria: Dynamics and responses. European View, 13, 217–224. Weenink, A. W. (2019). Adversity, criminality, and mental health problems in Jihadis in Dutch police files. Perspectives on Terrorism, 13(5). Weijenberg, G., & Roy van Zuijdewijn, J. De. (2021). The forgotten front: Dutch fighters in Ukraine. In ICCTcommentary. https://icct.nl/publication/the-forgotten-front-dutch-fighters-in-ukraine/ Whittaker, J. (2018). Countering terrorist activities in cyberspace. In Z. Minchev & M. Bogdanoski (Eds.), Countering terrorist activities in cyberspace. IOS Press. Wright, L. (2006). The looming tower: Al-Qaeda’s road to 9/11. Penguin Books.

244

PART III

Countering radicalisation Key debates

16 COUNTERING VIOLENCE OR IDEAS? THE POLITICS OF COUNTER-RADICALISATION1 Kodili Chukwuma and Lee Jarvis

Introduction This chapter focuses on a question that is central to any evaluation of counter-radicalisation programmes or initiatives: what is their purpose or target? To explore this question, we distinguish between narrower approaches to counter-radicalisation, which focus on countering violence, on the one hand. And, on the other, broader approaches which (also) attempt to counter violent, radical, or extremist beliefs. The differences between these approaches, we argue, allow us to think more carefully about (i) the conceptual relationship between ideas and behaviour; (ii) the effectiveness of counter-radicalisation policy; and (iii) the legitimacy of specific efforts to counter radicalisation. The chapter begins by sketching the diversity of ambitions within counter-radicalisation programmes, linking this to different understandings of the meaning and process of radicalisation itself. From here, we elaborate on the distinction between narrow and broad approaches as a heuristic for comparing and contrasting different counter-radicalisation initiatives. The chapter then illustrates the importance of this diversity by drawing on relevant ‘real world’ examples before turning to the conceptual, practical, and political significance it poses. The chapter concludes by arguing that the framework of narrow and broad approaches enables researchers, students, and those in the policy world to reflect further on key strategic, political, and normative questions around the purposes, effectiveness, and problems of counter-radicalisation programmes.

Narrow and broad approaches to counter-radicalisation Contemporary approaches to counter-radicalisation are surprisingly heterogeneous in their understanding of the meaning, drivers, and consequences of radicalisation. They are also, at times, ambiguous and fluid, shifting in focus either in specific formulations (for instance, in one policy document) or over time (for instance, where a particular counter-radicalisation policy goes through multiple revisions). These variations may be linked to the newness of this policy arena. As other chapters document (Coolsaet, this volume), radicalisation as a problem is a relatively recent invention and an outcome, in part, of a series of attacks against Western states in the early 21st century (Kundnani, 2009; Sedgwick, 2010, p. 480). Variations in counter-radicalisation policy also, though, derive from the complex institutional and political contexts in which ‘radicalisation’ as 247

DOI: 10.4324/9781003035848-19

Kodili Chukwuma and Lee Jarvis

a problem is produced, with higher education institutions, think tanks, government departments, and defence establishments all important contributors to the production or manufacturing of this phenomenon via the knowledge they create (see also Herman & O’Sullivan, 1989; Stampnitzky, 2013). Different norms, expectations, and interests across these institutions therefore introduce complexity here, although a shared broad concern for attempting to understand radicalisation in order to counter it (Heath-Kelly et al., 2015, p. 3) is reflective of a wider problem-solving orientation within counter-terrorism politics and scholarship (Gunning, 2007; Jarvis, 2009) and key to the analysis we develop later. To think about the implications and importance of these heterogeneities, we use the term ‘narrower approaches’ to counter-radicalisation in this chapter to describe approaches with a very specific focus on preventing violence (compare with Berger, 2016a). Such approaches often take an interest in the psychological, theological, ideological, and social dynamics within which violence emerges, but primarily in relation to the causal role of such dynamics in leading people to act violently. Broader approaches, in contrast, have a much wider purview within our framework, which expands beyond preventing violence to the countering of ‘radical’ or ‘extreme’ ideas and views. Radical ideas, in these approaches, matter in their own right, not (only) because of their causal importance in driving violent behaviours. That narrow and broad approaches to counter-radicalisation can exist simultaneously within this policy domain is a product, in large part, of conceptual disagreement over the meaning of the term ‘radicalisation’: a concept that can focus on people’s thoughts or actions. Cassam (2018, p. 195), for instance, drawing on Marc Sageman’s work, argues that ‘radicalisation can be understood in at least two different ways. One type of radicalisation is cognitive and involves the formation or acquisitions of extremist beliefs. Another type is behavioural radicalisation, which involves a turn to violence’. Although Cassam (in Sardoč, 2020, p. 167) questions the neatness of this distinction, it is a common one and therefore offers a helpful heuristic for capturing different uses of the same term. Clubb and McDaid (2019), for example, similarly distinguish between ‘the cognitive dimensions of ideology (i.e. holding a radical ideology) and behavioural radicalisation (i.e. engaging in terrorism)’, in their critical realist re-conceptualisation of this term. Keiran Hardy (2018, pp. 79–80), discussing Peter Neumann’s (2013) work, argues in related terms: An important distinction is often drawn between ‘cognitive’ radicalisation, which focuses on extremist beliefs, and ‘behavioural’ radicalisation, which focuses on extremist behaviour. . . . This gets at the question: has somebody radicalised once they adopt and internalise extremist views, or only if they engage in some criminal conduct (such as training for terrorism) as a result? From a public policy perspective, extremist behaviour would appear to be the primary concern, but the development of extremist views is also clearly relevant. Working with this broad distinction, we can begin to nuance our understanding of narrow and broad approaches. Narrow approaches, focused on countering violence, may draw upon cognitive conceptions of violence as something that emerges from the psyche of individuals, for instance due to pathologised decision-making processes (Monaghan & Molnar, 2016). Here, counter-radicalisation suggests an aspiration to detect troubling symptoms in the psychological realm of vulnerable or at-risk individuals in order to prevent future harm to others (Laqueur, 2004). Sewell and Hulusi (2016), for example, point out that ‘individuals become radicalised to provide a surety which reduces the anxiety felt because of personal uncertainty’. Thus, intervention strategies for them might productively focus on developing sensitivity to one’s anxiety and vulnerability, building cognitive and meta-awareness skills, and developing functional responses to uncertainty. 248

The politics of counter-radicalisation

Other approaches look beyond individual psychological dispositions to include ideological or theological beliefs and the social networks through which these are sustained as drivers of violence. In these accounts (e.g. Sageman, 2004, 2008), ideological or theological beliefs become more extreme in response to relevant ‘cognitive openings’, such as an individual’s identity crisis or dynamics of group interaction, which make terrorism increasingly possible (e.g. Wiktorowicz, 2005). Walter Laqueur (2004), for instance, observed that radicalisation has been facilitated by the growth of Muslim communities, the multiculturalism that resulted (especially in European societies), and the relative freedom of individuals to organise and prepare for violent acts in mosques, madrassas, and other cultural spaces. This social-network model of radicalisation is perhaps most strongly associated with the work of Marc Sageman, whose ‘bunch of guys’ thesis shifts analytical focus away from potentially relevant economic, political, and religious factors as direct causes of terrorism (Sageman, 2004, 2008). Drawing on a sample of 500 individuals linked to the 9/11 attackers, Sageman instead argues that ‘the most striking feature of the jihadist profile is that joining the global Islamist terrorism social movement was based to a great degree on friendship and kinship’ (Sageman, 2004). Sageman therefore suggests that policymakers should understand that the war against the terrorism of organisations such as Al-Qaeda fundamentally involves a battle for the hearts and minds of Muslim communities. As instances of relatively narrow approaches to counter-radicalisation, these examples share broad assumptions about the linearity of radicalisation processes through which certain individuals adopt radical ideas and commit violent acts. Overall, the aim of counter-radicalisation, here, is to prevent future violence, including through early detection of potentially pertinent warning signs. Van de Weert (2022), for instance, notes that early detection or prevention reflects a paradigm shift towards the risk management of potentially dangerous people and groups and away from the prosecution of crimes after the fact. Broader approaches towards counter-radicalisation have, as the name implies, a much wider set of aspirations aimed at countering radical ideas or beliefs in their own right. Identifying what counts as a radical belief, however, is far from straightforward, and depends, in part, on the wider contexts of those decisions, and the interests informing them (Sedgwick, 2010). Marret et al. (2013), for example, distinguish ‘non-violent radicalism’ as the holding of radical views, on the one hand, from ‘violent radicalisation’, which relates to an individual’s acting upon those radical views. Bartlett and Birdwell (2010, pp. 8–9) offer a similar distinction in their claim that although ‘all terrorists are radicals, many radicals never make the leap into violence’. John Horgan (2009, p. 152), for whom ‘current understandings of de-radicalisation fail to distinguish between cognitive and behavioural dimensions’, argues it is useful to distinguish de-radicalisation – referring to the re-orientation of worldviews and cognitive shifts of (potential) terrorists – from dis-engagement, which suggests the renunciation of violence without necessarily giving up ones’ (extremist) beliefs (Horgan, 2009, p. 17). Contra Bartlett and Birdwell (2010, pp. 8–9), this distinction here is important because, ‘the evidence from social science continues to assert that, just as not all radicals will seek out involvement in terrorism, not all terrorists hold radical beliefs’ (Horgan & Altier, 2012, p. 88). Distinguishing radical views from behaviour in broader approaches has importance because it helps to differentiate (i) the causal significance of ideas as drivers of violence from (ii) the specific threat that ideas are seen to pose for various entities or referents, which may include social cohesion, multiculturalism, and liberalism. It is this latter threat, within broader approaches, that justifies state intervention to prevent the diffusion and spread of radical ideas as a crucial counterradicalisation objective in its own right. Vidino and Brandon (2012), for instance, discuss general 249

Kodili Chukwuma and Lee Jarvis

preventive strategies aimed at challenging extremist views and influences and promoting tolerant, moderate, and democratic principles in this context. For Peter Neumann (2013, p. 445), similarly: government should play a more energetic role by encouraging civic challenges to extremist narratives and by promoting awareness and education of young people . . . and spreading awareness about online radicalisation among parents, teachers, and community leaders, so they are able to detect, report, and – if necessary – intervene in processes of online radicalisation.

Counter-radicalisation in practice This second section of our chapter explores the distinction between narrow and broad approaches through reference to prominent examples of counter-radicalisation initiatives. Reflecting on similarities and differences between these initiatives, we argue, demonstrates considerable diversity of ambition and enactment in this context, offering empirical illustration of the distinctions considered above and their importance. Let us begin with the widely known Prevent Strategy, the counter-radicalisation plank of the UK’s counter-terrorism strategy, CONTEST. Unlike many other examples in this policy arena, counter-radicalisation is explicitly defined within the strategy itself; its glossary of terms describes it as follows: ‘Counter-radicalisation usually refers to activity aimed at a group of people intended to dissuade them from engaging in terrorism-related activity’ (UK Home Office, 2011, p. 107). As a strategy for achieving this, the most recent incarnation of Prevent, published in 2011, outlines three overarching objectives:2 • Respond to the ideological challenge of terrorism and the threat we face from those who promote it; • Prevent people from being drawn into terrorism and ensure that they are given appropriate advice and support; and • Work with sectors and institutions where there are risks of radicalisation which we need to address (UK Home Office, 2011, p. 7). As the above indicates, the UK Prevent strategy looks, initially, like a relatively narrow approach to counter-radicalisation with its emphasis on addressing the threat of terrorism, a threat that is typically understood, of course, as an act of violence (see Dexter, 2012). If we unpack the strategy a little, however, it soon appears much broader in scope than we might first think, in large part because of the emphasis on ideas throughout. To begin with, terrorism is posed in the Prevent Strategy as an ‘ideological challenge’ (rather than, say, a behavioural challenge or a material threat) that requires dissuasive rather than coercive countering. Radicalisation, too, is understood in ideational terms as ‘the process by which a person comes to support terrorism and forms of extremism leading to terrorism’ (UK Home Office, 2011, p. 108, our emphasis). This emphasis on the ideational nature of terrorism and radicalisation already indicates a more expansive counter-radicalisation approach, yet Prevent goes further here in two important ways. The first is by concerning itself with ‘terrorism-related activity’ (our emphasis). The expansiveness of this compound adjective becomes clear when we note that terrorism-related offences in UK law include, among other things, encouragement of terrorism, dissemination of terrorist publications, and collecting information likely to be useful to a person committing or preparing acts of terrorism (see Macdonald, 2015). Second, and more importantly, 250

The politics of counter-radicalisation

Prevent, as Hardy (2018, p. 95) notes, also incorporates non-violent ideas within its conception of ‘extremism’, interpreted as ‘vocal or active opposition to fundamental British values, including democracy, the rule of law, individual liberty and mutual respect and tolerance for different faiths and beliefs’ (UK Home Office, 2011, p. 107). Notwithstanding the contestability of this characterisation of British values (see Lockley-Scott, 2019; Jarvis et al., 2020; Winter et al., 2021; Marsden et al., 2022), we see here a clearly expansive approach to counter-radicalisation in which ‘the problematic end-state to be deterred [is] neither active terrorist involvement nor ideological radicalism, but ideological alignment with terrorism’ (Edwards, 2015, p. 57, our emphasis). A similar, but more explicitly broad, approach to counter-radicalisation is that of the Danish government. As set out in the National Action Plan for Preventing and Countering Extremism and Radicalisation, radicalisation refers to ‘a short- or long-term process where persons subscribe to extremist views or legitimise their actions on the basis of extremist ideologies’ (Danish Ministry of Immigration, Integration and Housing, 2016, p. 7). With extremism here understood as the legitimisation of ‘violence or other illegal acts’ (Danish Ministry of Immigration, Integration and Housing, 2016, p. 7), the Danish approach presents its ambitions in the following way: the aim of preventing extremism and radicalisation is not merely to ward off terrorism – it also has wider implications for society. We need to stem the negative influence of extremist groups in local communities and on social media, in order to protect our general welfare and safety. The independence, critical thinking and democratic skills of our children and young people must be nourished, as it helps to increase their resilience to radicalisation. It also constitutes an intrinsic goal in a dynamic, democratic society to secure the development and active citizenship of our children and young people. (Danish Ministry of Immigration, Integration and Housing, 2016, p. 6) As with the UK Prevent Strategy, counter-radicalisation receives a very broad remit here, involving countering undesirable ideas as part of a wider democratic safeguarding effort. As Lindekilde (2016, p. 250) summarised the Danish government’s original, 2009, definition of radicalisation: ‘Being radical is here the opposite of being democratic, peaceful, pro-equality, tolerant, and respectful of law and order. In fact one sometimes gets the impression that radicalism is used to denote everything liberal democrats dislike’. An even more overtly expansive example of counter-radicalisation can be found within the European Union’s recent Counter-Terrorism Agenda. This Agenda incorporates a very explicit focus on the defence of (ostensibly) shared liberal values, arguing that these are vital to the very identity of the Union: The European Union is founded on a strong set of values. Our education, health and welfare systems are inclusive by nature but they come part and parcel with acceptance of the values that underpin them. Our European way of life – emblematic of inclusive and tolerant societies – is not optional and we must do all in our power to prevent those that seek to undermine it – from within or without. (European Commission, 2020, p. 6) This broad emphasis on liberal values supports a wide range of counter-radicalisation activities targeting ideas and violence, including countering online terrorist content and hate speech; bottom-up support for community resilience initiatives; and targeted interventions for individuals in prisons and detention centres (European Commission, 2020, pp. 6–10). 251

Kodili Chukwuma and Lee Jarvis

It is important to note that counter-radicalisation programmes are not the preserve of Western states or intergovernmental organisations, with prominent examples of related initiatives also evident within Asia, Africa, and other regions of the world (see Breidlid, 2021; Spalek & Lambert, 2008). Nigeria’s approach in its National Counter-terrorism Strategy offers a useful contrast to the British and Danish examples, with its description of radicalisation as ‘the main factor behind faith-based terrorism in Nigeria’ (NACTEST, 2016, p. 28). This emphasis on religion as a causal factor for violence leads here to a comparatively narrow, focused approach to counter-radicalisation: To address this trend with a view to its reversal, the Counter-terrorism Centre, within the Office of the National Security Advisor, will develop measures towards counter radicalization and de-radicalization of extreme ideologies. The Government will carry out effective monitoring of the radicalization processes and its indicators, systematic identification of the highly vulnerable groups for assimilation into the society, and launching of anti-radicalization campaigns. (NACTEST, 2016, p. 29) The purpose of counter-radicalisation, in the Nigerian context, is chiefly to prevent violence through the targeting of ‘extreme ideologies’ that may lead to violence or support for terrorism. This focus is evident, too, in Nigeria’s Policy Framework and National Action Plan for Preventing and Countering Violent Extremism, which states: ‘there is urgent need to sustain the momentum in preventing and countering not only the insurgency of Boko Haram but all other manifestations of violent extremism in Nigeria’ (PCVE, 2017, p. 3). Here, we see how a relatively narrow focus on violence can be sustained even when the specific actors to be addressed are expanded. A similarly narrow focus is evident, finally, in the United Nations’ recent Plan of Action on Preventing Violent Extremism, which is positioned as an example of a ‘re-energized’ prevention agenda within the UN’s wider approach to violence (UN General Assembly, 2015). Here, ‘violent extremism’ is restricted to activities ‘conducive to terrorism’, although the specific meaning of terrorism is devolved to member states and their obligations under international law. Countering violent extremism, in this approach, involves two strands of activity, each of which focuses on the prevention of violence: ‘security-based counter-terrorism measures’, and, ‘systematic preventive measures which directly address the drivers of violent extremism that have given rise to the emergence of these new and more virulent groups’ (UN General Assembly, 2015). This narrow focus on preventing violence and its protagonists is, though, interestingly underpinned by a seemingly progressive understanding of violence as the outcome of structural inequalities and frustrations: the creation of open, equitable, inclusive and pluralist societies, based on the full respect of human rights and with economic opportunities for all, represents the most tangible and meaningful alternative to violent extremism and the most promising strategy for rendering it unattractive. (UN General Assembly, 2015) The examples from Nigeria and the United Nations, we suggest, are indicative of what we term a narrow approach that is focused, ultimately, on the prevention of violence. As such, they differ from the earlier British, Danish, and EU examples which are illustrative of a much broader approach that involves targeting ideas, beliefs, or views seen to be anathema to those societies because those ideas, beliefs, or views are anathema. Although there may at times be a convergence 252

The politics of counter-radicalisation

of means across these approaches, the difference in their ends opens important conceptual, strategic, political, and moral questions to which we now turn.

Targeting ideas or violence? In this final section of our chapter, we now explore the implications of narrow and broad approaches to counter-radicalisation in further detail. To do so, we focus on three core sets of questions. First are conceptual questions on the relationship between ideas and behaviour, including how radical ideas and behaviour are identified and how they are linked in counterradicalisation approaches. We then turn to the thorny question of effectiveness in this context, asking how narrow and broad approaches to counter-radicalisation are (and, perhaps, should be) evaluated. And, third, we look at important political and moral questions about the legitimacy of different approaches to counter-radicalisation, with particular emphasis on the targeting of ideas that do not generate violence.

Ideas, behaviour, and their relationship To begin answering the first of our questions, it is important to reiterate that narrow and broader approaches often share important assumptions. The first of these, as we have seen, is that ideas matter. In narrower approaches, ideas have instrumental significance because they are capable of driving individuals to behave in particular ways, such as committing violence. In broader formulations, ideas matter because they threaten core liberal values and virtues, including moderation, tolerance, and respect for difference. As the above suggests, a key feature of narrower formulations is their frequent positing of a causal relationship between (certain) radical ideas and violence, which is, at best, heavily contested, both in general terms and in relation to terrorism more specifically. For Schuurman and Taylor (2018), for instance, the move towards violence should be understood in relation to the strength rather than the content of ideas. David Stevens (2009), however, goes further still here, expressing scepticism towards the very notion that ideas drive (radical) behaviour. Drawing on scholarship around conversion to minority religious groups, Stevens (2009, pp. 519–520) argued in his early critique of the UK’s Prevent Strategy: Research from sufficiently similar areas over the past two or more decades shows that the significance of the actual religious message as part of the decision of an individual to join a radical religious group is, at best, slight. . . . A major concern with the PVE [Preventing Violent Extremism] agenda, therefore, is that it concentrates a significant amount of its efforts on the wrong driver of extremism. Those who commit themselves to participation in radical communities, Stevens (2011, p. 170) argues, actually tend to do so following far more utilitarian calculations around the costs and benefits of membership in a community, which may include ‘friendship, social interaction, a sense of belonging or solidarity, self-worth or being valued by others’. Interests rather than ideals, in other words, may be a more pertinent explanation of the drivers of violence (Stevens, 2011), although individuals may learn to make sense of, or describe, their motivations in religious or ideological terms retroactively. Crone (2014) offers a related but distinct criticism in this regard, arguing not against the directionality of the ideas/violence relationship implicit in counter-radicalisation initiatives but rather 253

Kodili Chukwuma and Lee Jarvis

the ontological separability of these two phenomena. For Crone, the very idea of a relationship between violence and ideas misses the point that these two phenomena are mutually entangled in that they emerge co-constitutively through ‘techniques of the body’, in Marcel Mauss’s (1934) terminology. Radicalisation, in turn, is something that is embodied in individual micro-practices: practical, embodied, and everyday ways of behaving that might be learned or imitated from others. In Crone’s (2014, p. 299) helpful summary: the process of embracing violent militancy is not necessarily an intellectual process, where young aspiring jihadis become radicalized by delving into highbrow discussions about the concept of jihad. Rather, they pick up specific ways of speaking, walking, dressing etc. that provide them with a specific appearance which materialize their being part of a specific militant subculture. Crone (2014) argues here that radicalisation might therefore better be theorised as a process of bodily transformation in which religious beliefs and violence are co-constitutive and always entangled. Thus, ‘If religious worldviews are not the driving force behind religious violence, then “counter-narratives” that set out to change the mindset of violent people by inculcating democratic values will probably not be very successful’ (Crone, 2014, p. 307). A third line of criticism argues that the targeting of ideas as drivers of violence risks neglecting the wider ideological contexts from which specific ideas emerge. Juergensmeyer and Sheikh (2013), for instance, argue that understanding the religion-violence nexus involves examining the wider ‘epistemic worldviews’ of violent actors and reframing our focus from the causal question ‘why do people commit violence’, to the constitutive question: how is the world viewed ‘in such a way that would allow these actions to be carried out’ (Juergensmeyer & Sheikh, 2013, p. 627). In this understanding, we seem to be moving from a ‘cognitive’ conception of ideas as variables that impact the world, its inhabitants, and their behaviour (Bieler, 2001, pp. 94–95) to a more intersubjective approach to ideas as collective or shared and constitutive of the world, its inhabitants, and their behaviour (Bieler, 2001, pp. 95–97). This shift, though, is not a purely analytical one. It creates an important policy shift in favour of refocusing the efforts of de-radicalisation actors towards large-scale, whole-of-society initiatives seeking to inoculate communities against extreme ideas and/or behaviours, which may not necessarily lead to terrorism or violence at all (Juergensmeyer & Sheikh, 2013). Recognition of the wider ideational contexts in which violent ideas and behaviour emerge takes us towards the territory of broader counter-radicalisation approaches aimed at tackling radical ideas in their own right. The most important issue here is who gets to determine which ideas are radical (and therefore require interventionist responses), given that many radical ideas – for example in pursuit of civil rights – neither pose a threat to liberal values nor lead to violent action (Berger, 2018, p. 37). Mark Sedgwick (2010, p. 481) helpfully distinguishes absolute from relative understandings of radicalisation. Where the former attempt to pin down the term’s meaning in specific ways as appropriate to particular circumstances and policy domains, relative understandings tend to position ‘radical’ on a continuum with ‘non-radical’ or ‘moderate’. As Sedgwick (2010) argues, relative understandings therefore require decisions on (i) where the line between radical and moderate lies and (ii) what line – what policy area – we are considering in designating something ‘radical’. The latter is particularly important because there may be important differences between areas of public policy; for instance, ‘a group or individual that is a threat in security terms may not be a problem in integration terms’ (Sedgwick, 2010, p. 489). Such decisions, in turn,

254

The politics of counter-radicalisation

return us to the question of counter-radicalisation’s purposes because designations and priorities in one policy area may not translate successfully to others: The conflicting agendas of the public and political discourse, of security and integration, and of foreign policy, then, are among the causes of the disagreements . . . in official definitions over whether it is thought or action that constitutes a threat, and whether non-violent radicalism is or is not a threat. The thought involved in non-violent radicalism may well be a threat to integration, but it is especially action that supports violence that is a security threat. (Sedgwick, 2010, p. 489) Sedgewick’s discussion is so significant because it highlights the contingent and therefore political foundations of broader counter-radicalisation programmes that rest upon contestable, perhaps even arbitrary, differentiations between types of ideas (see Heath-Kelly et al., 2015, p. 5).

Questions of effectiveness A second important, but no less controversial, question concerns the effectiveness of counterradicalisation, and how this might vary between narrow and broad approaches. To begin, let us note the continuing scarcity of meaningful evaluative efforts in this context (Gielen, 2019, p. 1149; Fisher  & Busher, this volume). One survey of 73 academic studies purporting to answer this question, for instance, found only 14 that were based on empirical evidence, with the remainder offering theory-driven recommendations (Gielen, 2019, pp. 1153–1155). Horgan and Braddock’s (2010, pp. 285–286) analysis of de-radicalisation initiatives in five different countries – Northern Ireland, Saudi Arabia, Yemen, Indonesia, and Colombia – similarly argued: Despite the highly publicized claims for success associated with some of these risk reduction initiatives, there are major barriers to even the most tentative of evaluations: 1. There are no explicit criteria for success associated with any initiative. 2. There is little data associated with any of these initiatives that can be reliably corroborated independently. 3. There has been no systematic effort to study any aspect of these programs, even individually, let alone collectively. For Alex Schmid (2013, pp. 49–50) likewise: knowledge about de-radicalisation programmes is at this stage still very fragmented and uneven, which makes the comparative evaluation of various local initiatives an almost impossible task. External, in-depth evaluations of individual country programmes or detailed comparative evaluations of a rigorous sort do not yet seem to exist in the available literature (and it is very doubtful that they exist at all) . . . too much of what we think we know is still built on sand and lacks solid foundations for policy recommendations. Some scholars (e.g. Berger, 2016b) identify advantages for narrower counter-radicalisation approaches because the prevention of violence is both a more achievable and more readily quantifiable goal than the changing of beliefs. In this scenario, counter-radicalisation success can

255

Kodili Chukwuma and Lee Jarvis

be quantified through a series of metrics, including disrupting terrorist recruitment, interdicting attacks, and producing counter-narratives to extremist propaganda. One specific challenge here, however, is discerning the impact such measures have upon dis-engagement from political violence, given that people cease participation in violent activities for various reasons, including burnout and disillusionment (Cherney, 2020). As with other policy contexts, the absence of counterfactual conditions means it is difficult, if not impossible, to know what would have happened to would-be violent protagonists in the absence of particular counter-radicalisation initiatives (Jarvis, 2014). Evaluating the effectiveness of broader efforts to change radical ideas or worldviews is, if anything, still more challenging because cognitive changes are difficult to measure accurately or objectively (El-Said, 2015; Horgan & Braddock, 2010; Schmid, 2013). The lack of consensus noted earlier around terms such as ‘radicalisation’, ‘radical’, or ‘extreme’ ideas introduces new challenges for identifying successful ‘de-radicalisation’, especially once we recognise that ‘radical’ is a relative rather than an absolute term: a matter of degree rather than a matter of kind. On top of this, as Horgan and Braddock (2010) note that challenges of insincerity and recidivism make it extremely difficult to know in practice if an offender has sincerely de-radicalised. In addition, there is a risk of governmental interventions exacerbating rather than diminishing the attractiveness of illicit or extreme ideas by pushing individuals towards them. As Stevens (2011, p. 180) put it, ‘By directly involving itself in religious affairs, [the UK government] . . . is actively fuelling the fires of religious extremism’. Thus, although a more thorough countering of ideas via de-radicalisation programmes is often deemed politically necessary to satisfy policymakers and funders (Cherney, 2020), broad ambitions around the promotion of ‘positive values’ may set such initiatives up for failure given the challenges in determining their success.

Questions of legitimacy We finish our discussion by turning, finally, to the legitimacy of counter-radicalisation programmes in this context, beginning with the targeting of radical or extreme ideas that is typical of broader approaches. Targeting ideas because they are deemed harmful raises vital political and moral questions around who has authority to identify radical ideas in specific contexts and on whose behalf such decisions are made. Juxtapositions between radical and moderate, as we have seen, can never be wholly objective; they will reflect the assumptions, suspicions, and biases of their authors, whether these are deliberate, recognised, or otherwise. In the first instance, the suspicion of ideas deemed insufficiently progressive, liberal, or tolerant can lead to an inherently conservative defence of liberal democracy and its values as understood at a particular point in time. This, therefore, risks making important and far-reaching moral decisions – for instance, around what can or should be censored in order to protect the public – on contemporary social, political, or religious mores, the injustice of which may only become apparent with hindsight. Consider how many progressive historical movements would have been – indeed, often were – designated radical or extremist in their own time: from socialist and feminist movements to social movements around race, LGBTQ+ rights, environmentalism, and beyond (see Lindekilde, 2016, p. 249). The targeting of ideas, in other words, risks reproducing the traditional liberal paradox of extending tolerance only to the tolerant, when we might instead argue that democracy relies (or should rely) upon pluralism, contestation, and indeed conflict (see Mouffe, 2011). As the UK’s Commission for Countering Extremism (2019, p. 5) argued in a report on ‘hateful extremism’, ‘protecting democratic debate and freedom of expression is vital. This includes defending speech and actions which can be offensive, shocking, dissenting and critical; or advocates for conservative religious beliefs’. 256

The politics of counter-radicalisation

The policing of radical ideas has implications for people living within democratic societies as much as for the organisation of those societies. For critics, such efforts may contribute to longstanding neoliberal efforts to produce passive and compliant citizens through controlling and encouraging particular ideas, values, and beliefs in the service of social order (Foucault, 2003). In Kühle and Lindekilde’s (2012, p. 1622) phrasing, this policing of the boundaries of tolerable ideas reflects ‘a “perfectionist liberalism” which aims to mould individuals into liberal citizens, who not only passively accept the liberal principles of the rule of law and democracy, but actively endorse them’. Such moves can have very real consequences for immediate and precious rights, liberties, and protections. As Arun Kundnani (2014, pp. 12–13) notes in discussing the impacts of ‘radicalisation’ discourse upon Muslims within the United States, ‘when the government widened the perceived threat of terrorism from individuals actively inciting, financing, or preparing terrorist attacks to those having an ideology, they brought constitutionally protected activities of large numbers of people under surveillance’. The UK’s Prevent Strategy has been widely critiqued on similar grounds, being famously described by one former senior police officer as a ‘toxic brand’, especially among Muslim communities (Halliday & Dodd, 2015). As this suggests, if the targeting of seemingly illiberal ideas within broader counter-radicalisation initiatives creates (or attempts to create) particular kinds of citizens, it is among minority religious or ethnic communities that their force may be most strongly felt. Although we need to be wary of generalisation here (Jarvis & Lister, 2013), such communities may already be stigmatised for their ideas or practices, and counter-radicalisation initiatives can draw upon and perpetuate fears around integration, cohesion, or multiculturalism. Bennett (2019) discusses this in the Australian context as ‘the securitization of unconventional views’ or ‘minority beliefs’. And high-profile failings in countries such as the United Kingdom show the impact such suspicions can have on what may be already vulnerable citizens, from the 11-year-old boy referred to in Prevent for advocating giving ‘alms to the oppressed’ (misinterpreted as ‘arms’ by the teacher) (Taylor, 2021) to the eight-year-old boy questioned by police officers for wearing a t-shirt reading, ‘I want to be like Abu Bakr al-Siddique’ (Mortimor, 2016): a reference to a much-esteemed companion of the Prophet Muhammad. High-profile errors such as the above are indicative of how the targeting of ideas as well as violence serves significantly to widen the spaces within which counter-radicalisation work is done. In countries such as the United Kingdom, schools, colleges, universities, doctors’ surgeries, workplaces, and beyond all now function as sites of counter-radicalisation, with legally proscribed ‘safeguarding’ duties attached to their professionals (Heath-Kelly, 2017). This creates significant questions of competence and responsibility among public sector employees who may lack the time, training, or expertise to monitor and police radical ideas, to say nothing of the complications it poses – and the potential damage it may do – to vital social relationships such as, say, between teacher and pupil (Moffat & Gerard, 2020; although compare Busher et al., 2019). Concerns around stereotyping are relevant, too, for narrower approaches to counter-radicalisation focused on the potential violence of those identified as ‘at risk of being risky’ (Heath-Kelly, 2013). Individuals (identified as being) from minority communities are far more likely to be targeted for intervention (Kundnani, 2014), although a growing engagement with ‘far-right’ radicalisation may be beginning to shift this balance in some contexts (Monaghan & Molnar, 2016; Jarvis, 2022). Such targeting can lead to the construction of ‘suspect communities’ (compare Pantazis & Pemberton, 2009; Ali, 2015; Ragazzi, 2015; Thomas & Grossman, this volume) that are believed to be home to and therefore in some way responsible for violent behaviours. Beyond the immediate and long-term harm done to the targets of suspicion (see Hillyard, 1993), such dynamics, in turn, can reinforce and add to inter-communal hostilities, anxieties, and fears (Hardy, 2018, pp. 98–99). 257

Kodili Chukwuma and Lee Jarvis

Conclusion In this chapter, we have focused on the ends or targets of counter-radicalisation initiatives. Drawing on prominent ‘real world’ examples from a range of contemporary policy contexts as well as academic scholarship on initiatives such as these, we have offered two overarching arguments. First, a provisional and heuristic distinction may helpfully be drawn between narrower and broader approaches to counter-radicalisation. Where the former prioritises the targeting of violence, the latter involves the targeting of ideas defined as extreme or radical. In the second half of the chapter, we then argued that narrow and broad approaches raise important, although often different, questions around the conceptual relationship between ideas and violence, the effectiveness of counterradicalisation policy, and the legitimacy of targeting particular ideas or behaviours. Such questions have a bearing on vitally important dynamics, including the organisation and governance of democratic societies, the protections that are afforded to citizens and their communities, and the responsibilities and reach of the state’s apparatuses such as schools and healthcare providers. On the basis of all of the above, it is vital, we argue, to remember that designating ideas or acts of violence as radical or extreme not only involves making contingent and contestable decisions that are structured by historical, social, and political circumstances. But, in addition, such decisions are deeply political and contribute to the (re)creation or (re)production of mainstream or moderate norms – of ideas and behaviour – that are to be tolerated, permitted, and defended.

Notes 1 Our sincere thanks to the editors and fellow contributors for their helpful and thorough feedback on earlier drafts of this chapter and the ideas herein. Any errors remain, of course, our own. 2 At the time of writing, the UK Prevent Strategy was subject to an independent review under Lord Shawcross. For further information, see www.gov.uk/government/collections/independent-review-of-prevent

References Ali, N. (2015). Mapping the Muslim community: The politics of counter-radicalisation in Britain. In C. Baker-Beall, C. Heath-Kelly, & L. Jarvis (Eds.), Counter-radicalisation: Critical perspectives (pp. 139– 155). Routledge. Bartlett, J., & Birdwell, J. (2010). From suspects to citizens: Preventing violent extremism in a big society. Demos. Bennett, N. (2019). One man’s radical. Security Challenges, 15(1), 47–62. Berger, J. M. (2016a). Making CVE work: A focused approach based on process disruption. International Centre for Counter-Terrorism Research paper. www.jstor.org/stable/resrep29442. Berger, J. M. (2016b). Promoting disengagement from violent extremism. International Centre for CounterTerrorism Research Paper. www.jstor.org/stable/resrep17489. Berger, J. M. (2018). What is extremism? In Extremism (pp. 26–43). The MIT Press. Bieler, A. (2001). Questioning cognitivism and constructivism in IR theory: Reflections on the material structure of ideas. Politics, 21(2), 93–100. Breidlid, T. (2021). Countering or contributing to radicalisation and violent extremism in Kenya? A critical case study. Critical Studies on Terrorism, 14(2), 225–246. Busher, J., Choudhury, T., & Thomas, P. (2019). The enactment of the counter-terrorism “Prevent duty” in British schools and colleges: Beyond reluctant accommodation or straightforward policy acceptance. Critical Studies on Terrorism, 12(3), 440–462. Cassam, Q. (2018). The epistemology of terrorism and radicalisation. Royal Institute of Philosophy Supplements, 84, 187–209. Cherney, A. (2020). Evaluating interventions to disengage extremist offenders: A study of the proactive integrated support model (PRISM). Behavioural Science of Terrorism and Political Aggression, 12(1), 17–36. Clubb, G., & McDaid, S. (2019). The causal role of ideology and cultural systems in radicalisation and deradicalisation. Journal of Critical Realism, 18(5), 513–528.

258

The politics of counter-radicalisation Commission for Countering Extremism. (2019). Challenging hateful extremism. Retrieved February  23, 2022, from https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_ data/file/874101/200320_Challenging_Hateful_Extremism.pdf Crone, M. (2014). Religion and violence: Governing Muslim militancy through aesthetic assemblages. Millennium: Journal of International Studies, 43(1), 291–307. Danish Ministry of Immigration, Integration and Housing. (2016). Preventing and countering extremism and radicalisation: National action plan. Retrieved July 20, 2021, from https://ec.europa.eu/ home-affairs/sites/default/files/what-we-do/networks/radicalisation_awareness_network/docs/ preventing_countering_extremism_radicalisation_en.pdf Dexter, H. (2012). Terrorism and violence: Another violence is possible? Critical Studies on Terrorism, 5(1), 121–137. Edwards, P. (2015). How (not) to create ex-terrorists: PREVENT as ideological warfare. In C. Baker-Ball, C. Heath-Kelly, & L. Jarvis (Eds.), Counter-radicalisation: Critical perspectives (pp. 54–70). Routledge. El-Said, H. (2015). New approaches to countering terrorism: Designing and evaluating counter radicalization and de-radicalization programs. Springer. European Commission. (2020, December 9). A counter-terrorism agenda for the EU: Anticipate, prevent, protect, respond. Retrieved July 20, 2021, from https://ec.europa.eu/home-affairs/sites/default/files/pdf/09122020_ communication_commission_european_parliament_the_council_eu_agenda_counter_terrorism_ po-2020-9031_com-2020_795_en.pdf Federal Republic of Nigeria. (2017). Policy framework and national action plan for preventing and countering violent extremism. Retrieved July 15, 2021, from https://ctc.gov.ng/pcve-national-framework-and-action-plan/ Federal Republic of Nigeria, National Counter-terrorism Strategy (NACTEST). (2016). Retrieved July 15, 2021, from https://ctc.gov.ng/national-counter-terrorism-strategy-2016/ Foucault, M. (2003). Society must be defended: Lectures at the College de France, 1975–1976. Picador. Gielen, A. J. (2019). Countering violent extremism: A realist review for assessing what works, for whom, in what circumstances, and how? Terrorism and Political Violence, 31(6), 1149–1167. Gunning, J. (2007). A case for critical terrorism studies? Government and Opposition, 42(3), 363–393. Halliday, J., & Dodd, V. (2015, March 9). UK anti-radicalisation prevent strategy a “toxic brand”. The Guardian. Retrieved February  23, 2022, from www.theguardian.com/uk-news/2015/mar/09/ anti-radicalisation-prevent-strategy-a-toxic-brand Hardy, K. (2018). Comparing theories of radicalisation with countering violent extremism policy. Journal for Deradicalization, 15, 76–110. Heath-Kelly, C. (2013). Counter-terrorism and the counterfactual: Producing the “radicalisation” discourse and the UK PREVENT strategy. The British Journal of Politics and International Relations, 15(3), 394–415. Heath-Kelly, C. (2017). The geography of pre-criminal space: Epidemiological imaginations of radicalisation risk in the UK prevent strategy, 2007–2017. Critical Studies on Terrorism, 10(2), 297–319. Heath-Kelly, C., Baker-Beall, C., & Jarvis, L. (2015). Introduction. In C. Baker-Beall, C. Heath-Kelly, & L. Jarvis (Eds.), Counter-radicalisation: Critical perspectives (pp. 1–13). Routledge. Herman, E., & O’Sullivan, G. (1989). The “terrorism” industry: The experts and institutions that shape our view of terror. Pantheon Books. Hillyard, P. (1993). Suspect community: People’s experience of the prevention of terrorism acts in Britain. Pluto Press. Horgan, J. (2009). Walking away from terrorism: Accounts of disengagement from radical and extremist movements. Routledge. Horgan, J., & Altier, M. (2012). The future of terrorist de-radicalization programs. Georgetown Journal of International Affairs, 13(2), 83–90. Horgan, J., & Braddock, K. (2010). Rehabilitating the terrorists? Challenges in assessing the effectiveness of de-radicalization programs. Terrorism and Political Violence, 22(2), 267–291. Jarvis, L. (2009). The spaces and faces of critical terrorism. Security Dialogue, 40(1), 5–27. Jarvis, L. (2014). How might we evaluate counterterrorism policy: The three W’s of counterterrorism. Eastminster. Retrieved February 23, 2022, from www.ueapolitics.org/2014/12/17/evaluate-counterterrorism-policy/ Jarvis, L. (2022). Critical terrorism studies and the far-right: Beyond problems and solutions? Critical Studies on Terrorism. https://doi.org/10.1080/17539153.2021.2017484 Jarvis, L., & Lister, M. (2013). Disconnected citizenship? The impacts of anti-terrorism policy on citizenship in the UK. Political Studies, 61(3), 656–675.

259

Kodili Chukwuma and Lee Jarvis Jarvis, L., Marsden, L., & Atakav, E. (2020). Public conceptions and constructions of “British values”: A qualitative analysis. The British Journal of Politics and International Relations, 22(1), 85–101. Juergensmeyer, M., & Sheikh, M. K. (2013). A sociological approach to understanding religious violence, in M. Juergensmeyer, M. Kitts, & M. K. Jerryson (Eds.), The Oxford handbook of religion and violence (p. 627). Oxford University Press. Kühle, L., & Lindekilde, L. (2012). Radicalisation and the limits of tolerance: A Danish case-study. Journal of Ethnic and Migration Studies, 38(10), 1607–1623. Kundnani, A. (2009). Spooked: How not to prevent violent extremism. Institute of Race Relations. Kundnani, A. (2014). The Muslims are coming! Islamophobia, extremism, and the domestic war on terror. Verso. Laqueur, W. (2004). The terrorism to come. Policy Review, 126, 49–64. Lindekilde, L. (2016). Radicalisation, de-radicalisation and counter-radicalisation. In R. Jackson (Ed.), The Routledge handbook of critical terrorism studies (pp. 248–259). Routledge. Lockley-Scott, A. (2019). Towards a critique of fundamental British values: The case of the classroom. Journal of Beliefs & Values, 40(3), 354–367. Macdonald, S. (2015). Prosecuting suspected terrorists: Precursor crimes, intercept evidence and the priority of security. In L. Jarvis & M. Lister (Eds.), Critical perspectives on counter-terrorism (pp. 130–149). Routledge. Marret, J. L., Feddes, A. R., Mann, L., Doosje, B., & Griffioen-Young, H. (2013). An overview of the SAFIRE project: A  scientific approach to finding indicators of and responses to radicalisation. Journal EXIT – Deutschland: Zeitschrift für Deradikalisierung und demokratische Kultur, 1(2), 123–148. Marsden, L., Jarvis, L., & Atakav, E. (2022). “That still goes on, doesn’t it, in their religion?” British values, Islam and vernacular discourse. Nations and Nationalism. https://doi.org/10.1111/nana.12849 Mauss, M. (1934). Les techniques du corps. Journal de Psychologie, XXXII(3–4). Moffat, A., & Gerard, F. J. (2020). Securitising education: An exploration of teachers’ attitudes and experiences regarding the implementation of the prevent duty in sixth form colleges. Critical Studies on Terrorism, 13(2), 197–217. Monaghan, J., & Molnar, A. (2016). Radicalisation theories, policing practices, and “the future of terrorism?” Critical Studies on Terrorism, 9(3), 393–413. Mortimor, C. (2016, August 1). Eight-year-old boy questioned after teachers mistake t-shirt slogan for Isis propaganda. Independent Online. Retrieved July 19, 2021, from www.independent.co.uk/news/uk/homenews/isis-propaganda-t-shirt-boy-east-london-teachers-mistake-terrorism-terror-a7164941.html Mouffe, C. (2011). On the political. Routledge. Neumann, P. (2013). Options and strategies for countering online radicalisation in the United States. Studies on Conflict and Terrorism, 36(6), 431–459. Pantazis, C., & Pemberton, S. (2009). From the “old” to the “new” suspect community: Examining the impacts of recent UK counter-terrorist legislation. The British Journal of Criminology, 49(5), 646–666. Ragazzi, F. (2015). Policed multiculturalism? The impact of counter-terrorism and counter-radicalization and the “end” of multiculturalism. In C. Baker-Beall, C. Heath-Kelly, & L. Jarvis (Eds.), Counter-radicalisation: Critical perspectives (pp. 156–174). Routledge. Sageman, M. (2004). Understanding terror networks. University of Pennsylvania Press. Sageman, M. (2008). Leaderless Jihad: Terror networks in the twenty-first century. University of Pennsylvania Press. Sardoč, M. (2020). Radicalisation, violent extremism and terrorism: An interview with Quassim Cassam. Critical Studies on Terrorism, 13(1), 166–174. Schmid, A. P. (2013). Radicalisation, de-radicalisation, counter-radicalisation: A conceptual discussion and literature review. ICCT Research Paper, 97(1). Retrieved July 19, 2021, from www.icct.nl/app/uploads/ download/file/ICCT-Schmid-Radicalisation-De-Radicalisation-Counter-Radicalisation-March-2013.pdf Schuurman, B., & Taylor, M. (2018). Reconsidering radicalisation: Fanaticism and the link between ideas and violence. Perspectives on Terrorism, 12(1), 3–22. Sedgwick, M. (2010). The concept of radicalisation as a source of confusion. Terrorism and Political Violence, 22(4), 479–494. Sewell, A., & Hulusi, H. (2016). Preventing radicalisation to extreme positions in children and young people: What does the literature tell us and should educational psychology respond? Educational Psychology in Practice, 32(4), 343–354.

260

The politics of counter-radicalisation Spalek, B., & Lambert, R. (2008). Muslim communities, counter-terrorism and counter-radicalisation: A critically reflective approach to engagement. International Journal of Law, Crime and Justice, 36(4), 257–270. Stampnitzky, L. (2013). Disciplining terror: How experts invented “terrorism”. Cambridge University Press. Stevens, D. (2009). In extremis: A self‐defeating element in the “preventing violent extremism” strategy. The Political Quarterly, 80(4), 517–525. Stevens, D. (2011). Reasons to be fearful, one, two, three: The “preventing violent extremism” agenda. The British Journal of Politics and International Relations, 13(2), 165–188. Taylor, D. (2021). Boy, 11, referred to prevent for wanting to give “alms to the oppressed”. The Guardian. Retrieved July 19, 2021, from www.theguardian.com/uk-news/2021/jun/27/boy-11-referred-to-preventfor-wanting-to-give-alms-to-the-oppressed UK Home Office. (2011). Prevent strategy. Retrieved July 20, 2021, from https://assets.publishing.service. gov.uk/government/uploads/system/uploads/attachment_data/file/97976/prevent-strategy-review.pdf UN General Assembly. (2015, December 24). Plan of action to prevent violent extremism. Retrieved July 20, 2021, from www.un.org/en/ga/search/view_doc.asp?symbol=A/70/674 Van de Weert, A. (2022). Justice and risk assessment: The subjectivity of pre-emptive screening of radicalisation processes. Journal of Policing, Intelligence and Counter Terrorism, 17(3), 287–300. Vidino, L., & Brandon, J. (2012). Europe’s experience in countering radicalisation: Approaches and challenges. Journal of Policing, Intelligence and Counter Terrorism, 7(2), 163–179. Wiktorowicz, Q. (2005). Radical Islam rising: Muslim extremism in the West. Rowman & Littlefield. Winter, C., Heath-Kelly, C., Kaleem, A., & Mills, C. (2021). A moral education? British values, colourblindness, and preventing terrorism. Critical Social Policy. https://doi.org/10.1177/0261018321998926

261

17 RESPONDING TO RADICALISATION WITH DIFFERENT IDEOLOGICAL ROOTS How similar is the ‘problem’? How similar are the ‘solutions’? Daniel Koehler Introduction Programmes designed to prevent and counter violent extremist radicalisation (P/CVE) have become cornerstones in many national and international counterterrorism policies, accompanied by a widespread proliferation of such initiatives worldwide in the decades following the London 7 July bombings in 2005 (Koehler, 2016). In recent years, however, (counter-)radicalisation research and practice have grown increasingly diverse, while previously they were mostly focused on specific forms of violent extremism, namely Islamist extremism. Except for a few countries (e.g. Germany, Sweden, and Norway), the extreme right (or, speaking more broadly, domestic extremism) was mostly ignored. Building and maintaining counter-radicalisation (including prevention and deradicalisation) programming across ideological milieus and environments is no easy task and puts significant additional demands for evidence-based decisions on a practical field notorious for the widespread lack of transparency as well as limited quality control, evaluation, and theoretically informed programme design (e.g. Feddes & Gallucci, 2015; Horgan & Braddock, 2010; Jugl et al., 2020; Koehler, 2017). In the following sections, this chapter therefore reviews some of the key similarities and differences that research on countering radicalisation across different forms of violent extremist milieus and ideologies, as well as practical field experiences in Germany, have brought to light. After sketching the main points on which general agreement exists in the academic perspective on countering radicalisation (e.g. the contextual and individual nature of interventions, the necessity to include the emotional and social environment, as well as providing multi-disciplinary and hand-tailored responses to individual radicalisation trajectories), the chapter crosses over into the practitioner field. Using Germany as a case study, which has one of the most diverse counterradicalisation landscapes in the world and, importantly, has aimed for a comprehensive approach to violent extremism, it is possible to draw some main lessons regarding the challenges and differences in practical work to prevent and intervene against different extremist ideologies. The chapter

DOI: 10.4324/9781003035848-20 262

Responding to radicalisation with different ideological roots

then links lessons learned from Germany to wider debates and points out where they might differ in important ways from other countries. Finally, the chapter closes with recommendations for bridging some of the gaps and methodical differences in these fields. While counter-radicalisation interventions in all cases need to preserve sensitivity to the peculiarities of each extremist milieu and potential needs for unique approaches, existing research and practical experiences strongly suggest that a main body of generally valuable tools and approaches does exist. Hence, some reflection is provided as to why these extremist milieus are sometimes treated differently. This also has implications for future research. The key question in every context is how the most adequate mixture of approaches and methods can be selected and adapted to each individual client.

Comparative extremism research Terrorism research has been criticised on many occasions for being ‘too narrow, too event-driven and too strongly tied to governments’ counterterrorism policies’ (Schuurman, 2019, p. 463). Furthermore, the strong focus on specific ideological families and certain types of violent extremism and terrorism has also been noted repeatedly. For example, in 2011, Alex Schmid found that only 0.6% of 4.458 peer-reviewed sources treated domestic terrorism (Schmid, 2011, p. 461). Even though this situation improved somewhat in 2019, Schuurman still pointed out that ‘only 1.9% of the literature on terrorism as represented by the 3.442 articles published between 2007 and 2016 in nine of the field’s leading academic journals focus on right wing extremism’ (Schuurman, 2019, p. 8). Nevertheless, the overall field of terrorism studies ‘made considerable steps towards addressing long-standing issues’ (Schuurman, 2018, p. 1011). Among those developments is also an increasing interest in a comparative perspective looking at potential differences and similarities between various ideological families and milieus within the violent extremist and terrorist spectrum in recent years. This comparative approach has also slowly begun to approach the issue of disengagement and deradicalisation, trying to find potential common threads and characteristics as well as key differences involved when leaving one extremist milieu or another. The following section outlines some of the key evolving themes and research discourses that are part of this comparative perspective. Roughly, one can distinguish four main foci of comparative terrorism and extremism research: common ideological features, personality traits, and entry and exit processes (i.e. radicalisation and deradicalisation/disengagement).

Common ideological features Even though no universally accepted definitions of terrorism and violent extremism exist, some scholars have argued to approach the issue by pointing out commonly shared psychological and ideological elements among violent extremists instead of focusing on the relative position of certain political systems or organisational features, for example. Here, Berger’s (2018) understanding of ‘extremism’ essentially highlights the importance of in-group bias and hostility towards a defined other or enemy: Extremism refers to the belief that an in-group’s success or survival can never be separated from the need for hostile action against an out-group. The hostile action must be part of the in-group’s definition of success. Hostile acts can range from verbal attacks and diminishment to discriminatory behavior, violence, and even genocide. (Berger, 2018, p. 44)

263

Daniel Koehler

Extremist milieus require an ‘unwavering commitment’ (Berger, 2018, p. 33) to that out-group hostility, as the group, in Berger’s view, cannot exist or be successful without it. It is therefore the inseparability of that enmity against outsiders from the very nature of what (all) extremist environments are all about that characterises them. While not erasing ideological differences and distinctiveness (e.g., the specific enemy identified by the ideology naturally differs), such an approach helps to establish a comparative scientific and practical discourse. Notwithstanding this discourse, it seems that some extremist milieus are ideologically closer to each other than the rest. In their seminal study on (jihadist) educational profiles overrepresenting the STEM fields (i.e. science, technology, engineering, and mathematics), Gambetta and Hertog (2016) also attempted an ideological comparison of far-right, far-left, and Islamist extremist milieus following from their hypothesis that certain personality types could be more or less attracted to different types of extremist ideologies. Especially they pointed out a stronger link between the far-right and Islamist extremists, since left-wing radicalism is rooted in the Enlightenment and in a belief in scientific progress (however oddly framed by dialectical materialism), right-wing and Islamist extremists share an antiscientific worldview. (Gambetta & Hertog, 2016, p. 93) On the basis of that assumption, they compiled a list of 15 Islamist extremist ideological core elements through consultation with academic literature and other experts. Comparing those with far-right and far-left milieus, 13 were shared with the extreme right and five with the extreme left (Gambetta & Hertog, 2016, p. 95). Clearly, this comparative analysis needs further exploration and refinement. Nevertheless, it seems that some extremist milieus are ideologically likeminded: If we abstract from the transcendental beliefs that by definition concern religiously inspired groups, such as the Islamists, there is a near complete coincidence between the ideological cocktail of Islamist extremism and that of right-wing extremism. (Gambetta & Hertog, 2016, p. 97) Other approaches to identifying potential ideological ‘pathways’ or ‘bridges’ between otherwise mutually exclusive extremist ideologies have found evidence for the hypothesis that shared ideological elements or narratives do exist between some of these milieus. Exploring extremist sideswitchers between extreme right, left, and Islamist milieus, Koehler (2019, 2021b) found that especially commonly shared anti-Semitism and devotion to fighting a mainstream political culture through living a radical and oftentimes violent lifestyle are used by many side-switchers to justify their defection to the enemy camp. Likewise, Meiering et al. (2020) argued that the identification of ‘bridging narratives’ could facilitate comparative radicalisation studies. In their study of shared ideological elements among various extremist environments, they identified commonly constructed enemy frames (e.g. modernity, Jewish people, and feminism) and idealistic collective identities (e.g. toxic masculinity and heroism). This further strengthens the notion that considerable ideological relations and connections might exist between various extremist ideologies. In addition, as has been pointed out by Moskalenko and McCauley (2009), a shared feature among extremists might be the readiness to use illegal and violent means to achieve goals (radicalism) versus participation in peaceful and legal political action (activism). This argument closes the circle to the initial extremism framing by Berger: as long as the ideological core evolves around 264

Responding to radicalisation with different ideological roots

illegality and violence, the nature of the various radicalisms or extremisms might be much closer related to each other than to their various non-radical (i.e. activist) forms. However, ideological commonalities are but one aspect. As noted by Gambetta and Hertog, certain ideological profiles might also attract specific personality types as potential followers and recruits. The issue of extremist ideologies or political attitudes in general and their connection to personality traits has received far more extensive attention among scholars than ideological commonalities.

Personality traits Gambetta and Hertog found confirmation for their hypothesis that different types of personalities are attracted to those ideological clusters sharing fewer elements. They suggest three pre-ideological personality traits and one plausible complement shared by those individuals more likely to be attracted to far-right and Islamist extremism: the proneness to experience disgust, a strong need for cognitive closure, the strict distinction between in-group and out-group, as well as simplism (Gambetta & Hertog, 2016, pp. 130, 150). On the other hand, Berger (2018) has argued that strict in-group versus out-group hostility is essential for all forms of extremism. Indeed, scholars of political psychology have attempted to identify personality traits closely connected to political attitudes in extensive studies. Hans Eysenck, for example, found similarities between fascists and communists regarding their tough-mindedness (Eysenck, 1968). Decades later, this notion was somewhat supported in the systematic literature review investigating the relationship between mental health factors and radicalisation by Misiak et al. (2019), who found that empirical research has provided evidence for personality traits associated with radicalisation such as identity fusion, the need for group identification, low levels of empathy, morbid transcendence, feelings of being treated unjustly together with harbouring high levels of grievance, rational decision-making, dependent decision style, cognitive complexity, uncertainty, and an analytical cognitive style. Furthermore, personality disorders below the threshold for diagnosable psychological illnesses among extremists in general could be paranoid personality disorder, self-defeating personality disorder, and schizotypal personality disorder (Misiak et al., 2019, p. 57). Other studies highlight the dominance of cognitive rigidity among people holding extremist political views (Zmigrod et al., 2019). Extremists across ideological milieus are usually distrustful and critical towards the political system they live in and often display a black-and-white thinking style by which they categorise the world, for example, as good versus evil (Van Prooijen & Krouwel, 2019). Furthermore, the mindset of extremists and terrorists is often characterised by patterns such as, for example, militaristic thinking and terminology, the glorification of the past and of dying for a cause, utopianising, or the dehumanisation of the enemy (Saucier et al., 2009). Most recently, Kruglanski et al. (2021) have argued that every form of extreme behaviour could be the result of ‘uniform psychological dynamics’ involving a ‘motivational imbalance’ (Kruglanski et al., 2021, p. 184), for example, caused by humiliation and deprivation. Many studies in the field of political psychology, however, have identified an ideological asymmetry between left and right political attitudes (e.g. Aichholzer et al., 2018; Vasilopoulos & Jost, 2020). Left-wing attitudes are oftentimes connected to open mindedness and intellectual curiosity, while right-wing views are seen to be more attractive to people scoring higher on conscientiousness (efficient/organised vs. extravagant/careless) and more rigid, inflexible thinking, which does align well with the hypothesis by Gambetta and Hertog introduced earlier. Nevertheless, recent research by Zmigrod and Goldenberg (2021) found evidence that the combination of cognitive rigidity and emotional sensation seeking can facilitate readiness for violent extremist action 265

Daniel Koehler

across the spectrum. This does lend further support to the hypothesis that individuals who engage in extremist behaviour or internalise extremist worldviews (compared to non-violent radicalism) might share certain psychological features or mechanisms, for which a proxy indicator might be the widely shared belief in conspiracy theories among many forms of violent extremism (Bartlett & Miller, 2010). Furthermore, certain personality traits that are potentially more attracted towards violent extremism and terrorism arguably exist in every society and time but have a very different scope and impact on their surrounding environments depending on the dominant collective mentality and other macrosocial factors. It has been shown, for example, that ostracism and social exclusion on a societal level consistently increase individual support for political violence (Treistman, 2021; Pfundmair, 2018). A variety of additional contextual factors could be conducive to the manifestation of such personality traits and their connection to violent extremist action, for example, domestic or transnational political conflict, immigration, economic conditions, or the socio-demographic composition of the population in question (e.g. Doering & Davies, 2019). Such conducive environments arguably also have similar or connected effects on violent extremist milieus and the people attracted to them across ideological boundaries. A prime example would be the German Nazi stormtroopers (Sturmabteilung, SA) and the communist Alliance of Red Front Fighters (Roter Frontkämpferbund, RFB), which both used the political and historical context of Weimar Republic Germany to recruit members from largely the same target group and with many similar arguments (Brown, 2009). From this, it follows that successful counter-radicalisation and P/CVE working in the long term should also include those societal factors contributing to individual-level radicalisation across ideological milieus.

Entry and exit Research on entry processes into violent extremism and terrorism mainly evolved around the controversial term ‘radicalisation’, especially after the London terror attacks on 7 July 2005 (Sedgwick, 2010). Typically, the concept describes a ‘process by which an individual adopts an extremist ideology’ (Braddock, 2014, p. 62). Whether or not the use of violence is a key aspect of radicalisation remains contested, and increasingly, even benevolent forms of this phenomenon are recognised (Reidy, 2018). A vast amount of radicalisation research has been conducted and published since the September 11 attacks, mostly looking at religious, or rather, Islamist ideologies. In recent years, the empirical base has widened significantly beyond this ideological milieu, and several meta-studies have attempted to summarise the state of the art at various times (e.g. Borum, 2011a, 2011b; Dalgaard-Nielsen, 2010). Gøtzsche-Astrup (2018), for example, collected those theories explaining radicalisation with the highest empirical validity to date. In his perspective, strong empirical evidence points to radicalisation as a normal psychological mechanism rather than psychopathology. He also sees enough support in the literature to focus on motivational processes rather than instrumental calculations of risk and reward, as well as negative life experiences that put the individual in flux in terms of fundamental questions. Experiences of fundamental uncertainty or loss of meaning or significance and shift in social identity towards a single social group rather than many and small group dynamics driving the process to behavioural extremes are equally key to understanding the term and its underlying process. Finally, Gøtzsche-Astrup highlights heightened dispositional anxiety, aggression, and impulsivity, and the role of ‘sacred values’ in developing extremist attitudes and actions. Even though the majority of radicalisation studies that were used

266

Responding to radicalisation with different ideological roots

in the assessment by Gøtzsche-Astrup arguably mainly focused on Islamist radicalisation, the empirical validity approach also includes much broader perspectives. In particular, framing radicalisation beyond the use of violence as the individual’s desire to fundamentally alter the surrounding environment has proven helpful for a comparative approach. Echoing this notion, Dalgaard-Nielsen defines ‘radicalisation’ as ‘a growing readiness to pursue and support far-reaching changes in society that conflict with, or pose a direct threat to, the existing order’ (Dalgaard-Nielsen, 2010, p. 798). Furthermore, numerous process models of radicalisation trajectories have been developed to show the different phases and motivational mechanisms in each step (for an overview, see De Coensel, 2018). All these models point to the fact that individual radicalisation pathways are gradual processes spanning over a certain period that involve different cognitive and behavioural steps one must take towards an end state that could be classified as extremist. This understanding also connects to wider sociological research highlighting the socialisation processes that people entering extremist milieus typically undergo (e.g. Orsini, 2020). Hence, many scholars would agree today that extremists of different ideological convictions share at least some fundamental psychological mechanisms and steps during their entry processes. In consequence, most theories attempting to explain and trace the pathways leading to extremist involvement aim for broader applicability beyond a specific ideological environment. Most recently, for example, the so called ‘Attitudes-Behaviour-Correctives (ABC) model’ by Khalil et al. (2019) approached the issue by mapping out different combinations and trajectories of individuals along opposition to or sympathy for ideologically justified violence (attitudinal level) on the one hand and the extent of involvement in ideologically justified violence (behavioural level) on the other. Through such models, it has become possible to chart various types and steps of involvement in ‘ideologically motivated’ or ‘justified’ violence beyond a specific milieu. Another model that helps to inform P/CVE interventions across ideological spectrums was suggested by Koehler (2016, pp. 65–94) and understands the phenomenon as a process of individual depluralisation of political concepts and values (e.g. justice, freedom, honour, violence, and democracy) on the one hand and an increase in ideological urgency to act against an individual or collective problem on the other. With a higher degree of individual internalisation of the notion that no other alternative interpretations of the (individually prioritised) political concepts and values exist (or are relevant), one can show (e.g. in syntax, language, and behaviour) the progression of the radicalisation process towards a mindset that is fully exclusive and relies on existential conflict for social and individual identity. Beyond such rather abstract conceptualisations of entry pathways, little comparative research exists that maps out the similarities and differences of these processes between specific ideological milieus (for a positive example, see Brown et al., 2021). On the basis of the extensive research focusing on extreme right radicalisation and the growing evidence base for Islamist entry processes in Germany, a national research project there concluded in 2021 and attempted to do just that (Freiheit et al., 2021). In their assessment, the authors found numerous similar factors involved in the radicalisation of right-wing and Islamist extremists. Among the shared factors that were identified are, for example, youth-specific vulnerabilities (e.g. developmental challenges and rebellion against parents and establishment), secondary importance of ideology or religion, search for respect, meaning, and orientation, protest and provocation, experiences of injustice, search for adventure and fun, group belonging, access to the extremist milieu via social networks such as friends and family, transgenerational transmission of ideological components, crisis and trauma experiences, and societal alienation resulting from a lack of education or individual dispositions (e.g. mental health problems or dissocial personalities). The project also identified certain

267

Daniel Koehler

differences between the two environments, such as (among others) marginalisation in school and social environments (extreme right) versus religion- and heritage-based discrimination (Islamists), relevance of religious and spiritual motives among Islamists, ethnically dominated recruitment (extreme right) versus integrative recruitment (Islamists), or differences in the specific ideological reference points (Freiheit et al., 2021, pp. 254–255). Similarly, comparative approaches to exit pathways (deradicalisation and disengagement) have increasingly moved towards identifying push and pull factors, trajectories, and underlying steps in the way out of extremism and terrorism beyond the specific characteristics of certain ideological milieus. In her study of exit processes from violent extremism and terrorism published in 2013, Dalgaard-Nielsen (2013) assessed 16 publications between 1990 and 2012 with primary data on exit based on 216 interviews. At that point, she identified three clusters of reasons for exit: doubts in ideology or group (leadership), personal issues, and practical issues. Doubt in ideology is often the result of violence, a gradual process, and is created through significant others. Disillusionment with other group members and the underlying ideology have been identified as particularly influential exit reasons across the ideological spectrum. In their analysis of 87 autobiographical accounts of former extremists and terrorists with various ideological backgrounds, Altier et al. (2017) concluded that push factors located within the inner workings of an extremist milieu are much more relevant for exit processes than pull factors (without discarding the latter). In particular, ‘disillusionment with the group’s strategy or actions, disagreements with group leaders or members, dissatisfaction with one’s day-to-day tasks, and burnout’ (Altier et al., 2017, p. 305) are oftentimes mentioned to drive the decision to leave the group and environment behind. Some studies have also approached the issue through milieu-specific comparisons regarding the relevance of certain pre-selected push and pull factors. In their exploration of the exit processes of extreme left and right activists, Windisch et al. (2017) noted that perceptions of poor planning and organisation, low-quality personnel, and vindictive behaviour within these milieus can generate organisational distrust and disillusionment. One important difference between the two environments is, according to the study, the lesser relevance of in-group violence as a motivator to exit among far-leftists. In one of the largest sample-based studies comparing extremist milieus regarding various aspects of exit processes, James et al. (2019) used 300 former extreme left, extreme right, Islamist, and single-issue extremists from the US Profiles of Individual Radicalization in the United States (PIRUS) database. This study provides valuable insights into specific similarities and differences among violent extremist milieus. According to the analysis, ‘positive advancements in socioeconomic standing’ and becoming a parent could be identified as important push factors to exit for almost 50% of the extremists in the sample (James et al., 2019, p. 2). Furthermore, over 35% ‘cited changed religiosity (i.e. an increase in, a decrease in, or an reinterpretation of religious beliefs/participation) as important to their disengagement from extremism’ (James et al., 2019, p. 2). New social relationships were also very important aspects of the exit pathways, as noted by 36.1% of the sample. Further strongly present factors (with around a quarter of the sample each) were, for example, decreased substance use, disillusionment, and physical separation from the extremist environment (James et al., 2019, p. 2). The authors also identified important differences between the milieus. For example, the development of positive personal relationships with non-radicals and/or the termination of personal relationships with radicals were present in the exit processes for a large number of far-right and far-left extremists (43.9% and 41.1%, respectively) but were less common for Islamist extremists (14.3%). Similarly, disillusionment with an extremist group or cause was cited by many far-right and Islamist extremists as a reason for disengagement (25.6% and 33.9%) but was less commonly cited by far-left extremists (18.3%) (James et al., 2019, p. 2). 268

Responding to radicalisation with different ideological roots

Such milieu-specific comparisons of exit push and pull factors as well as pathway characteristics are still rare. However, driven by the widening of P/CVE programming in many countries to cover more than one extremist milieu in recent years, the need for more differentiated knowledge about similarities and differences to inform those interventions and prevention activities has become more pressing. Practically, however, certain key methods and types of activities have become standard repertoires among P/CVE programmes across the ideological spectrum. The use of mentoring and counselling or the effectiveness of including the close emotional and social environment of persons exiting such environments could be named here (e.g. El-Amraoui & Ducol, 2019; Spalek & Davies, 2012). In an extensive comparison of programme-level P/CVE practices, Koehler (2016) identified five main pillars of methods and tools generally used in this field around the world and across the ideological spectrum: ideological/theological, creative arts and sports, social work, psychological, and educational. Similarly, Dalgaard-Nielsen identified several generally recommended practices for P/CVE interventions based on the empirical exploration of exit processes: A humanization of the enemy, de-idealizing violence, leveraging internal strain in the extremist groups, leveraging bad leadership and/or personal and practical issues such as guilt feelings, longing for a normal life, and burnout. Moreover, the case studies indicate that increased contact with the world outside the extremist environment is a frequent trigger or accelerator of an exit process – a fact that supports the notion that external actors and external interventions can make a difference. Social influence – direct or indirect – appears to be at work. (Dalgaard-Nielsen, 2013, p. 106) Furthermore, mandatory ideological confrontation is strongly discouraged. To avoid reactance and cognitive dissonance, programmes should strive to achieve a match between agent and target. Such practical recommendations appear to be valid regardless of the ideological milieu targeted by the intervention. Research on counter-radicalisation or prevention of radicalisation is so far much less evidencebased than its counterpart on radicalisation processes. Nevertheless, some scholarship points to the existence of general resilience promoting factors across all forms of violent extremism, such as individual freedom to express one’s norms and values, critical reflection skills, mental flexibility and adaptability, psychological robustness, and strong social ties to positive pro-social environments (Stephens & Sieckelinck, 2021). Since the evidence on risk for radicalisation has pointed out the dominance of classic criminogenic factors across extremist milieus, it follows that protective factors suggested in the literature can also be applied across the spectrum with some reason to expect positive results. Those protective factors are, for example, political participation and socio-demographic factors such as socio-economic status, education, marital status, and age. Furthermore, political satisfaction, general trust, depression, and out-group friendships, as well as school bonding and performance, parental involvement, institutional trust, and law abidance, have been suggested with small to medium and large effect sizes across the available empirical literature (Wolfowicz et al., 2020, p. 424). Another promising approach in counter-radicalisation across ideological milieus appears to be that of communicative inoculation, which has been demonstrated to create strong and positive effects against various kinds of extremist recruitment efforts (e.g. Saleh et al., 2021). Only in recent years have calls to broaden P/CVE practices to cover multiple extremist milieus been raised internationally and impacted national strategies. One of the few countries with longstanding experience in P/CVE targeting multiple extremist milieus (in this case, extreme right and 269

Daniel Koehler

Islamist) is Germany. Hence, practitioners and policy makers have debated the potential overlap and similarities between the different approaches. In short, German P/CVE experts have discussed the value of transferring certain lessons learned from interventions targeting the extreme right to the field of Islamism, as well as the necessity for milieu-specific approaches. It is, therefore, promising to take a closer look at the German P/CVE landscape and discourse regarding similarities and differences in cross-ideology interventions and prevention.

The German case study Beginning in the late 1980s with exit programmes targeting left-wing terrorists, Germany can look back on one of the most diverse and longest traditions in this field worldwide. In addition to its long history, P/CVE in Germany is also coined by the strong involvement of governmental and non-governmental actors, or, in other words, a ‘hybrid model of shared responsibility between CSOs and state actors [that] has led to the formation of a diverse and regionally differentiated landscape’, according to Baaken et al. (2020, p. 13). The German P/CVE field ‘developed organically rather than having been designed through the support of academic research and advice’ and is mostly built on pre-existing practical experience from social work (Baaken et al., 2020, p. 12). Historically, governmental (i.e. intelligence-led) reintegration assistance for left-wing terrorists came first in the late 1980s, followed by large-scale publicly funded youth social work targeting right-wing youths after the German reunification in 1990. Beginning in 2000, deradicalisation or exit programmes for neo-Nazis became widely popular among civil society and governmental actors. The field of Islamist extremism became a major concern for the German P/CVE landscape following the stark increase of foreign fighters travelling to Syria and Iraq to join jihadist groups beginning in 2011. In consequence, a national support hotline for relatives of Islamists was introduced in 2012 as a public-private partnership model. Since then, multiple state and municipalitylevel projects have been introduced, and the overall landscape has reached a significant degree of heterogeneity (for a detailed account of the German P/CVE landscape’s history, see Kruglanski et al., 2019, p. 39 et seq.). In 2018, the German Federal Criminal Police (Bundeskriminalamt, BKA) counted 1,642 active P/CVE projects and programmes, of which 60% were run by non-governmental organisations, which leaves a substantial role for governmental P/CVE (Lützinger et al., 2020, p. 602). The number of active projects and programmes has more than doubled in only a short time, as just a couple of years earlier, in 2015, the BKA had counted 721 such initiatives (Lützinger et al., 2016, p. 9). This stark increase, especially in non-governmental programmes (in 2015, the share of NGOs in the P/CVE field was 53.4%), is indicative of the effects of significantly increased public spending in this field. According to the German government, federal funding (excluding any state-level funding) for counter-radicalisation programmes increased from €42.8 million in 2015 to €147.7 million in 2019, after having reached a peak in 2018 with €151.3 million (Bundesregierung, 2020, p. 2). The triplication of P/CVE funding was even surpassed by the available resources set aside for deradicalisation programmes in particular. From €300,000 in 2015, this funding was increased to €7.5 million (Bundesregierung, 2020, p. 2). Far-right extremism clearly dominates the programme landscape as the main target ideology, accounting for 64% of all initiatives, while 32% focus on Islamist extremism and merely 7% on the extreme left (Lützinger et al., 2020, p. 600). Beyond any ideological focus in their core work, however, around 62% of the programmes also aim to counter extremism and radicalisation in general, without a specific target ideology. According to another P/CVE mapping project finished in 2021, out of 555 assessed prevention and intervention services in Germany, 392 (71%) included 270

Responding to radicalisation with different ideological roots

far-right and Islamist extremism. Of the sample, 298 (54%) projects and programmes were deemed fully comprehensive, as they also included far-left extremism, meaning that more than half of the German P/CVE landscape (as per this study) can be considered cross-ideologically active in one way or another (Freiheit et al., 2021, p. 231). The German P/CVE landscape has therefore become much more diverse regarding the forms of extremist ideologies addressed through prevention and intervention efforts. Regarding the practical activities of German P/CVE programmes, many opt for a wide array of services spanning across the prevention triad. Around 85% of German P/CVE programmes in 2018 were active in the primary prevention field, meaning they engaged in activities such as awareness raising, capacity building, or general youth work with adolescents, for example. Of the sample, 47% of programmes engaged in secondary prevention and 35% in tertiary prevention, which shows that around half of the German P/CVE landscape works with at least some form of at-risk or early-stage radicalisation and is engaged in multiple prevention fields at the same time. Deradicalisation programmes in the strict sense make up only 6% of all initiatives, indicating the legal, professional, and public complexities involved in this kind of work (Lützinger et al., 2020, pp. 605–606). Most German P/CVE programmes (77%) address people directly affected by radicalisation and extremism, in particular family members (25%). In addition, 80% of programmes also offer to support professionals (e.g. teachers, psychologists, and municipality personnel) when they are confronted with cases of radicalisation. The main methods of delivering support are training and education (48%), followed by producing informational material such as leaflets or brochures (27%). Summarising the most common types of programmes in Germany, Baaken et al. list ‘social environment support, counselling, and support for those at risk of radicalization and (partially) radicalized individuals, exit support and stabilization, and deradicalization in prisons’ (Baaken et al., 2020, p. 13). It therefore seems that at least those basic services of P/CVE are well established in Germany across ideological milieus. Two additional sources help to gain an even more detailed understanding of the methodological similarities and differences. As part of the previously mentioned MAPEX P/CVE mapping project, Waleciak (2021) assessed the specific tools and methods used by German practitioners, regardless of their ideological focus. He identified four main types of methods: (1) socio-economic (e.g. economic support, vocational training, and employment support); (2) systemic (e.g. family therapy and group therapy); (3) psycho-social (e.g. trauma therapy, anti-aggression training, and addiction treatment); and (4) ideological methods (e.g. theological and ideological discourses and debates). Furthermore, the permanent conference of the Interior Ministers (Ständige Konferenz der Innenminister und -senatoren der Länder – IMK) was tasked in 2014 to deliver a report on lessons learned from prevention and intervention targeting the extreme right and their relevance for the field of Islamist extremism. The report was published in 2016 and lists several recommended and transferable practices, as well as necessary components unique to P/CVE against Islamism. Underlying this report is the finding (based on extensive consulting with governmental and civil society practitioners) that psycho-social dynamics and pathways leading into and out of violent extremism are largely identical (Innenministerkonferenz, 2016, p. 17). Among the key ‘success factors’, the report lists (1) permanent funding to avoid any disruption of counselling processes due to a lack of resources. Here, continuity of personnel and programme structure should be essential. Furthermore, the report highlights the importance of different client contact approaches, voluntariness of participation, and systemic methods. In order to achieve such broad programme setting and competency, (2) multi-disciplinary counselling teams are pointed out as core success factors. The report especially names the expertise in language, culture, and theology of Islamist extremists, social and youth work, as well as intelligence and policing. (3) Wide spanning networks of 271

Daniel Koehler

service providers, together with case-by-case cooperation with the authorities and individually tailored mentoring, are also identified as essential success factors. Finally, (4) the report finds that a continuously monitored and adapted counselling plan that is based on the increase of the personal and social skills of the client should be a programme standard (Innenministerkonferenz, 2016, pp. 18–20). However, the authors also note significant aspects likely unique to countering Islamist extremism. Most notably, they assumed a potentially higher security risk among clients stemming from associations with international terrorist organisations such as Al-Qaeda or the Islamic State in Iraq and Syria (ISIS). Furthermore, they found it evident that programme staff would need specific knowledge and training, for example, in Islamic theology, history, and culture. Programmes were also seen to be much more likely to need to bring religious authorities into the counselling process and be able to conduct their work in different languages (e.g. Arabic and Turkish) (Innenministerkonferenz, 2016, p. 17).

Discussion and conclusion Taking into account the available evidence base from existing comparative extremism, radicalisation, and P/CVE research, as well as seemingly established prevention and intervention practices across ideological milieus in countries such as Germany, it is fair to say that a core of transferrable practices in P/CVE appears to exist. Radicalisation and deradicalisation pathways, as well as push and pull factors, appear to share many fundamental elements across the different extremist environments, as do the potential solutions. Among those practices that are generally considered to be of universal value are individually tailored interventions that include the emotional and social networks, address socio-economic needs, advance individual critical reflection, and provide alternatives to needs otherwise satisfied in extremist milieus. P/CVE programmes should be financially stable, use multi-disciplinary teams that are well trained, and tap into methods from various categories (e.g. mental health, educational, social work, creative arts and sports, and ideological) based on an individual assessment of the participant’s radicalisation profile. Among researchers and practitioners, it has become widely accepted that radicalisation and the responses to it are highly complex and context-specific phenomena that defy simple or ‘one size fits all’ solutions. In its complexity and individuality, P/CVE can nevertheless draw on methods and tools that have been established in many adjacent fields (e.g. trauma therapy, criminal desistance programmes, and debiasing tools) and have been found to be of relevance in fighting extremist radicalisation as well. As has been noted earlier (Koehler, 2017), the arguably most important transferable lesson learned from decades of research and practice in this field is to adapt interventions to individual needs based on a variety of fundamental principles, such as including pro-social networks, attaining socio-economic stability, and increasing self-efficacy. Nevertheless, counter-radicalisation interventions need to preserve sensitivity to the peculiarities of each extremist milieu and the potential need for unique approaches. This does include not only the potentially different needs and characteristics in each milieu or individual entry and exit process as listed by the German IMK report but also the societal, cultural, and political perceptions of different threats and the adequacy of certain responses. It has been demonstrated, for example, that Western public perception of terrorist attacks is significantly skewed by the ideological and ethnic background of the perpetrators (e.g. Kearns et al., 2018). Different public perceptions of the threat posed by certain extremist milieus might also produce different response priorities among policy makers. One example can be taken from the German case study as well, where left-wing extremism is significantly underrepresented in the P/CVE landscape compared to the level of violent crimes attributed to this environment (Koehler, 2021a). Maybe due to the larger acceptance 272

Responding to radicalisation with different ideological roots

of many ideological concepts and goals of this milieu in the wider population, it is seen as a lesser antithesis to overall society. Dedicated prevention and intervention directed against this environment might therefore encounter more resistance when arguing to use resources against those extremists who are seen as much more dangerous. Another caveat that must be pointed out here is the fact that even in a country like Germany, the available evidence base for the effectiveness of P/CVE methods is very limited (e.g. Kober, 2017). This holds true for the international landscape as well (Feddes & Gallucci, 2015; Jugl et al., 2020). This means that even though some practices seem to have become established and many practitioners consider certain lessons transferrable, there is little empirical evidence to support this. To put it simply, just because something has been done for a long time in different contexts does not automatically mean it is effective. While the available research briefly discussed in this chapter does suggest that some similarities exist between different extremist milieus, the missing link to transferrable P/CVE solutions is the lack of rigorous scientific evaluations, monitoring, and (quasi) experimental designs to significantly improve our understanding of what works in P/CVE, when, how, for whom, and why.

References Aichholzer, J., Danner, D., & Rammstedt, B. (2018). Facets of personality and “ideological asymmetries”. Journal of Research in Personality, 77, 90–100. Altier, M. B., Leonard Boyle, E., Shortland, N. D., & Horgan, J. G. (2017). Why they leave: An analysis of terrorist disengagement events from eighty-seven autobiographical accounts. Security Studies, 26(2), 305–332. https://doi.org/10.1080/09636412.2017.1280307 Baaken, T., Korn, J., Ruf, M., & Walkenhorst, D. (2020). Dissecting deradicalization: Challenges for theory and practice in Germany. International Journal of Conflict and Violence (IJCV), 14(2), 1–18. Bartlett, J., & Miller, C. (2010). The power of unreason: Conspiracy theories, extremism and counter-terrorism. Demos. Berger, J. M. (2018). Extremism. The MIT Press. Borum, R. (2011a). Radicalization into violent extremism I: A review of social science theories. Journal of Strategic Security, 4(4), 7–36. Borum, R. (2011b). Radicalization into violent extremism II: A review of conceptual models and empirical research. Journal of Strategic Security, 4(4), 37–62. Braddock, K. (2014). The talking cure? Communication and psychological impact in prison de-radicalisation programmes. In A. Silke (Ed.), Prisons, terrorism and extremism: Critical issues in management, radicalisation and reform (pp. 60–74). Routledge. Brown, R. A., Helmus, T. C., Ramchand, R., Palimaru, A. I., Weilant, S., Rhoades, A. L., & Hiatt, L. (2021). Violent extremism in America: Interviews with former extremists and their families on radicalization and deradicalization. RAND Corporation. Brown, T. S. (2009). Weimar radicals: Nazis and communists between authenticity and performance. Berghahn Books. Bundesregierung. (2020). Antwort der Bundesregierung auf die Kleine Anfrage der Abgeordneten Benjamin Strasser, Stephan Thomae, Grigorios Aggelidis, weiterer Abgeordneter und der Fraktion der FDP – Drucksache 19/22753-Drucksache 19/23540. https://dserver.bundestag.de/btd/19/235/1923540.pdf Dalgaard-Nielsen, A. (2010). Violent radicalization in Europe: What we know and what we do not know. Studies in Conflict & Terrorism, 33, 797–814. Dalgaard-Nielsen, A. (2013). Promoting exit from violent extremism: Themes and approaches. Studies in Conflict & Terrorism, 36(2), 99–115. https://doi.org/10.1080/1057610x.2013.747073 De Coensel, S. (2018, Winter). Processual models of radicalization into terrorism: A best fit framework synthesis. JD: Journal for Deradicalization, 19(17), 89–127. Doering, S., & Davies, G. (2019). The contextual nature of right-wing terrorism across nations. Terrorism and Political Violence, 33(5), 1071–1093. El-Amraoui, A., & Ducol, B. (2019, Fall). Family-oriented P/CVE programs: Overview, challenges and future directions. JD: Journal for Deradicalization, 19, 190–231.

273

Daniel Koehler Eysenck, H. J. (1968). The psychology of politics. Transaction Publishers. Feddes, A., & Gallucci, M. (2015, Winter). A literature review on methodology used in evaluating effects of preventive and de-radicalisation interventions. JD: Journal for Deradicalization, 16(5), 1–27. Freiheit, M., Uhl, A., & Zick, A. (2021). Phänomenübergreifende Radikalisierungsprävention – Perspektiven aus Praxis und Forschung. In MAPEX-Forschungsverbund (Ed.), Radikalisierungsprävention in Deutschland: Mapping und Analyse von Präventions- und Distanzierungsprojekten im Umgang mit islamistischer Radikalisierung (pp. 223–266). MAPEX. Gambetta, D., & Hertog, S. (2016). Engineers of Jihad: The curious connection between violent extremism and education. Princeton University Press. Gøtzsche-Astrup, O. (2018). The time for causal designs: Review and evaluation of empirical support for mechanisms of political radicalisation. Aggression and Violent Behavior, 39, 90–99. https://doi. org/10.1016/j.avb.2018.02.003 Gøtzsche‐Astrup, O., & Lindekilde, L. (2019). Either or? Reconciling findings on mental health and extremism using a dimensional rather than categorical paradigm. Journal of Forensic Sciences, 64(4), 982–988. https://doi.org/10.1111/1556-4029.14014 Horgan, J., & Braddock, K. (2010). Rehabilitating the terrorists? Challenges in assessing the effectiveness of de-radicalization programs. Terrorism and Political Violence, 22(2), 267–291. https://doi. org/10.1080/09546551003594748 Innenministerkonferenz. (2016). Bericht der Arbeitsgruppe des AK IV unter Beteiligung des AK II zur Bekämpfung des gewaltbereiten islamistischen Extremismus – Erfolgsfaktoren für Aussteigerprogramme “Gewaltbereite Islamisten”. Retrieved May  23, 2021, from www.innenministerkonferenz.de/IMK/DE/ termine/to-beschluesse/2016-11-29_30/nummer%204%20abschlussbericht%20aussteigerprogramme. pdf?__blob=publicationFile&v=4 James, M., James, P., & Yates, E. (2019). Profiles of individual radicalization in the United States – desistance, disengagement, and deradicalization (PIRUS-D3). University of Maryland, The National Consortium for the Study of Terrorism and Responses to Terrorism (START). Jugl, I., Lösel, F., Bender, D., & King, S. (2020). Psychosocial prevention programs against radicalization and extremism: A meta-analysis of outcome evaluations. European Journal of Psychology Applied to Legal Context, 13(1), 37–46. Kearns, E., Betus, A., & Lemieux, A. (2018). Why do some terrorist attacks receive more media attention than others? Justice Quarterly. http://dx.doi.org/10.2139/ssrn.2928138 Khalil, J., Horgan, J., & Zeuthen, M. (2019). The attitudes-behaviors corrective (ABC) model of violent extremism. Terrorism and Political Violence, 1–26. https://doi.org/10.1080/09546553.2019.1699793 Kober, M. (2017, Summer). Zur evaluation von massnahmen der prävention von religiöser radikalisierung in Deutschland. JD: Journal for Deradicalization, 11, 219–257. Koehler, D. (2016). Understanding deradicalization: Methods, tools and programs for countering violent extremis. Routledge. Koehler, D. (2017). How and why we should take deradicalization seriously. Nature Human Behaviour, 1, 0095. https://doi.org/10.1038/s41562-017-0095 Koehler, D. (2019). Switching sides: Exploring violent extremist intergroup migration across hostile ideologies. Political Psychology, 41(3), 499–515. https://doi.org/10.1111/pops.12633 Koehler, D. (2021a). Disengaging from left-wing terrorism and extremism: Field experiences from Germany and research gaps. Studies in Conflict  & Terrorism, 1–21. https://doi.org/10.1080/10576 10X.2021.1917639 Koehler, D. (2021b). From Traitor to Zealot: Exploring the phenomenon of side-switching in extremism and terrorism. Cambridge University Press. Kruglanski, A. W., Szumowska, E., & Kopetz, C. (2021). The call of the wild: How extremism happens. Current Directions in Psychological Science, 30(2), 181–185. https://doi.org/10.1177/0963721421992067 Kruglanski, A. W., Webber, D., & Koehler, D. (2019). The radical’s journey: How German neo-Nazis voyaged to the edge and back. Oxford University Press. Lützinger, S., Gruber, F., & Hedayat, A. (2020). Extremismuspräventionslandschaft – eine Bestandsaufnahme präventiver Angebote in Deutschland sowie ausgewählter Präventionsstrategien aus dem europäischen Ausland. In B. Ben Slama & U. Kemmesis (Eds.), Handbuch Extremismusprävention: Gesamtgesellschaftlich, Phänomenübergreifend (pp. 597–626). Bundeskriminalamt (BKA). Lützinger, S., Gruber, F., & Kemmesis, U. (2016). Extremismusprävention in Deutschland – Erhebung und Darstellung der Präventionslandschaft. Retrieved May  23, 2021, from www.bka.de/nn_205960/

274

Responding to radicalisation with different ideological roots SharedDocs/Downloads/DE/Publikationen/Publikationsreihen/SonstigeVeroeffentlichungen/2016Extrem ismuspraeventionInDeutschland,templateId=raw,property=publicationFile.pdf/2016Extremismuspraeven tionInDeutschland.pdf Meiering, D., Dziri, A., & Foroutan, N. (2020). Connecting structures: Resistance, heroic masculinity and anti-feminism as bridging narratives within group radicalization. International Journal of Conflict and Violence (IJCV), 14(2), 1–19. Misiak, B., Samochowiec, J., Bhui, K., Schouler-Ocak, M., Demunter, H., Kuey, L., Raballo, A., Gorwood, P., Frydecka, D., & Dom, G. (2019). A systematic review on the relationship between mental health, radicalization and mass violence. European Psychiatry, 56, 51–59. https://doi.org/10.1016/j.eurpsy.2018.11.005 Moskalenko, S., & McCauley, C. (2009). Measuring political mobilization: The distinction between activism and radicalism. Terrorism and Political Violence, 21(2), 239–260. Orsini, A. (2020). What everybody should know about radicalization and the DRIA model. Studies in Conflict & Terrorism, 1–33. Pfundmair, M. (2018). Ostracism promotes a terroristic mindset. Behavioral Sciences of Terrorism and Political Aggression, 11(2), 134–148. Reidy, K. (2018, Spring). Radicalization as a vector: Exploring non-violent and benevolent processes of radicalization. Journal for Deradicalization, 14, 249–294. Saleh, N. F., Roozenbeek, J. O. N., Makki, F. A., McClanahan, W. P., & Van Der Linden, S. (2021). Active inoculation boosts attitudinal resistance against extremist persuasion techniques: A novel approach towards the prevention of violent extremism. Behavioural Public Policy, 1–24. Saucier, G., Akers, L. G., Shen-Miller, S., Kneževié, G., & Stankov, L. (2009). Patterns of thinking in militant extremism. Perspectives on Psychological Science, 4(3), 256–271. Schmid, A. (2011). The literature on terrorism. In A. Schmid (Ed.), The Routledge handbook of terrorism research (pp. 457–474). Routledge. Schuurman, B. (2018). Research on terrorism, 2007–2016: A review of data, methods, and authorship. Terrorism and Political Violence, 32(5), 1011–1026. https://doi.org/10.1080/09546553.2018.1439023 Schuurman, B. (2019). Topics in terrorism research: Reviewing trends and gaps, 2007–2016. Critical Studies on Terrorism, 12(3), 463–480. https://doi.org/10.1080/17539153.2019.1579777 Sedgwick, M. (2010). The concept of radicalization as a source of confusion. Terrorism and Political Violence, 22(4), 479–494. https://doi.org/10.1080/09546553.2010.491009 Spalek, B., & Davies, L. (2012). Mentoring in relation to violent extremism: A study of role, purpose, and outcomes. Studies in Conflict & Terrorism, 35(5), 354–368. https://doi.org/10.1080/1057610x.2012.666820 Stephens, W., & Sieckelinck, S. (2021). Resiliences to radicalization: Four key perspectives. International Journal of Law, Crime and Justice, 66, 100486. Treistman, J. (2021). Social exclusion and political violence: Multilevel analysis of the justification of terrorism. Studies in Conflict & Terrorism, online first, 1–24. Van Prooijen, J.-W., & Krouwel, A. P. M. (2019). Psychological features of extreme political ideologies. Current Directions in Psychological Science, 28(2), 159–163. https://doi.org/10.1177/0963721418817755 Vasilopoulos, P., & Jost, J. T. (2020). Psychological similarities and dissimilarities between left-wing and right-wing populists: Evidence from a nationally representative survey in France. Journal of Research in Personality, 88, 104004. Waleciak, J. (2021). Die Handlungspraxis der Deradikalisierungsarbeit in Deutschland – Eine explorative systematisierung der praktischen Ansätze. In MAPEX-Forschungsverbund (Ed.), Radikalisierungsprävention in Deutschland: Mapping und Analyse von Präventions- und Distanzierungsprojekten im Umgang mit islamistischer Radikalisierung (pp. 115–142). MAPEX. Windisch, S., Ligon, G. S., & Simi, P. (2017). Organizational [dis]trust: Comparing disengagement among former left-wing and right-wing violent extremists. Studies in Conflict  & Terrorism, 42(6), 559–580. https://doi.org/10.1080/1057610X.2017.1404000 Wolfowicz, M., Litmanovitz, Y., Weisburd, D., & Hasisi, B. (2020). A field-wide systematic review and metaanalysis of putative risk and protective factors for radicalization outcomes. Journal of Quantitative Criminology, 36(3), 407–447. Zmigrod, L., & Goldenberg, A. (2021). Cognition and emotion in extreme political action: Individual differences and dynamic interactions. Current Directions in Psychological Science. https://doi. org/10.1177/0963721421993820 Zmigrod, L., Rentfrow, P. J., & Robbins, T. W. (2019). Cognitive inflexibility predicts extremist attitudes. Frontiers in Psychology, 10, 989.

275

18 WHO SHOULD BE INVOLVED WITH COUNTER-RADICALISATION POLICY AND PRACTICE? Francesco Ragazzi and Josh Walmsley

Introduction Is preventing or countering violent extremism the task of the police, intelligence services, and judges alone? Or should religious leaders – in particular those from Muslim communities – also be involved? What about teachers, social workers, and educators? Is it also their role to monitor potentially risky individuals and carry out interventions? And if so, what does it mean for a society that so many of its sectors are suddenly infused with priorities and obligations emanating from the field of security? The question of who should be involved in counter-radicalisation policy and practice appears to be a technical one but is in fact a political one; it underpins the deeply transformative nature of counter-radicalisation policies. Once governments frame the question of political violence as one of radicalisation (as seen in Chapters 2 and 3 of this volume, it is not an obvious choice) and administrative decision-making ensues, the question that occupies this chapter becomes a pressing one. And how state authorities decide who should be involved in countering radicalisation largely depends on how they answer these questions in the first place. This, in turn, opens another set of questions. Who will be planning the policy at the ministries’ headquarters and delivering it in the streets? Which administrative bodies will be used to channel budgets and lines of reporting? What professional skills are required? And whose expertise should inform these choices? In this chapter, we pursue two inter-related goals. First, we provide an overview of the main policy and scholarly approaches to counter-radicalisation. As we will demonstrate, counter-radicalisation is a moving target, and the definitions of the problem have varied over time and across countries. This, in turn, means that several different answers have been offered to the question of who should be involved in counter-radicalisation work. Yet while many authors have traced the expansion of P/CVE, there has been no substantial effort to theorise the political and bureaucratic transformations it has generated. Our second ambition is to fill this gap. We evaluate and reflect upon the choices made by governments, often justified by or aligned with the options suggested in the first strand of literature. Asking the question of why P/CVE policies have taken hold in some contexts and not others required us to go to the political and socio-institutional roots of counter-radicalisation. The scope of the chapter is therefore limited to the emergence (or not) of P/CVE policies in European DOI: 10.4324/9781003035848-21 276

Who should be involved with counter-radicalisation policy?

countries while recognising that counter-radicalisation has a global reach (Kundnani & Hayes, 2018; Badurdeen, this volume; Mwangi, this volume). Rather than a comprehensive discussion of all European states’ policies, which would inevitably lead to a more superficial overview, we chose to focus in depth on a limited number of representative cases. Drawing on sociological approaches to security studies (Bigo, 2011; Huysmans, 2006), we show that P/CVE policies are related to three interrelated process: (1) the re-balancing of struggles in the field of (in)security professionals; (2) the creation of a new field of professional practice and expertise; and (3) the social engineering of a broad number of non-security fields (academia, education, social work, and health care) to privilege a logic of security over the logics specific to these professional fields – what we call a process of heteronomisation. In conclusion, we reflect briefly on the resulting political consequences for the democratic order and the implications for doing ‘policy-relevant’ research in this area.

Shifting definitions of radicalisation and who should be in charge As several authors in this handbook have shown, counter-radicalisation policies do not appear homogeneously as a single block of security doctrine (Augestad Knudsen, this volume; Coolsaet, this volume; Thomas & Grossman, this volume). They have a specific geography and history and evolve over time (Lindekilde, 2012; Vermeulen & Bovenkerk, 2012). Most authors agree in locating their emergence in the discussions among Dutch and British intelligence services in the early 1990s, who both argued that terrorism should be understood in a new light: not only as a matter of clearly organised networks mostly located overseas (i.e. the Al-Qaeda operatives behind the 11 September 2011 attacks on the United States) but also as a set of broader, diffused, societal waves located within western countries and posing a threat from within (Coolsaet, 2010; Fadil et al., 2019). This view quickly gained traction in security and policy circles. Yet the specific definition of the problem and, consequently, the determination of who should be responsible for it remained unspecified, becoming a source of lively debate within and between a broad range of sectors, from security professionals to academic experts to political parties (European Commission’s Expert Group on Violent Radicalisation, 2008; HM Government, 2011; Schmid, 2013). In this section, we focus on the four most significant interpretations of radicalisation: (1) as solely a law-enforcement problem; (2) as mainly a social problem; (3) as mainly an ideological problem; and (4) as mainly a problem of laïcité.1 Each of these framings, as we will see, has dramatic consequences not only for who should be in charge but also for democratic issues of trust and respect for fundamental freedoms.

Radicalisation as solely a law-enforcement problem Historically, European states have considered terrorism a tactic used by political groups to advance a political agenda (Stampnitzky, 2013) and have framed it accordingly as either an act of war – in colonial and post-colonial settings, such as the conflict in Algeria or Northern-Ireland, for example (Dixon, 2012) – or as a form of (political) crime (Red Brigades, Rote Armee Fraktion) (Della Porta, 2013; Crenshaw, 2010). The violence of the groups typically determined which state bodies should respond: either the army or the criminal justice system. In some countries, such as France before 2014 or Spain or Italy today, the security actors are solely responsible for the policy response. Italy makes for an interesting case to show that counter-radicalisation policy is not an ‘inevitable’ response to ‘home-grown’ terrorism. It experienced important waves of right-wing and left-wing terror attacks in the 1970s and 1980s as well as its own share of Al-Qaeda-inspired violence. In the 2010s, 125 fighters left for Syria. Yet the only attempt at developing a dedicated 277

Francesco Ragazzi and Josh Walmsley

PVE/CVE policy failed in 2016 and never returned to the political agenda. Who, then, we might ask, is responsible for counterterrorism and counter-radicalisation in Italy2? The judicial repression of terrorism offences is coordinated by the National Anti-Mafia and Counter-terrorism Prosecution Office, a centralised judicial authority with coordination powers (Maniscalco & Rosato, 2019, p. 72). The judicial authority relies on several dedicated counter-terrorist units. Among the National Police, the Central Directorate of Prevention Police (DCPP) and the local General Investigations and Special Operations Divisions (DIGOS) units are in charge. Within the Carabinieri force, the Special Investigative Department (ROS) specialises in complex organised crime and terrorism investigations. The Committee for Financial Security (CFS) of the Minister for the Economy and Finance can intervene, and the Anti-Terrorism and Rapid Response Service (ATPI) of the Finance Corps is also available for counter-terrorism activity, supervising airports, ports, and sensitive targets (Maniscalco & Rosato, 2019, p. 77). Judicial prosecution is often preceded by intelligence work. Italian authorities rely on the External Intelligence and Security Agency (AISE) and the Internal Intelligence and Security Agency (AISI). Meanwhile, the Department of Information and Safety (RIS) collects military intelligence. As is increasingly the case in Western countries (Human Rights Watch & Sunderland, 2008; McCulloch & Wilson, 2015), Italian counter-terrorism does not follow a criminal justice itinerary but is managed through administrative measures designed to disrupt activities considered to be suspicious but which do not meet the evidentiary thresholds necessary to obtain a judicial indictment, or sometimes even to initiate an inquiry (de Goede, 2008; Heath-Kelly, 2012). On top of the list is the administrative expulsion of foreign nationals, carried out either by the Ministry of Interior or by the Prefect. Personal preventive measures, such as the compulsory removal from the municipality of residence, the ban on access to public spaces, and the obligation to appear at least twice a week at a local police station, can be imposed either by the National Police, the Carabinieri, or by local police forces (Maniscalco & Rosato, 2019, p. 82). The Finance Corps can intervene to freeze assets and restrict the financial capabilities of individuals considered suspect. In prison, guards are responsible for monitoring ‘Islamic radicalism’ and in cases of perceived radicalisation, the Central Unit for Investigations (NIC), a special Rome-based unit, can intervene. Finally, the Postal and Communications Police (part of the National Police) can intervene to remove content accessible to the public and considered illegal (Maniscalco & Rosato, 2019, p. 86). As in most European states, counterterrorism in Italy involves a broad security apparatus that extends through several branches of the security and intelligence community. The Italian case shows that a European state can have a quite successful counter-terrorist policy – Italy has not experienced terror attacks on its soil since 2003 – without adopting the society-oriented model originating in the United Kingdom and the Netherlands that has now become the European and international norm. We now turn to them in more detail.

Radicalisation a social problem The Dutch domestic intelligence agency (AIVD) was one of the first security agencies to publish a significant report mentioning the notion of ‘radicalisation’, conceived as a complex problem yet mostly related to social roots (AIVD, 2004; Coolsaet, 2010). Exclusion and discrimination determined by both ethno-religious markers and socio-economic factors were presented as the main drivers of violent ideas among certain fringe Muslim groups (Fadil et al., 2019). Following the events of 2004–5 (The murder of Theo van Gogh and London Bombings), this approach became dominant in the Netherlands and in the United Kingdom. 278

Who should be involved with counter-radicalisation policy?

The redefinition of the problem gave rise to ‘community-oriented’ approaches, putting new actors at the centre of counter-terrorism: (Muslim) communities (Ragazzi, 2016). Funding was initially aimed at social assistance programmes and intercultural dialogue, targeting increased ‘resilience’ against extremism. Projects like ‘Wij Amsterdammers’ saw imams and other Muslim ‘community leaders’ enlisted to foster intercommunity exchange and promote trust in the authorities in return for ‘prevention’ funding for cultural events, dinners, debates, and sporting activities (Vermeulen & Bovenkerk, 2012). In the United Kingdom, the Department for Communities and Local Government (DCLG) similarly led an explicitly ‘hearts and minds’ strategy, with 50,000 young Muslims participating in similar localised projects in its initial pilot year in 2007–8 (DCLG, 2008; Thomas, 2014). In both countries, this shift meant strengthening ties between the police and other public authorities (Spalek, 2010). In return for funding, certain religious representatives provided a direct line for the police in previously ‘hard to reach’ neighbourhoods, designed to facilitate new flows of intelligence about the beliefs, networks, and activities of community members (Kundnani, 2009). Meanwhile, for example, the City of Amsterdam formed an Information Centre on Radicalisation (Informatiehuishouding Radicalisering) as a hub for police, teachers, social workers, and other local actors to manage individuals flagged as at-risk (Amsterdam Municipality, 2007). Similar ‘multiagency’ work was undertaken in the United Kingdom through the Channel programme, established in 2007 by the Home Office, which continues to employ Islamic experts, ‘ex-extremists’, social workers, psychiatrists, and others with close ties to reported individuals to conduct mentoring interventions for ‘de-radicalisation’ (HM Government, 2020). By 2011, the DCLG policy was heavily criticised from the left (accused of targeting only Muslims) and from the right (accused of squandering public monies) (Birt, 2009; HM Government, 2011; Kundnani, 2009). Funding was eventually removed from the DCLG and emphasis placed on centralised national coordination via the Home Office’s Office for Security and Counter-terrorism, in an overt attempt to repair the ostensibly counter-productive effects.

An ideological problem Dissatisfied with the ‘social explanations’ at the turn of the 2010s, governments across Europe shifted to a different explanation: ideology. In 2004, the AIVD had already described the spread of a virulent ‘Salafist’ strain of radical Islam targeting the ‘re-Islamisation’ of Dutch Muslims and promoting a violent Jihad (AIVD, 2004). The UK’s 2006 CONTEST strategy, meanwhile, described ‘a battle of ideas in which success will depend upon all parts of the community challenging the ideological motivations used to justify the use of violence’ (HM Government, 2006, p.  3). This can be understood as a mainstreaming of links between radical/extremist ideas and terrorist violence first made by European security professionals in the 1970s (Bigo, 2019, p. 270). Into the 2010s, radicalisation increasingly came to be described as a virus infecting the minds of second- and third-generation European Muslims (Heath-Kelly, 2017; Thomas, 2014). Mosques and Islamic schools were the obvious points of contagion in the official narrative, but so too were the bedrooms where ‘at risk’ youth would ‘self-radicalise’ following exposure to extremist content online. As many critics pointed out, the result was to reduce radicalisation to an individual, psychological phenomenon, displacing the structural and political factors long recognised as central to political violence (Heath-Kelly, 2013; Kundnani, 2015). Today, P/CVE policies customarily target ‘right-wing’ and ‘left-wing’ extremisms, and in the United Kingdom, practitioners have widened the net further by contriving the category of ‘mixed, unstable, or unclear’ ideologies (Home Office, 2020). 279

Francesco Ragazzi and Josh Walmsley

One key dimension of the ideology-oriented approaches developed in the Netherlands and the United Kingdom is a focus on ‘counter-narratives’ for delegitimising extremist ideas. Early on, part of this work was undertaken by the police, seeking partnerships with radical but non-violent Muslim groups, as was the case with the Muslim Contact Unit in London. The task was also taken up around 2007 by ostensibly more ‘mainstream’ Muslim community leaders, NGOs, and think tanks, such as the Radical Middle Way and the Quilliam Foundation in the United Kingdom, positioned as independent initiatives to promote ‘moderate’ Islam (despite close ties to the government) (Ragazzi, 2014, pp. 32–33). Meanwhile, the Home Office’s clandestine Research Information and Communications Unit (RICU) was found to have deliberately concealed the funding sources of campaigns run by private PR firms, promoting a ‘liberal, British’ Islam as the antidote to radicalisation (Cobain et al., 2016). By the mid-2010s, the departure from community-oriented framings crystallised around the idea that radicalisation represented a unique form of ideological abuse, both emotional and psychological, inflicted on ‘at risk’ individuals (Coppock & McGovern, 2014). Following the departure to Syria of 15-year-old Shamima Begum and her two London friends in February 2015, the UK government threw its weight behind this framing to justify the Prevent Duty, introducing a legal requirement for all public professionals nationwide, including doctors and nurses, teachers, social workers, and others, to ‘safeguard’ – detect and report – individuals deemed ‘at risk’ of ‘being drawn into terrorism’ (HM Government, 2016, 2018). Despite criticism from academic and activist communities that this framing pathologises young people, denying them agency and the right to freedom of expression (Coppock & McGovern, 2014; Durodie, 2016; Rightswatch, 2016), it has been widely rolled out across public services (Busher et al., 2017; Heath-Kelly & Strausz, 2018).

A problem of laïcité The visions of the British and Dutch frontrunners were proven not to be the only viable options. In countries where ‘communities’ are officially not recognised as legitimate policy actors, several challenges appeared. For example, until 2014, France – like Italy and Spain – considered that there was no need to re-organise the public response to terrorism and that counter-radicalisation doctrines contradicted its fundamental principles of equality (Approche Française Du Phénomène De Radicalisation, 2019; cited in Bonelli & Ragazzi, 2019, p. 488). This principled stance would, however, only hold for so long. Since the 1980s, France’s counter-terrorism apparatus has traditionally relied on three exclusively ‘security’ pillars: specially mandated investigative judges, intelligence services (DST and RG), and specialised police units (SDAT and BLAT) (Bonelli, 2008; Bonelli & Ragazzi, 2014).3 In this institutional framework, imams, religious organisations, and local social work actors were absent from the picture; in official terms, it was inconceivable that they could be involved in counter-terrorism. In practice, however, the French security apparatus had been, much earlier than 2014, substantially reliant on community representatives: through interpersonal contacts, the management of religious dignitaries’ visa applications and deportations (Jouanneau, 2009), bilateral diplomatic relations with countries of origin (Bruce, 2010; Kastoryano, 2004), the governance of Islam in prisons (Béraud et al., 2013), or administrative interventions in businesses suspected of harbouring radical activists (pôles de lutte contre l’Islam radical) (Bonelli, 2008). This official approach shifted in 2012 with the perceived policy failure that led to the terror attacks committed by Mohammed Merah in Toulouse and Montauban. In October 2013, a new 280

Who should be involved with counter-radicalisation policy?

report (Jounot, 2013) suggested introducing in France the principles that had become the staple of counter-radicalisation policies abroad.4 The document suggested involving ‘non-security oriented’ actors to best detect, report, and thus anticipate this dispersed threat. Yet despite repeatedly defining the issue as a problem of ‘radical Islam’, the Dutch and British-style community-oriented approaches were considered unsuitable to the French institutional tradition because ‘in [a policy] aimed at the religious sphere, the authorities cannot intervene directly because of the principle of laïcité’ (Jounot, 2013, p. 13). The solution was to disaggregate the problem of radicalisation into three sub-questions: one of policing, one of dérive sectaire (a risk of involvement in a cult), and one of crime prevention. This meant that the Miviludes, the government agency for sectarian movements, took charge of radicalisation prevention. Similarly, all the agencies involved in crime prevention (youth and social work, probation services) were now tasked with reporting and dealing with cases of radicalisation. Finally, the problem was framed as a question of relations with religious representatives (Jounot, 2013, pp. 6–8).5 As a concrete example, in four of the cities in which we conducted fieldwork, counterradicalisation is coordinated within a multi-agency departmental monitoring unit for preventing radicalisation and supporting families (CPRAF)6 composed of the following institutions: judicial (prosecutor’s office, children’s tribunal, juvenile protection service, prison services, and probation services), education (ministry of education), child protection, welfare, and social cohesion (CAF, Préfet pour l’égalité des chances, politiques de la ville), NGOs (ADSEA), police (DGSP), and healthcare (agence régionale de santé) (Lamy & Reixache, 2015, p. 4). These units discussed individual cases, shared the names of potentially radicalised individuals, and determined the best measures to apply. Religious representatives are handled separately through meetings and instances de dialogue (forums for dialogue) and do not participate directly in coordination.

Sociological analysis of transformations While we have traced several distinct approaches as to who should be responsible for tackling radicalisation in Europe, the question remains as to why these divergences have emerged, both in time and across states. Why have large sections of the civil society and civil servant workforce been mobilised in some countries but not others? In the following paragraphs, we assess the prevailing, and ultimately unconvincing, explanations and illustrate how an alternative conceptual framework in terms of the sociology of fields can provide more explanatory value.

The puzzle of counter-radicalisation policy One common-sense explanation of the variation in counter-radicalisation policies is that governments simply react to the threats they face (Rubin et al., 2011) and that the traditional judicialadministrative counterterrorist response is not adapted to face a threat that is diffused throughout society (Neumann, 2009). Another part of the argument is to say that it is not only about anticipating possible threats before they emerge but also about finding a non-repressive – some would say liberal – solution to deal with individuals who are involved in violence but are more victims than perpetrators (Spalek & Weeks, 2017). As we have seen with the example of Italy in this chapter, it seems quite possible to deal with the issue of terrorism without developing new policies and involving non-security actors. Of course, one could retort that Italy has not faced the same level of threat. But it is questionable that the attacks of 11 March 2004 were of less intensity than the 2004 events that triggered the adoption of P/CVE measures in the Netherlands. In other words, the 281

Francesco Ragazzi and Josh Walmsley

‘reaction’ hypothesis can be a partial explanation, but does not suffice to explain the adoption and the choice of the P/CVE policy. Another possible explanation lies in shifting power in party politics. If one follows the UK example, the original version of P/CVE under Labour emphasised the social causes of radicalisation, while the revised PREVENT policy implemented by the subsequent Conservative/Liberal Democrat coalition government favoured a more hard-nosed interpretation, de-emphasising social causes and focusing on extremist ideology (HM Government, 2011). While this argument certainly bears some weight in the British context, it does not travel very well outside the United Kingdom. The various approaches to PVE adopted in several EU countries such as Belgium, France, Switzerland, Denmark, Sweden, or Norway have not followed such partisan lines (HighLevel Commission Expert Group on Radicalisation, 2017). A final argument is that the P/CVE approach is a model that fits ‘Northern European’ countries, where there are higher levels of trust in the police and the social services, which happen to also be countries following a ‘multicultural model’ with a history of managing diversity through institutions based on community representation (Ragazzi, 2022). While this argument has initial appeal as there does seem to be a North/South divide in Europe when it comes to the diffusion and implementation of counter-radicalisation policies, again it does not chime with the empirical evidence. First, one would be hard-pressed to find a single ‘Northern’ model. While there are similarities and a certain degree of coordination between those countries, there are also significant differences. Citizens in Norway and Finland would be very uncomfortable with the degree of information sharing that takes place in Denmark or the Netherlands, for example (Andersson Malmros & Sivenbring, this volume, Sivenbring, 2019). Finally, as we have seen, France, which in principle opposes multiculturalism, has ended up adopting a C/PVE model very much inspired by the United Kingdom and the Netherlands (Bonelli & Ragazzi, 2019). What, then, can provide a more accurate explanation? Drawing on the work of Didier Bigo (2008, 2011) and more broadly Bourdieusian approaches in International Political Sociology (Basaran et al., 2017), we foreground three main processes.

A few conceptual distinctions Before developing our arguments, we first need to define a few concepts. Drawing on the work of Pierre Bourdieu, Didier Bigo (2008) introduced the notion of a ‘field of (in)security professionals’ in his analysis of security politics. Bigo emphasised the sociological grounding of views and practices of security and insecurity – which he conceives as co-constituted and inseparable – hence the shortcut ‘(in)security’. According to Bigo, the adoption of such views and practices should always be understood as the outcome of struggles inside of, as well as between, various professional arenas (fields) that have common objectives. The core argument is that professionals with similar sociological characteristics, often from a similar economic background, sharing a similar education or career path, and occupying similar positions in their professional sector (such as dominated or dominant, marginal or central), will tend to align in their support and opposition for certain perspectives. Struggles can occur inside professional fields enrolled in the ‘game’ of (in)security politics, that is, between politicians and police forces, or between those fields, that is, between politicians and the police and between the police and social workers. The second important conceptual pair, for our purposes here, is what Bourdieu defines as the dual dynamic of autonomisation and heteronomisation. According to Bourdieu, struggles occurring between professional fields are struggles over the autonomy of specific professions to determine the appropriate course of action (Bourdieu, 1983). The more a field is autonomous (e.g., the 282

Who should be involved with counter-radicalisation policy?

field of theoretical mathematics), the more it is able to set its own priorities. The more a field is heteronomous (e.g., the field of journalism), the more it depends on other fields (i.e. the field of politics or the field of economics) to set its priorities. Bigo explains that the field of (in)security professionals possesses a high degree of autonomy in relation to the political field that it is hierarchically supposed to serve. This is not to deny that the work of (in)security professionals is, to a certain extent, defined by the nature of certain ‘threats’ (i.e. P/CVE emerges only after significant terrorist events), as well as political pressures to be seen as providing solutions. Rather, it acknowledges that they also possess significant autonomy in shaping dominant ideas about how those problems should be understood, as well as prioritising and defining what the other important threats on the agenda should be (i.e. road security or domestic violence). As Bigo and others have shown empirically, professionals generally do so on the basis of the sociological characteristics described earlier. And since there are multiple specialisations within the field, there are constant struggles between the various security professionals to define what should be at the top of the agenda (Bigo, 2008).

Struggles in the field of security professionals Our first argument is that the successive and haphazard adoption of counter-radicalisation policies is the result of different outcomes of struggles in the field of (in)security in different countries. The focus on these struggles, we argue, is a powerful explanation for the development of the counterradicalisation agenda by those intelligence services that are specialised in societal surveillance and boast a ‘capillary’ presence in various sectors of society, as opposed to counter-espionage intelligence services, which are more focused on individuals and networks, or judicial police with expertise in investigative practices (Bonelli & Ragazzi, 2014). This is how one can explain the very late arrival of France into the counter-radicalisation ‘club’ in Europe: the dominant actors within the field of (in)security (e.g. the secretive French counter-intelligence agency DST) were largely against what they perceived to be a threat to their monopoly over the issue of terrorism. And while other actors (such as the political intelligence agency RG or the representative of the directorate for religions in the Ministry of Interior) clearly saw an opportunity for empowerment, it was not until the government set in motion a broad institutional reform that they were able to assert themselves (Bonelli & Ragazzi, 2019). This also accounts for Italy’s aversion to P/CVE policies; while it has faced terrorism for many years, and has among the most advanced institutional setups to deal with actors that have been active in attempting to undermine state institutions, no actors were strong enough to tilt the balance towards the adoption of the EU shared agenda.

The emergence of a new field of professionals of (de)radicalisation A second key explanation for the emergence and institutionalisation of P/CVE policies is what we can describe sociologically as the birth of an ‘interstitial’ (Stampnitzky, 2013) or ‘weak’ field (Topalov, 1994; Vauchez, 2008). In Bourdieu’s understanding, fields are relatively stable configurations of actors, each with its own entry barriers (i.e. the bar exam for lawyers), its own preferred path (specific studies and training), and ultimately its own institutional recognition through official positions (Bourdieu & Wacquant, 1992). But if terrorism is defined as an entirely new problem, questioning the expertise of established professionals and transcending the existing division of labour, who should take charge of counter-radicalisation? Since there are no established academic trajectories or institutional qualifications recognising ‘radicalisation expertise’, professionals from the fields of security, social 283

Francesco Ragazzi and Josh Walmsley

work, education, or academia see the potential to become key players in a ‘new’ potential field of (de)radicalisation. As ‘interstitial’ or ‘weak’, it sits ‘in-between’ established, more autonomous fields. In it, skills and competences acquired in other fields can be ‘converted’ to achieve a better social position than one previously occupied (Vauchez, 2008). We highlight two figures to illustrate the possibilities this generates. In France, the re-framing of terrorism as radicalisation in 2015 meant that security services had to compete, in the public discourse, with a new figure: the ‘(de)radicalisation expert’. Dounia Bouzar, for example, drew on her Algerian and French-Corsican heritage, her conversion to Islam, her education as an anthropologist, and her training as a youth worker (Prévention Judiciaire de la Jeunesse) to gain a prominent position in French public discussions on radicalisation. In 2012, she created a private consulting company ‘Bouzar Expertise’, ‘specialising in the application of secularism (laïcité) and sell[ing] her services to businesses, public institutions and elected representatives’. In 2014, she co-founded the non-profit ‘Centre de prévention des dérives sectaires liées à l’Islam’ (CPDSI), offering deradicalisation services to public institutions. In 2015–6, she was omnipresent in the media landscape. Her controversial hiring of Farid Benyettou, a former associate of the Kouachi brothers condemned in 2005 for terrorist recruitment, and a subsequent inquiry into the conflicts of interests between her private sector and non-profit work led to her eventual downfall. Although presenting a quite different biography, Maajid Nawaz’s rise in the United Kingdom similarly illustrates some of the stakes in play for ‘(de)radicalisation professionals’. Nawaz broke into the public eye in 2008 as co-founder of the Quilliam Foundation, self-proclaimed as the world’s first counter-extremism think tank, months after renouncing his membership in the ‘Islamist’ group Hizb ut-Tahrir. Nawaz’s trajectory as a ‘reformed extremist’ made him the ideal poster boy for the government’s P/CVE agenda. In return for millions of pounds of Home Office funding, Nawaz became a leading voice in the PREVENT polemic, publicly fending off the concerns of Muslim communities and rights groups offering himself as living proof of the successes of deradicalisation. For a brief period, this status as a voice for a liberal Islam was lucrative, bringing book contracts, a high-profile radio show, and a political candidature. In Bourdieusian terms, we understand this as a return on the investment of a certain type of capital acquired elsewhere, namely, the perceived ‘authenticity’ that Nawaz cultivated as an activist/spokesperson for British Muslims. Throughout the 2010s, however, this authenticity capital depreciated as Nawaz struggled to shake off criticisms that Quilliam worked against Muslim communities, particularly following an alliance with far-right agitator Tommy Robinson in 2013 and a 2018 report stereotyping ‘Asian grooming gangs’. Quilliam folded in April  2021, citing COVID-19-related financial hardship, when really its funding and policy influence had waned for several years.

Taking over other fields: heteronomisation The third process key to explaining the trajectories of P/CVE policies pertains to how the field of security imposes its interests on others. While the IPS (International Political Sociology)-inspired literature has elaborated on the competition among security elites, the question of what happens when the (in)security logic is transposed onto the lifeworlds of the rank-and-file members of nonsecurity fields – a central one posed by counter-radicalisation policies – remains underexplored. Put in Bourdieu’s terms (1983), we argue that P/CVE legitimises new forms of heteronomisation of the fields adjacent to the field of (in)security, privileging the external interests of security professionals (suspicion, threat assessment, and public protection) over the autonomous interests of nonsecurity fields (i.e. the principles of doctor-patient confidentiality and ‘do no harm’ in the medical field). P/CVE proposes to subordinate the values, professional norms, and logics of non-security 284

Who should be involved with counter-radicalisation policy?

fields to a logic of intelligence gathering, such as when doctors are asked to forego confidentiality to privilege the reporting of potential extremists (Ragazzi, 2017). These heteronomising influences work through specific logics of (dis)empowerment that reorganise positions of authority in the targeted fields. We highlight some examples below. One core facet of this heteronomisation is the disempowerment of actors who represent the traditional values of non-security fields. In education, the reframing of old problems – discrimination, emotional issues, behavioural changes – as ‘radicalisation’ in the mid-2010s undermined the confidence of many teachers who questioned where religious conservatism ended and extremism began. When, for example, the Association for Citizenship Teaching (ACT) in England attempted to reassert pedagogical autonomy by producing resources to aid balanced and open discussions of ‘political extremism’, their funding was cut by the Department for Education, which favoured a neo-conservative ‘British values’ curriculum that narrowed the space for such democratic pedagogical exchanges in line with the Home Office’s PREVENT programme (Ragazzi & Walmsley, 2020). Similarly, front-line social workers and youth workers in the Netherlands have confronted the central contradiction of P/CVE: that the practice of detecting and reporting potential ‘radicals’ undermines the very trust on which counter-extremism supposedly relies, damaging the community credibility that is a unique capital in their fields and which distinguishes them from security professionals (van de Weert & Eijkman, 2019). Simultaneously, newfound legitimacy is granted to previously marginal actors who play the ‘game’ of counter-radicalisation. In schools, autonomy has shifted further from the classroom to non-teaching professionals already operating as de-facto social workers and as police contacts in schools (i.e. Principal Education Advisers in France and Designated Safeguarding Leads in England). As teachers seek to unburden themselves of the moral and practical complexity of detection, these actors inherit the primary responsibility for interpreting and externally sharing cases, assuming the authority that comes with doing high-stakes work in an interdependent relationship with the police (Ragazzi & Walmsley, 2020). In the prison system, counter-radicalisation policies have produced two phenomena in several EU countries. First, in probationary services, a profession historically endowed with a double role of surveillance and rehabilitation of potential or past offenders, a renewed emphasis on the former has led probation officers to increasingly perceive themselves as doing security (not social) work, continuing trends that preceded counter-radicalisation policy (Bouagga, 2012). Second, for prison intelligence services – a traditionally marginal service both among the carceral professions (prison guards, social workers, and probation officers) and the various national intelligence communities – counter-radicalisation policies have provided unprecedented resources, status, and ‘rights to entry’ into the field of security professionals. For example, France’s Service National du Renseignement Pénitentiaire was initially a modestly staffed office of the Penitentiary Security (Sécurité Pénitentiaire) division of Prison Administration (Direction de l’Administration Pénitentiaire) (Scheer & Chantraine, 2021). Following various counter-radicalisation plans, it progressively acquired independent national competence, a budget, and a seat at the intelligence community table.

Conclusion Throughout this chapter, we have shown that the answer to the question of who, if anyone, should be involved in counter-radicalisation is not a self-evident one. Although initially we might assume that contrasting approaches to P/CVE in European countries are explained by their different experiences of terrorism, this does not match the empirical evidence. Where we find a divergence in policy trajectories, between governments that have rejected P/CVE’s societal orientation in favour 285

Francesco Ragazzi and Josh Walmsley

of traditional counterterrorist methods and those that have embraced it (again, in different ways), each is better explained by its own political and institutional socio-history. To understand why, for example, a whole-of-society approach was embraced in the United Kingdom and the Netherlands but not in Spain or Italy, we have foregrounded three interrelated processes. First, the rebalancing of the field of (in)security professionals: the displacement of the ‘old order’, specialised in targeted operations against finite suspects, and the legitimation of actors trained in societal methods (counterinsurgency and community policing). Second, and consequently, the emergence of an interstitial field of (de)radicalisation professionals, offering new opportunities for marginal actors from more established fields to become key players. Finally, we can trace a resulting process of heteronomisation, whereby the relations, practices, and logics of non-security fields are challenged and often reframed to serve the interests of security professionals. What then are the implications of this radically transformative societal project for the liberal democratic order? While P/CVE policies vary, they ultimately share the same impetus: to instrumentalise the trust relations that underpin the social fabric in democratic societies for the purposes of social control. The logic of counter-radicalisation is to tap into the intimate relations built by community representatives, teachers, social workers, and others – the very source from which these groups derive the ‘feel’ for their communities required to provide certain services – but then convert it into a mechanism for constructing and coercively managing ‘suspect’ categories of the population (Ragazzi, 2016). This is not to suggest that the fields of healthcare, education, or social work are inherently benign; their role in regulating acceptable forms of, for example, citizenship or religious identity is well established in the literature. Yet, by refiguring trust as a tool of intelligence gathering, P/CVE proposes to all but eliminate key spaces in which these categories can be legitimately contested by rendering such contestation as threat to society and/or vulnerable individuals while disempowering actors who resist the authoritarian logic of suspicion, often at the expense of fundamental rights like freedom of expression, the right to privacy, or the best interests of the child. What does this mean for future research on counter-radicalisation? Or, more pertinently, what is the role of social scientific research in not only producing knowledge about P/CVE practices but perhaps also influencing policy in this area? This second question strikes at the heart of debates that have been ongoing in Critical Terrorism Studies (CTS) since the mid-2000s (for a synthesis, see Fitzgerald et al., 2016). What are the stakes of producing ‘policy-relevant’ research on (counter)terrorism and (counter)radicalisation? If we accept a core insight of CTS, that the dominant approaches to fighting terrorism work to uphold a racialised order of violence and exclusion that is inextricable from the very knowledge categories upon which their existence is based (i.e. ‘terrorism’ and ‘radicalisation’), how ­– if at all – can we produce scholarship that can be ‘heard’ by policymakers and practitioners without contributing to these harmful effects? Yet, at the same time, what are the costs of not producing insights that have value beyond the confines of academic conferences and journals? We would argue that the political and ethical stakes are too high for the knowledge produced in counter-radicalisation policy/practice and academic scholarship to be syphoned off into silos. While these debates are far from being resolved, there are important insights to bring to bear on future investigations into counter-radicalisation practices. In line with discussions in CTS and IPS, we argue that practices of researcher reflexivity are indispensable for navigating the testing political and ethical terrain that comes with trying to do policy-relevant social scientific work on P/ CVE. By reflexivity, here we are referring to a double move in research practice, namely, (1) abandoning ‘intuitive readings’ and ready-made classifications of our research objects (i.e. a policy), in order to construct them empirically through inductive observations, and (2) a ‘critical examination 286

Who should be involved with counter-radicalisation policy?

of the dominant academic preconstructions of the specific subject area in question, that is, a critical analysis of the research tradition and the application of that tradition by the researcher’ (Madsen, 2011, p. 262, emphasis added; see also Bourdieu et al., 1991). In other words, it requires not only for us as researchers to critically examine taken-for-granted knowledge about P/CVE practices but also to explicitly account for the fact that we – like the police officers, intelligence agencies, community representatives, social workers, teachers, and other ‘practitioners’ we may study – are active agents in the political struggle over the legitimacy of ‘societal’ approaches to terrorism. So too does it demand a recognition from the start that choices of research design and methodology are not value-neutral, as much policy-oriented research would imply. Rather, the methods we deploy are ‘instruments . . . for power struggles, competing enactments of worlds and or creative disruptive positions in the worlds of international politics’ (Aradau & Huysmans, 2014, p. 599). This has important consequences for research on P/CVE policies, two of which we highlight here. First, it requires those wishing to impact the course of policy to balance a need to be ‘legible’ beyond academia with caution about how the reliance on certain knowledge categories and assumptions can result in the co-production of counter-productive discourses on radicalisation. Part of this equation is the need for researchers to reflexively analyse their own position in the ‘game’ and the stakes of, for example, adopting one terminology over another. This includes material stakes, in terms of the research funding required to have a chance of ‘bridging the gap’ for policymakers and professionals, and ethical stakes, such as those related to the lending of scientific legitimacy to racialised categories of suspicion that contribute to harms imposed on marginalised communities that bear the brunt of security practices and technologies. As we have argued elsewhere (Bigo et al., 2014, p. 31), this means exercising the utmost caution around the terminology of radicalisation, and in particular its inherent, empirically false assumption of an atomised, linear, ‘pre-terrorist’ process towards political violence that can be foreseen, rationally predicted, and intervened with. A second implication is that, on its own, analysing the shifting discursive constructions of radicalisation and the linguistic registers that render it amenable to expert intervention is not enough. As we have shown in this chapter, the transformative effects of the radical societal project of P/ CVE owe much more to the tacit practices and biographical trajectories of social agents than to discourses on security, public protection, and welfare. As Bigo (2016, pp. 1078–1079, 2011) has argued consistently, such discourses usually operate as post facto justifications deployed by security actors to preserve or transform their social position. We would argue that while discoursebased analyses have enriched understandings of counter-radicalisation policies, successfully challenging the false orthodoxies of mainstream radicalisation scholarship, a reluctance to go beyond such approaches has led to key misrepresentations becoming embedded in ‘critical’ scholarship. In particular, in pointing to the pernicious ‘securitising’ effects of official policy discourses, it has often reproduced an image of the benign ‘left hand’ of the state (education, social work, healthcare) being colonised by its violent and repressive ‘right hand’ (police, intelligence services). Yet, as we have shown in this chapter, the imperatives of P/CVE have created a marketplace of incentives and sanctions where the configuration of winners and losers does not conform to overarching blocs of state activity or even to specific institutions (schools, hospitals). We would argue that this only becomes visible empirically through a reflexive and transdisciplinary approach to researching counter-radicalisation: one premised on reconstructing social structures of competition and collective biographies as well as analysing discursive strategies, using a range of methods including discourse analysis, qualitative interviews, ethnographic observation of practices, and prosopography. From our research in this vein, we have made several practical recommendations, such as the idea of the ‘firewall’ between police and social work and ‘grey zones’ of anonymous referrals 287

Francesco Ragazzi and Josh Walmsley

designed to protect fundamental rights (i.e. confidentiality) and professional autonomy (see Bigo et al., 2014; Ragazzi & Walmsley, 2020). Future research priorities will include mapping the field of professionals of (de)radicalisation and ethnographically examining their practices of enactment, translation, improvisation, and resistance; situated encounters between fields (i.e. ‘multi-agency’ interventions) and the re-balancing of power relations in Europe and beyond. Finally, though we did not have space to discuss it here, much remains to be understood about the role of technology in instrumentalising trust, such as ‘low-tech’ risk-assessment tools that construct the ‘warning signs’ of radicalisation.

Notes 1 Laïcité is the French principle of separation of church and state, similar in some respects to the notion of secularism. On the notion of laïcité, see Baubérot (2012) and for its relation to counter-terrorism and counter-radicalisation, see Ragazzi (2022). 2 For a detailed analysis of the legislative framework in Italy, see Maniscalco and Rosato (2019, p. 71). 3 For a detailed analysis of the evolution, see Bonelli and Ragazzi (2019). 4 For a comparative analysis of UK, French, and Dutch counter-radicalisation policies, see Ragazzi (2014). 5 For a summary of France’s various counter-radicalisation strategies and their main features, see Pawella (2019). 6 Cellule départementale de suivi pour la prévention de la radicalisation et l’accompagnement des familles (CPRAF).

References AIVD. (2004). From Dawa to Jihad: The various threats from radical Islam to the democratic legal order. General Intelligence and Security Service. Amsterdam Municipality. (2007). Amsterdam tegen radicalisering. www.politieacademie.nl/kennisenonderzoek/ kennis/mediatheek/PDF/45522.pdf Approche française du phénomène de radicalisation. (2019). Typed manuscript. Aradau, C., & Huysmans, J. (2014). Critical methods in international relations: The politics of techniques, devices and acts. European Journal of International Relations, 20(3), 596–619. https://doi. org/10.1177/1354066112474479 Basaran, T., Bigo, D., Guittet, E.-P., & Walker, R. B. J. (Eds.). (2017). Routledge studies in international political sociology: International political sociology: Transversal lines. Routledge Taylor & Francis Group. Baubérot, J. (2012). French secularism: Republican, indivisible, democratic, and social. Cites, 52(4), 11–20. Béraud, C., Galembert, C. D., & Rostaing, C. (2013). Des Hommes et des Dieux en Prison. Ministère de la Justice. Bigo, D. (2008). Globalized (in)security: The field and the ban-opticon. In D. Bigo & A. Tsoukala (Eds.), Terror, insecurity and liberty: Illiberal practices of liberal regimes after 9/11. Routledge Taylor & Francis Group. Bigo, D. (2011). Pierre Bourdieu and international relations: Power of practices, practices of power. International Political Sociology, 5(3), 225–258. Bigo, D. (2016). Rethinking security at the crossroad of international relations and criminology. The British Journal of Criminology, 56(6), 1068–1086. Bigo, D. (2019). The maze of radicalization: Justification and professional interests. In N. Fadil, M. de Koning, & F. Ragazzi (Eds.), Radicalization in Belgium and the Netherlands: Critical perspectives on violence and security (pp. 269–282). I. B. Tauris. Bigo, D., Bonelli, L., Guittet, E.-P., & Ragazzi, F. (2014). Preventing and countering youth radicalisation in the EU. European Union. Birt, Y. (2009). Promoting virulent envy? reconsidering the UK’s terrorist prevention strategy. The RUSI Journal, 4(154), 52–58. https://doi.org/10.1080/03071840903216460 Bonelli, L. (2008). Les caractéristiques de l’antiterrorisme français: “Parer les coups plutôt que panser les plaies”. In D. Bigo, L. Bonelli, & T. Deltombe (Eds.), Cahiers libres. Au nom du 11 septembre: Les démocraties à l’épreuve de l’antiterrorisme/sous la direction de Didier Bigo, Laurent Bonelli et Thomas Deltombe (pp. 168–187). Découverte.

288

Who should be involved with counter-radicalisation policy? Bonelli, L., & Ragazzi, F. (2014). Low-tech security: Files, notes, and memos as technologies of anticipation. Security Dialogue, 45(5), 476–493. https://doi.org/10.1177/0967010614545200 Bonelli, L., & Ragazzi, F. (2019). La lutte contre la “radicalisation”: Genèse et expansion d’UN nouveau lieu commun administratif en France et dans l’union européenne. Archives De Politique Criminelle, 41(1), 119. https://doi.org/10.3917/apc.041.0119 Bouagga, Y. (2012). Le métier de conseiller d’insertion et de probation: Dans les coulisses de l’État pénal? Sociologie du Travail, 54(3), 317–337. https://doi.org/10.1016/j.soctra.2012.06.003 Bourdieu, P. (1983). The field of cultural production, or: The economic world reversed. Poetics, 311–356. Bourdieu, P., Chamboredon, J. C., Passeron, J. C., & Krais, B. (1991). The craft of sociology: Epistemological preliminaries. Walter de Gruyter. Bourdieu, P., & Wacquant, L. J. D. (1992). An invitation to reflexive sociology. University of Chicago Press. Bruce, B. (2010). Promoting belonging through religious institutionalisation? The French Council of the Muslim Faith (CFCM) and the German Islamkonferenz. Political Perspectives, 4(2), 49–69. www.politicalperspectives.org.uk/ wp-content/uploads/vol-4-2-2010-Bruce.pdf Busher, J., Choudhury, T., Thomas., P., & Harris, G. (2017). What the prevent duty means for schools and colleges in England: An analysis of educationalists’ experiences. Centre for Trust, Peace, and Social Relations, Coventry University. Cobain, I., Ross, A., Evans, R., & Mahmood, M. (2016, May 2). Inside Ricu, the shadowy propaganda unit inspired by the cold war. The Guardian. www.theguardian.com/politics/2016/may/02/inside-ricu-theshadowy-propaganda-unit-inspired-by-the-cold-war Coolsaet, R. (2010). Eu counterterrorism strategy: Value added or chimera? International Affairs, 86(4), 857– 873. https://doi.org/10.1111/j.1468-2346.2010.00916.x Coppock, V., & McGovern, M. (2014). “Dangerous minds”? Deconstructing counter-terrorism discourse, radicalisation and the “psychological vulnerability” of Muslim children and young people in Britain. Children & Society, 28(3), 242–256. https://doi.org/10.1111/chso.12060 Crenshaw, M. (2010). The consequences of counterterrorism. Russell Sage Foundation Publications. http:// books.google.nl/books?id=eqrbWrjrvDAC&pg=PA158&dq=The+Psychology+of+Counter+terrorism&i e=ISO-8859-1&cd=12&source=gbs_api DCLG. (2008). Prevent pathfinder fund – mapping of project activities 2007/08. DCLG. de Goede, M. (2008). The politics of preemption and the war on terror in Europe. European Journal of International Relations, 14(1), 161–185. https://doi.org/10.1177/1354066107087764 Della Porta, D. (2013). Clandestine political violence. Cambridge University Press. http://books.google.nl/book s?id=qgAh5BNJ2DEC&pg=PA302&dq=clandestine+political+violence+inauthor:della+porta&hl=&cd= 1&source=gbs_api Dixon, P. (2012). The British approach to counterinsurgency: From Malaya and Northern Ireland to Iraq and Afghanistan. Palgrave Macmillan. http://books.google.fr/books?id=UfLqaCH4OAgC&printsec=frontcov er&dq=isbn:0230293476&hl=&cd=1&source=gbs_api Durodie, B. (2016). Securitising education to prevent terrorism or losing direction? British Journal of Educational Studies, 64(1), 21–35. https://doi.org/10.1080/00071005.2015.1107023 European Commission’s Expert Group on Violent Radicalisation. (2008). Radicalisation processes leading to acts of terrorism. European Commission’s Expert Group on Violent Radicalisation. Fadil, N., de Koning, M., & Ragazzi, F. (Eds.). (2019). Radicalization in Belgium and the Netherlands: Critical perspectives on violence and security. I. B. Tauris. Fitzgerald, J., Ali, N., & Armstrong, M. (2016). Editors’ introduction: Critical terrorism studies: Reflections on policy-relevance and disciplinarity. Critical Studies on Terrorism, 9(1), 1–11. https://doi.org/10.1080/ 17539153.2016.1151306 Heath-Kelly, C. (2012). Reinventing prevention or exposing the gap? False positives in UK terrorism governance and the quest for pre-emption. Critical Studies on Terrorism, 5(1), 69–87. https://doi.org/10.1080/1 7539153.2012.659910 Heath-Kelly, C. (2013). Counter-terrorism and the counterfactual: Producing the “radicalisation” discourse and the UK prevent strategy. The British Journal of Politics and International Relations, 15(3), 394–415. https://doi.org/10.1111/j.1467-856X.2011.00489.x Heath-Kelly, C. (2017). The geography of pre-criminal space: Epidemiological imaginations of radicalisation risk in the UK prevent strategy, 2007–2017. Critical Studies on Terrorism, 2(10), 297–319. https://doi.org/ 10.1080/17539153.2017.1327141 Heath-Kelly, C., & Strausz, E. (2018). Counter-terrorism in the NHS. University of Warwick.

289

Francesco Ragazzi and Josh Walmsley High-Level Commission Expert Group on Radicalisation. (2017). Interim report (preliminary findings and recommendations). High-Level Commission Expert Group on Radicalisation. HM Government. (2006). Countering international terrorism: The United Kingdom’s strategy. HM Government. HM Government. (2011). Report to the home secretary of independent oversight of prevent review and strategy. HM Government. HM Government. (2016). Revised prevent duty guidance for England and Wales. HM Government. HM Government. (2018). Contest 3.0. HM Government. HM Government. (2020). Channel duty guidance: Protecting people vulnerable to being drawn into terrorism. HM Government. Home Office. (2020). Individuals referred to and supported through the prevent programme: England and Wales, April 2019 to March 2020. Home Office. Human Rights Watch, & Sunderland. (2008). Preempting justice: Counterterrorism laws and procedures in France. Human Rights Watch. http://books.google.com/books?hl=en&lr=&id=2aLAvtKhAMwC&oi =fnd&pg=PP1&dq=Counterterrorism+Laws+and+Procedures+in+France&ots=yTGSB7x6FK&sig=otR jJE_mDMJRUMC9AuKFrPyMns0 Huysmans, J. (Ed.). (2006). The politics of insecurity: Fear, migration and asylum in the EU. Routledge. Jouanneau, S. (2009). Régulariser ou non UN imam étranger en France: Droit au séjour et définition du “bon imam” en pays laïque. Politix, 2, 147–166. Jounot, Y. (2013). Prévention de la radicalisation: Rapport, Rédigé par Yann Jounot, Directeur de la protection et de la sécurité de l’Etat, au titre du SGDSN. https://static.mediapart.fr/files/2016/01/12/295128767rapport-jounot.pdf Kastoryano, R. (2004). Religion and incorporation: Islam in France and Germany. International Migration Review, 38(3), 1234–1255. Kundnani, A. (2009). Spooked! How not to prevent violent extremism. www.irr.org.uk/pdf2/spooked.pdf Kundnani, A. (2015). The Muslims are coming! Islamophobia, extremism, and the domestic war on terror (Paperback ed.). Verso. Kundnani, A., & Hayes, B. (2018). The globalisation of countering violent extremism policies. The Transnational Institute. Lamy, L., & Reixache, C. (2015, April 5). Interview by A. Piettre [Audio recording, transcribed]. Pantin. Lindekilde, L. (2012). Introduction: Assessing the effectiveness of counter-radicalisation policies in Northwestern Europe. Critical Studies on Terrorism, 5(3), 335–344. https://doi.org/10.1080/17539153.2012.7 23522 Madsen, M. R. (2011). Reflexivity and the construction of the international object: The case of human rights. International Political Sociology, 5(3), 259–275. https://doi.org/10.1111/j.1749-5687.2011.00133.x Maniscalco, M. L., & Rosato, V. (Eds.). (2019). Preventing radicalisation and terrorism in Europe: A comparative analysis of policies. Cambridge Scholars Publishing. McCulloch, J., & Wilson, D. (2015). Pre-crime: Pre-emption, precaution and the future (Routledge frontiers of criminal justice: Vol. 28, 1st ed.). Routledge. Neumann, P. (2009). Old and new terrorism: Late modernity, globalization and the transformation of political violence. Polity. Pawella, J. (2019). Plans d’action contre la radicalisation & le terrorisme * France * 2014–2018. https:// radical.hypotheses.org/13661 Ragazzi, F. (2014). Towards “policed multiculturalism”? Counter-radicalization in France, the Netherlands and the United Kingdom. Etudes du CERI. Ragazzi, F. (2016). Suspect community or suspect category? The impact of counter-terrorism as “policed multiculturalism”. Journal of Ethnic and Migration Studies, 42(5), 724–741. https://doi.org/10.1080/136 9183X.2015.1121807 Ragazzi, F. (2017). Countering terrorism and radicalisation: Securitising social policy? Critical Social Policy, 37(2), 163–179. https://doi.org/10.1177/0261018316683472 Ragazzi, F. (2022). Counter-Radicalisation, Islam and Laïcité: Policed multiculturalism in France’s Banlieues. Ethnic and Racial Studies, 46 (exact ref soon known). Ragazzi, F., & Walmsley, J. (2020). Counter-radicalisation in the classroom: Insights from eight grassroots projects in the Council of Europe member states. Council of Europe. Rightswatch. (2016). Preventing education? Human rights and UK counter-terrorism policy in schools. Rightswatch.

290

Who should be involved with counter-radicalisation policy? Rubin, L., Gunaratna, R., & Jerard, J. (2011). Terrorist rehabilitation and counter-radicalisation. Routledge. https://doi.org/10.4324/9780203830482 Scheer, D. H., & Chantraine, G. (2021). Intelligence and radicalization in French prisons: Sociological analysis bottom-up. Security Dialogue. https://doi.org/10.1177/09670106211004824 Schmid, A. P. (2013, March). Radicalisation, de-radicalisation, counter-radicalisation: A conceptual discussion and literature review. International Centre for Counter-terrorism. Sivenbring, J. (2019). Democratic dilemmas in education against violent extremism. In J. Lunneblad (Ed.), Policing schools: School violence and the juridification of youth (Vol. 2, pp. 173–186). Springer International Publishing. https://doi.org/10.1007/978-3-030-18605-0 Spalek, B. (2010). Community policing, trust, and Muslim communities in relation to “new terrorism”. Politics & Policy, 38(4), 789–815. https://doi.org/10.1111/j.1747-1346.2010.00258.x Spalek, B., & Weeks, D. (2017). The role of communities in counterterrorism: Analyzing policy and exploring psychotherapeutic approaches within community settings. Studies in Conflict  & Terrorism, 40(12), 991–1003. https://doi.org/10.1080/1057610X.2016.1253985 Stampnitzky, L. (2013). Disciplining terror: How experts invented “terrorism”. Cambridge University Press. Thomas, P. (2014). Divorced but still co-habiting? Britain’s prevent/community cohesion policy tension. British Politics, 9(4), 472–493. www.palgrave-journals.com/bp/journal/v9/n4/full/bp201416a.html Topalov, C. (1994). Le champ réformateur, 1880–1914: UN modèle (pp. 1–15). Les Éditions de l’EHESS. van de Weert, A., & Eijkman, Q. A. (2019). Subjectivity in detection of radicalisation and violent extremism: A youth worker’s perspective. Behavioral Sciences of Terrorism and Political Aggression, 11(3), 191–214. https://doi.org/10.1080/19434472.2018.1457069 Vauchez, A. (2008). The force of a weak field: Law and lawyers in the government of the European Union (for a renewed research agenda). International Political Sociology, 2(2), 128–144. http://onlinelibrary.wiley. com/doi/10.1111/j.1749-5687.2008.00040.x/full Vermeulen, F., & Bovenkerk, F. (2012). Engaging with violent Islamic extremism: Local policies in Western European cities. Eleven International Publishing.

291

19 WORKING WITH COMMUNITIES TO COUNTER RADICALISATION Paul Thomas and Michele Grossman

Introduction This chapter explores how international policy efforts to counter radicalisation across Western states have viewed, approached, and interacted with communities and community organisations on preventing and countering violent extremism (P/CVE). After first reviewing the broader international policy contexts in which community engagement in countering radicalisation has been foregrounded, we define the scope of what ‘community’ means and how ‘communities’ have been constructed in the context of P/CVE theory, policy, and practice. Third, we review what we know about how and why relationships between communities and the state can actually be effective in preventing radicalisation. Finally, drawing on examples from the United Kingdom, Australia, North America, and European Union states, we reflect on the evolving strengths and weaknesses of P/CVE state-community policy relationships and outcomes, the extent to which research and policy have engaged with and informed each other over time, and whether this focus on ‘community’ can be effective in the face of a growing extreme right-wing (XRW) threat and the increasing centrality of digital media to terrorist organisations.

The policy landscape of community involvement in P/CVE Following 9/11, a range of national P/CVE policy strategies identified ‘communities’ as a cornerstone of efforts to broaden the policy scope from a primarily securitised agenda to a ‘whole of society’ effort (Rosand, 2017; Levitt, 2017; European Commission, 2020). This evolving approach to understanding communities as critical partners in P/CVE rests on several insights. These insights include an understanding that local communities often have the earliest intelligence around people who may be radicalising, or about developing plots and attacks, that community-based services and social networks are often best positioned to intervene in diverting or disengaging those on violent extremist pathways, and that there are limits to the more security-oriented logics (Grossman, 2018) of dealing with terrorism, which are not well equipped to address factors such as the role of social influence and support. The observation that it is impossible to simply arrest or censor our way out of violent extremist ideology and action (Holt et al., 2017) reflects increasing awareness of the critical role played by

DOI: 10.4324/9781003035848-22 292

Working with communities to counter radicalisation

social processes that enable and support radicalisation to violence. Yet if radicalisation to violence is a social process, then it follows that the process of preventing or countering radicalisation must also draw on social energies and capital (Day & Kleinmann, 2017). It is in this context that the role of communities in P/CVE policy and practice has emerged since 9/11. Many national and international counter-radicalisation policy strategies stress the important role of ‘communities’. For instance, the UK’s CONTEST strategy (Her Majesty’s Government, 2018, pp. 9, 33) promises to ‘prioritise strengthening the resilience of local communities to terrorism’ and to ‘build stronger partnerships with communities’. Australia has had a long-standing focus on enhancing community engagement and social cohesion as a CVE tool, beginning with the National Action Plan to Build on Social Cohesion, Harmony, and Security in 2006 as part of the country’s national counter-terrorism strategic framework, and has continued to focus strongly on promoting resilience to violent extremism as part of a broader effort to create more generally resilient communities that can successfully manage a range of 21st-century dynamics, challenges, and transformations (Grossman et al., 2016; Department of Home Affairs, n.d.). In the United States, the 2016 Strategic Implementation Plan for Empowering Local Partners to Prevent Violent Extremism asserts that ‘strong and resilient local communities are the most effective means of safeguarding individuals in the United States against violent extremist recruitment and radicalization’ (Executive Office of the President of the United States, 2016, p. 1). Similarly, international bodies foreground communities and those who work with them; for example, the Global Counterterrorism Forum (GCTF) has prioritised the engaging of communities regarding violent extremist risks and threats, arguing that ‘community engagement and community-oriented policing’ help to build trust and foster engagement with community partners in order to ‘raise community awareness about the threat of violent extremism, to provide them with the necessary tools, and to empower them to intervene and prevent radicalization and violence’ (Global Counterterrorism Forum, 2013, p. 1). This emphasis stems in part from evidence of the distinctive social capital that communities can bring to P/CVE models and initiatives, coupled with ‘growing awareness among policy-makers, practitioners, and experts that locally-rooted actors, such as civil society organisations (CSOs) have huge comparative advantages when it comes to P/CVE’ (Rosand et al., 2018, p. 1). The relevant social capital of community actors can include local knowledge and expertise in identifying and addressing drivers of individual involvement in extremism, greater trust and credibility in comparison to local and national state bodies, and being more flexible and adaptive than the state. These capacities can span both the prevention and disruption of actual extremist recruitment and activity, as Huq (2017, p. 1041) observes: ‘Community-level counterterrorism focuses on the possibility that private individuals – typically enmeshed in a social network that includes recruits for terrorism – can take steps to reduce the expected success rates of recruiting tactics’. P/CVE policies have frequently identified that such community capital may need to be further bolstered, particularly among socially marginalised Muslim communities, by the further strengthening of domestic civil society (Thomas, 2012) and the linking capital conferred by deepening connections communities have with external state organisations (Vermeulen & Bovenkerk, 2012). However, the recognition of community capital seemingly conferred by P/CVE policymakers’ enthusiasm for engaging with communities has also generated sustained critique across various Western countries, particularly in relation to how P/CVE conceptualises and interacts with relationships between Muslim communities and the state. For the most part, these criticisms have involved the way in which P/CVE policy and practice frameworks have overwhelmingly identified Muslim communities as their key engagement

293

Paul Thomas and Michele Grossman

targets. Until recently, the perception that the key terrorist threat was generated first by Al-Qaeda and its global affiliates, then by the emergence of Islamic State, meant that ‘community’ was most often defined as visibly racialised and often spatially concentrated Muslim minority communities within broader Western societies. Simultaneously solicited as ‘partners’ by governments and blamed as ‘enablers’ in media (and sometimes political) discourse, state-based policy attempts in various Western societies to engage with Muslim communities have led to persistent, high-profile accusations that such engagement merely camouflaged the covert surveillance and securitisation of Muslim communities perceived as a threat to national security (Kundnani, 2014). Both Muslim and non-Muslim critics have argued that these communities are seen by their own governments as the new subjects of Hillyard’s (1993) ‘suspect communities’ thesis, a concept first applied to Irish-origin communities in Britain and the securitised focus on them during the ‘Troubles’ (Ragazzi, 2016). Alongside this, they suggest, Muslim communities have been ‘responsibilised’ (Thomas, 2017), given prime, public responsibility by national states and their P/CVE policies as agents for both preventing and building resilience to terrorism (Grossman, 2021) at the community level. Such accusations are underpinned by broader scepticism about whether ‘whole of society’ P/CVE approaches that overtly address entire ‘communities’ as a unit of policy concern can ever avoid stigmatising, securitising, and damaging state-community relationships, as we detail below. This is particularly the case when the risks and threats being addressed – however deadly their impacts may be – involve very small numbers of individuals who radicalise to violence. Berger (2016, p. 8) has critiqued the ‘whole of society’ model of community engagement on the basis of its ‘unreasonably wide scope of activity’, leading to unproductive outcomes and unintended consequences. He argues that the focus of P/CVE efforts should instead concentrate on those who are actively engaged in radicalised activity and exclude ‘purely preventative activities . . . seeking to “inoculate” communities against extremism’ (Berger, 2016, p. 3). Moreover, the drive to target entire communities as both ‘suspect’ enablers of and potential ‘partners’ in countering radicalisation has frequently seen Muslim communities constructed as homogenous rather than heterogeneous entities, ‘imagining’ (Anderson, 1983) into being a unitary, monolithic ‘Muslim community’ (Malik, 2009) that belies their considerable diversity and contestation in various national settings. One impact of this has been the perception that P/CVE is something done ‘to’ Muslim communities rather than with them, a model that neither invests in longer-term community-based solutions (Hardy, 2018) that respond to genuine community needs nor acknowledges community agency and existing resources in addressing radicalisation to violent extremism. Yet the question of how ‘communities’ are defined and navigated by P/CVE policy and practice has now become further complicated by the growing threat of XRW terrorism. How can and should policy engagement with ‘communities’ be understood and deployed in settings where terrorist actors are from majority rather than minority ethnic/faith backgrounds? When their social networks and trajectories towards violence are occurring digitally as communities of interest or identity rather than as spatially bounded collectives? What challenges does this create for the orientation of existing P/CVE community-based policy frameworks towards temporally and spatially situated responses rather than ‘mainstreamed’, sustainable approaches equipped to meet the complexities of how both community and communication are transforming?

Sociological approaches to conceptualising ‘community’ These questions raise the more fundamental issue of how P/CVE thinking has conceptualised and defined ‘community’ to date and how and by whom such communities have been ontologically and instrumentally ‘imagined’ into being (Anderson, 1983). 294

Working with communities to counter radicalisation

Sociologically, communities are understood in many ways, from ‘natural’ or local spatial/geographic communities to interest or ‘identity’ communities based on faith, ethnicity, sexual identity, or other in-group/out-group dynamics that may not be spatially proximate. Sub-cultural communities, such as skinhead movements, goths, or football fans, may similarly be geographically anchored only partially or not at all. In attempting to define ‘community’ in relation to extremist recruitment and activity, the varying spatial reality of sub-cultural communities has generated use of the term ‘milieu’ (Malthaner & Waldmann, 2014), a useful way of describing how online communities and networks develop and move between online and offline interactions. Such spatial or sub-cultural identifications link in turn to broader, superordinate ‘imagined communities’ (Anderson, 1983) of nation, ethnicity, or faith. In this sense, communities are not just ‘imagined’ by those who create and belong to such formations but also by national policymakers who identify and target work with these imagined communities in ways that move beyond the local or geographically defined. The meanings of ‘community’ in P/CVE contexts have at times been applied spatially, for example, in geographical areas like local suburbs or neighbourhoods; ethnoculturally, for example, among faith-based or ethnic formations, or relationally, in terms of a sense of belonging and psychosocial group affiliation. One of the distinguishing features of many diaspora Muslim communities in the West is the way in which they constitute ‘lived communities’, in which geographical and socio-spatial proximities converge. This comes both through the organic congregation stage that most migrant-background communities experience (Scheffer, 2011) as they navigate social belonging in a new country and also from the socially imposed constraints of racial discrimination and poverty that limit or concentrate demographic distribution. This has in turn shaped how P/CVE policy has conceptualised, operationalised, and enacted its engagement with the civil society infrastructure of minority, migrant-background communities. This engagement has been not only prosocial and overt but also anti-social and covert. Examples of prosocial engagement include various efforts to partner with community organisations and groups through ‘Community Support Group’ programmes in Australian capital cities or Prevent funding of local Muslim groups in the United Kingdom. Covert initiatives have included the 2010 spy camera controversy in Birmingham, the United Kingdom, where majority Muslim community areas were ringed by surveillance cameras funded by counter-terrorism (Thomas, 2012), and the sustained NYPD covert community surveillance and infiltration operating in New York City mosques, restaurants, schools, and community groups in the early to mid-2000s (Apuzzo & Goldman, 2013). Allied to this is the significant number of Western states, including Australia, Canada, the Netherlands, and the United Kingdom, which have enacted broader forms of ‘multicultural’ state policy approaches (Modood, 2013) to identify minority ethnic or faith communities and their nominal ‘leaders’. In these national settings, ‘multiculturalism’ and social cohesion have provided a pre-existing policy context for defining and operationalising ‘community’ in relation to P/CVE implementation. Such multiculturalist policy approaches, even when termed otherwise in response to backlash (Thomas, 2011), acknowledge and work with understandings of the specific characteristics and needs of distinct ethnic or faith ‘communities’ within broader society. As such, they provide both policy-based recognition of distinct identities and a vehicle for addressing real and substantial ethnic societal disadvantages and discrimination. Such framing has, however, generated charges of ‘essentialising’ ethnic minority communities as homogenous and static entities with uniform interests and needs dictated by the primacy of their ethnic or faith background (Malik, 2009). Bauman (2001, p. 1) identifies the increasing focus on ‘community’ in the broader policymaking discourse of modern Western societies but observes that a natural, whole-world community, or ‘Gemeinschaft’, has not actually existed in such countries for a long time. Instead, modern 295

Paul Thomas and Michele Grossman

concepts of community must be consciously and continually remade, underpinned by societal and individual tensions between desires for the security of ‘community’ and the freedom of individual agency. More particularly, ‘community’ has become a crucial element of governance. Arguably, societal racism, the state multiculturalist policy approaches discussed earlier, and community responses to both – including backlash against efforts to remediate racism, disadvantage, and discrimination – have significantly shaped an idea of the ‘communities’ that P/CVE has sought to engage with. Ragazzi (2016, p. 729) identifies this Weberian understanding of community in his assessment of how P/CVE frames communities, saying ‘the community does not exist as such, it is a product of political processes of identification and categorisation’. On the one hand, this framework argues the experience of ‘community’ may develop internally in response to external forces such as state policies through active social and political interaction as well as internal conflict, producing different or contested versions of communities in the plural. On the other hand, advocates for the ‘suspect community’ thesis argue that P/CVE approaches in many countries have reinforced the creation of a single unified (Muslim) community through a common experience of societal discrimination and securitised P/CVE suspicion. This accords with the Durkheimian view of community as naturally existing structures of commonality and shared life that shape individual perspectives. These differing conceptions of ‘community’ speak directly to how communities are conceptualised, identified, and prioritised, who ‘leads’ and speaks for communities, and the impacts for communities of engaging with and contributing to P/CVE state policies. New challenges around conceptualising and defining ‘communities’ are also emerging as problematic in the context of growing XRW threats, albeit in different ways and for different reasons than those associated with the construct of ‘Muslim communities’. These problems include what White (2020, p. 77) identifies as ‘the disparate nature of “community” often associated with the extreme right wing’. They also include the massive shift towards digital communities of interest that are now more fragmented, fluid, and porous (Peucker & Smith, 2019), more loosely structured and volatile in terms of collective identities and group formations (Centre for Resilient and Inclusive Societies, 2021), and in some cases more indebted to multiple and even contradictory ideological and motivational vectors (Baldino, 2021). While some analyses suggest that XRW growth emerges from de-industrialised, ‘White working class’ areas that are increasingly economically and socially marginalised within progressively multi-ethnic societies, others believe it is ‘lazy to equate the radical right solely with White, working class communities’ (Arbuthnot, 2021). The emergence of XRW extremism has created new pressures on how Western P/CVE frameworks understand and operationalise the meanings of ‘community’, and on how they envisage the prospects for building ‘community resilience’ to violent extremism when the ‘communities’ in question derive from a (White) majority rather than a minority identity group. This suggests that policymakers need to rethink their foregrounding of community both conceptually and practically within P/CVE policies.

Building community resilience against violent extremism Many national P/CVE frameworks have centred their efforts on strategies designed to build resilience to violent extremism.1 This national policy focus has been reinforced by multilateral organisations including the European Commission’s Radicalisation Awareness Network (RAN), the Global Community Engagement and Resilience Fund (GCERF), and the GCTF, all of which have developed community resilience-oriented models to counteract radicalisation to terrorist ideology and action in a variety of settings and contexts. 296

Working with communities to counter radicalisation

Scholars have noted the varying ways in which ‘resilience’ is defined in PVE policy. Stephens and Sieckelinck (2020) discern two key trends in the use of resilience as a metaphor in relation to resilience as a ‘shield’ and resilience as ‘connection’; similarly, Grossman (2021) identifies resistance, prevention, adaptation, and recovery as the four conceptual pillars by which the deployment of P/CVE community resilience is anchored. There have been continuing critiques of the imprecise ways in which ‘resilience’ has been mobilised as a concept in P/CVE discourse, with some arguing that its imprecision and contested meanings as a term have resulted in ‘community resilience to violent extremism’ becoming little more than a ‘weak byword for P/CVE-think’ (Grossman, 2021, p. 297). Notwithstanding such critiques, efforts to develop community-based or focused resilience to violent extremism have been pursued vigorously over time, in particular by drawing on public health models (Ellis & Abdi, 2017; Weine et al., 2017), although disaster and crisis recovery models have also informed some thinking around community programmes and responses (Ranstorp, 2018). While Coaffee (2006, p.  397) has argued that the focus on building community resilience to radicalisation was designed to help policymakers be more adaptable to an increasingly volatile and unpredictable threat landscape, it has also been asserted that the policy drive towards resilience-building reflected a desire by governments to move towards a more prosocial framework of engagement that deflected, without entirely repudiating, the securitised logics of more conventional counter-terrorism approaches to interacting with community stakeholders (Hardy, 2015). Similarly, Grossman (2021, p. 296) has argued that the developing P/CVE community resilience agenda in the late 2000s reflected in part the strategic effort by government and law enforcement agencies to make P/CVE ‘more palatable and less threatening to communities (and Muslim communities in particular) who felt chronically stigmatised and besieged by conventional counterterrorism discourse’. However, such efforts did not entirely dispel community perceptions that the ‘resilience’ narrative merely camouflaged more securitised agendas that still emphasised the monitoring and surveillance of communities under a ‘resilience’ rubric. The result was that governmentled community resilience strategies came to be perceived by many in Muslim communities as proxies for government efforts to secure and control communities rather than building genuine community awareness of and resistance to social harms and threats from violent extremism (Hardy, 2015; Council on American-Islamic Relations Minnesota, 2016). Grossman (2021, pp. 299–300) highlights the prevalence of focusing on community-level resilience to violent extremism compared to individual-level resilience within the P/CVE field, suggesting this prevalence reflects four continuing emphases in ‘how P/CVE analysts think about community resilience: 1. The extent to which terrorist trajectories themselves have been conceptualised as group-level rather than individual-level processes, involving an understanding of individuals who radicalise to violence as embedded within group-level socio-ideological processes and networks of various kinds and to various degrees; 2. Following from this, the extent to which social-ecological paradigms of resilience, which stress the complex interdependency between individuals and their collective social systems, have resonated most strongly in P/CVE thinking to date; 3. The responsiveness of CT and P/CVE scholars to the needs of policymakers and security agencies, which have been interested in the economies of scale and impact offered by building collective resilience to violent extremism through programming, planning and resourcing by governments, and 297

Paul Thomas and Michele Grossman

4. The tendency to attribute violent extremist ideologies and behaviours to communal identity structures (e.g., Muslims, Whites, men) rather than to communal ideological or belief structures (e.g., right-wing, Islamist, ecological, misogynist)’. The final point illustrates some of the challenges involved in policies targeting community-level resilience. First, there is an uneven awareness that while community resilience to various social harms may be built or strengthened, it can also be eroded if systemic risk factors overwhelm the protective factors conferred by existing community (and government) social, cultural, and economic capital. Second, targeting specific identity-based communities for P/CVE resiliencebuilding initiatives assumes a deficit of resilience in these communities rather than an effort to map existing resilience assets that can be drawn on in developing genuine partnerships in relation to counter-radicalisation efforts. Third, as we will see later, P/CVE resilience models tend to disproportionately responsibilise such communities by implying that they are not ‘resilient enough’ as a community when radicalisation threats emerge. Fourth, some policies designed to build resilience to violent extremism can actually create new adversities that undercut the very resilience they are trying to promote: for example, by stigmatising particular groups, unevenly distributing resources, or failing to implement safeguards against concerns around community monitoring and surveillance under the guise of resilience-building initiatives. A final critique of existing P/CVE community resilience paradigms is their reliance on creating and sustaining a construct of out-group communities that require integration and strengthening, with comparatively little attention paid to the resilience of those belonging to a society’s in-group. Stephens and Sieckelinck (2020) argue that this tends to elevate the status quo of a society’s mainstream norms, values, and behaviours to a level that excludes these from critique or scrutiny in relation to majority-culture individual and systemic resilience to violent extremism – an observation thrown into sharp relief by the intensified emergence of XRW in various Western democracies. The same may be said for governments themselves, which for the most part see resilience as a one-way trajectory focused on communities, without examining whether their own systems are as resilient as they need to be (Grossman, 2021).

Responsibilised or ‘suspect’ communities? In placing communities at the forefront of P/CVE efforts in different national settings, governments are arguably ‘responsibilising’ them (Thomas, 2017), making communities responsible for terrorism prevention and for disrupting terrorist recruitment within ‘their’ communities. It is beyond dispute that P/CVE has foregrounded and ‘responsibilised’ communities and their CSOs, whether through the USA’s CVE programme (Salyk-Virk, 2020) or the ‘Prevent 1’ phase of the UK’s Prevent programme, which prioritised funding local government and CSOs in areas of greatest Muslim populations (Thomas, 2012). Responsibilisation is a communitarian-aligned approach that has become increasingly prominent in modern public policy. Etzioni (1995) identifies it as a necessary re-balancing of rights and responsibilities within modern welfare societies, where individuals and communities expect the state to address and solve all social challenges. Highly influential in the ‘third way’ politics of Clinton in the United States and New Labour in the United Kingdom, responsibilisation makes significant assumptions about the ability of individuals and communities to impact the structural drivers of social problems. For Giddens (1998), it reflects the reality of the ‘risk society’ and the government’s increasingly limited ability to control social change. Viewed through this broader public policy lens, responsibilising Muslim communities for undertaking P/CVE roles in partnership with the state is arguably better than doing policy for them – an 298

Working with communities to counter radicalisation

avoidance of the ‘suspect communities’ approach discussed above and below. Alongside this, P/ CVE responsibilisation for communities has arguably represented governmental recognition of their own limited understanding of and influence upon dynamics and processes within communities, thereby necessitating the ‘civilianisation’ of security (Sliwinski, 2012). The P/CVE approach of making responsibilisation a key element of working with communities has been a new direction within the longer history of terrorism prevention (Spalek, 2013), bringing its own problems and challenges. As noted earlier, a persistent allegation in relation to communitylevel P/CVE policy efforts has been that they represent and operationalise the concept of ‘suspect communities’. Originally coined by Hillyard (1993) to describe the way Irish-origin communities living in Britain during the Northern Ireland Troubles were both surveilled and discriminated against under the ‘Prevention of Terrorism’ Acts, this concept has since been applied (and highly contested; see Ragazzi, 2016) to Muslim minority communities targeted by P/CVE policies. Thomas (2012) details how the initial iteration of Britain’s Prevent programme was explicitly and exclusively about Muslims and enacted on a very large scale, aiming to produce ‘demonstrable changes in attitudes among Muslims’ (DCLG, 2007, p. 7). Early Prevent documents talked of ‘winning hearts and minds’, a phrase straight from neo-colonial counter-insurgency approaches, suggesting that entire communities were the target. Such perspectives have been reinforced by evidence of police and security agencies overtly directing P/CVE programmes and accusations of ‘spying’ within such programmes (Kundnani, 2014; Apuzzo & Goldman, 2013) and perceptions that police have seen CSOs as ‘information conduits’ (Huq, 2017; Spalek, 2013). This analysis has been problematised, however, by several factors. One has been the extension of P/CVE in a number of countries to different threats beyond AQ/IS-inspired terrorism (Her Majesty’s Government, 2018). Alongside this has been the significant number of Muslim CSOs willing to engage with and accept funding from P/CVE programmes and the number of Muslim-background individuals active in P/CVE work. For Ragazzi (2016), these factors support a more nuanced concept of ‘suspect categories’ than is countenanced by the ‘suspect community’ thesis, which indicates the insidious effects of Muslims involved ‘in their own policing’ (ibid. 724). The ‘suspect community’ thesis fails to account for Muslim community agency and for contestation between different conceptions of locally ‘imagined community’ so that P/CVE becomes ‘a specific form of “government through community”, based on a differentialist approach to suspicion and recognition, characterised by the blurring of two sectors of government activity: that of policing and that of traditional functions of the multicultural welfare state’ (ibid. 732). In some Western countries, this has arguably constructed P/CVE as a form of ‘policed multiculturalism’ that is enacted through struggles between different types of (Bourdieusian) professionals and their fields and through differential categorisations of suspicion aimed at elements within communities (Ragazzi, 2016). As noted earlier, there has been vociferous anti-P/CVE campaigning in several countries by critics regarding the complicity of P/CVE policies in securitising and surveilling entire Muslim communities. Yet it is questionable how representative of broader Muslim opinion such claims are. For instance, recent survey evidence in the United Kingdom highlights that most British Muslims have not even heard of the Prevent programme. . . . When it is communicated to them, British Muslims support, in principle, the rationale for a safeguarding programme, even when it is mostly targeted towards their own community. (CREST Advisory, 2020, p. 7) This highlights the significant gap between external political perceptions and internal realities of opinions over P/CVE activity within Muslim communities, with British Muslims trusting police 299

Paul Thomas and Michele Grossman

more to counter extremism and tension than to tackle acquisitive crimes. This finding is consistent with evidence from other robust surveys of British Muslim community attitudes towards P/CVE policy efforts. Directly challenging the ‘alienation thesis’, this analysis identifies that ‘although a minority show signs of alienation, most British Muslims are satisfied with and trust counterterrorism policies as well as the government and the police’ (Shanaah, 2019, p. 1). Clearly, though, such community perceptions of and involvement in P/CVE activity will be impacted by experiences of being part of larger, superordinate national communities. While drivers of pathways towards individual radicalisation are contested, perceptions of societal discrimination and denial of equal access to citizenship are seen as significant causal factors in community alienation. Islamophobia/anti-Muslim racism can block the development of shared national identity (Ellis & Abdi, 2017), and the framing by politicians of Muslims as a collective societal threat can significantly undermine P/CVE policy efforts, for example through community policing approaches, to build relationships of trust and partnership with such communities (Cherney & Hartley, 2017). Similarly, ‘hard’ counter-terrorism policing within communities has damaged P/ CVE efforts at building community cooperation, as shown in the US CVE pilot programme in Minnesota, where significant raids and arrests by the FBI ‘created an enormous amount of suspicion and mistrust regarding the community’, exacerbated by the subsequent use of community informers (Salyk-Virk, 2020, p. 1024).

Challenges in building community resilience against violent extremism The implications of funding for CSOs in community-focused P/CVE work have raised both specific and generic issues internationally for state-community relationships. Britain’s Prevent initially funded community activity on the crude basis of their Muslim population (Thomas, 2012). Responses varied widely, with some Muslim CSOs refusing such funding as ‘tainted’, while others happily took the ‘Muslim money’ (Lowndes & Thorp, 2010) for a wide variety of community development activities. While the former response attracted media attention, the latter response was surprisingly common at a time of wider public funding cuts. Empirical research identified this funding as enabling capacity building, pragmatic state-community engagement (Vermeulen & Bovenkerk, 2012), and a stronger role for Muslim faith-based CSOs vis-à-vis local state governance (O’Toole et al., 2016). Such approaches are not without risks. Huq’s (2017) analysis of the USA’s CVE programme identifies that funding non-credible individuals or CSOs can alienate communities, while, conversely, state funding for genuinely credible CSOs can delegitimise them in the eyes of some community members, as was the case in Minneapolis-St. Paul (Salyk-Virk, 2020). In his study of British mosque involvement in P/CVE, Weeks (2019) perceives a similar cleavage between CSOs willing to engage with government and those increasingly reluctant to do so, while the response to this dilemma among Dutch mosques has been to do preventive work against radicalisation without any recourse to either generic or P/CVE-specific state funding (Welten & Abbas, 2021). A different funding issue involves who states are willing to fund and work with. Birt (2009) identifies a tension between ‘means-based’ and ‘values-based’ P/CVE approaches. The former is a pragmatic approach, whereby states have been willing to work with sections of Muslim communities, particularly Salafists, who may hold very different social attitudes but have the ability to divert individuals from AQ/IS-inspired violent pathways. This was the initial approach taken by the UK’s Prevent through the police’s ‘Muslim Contact Unit’ partnership with Salafists aimed at driving violent extremist supporters from London mosques (Lambert, 2011). Under a new 300

Working with communities to counter radicalisation

government, this approach changed to a ‘values-based’ model, withdrawing funding from CSOs perceived as Salafist. Wimelius et al. (2020) identify similar policy tensions in Sweden over state funding of and P/CVE cooperation with Salafist-identified Muslim CSOs. Such selective P/CVE approaches to funding communities can fuel allegations that P/CVE strategies have indeed sought to stratify communities into ‘good’ or ‘bad’ Muslims (Ragazzi, 2016). Closely allied are the issues of what kinds of community-based P/CVE activities states will fund. For Huq (2017, p. 25), the willingness of local communities to participate in P/CVE initiatives depends on how the issue is framed, what terminology is used, and whether they are perceived as active co-creators or instead feel P/CVE is being ‘done’ to them. In Britain, ‘Prevent 1’ saw a significant variety of community-work approaches conceived and designed at the local level. Yet the nature and effectiveness of these activities – and who led them – were increasingly contested and curtailed by police and security agencies, as a Parliamentary Inquiry laid bare (House of Commons, 2010). This goes to the heart of what community-based P/CVE activity should look like. The demand from UK police was for literal and explicit counter-terrorism activities within communities, but a review of P/CVE research evidence highlights that ‘some organisations have typically focused on trying to build resilience to violent extremism through their existing work streams, rather than specific P/CVE interventions’ (Lewis & Marsden, 2021, p. 37). This focus on protecting and nurturing the social ecology of peer and community networks is echoed by faith-based Swedish CSOs (Wimelius et al., 2020), who argued that ‘what was done on a daily basis should be regarded as part of the [P/CVE] solution rather than part of the problem’ (ibid. 140). Relevant also is the issue of who speaks for communities when engaging (or not) with P/CVE efforts, including the challenge of not taking claims of ‘knowing’ or being able to speak for communities at face value (Rosand et al., 2018). Government stakeholders have sometimes neglected to invest in the relationships necessary to navigate how deeply they are able or willing to go within communities in order to build sustainable state-community P/CVE partnerships, often stopping at the door of self-nominated (usually older male) community leaders or gatekeepers. This has had the effect of eclipsing or marginalising the voices, perspectives, skills, and concerns of women and young people in particular (Grossman et al., 2018). This is now starting to change, with stronger youth engagement and several high-level policy statements asserting the importance of ‘mainstreaming’ gendered approaches within broad P/CVE frameworks (e.g. the 6th Review of the UN Global Counter-Terrorism Strategy, 2018). The early stages of Britain’s Prevent strategy did see overt attempts to create more polyphonic leadership, itself an acknowledgement of the problematic ‘community leader’ model inherent in multiculturalist policy approaches. Some literature advocates a greater focus on key individuals within communities in developing community-level P/CVE activity, people Spalek (2017) terms ‘connectors’. Such individuals may not be ‘community leaders’ and can actually be dissenters, but they are influential individuals who can build ‘bridging’ relationships with marginalised sections of communities – such as young Muslims – and aid prevention efforts in places where there is low trust in the state. This is endorsed by research around ‘community reporting thresholds’ (Grossman, 2015; Thomas et al., 2020) for when or whether community members would share with authorities their concerns about a close friend or family member becoming involved in extremism. This research identified that community members would share their concerns through a ‘staged process’ that included drawing on the advice of trusted community figures resembling Spalek’s ‘connectors’. Resulting P/CVE initiatives to support community reporting (for instance, the UK Counter-Terrorism Police’s ‘Act Early’ campaign and Australia’s ‘Step Together’ advice and support service) are focused much more on families and close friends than on an amorphous construct of ‘community’. 301

Paul Thomas and Michele Grossman

Conclusion: P/CVE – a whole of society approach? We have been exploring the ways in which communities (real and ‘imagined’) have been portrayed as central to P/CVE strategies in many Western countries, guided by the mantra ‘communities defeat terrorism’. However, it is this very focus on ‘communities’ that has generated enduring controversy around P/CVE, particularly in terms of arguably stigmatising entire Muslim communities as ‘suspects’ and the broader securitisation of state-Muslim community relations. These controversies have been driven in part by the explicit naming and responsibilising of Muslim faith communities in various national policy documents and by the targeting of P/CVE activity at specific geographies of Muslim concentration in many countries as the basis for communitylevel activity. Many of these community-level approaches only appeared obvious because they linked the ideology driving the post-9/11 domestic terror threat to entire visible minority (Muslim) communities that were already ‘imagined’ and engaged with as distinct yet homogenous entities shaped overwhelmingly by their collective faith background. While overtly named P/CVE policies implemented at the level of ‘communities’ have also had some positive results, especially in disrupting travel to conflict zones, they have inevitably attracted or exacerbated the controversies outlined earlier. Some communities have tried to avoid this by implementing their own P/CVE programmes without any recourse to the state; however, such programmes may then be seen by the state itself (and by some within communities) as lacking legitimacy or effectiveness. Such controversies and dilemmas run alongside the issues we canvassed above, including who within communities attracts P/CVE engagement and funding, how such funding impacts their standing within broader community contexts, how state-community P/CVE relationships and programmes should actually work, and who directs and controls their content and strategies. These challenges around implementation and effectiveness, alongside the controversies generated by broad community-level targeting, underpin Berger’s (2016) assertion that such ‘whole of society’ approaches represent a counter-effective overreach by existing P/CVE frameworks. Our analysis of community-level activity as a specific spatial and temporal response is underpinned by two emerging trends. One trend is the already present but now rapidly increasing XRW terror threat. What does ‘community’ mean when XRW actors are from the majority ethnic/religious background and are not domiciled in bounded geographical places? Second is the increasing centrality of online activity to terror recruitment, networking, and planning across the ideological spectrum, which, while supported by off-line associations, complicates notions of geographically focused P/CVE activity. As a result, an effective retreat from such overt community-specific programming towards more ‘mainstreaming’ approaches is already discernible in some UK policy domains, for example, in the responsibilising of all front-line professionals and, more recently, ‘intimates’ (Thomas et al., 2020) to identify and refer individuals vulnerable to radicalisation through a ‘safeguarding’ paradigm. A broader shift towards the mainstreaming of prevention activities that are neither identified as nor funded under a P/CVE rubric is now needed, as is the recognition that this will require states to trust both CSOs and broad resilience-building activity more generally. Several key issues in P/CVE-based work with communities remain alive in the academic literature. One is the surprisingly limited body of genuinely empirical research around what communitylevel P/CVE activity has actually involved and how its implementation has been experienced by different actors. Another is that much of the empirical literature draws on qualitative research with community activists that highlights their criticisms of P/CVE policy targeting, with comparatively little attention given to the very different insights on ‘community opinion’ that emerge from the 302

Working with communities to counter radicalisation

smaller body of wider research with Muslim community members in general. The latter shows both low awareness of P/CVE policies and broad support for what they represent. This disjuncture is important but remains under-examined. A third involves how we understand which community-level P/CVE activities ‘work’ and how we can reliably measure their effectiveness. Many states have only patchily evaluated such programmes, if at all, making P/CVE evaluation studies a late-emerging element of the field. Future evaluation efforts could learn from more generic assessments of how community-level preventive measures against any significant social harm can be measured and over what timescale. Finally, there is the issue of trust: not only whether communities trust the state in countering radicalisation but also whether the state trusts communities, and beyond this, trusts in the importance of preventive activities and the investment needed to sustain them. Critical appraisals of what has gone wrong in policy approaches to engaging communities in P/ CVE have tended to prevail, with correspondingly less attention paid to examples of successful P/ CVE community-state partnerships and the features required to nourish them (Ellis & Abdi, 2017). Yet working with communities, whether mainstreamed or explicitly P/CVE-focused, is likely to remain essential in the future. Also, essential will be proactively engaging with changing definitions and dynamics of ‘communities’ and bringing to the fore the actual experience of communities on the ground who have led or contributed to P/CVE initiatives, whether in partnership with or defiance of the state contexts in which they operate.

Note 1 Examples include Public Safety, Canada’s Building Resilience Against Terrorism: Canada’s Counter-Terrorism Strategy (2013); the Council of Australian Governments (COAG) Australia’s Counter-Terrorism Strategy: Strengthening our Resilience (2015); the USA’s Strategic Implementation Plan (SIP) for Empowering Local Partners to Prevent Violent Extremism in the United States (2011); and the UK’s focus on resilience as a central organising concept.

References Anderson, B. (1983). Imagined communities: Reflections on the origin and spread of nationalism. Verso. Apuzzo, M., & Goldman, A. (2013). Enemies within: Inside the NYPD’s secret spying unit and Bin Laden’s final plot against America. Simon & Schuster. Arbuthnot, S. (2021, March 24). Can community engagement counter the radical right? Center for the Analysis of the Radical Right (CARR). www.radicalrightanalysis.com/author/sean-arbuthnot/ Baldino, D. (2021). ASIO, sticky ideologies and anti-government movements. Australian Institute of International Affairs, www.internationalaffairs.org.au/australianoutlook/asio-sticky-ideologies-and-antigovernment-movements/ Bauman, Z. (2001). Community: Seeking safety in an insecure world. Polity Press. Berger, J. M. (2016). Making CVE work: A focused approach based on process disruption. ICCT. Birt, Y. (2009). Promoting virulent envy – reconsidering the UK’s terrorist prevention strategy. Royal United Services Institute (RUSI) Journal, 154(4), 52–58. Centre for Resilient and Inclusive Societies (CRIS). (2021, February). Submission to the Parliamentary Joint Committee on Intelligence and Security (PJCIS) inquiry into extremist movements and radicalism in Australia. www.aph.gov.au/Parliamentary_Business/Committees/Joint/Intelligence_and_Security/ ExtremistMovements/Submissions Cherney, A., & Hartley, J. (2017). Community engagement to tackle terrorism and violent extremism: Challenges, tensions and pitfalls. Policing and Society, 27(7), 750–763. Coaffee, J. (2006). From counterterrorism to resilience. The European Legacy, 11(4), 389–403. Council on American-Islamic Relations Minnesota. (2016). Countering violent extremism program (CVE): What you need to know. www.cairmn.com/index.php/civil-rights/cve-toolkit

303

Paul Thomas and Michele Grossman CREST Advisory. (2020). Listening to British Muslims: Policing, extremism and Prevent. CREST Advisory. Day, J., & Kleinmann, S. (2017). Combatting the cult of ISIS: A social approach to countering violent extremism. The Review of Faith and International Affairs, 15(3), 14–23. Department for Communities and Local Government (DCLG). (2007). Preventing violent extremism: Guidance note for government offices and local authorities in England. DCLG. Department of Home Affairs (Australia). (n.d.). Living safe together. www.livingsafetogether.gov.au Ellis, B. H., & Abdi, S. (2017). Building community resilience to violent extremism through genuine partnerships. American Psychologist, 72(3), 289–300. Etzioni, A. (1995). The spirit of community: Rights, responsibilities and the communitarian agenda. Fontana. European Commission. (2020). A counter-terrorism agenda for the European Union: Anticipate, prevent, protect, respond. European Commission. Executive Office of the President of the United States. (2016). Strategic implementation plan for empowering local partners to prevent violent extremism. Executive Office of the President of the United States. Giddens, A. (1998). The third way: The renewal of social democracy. Polity Press. Global Counter Terrorism Forum. (2013). Good practices on community engagement and community-oriented policing as tools to counter violent extremism. www.thegctf.org/documents/10162/159885/13Aug09_ EN_Good+Practices+on+Community+Engagement+and+Community-Oriented+Policing.pdf Grossman, M. (2015). Community reporting thresholds: Sharing information with authorities concerning violent extremist activity and involvement in foreign conflict. Countering Violent Extremism Subcommittee, ANZCTC. Grossman, M. (2018). The role of families and civil society in detecting radicalisation and promoting disengagement from violent extremism. In C. Echle, R. Gunaratna, P. Rueppel, & M. Sarmah (Eds.), Combatting violent extremism and terrorism in Asia and Europe: From cooperation to collaboration (pp. 155–170). Konrad-Adenauer-Stiftung and S. Rajaratnam School of International Studies. Grossman, M. (2021). Resilience to violent extremism and terrorism: A multi-systemic analysis. In M. Ungar (Ed.), Multisystemic resilience: Adaptation and transformation in changing contexts (pp.  293–317). Oxford University Press. Grossman, M., Carland, S., Tahiri, H., & Zammit, A. (2018). The role of women in supporting and opposing violent extremism: The Australian landscape. Countering Violent Extremism Subcommittee, ANZCTC. Grossman, M., Peucker, M., Smith, D., & Dellal, H. (2016). Stocktake research project: A systematic literature and selected program review on social cohesion, community resilience and violent extremism 2011–2015. State of Victoria. Hardy, K. (2015). Resilience in UK counter-terrorism. Theoretical Criminology, 19(1), 77–94. Hardy, K. (2018, October 15). How the Australian government is failing on countering violent extremism. The Conversation. https://theconversation.com/how-the-australian-government-is-failing-oncountering-violent-extremism-104565 Her Majesty’s Government. (2018). CONTEST: The United Kingdom’s strategy for countering terrorism. Her Majesty’s Government. Hillyard, P. (1993). Suspect community: Peoples’ experience of the prevention of terrorism acts in Britain. Pluto Press. Holt, T., Freilich, J. D., & Chermak, S. (2017). Can taking down websites really stop terrorist and hate groups? The Conversation. http://theconversation.com/can-taking-down-websites-really-stop-terroristsand-hate-groups-84023 House of Commons Communities and Local Government Committee. (2010). Preventing violent extremism: Sixth report of session 2009–10. The Stationary Office. Huq, A. (2017). Community-led counter-terrorism. Studies in Conflict and Terrorism, 40(12), 1038–1053. Kundnani, A. (2014). The Muslims are coming! Islamophobia, extremism and the domestic war on terror. Verso. Lambert, R. (2011). Countering Al-Qaeda in London: Police and Muslims in partnership. Hurst. Levitt, M. (2017). Recent trends in terrorism and counterterrorism: National practices in countering violent extremism. Counter-Terrorism Committee Executive Directorate (CTED), United Nations. www.un.org/ sc/ctc/wp-content/uploads/2017/11/8-Statement-as-delivered-by-Mr.-Matthew-Levitt-The-WashingtonInstitute.pdf Lewis, J., & Marsden, S. (2021). Countering violent extremism interventions: Contemporary research. CREST (Centre for Research and Evidence on Security Threats).

304

Working with communities to counter radicalisation Lowndes, V., & Thorp, L. (2010). Preventing violent extremism – why local context matters. In R. Eatwell & M. J. Goodwin (Eds.), The new extremism in 21st century Britain (pp. 123–141). Routledge. Malik, K. (2009). From Fatwa to Jihad: The Rushdie affair and its legacy. Atlantic. Malthaner, S., & Waldmann, P. (2014). The radical milieu: Conceptualizing the supportive social environment of terrorist groups. Studies in Conflict and Terrorism, 37(12), 979–998. Modood, T. (2013). Multiculturalism: A civic idea (2nd ed.). Polity. O’Toole, T., Meer, N., DeHanas, D., Jones, S., & Modood, T. (2016). Governing through Prevent? Regulation and contested practice in state-Muslim engagement. Sociology, 50(1), 160–177. Peucker, M., & Smith, D. (Eds.). (2019). The far right in contemporary Australia. Palgrave Macmillan. Ragazzi, F. (2016). Suspect community or suspect category? The impact of counter-terrorism as “policed multiculturalism”. Journal of Ethnic and Migration Studies, 42(5), 724–741. Ranstorp, M. (2018). Ex post paper “research seminar”. RadicalisationAwareness Network Centre of Excellence. https://ec.europa.eu/home-affairs/sites/default/files/what-we-do/networks/radicalisation_awareness_ network/ran-papers/docs/ran_research_seminar_17102018_en.pdf Rosand, E. (2017). Fixing CVE in the United States requires more than just a name change. Brookings Institution. www.brookings.edu/blog/order-from-chaos/2017/02/16/fixing-cve-in-the-united-states-requiresmore-than-just-a-name-change/ Rosand, E., Winterbotham, E., Jones, M., & Praxl-Tabuchi, F. (2018). A roadmap to progress: The state of the global C/PVE agenda. Royal United Services Institute. Salyk-Virk, M. (2020). Building community resilience? Community perspectives of the countering violent extremism pilot program in Minneapolis/St. Paul. Studies in Conflict and Terrorism, 43(11), 1011–1042. Scheffer, P. (2011). Immigrant nations. Polity. Shanaah, S. (2019). Alienation or cooperation? British Muslims’ attitudes to and engagement in counterterrorism and counter-extremism. Terrorism and Political Violence, 34(1), 71–92. Sliwinski, K. F. (2012). Counter-terrorism – a comprehensive approach. Social mobilization and “civilianisation” of security: The case of the UK. European Security, 22(3), 288–306. Spalek, B. (2013). Terror crime prevention with communities. Bloomsbury Academic. Spalek, B. (2017). Community engagement for counter-terrorism in Britain: An exploration of the role of “connectors” in countering Takfiri Jihadist terrorism. Studies in Conflict and Terrorism, 37(10), 825–841. Stephens, W., & Sieckelinck, S. (2020). Being resilient to radicalisation in PVE policy: A critical examination. Critical Studies on Terrorism, 13(1), 142–165. Thomas, P. (2011). Youth, multiculturalism and community cohesion. Palgrave Macmillan. Thomas, P. (2012). Responding to the threat of violent extremism – failing to Prevent. Bloomsbury Academic. Thomas, P. (2017). Changing experiences of responsibilisation and contestation within counter-terrorism policies: The British Prevent experience. Policy and Politics, 45(3), 305–322. Thomas, P., Grossman, M., Christmann, K., & Miah, S. (2020). Overcoming barriers to community reporting on violent extremism by “intimates”: Emergent findings from international evidence. Critical Studies on Terrorism, 13(4), 638–659. United Nations. (2018). 6th review of the UN global counter-terrorism strategy. www.un.org/counterterrorism/ events/un-global-counter-terrorism-strategy-advancing-women%E2%80%99s-leadership-and-participation Vermeulen, F., & Bovenkerk, F. (2012). Engaging with violent Islamic extremism: Local policies in Western European cities. Forum. Weeks, D. (2019). Barking Mosque and quintessential insight: Overcoming the problematic government: Community counter-terrorism partnership in the UK. Studies in Conflict and Terrorism, 42(8), 735–754. Weine, S., Eisenman, D. P., Kinsler, J., Glick, D. C., & Polutnik, C. (2017). Addressing violent extremism as public health policy and practice. Behavioral Sciences of Terrorism and Political Aggression, 9(3), 208–221. Welten, L., & Abbas, T. (2021). “We are already 1–0 behind”: Perceptions of Dutch Muslims on Islamophobia, securitisation and de-radicalisation. Critical Studies on Terrorism, 14(1), 90–116. White, J. (2020). Community and gender in counter-terrorism policy: Challenges and opportunities for transferability across the evolving threat landscape. Evolutions in Counter-Terrorism, 2, 77–100. Wimelius, M., Erikisson, M., Strandh, V., & Ghazinour, M. (2020). “They think of us as part of the problem instead of part of the solution”: Swedish civil society and faith-based organizations in resilience building and prevention of radicalization and violent Islamist extremism. Journal for De-Radicalization, 22, 122–155.

305

20 THE ROLE OF RESEARCH AND RESEARCHERS IN COUNTER-RADICALISATION POLICY AND PRACTICE Alice Martini and Laura Fernández de Mosteyrín Introduction In the last decades, radicalisation studies have become a strong sub-field within terrorism studies, as the chapters in this volume illustrate. Focusing on this sub-field, this chapter provides an overview of the main developments, debates, and challenges that shape researchers’ positions when studying counter-radicalisation policy and practice. In other words, when conducting research within this field, researchers encounter various challenges that will shape the way they conduct research and their career path. These decisions are not solely linked to research issues but also related to the broader policy-relevant nature of the field. Overall, radicalisation studies inherit some of the characteristics that shaped the field of terrorism studies. The latter was born as a policy-oriented field of inquiry, and initially, many researchers in this field were working in close connection with policy-making spheres (Fitzgerald et al., 2016, p. 2; Stampnitzky, 2013). This shaped the need to orient research on terrorism and counterterrorism mostly towards problem-solving analyses. The policy-relevant nature of the field also implied that various dilemmas shaped research on terrorism and counter-terrorism, dilemmas that were mostly shaped by researchers’ interactions with practitioners and policy-makers and, in consequence, by the political nature of knowledge production. These characteristics have been inherited by the field of radicalisation studies (Russo & Selenica, 2021). Countering and preventing radicalisation entails the need to conceptualise and provide empirical evidence of a very complex phenomenon, its causes and consequences, and, overall, the formulation of intervention strategies. Therefore, all stages of research on counter-radicalisation – from the formulation of research objects to the connection between policy-orientedness and interaction with practitioners and policy-makers – not only offer various opportunities but also represent different challenges for researchers. It is these questions that this chapter describes, with the aim of illustrating not only scholarly debates but also research decisions that will orient researchers’ work in many ways. In other words, these decisions will not only shape their research but also, overall, the role researchers will have in counter-radicalisation policy and practice. To do this, in the chapter, we first approach the dilemmas researchers will have to face in the initial steps of their research project. Second, we will reflect on the various challenges researchers will encounter when accessing the field, collecting data, and then analysing their findings. DOI: 10.4324/9781003035848-23

306

Counter-radicalisation policy and practice

Lastly, we will discuss the key debates emerging in relation to the process of communicating research findings to targeted audiences. In all these stages of the research process, researchers will be embedded in a policy-relevant field; therefore, they will need to confront three main questions: what does the research study, what is its aim, and who is the research produced for.

Navigating a sensitive and policy-relevant field while researching counter-radicalisation This section focuses on the very first stages of research planning and design. It starts by discussing the various questions arising when developing the research plan; it then discusses conducting research in a very heterogeneous field such as radicalisation studies. The section then turns to the possibilities for funding and the challenges researchers will encounter when collecting data, challenges that arise primarily from the political and sensitive nature of the field.

Thinking about your research object: elusive concepts and problem-solving approaches The first challenge researchers encounter when entering the field of radicalisation studies is the definition of the object of study – that is, not only radicalisation but also counter-radicalisation. For decades, the definitional issue has been one of the overall challenges in the field of terrorism studies, and it is one of the features radicalisation studies has inherited. At the same time, however, the radicalisation studies are based on more elusive concepts, as processes of radicalisation and counter-radicalisation are strongly challenging to define. As discussed in other chapters, radicalisation and counter-radicalisation can include a wide variety of different processes, drivers, and individual and collective behaviours. This depends on the fact that there is no simple answer – if there is one – to the question of ‘what leads a person to turn to political violence?’ (Morrison, 2020). So far, no consensus has been reached on the definition or the specific drivers of radicalisation, rendering it very difficult to achieve a consensus not only on the very meaning of radicalisation but also on what counter-radicalisation should focus on – and therefore, what should be the specific focus of the sub-field. The definition of the object of the study will be the first thing researchers will have to focus on. The question that arises here is not merely whether they will focus their research on radicalisation or counter-radicalisation but, more specifically, how they will understand their key concepts and research object. In other words, when devising their research – theoretically, analytically, and methodologically – researchers need to be aware of not only what their research can tell them but also, above all, what it aims to do (Morrison, 2020). The way radicalisation is defined will shape the formulation of research and the production of counter-radicalisation strategies (Coolsaet, 2016; A. Schmid, 2013). Also, because of the subfield’s specificities, the conceptualisation given will somewhat position researchers and their research politically within the field. In other words, the research-policymaking-impact link that distinguishes radicalisation studies will take different forms depending on the researchers’ position towards their object of study and will shape the target audience for the consumption and dissemination of their findings – for example, state authorities and police agencies for prevention, and civil society advocates for rights watch. These questions are linked to a broader division among scholarly positions that already defined the field of terrorism studies and that now also shapes radicalisation studies. This debate divides scholars who advocate for a problem-solving and normative approach to radicalisation and 307

Alice Martini and Laura Fernández de Mosteyrín

counter-radicalisation from those who defend a critical stance. Simplifying intellectual positions (for further details, see, among others, Muro & Wilson, 2022; Silke, 2019; Jackson, 2016a), the former stance is usually closer to policy-making as it mostly focuses on understanding, analysing, and producing knowledge on radicalisation with the aim of formulating and improving counterradicalisation strategies (A. Schmid, 2013). More critical positions usually start their reflections with the political nature of concepts such as radicalisation and counter-radicalisation strategies (Jackson, 2016a). They argue that the state and international organisations are still the biggest organisers and sponsors of research on counter-radicalisation and also the biggest consumers of research (Russo & Selenica, 2021, p. 10; Stampnitzky, 2013, p. 7). This issue may also lead to possible abuses from powerful actors – for example, allowing a certain kind of control or surveillance of societies (Russo & Selenica, 2021, p. 10; Lindekilde, 2016). This issue was already present with counter-terrorism and has become stronger with counter-radicalisation because of the broader state/society interaction these practices involve (Russo & Selenica, 2021, p. 10; Lindekilde, 2016). Clearly, the scholarly different focus does not imply that critical researchers will automatically reject interaction with practitioners and policy-makers or policy impact. However, their research focus will shape this engagement differently. Compared to problem-solving and normative research, critical scholars would, for example, focus on the impact and consequences of counterradicalisation programmes on societies, while the question of why and how people engage in violent extremism and violence would be examined through more sociopolitical approaches that incorporate identity, ideology, and politics behind violence. Overall, researchers will need to reflect carefully on what will be their object of study and what will be the aim of their research. Regardless of their academic position, it is worth remembering that the study of radicalisation and counter-radicalisation is a political topic – like many others within the social sciences. Specifically, the focus on individuals – and the implementation of policies and programmes across society – renders radicalisation studies very sensitive. Therefore, regardless of their specific object of study, researchers will always need to carefully scrutinise the relationship between conceptualisations, models of analysis, levels of analysis, and even units of analysis by carefully following research ethics along the way, and also carefully determining who the beneficiaries and targeted audiences of their findings are.

Producing research on counter-radicalisation in a heterogeneous field The question of the definitions and the research aims is also linked to the researcher’s discipline and disciplinary approach. When researchers enter the field of radicalisation, they encounter a very heterogeneous landscape. Terrorism studies have always been positioned at the crossroads of many disciplines, and radicalisation studies are an even broader and more interdisciplinary field, as many approaches are needed to study how to counter processes of radicalisation and to produce counter-radicalisation strategies (Gordon, 2010). Increasingly, many are the disciplines that have the potential to documenting and explaining the complexity of the impacts of many counter-radicalisation programmes on society. In the last decade, the need to study radicalisation from different perspectives has led to a rise in interest in this topic from a wide variety of disciplines within the humanities and social sciences, including, for example, computer science and mathematics (Morrison, 2020, p. 57). Furthermore, within the subfield, psychologists have become the most numerous figures, displacing political scientists who, previously, dominated the conversation on terrorism (Phillips, 2021). All these developments have rendered the radicalisation studies multi-disciplinary and increasingly interdisciplinary. Interdisciplinarity has also brought more attention to methodological training in universities (Schuurman, 308

Counter-radicalisation policy and practice

2020) and methodological diversity. This has also implied that both quantitative and qualitative methods are widely used for the study of radicalisation and counter-radicalisation. Overall, this has allowed the production of more sophisticated data and more holistic knowledge of radicalisation and counter-radicalisation. This detailed and specific focus on these topics has gathered researchers from different disciplines into radicalisation studies (Boyle, 2012). However, while allowing a focus on micro-level data, it also led to the fragmentation of the study of political conflicts ‘into highly specialized studies of political violence and, somewhat, a loss of cross-pollination of ideas’ (Boyle, 2012, p.  528). Just as terrorism studies was a highly fragmented field (Richard English as in Morrison, 2020, p. 15), radicalisation studies reproduces but also further fractures the study of terrorism, political violence, and conflict. The high specialisation has thus implied a broader and more holistic approach to the study of radicalisation, with researchers from various disciplines working together. Nevertheless, the specialisation process and interdisciplinarity raise challenging issues. First, every discipline is strong in specific methods while weak in others, and each perspective has its own core research domains, conceptual apparatuses, vocabularies, and methodological strengths that set the limits of their explanatory potentials (Alford & Friedland, 1985). As the field of radicalisation studies is populated by researchers from very different traditions, a topic-centred field might decrease the methodological standards of each discipline. Second, the topic-centred nature of the field and its policy-orientedness have also rendered more challenging the possibilities of dialogue with related and broader fields. This implies that certain knowledge that could be relevant for the study of radicalisation and counter-radicalisation, for example, social movements, social mobilisation, and the broader field of the sociology of conflict being representative, does not always reach the researcher too easily because it may be produced under a different sub-field. Along these lines, and as an example, the definition of the object of research also implies defining the level of analysis, for example, micro (individuals), meso (organisations), macro (policies). It is a combination of these levels of analysis that will produce the most holistic understanding of radicalisation and thus more sophisticated counter-radicalisation strategies. The micro level focuses on the individual and the problems a person may experience that may lead them to terrorism, namely, identity issues, marginalisation, discrimination, failed integration, and even psychological processes (A. P. Schmid, 2016, p. 5). So far, definitions of radicalisation and the specific fields scholars belong to have led research to focus mostly on the individual and what may lead an individual to embrace violence, thus creating an unbalance in the focus of research adopted so far (Morrison, 2020; A. Schmid, 2013; A. P. Schmid, 2016; Schuurman, 2020). Contrastingly, studies of the meso level and, for example, milieus’ roles have been produced mostly in the field of social mobilisation (Della Porta, 2013; Della Porta & LaFree, 2012). While these works are acknowledged in radicalisation studies, their influence and dialogue with these different fields could be more substantial. For example, radicalisation studies still lack a robust account of the process of organisational radicalisation (Della Porta, 2013), such as the role of organisations and social actors in recruiting, driving, or preventing radicalisation. The focus on long-term sociohistorical changes could also be stronger (Malešević, 2017) and provide a long-term evolution of the causes and context of mobilisation, organisation of violence, or means of action to assess counter-radicalisation practices. A similar problem exists for macro-level focus. Centred on the study of conflict and conflict dynamics, policy-level analyses may provide a broad approach to conflicts and international and domestic politics, shaping conditions for radicalisation processes. Lastly, there is a final issue that needs to be discussed in relation to the heterogeneous composition of the field (Russo & Selenica, 2021). This is due to the fact that very heterogeneous professional figures are involved in it. As detailed in the previous chapters of this volume, the countering 309

Alice Martini and Laura Fernández de Mosteyrín

and preventing of radicalisation is based on early detection and anticipation, strategies that cannot be implemented solely in the state-led security domain. The logics of countering radicalisation have diversified the actors and spaces of counter-terrorism, now implemented also through and by society (Lindekilde, 2016). By doing so, they have also diversified the expertise needed to understand radicalisation and the production of counter-radicalisation strategies. In other words, this differentiation of actors and spaces has significantly broadened the strand of radicalisation research and the researchers involved within it. New figures of experts have emerged inside academia or research spheres but, importantly, also outside of them (Russo & Selenica, 2021). The broadening of the field has also led to including, in a more systematic way, civil society organisations, charities, NGOs, and citizens’ groups. Some of these groups are focused on local realities, while some others have a transnational scope – for example, the European Network Against Racism; Women Without Borders/Sisters Against Violent Extremism; and the European Forum for Urban Security (as identified by Russo & Selenica, 2021, p. 10). This has led to the formation of ‘professionalised counter-radicalisation networks of practitioners’ (Stanley et al., 2017), who can thus add knowledge on radicalisation. One of the more prominent platforms in this regard is the EU Radicalisation Awareness Network (RAN). This brings together ‘first-line practitioners, field experts, social workers, teachers, NGOs, civil society organisations, victims’ groups, local authorities, law enforcement, academics and others’1 to work together on countering radicalisation. In this sense, RAN is an example of interaction between policy-makers, practitioners, and academics, among other figures, and it can be considered a prime example of places where such interaction produces knowledge outside academia. Within these networks, practitioners have been claiming further inclusion in the production of knowledge and strategies so as to produce counter-radicalisation policies based on their experience (Stanley et al., 2017). Consequently, dialogue with grassroots practitioners is not only sought but also envisaged as a key pillar of counter-radicalisation (Parker et al., 2019) – another of the specificities of this sub-field. Nonetheless, while holistic knowledge of radicalisation processes and how to counter them would be a welcome development, the exchange of knowledge with these different sectors does not always work smoothly and has sometimes revealed itself to be very challenging. This is a feature researchers will observe and will have to deal with at all different stages of their research activity – that is, when designing their research, as they may have to collaborate with non-academic stakeholders; when collecting data, as some data may be held by these stakeholders; and even when disseminating their research, as they may have to carry out this activity not only with scholars but also with other professional figures. Overall, this shift has implied that researchers do not hold a monopoly on the knowledge production on radicalisation and counterradicalisation, but that this knowledge is produced in various spheres and by various professional figures. This diversified the spaces of production of knowledge, transforming it into a more multilayered process that should involve dialogue among academics, practitioners, and policy makers (Martini & Fernández de Mosteyrín, 2021) – all of these figures are usually focused on policy impact. Along these lines, more questions arise when the issue of getting funding is discussed, as this also requires the researcher to decide on various matters, as illustrated in the next section.

Dilemmas of securing funding and the policy relevance of the research produced The consolidation of the field has also implied that more resources were allocated for research and the opening of research opportunities within the process of specialisation of researchers and 310

Counter-radicalisation policy and practice

research. Therefore, the sub-field of radicalisation studies has slowly been institutionalised. An essential process at the core of the possibilities of institutionalising the field and researching this topic has been the increase in funding for the study of radicalisation and counter-radicalisation (Gordon, 2010) within universities, think tanks, and even international organisations such as the EU (Russo & Selenica, 2021) or the UN (Martini, 2021). Getting funding for research is a fundamental condition for research to take place, for research agendas to consolidate, and for researchers to enhance their autonomy. However, it also implies that researchers will have to make some decisions regarding their research design to access funding schemas. In this sense, getting funded also influences the definition of the research and the role researchers will adopt in counter-radicalisation policy and practice. While scientific policies vary and funding schemas are quite diverse, there are some paradigmatic consensuses that most (Western) funding avenues share. The way these questions are addressed is core to pulling resources and building autonomous research agendas. To mention a few examples, most funding schemas assess policy relevance by calling for research focusing on social priorities and policy issues. They also assess interdisciplinarity as a guarantee of more holistic knowledge. Gender approaches become core, not just to evaluate the gender composition of the research team but also to assess how gender inequality will be examined in any research project in any scientific domain. And research schemas usually also ask for a detailed ethics plan, especially, when researching sensitive topics. Lastly, a strong funding proposal will also need to consider questions of findings’ communication beyond scientific outlets. In reflecting on all these issues, researchers need to carefully consider the three big questions that we mentioned at the beginning: what does the research study, what is its aim, and who will the knowledge produced be for. It should be noted that access to funding varies according to patterns of inequality within and across countries. Moreover, in addition to typical socioeconomic individual-group conditions, there are at least three positions that shape researchers’ possibilities: career stage; affiliation to university versus private research institutions; and the researcher’s country or region. Researchers enter the field, primarily, as PhD candidates and early-career scholars. Additionally, they do so either as members of universities or as non-academic research institutes, think tanks, or public/ private agencies. These positions in the social structure of the field may grant access to funding, data, and publication opportunities or become an obstacle to research. The issues that arise from their position (Liebling, 2001) are related not only to the conditions of access to knowledge, that is, access to data and fieldwork, spaces for the transfer of knowledge, but also to the ends of the knowledge produced, and thus to funding and funding schemas. Finally, in a field where problem-solving and oriented knowledge are core, the fieldwork production/findings might become attractive to official penetration, making the thin line between research’s independence and autonomy and utility and impact an important source of necessary reflection. In this sense, there is also a question of sponsorship and the sphere of production of knowledge, and how these may influence research. Research is produced in many different spheres, all of which are shaped by different forces and needs. For example, NGOs, civil society organisations, and even non-governmental think tanks may need to apply for funding from external agencies or regional institutions. They may thus be forced to adapt their projects to these funding schemas. This may not be an issue for governmental think tanks and research centres, but these may also see their research influenced by the state’s needs or the funding agencies’ interests. Lastly, university scholars may also suffer from the pressure of the potential attraction of research funding (Fitzgerald et al., 2016, p. 2). Furthermore, varying hugely for different regions and countries, in all these contexts, there may be pressures for scholars to become policy-relevant, for example, through funding schemas aimed at the production of problem-solving knowledge 311

Alice Martini and Laura Fernández de Mosteyrín

and linking these, for example, to career opportunities. These different positions within the field and the access to specific funding schemas will also shape research and may, for example, focus it more on problem-solving and policy-making approaches. At this crossroads, researchers face a thin red line. This is very much contextual in each country, and it ends up shaping the research process, research results, and career outcomes. As all these issues are present in any field of science, the specificities of the field of radicalisation studies become powerful forces that shape researchers’ careers. These are issues that emerge in all phases of the research, and they are also linked to the data collection opportunities, which are further discussed below.

Data collection challenges Once the previous steps have been solved, researchers must navigate one of the core challenges of radicalisation studies, namely, how to retrieve or produce data. While evidence on the process of radicalisation and the effects of counter-radicalisation is increasing, producing empirical evidence for radicalisation research is still a challenge, and, for some scholars, whether the field has the correct data and the adequate instruments to analyse radicalisation and assess counter-radicalisation is still an open question (Miller & Mills, 2009). Starting with the matter of primary data, the availability, access to, and retrievability of empirical evidence still present important challenges. Compared to the initial situation in terrorism studies, radicalisation studies managed to offer more first-hand empirical data. While, in some cases, this still relies on secondary sources, the use of primary data has increased considerably, and ‘Scholars have also begun to adopt a wider variety of data-gathering techniques, greatly diminishing the overreliance on literature reviews that was noted from the 1980s through to the early 2000s’ (Schuurman, 2020, p. 1012). Nevertheless, access to certain data is still challenging, and many are the factors that represent an obstacle to data collection here. First, data on radicalisation and counter-radicalisation is mostly held within the power of the state, and a lot of information is still kept secret by governmental, intelligence, and law enforcement agencies, which are not always willing to disclose sensible data to scholars (Max Taylor, as reported in Morrison, 2020, p. 88). At the same time, they do not always have the expertise to exploit that data. Some authors have been strongly critical of states’ gatekeeping of data and have argued that governmental reticence to share primary data with academia has created a ‘stagnation’ in terrorism studies (Sageman, 2014). As Sageman puts it, ‘intelligence analysts know everything but understand nothing, while academics understand everything but know nothing’ (Sageman, 2014, p. 576). Some other scholars have taken a more nuanced position on this issue, but they have still highlighted that it is of paramount importance to narrow the gap between academic research and counter-terrorism intelligence (A. P. Schmid, 2016, p. 28; Sageman, 2014). Second, data collecting struggles are sometimes aggravated by the difficulties in getting security clearances, an issue that makes it very difficult for researchers to work with primary sources (A. P. Schmid, 2016, p. 28). For example, it is not always easy or possible to interview terrorists or extremists in jails (Gary LaFree, as in Morrison, 2020, p. 88). Consequently, it is more challenging to gather first-hand data and almost impossible to test and assess the validity of the theories produced (Bart Schuurman, as in Morrison, 2020, p. 18). Lastly, while some scholars have managed to access closed data (see, among others, Gill et al., 2022), they have also highlighted that its real utility is an open question. Closed data was not collected or developed for research purposes; therefore, it may not always be exploitable by researchers. Overall, thus, whether closed data is really needed and whether accessing it is feasible is an element the researcher should consider before devoting efforts to retrieve it. 312

Counter-radicalisation policy and practice

Moreover, if transparency and governmental gate-keeping are problems in terms of data retrievability, data production is also challenging. Like other social sciences, ethnographic research is a very common research instrument, but fieldwork in radicalisation and counter-radicalisation faces obstacles in terms of access, gatekeeping practices, and ethics (Russo & Strazzari, 2020). Access to the penal system and to specific populations such as convicted people, activists, and groups, or with frontline practitioners involved in counter-radicalisation may prove difficult. In addition to access obstacles, ethical dilemmas also arise in fieldwork. First and foremost, the do-no-harm principle should arise anytime researchers target vulnerable individuals or communities already labelled by state policies, as engaging people’s political views is always sensible. Furthermore, as universities and other funding agencies require ethics committees to approve and oversee fieldwork, the researcher’s safety might be a matter impeding research in specific settings and countries (Russo & Strazzari, 2020; Dolnik, 2013). On this, it is important to note the possible solutions new technologies offer that may be of help to the researcher in navigating around certain issues. Already extended before the coronavirus pandemic but then strengthened by it, some researchers are opting to conduct fieldwork with stakeholders via online interviews. This clearly presents some limitations, as, for example, personal reactions and some features of the immediacy of a face-to-face conversation may be lost. However, it may allow researchers to conduct interviews without having to travel to a specific country and to be able to reach out to a wider number of interviewees, among others. There is then the possibility of conducting various forms of digital ethnographic research (Pink et al., 2016). For example, using mostly a sociological and anthropological approach, netnography focuses on the qualitative study of online communities and has become an important approach to studying the interactions and interrelations of online digital communities. Another interesting option for researchers in this field is Social Network Analysis (SNA)which, bringing together sociological and mathematical approaches, maps the structure of communities on social networks and their social ties. It thus traces the spread of messages online and how these may influence beliefs, behaviours, or radicalisation processes. However, when conducting digital ethnography, researchers will have to scrutinise the reliability and veracity of the online sources and be aware of the fact that this method is not exempt from ethical and political risk dilemmas. Moreover, the technological advances that the field has been experiencing in the last few years – for example, automated data collection – will further improve and broaden the data available in the field. Various scholars remark that this is a considerable break from the past, when terrorism studies was a field that experienced the issue of the absence of data because of the many difficulties in retrieving it (Paul Gill in Morrison, 2020, p. 87). Nowadays, both terrorism studies and radicalisation studies are undergoing a sophistication of the data that is produced and of the questions that are asked – also in methodological terms (Paul Gill in Morrison, 2020, p. 87). In this sense, while issues remain, there is also a proliferation of new, higher-quality data and contemporary techniques for research that are constantly improving the evidence-based knowledge of these three levels and how they interact (Morrison, 2020, p. 82; Schuurman, 2020). Lastly, there are some other challenges researchers will need to consider when designing their fieldwork and conducting their data collection. Both official data and fieldwork challenges also raise questions for research devoted to comparison among countries or regions. While counterradicalisation paradigms are global, they are transposed to different countries. However, largescale comparisons are difficult because, while certain countries are more open to transparency, others are not (Russo & Strazzari, 2020). Legal and policy evaluation cultures vary across countries and regions of the world. In consequence, comparability depends very much on context specificities like legal systems, system transparency, security agencies, democratic cultures, or national 313

Alice Martini and Laura Fernández de Mosteyrín

security arrangements. In this sense, researchers will experience differences in access to information in different regions because of structural obstacles. As said, conducting fieldwork in conflict areas and with violent non-state actors may prove very challenging (Toros, 2008) in terms of both access and the researchers’ safety (Grimm et al., 2020; Sriram, 2009; Nordstrom & Robben, 1995). However, researchers will need to consider whether the state will also be a security challenge – for example, in some non-democratic settings (Bliesemann de Guevara & Bøås, 2020). These issues will also depend on the researchers’ professional affiliation – that is, different professional spheres may gain different access – and on the decision on how to access the field – whether as a lowprofile individual or a high-profile researcher, both options entailing benefits but also limitations. Lastly, researchers will also need to conduct a reflexive exercise and carefully consider how their subjectivity will impact their fieldwork, data collection, and overall research and analysis. Researchers’ identities influence their options, methods, and access to the field. Gender, race, country of origin, and even university affiliation are elements that will impact the different possibilities of accessing the field. For example, racialised scholars may find it difficult to access certain kinds of extremist groups, while gender and race relations, among others, will structure – and thus impact – the ways interviews will be conducted. Moreover, and more broadly, researchers’ cultural, social, and political backgrounds will also influence the analysis of the data and the overall research.

Dissemination of research results and impact on policy making The last steps researchers will have to take are those of dissemination of research results and possible engagement with and impact on policy-making. These are not only separate but also interrelated issues, as they are strongly linked with researchers’ understanding of the aim of their research – and, overall, the position they want to occupy in society. The debate on this issue is well summarised by the ‘public sociologies’ and the broad question of knowledge for what and for whom, as formulated by Burawoy (Burawoy, 2004). Starting from the process of knowledge transference and dissemination and the knowledge for whom question, it should be mentioned that researchers are also increasingly called to transfer knowledge and share their activity beyond the Ivory Tower, not just with the state or governmental agencies but also with broader society. Public knowledge raises the question of whether researchers have the duty to engage with society beyond state structures and if, in this case, researchers should avoid positioning themselves politically (Burawoy, 2004, p. 1606). For Burawoy, moral stances, values, and political commitment are not just impossible to escape but also they are ‘a sine qua non of any research program’ (Burawoy, 2004, p. 1605). In this sense, there is no inherent contradiction in publicly declaring the researchers’ underlying commitment and values, and, in fact, the issue may reside in research that is presented as above-politics, apolitical, and objective. Moreover, claims of apolitical and objective research may become particularly problematic in the social sciences because of the knowledge’s political, historical, and cultural contingencies (Lindekilde, 2016). Furthermore, universities, centres, or international studies associations are also actors in the landscape of civil society. Therefore, researchers may have ‘a right and an obligation to participate in politics’ (Burawoy, 2004, p. 1605), to adopt a position informed by their distinctive expertise, and to disseminate knowledge to society. These points become particularly relevant when looking at counter-radicalisation policies and practices because of the specific knowledge that is being transferred to the public. As this volume explains, early identification of radicalisation – and preventing and countering it – require a turn towards society. Various figures are called into collaboration with preventive and countering 314

Counter-radicalisation policy and practice

strategies. The lay citizen is thus asked to become ‘ever vigilant’ and to identify and denounce possible signs of radicalisation (Emerson, 2019). As it has been argued, in the expansion of state powers, the accumulation of coercion is heavily dependent on social legitimisation (Fernández de Mosteyrín, 2020, p. 201). Communication with lay citizens becomes thus key to legitimisation, but it is also a tool to raise awareness and empower citizens to spot signs of radicalisation and collaborate in prevention. This reflection shows how, also within radicalisation studies, the question of knowledge for whom is strictly linked to the question of knowledge for what. On this, Burawoy asked: Do we take the values and goals of our research for granted, handed down to us by some external (funding or policy) agency? Should we only concentrate on providing solutions to predefined problems, focusing on the means to achieve predetermined ends, on what Weber called technical rationality and what I call instrumental knowledge? In other words, should we repress the question of ends and pretend that knowledge and laws spring spontaneously from the data, if only we can develop the right methods? Or should we be concerned explicitly with the goals for which our research maybe mobilized, and with the values that underpin and guide our research? (Burawoy, 2004, p. 1606) Linked to this issue, two important reflections emerge for researchers in radicalisation studies, that is, whether to aim for policy impact and whether to engage with the state and society. Starting from the former, the imperatives of research impact and transfer are debates that affect deeply the social sciences and the politics surrounding the production of knowledge and are thus crucial when looking at the role of research and researchers in counter-radicalisation policy and practice. Impact and transferring policies give visibility to research and, more generally, to the use of public investment in research. Nevertheless, they may also have unintended consequences and affect the quality of research and research cultures (Chubb & Reed, 2018). Overall, these politics end up prioritising not only funding for certain disciplines. They also influence researchers’ choices that may need to be made concerning funding opportunities – thus ‘altering perceptions of self-determination’ (Chubb & Reed, 2018, p. 295) and even re-directing researchers’ interests to ‘here and now’, rather than producing long, historical, and social assessments of the phenomenon studied that would allow a better understanding of it. This may result in the decontextualising of research and, overall, counter-radicalisation practices and policies. Chubb and Reed also warn about how institutions’ calls for impact and transference have been leading to a growing politicisation of knowledge that may result in the prioritisation of certain kinds of ‘useful’ knowledge – that is, problem-solving knowledge – rather than critical engagements seen as less ‘useful’. Overall, societal impact and the transfer of knowledge are understood as linked to demonstrating policy relevance and engagement (Fitzgerald et al., 2016, p. 2). As mentioned, terrorism studies has been a policy-oriented field since its establishment (Fitzgerald et al., 2016, p. 2; Stampnitzky, 2013). The politics of impact and calls for the transfer of knowledge lie on their own terms. The politicisation of knowledge, therefore, is not a new dynamic, and it could even be argued that the demand for research on counter-terrorism and, more recently, counter-radicalisation is one of the forces shaping the field – and its institutionalisation (Stampnitzky, 2013). However, some scholars argue that heterogeneity of views and profoundness of analyses may thus be lost in favour of easy-to-access and fast-produced research aimed at policy-making and policy impact (Berling & Bueger, 2015; Jackson, 2016b). Moreover, if placed as a goal, the 315

Alice Martini and Laura Fernández de Mosteyrín

production of ‘policy-relevant’ research may have a detrimental effect on research on radicalisation and counter-radicalisation. This may push researchers to ask particular kinds of questions, which may lead them to look for specific types of evidence (Russo & Selenica, 2021). Framing research as problem-solving may thus lead researchers to ask questions about how to counter radicalisation, already taking some underlying assumptions behind ‘radicalization’ for granted (Jackson, 2016b; Lindekilde, 2016). Furthermore, the increased demand for policy impact and transfer of knowledge reflects the turn towards the commodification of knowledge, the neoliberal logics driving funding, and, as mentioned, the idea that research needs to be ‘useful’ (Thornton, 2015). This is why some scholars defend that the aim of the production of knowledge on radicalisation and counter-radicalisation is its implementation, that is, its ‘usefulness’. Overall, this raises the question of what kind of audience is knowledge for. However, some types of analysis, for example, the formulation of radicalisation models and conceptualisations of radicalisation, will be directed at the state. This depends on the fact that its strengths reside in the formulation of policies, thus, rendering policy engagement possible and, in some cases, needed. In this sense, this knowledge will be mostly directed to practitioners and policy-makers, as they are mostly in charge of counterradicalisation policy – and thus policy impact will usually be sought by researchers. However, as said, not all scholars agree on interaction with state agencies and institutions, and, in fact, interaction with policy-makers is among the main topics that divide the field. Their criticism is usually focused on the use the state may make of this knowledge to implement and legitimise certain kinds of responses that strengthen the control of societies in various ways (Jackson, 2007, p. 123). Therefore, they argue for a primary effort to dialogue with advocates and activists within civil society to reduce the legitimation of state policies (Jackson, as in Martini, 2017). In their view, this would also allow for a more autonomous production of critical knowledge because, as Russo and Selenica argue, the involvement of state agents, the involvement of state agents, officials, representatives of international organisations, whose political agendas and positions are less and less prone to contestation and resistance, seems to contradict the tenets of critical epistemologies (sought in social sciences). (Russo & Selenica, 2021, p. 12) Contrastingly, for some other scholars, since the state is still among the leading implementers, engagement with policy-makers and official operators seems to be the natural end of research (Fitzgerald et al., 2016, p. 2; for a detailed discussion, see Toros, 2016, p. 126). Scholars identifying with this position defend that it is possible to find a – delicate – balance between intellectual independence, the promotion of emancipation, and the production of critical and sophisticated scholarly knowledge, and engagement with the state (Toros, 2016, p. 126). This balance is said to be needed as these scholars understand that sharing their research with the state and state agencies will be a way to shape, formulate, and improve counter-radicalisation policies and practices. Therefore, it is important that researchers dialogue and interact with state structures to offer alternative views on counter-radicalisation practices (Toros, 2016, p. 127) while always keeping in mind the balance needed to produce independent research. Nevertheless, this scholarly division on engagement with policy-making spheres needs to be problematised. Positioning their research may not always be a straightforward and independent choice for all researchers. As Fitzgerald et al. remind us, ‘Perhaps it is naïve, however, to assume that the choice of whether to be policy-relevant (or not) really exists in such stark terms . . . as myriad other factors must also be considered’ (Fitzgerald et al., 2016, p. 2). Scholars do not always 316

Counter-radicalisation policy and practice

have control over their knowledge and cannot always predict who will pick up their research and how it will be used. Furthermore, deciding not to engage may even further marginalise researchers and move them further away from the core of the field. Moreover, there are other potential costs to not producing policy-relevant work. Here, the increasing pressure that permeates research to achieve social impact needs to be considered when looking at funding and career opportunities in an increasingly precarious and neoliberal high education system (Thornton, 2015). Engagement with policy-makers and practitioners and the production of ‘ready-to-use’ knowledge must also be contextualised within the different national settings all around the world, as countries may have different systems of production of knowledge and the possibility of producing research in an institution or research centre independent from the state should not always be taken for granted. All in all, therefore, the calls for impact, policy relevance, and policy engagement become a sword of Damocles that shapes the researchers’ choices, approaches, and roles in counter-radicalisation policies and practices.

Conclusion As knowledge and societies evolve, research and researchers become key players in the formulation and evaluation of policies and practices. It is in this context that radicalisation studies emerged and became institutionalised as a field. Knowledge is very much needed, not only to understand but also to counter and prevent violence from happening. Research is also necessary to comprehend counter-radicalisation and assess its impacts, counter-effects, and long-term consequences for society. Therefore, evidence will continue to be collected in all spheres where counter-radicalisation is implemented. Policy-makers, practitioners, and police forces will keep demanding policy advice and training, while democratic societies will demand science-oriented transparency and democratic policies. Radicalisation studies is acquiring maturity as an institutionalised field, and researchers will keep being called into the understanding of radicalisation and designs of counter-radicalisation. The phenomenon of radicalisation, the sources of extremism, and the conditions for violence are becoming more complex. Therefore, the field will become more sophisticated in the future. New debates and dilemmas will emerge on the way researchers manage the conditions and mandates of knowledge production and their role as producers of evidence on counter-radicalisation policies and practices. This chapter has described some of the main questions, challenges, and dilemmas researchers will face when conducting research on counter-radicalisation policies and practices. Very broadly, it has sought to provide a picture of how researchers’ individual trajectories will encounter and need to deal with more structural features shaping the field. First, the fact that this is an essentially policy-oriented field. As discussed, this feature gives rise to many opportunities and challenges for researchers that will emerge at all the various research and knowledge production steps. How the research is formulated, how funding is secured for researchers’ agendas, how data is produced, and how and to whom findings are communicated are all questions that are crossed by the endeavour of producing useful knowledge to build secure societies while preserving liberties and justice. Hopefully, the chapter will be a helpful guide for researchers approaching these questions when dealing with counter-radicalisation policy and practice.

Acknowledgements The authors are incredibly grateful to the editors of this volume for their valuable feedback on the previous versions of this chapter and their very patient, kind, and helpful editorial style. 317

Alice Martini and Laura Fernández de Mosteyrín

Note 1 Information RAN, retrieved 14 July 2022 from https://home-affairs.ec.europa.eu/networks/radicalisationawareness-network-ran/topics-and-working-groups_en

References Alford, R. R., & Friedland, R. (1985). Powers of theory: Capitalism, the state, and democracy. Cambridge University Press. Berling, T. V., & Bueger, C. (Eds.). (2015). Security expertise: Practice, power, responsibility. Routledge. Bliesemann de Guevara, B., & Bøås, M. (Eds.). (2020). Doing fieldwork in areas of international intervention: A guide to research in violent and closed contexts. Bristol University Press. Boyle, M. J. (2012). Progress and pitfalls in the study of political violence. Terrorism and Political Violence, 24(4), 527–543. https://doi.org/10.1080/09546553.2012.700608 Burawoy, M. (2004). Public sociologies: Contradictions, dilemmas, and possibilities. Social Forces, 82(4), 1603–1618. Chubb, J., & Reed, M. S. (2018). The politics of research impact: Academic perceptions of the implications for research funding, motivation and quality. British Politics, 13(3), 295–311. https://doi.org/10.1057/ s41293-018-0077-9 Coolsaet, R. (2016). “All radicalisation is local”: The genesis and drawbacks of an elusive concept (No. 84; Egmont Paper, p. 48). Royal Institute for International Relations. www.egmontinstitute.be/content/ uploads/2016/05/ep84.pdf?type=pdf Della Porta, D. (2013). Clandestine political violence. Cambridge University Press. Della Porta, D., & LaFree, G. (2012). Processes of radicalisation and de-radicalisation. International Journal of Conflict and Violence, 6(1), 4–10. Dolnik, A. (Ed.). (2013). Conducting terrorism field research: A guide. Routledge. Emerson, R. G. (2019). Vigilant subjects. Politics, 39(3), 284–299. https://doi.org/10.1177/0263395717747129 Fernández de Mosteyrín, L. (2020). Communication as legitimation in Spanish CVE: Bringing lessons from the past. In A. Martini, K. Ford, & R. Jackson (Eds.), Encountering extremism (pp. 200–220). Manchester University Press. https://doi.org/10.7765/9781526136619.00017 Fitzgerald, J., Ali, N., & Armstrong, M. (2016). Editors’ introduction: Critical terrorism studies: Reflections on policy-relevance and disciplinarity. Critical Studies on Terrorism, 9(1), 1–11. https://doi.org/10.1080/ 17539153.2016.1151306 Gill, P., Corner, E., McKee, A., Hitchen, P., & Betley, P. (2022). What do closed source data tell us about lone actor terrorist behavior? A research note. Terrorism and Political Violence, 34(1), 113–130. https://doi.org /10.1080/09546553.2019.1668781 Gordon, A. (2010). Can terrorism become a scientific discipline? A diagnostic study. Critical Studies on Terrorism, 3(3), 437–458. https://doi.org/10.1080/17539153.2010.521644 Grimm, J., Koehler, K., Lust, E., Saliba, I., & Schierenbeck, I. (2020). Safer field research in the social sciences: A guide to human and digital security in hostile environments (1st ed.). Sage Publishing. Jackson, R. (2007). An analysis of EU counterterrorism discourse post-September 11. Cambridge Review of International Affairs, 20(2), 233–247. https://doi.org/10.1080/09557570701414617 Jackson, R. (Ed.). (2016a). Routledge handbook of critical terrorism studies. Routledge. Jackson, R. (2016b). To be or not to be policy relevant? Power, emancipation and resistance in CTS research. Critical Studies on Terrorism, 9(1), 120–125. https://doi.org/10.1080/17539153.2016.1147771 Liebling, A. (2001). Whose side are we on? Theory, practice and allegiances in prisons research. British Journal of Criminology, 41(3), 472–484. https://doi.org/10.1093/bjc/41.3.472 Lindekilde, L. (2016). Radicalization, de-radicalization and counter-radicalization. In R. Jackson (Ed.), Routledge handbook of critical terrorism studies (pp. 248–259). Routledge. Malešević, S. (2017). The rise of organised brutality: A historical sociology of violence. Cambridge University Press. Martini, A. (2017, April 27). Critical terrorism studies: A conversation with Richard Jackson. Security Praxis. https://securitypraxis.eu/critical-terrorism-studies-richard-jackson/ Martini, A. (2021). The UN and counter-terrorism: Global hegemonies, power and identities. Routledge.

318

Counter-radicalisation policy and practice Martini, A., & Fernández de Mosteyrín, L. (2021). Del terrorismo al extremismo: Las políticas de prevención del extremismo violento en Europa. Revista CIDOB d’Afers Internacionals, 128, 15–37. https://doi.org/ 10.24241/rcai.2021.128.2.15 Miller, D., & Mills, T. (2009). The terror experts and the mainstream media: The expert nexus and the dominance in the news media. Critical Studies on Terrorism, 2(3), 414–437. Morrison, J. F. (2020). Talking stagnation: Thematic analysis of terrorism experts’ perception of the health of terrorism research. Terrorism and Political Violence, 1–21. https://doi.org/10.1080/09546553.2020.1804879 Muro, D., & Wilson, T. (Eds.). (2022). Contemporary terrorism studies. Oxford University Press. Nordstrom, C., & Robben, A. C. G. M. (Eds.). (1995). Fieldwork under fire: Contemporary studies of violence and survival. University of California Press. Parker, D., Pearce, J. M., Lindekilde, L., & Rogers, M. B. (2019). Challenges for effective counterterrorism communication: Practitioner insights and policy implications for preventing radicalization, disrupting attack planning, and mitigating terrorist attacks. Studies in Conflict & Terrorism, 42(3), 264–291. https:// doi.org/10.1080/1057610X.2017.1373427 Phillips, B. J. (2021). How did 9/11 affect terrorism research? Examining articles and authors, 1970–2019. Terrorism and Political Violence, 1–24. https://doi.org/10.1080/09546553.2021.1935889 Pink, S., Horst, H. A., Postill, J., Hjorth, L., Lewis, T., & Tacchi, J. (Eds.). (2016). Digital ethnography: Principles and practice. Sage. Russo, A., & Selenica, E. (2021). Actors and sites for knowledge production on radicalisation in Europe and beyond. Journal of Contemporary European Studies, 1–19. https://doi.org/10.1080/14782804.2021.1997729 Russo, A., & Strazzari, F. (2020). The politics of safe research in violent and illiberal contexts. In B. Bliesemann de Guevara & M. Bøås (Eds.), Doing fieldwork in areas of international intervention: A guide to research in violent and closed contexts (pp. 75–91). Bristol University Press. Sageman, M. (2014). The stagnation in terrorism research. Terrorism and Political Violence, 26(4), 565–580. Schmid, A. (2013). Radicalisation, de-radicalisation and counter-radicalisation: A conceptual discussion and literature review. International Centre for Counter-Terrorism – The Hague, ICCT, Research Paper. Schmid, A. P. (2016). Research on radicalisation: Topics and themes. Perspectives on Terrorism, 10(3), 26–32. Schuurman, B. (2020). Research on terrorism, 2007–2016: A review of data, methods, and authorship. Terrorism and Political Violence, 32(5), 1011–1026. https://doi.org/10.1080/09546553.2018.1439023 Silke, A. (Ed.). (2019). Routledge handbook of terrorism and counterterrorism. Routledge. Sriram, C. L. (Ed.). (2009). Surviving field research: Working in violent and difficult situations. Routledge. Stampnitzky, L. (2013). Disciplining terror: How experts invented “terrorism”. Cambridge University Press. Stanley, T., Guru, S., & Coppock, V. (2017). A risky time for Muslim families: Professionalised counterradicalisation networks. Journal of Social Work Practice, 31(4), 477–490. https://doi.org/10.1080/02650 533.2017.1394829 Thornton, M. (Ed.). (2015). Through a glass darkly: The social sciences look at the neoliberal university (2nd ed.). ANU Press. https://doi.org/10.22459/TGD.11.2015 Toros, H. (2008). Terrorists, scholars and ordinary people: Confronting terrorism studies with field experiences. Critical Studies on Terrorism, 1(2), 279–292. https://doi.org/10.1080/17539150802184652 Toros, H. (2016). Dialogue, praxis and the state: A response to Richard Jackson. Critical Studies on Terrorism, 9(1), 126–130. https://doi.org/10.1080/17539153.2016.1147775

319

21 HOW CAN WE MEANINGFULLY EVALUATE THE EFFECTS AND EFFECTIVENESS OF PROGRAMMES TO PREVENT OR COUNTER RADICALISATION? Tom Fisher and Joel Busher Introduction: the rise and rise of evaluation There has been a surge of interest in recent years in how to evaluate and better understand the effects of policies and programmes intended to prevent and counter radicalisation into violent extremism (P/CVE). There has been a rapid increase in the number of projects and publications dealing with the topic, as well as a growing emphasis on evaluation among policy planners and organisations that design and deliver P/CVE programmes (e.g. RAN, 2017; UNDP, 2018). More generally, researchers and policy practitioners have increasingly exhorted one another to ensure evaluation sits at the heart of this field (e.g. Schmid, 2016; Weine et al., 2017). John Horgan has even argued that a ‘CVE programme without an evaluation component is, in my view, worthless’ (European Eye on Radicalization, 2018). There are a number of drivers for this turn towards evaluation. The scale of investment in P/CVE programmes, the persistent threat of violent extremism, and a wider backdrop of contention around counter-radicalisation programmes have meant policy planners at all levels have found themselves under pressure to justify policies and programmes (Mastroe, 2016; Thomas, 2020). The emphasis on evaluation also reflects a wider trend in public policy. Evaluation has been integral to the advance of evidence-based policy agendas that have become the norm since the expansion of welfare states in the 1960s (Wollman, 2007). In most areas of public policy today, it is broadly accepted that evaluation is a vital part of the programme or policy cycle (Hall, 2017; Vaessen & Leeuw, 2011). If anything, it is surprising the P/CVE field took as long as it did to begin to align with this broader trend. There has also been growing awareness among policy, practitioner, and research communities about the paucity of evidence around P/CVE programmes and the need to address this if P/CVE is to become more effective (Baykal et al., 2021). As terrorism research expanded rapidly in the early 2000s, concerns mounted about the relative absence of reliable evaluation research (Horgan, 2008; Romaniuk & Fink, 2012). A major systematic review in 2006 found an ‘almost complete lack of evaluation research on counter-terrorism interventions’ (Lum et al., 2006, p. 489). Until at least the early 2010s, a similar observation could be made about the subfield of P/CVE (Christmann, 2012; Horgan & Braddock, 2010; Veldhuis, 2012), a situation at least partly driven by a DOI: 10.4324/9781003035848-24

320

Effects and effectiveness of programmes to prevent or counter radicalisation

widespread perception that evaluating such programmes was simply an ‘inherently difficult task’ (Vidino & Brandon, 2012, p. 177). A systematic review published in 2015 found the majority of P/ CVE programmes had not been evaluated and that, where evaluations had happened, they ‘often do not meet scientific standards’ (Feddes & Gallucci, 2015, p. 4). Since then, there has been a rapid expansion in evaluation research (El-Said, 2015; Koehler, 2017, 2019; Madriaza et al., 2022). For the most part, this has consisted of evaluations of specific P/CVE interventions or programmes (e.g. Brett & Kahlmeyer, 2017; Cherney, 2018; HarrisHogan, 2020; Fisher et al., 2020; Sahgal & Kimaiyo, 2020; Williams et al., 2016). There have also been attempts to review national P/CVE policies (e.g. Finn et al., 2016; Van der Heide & Schuurman, 2018) and even regional-level analyses (e.g. Glazzard & Reed, 2018). With this expansion, we have seen significant advances in evaluation methods and practice. There remains, however, a broad consensus that the field still requires a greater quantity and quality of evaluations if it is to advance (e.g. Baruch et al., 2018; Bellasio et al., 2018; Cherney, 2018; Gielen, 2019; Jugl et al., 2021; Lewis et al., 2020; Madriaza et al., 2022; Pistone et al., 2019; Schuurman, 2020; Silke & Velduis, 2017). It has been observed, for example, that many evaluations still comprise ‘process’ or ‘formative’ evaluations, and while such evaluations provide valuable insight about how interventions are delivered, the appropriateness of their design, and whether they have reached their intended beneficiaries, they provide at most only limited insight about programme outcomes or impact (Cherney, 2020). Furthermore, even where evaluations discuss impacts, they still tend to be largely descriptive, with relatively few employing robust methods to interpret impact and outcomes (Mastroe & Szmania, 2016; Lewis et al., 2020). The purpose of this chapter then is to survey the evolution of evaluation research on P/CVE programmes and policies. It first outlines some of the main challenges encountered by researchers, policy planners, and practitioners as they have sought to advance an evaluation agenda. It then discusses some of the key developments in the field, how they respond to these challenges, and the extent to which they have been successful. The chapter concludes with a short reflection on issues for consideration going forward. The chapter is based on a narrative review of the literature on the evaluation of P/CVE programmes and policies. It also draws on the authors’ own experience evaluating and assessing the impacts of such policies and programmes. For both authors, this includes evaluating P/CVE programmes in the United Kingdom. For Fisher, it also includes evaluating P/CVE programmes in other European countries, East Africa (Fisher et al., 2020) and, currently, Afghanistan. For Busher, it also includes research projects examining the enactment and impacts of P/CVE-related legislation in educational settings in the United Kingdom (Busher et al., 2017) and public perceptions of violent extremism and P/CVE programmes in the Sahel.

Key challenges in evaluating programmes to prevent or counter radicalisation As noted earlier, early attempts to evaluate P/CVE programmes, or encourage the evaluation of such programmes, were often met, and sometimes discouraged, with the view that this was an ‘inherently difficult task’. As such, it is useful to set out what are often identified as the main challenges in evaluating such programmes.

A lack of clarity around the scope and objectives of P/CVE programmes Effective evaluation usually requires a clear, shared, and realistic understanding of what success looks like (Gajda & Jewiss, 2004, p. 2). Yet beyond the high-level and arguably somewhat vague 321

Tom Fisher and Joel Busher

aspiration to prevent or counter violent extremism, P/CVE programmes have been characterised by significant ambiguity around what they are trying to achieve and how they intend to achieve it (Harris-Hogan, 2020; Madriaza et al., 2022). This lack of clarity is related partly to the sheer diversity of interventions. As Berger (2016, p. 3) notes, a quite ‘astonishing range of activities’ are sometimes framed as P/CVE programmes. This includes, but is not limited to, education initiatives and life skills programmes to build resilience to violent extremism, awareness-raising initiatives, counter-narrative campaigns, mentoring for ‘at risk’ individuals, programmes on religious argumentation, exit programmes, and even economic development programmes. While such programmes might share a higher-order goal around reducing or stopping radicalisation, they will almost inevitably be oriented towards a range of outcomes, underpinned by diverse assumptions about how to achieve them and how they contribute towards the goal of reducing violent extremism. The lack of clarity around the scope and objectives of P/CVE programmes is also rooted in more fundamental instabilities in how P/CVE is understood, and in particular ambiguity around the extent to which P/CVE programmes are or are not, or should or should not be, concerned with behavioural outcomes (e.g. disengagement from extremist milieus), attitudinal or ideological outcomes (e.g. not embracing a particular worldview), or a combination of the two (Chukwuma & Jarvis, this volume; Clubb, 2015; Horgan, 2008) – ambiguity at least partly rooted in differing conceptions of radicalisation itself (Dawson, 2021; Ramakrishna, this volume; Schuurman, 2021). Indeed, it is not uncommon to find different interpretations of goals even within the same project. For example, in a mentoring project evaluated by the authors, while senior policy planners focused specifically on outcomes around radical ideology and attitudes towards violence, those delivering the interventions hoped for a much broader range of changes for mentees, including enhanced self-efficacy, signs of stability in their everyday lives, and better connections with family and friends. Aside from any problems that such a lack of clarity can generate in programme design and implementation (Pressman & Wildavsky, 1984; Thiessen, 2022), it has also been problematic for evaluation. Divergent interpretations of a programme’s scope and objectives make it difficult to construct the kind of shared theories of change that typically underpin effective evaluation (Koehler, 2017). In our experience, what usually happens is that the evaluation is either positioned awkwardly between competing versions of success, or the frames and understanding of the programme funders prevail, both of which comprise problematic outcomes. The sheer diversity of P/CVE programmes has also created problems for comparison and effective meta-analysis (McBride et al., 2022). Evaluation is often used to underpin decisions about future investment: from their limited pool of resources, should policy planners invest more time and money in counter-narrative campaigns, one-to-one mentoring, or community development programmes? Yet if programmes are oriented towards quite different outcomes, they will almost inevitably have different ways of defining and demonstrating success, making meaningful comparison difficult. One solution pursued by policy planners has been to develop standardised or common assessment instruments deployable across multiple, quite different interventions (Marret et al., 2017; Romaniuk, 2015). As far as we are aware, however, no such efforts to date have been particularly successful. Indeed, from our experience, some have actually damaged relationships between policy and practitioner communities, particularly where practitioners perceived that such approaches failed to understand or engage with the logic, context, or value of their programmes (see also ­Clément et al., 2021), thereby actually undermining opportunities for effective evaluation.

322

Effects and effectiveness of programmes to prevent or counter radicalisation

Other solutions have focused on trying to bring the scope and objectives of programmes into sharper focus. For example, it has become relatively common to break programmes down into primary, secondary, and tertiary interventions – often referred to as a public health model – thereby enabling more meaningful comparison between projects with broadly similar goals (e.g. Gielen, 2019; Hassan et al., 2021a, 2021b; Madriaza et al., 2022). Others, such as Berger (2016), have argued in favour of narrowing the focus of P/CVE interventions to behavioural outcomes precisely because it makes it easier to define success and evaluate programme effectiveness. Policy planners continue, though, to invest in an array of activities that tackle both ideas and behaviours, often without explicit articulation of their objectives or how they believe the interventions will produce that change (Harris-Hogan, 2020; Madriaza et al., 2022). We return to this below.

Access to data A second challenge concerns access to data. This challenge is not unique to P/CVE (Ris & Ernstorfer, 2017; Davies et al., 2017), but is arguably particularly acute here. In most evaluations, the most valuable data comes from those taking part in the project, such as beneficiaries, delegates, mentees, and participants (Johnson et al., 2009; Mertens, 1999; OECD, 1991). Accessing such data in relation to P/CVE programmes has been difficult. The P/CVE sector is saturated with risk, and programmes are often highly sensitive. Policy planners and programme gatekeepers are cautious about sharing data or granting access to participants due to concerns about potential security, political and reputational costs, and the well-being of those involved in the programmes (Ahmed, 2016; Glazzard, 2022). Evaluators, policy planners, and programme gatekeepers must pick their way through complex questions about the legality and ethics of sharing information; and gatekeepers have to consider potential operational impacts if, for example, participants are still persons of interest for the security or intelligence community (Madriaza et al., 2022, p. 9; Pistone et al., 2019, p. 22). Overcoming these issues requires one or more stakeholders to place their trust in the other stakeholders and assume a certain level of risk (Madriaza et al., 2022, p. 42). Yet there are multiple barriers to this. Many policy planners, practitioners, and researchers have direct or vicarious experience of situations where such leaps of faith have not worked out as hoped – experiences rarely if ever written about but much discussed in corridors or over post-meeting cups of coffee. Furthermore, most actors involved are under significant budgetary, legal, and policy pressures within their own organisations, all of which encourage risk aversion (Baykal et al., 2021). This has undoubtedly had a negative effect on the number and quality of evaluations. It has also significantly limited the pool of people able to undertake such research – something not conducive to raising standards, transparency, and public confidence. Even when researchers have a good working relationship with suitable gatekeepers, it can prove difficult for external evaluation teams to gain meaningful access to those directly involved in projects due to the particular and often significant vulnerabilities of those involved in the programmes (Madriaza et al., 2022, p. 19). Programme participants might be reluctant to share their experiences with external evaluators – hardly a surprise given the heavy stigma associated with ‘terrorism’, ‘radicalisation’, and ‘violent extremism’ and the fact that those involved in these programmes are likely to be suspicious of anyone they associate with the state or the security apparatus (Aistrope, 2016; Barkindo & Bryans, 2016, p. 12; Dunn et al., 2015). In our experience, P/CVE practitioners also often harbour concerns that evaluations might produce detrimental effects for programme participants by, for example, making them question the degree of confidentiality within programmes

323

Tom Fisher and Joel Busher

or more generally disrupting the intervention process, particularly for interventions that comprise confidential conversations behind closed doors, such as mentoring programmes.

The problem of attribution A third challenge concerns the problem of attribution: how can we be confident that the observed outcomes are a product of the interventions under analysis and not of an external factor? P/CVE projects take place within complex security, political, institutional, and economic environments with multiple variables that affect how, if at all, the interventions work on the ground (Glazzard, 2022). Effective evaluation cannot overlook such contextual complexity (Christmann, 2012; Clubb & Tapley, 2018; Gielen, 2019). In addition, policies, programmes, and projects often look quite different in practice than on paper. In any policy field, practitioners will have their own ideas about what works and under what circumstances – they ‘may narrow the goals of policy makers or elaborate upon them, or substitute entirely different goals’, and are likely to make adjustments as they seek to make these policies or programmes work within their particular settings (Lub, 2014, p. 48; see also Ball et al., 2011). P/ CVE is no different (Busher & Jerome, 2020). This can make it difficult to identify what it is about the intervention that has generated the observed outcomes. Again, the problem of attribution is not unique to the P/CVE field (Ris & Ernstorfer, 2017). There are, however, factors that magnify the challenge here. Most researchers and policy planners have taken the view that randomised control trials (RCTs), widely considered the ‘gold standard’ for evaluations, are unlikely to be feasible or publicly acceptable (Gielen, 2019; Cherney, 2018; Nehlsen, 2020), except in some broadly targeted primary interventions (e.g. Parker & Lindekilde, 2020; Kurtz et al., 2018). This is partly due to ethical concerns around withholding support from individuals who need it (Cherney, 2020; Lewis et al., 2020, p. 4). It is also about security and reputational concerns. In Kutner’s discussion of evaluation strategies for Saudi Arabia’s deradicalisation programme, for example, they conclude ‘recidivism risks are too high to support a control or waitlist group’ (2016, p. 119), and one can only imagine the headlines if it emerged a terrorist attack had been perpetrated by somebody in the control group of a counter radicalisation RCT. Radicalisation into violent extremism also remains, in most settings, a very low frequency event. While we might by and large be grateful for this state of affairs, it generates a problem for evaluators (McBride et al., 2022). In fields such as public health, drug use reduction, or general crime prevention, one way to address the problem of attribution, even where RCTs are not possible, is through large-N quasi-experimental studies that permit the use of powerful statistical techniques to make inferences about causal relationships with a high degree of confidence (Pawson & Tilley, 1994). Such strategies are generally not available for programmes aimed at countering radicalisation, however, because there are simply too few cases – for our analytical interests, at least – in which the outcome is violent extremism.

How to measure a negative Perhaps the thorniest challenge when it comes to evaluating counter radicalisation programmes has been the view that, in the final analysis, the ‘desired outcome’ of most P/CVE projects is ‘a non-event’ (Fink et al., 2013, p. 2): that somebody did not (continue to) engage in violent extremism who would most likely have done so otherwise. Yet how does one evidence such an outcome? As Hassan and colleagues ask, ‘[W]hat empirical indicators and sources of information can be 324

Effects and effectiveness of programmes to prevent or counter radicalisation

used to measure the absence of radical ideas?’ (2021b, pp. 13–14). And on what basis can we say that the individual or group would otherwise have engaged in violent extremism? Again, this problem is not unique to P/CVE. Any field concerned with prevention faces a similar challenge – whether that is violence reduction (Roman et al., 2018), environmental policy (Ferraro, 2009), or conflict prevention (Ris & Ernstorfer, 2017). As in those areas, steps can be taken to address such issues. The most obvious of these is to focus on identifying, articulating, and measuring positive P/CVE-related outcomes that we believe are likely to generate the desired higher-order impacts (e.g. Fisher et al., 2020; Helmus et al., 2017). Yet, as with the other issues, there are aspects of the P/CVE field that make this issue particularly challenging. Perhaps the most intransigent of these is the relative immaturity of the science of radicalisation itself. As long as there is significant disagreement about how the process of radicalisation itself works (see the first two sections of this volume), claims that observed outcomes do indeed represent a reduction in the threat of violent extremism or that violent extremism was ever likely in the first place, are always likely to be subject to contestation. Furthermore, in our experience, it is often precisely those larger claims about impact, about threat reduction, that policy planners and programme leads want to be able to make. For those involved in delivering programmes, such claims can seem like an essential means to securing future funding; for policy planners, such claims can be seen as crucial to their justification of policy direction and investment (Baykal et al., 2021).

Addressing these challenges The challenges outlined above are significant. They are not without solutions, however. In this section, we turn to three major developments in P/CVE evaluation research and practice and discuss how and to what extent they have begun to address these challenges.

Towards the use of more sophisticated evaluation designs One of the most obvious developments has been the use of more sophisticated evaluation designs. This has partly been about the growing use of multi- and mixed-method approaches. One of the criticisms of the first generation of P/CVE evaluation was that it often relied on a single research instrument (Feddes & Gallucci, 2015). In recent years, however, there has been a growing trend for evaluations to use multiple methods, often combining qualitative and quantitative strategies, to generate more compelling evidence about impacts and a richer understanding of causal pathways between interventions and outcomes. Williams and colleagues (2016), for example, integrated focus groups and surveys through a grounded theory approach across a multi-phase evaluation of a community-based CVE programme in the United States. In their study of the outcomes of an intervention with radicalised offenders in Australia, Cherney and Belton (2020) combined client and staff interviews with progress reports written by programme psychologists and case notes compiled by a range of programme staff. In their evaluation of the EU-funded STRIVE II programme in Kenya, Fisher and colleagues (2020) combined baseline, midline, and end-line participant questionnaire data from internal evaluation efforts with data from focus groups, interviews, and ethnographic observations in order to evaluate process, outcomes, and broader (unexpected) impacts. This growing diversity of methods and approaches can bring its own challenges. For example, without coordination, there is potential for the growing abundance of different approaches and methods to exacerbate further challenges around comparison across different programmes (McBride et al., 2022; Nehlsen et al., 2020). Nonetheless, such multi-method approaches offer significant advantages in enabling triangulation, giving greater confidence around findings, and 325

Tom Fisher and Joel Busher

meeting the needs of a wider range of stakeholders (Patton, 2008). Moreover, evaluation strategies that limit reliance on one instrument or data type can help address data access issues. Casting a broader net across potential data sources and methods means that evaluators have alternative options if one data type or participant group is unavailable. For example, in our evaluation of a UK mentoring programme, we were denied access to internal client questionnaire data. The evaluation was able to continue, however, using data from other methods, such as interviews with mentors and mentees. There have also been other notable advances in evaluation design. One of these has been the growing use of experimental and quasi-experimental designs to enable the making of stronger causal claims. As noted earlier, experimental designs carry ethical, security, and reputational challenges, especially outside the context of broadly targeted primary interventions. Nonetheless, we have seen some use of experimental designs in contexts where such an approach is possible, such as primary interventions in educational settings. Perhaps the best example to date for a detailed exposition of such methods is Parker and Lindekilde’s (2020) use of an RCT design with 1,931 Danish youths to study the effects of using former extremists to reduce violent extremism in Danish schools. Quasi-experimental designs are also being encouraged and used more widely (Braddock, 2020). In some cases, this has been done using naturally occurring variation to achieve a suitable comparison. Aldrich (2014), for example, compared relevant attitudes and behaviours in one town in northern Mali that had been exposed to a P/CVE intervention with those of another nearby town that had not. In Sri Lanka, Webber and colleagues (2018) used variation in processing times across rehabilitation centres to construct a comparison between militants who went through a full rehabilitation programme and those who had only been through a minimal rehabilitation programme. Alongside such use of naturally occurring variation, we are seeing exploration of how innovative control group designs from other areas characterised by significant ethical challenges, such as medical trials, might enable greater use of experimental and quasi-experimental designs in the P/CVE field. Lewis and colleagues (2020, pp. 21–22), for example, point to the promise of the ‘stepped wedge’ or ‘dynamic wait list’ design (Li et al., 2018). In such a design, all participants are initially allocated to the control group, and then randomly allocated clusters are added to the treatment group until all participants are in the treatment group. Such an approach addresses some of the ethical and political concerns associated with the use of control groups by ensuring that all participants receive the intervention, although Lewis and colleagues question whether it would be suitable for the highest risk populations as it may increase the average wait time for inclusion across the sample as a whole. Similarly, Braddock (2020) advocates for ‘switching-replication’ designs, in which all participants are at some point allocated to both treatment and control groups across a multi-phase intervention, although Braddock also cautions that to ensure all participants are exposed to the intervention in question and relevant outcomes are measured, it usually requires extended access to participants. As discussed earlier, this is currently rare in P/CVE interventions. In a similar vein, there have also been attempts to use longitudinal designs, albeit they are still rare (Madriaza et al., 2022, p. 26), and they tend to have been deployed in settings and with populations where there are fewer access issues, such as in educational settings or around training programmes. For example, Feddes et al. (2015) undertook a longitudinal evaluation of a resilience training programme delivered to young people in the Netherlands with a migrant background, taking measurements before, during, and after a three-module programme administered over a period of three months. Similarly, Savoia and colleagues (2019) carried out a longitudinal cohort study of a programme in schools in the United States aimed at ‘decreasing exposure to hate messages in the school environment’ and ‘improving students’ attitudes towards ethnocultural diversity’ (p. 43). In 326

Effects and effectiveness of programmes to prevent or counter radicalisation

the tertiary space, Cherney and Belton (2020) used a combination of interviews and programme data, such as case notes, to trace the outcomes for three individuals who had participated in the Proactive Integrated Support Model (PRISM) intervention in Australia. Such studies offer obvious advances in terms of being able to trace change over time and make stronger inferences about possible causal pathways. Nonetheless, there are a number of obstacles to the wider use of such designs. Some of the most significant are about time and financial resources – longitudinal designs tend to be more resource intensive than cross-sectional designs (Madriaza et al., 2022, p. 26). As Cherney and Belton point out, another challenge lies in the fact that in the P/CVE field, the priority tends to be ‘to get interventions up and running with little thought given to how they will be evaluated’ (2020, p. 207; also Baykal et al., 2021; Williams et al., 2016). This means that the ideal of working evaluation in from the inception of the programme rarely occurs in practice. As Cherney and Belton show, however, in some situations and to some extent at least, this issue can be addressed through the effective use of programme data. Finally, in response to the first challenge discussed earlier, evaluation researchers have also been exploring strategies to negotiate ambiguity around programme objectives and enable meaningful comparison across programmes. Perhaps the most influential of such strategies to date, certainly in the area of tertiary interventions, has been the use of ‘Multi-attribute Utility Technology’ (MAUT) or ‘Multi-attribute Evaluation’ (Edwards & Newman, 1982), instigated by Horgan and Braddock (2010). MAUT is a structured methodology designed to handle multiple, and at times competing, objectives or ‘attributes’. Once stakeholders are identified, the intervention attributes are identified, their relative importance assessed, and they are organised into a hierarchical ‘value tree’, with specific metrics linked to each. This can enable analysis of the effectiveness of a given intervention across different attributes and, where this approach is taken across multiple projects, can also provide a basis for a structured comparison of the relative effectiveness of different interventions with regard to a particular attribute (Lewis et al., 2020, p. 13). As Marsden (2015) observes, however, the still limited empirical and theoretical knowledge of the processes of interest in P/CVE makes it difficult to develop with confidence the sort of quantitative indicators usually developed within a MAUT framework in other policy areas. Ambiguity and instabilities in the basic understanding of P/CVE programmes also make it difficult to form the kind of agreement around suitable metrics (see Mastroe & Szmania, 2016) that would be required in order to use MAUT effectively across multiple interventions. In our own evaluations, we have found success in facilitating discussions on suitable impacts and outcomes but have often struggled to achieve stakeholder consensus when it comes to identifying metrics. In summary, while important challenges remain, a number of solutions to the technical challenges of evaluating P/CVE programmes are being identified and have the potential to significantly enhance our understanding of the effects and effectiveness of these programmes and policies.

Developing more precise and testable claims about the effects and effectiveness of P/CVE programmes Another major area of development has been around how to establish more precise and testable claims about the effects and effectiveness of P/CVE programmes in the first place, an essential basis for effective evaluation. Here, drawing on wider advances in evaluation science and experience from other policy domains, evaluation researchers in the P/CVE field have increasingly turned to ‘theory-driven’ (Chen, 1990; Coryn et al., 2011) or ‘theory-based’ (Weiss, 1995) approaches. Two such approaches warrant particular discussion, both for their impact in the field and the way they respond to some of the aforementioned challenges: ‘Theory of Change’ (Weiss, 1995) and 327

Tom Fisher and Joel Busher

the Realist (or Realistic) evaluation framework (Pawson & Tilley, 1997) – approaches often in practice used together within evaluations (Blamey & Mackenzie, 2007; Ris & Ernstorfer, 2017). The Theory of Change is, in essence, an outcomes framework used to guide programme evaluation and planning. Programme stakeholders develop a detailed description of how and why a desired change can be achieved in a particular context as a result of a particular set of activities. Put another way, they outline the anticipated causal chains between inputs, specific outcomes, and longer-term goals. In doing so, they make explicit assumptions about how and under what conditions these causal chains are expected to operate. For example, if a programme provides beneficiaries with information about the tactics of extremist recruiters, it is assumed that participants will be better able to identify recruitment tactics early and thus reduce their effect, combining with other outcomes to contribute to an overall goal such as ‘young people in Region A having greater resilience to violent extremist recruitment and radicalisation’, provided that certain factors hold, for example, there is not a dramatic change in recruitment tactics in the interim. Once developed, the ‘theory’ can then be used to guide the evaluation, interrogating whether the intervention delivers the expected outcomes and developing and testing hypotheses as to why it does or does not. There are several instances of Theory of Change, or Theory of Change-like, approaches being used in the P/CVE field. Fisher and colleagues (2020), for example, use such an approach in their evaluation of the STRIVE II programme in Kenya; Schuurman and Bakker (2016) adopt a similar approach, using the terminology of ‘program theory’, in their evaluation of a reintegration programme for former extremists in the Netherlands; and Weine and colleagues’ (2018) formative evaluation of the use of a targeted violence prevention programme to prevent violent extremism in Los Angeles also uses a similar approach, referred to as a ‘participatory logic model’. Such approaches are gaining traction in the field of P/CVE for the same reasons they have in other fields. They effectively ground the evaluation in the logic of programme stakeholders and their local context, thereby responding to issues around the ambiguity of programme goals. They also make explicit and open to interrogation the anticipated causal pathways from intervention to outcomes to those goals, addressing what is sometimes referred to as the ‘missing middle’ or ‘black box’ between programme activities and goals (De Silva et al., 2014; Khalil & Zeuthen, 2016; Koehler, 2017; Ris & Ernstorfer, 2017), thereby helping to respond to the problem of attribution. A well-constructed Theory of Change, with carefully developed assumptions and conditions, can also provide an effective way of engaging with negative outcomes without necessarily writing programmes off as ‘failures’. For example, Schuurman and Bakker (2016, p. 79) identify that, while undesirable, the low rate of recidivism observed was not incompatible with the programme theory of the reintegration programme under analysis and that, therefore, such instances ‘should not be taken as an indictment of the project’s overall effectiveness’. From our experience, such approaches can also render significant process benefits. In a sector in which programmes often lack a clear articulation of how and the conditions under which their activities lead towards their goals (Baykal et al., 2021; Madriaza et al., 2022; Thompson & Leroux, 2022), working with stakeholders to develop a Theory of Change can not only provide a stronger basis for meaningful evaluation but also help stakeholders develop a better and more critical understanding of the strengths and limitations of their own projects and provide an otherwise rare opportunity to strengthen a sense of common purpose across diverse stakeholders with sometimes differing interests. For such process benefits to be achieved, however, Theories of Change must be developed with direct, inclusive engagement with those closest to programme delivery rather than as a top-down exercise (Ho et al., 2020; Mason & Barnes, 2007; Mulgan, 2016). Whereas Theory of Change can be used to better articulate and interrogate the internal programme logic, the strength of the realist evaluation framework is usually seen to lie in how it 328

Effects and effectiveness of programmes to prevent or counter radicalisation

unpacks the relationship between interventions, outcomes, and context (Veldhuis, 2012; Glazzard, 2022). Realist evaluation recognises that the effects of programmes and policies vary markedly depending on a range of factors, such as, but not limited to, changes in the socio-political context, the programme setting, and participant demographics, and that meaningful evaluation must find a way to accommodate this (Blamey & Mackenzie, 2007). To do this, it moves evaluation beyond popular but overly simplistic questions about ‘what works’ to a more nuanced set of questions about ‘what works, for whom, in what circumstances and how’ (Gielen, 2019, p. 1149; Veldhuis, 2012). As well as helping to formulate more precise and meaningful questions for evaluation, the realist approach can also enrich our understanding of how programmes work in practice. Of particular importance, the emphasis realist evaluation places on understanding the interactions between context and programme outcomes encourages the ongoing building and testing of ‘mini theories’ (Blamey & Mackenzie, 2007, p. 444) about how the programme works in those contexts, leading to a ‘refined programme theory . . . which in turn can be used for empirical testing’ (Gielen, 2018, pp. 467–468). Furthermore, the more specific claims that the realist evaluation framework entails provide an effective means of managing expectations among policy actors and donors about the types of claims that might reasonably be made on behalf of P/CVE programmes (Ris & Ernstorfer, 2017) There are, again, challenges associated with the use of the realist framework. Glazzard (2022), for example, cautions that the sometimes broad use of the term ‘context’ can be unhelpful and risks a potentially significant misunderstanding of some programmes and how they work. For example, some of the immediate contextual aspects, such as the programme setting or the social capital of key stakeholders, which are sometimes bundled into the category ‘context’, are in fact integral to the programme itself. Yet such problems are not insurmountable; Glazzard himself outlines a basic framework that would appear to deal with such issues by differentiating between forms of context. Again, while there are limits and challenges associated with both the Theory of Change and the Realist framework, both offer useful routes by which to enhance the evaluation of P/CVE programmes through the construction of more specific and testable claims. Crucially, the approaches are compatible, have been tried and tested in other fields, and do not prescribe or preclude specific evaluation methods. Moreover, in relation to the data access challenges described earlier, by requiring the articulation of a programme’s logic, theory-driven approaches encourage the incorporation of evaluation from the outset of the programme, which in turn generates more time and scope to plan the relationship between the programme and its evaluation and build in routes for appropriate data access and sharing from the start.

Strengthening collaboration between researchers, policy planners and practitioners, and growing attempts to share and publish data from evaluations The third major area of development relates to growing efforts to foster collaboration between researchers, policy planners, and practitioners and to increase the sharing and/or publishing of evaluation data. Internal self-evaluation remains the most common form of evaluation in the field (Lewis et al., 2020, p. 7). This is perhaps unsurprising, given both the sensitivities around these programmes and the size of the P/CVE sector. Nonetheless, there is broad agreement today among researchers, policy planners, and practitioners alike about the importance of working together to enhance understanding of P/CVE interventions, at least in principle. There is also growing agreement around the importance of increased sharing of data and insights from evaluation if the field is to move forward (Clément et al., 2021; Madriaza et al., 2022). This agenda has been taken forward by a number of substantial international initiatives. Within Europe, for example, the EU has funded major projects concerned with enhancing P/CVE 329

Tom Fisher and Joel Busher

evaluation. Among the most notable of these are the IMPACT Europe project, from 2014 to 2017 (Marret et al., 2017), and the INDEED project, 2021–4. Both projects entail a wide range of professionals from research, policy, and practitioner backgrounds working together to develop common evaluation guidance, toolkits, and frameworks. Similarly, the EU-funded Radicalisation Awareness Network (RAN) has brought practitioners and policy planners together with academics to share experience and develop good evaluation practice (e.g. RAN, 2021). Meanwhile, the Canadian Practitioners Network for the Prevention of Radicalization and Extremist Violence (CPN-PREV) and the UNESCO Chair in Prevention of Radicalisation and Violent Extremism (UNESCO-PREV Chair) have led national and international initiatives to foster collaboration between practitioners, policy planners, and academics to advance P/CVE evaluation science and practice (Clément et al., 2021; Hassan et al., 2021a, 2021b; Madriaza et al., 2022). At the national and local levels, we are also seeing concerted attempts by policy planners, practitioners, and researchers to collaborate on P/CVE evaluations. In the United Kingdom, for example, the Home Office commissioned external evaluations of several local Prevent initiatives for the first time in 2015. In 2017, it began publishing data on referrals into the Channel de-radicalisation programme, has actively supported research on programme outcomes, and, at the time of writing, has recently issued a call to tender for an external evaluation of Channel. In Canada and the Netherlands, central knowledge hubs have been established to build evaluation capabilities at the sector level and foment a positive evaluation culture (Baykal et al., 2021). While there are still plenty of calls for further openness and data sharing, these developments are significant and potentially game-changing. By providing evaluation researchers with greater access to programme data, stakeholders, and participants, they are enabling those researchers to interrogate more effectively the ways in which P/CVE programmes do or do not work; are helping policy planners develop a stronger understanding about how and under what conditions programmes do and do not work; and are fostering faster and more widespread sharing of good practice and innovation across policy, practitioner, and research communities. There are still challenges, however. Even where researchers are able to access programme data, that access is often only partial. For example, some of the primary data might still be deemed too sensitive to share by governments, or researchers might only have access to intervention records rather than directly to the participants (Madriaza et al., 2022; Schuurman & Bakker, 2016). And while there might be growing normative agreement about the value of sharing and publishing, the low numbers of evaluations made public indicate that there is still a long way to go on this (Nehlsen et al., 2020, p. 8). There are also a number of persistent structural and cultural challenges that inhibit effective collaboration between policy planners, practitioners, and evaluation researchers. For example, there is a tendency, outside exemplary cases, for evaluation not to be given significant consideration until late in programme delivery or towards the end of a funding cycle, rather than being worked in from the outset, thereby hindering effective collaboration (Madriaza et al., 2022, p. 17). There is evidence that, in some countries at least, significant financial insecurity associated with the institutional structures of the P/CVE sector – such as fears among practitioners that their funding might be withdrawn as a result of poor performance in a one-off evaluation – might be driving reluctance to engage openly and honestly with evaluations (Baykal et al., 2021; also our own experience). There are also questions coming to the surface about how to reconcile competing interests between evaluation stakeholders and navigate the complex power dynamics around P/CVE programmes in ways that are conducive to good evaluation practice. Stakeholders frequently struggle to reach agreement about evaluation design or methods due to a combination of differing interests

330

Effects and effectiveness of programmes to prevent or counter radicalisation

as well as wider ‘paradigm wars’ (Madriaza et al., 2022, p. 21). There is often significant pressure within policy environments to favour quantitative methods with defined metrics (Morçöl, 2001), which are seen by many as more categorical and efficient. Those involved in programme delivery, however, often lean more towards qualitative methods, which are seen as ‘more human and less mechanistic’ (Madriaza et al., 2022, p. 24) and better suited to capturing the nuance of the experiences of their teams and beneficiaries (Clément et al., 2021). In our own evaluation practice, we have seen practitioners involved in highly successful community development or youth leadership initiatives deliver impressive outcomes around improved attitudes towards out-groups, improved relationships with friends and family, or more positive perspectives on the world, yet feel that their work was not valued because they lacked explicit evidence that violence had been reduced as a direct result of their work. We have also encountered difficulties as evaluators in arriving at a shared understanding with ground-level practitioners about sampling frames. While an evaluation team might favour sampling frames with the sort of clear and systematic inclusion and exclusion criteria that are usually required if findings are to be published in peer-reviewed journals, practitioners might be wary of such an approach, preferring instead to select ‘good cases’, either due to a belief that specific known cases best illustrate how their programme operates or due to concerns about how ‘negative’ results would reflect on their reputation within the sector. In their research with P/CVE practitioners, Clément et al. (2021) find that another driver of practitioner ambiguity towards evaluation is mundane, but no less important, considerations such as the time burden that evaluation can place on programme staff. Finally, there are important questions to take into consideration about how closer collaboration between policymakers, practitioners, and researchers affects the positionality of those researchers. In particular, it raises questions about the extent to which, as the practice and indeed the careers of those researchers become intertwined with those of members of the security and intelligence apparatus, they can still maintain their intellectual independence (see Martini & Fernández de Mosteyrín, this volume). In a recent report on the institutional structures of P/CVE evaluation, Baykal and colleagues argue that it is important to focus not only on how we might better design individual evaluations but also on how we go about ‘shaping an overall system of constructive P/CVE evaluation practice’ (2021, p. 7). Effective collaboration between researchers, policy planners, and practitioners will be central to this, and it is clear that there are important challenges that will require shared attention over the coming years. Nonetheless, the fact that such collaborations exist and that the focus of conversation has now moved more towards how to make them work in itself marks a significant step forward for the field.

Conclusions Researchers, policy planners, and practitioners alike are in broad agreement that effective evaluation is essential if the P/CVE field is to advance and that there is still a long way to go in terms of the quantity and quality of evaluations. Early attempts to advance an evaluation agenda were surprisingly slow to materialise, and when they did, they ran into a number of challenges. Foremost among these were a common lack of clarity around programme objectives, limited access for researchers to primary data, the problem of attribution, and challenges associated with the fact that initial evaluation efforts became ensnared in questions about how to demonstrate a negative. To some extent, none of these challenges were

331

Tom Fisher and Joel Busher

unique to the P/CVE field. There were, however, factors that made them particularly acute here, including the fact that radicalisation into violent extremism is such a low-frequency event; that the sector is so young and arguably particularly risk-averse; and the immaturity and contested nature of the underlying science of radicalisation itself. As investment in and commitment to evaluation have grown, however, a number of developments have gone at least some way towards addressing these challenges. Drawing on innovations and good research and evaluation practice in other fields, such as criminology, public health, and peace and conflict programming, growing use has been made of more sophisticated evaluation designs, including experimental, quasi-experimental, and longitudinal designs. Such approaches have resulted in some significant strengthening of the evidence base around P/CVE programmes, including helping to generate more compelling claims about programme impacts and causal pathways. A turn towards theory-driven evaluation has helped formulate clearer, more testable claims for evaluation and elucidate the logic of these programmes, thereby helping to address the lack of clarity around P/CVE programme objectives that often undermined early evaluation efforts. By engaging explicitly with questions about the role of context in shaping intervention outcomes, theory-driven approaches are also helping to address issues around attribution. Moreover, theorydriven approaches provide a potentially effective mechanism for moving the field past preoccupations with measuring the absence of a negative event, focusing attention instead on how positive outcomes from programmes, such as self-esteem or greater stability, contribute to reducing risk and harm. We are also seeing significant efforts to foster more effective collaborations between policy planners, practitioners, and evaluation researchers. While challenges remain here and are likely to persist given the sensitivities that surround P/CVE programming, these efforts are at least improving researcher and wider public access to relevant data, creating a greater common understanding about these programmes, and fostering the sharing of innovative approaches and good practice. There is, then, some cause for optimism in the P/CVE evaluation landscape. There is also a series of clear, simple messages that can be gleaned from the literature and the collected pool of experience that has developed over the last decade or so: • Evaluation should be incorporated into programming as soon as possible, ideally at the design/ commissioning stage, and should support programmes to clearly articulate and test their logic and claims; • Evaluation and the theories of change that underpin evaluation should be explicit about the role of (different forms of) context in enabling and inhibiting programme effectiveness; • Evaluations should embrace emergent good practice and take up the mantle of more sophisticated research designs that are increasingly being used in this field; • Evaluation should be inclusive, ensuring that the perspectives and understandings of all stakeholders are incorporated, and should also be explicit about any possible conflicts of interest; • Wherever possible, evaluations should be published, including a detailed description of the evaluation design and methods. Such messages might appear straightforward. The challenge, however, lies in turning such seemingly simple, even obvious, messages into norms and embedding them within the emergent sectoral culture. Going forward, we believe the key issues are likely to lie not so much in questions about how we address particular technical questions around research design but rather in how we generate

332

Effects and effectiveness of programmes to prevent or counter radicalisation

institutional structures and the kinds of behaviours and relationships that give rise to the sort of ‘overall system of constructive P/CVE evaluation practice’ to which Baykal and colleagues encourage us to aspire. The evaluation field has demonstrated that expertise, curiosity, and potential for innovation are present; a bigger cultural shift, of which we are seeing the sparks, must now follow.

References Ahmed, W. (2016, August 26). Prevent: Where perception is more important than reality. Huffington Post. www.huffingtonpost.co.uk/waqar-ahmed/prevent-where-perception-_b_11722602.html Aistrope, T. (2016). The Muslim paranoia narrative in counter-radicalisation policy. Critical Studies on Terrorism, 9(2), 182–204. Aldrich, D. P. (2014). First steps towards hearts and minds? USAID’s countering violent extremism policies in Africa. Terrorism and Political Violence, 26(3), 523–546. Ball, S. J., Maguire, M., & Braun, A. (2011). How schools do policy: Policy enactments in secondary schools. Routledge. Barkindo, A. B., & Bryans, S. (2016, Summer). De-radicalising prisoners in Nigeria: Developing a basic prison based de-radicalisation programme. Journal for Deradicalization, 7. Baruch, B., Ling, B., Warnes, R., & Hofman, J. (2018). Evaluation in an emerging field: Developing a measurement framework for the field of counter-violent-extremism. Evaluation, 24(4), 475–495. Baykal, A., Bressan, S., Friedrich, J., Pasquali, G., Rotmann, R., & Wagner, M. (2021). Evaluating P/CVE: Institutional structures in international comparison. Global Public Policy Institute. www.gppi.net/media/ Friedrich-et-al_PrEval_ENG_UPDATED.pdf Bellasio, J., Hofman, J., Ward, A., Nederveen, F., Knack, A., Meranto, A. S., & Hoorens, S. (2018). Counterterrorism evaluation: Taking stock and looking ahead. RAND Corporation. www.rand.org/pubs/research_ reports/RR2628.html Berger, J. M. (2016). Making CVE work: A focused approach based on process disruption (ICCT Research Paper). ICCT. Blamey, A., & Mackenzie, M. (2007). Theories of change and realistic evaluation: Peas in a pod or apples and oranges? Evaluation, 13(4), 439–455. Braddock, K. (2020). Experimentation & quasi-experimentation in countering violent extremism: Directions of future inquiry. RESOLVE Network. Brett, J., & Kahlmeyer, A. (2017). Strengthening resilience to violent extremism – STRIVE (Horn of Africa): Evaluation report. https://ct-morse.eu/wp-content/uploads/2017/04/170124-STRIVE-evaluation-ReportFinal.pdf Busher, J., Choudhury, T., Thomas, P., & Harris, G. (2017). What the prevent duty means for schools and colleges in England: An analysis of educationalists’ experiences. Coventry University. https://pureportal. coventry.ac.uk/files/11090509/Busher_Choudhury_Thomas_Harris_What_the_Prevent_duty_means_ for_schools_and_colleges.pdf Busher, J., & Jerome, L. (2020). The prevent duty in education: Impact, enactment and implications (p. 176). Palgrave Macmillan. Chen, H. T. (1990). Theory-driven evaluations. Sage. Cherney, A. (2018). Evaluating interventions to disengage extremist offenders: A study of the proactive integrated support model (PRISM). Behavioral Sciences of Terrorism and Political Aggression, 1–20. Cherney, A. (2020). Evaluating interventions to disengage extremist offenders: A study of the proactive integrated support model (PRISM). Behavioral Sciences of Terrorism and Political Aggression, 12(1), 17–36. Cherney, A., & Belton, E. (2020). Assessing intervention outcomes targeting radicalised offenders: Testing the pro integration model of extremist disengagement as an evaluation tool. Dynamics of Asymmetric Conflict, 13(3), 193–211. Christmann, K. (2012). Preventing religious radicalisation and violent extremism: A systematic review of the research evidence (Research Report). Youth Justice Board. Clément, P.-A., Madriaza, P., & Morin, D. (2021). Constraints and opportunities in evaluating programs for prevention of violent extremism: How the practitioners see it. UNESCO Chair in Prevention of Radicalisation and Violent Extremism (UNESCO-PREV Chair), Sherbrooke University. https://chaireunesco-prev.ca/ wp-content/uploads/2021/07/FINAL_UnescoPrev_Rapport_PREV-IMPACT_EN.pdf?

333

Tom Fisher and Joel Busher Clubb, G. (2015). De-radicalization, disengagement and the attitudes-behavior debate. In C. Kennedy-Pipe, G. Clubb, & S. Mabon (Eds.), Terrorism and political violence (pp. 258–266). Sage. Clubb, G., & Tapley, M. (2018). Conceptualising de-radicalisation and former combatant re-integration in Nigeria. Third World Quarterly, 39(11), 2053–2068. Coryn, C. L. S., Noakes, L. A., Westine, C. D., & Schröter, D. C. (2011). A systematic review of theory-driven evaluation practice from 1990 to 2009. American Journal of Evaluation, 32(2), 199–226. Davies, M., Warnes, R., & Hofman, J. (2017). Exploring the transferability and applicability of gang evaluation methodologies to counter-violent radicalisation. RAND Corporation. Dawson, L. (2021). Bringing religiosity back in. Perspectives on Terrorism, 15(2), 2–22. De Silva, M. J., Breuer, E., Lee, L., Asher, L., Chowdhary, N., Lund, C., & Patel, V. (2014). Theory of change: A theory-driven approach to enhance the medical research council’s framework for complex interventions. Trials, 15, 267. https://doi.org/10.1186/1745-6215-15-267 Dunn, K. M., Atie, R., Kennedy, M., Ali, J. A., O’Reilly, J., & Rogerson, L. (2015). Can you use community policing for counter terrorism? Evidence from NSW, Australia. Police Practice and Research, 17(3), 196–211. Edwards, W., & Newman, J. R. (1982). Multiattribute evaluation. Sage. El-Said, H. (2015). New approaches to countering terrorism: Designing and evaluating counter radicalization and de-radicalization programs. Springer. Eye on Radicalization. (2018). An interview with Dr. John Horgan – terrorism, psychology, and major issues in the field. https://eeradicalization.com/an-interview-with-dr-john-horgan-terrorism-psychology-and-majorissues-in-the-field/ Feddes, A. R., & Gallucci, M. (2015). A literature review on methodology used in evaluating effects of preventive and de-radicalisation interventions. Journal for Deradicalization, 15–16(5), 1–27. Feddes, A. R., Mann, L., & Doosje, B. (2015). Increasing self‐esteem and empathy to prevent violent radicalization: A longitudinal quantitative evaluation of a resilience training focused on adolescents with a dual identity. Journal of Applied Social Psychology, 45(7), 400–411. Ferraro, P. J. (2009). Counterfactual thinking and impact evaluation in environmental policy. New Directions for Evaluation, 2009(122), 75–84. Fink, N. C., Romaniuk, P., & Barakat, R. (2013). Evaluating countering violent extremism programming: Practice and progress. Center on Global Counterterrorism Cooperation. www.cve-kenya.org/media/ library/Fink_et_al_2013._Evaluating_CVE_Programming.pdf Finn, M., Momani, B., Opatowski, M., & Opondo, M. (2016). Youth evaluations of CVE/PVE programming in Kenya in context. Journal for Deradicalization, 7, 164–224. Fisher, T., Range, D., & Cuddihy, J. (2020). Evaluation of “violent extremism strengthening resilience” (STRIVE II) in Kenya: Final report. https://ct-morse.eu/evaluation-of-strive-ii-in-kenya/ Gajda, R., & Jewiss, J. (2004). Thinking about how to evaluate your program? These strategies will get you started. Practical Assessment, Research, and Evaluation, 9(1), 8. Gielen, A. J. (2018). Exit programmes for female Jihadists: A proposal for conducting realistic evaluation of the Dutch approach. International Sociology, 33(4), 454–472. Gielen, A. J. (2019). Countering violent extremism: A realist review for assessing what works, for whom, in what circumstances, and how? Terrorism and Political Violence, 31(6), 1149–1167. Glazzard, A. (2022). Violent extremist disengagement and reintegration: A framework for planning, design and evaluation of programmatic interventions. Studies in Conflict & Terrorism, 1–20. Glazzard,A., & Reed,A. (2018). Global evaluation of the European Union engagement on counter-terrorism. ICCT. https://icct.nl/publication/global-evaluation-of-the-european-union-engagement-on-counter-terrorism/ Hall, I. (2017). Evaluation and social research. Bloomsbury. Harris-Hogan, S. (2020). How to evaluate a program working with terrorists? Understanding Australia’s countering violent extremism early intervention program. Journal of Policing, Intelligence and Counter Terrorism, 15(2), 97–116. Hassan, G., Brouillette-Alarie, S., Ousman, S., Kilinc, D., Savard, É. L., Varela, W., Lavoie, L., Fetiu, A., Harris-Hogan, S., Borokhovski, E., Pickup, D., Madriaza, P., Rousseau, C., Thompson, S. K., McCoy, J., Venkatesh, V., Boivin, M., Srimathi Narayana, M., Morin, D., . . . The CPN-PREV Team. (2021a). A systematic review on the outcomes of primary and secondary prevention programs in the field of violent radicalization. Canadian Practitioners Network for the Prevention of Radicalization and Extremist Violence. https://cpnprev.ca/wp-content/uploads/2021/03/CPN-PREV-2nd-Systematic-Review-2.pdf

334

Effects and effectiveness of programmes to prevent or counter radicalisation Hassan, G., Brouillette-Alarie, S., Ousman, S., Savard, É. L., Kilinc, D., Madriaza, P., Varela, W., Pickup, D., Danis, E., & The CPN-PREV team. (2021b). A systematic review on the outcomes of tertiary prevention programs in the field of violent radicalization. Canadian Practitioners Network for the Prevention of Radicalization and Extremist Violence. https://cpnprev.ca/wp-content/uploads/2021/11/Intervention-reportFINAL-2021.pdf Helmus, T. C., Matthews, M., Ramchand, R., Beaghley, S., Stebbins, D., Kadlec, A., Brown, M. A., Kofner, A., & Acosta, J. D. (2017). RAND program evaluation toolkit for countering violent extremism. RAND Corporation. Ho, W., van Wessel, M. G. J., & Tamas, P. A. (2020). The hidden life of theories of change. WUR/Hivos. Horgan, J. (2008). Deradicalization or disengagement?: A process in need of clarity and a counterterrorism initiative in need of evaluation. Perspectives on Terrorism, 2(4), 3–8. Horgan, J., & Braddock, K. (2010). Rehabilitating the terrorists? Challenges in assessing the effectiveness of de-radicalization programs. Terrorism and Political Violence, 22(2), 267–291. Johnson, K., Greenseid, L. O., Toal, S. A., King, J. A., Lawrenz, F., & Volkov, B. (2009). Research on evaluation use: A review of the empirical literature from 1986 to 2005. American Journal of Evaluation, 30(3), 377–410. Jugl, I., Lösel, F., Bender, D., & King, S. (2021). Psychosocial prevention programs against radicalization and extremism: A meta-analysis of outcome evaluations. The European Journal of Psychology Applied to Legal Context, 13(1), 37–46. Khalil, J., & Zeuthen, M. (2016). Countering violent extremism and risk reduction: A guide to programme design and evaluation (RUSI Whitehall Report 2–16). RUSI Whitehall. Koehler, D. (2017). Preventing violent radicalisation: Programme design and evaluation. Competence Center for the Coordination of the Prevention Network against (Islamic) Extremism in Baden-Württemberg (KPEBW), 91–98. www.cidob.org/en/articulos/monografias/resilient_cities/preventing_violent_radica lisation_programme_design_and_evaluation Koehler, D. (2019). Stand-alone de-radicalisation programme evaluation tool for stakeholders. Dialogue and Radicalisation and Equality. www.dare-h2020.org/uploads/1/2/1/7/12176018/dare-de-radicalisation_ evaluation_tool.pdf Kurtz, J., Tesfaye, B., & Wolfe, R. J. (2018). Can economic interventions reduce violence? Impacts of vocational training and cash transfers on youth support for political violence in Afghanistan. Mercy Corps. www.voced.edu.au/content/ngv:79236 Kutner, S. (2016). The call for component analyses of the Saudi Arabian risk reduction initiative: An examination of religious re-education’s role in the deradicalization and disengagement process. Journal for Deradicalization, 7, 107–123. Lewis, J., Marsden, S., & Copeland, S. (2020). Evaluating programmes to prevent and counter violent extremism. CREST. https://crestresearch.ac.uk/resources/evaluating-programmes-to-prevent-and-counter-extremism/ Li, Y. H., Mullette, E., & Brant, J. M. (2018). The stepped-wedge trial design: Paving the way for cancer care delivery research. Journal of the Advanced Practitioner in Oncology, 9(7), 722–727. Lub, V. (2014). The plausibility of policy: Case studies from the social domain. Eleven International. Lum, C., Kennedy, L. W., & Sherley, A. (2006). Are counter-terrorism strategies effective? The results of the Campbell systematic review on counter-terrorism evaluation research. Journal of Experimental Criminology, 2, 489–516. Madriaza, P., Morin, D., Ousman, S., Autixier, C., Hassan, G., & Venkatesh, V. (2022). Improving evaluations of programs for prevention of radicalization and violent extremism: An exploratory international study. UNESCO Chair in Prevention of Radicalisation and Violent Extremism (UNESCO-PREV Chair) in Collaboration with the International Centre for the Prevention of Crime, University of Sherbrooke. Marret, J. L., Bellasio, J., van den Berg, H., van Gorp, A., van Hemert, D., Leone, L., Meijer, R., Warnes, R., Van Wonderen, R., & Longo, C. (2017). Final report providing background for using and further developing the validated toolkit. Impact Europe. http://impacteurope.eu/wp-content/uploads/2017/11/D6.2-Final-Synthesisreport.pdf Marsden, S. V. (2015). Conceptualising “success” with those convicted of terrorism offences: Aims, methods, and barriers to reintegration. Behavioral Sciences of Terrorism and Political Aggression, 7(2), 143–165. Mason, P., & Barnes, M. (2007). Constructing theories of change: Methods and sources. Evaluation, 13(2), 151–170. Mastroe, C. (2016). Evaluating CVE: Understanding the recent changes to the United Kingdom’s implementation of prevent. Perspectives on Terrorism, 10(2), 50–60.

335

Tom Fisher and Joel Busher Mastroe, C., & Szmania, S. (2016). Surveying CVE metrics in prevention, disengagement and de-radicalization programs. www.start.umd.edu/pubs/START_SurveyingCVEMetrics_March2016.pdf McBride, M. K., Carroll, M., Mellea, J., Hughes, D., & Savoia, E. (2022). Evaluating terrorist and extremist reintegration programming: A systematic literature review. Journal for Deradicalization, 32, 35–75. Mertens, D. M. (1999). Inclusive evaluation: Implications of transformative theory for evaluation. American Journal of Evaluation, 20(1), 1–14. Morçöl, G. (2001). Positivist beliefs among policy professionals: An empirical investigation. Policy Sciences, 34(3), 381–401. Mulgan, G. (2016, September 1). What’s wrong with theories of change? Nesta Blog. www.nesta.org.uk/blog/ whats-wrong-with-theories-of-change/ Nehlsen, I., Biene, J., Coester, M., Greuel, F., Milbradt, B., & Armborst, A. (2020). Evident and effective? The challenges, potentials and limitations of evaluation research on preventing violent extremism. International Journal of Conflict and Violence (IJCV), 14, 1–20. Organisation for Economic Co-operation and Development. (1991). Development assistance committee principles for evaluation of development assistance. www.oecd.org/dac/evaluation/2755284.pdf Parker, D., & Lindekilde, L. (2020). Preventing extremism with extremists: A double-edged sword? An analysis of the impact of using former extremists in Danish schools. Education Sciences, 10(4), 111. Patton, M. Q. (2008). Utilization-focused evaluation. Sage. Pawson, R., & Tilley, N. (1994). What works in evaluation research? The British Journal of Criminology, 34(3), 291–306. Pawson, R., & Tilley, N. (1997). Realistic evaluation. Sage. Pistone, I., Eriksson, E., Beckman, U., Mattson, C., & Sager, M., (2019). A scoping review of interventions for preventing and countering violent extremism: Current status and implications for future research. Journal for Deradicalization, 19, 1–84. Pressman, J. L., & Wildavsky, A. (1984). Implementation: How great expectations in Washington are dashed in Oakland; or, why it’s amazing that federal programs work at all, this being a saga of the economic development administration as told by two sympathetic observers who seek to build morals on a foundation. University of California Press. Radicalisation Awareness Network. (2017). RAN handbook on CVE/PVE training programmes: Guidance for trainers and policy makers. https://home-affairs.ec.europa.eu/document/download/6b0ade34-feaf-45e2851a-14d942dcdedd_en?filename=ran_handbook_on_cve_pve_training_programmes_12-2017_en.pdf Radicalisation Awareness Network. (2021). Effective and realistic quality management and evaluation of P/CVE. RAN Conclusion Paper. https://home-affairs.ec.europa.eu/document/download/61b2db51-68c34c44-9340-8a9997aec4eb_en?filename=ran_small-scale_expert_meeting_quality_management_evaluation_05032021_en.pdf Ris, L., & Ernstorfer, A. (2017). Borrowing a wheel: Applying existing design, monitoring, and evaluation strategies to emerging programming approaches to prevent and counter violent extremism. Peacebuilding Evaluation Consortium Briefing Paper. www.cdacollaborative. org/?smd_process_download=1&download_id=5325 Roman, C. G., Klein, H. J., & Wolff, K. T. (2018). Quasi-experimental designs for community-level public health violence reduction interventions: A case study in the challenges of selecting the counterfactual. Journal of Experimental Criminology, 14(2), 155–185. Romaniuk, P. (2015). Does CVE work? Lessons learned from the global effort to counter violent extremism. Global Center on Cooperative Security. www.globalcenter.org/wp-content/uploads/2015/09/Does-CVEWork_2015.pdf Romaniuk, P., & Fink, C. N. (2012). From input to impact: Evaluating terrorism prevention programs. Global Center on Cooperative Security. http://globalcenter.org/wpcontent/uploads/2012/10/CGCC_EvaluatingTerrorismPrevention.pdf Sahgal, G., & Kimaiyo, T. (2020). Youth resilience to violent extremism: An evaluation of a mentorship intervention in Kenya. Journal for Deradicalization, 24, 113–160. Savoia, E., Su, M., Harriman, N., & Testa, M. A. (2019). Evaluation of a school campaign to reduce hatred. Journal for Deradicalization, 21, 43–83. Schmid, A. P. (2016). Research on radicalisation: Topics and themes. Perspectives on Terrorism, 10, 26–32. Schuurman, B. (2020). Research on terrorism, 2007–2016: A review of data, methods, and authorship. Terrorism and Political Violence, 32(5), 1011–1026.

336

Effects and effectiveness of programmes to prevent or counter radicalisation Schuurman, B. (2021). The role of beliefs in motivating involvement in terrorism. Perspectives on Terrorism, 15(5), 85–92. Schuurman, B., & Bakker, E. (2016). Reintegrating Jihadist extremists: Evaluating a Dutch initiative, 2013– 2014. Behavioral Sciences of Terrorism and Political Aggression, 8(1), 66–85. Silke, A., & Veldhuis, T. (2017). Countering violent extremism in prisons: A review of key recent research and critical research gaps. Perspectives on Terrorism, 11(5), 1–10. Thiessen, C. (2022). The strategic ambiguity of the United Nations approach to preventing violent extremism. Studies in Conflict & Terrorism, 45(2), 111–132. Thomas, P. (2020). Britain’s prevent strategy: Always changing, always the same? In J. Busher & L. Jerome (Eds.), The prevent duty in education: Impact, enactment and implications (pp. 11–31). Palgrave Macmillan. Thompson, S. K., & Leroux, E. (2022). Lessons learned from dual site formative evaluations of countering violent extremism (CVE) programming co-led by Canadian police. Journal of Policing, Intelligence and Counter Terrorism, 1–19. UNDP. (2018). Improving the impact of preventing violent extremism programming: A toolkit for design, monitoring and evaluation. United Nations Development Programme. Vaessen, J., & Leeuw, F. L. (Eds.). (2011). Mind the gap: Perspectives on policy evaluation and the social sciences (Vol. 1). Transaction. Van der Heide, L., & Schuurman, B. (2018). Reintegrating terrorists in the Netherlands: Evaluating the Dutch approach. Journal for Deradicalization, 17. Veldhuis, T. (2012). Designing rehabilitation and reintegration programs for violent extremist offenders: A realist approach. International Centre for Counter-Terrorism. Vidino, L., & Brandon, J. (2012). Europe’s experience in countering radicalisation: Approaches and challenges. Journal of Policing, Intelligence and Counter Terrorism, 7(2), 163–179. Webber, D., Chernikova, M., Kruglanski, A. W., Gelfand, M. J., Hettiarachchi, M., Gunaratna, R., Lafreniere, M.-A., & Belanger, J. J. (2018). Deradicalizing detained terrorists. Political Psychology, 39, 539–556. https://doi.org/10.1111/pops.12428 Weine, S., Eisenman, D. P., Glik, D., Kinsler, J., & Polutnik, C. (2018). Leveraging a targeted violence prevention program to prevent violent extremism: A formative evaluation in Los Angeles. www.dhs.gov/ publication/st-leveraging-targeted-violence-prevention-program-prevent-violent-extremism-formative Weine, S., Eisenman, D. P., Kinsler, J., Glik, D. C., & Polutnik, C. (2017). Addressing violent extremism as public health policy and practice. Behavioral Sciences of Terrorism and Political Aggression, 9(3), 208–221. Weiss, C. H. (1995). Nothing as practical as good theory: Exploring theory-based evaluation for comprehensive community initiatives for children and families. In J. Connell, A. C. Kubisch, L. B. Schorr, & C. H. Weiss (Eds.), New approaches to evaluating community initiatives: Concepts, methods, and contexts (pp. 65–92). Aspen Institute. Williams, M. J., Horgan, J., & Evans, W. (2016). Evaluation of a multi-faceted, U.S. community based, Muslim-led CVE program. www.ncjrs.gov/pdffiles1/nij/grants/249936.pdf Wollmann, H. (2007.) Policy evaluation and evaluation research. In F. Fischer & G. J. Miller (Eds.), Handbook of public policy analysis: Theory, politics, and methods (1st ed., pp. 393–402). Routledge.

337

22 RE-COLONISING THE FIELD OF EVALUATION OF PREVENTION OF VIOLENT RADICALISATION A critical, cross-regional perspective Pablo Madriaza, David Morin, and Vivek Venkatesh Introduction The notion of violent radicalisation and, consequently, that of prevention of violent radicalisation and violent extremism1 have been debated in the literature to try to understand a situation that was initially problematised in the West in the early years of the new millennium (cf. Kundnani & Hayes, 2018). For example, PVR frameworks have been used to explain why young Muslims in Europe were alienated from their societies, making them prone to support extremist groups or commit terrorist acts (Coolsaet, 2016). This emerging conceptualisation, even if not very precise, triggered the development of a revolution in a sector historically dominated by a rationale that emphasises national security. The idea was, as Kundnani and Hayes indicate, to change the focus from the military, law enforcement, and intelligence response to the support that the Muslim world could eventually give to radical Islam: ‘In other words, the ‘battle of ideas’ would be engaged alongside the battle for territory; the fight against ‘radical Islam’ would be cultural as much as corporeal, preventative as much as reactive’ (2018, p. 4). This shift involved the progressive integration of psychosocial prevention components and the addition of new social actors working beyond the national security sector in prevention activities, including the mental health, education, and social and community services sectors (Hassan et al., 2020; Madriaza et al., 2017). Since 2005, a more novel field of study and social intervention, virtually non-existent until then, has been developed, responding particularly to perceived needs and threats in the West. This has been accompanied by new theories and concepts that are part of the cultural traditions of the countries of the global North and have been supported by a considerable amount of empirical evidence, mostly from these countries. In a recent study, Ilyas (2021a) shows, for example, that between the years 2015 and 2019, 83% of the articles published in seven specialised journals came from Western scholars. In another review conducted on nine specialised journals, it was found that during 2020, 83% of the publications came from scholars from only four countries: the United States, the United Kingdom, Australia, and Canada (Malo et al., 2021), four of the five countries that are grouped under the international intelligence alliance known as the ‘five eyes’. Kundnani and Hayes (2018) clearly demonstrate how the conceptual, technical, and practical frameworks governing PVR are firmly ensconced in the perspectives of the Global North and are transferred DOI: 10.4324/9781003035848-25

338

Re-colonising the field of evaluation of PVR

to other regions both within and outside the scope of the United Nations organisation. Hence, a growing and unidirectional flow developed from the Western world both to finance new prevention programmes in these regions, particularly Asia and Africa and to transfer knowledge, techniques, practices, and human resources – scientists, consultants, and practitioners – that would support and facilitate the effective implementation of these programmes. A mostly unidirectional knowledge transfer has led to a Western-centric view of the way radicalisation is understood and prevented in countries that, at one time in their history, were European colonies. The implications of this last statement are clear: the flow of knowledge, practices, and knowhow surrounding the phenomena of radicalisation and extremism has followed a pattern of coloniality, reproducing the primacy of the West in the construction of universal narratives that respond, in many cases, to the local needs of the countries of the Global North rather than to those that arise in those regions where these universal narratives are applied. The globalisation of radicalisation prevention strategies from the West has been undertaken, as Kundnani and Hayes (2018) point out, even though many of these strategies have been contested regarding their actual effectiveness and legitimacy in their countries of origin, as is the oft-quoted case of the United Kingdom’s PREVENT strategy. The colonial-postcolonial-decolonial conceptual lens, as we will examine below, has, however, been scarcely applied in the domain of security studies and, surprisingly, is almost non-existent in critical studies on these issues. Ilyas (2021b) notes, for example, that a quick keyword search for the concept ‘decolonial’ yielded no results in the Critical Terrorism Studies Journal.2 The critical nodes of the conceptualisation around radicalisation and its prevention are, however, not hardly interpretable from the point of view of decolonial and postcolonial studies. The creation of the concept of radicalisation around Islam (Coolsaet, 2016; Kundnani, 2012; Kundnani & Hayes, 2018) and the negative consequences that some prevention models have had on Muslim communities, particularly within the Western world (Kundnani, 2009; Ragazzi, 2014; Taylor, 2020), reveal to us a new form of the orientalism denounced by Said (1995) in the late 1970s, this time taking the form of a threat, as well as the construction of categories of social subordination in relation to the Western world, which can be associated with the notion of coloniality of power in Quijano (2007). On the other hand, the Western world has been unwilling to apply the same definitional notions to extreme right-wing groups (Nesbitt, 2021), and only recently have a number of these groups begun to be included alongside terrorist groups that were already replete with those of Arab-Muslim heritage. The differential treatment of Islamist and right-wing extremism phenomena in a decolonial frame far exceeds the scope of this chapter. However, the importance of adopting a decolonial interpretation in this domain should not be ignored and may serve as an opportunity to reinterpret the emergence of these concepts from the margins of the West (Mignolo, 2011) and not towards the margins, as has been the tendency so far. In this chapter, we will try to open new avenues of discussion on the transregional flows of knowledge and power in the evaluation of PVR programmes, as well as on the repercussions that these flows could have on research and practice in different regions of the world. For this purpose, we will draw on a recent systematic methodological review (Madriaza et al., 2022), in which we identified 219 evaluative studies of PVR programmes up to 2019, including 98 studies in Europe, 50 in Africa, 42 in Asia, 18 in North America, and seven in Oceania. The lens on which we will base our analyses will obviously be the decolonial perspective. Thus, the following section will be devoted to describing the main features of this approach. The third section will address how a handful of studies have integrated these notions into the field of violent extremism and its prevention. In the fourth section, we will present some results of our research that illustrate this discussion, and in the last part of this chapter, we will conclude with some elements for discussion and future research. 339

Pablo Madriaza, David Morin, and Vivek Venkatesh

The decolonial turn The decolonial turn is the name given to the work of a group of Latin American intellectuals who have criticised and reinterpreted, through the notion of modernity/coloniality, the predominant use in the social sciences of categories of analysis derived from Western social reality, which emerge particularly in the industrialised European countries of the North and which were universalised through the process of colonisation (Escobar, 2007). According to these authors, this Westerncentric perspective places modern Western countries at the forefront of a supposedly universal, linear, and homogeneous historical process (Dussel, 2000). Modern Western society would then be the only type of legitimate and civilised society based on a cultural project of ordering the world in rational, objective, neutral, and predictable terms (Escobar, 2007). A key concept for these authors is coloniality (Quijano, 2007). If ‘colonialism’ has been the military occupation and annexation of a territory and its inhabitants by a foreign imperial force, ‘coloniality’, on the other hand, refers to the ‘cultural logic’ of colonialism, that is to say the types of colonial legacies that persist and multiply even after colonialism has ended (Mignolo, 2007). Although there are various ways of understanding coloniality, there are two manifestations of coloniality that are relevant to this chapter and, as we will see in the next section, are at the centre of the discussion on the decolonial approach in critical security studies: the coloniality of power and the coloniality of knowledge. Quijano (2007) explains that the concept of coloniality of power is a form of social classification in contemporary societies that was born during the colonial period and is defined according to the race and phenotype of individuals (white, aboriginal, black, brown, etc.). The coloniality of knowledge (Dussel, 2000; Quijano, 2007) refers to the techno-scientific rationality developed in Europe during the industrialisation process, which becomes on the one hand a determining factor in the expansion of European colonialism and, on the other, transforms the local rules of European and then Western knowledge into the only valid, objective, and neutral model of knowledge production, leaving aside any other type of ‘episteme’ (traditional or ancestral) generated outside the Western world. The unidirectional flow of Western-centric, North-to-South knowledge and practices, which began in the colonial period, has played in this case a central role in the universalisation of Western thought, covering up, obfuscating, and delegitimising alternative forms of thought, particularly in the formerly colonised countries. Thus, although decolonisation has taken place since the 19th century in much of the world, coloniality remains a contemporary issue in our societies, both within Western countries and in their former colonies. The decolonial turn has implied a new view of the impact of subordination relationships both within Western societies and between the latter and their former colonies and has opened up new avenues for interpreting complex and diverse phenomena in a different and critical manner. One of its greatest contributions is perhaps the idea of transmodernity and border thinking (Grosfoguel, 2009; Mignolo, 2011). With these concepts, Dussel and Mignolo, respectively, open, on the one hand, the gates to deconstruct and go beyond the myth of a single universal modernity and give way to the idea of multiple modernities, based on new knowledge and epistemes from the margins of the West, as well as hybrid modernities born from the encounter between the West and its alterity.

Decolonial perspective in preventing violent extremism The decolonial approach has been, as we have said, less applied in ‘security studies’ or ‘critical security studies’ on terrorism, radicalisation, or violent extremism. Reproducing the rapid keyword search conducted by Ilyas (2021a) for the term ‘decolonial*’ in the journal ‘Critical studies

340

Re-colonising the field of evaluation of PVR

on terrorism’ in February 2022, we found only eight texts that have used this term in some section of the text, of which only three effectively included a decolonial perspective3 (Crosby, 2021; Khan, 2021; Oando & Achieng’, 2021). Khan (2021) calls for opening the field of critical studies to a perspective that more deeply addresses the relationships between race, coloniality, and terrorism; she explains that critical studies has merely denounced and responded to neo-Orientalism and Islamophobia, unintentionally reproducing in reverse the colonial response of mainstream terrorism studies. Khan (2021) found, through the analysis of 296 articles, that the same colonial language was used in both critical and mainstream studies in this field. Among the handful of studies that address these issues in more depth,4 three axes of analysis are frequently invoked: the construction of the ‘terrorist’ category associated with Islam, the criminalisation of Muslims in the West that reproduces colonial patterns, and the analysis of the unidirectional flow of North-to-South knowledge and practices, particularly in reference to prevention practices in non-Western countries. The first two axes address elements like those that have been addressed by critical studies in general, namely, the association between Islam and terrorism, and can be associated from a certain point of view with an application of the concept of coloniality of power and the categories of hierarchies of knowledge and power, otherwise called subalternisation. Majozi (2018), for example, critically analyses the Western discourse around the conceptualisation of the Islamic State (IS) as a terrorist entity. He argues that this dominant view is a racist and Islamophobic Western epistemological narrative that seeks to create a natural link between terrorism and Islam through the supposed backwardness of Islamic culture while denying the European roots of terrorism that lie at the very foundation of Western modernity. This narrative would aim to ensure the political, economic, cultural, and moral supremacy of the West over other regions of the world. The IS, he explains, has, on the contrary, violated all the norms defining the self-declaration of a caliphate in accordance with the first centuries of Islam. IS has thus reproduced the way in which Western nation-states were violently constructed and illustrates this claim through its use of Western propaganda. Therefore, IS would not be an alternative to the Western world but an inverted mimicry of the Western colonial system of domination already present in Syria and Iraq. Crosby’s (2021) study analyses Canadian government records to demonstrate how surveillance and criminalisation techniques and discourses surrounding Tamil refugees were used to construct an image of them as terrorist sympathisers and as a threat and danger to Canadian society. This is one of the few studies that analyses, from a decolonial perspective, the construction of discourses in a Western country. It concludes that these policies, seen by the West as a depoliticised, neutral, and de-racialised process, have been implemented as a means to exclude racialised people considered undesirable from white settlements. This racialisation, according to Crosby, would be a social process carried out through racialised security technologies. By these, he means: those methods and techniques of exclusion, restriction, and containment enacted through the contemporary border regimes of white settler or postcolonial states and practiced by its agents and agencies of security that prevent non-preferred racialised migrants from accessing rights to citizenship, health care, and a viable life. (Crosby, 2021, p. 191) Elsewhere, while not explicitly addressing colonialism as a term in their publication, Venkatesh et al. (2020) have examined aspects of eight violent, traumatic terrorist video propaganda produced by IS within the theoretical confines of Jean Baudrillard’s hyperrealism, Western cinema

341

Pablo Madriaza, David Morin, and Vivek Venkatesh

studies, abjection, and the use of utopian/dystopian themes. Their research elucidates how IS uses specific techniques and narratives – grounded in a co-opting of Western media tropes of ultraviolence, religious intolerance, and conservatism that are used to frame videos that depict horrific instances of the murder of alleged apostates and combatant enemies. The third axis, that is, the analysis of the unidirectional flow of North-South knowledge and practices, is probably the most concrete example of what we have defined above as the ‘coloniality of knowledge’, or, in other words, the process through which Western knowledge is transformed into the only valid, objective, universal, and neutral model of knowledge production. A central element in these studies is the critique of the dependence of non-Western countries on the knowledge produced in this region (Ilyas, 2021b; Oando & Achieng’, 2021). This dependence is based on the supposed supremacy of the Western episteme constructed through colonialism and coloniality, which not only allows an overrepresentation of Western researchers in this field but also marginalises local authors and knowledge and discourages the production of indigenous knowledge (Ilyas, 2021a). Local researchers are relegated to a data ‘mining’ role in their countries of origin, data that are then processed and analysed by Western researchers and then returned unidirectionally to these regions as an objective form of scientific output (Ilyas, 2021b). This kind of knowledge transfer masks the fact that the Western point of view is a local point of view that, through the coloniality of knowledge, becomes a universal, rootless, and ahistorical knowledge that can therefore be applied to any context. Ilyas (2021a) analyses, for example, the country of origin of authors who published between 2015 and 2019 in seven prestigious journals in this domain. For this purpose, he classifies the authors into three categories: those with a Western heritage; those who are located in non-Western countries but have a standard of living similar to Western countries; and those authors who do not have a Western heritage. His results reveal that 83% of the articles were written by scholars with a Western heritage and that 90% of the articles were published by authors who were based in universities or think tanks located in Western countries. It concludes that despite the fact that most terrorist attacks occur in non-Western countries, the production and flow of knowledge are unbalanced in the Western world. This Western overrepresentation results in the greater power of scholars from these countries to determine and classify what is terrorism, what actions constitute this phenomenon, and to influence the development of prevention policies and practices. Oando and Achieng’s (2021) study demonstrates how security strategies in Kenya have depended on Western conceptual and strategic frameworks, which in turn reproduce colonial frameworks that have neglected the particularities of the local context. This dependence is seen in both hard and soft security approaches. According to these authors, the military approach against Al-Shabaab draws on the colonial heritage of the British Empire, which used a bloody military strategy to combat the Mau Mau insurgency in the mid-20th century. Today, this dependence can be seen in the influence of the Global War on Terrorism since the attack on the Twin Towers in 2001. At least two bills, which exactly copied the US Patriot Act, were rejected at the last minute by the Kenyan parliament in 2003 and 2006. Kenya’s counter-terrorism approach is therefore mainly Eurocentric, they conclude, as it does not reflect the particularity of the country’s history and struggles, but rather Western interests. In terms of soft approaches, Oando and Achieng’ note a similar problem. The Kenyan state, less interested in investing in prevention, has left the funding of these approaches in the hands of Western donors, who have promoted models developed in Europe and North America and fostered dependence on Western knowledge. Most of these interventions in Kenya, they argue, are based on an ‘aggressor-victim’ criminological model, which facilitates the labelling of alleged ‘criminals’ in geographical, ethnic, and religious terms. This dependence is also seen in the discourses associated with the causes of terrorism in the country, 342

Re-colonising the field of evaluation of PVR

which are based on the Western diagnosis that endemic poverty, poor governance, and fragile governments are the predominant factors in the emergence of terrorism in the region. This diagnosis hides, however, the fact that these factors are not specific to the continent but are the result of its colonial heritage. All these elements allow Oando and Achieng’ to conclude on the need to develop an indigenous PVR strategy in the Kenyan context. In a study applied to the Indonesian context, Ilyas (2021b) analyses the dependency relationship between the South and Western countries in terms of knowledge and intervention models in PVR. To this end, he conducted interviews with NGO representatives in Indonesia who worked with Western researchers and agencies. Ilyas addresses three main points in his analysis: the dependence of Indonesian NGOs on Western funding, particularly from the United States and Australia; the unequal relationship in the field of research between local and Western actors; and the unidirectional return of this knowledge. Among the points addressed by Ilyas, the strong influence of Western funders is the starting point. Donors, Western researchers, and universities have complete control over the nature of the prevention programme implemented and often impose their own objectives, which do not always meet the needs of the Indonesian people. These objectives respond exclusively, according to Ilyas, to the security needs of these countries and to securing economic interests that terrorist groups could jeopardise. This top-down approach demonstrates for this author that colonialism is still very much in force in the country. The relationship between Western researchers and local actors is the second element in which this coloniality appears. The latter are often used as simple gatherers of information, which reproduces the extractivist relationship of colonial times. Local stakeholders often complain that they do not know how this information will be used and that they rarely participate in the final analysis of this data. After this extractivist phase, the information is processed and analysed by Occidental researchers, transforming it into Western knowledge that is then imported in a unidirectional manner by the very countries where this knowledge was collected. This is the third element addressed by Ilyas. Most of the knowledge that is taught in this field in Indonesia, both in universities and through trainings, is indeed knowledge coming from the West. This process (financial dependence, extractivism, information processing in the West, and unidirectional return) results, according to this author, in dependence on Western ideas and knowledge and the passive position of local actors in Indonesia, who are engaged in the extraction and reception of knowledge rather than its production. This situation also creates a kind of monoculture of thought around terrorism, which reproduces the intellectual imperialism of Western countries in Indonesian education, marginalising local actors and discouraging indigenous knowledge production. The few studies presented in this section are not only scarce and extremely recent, but they are also the product of early-career researchers, such as PhD candidates, who are not affiliated with large research centres and who have not received funding to carry out these investigations. These analyses thus come from the margins of Western and mainstream academia, and it is possibly because of this that they have been able to incorporate new interpretative elements into the production of knowledge between the West and its former colonies.

The evaluation of PVR programmes: a colonial relationship? To illustrate the flow of knowledge and practices between the global North and South, we will use the results of a recent systematic methodological review of PVR programme evaluations, in which we identified 219 studies published up to 2019 in English, French, and Spanish (Madriaza et al., 2022).5 Before beginning the presentation of the results, a few remarks are in order. First, this systematic review was not intended to demonstrate North-to-South flows of knowledge and practices, 343

Pablo Madriaza, David Morin, and Vivek Venkatesh

but rather this analysis arises from the results obtained and the evident imbalance in knowledge production between the different regions. On the other hand, these results refer particularly to the flows of knowledge and practices regarding programme evaluation and cannot be directly applied to the practice of prevention as a whole, although it can be concluded that there are obvious links. Nevertheless, we believe that these results are useful in describing the development and flow of knowledge and practices currently available in this field. Hence, these results are not intended as a demonstration of the decolonial approach, but as an illustration of how these flows work and can be interpreted through a decolonial lens. This section thus seeks to open up new questions and insights regarding the production and transfer of knowledge between regions in this field. In this regard, our analyses are an example of how the coloniality of knowledge can serve to interpret the existing imbalance in these flows of knowledge and practices.

The primacy of the west As shown in Table 22.1, most of the studies evaluate programmes located in Western Europe, followed by studies that have evaluated programmes located in sub-Saharan Africa. Seen in this way, despite the overrepresentation of Western Europe within the set of studies, studies are relatively present in almost all regions of the world, and in fact, the presence of studies in sub-Saharan Africa, Southeast Asia, the Middle East, and North Africa could demonstrate a certain balance in the production of knowledge on these issues in the world. However, this portrait of reality should be viewed with caution, as it conceals other elements that reveal a Western-centric approach to the production of knowledge in this field. If we compare, for example, this information with the number of terrorist incidents collected in the Global Terrorism Database (GTD) in the same regions and during the same years in which these studies were published, that is, between 2001 and 2019, the overrepresentation of programmes evaluated in Western countries is evident. One might expect, for example, that the most affected regions should also be leaders in the production of knowledge that seeks to prevent these incidents. However, the regions where most terrorist incidents have occurred during this period are not those where there has been the most knowledge production around the programmes implemented, namely, South Asia, Southeast Asia, the MENA, and sub-Saharan Africa. On the contrary, regions where proportionally fewer incidents have occurred are overrepresented in the production of knowledge about extremism prevention practices. Western Europe, for example, which has experienced 2.65% of terrorist incidents during this period, accounts for 42.5% of the scientific output in this area. The Australasia and Oceania region, where the only country represented in the published studies is Australia, suffered 13 incidents per published study, while the region with the most incidents, the MENA region, suffered 2749.8 terrorist incidents per published study. This overrepresentation has obvious consequences for the way prevention is understood around the world. If 53.9% of the studies published in this field have evaluated programmes located in the West (Australasia and Oceania, Western Europe, and North America), then what we know about the practices that work or do not work about the factors that influence the success or failure of the implementation of a programme, comes mostly from a particular cultural world and responds to the specific needs of its local level. A local knowledge that, due to its majority presence in the production of knowledge or to the power relations that exist between the various regions of the world, is converted into a universal knowledge and applicable to any cultural or geographical context. Programmes in other regions that are based on this universalised knowledge cannot help but fail to reduce these incidents at the local level. This consideration does not obviously seek to reduce the number of publications in the West, but to

344

Re-colonising the field of evaluation of PVR Table 22.1 Region of Origin of the Evaluated Programme, of the Authors of the Evaluation, and of the Terrorist Incidents

Australasia and Oceania Central Asia East Asia South Asia Southeast Asia Middle East and North Africa Sub-Saharan Africa Eastern Europe Western Europe North America South America Central America and Caribbean Unspecified region6 Total

Origin Region of the Evaluated Programme*

Origin Region of the Evaluation Authors*

Region Where a Terrorist Incident Occurred (2001–19)**

N

N

N

%

%

%

7 3 – 11 21 17 41 2 93 18 – –

3.2 1.4 – 5.0 9.6 7.8 18.7 0.9 42.5 8.2 – –

27 5 – 22 33 12 15 9 143 108 – –

6.9 1.3 – 5.7 8.5 3.1 3.9 2.3 36.8 27.8 – –

92 218 202 43185 10900 46746 17045 3901 3344 837 2952 105

0.07% 0.17% 0.16% 33.34% 8.42% 36.09% 13.16% 3.01% 2.58% 0.65% 2.28% 0.08%

6 219

2.8 100

15 389

3.9 100

– 126268

– 100.00%

Source: Created by the authors based on (*) Madriaza et al., 2022 and (**) START, 2022

understand how the specific weight of local knowledge can be counterproductive when it becomes universal or the only source of data.

Who publishes and what is its origin? A second element to consider in the Western-centric approach to knowledge on prevention practices is knowing who publishes and what their origin is. We were able to identify 389 authors, 71.5% of whom came from Western countries, even though only 53.9% of these studies evaluated programmes in these regions (Table 22.1). With this overall result, we can already hypothesise preliminarily that an important part of the studies in other regions have been conducted by Western researchers, but to confirm this hypothesis, we need to go a little further. The authors come mainly from Western Europe (36.8%) and North America (27.8%). The first case is logical since most of the programmes evaluated are European. On the other hand, the second is quite surprising if we consider that only 8.2% of the studies have evaluated programmes in this region, that is, for each study in our database, there are six North American researchers involved. Indeed, as we will see later, many programmes in other regions, particularly in sub-Saharan Africa, have been evaluated by American researchers (see Table 22.2); however, before changing the line of argument, let us analyse in more detail the ratios by region between the number of publications and the number of authors. The next regions with a high ratio are Eastern Europe (4.5) and Australasia and Oceania (3.9). Of the nine authors from the first region, eight are authors of the two studies identified in Eastern Europe, and one participated in a study in Western Europe. In the case of the latter, on the other hand, the situation is relatively different. Of the 27 authors from this region (Australia), 17 conducted studies in

345

Table 22.2  Origin Region of the Evaluation Authors by Region of First Publication Region of First Publication

Origin Region of the Evaluation Authors Australasia and Oceania

Central Asia

South Asia

Southeast Asia

Middle East and North Africa

SubSaharan Africa

Eastern Europe

Western Europe

North America

Total

N

17

0

0

0

0

0

0

1

0

%

89.5%

0.0%

0.0%

0.0%

0.0%

0.0%

0.0%

5.3%

0.0%

100.0%

Central Asia

N %

0 0.0%

5 62.5%

1 12.5%

0 0.0%

0 0.0%

0 0.0%

0 0.0%

1 12.5%

1 12.5%

8 100.0%

South Asia

N %

0 0.0%

0 0.0%

20 58.8%

1 2.9%

0 0.0%

0 0.0%

0 0.0%

0 0.0%

10 29.4%

34 100.0%

Southeast Asia

N %

6 13.3%

0 0.0%

1 2.2%

31 68.9%

1 2.2%

0 0.0%

0 0.0%

2 4.4%

4 8.9%

45 100.0%

Middle East and North Africa

N

0

0

0

0

10

0

0

2

12

%

0.0%

0.0%

0.0%

0.0%

38.5%

0.0%

0.0%

7.7%

46.2%

100.0%

Sub-Saharan Africa

N %

2 2.7%

0 0.0%

0 0.0%

0 0.0%

15 20.3%

0 0.0%

16 21.6%

31 41.9%

74 100.0%

Eastern Europe

N %

0 0.0%

0 0.0%

8 88.9%

1 11.1%

0 0.0%

9 100.0%

Western Europe

N %

1 0.8%

0 0.0%

0 0.0%

0 0.0%

1 0.8%

0 0.0%

1 0.8%

109 88.6%

11 8.9%

123 100.0%

North America

N %

1 2.5%

0 0.0%

0 0.0%

0 0.0%

0 0.0%

0 0.0%

0 0.0%

2 5.0%

37 92.5%

40 100.0%

Total

N %

27 6.9%

5 1.3%

22 5.7%

33 8.5%

12 3.1%

15 3.9%

9 2.3%

143 36.8%

108 27.8%

389 100.0%

0 0,0%

Source: Created by the authors based on Madriaza et al. (2022)

1 1,4% 0 0,0%

0 0,0%

0 0,0%

19

26

Pablo Madriaza, David Morin, and Vivek Venkatesh

346

Australasia and Oceania

Re-colonising the field of evaluation of PVR

this region, while the others conducted evaluations in Southeast Asia and sub-Saharan Africa. This information supports the study by Ilyas (2021b) on Indonesia regarding the significant presence of Australians in the prevention industry in the country. On the other side of the equation, the regions with the least presence of local authors compared to publications that have evaluated programmes in these regions are sub-Saharan Africa (0.4) and the MENA (0.7), which confirms the fact that the evaluations conducted in these regions are mainly done by researchers from the West. Table 22.2 shows a comparison between the author’s region of origin and the region in which they first published in our database. Of the authors, 89.5% made only one evaluation, so this criterion seems to be sufficient. This table allows us to see the specific weight of each region in relation to the origin of the authors. The first element that stands out is the fact that, in the western regions, the authors are mainly local (Australasia and Oceania = 89.5%; Western Europe = 88.6%; North America = 92.5%) and thus have little influence from other regions. Authors from other regions who evaluate programmes in the West are also mostly Western. Only one author from the MENA region and one from Eastern Europe participated in an evaluation of Western Europe. In other words, knowledge in the West is mainly endogamous and has little permeability to other cultural and geographical contexts. This is important because it relates to the value placed on a specific type of knowledge, that is, knowledge produced in the West in relation to knowledge from other regions. Why, for example, would a sub-Saharan researcher have the right to conduct an evaluation in the United States on a local programme? What could be the added value in relation to local researchers? The answers to these questions are obvious: they lack in-depth knowledge of the local level, they do not necessarily speak the language, and there are local researchers with similar skills to conduct this evaluation. However, going outside Western space, these answers seem less obvious. The most dramatic cases are precisely sub-Saharan Africa, where barely 20% of the authors are local, and in the MENA region, where this percentage reaches 38.5% of the authors of these evaluations. In both cases, it is mainly North Americans (sub-Saharan Africa = 41.9%, MENA = 46.2%) and Europeans (sub-Saharan Africa = 21.6%, MENA = 7.7%) who carry out the evaluations. These regions are thus extremely permeable to knowledge that responds to criteria and needs that are not specific to them and which reflects the imbalance and hierarchy in the production of knowledge in which the West appears to have an overwhelming supremacy. One of the few non-Western regions where this permeability is less important is Southeast Asia, where 68.9% of the studies were conducted by local authors, a percentage that partly contradicts what Ilyas (2021b) said in the case of Indonesia. Despite this, 26.6% of the studies involved Westerners, mainly from Australia (13.3%) and North America (8.9%).

The need for a linguistic turn A final element to consider with regard to a Western-centric view is linguistic. English was evidently the language of writing for most of the studies (only five studies were published in French and none in Spanish). The fact that these studies are written in English has the obvious advantage of potentially greater dissemination of the results and scientific exchange. However, at the same time, this is an obstacle for many actors in the field who cannot necessarily access them. For example, if we take the African and Asian continents as a whole, out of the total number of evaluative studies, in 31 of the 50 studies identified in Africa and 29 of the 42 in Asia, the official language of the country concerned was not English. Despite this, all the reports were written in English. In Africa, although almost half of the studies evaluated programmes in countries where French is the official language or language of use, none of them were written in French. In many cases, as referred to by the evaluators themselves in the limits of their studies, the local language 347

Pablo Madriaza, David Morin, and Vivek Venkatesh

was a barrier to the evaluation. In other words, the evaluators, almost exclusively Westerners, did not speak the language of the country where they conducted the evaluation and had to rely on translators or interlocutors to conduct interviews in the field. Apart from this problem of access to the field, these researchers had to rely on the reflections of intermediaries and could not read the programme documentation written in a language other than their own, which in most cases was English. Would a similar situation be acceptable in a Western country? Could a researcher who does not speak English or French conduct an evaluation of a programme in Canada, then publish its results in a language other than English or French, and still be considered a legitimate evaluation? As we have indicated earlier, this has nothing to do with the quality of the methods used, with most of these evaluations having an excellent assessment of the quality of the methods used. However, beyond quality, this information tells us about the legitimacy of these studies, which were not designed to provide information to programme designers and implementers on the ground and thus have an impact on improving practice, especially in non-Western countries, but rather to meet the needs of Western donors and Western scientific criteria and needs.

Conclusion This analysis has been used as an example to illustrate, based on the evaluated programmes, the primacy of the West in the production of knowledge in the world in relation to PVR and the relationship of subalternisation and dependence that most of the other regions have in relation to this knowledge. More specifically, the knowledge we have globally on the practice of preventing violent extremism comes mainly from one region, the West. This is true if we consider that most of the evaluations have been conducted in Western programmes, but it is also true of a significant portion of the programmes evaluated outside this region, which have also been conducted primarily by Western researchers. This is especially dramatic in the case of sub-Saharan Africa and the MENA region, but it is also true to varying degrees in all non-Western regions. However, if we consider these data by the relative weight of each knowledge, according, for example, to the sources of information used in each of the publications that have evaluated PVR programmes, it is likely that, given the influence and the greater scientific production of Western thought, the numbers presented here will increase on a logarithmic basis. As a result, on the one hand, the indigenous knowledge of these regions is little known or is simply excluded, and, therefore, the main knowledge available in the regions most affected by terrorist incidents is knowledge that has little to do with their local needs or that uses theoretical and empirical lenses belonging to the particularities of another region: the West. On the other hand, the flow of Western-centric North-South knowledge and practices has been fundamentally unidirectional. The West draws little from the knowledge produced in other regions. It is clear from this analysis that the evaluation of programmes for the prevention of violent extremism is a territory largely colonised by Western countries. As Spivak (1988) mentions, knowledge is never innocent since it expresses the interests of those who have produced it, and as we have seen in other decolonial studies (Ilyas, 2021b; Oando & Achieng’, 2021), it is also not trivial since it is associated with the capacity to exercise power. This dependence thus makes it possible to legitimise certain points of view and delegitimise those that could potentially contradict or propose alternative visions, and in this way makes it possible to justify political agendas in other regions that align with the policies of Western countries. The origin of the evaluator and the language used are, however, extremely limited indicators of the coloniality of knowledge in this field. We were able, for example, to identify only studies in English, French, or Spanish. It is quite possible that studies written in other languages could not have been identified with this systematic review. This is particularly true for studies in European countries 348

Re-colonising the field of evaluation of PVR

such as Germany, the Netherlands, or the Nordic countries. However, this does not seem to be substantially likely in non-Western countries, given the dependence of these countries on international funding and the use of English as a lingua franca. We also do not know the exact origin of the evaluators, but rather the country in which they work. Western countries are increasingly multicultural, and many researchers working in Western centres have a different origin from this region and could have a more in-depth view of the terrain in which they evaluate. A relatively deeper analysis would have considered this factor, as well as the authors cited within the papers, to have a broader perspective on this coloniality of knowledge. However, this does not necessarily contradict the hegemony of Western thought. Theoretically, a Western individual can have a decolonial way of seeing the world, and a non-Western individual can have a captive mind, as mentioned by Alatas (2000), of Western categories and modes of thought. It is not a question of individuals but of a system of colonialism that has repercussions at the individual, institutional, or societal level in different regions of the world. This analysis is thus a starting point that has served to illustrate how the transregional flows of knowledge and practices function in terms of the PVR programme evaluation, but it obviously does not exhaust the subject. It has allowed us, for the moment, to timidly establish a questioning of the widespread influence that Western thought has in this area and of the consequences that this universalisation may have on the definition of concepts and, therefore, on the actions and policies that may result from it. New questions are thus raised that will require a diversity of studies to show, first, the forms of coloniality that emerge as universal narratives in the production of knowledge as well as in the practices and public policies for the prevention of this violence, both in Western countries and from a transregional comparative perspective. Second, it should analyse the impact that this colonial interpretation has had on ethnocultural communities in Western countries, as well as on the populations of the South, where this universal knowledge has been applied. Finally, it should seek to go beyond the North-South dichotomy and look for conceptual and practical models of hybridisation and mestisation of the encounter between the West and other regions that account for alternative ways out of this universal thinking.

Notes 1 In this chapter, we will understand violent radicalisation and violent extremism as synonyms, as well as preventing violent radicalisation (PVR) and preventing violent extremism (PVE). In this latter case, we will favour the use of the acronym PVR. 2 There is no scientific journal with this name. Ilyas is probably referring to the journal ‘Critical Studies on Terrorism’. 3 Performing the same exercise in relation to postcolonial studies [postcolon*], the search yielded 41 articles when this word was mentioned in the body of the text and only three when it was searched exclusively in the abstracts. 4 Other studies use the term ‘decolonial’ rather anecdotally. This is the case in the study by Rapanyane and Sethole (2021), who, despite only superficially addressing this approach, justify their belonging to this new paradigm by the simple fact of being non-Western researchers. 5 The complete methodology of this systematic review can be found in Annex B of this report. 6 These programmes target digital interventions through the Internet and thus do not have a specific region of origin.

References Alatas, S. H. (2000). Intellectual imperialism: Definition, traits, and problems. Southeast Asian Journal of Social Science, 28(1), 23–45. Coolsaet, R. (2016). “All radicalisation is local”: The genesis and drawbacks of an elusive concept (Vol. 84). Egmont-Royal Institute for International Relations.

349

Pablo Madriaza, David Morin, and Vivek Venkatesh Crosby, A. (2021). Reverberations of empire: Criminalisation of asylum and diaspora dissent in Canada. Critical Studies on Terrorism, 14(2), 179–200. https://doi.org/10.1080/17539153.2021.1899598 Dussel, E. D. (2000). Europe, modernity, and eurocentrism. Nepantla: Views from South, 1(3), 465–478. Escobar, A. (2007). Worlds and knowledges otherwise: The Latin American modernity/coloniality research program. Cultural Studies, 21(2–3), 179–210. https://doi.org/10.1080/09502380601162506 Grosfoguel, R. (2009). A decolonial approach to political-economy: Transmodernity, border thinking and global coloniality. Kult, 6(1), 10–38. Hassan, G., Ousman, S., Madriaza, P., Fetiu, A., Boily, L.-A., Levesque, F., Squalli, Z., Ajrouche, K., ElTahry, N., Lampron-De Souza, S., Desmarais, L., Duong, E., & Moyano, R. (2020). D’un océan à l’autre: Cartographie des initiatives de prévention secondaire et tertiaire oeuvrant dans un contexte de radicalisation et d’extrémisme violent au Canada. Réseau des Praticiens Canadiens pour la Prévention de la Radicalisation et de l’Extrémisme Violent (RPC-PREV). https://cpnprev.ca/wp-content/uploads/2020/12/ FR-Rapport-mapping-final-1.pdf Ilyas, M. (2021a). Decolonising terrorism journals. Societies, 11(1), 6. https://doi.org/10.3390/soc11010006 Ilyas, M. (2021b). Decolonising the terrorism industry: Indonesia. Social Sciences, 10(2), 53. https://doi. org/10.3390/socsci10020053 Khan, R. M. (2021). Race, coloniality and the post 9/11 counter-discourse: Critical terrorism studies and the reproduction of the Islam-terrorism discourse. Critical Studies on Terrorism, 1–4. https://doi.org/10.1080 /17539153.2021.1983112 Kundnani, A. (2009). Spooked! How not to prevent violent extremism. Institute of Race Relations. Kundnani, A. (2012). Radicalisation: The journey of a concept. Race & Class, 54(2), 3–25. International Bibliography of the Social Sciences (IBSS); Sociological Abstracts. https://doi.org/10.1177/0306396812454984 Kundnani, A., & Hayes, B. (2018). The globalisation of countering violent extremism policies: Undermining human rights, instrumentalising civil society. Transnational Institute. Madriaza, P., Morin, D., Hassan, G., Venkatesh, V., Plaude, M., Deli, C., Girard, M., Durocher-Corfa, L., Grijalva-Lavallée, R., & Poulin, K. (2022). Que savons-nous de l’évaluation de programmes de prévention de l’extrémisme violent: Une revue systématique méthodologique des évaluations des programmes de prévention de l’extrémisme violent. Chaire UNESCO en prévention de la radicalisation et de l’extrémisme violents (Chaire UNESCO-PREV). Madriaza, P., Ponsot, A.-S., Marion, D., Monnier, C., Ghanem, A., Zaim, B., Nait-Kac, W., Hassani, N., & Autixier, C. (2017). The prevention of radicalization leading to violence: An international study of frontline workers and intervention issues. International Centre for the Prevention of Crime. Majozi, N. (2018). Theorising the Islamic state: A decolonial perspective. ReOrient, 3(2), 163–184. https:// doi.org/10.13169/reorient.3.2.0163 Malo, A., Morin, D., Audet-Gosselin, L., Madriaza, P., & Al Baba Douaihy, S. (2021). La prévention de l’extrémisme violent dans les revues scientifiques – REVUE 2020. Chaire UNESCO en prévention de la radicalisation et de l’extrémisme violents (Chaire UNESCO-PREV) et Centre de prévention de la radicalisation menant à la violence (CPRMV). Mignolo, W. (2007). Delinking: The rhetoric of modernity, the logic of coloniality and the grammar of decoloniality. Cultural Studies, 21(2–3), 449–514. https://doi.org/10.1080/09502380601162647 Mignolo, W. (2011). Geopolitics of sensing and knowing: On (de)coloniality, border thinking and epistemic disobedience. Postcolonial Studies, 14(3), 273–283. https://doi.org/10.1080/13688790.2011.613105 Nesbitt, M. (2021). Violent crime, hate speech or terrorism? How Canada views and prosecutes far-right extremism (2001–2019). Common Law World Review, 50(1), 38–56. https://doi.org/10.1177/1473779521991557 Oando, S., & Achieng’, S. (2021). An indigenous African framework for counterterrorism: Decolonising Kenya’s approach to countering “Al-Shabaab-ism”. Critical Studies on Terrorism, 14(3), 354–377. https:// doi.org/10.1080/17539153.2021.1958182 Quijano, A. (2007). Coloniality and modernity/rationality. Cultural Studies, 21(2–3), 168–178. https://doi. org/10.1080/09502380601164353 Ragazzi, F. (2014). Towards a policed multiculturalism? Counter-radicalization in France, Netherlands and the United Kingdom. Études Du CERI, 206. Rapanyane, M. B., & Sethole, F. R. (2021). A decolonial perspective on Boko Haram’s campaign of terror, sources of funding, mobilisation strategies and major attacks since 2009. The Strategic Review for Southern Africa, 43(1), Article 1. https://doi.org/10.35293/srsa.v43i1.339 Said, E. W. (1995). Orientalism: Western conceptions of the orient (Reprinted with a new afterword). Penguin Books.

350

Re-colonising the field of evaluation of PVR Spivak, G. C. (1988). Can subaltern speak ? In C. Nelson & L. Grossberg (Eds.), Marxism and the interpretation of culture (pp. 271–313). University of Illinois Press. START. (2022). Global terrorism database (GTD). University of Maryland. https://start.umd.edu/gtd/ Taylor, J. D. (2020). “Suspect categories,” alienation and counterterrorism: Critically assessing PREVENT in the UK. Terrorism and Political Violence, 32(4), 851–873. https://doi.org/10.1080/09546553.2017.14 15889 Venkatesh, V., Podoshen, J., Wallin, J., Rabah, J., & Glass, D. (2020). Promoting extreme violence: Visual and narrative analysis of select ultraviolent terror propaganda videos produced by the Islamic state of Iraq and Syria (ISIS) in 2015 and 2016. Terrorism and Political Violence, 32(8), 1753–1775. https://doi.org/10.1080/ 09546553.2018.1516209

351

PART IV

Countering radicalisation Strategies and challenges

23 RETHINKING CVE AND PUBLIC HEALTH PREVENTION Towards health promotion Keiran Hardy

Introduction Countering violent extremism (‘CVE’) refers to diverse strategies for preventing radicalisation and terrorism. The acronym is not typically used by governments, which state various goals in their counter-terrorism policies. These include building resilience (Government Offices of Sweden, 2016), promoting democracy (German Federal Government, 2016), and ‘confronting longterm contributors to domestic terrorism’ (The White House, 2021). However, CVE has become a common label – an umbrella term or catch-all – for such strategies. Essentially, CVE captures ‘soft’ approaches to counter-terrorism as distinct from more typical ‘hard’ approaches, such as prosecution and imprisonment (Aly et al., 2015). The distinction between ‘soft’ and ‘hard’ counterterrorism is not always clear, creating suspicion around the underlying motives of some CVE programmes (Kundnani, 2009; Thomas, 2010). As a result, CVE remains poorly defined (Gielen, 2019). Nonetheless, the initialism has stuck; at least, it groups together research and practice that addresses long-term causes of terrorism. Probably the most influential conceptual framework in CVE is the public health model. This is said to divide prevention into three types: primary, secondary, and tertiary (Challgren et al., 2016; Harris-Hogan et al., 2016). Primary prevention targets communities or society as a whole (e.g. through education and social cohesion programmes). Secondary prevention targets those at risk of radicalisation (e.g. through mentoring programmes for vulnerable youth). Tertiary prevention targets those who have already radicalised (e.g. through prison deradicalisation and EXIT programmes that help people leave extremist groups). These describe a pyramid structure, with community-based prevention as a foundation, moving up through targeted interventions to the pointy end of CVE, targeting more hardened offenders. The analogy is to public health strategies for preventing disease in a population (see Harris-Hogan et al., 2016). For example, to address the risks of heart disease, governments can educate the public about the benefits of exercise and healthy eating (primary prevention). Screening programmes and interventions target people at higher family or statistical risk of developing the disease (secondary prevention). Treatment and rehabilitation programmes help those who are diagnosed and experiencing symptoms (tertiary prevention).



355

DOI: 10.4324/9781003035848-27

Keiran Hardy

There is significant scholarly support for public health prevention as a guiding framework for CVE (Bhui et al., 2012; Challgren et al., 2016; Harris-Hogan et al., 2016; Shortland et al., 2021). Others warn against it (Aggarwal, 2019; Heath-Kelly, 2017), and the model certainly raises some doubts. We might first ask how well the analogy holds up. Medical conditions can and will affect everyone at some point in their lives, so the general public can benefit from national campaigns around exercise, dieting, and the dangers of smoking. On the other hand, most people in society will never have any propensity for violent extremism. Can we all benefit directly from CVE efforts? Of greater concern is the use of language around disease, contagion, and infection to describe risky individuals and the spread of extremist ideas. These concepts might accurately describe health problems, but are they appropriate for social problems that are frequently politically, culturally, and religiously sensitive? Finally, we might ask, what principles and values does public health prevention draw upon beyond the pyramid structure? Does the model, as some scholars have suggested (Aggarwal, 2019; Heath-Kelly, 2017), encourage risk analysis and surveillance, rendering everyone a potential suspect? In this chapter, I investigate these and related questions. My aim is to assess whether public health prevention provides a useful and appropriate conceptual framework for CVE. In Part II, I assess the strengths and weaknesses of three arguments supporting the model. First is a conceptual argument that public health prevention helps to analyse CVE efforts and achieve greater balance in programme design (Challgren et al., 2016; Harris-Hogan et al., 2016). The second is an ethical argument that it avoids creating suspect communities (Bhui et al., 2012; Shortland et al., 2021). The third is an empirical argument that it contributes to successful CVE outcomes (Bhui et al., 2012; Challgren et al., 2016). In these three respects, there are not only strengths to a public health approach but also limitations and several unknowns. For example, regarding the empirical argument, we know little about the overlap in the social determinants that create a greater propensity for both poor health outcomes and violent extremism. In Part III, I argue that public health approaches to CVE should focus on ‘health promotion’. In this part, I examine prevention from the perspective of public health scholarship and key bodies such as the World Health Organization (WHO). From these sources, it becomes clear that the pyramid comprising primary, secondary, and tertiary approaches is a feature of public health prevention but not its core focus. Instead, ‘health promotion’ through primary prevention, combined with robust evaluation, is the hallmark of successful public health campaigns. Rethinking CVE along these lines would ensure that public health approaches have long-term utility for the field. This would mean moving away from a ‘disease prevention’ mindset, which aims to combat violent extremism, towards a more positive goal: creating the kinds of healthy societies in which violent extremism is less likely to develop.

Public health prevention and CVE There is widespread support in the literature for public health prevention as a conceptual framework for CVE. According to the WHO (2021b, 2012), which relies on the classic definition offered by Acheson (1988), ‘public health is the science and art of preventing disease, prolonging life and promoting health through the organized efforts of society’. The field has its origins in 19th-century attempts to eradicate disease by improving living conditions (e.g., through improved sanitation and housing) (Fraser, 2005). Now, public health has wider goals across ‘the entire spectrum of health and wellbeing’ (WHO, 2021b). It is difficult to tie down the first discussion of public health and CVE, though Bhui et al. (2012) provide an early example before wider uptake in the mid-2010s (Challgren et al., 2016; 356

Rethinking CVE and public health prevention

Harris-Hogan et al., 2016; Weine & Eisenman, 2016). The Danish government offered a similar pyramid – comprising preventive, anticipatory, and direct interventions – in its Action Plans for preventing radicalisation and extremism, though those documents did not identify public health as a guiding rationale (Danish Government, 2016, 2014). Growing support for a public health approach aligned roughly with the threat from Islamic State, though research on foreign fighters tended to focus on risk factors and reintegration (Borum & Fein, 2017; Dawson & Amarasingam, 2017; Lindekilde et al., 2016). The genesis likely had more to do with developing a ‘second generation’ of more successful, whole-of-government approaches to CVE, following the perceived failures of the UK’s police-led Prevent strategy (Home Office, 2011; Kundnani, 2009; Thomas, 2010). Support for a public health approach continues, focusing largely on the prevention pyramid (Garcia, 2019; Shortland et al., 2021). Advocates argue that CVE should balance primary prevention (e.g. social cohesion and education programmes), secondary prevention (e.g. mentoring or other support for at-risk individuals), and tertiary prevention (e.g. disengagement and deradicalisation efforts). Figure 23.1 shows this tripartite structure. Public health also encourages a multi-disciplinary, multi-agency approach (Challgren et al., 2016) rather than one led by police or counter-terrorism agencies. This reflects what is happening on the ground, with policymakers in Denmark, the United States, and other countries relying on similar frameworks – whether based explicitly on public health or not – to diversify their CVE efforts (Andersson Malmros, 2021; Challgren et al., 2016).

Figure 23.1  Public health prevention in CVE Source: Created by author

357

Keiran Hardy

Support for a public health approach can be broken down into three more specific arguments. First is a conceptual argument that the model provides greater clarity in an uncertain field. The second is an ethical argument that it avoids creating suspect communities, and the third is an empirical argument that it contributes to successful CVE outcomes. Below, I address the strengths and weaknesses of these arguments in turn.

Conceptual clarity? A key strength of the public health model is that it allows researchers and policymakers to understand and analyse CVE efforts. In an influential article, Harris-Hogan et al. (2016) argue that public health prevention ‘distinguishes key components of CVE and presents a framework to categorise the range of programs within this field’. They use the framework to analyse CVE programmes in Australia, arguing that the country has focused disproportionately on primary prevention efforts, which relate only tangentially to counter-terrorism (Harris-Hogan et al., 2016). They recommend a shift towards secondary prevention, which would direct resources to those most in need of support. In addition, they warn against focusing too heavily on tertiary prevention, which comes into play only once someone has already radicalised (Harris-Hogan et al., 2016). Used in this way, the public health pyramid provides an analytical framework that allows researchers to assess the relative balance of CVE strategies, identify preventive gaps, and recommend strategic readjustments. Challgren et al. (2016) highlight similar benefits. Writing as members of a National Security Critical Issues Task Force, they explain that ‘CVE has an unclear definition, mission, and leadership structure’ (Challgren et al., 2016). They argue that public health frameworks allow governments to define their ‘CVE mission’ more clearly, ‘assign responsibilities for its implementation, and engage effectively in combined action’ (Challgren et al., 2016). In other words, the model helps not only to analyse the relative balance of different preventive strategies but also to clarify and coordinate CVE responsibilities across different agencies. Andersson Malmros (2021) uses the prevention pyramid to categorise actors and practices involved in 60 municipal CVE policies across Norway, Denmark, and Sweden. Similar to the Australian study, he finds the greatest focus on primary prevention. He identifies police and corrections as the most important tertiary actors, compared to primary prevention efforts, which are typically led by local governments, social services, schools, libraries, and youth centres. He demonstrates that ‘policy on violent extremism is becoming increasingly localised’ and highlights problematic practices of surveillance, which require teachers, social workers, and other public employees to identify radicalisation risks (Andersson Malmros, 2021). In this way, the public health model provides a springboard for deeper critical analysis of CVE practices, their impacts, and the levels of governance at which these occur. Drilling down further reveals some notable limitations. The pyramid approximately divides CVE efforts, but fitting a programme into one of the three types can be difficult. Some programmes aim to educate communities or improve community cohesion, but they target groups or communities perceived to be at greater risk of violent extremism, either geographically or demographically (Harris-Hogan et al., 2016). Should these be classified as primary prevention because of the wider community-based approach or secondary because risk characterises the selection? Heath-Kelly (2017) goes further, arguing that primary and secondary prevention merge beyond distinction under public health approaches to CVE because entire populations are assumed to be at risk. In the UK’s Prevent strategy, she argues, ‘everyone’ and ‘at risk individuals’ are one and the same (Heath-Kelly, 2017). 358

Rethinking CVE and public health prevention

Distinguishing between secondary and tertiary prevention can be equally difficult. A  clear example of tertiary prevention is when corrective services aim to rehabilitate a terrorist offender in prison, but the lines are not so clear elsewhere. The UK’s Channel programme, for example, involves interventions for individuals at risk of radicalisation, those showing signs of radicalisation, and those already radicalised (Home Office, 2020, 2012). Even if a clear line could be drawn between these three sets of circumstances, details of individual cases are not made public, so researchers cannot know the extent to which Channel is mostly, in practice, an example of secondary or tertiary prevention. The targets of CVE strategies can also be fluid, depending on who they speak to, who receives the message, or who chooses to use a service. One example is counter-narratives (Braddock, 2019; Braddock & Horgan, 2016). These could be classified as primary or secondary prevention, depending on whether they target an entire population or at-risk groups or who sees the messaging online. Other initiatives, such as telephone helplines, are available for entire populations but provide confidential support either to concerned family members or to individuals at risk of extremism (Department of Communities and Justice, 2022). Further examples defy categorisation. Governments are increasingly regulating social media companies, for example, to address the spread of online extremism (He, 2020). These laws do not target populations, at-risk individuals, or those already radicalised, though the impacts could be felt by all three. The public health pyramid therefore lacks some of the nuance required to understand CVE programmes on a deeper level, especially counter-narratives and platform regulation. Nonetheless, it provides an intuitive template and common language to categorise CVE efforts across different countries and levels of governance. It usefully captures the difference between preventive strategies targeting populations, at-risk individuals, and those already radicalised even if finer distinctions might be drawn. It sheds light on the shape and structure of these CVE efforts and aids critical analysis. Importantly, as Andersson Malmros (2021) shows, it reflects how policymakers have been shaping CVE efforts on the ground. The framework provides much-needed clarity to the CVE field, which remains poorly defined and theorised (Gielen, 2019). Without it, researchers and policymakers would lack a common framework to guide critical discussions of counter-terrorism’s ‘softer’ side.

Enhanced ethics? A second argument in favour of the public health model is that it avoids known ethical problems with some CVE strategies. A key criticism of both counter-terrorism laws and CVE programmes is that they construct Muslim communities as inherently suspicious (Cherney & Murphy, 2016; Kundnani, 2009; Pantazis & Pemberton, 2009; Thomas, 2010). Bhui et al. (2012) believe a public health approach avoids these damaging perceptions because it focuses on social determinants of extremism, including marginalisation and discrimination. Rather than framing Muslim or other communities as inherently risky, addressing these underlying causes is more likely to ‘foster social inclusion and social justice in communities that feel threatened by terrorism’ (Bhui et al., 2012). They believe that this can ‘help destigmatize “suspect communities” ’ created by counter-terrorism laws and CVE programmes (Bhui et al., 2012). Shortland et al. (2021) favour a public health approach when addressing other ethical questions. They argue that public health ethics can help determine if CVE interventions affecting individual liberty are justified and proportionate. They offer similar criteria to those offered by Ashworth and Zedner (2014) in their work on preventive justice. CVE measures will be justified, according to Shortland et al. (2021), if they are ‘necessary, effective, minimally infringing on individual rights, and subject 359

Keiran Hardy

to public accountability’. A preventive justice approach shows that public health frameworks are not needed to develop ethical approaches to CVE (Ashworth & Zedner, 2014; Hardy, 2017). However, as the world has seen through the COVID-19 pandemic, public health deals with complex ethical problems that require individual liberties to be weighed against a wider public benefit. More research into the relevance of public health ethics for CVE could solidify these possible benefits. These benefits are largely speculative, though, and must be weighed against some possible dangers. One is the risk, highlighted by Heath-Kelly (2017) and Aggarwal (2019), that a public health approach encourages population surveillance, leading to ‘false positives’. Heath-Kelly (2017) traces the development of public health approaches and their historical translation into the criminal justice system. She argues that Prevent relies on an epidemiological model that frames Muslims as inherently suspicious and that the ‘deployment of Prevent through the entire social services network renders the whole-of-population as an object of surveillance’ (Heath-Kelly, 2017). She points to a significant increase in Channel referrals, even though 90% of the individuals did not ultimately require support (Heath-Kelly, 2017). Similarly, Aggarwal (2019) argues that many recognised problems with Prevent, including the violation of patient confidentiality, derive from a medicalised approach. He argues that CVE, like psychiatry, involves ‘judgments of normal and abnormal behaviours’ and that this has justified widespread community surveillance and intrusive state action (Aggarwal, 2019). In these respects, a public health approach to CVE may have helped precisely to build ‘suspect communities’ – not destigmatise them. The spread of CVE through multiple institutions – reflecting a multi-disciplinary, public health approach – has certainly contributed to damaging perceptions of spying and surveillance. The revised Prevent strategy (Home Office, 2011) aimed to avoid perceptions that CVE was too closely linked to coercive counter-terrorism (Thomas, 2010). It did so by placing obligations, formalised through the 2015 Prevent Duty, on hospitals, schools, universities, and other institutions to identify at-risk individuals (Home Office, 2021). This in turn generated significant criticism that the government was impinging on academic and institutional freedom by requiring all public employees to be on the lookout for potential terrorists without any consensus on identifiable risk factors (Hubble, 2015). It raises significant ethical concerns about doctors, teachers, social workers, and others not in law enforcement or intelligence who are assigned some responsibility, legally or professionally, to assess individual propensities for violent extremism (Hubble, 2015; Mattsson & Säljö, 2018; McKendrick & Finch, 2017; van der Weert & Eijkman, 2019; Younis & Jadhav, 2020). Labelling at-risk individuals and the spread of extremism with medicalised language raises additional concerns. Shortland et al. (2021), for example, frame violent extremism as a form of ‘social contagion’. Weine et al. (2017) speak of ‘diagnosing violent extremism’. In his influential work on counter-messaging, Braddock (2019) relies on the metaphors of ‘inoculation’ and ‘vaccination’. This research and the messages underlying it are significant, but the choice of language may create damaging perceptions that people who hold extremist beliefs are medically ill. This conflicts with the consensus that there is no psychological profile or medical pathology that can predict violent extremism (Horgan, 2014; Silke, 1998). It is also perhaps unnecessarily alarmist because violent extremism cannot spread exponentially, like a virus, among entire populations. The language also implies a level of scientific knowledge that is not yet available in CVE. Medical analogies appear more damaging when viewed in the shadow of colonial history. Similar metaphors have a long history in counter-insurgency military operations, in which insurgents are treated as a ‘virus’ to be eradicated from a healthy population. FM 3-24, the controversial counter-insurgency manual used by the United States military in Iraq and Afghanistan, likened these operations to three stages of medical intervention: (1) administer emergency first 360

Rethinking CVE and public health prevention

aid to stop the bleeding; (2) administer inpatient care to help the patient through long-term recovery; and (3) administer outpatient care to help the patient move towards self-sufficiency (United States Army, 2007). Kilcullen (2009) wrote about insurgencies as a ‘virus’ or ‘bacterium’ that infects a host population’s immune system. He described the walls around Baghdad’s gated communities as an ‘urban tourniquet’ that would stop a ‘life-threatening hemmorhage’ of refugees (Kilcullen, 2007). These metaphors have colonial roots (French, 2011; Newsinger, 2011) and reflect a discredited past in which foreign populations with ideologies opposed to the West were viewed as ‘infected’ with dangerous thoughts. CVE researchers and policymakers supporting a public health approach should acknowledge the discriminatory, historical practice of dividing populations into ‘healthy’ and ‘infected’, depending on whether their beliefs were acceptable to a colonial administration. These risks do not undermine the need to address social determinants of extremism, as a public health approach suggests (Bhui et al., 2012). There is also ongoing value in asking whether CVE interventions affecting individual liberty are necessary and proportionate (Shortland et al., 2021). However, it is worth recognising that the ethical case for a public health approach is not clear-cut and raises some notable dangers. Compared to the analytical benefits, the ethical case for a public health approach to CVE is less clear.

Successful outcomes? A third argument in favour of a public health approach is that it contributes to successful CVE outcomes. There are two aspects to this argument. First, public health programmes may contribute directly to CVE because there is overlap between risk factors for extremism and other ‘social ills’ (Challgren et al., 2016). According to Challgren et al. (2016), established prevention programmes addressing mental health issues, marginalisation, social grievances, and unemployment can be used to combat radicalisation. Bhui et al. (2012) make a similar case, arguing that a public health approach can ‘address common issues of grievance or marginalisation’. Public health programmes addressing marginalisation, social grievances, and unemployment may well contribute to positive CVE outcomes, given their relevance as risk factors for radicalisation. Indeed, many CVE programmes are designed with these risk factors in mind (Home Office, 2012). The links between mental health and terrorism are less clear; they continue to be investigated, with stronger evidence of mental illness playing a causal role in lone-actor terrorism (Corner & Gill, 2015). More evidence is needed, however, on the relevance of mental illness across the spectrum of cognitive to behavioural radicalisation (Bhui et al., 2020). Links between mental health, social ills, and terrorism are further tempered by the lack of scholarly consensus on psychological risk factors (Horgan, 2014; Neumann, 2013; Silke, 1998). The idea that prevention programmes across public health and CVE can address shared risk factors (Bhui et al., 2012; Challgren et al., 2016) is therefore encouraging but not yet supported by the available evidence. This is conceded by those who support a public health approach to CVE on empirical grounds (Bhui et al., 2012; Weine et al., 2017). More convincing is a second limb of the empirical argument: that CVE strategies are most likely to be effective if they draw upon similar multi-agency, multi-disciplinary approaches to those found in public health. As Weine et al. (2017) explain: Public health involves diverse disciplines relevant to CVE such as psychiatry, psychology, sociology, communications, education, and public policy. Public health professionals carry out their work through many relevant approaches too, including developing and 361

Keiran Hardy

implementing community-based programs, administering services, conducting research and evaluation, and recommending policies. All these characteristics make public health a potential framework for understanding the multiple aspects of violent extremism. Given the lack of a clear psychological profile for who is likely to radicalise (Horgan, 2014; Silke, 1998), addressing diverse risk factors through multiple disciplinary lenses is an appropriate, overarching strategy. It is consistent with the suggestion that CVE should balance primary, secondary, and tertiary prevention (Challgren et al., 2016; Harris-Hogan et al., 2016). It must be recognised, however, that public health does not provide any direct evidence suggesting how CVE programmes should be designed. The field has its own extensive evidence base demonstrating continued successes, including in vaccination programmes and anti-smoking campaigns (Australian Institute of Health and Welfare [AIHW], 2020; Public Health Association of Australia, 2018). These do not speak directly to CVE, or even suggest that programmes based on similar principles are likely to be effective. CVE programmes that draw upon aspects of public health prevention – including a blend of primary, secondary, and tertiary approaches – must be assessed on their own terms through robust evaluation frameworks. Evaluation research is improving significantly (Carthy et  al., 2020; Cherney, 2020; Winterbotham, 2020a, 2020b), though it may be difficult for CVE to reach a similar scientific standard for two reasons. The first is that social problems are less amenable to quantification. For example, it can be known, according to recognised medical criteria, whether someone has heart disease, diabetes, or another medical condition. It cannot be said with the same level of certainty whether someone has radicalised, is radicalising, or is at risk of radicalising, or how many of each there might be in a population. Definitions of radicalisation and violent extremism vary across governments (Hardy, 2018), and there is no scholarly consensus on when someone has radicalised to a point of concern (Neumann, 2013). Even if rates of radicalisation could be accurately measured, they are far lower than those for common diseases. The number of violent extremists in a country might be in the hundreds or low thousands. In Australia, heart disease alone affects around 5% of the population – more than a million people each year (Australian Bureau of Statistics [ABS], 2018b). It kills around 50 people per day, or one person every 30 minutes (Heart Foundation, 2020). These mortality rates have been measured for more than half a century (ABS, 2018a). CVE may never reach these levels of precision in statistical mapping for which public health is known. Second, due to increased levels of secrecy around counter-terrorism matters, CVE research suffers from a lack of available data (Mastroe, 2016; Safer-Lichtenstein et al., 2017; Sageman, 2014). This puts CVE researchers in a vastly different position from public health researchers, who often draw upon open-source, national survey data (ABS, 2018a). This does not discount the overall benefit of a multi-pronged strategy based on the prevention pyramid, but it cannot be assumed that public health provides direct evidence for CVE or that CVE as a field could achieve the same high standards of scientific knowledge. A public health approach does, however, affirm the need for rigorous, ongoing programme evaluation. It favours interventions that are ‘evidence-based, cost-effective and applicable on a large scale’ (Rivara & Johnston, 2013). As risk factors for poor health outcomes are identified, interventions to address them are developed, monitored, and evaluated, ideally in randomised control trials and longitudinal studies (Rivara & Johnston, 2013). These processes are designed to achieve ‘long-term sustainability of outcomes’ (AIHW, 2014). CVE may lag public health in terms of the available evidence base, but developing it remains a key goal.

362

Rethinking CVE and public health prevention

Towards health promotion The concerns outlined earlier suggest that researchers and policymakers should be cautious about applying a public health approach to CVE – or at least transplanting it directly. However, two key features of public health prevention, separate from the prevention pyramid and overlooked by the CVE literature to date, may have greater long-term utility for the field. These are a focus on the general public and the idea of ‘health promotion’. CVE scholars typically overstate the role of the prevention pyramid in public health. Some public health sources offer the archetypal pyramid (Centers for Disease Control and Prevention [CDC], 2017), but others blur the lines significantly across the three levels. Definitions of secondary prevention, for example, commonly refer to screening programmes (CDC, 2017; WHO, 2005). These can be either population-based or targeted towards high-risk groups (Australian Government Department of Health, 2018), suggesting they straddle primary and secondary prevention. Other definitions of secondary prevention include early treatment, such as drugs to prevent symptoms from worsening (WHO, 2021a). This blurs the lines with tertiary prevention. Other formulations leave tertiary prevention off altogether or define it to include planning, compliance, and administration in addition to rehabilitation and treatment (WHO, 2021a; Commonwealth Health Hub, 2021). In other words, the prevention pyramid lacks clarity not only in CVE but also in its field of origin. Two features of public health prevention characterise the field more clearly. The first, as the name suggests, is that public health is for the public’s benefit. In contrast to medical treatment provided by General Practitioners and hospitals, public health focuses on populations rather than individual patients or cases of disease (Marks et al., 2011). This is a ‘unifying principle’ of public health (WHO, 2012) in a way that the prevention pyramid, with its various formulations, arguably is not. It implies that the focus of prevention programmes should be at the primary level (Fraser, 2005; Marks et al., 2011; WHO, 2012). By contrast, Harris-Hogan et al. (2016) suggest that primary prevention is less relevant to CVE and that favour should be given to secondary prevention programmes targeting ‘at-risk’ individuals. A second key feature, reflecting a population-based approach, is the idea of ‘health promotion’. The term was coined in 1945 by Henry E. Sigerist, a medical historian who defined the four major tasks of medicine as promoting health, preventing illness, restoring the sick, and rehabilitation (Kumar & Preetha, 2012). Health promotion involves efforts to improve health literacy, but it is not the same thing as health education (Kumar & Preetha, 2012). Rather, it recognises that the circumstances leading to negative health outcomes are socio-economic, often generational, and not the result of poor individual choices. Health promotion aims to educate and empower communities. It is about ‘promoting and encouraging healthy behaviours and environments’ (Kumar & Preetha, 2012). Health promotion addresses the social determinants of negative health outcomes, as previously suggested for CVE (Bhui et al., 2012). It reflects a multi-disciplinary, multi-agency approach for which CVE is increasingly known (Shortland et al., 2021). However, embracing health promotion would reorient the main task away from preventing radicalisation towards creating the kinds of healthy societies in which radicalisation is less likely to occur. Health promotion aims to build healthy communities by improving the conditions in which ‘good health’ can develop (Kumar & Preetha, 2012). Many successful health promotion campaigns involve a simple message for behavioural change: get active, wear a helmet, put on sunscreen, stop smoking. These are effective and avoid stigmatising at-risk groups when they are advertised nationally to an entire population (Rivara &

363

Keiran Hardy

Johnston, 2013). Examples in Australia include ‘Slip Slop Slap’ for skin cancer and ‘Life Be In It’ for healthy lifestyles. Another from the United States was a campaign reminding parents to put their children ‘Back to Sleep’ to avoid Sudden Infant Death Syndrome (SIDS) (Rivara & Johnston, 2013). Strategic decisions to ‘go national and go broad’ are a defining feature of many successful public health campaigns (Rivara & Johnston, 2013). In CVE, a similar national campaign could encourage community members, for example, to meet people from different cultural backgrounds and learn their customs and traditions. Clearly, social inclusion policies must be supplemented with other, more targeted strategies and interventions. The prevention pyramid still has utility in categorising CVE efforts and identifying strategic gaps. However, a broader, population-focused approach would more accurately reflect successful public health approaches compared to the task of balancing primary, secondary, and tertiary prevention (Harris-Hogan et al., 2016; Challgren et al., 2016). Above all, a public health approach to CVE should aim to improve social cohesion and build inclusive societies. It should discourage a ‘disease prevention’ mindset, which aims to reduce instances of radicalisation. Good health means not only the absence of illness or infirmity (i.e. a lack of extremism and radicalisation) but also the presence of physical, social, and mental wellbeing (WHO, 2005). This is reflected globally in United Nations Sustainable Development Goal #3, which aims to ‘ensure healthy lives and promote well-being for all at all ages’ (United Nations Department of Economic and Social Affairs, 2021).

Conclusion Public health prevention is an influential model for CVE. Given that the field lacks clarity and definition, organising CVE programmes into three broad categories is a useful task. A public health approach aims to identify and address social risk factors through a holistic, multi-disciplinary lens. These features are consistent with the CVE agenda. The significant evidence base from public health does not speak directly to CVE, but it provides insights into effective harm prevention. At the same time, applying the public health model to CVE raises some difficulties and dangers. It can be difficult to categorise CVE efforts neatly into primary, secondary, and tertiary approaches. The medical metaphors of disease, infection, and contagion are inappropriate and inaccurate when applied to the spread of violent extremism. Importantly, CVE researchers currently lack the evidence base to know if prevention programmes built on similar principles are likely to be effective over the long term. Given these benefits and dangers, what utility does public health prevention hold for the future of CVE? A starting point is to recognise that public health is not only, or even primarily, about balancing primary, secondary, and tertiary prevention. Rather, public health aims to improve the physical, social, and mental well-being of populations. It seeks to reduce harm by educating and empowering communities. It is about ‘encouraging healthy behaviors and environments’ (Weine et al., 2017), not assessing risk or treating sickness in individual cases. Framing public health in this way has very different implications for CVE. A  public health approach does not mean striking some arbitrary balance between primary, secondary, and tertiary prevention. It should mean improving the well-being of populations so that radicalisation in individual cases is less likely to occur. In terms of the well-known pyramid model, the strongest case is for primary prevention. This conflicts with previous suggestions that primary prevention is too broad to be effective in countering extremism (Harris-Hogan et al., 2016). An ongoing need is to avoid stigmatising at-risk groups. Successful public health campaigns achieve this by making a strategic decision to ‘go national and go broad’ (Rivara & Johnston, 2013). 364

Rethinking CVE and public health prevention

In CVE, similar campaigns could promote the kinds of behaviours that contribute to a healthier, more inclusive society. These could be as simple as meeting your neighbours, attending cultural events, or volunteering in your community – anything that could reduce division and hatred in our societies. Such strategies do not replace the need for more targeted interventions, prison deradicalisation, and EXIT programmes. But they could provide the foundation for a preventive approach more accurately based on public health campaigns. Understood in these terms, the public health model can continue to provide significant benefits for CVE research and policy. It requires only some rethinking and refocusing, away from the need to prevent disease and identify risky individuals, towards a health promotion agenda that aims to improve social conditions over the long term.

References Acheson, D. (1988). On the state of the public health [the fourth Duncan lecture]. Public Health, 102(5), 431–437. Aggarwal, N. K. (2019). Questioning the current public health approach to countering violent extremism. Global Public Health, 14(2), 309–317. Aly, A., Balbi, A., & Jacques, C. (2015). Rethinking countering violent extremism: Implementing the role of civil society. Journal of Policing, Intelligence & Counter-Terrorism, 10(1), 3–13. Andersson Malmros, R. (2021). Mobilizing to counter violent extremism: Variance and translation in municipal policies. International Journal of Public Administration, forthcoming (copy on file with author). Ashworth, A., & Zedner, L. (2014). Preventive justice. Oxford University Press. Australian Bureau of Statistics. (2018a). Changing patterns of mortality in Australia. www.abs.gov.au/ statistics/health/causes-death/changing-patterns-mortality-australia/latest-release Australian Bureau of Statistics. (2018b). National health survey: First results. www.abs.gov.au/statistics/ health/health-conditions-and-risks/national-health-survey-first-results/latest-release Australian Government Department of Health. (2018). Population based screening framework. www.health. gov.au/resources/publications/population-based-screening-framework Australian Institute of Health and Welfare. (2014). Australia’s health 2014. www.aihw.gov.au/reports/ australias-health/australias-health-2014/contents/table-of-contents Australian Institute of Health and Welfare. (2020). Australia’s health 2020 in brief. www.aihw.gov.au/reports/ australias-health/australias-health-2020-in-brief/contents/summary Bhui, K. S., Hicks, M. H., Lashley, M., & Jones, E. (2012). A public health approach to understanding and preventing violent radicalization. BMC Medicine, 10(16), 1–8. Bhui, K. S., Otis, M., Joao Silva, M., Halvorsrud, K., Freestone, M., & Jones, E. (2020). Extremism and common mental illness: Cross-sectional community survey of White British and Pakistani men and women living in England. British Journal of Psychiatry, 217(4), 547–554. Borum, R., & Fein, R. (2017). The psychology of foreign fighters. Studies in Conflict & Terrorism, 40(3), 248–266. Braddock, K. (2019). Vaccinating against hate: Using attitudinal inoculation to confer resistance to persuasion by extremist propaganda. Terrorism and Political Violence. https://doi.org/10.1080/09546553.2019 .1693370 Braddock, K., & Horgan, J. (2016). Towards a guide for constructing and disseminating counternarratives to reduce support for terrorism. Studies in Conflict & Terrorism, 39(5), 381–404. Carthy, S. L., Doody, C. B., Cox, K., O’Hora, D., & Sarma, K. M. (2020). Counter-narratives for the prevention of violent radicalisation: A systematic review of targeted interventions. Campbell Systematic Reviews, 16(3), e1106. Centers for Disease Control and Prevention. (2017). Picture of America: Prevention. www.cdc.gov/pictureofamerica/ pdfs/Picture_of_America_Prevention.pdf Challgren, J., Kenyon, T., Kervick, L., Scudder, S., Walters, M., Whitehead, K., Connor, J., & Rollie Flynn, C. (2016). Countering violent extremism: Applying the public health model. https://georgetownsecuritystudiesreview.org/wp-content/uploads/2016/10/NSCITF-Report-on-Countering-Violent-Extremism.pdf Cherney, A. (2020). Evaluating interventions to disengage extremist offenders: A study of the proactive integrated support model (PRISM). Behavioral Sciences of Terrorism and Political Aggression, 12(1), 17–36.

365

Keiran Hardy Cherney, A., & Murphy, K. (2016). Being a “suspect community” in a post-9/11 world: The impact of the war on terror on Muslim communities in Australia. Australian and New Zealand Journal of Criminology, 49(4), 480–496. Commonwealth Health Hub. (2021). Prevention. www.thecommonwealth-healthhub.net/prevention/ Corner, E., & Gill, P. (2015). A false dichotomy? Mental illness and lone-actor terrorism. Law and Human Behavior, 39(1), 23–34. Danish Government. (2014). Prevention of radicalisation and extremism: Action plan. www.justitsministeriet.dk/ sites/default/files/media/Pressemeddelelser/pdf/2015/SJ20150422125507430%20%5BDOR1545530% 5D.PDF Danish Government. (2016). Preventing and countering extremism and radicalisation: National action plan. https://ec.europa.eu/home-affairs/sites/default/files/what-we-do/networks/radicalisation_awareness_ network/docs/preventing_countering_extremism_radicalisation_en.pdf Dawson, L. L., & Amarasingam, A. (2017). Talking to foreign fighters: Insights into the motivations for Hijrah to Syria and Iraq. Studies in Conflict & Terrorism, 40(3), 191–210. Department of Communities and Justice. (2022). Step together. NSW Government. https://steptogether.nsw. gov.au/ Fraser, J. (2005). Population and public health in Australian general practice: Changes, challenges and opportunities. Australian Family Physician, 34(3), 177–179. French, D. (2011). The British way in counter-insurgency: 1945–1967. Oxford University Press. Garcia, M. (2019, April 3). A public health approach to countering violent extremism. Just Security. www. justsecurity.org/63455/a-public-health-approach-to-countering-violent-extremism/ German Federal Government. (2016). Federal government strategy to prevent extremism and promote democracy. www.bmfsfj.de/bmfsfj/meta/en/publications-en/federal-government-strategy-to-preventextremism-and-promote-democracy-115450 Gielen, A. (2019). Countering violent extremism: A realist review for assessing what works, for whom, in what circumstances, and how? Terrorism and Political Violence, 31(6), 1149–1167. Government Offices of Sweden. (2016). Actions to make society more resilient to violent extremism. www. government.se/legal-documents/2016/01/skr.-201415144/ Hardy, K. (2017). Preventive justice principles for countering violent extremism. In T. Tulich, R. AnanianWelsh, S. Bronitt, & S. Murray (Eds.). Regulating preventive justice: Principle, policy and paradox (pp. 117–135). Routledge. Hardy, K. (2018). Comparing theories of radicalisation with countering violent extremism policy. Journal for Deradicalisation, 18, 76–110. Harris-Hogan, S., Barrelle, K., & Zammit, A. (2016). What is countering violent extremism? Exploring CVE policy and practice in Australia. Behavioral Sciences of Terrorism and Political Aggression, 8(1), 6–24. He, D. (2020). Governing hate content online: How the Rechtsstaat shaped the policy discourse on the NetzDG. International Journal of Communication, 14, 3746–3768. Heart Foundation. (2020). Key statistics: Coronary heart disease. www.heartfoundation.org.au/ activities-finding-or-opinion/coronary-heart-disease-key-stats Heath-Kelly, C. (2017). The geography of pre-criminal space: Epidemiological imaginations of radicalisation risk in the UK prevent strategy, 2007–2017. Critical Studies on Terrorism, 10(2), 297–319. Home Office. (2011). Prevent strategy (Cm 8092). www.gov.uk/government/publications/preventstrategy-2011 Home Office. (2012). Channel: Protecting vulnerable people from being drawn into terrorism: A guide for local partnerships. HM Government. Home Office. (2020). Channel duty guidance: Protecting people vulnerable to being drawn into terrorism. www.gov.uk/government/publications/channel-and-prevent-multi-agency-panel-pmap-guidance Home Office. (2021). Revised prevent duty guidance: For England and Wales. www.gov.uk/government/ publications/prevent-duty-guidance/revised-prevent-duty-guidance-for-england-and-wales Horgan, J. (2014). The psychology of terrorism (2nd ed.). Routledge. Hubble, S. (2015). Freedom of speech and preventing extremism in UK higher education institutions (Briefing Paper CBP 7199). House of Commons Library. Kilcullen, D. (2007, April 27). The urban tourniquet: “Gated communities” in Baghdad. Small Wars Journal. https://smallwarsjournal.com/blog/the-urban-tourniquet-gated-communities-in-baghdad Kilcullen, D. (2009). The accidental guerrilla: Fighting small wars in the midst of a big one. Oxford University Press.

366

Rethinking CVE and public health prevention Kumar, S., & Preetha, G. S. (2012). Health promotion: An effective tool for global health. Indian Journal of Community Medicine, 37(1), 5–12. Kundnani, A. (2009). Spooked! How not to prevent violent extremism. Institute of Race Relations. Lindekilde, L., Bertelsen, P., & Stohl, M. (2016). Who goes, why, and with what effects: The problem of foreign fighters from Europe. Small Wars & Insurgencies, 27(5), 858–877. Marks, L., Hunter, D. J., & Alderslade, R. (2011). Strengthening public health capacity and services in Europe. World Health Organization. www.euro.who.int/__data/assets/pdf_file/0007/152683/e95877.pdf Mastroe, C. (2016). Evaluating CVE: Understanding the recent changes to the United Kingdom’s implementation of prevent. Perspectives on Terrorism, 10(2), 50–60. Mattsson, C., & Säljö. (2018). Violent extremism, national security and prevention: Institutional discourses and their implications for schooling. British Journal of Vocational Studies, 66(1), 109–125. McKendrick, D., & Finch, J. (2017). “Under heavy manners?” Social work, radicalisation, troubled families and non-linear war. British Journal of Social Work, 47, 308–324. Neumann, P. (2013). The trouble with radicalization. International Affairs, 89(4), 873–893. Newsinger, J. (2011). British counterinsurgency (2nd ed.). Palgrave Macmillan. Pantazis, C., & Pemberton, S. (2009). From the “old” to the “new” suspect community: Examining the impacts of recent UK counter-terrorist legislation. British Journal of Criminology, 49(5), 646–666. Public Health Association of Australia. (2018). Top 10 public health successes over the last 20 years (PHAA Monograph Series No. 2). www.phaa.net.au/documents/item/3241 Rivara, F. P., & Johnston, B. (2013). Effective primary prevention programs in public health and their applicability to the prevention of child maltreatment. Child Welfare, 92(2), 119–139. Safer-Lichetenstein, A., LaFree, G., & Loughran, T. (2017). Studying terrorism empirically: What we know about what we don’t know. Journal of Contemporary Criminal Justice, 33(3), 273–291. Sageman, M. (2014). The stagnation in terrorism research. Terrorism and Political Violence, 26(4), 565–580. Shortland, N., Evans, N., & Colautti, J. (2021). A public health ethics model of countering violent extremism. Terrorism and Political Violence, 33(2), 324–337. Silke, A. (1998). Cheshire-cat logic: The recurring theme of terrorist abnormality in psychological research. Psychology, Crime and Law, 4(1), 51–69. Thomas, P. (2010). Failed and friendless: The UK’s “preventing violent extremism” programme. British Journal of Politics and International Relations, 12(3), 442–458. United Nations Department of Economic and Social Affairs. (2021). Goal 3. https://sdgs.un.org/goals/goal3 United States Army. (2007). Counterinsurgency field manual: US army field manual no. 3–24. University of Chicago Press. van der Weert, A., & Eijkman, Q. A. M. (2019). Subjectivity in detection of radicalisation and violent extremism: A youth worker’s perspective. Behavioral Sciences of Terrorism and Political Aggression, 11(3), 191–214. Weine, S., & Eisenman, D. P. (2016, April 5). How public health can improve initiatives to counter violent extremism. START: National Consortium for the Study of Terrorism and Responses to Terrorism. www. start.umd.edu/news/how-public-health-can-improve-initiatives-counter-violent-extremism Weine, S., Eisenman, D. P., Kinsler, J., Glik, D. C., & Polutnik, C. (2017). Addressing violent extremism as public health policy and practice. Behavioral Sciences of Terrorism and Political Aggression, 9(3), 208–221. TheWhite House. (2021). National strategy for countering domestic terrorism. www.whitehouse.gov/wp-content/ uploads/2021/06/National-Strategy-for-Countering-Domestic-Terrorism.pdf Winterbotham, E. (2020a, September  30). How effective are mentorship interventions? Assessing the evidence base for preventing and countering violent extremism (RUSI Occasional Papers). https://rusi.org/explore-our-research/publications/occasional-papers/how-effective-are-mentorshipinterventions-assessing-evidence-base-preventing-and-countering-violent Winterbotham, E. (2020b, 11 May). What can work (and what has not worked) in women-centric P/CVE initiatives: Assessing the evidence base for preventing and countering violent extremism. RUSI Occasional Papers. https://rusi.org/explore-our-research/publications/occasional-papers/what-can-work-andwhat-has-not-worked-women-centric-pcve-initiatives-assessing-evidence-base World Health Organization. (2005). Final report: WHO global forum IV on chronic disease prevention and control. https://apps.who.int/iris/handle/10665/43208 World Health Organization. (2012). European action plan for strengthening public health capacities and services (Regional Committee for Europe, 62nd session). www.euro.who.int/en/health-topics/Health-systems/

367

Keiran Hardy public-health-services/publications/2012/european-action-plan-for-strengthening-public-health-capacities-and-services World Health Organization. (2021a). About us: Health promotion and disease prevention through populationbased interventions, including action to address social determinants and health inequity. www.emro.who. int/about-who/public-health-functions/health-promotion-disease-prevention.html World Health Organization. (2021b). Public health services. www.euro.who.int/en/health-topics/ Health-systems/public-health-services Younis, T., & Jadhav, S. (2020). Islamophobia in the national health service: An ethnography of institutional racism in PREVENT’s counter-radicalisation policy. Sociology of Health & Illness, 42(3), 610–626.

368

24 MULTI-AGENCY APPROACHES TO COUNTERING RADICALISATION Robin Andersson Malmros and Jennie Sivenbring

Introduction A report on prevention of radicalisation from the European Commission suggests that ‘prevention work at the local level must be conducted jointly by several actors: schools, social services and the police are some of the main actors to be included in a multi-agency approach’ (European Commission, 2019, p. 7). The quote successfully highlights the taken-for-granted role of multi-agency (MA) approaches in contemporary local efforts to prevent or counter violent extremism. This paradigm extends beyond Europe’s borders to, for example, the United States (Ambrozik, 2018), Australia (Cherney, 2016; Cherney & Belton, 2021; Harris-Hogan et al., 2016), Asia (Crone & Nasser, 2018; El-Said, 2015), and Africa (Breidlid, 2021). Organising work on countering violent extremism (CVE) using MA structures is commonly framed as a response to the wickedness of radicalisation and extremism (Dahlgaard-Nielsen, 2016; Rittel & Webber, 1973) and functions as an organisational platform for strategic, interventional, and intelligence gathering work (Mazerolle et al., 2021; Sivenbring & Andersson Malmros, 2019, 2021; Stephens & Sieckelinck, 2019). In MA structures, various efforts that target the public, specific ‘risk’ communities or groups, and/or individuals (Gielen, 2019) are planned, designed, and implemented, often based on the information shared between the involved actors. van de Weert and Eijkman (2019) stated that an important objective of MA structures is to ‘increase information-sharing and monitoring of violent extremism at the local level’ (p. 191), which means that ‘municipalities have been given joint responsibility for the early detection of potentially risky individuals’ (p. 191). While MA approaches have long constituted a bureaucratic response to complex policy issues, not least in the related issue of local crime prevention (e.g. Cherney, 2016; Crawford, 1994; Hardy, 2022), the application of MA approaches to CVE is still an emerging but vibrant stream of research. This chapter aims to provide an organisational theory perspective, based on insights from new institutional theory, on what we call ‘local MA approaches’ applied to CVE. More specifically, we will concentrate on describing and analysing how MA approaches are organised in administratively defined local areas and arenas (e.g. a specific residential area, a school, or a municipality). Our focus is on locally oriented public agencies, which we define as the formal organisations of the public sector that predominantly pursue the collective interests of local administrations and communities (Christensen et al., 2020). Consequently, ‘multi-agency’ requires two or more such

369

DOI: 10.4324/9781003035848-28

Robin Andersson Malmros and Jennie Sivenbring

agencies (e.g. agencies in the school sector and police and social services) to collaborate in a local administrative setting (e.g. in a local government or municipality). Collaboration, another key term in understanding MA approaches, is a complex, ambiguous, and multifaceted term, but we use it to describe ‘a cooperative, inter-organizational relationship that is negotiated in an ongoing communicative process that relies on neither market nor hierarchal mechanisms of control’ (Lawrence et al., 2002, p. 282). The choice to use the new institutional theory approach to analyse contemporary MA approaches is based on two observations. First, we build on the statement of Mazerolle et al. (2021) in their systematic literature review that there ‘are a lot of unsubstantiated assumptions about why multiagency approaches are good for community safety, yet organisational theories are not well integrated into the logic models that underpin the purpose, capacities, functioning and stated outcomes of these collaborations’ (p. 79). Second, and connected to that observation, we argue that CVE research lacks an understanding of the social and cultural rationales that underpin MA approaches. Here, new institutional theory constitutes a dominant theoretical framework in studies of organisations and how they are affected by and affect their regulative, technical, normative, and cognitive – cultural institutional environment (DiMaggio & Powell, 1983; Meyer & Rowan, 1977). We now outline the four-section structure of this chapter. In the first section, we describe the foundations of MA approaches and the processes that led to their becoming an institutionalised way of organising local CVE work. In the second section, we review two of the most prominent examples of local MA approaches (the Danish Aarhus model and the British Channel programme) and analyse their respective purposes, organisational designs, embedded practices, and identified effects and consequences. In the third section, based on the information in the second section, we address the following key debates in this stream of research: (1) the translation of ‘best’ practices between organisational contexts; (2) the social and cultural bases of information gathering and sharing practices; (3) conflicts between involved professionals; and (4) the appropriateness and constituents of MA approaches. Finally, we conclude the chapter and identify some directions for future research.

Foundations Addressing wicked problems such as CVE has called for public officials to forge management structures that recognise and correspond with the degree of complexity inherent to the challenge (Weber & Khademian, 2008). Hence, the most common structural approach to such problems has been the utilisation of diverse types of collaborative governance structures (Head & Alford, 2015). Such governance structures typically have both cross-sectoral and multi-level members and are formalised, consensus-oriented, knowledge-intensive, and focused on implementing a particular policy or solving a public issue (Ansell & Gash, 2008; Weber & Khademian, 2008). One of the earliest policy issues for which MA approaches were utilised, and perhaps the most comparable and influential to CVE, is crime prevention. For example, Crawford (1994) demonstrated that MA approaches to crime prevention have had a prominent place in British policy since the 1980s. Ragazzi (2017) argued that the transition to MA approaches was the result of a managerial attempt to protect society from crime rather than address the root causes of crime. This implied a pluralisation of social control (cf. Jämte & Ellefsen, 2020; Johansen, 2020) that had the following organisational consequences: ‘Joined-up’ teams, multi-agency and community – police partnerships, became the new model of crime prevention. . . . Narrow, professionalised groups of individuals (police, 370

Multi-agency approaches to countering radicalisation

social workers [and] youth workers) should not have a monopoly on the maintenance of social order. Instead, the entire spectrum of professions that are in contact with the public should be involved. (Ragazzi, 2017, pp. 169–170) Accordingly, it is not surprising to see prominent criminologists theorising on and studying MA approaches to CVE (e.g. Bjørgo, 2016; Cherney, 2016; Hardy, 2022; McCulloch & Pickering, 2009). Bjørgo (2016) suggested that ‘within democratic societies, counter-terrorism is almost exclusively about crime prevention’ (p. 25). On the same note, Hardy (2022) argued that ‘researchers and policymakers should think about CVE as a collection of diverse preventive strategies designed to reduce risks of criminal behaviour’ (p. 643). Analogous to counter-terrorism and CVE, crime prevention strategies have both a ‘hard’ and a ‘soft’ side to them. For example, Bjørgo (2016) refers to social crime prevention as attempts to affect ‘the conditions that make people criminals or get them involved in crime, addressing risk factors as well as protective factors’ (p. 26). Given the ideational and ideological similarities between CVE and crime prevention, it is therefore not surprising to see that municipalities embed CVE efforts in their crime prevention strategies. This has been noticed in Scandinavia (Andersson Malmros, 2021; Andersson Malmros & Mattsson, 2017; Hemmingsen, 2015; Johansen, 2020; Lindekilde, 2012a), Canada (O’Halloran, 2021), and the United Kingdom (e.g. Ragazzi, 2017).

Multi-agency approaches to Countering Violent Extremism (CVE) Dahlgaard-Nielsen (2016) suggested that MA approaches ‘[offer] the best hope of matching the complexity and dynamism of the phenomenon of violent extremism’ (p. 136) because actors working together are ‘much more likely to be able to intervene effectively to counter individual cases of violent extremism than any one single government agency acting on its own’ (p. 136). From a more strategic, goal-oriented perspective, Stephens and Sieckelinck (2019) identified two broad purposes of MA approaches to CVE. The first purpose is to strengthen society by (1) building organisational capacity through the development of a diverse toolbox in which different actors (e.g. schools, civil society organisations, the private sector, and police) can assist; (2) improving dialogue and networks between groups perceived to be opposed to or unlike each other; and (3) building trust and legitimising efforts by creating links between ‘grassroots’ and local public-sector actors (e.g. municipal employees). The second purpose is intelligence sharing and intervention centred on (1) creating links that can be used to facilitate the flow of information concerning perceived risks of radicalisation; (2) assessing identified risks and formulating appropriate interventions; and (3) facilitating training and sharing of best practices (Stephens & Sieckelinck, 2019). While MA approaches have gained an institutionalised, taken-for-granted status in large parts of the global North, strong scientific evidence of their effectiveness is lacking. Mazerolle et al. (2021) found, from their review of literature on MA approaches involving the police, that ‘there is no clear evidence [that] multiagency partnerships involving [the] police are effective approaches [to] reducing radicalisation to violence’ (p. 76). In another literature review, Pistone et al. (2019) concluded that there are ‘no evidence-based interventions that prevent and counter the development of the intention to commit acts of violent extremism’ (p. 22). On the basis of these conclusions, it is difficult to identify the types of MA approaches that might be considered more or less effective. Instead, we will use a multiple case-study design to describe 371

Robin Andersson Malmros and Jennie Sivenbring

two formalised local MA approaches – the Aarhus model from Denmark and the British Channel programme – and highlight their organisational features in relation to five variables: main purposes, organisational design, important practices, power relations, and effects or consequences. The cases were strategically chosen based on three criteria. First, they demonstrate how MA approaches have been implemented in various ways, depending on the context, despite building on the same tradition of crime prevention with the police as the central actor. Second, both the Channel programme and the Aarhus model have served as inspirations for MA approaches developed in other countries, which suggests that their organisational features and rationales are representative of other contexts. Third, the two approaches have been widely discussed in academic literature. While we are aware of our highly northern European-centric selection of cases, we believe that the cases reflect the status of policy development and research in the field. However, we are seeing more and more empirical research on local MA approaches coming from other parts of the world (e.g. Breidlid, 2021; Cherney et al., 2018; Ellis et al., 2020).

The Aarhus model As mentioned in the previous section, the Aarhus model is, together with the British Channel programme, perhaps the most well-known example of a local MA approach and has served as an inspiration for many countries. Today, after spreading throughout Denmark’s police districts, it is known simply as the Danish approach (Bertelsen, 2015; Hemmingsen, 2015; Johansen, 2020). The Aarhus model is based on an already established crime prevention model: the School, Social Services, and Police (SSP) model. In 2009–10, it was redeveloped for CVE (Johansen, 2020). The CVE organisational design integrates what has been termed Info-houses into the SSP model. The term house is misleading since Info-houses are not physical buildings but: serve as ‘preventive hubs’ anchored regionally in (12) police districts . . . centers for sharing knowledge and expertise, and a forum where the police and municipal actors in the area can discuss methods, actions and challenges in general preventive work and specific cases. (Sivenbring & Andersson Malmros, 2019, p. 54) Besides integrating municipal professionals (e.g. social workers and teachers) with regional professionals (i.e. police), the Info-houses also have a national component: the Danish Security Intelligence Service and its prevention unit (Hemmingsen, 2015; Tammikko, 2018). Occasionally, the Info-houses also involve other actors and collaborative networks. For example, the Psychiatry, Social Services, and Police (PSP) network has been used to forward cases of concern to the Infohouses and as a forum for coordinating interventions (Sestoft et al., 2014, 2017). Taken together, these features suggest that the Aarhus model is a highly pluralised and expansive effort that engages a wide range of actors as ‘preventive agents that depend on the frontline as a monitoring system’ (Johansen, 2020). The model aligns the aforementioned two purposes of MA approaches to CVE: strengthening society and intelligence sharing and intervention (Stephens & Sieckelinck, 2019). However, the Aarhus model is predominantly associated with intelligence sharing and intervention. Sestoft et al. (2017) described the following typical process in the model when working with a referral: When PET’s Prevention Centre receives a report of possible radicalization, they conduct an intelligence-led assessment. If this assessment evidences that there is no threat to national 372

Multi-agency approaches to countering radicalisation

security, but there is still a case of radicalization risk, the case is referred back to the Infohouse for rehabilitation measures. (p. 352) Here, it should be noted that other evidence suggests that the flow of intervention and guidance between PET and Info-houses is bidirectional (Crone & Nasser, 2018; Hemmingsen, 2015). Crone and Nasser (2018) referred to an ‘open’ and ‘classified’ system of referrals; the open system derives referrals from all frontline workers, and the classified system from PET’s use of surveillance techniques and information. Lindekilde (2014) complemented this practice by outlining what happens in the next phase of the process: Trained personnel screen notifications and, in cases of concern, pass information on to a steering committee consisting of municipality employees with special training in radicalisation prevention, the local SSP (School – Social services – Police collaboration) coordinator and specially trained police officers, who further investigate the case by, for example, contacting teachers, parents [et al.]. If the steering committee finds that intervention is called for, the type of intervention is discussed and the individual may be offered a mentor. (p. 230) Depending on the nature of the concern, this process can lead to interventions that range from different forms of counselling, financial support, housing, therapy, Exit programmes, and mentoring programmes to PET’s so-called preventive interviews (Bertelsen, 2015; Johansen, 2020; Lindekilde, 2014; Sivenbring & Andersson Malmros, 2019; Tammikko, 2018). Lindekilde (2014) argued that the individual mentoring programmes have grown in importance and have become central to the current approach. There are also practices associated with strengthening society that are embedded in the Aarhus model. For example, Bertelsen (2015) advocated the use of a series of two-hour workshops directing schools to help ‘pupils and teachers recognise risk factors and markers of possible radicalisation processes among their peers to be able to spot lurking radicalising influences from recruiting groups’ (p. 244). Other practices include a parenting network to support parents so that they may contribute to anti-radicalisation efforts; dialogues ‘especially with Muslim communities’ (Bertelsen, 2015, p. 244); and advanced training for the involved professionals (especially the mentors, workshop instructors, and network facilitators). Concerning power relations, the police have a dominant position (Sivenbring & Andersson Malmros, 2019). While social welfare actors play an important role in designing and implementing practices, the police have organisational responsibility for setting up and leading both SSP networks and Info-houses (Sivenbring & Andersson Malmros, 2019). Evidence also suggests that police agencies have considerable influence on the types of efforts that will be implemented. Sestoft et al. (2017) stated, ‘PET offers advice on which rehabilitation measures should be used by the local authorities through the Info-House’ (p. 352). The role of ordinary social welfare professionals in this model is most often limited to information gathering and the implementation of efforts designed by others. The consequences or effects of the Aarhus model are not yet well understood. No systematic and independent research has displayed any causal effect of the model on radicalisation in Danish municipalities. However, as Lindekilde (2012b) stressed, isolating the effect of specific ‘preventive measures is close to impossible’ (p. 395). Instead, the effects discussed in literature are organisational in character, and often anecdotal, and not independent (Bertelsen, 2015; Hemmingsen, 373

Robin Andersson Malmros and Jennie Sivenbring

2015). In particular, studies have stressed how the legal framework and the institutional set-up help facilitate, assess, and manage free horizontal and vertical flows of information between the organisational actors involved (Sivenbring & Andersson Malmros, 2019; Tammikko, 2018). Johansen (2020) stated a critical perspective on the consequence of such extensive information sharing – that the Aarhus model is ‘blurring boundaries between security and other public services’ (p. 37), which assists in ‘the co-production of intensified territorial control and de-territorialized policing’ (p. 38).

The Channel programme The British Channel programme was established in 2011. Since 2015 (when it was placed under statutory footing), it has become one of the central components of the British Prevent strategy, which, in turn, is part of the UK counterterrorism strategy, CONTEST. The Channel programme is debated and has been occasionally labelled a ‘damaged brand’ (Elshimi, 2017; Thornton & Bouhana, 2017) and part of ‘a sinister state programme of stigmatising and repressing political dissidents’ (Richards, 2017, p. 190). Richards (2017) described the Channel programme as follows: Essentially, Channel operates very much in the ‘pre-criminal’ space and is about spotting and mitigating potential cases of concern before they enter the criminal realm. The nature of intervention will vary a great deal and will depend on the particular case in question: some cases may be issues of mental health; others to do with support in the educational environment; and others might be considered suitable for ‘theological intervention’ in some shape or form. (Richards, 2017, p. 189) The so-called Prevent Duty, which obliges employees such as teachers, social workers, nurses, and other public employees to assess, identify, and report suspected individuals, is one of the prerequisites of Channel. The programme’s organisational design is based on locally administered panels, usually chaired by a local authority representative (typically representing the county, district, or borough council), and should involve a local police practitioner. These obligatory partners are then supplemented with suitable actors from relevant agencies subject to the Prevent Duty (e.g. agencies offering youth offending services, health services, prisons, and social services). Channel panels meet regularly to discuss and evaluate individual cases and determine whether individuals are ready to leave the programme (which is voluntary). The official purpose of the Channel programme is to ‘safeguard children and adults from being drawn into committing terrorist-related activity’ (HM Government, 2012, p. 2) or, as further elaborated in 2020, to [provide] early support for anyone who is vulnerable to being drawn into any form of terrorism or supporting terrorism organisations. . . . Individuals can receive support before their vulnerability [is] exploited by those that want them to embrace terrorism. (HM Government, 2020, p. 7) Such an approach calls for instruments and techniques that can make those who are vulnerable visible (Martin, 2018). Consequently, the Channel programme centres on information gathering and sharing, which form the basis of interventions for those who are considered at risk.

374

Multi-agency approaches to countering radicalisation

Regarding important practices, Thornton and Bouhana (2017) explained that information on the operations of Channel is scarce. The authors’ interviews with practitioners involved in Channel showed that the programme’s operations vary significantly between local authorities. However, the process of intervention seems more standardised. The first step involves an assessment of an individual’s vulnerability to becoming involved in violent extremism (Gill & Marchment, 2022). The assessment criteria are standardised and are called the Vulnerability Assessment Framework (VAF). VAF consists of 22 factors, according to which individuals are given a score of 0–2. The second step is interventional. While the Channel Duty Guidance (HM Government, 2020) reiterates that Channel is about supporting vulnerable individuals, the guide provides limited information about what interventions or actions are suitable. Gill and Marchment (2022) pointed out that practitioners are requesting more information and guidance about which interventions to use. Following this institutional void, Vidino and Brandon (2012) mentioned that ‘Channel coordinators are largely free to experiment with different forms of intervention as they see fit’ (p. 20). This was affirmed by Richards (2017), who recognised that different regional and local authorities deal with differing crime priorities, leading to considerable variances. The power relations in the Channel programme clearly lean towards the police. While the chairpersons of Channel Panels are from local authorities, the Channel Duty Guidance puts the police in charge of directing the work and managing resources. The section of the Guidance dedicated to the Panel members states that MA involvement ‘is essential to ensure that the full range of information is accessible to the panel’ (Richards, p. 14). Social welfare professionals are repeatedly recognised as important sources of information. Given the prominent and extensive role of information gathering in the Channel programme, it is not surprising to find a significant body of literature describing the consequences or effects of the programme as highly problematic. Most significantly, the Channel programme has been accused of contributing to the stigmatisation and securitisation of Muslim youth and communities (Awan, 2014; Coppock & McGovern, 2014; Heath-Kelly, 2013; Kundnani, 2012). Another line of criticism points to the impact on professional work. More specifically, the statutory duty to detect and report clients, patients, students, and so forth puts important relationships at risk. As O’Donnell (2015) phrased, such practices could border on the policing of ‘thought-crime’ since practitioners are supposed to observe, assess, and report signs of ‘engagement with a group, cause or ideology’, ‘a need for identity, meaning and belonging’, or ‘a desire for political or moral change’ (HM Government, 2020, p. 51). Nonetheless, Vidino and Brandon (2012) described Channel as ‘the clearest success of the UK’s Prevent strategy. Channel is also cheap’ (p. 21). Thornton and Bouhana (2017) pointed out the absence of literature describing the programme’s mechanisms or measures of success and suggested best practices.

Key debates concerning multi-agency approaches: insights from the new institutional theory In this section, we will highlight how concepts developed in the new institutional theory, as applied in the tradition of organisational studies (Greenwood et al., 2017), are suited to illuminating some of the key debates concerning MA approaches in the field of CVE. These debates are centred on (1) the translation of ‘best’ practices between organisational contexts; (2) the social and cultural bases of information-gathering and sharing practices; (3) conflicts between the involved professionals; and (4) the appropriateness and features of MA approaches.

375

Robin Andersson Malmros and Jennie Sivenbring

Diffusion of ‘best’ practices between organisational contexts As touched upon in the previous section, some established models of MA collaboration have served as ‘inspirations’ or ‘role models’ for other countries’ development of their CVE approach. We argue that most CVE studies described this transfer of ideas using sweeping and unfocused terms (i.e. the practices, models, or approaches are ‘spread’) and did not pay enough attention to what happens to the model when it is implemented in new settings. In this chapter, we affirm our belief that the concept of translation (Czarniawska & Joerges, 1996; Sahlin-Andersson, 1996) can help direct interest towards and theorise the micro-level changes that occur when practices are moved from one local context to another or from one purpose to another (e.g. from promoting democratic values to CVE; Clarke, 1999). According to the translation perspective, organisational ideas, objects, and practices do not remain the same when moved in time and space – from one organisation to another – but become locally translated and edited in relation to pre-existing local conditions in the adopting organisation (Czarniawska & Joerges, 1996; Sahlin-Andersson, 1996) or even by actors in the same organisation (Waldorff, 2010). In this regard, the Aarhus model is an obvious example of the benefits of exploring MA approaches from a translation perspective. First, the model itself can be considered the result of multiple translation processes. Tammikko (2018) noted the following about the model: [It was] inspired in part by the British programme, and in part by a similar programme in the Netherlands. . . . This allowed the Danes to take into account some of the criticism that the British policies had faced, primarily regarding the problem with creating suspect communities. (p. 105) Hence, in the translation process, the original model was edited (Sahlin-Andersson, 1996) to make it more legitimate in the context where it was to be adopted. Johansen (2020) mentioned another aspect of translation: that the Aarhus model was transformed to incorporate the ‘new’ CVE purpose. The translation perspective, through the concept of editing, provides a strong theoretical approach to the study of how practices are discursively and materially reformed to serve CVE purposes. Another interesting aspect of translation is the focus on how things are translated (i.e. by what means and in what forums) and the conceptualisation of the actors involved (e.g. translators, editors, and broadcasters; Sahlin-Andersson, 1996). Translation studies showed that models do not become fashionable or interesting because of their economic or technical effectiveness but because of how well they resonate with institutional conditions in the adopting context and as a result of who advocates them as well as the material support given them. The description by Tammikko (2018) of how the Aarhus model became a global phenomenon serves as a good example: The Aarhus policies received a great deal of international attention as a PVE success story. Seven months later, in February 2015, the Mayor of Aarhus, Jacob Bundsgaard (Social Democratic Party), gave a presentation at a White House Summit, making the ‘Aarhus model’ world famous in the process. (p. 114) The Aarhus model is now spreading in time and space. Breidlid (2021) described how it is being referred to in Kenyan counties’ CVE policies, an institutional context that differs greatly from the Danish context. Crone and Nasser (2018) discussed the possibilities of the Aarhus model being 376

Multi-agency approaches to countering radicalisation

implemented in Lebanon and acknowledged a ‘need to be realistic and not attempt to apply Danish methodologies requiring a well-developed, decentralised welfare state along with [a] tradition [of] multi-agency coordination on crime prevention as their backbone’ (p. 30). A geographically closer example of the translation of the Aarhus model to another context is the attempt of the Swedish national CVE coordinator to establish Kunskapshus (Knowledge-houses) in four Swedish cities. While the concept was heavily inspired by the Danish use of Info-houses, evaluators (Frisk et al., 2017) found extensive ambiguity in the municipal Knowledge-houses’ management of individual referrals and cases, which is one of the key functions of the Aarhus model. The interviews with officials from the Knowledge-houses pointed to the fact that Swedish laws did not permit individual case management based on mere suspicion of ongoing radicalisation. Instead, the Knowledge-houses resorted to focusing on efforts to strengthen society, or primary prevention, rather than information sharing and intervention. Since such efforts already existed in the municipalities, it was unclear to some of the municipalities what the Knowledgehouses contributed with (Frisk et al., 2017). Again, this shows how local conditions (i.e. the legal frameworks) affect the translation process and, to some degree, the perceived success of approaches taken from other contexts. On the basis of these findings, we believe that Sweden’s incentive for translating the Info-houses in Denmark to its Knowledge-houses was its desire to associate its approach with the Danish approach to gain or regain legitimacy (see Andersson Malmros & Mattsson, 2017) after public and political criticisms of its earlier initiatives. All in all, we believe that translation can help to highlight: (1) the genealogy of fashionable MA approaches by studying how they travel and by what means; (2) how they are made fashionable by translators, broadcasters, and other actors with resources to spread practices and why these actors are involved; (3) how pre-existing practices (e.g. the SSP network) are edited to serve CVE purposes (see also Sivenbring, 2019, on editing in schools); and (4) how local conditions affect how a specific model is implemented. Consequently, neither researchers nor policymakers should be misled by semantic similarities or appealing labels and formulations but must pay closer attention to the micro-level implementation of MA approaches. Translation stipulates that the Aarhus model will mean one thing in Aarhus, another in Esbjerg, and yet another in Sweden or Lebanon. This is especially important if policymakers are expecting a certain output when translating a given practice, as the micro-level changes that occur when practices are moved from one place to another set the stage for disappointments. On the other hand, translation processes, especially in MA settings (Waldorff, 2013), can lead to innovation that can help spearhead development. New contexts can benefit from lessons learned from experiences with a model in another setting, which can be refined or ‘force’ existing practices to change in accordance with local conditions, which might assist in developing the model.

The cultural basis of information-gathering and -sharing practices A generally highly discussed area in the debate on CVE approaches is information-gathering and sharing practices. Such practices are often organised as part of formal MA approaches, and are instrumental features of both the Aarhus model and the Channel programme. However, the types of information gathered and shared have been heavily criticised in the academic community for securitising social and public spaces, communities, and individuals (e.g. Elshimi, 2017; HeathKelly, 2013; Johansen, 2020; Ragazzi, 2017; Kundnani, 2012; Lindekilde, 2012a). A central issue here has been what and why some attitudes or behaviours are forwarded to and assessed by MA structures. Research has indicated that the interpretation of radicalisation at the micro level seems 377

Robin Andersson Malmros and Jennie Sivenbring

to be a matter of subjective individual, group, or organisational perspectives and ideas (Andersson Malmros, 2019; Lindekilde, 2014; van de Weert & Eijkman, 2019, 2020). Lindekilde (2014) noted that: The diversity of professional groups involved in assessing individual cases may also cause problems, because what counts as ‘radicalisation’ or ‘signs of radicalisation’ on the one hand and protective ‘resilience factors’ on the other hand – who qualifies as ‘vulnerable individuals’ – varies considerably across professional groups. (pp. 230–231) Consequently, we can expect identities, norms, bodies of knowledge, values, assumptions, and intuition to influence professionals’ reporting of radicalisation. For example, teachers and the police will look upon troublesome behaviour from different perspectives, with the teacher focusing on the individual’s well-being and the police primarily assessing if the behaviour is breaking any laws (Lindekilde, 2014). In turn, this indicates that we should look ‘above’ the individual level and beyond the national context to understand professionals’ reporting. This argument is strengthened by the comparative study of Parker et al. (2021) of teacher reporting in Denmark and the United Kingdom, which found ‘a fundamentally shared understanding of concerning signs of student radicalisation’ (p. 649) that challenges the assumption of reporting as ‘entirely subjective or completely context’ (p. 634). Hence, an important objective of research is to understand what these professional interpretations are based on. We believe that the concept of institutional logic can inform and explain differences in interpretations of what information should be shared and why. Institutional logics are supra-organisational ‘socially constructed, historical patterns of cultural symbols and material practices, including assumptions, values, and beliefs’ (Thornton et al., 2012, p. 2) that provide meaning and guidance for individuals and groups of actors on how to act in organisational contexts. More specifically, institutional logics help individuals, groups, or organisations to understand the type of collective they belong to (e.g. their professional identity); what their collective is striving to achieve; which problems or situations are important and how they are to be interpreted; which strategies and tools are appropriate to apply in relation to the problem or situation; and how decision-making is performed (e.g. through a hierarchical chain of command or a consensus-oriented approach; Sivenbring & Andersson Malmros, 2019). A better understanding of how institutional logics guide professionals’ thinking and action can also help to avoid ‘disagreement, miscommunication and the risk of both “false negatives” and “false positives” ’ (Lindekilde, 2014, p. 237), which otherwise carries an obvious risk when professional groups guided by diverse logics are brought together in MA approaches.

Paradoxes and challenges in strategic and professional CVE work Building on the same theoretical framework as that in the previous section, we believe that institutional logics, together with the concept of hybridity (Pache & Santos, 2013; Pache & Thornton, 2020), can help illuminate some of the paradoxes and challenges of MA approaches. Hybridity can be defined as ‘the mixing of core organizational elements that would not conventionally belong together’ (Battilana et al., 2017, p. 138), such as different institutional logics (Pache & Thornton, 2020). Sivenbring and Andersson Malmros (2019) used this framework to define MA approaches as hybrid organisations that consist of two logics: a societal security logic and a social care logic. The balance or struggle between what has been described as more repressive approaches, which 378

Multi-agency approaches to countering radicalisation

emphasise restrictive and punitive measures, and welfare approaches (i.e. the so-called soft measures that mimic programmes and methods designed to handle other forms of youth delinquency) is a subject of debate in studies of CVE (Mattsson & Säljö, 2018; Sivenbring & Andersson Malmros, 2019). We believe that an institutional logic approach to the hybridity inherent in MA approaches can help explain why conflicts and tensions can not only easily arise but also provide an analytical tool for discerning how these logics mix in everyday work. Concerning the latter, Sivenbring and Andersson Malmros (2019) mentioned Norwegian and Danish police units that have goals and use techniques that are typically associated with a social care logic. Mixing logics in MA approaches to CVE can, however, also create paradoxes that lead to unintended or problematic consequences. Schools and other educational institutions are, on the one hand, to (1) continue with business as usual, for example, teaching, developing democracy, and safeguarding their pupils, and on the other hand, to (2) make risk assessments and identify individuals at risk of radicalisation in their own classrooms. If social welfare professionals engage in such controlling activities, they risk losing the trust of the communities in which they work (van de Weert & Eijkman, 2019). While the societal security and social care logics overlap in some respects, they also reflect considerable differences. We believe that this helps to explain why different professional actors have diverse perspectives about the purpose of CVE, what behaviour is considered problematic, how violent extremism is to be countered, and how and by whom decisions are to be made. Hence, when these logics collide and compete rather than mix and co-exist, it is likely that tensions and conflict will follow. Besides the organisational consequences, it is possible that a situation where one logic becomes hegemonic over the other can lead to misconceptualisations of radicalisation cases and ineffective and counterproductive measures (see Lindekilde, 2014). On the other hand, it is also possible that MA settings that successfully balance these logics can mobilise a diverse toolbox and different bodies of knowledge that can be applied in both general strategies and individual cases (Stephens & Sieckelinck, 2019). Consequently, the institutional logics perspective provides fertile theoretical ground not only for use in the evaluation of MA approaches but also for other researchers interested in exploring MA settings.

The appropriateness and constituents of MA approaches Finally, and on the basis of the literature touched upon in this chapter, we believe that there is a reason to question the MA paradigm within CVE policies and that concepts from the new institutional theory could aid in that quest. Organisations strive to be considered socially legitimate in their institutional environment, which, in turn, enhances their chances of survival (Meyer & Rowan, 1977). As part of this endeavour, organisations change their forms and behaviours in relation to rationalised myths – that is, the forms, symbols, and practices perceived as rational and legitimate by their environment (Meyer & Rowan, 1977). MA approaches and collaboration seem to have evolved into such a myth: a highly legitimate, taken-for-granted way to organise CVE locally. Considering the unclear evidence of the success of the MA approach, this status is, in many respects, undeserved. For the time being, there is almost a ‘the more the merrier’ attitude towards MA approaches, where collaborative structures are expanded to involve more and more actors (Andersson Malmros & Mattsson, 2017; Sivenbring & Andersson Malmros, 2019). On that note, Mazerolle et al. (2021) argued, based on their systematic literature review, that: Some evidence from this review suggests that when the capabilities of the various agencies are well aligned, then there might be a greater chance [of] better working relationships 379

Robin Andersson Malmros and Jennie Sivenbring

between partners. . . . While the degree of alignment of capabilities may limit the breadth [of] multiagency interventions, it needs to be balanced against the need to involve an optimal (not too large [and] not too small) number of partners in a collaborative programme. (p. 78) As mentioned earlier in this chapter, researchers have mainly questioned the MA approach because of its consequences (e.g. securitisation of target groups and social welfare operations). By departing from the assumption of MA approaches as a rationalised myth, we unlock the potential to critically engage in discussing both the general appropriateness of MA approaches as an organisational structure for dealing with CVE and the constituents of such structures. The risk of MA approaches becoming ‘locked in’ as an organisational solution is that it risks delimiting our availability to develop and choose alternative approaches that might be more effective in relation to the adopting context. Also, and in line with the suggestion of Mazerolle et al. (2021), there is a need to discuss which actors are suitable to be involved as well as which practices are utilised in MA structures. An expanding set of actors also means that the toolbox is expanded to a degree at which there is an increase in the risk of means – ends decoupling (Bromley & Powell, 2012) – that is, the adoption of practices where the outcomes or ends of the practices have a weak relationship to the organisational rationale or goals behind adopting the means (i.e. CVE).

Conclusion MA approaches today constitute an institutionalised way of organising CVE efforts in local public administrations. In this chapter, we highlighted this phenomenon through an extensive literature review that explored the foundations, main purposes, and popular applications of MA approaches in the context of CVE. Although research had successfully highlighted the consequences and effects, or lack of effects, of such collaborative arrangements, we still know little about the social and cultural processes related to this type of organising. Therefore, in this chapter, we introduced and discussed how a new institutional theory perspective can help analyse how MA approaches change as they travel between contexts, how institutional logics affect professional work and collaboration in MA structures, and the potentially negative effects of MA approaches becoming ‘locked in’ as a rationalised myth of how to organise CVE efforts.

References Ambrozik, E. (2018). Countering violent extremism locally [Unpublished doctoral dissertation, Cornell University]. Andersson Malmros, R. (2019). From idea to policy: Scandinavian municipalities translating radicalization. Journal for Deradicalization, 18, 38–73. Andersson Malmros, R. (2021). Translating ideas into actions: Analyzing local strategic work to counter violent extremism. Democracy and Security, 1–28. Andersson Malmros, R., & Mattsson, C. (2017). Från ord till handlingsplan: En rapport om kommunala handlingsplaner mot våldsbejakande extremism [From words to action plans: A report on municipal action plans against violent extremism]. Sveriges Kommuner och Landsting. Ansell, C., & Gash, A. (2008). Collaborative governance in theory and practice. Journal of Public Administration Research and Theory, 18(4), 543–571. Awan, I. (2014, September). Operation “trojan horse”: Islamophobia or extremism? Political Insight, 38–39. Battilana, J., Besharov, M., & Mitzinneck, B. (2017). On hybrids and hybrid organizing: A review and roadmap for future research. In R. Greenwood, C. Oliver, T. B. Lawrence, & R. E. Meyer (Eds.), The Sage handbook of organizational institutionalism (pp. 133–169). Sage.

380

Multi-agency approaches to countering radicalisation Bertelsen, P. (2015). Danish preventive measures and de-radicalization strategies: The Aarhus model. Panorama: Insights into Asian and European Affairs, 1(241), 53. Bjørgo, T. (2016). Counter-terrorism as crime prevention: A holistic approach. Behavioral Sciences of Terrorism and Political Aggression, 8(1), 25–44. Breidlid, T. (2021). Countering or contributing to radicalisation and violent extremism in Kenya? A critical case study. Critical Studies on Terrorism, 14(2), 225–246. Bromley, P., & Powell, W. W. (2012). From smoke and mirrors to walking the talk: Decoupling in the contemporary world. Academy of Management Annals, 6(1), 483–530. Cherney, A. (2016). Designing and implementing programmes to tackle radicalization and violent extremism: Lessons from criminology. Dynamics of Asymmetric Conflict, 9(1–3), 82–94. Cherney, A., & Belton, E. (2021). Evaluating case-managed approaches to counter radicalization and violent extremism: An example of the proactive integrated support model (PRISM) intervention. Studies in Conflict & Terrorism, 44(8), 625–645. Cherney, A., Sweid, R., Grossman, M., Derbas, A., Dunn, K., Jones, C., Hartley, J., & Barton, G. (2018). Local service provision to counter violent extremism: Perspectives, capabilities and challenges arising from an Australian service mapping project. Behavioral Sciences of Terrorism and Political Aggression, 10(3), 187–206. Christensen, T., Lægreid, P., & Røvik, K. A. (2020). Organization theory and the public sector: Instrument, culture and myth. Routledge. Clarke, L. (1999). Mission improbable: Using fantasy documents to tame disaster. University of Chicago Press. Coppock, V., & McGovern, M. (2014). “Dangerous minds”? Counter-terrorism discourse, radicalisation and the “psychological vulnerability” of Muslim children and young people in Britain. Children & Society, 28, 242–256. Crawford, A. (1994). Appeals to community and crime prevention. Crime, Law and Social Change, 22(2), 97–126. Crone, M., & Nasser, K. (2018). Preventing violent extremism in Lebanon: Experience from a Danish-Lebanese partnership (IEMed Papers 38, EuroMeSCo Series). European Institute of the Mediterranean. Czarniawska, B., & Joerges, B. (1996). Translating organizational change. De Gruyter. Dahlgaard-Nielsen, A. (2016). Countering violent extremism with governance networks. Perspectives on Terrorism, 10(6), 135–139. DiMaggio, P. J., & Powell, W. W. (1983). The iron cage revisited: Institutional isomorphism and collective rationality in organizational fields. American Sociological Review, 147–160. Ellis, B. H., Miller, A. B., Schouten, R., Agalab, N. Y., & Abdi, S. M. (2020). The challenge and promise of a multidisciplinary team response to the problem of violent radicalization. Terrorism and Political Violence, 1–18. El-Said, H. (2015). New approaches to countering terrorism: Designing and evaluating counter radicalization and de-radicalization programs. Springer. Elshimi, M. S. (2017). De-radicalisation in the UK prevent strategy: Security, identity, and religion. Routledge. European Commission. (2019). Reports of the project-based collaborations on prevention of radicalisation. https://ec.europa.eu/home-affairs/system/files/2020-11/2019_local.pdf Frisk, L., Axelson, T., Hylén T., & Stier, J. (2017). Att lägga rälsen medan man kör: En utvärdering av kunskapshusens arbete i fyra pilotkommuner – Örebro, Göteborg, Borlänge och Stockholm [Laying the rails while driving: An evaluation of the work of the knowledge centers in four pilot municipalities – Örebro, Gothenburg, Borlänge and Stockholm] (Rapport IKUD nr 3). Nationella Samordnaren mot Våldsbejakande Extremism/Högskolan Dalarna. Gielen, A. J. (2019). Countering violent extremism: A realist review for assessing what works, for whom, in what circumstances, and how? Terrorism and Political Violence, 31(6), 1149–1167. Gill, P., & Marchment, Z. (2022). Evaluating the channel programme’s vulnerability assessment framework. CREST Security Review. Greenwood, R., Oliver, C., Lawrence, T. B., & Meyer, R. (2017). Introduction: Into the fourth decade. In R. Greenwood, C. Oliver, K. Sahlin, & R. Suddaby (Eds.), The Sage handbook of organizational institutionalism (pp. 1–24). Sage. Hardy, K. (2022). A crime prevention framework for CVE. Terrorism and Political Violence, 34(3), 633–659. Harris-Hogan, S., Barrelle, K., & Zammit, A. (2016). What is countering violent extremism? Exploring CVE policy and practice in Australia. Behavioral Sciences of Terrorism and Political Aggression, 8(1), 6–24. Head, B. W., & Alford, J. (2015). Wicked problems: Implications for public policy and management. Administration & Society, 47(6), 711–739.

381

Robin Andersson Malmros and Jennie Sivenbring Heath-Kelly, C. (2013). Counter-terrorism and the counterfactual: Producing the “radicalisation” discourse and the UK PREVENT strategy. The British Journal of Politics and International Relations,  15(3), 394–415. Hemmingsen, A. S. (2015). An introduction to the Danish approach to countering and preventing extremism and radicalization (DIIS Report No. 2015: 15). DIIS. HM Government. (2012). Channel: Vulnerability assessment framework. HM Government. HM Government. (2020). Channel duty guidance: Protecting vulnerable people from being drawn into terrorism. Statutory guidance for channel panel members and partners of local panels. https://assets. publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/964567/6.6271_ HO_HMG_Channel_Duty_Guidance_v14_Web.pdf Jämte, J., & Ellefsen, R. (2020). Countering extremism (s): Differences in local prevention of left-wing, rightwing and Islamist extremism. Journal for Deradicalization, 24, 191–231. Johansen, M. L. (2020). Epidemic policing: The spreading of counter-radicalization in Denmark. Environment and Planning D: Society and Space, 38(3), 472–489. Kundnani, A. (2012). Radicalisation: The journey of a concept. Race & Class, 54(2), 3–25. Lawrence, T. B., Hardy, C., & Phillips, N. (2002). Institutional effects of interorganizational collaboration: The emergence of proto-institutions. Academy of Management Journal, 45(1), 281–290. Lindekilde, L. (2012a). Neo-liberal governing of “radicals”: Danish radicalization prevention policies and potential iatrogenic effects. International Journal of Conflict and Violence (IJCV), 6(1), 109–125. Lindekilde, L. (2012b). Introduction: Assessing the effectiveness of counter-radicalisation policies in northwestern Europe. Critical Studies on Terrorism, 5(3), 335–344. Lindekilde, L. (2014). Refocusing Danish counter-radicalisation efforts: An analysis of the (problematic) logic and practice of individual de-radicalisation interventions. In Counter-radicalisation (pp. 223–241). Routledge. Martin, T. (2018). Identifying potential terrorists: Visuality, security and the channel project. Security Dialogue, 49(4), 254–271. Mattsson, C., & Säljö, R. (2018). Violent extremism, national security and prevention. Institutional discourses and their implications for schooling. British Journal of Educational Studies, 66(1), 109–125. Mazerolle, L., Cherney, A., Eggins, E., Hine, L., & Higginson, A. (2021). Multiagency programs with police as a partner for reducing radicalisation to violence. Campbell Systematic Reviews, 17(2), e1162. McCulloch, J., & Pickering, S. (2009). Pre-crime and counter-terrorism: Imagining future crime in the “war on terror”. British Journal of Criminology, 49(5), 628–645. Meyer, J. W., & Rowan, B. (1977). Institutionalized organizations: Formal structure as myth and ceremony. American Journal of Sociology, 83(2), 340–363. O’Donnell, A. (2015). Securitisation, counterterrorism and the silencing of dissent: The educational implications of prevent. British Journal of Educational Studies, 64(1), 53–67. O’Halloran, P. J. (2021). Assessing the integration of cross-sectoral policy issues: A case study of Canada’s approach to countering radicalization to violence. Policy Sciences, 54(1), 183–208. Pache, A. C., & Santos, F. (2013). Inside the hybrid organization: Selective coupling as a response to competing institutional logics. Academy of Management Journal, 56(4), 972–1001. Pache, A. C., & Thornton, P. H. (2020). Hybridity and institutional logics. In M. L. Besharov & B. C. Mitzinneck (Eds.), Organizational hybridity: Perspectives, processes, promises (pp. 29–52). Emerald Publishing Limited. Parker, D., Lindekilde, L., & Gøtzsche‐Astrup, O. (2021). Recognising and responding to radicalisation at the “frontline”: Assessing the capability of school teachers to recognise and respond to radicalisation. British Educational Research Journal, 47(3), 634–653. Pistone, I., Eriksson, E., Beckman, U., Mattson, C., & Sager, M. (2019). A scoping review of interventions for preventing and countering violent extremism: Current status and implications for future research. Journal for Deradicalization, 19, 1–84. Ragazzi, F. (2017). Countering terrorism and radicalisation: Securitising social policy? Critical Social Policy, 37(2), 163–179. Richards, J. (2017). State policy and strategy: Prevent, “multi-agency” responses, and the way forward. In Extremism, radicalization and security (pp. 173–201). Palgrave Macmillan. Rittel, H. W., & Webber, M. M. (1973). Dilemmas in a general theory of planning. Policy Sciences, 4(2), 155–169.

382

Multi-agency approaches to countering radicalisation Sahlin-Andersson, K. (1996). Imitating by editing success: The construction of organizational fields. In B. Czarniawska & B. Joerges (Eds.), Translating organizational change (pp. 69–92). De Gruyter. Sestoft, D., Hansen, S. M., & Christensen, A. B. (2017). The police, social services, and psychiatry (PSP) cooperation as a platform for dealing with concerns of radicalization. International Review of Psychiatry, 29(4), 350–354. Sestoft, D., Rasmussen, M. F., Vitus, K., & Kongsrud, L. (2014). The police, social services and psychiatry cooperation in Denmark – a new model of working practice between governmental sectors: A description of the concept, process, practice and experience. International Journal of Law and Psychiatry,  37(4), 370–375. Sivenbring, J. (2019). Democratic dilemmas in education against violent extremism. In J. Lunneblad (Ed.), Policing schools: School violence and the juridification of youth (pp. 173–186). Springer. Sivenbring, J., & Andersson Malmros, R. (2019). Mixing logics: Multiagency approaches for countering violent extremism. The Segerstedt Institute, University of Gothenburg. Sivenbring, J., & Andersson Malmros, R. (2021). Collaboration in hybrid spaces: The case of Nordic efforts to counter violent extremism. Journal for Deradicalization, 29, 54–91. Stephens, W., & Sieckelinck, S. (2019). Working across boundaries in preventing violent extremism: Towards a typology for collaborative arrangements in PVE policy. Journal for Deradicalization, 20, 272–313. Tammikko, T. (2018). The political challenges of community-level PVE practices: The Danish case of Copenhagen vs. Aarhus on dialoguing with extremist milieus. Journal for Deradicalization, 16, 103–124. Thornton, A., & Bouhana, N. (2017). Preventing radicalization in the UK: Expanding the knowledge-base on the channel programme. Policing, 13(11), 331–344. Thornton, P. H., Ocasio, W., & Lounsbury, M. (2012). The institutional logics perspective: A new approach to culture, structure, and process. Oxford University Press. van de Weert, A., & Eijkman, Q. A. (2019). Subjectivity in detection of radicalisation and violent extremism: A youth worker’s perspective. Behavioral Sciences of Terrorism and Political Aggression, 11(3), 191–214. van de Weert, A., & Eijkman, Q. A. (2020). Early detection of extremism? The local security professional on assessment of potential threats posed by youth. Crime, Law and Social Change, 1–17. Vidino, L., & Brandon, J. (2012). Countering radicalization in Europe. International Centre for the Study of Radicalisation and Political Violence (ICSR). Waldorff, S. (2010). Emerging organizations: In between local translation, institutional logics and discourse [Dissertation, Copenhagen Business School (CBS)]. Waldorff, S. (2013). Accounting for organizational innovations: Mobilizing institutional logics in translation. Scandinavian Journal of Management, 29(3), 219–234. Weber, E. P., & Khademian, A. M. (2008). Wicked problems, knowledge challenges, and collaborative capacity builders in network settings. Public Administration Review, 68(2), 334–349.

383

25 NATURALISATION THROUGH MAINSTREAMING Counter-terrorism and counter-radicalisation in UN and EU discourse Rita Augestad Knudsen Introduction With the 11 September 2001 attacks in the United States, counter-terrorism abruptly became an exceptional global priority.1 At the forefront of many states’ and international organisations’ policy agendas during the early years following this watershed moment were ‘hard security’ measures such as defence-led military interventions against specific targets, increased border controls and other law enforcement action, and clampdowns on ‘safe havens’ from which terrorists could train and direct attacks. Most of these counter-terrorism measures were introduced with reference to the unique nature of the terrorist threat; across the world, official documents described terrorism as a special, fundamental danger, one with the potential to disturb the very foundations of societies. In consequence, fighting terrorism came to be portrayed as essential for reasons of security, ‘values’, and the survival of individuals, states, and the international system. Nations and international institutions soon produced dedicated strategies, budgets, systems, structures, and professionals to deal with this singular and particularly threatening challenge. As has been analysed in depth in other contributions to this volume, over the course of the 2000s, combating ‘radicalisation’ became a central part of counter-terrorism both conceptually and in practice. Defined in various ways and subjected to critiques that will not be re-rehearsed here, the most widespread understanding of radicalisation is as a process in which an individual comes to accept or becomes ready to use violence to achieve ideological, religious, and/or political ends.2 Especially following the mid-2000s ‘homegrown’ attacks in Europe, states and institutions within Europe became active in putting radicalisation on national and international agendas as an assumed necessary precursor of terrorist attacks. While radicalisation was seen as something that could happen fast or slowly, it was generally conceptualised as a step-by-step process (Moghaddam, 2005; McCauley & Moskalenko, 2008; Borum, 2003, 2011) preceding terrorist attacks. Rooting it out would stop terrorism from occurring. As a consequence, counter-radicalisation became an increasingly central part of counter-terrorism. This chapter is interested in the increasing centrality of counter-radicalisation within counterterrorism and even more so in the inclusion within counter-terrorism and counter-radicalisation discourse of ever more sectors, structures, and systems, including ones outside the conventional reach of ‘hard security’. It argues that the best framing device for these developments of growth DOI: 10.4324/9781003035848-29

384

Naturalisation through mainstreaming

and interconnectedness is mainstreaming; mainstreaming is what has taken place when counterterrorism has expanded through counter-radicalisation. This chapter starts by explaining the concept of mainstreaming as a process of increasing incorporation within a field or policy area of other sectors and priorities and the growing presence of this field in other sectors – revealing the concept’s usefulness for the developments in focus here. The chapter then provides discursive examples of how the mainstreaming of counter-terrorism through the foregrounding of counterradicalisation has unfolded in the official counter-terrorism discourse of the UN and the EU. Inspired by parts of critical discourse analysis, this chapter views language and discourse as social practices reflecting their contexts’ dominant concepts and embodying the settings in which they are produced (e.g. Janks, 1997; Fairclough, 2010). However, it does not share critical discourse analysis’ traditional concentration on the possible structures of power behind discourses – neither its occupation with their underlying intentionalities nor agency (Sayer, 2003). Instead, the chapter ‘simply’ proceeds on the assumption that language and the concepts expressed through it live at the social level rather than ‘one-dimensionally’ on the page or in the sounds or signs of human communication. When analysing the mainstreaming of counter-terrorism through counterradicalisation in the discourse of official strategies, this chapter considers language as an essential and essentially revealing signifier of the political and social reality in which it is produced. This version of discourse analysis considers the language of counter-terrorism and counter-radicalisation as signifying contextually determined multifaceted policy and social developments, as something both emerging from and continuing to shape the non-linguistic domains it is formally ‘about’. Relying upon such a method of qualitative and example-based discourse analysis, this chapter, however, lingers at the level of text rather than subjecting these deeper societal structures themselves to scrutiny. More concretely, this chapter tracks counter-terrorism and counter-radicalisation’s mainstreaming through their ever-increasing discursive incorporation of other sectors, systems, and practices. This does not entail a quantitative or sector-by-sector-based survey of counter-radicalisation’s policy mainstreaming or of the mechanics of its application within the different societal fields it has come to encompass. Rather, this chapter casts mainstreaming as conceptually relevant to understanding the process of extension and integration of counter-terrorism through counterradicalisation – from an ‘exclusive’ and exceptionally prioritised operational field of hard security to a broadly encompassing idea – and provides examples of how this has unfolded in discourse. The chapter finds that the apparent irreversibility and ‘self-evidentness’ of these developments suggest a discursive ‘naturalisation’ of counter-terrorism and counter-radicalisation over the past 20 years – although sometimes unevenly paced and not occurring at the cost of simultaneously undertaking hard security measures. Even today, the increasing merger of counter-terrorism with counter-radicalisation, and the latter’s claim to an ever-wider range of societal fields continue on their naturalised journey throughout the world.

Mainstreaming and radicalisation The concept of mainstreaming is central to understanding the evolution of counter-terrorism and counter-radicalisation over the past 20 years. Mainstreaming is already a well-established term within international affairs and international relations (IRs). However, since the mid-to-late 1990s, it has been best known in the context of ‘gender mainstreaming’. In simplified terms, this is taken to mean a process in which considerations around gender and gender equality have been taken more and more into account in other policy domains and initiatives nationally and internationally – and where action and strategic documents on gender have come to encapsulate ever more other 385

Rita Augestad Knudsen

sectors too. Although much criticised (and today also strikingly binary), the following definition of gender mainstreaming from 1997 still sums up the customary understanding: The process of assessing the implications for women and men of any planned action, including legislation, policies or programmes, in all areas and at all levels. (Mainstreaming) is a strategy for making women’s as well as men’s concerns and experiences an integral dimension of the design, implementation, monitoring and evaluation of policies and programmes in all political, economic and societal spheres so that women and men benefit equally and inequality is not perpetuated. The ultimate goal is to achieve gender equality. (UN, 1997a) In 2001, the landmark document on gender mainstreaming, UN Security Council Resolution 1325, affirmed this understanding. Praised by some activists and condemned by others, Resolution 1325 made explicit reference to ‘mainstreaming’ and called for the incorporation of gender equality perspectives and considerations by the UN and its member states when implementing peacekeeping and other activities around the world. To be sure, within the broad field of IR and related scholarship, gender mainstreaming is the topic subjected to most research and critique among all issues possibly ‘mainstreamed’ over the past decades (e.g. Rees, 2005; Daly, 2005; Squires, 2005; Rai, 2003; True & Mintrom, 2002). The centrality of gender within the discourse and scholarship of mainstreaming is also evident from the fact that the explicit use of the term ‘mainstreaming’ within the field of counter-terrorism and counter-radicalisation is also primarily reserved for the mainstreaming of gender within counterterrorism or ‘Preventing/Countering Violent Extremism’ (P/CVE) (Rothermel, 2020). With regard to international development policies for instance, Western donors have called upon ‘gender’ to be mainstreamed into different aspects of CVE work (Henshaw, 2021), and only sometimes required developing countries to ‘mainstream’-specific CVE programmes into their overall security structures (Akwanda, 2020). The possible mainstreaming of counter-terrorism or counter-radicalisation as fields of either discourse or policy has been less of a focus thus far. Although more widely cited since the late 1990s, ‘mainstreaming’ is often both connected to gender and applied to the developing world – perhaps due to the centrality of gender mainstreaming in UN peacekeeping to the conceptual history of ‘mainstreaming’. All the same, the term is increasingly being applied to other issues and processes, including ones within the West. Also there, the term is taken to signify a process in which an issue both gradually adopts itself within it and becomes ever more engrained in other policies, discourses, and structures, reshaping their meanings and agendas. Recent issues discussed with the language of mainstreaming include action against climate change, the integration of refugees and immigrants, nutrition, disaster relief, intersectionality, green production, and many others. Using ‘mainstreaming’ to capture the evolution of a policy phenomenon or idea commonsensically implies a process where the issue at hand starts off as not inherent to the ‘mainstream’ or core of an authority’s priorities. With time, however, and through the mainstreaming process, the issue becomes both more central to and increasingly integrated with existing structures. Within official discourse, as is the focus here, mainstreaming can happen both through the citation of the topic in question outside its own field through its adoption within other agendas and, conversely, through the reference of other fields within the official discourse of the issue that is being mainstreamed. Later, this chapter will turn to the second of these dynamics with examples of the increasing incorporation of a range of sectors within official counter-terrorism and counter-radicalisation discourse, using ‘mainstreaming’ to make sense of both the fusing of counter-terrorism with 386

Naturalisation through mainstreaming

counter-radicalisation, and of how these priorities have seeped into and included other domains in UN and EU official discourse. Regardless of its burgeoning use in policy discourse and scholarship since the 1990s, the meaning and implications of the concept of mainstreaming are still loaded with complexities and controversies (e.g. Thiessen, 2019). For one, mainstreaming could signify a deep and genuine commitment to an issue area – and at the same time, its possible ‘watering down’ through making it generic and spreading its prioritisation too thinly. Mainstreaming could simultaneously mean that something previously neglected is finally being brought deservedly to light and that its integration into the ‘mainstream’ might come at the cost of devoted, special attention. Clearly, mainstreaming cannot meaningfully be discussed in isolation without regard for what exactly is being mainstreamed, how, and to the benefit of whom. As Martti Koskenniemi has put it, mainstreaming can be seen as: a project for seizing institutional power that is profoundly ambiguous in its effects. There is no reason to be either “for” or “against” mainstreaming without a clear sense of what priorities it is intended or likely to support and what foreseeable effect it might have on the allocation of resources among human groups. (Koskenniemi, 2010) In this light, it becomes obvious that applying the concept of mainstreaming to the fields of counter-terrorism and counter-radicalisation means something very different from using the term with regard to ‘gender’. Gender mainstreaming has the explicitly emancipatory and rights-based aim of gender equality and is actively seeking the empowerment of previously marginalised groups. Meanwhile, counter-terrorism and counter-radicalisation/P/CVE are their core security-oriented policy concepts and measures. Although often involving ‘softer’ societal sectors and appealing to values beyond those typically encapsulated by hard security initiatives, preventing and countering violent extremism, as is what P/CVE stands for, is ultimately intended to aid states in protecting themselves and their citizens from violence – the perhaps most conventional security-oriented aim of all. While often cited as secondary or preliminary goals or as welcome by-products, the ultimate ends of P/CVE measures are not in and of themselves to even out inequalities, benefit any marginalised group, or liberate individuals or communities. Before returning to the potential implications of applying the concept of mainstreaming to counter-terrorism/counter-radicalisation/P/CVE, it is worth pausing to underline the importance of ideas of ‘radicalisation’ to these developments. Indeed, the discursive mainstreaming of counterterrorism was enabled by a post-9/11 historical context of growing national and international consensus on the importance of detecting and acting against radicalisation as a key to stopping terrorist violence. In a setting where states and institutions sought to explain how ‘ordinary’ citizens could be involved in terrorism, radicalisation emerged as a central part of the explanation. Radicalisation – understood as an increasing willingness to accept or use violence to achieve ideological, political, or religious aims – came to be seen as a necessary component in an assumed pre-terrorist attack chronology. In consequence, stopping radicalisation became central to stopping terrorist attacks. For many, if not most, international and national authorities, countering ‘radicalisation’ became a given part of counter-terrorism. It is worth noting in brief that ‘P/CVE’ is still more often used by the UN and with reference to the developing world, while ‘radicalisation’ remains widely cited, especially in Europe, as will be evident from the references below. The differences between the language of ‘radicalisation’ and that of ‘Preventing/Countering Violent Extremism (P/CVE)’ are not the focus of this chapter, but if one were to speculate on why this might be the case, it might 387

Rita Augestad Knudsen

be that the well-known and ongoing disagreements and controversies around the language of ‘radicalisation’ have registered and been taken on board in different ways at the UN and elsewhere. The growing centrality of radicalisation and P/CVE in international discourse and within models seeking to explain terrorism was essential to counter-terrorism’s mainstreaming, as this paved the way for its involvement in an array of areas outside the strictly hard security domain. Considered in ‘isolation’, terrorist violence could be seen first and foremost as a matter of straightforward hard security and as a phenomenon fought through hard security counter-terrorism measures, without a clear claim to sectors outside this scope. Counter-terrorism is strictly defined, hence it does not seem especially fit for ‘mainstreaming’. Radicalisation, meanwhile, is a more woolly concept with no obvious temporal or sectoral boundaries. Targeting this likely multifaceted process in which an individual takes on a certain mindset, possibly connected with a plethora of ‘hard’ and ‘soft’ under and overlying conditions including family circumstances, educational and psychological issues (perceived or real), discrimination, and more, would lay claim to a range of domains beyond the hard security realm. The expanding scope and involvement within other sectors of counter-terrorism and counterradicalisation have already been noted and criticised by scholars of Securitisation and Critical Terrorism Studies. Although rarely evoking ‘mainstreaming’ directly, such scholarship has sometimes insightfully problematised what it sees as an ever-growing involvement of counter-terrorism and counter-radicalisation in previously less- or non-securitised fields, such as education and health care. Researchers attentive to the potential for authorities’ overreach and abuse of powers have also warned that the increasing reliance on counter-radicalisation within counter-terrorism may encroach on people’s liberties, work unfairly, and violate human rights.3 As a leading country within European counter-terrorism and counter-radicalisation, the United Kingdom has often been studied in particular depth. Some analyses have centred on the counter-radicalisation ‘Prevent’ strand of the UK’s counter-terrorism strategy and its statutory emplacement of responsibilities on institutions outside the hard security realm. As a possible policy example of the in-practice mainstreaming of counter-terrorism through counter-radicalisation, the ‘Prevent Duty’ in 2015 put a legal duty on UK health, educational, and social care institutions to report concerns over the potential ‘radicalisation’ of individuals under their care.4 To be sure, although sometimes implemented with reference to individuals’ vulnerability and safeguarding, the end goal of counter-radicalisation as part of counter-terrorism – in the United Kingdom and elsewhere – remains one of societal and state security – keeping nations and citizens safe from violent terrorist attacks. Although neither securitisation nor the real-life policy consequences of mainstreaming processes are the focus here, it does seem undisputable that the infusion of counter-radicalisation agendas in other fields does represent a ‘securitisation’ of sorts of these sectors. Whether or not this is warranted, proportional, or a price worth paying for keeping societies, communities, and individuals safe are questions to be pursued elsewhere. Given the critical interest in the on-the-ground policy mainstreaming of counter-terrorism and counter-radicalisation, the practice and consequences of such policies have already been subjected to much analysis – although rarely under the label of ‘mainstreaming’. The scholarship on mainstreaming, however, has alerted us to some of the general risks associated with real-life mainstreaming processes. Since this research in the main concentrates on empirical questions related to the practicalities of mainstreaming rather than its discourse (as is the focus here), its occupations largely fall outside the scope of this chapter – but two dynamics are nonetheless worth noting as potentially relevant. One is the possible risk of the newly prioritised ‘mainstreamed’ field being overflowed by (often self-appointed) ‘experts’. Mainstreaming typically means more attention, and larger budgets set aside for the mainstreamed field, and a requirement to do something new; 388

Naturalisation through mainstreaming

policymakers might naturally be inclined to turn to those quickly putting themselves forward with solutions. A certain risk of ‘expert co-option’ hence seems bound up with the phenomenon of mainstreaming and to be expected with regard to the mainstreaming of counter-terrorism and counter-radicalisation as well (Paterson, 2010). A second possible dynamic is the danger of the mainstreaming agenda being co-opted to benefit established agents, agendas, and power structures that perhaps would be better suited candidates for being challenged (Thiessen, 2019, unpaginated; Barnett & Finnemore, 2004). Developments of this kind should be well known to those following earlier IR debates on peacebuilding, statebuilding, and humanitarian interventions (for a summary, see Augestad Knudsen, 2013) and would put calls for ‘local ownership’ of international agendas in a tricky light, especially where this would mean the de facto propping up of illegitimate local elites (see also Thiesen, 2019). Such domestic or international ‘elite co-option’ could result in the reification of problematic authorities and in the context of counter-terrorism and counter-radicalisation, a possible stigmatisation of marginalised individuals and groups or of those ‘simply’ in opposition. To be sure, no fixed template exists on how to implement counter-terrorism or, for that matter, on how ‘radicalisation’ or ‘extremism’ may materialise, be addressed, or play a role towards the execution of terrorist attacks. These uncertainties seem to leave much room for those in power to take liberties when executing mainstream agendas. When this is said, the next part of this chapter puts such aspects of policy to one side and instead concentrates on the comparatively less-studied international discourse of two central institutions with important and expanding roles within counter-terrorism and counter-radicalisation: the UN and the EU. As already indicated, the focus here is not so much on how counter-terrorism and counter-radicalisation have infused other sectors and possibly securitised them as on how other sectors have been increasingly incorporated within such UN and EU discourse. A securitisationrelated question arising from this that could be posed in brief might be whether the incorporation into such discourse of ‘softer’ fields such as education, health, and care could be seen as ‘desecuritising’ counter-terrorism and counter-radicalisation – or if this in fact conceals or masks processes of actually securitising and co-opting these softer domains. As indicated, processes of mainstreaming, discursively and in practice, often unfold without any explicit mentioning of the term ‘mainstreaming’. So when now proceeding with examples from the discursive trajectory of counter-terrorism’s and counter-radicalisation’s mainstreaming, the next section does not primarily look for explicit citations of ‘mainstreaming’. Instead, it pinpoints this process by identifying examples of how counter-terrorism and counter-radicalisation discourse have moved beyond their original ‘hard security’ constraints and into other policy and societal domains. Specifically, it identifies the ever-expanding number of other fields, sectors, practitioners, institutions, values, and priorities enlisted in the service of counter-terrorism and counter-radicalisation in key UN and EU documents.

Mainstreaming of counter-radicalisation in official UN and EU discourse Counter-terrorism has certainly been a concern for international organisations for decades,5 but the 9/11 attacks in the United States and the subsequent mid-2000s attacks in the Netherlands, Madrid, and London were significant in concentrating minds on formulating joint international efforts. Especially for Western decision makers, these attacks brought into sharp focus that terrorism was very much an international and not only nationally located threat and one that required joint inter-state action. These attacks also underlined that individuals born and raised in any state in the world could become motivated to take violent action against their homeland to achieve 389

Rita Augestad Knudsen

political, religious, and ideological ends. The emerging global near-consensus on these assumptions led states to seek renewed international cooperation on how to fight terrorism. With growing agreement on the role believed to be played by radicalisation in this process, international initiatives soon turned towards earlier (and ever earlier) stages of an assumed chronological line from radicalisation – and the development of a ‘terrorist mindset’ – to the execution of terrorist violence. This section of the chapter illustrates how, in this context, the mainstreaming of counter-terrorism and counter-radicalisation in international discourse was facilitated by this growing occupation with radicalisation, enabling it to encapsulate ever more traditionally non-security-oriented domains. It traces this process in some of the key UN and EU official documents on counter-terrorism, counterradicalisation, and P/CVE since the early- to mid-2000s, when these organisations’ attention really turned towards these issues. The UN and EU – two institutions positioned at different places in the global security environment – are certainly not the only organisations that have worked on counterterrorism and counter-radicalisation since the early 2000s. But both have been active and influential throughout the world in formulating joint efforts. Together, the examples from their official discourse provide telling insights into how the discursive mainstreaming of counter-terrorism through counter-radicalisation has unfolded internationally during the last two decades.

United Nations discourse Ever since at least the 1960s, the UN has been acting to seek to condemn or prohibit various methods that could possibly be used to carry out terrorist actions, such as airline hijacking (1963, 1970, 1971), the taking of hostages (1979), and bombings (1997b) (UN, 1963, 1970, 1971, 1979). In 1999, the UN for the first time passed a Security Council Resolution seeking to condemn terrorism as such rather than chiefly the various means by which it could be carried out (UN Security Council, 1999). The UN’s early action on the matter was, however, and still partly is, characterised by conceptual and terminological disagreement on how to define ‘terrorism’ and on whether, for instance, states can be defined as ‘terrorists’. Controversies around the term and possible process of ‘radicalisation’ have been evident too at the UN, and various UN members have taken different views on whether the organisation should focus on preventing (imminent) attacks or on terrorism’s ‘root causes’ (Ruperez, 2017). The day after the 9/11 attacks, the UN Security Council passed a Resolution condemning these attacks, and during the years that followed, UN counter-terrorism discourse accelerated in both intensity and scope (UN Security Council, 2001; UN, 1997b). The UN passed its first proper dedicated counter-terrorism strategy in 2006 as a General Assembly Resolution and a Plan of Action (UN General Assembly, 2006). Already then, the UN endorsed both the idea that counter-terrorism is a broad-based effort in line with the UN’s wider mandate and purpose and the importance of combating ‘conditions conducive’ to terrorism in order to prevent violent attacks from occurring (UN, 2006). Not relying on the language of ‘radicalisation’, the strategy underlined the importance of ‘Affirming Member States’ determination to continue to do all they can to resolve conflict, end foreign occupation, confront oppression, eradicate poverty, promote sustained economic growth, sustainable development, global prosperity, good governance, human rights for all and rule of law, improve intercultural understanding, and ensure respect for all religions, religious values, beliefs, or cultures, and indeed, to promote a ‘culture of peace’ – as part of fighting terrorism. These ends, as formulated in the 2006 strategy, however, were expressed at a general and declarative level in standard UN parlance and did not include specific measures or anything resembling a roadmap for their achievement. In terms of concrete measures, this UN counter-terrorism discourse concentrated on the setting up of specialised mechanisms to combat terrorism, including 390

Naturalisation through mainstreaming

an ‘international centre’ and an ‘international conference’. Over the years and especially from the mid-2010s onwards, the UN’s regular reviews of its terrorism strategy have both specified the measures it has called for against terrorism and expanded the scope of these specified measures from hard security-concentrated initiatives to more encompassing ones. For instance, in 2014, the fourth review of the strategy both put the spotlight on the hard security threat of FTFs and the links between terrorism and organised crime, and encouraged broad gender perspectives in counterterrorism, and underlined the importance of addressing ‘conditions conducive to the spread of terrorism’.6 In 2015, the UN’s discourse went even more expansive and specific with the Secretary-General’s dedicated report on PVE (UN General Assembly, 2015b). This report is especially interesting as it marks the shift in UN discourse away from primarily hard security measures and general statements of disapproval to a concentration on softer, wide-reaching, and specialised measures of counter-radicalisation (PVE in UN parlance). The report condemned violent extremism as ‘an affront to the purposes and principles of the’ UN as well as to ‘peace and security, human rights and sustainable development’. Furthermore, this PVE report claimed that fighting violent extremism was important not only to stop terrorist attacks from occurring. PVE, it said, was necessary also because violent extremism hinders efforts to meet the Millennium Development Goals (MDGs), and hampers educational attainment, gender equality, peacekeeping aims, human rights, rule of law, and fundamental freedoms. The strategy portrayed the threat of VE as extraordinary yet also enormously wide-reaching. Hence, this threat required both comprehensive and especially prioritised counter-action, including ‘systematic preventive measures which directly address the drivers of violent extremism’ and creating ‘open, equitable, inclusive and pluralist societies’. This discursively significant UN report on PVE further stated that there is ‘no authoritative statistical data on the pathways towards individual radicalisation’ – but that combating it nonetheless required addressing an array of detailed ‘push’ and ‘pull’ factors implicating more or less all of an individual’s personal, family, or citizen life. This would, according to the report, require the involvement of numerous specific agents, actors, and industries in all of the UN’s member states. This PVE report shows that by 2015, UN discourse had warmed to the language of ‘radicalisation’. Still, it is notable that the report, as characteristic of UN parlance overall, continued to rely primarily on the phrases ‘violent extremism’ and ‘conducive conditions’. Violent extremism, the report said, is problematic ‘as, and when, conducive to terrorism’(UN General Assembly, 2015b). Echoing this position, in December the same year, the UN Secretary-General called for a ‘whole of UN’, ‘all-of-society’, and ‘all-of-government’ approach to PVE within seven specified focus areas encompassing good governance, human rights, and the rule of law, engaging communities, empowering youth, gender equality, and empowering women, education, skill development, and employment facilitation (UN General Assembly, 2015a). This direction of UN discourse of first broadening the scope of counter-terrorism through foregrounding counter-radicalisation/C/PVE and second specifying the measures this broadening scope entailed was perpetuated in subsequent UN documents. In 2016, the fifth review of the UN’s counter-terrorism strategy significantly expanded the purview of the UN’s discourse and action on CVE/counter-radicalisation, with a total of 72 items. Fundamentally, targeting ‘conducive conditions’ was underlined as a cornerstone of UN involvement. In the words of the document, ‘since wars begin in the minds of men, it is in the minds of men that the defences of peace must be constructed’. This fifth review of the strategy detailed a plethora of ever more wide-ranging and concrete measures encompassing the formulation of counter-narratives, addressing child abuse with regard to extremism, educational opportunities, gender equality, Internet and prison recruitment, post-release rehabilitation, rule of law capacity building, the strengthening of border controls, and 391

Rita Augestad Knudsen

the inclusion of CVE considerations in the handling of applications for asylum. All member state governments and all of civil society all over the world should be involved. The final telling example of UN discourse to be mentioned here is the sixth review of the UN’s counter-terrorism strategy from 2018, which also cited the 2017 establishment of the UN Office of Counter-Terrorism. Significantly, this sixth review added to its list of specific terrorism-related security concerns possible attacks on critical infrastructure, illicit trade in natural resources, and the exploitation of online safe havens for the purposes of promoting VE. This sixth review also called for the proper follow-up of terrorism victims and outlined broad-ranging hard security measures to be undertaken by member states, such as intelligence gathering on travel movements, capacity building on databases, an increased role of Interpol, follow-up of social media, and the criminalisation of terrorist acts. Increased international cooperation and better publicprivate partnerships, including in the formulation of counter-narratives, were emphasised as key. Interestingly, the document also cemented in UN discourse the language of ‘vulnerability’ alongside ‘radicalisation’ as an area of concern and pertinence with regard to counter-terrorism and counter-radicalisation/P/CVE. Discursively targeting ‘vulnerability’ in a terrorism strategy certainly marked the UN’s expansion of counter-terrorism’s conceptual scope beyond that of hard security and possibly into domains of welfare provision and care, where references to ‘vulnerability’ are generally more common.

European Union discourse The official discourse of the EU on counter-terrorism has largely moved along the same arc as that of the UN over the post-9/11 period. As already indicated, the ‘homegrown’ attacks in the Netherlands, Madrid, and London brought urgency to calls for Europe-wide joint counter-terrorism action and enabled increased attention to ‘radicalisation’ as a key link in explaining how ‘ordinary’ citizens could conduct terrorist attacks. Shortly after the 7/7 attacks in London in 2005, the EU formulated a joint counter-terrorism strategy, and the same year, the European Council adopted a plan of action containing over 100 initiatives (Council of the EU, 2005). This action plan listed as EU priorities a number of ‘hard security’ issues, including information sharing, combating terrorist financing, better incorporation of counter-terrorism in the EU’s external relations, and improving systems for protection (Vries, 2005; Lugna, 2006; Monar, 2007, 2015; Wolff, 2008). At the same time, European authorities seemed increasingly convinced by the idea that radicalisation played a key role in leading to terrorist attacks. In 2011, the EU established the Radicalisation Awareness Network (RAN).7 The RAN was and still is a large international network of scholars and practitioners exchanging views and insights into the role of radicalisation and how to prevent violent extremism, especially in a European context. Perhaps reflecting the many controversies over the term ‘radicalisation’ and its vagueness in terms of both foundations and implications, the RAN today, seemingly in contrast to other EU discourse, relies primarily on the language of ‘violent extremism’ rather than ‘radicalisation’. Regardless of terminology, however, the establishment of RAN, its growth – now connecting more than 6,000 experts – and the EU’s continuing reliance on it underline the importance the EU places on the assumed role of radicalisation in leading to terrorism and violent extremism and in counter-radicalisation as a central tenet of counter-terrorism. Addressing the development of attitudes and mindsets, as the EU seems to assume, is key to stopping terrorist attacks. In line with such assumptions, ideas of taking early action against terrorism have taken hold within Europe as well as elsewhere in the world. This is reflected in EU counter-terrorism action, especially from the mid-2010s, which by then had placed counter-radicalisation at its core. In June 2014, the EU revised and propped up its 392

Naturalisation through mainstreaming

strategy for ‘combating Radicalisation and Recruitment’, and in 2015, the EU set up the RAN Centre of Excellence. In 2018, the European Commission established a separate Steering board on Radicalisation (European Commission, 2018). Over the last few years, the EU has been active in trying to formulate a shared vision for how its member states should combat radicalisation in practice on a growing number of fronts: in prisons, online, and in schools, among other areas. The guiding principle seems to have been that the earlier intervention, the better, in rooting out the potential for terrorist violence. Like with UN counter-terrorism discourse, the EU’s language on counter-terrorism through the foregrounding of counter-radicalisation has become ever more encompassing as well as specific (e.g. European Commission, 2014). European initiatives caught particular speed with the onset of the war in Syria, the declaration of the ‘Islamic State’, the associated contingents of foreign fighters from Europe to the region, and the attacks in Europe in 2015–6 (Council of the EU, 2015a, 2014). The EU’s renewed focus on radicalisation seems to have been particularly sharpened in the aftermath of the attacks in France and Belgium (Council of the EU, 2016). In 2015, the EU Internet Forum was set up, and in 2018, the Commission proposed that member states adopt binding legislation to remove terrorist content. In 2015, 2017, and 2019, the EU paid special attention to the online domain as well as prisons as possible arenas for radicalisation (Council of the EU, 2015b). At the same time, more consideration has been paid to so-called lone wolves and the assumed mental health issues involved with such individuals’ engagement with extremism (European Commission, n.d.b; Schuurman et al., 2019). Risk assessment within the field of counter-terrorism and counter-radicalisation is another field of much European engagement and interaction, with current discussions often centred around how such assessments should be structured and used and which target groups they should be applied to.8 Along with the discursive weight placed on broad-based counter-radicalisation action, the EU has also made significant hard security counter-terrorism efforts, working on initiatives such as better border controls, intelligence and information sharing, and specific systems such as the Passenger Name Record (PNR) (Council of the EU, 2016). Especially following the Charlie Hebdo attacks in Paris in 2015, as part of the Commission’s European Agenda on Security, the Union proposed to set up a European Counter Terrorism Centre (ECTC) within Europol to aid coordination and cooperation on the field (Europol, n.d.). This was followed by the EU’s 2017 directive on combating terrorism and, more recently, the EU Security Union strategy, adopted in July 2020, both outlining an array of specific hard security actions intended to stop terrorism (Council of the EU, 2017; European Commission, 2020b). As the EU has become more globally ambitious, it has also become more active in counterterrorism and counter-radicalisation outside its borders. In 2016, the Global Strategy for the EU’s Foreign and Security Policy stated that the EU considered peace and security essential to development and that this was why the EU chose to promote both development and P/CVE in tandem around the world (High Representative of the Union for Foreign Affairs and Security Policy, 2016). It defined its scope broadly: P/CVE can be understood to constitute all actions that strengthen the resilience of individuals and communities to the appeal of radicalisers and violent extremism, from interfaith dialogues to vocational training, mentorship programmes, training of state governance and security actors and community debates on sensitive topics. Interestingly, choosing primarily the language of ‘P/CVE’ rather than ‘radicalisation’ in this nonEuropean context, the EU called for working ‘to reduce the incidents of VE and support for such 393

Rita Augestad Knudsen

acts’. The mainstreaming of counter-terrorism through this emphasis on counter-radicalisation within the EU’s external relations and the foreign policy of its member states has involved an increasing number of sectors, including the allocation of development aid (Cross, 2017). Seen as one long development, the EU’s evolving discourse on counter-terrorism, like that of the UN, reveals a significant broadening of the scope of counter-terrorism enabled by an everheavier weight placed on counter-radicalisation. However, given the nature of the EU, it has had more space than the UN to initiate specific hard security measures for its member states, alongside its counter-radicalisation focus. All the same, early and comprehensive intervention against radicalisation remains a core principle in EU discourse. Last year, the EU’s new counter-terrorism agenda closely echoed UN discourse in citing a ‘whole-of-society’ approach to security, counterterrorism, counter-radicalisation, and P/CVE. It called for ‘a further streamlining of EU policies, initiatives and funds’ in combating radicalisation as pivotal to counter-terrorism and stressed that this included ‘promoting inclusive society, education and EU common values’ (European Commission, 2020a). Combating terrorism through counter-radicalisation, as this EU agenda expresses, means nothing less than an approach ‘that safeguards and upholds our pluralistic society, our common values and our European way of life’ – leaving no area of member state or partner state action unaffected. It is worth noting that the expansion of counter-terrorism and counter-radicalisation’s domain in EU discourse has not come at the expense of arguing for the ‘exceptionalism’ of the threat of terrorism. Counter-terrorism’s and counter-radicalisation’s scope has grown ever larger, exactly with reference to terrorism’s assumedly uniquely damaging potential. Meanwhile, counterterrorism discourse has increasingly involved the whole of societies intended to combat this threat, with legitimating reference to the all-encompassing and often generic factors that may lead to radicalisation.

Conclusion The idea that radicalisation is an essential precursor to terrorism and that counter-radicalisation is therefore crucial to counter-terrorism has enabled the expansion of these fields into a large range of other sectors of discourse as well as policy, law, and communities over the past two decades. With time, counter-terrorism and counter-radicalisation agendas have implicitly made ever stronger claims to infuse and adopt other domains, including ones initially not associated with ‘hard security’ agendas. Especially since the mid-2000s, what has been seen as the exceptional nature of the terrorist threat combined with the infinity of possible causes of radicalisation have enabled both the special prioritisation and apparently boundaryless expansion of counter-terrorism and counter-radicalisation. Through its increasing foregrounding in discourse and policy, counter-radicalisation has thus become a way for counter-terrorism to grow out of its hard security frame and become a vector for blowing open what could be relevant to counteract it. As narrowly understood, counter-terrorism belongs to the field of defence and law enforcement and is a matter of ‘hard security’: the aim is to stop violent attacks of at least some degree of imminency. Counter-radicalisation, meanwhile, is both a more undefined concept and one concentrated at an earlier but unspecified stage of an assumed pre-attack chronology: like the ‘radicalisation’ it intends to root out, counter-radicalisation lives at the plane of ideas, values, and attitudes without clearly delineated conceptual or practical boundaries. The perceived importance of fighting terrorism through counter-radicalisation allowed for the increasing enlistment of societal and policy sectors far outside the hard security domain, both nationally and internationally. 394

Naturalisation through mainstreaming

These developments could certainly be seen in both UN and EU official discourse over the past 20 years. When combating terrorism meant fighting ‘radicalisation’, there seemed no natural end point for where to target interventions. As shown in this chapter, this seems to have spurred international discourse, citing broad-based and encompassing programmes often including numerous sectors of society, such as health, education, immigration, integration, and forms of welfare provision. Now, 20 years after the 9/11 attacks, counter-terrorism and counter-radicalisation/P/ CVE have also, in practice, come to encompass security, legal, welfare, care, health, equality, and educational agendas, among others. And as UN and EU discourse shows, both traditional and newly created institutions, agencies, coalitions, and specific individuals are called on to join the fight against terrorism, and ever more policy and societal fields today have a counter-terrorism or counter-radicalisation/P/CVE component. This chapter has argued for applying the concept of ‘mainstreaming’ to these processes of counter-terrorism’s and counter-radicalisation’s growth, incorporation, and integration. While generally used with regard to other topics of policy and discourse, ‘mainstreaming’ works well as a framing device for counter-terrorism and counter-radicalisation’s striking evolution over the past 20 years. The increasing expansion and interconnectedness of counter-terrorism and counterradicalisation discourse with other fields and their ever-growing adoption of sectors outside their narrow starting point were in need of a name, and ‘mainstreaming’ fit the label. Going through key understandings of mainstreaming and its possible implications, this chapter has shown how the concept makes sense of the discursive evolution of counter-terrorism and counter-radicalisation. It signifies both the integration of such discourses into existing programmes and priorities and the ‘naturalisation’ of this development through its non-reversed and often unquestioned (although sometimes slow or even temporarily halted) progression. In sum, counter-terrorism needed counter-radicalisation for it to be mainstreamed, which again enabled both to be ‘naturally’ incorporated in states’ and institutions overall functioning: Counter-terrorism’s narrow focus on attack prevention would not allow it seamless entry into other societal domains, whereas counter-radicalisation could easily find a home and lay claim to any other sector in its service. The one-directionness, apparent self-evidentness, and the seeming nonfinality of counter-terrorism’s mainstreaming through counterradicalisation all suggest a naturalisation of these agendas. In the discourses investigated here, the growth of counter-radicalisation into other fields and its accumulative claim to be a ‘stakeholder’ within a wide array of other sectors has sometimes been controversial, but it has not stopped. That today’s discourses and approaches to counter-radicalisation have become less exceptional and more generically oriented should not be taken as a sign of their downgrading or of their having been set aside. Rather, their so-far irreversible blending into other agendas and fields seems to be evidence of their perpetual naturalisation in international and policy discourse overall.

Notes 1 The author wants to thank the editors of this volume for useful feedback on early abstracts of this chapter, as well as Cristiana Maglia and Gabriella Bolstad for invaluable research assistance. 2 For an influential understanding, see Schmid (2013). 3 For this strand of the literature’s flagship journal, see Critical Studies on Terrorism. Retrieved 27 June 2021, from www.tandfonline.com/toc/rter20/current 4 For education, see the UK Department for Education (2015). 5 See, for example, the League of Nations 1937 initiative to draft a convention for the prevention and punishment of terrorism; Saul (2011) provides an overview.

395

Rita Augestad Knudsen 6 For this and the subsequent reviews and accompanying reports, see the UN Global Counter-Terrorism Strategy. 7 For details, see the Radicalisation Awareness Network (RAN). Retrieved 27 June 2021, from https:// ec.europa.eu/home-affairs/what-we-do/networks/radicalisation_awareness_network_en 8 For the prison context, see, for example, the Council of Europe (2016).

References Akwanda, S. (2020, August 12). Mainstream CVE into your programmes, country governments told. Kenya News Agency. www.kenyanews.go.ke/mainstream-cve-into-your-programmes-county-governments-told/ Augestad Knudsen, R. (2013). Privatisation in Kosovo: “Liberal peace” in practice. Journal of Intervention and Statebuilding, 7(3), 287–307. https://doi.org/10.1080/17502977.2012.734559 Barnett, M., & Finnemore, M. (2004). Rules for the world: International organisations in global politics. Cornell University Press. Borum, R. (2003). Understanding the terrorist mindset. FBI Law Enforcement Bulletin, 72(7), 7–10. https:// scholarcommons.usf.edu/cgi/viewcontent.cgi?article=1227&context=mhlp_facpub Borum, R. (2011). Radicalisation into violent extremism II: A review of conceptual models and empirical research. Journal of Strategic Security, 4(4), 37–62. http://dx.doi.org/10.5038/1944-0472.4.4.2 Council of the EU. (2005, November 30). The European Union counter-terrorism strategy. https://data.consilium.europa.eu/doc/document/ST%2014469%202005%20REV%204/EN/pdf Council of the EU. (2014, October 9–10). Council discusses further action regarding foreign fighters. www. consilium.europa.eu/en/meetings/jha/2014/10/09-10/ Council of the EU. (2015a, March 16). Council conclusions on the EU regional strategy for Syria and Iraq as well as the ISIL/Daesh threat. https://data.consilium.europa.eu/doc/document/ST-7267-2015-INIT/en/pdf Council of the EU. (2015b, November 15). Conclusions of the council of the European Union and of the member states meeting within the council on enhancing the criminal justice response to radicalisation leading to terrorism and violent extremism. www.consilium.europa.eu/en/press/press-releases/2015/11/20/ conclusions-radicalisation/ Council of Europe. (2016, December 1). Council of Europe handbook for prison and probation service regarding radicalisation and violent extremism. https://rm.coe.int/16806f9aa9 Council of the European Union. (2016a, April 21). Council adopts EU passenger name record (PNR) directive. www.consilium.europa.eu/en/press/press-releases/2016/04/21/council-adopts-eu-pnr-directive/ Council of the European Union. (2016b, November 23). Conclusions of the council and of the representatives of the governments of the member states, meeting within the council, on the prevention of radicalisation leading to violent extremism. https://data.consilium.europa.eu/doc/document/ST-14276-2016-INIT/en/pdf Council of the European Union. (2017, March 7). EU strengthens rules to prevent new forms of terrorism. www.consilium.europa.eu/en/press/press-releases/2017/03/07/rules-to-prevent-new-forms-of-terrorism/ Cross, M. D. (2017). Counter-terrorism in the EU’s external relations. Journal of European Integration, 39(5), 609–624. https://doi.org/10.1080/07036337.2017.1327524 Daly, M. (2005). Gender mainstreaming in theory and practice. Social Politics: International Studies in Gender, State & Society, 12(3), 433–450. https://doi.org/10.1093/sp/jxi023 Department for Education. (2015, June). The prevent duty: Departmental advice for schools and childcare providers. https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/ file/439598/prevent-duty-departmental-advice-v6.pdf European Commission. (2014, January 15). Preventing radicalisation to terrorism and violent extremism (COM/2013/941). https://ec.europa.eu/home-affairs/sites/default/files/e-library/documents/policies/crisis-and-terrorism/radicalisation/docs/communication_on_preventing_radicalisation_and_violence_promoting_extremism_201301_en.pdf European Commission. (2018, October 24). Steering board on radicalisation. https://ec.europa.eu/transparency/ expert-groups-register/screen/meetings/consult?lang=en&meetingId=11350&fromExpertGroups=true European Commission. (2020a, December 9). A counter-terrorism agenda for the EU: Anticipate, prevent, protect, respond (COM/2020/795). https://ec.europa.eu/home-affairs/sites/default/files/pdf/09122020_ communication_commission_european_parliament_the_council_eu_agenda_counter_terrorism_ po-2020-9031_com-2020_795_en.pdf European Commission. (2020b, July 24). On the EU security union strategy (COM/2020/605). https://eur-lex. europa.eu/legal-content/EN/TXT/PDF/?uri=CELEX:52020DC0605&from=EN

396

Naturalisation through mainstreaming European Commission. (n.d.a). Radicalisation awareness network. European Commission. https://ec.europa. eu/home-affairs/what-we-do/networks/radicalisation_awareness_network_en European Commission. (n.d.b). RAN mental health working group (RAN HEALTH). https://ec.europa.eu/ home-affairs/what-we-do/networks/radicalisation_awareness_network/topics-and-working-groups/ ran-health_en Europol. (n.d.). European counter terrorism centre (ECTC). www.europol.europa.eu/about-europol/ european-counter-terrorism-centre-ectc Fairclough, N. (2010). Critical discourse analysis: The critical study of language (2nd ed.). Routledge. Henshaw, A. (2021, March 29). Bringing women, peace and security online: Mainstreaming gender in responses to online extremism. Global Network on Extremism & Technology. https://gnet-research.org/ wp-content/uploads/2021/03/GNET-Report-Women-Peace-And-Security.pdf High Representative of the Union for Foreign Affairs and Security Policy. (2016). Shared vision, common action. A Global Strategy for the European Union’s Foreign and Security Policy. https://eeas.europa.eu/ archives/docs/top_stories/pdf/eugs_review_web.pdf Janks, H. (1997). Critical discourse analysis as a research tool. Discourse: Studies in the Cultural Politics of Education, 18(3), 329–342. https://doi.org/10.1080/0159630970180302 Koskenniemi, M. (2010). Human rights mainstreaming as a strategy for institutional power. Humanity, 1(1), 47–58. https://doi.org/10.1353/hum.2010.0003. Lugna, L. (2006). Institutional framework of the European Union counter-terrorism policy setting. Baltic Security & Defense Review, 8, 101–127. www.baltdefcol.org/files/docs/bsdr/6-EU%20Counterterrorism%20policy-Institutional%20Framework-Lauri%20Lugna.pdf McCauley, C., & Moskalenko, S. (2008). Mechanisms of political radicalisation: Pathways toward terrorism. Terrorism and Political Violence, 20(3), 415–433. https://doi.org/10.1080/09546550802073367 Moghaddam, F. M. (2005). The staircase to terrorism: A psychological exploration. American Psychologist, 60(2), 161–169. https://doi.org/10.1037/0003-066X.60.2.161 Monar, J. (2007). Common threat and common response? The European Union’s counter‐terrorism strategy and its problems. Government and Opposition, 42(3), 292–313. https://doi. org/10.1111/j.1477-7053.2007.00225.x Monar, J. (2015). The EU as an international counter-terrorism actor. Intelligence and National Security, 30(2), 333–356. https://doi.org/10.1080/02684527.2014.988448 Paterson, S. (2010). What’s the problem with gender-based analysis? Gender mainstreaming policy and practice in Canada. Canadian Public Administration, 53(3), 395–416. https://doi.org/10.1111/ j.1754-7121.2010.00134.x Rai, S. (2003). Mainstreaming gender, democratising the state. Manchester University Press. Rees, T. (2005). Reflections on the uneven development of gender mainstreaming in Europe. International Feminist Journal of Politics, 7(4), 555–574. https://doi.org/10.1080/14616740500284532 Rothermel, A. K. (2020). Gender in the United Nations’ agenda on preventing and countering violent extremism. International Feminist Journal of Politics, 22(5), 720–741. https://doi.org/10.1080/14616742.2020 .1827967 Ruperez, J. (2017, January 20). The United Nations in the fight against terrorism. Executive Director CounterTerrorism Committee, unspecified. Saul, B. (2011). The legal response of the league of nations to terrorism. Journal of International Criminal Justice, 4(1), 78–102. https://doi.org/10.1093/jicj/mqi096 Sayer, A. (2003). (De)commodification, consumer culture, and moral economy. Environment and Planning D: Society and Space, 21(3), 341–357. https://doi.org/10.1068/d353 Schmid, A. P. (2013). Radicalisation, de-radicalisation, counter-radicalisation: A conceptual discussion and literature review. The International Centre for Counter-Terrorism. http://dx.doi.org/10.19165/2013.1.02 Schuurman, B., Lindekilde, L., Malthaner, S., O’Connor, F., Gill, P., & Bouhana, N. (2019). End of the lone wolf: The typology that should not have been. Studies in Conflict & Terrorism, 42(8), 771–778. https://doi. org/10.1080/1057610X.2017.1419554 Squires, J. (2005). Is mainstreaming transformative? Social Politics: International Studies in Gender, State & Society, 12(3), 366–388. https://doi.org/10.1093/sp/jxi020 Thiessen, C. (2019). The strategic ambiguity of the United Nations approach to preventing violent extremism. Studies in Conflict and Terrorism. https://doi.org/10.1080/1057610X.2019.1647685 True, J., & Mintrom, M. (2002). Transnational networks and policy diffusion: The case of gender mainstreaming. International Studies Quarterly, 45(1), 27–57. https://doi.org/10.1111/0020-8833.00181

397

Rita Augestad Knudsen UN. (1963, September  14). Convention on offences and certain other acts committed on board aircrafts (Treaty Series, 704). https://treaties.un.org/doc/db/Terrorism/Conv1-english.pdf UN. (1970, December 16). Convention for the suppression of unlawful seizure of aircraft (Treaty Series, 860). https://treaties.un.org/doc/db/Terrorism/Conv2-english.pdf UN. (1971, September 23). Convention for the suppression of unlawful acts against the safety of civil aviation (Treaty Series, 974). https://treaties.un.org/doc/db/Terrorism/Conv3-english.pdf UN. (1979, December 17). International convention against the taking of hostages (Treaty Series, 1316). https://treaties.un.org/doc/db/Terrorism/english-18-5.pdf UN. (1997a, September 18). Gender mainstreaming (A/52/3). www.un.org/womenwatch/daw/csw/GMS.PDF UN. (1997b, December 15). International convention for the suppression of terrorist bombings (Treaty Series, 2149). https://treaties.un.org/doc/db/Terrorism/english-18-9.pdf UN. (2006). UN global counter-terrorism strategy. www.un.org/counterterrorism/un-global-counterterrorism-strategy UN General Assembly. (2006, September 20). The United Nations global counter-terrorism strategy (A/ RES/60/288). https://documents-dds-ny.un.org/doc/UNDOC/GEN/N05/504/88/PDF/N0550488.pdf? OpenElement UN General Assembly. (2015a, December 24). Letter dated 22 December 2015 from the secretary-general to the president of the general assembly (A/70/675). www.un.org/en/ga/search/view_doc.asp?symbol=A/70/675 UN General Assembly. (2015b, December 24). Plan of action to prevent violent extremism: Report of the secretary-general (A/70/674). www.un.org/en/ga/search/view_doc.asp?symbol=A/70/674 UN Security Council. (1999, October 19). Resolution 1269 (S/RES/1269). https://digitallibrary.un.org/ record/287509 UN Security Council. (2001, September 12). Resolution 1368 (S/RES/1368). www.un.org/en/ga/search/ view_doc.asp?symbol=S/RES/1368%282001%29 Vries, G. (2005). The European Union’s role in the fight against terrorism. Irish Studies in International Affairs, 16, 3–9. www.jstor.org/stable/30001931 Wolff, S. (2008). The Mediterranean dimension of EU counter‐terrorism. Journal of European Integration, 31(1), 137–156. https://doi.org/10.1080/07036330802504013

398

26 COUNTERING RADICALISATION WHILE EXPANDING THE CRIMINAL LAW Tufyal Choudhury

Introduction After the 2001 terrorist attacks on New York and Washington, DC, counter-terrorism institutions and officials rapidly recalibrated their policies towards stopping future attacks from coming to fruition. The US military-led response placed the initial focus on changes in defence doctrines and recasting the rules on the use of force in international law. Parallel to the military response was an emphasis on prosecution through the expansion of the criminal law, the increased use of executive and administrative powers, and the development of social policy responses to address conditions conducive to terrorism. This chapter focuses on the development and use of legal instruments to prevent terrorism and counter-radicalisation. The dominant mode of legal regulation is the expansion of the reach of criminal law through the development of ‘pre-emptive’ or ‘pre-crime’1 offences. These criminalise actions and activities that are some distance, causally and temporally, from the harm of a terrorist attack. They prohibit conduct associated with the activities of terrorists or actions that enable and encourage radicalisation. Pre-emptive terrorism offences are an important feature of the post-2001 global counter-terrorism architecture. UN Security Council resolutions require all states to criminalise an increasing range of pre-emptive and preparatory actions related to terrorism.2 Regionally, both the Council of Europe and the European Union have contributed to the normalisation of pre-emption. In 2005, the Council of Europe adopted the Convention of the Prevention on Terrorism and a CounterTerrorism Strategy in 2018.3 The EU instigated the introduction of terrorism offences in the penal codes of many Member States through its Framework Decision on Combating Terrorism.4 This required states to criminalise nine ‘terrorist acts’,5 including public provocation to commit a terrorist act,6 recruitment for terrorism,7 providing or receiving training for terrorism,8 travelling9 or supporting or otherwise facilitating travel for the purpose of terrorism,10 and financing terrorism.11 There is evidence of the increasing use of pre-emptive offences in some European states. In France, following changes to the offence of ‘apology for terrorism’ in 2014, prosecutions have leapt from three convictions in 2014 to 306 in 2016 (Commissioner for Human Rights, 2018). These were the results of 1,850 police investigations, one in five of which involved young people

399

DOI: 10.4324/9781003035848-30

Tufyal Choudhury

under the age of 18 (Houry, 2018). By 2018, ‘apology for terrorism’ was the most frequently used counter-terrorism criminal measure in France (UN OHCHR, 2018). In Spain, the criminal law code was expanded in 2015 to cover a wider range of pre-emptive terrorism offences,12 focusing on prohibiting ‘training and indoctrination in hate’. This extends beyond training in practical skills to encompass ‘ideological training’ through indoctrination. Controversially, this covers not only the indoctrination of others but also self-indoctrination and makes it unlawful to repeatedly access online material that incites terrorism. It also prohibits the acquisition or possession of documents of this nature for the purpose of self-indoctrination in order to commit any terrorist offence. The pre-crime offences are so wide-ranging that an individual may be found guilty of an offence if they are simply in possession of a book with the aim of indoctrinating themselves to commit a terrorist offence, even if they have not read it (Rimo, 2021). The attacks on Madrid in 2004 and London in 2005 evidenced the transnational reach of alQaida and its ability to direct and inspire young Europeans to murder their fellow citizens; this instigated greater emphasis on international policy coordination and cooperation on interventions outside the criminal law framework. Across Europe, ‘radicalisation’ swiftly surfaced as the dominant conceptual framework favoured by politicians and policymakers grappling to understand and explain so-called “home-grown” terrorism and soon garnered a policy response aimed at preventing radicalisation (Coolsaet, 2005). Since then, both the UN and the EU have promoted the adoption of policies that aim at preventing radicalisation through support for individuals and communities assessed as vulnerable to terrorist recruitment (Hayes & Kundnani, 2018). The conflicting imperatives between policies of deterrence through prosecution and prevention through support reveal tensions at the point of policy enactment and implementation. To understand how these contradictions and strains are manifested and managed, this chapter focuses on the development and implementation of policies in the United Kingdom. While international and regional organisations generate common frameworks, understandings, and policy approaches, States retain significant discretion in their implementation of international obligations. This can lead to divergent approaches between states. For example, an analysis of terrorism related criminal offences in Germany and the UK suggested that the German constitutional law principle of legal certainty contributed to a significantly narrower scope for such offences in Germany compared to the United Kingdom (Cornford & Petzsche, 2020). This chapter begins by examining the UK counter-radicalisation policy Prevent, which features prominently in Britain’s claims to global leadership in the development of counter-radicalisation policies.13 International interest in the UK Prevent policy intensified after it augmented its approach to countering radicalisation through a legal duty compelling key institutions and organisations to have policies and training in place to identify and, if necessary, report individuals deemed to be at risk of radicalisation.14 When it was introduced, there were concerns that the duty would alienate and stigmatise Muslims, undermine trust in public services, and conflict with other legal obligations on human rights, equality, and data protection. Central to garnering acceptance of the Prevent Duty among professionals tasked with its implementation has been its framing as a safeguarding measure that operates in a ‘non-criminal’ space (Heath-Kelly & Strausz, 2017; Busher et al., 2019). Given the importance attached to framing Prevent as a non-criminal measure, it is important to understand the ways in which the concerns of criminal laws overlap with those of the Prevent Duty. The chapter details the development of pre-emptive criminal law offences to counter radicalisation. While many originate in legislative responses to political violence in Northern Ireland and were included in the Terrorism Act 2000 (TA 2000), since 2001, the number and scope of pre-emptive terrorism offences have expanded significantly to keep pace with evolving fears and 400

Countering radicalisation while expanding the criminal law

understandings of radicalisation pathways and processes. The continual expansion of the criminal law through pre-emptive terrorism offences raises concerns about the infringement of human rights and their consistency with the core principles of the rule of law. While policy documents, guidelines, and training accompanying the Prevent Duty narrate a sharp boundary between the criminal and ‘non-criminal’ spaces, the relentless expansion of pre-emptive offences has drastically contracted, if not entirely eroded, the non-criminal spaces in which the duty seeks to operate.

Countering radicalisation through the Prevent strategy and duty Countering radicalisation lies at the core of Prevent, a pillar of the UK’s Counter-Terrorism strategy, CONTEST. The focus and scope of Prevent have changed significantly over time. The initial strategy, developed after the London 2005 bombings, focused only on Muslim communities and prioritised engagement and delivery through partnerships with community organisations. It supported projects aimed at capacity building in Muslim communities and activities overcoming barriers to integration (Thomas, 2015). A revised strategy published in 2011 recast its scope to take in all forms of terrorism and confirmed the expansion of Prevent from concern with ‘violent extremism’ to the broader notion of ‘extremism’.15 The latter is defined as, ‘vocal or active opposition to fundamental British values, including democracy, the rule of law, individual liberty and mutual respect and tolerance of different faiths and beliefs’. There was now a greater emphasis on challenging ideology and the broader set of ideas that, it was argued, underpinned radicalisation. Thus, ‘preventing terrorism’ involved challenging extremist (and non-violent) ideas that are also part of a terrorist ideology. Commitment to the rule of law and liberal values was now a bulwark against radicalisation, while a lack of commitment to them was a sign of extremism. The 2011 revision signalled a greater focus on ‘sectors and institutions where there are risks of radicalisation’. However, reluctance and reticence on the part of some organisations and institutions to work on Prevent contributed, in part, to the introduction of the statutory ‘Prevent Duty’ to compel cooperation by specified authorities in Great Britain.16 In educational institutions, initial opposition and anxiety towards the duty among professionals tasked with its implementation appear to have given way to reluctant accommodation and rapid normalisation of the duty into everyday practice (Jerome et al., 2020). An important contributor to the professional acceptance of the duty has been its framing and enactment through existing policies and practices of ‘safeguarding’ vulnerable individuals (Busher et al., 2019). Critics argue that the national security imperatives of Prevent sit uncomfortably with traditional notions of ‘safeguarding’, suggesting that it is not just the ‘at risk’ individual who is being protected from radicalisation but also the society that may need to be protected from the ‘risky’ individual (Rizq, 2017). In implementing the duty, practitioners see their subjects as simultaneously at risk of radicalisation and posing a risk of terrorism (Pettinger, 2020). Structural and discursive analysis of Prevent as well as empirical research on implementation highlight the difficulties in distinguishing between the would-be terrorist, who engages in lawful but radicalising activities, placing them on the pathway towards terrorism, and the alleged terrorist, who has crossed the Rubicon and engaged in unlawful terrorismrelated activities. Counter-terrorism policy documentation and discourse narrate and sustain a sharp rhetorical distinction between the non-criminal and the criminal, between becoming and being a terrorist. The UK National Security Strategy declares confidently that the Prevent Duty will enable interventions with individuals at risk of radicalisation ‘to stop them before they begin to engage in any terrorist-related activity’ (HM Government, 2015, p. 38, emphasis added). A cordon sanitaire is placed around Prevent and the Channel programme. The Prevent Strategy recognises that in the 401

Tufyal Choudhury

course of Prevent-related work, practitioners may ‘identify someone who may already be engaged in illegal terrorism-related activity’; however, at such a point, the matter must be referred to the police for investigation outside the framework of Prevent (HM Government, 2011, p. 32). Police involved in Prevent aim to ensure that people engaged in terrorism-related ‘criminal activity’ are not placed in the Channel programme (HM Government, 2011, p. 61). The Channel Duty Guidance reiterates this distinction; it stresses that it ‘provides early support for anyone who is vulnerable to being drawn into any form of terrorism or supporting terrorist organisations’ and that the support it provides is targeted at individuals ‘before they become involved in criminal terroristrelated activity’ (HM Government, 2020, p. 7, emphasis added). A similar demarcation is evident in documents guiding the implementation of the Prevent Duty in healthcare, which declare that Prevent operates in a ‘pre-criminal space’ (Goldberg et al., 2017). For Heath-Kelly (2017), while Prevent has always located itself in the pre-criminal space, the revised strategy and duty reflect an epistemological shift and ‘reconceptualization of pre-crime’ away from prediction of risk and a focus on particular neighbourhoods towards a view of risk based on big data and algorithmic logic applied across the whole of Britain. Researchers have mapped the varied terrain in the emerging topography of the pre-crime space, from the therapy room (Rizq, 2017) and classroom, to the family home (Fernandez, 2018; Abbas, 2019) and spaces of associational community activities. It covers both temporal and spatial dimensions and ‘denotes that time, space and inter-professional activity involved in planning, coordinating support and possibly monitoring in the NHS England Prevent programme’ (Goldberg et al., 2017, p. 209). Empirical research exploring how the duty is interpreted and implemented finds the narration of Prevent as pre-crime reproduced by practitioners. In interviews with Pettinger (2020, p. 128), Channel mentors describe the counter-radicalisation programme as operating ‘fundamentally . . . in the pre-crime space’ so that they are ‘not talking about people who have broken the law yet, but there are concerns’. Operating in a ‘pre-crime’ space does not take Prevent beyond the gaze or reach of the widening array of institutions and individuals enrolled into the apparatus of the national security state through the Prevent Duty. The strength of the emphatic government rejection of allegations that Prevent involves or encourages ‘spying’ on Muslims is blunted by its careful confinement to ‘covert spying’ and the absence of any clarification on the distinction between spying and surveillance (HM Government, 2011). The Prevent Duty requires specified authorities to be able to identify those ‘at risk of radicalisation’. The delivery of Prevent training plays a fundamental role in the Duty’s implementation. For the Home Office, one measure of success in preventing terrorism is the number of times Prevent training is completed (Home Office, 2021, p. 4). This training identifies the signs and indicators frontline professionals must be alert to while warning against discrimination and profiling. Much of the critique of the implications of Prevent has drawn on theories of governmentality to emphasise the disciplinary nature of governance through Prevent (Heath-Kelly, 2012; Martin, 2014). Research on the implementation of Prevent documents the ways in which it embeds surveillance by calling on frontline staff to look out for signs of radicalisation that are diffuse and ever-present. Researchers have emphasised the banality (Coppock & McGovern, 2014) of the indicators, questioned their evidence base (Qureshi, 2016), reliability (Sageman, 2021), and the inherently speculative endeavour of identifying a genuine case, a process of ‘brokering’ (Verfaille et al., 2019) or ‘fixing’ (Pettinger, 2020) identity. For some, racialised discrimination is a significant risk that arises from unbounded discretion (Verfaille et al., 2019), while others see this as embedded and inherent to imagining future risks (Ali, 2020; Younis, 2021). Human rights organisations have drawn attention to the ways in which Prevent infringes on fundamental rights, particularly freedoms of expression, religion, and association (Rights Watch 402

Countering radicalisation while expanding the criminal law

UK, 2016). However, it is difficult to prove that an individual infringement reaches the level of intensity required to constitute an unlawful violation of such rights. The rights in question are often qualified rights, so states can justify an interference on the grounds of protecting national security. Judicial bodies also show deference to the executive in judging the necessity of actions required to protect national security. Furthermore, unlike criminal law, where the punitive assertion of state power through conviction and imprisonment, is clear, the impact of Prevent policies is more diffuse. Notwithstanding these limitations, Walker and Cawley (2020) identify important benefits that arise from the ‘juridification’ of Prevent through the duty, including greater transparency and accountability as implementation generates guidance and enables space for legal, including rightsbased challenges. While legal analysis on the implications and impact of the Prevent Duty on fundamental rights and the rule of law has been muted, legal scholarship has developed an extensive critique of the expansion of criminal law.

Criminal law offences and the preventive state Immediately after the July 2005 London bombings, then Prime Minister Tony Blair, drawing on the discourse of the ‘new terrorism’ thesis, argued that the scale of violence and level of threat posed by Al-Qaeda differed qualitatively from nationalist and separatist terrorism, and so the ‘rules of the game’ had changed (Brown & Wolf, 2013). The government proposed a raft of new pre-emptive offences. These were needed as the scale of causalities from a successful attack required offences enabling earlier intervention to allow the state to ‘defend further up the field’ (Anderson, 2013). To appreciate the expansive reach of pre-emptive terrorism offences, it is necessary to understand how they differ from and develop beyond the scope and focus of most ordinary criminal offences. Specific terrorism offences are not necessary to prosecute terrorists for the direct harm resulting from a terrorist attack. Such harms are covered by established general laws proscribing murder, the destruction of property, and the use or possession of explosives and weapons. Ordinary criminal laws also provide some protection from harm before it occurs by enabling prosecution of incomplete or inchoate offences; they attach liability to attempting, assisting, encouraging, or aiding and abetting another in committing an offence. The broad approach, underpinned by the need to respect individual autonomy, aims to limit liability for incomplete offences to individuals who demonstrate a determination to commit an offence while leaving room for individuals to change their mind. Thus, to commit an ‘attempt’, a defendant must engage in actions that are more than ‘merely preparatory’. For example, a person may drive to the bank and park outside with the intention of robbing it. However, until they enter the bank, the possibility that they may change their mind is deemed to exist, and so their conduct is viewed as preparatory and not yet an attempt.17 By contrast, some pre-emptive terrorism offences criminalise conduct well before a specific plan is determined or developed. Others prohibit conduct that relies on further action by third parties for harm to occur, even though there is no plan, agreement, or intention to encourage any third party. Pre-emptive terrorism offences thus extend the temporal and causal reach of the criminal law beyond the actions covered by inchoate liability, creating ‘pre-inchoate’ or ‘pre-crime’ offences (Zedner, 2014). Criminologists locate the development of pre-emptive terrorism offences in the context of deeper transformation in modern society, generating a diffuse sense of insecurity and the concomitant rise of social policies directed towards the management of chronic risks endemic to post-industrial societies (Beck, 1992). This raises new challenges for policing and criminal justice (Ericson & Haggerty, 1997). One response has been the emergence of ‘actuarial justice’, an approach that seeks to identify the level of risk posed by individuals or groups and to manage the danger they pose rather than punish or reform individuals (Feeley & Simon, 1994). The model 403

Tufyal Choudhury

of actuarial justice is not an easy fit for counter-terrorism, as the low number of incidents of and individuals involved in terrorism precludes the development of adequate models for risk assessment. Yet, as Murphy (2012, p. 11) observes, ‘in the face of these unknowns, government agencies have taken to imagining potential future attacks; replacing the statistical basis of actuarial justice with worst case scenario hypothesis’, leading to a shift from what is probable to what is possible. The development of new terrorism offences is therefore part of a shift from a post-crime society that punishes culpable and harmful actions to a ‘pre-crime’ society that criminalises actions that generate risks of future harm (Zedner, 2007). The new offences are located in a broader trend of the development of the ‘preventive state’ (Steiker, 1998) that has prompted a burgeoning legal scholarship identifying and articulating normative limits for preventive measures (Ashworth & Zedner, 2014) that some courts are beginning to explicitly engage with (Tulich et al., 2021). McCulloch and Pickering (2010) suggest that measures for preventing terrorism, while a manifestation of the ‘Preventive state’, operate in ways that distinguish them from traditional crime prevention. The latter, aimed at addressing the environment and social conditions that give rise to crime, is ‘non-punitive’. By contrast, counter-terrorism measures retain a ‘coercive or punitive accent’ (McCulloch & Pickering, 2010, p. 15). This is clearly true of criminal offences that carry the possibility of a significant custodial sentence, but it is also true of other non-criminal measures that can involve restrictions on liberty or authorise extensive surveillance. They also note the paucity of evidence that preventive counter-terrorism measures prevent terrorism. Thus, they argue, calling counter-terrorism measures preventive ‘asserts an outcome that cannot be shown to be evidenced’. Instead, the concept of ‘pre-emption’ offers a more accurate label, not least because it makes explicit the ‘strategy behind the legislation’ (McCulloch & Pickering, 2010, p. 17).

The expansive definition of terrorism The Prevent strategy and duty aim to stop people from becoming terrorists; however, anyone who is engaged in the commission, preparation, or instigation of ‘acts of terrorism’ is already a terrorist.18 Knowing whether a person is committing an act of terrorism is particularly important for frontline practitioners involved in counter-radicalisation, as failure to report to the police ‘any information’ they know or believe may be of ‘material assistance in preventing the commission by another person of an act of terrorism’ is a criminal offence.19 The problem for those working in counter-radicalisation is the vast range of conduct that constitutes terrorism and therefore requires reporting to the police. The extensive reach of terrorism offences rests both on a broad definition of terrorism and on the wide range of activities and actions that are criminalised as related to or predictive of future terrorism. Terrorism is not a specific offence; however, the definition of terrorism provides the basis for identifying specific ‘acts of terrorism’. Identifying the range of actions and activities that constitute an ‘act of terrorism’ is important in determining whether conduct crosses the line between non-criminal and criminal. In UK law, terrorism entails the ‘use or threat of action’ that is ‘designed to influence the government’, ‘an international governmental organisation’, or ‘the public or a section of the public’. Furthermore, the use or threat of action must be ‘made for the purposes of advancing a political, religious, racial, or ideological cause’.20 Thus, the purpose of the action is pivotal in turning it into a terrorist act. The ‘actions’ constitutive of terrorist acts include serious violence against a person, actions endangering the life of a person, serious violence to property, and creating a serious risk to the health and safety of the public or a section of the public.21 Furthermore, any action taken for the benefit of a proscribed organisation is also deemed to be ‘an action for the purposes of terrorism’.22 404

Countering radicalisation while expanding the criminal law

This is particularly important for countering radicalisation as ‘proscribed organisations’ are not limited to those engaged in terrorist violence but can include organisations believed to be involved in radicalisation. The shift away from traditional criminalisation of directly harmful acts towards pre-emption begins with the seemingly innocuous extensions of the definition of terrorism beyond acts of violence to include ‘threatening’ to commit any terrorist acts. This expands the reach of the criminal law to bring future possible action within the scope of a prosecution for terrorism. It extends liability to words that have not manifested as actions sufficient to constitute an attempt and to plans that do not amount to a conspiracy. Furthermore, the definition of terrorism is not limited to actions or threats in the United Kingdom but extends to any actions or threats directed at any government, even authoritarian dictatorships. The UK Supreme Court rejected adopting a narrower reading of the territorial scope, accepting that the definition ‘was indeed intended to be very wide’.23 Pre-emptive offences build on this broad definition of terrorism. The intersection of evidence relevant to identifying individuals who are believed to be in the process of radicalisation and those engaging in actions and activities that constitute terrorist acts arises most clearly in relation to possession, preparation, encouragement, and offences linked to proscribed organisations.

Possessing or viewing material connected to terrorism A person becomes a terrorist if they possess material or information that can be connected to terrorism. There are two ways this can happen. First, the accused may ‘possess an article in circumstances which give rise to a reasonable suspicion that his possession is for a purpose connected with the commission, preparation or instigation of an act of terrorism’.24 The range of material that can make the possessor a terrorist is particularly wide, as it includes any item that can be used for the ‘preparation’ of an act of terrorism. This appears to include the possession of propaganda material that offers political or moral justification for terrorism, as such justification can be an important element in the preparation for an act of terrorism.25 Second, an individual becomes a terrorist if they collect, possess, or view information ‘of a kind likely to be useful to a person committing or preparing an act of terrorism’. In 2019, this offence was dramatically expanded with the addition of online viewing as an offence, which widens the net of people who have the potential to be treated as terrorists. This extension was justified as essential to ensure the criminal law kept pace with evolving methods of radicalisation, in particular the escalating fear of online radicalisation. The nature of the information targeted by the offence is potentially broad, and its scope remains unclear. One review of the case law concluded that ‘exactly what information is targeted and why remains mysterious’, and consequently ‘it is hard to know whom we ought to prosecute for what’ (Cornford, 2020, p. 682). There is a ‘reasonable excuse’ defence that includes not knowing or having no reason to believe that the document or record contained or was likely to contain prohibited information. However, it is not clear whether the reasonable excuse defence extends to a person who, knowing the material they are about to view contains prohibited information, nevertheless chooses to view it out of personal curiosity or interest (Zedner, 2021). The UK’s Independent Reviewer of Terrorism Legislation noted that prosecutors do ‘not take the view that mere curiosity will always be a reasonable excuse’, and so ‘the curious must thus place their faith in the restrained exercise of prosecutorial discretion’ (Anderson, 2012, p. 125). In relation to both offences, a person can be a terrorist even if they lack any terrorist intent. For the first offence, the required mental state is only that the accused knows that they possess or 405

Tufyal Choudhury

have control over the prohibited material. It is not necessary to show that the accused intended to commission, prepare, or instigate a terrorist act; it is sufficient that the material in their possession gives rise to a reasonable suspicion that it is for this purpose. As Walker notes, such ‘reasonable suspicion’ can be fashioned from the circumstantial evidence of the accused’s expressions of extremist ideas and beliefs alongside their interactions and association with a presumed extremist (Walker, 2011, p. 214). Similarly, there is no requirement for the person viewing or collecting the information to intend to commission, prepare, or instigate a terrorist act; it is sufficient that they intended to collect, possess, or view the prohibited information.

Preparation for terrorism As noted earlier, ordinary criminal law prohibits any ‘attempt’ at a crime but does not criminalise a person if their actions are limited to ‘mere preparation’. By contrast, preparatory acts are sufficient to render a person a terrorist if the actions are to give effect to an intention to commit acts of terrorism or to assist another to commit such acts.26 In the clearest example of the extension of the temporal reach of the criminal law, the offence widens the net of the criminal law to capture preparation. The vast range of actions that can be regarded as preparation leaves the scope of the law unclear and uncertain. The offence has been used in a wider range of contexts, from a former British Marine who had assembled sufficient materials to build 19 pipe bombs and possessed several IEDs to actions by individuals preparing to travel to ISIS-controlled territory in Syria. Unlike the two possession offences, here there is a requirement for the terrorist suspect to have terrorist intent. However, the required intention is not an intention to carry out a specific and determined terrorist act; the offence only requires proof of a more general intention to engage in terrorist actions. The overlap with the Prevent Duty arises as proof of a general intention to commit acts of terrorism that may be knitted together by bringing together circumstantial evidence of the viewing or possession of extremist videos, the expression of legal but ‘extremist’ ideas, or the interactions with friends and associates who are linked to terrorism or extremist groups (Cornford, 2020).

Encouragement of terrorism The publication or dissemination of a statement that may directly or indirectly encourage or induce others to commit, prepare, or instigate acts of terrorism also places a person at risk of being a terrorist.27 The statement need not be particularly strong or emphatic to constitute encouragement; it is sufficient that the statement is one that ‘may’ have an encouraging effect. The range of statements that risk rendering a person a terrorist is wide since it includes not just direct encouragement but the more amorphous notion of ‘indirect’ encouragement. Furthermore, there is no requirement for the publication or dissemination to actually encourage anyone. In fact, there have been prosecutions for the publication of statements where it was clear that no one was ever encouraged. What makes this offence even broader is that the prohibited encouragement does not need to relate to a specific terrorist plot or plan; instead, the offences aim towards proscribing any general ‘encouragement’ of acts of terrorism. An analysis of the actual prosecutions conducted using this offence suggests that it covers material that ‘normalises terrorism or casts it in a positive or sympathetic light’ that it has been used against the publication of material that does not explicitly encourage or glorify terrorism but may have ‘potentially encouraging effects’ (Cornford, 2020, p. 672). For example, it was used against the publication of photos of an accused wearing what appear to be suicide vests (Cornford, 2020). Online social media posts, even to a private group, amount to publication, and sharing those posts can be dissemination. Carlile and MacDonald (2014, p. 165) suggest 406

Countering radicalisation while expanding the criminal law

that a person who tweets that ‘terrorists are incredibly brave to give up their lives for their cause’, and anyone who retweets this without condemnation, is liable to have committed a terrorist offence.

Proscribed organisations The UK Home Secretary is able to proscribe any organisation if she believes the organisation is ‘concerned with terrorism’.28 While proscription is used against groups like ISIL and Al-Qaeda that are directly engaged in terrorist violence, being ‘concerned’ with terrorism is broader than being ‘engaged’ in terrorism and enables the banning of groups primarily connected to radicalisation through, for example, supporting, condoning, or glorifying terrorism. Thus, proscription is used ‘to curtail radicalising activity’ by such organisations (HM Government, 2011, p. 26). The use of proscription-related offences has increased recently, with 21 prosecutions in Britain in 2011, compared to a total of 25 in the preceding six years (Hall, 2020, p. 38). Proscription gives rise to three potential terrorist offences. First, membership in a proscribed organisation is a terrorist offence; in fact, merely professing membership, even if the claim itself is unproven, is an offence. The offence does not require any further conduct, involvement, or active participation in the organisation. A Palestinian student who proclaimed in class that ‘I am Hamas’ committed an offence even though it was not known if the claim was true.29 Such an approach reflects the legislative objective since ‘it is dangerous even to pretend that a proscribed organisation is tolerable’ (Walker, 2011, p. 355). Second, anyone who ‘invites support’ for a proscribed group may be regarded as a terrorist.30 However, as individuals were found to be adept at carefully treading the line between expressing and inviting support for a proscribed group, the legal net was widened to make it illegal to express ‘an opinion or belief that is supportive of a proscribed organisation’.31 This expands the scope of the offence significantly and even risks covering statements made in debates on whether a group is a terrorist organisation or should be proscribed (Joint Committee on Human Rights, 2018). Third, it is an offence to wear, carry, or display, in a public place, articles that arouse suspicion of support for or membership in a proscribed group. This was further expanded to include publishing images of ‘an item of clothing, or any other article, in such a way or in such circumstances as to arouse reasonable suspicion that the person is a member or supporter of a proscribed organisation’. This would cover images taken in private and made public through publication. It is not necessary to intend to encourage support; the offence is committed if the display would arouse ‘reasonable suspicion’ that the person is a member or supporter of a proscribed organisation. An example of the kind of action targeted through this offence is the publication of an image of a Daesh flag hanging in a private bedroom. The criminalisation of such conduct is justified as necessary to deter people from engaging in activities that may encourage others to support a terrorist group (Home Office, 2018). Furthermore, in relation to all of the pre-emptive terrorist offences (possession, preparation, or encouragement), ‘any action’ taken for the benefit of a proscribed organisation is also deemed to be ‘an action for the purposes of terrorism’.32 This potentially brings a wide range of actions into the scope of terrorism, for example a person who provides space to meet with a proscribed organisation is liable for committing an act preparatory to terrorism since their action is one that benefits a proscribed organisation and actions benefiting a proscribed organisation are terrorist actions (Hall, 2020).

Criticism of pre-emptive offences For liberal states, pre-emptive offences present a significant departure from traditional principles of criminalisation centred on culpable harm. The scope of offences of terrorism-related possession, 407

Tufyal Choudhury

preparation, encouragement, and proscription that criminalise activities in the absence of directly harmful culpable conduct has attracted anxious scrutiny from criminal law theorists and disagreements as to their legitimacy (Cornford, 2020). For example, Ashworth and Zedner (2014, p. 180) find that the offence of indirect encouragement fails to meet the normal requirements of harm required in criminal law, as it criminalises conduct that may not be harmful and is only ‘tenuously causally related to the consummate harm’, particularly as the harm arises from the subsequent conduct of a third party. While Walker (2011, p. 225) warns that ‘the broad nature of “preparations” comes close to penalizing criminal thoughts rather than harms and leaving no possibility of withdrawal’. As these offences capture neutral and non-harmful actions, much of the concern centres on the weight they place on the mental state of the accused. The concern is particularly acute for offences that can be committed recklessly, that is, without requiring the accused to form any terrorism related intention. Even where a terror-related intention is required, this does not entail an intention fixed upon a specific plan or plot. Rather, it requires a general intention to engage in acts of terrorism. The effect is to extend the reach of the criminal law to situations where ‘the prospect of harm is uncertain and where the only immorality has been the imagining of wickedness rather than its infliction or a fair imputation that it will occur’ (Walker, 2011, p. 204). Second, in the absence of any evidence demonstrating the intention to carry out a particular terrorist act, there is concern that circumstantial evidence from a wide range of actions and activities is knitted together to show the accused possesses the required general intention to commit terrorist acts. The criminalisation of ‘guilty minds’ rather than ‘guilty acts’ usurps the principle that people should be accountable for their bad actions, not their bad thoughts. A significant proportion of the legal analysis of pre-emptive offences focuses on their incongruence with rule of law principles. In particular, the vague and broad boundaries of ‘terrorism’ and the range of conduct captured by terrorist offences appear to undermine the requirement that prohibited conduct should be ‘foreseeable’ so as to enable law-abiding citizens to plan their actions to remain within the boundaries of the law. Edwards suggests that pre-emptive offences constitute an example of ‘counter-law’, a situation where ‘legal powers are used to circumvent legal standards and principles’ (2018, p. 283). Broad and expansive pre-emptive offences criminalise as ‘terrorist actions’ a huge swathe of activities. However, only a fraction of individuals who could be charged with pre-emptive terrorism offences are in fact subject to prosecution. The judicious exercise of prosecutorial discretion is identified as an important constraint on the use of the legislation, counteracting the definitional breadth of offences (Anderson, 2012). Reliance on the exercise of executive discretion to limit prosecutions to those who are the ‘real terrorists’ constitutes an admission that the definitions are too broad and remain problematic from the viewpoint of the rule of law. A review of recent cases suggests that executive or judicial discretion has not limited prosecutions but has in fact extended the scope of some offences beyond the original legislative goal (Cornford, 2020). Using pre-emptive offences to target those who are preparing a future terrorist attack is difficult as the evidence used to prove that the accused harbours a general terrorist intent can be indistinguishable from individuals who nurture peculiar, but completely lawful and harmless, obsessions and fantasies (Cornford, 2020). Furthermore, if the basis on which the accused is selected for prosecution is not found in their breach of the law alone, then the actual reason for their prosecution is never put forward to be challenged in the adversarial criminal justice system and thus remains beyond the reach of the courts (Edwards, 2010). The potential for violation of human rights from the prosecution of pre-emptive offences is blunted, but not entirely negated, by the obligation on courts to interpret legislation in a way that is compatible with human rights (Kelly, 2017).33 Human rights analysis therefore extends beyond the 408

Countering radicalisation while expanding the criminal law

violation of the rights of the accused to the chilling impact on freedoms of expression, association, and religion of individuals who are members of groups and communities that are most likely to come under suspicion as they seek to avoid lawful conduct that could be mistakenly construed as evidence of a general terrorist intention (Cornford, 2020). In this way, the diffuse impact of such offences is to circumscribe the activities and beliefs of minority groups and communities that fear being the focus of counter-terrorism investigations. Critical race theorists draw attention to the role of pre-emptive offences in constructing and sustaining a moderate/extremist binary. They draw attention to the ways in which counter-terrorism policies and practices intensify and extend existing processes of the racialisation of Muslims (Garner & Selod, 2015; Smith, 2018; Cainkar & Selod, 2018; Abu-Bakare, 2020). The notion of the radical Islamist resonates as it is inscribed onto powerful existing Orientalist tropes and stereotypes of Muslim men and Islam as innately violent (Said, 1978). Such tropes persist, reconfigured in claims that the ‘new terrorism’ of Al-Qaeda and ISIS, driven by ‘religious fanaticism’ decoupled from wider geo-political conflicts, differs qualitatively from other forms of terrorism, and therefore justifies the expanded criminalisation of pre-emptive offences. In the absence of an intention to commit a specific terrorist plot or plan, pre-emptive offences rely on the accused’s commitment to extremist beliefs as evidence of ‘incipient violence’ so that ‘being Muslim becomes evidence of planning to do something’ (Sentas, 2014). In the Farooqi case, for example, two years of undercover policing investigations found no evidence of a specific terrorist plot or attack, but they revealed the defendant’s belief in ‘violent jihad’ and attempts to persuade others that fighting against British forces in Afghanistan was a religious obligation. As the police commented after the men were found guilty, ‘we did not recover any blueprint, attack plan or endgame for these men. However, what we were able to prove was their ideology’ (BBC News, 2011).

The overlap of Prevent Duty and pre-emptive offences Although the dominant narrative in the policy discourse draws a sharp distinction between the criminal and non- or pre-criminal spaces in which the Prevent Duty is supposed to operate, there are fissures in the official documentation that hint towards the complexity rising from the broad range of conduct viewed as constituting terrorism. For example, the Channel Duty Guidance states that association with or support for a proscribed organisation is a criminal offence, but nevertheless goes on to suggest that those on the ‘periphery of proscribed organisations’ may be referred to Prevent for support instead of facing prosecution (HM Government, 2020, p. 26). The recent expansion of the pre-emptive offences to include the act of viewing online extremist material and expressing any opinion or belief that is supportive of a proscribed organisation further restricts the area free from the threat of criminal sanctions. Legal scholars, aware of the expansive reach of pre-emptive offences, have begun to identify and examine the overlap between these offences and Prevent. Simon (2015) details the significant overlap between Channel indicators and pre-emptive offences, noting that the list of indicators used in the Channel programme to assess vulnerability to radicalisation includes the possession of material or symbols associated with an extremist cause, condoning or supporting violence towards others, and even plotting or conspiring with others. Such indicators overlap with pre-emptive offences, and so the individual is already a terrorist to the extent that they are engaging in actions deemed to be acts of terrorism. The challenge of locating the non-criminal criminal space in which Prevent is argued to operate is evident in the case studies of Prevent referrals and in documented examples noted in interviews with practitioners. In fieldwork on the implementation of the Prevent Duty in education, Busher 409

Tufyal Choudhury

et al. (2017) report a case in which students made a video with friends, claiming to be members of ISIS and offering a (fictitious) number that viewers can use to contact ISIS. The practitioner dealing with the case recognised that the video was a joke, although some further action was taken in order to make it clear to the students that this was not a matter that they could joke about. In fact, the students may have committed a terrorist offence, as ‘professing membership’ in a proscribed organisation is an act of terrorism, irrespective of whether the person is actually a member of the group. The overlap between Prevent and pre-emptive offences is deleterious for both prevention and prosecution. Challenging extremist ideas is a central theme of the Prevent strategy. When the introduction of the Prevent Duty raised concerns that it would diminish freedom of expression, the UK government insisted that the Duty ‘doesn’t and shouldn’t stop schools from discussing controversial issues’ (Williams, 2015). Yet the non-criminal space for such conversations has closed significantly. The criminalisation of indirect encouragement already made this uncertain terrain difficult for practitioners to navigate; the criminalisation of any opinion supportive of a proscribed organisation further erodes the space for meaningful dialogue. For example, in discussions of Israel and Palestine, educationalists must now be alert to any expression of opinions supportive of Hamas or Hezbollah, as both are proscribed organisations. Even speaking in favour of restoring a distinction between their military units and their wider organisations or movements may constitute a terrorist act. Relying on the discretion that organisations subject to the Prevent Duty retain on whether they refer cases to the police must be read in line with the legal duty to report any act of terrorism, where the indicator giving rise to the concern of potential radicalisation is also a potential terrorist act. The overlap between indicators used for Prevent assessments and conduct criminalised by preemptive offences also undermines claims that participation in Channel is entirely voluntary. The indicators that lead to the assessment that an individual is offered support may also provide the basis for criminal prosecution. The threat of possible prosecution, even if unspoken, may create undue pressure to participate in the programme (Hardy, 2017). Furthermore, the open and honest disclosure by a Channel mentee of their beliefs and online activities is constrained if these disclosures also reveal criminal conduct. Where this happens, it can trigger a police investigation, making it difficult to sustain the claim that Prevent will not lead to surveillance or investigation by the police (Simon, 2015). The threat to the rule of law arising from the wide discretion conferred on the executive branch of government in deciding whether to pursue criminal prosecutions, offer support through the Channel intervention programme, or take no action has promoted reflection by jurists on the principles that should constrain state action. Ashworth and Zedner (2014) have posited 25 principles of preventive justice to limit the use of coercive preventive measures. For example, liability for current acts, on the basis of what a person ‘may do at some time in the future’, should only occur ‘if the person has declared an intent to do those further acts’; or, a person should only be held liable for the future acts of others if those actions are sufficiently foreseeable or there is ‘sufficient normative involvement in those acts’ through encouragement, assistance, or facilitation’. Hardy (2017), building on this approach, identifies principles that can be applied to programmes countering radicalisation.

Conclusion Counter-terrorism strategies approach counter-radicalisation through both Prevent policies and the use of prosecution powers. Over time, the UK Prevent policy shifted its focus from wider initiatives addressing community capacity building and barriers to integration, towards a greater emphasis on challenging ideology and identifying individuals at risk of radicalisation for participation in the 410

Countering radicalisation while expanding the criminal law

Channel. This culminated in the Prevent Duty. The acceptance and integration of Prevent rested in its framing as a safeguarding measure that operates in a non-criminal space, diverting people away from entanglement with the criminal justice system. At the same time, since 2001, the reach of the pre-emptive terrorism offences, building on a broad definition of terrorism, has expanded and extended so that many of the actions and activities criminalised through these offences captures conduct that is used to assess suitability for participation in the counter-radicalisation Channel programme. The overlap between prevention and prosecution and its implications remain unacknowledged in the policy discourse, which continues to imagine a bright and largely clear boundary between being and becoming a terrorist. Academic disciplinary boundaries and silos may have enabled the unnoticed expansion and overlap between the two policies. The impact of Prevent has been the domain of social scientists, often building on the insights from theories of governmentality and bringing much-needed attention to the disciplinary impact of surveillance, suspicion, and responsibilisation. Not surprisingly perhaps, the intricate and detailed legal analysis needed to appreciate the expansive reach of terrorism offences has been limited to legal scholars, particularly criminal law specialists. Legal scholarship has, until recently, left the Prevent policy unexamined, reflecting both the absence of an explicit statutory basis for Prevent and, compared to criminalisation, the direct threat from the state’s coercive force. The examination of Prevent by preventive justice scholars reveals the dangers to rule of law principles from this expansion. However, an effective examination of the impacts of counter-terrorism measures on fundamental rights cannot just be a disembodied analysis of legal principles; it also needs to engage with and understand the lived experience of affected individuals, groups, and communities. This requires greater engagement across disciplinary boundaries between legal scholars and social scientists.

Notes 1 The term ‘pre-crime’ is used by legal scholars to refer to the expansion of the criminal law to actions that do not normally come within its scope; these are also referred to as pre-emptive offences. By contrast, social scientists examining counter-radicalisation policies use the term ‘pre-crime’ space to identify activities that lie just outside the reach of criminal law. This can also be referred to as the ‘non-criminal’ space; however, the notion of ‘pre-crime’ makes clear that these actions may attract state attention as they are seen as precursors to future criminal actions. Where possible, this chapter uses the terms pre-emptive offences rather than pre-crime offences, and non-criminal space rather than pre-criminal space. But where ‘pre-crime’ is used, the sense in which it is used should be clear from the context. 2 UN SC Res 1373 (28 September 2001); UN SC Res. 1566 (8 October 2004); UN SC Res 1624 (14 September 2005); UN SC Res. 2178 (14 September 2014). 3 Council of Europe, ‘Counter-Terrorism Strategy (2018–2022)’ CM (2018) 86-addfinal 4 Council Framework Decision 2002/475/JHA of 13 June 2002 on combating terrorism [2002] OJ L164/3, amended by Council Framework Decision 2008/919/JHA of 28 November 2008 amending Framework Decision 2002/475/JHA on combating terrorism [2008] OJ L330/21. 5 The offences are set out in Arts.3(1)(a)-(i). 6 DCT Art.5 7 DCT Art.6 8 DCT Art.7–8 9 DCT Art.9 10 DCT Art.10 11 DCT Art.11 12 Organic Law 2/2015 modified measures on terrorism offences in Organic Law 10/1995 13 According to Security Minister Wallace, The UK is leading the world in preventing people being drawn into terrorism and many countries are starting to copy our approach. See www.theguardian.com/ uk-news/2016/sep/05/prevent-strategy-has-more-pros-than-cons

411

Tufyal Choudhury 1 4 Counter-Terrorism and Security Act 2015 (CTSA 2015) 15 This shift was first noted in the 2009 Contest Strategy. See HM Government (2009, p. 13). 16 CTSA 2015, s26. The Prevent Duty does not apply to Northern Ireland, CTSA 2015, s51(1) 17 R v Campbell (1990) 93 Cr App R 350 18 TA 2000, s40 19 TA 2000, s38(b) 20 TA 2000 s1(1) 21 TA 2000 s1(2) 22 TA 2000 s1(5) 23 R v Gul [2013] UKSC 64 para.38 24 TA 2000, s57 25 R v Sultan Mohammed and Aabid Hussain Khan [2009] EWCA Crim 2653, cited in Walker (2011, p. 215) 26 Terrorism Act 2006, s5 (TA 2006) 27 TA 2006, ss1–2 28 TA 2000, s3 29 Attorney General’s Reference (No 4 of 2002) [2004] UKHL 43 30 TA 2000 s12(1). The offence also covers arranging or speaking at a meeting. 31 TA 2000 s12(1A), as amended by Counter-Terrorism and Security Act 2019, s1. 32 TA 2000 s1(5) 33 Human Rights Act 1998, s3

References Abbas, M. S. (2019). “I grew a beard and my dad flipped out!” Co-option of British Muslim parents in countering “extremism” within their families in Bradford and Leeds. Journal of Ethnic and Migration Studies, 45(9), 1458–1476. Abu-Bakare, A. (2020). Counterterrorism and race. International Politics Review, 8, 79–99. https://doi. org/10.1057/s41312-020-00074-x Ali, N. (2020). Seeing and unseeing prevents racialized borders. Security Dialogue, 51(6), 579–596. Anderson, D. (2012). The terrorism acts in 2011 – report of the independent reviewer on the operation of the terrorism act 2000 and part 1 of the terrorism act 2006. www.gov.uk/government/publications/ the-terrorism-acts-in-2011-report-of-the-independent-reviewer Anderson, D. (2013). Shielding the compass: How to fight terrorism without defeating the law. http:// dx.doi.org/10.2139/ssrn.2292950; https://terrorismlegislationreviewer.independent.gov.uk/wp-content/ uploads/2013/04/shielding-the-compass1.pdf Ashworth, A., & Zedner, L. (Eds.). (2014). Preventative justice. Oxford University Press. BBC News. (2011, September 9). Munir Farooqi given four life sentences for terror charges. BBC News. www.bbc.co.uk/news/uk-england-manchester-14851811 Beck, U. (1992). The risk society towards a new modernity. Sage. Brown, C., & Wolf, M. (2013, April 29). Rights laws to be overhauled as Blair says “the game has changed”. The Independent. www.independent.co.uk/news/uk/politics/rights-laws-to-be-overhauled-as-blair-saysthe-game-has-changed-304031.html Busher, J., Choudhury, T., & Thomas, P. (2019). The enactment of the counter-terrorism “prevent duty” in British schools and colleges: Beyond reluctant accommodation or straightforward policy acceptance. Critical Studies on Terrorism, 12(3), 440–462. Busher, J., Choudhury, T., Thomas, P., & Harris, G. (2017). What the prevent duty means for schools and colleges in England: An analysis of educationalists’ experiences. Centre for Trust, Peace and Social Relations, Coventry University. https://pureportal.coventry.ac.uk/en/publications/what-the-prevent-dutymeans-for-schools-and-colleges-in-england-a Cainkar, L., & Selod, S. (2018). Review of race scholarship and the war on terror. Sociology of Race and Ethnicity, 4(2), 165–177. Carlile, A., & MacDonald, S. (2014). The criminalisation of terrorists’ online preparatory acts. In T. Chen, L. Jarvis, & S Macdonald (Eds.), Cyberterrorism: Understanding, assessment and response (pp. 155–173). Springer.

412

Countering radicalisation while expanding the criminal law Commissioner for Human Rights. (2018, December 4). Misuse of anti-terror legislation threatens freedom of expression. www.coe.int/en/web/commissioner/-/misuse-of-anti-terror-legislation-threatens-freedomof-expression Coolsaet, R. (2005). Radicalisation and Europe’s counter-terrorism strategy: The transatlantic dialogue on terrorism. CSIS/Clingendael. www.egmontinstitute.be/content/uploads/2014/01/CSIS_081205_Rev. pdf?type=pdf Coppock, V., & McGovern, M. (2014). “Dangerous minds”? Deconstructing counter-terrorism discourse, radicalisation and the “psychological vulnerability” of Muslim children and young people in Britain. Children and Society, 28, 242–256. Cornford, A. (2020). Terrorist precursor offences: Evaluating the law in practice. Criminal Law Review, 8, 663–685. Cornford, A., & Petzsche, A. (2020). Terrorism offences. In K. Ambos, A. Duff, J. Roberts, & T. Weigend (Eds.), Core concepts in criminal law and criminal justice (Vol. 1, pp. 172–210). Cambridge University Press. Edwards, J. (2010). Justice denied: The criminal law and the ouster of the courts. Oxford Journal of Legal Studies, 30(4), 725–748. Edwards, P. (2018). Counter-terrorism and counter-law: An archetypal critique. Legal Studies, 38(2), 279–297. Ericson, R., & Haggerty, K. (1997). Policing the risk society. Oxford University Press. Feeley, M., & Simon, J. (1994). Actuarial justice: The emerging new criminal law. In D. Nelken (Ed.), The futures of criminology (pp. 173–195). Sage. Fernandez, S. (2018). The geographies of prevent: The transformation of the Muslim home into a pre-crime space. Journal of Muslim in Europe, 7(2), 167–189. Garner, S., & Selod, S. (2015). The racialization of Muslims: Empirical studies of Islamophobia. Critical Sociology, 41(1), 9–19. Goldberg, D., Jadhav, S., & Younis, T. (2017). Prevent: What is pre-criminal space? BJPsych Bulletin, 41(4), 208–211. Hall, J. (2020). The terrorism acts in 2018 – the report of the independent reviewer of terrorism legislation on the operation of the terrorism act 2000 and 2006. https://terrorismlegislationreviewer.independent.gov.uk/ wp-content/uploads/2020/03/Terrorism-Acts-in-2018-Report.pdf Hardy, K. (2017). Preventative justice principles for countering violent extremism. In T. Tulich, R.Ananian-Welsh, S. Bonnitt, & S. Murphy (Eds.), Regulating preventive justice: Principle, policy and paradox. Routledge. Hayes, B., & Kundnani, A. (2018). The globalisation of countering violent extremism policies: Undermining human rights, instrumentalising civil society. The Transnational Institute. Heath-Kelly, C. (2012). Counter-terrorism and the counterfactual: Producing the “radicalisation” discourse and the UK prevent strategy. The British Journal of Politics & International Relations, 15(3), 394–415. Heath-Kelly, C. (2017). The geography of pre-criminal space: Epidemiological imaginations of radicalisation risk in the UK prevent strategy, 2007–2017. Critical Studies on Terrorism, 10(2), 297–319. Heath-Kelly, C., & Strausz, E. (2017). Counter-terrorism in the NHS: Evaluating prevent duty safeguarding in the NHS. University of Warwick. HM Government. (2009). Pursue prevent protect prepare – the United Kingdom’s strategy for countering international terrorism. The Stationary Office. HM Government. (2011). The prevent strategy (Cm 8092). The Stationary Office. HM Government. (2015). National security strategy and strategic defence and security review 2015: A secure and prosperous United Kingdom. The Stationary Office. HM Government. (2020). Channel duty guidance: Protecting people vulnerable to being drawn into terrorism. The Stationary Office. Home Office. (2018). Counter-terrorism border and security bill – European convention on human rights memorandum by the home office. https://assets.publishing.service.gov.uk/government/uploads/system/ uploads/attachment_data/file/713837/20180601_ECHR_Memorandum.pdf Home Office. (2021). Prevent handbook for elected members. www.hounslow.gov.uk/downloads/file/3077/ prevent_-_elected_members_handbook Houry, N. (2018, May 30). France’s creeping terrorism laws: Restricting free speech. Just Security. www. justsecurity.org/57118/frances-creeping-terrorism-laws-restricting-free-speech/ Jerome, L., Busher, J., Armstrong, M. A., Choudhury, T., da Silva, R., Elwick, A., Fontana, G., James, N., Lewis, J., Robson, J., Svennevig, H., & Thomas, P. (2020). Conclusion: Reflections on the first five years of

413

Tufyal Choudhury the prevent duty in early years, primary, secondary and further education. In J. Busher & L. Jerome (Eds.), The prevent duty in education: Impact, enactment and implications (pp. 159–170). Palgrave Macmillan. Joint Committee on Human Rights. (2018). Legislative scrutiny: Counter-terrorism and boarder security bill (9th Report of Session 2017–19, HC 1208, HL Paper 167). The House of Commons. Kelly, R. (2017). The right to a fair trial and the problem of pre-inchoate offences. European Human Rights Law Review, 6, 596–602. Martin, T. (2014). Governing an unknowable future: The politics of Britain’s prevent policy. Critical Studies on Terrorism, 7(1), 62–78. https://doi.org/10.1080/17539153.2014.881200 McCulloch, J., & Pickering, S. (2010). Counter terrorism: The law and policing of pre-emption. In N. McGarrity, A. Lynch, & G. Williams (Eds.), Counter-terrorism and beyond: The culture of law and justice after 9/11 (pp. 13–29). Routledge. Murphy, C. (2012). EU counter-terrorism law: Pre-emption and the rule of law. Hart Publishing. Pettinger, T. (2020). CTS and normativity: The essentials of preemptive counter-terrorism interventions. Critical Studies on Terrorism, 13(1), 118–141. Qureshi, A. (2016). The science of pre-crime: The secret “radicalisation” study underpinning prevent. Cage Advocacy. Rights Watch UK. (2016). Preventing education? Human rights and UK counter-terrorism policy in schools. www.rightsandsecurity.org/assets/downloads/preventing-education-final-to-print-3.compressed-1_.pdf Rimo, A. (2021). Is prevention better than cure? The ever-increasing criminalisation of acts preparatory to an offence in Spain. International Journal for Crime, Justice and Social Democracy, 10(1), 1–15. Rizq, R. (2017). “Pre-crime”, prevent, and practices of exceptionalism: Psychotherapy and the new norm in the NHS. Psychodynamic Practice, 23(4), 336–356. Said, E. (1978). Orientalism. Pantheon Books. Sageman, M. (2021). The implication of terrorism’s extremely low base rate. Terrorism and Political Violence, 33(2), 302–311. https://doi.org/10.1080/09546553.2021.1880226 Sentas, V. (2014). Traces of terror: Counter-terrorism law, policing, and race. Oxford University Press. Simon, J. (2015). Preventative terrorism offences [DPhil, University of Oxford]. Smith, C. (2018). Race and the logic of radicalisation under neoliberalism. Journal of Sociology, 54(1), 92–107. Steiker, C. (1998). Forward: The limits of the preventive state. Journal of Criminal Law and Criminology, 88(3), 771–808. Thomas, P. (2015). UK prevent and community cohesion: Conflict or synergy? In C. Baker-Beall, C. HeathKelly, & L. Jarvis (Eds.), Counter-radicalisation: Critical perspectives (pp. 36–53). Routledge. Tulich, T., Murray, S., & Skead, N. (2021). Antipodean perspectives on preventive justice: The high court and serious crime prevention orders. Griffith Law Review. https://doi.org/10.1080/10383441.2021.1925411 UN OHCHR. (2018, May  23). Preliminary findings of the visit: UN special rapporteur on the promotion and protection of human rights and fundamental freedoms while countering terrorism concludes visit to France. www.ohchr.org/EN/NewsEvents/Pages/DisplayNews.aspx?NewsID=23128&LangID=E Verfaille, K., De Kimpe, S., & Cools, M. (2019). Counter-terrorism policing and the prevention paradox. In N. Clycq, C. Timmerman, D. Vanheule, R. van Caudenberg, & S. Ravn (Eds.), Radicalisation: A marginal phenomena or a mirror to society? (pp. 193–213). Leuven University Press. Walker, C. (2011). Terrorism and the law. Oxford University Press. Walker, C., & Cawley, O. (2020). The juridification of the UK’s counter terrorism prevent policy. Studies in Conflict & Terrorism. https://doi.org/10.1080/1057610X.2020.1727098 Williams, R. (2015, June 9). School heads raise alarm over new duty to protect students from extremism. The Guardian. www.theguardian.com/education/2015/jun/09/schools-duty-police-extremism-anti-terrorism-laws Younis, T. (2021). Counter-radicalization, public health, and racism: A case analysis of prevent. In K. Bhui & D Bhugra (Eds.), Terrorism, violent radicalisation and metal health (pp. 203–206). Oxford University Press. Zedner, L. (2007). Pre-crime and post-criminology? Theoretical Criminology, 11(2), 261–281. Zedner, L. (2014). Terrorizing criminal law. Criminal Law and Philosophy, 8, 99–121. Zedner, L. (2021). Countering terrorism or criminalizing curiosity? The troubled history of UK responses to right-wing and other extremism. Common Law World Review, 50(1), 57–75.

414

27 RESPONDING TO RADICALISATION THROUGH EDUCATION Stijn Sieckelinck and William Stephens

In October 2020, the French teacher in history and civic education, Samuel Paty, was assassinated outside his school. The perpetrator was an 18-year-old Russian of Chechen descent. He had learned on social media about this teacher’s alleged habit of showing derogatory cartoons of Prophet Muhammad in class. Therefore, the act appears to have been motivated by political-religious hatred and was characterised as violent extremism. The incident sent shock waves through schools, teachers, and pupils around the globe. In the aftermath of the events in France, all over the European continent, incidents were reported where pupils and parents disputed with and even threatened teachers for their use of the cartoons in class. The killing of Mr. Paty was a recent example of ideologically motivated schoolrelated violence. For decades, schools and non-formal youth educational sites have made headlines as loci of shooting (Pfeifer, 2017), hostage taking (Beslan, Russia), and killing sprees (Utøya, Norway), and are unfortunately globally considered soft or easy targets for extremist violence (UNESCO, 2017). Yet, these incidents are rare, whereas many places of education are easy targets in a different way. Over the years, a whole branch of social strategies aimed at prevention of violent extremism (PVE) has emerged (see also Augestad; Ragazzi & Walmsey, both in this volume). As a result, schools and youth work have also been expected to take part in the efforts to prevent and counter terrorism. Hence, the threat of extremism in broader society has rendered schools and youth work targets for policies aimed at preventing or countering extremism. Governments across the world want to use education to prevent extremism, whether it is religiously motivated, left- or right-wing-oriented, or COVID-related anti-science activism. Since it started developing counterradicalisation courses for schools, UNESCO has described radicalism as a learned behaviour. This characterisation implies that it can be ‘unlearned’. While it is broadly agreed that education can play an important role in addressing the challenges surrounding extremism, there are deep concerns about how places of education are called upon to do this. Too often, it is argued, preventing extremism in education is approached from a security paradigm (Waever, 1995), with the central aim being to identify threat. This paradigm can undermine broader educational goals and pedagogical relationships and lacks a clear educational core (Glaser, 2017; O’Donnell, 2016; Sieckelinck et al., 2015; Stephens & Sieckelinck, 2020). Still today, for many people, PVE-E entails the obligation/expectation of teachers to report any potentially radicalised/radicalising individuals. This security lens raises the question, which lies at the heart of this chapter: is it possible for the prevention of extremism to find a pedagogical core? 415

DOI: 10.4324/9781003035848-31

Stijn Sieckelinck and William Stephens

On various levels, from local governments to transnational institutions, programmes have been developed for schools that can be used to deal with extremism. At the time that this branch of preventing violent extremism in education (PVE-E) became operational in many countries, scientific findings about its benefits were still limited, and assumptions were fraught with many complexities. Notwithstanding the inconclusiveness of the academic debate, the importance of education in prevention strategies became widely recognised in policy and practitioner circles. Around 2013, while many interventions were still in the pilot phase and conceptual debates about the key terminology of radicalisation were far from concluded, governments took decisive measures around PVE-E, and for many schools and neighbourhoods, doing nothing in the face of a huge escalation or extremist threat was not an option as the safety of their pupils was at stake. Consequently, a multitude of approaches, methods, and interventions were developed and tried out. However, in general, schools were not given time to reflect on the assumptions of their responses to political and religious violence. Since then, all over the globe, new technologies have been high on teachers’ and educators’ lists of concerns (Valkenburg & Piotrowski, 2017). Teachers generally feel they lack information, instruments, and control to address what their pupils are experiencing online (ibid.). For all the affordances of new technology, teachers are faced with a new and constantly changing arena of engagement and exchange, which they can quickly feel incapable of keeping up with, let alone addressing (Katz & El Asam, 2021). These rapidly developing technologies mean children and youth are easily exposed to roughly the same information as adults, albeit packed in often less nuanced forms (Katz & El Asam, 2021). It is through these technologies that societies’ most polarising issues can generate the greatest heat, with the potential to drive young people to more extreme corners (Valkenburg & Piotrowski, 2017). A non-exhaustive list: the regulations regarding religious practice and symbols in public space (headscarves, ritual slaughter, the ‘burkini issue’), racism and the pain of colonialism and white privilege, political agendas related to gender and sexism, the dilemmas regarding freedom of expression. These are all major themes that, in a polarised society, usually in a raw, unmediated, and often online form, penetrate students’ hearts and minds and can trigger a chain reaction of intense emotions (Katz & El Asam, 2021). It has even been argued that in this hypermediated era, adolescence itself, as a bridge between childhood and adulthood, is on the brink of collapse (Koops & Zuckerman, 2003). Considering this context, it is perhaps unsurprising that some young people are already exploring the more radical spectrum. Even more so when this attraction can be understood in light of their identity development (see further). Many authors from different fields of interest note, based on various studies, that an important difficulty with discourses of radicalisation and extremism is that they have been too closely linked with counterterrorism rather than education or pedagogy (Davies, 2018; Glaser, 2017; O’Donnell, 2016; Sieckelinck et al., 2015). A significant body of literature thematises the tensions of securitised educational efforts in the context of preventing violent extremism. The underlying logics of educational and security paradigms overlap but are sometimes diametrically opposed: idealistic educators are committed to independent critical thinking and the questioning of taken-for-granted assumptions, while security actors want to gather intelligence and detect possible deviants (Halstead & Pike, 2006; O’Donnell, 2016; Sieckelinck et al., 2015). Interpreting ideas and discussions in classrooms through security-driven frameworks can undermine pedagogical efforts. Most of this literature reflects on the UK’s Prevent strategy or the specific dilemmas of educators or youth workers. But also more broadly, Christodoulou (2020) observes how the securitisation of education can have ethical, pedagogical, social, and political repercussions. Some have warned of the ‘instrumentalization of civil society’ (Kundani & Hayes, 2018) and its co-optation in security 416

Responding to radicalisation through education

paradigms. Therefore, we argue that schools can only contribute to the agenda of tackling extremism if radicalisation is considered a pedagogical challenge, that is, if the tackling of radicalisation is organised through a pedagogical lens in formal and non-formal educational settings. After nearly 20 years of educational responses to extremism, it is about time to take stock. What are the main arguments for paying attention to extremism in education? What are the pitfalls when this ambition is pursued? Educational thinkers have discussed and underlined the value of educating against extremism at length (Davies, 2008; Sieckelinck et al., 2015). However, the political expectation that education should be able to defeat extremism raises doubts (O’Donnell, 2016). Not in the least because some of these politics may have a radicalising effect themselves (Jackson et al., 2011; Sieckelinck, 2017). Moreover, PVE-E risks instrumentalising the authentic educational relationship (Biesta, 2014) and stigmatising members of marginalised groups (O’Donnell, 2016). All these considerations raise another question: Can/ought it be prevented in schools at all? Educational actions start from diverse and even opposing assumptions, target different groups, have a range of foci, and aim for different outcomes. Though most PVE-E interventions are seldom properly tested or evaluated, many schools and teachers across Europe seem to have tackled these challenges with increasing confidence. The experiences of these educators can help identify and address problems and challenges and can also offer guidelines for other educators (Nordbruch & Sieckelinck, 2018). We will examine more closely the precise role education is supposed to play and what a genuinely pedagogical/educational approach to PVE could look like. It will be argued that schools can only contribute to this agenda if radicalisation is considered a pedagogical challenge and the tackling of radicalisation is organised through a developmental lens with an eye for individual and societal development, materialising in formal and non-formal educational programmes. Hence, the importance of resetting the pedagogical parameters of PVE-E as a truly educative response cannot be predicated on fear.

Preventing radicalisation through education: a contested concept in a complex field of practice Before outlining PVE-E policies, however, we must first contextualise the concept of radicalisation in the sphere of education. Most people regard radicalisation as a process in which an individual or group increasingly embraces extreme political, social, or religious ideals that threaten to undermine the established or social-democratic order. Major attention has been given to jihadi extremism as a radical evil (Schuurman, 2019). More recently, xenophobia and white supremacy have gained traction among young people (Krall, 2021). Further, environmental or animal rights may lead people onto an extremist path, while others target abortion clinics or expose virologists, spurred by conspiracy theories on the coronavirus (Nogrady, 2021). The banner of radicalisation conceals many divergent practices. It is safe to say, though, that when someone radicalises, they become increasingly resentful and hateful towards a section of society and anyone who defends the status quo or the elite. However, it is far from clear what process is exactly being suggested by the idea of radicalisation. Empirical and theoretical research into radicalisation processes has shown that there are many different pathways towards extremism and violent extremism, many of which do not start with ideological affinity/attraction (Bertelsen, 2018) and do not even necessarily lead to extremist attitudes or an extremist mindset (Roy, 2014); young people also join extremist groups without becoming ideological extremists (Horgan, 2008). Radicalisation is the result of a multitude of political-societal conditions, troubled relationships, and behavioural antecedents (de Ruyter & Sieckelinck, 2023). 417

Stijn Sieckelinck and William Stephens

On a more fundamental level, the notion of radicalisation has been regarded as the master signifier of the late ‘war on terror’ (Kundani, 2014, p. 3), providing a ‘vehicle for policy-makers’ who required a discourse through which they could discuss the supposed ‘causes of terrorism’ while at the same time delimiting the boundaries of what it was permissible to include in any explanatory framework (Heath-Kelly, 2013). Namely, giving primary attention to individual and ideological factors to the exclusion of structural and state-related factors. Most authors agree that radicalisation is a process, be it a staircase, pyramid, or pathway, but apart from this, there is little consensus on when radicalisation is at play, why it occurs, and how it can be prevented (de Graaf, 2011; Fadil et al., 2019). In relation to young people in particular, the prevention of radicalisation has been characterised as ‘a solution in search of a problem’ (Ragazzi & Walmsley, 2020). That is to say, there are pressures from above and below to ‘do something’ about radicalisation while this challenge is numerically marginal in comparison to the usual, regular issues faced by schools. Additionally, some public policies on radicalisation have undermined the confidence of many educators in their ability to respond to traditional pedagogic challenges when these are reframed in terms of ‘radicalisation’. As Ragazzi and Walmsley (2020) put it, ‘the discourse of radicalisation generates in part its own reality’ (p. 14). These critiques, as we will argue below, underline the importance of studying and approaching the prevention/countering of extremism through an educational lens rather than the other way around. A  pedagogical perspective should allow us to treat what is often defined as a security concern as a developmental concern, as understood in positive youth development. One issue this brings to the fore is that of definition. As Marsden (2020) argues, different actors and communities of practice have different definitional needs: ‘it is useful to recognize the differing priorities, scope and real-world implications of each community’s engagement with the definitional question’. A pedagogical approach to preventing extremism requires a pedagogical definition of the issue at hand. To this end, a definition was constructed in an attempt to be more attentive to the meaning experienced by actors in an educational environment (youth and teachers alike) and puts the pedagogical relationship at the heart of the concept: ‘when a child or adolescent starts to develop strong political or religious ideas and agency that are so fundamentally at odds with the educational environment or mainstream expectations that the pedagogical or educational relationship is increasingly put at stake’ (Sieckelinck, 2017). This can be contrasted with a more securityoriented definition, such as ‘The process by which people come to support terrorism and violent extremism and, in some cases, then join terrorist groups’ (UK Home Office, 2011). While this definition is potentially valuable for security actors, it cannot provide a sound basis for an educational response. It is not unthinkable that the use of the concept ‘radicalization’, with all its discursive flaws (Heath-Kelly, 2013) and potential biases (Kundani, 2012), may in some cases have been useful to thoughtful social and educational professionals such as teachers and youth workers in better addressing their concerns about some of their students or clients. The radicalisation frame enabled some social professionals to become more attentive to a process of moral disengagement at a stage in which they were often still able to make a difference. The question taken up here is, however, to what extent this is still true today.

The role of PVE-E in counter-radicalisation policies It is clear by now that since PVE-E policies first emerged, the landscapes of security, intelligence, and education have started to shift. This shift determines, to a large extent, as we will see, the educational space to work on countering extremism (Sieckelinck, 2017). First, we will take a brief 418

Responding to radicalisation through education

look at some prevailing education policy initiatives on a transnational, national, and local level in order to trace how the role of education has been framed. Education has often been included as one of the central tools for preventing extremism within a general strategy for the prevention of extremism. That is, education is adopted as one of the forces within the control of the government that can be drawn upon to achieve the overall goal of prevention. We define PVE-E policy, then, as the ‘set of actions taken by a government’ (Walker, 2000, p. 13) to utilise education as a tool for prevention. Reaching a set of actions involves a process of problem definition, and as such, the framing of the problem is inherent to and shapes the policy (Walker, 2000). The role of education in prevention has been established in the European context through a number of EU-wide policy initiatives (O’Donnell et al., 2021a). The inclusion of education as a pillar of prevention was evident in the establishment of the Radicalisation Awareness Network (RAN), which drew frontline workers, including teachers and youth workers, into a network of practitioners addressing the prevention of radicalisation. Between 2015 and 2020, a specific working group focused on education as a preventive instrument. In 2017, the European Commission zoned in specifically on the role of youth work in ‘The Contribution of Youth Work to Preventing Marginalisation and Violent Radicalisation’ (European Commission, Directorate-General for Education, Youth, Sport, & Culture, 2017). Through these initiatives, an array of educational materials and practices have emerged, with their influence spreading across the EU through the RAN. In a review of this process, it was concluded that ‘over the last few years, an impressive corpus of educational methods and tools has been developed by educators and researchers across different fields .  .  . promising an evidence-based responses to various dimensions of radicalisation’ (Nordbruch & Sieckelinck, 2018, p. 23). Yet teachers and school officials generally see room for improvement and often voice concern over a lack of resources and opportunities to implement these strategies in a sustainable and wide-reaching way (Radicalisation Awareness Network, 2016). This central role given to education was consolidated when the High-Level Commission Expert Group on Radicalisation (HLCEG-R) identified education ‘as a cornerstone for effective prevention of radicalisation’ and framed the role of education as ‘strengthening resilience against radicalisation and recruitment’, pointing specifically to the importance of ‘fostering social inclusion, promoting common democratic values and managing controversial issues with open discussions in safe classrooms’ (European Commission, 2018, p. 12). Unsurprisingly, this attention to the role of education also finds expression in national strategies, many of which similarly frame the role of education in terms of building resilience and promoting democratic values (Stephens & Sieckelinck, 2020). The specific role of education is also shaped by differing conceptualisations of the exact nature of the problem, as captured by the differing emphasis in definitions of radicalisation evident in different national policies (Hardy, 2018). It appears that the actual government sector/ministry (or combined sectors of government) that are involved in P/CVE depends on the concerns that country has within the area of P/CVE, as well as perceptions and realities of risk and threat there. For example, the UK Prevent policy, with its primary emphasis on the role of ideology in radicalisation (Hardy, 2018), focused in 2015 on the role of education in building resilience by ‘promoting fundamental British values and enabling these to challenge extremist views’ (Department of Education, 2015, p. 5). Given its different context, Germany’s PVE-E approach has evolved from a primary focus on right-wing extremism and places emphasis on promoting democracy among young people (Glaser & Soyler, 2021). In the Netherlands, preventing radicalisation became a distinct and prominent element of the national policy agenda. In 2005, a government-wide approach to prevention was introduced, calling not only for government but also for society to be involved in ‘preserving and creating an open, 419

Stijn Sieckelinck and William Stephens

tolerant, and peaceful society’ (Sieckelinck & Stephens, 2021). As a result, local action plans were drafted in which educational/pedagogical organisations such as schools, family support centres, and youth work services were mentioned as critical partners in keeping local communities safe. In 2015, schools became legally obligated to institute social safety policies, including the prevention of radicalisation. This work in schools is supported by the ‘School and Safety’ organisation, which serves as a national platform and knowledge base. The role of educational professionals towards radicalisation is given shape largely in three related areas: first, as having a responsibility to be alert to and identify signs of radicalisation; second, as creating a safe environment for the discussion of difficult topics to diffuse their power; and third, as fostering citizenship and promoting democratic values (Sieckelinck & Stephens, 2021). Preventive interventions are generally called for at the local level in areas deemed ‘at risk’. These interventions involve strengthening social infrastructure and local networks while applying the ‘personal approach’, which involves creating an alternative for young people at risk, such as a job, school, or sport (Sieckelinck & Stephens, 2021; van Heelsum & Vermeulen, 2017). In the Belgian city of Mechelen, for example, a policy programme called ‘Positive Identity and Society’ (PiM) became operational, in which social alienation is seen as the core problem underlying ideological radicalisation (Vergani et al., 2021; Van Leuven, 2023). In this programme, radicalisation is seen as a process of negative identity development, and this insight leads to practices that transcend the security domain. The programme looks for important key figures around youth and supports those people to help the young person find his or her place. They coach and empower social workers, schools, families, and other organisations, such as martial arts clubs, to prevent alienation. PiM coaches reinforce or build up the social safety net. In doing so, they support both the young people and the institutions. In this way, they aim to reverse the process of social alienation and focus on a positive identity in the young person. When asked if PiM effectively prevents extremism, a youth worker involved in the programme is ‘not sure whether we help prevent attacks, but we do help to transcend the dividing lines that terror creates’. By embedding a social-educational mindset in the municipal system and maintaining balanced contacts with the police and services, Mechelen works on an integral inclusive policy that helps tackle polarisation and radicalisation. In sum, education has been positioned at all levels as one of the key means through which the government can shape actions in order to prevent extremism, positioned within a wider apparatus of prevention measures. The specificities of PVE-E policy are shaped by national and local differences in problem framing; however, broadly, education is positioned as a prime means for ‘building resilience’ and promoting and developing democratic attitudes and competencies. This framing of the role of education raises important questions about the instrumentalisation of education for prevention in a manner that can undermine broader educational goals (O’Donnell, 2016; Stephens & Sieckelinck, 2020) and, as we will unpack further, may fail to realise the true potential of a pedagogical, developmental response to the issues underlying and surrounding extremism. First, however, it is instructive to turn attention to the range of practices that have arisen, some of which are direct outcomes of government action and others that have emerged from grassroots initiatives, albeit within the prevailing PVE-E milieu.

PVE-E in practice: mapping the heterogeneous field As a result of the contradictory national and international dynamics of counter-radicalisation in the education sector, the range of practices is broad. Ragazzi and Walmsley (2021) divide practices into two categories: (1) projects aimed broadly at raising awareness around key societal issues such as nationalism, racism, and discrimination; and (2) projects aimed at dealing with individual cases that are 420

Responding to radicalisation through education

deemed to need special attention. The first category of awareness-raising projects is generally in line with the principles and philosophy of education for democratic citizenship (EDC) and human rights education (HRE), and the core values of the Reference Framework of Competences for Democratic Culture (RFCDC).3 The authors conclude that: ‘such projects generally emanate from NGOs that are themselves at the core of the conversation on citizenship education. The second category of projects, which they refer to as “casework based”, poses a different set of ethical and political questions’. If we zoom in from the policies to the practices on the ground, we find, not surprisingly, a myriad of educational practices reflecting different points of entry. Davies (2018) has given an overview of what educational institutions are doing in different countries in a sustained way to tackle violent extremism and whether there is any evidence of the impact of these activities. In her synopsis, identifying the entry points for a wide range of programmes or projects enables some analysis of the different directions taken and what is realistic and effective in conventional educational settings. Davies justifiably categorises four main approaches based on the topic of engagement: Islamism, extreme right, Islamophobia, dialogue. However, there are many more, very different angles from which to approach PVE-E. A great variety can be found in the central principles of different practices, including democracy (RAN ‘labs for democracy’), integral complexity (Cambridge University), peace building (many initiatives, mostly located in African contexts), inclusiveness and empowerment (UNESCO), informing pupils about (the history of) terrorism (Ter Info), and debunking conspiracies (Think, UK). Moreover, one cannot but take note of the vast scope of issues addressed: some programmes focus on strengthening personal identity; other programmes focus on equipping youth with life skills; some programmes invest in strengthening the local city identity; others in inculcating so-called national values. Some programmes teach human rights, while others aim to tackle online threats. It is an understatement to say that the term ‘counter-radicalisation’ covers several types of practices (Ragazzi & Walmsley, 2020). Despite this wide diversity of practices, many essentially focus on protection or detection rather than alternative mobilisation or transformation. Young people are, for example, shielded from harmful content, warned against delusional ideologies, or cautioned about the illegality of spreading propaganda. These programmes are mainly found in disadvantaged areas and neighbourhoods where victimisation is a real issue. Other programmes take a largely cognitive angle, aimed at informing youth about, for example, the mechanisms of polarisation or recruitment. Or the importance of knowing facts from fiction about terrorist attacks. But also: what do pupils know about the constitutional-democratic systems of their countries? What should they know about the religion of the ‘other’? A subcategory is formed by programmes around more deliberative epistemological questions: how to build logical, reasonable arguments. Fewer programmes work on the affective or emotional dimension of radicalisation: how do certain ideas make students feel? (de Ruyter & Sieckelinck, 2023; O’Donnell, 2016). What politics trigger social pain? (Eisenberger & Lieberman, 2004). The same is true for more mobilising programmes, relating to youth’s ideals and agency, in which youth are empowered to turn their ideas into political or creative action. Of high importance from a pedagogical point of view is the role of the teacher. In relation to radicalisation many educators fall back on a disciplining role: ‘You better not, . . . or you will feel the consequences’. Or a more subtle version reminiscent of Plato’s (prob. 369 BC) warning: You are young, my son, and, as the years go by, time will change and even reverse many of your present opinions. Refrain therefore a while from setting yourself up as a judge of the highest matters. 421

Stijn Sieckelinck and William Stephens

The problem is that many discipliners lack the informal authority necessary to be seen as credible messengers (Samuel, 2020). Other educators present themselves more as allies; they offer a mental and/or physical refuge where young people can find temporary shelter against threats or hardships. And then there are educators who try to learn from the success of extremist movements and copy some of their strategies to be active recruiters for a more inclusive democratic way of living (Niconchuk et al., 2018). Next to these variables, one can discern different kinds of activities: one-on-one mentoring projects, group activities (recommended to educate young people about peer pressure and other group dynamics), and socio-cultural or artistic activities (Sieckelinck & Kaulingfreks, 2021). In addition to the many offline programmes, there has been an increase in online programmes in which all these different approaches are transferred into the online world of youth (Jayakumar, 2023). Another typology can be made on the basis of time and age: apart from the incidental and rather absurd pre-school initiative against radicalisation (Belga, 2017), most programmes target pupils in secondary school and students in higher education. However, some call for more attention to be given to primary schools, not addressing extremism as such but, for example, the ability to deal with conflict (Macaluso, 2016). Because education is ‘paramount to shape values and behaviour and to favour identity formation’, Macaluso advises shifting the focus of such preventive policies from secondary to primary education. And indeed, there may be compelling reasons to pay attention at younger ages. Not because 12-year-olds might become terrorists overnight. But because in vulnerable situations, young people from this age already start consuming unchecked or fearmongering political and religious messages, hear the same conspiracy theories, and watch the same propaganda videos as their 16-year-old siblings (Howard et al., 2021). For many parents and educators, attending to how these messages impact youth’s ideas, emotions, and feelings of belonging is very demanding. It is fair to say that this last decade has witnessed the emergence of educational practices framed in terms of their potential to prevent radicalisation. Some of these, such as teaching how to debunk conspiracies or engage in discussion on issues around terrorism, are explicitly and directly developed for this purpose. Others, such as education in democratic citizenship, can be seen as broader educational projects that are reframed in terms of their potential for preventing radicalisation. These face the same challenge as most PVE activities, with the extent to which they actually contribute to preventing extremism being notoriously hard to demonstrate (Gielen, 2020). Yet, as educational practices they have another test to live up to: the extent to which they are coherent with the broader educational goals being pursued. It is simply not enough to assess an educational practice on its ability to prevent extremism alone (Stephens & Sieckelinck, 2020).

An educational/pedagogical perspective to (preventing) radicalisation Theoretical debates on the (questionable) value of education in countering radicalisation have long drawn from empirically based theories on the process of radicalisation on the one hand (Borum et  al., 2003; Dalgaard-Nielsen, 2008; Moghaddam, 2005) and philosophical and sociological examinations on the other hand (Jackson et al., 2011). Only recently has a more pedagogical line of investigation been initiated towards a more comprehensive pedagogical or educational account of what extremism is and what schools and youth work might do to pre-empt and counter it. This educational account requires drawing not only on theories of radicalisation but also taking seriously broader educational theories. In this section, we make a modest attempt to advance a line of thinking in this regard by both drawing out the educational implications of radicalisation theories and drawing on pedagogical theory to identify features of a truly educational response to extremism. 422

Responding to radicalisation through education

From an educational perspective, the influential social psychological work on the role of the quest for significance among extremists raises important considerations. In ‘The Three Pillars of Radicalization: Needs, Narratives, and Networks’ Kruglanski et al. (2019) propose that radicalisation is enabled by the coming together of three factors (‘three Ns’): the need for significance – to matter and have respect, a narrative that identifies extreme violence as a means of achieving significance, and a social network that reinforces, validates, and rewards this narrative justification for the use of extreme violence. The educational significance of this theory of radicalisation is twofold: on the one hand, sites of education can contribute to the conditions in which young people experience a loss of significance, and on the other, we can view the role that education can play in fulfilling this quest for significance. A loss of significance, it is proposed, can be brought about by a sense of failure or an affront to one’s social identity. Schooling can easily become a site of failure; indeed, it is inherent in some modes of educational organisation, and schooling can certainly be a setting in which dominant narratives and values can lead to experiences of exclusion and marginalisation of various social identities (e.g. Keddie, 2014; Valenzuela, 2010). This is to say, flawed education may contribute to feelings of insignificance, frustration, and exclusion. Yet education in assisting pupils to become at home in the world (Arendt, 1994), has the possibility of creating the conditions in which significance, narrative, and connection arise. Education can bestow meaning and respect when talents are nurtured, difference is held with regard, and space for the exploration of social identities is provided. Without going further at this point, what becomes clear is that an educational response to extremism cannot be reduced to developing skills to deconstruct narratives or becoming committed to ‘national values’. An educational response in its fullest sense involves tending to this legitimate need for significance and, at the very least, being attentive to those situations that can occasion a loss of significance. Where social psychologist Kruglanski and his colleagues offer an intelligible theory of why people end up in extremist milieus, philosopher Quassim Cassam (2021) has made an effort to dissect the concept of extremism in a way that might be highly relevant for the debate on the role of education. He argues that we need to go beyond focusing on what and even how extremists may believe, and look rather at the ‘cast of mind’ that characterises extremism, which includes particular preoccupations, attitudes, thinking styles, and emotions. Cassam refers to this as the ‘extremist mindset’. This mindset involves a preoccupation with victimhood and purity. Extremists are extremely focused on the belief that they are victims of oppression or lack of recognition (primarily in society) and, at the same time (maybe in response to this felt victimisation or humiliation), believe they are superior with their striving for moral, ideological, or racial purity. Such preoccupations are unlikely to be unfamiliar to teachers and youth workers. Similarly, many teachers have come across moments where attitudes characteristic of this mindset are at work: unwillingness to compromise, rejection of pluralism, intolerance, and indifference to the consequences of their actions for ‘the others’. The third characteristic of the extremist mindset is an extremist thinking style, characterised by utopian thinking or conspiratorialism. This problematic thinking style is not limited to students and youth, but its increase certainly impacts their behaviour. Finally, extremists have typical emotions. Cassam primarily mentions negative emotions such as anger, resentment, and self-pity. He believes that these emotions must be interconnected with the extremists’ preoccupations because, in a different context, anger may also be justified and resentment may be an appropriate emotion to have against others. Crucially, for an educational perspective on extremism, Cassam carefully disentangles extremism from radicalism. While extremism is always reprehensible, radicalism is not. A radical may seem to evince some of the characteristics of an extremist mindset, particularly the unwillingness to compromise. Although for an educator the unwillingness of a student to compromise may be 423

Stijn Sieckelinck and William Stephens

considered quite concerning, when we consider, for example, the emancipation of women, we are surely thankful that there was a refusal to compromise on the assertion that women must have equal rights to men. The key consideration here is that while education may play a role in averting the development of an extremist mindset, we likely don’t want education to stamp out radicalism. Indeed, the whole thrust of an important pedagogical movement – critical pedagogy developed by Paulo Freire – is directed towards the development of some form of radicality. Cassam’s broadening of the psychology of extremism from what and how people believe to the notion of a mindset, along with this careful distinguishing of extremism from radicalism is potentially productive as inspiration for an educational response to extremism. The account of the extremist mindset is comprehensive and gives ample attention to the oft-neglected affective dimension of extremism (de Ruyter & Sieckelinck, 2023). Yet, a note of caution is in place: many educators, with good reason, are cautious of labelling, particularly if labelling implies some fixed reality unamenable to educational intervention. It would clearly be a mistake to quickly view students as having an extremist mindset. While there may be a small few who develop a fullblown extremist mindset (O’Donnell et al., 2021b) and require specific interventions, the broader educational significance of Cassam’s theory lies in directing attention to attitudes, thinking styles, preoccupations, and emotions. We can continue to lay the theoretical foundations of a distinctly educational perspective by broadening this examination of theories of extremism to consider pedagogical theories. The work of Paulo Freire has clear relevance for the important distinction Cassam makes between radicalism and extremism. The preface to his seminal work, Pedagogy of the Oppressed (1970), makes a pivotal distinction between the radical and the ‘sectarian’: On one side, there is radicalisation. Radicalisation, nourished by a critical spirit, is always creative . . . [it] criticizes and thereby liberates. On the other side, there is sectarian- ism. Sectarianism, fed by fanaticism, is always castrating . . . [it] mythicizes and thereby alienates (p. 37). The pedagogy of the oppressed . . . is a task for radicals; it cannot be carried out by sectarians. (p. 39) What is significant here is that the educational journey is not one of acquiring passivity; rather, it locates educational significance in enabling young people to engage with the world and find their ability to make an impact on the world. Further, from a critical pedagogical point of view, it is no use trying to convince students that their political or religious ideas are mistaken. By trying to deposit her own ideas, a teacher will be less successful than by engaging in dialogue (Nieto Ángel et al., 2020). From this perspective, awareness programmes should go hand in hand with exercises and activities that help students acquire voices and skills for action. This strategy is oriented towards active peacebuilding rather than pacifying conflicts. In Freire’s terms, to build on a future in which my actions matter. A fourth strand of theory that lends itself to developing an educational response to extremism can be found within a body of work around history education. In educating students around extremism, teachers are dealing with value-laden facts in a world without a clear epistemological centre. Most claims that build students’ worldviews are rather true, or rather false. The approach to historical events in which different narratives or ‘versions of the same truth’ are recognised is called ‘multiperspectivity’ and has, with the advent of the information society (Castells, 2007), gained a more important place in history education (Nordgren & Johansson, 2015). This should not lead to a relativism in which all perspectives are considered of equal merit, but students should 424

Responding to radicalisation through education

also be able to assess different perspectives (Seixas, 2017; Stoel et al., 2017). Grever (2012) argues that multiperspectivity is important because recognising different perspectives increases awareness that there are (have been) other people with different views and promotes self-reflection. It also helps identify and articulate different perspectives, deepens knowledge and insights, and provides insight into the complexity of reality. In relation to the shift of a relatively static and hierarchic society into our current information society (Castells, 2007) and the difficulty of telling truth from falsehood in an increasingly complex society, the educational debate can benefit from the work on ‘multiperspectivity’. With former hierarchies of knowledge questioned or even overthrown, teachers will increase the quality of their courses by allowing for multiple stories and interpretations about the same events. In work by Wansink et al. (2018), teachers’ approaches to multiperspectivity in lessons on three topics varying in moral sensitivity (i.e. the Dutch Revolt, Slavery, and the Holocaust) were examined. Four categories of considerations for or against introducing specific subjects’ perspectives were found: functional, moral, pedagogical, and practical. However, teachers engaged in ‘normative balancing’, meaning that not all perspectives were perceived as equally valid or politically desirable, showing where multiperspectivity ends. This rough overview of relevant theories for an educational response to extremism is far from comprehensive, but it is a modest attempt at drawing together these various threads. In the works of Kruglanski and Cassam, we find a clear broadening of the role of education relevant to extremism. It cannot be concerned with a narrow focus on disrupting engagement with extreme ideologies but must address the need for significance and the various preoccupations and emotions that extreme narratives can fuel and harness. In the works on critical pedagogy and multi-perspectivity, we find insights into the kinds of educational experiences that can foster a broader and more flexible mindset while also retaining the educational significance of the kind of radicalism in young people that refuses to be content with the perpetuation of injustice.

Resetting education against extremism This final section will discuss how, in education, dealing with (the onset of) radicalisation requires a pedagogical point of view. As Beelmann and Jonkman (2021) have argued: [p]lacing the problem of radicalization and extremism in a development-oriented perspective broadens the perspective on the problem, its origin and its approach. A positive effect here is that an approach as part of a preventive youth policy is less stigmatizing. Moreover, for signals of radicalization and extremism as well as for underlying factors found, the outcome does not obviously funnel towards one specific problem, but rather keeps the broad development of young people in mind and relates it to social, relational and individual factors. More than its focus on prevention suggests, this developmental perspective displaces ‘threat’ as the organising principle and instead puts ‘growth’ at the centre of its attention. Remember how Freire (1970) characterised education of the oppressor as ‘essentially an act that hinders the intellectual growth of students by turning them into, figuratively speaking, comatose “receptors” and “collectors” of information that have no real connection to their lives’. In this view, Freire claims that by assuming the roles of teachers as depositors and students as receptors, humans are made into objects (Micheletti, 2010). Humans as objects have no options for development or growth. When we don’t regard humans as objects, then, developmentally, we can view the potential radicality 425

Stijn Sieckelinck and William Stephens

of youth as an expression of the human urge to act upon the world. Pedagogically, then, we don’t approach this urge with fear or anger but with a desire to explore, understand, and, if possible, give shape to it. This would be grounded in the view that being young is not merely a transitional period of life to be ‘gotten through’, but that the youth are an important segment of society with a distinct role to play in social change (Wyn & Woodman, 2006). This radicality in a youth might be part of his or her coming of age and can reasonably be expected to alter or disappear with (young) adulthood. As Kruglanski states in Greenbaum (2018): The need for significance has to be satisfied in socially constructive ways. So, first of all, it would behoove governments not to create loss of significance by promoting prejudice and inequality against certain groups. The second recommendation is to emphasize narratives that are generally accepted, compelling and delivered by charismatic communicators who promote these constructive ways of gaining significance. . . . And finally, governments should create movements in which people will be accepted, revered and respected for counterviolence activities. The common security angle of preventing risk in education runs counter to what education is basically about (Biesta, 2014). The educational question is also an existential question. Getting into an open conversation about one’s ideas and feelings is precisely what education is about. The occurrence of radicalisation from a pedagogical point of view is an educational opportunity to teach about diversity, democracy, the rule of law, freedom of speech, and freedom of religion. According to Davies, CVE/PVE is more successful ‘when a programme is non-prescriptive, not moralising, but leads to independent thinking and reflection on ethical dilemmas and concerns; when learners are listened to’ (Davies, 2019). Young people may demonstrate an understandable preoccupation with victimhood (Cassam, 2021). Rather than dismissing these concerns, the more pupils’ realities (of victimisation) are acknowledged (not necessarily solved!), the less they will be grabbed by fantasies of persecution. For example, talking to young people about the threat of terrorism will have little impact unless their own feelings or lack of protection in their lives are addressed. Especially in disadvantaged neighbourhoods, families have a higher chance of being terrorised by tax agencies or fearing expulsion. Hence, effective educational PVE programmes do not only warn against false claims and delusional ideologies. They also address the underlying grievances of students (who may have good reasons to not trust the authorities’ interference with what they think). Davies (2018) formulates another compelling advice: CVE/PVE work has in fact to be wary of the notion of an ‘intervention’, with the implication of a one-off attempt at interrupting a negative pathway towards radicalisation and should be more about building a permanent culture in schools where resilience to extremism is just one aspect of a fuller learning of rights, history, religious and ethnic conflict, and community dynamics. (p. 49) In other words, prevention of extremism, rather than incidental interventions or specific programmes (although they can be helpful), requires sustained attention to the quality of the ethos of the school. As an alternative to the narrow security approach, schools may become places where all students can: discover a sense of purpose; share and explore their ideals and their identities without fear of ridicule; come to realise that the school cannot accept everything that they may value; and through this, learn to take into account the wide array of interests and values held by other students and in society at large (de Ruyter & Sieckelinck, 2013). 426

Responding to radicalisation through education

Having said this, it is important to understand that education is not just about schools. This is especially important as schools report widely that they already have ‘too much on their plate’ to prioritise PVE-E. This burden was only exacerbated by the COVID crisis. Whenever schools feel strained or pressured by requests to help counter extremism, many well-intended programmes to support them in this task will remain ineffective. To compensate for this overburdening of the school programme, it is important to take a broader view of education. To shift focus away from formal education to other domains. In most societies, educational activities are not limited to schools. Apart from formal schooling, one can think of organising informal educational support in the neighbourhood. And, mostly underexamined, many European countries have a non-formal educational work force that specialises in organised youth work/in service learning/running social movements with a training structure (European Commission, Directorate-General for Education, Youth, Sport, & Culture, 2017). Youth workers who understand the lives of children and youth ‘from the inside’ can act as benign recruiters for a better future. They are the living proof that the best (primary) prevention against violent radicalisation is often not labelled as such: ‘while the term “prevent” is understandable’, Davies’ review confirms that ‘PVE has to focus less on what students should not become and more on what they actively become’. Future research on the role of education against extremism takes the critique on securitisation seriously and builds on the proposed ‘re-pedagogization’ of radicalisation studies: if the study of (the prevention of) radicalisation is organised through a pedagogical lens, in formal and non-formal educational settings, teachers, school staff, youth workers, and citizens will contribute more to the developing knowledge base, feel more respected in their big challenges, and acquire more skills to make a real difference in the lives of their pupils and youth. It also brings to the fore the need to address in more depth support for the educator: a pedagogical response cannot be taught through the demonstration of techniques or the provision of materials alone. It calls us to ask, what forms of professional development render educators feeling equipped to make sound pedagogical choices when seeking to address issues relevant to extremism in their practice? And lest we forget, radicality is commonly linked to fear and anger, but it may also announce hope. Crises of the environment, of public health, of democratic culture, and of humanity as a whole spread across the globe and require some radical responses that will no longer sustain the current political and financial status quo. Activism and idealism are both characterised by extreme preoccupations. So, although we share Cassam’s concern about the extremist mindset’s black-and-white worldview (lack of compromise), teachers should also leave some space for the possibility of radical change as an outcome of a learning process. Initiatives such as School strike for the climate or Black Lives Matter were only possible through a non-compromising stance. Teachers can trust that skilfully carving out space for this dialogue will undermine the growth of an extremist mindset. Seen in this light, PVE-E actions are no goal in themselves but one of the many means to help students learn and fulfil their human potential.

References Arendt, H. (1994). Understanding and politics (the difficulties of understanding). In J. Kohn (Ed.), Essays in understanding 1930–1954 (pp. 203–327). Brace and Company. Beelmann, A., & Jonkman, H. (2021). Preventie van radicalisering en extremisme: Een kennisbestand voor het Hoger Onderwijs. Verwey-Jonker Instituut. Belga. (2017, August 2021). Zelfs radicalisering bij kleuters. De Standaard. Bertelsen, P. (2018). The fight against violent extremism: The Aarhus model. In J. Kargel (Ed.), They have no plan B: Radicalization, departure, return – between prevention and intervention (pp. 154–169). Bundeszentrale für Politische Bildung.

427

Stijn Sieckelinck and William Stephens Biesta, G. (2014). The beautiful risk of education. Taylor & Francis. Borum, R., Bartel, P., & Forth, A. (2003). Manual for the structured assessment violence risk in youth (SAVRY) version 1. Louis de la Parte Florida Mental Health Institute, University of South Florida. Cassam, Q. (2021). Extremism: A philosophical analysis. Routledge. Castells, M. (2007). Communication, power and counter-power in the network society. International Journal of Communication, 1. Christodoulou, E. (2020). “Boosting resilience” and “safeguarding youngsters at risk”: Critically examining the European Commission’s educational responses to radicalization and violent extremism. London Review of Education, 18(1), 18–34. https://doi.org/10.18546/LRE.18.1.02 Dalgaard-Nielsen, A. (2008). Studying violent radicalization in Europe I: The potential contribution of social movement theory (No. 2). CDR. Davies, L. (2008). Educating against extremism. Trentham Books. Davies, L. (2018). Review of educational initiative in counter-extremism internationally: What works? The Segerstedt Institute. Davies, L. (2019). Ex post paper: Education and radicalisation prevention: Different ways governments can support schools and teachers in preventing/countering violent extremism. Sage. de Graaf, B. (2011). Evaluating counterterrorism performance: A comparative study. Routledge. Department of Education. (2015). The prevent duty: Departmental advice for schools and childcare providers. Department of Education. https://assets.publishing.service.gov.uk/government/uploads/system/ uploads/attachment_data/file/439598/prevent-duty-departmental-advice-v6.pdf de Ruyter, D. J., & Sieckelinck, S. (2023). Creating caring and just democratic schools to prevent extremism. Educational Theory, 73(3), 413–433. Eisenberger, N. I., & Lieberman, M. D. (2004). Why rejection hurts: A common neural alarm system for physical and social pain. Trends in Cognitive Sciences, 8(7), 294–300. https://doi.org/10.1016/j.tics.2004.05.010 European Commission. (2018). High-Level commission expert group on radicalisation (HLCEG-R) for the European Commission: Final report. Publications Office of the European Union. European Commission, Directorate-General for Education, Youth, Sport, & Culture. (2017). The contribution of youth work to preventing marginalisation and violent radicalisation: A practical toolbox for youth workers & recommendations for policy makers: Results of the expert group set up under the European Union work plan for youth for 2016–2018. Publications Office. Fadil, N., Ragazzi, F., & de Koning, M. (2019). Radicalization in Belgium and the Netherlands: Critical perspectives on violence and security. Bloomsbury Publishing. Freire, P. (1970). Pedagogy of the oppressed. Herder and Herder. Gielen, A.-J. (2020). Cutting through complexity: Evaluating countering violent extremism (CVE). University of Amsterdam. Glaser, M. (2017). Extremist, militant, radicalised? In M. Glaser, F. Greuel, M. Herding, S. Hohnstein, & J. Langner (Eds.), Young and radical: Political violence during adolescence. Deutsches Jugendinstitut. Glaser, M., & Soyler, S. (2021). EDURAD mapping and research report – Germany. http://edurad.eu/wpcontent/uploads/2021/03/Germany-Country-Report.pdf Greenbaum, Z. (2018). 5 questions for Arie W. Kruglanski: The social psychologist explains the psychology of violent extremism and how governments can counteract it. Monitor on Psychology, 50(4). www.apa. org/monitor/2019/04/conversation-kruglanski Grever, M. (2012). Dilemmas of common and plural history. Reflections on history education and heritage in a globalizing world. In M. Carretero, M. Asensio, & M. Rodriguez-Moneo (Eds.), History education and the construction of national identities (pp. 75–91). Information Age Publishing. Halstead, M., & Pike, M. (2006). Citizenship and moral education: Values in action (1st ed.). Routledge. Hardy, K. (2018). Comparing theories of radicalisation with countering violent extremism policy. Journal for Deradicalization, 15, 76–110. Heath-Kelly, C. (2013). Counter-terrorism and the counterfactual: Producing the “radicalisation” discourse and the UK prevent strategy. British Journal of Politics and International Relations, 15(3), 394–415. https://doi.org/10.1111/j.1467-856X.2011.00489.x Horgan, J. (2008). From profiles to pathways and roots to routes: Perspectives from psychology on radicalization into terrorism. The Annals of the American Academy of Political and Social Science, 618(1), 80–94. https://doi.org/10.1177/0002716208317539

428

Responding to radicalisation through education Howard, P., Neudert, L.-M., & Prakash, N. (2021). Digital misinformation/disinformation and children. UNICEF. Jackson, R., Jarvis, L., Gunning, J., & Breen-Smyth, M. (2011). Terrorism: A critical introduction. Palgrave Macmillan. Jayakumar, S. (Ed.). (2023). Diversion: Perspectives on PVE from around the globe. Palgrave Macmillan. Katz,A.,  & ElAsam,A. (2021). Refuge and risk: Life online for vulnerable young people. www.internetmatters.org/ wp-content/uploads/2021/01/Internet-Matters-Refuge-And-Risk-Report.pdf Keddie, A. (2014). The politics of Britishness: Multiculturalism, schooling and social cohesion. British Educational Research Journal, 40(3), 539–554. https://doi.org/10.1002/berj.3101 Koops, W., & Zuckerman, M. (2003). Introduction: A historical developmental approach to adolescence. The History of the Family, 8(3), 345–354. https://doi.org/10.1016/S1081-602X(03)00041-1 Krall, H. (2021). Youth cultures, right-wing extremism and violence. In G. Knapp & H. Krall (Eds.), Youth cultures in a globalised world: Developments, analyses and perspectives. Springer. Kruglanski, A. W., Bélanger, J. J., & Gunaratna, R. (2019). The three pillars of radicalization: Needs, narratives, and networks. Oxford University Press. Kundani, A. (2012). Blind spot? Security narratives and far-right violence in Europe. International Centre for Counter-Terrorism. Kundani, A. (2014). Radicalisation: The journey of a concept. In C. Heath-Kelly & L. Jarvis (Eds.), Counterradicalisation: Critical perspectives (pp. 14–35). Taylor and Francis. Kundani, A., & Hayes, B. (2018). The globalisation of countering violent extremism policies. Transnational Institute. Macaluso, A. (2016). From countering to preventing radicalization through education: Limits and opportunities. Institute for Global Justice. Marsden, S. (2020). In response – defining terrorism: Communities and context. Contemporary Voices: St Andrews Journal of International Relations, 2(1), 1–88. https://doi.org/10.15664/jtr.1601 Micheletti, G. (2010). Re-envisioning Paulo Freire’s “banking concept of education”. Inquiries Journal/Student Pulse, 2(2). www.inquiriesjournal.com/a?id=171 Moghaddam, F. M. (2005). The staircase to terrorism: A psychological exploration. American Psychologist, 60, 161–169. https://doi.org/10.1037/0003-066X.60.2.161 Niconchuk, M., Dietrich, K., Greiner, K., Wood, D., & Couturier, G. (2018). Two sides of the same coin? www.equalaccess.org/wp-content/uploads/EAI-YouthPower-Report-1-FINAL.pdf Nieto Ángel, M. C., Maciel Vahl, M., & Farrell, B. (2020). Critical pedagogy, dialogue and tolerance: A learning to disagree framework. In S. L. Macrine (Ed.), Critical pedagogy in uncertain times: Hope and possibilities (pp. 139–158). Springer International Publishing. Nogrady, B. (2021). “I hope you die”: How the COVID pandemic unleashed attacks on scientists. Nature, 598. https://media.nature.com/original/magazine-assets/d41586-021-02741-x/d41586-021-02741-x.pdf Nordbruch, G., & Sieckelinck, S. (2018). Transforming schools into “labs for democracy”: A companion to preventing violent radicalisation through education. RAN. Nordgren, K., & Johansson, M. (2015). Intercultural historical learning: A conceptual framework. Journal of Curriculum Studies, 47(1), 1–25. https://doi.org/10.1080/00220272.2014.956795 O’Donnell, A. (2016). Securitisation, counterterrorism and the silencing of dissent: The educational implications of prevent. British Journal of Educational Studies, 64(1), 53–76. https://doi.org/10.1080/00071005 .2105.1121201 O’Donnell, A., Hamilton, C., Malone, A., & Melaugh, B. (2021a). EDURAD mapping and research report – Europe. http://edurad.eu/wp-content/uploads/2021/03/European-Report.pdf O’Donnell, A., Hamilton, C., Malone, A., & Melaugh, B. (2021b). EDURAD: Addressing violent radicalisation: A multi-actor response through education – transnational report. http://edurad.eu/wp-content/ uploads/2021/03/Transnational-Report.pdf Pfeifer, B. (2017). School shootings: Existential concerns and implicit religion. Vrije Universiteit Amsterdam. https://research.vu.nl/en/publications/school-shootings-existential-concerns-and-implicitreligion Plato. (prob. 369 BC). Dialogues, Theatus. Radicalisation Awareness Network. (2016). Empowering and supporting teachers: Pedagogical role requires time and training. RAN.

429

Stijn Sieckelinck and William Stephens Ragazzi, F., & Walmsley, J. (2020). The European Union and “foreign terrorist fighters”. In D. Bigo, T. Diez, E. Fanoulis, B. Rosamond, & Y. Stivachtis (Eds.), The Routledge handbook of critical European studies (1st ed.). Routledge. Ragazzi, F., & Walmsley, J. (2021). Counter-radicalisation in the classroom. Council of Europe. Roy, O. (2014). Holy ignorance: When religion and culture part ways. Columbia University Press. Samuel, T. (2020). At the crossroads: Rethinking the role of education in preventing and countering violent extremism. In A. P. Schmid (Ed.), Handbook of terrorism prevention and preparedness. ICCT Press. Schuurman, B. (2019). Topics in terrorism research: Reviewing trends and gaps, 2007–2016. Critical Studies on Terrorism, 12(3), 463–480. https://doi.org/10.1080/17539153.2019.1579777 Seixas, P. (2017). A model of historical thinking. Educational Philosophy and Theory, 49(6), 593–605. https:// doi.org/10.1080/00131857.2015.1101363 Sieckelinck, S. (2017). Reradicaliseren: Rondsolen voor een betere wereld. Lannoo Campus. Sieckelinck, S., & Kaulingfreks, F. (2021). Speelruimte voor identiteit. Amsterdam University Press. Sieckelinck, S., Kaulingfreks, F., & de Winter, M. (2015). Neither villains nor victims: Towards an educational perspective on radicalisation. British Journal of Educational Studies, 63(3), 329–343. https://doi. org/10.1080/00071005.2015.1076566 Sieckelinck, S., & Stephens, W. (2021). EDURAD mapping and research report – The Netherlands. http:// edurad.eu/wp-content/uploads/2021/03/The-Netherlands-Country-Report.pdf Stephens, W., & Sieckelinck, S. (2020). Being resilient to radicalisation in PVE policy: A critical examination. Critical Studies on Terrorism, 1–24. https://doi.org/10.1080/17539153.2019.1658415 Stoel, G., Logtenberg, A., Wansink, B., Huijgen, T., van Boxtel, C., & van Drie, J. (2017). Measuring epistemological beliefs in history education: An exploration of naïve and nuanced beliefs. International Journal of Educational Research, 83, 120–134. https://doi.org/10.1016/j.ijer.2017.03.003 UK Home Office. (2011). CONTEST: The United Kingdom’s strategy for countering terrorism (Cm 8123). The Stationery Office. UNESCO. (2017). Preventing violent extremism through education: A guide for policy-makers. UNESCO. https://unesdoc.unesco.org/ark:/48223/pf0000247764 Valenzuela, A. (2010). Subtractive schooling: US-Mexican youth and the politics of caring. SUNY Press. Valkenburg, P. T., & Piotrowski, J. T. (2017). Plugged in: How media attract and affect youth. Yale University Press. van Heelsum, A., & Vermeulen, F. (2017). Cities’ policies: The work of European cities to counter Muslim radicalisation. Journal of International Migration and Integration, 1–19. https://doi.org/10.1007/ s12134-017-0533-1 Van Leuven, A. (2023). Evidence-inspired Prevention of Violent Extremism in the Belgian City of Mechelen: Evaluation and Recommendations. In S. Jayakumar (Ed.): Perspectives on Countering Extremism Diversion and Disengagement. London: Bloomsbury. Vergani, M., Iqbal, M., O’Brien, K., Lentini, P., & Barton, G. (2021). Examining the relationship between alienation and radicalization into violent extremism. In S. Bonino & R. Ricucci (Eds.), Islam and security in the West (pp. 115–138). Springer International Publishing. Waever, O. (1995). Securitization and desecuritization. In R. Lipschutz (Ed.), On security. Columbia University Press. Walker, W. (2000). Policy analysis: A systematic approach to supporting policymaking in the public sector. Journal of Multi-Criteria Decision Analysis, 9(1), 11–27. https://doi.org/10.1002/10991360(200001/05)9:1/3  3.0.CO;2-3 Wansink, B., Akkerman, S., Zuiker, I., & Wubbels, T. (2018). Where does teaching multiperspectivity in history education begin and end? An analysis of the uses of temporality. Theory & Research in Social Education, 46(4), 495–527. https://doi.org/10.1080/00933104.2018.1480439 Wyn, J., & Woodman, D. (2006). Generation, youth and social change in Australia. Journal of Youth Studies, 9(5), 495–514. https://doi.org/10.1080/13676260600805713

430

28 COUNTER-NARRATIVES AND STRATEGIC COMMUNICATIONS, OFFLINE AND ONLINE Benjamin Lee

Funding This research was funded by the Centre for Research and Evidence on Security Threats, an independent centre commissioned by the Economic and Social Research Council (ESRC Award: ES/ N009614/1) and funded in part by the UK Security and intelligence agencies and the Home Office.

Introduction This chapter provides a short introduction to the conceptualisation, theory, and implementation of ­counter-narratives. The key theme of this chapter is that counter-narratives are not yet fully formed, leaving considerable conceptual and theoretical room around what they are and how they are supposed to work. Despite this confusing picture, counter-narrative remains firmly embedded among policy makers, and is still an important dimension of counter-extremism. Consequently, counter-narrative campaigns seem likely to remain on the research agenda for some time to come. Added to this, governmentinitiated campaigns, while drawing the bulk of interest, are not the only form of counter-narrative. Narratives that contest extremist claims are commonplace and a feature of everyday life, embedded as they are in mainstream media, communities, and relationships that we all encounter every day. Reflecting the origins of counter-narrative and their embrace by governments and NGOs, the literature in the field is scattered. Much of it is grey literature, reports, and evaluations undertaken by NGOs, while a smaller amount is peer-reviewed work in academic journals. What follows roughly traces the debates around counter-narrative as a concept before engaging with the theories underpinning counter-narrative, divided roughly between audiences, messengers, and practices. The chapter concludes by highlighting some of the research areas likely to be of interest in the future. This chapter should be treated as an overview only rather than a comprehensive account; it is a starting point for further inquiries as opposed to the entire story.

Counter-narrative, counter-message, or something else? Counter-narrative as a concept is intimately connected to the period of conflict that emerged following the September 11 attacks on the United States (Schmid, 2014, p.  9). Early efforts and

431

DOI: 10.4324/9781003035848-32

Benjamin Lee

understanding of counter-narratives are closely connected with what was talked about as the ‘battle of ideas’: the pressing need to contest the messaging being produced by Al-Qaeda, the dominant Salafist Jihadist network at the time and the principal opponent (at the outset of the so-called Global War on Terror at least). While the seed of the idea was already well-established, Glazzard (2017, p. 4) suggests that the first explicit use of the term counter-narrative emerged around the UK’s 2005 counter-terrorism strategy (CONTEST). In contrast to the strategic messaging directed abroad of by United States, the United Kingdom was more focused on reaching Muslim audiences within the United Kingdom, especially in the wake of the 7/7 bombings (Schmid, 2014, p. 13). The term was rapidly picked up by civil society groups, such as the Quilliam Foundation and the Institute for Strategic Dialogue, subsequently becoming adopted into both US and EU policy (see also Ferguson, 2016, p. 8). Conceptually, a counter-narrative was presented as a sort of enlightened counter-terrorism. The consequence of being unable to defeat an idea and ideology, Salafi Jihadism, militarily was the need to pay more attention to soft power alternatives (Jacobson, 2009, p. 12). Counter-narrative also emerged in response to the recognition that Jihadists were able to effectively recruit others by propagating their own narratives of the West’s ‘war’ on Islam (Glazzard, 2017, p. 4). Early efforts at counter-narrative embedded in US strategic communications seemed to largely fail in dispelling the persistent idea, particularly in the Muslim world, that the ‘war on terror’ was simply a war on Islam (Leuprecht et al., 2009). The earliest ‘shared values initiative’ put out by the Council of American Muslims for Understanding was a Madison Avenue-style advertising campaign that gained little traction in the Muslim world (Schmid, 2014, p. 10). More recently, counter-narrative emerged as a tool for dealing with extremism more generally, following the widespread digitisation of extremism and, in particular, a growing recognition that both Jihadists (particularly the group often referred to as Islamic State) and the extreme right had been able to use the Internet and social media to expand their reach (Warrington, 2018; Briggs & Feve, 2013). Even though the idea of a counter narrative is now firmly embedded within policy, there has been little evidence of a common conceptualisation of the term (Ferguson, 2016, p. 2). As is often the case in terrorism studies, terminology is disputed and seldom fully agreed upon. Counter-­narrative has been variously described as counter-messaging (Lee, 2019), counter-terrorism messaging (Aggarwal, 2017), counter ideological support for terrorism (Aldis & Herd, 2007), counter-speech (Bartlett & Krasodomski-Jones, 2015), the battle of ideas (Payne, 2009), alter-messaging (Pizzuto, 2013), and counter-ideology (Sedgwick, 2012). The choice of the term counter-narrative in this chapter reflects that, for most people, this is the most recognisable and accessible description for all of these activities while acknowledging that disputes exist (see below). Collectively, these activities constitute a distinctive branch of countering violent extremism (CVE) in which communication is central (Braddock, 2020, p. 58). The most readily available critique is around the term narrative itself, which is often used in the loosest sense (Schmid, 2014, p. 3). Many have noted that, strictly speaking, narratives amount to more than simply messages; they are storylines that help make sense and explain events and allow audiences to draw inferences (Leuprecht et al., 2009, p. 26; Glazzard, 2017, p. 11). Narrative reasoning specifically is important as it can supplant decision-making based on rationality (Schmid, 2014, p. 3). Narrative is distinctive from other forms of messaging and exists as a system of stories with common underlying themes (Schmid, 2014, p. 3). Drawing on literary theory, Glazzard emphasised that narrative requires not just messages and events but also an understanding of how they are presented: that they be organised as a distinct thread and be experienced and commented on by characters (2017, pp. 7–8).

432

Counter-narratives and strategic communications

Narratives can be further categorised on the basis of their content: either explicitly fact-based, attempting to discredit other messengers and messages, or attempting to promote a more positive narrative. While myth-busting and fact-checking-style approaches are often seen as less controversial, inserting alternative ideas is more controversial, particularly in the context of state-backed campaigns that run the risk of being equated to propaganda and social engineering (Corman, 2006, p. 101). What Briggs and Feve termed the counter-messaging spectrum (2013, p.  6) identified three broad programmes of counter-messaging work: • strategic communications, focused on awareness raising, highlighting what the government is doing, and correcting misinformation; • alternative narratives that set out positive narratives aiming to demonstrate ‘what we are for’ rather than against; • and finally counter-narratives that are presented as challenging extremist narratives directly by discrediting and undermining them. In truth, many of the examples of counter-narratives included in this chapter fall short of the stricter definitions of narratives, instead limiting themselves to fact-checking or pointing out hypocrisy rather than crafting more elaborate, complete stories. The following section highlights some of the available theory that has been applied to counter-narratives, while at the same time, the key message should be that there is only limited theory underpinning most counter-narrative campaigning that has gone on.

Counter-narrative theory and practice A major criticism of counter-narrative campaigns has evolved from the perception that strategies have been developed by policy-makers and civil society as opposed to academia and therefore lack a solid theoretical base (Glazzard, 2017, p. 3; Braddock, 2020, p. 63; Malet, 2021). The majority of research work on counter-narrative reflects more pragmatic origins, focusing heavily on practical guidance, best practice, and ‘what works’. In many cases, these accounts are pitched at the strategic level, making the case for counter-narrative and how it should be undertaken (Payne, 2009; Sedgwick, 2012). Several accounts are more in-depth and take the form of toolkits and how-to guides designed to guide campaigners (e.g. Ashour, 2010; Allchorn, 2020; Sim, 2013; Silverman et al., 2016). Lastly, there is a growing literature of write-ups and reports offering practical examples of how counter-narrative campaigns have developed previously; a notable example is the Radicalisation Awareness Network’s database of ‘Inspiring Practices’, which includes some examples of counter-narrative alongside a broader CVE portfolio (European Commission n.d.). This pragmatic focus does not mean, however, that a theoretical dimension is completely absent, and accounts often walk the line between abstract theory and practical guidance, sometimes offering connections to deeper ideas of persuasion and audiences. Researchers have drawn on a range of theoretical arguments to try and explain how counternarrative campaigns may work. Corman (2006) makes a helpful distinction between audiences, agents, and systems (interpreted here more as content than the social systems originally used by Corman) in their discussion of contesting Salafist ideologies. This, in turn, offers a relatively neat way to divide up the different aspects researchers have addressed, focusing variously on audiences, individuals, and the practice or content of campaigns.

433

Benjamin Lee

Audiences The earliest counter-narrative campaigns often targeted the ‘Muslim world’, seeking to counter the idea central to the narrative of Salafi Jihadists that the West was at war with Islam. As time and the ‘war on terror’ have worn on, however, thinking about audiences has steadily become more granular. More practical advice on counter-narrative often appeals to campaigners to tailor their narratives to particular audiences (Payne, 2009, p. 110; Pizzuto, 2013, p. 2). A major criticism of early counternarrative efforts from the Bush era was the focus on a single counter-narrative to contest the claims being made by Al-Qaeda and Salafi-Jihadists, overlooking the variety of audiences they were trying to reach. As understandings of radicalisation and radical milieus have evolved, so too have understandings of counter-narratives. In particular, the identification of large audiences that may be ambiguous about, or sympathetic to, violence without necessarily endorsing it or participating opened a number of audiences for counter-narrative campaigns, all of whom would require their own specific messaging (Leuprecht et al., 2009, p. 29; see also Malthaner & Waldmann, 2014; Conway, 2012). The public health model of countering violent extremism (CVE) is helpful in that it allows for the attempted categorisation of audiences according to their perceived proximity to violent extremism (Harris-Hogan, 2020, p. 99; see Hardy’s chapter in this volume). Primary interventions that are purely preventive are directed at those who are as yet unconnected to violent extremism, even where they may be a member of a broader population considered at risk (most typically the young). Examples of primary interventions would be the inoculation-based approaches to counter-narrative that seek to reach audiences before they are exposed to violent extremism, highlight the kinds of arguments they are likely to encounter, and thereby prime audiences ‘defences’ (see below). Secondary interventions are those that focus on those considered to be at risk or already on the margins of violent extremism. Even within the boundary, targeting can be difficult. Work by researchers associated with Facebook revealed that even with the level of deep insight possessed by Facebook’s marketing tools, finding those engaging with extremist content can be difficult (Saltman et al., 2021). In practice, secondary audiences may be relatively contained and difficult to reach. They are also potentially more inured to counter-narrative campaigns, steeped as they are in oppositional subcultures that provide them with the tools they need to ‘fight off’ persuasive attempts. Tertiary interventions are those that occur after full engagement with violent extremist narratives and are generally the focus of ‘exit’-style programmes designed to work with those seeking to disengage. Differing potential audiences have seldom been explicitly theorised in the literature, but they do heavily overlap with broader questions over the content and practice of counter-narratives. Different audiences are noted as likely having different opinions on what makes a message or messenger credible, which therefore can have a large impact on the potential success of a counter-narrative (Braddock & Morrison, 2020, p. 474). Counter-narrative is generally pitched at primary and secondary audiences: either those who have nothing to do with violent extremism but plausibly one day might or those who are already considered at risk in some respect, for example by actively engaging with violent extremists.

Agents The question of who should speak has a better developed theoretical basis. Schmid (2014, p. 14) describes credibility and legitimacy as some of the key ingredients in narrative. Others link the messengers to the concept of trust in counter-narratives, arguing that without a level of trust, counter-narratives are likely to fail and be ignored or deemed illegitimate by audiences. Furthermore, trust takes on many forms, ranging from the abstract (e.g. trust in systems) to the more 434

Counter-narratives and strategic communications

interpersonal (e.g. trusting relationships with others or believing someone to be reliable) (Braddock & Morrison, 2020, p. 469). Many acknowledge that governments, particularly the US and UK governments, may not be the most trusted messengers in some parts of the world (Braddock & Morrison, 2020). Where messages are seen as originating with untrustworthy messengers in the eyes of target audiences, they may fail to make headway or persuade (Corman, 2006, p. 98). This variation means that audiences are likely to respond very differently to a message that comes from a trusted friend than, say, a mistrusted government. This belief has translated into a range of practical recommendations for counter-narratives. Although governments are recognised as having a valuable role to play in strategic communications and getting positive messages out about what the government is doing, this is unlikely to translate to all audiences. Especially where messages are targeted at extremist narratives, the advice from some quarters has been that governments need to take a more hands-off approach, enabling civil society and facilitating networks and investment rather than attempting to get their hands dirty by rolling around in the mud with extremists (Briggs & Feve, 2013; Lee, 2019). In response, governments have frequently sought to enlist the help of outside agents in engaging in counter-narratives. Even where the NGO sector has stepped in to deliver counter-narrative campaigns, this has often been with government funding (Schmid, 2014, p. 15). In cases where governments have attempted to conceal their involvement, the exposure of a counter-narrative campaign as being at the behest of a government is often framed as a ‘gotcha’ moment by the press. In one memorable example, the UK’s Research Information and Communication Unit (RICU), the Home Office unit responsible for counter-narrative work, was branded a ‘shadowy propaganda unit inspired by the cold war’ (Cobain et al., 2016). In contrast, messages coming from those with authority, knowledge, and status among target audiences may make more headway (Ashour, 2010). Early approaches to counter-messaging not only acknowledged the role of the government, particularly the US government, in the battle of ideas but also suggested exploring alternative messengers, with former Jihadist terrorists being a key recommendation (Jacobson, 2009). Former extremists, in some cases exJihadists and in others former right-wing extremists, are often identified as being able to relate to present-day extremists and more likely to be trusted by target audiences (Braddock, 2020, p. 60; Braddock & Morrison, 2020). Others have noted the value of victims and survivors of terrorism as credible voices on the contradictions often embedded in extremist narratives, in particular highlighting the instances where Islamist extremist violence has targeted fellow Muslims (Schmid, 2014, p. 16). Despite the overall keenness to incorporate formers into counter-narrative work, evidence from the wider CVE sphere suggests that careful analysis is needed before working with formers. Analysis of former ETA terrorists in Spain suggested that, despite disengaging from violence, some did not fully repudiate their views (Alonso & Díaz Bada, 2016, p. 17). One study of the use of formers in CVE interventions in Denmark found mixed effects, with the formers’ participation found to have a positive effect on delegitimising violence but not delivering benefits around political efficacy and political tolerance. In their conclusions, the authors suggested that the use of formers could potentially be something of a ‘double edged sword’ (Parker & Lindekilde, 2020, p. 15). A more controversial question is what compromises authorities are prepared to make to reduce the possibility of violence. Schmid (2014, p. 29) suggests that successful campaigns may seek to incorporate radical but not extreme messengers into them. Messengers can share the values and maybe even the grievances of target audiences but then divert them from violent solutions. 435

Benjamin Lee

Content The final of the three issues here covers the message – in essence, what do people say and how do they say it. The majority of research and advice on counter-narrative focuses on the content of narratives and offers different takes on what makes a compelling narrative likely to win favour. Research on persuasion is relatively well developed and holds interest in a wide array of fields, including public health research (Cameron, 2009). In terms of counter-narrative, one of the most promising theories is narrative itself. As already noted, narratives are considered to be more than simply stories, and a more developed understanding of what narratives are and how they function has been presented as both a natural persuasive tool (Lee & Leets, 2002) and as essential to developing counter-narratives (Glazzard, 2017). Researchers have reached back as far as Aristotle and more recently into disciplines like literary studies and narratology to argue for the need to go beyond the main themes presented in communications and pay more attention to their broader presentation (Glazzard, 2017, p. 8). A more literary analysis of the overarching geo-political fables embedded in AlQaeda’s narratives, for example, can potentially move beyond political science approaches that tend to overlook the more abstract and textural aspects of how narratives are presented (Glazzard, 2017, p. 15). Braddock provides an overview of some of the aspects of counter-narrative that can aid in persuasion, breaking apart analogies and binary distinctions that often feature in extreme narratives, noting in particular the role of highlighting contradictory themes in extremist narratives, providing speakers and characters that the target audience can identify with, and creating convincing world and situations, in particular the transformative effects of narrative, that is, transporting audiences to different situations (Braddock, 2020, pp. 73–83). In a related vein, the Elaboration Likelihood model of persuasion is widely used in health research and argues that persuasion can occur via differing routes, known as central and peripheral processing (Cameron, 2009, p. 312; Lee & Leets, 2002, p. 931). Audiences are theorised to respond to different forms of persuasion differently. Central routes, where audiences are alive to the persuasive attempt and are able to think up counter-arguments, are thought to lead to resistance from audiences. Peripheral processing refers to subtler means of persuasion where audiences are encouraged to go along with messages based on cognitive cues such as the perceived trustworthiness of the messenger (Cameron, 2009, p. 312). This idea is developed further in research that has not only assessed the impact of white-nationalist ‘persuasive story-telling’ on audiences (Lee & Leets, 2002) but also led to recommendations for using drama and other less obvious tools as potential tools for counter-narrative (Ferguson, 2016). Inoculation theory is distinctive from other theories as it deals with resistance to persuasion rather than persuasion itself. Making a biological analogy, inoculation theory suggests that beliefs can be strengthened by exposing the holder to weakened versions of the arguments that may be brought against them (Saleh et al., 2021, p. 4). In this way, those holding beliefs can then develop their own counter-arguments and resist persuasive ‘attacks’ (Cameron, 2009, p. 314). In the setting of extremism and counter-narrative interventions, inoculation-style approaches are readily seen in extremist narratives that seek to forewarn and condition believers against buying into opposing narratives. Attitudinal inoculation works in the same way. When individuals have ‘healthy’ beliefs and attitudes about an issue, they can be exposed to messages that trigger the development of psychological defenses that prepare them to reject persuasive attempts to change their 436

Counter-narratives and strategic communications

positions on that issue. In the context of violent extremism, ‘healthy’ beliefs and attitudes would be those that are not yet supportive of violence in support of an extremist ideology. (Braddock, 2020, p. 116) For more on how inoculation theory applies to counter-narrative campaigns, see Braddock (this volume). As well as the content of messages, theoretical insights have also come from research that considered the modes of presentation. Braddock and Morrison (2020) focused heavily on trust as their main theoretical construct but developed the idea that trust could be affected by the various affordances built into the design of platforms. Drawing on the work of Sundar, in particular the MAIN (modality, agency, interactivity, and navigability) model, different design decisions were thought to feed into how audiences interpreted the overall credibility of the messages they were consuming (Braddock & Morrison, 2020, p. 474). They went on to outline a series of recommendations or messengers to build trust based on providing users with specific cues such as allowing users to navigate information easily to encourage a feeling of agency and control over processing, limiting the amount of off-putting navigational steps (‘click for next slide’), and recruiting and targeting peers of target audiences (see Braddock & Morrison, 2020, pp. 480–486). Former terrorists were also a source of inspiration for early approaches to counter-messaging; looking at what made some terrorists quit the field was thought likely to be instructive in convincing others (Jacobson, 2009). On the basis of an analysis of formers, one researcher suggested two general approaches: one to undermine the credibility of leadership figures to provoke internal dissent, and another to highlight the reality of life as a Jihadist, including the threat of death, harsh conditions, and the high levels of violence often inflicted on fellow Muslims by Jihadists. A more pragmatic account based on Somali-Canadians identified the availability of counternarratives as a key tool for empowering the diaspora community to reject the narratives of the terror group al-Shabaab. Reversing the ‘why they joined’ question via narrative criminology, Joosse et al. (2015) found that non-recruits were able to equip narratives to resist persuasive attempts, in particular by discrediting al-Shabaab and linking them to the bogeyman idea found in almost all cultures. Among non-recruits, recruiters were rejected on the grounds that they were tricksters and that they perverted religion. Despite these accounts, it is difficult to escape the conclusion that counter-narrative as a whole lacks an overarching theory connecting it solidly to either radicalisation theories or related literature on persuasion and narrative. Some researchers have begun the process of building bridges to outside theoretical work, but this is incomplete, and there is nothing like a consensus yet on how narratives persuade or how this should be reflected in campaigns.

Emergent issues and critiques The above has outlined some of the key concepts, theoretical takes, and practical examples of counter-narrative as it currently stands in the literature. This section outlines some of the ongoing debates relating to counter-narrative and, in particular, some of the enduring critiques of counternarrative as a component of CVE. The most fundamental critique of counter-narratives comes from radicalisation theory, in particular the ambiguous role of communication and socialisation in radicalisation. The idea that violent words lead to violent deeds is still under debate (Ferguson, 2016, p. 2). While the link between violent extremist content and violence features heavily in current debates (Sageman, 2008; Hamm & Spaaij, 2017; Gaudette et al., 2020), researchers also 437

Benjamin Lee

emphasise that exposure alone is insufficient and other prior vulnerabilities and grievances need to be present (Rottweiler et al., 2021). A secondary critique, already discussed earlier, is that the practice is under-theorised. Aside from these complaints, four distinct issues stand out as being the main points of friction around counter-narratives: evaluation, informal approaches, normative questions, and online/offline dynamics.

Evaluation One of the most enduring issues associated with counter-narratives is the overall lack of evaluation and empirical evidence (Parker & Lindekilde, 2020; Silverman et al., 2016; Malet, 2021). Counter-narratives are intended to be preventive, and so evaluation often relies on assessing the absence of activity (Schmid, 2014, p. 12; see also Fisher & Busher, Chapter 21 in this volume on evaluation and CVE more broadly). Added to this, counter-narratives operate in noisy environments, and researchers have pointed out the difficulties in identifying effects and thereby ­evaluating counter-messages: Attributing the successful prevention of violence to a single factor is always difficult, which means isolating and measuring the success or impact of preventative media and communication strategies is particularly challenging. Nevertheless, the absence of methodologically robust monitoring and evaluation (M&E) practices with regard to CVE counter-narratives is striking. (Ferguson, 2016, p. 8) Write-ups of counter-narrative work may fall back on measures of impressions and reach to make their point, but these only go so far if they cannot be linked to behavioural or attitudinal change (Silverman et al., 2016, p. 6). Research on CVE more broadly has described a divide between academia and practitioners, which prevents theoretical accounts from connecting with real-world implementations (Malet, 2021). Some of the most interesting evidence around the effectiveness of counter-narrative (specifically counter-speech) has come from studies carried out within Facebook itself. Saltman et al. (2021) set out the findings from two counter-speech initiatives in which Facebook partnered with outside NGOs to test two different approaches to counter-speech. The results of the studies were mixed. An A/B study testing different approaches to counter-speech that ‘surfaced’ to users identified as being at risk of Jihadist extremism identified no significant change in the targeted audience’s behaviours after exposure to counter-speech. It also highlighted that even Facebook itself is unable to compel users to pay attention to specific content. Of more than 40 minutes of total content, just 23% of those targeted were confirmed as having viewed more than a minute, and only 5% more than three minutes (Saltman et al., 2021, p. 12). One possible reason for this is the difficulty in identifying audiences, with Facebook’s own marketing tools proving too blunt to target the most at-risk audiences. A more focused qualitative analysis of the profiles thought to be most at risk identified some examples of individuals seemingly desisting after exposure to counter-speech content (Saltman et al., 2021, p. 14). Likewise, a parallel study focused on right-wing extremist content used a ‘re-direct’ method tied to specific search terms and found improved engagement with an external partner NGO as a result (Saltman et al., 2021, p. 21). Despite this, the lead finding offered by the paper was still that being targeted for counter-speech did not make audiences posting habits any worse.

438

Counter-narratives and strategic communications

This study, supported as it was by Facebook, offers some of the best evidence on the actual effects of counter-speech, or counter-narratives as described here, and it is fair to say that the evidence is fairly underwhelming. However, this evidence is also limited. First, it deals only with Facebook’s walled garden, where a great deal of violent extremist content has been forced off of mainstream platforms and onto smaller and less popular alternatives, suggesting that Facebook is unlikely to be the home of the most committed or the most threatening. Second, the approaches being tested are heavily focused on what Briggs and Feve (2013) identified as counter-narratives, debunking the extremist message to extremists as opposed to instilling a more positive alternative narrative. As a result, there is little evidence that speaks either to the idea of inoculation suggested by Braddock (2020) or preventing audiences from entering the ‘at risk’ category in the first place. There is some positive evidence to support inoculation-based approaches, in particular approaches that shift from passive (e.g. reading) to more active modes of communication such as interaction that aim to generate greater cognitive engagement (Saleh et al., 2021). ‘Gamification’, or the use of game-like mechanisms in counter-narrative, has been suggested by some to be one area of future innovation in counter-narrative (Saleh et al., 2021, p. 18). One well-designed trial used an online game, Radicalise, to sensitise players to extremist recruitment and messaging. It is a 15-minute online game in which participants take on the role of (fictional) extremists aiming to win over converts. Taking the perspective of the extremists in this instance was thought to be particularly valuable as it avoided branding participants as vulnerable (Saleh et al., 2021, p. 6). Participants were found to significantly improve in their ability and confidence in spotting manipulative messaging following exposure, compared to a group who had instead played Tetris, which the researchers translated as evidence that this kind of intervention could limit the effectiveness of online manipulation and misinformation. Another intervention, designed by Moonshot, targeted at-risk groups in Indonesia and found that adverts targeting loneliness and employment support were more likely to engage users than those offering more traditional ideological support. This was interpreted as evidence that psychosocial support was an area of unmet need in at-risk populations and may provide greater benefits than attempting to counter ideology directly through online interventions (Moonshot, 2020). Overall, however, the evidence base is lacking. A Campbell systematic review provided an overview of experimental and quasi-experimental research on the effectiveness of counter-narratives. The combined evidence of 19 studies revealed a mixed picture for counter-narrative interventions. Some specific approaches, such as interventions designed to counter stereotypes, were found to reduce in-group favouritism and out-group hostility, as well as foster a more realistic threat perception. In contrast, there was little evidence for any effect relating to ‘primary outcomes’ related to violent extremism, such as the intent to act violently (Carthy et al., 2020). While the evidence base for counter-narratives is improving, the new insights provided by these studies also serve to highlight how much we do not know. Inside data from Facebook reveals the limits of tracking users, let alone linking their online and offline behaviours. This raises issues for public health-based models, as it seems that even with access to relatively granular data, identifying populations ‘at risk’ of extremism is extremely difficult. Likewise, experimental approaches are also hard to link to real-world outcomes with certainty, operating as they do with self-selecting samples rather than actual extremists. While things have certainly improved, a great deal of variance is likely to remain unexplained.

Informal approaches The examples discussed so far have all originated with either the state or non-governmental organisations. However, counter-narratives are not just the business of governments and the

439

Benjamin Lee

organisations that orbit them; there is no shortage of actors implementing a variety of persuasive strategies to counter extremist messaging. Neckbeard Deathcamp is a Black Metal band that explicitly parodies and undermines tropes associated with the online extreme right. Album titles such as White Nationalism Is for Basement Dwelling Losers and So Much for the Tolerant Left are accompanied by song titles that aim to deflate the seriousness of the extreme-right, including Zyklon /b/, MAGAPHOBE, and XXXL Obersturnfuhrer Leather Duster. Cover art includes parodies of fascist eagles, replacing the eagle’s head with a penis, and a (sold-out) t-shirt advertised bears the slogan ‘exterminate the bedroom reich’. Neckbeard Deathcamp is one example in a long list of projects, campaigns, and everyday activities that have sought to speak out against extremism in different forms. Examples include the use of Anime-inspired memes to disrupt Islamic State propaganda (Johansson, 2018); launching campaigns aimed variously at mocking, exposing, and, in some cases, even discussing issues with extremists (Lee, 2020); and grassroots online campaigns (Macnair & Frank, 2017). There has also been interest in more spontaneous and organic forms of counter-messaging made possible by the design of social media platforms like Twitter. A study of the Twitter hashtag #stopislam that became prominent in the aftermath of the March 2016 Islamist attack in Brussels that killed 32 noted that despite the original intention of the hashtag, it was appropriated by opponents, and nine of the ten most popular messages on the hashtag were critical of the call the #stopislam (Giraud & Poole, 2021). Citizen-led approaches to countering narratives open up interesting possibilities. Theoretically, they mesh well with wider theories around increasing citizen’s roles in security provision, in which private citizens contribute the resources, expertise, and symbolic capital no longer present within governmental networks (Dupont, 2004; see also Nhan et al., 2015; Huey et al., 2013). In the case of counter-messaging, individual citizens, potentially with backgrounds and perspectives much closer to potential audiences than civil servants and NGO employees, may contribute significantly to counter-narrative campaigns. Things like credibility, trustworthiness, and social status (symbolic capital in Dupont’s language), however, are fragile, and bringing informal actors closer to formal CVE risks undermining source credibility. In addition, large and well-insulated civil society actors may unintentionally outsource the risk that comes with engaging with extremists with uncertain results (Lee, 2019).

Normative critiques Many of the issues that have emerged around counter-narrative campaigns are normative and speak directly to concerns over the proper role of the state, social engineering, propaganda, and dealing with dissent. In common with CVE more generally, there have been concerns that the counter-narrative focuses excessively on Islamists, and more problematically, Muslims, to the exclusion of other types of extremism (especially the extreme right) (Tierney, 2017, p. 69). Others have focused on the ‘credibility gap’ between the words and deeds of counter-narrative actors. This mismatch between rhetoric and reality was referred to by one research team as the ‘say-do gap’ and was presented as an important reason why governments in particular needed to tread carefully in the counter-narrative space (Briggs & Feve, 2013, p. 11). These critiques are a natural fit for those with a critical perspective on the ‘war on terror’, although they are seldom rolled out in response to programmes directed at the far right. However, if counter-narratives have an original sin, it is the inescapable link to propaganda. The origins of counter-narratives present, perhaps a little ironically, themselves as a more open and liberal approach to emerging issues of online extremism. Rather than censoring and banning 440

Counter-narratives and strategic communications

content, counter-narratives allowed a more liberal approach whereby dissent was tolerated, but extremist appeals were mitigated by effective counter-narratives (Briggs  & Feve, 2013). However, for some researchers, the lack of explicit conceptualisation and the proximity to large state organisations leave the counter-narrative as a ‘euphemism for state propaganda’ (Glazzard, 2017, p. 6). Propaganda, with its inevitable connections to totalitarian regimes, casts a long shadow over counter-narratives, strongly suggesting that this kind of campaign is somehow illegitimate (Payne, 2009, p. 116). The response has been either to carve out some kind of distinction for counter-­ narrative, marking it out as being different from propaganda (when done properly) (Glazzard, 2017), or to try and rehabilitate the idea of propaganda in liberal settings (Payne, 2009).

Online/offline dynamics A further issue concerns the division between online and offline counter-narratives. As already noted, counter-narratives began firmly embedded in ‘conventional’ modes of communication such as the media and advertising. When extremists shifted online with seeming ease, counter-­ narratives too went online. At this point, counter-narrative is almost synonymous with online communications. This is perhaps less surprising given the broader shift of communications to the ‘post-digital’; increasingly, online and offline are artificial distinctions, and the online and offline are increasingly interdependent. As one easily accessible example, the 6 January 2021 riot instigated by Trump supporters in Washington, DC was filmed from so many different angles and devices that New York Times Visual Investigations was able to put together a stunningly detailed account assembled from the footage (New York Times, 2021). This wider shift to a digitally infused communications environment creates several issues. Whereas counter-narratives backed by government-scale budgets and on friendly terms with powerful media giants could rely on an almost captive audience in the early days, the proliferation of digital technology in the 21st century, even for populations considered marginalised as refugees in transit (Kaplan, 2018), means that ‘official’ narratives must now compete against millions of other potential sources of information as consumers have far more control over what they view (Payne, 2009, p. 117). Although they remain large players, governments no longer enjoy the direct line to their citizens that they once did. In other words, any counter-narrative campaign is now in a contest for attention with thousands of other content creators, many of them with material that is far more readily engaging than worthy counter-narrative campaigns. A related point surrounds the mode of delivery of counter-narrative content. While online delivery has come to dominate the counter-narrative space to the point that it is almost synonymous with counter-narrative, the technology itself can impact how messages are received and interpreted by audiences, with the design of sites and content potentially allowing messages to activate specific persuasive heuristics that may improve the reception of content. An example is providing users with a choice over how they access content, allowing them to interpret information at their own pace, and preventing them from being overwhelmed by too much information (Braddock & Morrison, 2020, p. 479). The strong commitment to digital methods to counter extremist narratives has the advantage of scalability and a degree of automation. However, this leaves counter-narrative campaigns exposed to shifts in the way extremist narratives are transmitted. A  return to more offline-centred communications by extremists or the adoption of alternative platforms, which are less concerned by their use by extremists, most notably Telegram, have the potential to derail, or at least reduce, the value of campaigns limited to mainstream digital platforms. Now that the Internet genie is out of the bottle, it is unlikely that any platform online will be free of extremist narratives, and yet, an 441

Benjamin Lee

online-only approach to countering narratives may struggle to reach those at risk of radicalisation in other arenas.

Unintended consequences Finally, a persistent spectre haunting the counter-narrative landscape is the idea that counter-­ narrative campaigns may do more harm than good for their intended audiences. While mainstream opinion may find it difficult to understand how messaging promoting tolerance and unity could possibly do harm, unintended consequences are a persistent risk. At worst, counter-narratives may come off as shrill overreactions to violent extremism, perhaps further reinforcing the reasons individual’s rejected mainstream cultures in the first place. Corman (2006) notes the degree to which Salafist ideologies are in dialogue with Western ones, and so arguments such as bringing democracy to a region can be easily reproduced to the opposite of the intended effect, with democratic reform becoming an attempt to undermine the primacy of Islam (see also Braddock, 2020). Democratisation was also a core theme of early US efforts at winning over the Muslim world but was undercut by the idea’s unpopularity with the (frequently anti-democratic) leadership in some Muslim countries, the fear of democratic elections putting ideological foes into power, and the extensive gap between the presentation of the democratic West and images from Abu Ghraib and Guantanamo Bay (Schmid, 2014, p. 9). An attempt at counter-narrative that seemingly fell into this trap came from the US government and the State Department’s Think Again, Turn Away campaign that sought to ridicule the Islamic State. The central video was a parody of an Islamic State recruitment video, offering new recruits the opportunity to learn new skills for the ‘Ummah’, including blowing up mosques and executing Muslims (Ramsay & Alkheder, 2020, pp. 18–19). The ironic tone of the video, combined with the inclusion of footage taken by Islamic State activists, prompted criticism from within the news media, including a notable segment from satirist John Oliver and others (McClennen, 2014). Beyond the rejection of the tone, however, Ramsay and Alkheder (2020, p. 19) note that the aspects of the Islamic State that the video chose to mock – specifically the violence and occupation of Syria and Iraq – were actually perfectly in line with the Islamic State’s narrative at the time. Violence and death were key parts of the appeal, as was the idea of a caliphate that included all Muslims. The unanticipated negative effects of counter-narratives may be exacerbated where audiences, especially those often assumed to be vulnerable to violent extremism (e.g. Muslims and young white males), are repeatedly targeted by counter-narrative campaigns that lack accurate targeting information or coherent understandings of radicalisation. Such groups are repeatedly asked to condemn terrorism or extremism that they may have little or no knowledge of (see Ferguson, 2016, p. 9). In essence, a badly implemented counter-narrative could conceivably work to sensitise audiences uninterested in violent extremism and act as a confirmation of some aspect of extremist narratives.

Conclusions Counter-narrative is not without its critics, and the debates start with conceptualisation. For those with a stronger grasp on literary theory, counter-narrative is often used too loosely to be worthy of the name, while a host of other researchers and NGOs have sought to bring their own perspectives and terminology. A second criticism linked to this has been that counter-narratives are under-­theorised. The majority of research work is forward-looking and concentrates on strategic and tactical accounts of the need for counter-narratives and their execution. Coherent theoretical explanations are less common, but nonetheless a body of work more closely related to existing 442

Counter-narratives and strategic communications

theories around literature and persuasion is emerging. The final major critique has been the lack of empirical evidence around the effectiveness of counter-narrative. Here too, the landscape is slowly improving, even if it is revealing new complexities as it does so. There are a range of secondary criticisms, including the tendency to overlook more grassroots approaches to counter-narrative, the risks of unintended consequences, and a fairly narrow focus on social media and the online realm. This review has offered a brief tour of a complex field. In summary, counter-narrative sits as a specific form of CVE centred on communication. Its origins stem from the policy world, where it has developed over time, first as a tool specific to the ‘war on terror’, and then as a broader pathway for challenging extremism. The legacy of these origins is stark in the terminology used in the literature: war metaphors are everywhere. Going forward, and as the theoretical and evidential bases improve, it is possibly normative questions around counter-narratives that seem likely to generate the most debate. What started, particularly in the United States, as a set of tools to be directed against both an ideological and military enemy is likely to be increasingly used against citizens at home. In the United Kingdom, critical commentary about the unfair targeting of British Muslims has made some headway, but as more and more extreme subcultures, most notably the extreme and radical right, are folded into the terrorism envelope, fresh controversies are likely to emerge. Further expansions of counter-narratives are likely to elevate as yet unanswered questions about the role of counter-narratives, and the role of governments in them.

References Aggarwal, N., (2017). Exploiting the Islamic state-Taliban rivalry for counterterrorism messaging. Journal of Policing, Intelligence and Counter Terrorism, 12(1), 1–15. https://doi.org/10.1080/18335330.2016.12 23868 Aldis, A., & Herd, G. (2007). The ideological war on terror: Worldwide strategies for counter-terrorism. Routledge. Allchorn, W. (2020). Building a successful radical right counter-narrative campaign: A how to guide. Hedayah. https://hedayahcenter.org/app/uploads/2021/09/2020DEC16_CARR_HowToGuide_FINALdouble-spread.pdf Alonso, R., & Díaz Bada, J. (2016). What role have former ETA terrorists played in counter terrorism and counter radicalization initiatives in Spain? Studies in Conflict & Terrorism, 39(11), 982–1006. https://doi. org/10.1080/1057610X.2016.1154365 Ashour, O. (2010). Online de-radicalization? Countering violent extremist narratives: Message, messenger and media strategy. Perspectives on Terrorism, 4(6). Bartlett, J., & Krasodomski-Jones, A. (2015). Counter-speech: Examining content that challenges extremism online. Demos. Braddock, K. (2020). Weaponized words: The strategic role of persuasion in violent radicalization and counterradicalization. Cambridge University Press. Braddock, K., & Morrison, J. (2020). Cultivating trust and perceptions of source credibility in online counternarratives intended to reduce support for terrorism. Studies in Conflict & Terrorism, 43(6), 468–492. https://doi.org/10.1080/1057610X.2018.1452728 Briggs, R., & Feve, S. (2013). Review of programs to counter narratives of violent extremism: What works and what are the implications for government? Institute for Strategic Dialogue. Cameron, K. A. (2009). A practitioner’s guide to persuasion: An overview of 15 selected persuasion theories, models and frameworks. Patient Education and Counseling, 74(3), 309–317. https://doi.org/10.1016/j. pec.2008.12.003 Carthy, S. L., Doody, C. B., Cox, K., O’Hora, D., & Sarma, K. M. (2020). Counter‐narratives for the prevention of violent radicalisation: A systematic review of targeted interventions. Campbell Systematic Reviews, 16(3). https://doi.org/10.1002/cl2.1106 Cobain, I., Ross, A., Evans, R., & Mahmood, M. (2016). Inside Ricu, the shadowy propaganda unit inspired by the cold war. Retrieved February 11, 2022, from www.theguardian.com/politics/2016/may/02/ inside-ricu-the-shadowy-propaganda-unit-inspired-by-the-cold-war

443

Benjamin Lee Conway, M. (2012). From Al-Zarqawi to al-Awlaki: The emergence of the internet as a new form of violent radical Milieu. CTX: Combating Terrorism Exchange, 2(4), 12–22. Corman, S. (2006). Weapons of mass persuasion: Communicating against terrorist ideology. Connections: The Quarterly Journal. http://connections-qj.org/article/weapons-mass-persuasion-communicatingagainst-terrorist-ideology Dupont, B. (2004). Security in the age of networks. Policing and Society, 14(1), 76–91. https://doi. org/10.1080/1043946042000181575 European Commission. (n.d.). RAN collection of practices search. Retrieved February 9, 2022, from https:// ec.europa.eu/home-affairs/networks/radicalisation-awareness-network-ran/collection-inspiring-practices/ ran-collection-search_en Ferguson, K. (2016). Countering violent extremism through media and communication strategies: A review of the evidence. Partnership for Conflict, Crime  & Security Research. www.paccsresearch.org.uk/wpcontent/uploads/2016/03/Countering-Violent-Extremism-Through-Media-and-Communication-Strategies-.pdf Gaudette, T., Scrivens, R., & Venkatesh, V. (2020). The role of the Internet in facilitating violent extremism: Insights from former right-wing extremists. Terrorism and Political Violence. https://doi.org/10.1080/09 546553.2020.1784147 Giraud, E., & Poole, E. (2021). Constructing digital counter-narratives as a response to disinformation and populism. In H. Tumber & S. Waisbord (eds), The Routledge companion to media misinformation and populism (pp. 529–538). Routledge. Glazzard, A. (2017). Losing the plot: Narrative, counter-narrative and violent extremism. ICCT. https://doi. org/10.19165/2017.1.08 Hamm, M., & Spaaij, R. (2017). The age of lone wolf terrorism. Columbia University Press. Harris-Hogan, S. (2020). How to evaluate a program working with terrorists? Understanding Australia’s countering violent extremism early intervention program. Journal of Policing, Intelligence and Counter Terrorism, 15(2), 97–116. https://doi.org/10.1080/18335330.2020.1769852 Huey, L. (2015). This is not your mother’s terrorism: Social media, online radicalization and the practice of political jamming. Journal of Terrorism Research, 6(22), 1–16. Huey, L., Nhan, J.,  & Broll, R. (2013). “Uppity civilians” and “cyber-vigilantes”: The role of the general public in policing cyber-crime. Criminology and Criminal Justice, 13(1), 81–97. https://doi. org/10.1177/1748895812448086 Jacobson, M. (2009). Terrorist drop-outs: One way of promoting a counter-narrative. Perspectives on Terrorism, 3(2). www.terrorismanalysts.com/pt/index.php/pot/article/view/66 Johansson, A. (2018). ISIS-Chan – the meanings of the manga girl in image warfare against the Islamic state. Critical Studies on Terrorism, 11(1), 1–25. https://doi.org/10.1080/17539153.2017.1348889 Joosse, P., Bucerius, S., & Thompson, S. (2015). Narratives and counternarratives: Somali-Canadians on recruitment as foreign fighters to Al-Shabaab. British Journal of Criminology, 55(4), 811–832. https://doi. org/10.1093/bjc/azu103 Kaplan, I. (2018). How smartphones and social media have revolutionized refugee migration. Retrieved October 26, 2021, from www.unhcr.org/blogs/smartphones-revolutionized-refugee-migration/ Lee, B. (2019). Informal countermessaging: The potential and perils of informal online countermessaging. Studies in Conflict & Terrorism, 42(1–2), 161–77. https://doi.org/10.1080/1057610X.2018.1513697 Lee, B. (2020). Countering violent extremism online: The experiences of informal counter messaging actors. Policy & Internet, 12(1), 66–87. https://doi.org/10.1002/poi3.210 Lee, E., & Leets, L. (2002). Persuasive storytelling by hate groups online: Examining its effects on adolescents. American Behavioral Scientist, 45(6), 927–957. https://doi.org/10.1177/0002764202045006003 Leuprecht, C., Hataley, T., Moskalenko, S., & McCauley, C. (2009). Winning the battle but losing the war? Narrative and counter-narratives strategy. Perspectives on Terrorism, 3(2). www.terrorismanalysts.com/ pt/index.php/pot/article/view/68 Macnair, L., & Frank, R. (2017, Spring). Voices against extremism: A case study of a community-based CVE counter-narrative campaign. Journal of Deradicalization, 10, 147–174. Malet, D. (2021). Countering violent extremism: Assessment in theory and practice. Journal of Policing, Intelligence and Counter Terrorism, 16(1), 58–74. Malthaner, S., & Waldmann, P. (2014). The radical milieu: Conceptualizing the supportive social environment of terrorist groups. Studies in Conflict & Terrorism, 37(12), 979–998. https://doi.org/10.1080/1057 610X.2014.962441

444

Counter-narratives and strategic communications McClennen, S. (2014). John Oliver is smarter than the state department: Irony will not defeat ISIS. Retrieved February 11, 2022, from www.salon.com/2014/10/06/john_oliver_is_smarter_than_the_ state_department_irony_will_not_defeat_isis/ Moonshot. (2020). Social grievances and violent extremism in Indonesia: Exploring the appetite for psychosocial support among at-risk audiences. Retrieved February 9, 2022, from https://moonshotteam.com/ wp-content/uploads/2020/12/Social-Grievances-and-Violent-Extremism-in-Indonesia_Moonshot.pdf New York Times. (2021). Day of rage: How Trump supporters took the U.S. capitol | visual investigations. New York Times. Retrieved February 9, 2022, from www.youtube.com/watch?v=jWJVMoe7OY0 Nhan, J., Huey, L., & Broll, R. (2015). Digilantism: An analysis of crowdsourcing and the Boston Marathon bombings. British Journal of Criminology, 57(2), 341. https://doi.org/10.1093/bjc/azv118 Parker, D., & Lindekilde, L. (2020). Preventing extremism with extremists: A double-edged sword? An analysis of the impact of using former extremists in Danish schools. Education Sciences, 10(4), 111. https://doi.org/ 10.3390/educsci10040111 Payne, K. (2009). Winning the battle of ideas: Propaganda, ideology, and terror. Studies in Conflict & Terrorism, 32(2), 109–128. https://doi.org/10.1080/10576100802627738 Pizzuto, M. (2013). Alter-messaging: The credible, sustainable counterterrorism strategy. Global Centre on Cooperative Security. www.jstor.org/stable/resrep20349?seq=1#metadata_info_tab_contents Ramsay, G., & Alkheder, M. (2020). Joking about Jihad: Comedy and terror in the Arab world. Hurst & Company. Rottweiler, B., Gill, P., & Bouhana, N. (2021). Individual and environmental explanations for violent extremist intentions: A German nationally representative survey study. Justice Quarterly, 1–22. https://doi.org/1 0.1080/07418825.2020.1869807 Russell, J., & Rafiq, H. (2016). Countering Islamist extremist narratives: A strategic briefing. Quilliam Foundation. www.academia.edu/download/50998161/Countering-islamist-extremist-narratives.pdf Sageman, M. (2008, July). A strategy for fighting international Islamist terrorists. Annals of the American Academy of Political and Social Science, 223–231. Saleh, N. F., Roozenbeek, J., Makki, F. A., Mcclanahan, W. P., & Van Der Linden, S. (2021). Active inoculation boosts attitudinal resistance against extremist persuasion techniques: A novel approach towards the prevention of violent extremism. Behavioural Public Policy, 1–24. https://doi.org/10.1017/bpp.2020.60 Saltman, E., Kooti, F., & Vockery, K. (2021). New models for deploying counterspeech: Measuring behavioral change and sentiment analysis. Studies in Conflict & Terrorism, 1–24. https://doi.org/10.1080/1057 610X.2021.1888404 Schmid, A. P. (2014). Al- Qaeda’s “single narrative” and attempts to develop counter-narratives: The state of knowledge. ICCT. www.icct.nl/app/uploads/download/file/Schmid-Al-Qaeda’s-Single-Narrative-andAttempts-to-Develop-Counter-Narratives-January-2014.pdf Sedgwick, M. (2012). Jihadist ideology, Western counter-ideology, and the ABC model. Critical Studies on Terrorism (April 2015), 1–14. https://doi.org/10.1080/17539153.2012.723520 Silverman, T., Stewart, C., Amanullah, Z., & Birdwell, J. (2016). The impact of counter-narratives: Insights from a year-long cross-platform pilot study of counter-narrative curation, targeting, evaluation and impact. Institute for Strategic Dialogue. www.isdglobal.org/wp-content/uploads/2016/08/Impact-ofCounter-Narratives_ONLINE_1.pdf Sim, S. (2013). Countering violent extremism: Leveraging terrorist dropouts to counter violent extremism in Southeast Asia. Qatar International Academy for Security Studies. Tierney, M. (2017). Combating homegrown extremism: Assessing common critiques and new approaches for CVE in North America. Journal of Policing, Intelligence and Counter Terrorism, 12(1), 66–73. https://doi. org/10.1080/18335330.2017.1293281 Warrington, A. (2018, Spring). “Sometimes you just have to try something” – a critical analysis of Danish state-led initiatives countering online radicalisation. Journal for Deradicalization, 14, 111–152.

445

29 DERADICALISATION AND DISENGAGEMENT Lessons from the Indonesian experience Julie Chernov Hwang

This chapter explores deradicalisation and disengagement conceptually and programmatically, drawing on both the wider secondary literature on the subject and the Indonesian case. In doing so, it will unpack the lessons that can be learned from the Indonesian experience with deradicalisation and disengagement. These include the importance of emphasising lifeskills training and professional development in programme design; the key role that formers can play in programme development and implementation; the need for programmes to be agile and adaptable in their design in order to accommodate new vulnerable populations and challenges; and the cruciality of research, time, and rapport building for successful programmatic interventions. This may raise the question of why the focus is on Indonesia, a country that is not commonly addressed in the literature on deradicalisation and disengagement. The answer is that Indonesia is the world’s largest Muslim nation and one of the Muslim world’s only democracies. It has struggled with terrorism more than any other nation in Southeast Asia and, importantly, has developed a wide variety of interventions, some of which have been successful and others of which have been far less so. It can offer a window into best practices that few countries can due to their varied experience with different types of programmes. It’s important to note that this chapter will centre on the experiences of Muslim nations, particularly their programmatic interventions, in order to offer a supplement to the often Europe-heavy literature on this subject. This chapter will first define the terms disengagement and deradicalisation. Then, it will outline the key debates in the literature on deradicalisation and disengagement, exploring these at the organisation level, individual level, and programmatic level. Finally, it will turn to the Indonesian experience.

Defining deradicalisation and disengagement There is a lack of agreement within the academic literature on what constitutes deradicalisation. However, that said, there is a fair amount of consensus on certain common elements, notably that the term refers to a change in ideology1 and beliefs,2 in contrast3 to disengagement, which addresses behavioural change.4 Moreover, there is a great deal of consensus that deradicalisation is a gradual process. Definitions vary based on whether scholars studied cases of collective deradicalisation5 (e.g. within a terrorist group) or individual-level6 deradicalisation. DOI: 10.4324/9781003035848-33 446

Deradicalisation and disengagement

Perhaps the most comprehensive study of collective deradicalisation comes from Omar Ashour (2010, 2009). He defines ideological deradicalisation as ‘the delegitimation of the ideology that underpins the use of violence to differentiate it from behavioral deradicalization, which involves cessation in violent actions’. Scholars studying individual-level deradicalisation emphasise different aspects of the process, with some highlighting violence7 and others highlighting changes in worldview more broadly as well as commitment to the group.8 Horgan and Braddock define it as ‘a social and psychological process whereby an individual commitment to and involvement in violent action is reduced to the extent that they are no longer at risk of involvement and engagement in violent activity’ (2010). Both Horgan and Braddock (2010), on the one hand, and Ashour (2010, 2009), on the other, conceptualise deradicalisation narrowly, in terms of revising previously held views pertaining to the use of violence. From a programmatic standpoint, there is tremendous utility in focusing definitions of deradicalisation around the use of violence (see Chukwuma & Jarvis, this volume). Such a narrow conceptualisation enables programmes to hone in on a distinct area where the threat to national security is greatest and where Islamist extremists already have an extensive literature debating the conditions under which violence is permissible. Alternatively, Bertjan Doosje, Fathali Moghaddam, Arie Kruglanski, Arjan de Wolf, Liesbeth Mann, and Allard Feddes conceptualise the term more broadly, asserting that deradicalisation involves lessening one’s commitment to the group and to the radical ideology they once embraced (2016). Here, the revision of previously held views on violence is implicit. Rabasa, Pettyjohn, Ghez, and Boucek characterise the term even more generally, asserting that deradicalisation involves changing an individual’s belief system, rejecting extremist ideology, and embracing mainstream values (2010). This latter definition conflates deradicalisation with moderation. Chernov Hwang chooses to define not deradicalisation but the related concept of disengagement, which she explains as the process via which members of terrorist groups, radical movements, gangs, or cults choose to cease participation in acts of violence (2018). She further stipulates that the process of disengagement is (a) gradual; (b) non-linear, meaning it can proceed in fits and starts; (c) and often conditional, meaning there may be conditions under which one might reengage, like the invasion of their country or the opportunity to take part in an international jihad experience (2018). Horgan emphasises that disengagement does not require one to leave the movement; one may remain in it in support roles (2009). Bjørgo terms this ‘role migration (2011)’. Kenney and Chernov Hwang found three ways in which Indonesian and British jihadists disengaged from violence: via leaving the movement; via role migration; and via going inactive (2021). It is important to note that one can disengage from violence without deradicalising.

Deradicalisation and disengagement pathways in the literature At the organisational level, exploring the cases of al-Gamaah al-Islamiyah and Islamic Jihad in Egypt as well as Armee Islam du Salut in Algeria, Omar Ashour found successful organisationallevel demobilisation to be a result of the interplay of four factors: charismatic leadership, selective inducements, state repression, and interaction with others (2010, 2009). According to Ashour, the linchpin of successful deradicalisation and demobilisation in the Egyptian and Algerian cases was charismatic leadership (2010, 2009). At the individual level, Daniel Koehler conceptualised how individuals deradicalise as a process of ‘repluralization’ of one’s values, beliefs, and approaches to solving problems (2017). Via external events, interventions, internal reflection, and shifts in priorities, one saw a decreased 447

Julie Chernov Hwang

urgency for violence as rising doubts led to a re-evaluation of commitments to the group of which they once were a part (2017). This allowed space for push and pull factors to enter (2017). The concept of ‘push and pull’ factors comes from Tore Bjørgo, a criminologist specialising in Scandinavian Neo-Nazi and far-right extremist movements (2009). He operationalised the disengagement process based on a series of factors that may ‘push’ the extremist away from the violent movement or ‘pull’ them towards a ‘normal’ life (2009). Notable push factors included the sense that ‘things have gone too far’, disillusionment with the inner workings of the group, negative social sanctions from family or community, and a loss of faith in the movement (Bjørgo, 2009). Fernando Reinares, a scholar examining the Euskadi Ta Askatasuna (ETA), highlighted the role of changing political context in disengagement, noting that after Basques had won regional autonomy, some fighters felt violent actions ‘just didn’t make sense anymore’ (2011). Drawing on cases outside Europe, Chernov Hwang’s research on Indonesia built on Horgan, Bjørgo, and Reinares in affirming the importance of disillusionment with tactics and leaders, the role of changing context, and shifts in priorities (2017, 2018). She emphasised alternative social networks as the linchpin of successful disengagement and reintegration (2017, 2018). New research has explored the mechanisms through which individuals disengage from violence. Especially notable in this regard is Altier, Thoroughgood, and Horgan’s use of Rusbult’s investment model. They have asserted that the propensity of a person to disengage from or remain committed to a violent extremist group is derived, in large part, from (a) the satisfaction a militant derives from participation in violent extremism; (b) the sunk costs that militants have incurred as a result of their involvement; and (c) any alternatives available to him/her outside that role (2014). The greater the dissatisfaction and the greater the availability of alternatives, the greater the likelihood of disengagement.

Deradicalisation and disengagement programmes in the literature Much of the literature on deradicalisation and disengagement in the Muslim world comprises evaluative studies, which assess the strengths and weaknesses of the Saudi, Pakistani, Yemeni, Singaporean, Egyptian, Malaysian, Moroccan, or Indonesian programmes.9 One common aspect of all such programmes is that they seek to reduce the risk that someone will engage in terrorism. Some of these programmes, notably the Saudi, Malaysian, and Singaporean programmes, take a more comprehensive approach, seeking to re-educate the offending terrorist away from their current Salafi-Jihadi interpretation of Islam towards a version of Islam that is more acceptable to the state (Fealy & Borgu, 2005). For example, the Malaysian government’s Religious Rehabilitation Program seeks to correct religious misperceptions, encourage militants to accept the argument that the Malaysian state is already an Islamic state, and provide assistance reintegrating back into society upon release (Aslam et al., 2016). However, the extent to which religious re-education programmes actually yield substantive ideological change is difficult to measure. How do you assess the extent to which someone has actually changed their thinking? One approach favoured in discussions of reintegration is the strategy of ‘reducing idleness’ (McRae, 2010; Gjelsvik & Bjorgo, 2012). The Danish programme for ISIS returnees is indicative of this model; those eligible receive assistance in obtaining employment, housing, and educational assistance (Horgan, 2014). Likewise, the Saudi, Singaporean, and Malaysian programmes also offer some aftercare assistants to newly released participants in their programmes (Abuza, 2009; Aslam et al., 2016). Even small-scale reintegration programmes can be useful in preventing recidivism, if they are well targeted and well planned (Gjelsvik & Bjorgo, 2012; McRae, 2010).

448

Deradicalisation and disengagement

However, if the programmes are not developed with an understanding of the needs of the target population, they are likely to fall short (Dennison, 2010). The experience of deradicalisation and disengagement programmes in Indonesia can offer an interesting comparison to highlight both best practices and those that are counter-productive. Going forward, this chapter will address four types of CVE initiatives in Indonesia: government-led, former-led, civil society-led, and state-civil society initiatives. In so doing, it will focus primarily on disengagement programmes, as deradicalisation initiatives have been fewer in number and far less successful. The most successful Indonesian programmes have focused on (a) rapport building; (b) encouraging the revision of previously held views regarding the conditions under which violence is permissible; and (c) providing life skills training and professional development.

Indonesian governmental programmatic interventions The first attempt at facilitating the disengagement of jihadists in Indonesia was an ad hoc initiative begun by Surya Dharma, then the head of Densus 88, the police counter-terrorism team, and continued by his successor, Tito Karnavian. This off-budget, grossly underfunded effort, known as ‘the soft approach’, made international news for the innovative idea that well-respected senior jihadists who had already disengaged could convince their subordinates to follow suit through formal and informal discussions. Key to this initiative were two senior members of Jemaah Islamiyah (JI), Nasir Abas, a Malaysian citizen and the former commander of Jemaah Islamiyah’s training region, and Ali Imron, a somewhat reluctant participant in the 2002 Bali bombing. The fact that these two senior JI members agreed to cooperate with Densus in this regard was something of a coup. Although Nasir Abas had not participated in any bombings, he was well respected, having been a trainer at the As Saadah camp on the Pakistan side of the border during the Soviet-Afghan war and having set up Camp Hudaibiyah, JI’s training camp in Mindanao. He says he opposed Osama Bin Laden’s 1998 fatwa legitimising the killing of civilians, but he believed in the use of military force to fight against the oppression of Muslims. Ali Imron was a highly respected Islamic scholar with jihad experience in Afghanistan and Ambon. He participated in several bombings, including the 2000 attack on the Philippine Ambassador’s residence, the 2000 Christmas Eve bombing, and the 2002 Bali bombing. He took a different view than Abas, contending that the bombers’ interpretation of jihad was actually correct, but they had acted precipitously without assessing whether they had popular support, the necessary strength to take on the state, a secure base from which to mount operations, whether the Muslim community would benefit from their actions, or whether they could achieve the same goals via religious outreach (Imron, 2007). It was this nuanced perspective, among other factors, that convinced key JI figures and contributed to the emergence of JI’s dominant anti-bombing-in-Indonesia narrative (International Crisis Group, 2007). The International Crisis Group explained the programme from the perspective of the police as follows: The police took a major gamble in giving [Abas and Imron] access to other detainees to engage in informal debates and encourage discussion about what was right and wrong in their approach to jihad. The police understood at the outset that any debates about right and wrong tactics had to take place within the movement itself. Jihadis were not going to listen

449

Julie Chernov Hwang

to moderates outside their circle. . . . The goal of the police in giving Ali Imron and Nasir Abas access to new detainees was to pick off important leaders of JI in the belief that given JI’s hierarchical structure, if the leader changed his mind, others would follow. (International Crisis Group, 2007) Within the parameters of this approach, Nasir Abas had a higher profile because he could freely move about the country and meet with jihadists both inside and outside prison. Ali Imron’s scope of influence was largely confined to his fellow terrorist prisoners serving time in the POLDA Metrojaya detention centre. In tandem with discussions held by Ali Imron and Nasir Abas, Densus 88 also put in place its own ‘soft approach’, where humane treatment, interaction, and the building of relationships were used to undermine the narrative that the police were inhuman and un-Islamic. When Surya Dharma was Densus chief, he led prayer sessions in prisons, and interrogation sessions were stopped for prayers (Abuza, 2009). During Ramadan, Densus members shared the iftar (break the fast) meal with the prisoners. On several occasions, Surya Dharma also took detainees at POLDA on outings to visit orphanages and scavengers to show how one could be a good Muslim and contribute to the welfare of the Muslim community without the use of violence.10 Tito Karnavian explained that after studying up on the background of the arrested jihadist, if he was amenable to cooperating, they would seek to build a relationship of trust where, in time, the detainee would feel comfortable sharing their problems and concerns (Karnavian, 2009). Certain jihadists were also singled out for special treatment, if their cases were high priority, if they cooperated with authorities, or if they showed promise. In a handful of instances, for example, Densus allowed jihadists to marry their girlfriends in ceremonies at the detention centre with family flown in for the occasion. Such humane treatment had an impact on certain jihadists. However, these programmes and policies were inconsistently applied, and torture was used far too often, leading to distrust of authorities. The issues with the soft approach lay not in the idea itself but with improper measurement of success, insufficient resources to build a proper programme, and inconsistent enforcement of norms of behaviour (Chernov Hwang, 2018). When asked about the success of the soft approach in 2010, Tito Karnavian estimated that of the 200 whom Densus 88 approached, 100 were cooperative (Abuza, 2009; Karnavian, 2009). The language Karnavian used highlights one of the core issues with the programme. The soft approach was never actually a disengagement programme, irrespective of how it was portrayed in the international media.11 The top priority of the ‘soft approach’ was to gather intelligence on Jemaah Islamiyah and its affiliates to better assist in heading off terrorist attacks and to understand the inner workings of the network. In short, the goal was to facilitate cooptation, not reintegration. Disengagement was a positive byproduct rather than the focal point. The conflation of cooptation with disengagement inhibited Densus from doing a systematic analysis of successful cases of disengagement and then building upon their best practices. Second, the programme was ad hoc, under-resourced, and underfunded. In contrast to the teams of social workers, religious scholars, and psychologists staffing programmes in Singapore and Saudi Arabia, a small sub-group of Densus officers, together with Nasir Abas and Ali Imron, comprised the entirety of the cooptation-cum-disengagement effort. Tito Karnavian lamented the scarcity of resources available for counter-terrorism: The police team had a tiny operating budget. This inhibited the optimum outcome. There were many things that could have been done that were beyond our capacity, for example, 450

Deradicalisation and disengagement

paying for the school fees for the children of imprisoned radicals or helping their families after imprisonment. (Karnavian, 2009) Another shortcoming of the soft approach was its inconsistent usage. Despite the obvious benefits of humane treatment and its use at the national level at POLDA and for those who surrendered to the police, the default behaviour of local police and local Densus branches was often torture without kindness (Chernov Hwang, 2012). In repeated interviews, it was stated that those who surrendered were treated humanely, while those who were arrested by the authorities, especially local authorities, faced up to a week of torture at the hands of their interrogators (Chernov Hwang, 2018). The use of torture led some, who may have been ripe for disengagement, to be susceptible to further radicalisation due to their treatment at the hands of the police. In the small-scale bombings that took place between 2009 and 2013, police were a frequent target (Chernov Hwang, 2018).

Indonesian government programmes post-2010: the role of the BNPT The Indonesian National Counter-Terrorism Agency (BNPT) was founded in July 2010, almost a year to the date following the 2009 bombings of the Marriot and Ritz-Carlton Hotels. It was tasked with coordinating among government agencies in three areas: intelligence gathering, monitoring, and operations; prevention and deradicalisation; and international cooperation. Densus 88 was incorporated into the BNPT, which, in theory, would have provided a wealth of experience upon which to draw. The BNPT has been quite ineffective, however. Part of the issue at the BNPT is one of manpower. At the formation of the BNPT, the military demanded a role for itself and was tasked with prevention and deradicalisation. Those law enforcement officers with experience, relationships, and direct field knowledge joined the intelligence, monitoring, and operations division, which left prevention and deradicalisation to newcomers who lacked experience in these areas (Institute for the Policy Analysis of Conflict, 2014). There were also problems with coordination. Counter-terrorism police, who had spent the better part of the past decade building up contacts and gathering information, were reluctant to share it (Institute for the Policy Analysis of Conflict, 2014). Had they been freer in sharing their information, it may have helped the prevention and deradicalisation division. However, it must also be noted that prevention and deradicalisation also neglected to do their own research. Had they reviewed the interrogation reports and court transcripts, they could have better understood the process of radicalisation, and, thus, both counter-radicalisation and disengagement programmes could have been better conceived (Institute for the Policy Analysis of Conflict, 2014). BNPT has also been reluctant to coordinate with the Department of Corrections (DGC), viewing it as part of the problem given Indonesia’s notoriously corrupt prison system. Yet, prison reform would be one component of a comprehensive disengagement effort, which would, in turn, necessitate such coordination. On the national level, the military took steps to justify its involvement in the BNPT. Tasked with prevention and deradicalisation, the latter has received short shrift. Moreover, they filtered their understanding of terrorism through the prism of their organisational identity. A 2010 strategic plan blamed terrorism on a weak sense of nationhood, weak citizenship education, the erosion of local values via modernisation, and a fanatical understanding of religion (Institute for the Policy Analysis of Conflict, 2014). As the Indonesian military sees itself as the guardian of nationhood, such a perspective validates its own presence, but it is at odds with the reality of radicalisation as a social experience where kinship ties, teacher-student ties, and the bonds formed in entry-level 451

Julie Chernov Hwang

study circles are often more binding than ideology (Chernov Hwang, 2023). Furthermore, it ignores how a politicised local conflict can engender radicalisation (Chernov Hwang et al., 2013; Chernov Hwang & Schulze, 2018). This misinterpretation had negative spillover effects on deradicalisation programme design, as initiatives did not address the needs of the target population or show sufficient understanding of that population. BNPT deradicalisation efforts are frequently criticised as poorly conceived and implemented, disconnected from the actual needs of the target population. In interviews with civil society actors working on disengagement, several common points were raised. First, BNPT programmes are too top-down, both in organisation and in style.12 Second, they do not do needs assessments among target populations prior to embarking on programming to see what types of programmes would be best received, nor do they examine what programming infrastructure already exists.13 Third, programmes are not sustained over time in a manner that builds trust between the staff and the participants.14 Fourth, after programmes are conducted, the BNPT conducts an outcome assessment to see if there is any change in the participants.15 Finally, aftercare is still being neglected.16 The Wayang Kulit shadow puppetry performances held in prisons in 2011 are an example of this weakness. The planners at the BNPT believed that seeing the non-violence-themed Wayang Kulit shadow puppet show would help convince the terrorists to return to their Javanese heritage. This shows a lack of understanding of the mechanisms behind disengagement and ideological reconsideration. When the terrorist detainees seemed unreceptive to the show, the programme was temporarily shelved and later reimagined as a counter-radicalisation programme for youths with the same justification. A more appropriate December  2013 initiative involved bringing in three well-respected hardline clerics from the Middle East: Ali Hasan al-Halabi, a Salafi cleric from Jordan; Dr. Najih Ibrahim, one of the founders of Islamic Jihad in Egypt; and Hisyam al Najjar, a former member of the Jihad Group who is now active in a Salafi political party. All three opposed violence against civilian targets as well as the declaration of takfir against other Muslims. The three clerics held short courses for terrorist detainees in Cipinang Prison in Jakarta as well as Pasir Putih Prison on Nusa Kambangan Island, where some of the most high-risk terrorists are held, in addition to participating in a conference at the University of Indonesia and sitting for a TV interview.17 While the detainees enjoyed the opportunity to engage the visiting clerics in a lively conversation, the discussion mirrored many points that were already taking place within the JI community itself. Drawing on the writings of Abu Musab al-Suri, who advocated assessing each potential activity undertaken based on its permissibility according to Islamic law, its benefit to the Muslim community and jihad movement, and whether the group possessed the capacities to successfully carry it out, JI leaders applied cost-benefit analysis to the terror attacks in Indonesia and decided the costs to the organisation were far too high.18 Many of these same leaders participated in the discussions with the Egyptian and Jordanian clerics. This thus raises the question of whether the visits actually contributed to disengagement or the reconsideration of previously held views. Rather than bring in Middle Eastern scholars, it may have been better to simply build upon what was already present and encourage these discussions and assessments within the Indonesian jihadist community, both in prison and outside it (Chernov Hwang, 2018). Another programme that has been beset with issues since its conception is the deradicalisation centre at Sentul. One problem has been that different agencies with vested interests in deradicalisation had vastly differing opinions on how the facility should best be used. On the one hand, the BNPT wanted to use it instead as a showcase for its most cooperative prisoners (Institute for the Policy Analysis of Conflict, 2016). On the other hand, the Department of Corrections wanted to 452

Deradicalisation and disengagement

use it as a facility for the most unrepentant terrorists (Institute for the Policy Analysis of Conflict, 2016). One prison official asserted, It’s no good for us if the cooperative prisoners are all moved to Sentul and we’re left with the hardcore. Isn’t the task of the BNPT to deradicalize the ISIS extremists? If the inmates are already not so radical, why are they being deradicalized? (Institute for the Policy Analysis of Conflict, 2016) While the Department of Corrections eventually gave into the BNPT, the BNPT plan evoked considerable ire from prison officials, who felt their job would be far more difficult as the BNPT was taking the most moderate prisoners and leaving them with the most intransigent pro-ISIS ones (Institute for the Policy Analysis of Conflict, 2016). In short, going back to the initial counter-terrorism efforts in the aftermath of the first Bali bombing, disengagement and aftercare have been given short shrift. Early initiatives by Densus 88 had a good base in understanding the jihadists and their needs, but they were ad hoc and underfunded, and humane treatment was inconsistently used. Promising local programmes ran into problems of implementation, for higher-ups demanded the funding be used up within a year, which led to a rush to find participants and carry out the programme irrespective of whether they were actually targeting those who needed the programmes most. BNPT’s efforts at deradicalisation that followed ran into problems in coordination, fighting between the police and military, and a general lack of knowledge of how to do the extraordinarily large job with which they were tasked. As a result, Indonesia still does not have a disengagement programme to speak of at the national level. However, it is important to note that the jihadists are still disengaging, irrespective of the degree of assistance they are receiving from the state. Moreover, there are reservoirs of knowledge and expertise within the BNPT, the Department of Corrections, the jihadi community, and civil society that can be tapped to improve upon the initial efforts.

Efforts among disengaged jihadists to facilitate disengagement and reintegration The most significant programme in this arena is a mutual aid society established by Ali Fauzi, a former trainer with the extremist group Mujahidin KOMPAK, for veterans of the Afghanistan, Moro, Poso, and Ambon conflicts living in his hometown of Lamongan. Lamongan has gained a reputation as a hub of radicalism in Indonesia, first being the hometown of Mukhlas, Amrozi, and Ali Imron, three key figures in the 2002 Bali bombing, and, more than a decade later, for the existence of several pro-ISIS groups within the town. Recognising the need to assist the families of veterans of jihads in Afghanistan, Ambon, Poso, and Mindanao who were living in Lamongan, Ali Fauzi started a mutual aid society. He identified about a dozen families in need of economic support. Their activities included a joint goat breeding initiative, giving small loans to members facing imminent cash shortages, and an informal religious study group where they could examine their prior religious beliefs and explore Islamic law more fully.19 If a family lacked the funds to pay their children’s school fees, Ali Fauzi arranged for them to attend Al Islam, his family’s boarding school, at a heavily subsidised rate. By 2014, the mutual aid society had begun to reach out to the families of detained terrorists from Lamongan, offering their children subsidised education at Al Islam, inviting their wives to join the goat breeding cooperative, and facilitating discussions where individuals with relatives in jail or fighting in Syria could express their views and their needs without fear.20 By empowering the 453

Julie Chernov Hwang

wives economically, it led those women to become more assertive in sharing their views, including opposition to ISIS.21 This is important, for, wives and mothers can have a discernable impact on their husband’s/children’s decision to disengage and reintegrate. Perhaps most important, the mutual aid society has provided an alternative social network of sorts for individuals who have decided to disengage and become more open in their perspective. They need not fear ostracism from the jihadi community on the whole because the mutual aid society constitutes a set of friends who support their decision. This provides a source of empowerment for facilitating reintegration. This mutual aid society in Lamongan is an important example of how disengagement and reintegration can be assisted and even facilitated through careful and targeted programming. Such programming points to the importance of recruiting the right people with legitimacy among those in the radical community, men like Ali Fauzi, who have jihad experience and religious training, who are respected in the network, who can empathise with their experiences and speak to the right touchstones, and who understand the journey on which they are embarking (Chernov Hwang, 2018). These can not only affect how one thinks but also, and in fact more importantly, provide a safe social network in which to reconsider previously held views and gain social, financial, and material assistance.

Civil society initiatives in Indonesia In addition to government programmes and initiatives coming from the disengaging jihadists themselves, we also find several initiatives coming out of local and international civil society. The most notable are the Indonesian Alliance for Peace (AIDA), Research Center for Police Studies (PRIK), the Institute for International Peace Building (YPP), and Civil Society Against Violent Extremism (C-SAVE). It is important to note at the outset that these are small-scale programmes. One point of consensus that has emerged from all programmes is that terrorist detainees do not want to discuss religious ideology with outsiders. They are far more open to lifeskills training and professional development programmes. This aligns with findings from the Sabaoon programme in Pakistan as well as the Arhaus program in Denmark, both of which have had success in disengagement and reintegration efforts combining professional development with counselling and eschewing efforts at worldview overhaul. AIDA developed a programme to facilitate dialogues between convicted terrorists and the victims of terrorist attacks (Institute for the Policy Analysis of Conflict, 2016). As of April 2017, AIDA had conducted one-on-one meetings in five Indonesian prisons (Sumpter, 2017). These have tended to be very emotional conversations; several have led to reconciliation between the victims and the bombers (Institute for the Policy Analysis of Conflict, 2016). However, few, if any, have involved ISIS inmates (Institute for the Policy Analysis of Conflict, 2016). The University of Indonesia’s Research Center for Police Studies (PRIK) is another organisation that runs programmes both in prisons and with parolees. Initially, PRIK’s head, then Sarlito Sarwono, attempted deradicalisation programmes, inviting religious scholars to converse with the prisoners, but they eventually realised these programmes were not productive; the militants were not open to those arguments (Sumpter, 2017). Thus, they moved to a different approach, building rapport with the prisoners and parolees by discussing their families and personal problems (Sumpter, 2017). In time, they moved on to discuss broader issues such as corruption, governance, social welfare, and eventually the counter-productivity of violence to achieve their goals (Sumpter, 2017). Another programme of note that works both inside and outside prisons is the Institute for International Peacebuilding (YPP). YPP was founded by Noor Huda Ismail, a graduate of the Al Mu’mim Islamic boarding school and former Washington Post journalist, and Taufik Andrie, a 454

Deradicalisation and disengagement

former journalist. It began as a somewhat ad hoc initiative where Andrie and Ismail reached out to their former journalist contacts among the jihadists to see how they could help them reintegrate successfully. Taufik Andrie explained, ‘We began with a needs assessment that we conducted between 2006 and 2008 and in 2009, we began formally helping two people to set up businesses with an end goal of facilitating successful rehabilitation’.22 Andrie explained their method as a three-step process. First, if someone sought their assistance, they did research on the militant to learn as much as possible about their case and background. Then, they approached the militant to build a relationship, calling on a weekly basis and visiting them once every two months. In doing so, they came to understand their goals and needs. Third, they endeavoured to build loyalty and trust in that relationship by fulfilling their needs – assisting with children’s school fees, for example – and helping them, upon release, put together a business plan, find seed capital for a small business, or obtain professional training. Andrie contends that this process often took up to two years. In time, as trust was built, affirmed, and reaffirmed through their rehabilitation programme, in certain instances they may begin to challenge their ideas on jihad or the Islamic state.23 However, YPP’s operations in this regard were fairly small in scope. Between 2008 and 2023, they assisted approximately 80 persons, including, most recently, some 12 juveniles former terrorists.24 The number was low because the process of helping the jihadists is quite labour-intensive.

State-civil society initiatives in Indonesia: C-SAVE, YPP, and the ISIS returnees The most notable state-civil society initiative has come about as Indonesia has struggled to deal with ISIS deportees and returnees. In January 2017, the Turkish government deported some 75 Indonesian nationals, 70% of whom were women and children; dozens more followed after the initial batch (Sumpter, 2018). The Indonesian government found itself overwhelmed by the influx. Those who were not incarcerated, overwhelmingly women and children, were sent to one of two centres: Panti Sosial Marsudi Putra Handayani for women with children and Rumah Perlindungan Trauma Center, if one was childless, both run by the Ministry of Social Affairs (Institute for the Policy Analysis of Conflict, 2016). There, they were required to take part in 30-day rehabilitation programmes before being released back to their villages. The problem was that the Ministry for Social Affairs had little to no experience in how to run these programmes. Therefore, C-SAVE, a coalition of Indonesia-based terrorist rehabilitation NGOs, and YPP offered assistance and advice, giving them a crash course in Salafi-Jihadi ideology, cautioning them against talking about religion, and encouraging staff to simply talk about daily issues and families (Sumpter, 2018). With this approach, some deportees began to communicate. However, the programme has run into several issues due to an insufficient budget and the 30-day time frame (Sumpter, 2018). For example, the Ministry did not have funds earmarked for bus tickets and reintegration assistance (Sumpter, 2018). Additionally, within the centres themselves, unrepentant pro-ISIS women and children may seek to re-radicalise or further radicalise those who may be having doubts. Once they returned to their villages, the task of reintegration assistance fell to local NGOs, YPP, and C-SAVE. Some local NGOS have had success with a small number of economic empowerment projects for women (Sumpter, 2017). However, there were also a considerable number of deportees who felt bitter at being forcibly repatriated, refused to engage with those who did not share their worldview, and remained estranged from their family of origin, considering them kafir (infidels) (Sumpter, 2017). All of this points to the importance of improving and upgrading aftercare capacities. 455

Julie Chernov Hwang

Conclusion Indonesia’s experience with deradicalisation and disengagement programmes can offer lessons for other countries in developing and improving their own programmes. First, the most successful programmes have been focused on lifeskills training, professional development, and the creation of support networks. These programmatic interventions provide people with a basic foundation to facilitate gainful employment and possibly, eventually, reintegration. The Indonesian experience has shown that militants may resist being lectured about their ideological views. However, when programmes see to their basic needs, it enables rapport building that can, in itself, eventually challenge some of those previously held views. Second, the government may not be best suited to provide those programmatic interventions. Instead, state-civil society partnerships and programmes run by civil society organisations may be better. These programmes, especially those centred in specific localities, can adapt to changing circumstances more easily than government programmes. They can take the time to do research on those they are helping and to build ties with their families. They can invest a year or two in a person and not be pressured to show results at the end of a fiscal year. Third, those programmes run by the government would do best to adopt best practices from civil society: needs assessments at the start; outcomes assessments at the end; research into the target populations and rapport building; showing adaptability and agility in programme design and adjustment; and understanding that disengagement and deradicalisation are slow processes where results don’t conform to the fiscal year. Moreover, these programmes are too important to be politicised or underfunded. Finally, former members of extremist groups are a valuable asset in facilitating disengagement. One of the best small-scale Indonesian programmes is run by a former member of an extremist group. His programme is noted especially for its agility and its understanding of the needs of the vulnerable populations in the town in which it operates. The goal has remained constant: to facilitate disengagement by helping veterans of the Afghan, Mindanao, Ambon, and Poso jihads earn an income, provide for their basic needs, and create a safe support network of like-minded individuals. The utility of formers is not just as employees of existing government programmes but also as leaders of their own programmes. This is not to say that every former can run their own programme, but there will be some. They show themselves trustworthy through their own disengagement and reintegration trajectories and through the work they do to help others follow their example. This finding that professional development should be prioritised over ideological deradicalisation runs contrary to the programmatic efforts pioneered by the Yemenis, the Saudis, the Singaporeans, and the Malaysians that combine ideological revision and professional development. However, they echo the findings of the Arhaus initiative in Denmark as well as Sabaoon in Pakistan, both of which have had success deprioritising ideological change in favour of lifeskills training, job training, and counselling. However, the finding that formers can be a vital part of programme design and development is new to the literature on deradicalisation programming in the Muslim world. This is a valuable addition to the conversation.

Notes 1 For authors who refer to ‘ideology’ as part of their definition of deradicalisation, please see Horgan, J., & Altier M. B. (2012, Summer–Fall). The future of terrorist deradicalization programs. Georgetown Journal of International Affairs, 13(2), 86; Altier, M. B., Thoroughgood, C., & Horgan, J. (2014). Turning away from terrorism: Lessons from psychology, sociology and criminology. Journal of Peace Research, 51(5), 647; Doosje, B., Moghaddam, F. M., Kruglanski, A. W., De Wolf, A., Mann, L., & Feddes, A. R. (2016).

456

Deradicalisation and disengagement Terrorism, radicalization and deradicalization. Current Opinion in Psychology, 11, 82; Islam, M. (2019, May). Deradicalization of terrorists: Theoretical analysis and case studies. Counter-Terrorists Trends and Analysis, 11(5); Ashour, O. (2009). Deradicalization of Jihadists in Egypt and Algeria. London: Routledge; Rabasa, A., Pettyjohn, S., Ghez, J., & Boucek, C. (2010). Deradicalizing Islamist extremists (p. xiii). RAND Corporation; Braddock, K. (2014). The talking cure? Communication and psychological impact in prison deradicalization programs. In A. Silke (Ed.), Prisons, terrorism and extremism: Critical issues in management, radicalization and reform (p. 60). London: Routledge; Morrison, J., Silke, A., Malberg, H., Slay, C., & Stewart, R. (2021, August). A systematic review of post-2017 research on disengagement and deradicalization. CREST; Koehler, D., & Fiebig, V. (2019, June). Understanding of how to counter violent radicalization. Perspectives on Terrorism, 13(3), 47. 2 For scholars who refer to belief as part of their definition of deradicalization, please see Islam, M. (2019, May). Deradicalization of terrorists: Theoretical analysis and case studies. Counter-Terrorists Trends and Analysis, 11(5); Rabasa, A., Pettyjohn, S., Ghez, J., & Boucek, C. (2010) Deradicalizing Islamist extremists (p. xiii). RAND Corporation; Morrison, J., Silke, A., Malberg, H., Slay, C., & Stewart, R. (2021, August). A systematic review of post-2017 research on disengagement and deradicalization. CREST; Koehler, D., & Fiebig, V. (2019, June). Understanding of how to counter violent radicalization. Perspectives on Terrorism, 13(3), 4. 3 For a discussion of the difference between disengagement and deradicalisation, see Della Porta, D., & LaFree, G. (2011). Guest editorial: Processes of radicalization and deradicalization. International Journal of Conflict and Violence, 6(1); Chernov Hwang, J. (2018). Why terrorists quit: The disengagement of Indonesian Jihadists. Cornell University Press. 4 For scholars who refer to behaviour as part of their definitions of disengagement, please see Chernov Hwang, J. (2018). Why terrorists quit: The disengagement of Indonesian Jihadists (p. 4). Cornell University Press; Altier, M. B., Thoroughgood, C., & Horgan, J. (2014). Turning away from terrorism: Lessons from psychology, sociology and criminology. Journal of Peace Research, 51(5), 647; Islam, M. (2019, May). Deradicalization of terrorists: Theoretical analysis and case studies. Counter-Terrorists Trends and Analysis, 11(5); Morrison, J., Silke, A., Malberg, H., Slay, C., & Stewart, R. (2021, August). A systematic review of post-2017 research on disengagement and deradicalization. CREST; Koehler, D. (2017). Understanding deradicalization (p. 3). Routledge. 5 Ashour, O. (2010). Deradicalization of Jihad? The impact of Egyptian Islamist revisionists on Al Qaeda. Perspectives on Terrorism, 2(5); Ashour, O. (2009). Deradicalization of Jihadists in Egypt and Algeria. Routledge; Matesan, I. (2020). The violence pendulum. Oxford University Press. 6 See Koehler, D. (2017). Understanding deradicalization (p. 3). Routledge; Altier, M. B., Thoroughgood, C., & Horgan, J. (2014). Turning away from terrorism: Lessons from psychology, sociology and criminology. Journal of Peace Research, 51(5), 647; Horgan, J., & Altier, M. B. (2012, Summer–Fall). The future of terrorist deradicalization programs. Georgetown Journal of International Affairs, 13(2), 86; Horgan, J., & Braddock, K. (2010). Rehabilitating the terrorists. Terrorism and Political Violence, 22, 280; Braddock, K. (2014). The talking cure? Communication and psychological impact in prison deradicalization programs. In A. Silke (Ed.), Prisons, terrorism and extremism: Critical issues in management, radicalization and reform (p. 60). Routledge; Doosje, B., Moghaddam, F. M., Kruglanski, A. W., De Wolf, A., Mann, L., & Feddes, A. R. (2016). Terrorism, radicalization and deradicalization. Current Opinion in Psychology, 11; Islam, M. (2019). Deradicalization of terrorists: Theoretical analysis and case studies. Counter-Terrorists Trends and Analysis, 11(5); Rabasa, A., Pettyjohn, S., Ghez, J., & Boucek, C. (2010). Deradicalizing Islamist extremists (p. xiii). RAND Corporation; Morrison, J., Silke, A., Malberg, H., Slay, C., & Stewart, R. (2021, August). A systematic review of post-2017 research on disengagement and deradicalization. CREST; Horgan, J., & Bjorgo, T. (2009). Introduction. In J. Horgan & T. Bjorgo (Eds.), Leaving terrorism behind. Routledge; Bjorgo, T. (2009). Processes of disengagement from violent groups of the extreme right. In J. Horgan & T. Bjorgo (Eds.), Leaving terrorism behind. Routledge. 7 See Altier, M. B., Thoroughgood, C., & Horgan, J. (2014). Turning away from terrorism: Lessons from psychology, sociology and criminology. Journal of Peace Research, 51(5), 647; Horgan, J., & Altier, M. B. (2012, Summer–Fall). The future of terrorist deradicalization programs. Georgetown Journal of International Affairs, 13(2), 86; Horgan, J., & Braddock, K. (2010). Rehabilitating the terrorists. Terrorism and Political Violence, 22, 280; Braddock, K. (2014). The talking cure? Communication and psychological impact in prison deradicalization programs. In A. Silke (Ed.), Prisons, terrorism and extremism: Critical issues in management, radicalization and reform (p. 60). Routledge; Horgan, J., & Bjorgo, T. (2009). Introduction. In J. Horgan & T. Bjorgo (Eds.), Leaving terrorism behind. Routledge; Bjorgo, T. (2009).

457

Julie Chernov Hwang Processes of disengagement from violent groups of the extreme right. In J. Horgan & T. Bjorgo (Eds.), Leaving terrorism behind. Routledge. 8 See Doosje, B., Moghaddam, F. M., Kruglanski, A. W., De Wolf, A., Mann, L., & Feddes, A. R. (2016). Terrorism, radicalization and deradicalization. Current Opinion in Psychology, 11; Islam, M. (2019, May). Deradicalization of terrorists: Theoretical analysis and case studies. Counter-Terrorists Trends and Analysis, 11(5); Rabasa, A., Pettyjohn, S., Ghez, J., & Boucek, C. (2010). Deradicalizing Islamist extremists (p. xiii). RAND Corporation; Morrison, J., Silke, A., Malberg, H., Slay, C., & Stewart, R. (2021, August). A systematic review of post-2017 research on disengagement and deradicalization. CREST; Koehler, D., & Fiebig, V. (2019, June). Knowing what to do: Academic and practitioner understanding of how to counter violent radicalization. Perspectives on Terrorism, 13(3), 47. 9 Abuza, Z. (2009). The rehabilitation of Jemaah Islamiyah detainees in Southeast Asia: A preliminary assessment. In T. Bjorgo & J. Horgan (Eds.), Leaving terrorism behind: Individual and collective disengagement (pp. 193–211). Routledge; Boucek, C. (2009). Extremist reeducation and rehabilitation in Saudi Arabia. In T. Bjorgo & J. Horgan (Eds.), Leaving terrorism behind: Individual and collective disengagement (pp. 212–223). Routledge; Boucek, C., Beg S., & Horgan, J. (2009). Opening up the Jihadi debate: Yemen’s committee for dialogue. In T. Bjorgo & J. Horgan (Eds.), Leaving terrorism Behind: Individual and collective disengagement (pp. 181–192). Routledge; International Crisis Group. (2007, November 19). ‘Deradikalisasi’ dan Lembaga Pemasyarakatan di Indonesia. Asia Report, 124; Schulze, K. (2008). Indonesia’s approach to Jihadist deradicalization. CTC Sentinel, 1(8); Karnavian, T. (2009). The “soft approach” strategy in coping with Islamist terrorism in Indonesia (Working Paper). Asian Association of Social Psychology; Khan, S. (2015). Deradicalization programming in Pakistan. US Institute of Peace. 10 These outings were cited by a handful of interviewees who served time at POLDA Metro. 11 Interview with Tito Karnavian, July 2010, Jakarta, Indonesia 12 Interviews with Agus Nahrowi, Search for Common Ground, Zora Sukabdi, Former Executive Director, INSEP, Taufik Andrie, Executive Director, Institute for International Peacebuilding, July 2014, Jakarta, Indonesia. 13 Interview with Zora Sukabdi, Former Executive Director of INSEP, July 2014, Jakarta, Indonesia 14 Interviews with Agus Nahrowi, Search for Common Ground, Zora Sukabdi, Executive Director, INSEP, Taufik Andrie, Executive Director, Institute for International Peacebuilding, July 2014, Jakarta, Indonesia. 15 Ibid. 16 Interview with Taufik Andrie, Executive Director of the Institute for International Peacebuilding, July 2023, Jakarta, Indonesia. 17 Ibid. 18 Jones, S. (Forthcoming). The rise and fall of Jemaah Islamiyah. In G. Fealy & S. White (Eds.), Indonesian terrorism (p. 22). Routledge. 19 Ibid. 20 Ibid., p. 18. 21 Ibid. 22 Interview, Taufik Andrie Susanto, Executive Director of the Institute for International Peacebuilding, July 2023, Jakarta, Indonesia. 23 Ibid. 24 Ibid.

References Abuza, Z. (2009). The rehabilitation of Jemaah Islamiyah detainees in Southeast Asia: A preliminary assessment. In T. Bjorgo & J. Horgan (Eds.), Leaving terrorism behind: Individual and collective disengagement (pp. 193–211). Routledge. Altier, M. B., Thoroughgood, C., & Horgan, J. (2014). Turning away from terrorism: Lessons from psychology, sociology and criminology. Journal of Peace Research, 51(5). https://doi.org/10.1177 %2F0022343314535946 Ashour, O. (2009). Deradicalization of Jihadists in Egypt and Algeria. Routledge. Ashour, O. (2010). Deradicalization of Jihad? The impact of Egyptian Islamist revisionists on Al Qaeda. Perspectives on Terrorism, 2(5), 11–14.

458

Deradicalisation and disengagement Aslam, M., Othman, I., & Rosili, N. (2016, June). Deradicalization programs in Southeast Asia: A comparative study of rehabilitation programs in Malaysia, Thailand, Indonesia and Singapore. Journal of Education and Social Sciences, 4, 154–160. Bjorgo, T. (2009). Processes of disengagement from violent groups of the extreme right. In J. Horgan & T. Bjorgo (Eds.), Leaving terrorism behind (pp. 30–48). Routledge. Bjorgo, T. (2011). Dreams and disillusionment. Crime, Law and Social Change, 55(4), 277–285. Boucek, C. (2009). Extremist reeducation and rehabilitation in Saudi Arabia. In T. Bjorgo & J. Horgan (Eds.), Leaving terrorism behind: Individual and collective disengagement (pp. 212–223). Routledge. Boucek, C., Beg, S., & Horgan, J. (2009). Opening up the Jihadi debate: Yemen’s committee for dialogue. In T. Bjorgo & J. Horgan (Eds.), Leaving terrorism behind: Individual and collective disengagement (pp. 181–192). Routledge. Braddock, K. (2014). The talking cure? Communication and psychological impact in prison deradicalization programs. In A. Silke (Ed.), Prisons, terrorism and extremism: Critical issues in management, radicalization and reform (pp. 60–74). Routledge. Chernov Hwang, J. (2012). Terrorism in perspective: An assessment of Jihad project trends. Asia-Pacific Issues, 104, 1–12. Chernov Hwang, J. (2017). The disengagement of Indonesian Jihadists: Understanding the pathways. Terrorism and Political Violence, 29(2), 277–295. Chernov Hwang, J. (2018). Why terrorists quit: The disengagement of Indonesian Jihadists. Cornell University Press. Chernov Hwang, J. (2023, forthcoming). Becoming Jihadis: Radicalization and commitment in Southeast Asia. Oxford University Press. Chernov Hwang, J., Panggabean, R., & Ali Fauzi, I. (2013). The disengagement of Jihadis in Poso, Indonesia. Asian Survey, 53(4), 754–777. Chernov Hwang, J., & Schulze, K. (2018). Why they join: Pathways into Indonesian Jihadist organizations. Terrorism and Political Violence, 30(6), 911–932. Della Porta, D., & LaFree, G. (2011). Guest editorial: Processes of radicalization and deradicalization. International Journal of Conflict and Violence, 6(1), 4–10. Dennison, M. (2010, April). Reintegrating ex combatants into civilian life. Peace and Change, 35(2), 328–352. Doosje, B., Moghaddam, F. M., Kruglanski, A. W., De Wolf, A., Mann, L., & Feddes, A. R. (2016). Terrorism, radicalization and deradicalization. Current Opinion in Psychology, 11, 79–84. Fealy, G., & Borgu, A. (2005). Local Jihad. Australia Strategic Policy Institute. www.aspi.org.au/report/ local-jihad-radical-islam-and-terrorism-indonesia Gjelsvik, I., & Bjorgo, T. (2012). Ex pirates in Somalia: Processes of engagement, disengagement and reintegration. Journal of Scandinavian Studies of Criminology and Crime Prevention, 13(2), 1–21. Horgan, J. (2014). What makes a terrorist stop being a terrorist? Journal for Deradicalization, (1), 1–4. Horgan, J., & Altier, M. B. (2012, Summer–Fall). The future of terrorist deradicalization programs. Georgetown Journal of International Affairs, 13(2), 83–90. Horgan, J., & Bjørgo, T. (2009). Introduction. In J. Horgan & T. Bjørgo (Eds.), Leaving terrorism behind (pp. 1–13). Routledge. Horgan, J., & Braddock, K. (2010). Rehabilitating the terrorists. Terrorism and Political Violence, 22, 267–291. Imron, A. (2007). Ali Imron Sang Pengebom. Republika. International Crisis Group. (2007). “Deradikalisasi” dan Lembaga Pemasyarakatan di Indonesia. Asia Report, 124. www.crisisgroup.org/id/asia/south-east-asia/indonesia/deradicalisation-and-indonesian-prisons Institute for the Policy Analysis of Conflict. (2014). Countering violent extremism: The need for a rethink (Report No. 11, pp. 1–21). Institute for the Policy Analysis of Conflict. Institute for the Policy Analysis of Conflict. (2016). Update on Indonesian Pro-ISIS prisoners and deradicalization efforts (Report No. 34, pp. 1–23). Institute for the Policy Analysis of Conflict. Islam, M. (2019, May). Deradicalization of terrorists: Theoretical analysis and case studies. Counter-Terrorists Trends and Analysis, 11(5), 6–12. Jones, S. (forthcoming). The rise and fall of Jemaah Islamiyah. In G. Fealy & S. White (Eds.), Indonesian terrorism. Routledge. Karnavian, T. (2009). The “soft approach” strategy in coping with Islamist terrorism in Indonesia (Working Paper). Asian Association of Social Psychology.

459

Julie Chernov Hwang Kenney, M., & Chernov Hwang, J. (2021). Should I stay or should I go? Understanding how British and Indonesian extremists disengage and why they don’t. Political Psychology, 42(4), 537–553. Khan, S. (2015). Deradicalization programming in Pakistan. US Institute of Peace. www.usip.org/ publications/2015/09/deradicalization-programming-pakistan Koehler, D. (2017). Understanding deradicalization. Routledge. Koehler D., & Fiebig, V. (2019, June). Knowing what to do: Academic and practitioner understanding of how to counter violent radicalization. Perspectives on Terrorism, 13(3), 1–62. Matesan, I. (2020). The violence pendulum. Oxford University Press. McRae, D. (2010). Reintegration and localized conflict. Conflict, Security and Development, 10(3), 403–440. Morrison, J., Silke, A., Malberg, H., Slay, C., & Stewart, R. (2021, August). A systematic review of post-2017 research on disengagement and deradicalization. CREST. Rabasa, A., Pettyjohn, S., Ghez, J., & Boucek, C. (2010). Deradicalizing Islamist extremists (pp. 1–244). RAND Corporation. Reinares, F. (2011). Exit from terrorism: A qualitative empirical study on disengagement and deradicalization among members of ETA. Terrorism and Political Violence, 23(5), 780–803. Schulze, K. (2008). Indonesia’s approach to Jihadist deradicalization. CTC Sentinel, 1(8), 1–2. Sumpter, C. (2017). Countering violent extremism in Indonesia. Journal for Deradicalization, 11, 97–121. Sumpter, C. (2018, July). Returning Indonesian extremists (ICCT Policy Brief). https://icct.nl/app/ uploads/2018/07/ICCT-Sumpter-Returning-Indonesian-Extremists-July2018.pdf

460

30 INTERVENTIONS WITH ‘AT RISK’ INDIVIDUALS Raquel da Silva

Introduction Interventions with ‘at risk’ individuals have become, over the last decade, central to national and global counterterrorism strategies. On the one hand, Western democracies, headed by the European Commission, have propelled ‘a broad approach to upstream prevention’ (Coolsaet, 2016, p. 19). So much so that prevention is the first of the four strategic objectives composing the EU Counter-Terrorism Strategy adopted by the heads of state and government in 2005. On the other hand, preventive interventions have also been making their way outside Western democracies, mainly driven by the exportation of European approaches (e.g., see Crone & Nasser, 2018). These interventions are targeted at individuals deemed at high risk of engaging in, or actively supporting, terrorism or violent extremism but whose actions have not reached a criminal threshold. To use the language of public health, this kind of intervention could be considered a secondary intervention, sitting between primary interventions, such as education and counter-narrative programmes, and tertiary programmes to ‘exit’ or ‘de-radicalise’ those who have crossed thresholds into the criminal space or are actively involved in the radical milieus to which counterterrorism strategies seek to respond (Gielen, 2019). In some respects, interventions with ‘at risk’ individuals have built on existing tertiary mentoring programmes (Bjørgo & Horgan, 2009). Yet seeking to reduce the threat of terrorist violence before the threshold into physical violence and crime has been crossed makes them different in important ways. They usually operate on a different legal basis and are, at least ostensibly, voluntary (Institute for Strategic Dialogue, 2010). The fact that they are carried out in a pre-crime space raises questions about how the authorities identify who is ‘at-risk’, particularly when the science underpinning the concept of radicalisation remains subject to contestation (Heath-Kelly, 2013; Mythen et al., 2017; Pettinger, 2020), and about how researchers, practitioners, and policymakers might and do understand and evaluate the results of such interventions (Rabasa et al., 2010; Thornton & Bouhana, 2017). Given these questions, it is unsurprising that this extension of targeted counter-terrorist interventions into pre-crime spaces has proved contentious and divided academic and policy opinions. While supporters of such programmes have argued they remain essential to combating terrorist and extremist threats (Koehler, 2015), critics consider that interventions with ‘at risk’ individuals 461

DOI: 10.4324/9781003035848-34

Raquel da Silva

are based on problematic counterfactuals and represent a dangerous, discriminatory, and counterproductive extension of state power (Heath-Kelly, 2013; Pettinger, 2017). Yet our ability as societies to effectively evaluate and tease out the nuance of these competing claims has been severely limited due to the scarcity of systematic and independent research that examines either how interventions with ‘at risk’ individuals operate in practice, or the effects, both intended and unintended, they have on programme participants and their families (Horgan & Braddock, 2010; Malet, 2021). Such a state of affairs is due at least in part to the challenges associated with academics gaining access to such programmes and their participants (El-Said, 2015; Thornton & Bouhana, 2017). This chapter is based on existing literature produced by academics and practitioners on programmes that aim to intervene with ‘at risk’ individuals, combined with the author’s observations from four years of advising civil society organisations engaged in concrete initiatives in the United Kingdom. The choice of the two national contexts, the United Kingdom and Denmark, is connected to the fact that most available literature is focused on the United Kingdom (Tammikko, 2018), which is considered a pioneer and ‘norm producer’ in the field of preventing and countering violent extremism (P/CVE) policies in Europe (Directorate-General for Internal Policies, 2014, p. 13; Bonelli & Carrié, 2018; Skleparis & Augestad Knudsen, 2020). This country is closely followed by Denmark, which, in 2008, was appointed the lead country in the prevention of radicalisation and de-radicalisation programmes by EU Counter-Terrorism Coordinator Gilles de Kerchove (Lindekilde, 2015). In the following years, Denmark’s P/CVE model became known as the ‘Aarhus model’, since in 2014, this municipality witnessed a drastic decrease in the number of people travelling from Aarhus to Syria or Iraq (Skjoldager & Sheikh, 2014), which some authors believe can also be due to other factors unrelated to the city’s P/CVE model (see Tammikko, 2018). Nonetheless, the municipality of Aarhus, the second largest city in Denmark, has undoubtedly been a pioneer regarding the development of an extensive P/CVE programme, which has been an inspiration at home and abroad, as the widespread recognition of the model has led to its exportation to other countries, such as Lebanon (see Crone & Nasser, 2018) and Australia (Lindekilde, 2014a). However, it is important to note that, to date, even regarding these two national contexts, most literature concentrates on broader policy analysis and theorisation of the logics that supposedly underpin these programmes, rather than on their effects or even indeed on what they look like in practice (Jerome & Busher, 2020).

Conceptions of ‘risk’: how do individuals become perceived as ‘at risk’? The anticipation of the terrorist threat lies at the heart of P/CVE policies in Europe, comprising ‘a non-coercive attempt to reduce involvement in terrorism’ (van de Weert & Eijkman, 2020, p. 492). As radicalisation is seen as the process preceding terrorist violence (Kundnani & Hayes, 2018; Romaniuk, 2015), such policies are based, according to Skleparis and Augestad Knudsen (2020, p. 5), on three basic (and very much contested) ‘truths’ about radicalisation: (1) ‘radicalisation is a risk that can lead to terrorism’; (2) ‘terrorism can be stopped by rerouting or reversing radicalisation at an early stage’; and (3) ‘radicalisation is measurable’. In this scenario, detecting signs of radicalisation (or, in other words, ‘at risk’ individuals) is key within P/CVE efforts. In most countries, the responsibility for such detection has fallen not only on the security apparatus but also on frontline practitioners working at the local level (Crone, 2016). According to van de Weert and Eijkman (2020, p. 493), ‘the municipality has become an extension of the intelligence and security services, we are, in fact, asking frontline professionals to be informants for the security apparatus’. Therefore, ‘at risk’ individuals are spotted both through policing

462

Interventions with ‘at risk’ individuals

and intelligence work (Brooks, 2011) and through concerns expressed by healthcare and social workers, educators, or members of the community (Vidino & Brandon, 2012). However, research has demonstrated that frontline practitioners’ judgement regarding who is ‘at risk’ is ‘unstructured and not entirely objective’ (van de Weert & Eijkman, 2020, p. 493; see also Aly et al., 2015; Mattsson, 2018; Pettinger, 2020). This is partly due to the general lack of consensus at the conceptual level regarding radicalisation and its signs, leaving it to different professionals with different experiences, levels of knowledge, and identities to pass judgement (Lindekilde, 2014a). This judgement tends to be based on knowledge gained through P/CVE training programmes and informed by checklists and hand-outs created by governments and civil society organisations, as well as by the application of risk assessment tools. The following sections focus on each of these issues in turn.

P/CVE training programmes The joint efforts of police and frontline workers in spotting ‘at risk’ individuals are backed up by taking part in P/CVE training programmes, which can be found in different countries (see de Goede & Simon, 2013; Monaghan & Molnar, 2016). This section focuses on the UK context due to the vast research effort that it comprises in the P/CVE arena when compared to other contexts, particularly following the introduction of the Prevent Duty in 2015. In the United Kingdom, the first iteration of Prevent (2006–11) was about community-based work with Muslim communities, violent extremism, and police surveillance, while the second iteration of Prevent (2011–present) focuses on individuals at risk of, or vulnerable to, radicalisation, who have to be identified and referred to the anti-radicalisation mentoring scheme Channel (Thomas, 2017). From 2015 onwards, a legal duty was placed on frontline education, health, and welfare workers to implement Prevent and sought to equip them to do so through WRAP (Workshop to Raise Awareness of Prevent) training, which aims to foster awareness of extremist ideologies and signs of vulnerability to radicalisation (Blackwood et al., 2012). In this context, research conducted in the education (Busher et al., 2019; Jerome et al., 2019; Jerome & Busher, 2020; Lakhani, 2020; Moffat & Gerard, 2020; Sian, 2015; da Silva et al., 2021) and healthcare (Heath-Kelly, 2017; Heath-Kelly & Strausz, 2019; Younis & Jadhav, 2019) sectors attests that most educators and healthcare workers receive P/CVE training at the workplace. In such training, the Prevent Duty is presented through a language of ‘care’ (see de Goede & Simon, 2013) and positioned under the safeguarding umbrella, which allows frontline workers to synthesise these additional statutory guidelines into their previous knowledge as just another form of safeguarding (Busher et al., 2019; da Silva et al., 2021) and a tick-box exercise (Younis & Jadhav, 2019). However, it is important to note that the type of P/CVE training and its delivery in the United Kingdom vary from institution to institution (da Silva et al., 2021; Younis & Jadhav, 2019). Some stick to WRAP training materials while others resort to other available resources; some do it in person while others do it online; some resort to internal facilitators while others invite external facilitators, which can be members of the police, the local authority, or civil society organisations (which has created a whole industry around P/CVE training). P/CVE training for frontline workers has been the object of several criticisms, including claims that it has furthered discriminatory practices against the Muslim community (Awan, 2012; Qureshi, 2015; Pantazis & Pemberton, 2009) and that it places educators and healthcare professionals in an ethical dilemma (for a discussion, see Reed, 2016; Summerfield, 2016; Younis & Jadhav, 2019). Additionally, it is considered inadequate, as it is not accredited or regulated and was developed using unreliable indicators (Open Society Justice Initiative, 2016).

463

Raquel da Silva

Risk assessment tools Developments in identifying the prevalence of risk factors and indicators for violent extremism have led to the creation of a number of risk assessment tools. This section covers the most widely used violent extremist risk assessment tools in different national contexts and finishes with a reflection on the issues their use entails. In the United Kingdom, the government departed from the Extremism Risk Guidance (EGR22+), which is used to assess sentenced terrorism offenders in prison, to develop the Vulnerability Assessment Framework (VAF), which is used to assess individuals ‘at risk’ of radicalisation into violent extremism (Augestad Knudsen, 2020; Sarma, 2017; Thornton & Bouhana, 2017). The VAF is a checklist of 22 factors, where individuals are scored from zero to two on each factor, and each factor score must be accompanied by justificatory evidence (HM Government, 2012). The VAF sections comprise: engagement indicators (with a group, cause, or ideology), which include feelings of grievance and injustice, a need for identity, meaning, and belonging, a desire for excitement and adventure, a desire for political or moral change, experiencing a transitional period in life, and any relevant mental health issues; intent to cause harm indicators, which include evidence of us versus them thinking, dehumanisation of an enemy, and attitudes that justify offending; and capability to do harm indicators, which include access to relevant networks, skills, and competencies, relevant individual knowledge, access to funding or equipment, or known criminal capability. In sum, if an individual ticks the boxes of several key indicators, it is then seen as ‘vulnerable to’ or ‘at risk from radicalisation’ and is provided with appropriate follow-up interventions (HM Government, 2012, p. 2). Also from the United Kingdom, there is another list of indicators organised in the Identifying Vulnerable People (IVP) system, which are ‘drawn from the existing literature on extremism and violence’ (Cole et al., 2013, p. 2). This system combines different concerning behaviours, such as membership in non-violent radical groups, contact with known extremists, advanced military training, and overseas combat experience, with other factors, such as cultural and/or religious isolation, isolation from family, risk-taking conduct, an isolated peer group, hate rhetoric, and political activism. According to its authors, the presence of ‘even a single factor should prompt the practitioner to seek advice from their line manager’ (Cole et al., 2013, p. 7). In Canada, items of potential relevance to the process of radicalisation leading to violent extremism have been identified in the literature and in the views of practitioners in the field, giving origin to the Violent Extremism Risk Assessment 2nd edition (VERA-2R), which is also applied to convicted terrorism offenders (Pressman, 2009; Pressman & Flockton, 2012; Pressman et al., 2016). The VERA-2R consists of a set of 34 risk factors across the domains of beliefs, attitudes, and ideology (e.g. commitment to an ideology that justifies the use of violence), social context and intention (e.g. seeker, user, or developer of violent extremist materials), history, action, and capacity (e.g. early exposure to violence-promoting, militant ideology), commitment and motivation (e.g. motivated by perceived religious obligation and/or glorification), as well as five protective/ risk mitigating indicators (e.g. reinterpretation of ideology), plus three demographic domains (e.g. criminal history). Three-point ratings are made on all but the three demographic domains – ratings of low, moderate, or high – to indicate the extent of their presence in the individual case. Subsequently, an overall or final summary judgement about risk is made, also on a three-point scale: low, moderate, or high. Another system used in Canada to assess the risk of violent extremism, borrowed from the structured professional judgement (SPJ) approach to assessing the risk of group-based violence,

464

Interventions with ‘at risk’ individuals

is the Multi-Level Guidelines (MLG) (Cook et al., 2013). However, as noted by Hart et al. (2017, p. 12), the concept of group-based violence includes the majority of terrorism, with the exception of some lone actor terrorism; but it also includes many forms of violence that are not terrorism (e.g., violence committed by criminal organisations, street gangs, new religious movements, clans, and ideologically-focused groups). In this vein, these authors consider that the MLG must be used in ‘conjunction with other relevant risk assessment tools to evaluate people who are known or suspected to have committed terrorist group-based violence, as well as those who may be at risk for terrorist group-based violence’ (Hart et al., 2017, p. 12). The 16 basic risk factors in the MLG are divided into four groups: individual (e.g. violent behaviour and mental health problems); individual-in-group (e.g. group-based identity, negative attitudes towards the out-group); group (e.g. group violence and strong violent leadership); and group-in-society (e.g. intergroup threat, perceived injustice, extreme social status of the group); and there is scope to include additional or other considerations. Ratings of relevance are also created on a three-point rating scale (low, moderate, or high), followed by the development of risk scenarios, case management, and concluding opinions. Finally, in the United States, there was the development of an investigative template to deal particularly with threats associated with lone actors with a terrorist motivation: the Terrorist Radicalisation Assessment Protocol (TRAP-18) (Meloy, 2018; Meloy & Gill, 2016; Meloy et al., 2015). This protocol is not considered a standalone risk assessment tool as it is thought that the field is too poorly developed to support such a purpose, leading its authors to recommend that its use should be combined with that of the VERA-2R or the MLG in order to ‘capitalise on the greater accuracy of multi-method assessment practice’ (Meloy et al., 2015, p. 149). The TRAP-18 is comprised of a total of 18 items organised into two clusters: eight indicators of proximal warning behaviours (e.g. pathway warning or attack planning behaviour, fixation warning behaviour, or evidence of preoccupation with a harmful person or cause) and ten dynamic and more distal characteristics of the individual (e.g. evidence of personal grievance and moral outrage, failure to affiliate with an extremist group, dependence on the virtual community, nexus of psychopathology, and ideology). Each item is rated as present, absent, or insufficient information, and the intention is to structure the evaluation of pre-offence behaviours with a view to better conceptualising their meaning and direction of travel, thus assisting case prioritisation. The last 20 years have been marked by empirical investments in developing risk assessment tools. However, as noted by Sarma (2017, p. 279), ‘risk assessment in prevention work poses a range of empirical, ethical, and pragmatic obstacles’, which apply to risk assessment with terrorist offenders but are even more significant when individuals are being assessed in the pre-criminal space. Indicators-based approaches in the field of P/CVE are built on factors and categories that lack substantial empirical validation (Augestad Knudsen, 2020; Desmarais et al., 2017; RTI International, 2018), leading to high rates of false positives (Baaken et al., 2020; Sarma, 2017; van de Weert & Eijkman, 2020), as well as to discrimination and stigmatisation of certain strands of society (Mythen et al., 2017; Pettinger, 2017). Therefore, it is important to underline that several issues remain problematic regarding the study of violent extremism and the implementation of risk assessment tools in this domain (for a deeper discussion, see Clemmow et al., 2020; Gill, 2015; Scarcella et al., 2016).

465

Raquel da Silva

Interventions with ‘at risk’ individuals at the national level The aim of interventions with ‘at risk’ individuals is ‘to prevent the process of radicalization from evolving’ (Nehlsen et al., 2020, p. 4). The investment in such types of interventions was strengthened after the European Commission’s release of a revised strategy to combat violent extremist radicalisation and recruitment to terrorism in January 2014 and by its continuous work on these issues (Koehler, 2017). Several countries have included specific interventions with ‘at risk’ individuals in their P/CVE practices, including Germany (Baaken et al., 2020; Berczyk & Vermeulen, 2014; Said & Fouad, 2018), the Netherlands (van de Weert & Eijkman, 2020; Vidino, 2008), and France (Hellmuth, 2015a, 2015b) (for a worldwide overview of P/CVE practices, see Koehler, 2017). The following sections will zoom in on the interventions offered to ‘at risk’ individuals in two countries seen as role models in this arena: the United Kingdom and Denmark.

The United Kingdom The United Kingdom was the European pioneer in the creation of a special policy programme on P/CVE, which explains the prevalence of P/CVE research in this context (Tammikko, 2018) as well as its influence abroad (see Institute for Strategic Dialogue, 2010; Skleparis & Augestad Knudsen, 2020). Prevent, the British pre-emptive programme, is one of the four strands of the UK counterterrorism strategy, CONTEST, which also includes Pursue, Protect, and Prepare. This strategy reflects a general shift in the understanding of terrorism in security and counterterrorism policy from an externally focused threat following 9/11 to a more internally focused consideration of ‘homegrown’ extremism after 7/7. As already mentioned, the second phase of Prevent (2011– present) has focused on individuals at risk of, or vulnerable to, radicalisation, who have to be identified and referred to the anti-radicalisation mentoring scheme Channel. In this phase, the strategy has also emphasised broader and different forms of extremism (Miah, 2017), and key public sector areas – education, health, and social welfare – have been involved in counterterrorism surveillance (Heath-Kelly, 2017; Thomas, 2017). Channel was introduced in 2007 as the intervention arm of Prevent and is an ostensibly voluntary programme that provides mentoring and support to people identified as being at risk of radicalisation to violent extremism. It operates through a multi-agency approach, with local Channel assessment panels encompassing a range of local authority actors (e.g. social and health services, and educational institutions) and the police, and interventions provided by independent, governmentapproved mentors (Weeks, 2018). The number of Channel referrals and interventions has increased considerably in recent years, particularly after the introduction of the Counter-Terrorism and Security Act 2015, which placed a statutory duty on health, education, and social service institutions to pay due regard to the risk of radicalisation towards violence. Recent literature has been able to pinpoint the general inner workings of Channel (Thornton & Bouhana, 2017; da Silva et al., 2021; Pettinger, 2020), which involve public sector workers (e.g. teachers and doctors) referring individuals who they suspect might be ‘at risk’ of radicalisation to the relevant safeguarding leads in their workplace (e.g. school and hospital), who in turn assess if the referral can be dealt with internally or needs to be communicated to the local Prevent police team. If the referral arrives at the police and concerns still hold, the case is sent to a local Channel panel. Here, individual risk factors identified in the information provided by the original referee and the police team are discussed in light of VAF parameters, and a decision is made regarding the kind of support, if any, that needs to be put into place, as it is the role of the Channel panel ‘to develop an appropriate support package to safeguard those at risk of being drawn into terrorism 466

Interventions with ‘at risk’ individuals

based on an assessment of their vulnerability’ (HM Government, 2012, p. 7). Such support can take the shape of assistance with employment, education, housing, and mental health, as well as a one-to-one intervention with a Channel mentor (HM Government, 2015), and it is delivered by statutory agencies and local intervention providers approved by the Home Office (Thornton & Bouhana, 2017). If the mentoring solution is accepted, mentor and mentee agree on a date, time, and place to meet, and, on average, six to eight sessions are sanctioned, taking place over the course of six to 12 months (Pettinger, 2020). However, due to the tailored nature of this kind of mentoring, there is quite a lot of variation in these numbers, as a case can be made to increase the number of sanctioned sessions, as well as in risk assessment practices (see HM Government, 2011).

Denmark Denmark took inspiration from P/CVE practices in the United Kingdom and the Netherlands, including their criticisms (Tammikko, 2018); however, it also built on its own broader structures of crime prevention and evolved according to different events taking place at both national and international levels (Crone & Nasser, 2018). Such anchoring of pre-emptive interventions in existing structures of crime prevention gave them legitimacy and eased their integration into frontline workers’ routines, leading them to treat signs of radicalisation as they would treat other signs of risk behaviour (Lindekilde, 2014a). Consequently, the Danish approach to P/CVE is founded on extensive multi-agency collaboration between various social-service providers, the educational system, the health-care system, the police, and the intelligence and security services, so much so that the 12 police districts in the country have ‘info-houses’, where both public servants and private citizens can bring concerns about radicalisation or particular worrying cases (Agershou, 2014; Lindekilde, 2012). It also combines soft and hard measures, taking into consideration citizens’ well-being and offering social support if needed, as well as national security matters (Hemmingsen, 2015). Such measures are organised according to a tripartite prevention triangle, which facilitates the assessment of the severity of each case and its referral process to the relevant actors working in the P/CVE network and is composed of: ‘General Prevention’ at the bottom level, where municipalities, civil society organisations, and the preventive department of the intelligence service engage with a wider audience aiming to develop societal resilience; ‘Anticipatory Intervention’ at the middle level, where municipalities and local police districts offer voluntary support to individuals considered ‘at risk’ of radicalisation to help reverse such a pathway and thrive in society; and ‘Direct Intervention’ at the top level, where intelligence service and law enforcement agencies target individuals who are already active or about to become active in violent extremist milieus to stop them (Crone & Nasser, 2018, pp. 14–16). The interventions with ‘at risk’ individuals placed in the middle level of the prevention triangle are the concern of this chapter. Such interventions are framed in a welfare state logic that perceives crime as a consequence of social issues and follows the motto ‘help to self-help’ (Hemmingsen, 2015, p.  8). In this vein, interventions attempt to fulfil individuals’ needs through regular social support, available to all Danish citizens, which can include access to education, housing, and work opportunities; however, they also offer specific de-radicalisation support, which can include mentoring and counselling opportunities (Agershou, 2014; Crone & Nasser, 2018; Lindekilde, 2012). According to Lindekilde (2014a, p. 229), in Denmark, ‘the mentoring strategy is considered the approach to dealing with individual cases of radicalisation’ (emphasis in original). Mentors are supposed to act as role models, supervisors, coaches, and significant others, which integrates their ‘elaborate mentor role’ (Lindekilde, 2014a, p. 232). In this middle level of the prevention triangle, relatives of ‘at risk’ individuals and communities can also find support. Both can access coaching 467

Raquel da Silva

to deal with the situation, and the former can become part of networks of relatives trying to support their loved ones’ de-radicalisation process (Agershou, 2014; Hemmingsen, 2015). Finally, it is also at this middle level that frontline workers receive training around radicalisation and how to deal with potential cases. Such training has targeted SSP (School, Social Service, and Police Collaboration) coordinators, who are taught how to include radicalisation as a new ‘parameter of concern’ alongside pre-existing ones regarding alcohol and drug abuse, crime, etc. (Lindekilde, 2012, p. 392).

Criticisms and limitations of interventions with ‘at risk’ individuals P/CVE strategies and policies have proved contentious due to a lack of transparency, and rigorous evaluations, and unconfirmed success, among other structural issues. As an example, this section covers the criticisms and limitations of the interventions offered to ‘at risk’ individuals in the United Kingdom and Denmark. On the one hand, despite being considered a voluntary programme, Channel also includes coercive components like monitoring and surveillance (Koehler, 2017), and in the case of referrals of individuals under 18 years old, parental consent is only requested at the intervention stage, not at the start of the process (Marsden, 2016). On the other hand, critics also argue that Channel rests on ill-defined and poorly evidenced assumptions about how people become involved in political violence; it further embeds religious and ethnic discrimination through its overwhelming focus on individuals identified as being at risk of radicalisation into Al-Qaeda- or ISIS-inspired violent extremism; and it represents a counter-productive extension of state power (Heath-Kelly, 2013; Kundnani, 2014, 2015; Martin, 2018; Mythen et al., 2017; Pettinger, 2020; Qureshi, 2015). The UK government has countered that Channel addresses all types of radicalisation, pointing to a sharp increase in the proportion of far-right-related referrals and interventions in recent years as evidence, and that the distribution of Channel interventions to date reflects the nature of the terrorist threat facing the United Kingdom. However, it is well established that Prevent funding has been allocated to local authorities according to ‘Muslim demographics’ (McGhee, 2011, p. 1; Qurashi, 2018), despite also being described as ‘intelligence-led’ and ‘proportionate to threat levels’ (HM Government, 2011, p. 11). What empirical research is available regarding Channel is based either on interviews with members of Channel panels and mentors (Pettinger, 2017, 2020; Thornton & Bouhana, 2017) or on information collected through Freedom of Information (FOI) requests (Dudenhoefer, 2018), excluding the experiences of actual Channel mentees or a public and transparent programme evaluation. In Denmark, targeted Muslim communities and frontline workers dealing with P/CVE practices agree with the criticisms raised in the UK context, considering ‘the Danish action plan against radicalization in general, and particular instruments, such as mentoring schemes, role model campaigns, targeted civic education and distribution of information material on citizenship, as misrecognizing Muslims’ (Lindekilde, 2014b, p. 60; see also Kühle & Lindekilde, 2010). This issue of problematising Muslims in general, which can be both found in Denmark and in the United Kingdom (Lindekilde, 2014b; Mythen et al., 2009), can lead to the stigmatisation of Muslim communities and consequent resistance to engage in P/CVE activities on the one hand and, on the other hand, to the use of P/CVE practices and policies as radicalisation and recruitment instruments (Lindekilde, 2014b). Moreover, Muslim communities also oppose the anchoring of P/CVE strategies in existing structures of crime prevention, ‘obscuring the political nature of de-radicalisation work – the attempt to “normalise” and battle political and religious attitudes and practices that are not unlawful as such but that are deemed risky by authorities’ (Lindekilde, 2014a, p. 235). 468

Interventions with ‘at risk’ individuals

Both countries’ interventions offered to ‘at risk’ individuals draw heavily on the use of mentors, which has also given rise to criticism and presented some limitations. In the United Kingdom, the apparent poor selection and training of mentors has been raised (Pettinger, 2020; Thornton & Bouhana, 2017), while in Denmark, this seems to be a very thorough process (see Lindekilde, 2014a). However, in Denmark, the way in which the mentoring offer is structured also presents some limitations. First, as already mentioned, the nature of the mentor’s role is widely encompassing (i.e. the elaborate mentor role positions mentors as role models, supervisors, coaches, and significant others), which can make the mentor walk a very fine line between advising, qualifying and creating reflections on the mentee’s engagement with extremist milieus and the consequences of extreme action repertoires on the one hand, and on the other hand sustaining trust and openness in the relationship so that the mentee continues to share thought, emotions and dreams of the future. (Lindekilde, 2014a, p. 233) Moreover, the termination of a mentoring relationship can have a negative impact on mentees, particularly in cases where mentors provided important practical but also emotional support (Lindekilde, 2014a). Second, the tailored approach of the mentoring relationship means that the aims of each intervention vary greatly and, consequently, so do the success criteria, which do not facilitate the nationally and internationally desired evidence-based policy evaluations, which are seen to confer legitimacy to such interventions and influence the (neo-liberal) value-for-money considerations around what type of interventions to support (Lindekilde, 2014a). Another issue related to mentoring ‘at risk’ individuals is its purpose: ‘whether mentoring in the area of violent extremism is about changing beliefs or changing behaviour (or both)’ (Spalek & Davies, 2012, p. 359). The United Kingdom, for instance, has taken a ‘values-based’ approach in the field of P/CVE, appearing, at times, as overtly shaping individuals’ religious practices (Thomas, 2010), which undermine liberal freedoms, particularly in a pre-criminal space, favouring ‘more or less arbitrary mainstream or “government sanctioned” politics and religious views’ (Koehler, 2017, p. 201).

Evaluation of interventions with ‘at risk’ individuals A United Nations report referring to national P/CVE plans of action underlined that ‘effective monitoring and evaluation mechanisms for these plans are essential to ensuring that policies are having the desired impact’ (United Nations, General Assembly, 2015, p. 13). Thus, it is essential not only to implement P/CVE programmes but also to understand if and why they work (Horgan & Braddock, 2010). However, rigorous programme evaluation and reporting of P/CVE practices are sparse (Bellasio et al., 2018; Madriaza & Ponsot, 2016). A systematic review of counterterrorism strategies worldwide only found 21 evaluated projects, and out of these, only seven studies applied sound methodological designs (Lum et al., 2006); a review on methodology used in evaluating the effects of preventive and de-radicalisation interventions found that only 12% of the 55 considered studies contained a systematic qualitative or quantitative evaluation (Feddes & Gallucci, 2015); a meta-evaluation of CVE metrics in prevention, disengagement, and de-radicalisation programmes found that only half of the practitioner teams provided any correlational data between programme and outcomes, and none published an experimental design to attempt to measure whether the treatment group produced different results when compared to the control group (Mastroe & Szmania, 2016); and a systematic international review and meta-analysis of outcome evaluations has only 469

Raquel da Silva

found nine more or less well-controlled outcome evaluations in seven different countries (Jugl et al., 2021). Lack of programme evaluation is related to different issues, including difficulties in determining valid and reliable indicators of successful de-radicalisation (Horgan & Braddock, 2010), particularly in the case of preventive interventions (Thornton & Bouhana, 2017); the lack of a framework of analysis for measuring programme effectiveness (Malet, 2021, p. 58); and difficulties in accessing data in such a securitised and sensitive field (El-Said, 2015; Thornton & Bouhana, 2017). Thus, without proper data collection, it is impossible to establish adequate ‘baselines’ and ‘benchmarks’ to assess how successful programmes are, what should be expanded, and what should be stopped (Rabasa et al., 2010, p. 188). In this vein, not only participant outcomes have to be the object of assessment (which at times almost looks like an afterthought), but ‘rigorous standards with clear expectations and empirical tests for validity’ must be planned from the beginning of the programme (Malet, 2021, p. 66). The production of such scientific evidence would result in ‘evidence-based policy making’ and avoid ‘policy-based evidence making’ (Mythen et al., 2017, p. 196). Nonetheless, there have been efforts to suggest the transposition of evidence-based programme evaluation from sectors like education (Malet, 2021) and public health (Baruch et al., 2018; Byrne-Diakun, 2016; National Academies of Sciences, Engineering, and Medicine, 2017) to the P/CVE field.

Ways forward Throughout this chapter, it has been possible to see that ideological radicalisation is at the heart of P/CVE strategies in general and of interventions with ‘at risk’ individuals in particular. However, most research as well as policy literature overemphasise the role of ideology, and specifically religious ideology, in the radicalisation process, ignoring other ‘systemic and structural factors that shape human motivations and influence (anti-)social behaviours’ (Mythen et al., 2017, p. 195). This tendency has led to disastrous practices such as the targeting and stigmatisation of specific communities, as explored earlier. Thus, instead of mostly focusing on the intellectual level of radicalisation, policymakers, practitioners, and researchers might benefit from also considering the socio-political factors leading individuals towards politically violent activities, such as sustained experience with non-political forms of violence, drive towards actions, and acts of heroism (see Crone, 2016), as well as state violence and discriminatory government policies. Such a more complex and nuanced understanding of radicalisation and, consequently, ‘risk’ would result in a broader approach more focused on social measures than on assessments drawing on risk indicators such as changes in dress code, developing IT skills (Rights Watch UK, 2016), or legitimate social grievances like voicing concerns with foreign policy (de Goede & Simon, 2013). Finally, further empirical research is very much needed regarding the effects of the interventions covered in this chapter on the lives of the individuals they target. Such research should provide insight about the nature of interventions with ‘at risk’ individuals, including the extent to which they function as counter- or de-radicalisation programmes (see Thornton & Bouhana, 2017); the different pathways through which people enter, travel through, and exit intervention programmes; the social, psychological, and narrative transitions that happen during this process; and the short-, medium-, and long-term effects of having experienced intervention programmes. This would enable policymakers, practitioners, and the wider public to better understand how this type of interventions work, its possible unintended consequences, and the kind of support that different types of ‘at risk’ individuals require in order to live peaceful and fulfilling lives. 470

Interventions with ‘at risk’ individuals

References Agershou, T. (2014). Preventing radicalization and discrimination in Aarhus. Journal of Deradicalization, 1, 5–22. Aly, A., Balbi, A. M., & Jacques, C. (2015). Rethinking countering violent extremism: Implementing the role of civil society. Journal of Policing, Intelligence and Counter Terrorism, 10(1), 3–13. Augestad Knudsen, R. (2020). Measuring radicalisation: Risk assessment conceptualisations and practice in England and Wales. Behavioral Sciences of Terrorism and Political Aggression, 12(1), 37–54. Awan, I. (2012). “I am a Muslim not an extremist”: How the prevent strategy has constructed a “suspect” community. Politics & Policy, 40(6), 1158–1185. Baaken, T., Korn, J., Ruf, M., & Walkenhorst, D. (2020). Dissecting deradicalization: Challenges for theory and practice in Germany. International Journal of Conflict and Violence, 14(2), 1–18. Baruch, B., Ling, T., Warnes, R., & Hofman, J. (2018). Evaluation in an emerging field: Developing a measurement framework for the field of counter-violent-extremism. Evaluation, 24(4), 475–495. Bellasio, J., Hofman, J., Ward, A., Nederveen, F., Knack, A., Meranto, A. S., & Hoorens, S. (2018). Counterterrorism evaluation: Taking stock and looking ahead. RAND Corporation. https://doi.org/10.7249/RR2628 Berczyk, J., & Vermeulen, F. (2014). Prevent abroad: Militant democracy, right-wing extremism and the prevention of Islamic extremism in Berlin. In C. Baker-Beall, C. Heath-Kelly, & L. Jarvis (Eds.), Counterradicalisation: Critical perspectives (pp. 88–105). Routledge. Bjørgo, T., & Horgan, J. (2009). Leaving terrorism behind: Individual and collective disengagement. Routledge. Blackwood, L., Hopkins, N., & Reicher, S. (2012). Divided by a common language? Conceptualising identity, discrimination and alienation. In K. Jonas & T. Morton (Eds.), The psychology of intervention and engagement following crisis (pp. 222–236). Wiley. Bonelli, L., & Carrié, F. (2018). La fabrique de la radicalité: Une sociologie des jeunes djiadistes français. Seuil. Brooks, R. A. (2011). Muslim “homegrown” terrorism in the United States: How serious is the threat? International Security, 36, 7–47. Busher, J., Choudhury, T., & Thomas, P. (2019). The enactment of the counter-terrorism “prevent duty” in British schools and colleges. Critical Studies on Terrorism, 12(3), 440–462. Byrne-Diakun, R. M. (2016). Countering violent extremism: Applying the public health model. Georgetown security studies review. Georgetown University. Clemmow, C., Schumann, S., Salman, N. L., & Gill, P. (2020). The base rate study: Developing base rates for risk factors and indicators for engagement in violent extremism. Psychiatry & Behavioral Science, 65(3), 865–881. Cole, J., Alison, E., Cole, B., & Alison, L. (2013). Guidance for identifying people vulnerable to recruitment into violent extremism. https://preventforfeandtraining.org.uk/wpcontent/uploads/2017/09/IVP_Guidance_Draft_v0.3_web_version.pdf Cook, A. N., Hart, S. D., & Kropp, P. R. (2013). Multi-level guidelines (MLG) for the assessment and management of group-based violence (User Manual). Mental Health, Law, and Policy Institute. Coolsaet, R. (2016). All radicalisation is local: The genesis and drawbacks of an elusive concept. Egmont. Crone, M. (2016). Radicalization revisited: Violence, politics and the skills of the body. International Affairs, 92(3), 587–604. Crone, M., & Nasser, K. (2018). Preventing violent extremism in Lebanon: Experience from a Danish-Lebanese partnership (IEMed Papers 38, EuroMeSCo Series). European Institute of the Mediterranean. da Silva, R., Fontana, G., & Armstrong, M. (2021). It’s about keeping children safe, not spying”: A governmentality approach to prevent in primary education. British Journal of Politics and International Relations. https://doi.org/10.1177/13691481211021212 de Goede, M., & Simon, S. (2013). Governing future radicals in Europe. Antipode, 45(2), 315–335. Desmarais, S. L., Simons-Rudolph, C. J., Brugh, E. S., & Hoggan, C. (2017). The state of scientific knowledge regarding factors associated with terrorism. Journal of Threat Assessment and Management, 4(4), 180–209. Directorate-General for Internal Policies. (2014). Preventing and countering youth radicalisation in the EU: Study for the LIBE committee. European Parliament. Dudenhoefer, A. (2018). Resisting radicalisation: A critical analysis of the UK prevent duty. Journal for Deradicalization, 14, 153–191. El-Said, H. (2015). New approaches to countering terrorism: Designing and evaluating counter radicalization and de-radicalization programmes. Palgrave Macmillan.

471

Raquel da Silva Feddes, A. R., & Gallucci, M. (2015). A literature review on methodology used in evaluating effects of preventive and de-radicalisation interventions. Journal for Deradicalization, 5, 1–27. Gielen, A. J. E. (2019). Countering violent extremism: A realist review for assessing what works, for whom, in what circumstances, and how? Terrorism and Political Violence, 31(6), 1149–1167. Gill, P. (2015). Toward a scientific approach to identifying and understanding indicators of radicalization and terrorist intent: Eight key problems. Journal of Threat Assessment and Management, 2(3–4), 187–191. Hart, S. D., Cook, A. N., Pressman, D. E., Strang, S., & Lim, Y. L. (2017). A concurrent evaluation of threat assessment tools for the individual assessment of terrorism. Canadian Network for Research on Terrorism, Security, and Society. Heath-Kelly, C. (2013). Counter-terrorism and the counterfactual: Producing the “radicalization” discourse and the UK prevent strategy. The British Journal of Politics and International Relations, 15(3), 394–415. Heath-Kelly, C. (2017). Algorithmic autoimmunity in the NHS: Radicalisation and the clinic. Security Dialogue, 48(1), 29–45. Heath-Kelly, C., & Strausz, E. (2019). The banality of counterterrorism “after, after 9/11”? Critical Studies on Terrorism, 12(1), 89–109. Hellmuth, D. (2015a, Fall). Countering Jihadi radicals and foreign fighters in the United States and France: Très similaire. Journal for Deradicalization, 4, 1–43. Hellmuth, D. (2015b). Countering Jihadi terrorists and radicals the French way. Studies in Conflict and Terrorism, 38(12), 979–997. Hemmingsen, A. (2015). An introduction to the Danish approach to countering and preventing extremism and radicalization. DIIS Report. http://pure.diis.dk/ws/files/470275/DIIS_Report_2015_15_2_ed..pdf HM Government. (2011). Prevent strategy. Home Office. www.homeoffice.gov.uk/publications/ counter-terrorism/prevent/prevent-strategy HM Government. (2012). Channel: Protecting vulnerable people from being drawn into terrorism. A guide for local partnerships. www.npcc.police.uk/documents/TAM/2012/201210TAMChannelGuidance.pdf+ &cd=7&hl=en&ct=cnk&gl=uk/ HM Government. (2015). Channel duty guidance: Protecting vulnerable people from being drawn into terrorism. https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/ file/42518 9/ChannelDutyGuidance_April_2015.pdf Horgan, J., & Braddock, K. (2010). Rehabilitating the terrorists? Challenges in assessing the effectiveness of de-radicalization programmes. Terrorism and Political Violence, 22(2), 267–291. Institute for Strategic Dialogue. (2010). PPN working paper: De-radicalization. www.counterextremism.org/ resources/details/id/115/channel-process Jerome, L., & Busher, J. (2020). The prevent duty in education: Impact, enactment and implications. Palgrave Macmillan. Jerome, L., Elwick, A., & Kazim, R. (2019). The impact of the prevent duty on schools. British Educational Research Journal, 45(4), 821–837. Jugl, I., Lösel, F., Bender, D., & King, S. (2021). Psychosocial prevention programs against radicalization and extremism: A meta-analysis of outcome evaluations. The European Journal of Psychology Applied to Legal Context, 13(1), 37–46. Koehler, D. (2015). Deradicalization. In N. Hall, A. Corb, P. Giannasi, & J. G. D. Gireve (Eds.), The Routledge international handbook on hate crime (pp. 420–429). Routledge. Koehler, D. (2017). Understanding deradicalization: Methods, tools and programs for countering violent extremism. Routledge. Kühle, L., & Lindekilde, L. (2010). Radicalization among young Muslims in Aarhus. Center for Studies in Islamism and Radicalisation. https://f.hypotheses.org/wp-content/blogs.dir/2725/files/2016/04/radicalization_aarhus.pdf Kundnani, A. (2014). The Muslims are coming: Islamophobia, extremism, and the domestic war on terror. Verso. Kundnani, A. (2015). A decade lost: Rethinking radicalisation and extremism. Claystone Publications. Kundnani, A., & Hayes, B. (2018). The globalisation of countering violent extremism policies. Undermining human rights, instrumentalising civil society. Transnational Institute. Lakhani, S. (2020). Social capital and the enactment of prevent duty: An empirical case-study of schools and colleges. Critical Studies on Terrorism, 13(4), 660–679. Lindekilde, L. (2012). Value for money? Problems of impact assessment of counter-radicalisation policies on end target groups: The case of Denmark. European Journal of Criminal Policy and Research, 18, 385–402.

472

Interventions with ‘at risk’ individuals Lindekilde, L. (2014a). Refocusing Danish counter-radicalisation efforts: An analysis of the (problematic) logic and practice of individual de-radicalisation interventions. In C. Baker-Beall, C. Heath-Kelly, & L. Jarvis (Eds.), Counter-radicalisation: Critical perspectives (pp. 223–241). Routledge. Lindekilde, L. (2014b). A typology of backfire mechanisms. In L. Bosi, C. Demetriou, & S. Malthaner (Eds.), Dynamics of political violence: A process-oriented perspective on radicalization and the escalation of political conflict (pp. 51–69). Ashgate. Lindekilde, L. (2015). Dansk forebyggelse af ekstremisme og radikalisering 2009–2014: Udviklingstendenser of fremtidige udfordringer. Politica, 47(3), 424–444. Lum, C., Kennedy, L. W., & Sherley, A. J. (2006). The effectiveness of counter-terrorism strategies. Campbell Systematic Reviews, 2. www.campbellcollaboration.org/library/effectiveness-of-counterterrorism-strategies.html Madriaza, P., & Ponsot, A.-S. (2016). Preventing radicalization: A systematic review. International Centre for the Prevention of Crime. Malet, D. (2021). Countering violent extremism: Assessment in theory and practice. Journal of Policing, Intelligence and Counter Terrorism, 16(1), 58–74. Marsden, S. V. (2016). Explainer: What happens to people who are suspected of being “radicalised”? The Conversation. https://theconversation.com/explainer-what-happens-to-people-who-are-suspected-of-being-radicalised-53652 Martin, T. (2018). Identifying potential terrorists: Visuality, security and the channel project. Security Dialogue, 49(4), 254–271. Mastroe, C., & Szmania, S. (2016). Surveying CVE metrics in prevention, disengagement and deradicalization programs: Report to the office of university programs, science and technology directorate. U.S. Department of Homeland Security, START. www.start.umd.edu/pubs/START_SurveyingCVEMetrics_ March2016.pdf Mattsson, C. (2018). Caught between the urgent and the comprehensible: Professionals’ understanding of violent extremism. Critical Studies on Terrorism, 11(1), 111–129. McGhee, D. (2011). New prevent: Different is not better. Muslim Council of Britain. http://soundings.mcb. org.uk/?p=38 Meloy, J. R. (2018). The operational development and empirical testing of the terrorist radicalization assessment protocol (TRAP–18). Journal of Personality Assessment, 100(5), 483–492. Meloy, J. R., & Gill, P. (2016). The lone-actor terrorist and the TRAP-18. Journal of Threat Assessment and Management, 3, 37–52. Meloy, J. R., Roshdi, K., Glaz-Ocik, J.,  & Hoffmann, J. (2015). Investigating the individual terrorist in Europe. Journal of Threat Assessment and Management, 2, 140–152. Miah, S. (2017). Muslims, schooling and security. Palgrave Pivot. Moffat, A., & Gerard, J. F. (2020). Securitising education: An exploration of teachers’ attitudes and experiences regarding the implementation of the prevent duty in sixth form colleges. Critical Studies on Terrorism, 13(2), 197–217. Monaghan, J., & Molnar, A. (2016). Radicalisation theories, policing practices, and “the future of terrorism?” Critical Studies on Terrorism, 9(3), 393–413. Mythen, G., Walklate, S., & Khan, F. (2009). “I’m a Muslim, but I’m not a terrorist”: Victimization, risky identities and the performance of safety. British Journal of Criminology, 49(6), 736–754. Mythen, G., Walklate, S., & Peatfield, E. J. (2017). Assembling and deconstructing radicalisation in prevent: A case of policy-based evidence making? Critical Social Policy, 37(2), 180–20. National Academies of Sciences, Engineering, and Medicine. (2017). Countering violent extremism through public health practices: Proceedings of a workshop. The National Academies Press. Nehlsen, I., Biene, J., Coester, M., Greuel, F., Milbradt, B., & Armborst, A. (2020). Evident and effective? The challenges, potentials and limitations of evaluation research on preventing violent extremism. International Journal of Conflict and Violence, 14(2), 1–20. Open Society Justice Initiative. (2016). Eroding trust: The UK’s prevent counter- extremism strategy in health and education. Retrieved July 5, 2021, from www.opensocietyfoundations.org/sites/default/files/erodingtrust-20161017_0.pdf Pantazis, C., & Pemberton, S. (2009). From the “old” to the “new” suspect community: Examining the impacts of recent UK counter-terrorist legislation. The British Journal of Criminology, 49(5), 646–666. Pettinger, T. (2017). De-radicalization and counter-radicalization: Valuable tools combating violent extremism or harmful methods of subjugation. Journal for Deradicalization, 12, 1–59.

473

Raquel da Silva Pettinger, T. (2020). British terrorism preemption: Subjectivity and disjuncture in channel “de-radicalization” interventions. British Journal of Sociology. https://doi.org/10.1111/1468-4446.12754 Pressman, D. E. (2009). Risk assessment decisions for violent political extremism 2009–02. Public Safety Canada, Government of Canada. www.publicsafety.gc.ca/cnt/rsrcs/pblctns/2009-02-rdv/2009-02-rdv-eng.pdf Pressman, D. E., Duits, N., Rinne, T., & Flockton, J. (2016). VERA-2R violent extremism risk assessment version 2 revised: A structured professional judgement approach. NIFP/DJI. Pressman, D. E., & Flockton, J. (2012). Calibrating risk for violent political extremists and terrorists: The VERA-2 structured assessment. British Journal of Forensic Practice, 14, 237–251. Qureshi, A. (2015). PREVENT: Creating “radicals” to strengthen anti-Muslim narratives. Critical Studies on Terrorism, 8(1), 181–191. Qurashi, F. (2018). The prevent strategy and the UK “war on terror”: Embedding infrastructures of surveillance in Muslim communities. Palgrave Communications, 4(17). https://doi.org/10.1057/s41599-017-0061-9 Rabasa, A., Pettyjohn, S. L., Ghez, J. J., & Boucek, C. (2010). Deradicalizing Islamist extremists. RAND Corporation. www.ppv.issuelab.org/resources/8976/8976.pdf Reed, S. (2016). The prevent programme: An ethical dilemma for teachers as well as psychiatrists. BJ Psych Bulletin, 40, 85–86. Rights Watch UK. (2016). Preventing education? Human rights and UK counter-terrorism policy in schools. Rights Watch UK. Romaniuk, P. (2015). Does CVE work? Lessons learned from the global effort to counter violent extremism. Global Center on Cooperative Security. www.globalcenter.org/wp-content/uploads/2015/09/Does-CVEWork_2015.pdf RTI International. (2018). Countering violent extremism: The application of risk assessment tools in the criminal justice and rehabilitation process: Literature review. U.S. Department of Homeland Security. Said, B. T., & Fouad, H. (2018). Countering Islamist radicalisation in Germany: A guide to Germany’s growing prevention infrastructure (ICCT Policy Brief). ICCT. Sarma, K. M. (2017). Risk assessment and the prevention of radicalization from nonviolence into terrorism. American Psychologist, 72(3), 278–288. Scarcella, A., Page, R., & Furtado, V. (2016). Terrorism, radicalisation, extremism, authoritarianism and fundamentalism: A systematic review of the quality and psychometric properties of assessments. PLoS ONE, 11(12), 1–19. Sian, K. P. (2015). Spies, surveillance and stakeouts: Monitoring Muslim moves in British state schools. Race Ethnicity and Education, 18(2), 183–201. Skjoldager, M., & Sheikh, J. (2014). Samarbejde mellem omstridt moske og politiet bremser Syriensfarere. Politiken. http://politiken.dk/indland/art5520181/ Samarbejde-mellem-omstridt-moske-og-politiet-bremser-Syriensfarere Skleparis, D., & Augestad Knudsen, R. (2020). Localising “radicalisation”: Risk assessment practices in Greece and the United Kingdom. The British Journal of Politics and International Relations. https://doi. org/10.1177/1369148120910987 Spalek, B., & Davies, L. (2012). Mentoring in relation to violent extremism: A study of role, purpose, and outcomes. Studies in Conflict and Terrorism, 35, 354–368. Summerfield, D. (2016). Mandating doctors to attend counter-terrorism workshops is medically unethical. BJ Psych Bulletin, 40, 87–88. Tammikko, T. (2018). The political challenges of community-level PVE practices: The Danish case of Copenhagen vs. Aarhus on dialoguing with extremist milieus. Journal of Deradicalization, 16, 103–124. Thomas, P. (2010). Failed and friendless: The UK’s “preventing violent extremism” programme. The British Journal of Politics and International Relations, 12(3), 442–458. Thomas, P. (2017). Changing experiences of responsibilisation and contestation within counter-terrorism policies. Policy and Politics, 45(3), 305–322. Thornton, A., & Bouhana, N. (2017). Preventing radicalization in the UK: Expanding the knowledge-base on the channel programme. Policing, 13(3), 331–344. United Nations, General Assembly. (2015). Plan of action to prevent violent extremism, report of the secretarygeneral (A/70/674). http://undocs.org/A/70/674 van de Weert, A., & Eijkman, Q. A. M. (2020). Early detection of extremism? The local security professional on assessment of potential threats posed by youth. Crime, Law and Social Change, 73, 491–507.

474

Interventions with ‘at risk’ individuals Vidino, L. (2008). A preliminary assessment of counter-radicalization in the Netherlands. CTC Sentinel, 1(9), 12–16. Vidino, L., & Brandon, J. (2012). Countering radicalization in Europe. http://icsr.info/wp-content/ uploads/2012/12/ICSRReport-Countering-Radicalization-in-Europe.pdf Weeks, D. (2018). Doing Derad: An analysis of the U.K. system. Studies in Conflict  & Terrorism, 41(7), 523–540. Younis, T., & Jadhav, S. (2019). Keeping our mouths shut: The fear and racialized self-censorship of British healthcare professionals in PREVENT training. Culture, Medicine and Psychiatry, 43, 404–424.

475

31 HOW CAN GENERAL VIOLENCE RISK ASSESSMENT AND MANAGEMENT INFORM THAT OF VIOLENT EXTREMIST RISK?1 Paul Gill, Zoe Marchment, Amber Seaward, Philip Doherty, and Kirsty Goodman Risk assessment and management are core components of policing, intelligence, judicial, and clinical settings. Risk assessment considers available information and determines an individual’s offending risk. Risk management acts to gain or maintain control over the risk posed. Although historically associated with judgements related to the risk of violence recidivism, risk assessment and management principles are increasingly applied to other types of violence (e.g. stalking) as well as newly emerging criminal ventures with high harm potential (e.g. cybercrime). Different forms of risk assessment exist. In ‘unstructured’ risk assessment, assessors are given no guidance, and simply gather information they feel is needed. Their judgement of risk is purely based on their professional opinion. With ‘actuarial’ risk assessment, assessors gather information on the presence of pre-assigned risk factors, which are typically summed to produce a risk ‘score’. The gold standard of risk assessment is the ‘structured professional judgement’ approach. Assessors gather information on the presence and relevance of a set of pre-assigned risk factors, formulate an individualised risk assessment based on both a structured guidance manual and their professional opinion, and develop suitable management plans. The major difference between actuarial and SPJ instruments is the role of the assessor. Actuarial risk assessment guides only allow for a final decision about the volume of risk in an individual case based on fixed rules and statistical calculations. In contrast, SPJ guidance on risk assessment and management supports the assessor to make decisions based on their understanding of the case but does not offer an actuarial formula or fixed rules by which any such decision may be made. Therefore, SPJ guidance supports the application of discretion in decision-making about risk, allowing case-specific factors and the context to have a bearing not only on the understanding of the case but also on the management of the risks posed, whatever those may be. The principles of risk assessment and management are also increasingly used in the pre-crime space, where they address those who are deemed ‘at-risk’ of crossing the threshold into offence. This is evidenced by PREVENT in the UK counterterrorism infrastructure (and its recent various offshoots in strategies focusing on preventing engagement in cybercrime and serious organised crime). Such ‘at risk’ individuals may have engaged with extremist materials or settings and have demonstrated attitudinal affinity with an extremist cause, but have yet to mobilise to violence. DOI: 10.4324/9781003035848-35 476

Risk assessment instruments can help predictions

In the United Kingdom, 4,915 such referrals were made to counterterrorism policing between April 2020 and March 2021 (Home Office, 2021). Approximately 14% of these referrals were taken seriously enough for interventions to be implemented to manage their risk. Plainly speaking, scrutiny of the effectiveness and quality of violent extremism risk assessment and management comes to the forefront when terrorist attacks occur. Unfortunately, this has been too common an occurrence. According to Europol, 2019 witnessed 119 foiled, failed, and completed terrorist attacks from a range of ideological influences across 13 EU Member States (Europol, 2020). Many recent attackers were known to intelligence services, had previously been closed cases, or had spent time in prison for terrorist offences. This naturally raises concerns about how these individuals were risk assessed and managed at all points in their trajectory towards offending (see, e.g., the inquest into the 2019 Fishmonger’s Hall attack). Similar concerns have also been evoked about the procedures in place for preventing recent high-profile acts of targeted violence, such as mass public murders, school shootings, and workplace violence. There are many challenges associated with risk assessment for violent extremism, including concerns regarding the empirical basis, ecological validity, and general suitability of the tools used. Sarma (2017) outlines four challenges for the assessment of the risk of violent radicalisation into terrorism: risk specification, identifying factors, screening out true negatives, and low base rates. First, it is important to differentiate between different forms of terrorism and focus on the behaviour of concern, as risk factors for one type may not be applicable or have predictive value across others. Second, there is a lack of evidence on risk and protective factors that influence an individual’s propensity for risk, especially when compared to those developed for general violence, such as the HCR-20. A third difficulty faced is the ability of risk assessment to accurately identify those at high risk. Existing tools have prioritised sensitivity at the expense of specificity. This can lead to a waste of resources and a reduction in the public’s trust in the police and security services. Finally, the low base rates associated with involvement in terrorism mean the accuracy of tools that rely on actuarial methods is very poor. For those crime types that are both on the increase and have the potential for high harm, be it physical, psychological, or economic, the onus on risk assessment and management will likely amplify and necessitate better links across practitioner agencies (both policing and non-policing), academia, and industry. Consequently, there has been a boom in the availability of risk assessment and management instruments over the past decade focused on forms of targeted violence, particularly violent extremism. Some were designed by government entities for in-house work; others are commercially available. There are several risk assessment tools currently used for the assessment and management of extremism risk, which vary in terms of context, purpose, and users. These include Identifying Vulnerable People (IVP), Violent Extremism Risk Assessment (VERA-2R), Extremism Risk Guidance (ERG22+), Multi-Level Guidelines (MLG), and Terrorist Radicalisation Protocol (TRAP-18). The IVP was designed as part of the UK’s Prevent strategy to assess the risk of vulnerability to radicalisation. It provides practitioners with 16 factors that have been theoretically linked to violent extremism, specifically those who ‘are at risk of being targeted or recruited by violent extremists’ (Cole et al., 2010, p. 3). The VERA-2R uses 34 specified dynamic indicators to analyse the risk of violent extremism. It can be used by psychologists and psychiatrists with knowledge of violent extremism, as well as analysts of security and intelligence services, forensic social workers, and police forces or others tasked with assessing people suspected of violent extremist or terrorist criminal offences. The ERG22+ uses SPJ to assess the risk of re-offending for already convicted extremist offenders. It consists of three sections: engagement, intent, and capability. ‘Engagement’ includes factors that motivate an individual to engage with a cause, ideology, or group; ‘intent’ consists of factors relating to the individual’s degree of readiness to offend; and ‘capability’ concerns their ability to carry out an act of terrorism. 477

Paul Gill et al.

The MLG is an SPJ approach to assessing the risk of group-based violence for criminal justice, mental health, and security professionals. The MLG contains 16 risk factors across four domains: individual, individual-group, group, and group-societal. For lone actor targeted violence risk, there is the TRAP-18 framework. This framework consists of 18 coded behavioural patterns for assessing the lone actor, consisting of two sets of indicators. The first set contains eight warning behaviours developed to identify patterns of proximal risk for intended or targeted violence. The second set contains ten chronic and distal aspects of lone-actor terrorists that may prompt further intelligence gathering and monitoring but may stop short of active risk management. Given both the emergent, growing demand and nascent supply of these instruments focused on violent extremism, it is worth taking stock of the more voluminous literature on general violence risk assessment and management. We ask: what works, for whom, and in what contexts? We conduct an umbrella review that synthesises the results of 23 meta-analyses of risk assessment instruments that were identified through a systematic search and sift (see Appendix 1).2 These meta-analyses collectively cover over 2,108 empirical studies on over 1.3 million individuals. Our umbrella review leads us to the following five conclusions, which we elaborate upon in turn in the sections that follow: 1. Risk assessment instruments can help make predictions. 2. Risk assessment performance can be moderated by several design features. 3. Blending insights from different instruments can improve predictive accuracy. 4. Prediction is not everything. 5. Assessments require multivariate explanations of risk. After outlining the contours of these five conclusions, we look at their implications for violent extremism risk assessment and management.

Risk assessment instruments can help predictions A common goal of all risk assessment instruments is to help an assessor grasp what level of risk is currently posed by the individual under assessment. But how accurate are different risk assessment instruments at prediction? Because risk assessment instruments are not designed equally, we should expect a great deal of variability in performance. A common method of testing the degree to which an instrument can discriminate between those who conduct the outcome of interest (e.g. violent recidivism) and those who do not (e.g. the nonrecidivists) is to use the Area Under the Curve (AUC). The result determines the probability that a randomly selected, in this case, violent recidivist was scored highly by the instrument. AUC ROC (Receiver Operating Characteristic) scores of 0.5 mean complete chance prediction (e.g. tossing a coin would be just as good), between 0.64 and 0.7 means average predictive ability, between 0.7 and 0.8 means good predictive ability, and between 0.8 and 0.9 means excellent and above 0.9 outstanding. Table 31.1 highlights the aggregated predictive performance (as measured by AUC) of a range of instruments across 18 meta-analyses. We can draw a number of conclusions from this table. First, the vast majority of meta-analyses place predictive performance in the average-to-good category, with none performing excellently or outstandingly. Therefore, there are limitations to what an instrument can do by itself in terms of prediction.

478

Risk assessment instruments can help predictions Table 31.1  Aggregated Predictive Performance of Risk Assessment Instruments Study

Instrument

Recidivism Type

Sample Specifics

AUC

Babchishin et al. (2012)

Static-99R

Sexual violence General violence Any Sexual violence General violence Any recidivism Sexual violence General violence Any recidivism Institutional violence General violence Institutional violence General violence Institutional violence General violence Institutional violence Institutional violence General violence Institutional violence General violence Sexual violence General violence Sexual violence General violence Domestic violence

Sex offenders

.684 .703 .718 .661 .614 .582 .686 .708 .732 .678 .642 .637 .642 .585 .637 .642 .545 .637 .596 .654 .670 .660 .660 .590 .647 .760 .715 .780 .690 .772 .555 .692 .619

RRASOR STATIC 2002R Campbell et al. (2009)

HCR 20 LSI/LSI-R PCL/PCL-R PCL: SV SIR Scale VRAG

Viljoen et al. (2012)

J-SOAP II ERASOR

Van Der Put et al. (2019)

van den Berg et al. (2018) Pusch and Holtfreder (2018)

Schwalbe (2007, 2008)

All Tools CRAT-P DVRAG IPPA-RAT ODARA P-TRAIT KSID Dynamic sex offender assessment instruments LSI

All Tools YLS PCL NCAR OCRA

Adult offenders

Sex offenders

N/A

Sexual violence General violence

Sex offenders

Any Any General violence General violence Any

Juvenile male Juvenile female Juvenile male Juvenile female Juveniles

.661 .644 .669 .644 .642 .641 .695 .603 .585

(Continued)

479

Paul Gill et al. Table 31.1 (Continued) Study

Instrument

Recidivism Type

Sample Specifics

AUC

Olver et al. (2009)

YA-LSI

Any Non-violent Violent Sexual Any Non-violent Violent Sexual Any Non-violent Violent Sexual Violent Any Any Violent Non-violent Sexual Verbal aggression Sexual Any aggression Violent Sexual Criminal offending Any Sexual Violent Any

Juveniles

.684 .678 .666 .608 .642 .585 .642 .540 .689 .719 .678 .534 .679 .642 .672 .619 .642 .579 .745 .705 .714 .720 .740 .660 .657 .659 .641 .780 .750 .740 .710 .700 .700 .700 .660 .692 .668 .684 .666 .707 .696 .638 .755 .737

YA-PCL

SAVRY

Rossdale et al. (2020)

HCR 20

Olver et al. (2014)

LSI

O’Shea et al. (2013) Helmus et al. (2012) O’Shea & Dickens (2014) Fazel et al. (2012)

HCR 20 Static 99R START 8 tools 3 tools 2 tools All tools

Hanson & Morton-Bourgon (2009) Singh et al. (2011)

Tully et al. (2013)

SVR20 SORAG VRAG SAVRY HCR SARA STATIC PCL Static 99 RRasor SORAG RM2000/S Static 2002 SVR20 MnSOST-R VRS-SO SRA

Sexual

Source: Created by the authors

480

Females N/A

Inpatients N/A N/A N/A Sex offenders N/A

Sex offenders

Risk assessment instruments can help predictions

Second, looking at the crime types listed, the predictive scores tend to err higher for higher base rate outcomes. For example, general violence occurs more often than sexual violence in the general population, and those meta-analyses focused on general violence recidivism prediction fare much better. This has major implications for the adoption of threat and risk assessment instruments for types of violence like terrorism, which have even lower base rates. Alternative measures like attitudinal affinity with extremist measures have higher prevalence rates (Wolfowicz et al., 2020), yet attitudes are typically not considered or tested in the broader risk assessment field, where the focus is largely upon overt behaviour. It warrants greater consideration, therefore, of how the assessment of attitudes should be conducted. Third, the only demonstration of a meta-analysis looking at the onset of offending compared to recidivism was that of van der Put et al.’s (2019) systematic review of domestic violence prediction. The results showed that those instruments predicting the onset of domestic violence scored significantly higher (AUC = 0.744) than those predicting domestic violence recurrence (AUC = 0.643), which points towards the importance of early detection and prevention and the potential of threat assessment instruments to provide capabilities for spotting this trajectory upstream.

Risk assessment performance can be moderated by several design features A range of meta-analyses investigated whether the predictive performance of risk assessment instruments is impacted (positively or negatively) by a range of potential moderators. The evidence suggests this is the case in terms of (1) the types of risk factors an instrument contains, (2) the types of information that informs an assessment, (3) the instrument’s development and risk specification, and (4) authorship bias. We look at each in turn.

Factor 1: risk factor types There are different ways of categorising risk factors, and a popular one is to split them into static and dynamic factors. Static risk factors relate to historical factors that are non-changeable aspects of the person being assessed. Dynamic risk factors relate to situational and other types of stressors that are amenable to change and occurred relatively recently to the problem behaviour. Campbell et al. (2009) examined the differential performance of instruments that (1) primarily contain static risk factors, (2) dynamic risk factors, and (3) an equal combination of static and dynamic risk factors. For institutional violence, static instruments had a significantly larger mean effect than primarily dynamic and mixed instruments. The surprising over-performance of static factors here is likely due to the level of control held and surveillance maintained within institutional settings on dynamic risk factors (e.g. access to substances, engagement with third parties). Moving to violence in the community, Campbell et al. (2009) found that for violent recidivism, dynamic instruments performed marginally better than instruments that primarily contain static risk factors and instruments that combine an equal combination of static and dynamic risk factors. Hanson and Morton-Bourgon (2009) also found instruments containing dynamic factors performed better for sexual violence recidivism prediction than those primarily containing static risk factors. Additionally, in a meta-analysis of a single instrument, the HCR-20, the results demonstrated that the dynamic components of the instrument produced larger effect sizes than the static items when predicting inpatient aggression in forensic settings (O’Shea et al., 2013). The strong performance of instruments with dynamic risk factors is important because, by definition, dynamic factors are the most amenable to change and typically form the targets of interventions and management strategies. 481

Paul Gill et al.

Factor 2: data inputs Information is crucial for assessments. Depending on the operational context, different types of data sources might inform risk assessments. Some contexts may rely on fully remote assessments with no first-hand contact with the individual being assessed (labelled ‘file review only’). Other contexts might rely solely on first-hand interviews with the individual, solicit self-reports from the individual absent an interview, or conduct a mixture of these practices. Campbell et al. (2009) tested whether the data sources for assessments impacted predictive performance. File review only and a mixture of file review plus interview produced the largest effect sizes for violent recidivism – almost double that of interview only and self-report only. Similarly, Walters’ (2006) meta-analysis demonstrated that combining data sources (in this case, file review, interview, and self-report) added incremental validity to predictions compared to relying on a sole source. Schwalbe (2007) demonstrated that ‘file review only’ and ‘direct interview only’ performed just as well as each other for juvenile justice risk assessment instruments but did not compare these with a mixture of the two. The conclusion from this is that assessments will only be as good as the variety of data sources that inform them. Some sources may more readily hold insights into particular factors than others, and it is important that different sources offset the potential biases generated from alternative sources. It may not always be possible to interview an individual under assessment (e.g. consent was not obtained or the assessment is part of a covert investigation), but assessors should try to consider offence paralleling behaviours (e.g. behaviours that are functionally similar to the risk outcome) as well as the risk outcome under consideration to get a more holistic understanding.

Factor 3: tool genesis Campbell et al. (2009) also distinguished risk assessment instruments between those that were derived from criminological theory and/or were created primarily as risk assessment instruments and those that were neither (e.g. instruments focused on issues such as personality, literacy, and self-esteem) but are used to predict violence outcomes. The former, expectedly, performed significantly better at predicting institutional violence and violent recidivism. Singh et al.’s (2011) meta-analysis of nine different risk assessment instruments consistently showed that instruments designed for specific populations produced higher rates of predictive validity than more generally oriented instruments. More specifically, Hanson and Morton-Bourgon’s (2009) meta-analysis similarly showed that risk assessment instruments designed specifically to assess the risk of sexual recidivism performed significantly better than instruments designed for violent recidivism when used to predict sexual recidivism. The inverse was also true, whereby sexual violence risk assessment instruments struggled with predicting violence recidivism compared to those tools designed to assess its likelihood. van den Berg et al.’s (2018) meta-analysis showed that instruments designed for sexual recidivism risk performed significantly better at predicting sexual risk compared to violent risk. Olver et al. (2009) conducted a meta-analysis of three risk assessment instruments not designed for sexual recidivism.3 For each instrument, performance was significantly greater in predicting general, non-violent, and violent recidivism compared to predicting sexual recidivism. In the case of SAVRY, the effect sizes were five times smaller for sexual recidivism, and in the case of PCL, they were two to three times smaller. However, van der Put et al.’s (2019) meta-analysis of domestic violence risk assessment demonstrated that those instruments developed specifically for domestic violence performed just as well as other more generic violence risk assessment instruments. 482

Risk assessment instruments can help predictions

The key conclusion, therefore, is that violence risk assessment instruments perform better than psychometric scales and that risk assessment instruments typically perform best when built specifically for a certain crime type and deployed against that crime type.

Factor 4: authorship bias Singh et al.’s (2013) meta-analysis of nine risk assessment instruments demonstrated a significant authorship effect on studies validating these instruments. Average effect sizes were significantly higher when instrument designers were involved in a validation study. Singh et al. (2013) propose two alternative explanations. First, the result may be incidental. Designers will be more familiar with the instrument, and those working with the instrument may more prodigiously stick to the its guidelines, thus producing a greater effect. Second, designers may be more likely to not submit studies for publication that show poor instrument performance. Either way, the effect size differential is so large that all designers should formally declare their engagement in the study and the potential for authorship bias to have occurred.

Blending insights from different instruments can improve predictive accuracy As previously mentioned, there is no shortage of violent extremism risk assessment instruments available (e.g. ERG 22+, IVP, MLG, OVEST, VERA 2R, and TRAP 18). The same is true for other forms of violence, including general violence (HCR 20 v3, BARR-2002R, VRAG, COVR, VRS, SAPROF, SAVRY), domestic violence (SARA v3, DASH, DVI, ODARA, B-SAFER), sexual violence (SVR 20 v2, RSVP, ERASOR, J-SOAP II, PROFESOR, and VASOR 2), and stalking (SRP, SAM). It naturally leads to questions about which instrument is best. If we judge ‘best’ naively by predictive validity, then there is no clear answer when there are two or more well designed instruments. Because each instrument tends to draw on a similar gene pool of scientific evidence, the items they include can often be definitionally similar, thus leading to a high level of correlation between the outcomes generated. For example, O’Shea and Dickens’ (2014) systematic review found multiple studies demonstrating positive correlations between the START instrument with SAPROF, PCR:SV, and individual domains of the HCR-20. Viljoen et al.’s (2012) meta-analysis found correlations between four sex offending risk assessment instruments (J-SOAP-II, ERASOR, J-SORRAT-II, and Static-99). Olver et al.’s (2009) meta-analysis of the performance of LSI, PCL, and SAVRY in youth samples similarly shows negligible differences in the performance between these instruments when applied to the same samples. By synthesising the results of 20 studies that used three actuarial sex offender risk assessment instruments, Babchishin et al. (2012) investigated whether combining risk scales improved performance. The meta-analysis showed that each scale consistently positively contributed to the other’s performance when predicting recidivism. van den Berg et al. (2018) replicated this finding in their meta-analysis of dynamic sex offence risk assessment instruments. Babchishin et al. (2012) also considered what is the most efficient process to consider multiple scales (e.g. take the highest, take the lowest, or average). The analysis demonstrated that it was better to average the risk ratings than to consistently choose the instrument that generated the highest score for an individual case. The combination of multiple instruments therefore managed to negate the risk of underestimating and overestimating the risk of recidivism from case to case. There are therefore potentially grounds to make use of multiple instruments in a single assessment, but assessors should have a thorough understanding of the different domains of risk/protective factors incorporated into different instruments and how this may relate to the outcome of interest. 483

Paul Gill et al.

Prediction is not everything There are three approaches to risk assessment: ‘unstructured’, ‘actuarial’, and ‘structured professional judgement’. Whereas the latter two require risk assessment instruments, the former sees practitioners rely solely on judgement. The identification of characteristics would be based on previous experience, the practitioner’s memory of previous empirical findings, and perhaps intuition (Hilton et al., 2006). This method has proven to be less accurate than assessments based on empirical evidence (Monahan, 2008). Meta-analyses have shown poor predictive performance for such unstructured approaches (Hanson & Morton-Bourgon, 2009), with effect sizes between two and five times smaller than alternative approaches (documented below). If the practitioner uses the actuarial method, the characteristic chosen will be based on empirical evidence from follow-up studies that identify the relationship of the characteristics to the outcome (Hilton et al., 2006). In this manner, the actuarial method differs from the clinical method since it uses statistics to identify the most powerful predictors of offence. However, this can only be done with outcomes that have an extensive and robust scientific basis underpinning them. Despite the reported success of using actuarial approaches to risk assessment, some note the potential pitfalls of strictly following statistical results (Douglas et al., 1999; Douglas & Skeem, 2005). Among these are the difficulty of generalisability beyond the samples used in the development of the tool, the challenge of applying statistical knowledge to a clinical setting, the propensity of actuarial methods to exclude potentially important risk factors, the rigidity of actuarial models, and their lack of space for change. Most importantly, actuarial methods fail to address prevention and risk management. The combination of statistical methods and clinical judgement methods resulted in a new type of risk assessment, namely, the structured professional judgement (SPJ) approach (Douglas & Kropp, 2002; Douglas et al., 2003). SPJ allows practitioners to review all available clinical data to identify any potential risk factors, which are found in a structured manual based on empirical evidence (Douglas et  al., 2003). On the basis of these factors, a final structured risk judgement is made, which indicates the risk of offending (Douglas et al., 2003). Unlike actuarial methods, the SPJ does not include fixed guidelines on how to calculate the level of risk. Instead, SPJ tools are structured to guide the decision-making process of practitioners. Tools in this category include a list of risk factors, all of which have been empirically supported, with guidelines on how these risk factors are coded and how to reach a final judgement of different gradations of risk (Douglas et al., 2003). The major difference between actuarial and SPJ tools is the final risk judgement: actuarial tools only allow for a final decision based on fixed rules and statistical calculations, while SPJ instruments allow the professional to make a final decision considering all risk variables that have been identified (Pedersen et al., 2010). Thus, the expertise of the practitioner is still valued alongside empirical evidence (Douglas et al., 2003). As Logan and Lloyd (2019, pp. 146–147) outline in regard to SPJ approaches to violence risk assessment and management: The substantial part of SPJ-informed risk assessment and management guidance is the identification of risk factors evidenced in the research as being the most relevant to the specific outcome to be prevented . . . and the review of evidence justifying their importance. . . . Depending on the guidance [document], support is offered on understanding the relevance of risk factors to the individual case, the meaning of violence to the person at risk, the scenarios in which that violence may become manifest, and risk management planning involving a range of strategies. 484

Risk assessment instruments can help predictions

SPJ can be completed in an abbreviated form whereby an appraiser identifies risk and protective factors and assigns a rating of high, medium, or low risk ‘based on the pattern of risk factors present in the case’. This is known as ‘SPJ-lite’. ‘Full’ SPJ, on the other hand, additionally requires scenario planning and formulation (Logan & Lloyd, 2019, p. 147). In terms of predictive performance, Singh et al. (2011) found no difference in predictive performance between actuarial and SPJ instruments across many different outcomes. Fazel et al. (2012) found similar. van der Put et al.’s (2019) meta-analysis of domestic violence prediction showed actuarial tools fared significantly better. However, the actuarial method has the potential to disregard the different dynamics of risk, including the nature, severity, imminence, duration, and frequency of future offences (Hart & Logan, 2011). So, while actuarial methods may aid assessment, they do next to nothing to aid management. Although prediction is important, the primary purpose of risk assessment should be prevention. This requires a greater understanding of the individual than a straightforward actuarial score can provide. Engaging in violence is a choice. It is carried out with a specific goal, reward, or benefit in mind. The decision might be made quickly, without much forethought, without knowledge of it being an offence, through carelessness or bad (mis)information – but it is still a decision (Hart & Logan, 2011). SPJ focuses on understanding the factors that led to that decision. Why did they cross the threshold into offence? What was happening for the individual at the time (internally and externally) that facilitated or justified the offence? What were the perceived benefits? What goal did they hope to achieve? Were the negative consequences of offending considered? What were the barriers that could have prevented the offence? With these answers, an assessor can draw inferences about the factors that influenced the onset and continued the problem behaviour, as well as consider the factors that might inhibit further problem behaviour.

Assessments require multivariate explanations of risk Gill et al. (2021a) conducted an umbrella review that synthesised the results of 27 meta-analyses and systematic reviews on predictors of violent recidivism. These 27 evidence syntheses examined at least 526,510 participants and 1,118 primary studies. They found 32 different factors that had been subjected to meta-analysis. Table 31.2 summarises their effect sizes and ranks them. Effect size ranges were consistently low (r < 0.3), with only one factor experiencing a range that went into the moderate range (r = 0.3–0.5). No individual factor had a high effect size (r > 0.5). At first, the consistently low effect sizes may appear underwhelming. But the opposite is true. The results demonstrate that no single factor will explain the emergence of the risk. The key is to think about multivariate explanations of risk, the inter-relationships between risk factors, and their cumulative impact upon recidivism likelihood. Putting gender aside, the strongest effect sizes consistently refer to aspects of personality disorders, prior anti-social/criminal/violent behaviour, the interaction of several poor life situations combined (e.g. living conditions, employment, and education), and other factors that impact selfregulation (e.g. substance abuse). On the other hand, socio-demographic characteristics like age, race, and others were among the smallest effect sizes. The umbrella review also highlights a number of areas where evidence is lacking. For example, no meta-analyses examined family criminality, self-esteem, social skills, or social ecological factors that consider the impact of community dynamics. Only one low-quality study examined the impact of prison sentence length and did not consider the moderating impact of age. Few studies examined the differential impact of either follow-up periods (e.g. do some risk factors matter more in different follow-up periods) or whether the index offence involved co-offenders or not. Few 485

Paul Gill et al. Table 31.2  Risk Factor Effect Size for Violence Perpetration Factor

Effect Size From Highest Quality Study (%)

Factor

Effect Size From Highest Quality Study (%)

Anti-social personality disorder Personality disorder (Any)

0.44 0.29

0.13 0.13

Anti-social behaviours Gender (male)

0.27 0.24

Psychopathic personality disorder Living situation (poor) Alcohol use Violent behaviour (previous) Employment (no)/poor education combined Self-regulation (poor) Drug use Leisure (poor) Anti-social attitudes

0.24 0.23 0.22 0.21 0.20

Substance abuse (any) Personal/psychological problems Anti-social personality traits Psychopathic personality disorder Criminal behaviour (previous) Childhood sexual abuse victim Family problems Relationship problems Married (no)

Age (older) Criminal peers Weapon use/threats/physical injuries caused

0.18 0.18 0.17 0.16

Race (minority) Age (younger) Socio-economic status (low) Childhood physical abuse victim Mental disorder (yes) Employment (no) Education (poor)

0.16 0.14 0.14

0.13 0.13 0.11 0.11 0.10 0.08 0.06 0.03 0.01 Non-significant Non-significant Non-significant Non-significant Non-significant

Source: Created by the authors

meta-analyses considered mediating or moderating effects across studies, and this demonstrates potentially major opportunities for future research. The definition of risk and protective factors was pretty uniform throughout the evidence syntheses. Simply put, studies noted the presence of a factor that has a theoretical or pre-existing empirical correlation with offending and/or recidivism. Often, syntheses focus on a single factor, like psychopathic traits (Asscher et al., 2011). Studies also simply analysed whether the presence of a ‘risk factor’ was greater in the violent recidivism group compared to the non-violent recidivism group. There was next to nothing that sought to understand the relevance of a risk factor to the outcome of violent recidivism. This is for two main reasons. First, presence suits idiographic research designs informed by the simple coding that actuarial risk assessment tools produce (e.g. everything coded as yes or no). Relevance is potentially much more suited to qualitative nomothetic research designs, which are unsuitable for meta-analyses. Second, SPJ manuals do not have an overt focus on prediction in the same way that actuarial tools do. SPJ assesses and manages, whereas actuarial tools predict. Therefore, actuarial tools and the data they use lend themselves better to meta-analyses.

Implications for violent extremism risk assessment and management The empirical evidence base for validated risk and protective factors for terrorism lags behind the evidence base for other forms of violence. This should be unsurprising given these are relatively 486

Risk assessment instruments can help predictions

new literatures, but both have been hampered by lengthy definitional debates. The contours of what demarcates (a) terrorist engagement from disengagement, (b) radicalisation from deradicalisation, and (c) re-engagement from recidivism remain heavily contested. However, this is not a new problem. Gill et al.’s (2021a) umbrella review of predictors of violent recidivism demonstrates that this is also true for the general violent recidivism literature, which means it is surmountable (Table 31.3). Transparent and flexible research practices are the key to pushing science forward in this space. For example, consider the grounds for a rigorous violent extremism re-engagement/recidivism study and set aside the research difficulties associated with its comparatively lower base rate of index offending, and potentially difficult access to official data. A major difficulty in such a study would be overcoming the boundary confusion in terms of where re-engagement starts and ends and when it transitions into violent extremist recidivism. One route to transparency would be to Table 31.3  Recidivism Definitions in Different Meta-Analyses Study

Parole Rearrest/ New Reconvic- Reincar- Recommitment Treatment Violations Re-offence Charges tion ceration to Psych Failure Facility

Asscher et al. (2011)

X

X

Bonta and Blais (2014)

X

X

Doherty & Bersani (2018)

X

Dowden and Brown (2002) Eher et al. (2016)

X

Eisenberg et al. (2019)

X

X

X

X

X X

X

Geerlings et al. (2020)

X

Gerhold et al. (2007)

X

Golenkov et al. (2014)

X

Hanson and Bussiere (1998) Hanson & MortonBourgon (2005) Kreis et al. (2014)

X X

Maille et al. (2011) McCann and Lussier (2008) Mohr-Jensen and Steinhausen (2016) Olver et al. (2014)

X

X

X

X

X

X

X

X

X

X X

X

X

X

X

X

X

X

X X

Olver et al. (2011) Viljoen et al. (2012)

X

Wibbelink et al. (2017)

X

Yu et al. (2012)

X

X

Source: Created by the authors

487

X

Paul Gill et al.

develop a standardised and widely acceptable definition of violent extremism re-engagement and recidivism prior to any data being collected. Such a definition would articulate inclusion criteria in terms of offence type and time span under consideration. From experience, this is likely to be an exhaustive process that is unlikely to achieve the desired consensus that manages to keep all academics and practitioners satisfied. Instead of this exhaustive and likely unsatisfying route, we would suggest a more pragmatic route that prioritises data collection and flexible definitions. Each small facet of a pre-determined definition has the potential to sway results in any given direction. Instead, a project could list all of the potential facets of a definition and dummy code (e.g. code as yes or no) for each one. Rather than checking whether an individual had recidivated against a predetermined definition, coders would answer separately whether there was a parole violation (this is the closest approximation to re-engagement), a re-arrest, a re-offence, new charges, reconvictions, reincarcerations, and so on. They could code whether the recidivism was a terrorism offence or not and whether it occurred within one, three, or five years, and so on. The outcomes of such a pragmatic, data-driven process would show how re-engagement/recidivism rates (and their risk factors) differ according to different definitional permutations. Next, we return to the five key conclusions we derived and consider the implications for the risk assessment and management of terrorism.

Risk assessment instruments can help predictions The emphasis here is ‘help’. Risk assessments are not foolproof because risk is thoroughly dynamic, subject to ebb and flow, information will be missing or lost between agencies, and there are only finite resources. This message is important for the general public, policymakers, and politicians alike because the expectation is often that 0% risk is the only acceptable outcome in terms of terrorism risk assessment and management. Given the oftentimes more sophisticated, predatory, and crafty behaviour associated with violent extremism versus other forms of criminal activity (which are often impulsive and substance abuse-driven), the expectation that an instrument will provide better results for the former than decades of research have demonstrated for the latter is fanciful. On top of that, risk assessment predictions generally work best for high volume incidents. It was striking to see meta-analyses refer to the sometimes poor performance of sexual violence recidivism prediction as being due to that crime type’s ‘low base rate’. Compared to terrorism, it is not a low base rate. So, this again should temper our expectations about the prediction of violent extremism, and this provides a large challenge for policy-makers to effectively communicate both the advantages and limits of such approaches.

Risk assessment performance can be moderated by several design features Traditionally, the focus of general violence risk assessment instruments has been on the differential importance of static and dynamic risk factors. This has happened in parallel in the study of radicalisation where focus has shifted away from root cause explanations to more group-/networklevel and dynamic influences (Sageman, 2004; Bouhana & Wikstrom, 2011). Given the social nature of terrorist engagement, it is likely that additional factors require consideration for this problem area also. Issues related to beliefs and attitudes are often overlooked elsewhere but may relate to the facets of an individual we would typically associate with being radicalised. Social context and capability factors are also often overlooked because of the simple nature of most forms of violence, but these factors move closer towards the individual’s immediate social network and

488

Risk assessment instruments can help predictions

capability for violence. More work is required to consider how this impacts threat assessments, whether this type of outcome is readily observed, and what types of systems are needed to watch for changes over time.

Blending insights from different instruments can improve predictive accuracy Existing violent extremism assessments and management guidance differ in their focus. The risk specifications of some tools vary in focus, such as assessing vulnerability to radicalisation (IMV), Jihadist extremist tendencies (EMI), and online radicalisation (OVEST). The MLG and TRAP-18 are limited to ‘group violence’ and lone-actor terrorism, respectively. Others are slightly broader, for example, ERG (risk of re-offending) and VERA (violent extremism). Table 31.4 notes the percentage of risk factors for each tool, further demonstrating the different types of focus (Gill et al., 2021b). It is likely that some assessment cases could benefit from multiple instruments being looked at. However, a bigger challenge likely awaits the development and validation of risk assessment instruments in this space. Repeatedly, it has been demonstrated that radicalisation and terrorist engagement should be viewed as a process with various discrete phases along the way. These discrete phases depict multiple types of ‘risk’ related to extremism and terrorism. These phases are labelled differently across various theoretical models and include Precht’s (2007) phases of preradicalisation, conversion and identification, conviction and indoctrination, and action; Borum’s (2003) terrorist mindset model, which includes the stages of grievance, injustice, target attribution, and devaluation; Wiktorowicz’s (2004) radicalisation model, which includes the stages of cognitive opening, religious seeking, frame alignment, socialisation, and joining; Moghaddam’s (2005) ‘staircase to terrorism’ model involving psychological interpretations of material conditions, perceived options to fight unfair treatment, displacement of aggression, moral engagement, solidification of categorical thinking, and the perceived legitimacy of the terrorist organisation and the terrorist act and sidestepping inhibitory mechanisms; Silber and Bhatt’s (2007) radicalisation model involving pre-radicalisation, self-identification, indoctrination, and jihadisation; de Coensel’s (2018) meta-framework of pre-radicalisation, cognitive opening, solution seeking, interest, targeting, indoctrination/acceptance, implementation, and post-implementation; Khalil et al.’s (2022) ABC model, which considers attitudes (including the extent of opposition to and sympathy Table 31.4  Distribution of Risk Factor Types by Violent Extremism Instrument Tool

Static Risk Factors

Dynamic Risk Factors

Beliefs, Attitudes, Behaviours

Social Context and Capability

Protective Factors

IVP EMI OVEST ERG22+ MLG MLG2 VERA VERA2 TRAP-18

0% 0% 7.69% 13.67% 30.43% 23.81% 12.5% 13.79% 16.66%

31.25% 0% 7.69% 4.55% 17.39% 19.05% 8.33% 0% 27.77%

25% 100% 30.77% 59.09% 21.74% 23.81% 37.5% 51.72% 44.44%

43.75% 0% 30.77% 27.27% 30.43% 33.33% 20.83% 17.24% 11.11%

0% 0% 23.07% 0% 0% 0% 20.83% 17.24% 0%

Source: Created by the authors

489

Paul Gill et al.

for ideologically justified violence), behaviours (including the extent of involvement in ideologically justified violence), and how these can be measured in a context specific manner. What constitutes a risk factor for vulnerability to radicalisation may not necessarily represent a risk factor for terrorist engagement. Similarly, risk factors may play different roles at different points in the phases. Risk assessment instruments in this space should therefore be mindful of what phase of the process they are considering the likelihood of harmful outcomes occurring.

Prediction is not everything Fortunately, for violent extremism risk assessment, prediction is not everything because the science lags far behind the other forms of violence outlined in the preceding sections. Few instruments here have demonstrated the ability to discriminate between persons of interest and those that are not; few have demonstrated inter-rater reliability; and certainly none have been tested to the extent that a meta-analysis is achievable. Fortunately, SPJ approaches are currently in vogue in this space and widely seen as best practice among those who have responsibility for risk assessment in a variety of clinical, forensic, and custodial settings. Each of the instruments in this space, either explicitly or implicitly (e.g. IVP), identifies itself as an SPJ approach. However, the grounds on which they self-identify as SPJ differ. The likes of IVP, EMI, OVEST, and VERA encourage the first two steps of SPJ (e.g. gather information and resources; determine the relevance of factors to individual risk) but fall far short of providing advice on formulation, scenario planning, and management strategies. IVP, for example, provides ‘useful and relevant information to practitioners about issues they should consider’ and encourages decisions to be made ‘in conjunction with practitioners’ professional judgement and expertise, and consultation with line management and agency policy’ (Cole et al., 2010, p. 2). IVP falls short of providing the full SPJ process of formulation, scenario planning, and management strategies. The ERG22+ goes a step further than the IVP and uses formulation but not scenario planning. The TRAP-18 and MLGs are the closest matches in terms of the degree to which they mirror SPJ development guidelines. This is particularly so with the MLG, whose developers were also heavily involved with Historical Clinical Risk-20 (HCR-20), the Guidelines for Stalking Assessment and Management (SAM), and other highly rated risk assessment tools for other types of violence. The TRAP-18 manual suggests formulations and management approaches. However, the guidance provided is comparatively very limited, with just one page in the guidance document, and appears underdeveloped compared with the MLG. Unlike actuarial approaches, SPJ approaches do not include fixed guidelines on how to calculate the level of risk and instead allow practitioners to review all available evidence to identify any potential risk (or, if applicable, protective) factors. SPJ approaches then typically direct a user to code the relevance of an individual indicator. By relevance, we mean the degree to which the item motivates, disinhibits, or destabilises the risks posed by the individual being assessed (e.g. its functional relation to offending). SPJ then typically involves the generation of a statement of understanding about the case (e.g. a formulation). It is imperative to synthesise the quality of formulations and determine whether (a) formulation quality impacts outcomes, (b) formulations conducted remotely differ from those informed by contact with the individual, (c) training and experience impact the quality of formulations. Additionally, we know very little right now about how risk assessments inform management approaches. Do the presence/volume of risk factors impact subsequent intervention intensity? Are interventions emanating from formulations evidence-based, tailored to culture and individual characteristics, and do they draw on building 490

Risk assessment instruments can help predictions

protective factors? Questions such as these are paramount for risk assessment and management, particularly for an understanding of what works, for whom, and in which contexts.

Assessments require multivariate explanations of risk Empirical research demonstrates that violent extremism is usually the culmination of a complex mix of personal, political, and social drivers that crystallise at the same time to drive the individual down the path of violent action. Whether the violence comes to fruition is usually a combination of the availability and vulnerability of suitable targets that suit the heady mix of personal and political grievances and the individual’s capability to engage in an attack from both psychological and technical capability standpoints. Many individual cases share a mixture of unfortunate personal life circumstances coupled with an intensification of beliefs that later developed into the idea of engaging in violence. What differed was how these influences were sequenced. Sometimes personal problems lead to a susceptibility to ideological influences. Sometimes long-held ideological influences became intensified after the experience of personal problems. This is why we should be wary of mono-causal master narratives. The development of these behaviours is usually far more labyrinthine and dynamic. Risk assessment instruments need to embrace this complexity rather than shy away from it with pre-determined algorithms. SPJ-informed assessments aim to make sense of an individual’s behaviour in the context of their unique circumstances. The ‘risk’ someone poses is subject to change. It is driven by situations and circumstances. This is particularly true for adolescent populations, who are subject to major changes in their lives. An instrument should help assessors determine the nature and degree of risk an individual poses for certain kinds of behaviour given their current and future circumstances. By understanding the individual’s trajectory into problematic behaviour, we might gain insight into the factors that might exacerbate or stem the issue. Assessments can inform our understanding of future likely scenarios and the tailored management plans that are required to avert harm. We need to look at each case on its individual merits by weighing and integrating the evidence and gaining an overall understanding of the individual, their behaviour, intentions, and motives. We seek to understand whether there is a risk of harm, its nature, scale, severity, and imminence. Through this, we seek to prevent potentially harmful outcomes. To do so, we require priorities for intervention that guide risk management. Identifying these priorities is contingent on getting a thorough understanding of the individual. Assessors should not weigh the presence of each risk factor equally. If an individual has a lot of risk factors but little proclivity or skills for designing a bomb, the risk of a bombing is likely low. On the other hand, there might be cases where only one factor is present but is highly influential in driving harmful risk. The purpose of an instrument should be to help assessors exercise their best judgement.

Conclusion The synthesis suggests there are several issues that require immediate focus from practitioners and academics alike in the coming years. First, we need to consider how users and their organisations interact with instruments. The meta-analyses and systematic reviews we synthesised throughout this chapter are overtly focused on the instruments, their psychometric properties, and their predictive capabilities. However, behind every risk assessment is an individual practitioner, behind them is an organisation, and in front of them both are individuals in need of assessment, many requiring some form of intervention. Practitioners will differ in experience, professional 491

Paul Gill et al.

background, and degree of training. Institutions will differ in terms of caseload, supervisory capacity, and data access. Those assessed as ‘high-risk’ will likely have greater wrap-around interventions, which, in theory, should disrupt the onset of offending (and make the overt focus on prediction even more curious). Yet none of these issues have been synthesised thus far, yet they are likely far more consequential for real-life harm reduction than an instrument’s predictive powers, and especially so when low base rate events are the subject of enquiry. Future work should endeavour to do so through a systematic review of the human factors side of risk assessment and management. Second, the move towards SPJ instruments is a welcome one. They often perform just as well as actuarial instruments on prediction but offer a lot more insight for management. However, they need to be developed specifically for certain types of harmful outcomes. Any violent extremism assessment and management guidance needs to make best use of available science and work closely with practitioners (e.g. through Delphi studies, interviews, and focus groups) regarding the types of information that typically may be available at different points of an assessment and intervention. This is likely to be highly context-specific and any learning from other domains and countries needs to reflect what is possible within a given domain. Finally, the vast majority of the research reviewed here focuses on recidivism, where individual cases are referred into a system because of some criminal wrongdoing. However, violent extremism assessment and management often work predominantly in the pre-crime space, and referrals arrive from a range of other sources in the community that may have different levels of veracity behind them. Unfortunately, little is currently known about the systems of triage that determine the suitability of a referral and prioritise competing referrals in situations where information is often incomplete and decisions are time-sensitive.

Notes 1 Portions of this material are based on work supported by the U.S. Department of Homeland Security, Science and Technology Directorate, Grant Award Number, 20STTPC00001‐02–01. The views and conclusions included here are those of the authors and should not be interpreted as necessarily representing the official policies, either expressed or implied, of the U.S. Department of Homeland Security. 2 These are denoted with an * in the reference list. 3 The LSI was developed to assess the risk of general recidivism; the PCL is a diagnostic tool to reliably measure the presence of psychopathic features; and the SAVRY was designed to assess the risk of violence.

References1 Asscher, J. J., van Vugt, E. S., Stams, G. J., Deković, M., Eichelsheim, V. I., & Yousfi, S. (2011). The relationship between juvenile psychopathic traits, delinquency and (violent) recidivism: A meta-analysis. The Journal of Child Psychology and Psychiatry, 52(11), 1134–1143. *Babchishin, K. M., Hanson, R. K., & Helmus, L. (2012). Even highly correlated measures can add incrementally to predicting recidivism among sex offenders. Assessment, 19(4), 442–461. Bonta, J., Blais, J., & Wilson, H. A. (2014). A theoretically informed meta-analysis of the risk for general and violent recidivism for mentally disordered offenders. Aggression and violent behavior, 19(3), 278–287 Borum, R. (2003). Understanding the terrorist mind-set. FBI Law Enforcement Bulletin, 72, 7. Bouhana, N., & Wikstrom, P. O. (2011). Al Qai’da-influenced radicalisation: A rapid evidence assessment guided by situational action theory. Home Office. *Campbell, M. A., French, S., & Gendreau, P. (2009). The prediction of violence in adult offenders: A metaanalytic comparison of instruments and methods of assessment. Criminal Justice and Behavior, 36(6), 567–590.

492

Risk assessment instruments can help predictions Cole, J., Alison, E., Cole, B., & Alison, L. (2010). Guidance for identifying people vulnerable to recruitment into violent extremism. University of Liverpool, School of Psychology. De Coensel, S. (2018). Processual models of radicalization into terrorism: A best fit framework synthesis. JD: Journal for Deradicalization, 17, 89–127. Doherty, E. E., & Bersani, B. E. (2018). Mapping the age of official desistance for adult offenders: Implications for research and policy. Journal of Developmental and Life-Course Criminology, 4(4), 516–551. Douglas, K. S., & Kropp, P. R. (2002). A prevention-based paradigm for violence risk assessment: Clinical and research applications. Criminal Justice and Behavior, 29(5), 617–658. Douglas, K. S., Ogloff, J. R., & Hart, S. D. (2003). Evaluation of a model of violence risk assessment among forensic psychiatric patients. Psychiatric Services, 54(10), 1372–1379. Douglas, K. S., Ogloff, J. R., Nicholls, T. L., & Grant, I. (1999). Assessing risk for violence among psychiatric patients: The HCR-20 violence risk assessment scheme and the psychopathy checklist: Screening version. Journal of Consulting and Clinical Psychology, 67(6), 917. Douglas, K. S., & Skeem, J. L. (2005). Violence risk assessment: Getting specific about being dynamic. Psychology, Public Policy, and Law, 11(3), 347. Dowden, C., & Brown, S. L. (2002). The role of substance abuse factors in predicting recidivism: A metaanalysis. Psychology, Crime and Law, 8(3), 243–264. Eher, R., Schilling, F., Hansmann, B., Pumberger, T., Nitschke, J., Habermeyer, E., & Mokros, A. (2016). Sadism and violent reoffending in sexual offenders. Sexual Abuse, 28(1), 46–72 Eisenberg, M. J., Van Horn, J. E., Dekker, J. M., Assink, M., Van Der Put, C. E., Hendriks, J., & Stams, G. J. J. (2019). Static and dynamic predictors of general and violent criminal offense recidivism in the forensic outpatient population: A meta-analysis. Criminal justice and behavior, 46(5), 732–750. Europol. (2020). European Union Terrorism situation and trend report. European Union Agency for Law Enforcement Cooperation. *Fazel, S., Singh, J. P., Doll, H., & Grann, M. (2012). Use of risk assessment instruments to predict violence and antisocial behaviour in 73 samples involving 24 827 people: Systematic review and meta-analysis. BMJ Journal, 345. Geerlings, Y., Asscher, J. J., Stams, G. J. J., & Assink, M. (2020). The association between psychopathy and delinquency in juveniles: A three-level meta-analysis. Aggression and violent behavior, 50, 101342. Gerhold, C. K., Browne, K. D., & Beckett, R. (2007). Predicting recidivism in adolescent sexual offenders. Aggression and Violent Behavior, 12(4), 427–438. *Gill, P., Kyprianides, A., & Marchment, Z. (2021a). A rapid evidence assessment of predictors of recidivism in violent offending and violent extremism. Report for UK Government. Gill, P., Zolghadriha, S., & Salman, N. (2021b). Testing the validity, reliability and equity of terrorist risk assessment tools. Report for Public Safety Canada. Golenkov, A., Nielssen, O., & Large, M. (2014). Systematic review and meta-analysis of homicide recidivism and schizophrenia. BMC Psychiatry, 14(1), 46. Hanson, R. K., & Bussiere, M. T. (1998). Predicting relapse: a meta-analysis of sexual offender recidivism studies. Journal of consulting and clinical psychology, 66(2), 348. Hanson, R. K., & Morton-Bourgon, K. E. (2005). The characteristics of persistent sexual offenders: A metaanalysis of recidivism studies. Journal of Consulting and Clinical Psychology, 73(6), 1154–1163. *Hanson, R. K., & Morton-Bourgon, K. E. (2009). The accuracy of recidivism risk assessments for sexual offenders: A meta-analysis of 118 prediction studies. Psychological Assessment, 21(1), 1. Hart, S. D., & Logan, C. (2011). Formulation of violence risk using evidence-based assessments: The structured professional judgment approach. Forensic Case Formulation, 83–106. *Helmus, L., Hanson, R. K., Thornton, D., Babchishin, K. M., & Harris, A. J. (2012). Absolute recidivism rates predicted by static-99R and static-2002R sex offender risk assessment tools vary across samples: A meta-analysis. Criminal Justice and Behavior, 39(9), 1148–1171. Hilton, N. Z., Harris, G. T., & Rice, M. E. (2006). Sixty-six years of research on the clinical versus actuarial prediction of violence. The Counseling Psychologist, 34(3), 400–409. Home Office. (2021). Individuals referred to and supported through the prevent programme, April 2020 to March 2021. HSAI Statistics. Khalil, J., Horgan, J., & Zeuthen, M. (2022). The attitudes-behaviors corrective (ABC) model of violent extremism. Terrorism and Political Violence, 34(3), 425–450.

493

Paul Gill et al. Kreis, M. K. F., Schwannauer, M., & Gillings, K. (2014). Relational risk factors for reoffending in women: A systematic review. International Journal of Forensic Mental Health, 13(4), 381–393. Logan, C., & Lloyd, M. (2019). Violent extremism: A comparison of approaches to assessing and managing risk. Legal and Criminological Psychology, 24(1), 141–161. Mallie, A. L., Viljoen, J. L., Mordell, S., Spice, A., & Roesch, R. (2011, October). Childhood abuse and adolescent sexual re-offending: A meta-analysis. In Child & Youth Care Forum (Vol. 40, pp. 401–417). Springer US. McCann, K., & Lussier, P. (2008). Antisociality, sexual deviance, and sexual reoffending in juvenile sex offenders: A meta-analytical investigation. Youth Violence and Juvenile Justice, 6(4), 363–385. Moghaddam, F. M. (2005). The staircase to terrorism: A psychological exploration. American Psychologist, 60(2), 161. Mohr-Jensen, C., & Steinhausen, H. C. (2016). A meta-analysis and systematic review of the risks associated with childhood attention-deficit hyperactivity disorder on long-term outcome of arrests, convictions, and incarcerations. Clinical Psychology Review, 48, 32–42. Monahan, J. (2008). Structured risk assessment of violence. Textbook of Violence Assessment and Management, 17–33. *Olver, M. E., Stockdale, K. C., & Wormith, J. S. (2009). Risk assessment with young offenders: A metaanalysis of three assessment measures. Criminal Justice and Behavior, 36(4), 329–353. Olver, M. E., Stockdale, K. C., & Wormith, J. S. (2011). A meta-analysis of predictors of offender treatment attrition and its relationship to recidivism. Journal of Consulting and Clinical Psychology, 79(1), 6. *Olver, M. E., Stockdale, K. C., & Wormith, J. S. (2014). Thirty years of research on the level of service scales: A meta-analytic examination of predictive accuracy and sources of variability. Psychological Assessment, 26(1), 156. *O’Shea, L. E., & Dickens, G. L. (2014). Short-term assessment of risk and treatability (START): Systematic review and meta-analysis. Psychological Assessment, 26(3), 990. *O’Shea, L. E., Mitchell, A. E., Picchioni, M. M., & Dickens, G. L. (2013). Moderators of the predictive efficacy of the historical, clinical and risk management-20 for aggression in psychiatric facilities: Systematic review and meta-analysis. Aggression and Violent Behavior, 18(2), 255–270. Pedersen, L., Rasmussen, K., & Elsass, P. (2010). Risk assessment: The value of structured professional judgments. International Journal of Forensic Mental Health, 9(2), 74–81. Precht, T. (2007). Home grown terrorism and Islamist radicalisation in Europe. www.justitsministeriet.dk/sites/ default/files/media/Arbejdsomraader/Forskning/Forskningspuljen/2011/2007/Home_grown_terrorism_ and_Islamist_radicalisation_in_Europe_-_an_assessment_of_influencing_factors__2_.pdf *Pusch, N., & Holtfreter, K. (2018). Gender and risk assessment in juvenile offenders: A meta-analysis. Criminal Justice and Behavior, 45(1), 56–81. *Rossdale, S. V., Tully, R. J., & Egan, V. (2020). The HCR-20 for predicting violence in adult females: A meta-analysis. Journal of Forensic Psychology Research and Practice, 20(1), 15–52. Sageman, M. (2004). Understanding terror networks. University of Pennsylvania Press. Sarma, K. M. (2017). Risk assessment and the prevention of radicalization from nonviolence into terrorism. American Psychologist, 72(3), 278. *Schwalbe, C. S. (2007). Risk assessment for juvenile justice: A meta-analysis. Law and Human Behavior, 31(5), 449. *Schwalbe, C. S. (2008). A meta-analysis of juvenile justice risk assessment instruments: Predictive validity by gender. Criminal Justice and Behavior, 35(11), 1367–1381. Silber, M. D., Bhatt, A., & Analysts, S. I. (2007). Radicalization in the West: The homegrown threat (pp. 1–90). Police Department. *Singh, J. P., Grann, M., & Fazel, S. (2011). A comparative study of violence risk assessment tools: A systematic review and metaregression analysis of 68 studies involving 25,980 participants. Clinical Psychology Review, 31(3), 499–513. *Singh, J. P., Grann, M., & Fazel, S. (2013). Authorship bias in violence risk assessment? A systematic review and meta-analysis. PLoS ONE, 8(9), e72484. *Tully, R. J., Chou, S., & Browne, K. D. (2013). A systematic review on the effectiveness of sex offender risk assessment tools in predicting sexual recidivism of adult male sex offenders. Clinical Psychology Review, 33(2), 287–316.

494

Risk assessment instruments can help predictions *van den Berg, J. W., Smid, W., Schepers, K., Wever, E., van Beek, D., Janssen, E., & Gijs, L. (2018). The predictive properties of dynamic sex offender risk assessment instruments: A meta-analysis. Psychological Assessment, 30(2), 179. *van Der Put, C. E., Gubbels, J., & Assink, M. (2019). Predicting domestic violence: A meta-analysis on the predictive validity of risk assessment tools. Aggression and Violent Behavior, 47. *Viljoen, J. L., Mordell, S., & Beneteau, J. L. (2012). Prediction of adolescent sexual reoffending: A metaanalysis of the J-SOAP-II, ERASOR, J-SORRAT-II, and static-99. Law and Human Behavior, 36(5), 423. *Walters, G. D. (2006). Risk-appraisal versus self-report in the prediction of criminal justice outcomes: A meta-analysis. Criminal Justice and Behavior, 33(3), 279–304. Wibbelink, C. J., Hoeve, M., Stams, G. J. J., & Oort, F. J. (2017). A meta-analysis of the association between mental disorders and juvenile recidivism. Aggression and Violent Behavior, 33, 78–90. Wiktorowicz, Q. (2004). Joining the cause: Al-Muhajiroun and radical Islam. In The roots of radical Islam. Department of International Studies, Rhodes College. Wolfowicz, M., Litmanovitz, Y., Weisburd, D., & Hasisi, B. (2020). A field-wide systematic review and metaanalysis of putative risk and protective factors for radicalization outcomes. Journal of Quantitative Criminology, 36(3), 407–447. Yu, R., Geddes, J. R., & Fazel, S. (2012). Personality disorders, violence, and antisocial behavior: A systematic review and meta-regression analysis. Journal of personality Disorders, 26(5), 775–792.

495

Paul Gill et al.

Appendix 1 We conducted a keyword search of titles and abstracts in ProQuest and PsycINFO for papers published from database inception until 28 November 2020, restricted by English language. For ProQuest, we used the following databases: PRISMA, Public Health Database Information, and Psychology and Social Science Premium Collection. For PsycINFO, we used the following databases: Journals, PsycArticles Full Text, PsycBooks, PsychExtra, PsycInfo, PsycTests, and Social Policy and Practice. We used the same key words for both databases. Key words searched for issues related to recidivism (recidiv* OR re?offen* OR desist* OR re?imprisonment OR re?conviction OR repeat offending OR recur* OR re?arrest OR re?incarceration OR violation OR relapse OR reoccur* OR re?institution* OR setback OR repetit* OR rehabili* OR recall), violence (violen* OR crim* OR offen* OR antisocial OR delinqu* OR aggress* OR abuse OR conflict OR threat OR harm OR assault OR force OR intent* OR serious bodily harm OR victim OR violation), risk or protective factors (risk OR protect* OR predict* OR moderat* OR mediat* OR need OR factor OR cause OR determinant OR indicator OR explain OR predecessor OR associated OR precede OR impact OR antecedent OR promot*), and publication types (meta* OR systematic review OR rapid evidence assessment OR scoping review OR evidence map OR rapid review OR research synthes* OR Campbell). We also conducted forward and backward citation searches to supplement the search. The inclusion criteria sought evidence syntheses on predictors/inhibitors of violent recidivism. For the purposes of the sift, we: • used the HCR-20’s definition of violence: ‘Actual, attempted or threatened infliction of bodily harm on another person. Bodily harm includes both physical and serious psychological harm so long as it substantially interferes with the health or well-being of an individual. Psychological harm includes fear of physical injury and other emotional, mental or cognitive consequences of the act in question’; • understood recidivism as a person’s relapse into criminal behaviour, often after the person receives sanctions or undergoes intervention for a previous crime; • defined a risk factor as something that is considered to contribute (directly or indirectly) to the perpetration of violence. This can be specific to the individual, their upbringing, their context, and their experiences; • broadly defined a protective factor as something that inhibits (directly or by buffering risk factors) the perpetration of violence; • focused on evidence syntheses such as meta-analyses and systematic reviews. For the purpose of the search and sift, we were purposefully broad in our definitions of what constitutes recidivism and what a risk/protective factor for it is. The purpose of this broadness was to give us the ability to flesh out the contours of definitional boundaries across relevant evidence syntheses. Figure  31.1 depicts the search process. Four researchers sifted through 2,854 studies and excluded 2,767 for not meeting all of the definitions above based on an assessment of the title and abstract. Eighty-seven full papers were analysed by one member of the team. Twenty-one were excluded because they were some forms of evaluation on a single type of intervention, not an instrument. Thirty-three were excluded because they solely focused on specific risk factors, not ones that

496

Risk assessment instruments can help predictions

Figure 31.1  Systematic search process and rates of attrition Source: Created by the authors

prescriptively belonged to a named risk assessment instrument. Eight were excluded because they did not distinguish violent recidivism from general recidivism, although the abstract suggested otherwise. Two were excluded because, although the abstract was in English, the full paper was not. The remaining papers and those identified in the forward citation search were analysed, and key sources of information (e.g. study design features, variables, and effect sizes) were extracted by a second member of the team. To allow for a standardised comparison, we converted all Cohen’s d, Odds Ratio, and Standard Normal Deviate z, and Pearson’s Correlation r to Area-under-Curve using one of two online effect size converters (www.escal.site, www.polyu.edu.hk/mm/effectsizefaqs/ calculator/calculator.html).

497

32 CONTENT MODERATION Social media and countering online radicalisation Bharath Ganesh

Introduction In the past decade, jihadist propaganda, the audiences of alt-right celebrities, conspiracy theories, and thriving hate speech cultures on social media platforms have led to serious questions about the role social media plays in terrorism, extremism, and radicalisation. In response to government pressure, platforms expanded their use of content moderation to reduce the presence and visibility of terrorist, extremist, and hateful content on social media platforms. Content moderation is the process of flagging, reviewing, and enforcing platform-defined rules or ‘community guidelines’ on content created by users on platforms or websites that host them (Gillespie, 2018). Violation of these guidelines can be grounds for a user’s post to be taken down or even result in a ban from the platform. Content moderation involves three steps: flagging, which refers to the identification of content that might violate community guidelines, an area that is increasingly automated; review, which refers to the process by which a human or a machine (or some combination of the two) makes a decision on whether or not a piece of content violates community guidelines; and enforcement, which refers to the process by which a decision is made following review and the platform takes action on a piece of content (or user). Content moderation that focuses on terrorist content, extremism, hate speech, and radicalisation has grown significantly since 2014, following ISIS’s use of social media platforms. Platforms have always engaged in content moderation of some form, enforcing their own community guidelines (such as Facebook’s policies about nudity), copyright infringement, and requests from law enforcement, but the current intersection of counter-extremism and social media is a more recent development (Borelli, 2021). Content moderation is now its own industry, with specialised businesses providing services to websites and platforms, though many employees in this industry face low pay, unsafe and unsustainable working conditions, and are poorly supported (Roberts, 2019). Content moderation is particularly expensive considering it requires a significant amount of human labour to review flagged content (see Common, 2020, pp. 128–129), but this is increasingly being augmented with automation (Macdonald et al., 2019). While platforms play an important role in both driving and hosting strategic communications solutions that are increasingly in use in CVE programmes across the world, this chapter focuses specifically on content moderation to provide readers with an in-depth understanding of this process and its implications for countering radicalisation. DOI: 10.4324/9781003035848-36 498

Content moderation

Synthesising an understanding of content moderation and radicalisation is necessary as key debates and evidence are spread across a variety of literatures, disciplines, and reports. This chapter orients readers to this complex field by providing a critical review of policy and academic literature. The goal of this chapter is to orient readers who are encountering the growing literature on content moderation and radicalisation with an exposition of the regulatory context, power dynamics between stakeholders, technologies, and key debates in the field. While this chapter makes passing reference to a variety of platforms, its focus is on three major companies: Facebook, Twitter, and Google (YouTube). Thus, the focus of this chapter is not on ‘alt-tech’ platforms (Donovan, 2018; Jasser et al., 2021) or platforms that extremists have recently exploited (e.g. Discord and Telegram). Instead, the chapter focuses on key industrial actors in the global tech space that have a track record of exploitation by extremists and are at the centre of contemporary governance arrangements to counter radicalisation. Despite the chapter’s focus, the description of the regulatory environment and mapping of content moderation stakeholders are relevant to a wide variety of digital communication services and software. Readers should remember that social media platforms’ counter-radicalisation efforts include strategic communication, counter-propaganda, lobbying, and relationships with government and international organisations (Archetti, 2019; Borelli, 2021; Ganesh & Bright, 2020). Discussing all of these aspects is too ambitious for this chapter, which focuses on content moderation because it constitutes the largest proportion of the actions that platforms take against radicalisation. Further, this is a review of emerging literature ranging across debates on platform governance in media and communications, policy practice, and scholarship on terrorism and extremism. While the actors, tactics, methods, and technologies of content moderation will evolve and change in the next decade, future research into content moderation and radicalisation will benefit from this chapter’s focused, critical exposition of dynamics between stakeholders and the regulatory environment of content moderation. Indeed, during the preparation of this chapter, with final submission in 2021 updates in 2022, and proofs in 2023, the context and actors have changed significantly. The EU passed the Digital Services Act (DSA), Facebook changed its name to Meta, Elon Musk bought Twitter and renamed it ‘X’, and financial tightening in Western economies led to thousands of layoffs at social media corporations, affecting content moderation and trust and safety divisions precipitously. Consequently, readers should focus on broad dynamics as this is a very precarious and unstable industry that is rapidly changing. This chapter begins with a detailed consideration of the changing regulatory context in which content moderation takes place. It focuses on the European Union, as the bloc has been the most active in the development of content moderation regulation concerning radicalisation. Content moderation remains a predominantly self-regulatory practice, with the EU working to shape the parameters of corporate practices and ensure compliance with national law and community guidelines. However, recent legislation on terrorist content and the Digital Services Act set stricter standards, obligations, and liabilities for platforms. Following this, the chapter moves on to discuss stakeholders and technologies involved in content moderation around extremism. The chapter provides a brief overview of automation in content moderation before discussing alternative content moderation strategies that are softer than content takedown. Finally, the chapter covers three key issues in the future of content moderation and radicalisation: deplatforming, lacking transparency, corporations throttling of research and accountability, and the inconsistency of platforms’ enforcement actions.

Content moderation in a changing regulatory context Content moderation is primarily a self-regulatory process. Self-regulation in media refers to governance arrangements in which industrial actors (individually or in concert) themselves set and 499

Bharath Ganesh

enforce rules and standards (Puppis, 2010). Self-regulation is advantageous in that industrial actors have technical expertise; it tends to be more flexible and applicable in areas such as content regulation (Latzer et al., 2013, p. 375). However, it has the disadvantage of relatively weak standards, a tendency towards anticompetitive behaviours, advancement of private interests, and decreased accountability and transparency (Latzer et  al., 2013, p.  375). Self-regulation in contemporary media governance involves a variety of stakeholders (Ginosar, 2013; Puppis, 2010), and governments and civil society have an important role to play in shaping self-regulation (Latzer et al., 2013, p. 377). Recently, research on content moderation policy has offered the concept of platform governance to describe contemporary multistakeholder arrangements such as the development of GIFCT and the Christchurch Call, particularly in the context of content regulation (Gorwa, 2019). While multistakeholder governance is a dominant paradigm for content moderation and recent EU regulation is moving in the direction of co-regulation, in which states shape ‘the formal basis for cooperation with private actors’ (Latzer et  al., 2013, p.  377), the legal environment around content moderation is primarily self-regulatory, though new obligations from the EU to take down terrorist and illegal content are creating liabilities for platforms and setting standards for content moderation. As all the major platforms considered in this chapter are based in the United States, their ability to self-regulate content is mandated by Section 230 of the Communications Decency Act (see Keller, 2018). Section 230 was introduced in 1995 as part of the Communications Decency Act and incorporated into the Telecommunications Act of 1996 (Citron, 2014, p. 170), providing websites with ‘immunity both from publisher and distributor liability for user-generated content’ (2014, p. 171) and protecting platforms if and when they choose to restrict access to ‘objectionable online material’ (2014, p. 170). Today, there are numerous proposals to reform Section 230 (Citron & Franks, 2020), many of which focus on the question of increasing liability for platforms’ algorithmic amplification of speech that is illegal or harmful (Keller, 2021). Recent EU policy is particularly relevant for understanding content moderation and countering extremism (see Keller, 2019). From 2016 to 2021, there have been three major developments that are changing the EU regulatory environment. The first major change came in 2016, when the EU agreed a ‘Code of conduct on countering illegal hate speech online’ with Facebook, Microsoft, Twitter, and YouTube. The code of conduct requires they review notifications for removal of illegal hate speech within 24 hours, have clear processes in place to review notifications, work with trusted flaggers, including civil society (discussed below), and intensify cooperation between platforms on hate speech (European Commission, 2016). As of 1 January 2018, Germany’s Network Enforcement Act (NetzDG) obligated social networking sites with at least two million users to examine reports on illegal hate speech and ‘manifestly unlawful’ content and delete it within 24 hours, with fines up to EUR 50 million for non-compliance (Heldt, 2019). NetzDG received significant criticism on the grounds that it shifts review of what counts as ‘manifestly unlawful’ content to private actors, could lead to excessive takedowns, and mandates too narrow a removal window (Heldt, 2019, pp. 4–5). While these are important concerns, they do not seem to have materialised; according to Wischmeyer’s (2020) analysis, the number of NetzDG-related complaints was lower than expected (2020, p. 53), and most content was removed due to violations of community guidelines (2020, p. 54). On the basis of platforms’ transparency data, Wischmeyer writes, ‘the reports can neither confirm nor refute the [private] “censorship” or the “over-blocking” claims. They only demonstrate that some of the fears associated with the law have clearly been exaggerated’ (2020, p. 55). Indeed, Tworek and Leerssen (2019) and Heldt (2019) make similar points, with their reviews of NetzDG highlighting that platforms have primarily leaned on community guidelines to take down content (Tworek & Leerssen, 2019, p. 6). This work stresses that improvements are still necessary to improve NetzDG 500

Content moderation

and cautions against viewing it as a panacea. They note that procedural improvements are required in terms of transparency, reporting and complaint tools, inconsistencies between platforms, and the inclusion of multiple stakeholders (Heldt, 2019; Tworek & Leerssen, 2019, p. 8). In April 2021, the European Parliament approved a new regulation on terrorist content online that came into effect in June 2022 (Regulation (EU) 2021/784 of the European Parliament and of the Council of 29 April 2021 on Addressing the Dissemination of Terrorist Content Online, 2021). This regulation goes significantly further than NetzDG, stipulating a window of one hour for ‘hosting service providers’ for the removal of terrorist content as defined in Article 3(1) of Directive (EU) 2017/541. Any content likely to constitute terrorist content certainly violates platform terms of service, though the language around advocacy, glorification, and solicitation of a person or group to engage in terrorist activity in the law will cause it to cover a large range of expression (Keller, 2019). Like NetzDG, this regulation sets standards for self-regulation on platforms and also includes further provisions that seek to include multiple stakeholders in the content moderation process. Article 3.1 stipulates that EU member states will set up ‘competent authorities’ with the power to issue removal orders to hosting service providers, who (according to Article 3.3) should remove the content within one hour across the EU (Regulation (EU) 2021/784). Competent authorities are required to provide a detailed explanation of why the content is considered to be terrorist content (Regulation (EU) 2021/784, Article 3.4). The law sets further standards on companies, requiring those exposed to terrorist content to implement sufficient technical and operational resources to ‘identify and expeditiously remove or disable access to terrorist content’, establish user-friendly mechanisms for users to report terrorist content, and leave space for hosting service providers to generate their own solutions so long as they ‘mitigate’ providers’ exposure to terrorist content, are sufficiently targeted and proportionate, and take users’ fundamental rights into account (Regulation (EU) 2021/784, Article 5). Finally, the regulation sets out specific requirements for member states and hosting platforms to publish transparency reports. Just as with NetzDG, there have been numerous, well-founded concerns raised with this regulation. Human rights and civil liberties organisations argued that the proposal incentivises automated content moderation, lacks independent judicial oversight, and instead relies on national competent authorities, and that the one-hour window means that one member state ‘can extend its enforcement jurisdiction beyond its territory without prior judicial review’ (Human Rights Watch, 2021). As the signatories of the letter note, national competent authorities in member states in which political actors compromise the rule of law raise significant risks to fundamental rights, a key area for future research and monitoring of this regulation. All three of these legal developments in the EU (the code of conduct, NetzDG, and terrorist content online regulation) are shaping the self-regulation of global tech platforms. Despite the muscularity of the EU’s terrorist content regulation, it mandates the removal of content that is undoubtedly in violation of platform guidelines. In terms of governance, however, recent legislation in the EU shows that the bloc is increasingly pushing to use statutory mechanisms (primarily fines) to ensure global tech’s self-regulation is consistent with both national law and their own community guidelines. However, outside setting statutory requirements for platforms to meet, the EU’s legal reforms around hate speech and terrorist content are primarily grounded in selfregulation while setting baseline standards. The EU’s Digital Services Act, which was agreed in August 2022 and comes into full effect in 2024, does codify and standardise many of the regulatory practices relevant to terrorist, extremist, and hateful content by creating liabilities and obligations for platforms that include mandatory takedown upon notification, transparency requirements, auditing, and solidifying the importance of national reporting authorities and trusted flaggers. This sets clear limits and obligations on content moderation, as well as setting clear requirements, 501

Bharath Ganesh

limits, and liabilities on self-regulation. Further, in the case of “very large online platforms” that this chapter is chiefly concerned with, the DSA does extend the kinds of content covered beyond “terrorist” or “illegal” content to a wide range of content and actors that might represent “systemic risks” in Article 34. While the Digital Services Act involves a range of new requirements, obligations, and liabilities, it standardises the emerging governance architecture that this chapter describes. Consequently, the role of trusted flaggers, automation, and content moderation interventions beyond takedown – which this chapter will discuss next – will remain crucial aspects of the content governance architecture in the EU.

Trusted flaggers and Internet referral units Historically, most platforms relied on crowdsourced flagging, which meant that their users would flag content that may violate community guidelines to platforms by using interfaces available to users. Crawford and Gillespie argue that through these interfaces, platforms frame flagging categories such that they can ‘retain the ability to make judgements on content removal, based on ad hoc and often self-interested assessments’ (2016, p. 419) and provide ‘little room for expressing the degree of concern, or situating the complaint, or taking issue with the rules’ (2016, p. 418). Platforms do continue to act on content flagged by their users, but today, some government, law enforcement, and civil society organisations are ‘trusted flaggers’, which platforms weigh when making enforcement decisions. This section contextualises trusted flagging and discusses government, law enforcement, and civil society organisations’ roles in content moderation. While trusted flagging has been developed since the 2016 Code of Practice, recent investigations suggest that trusted flagging relationships are tenuous, and following layoffs in the industry, being compromised (Internews, 2023). Trusted flaggers generally come from government, law enforcement, think tanks, journalism, and civil society.1 Trusted flaggers have a privileged relationship with platforms to ensure that their flagged content is reviewed by the platform. It is important to remember that these flaggers are meant to report content that likely violates local laws, human rights norms, or the platforms’ own community guidelines. In all cases, trusted flaggers’ referrals are reviewed by platforms based on whether or not the content actually violates terms of service, and platforms may, in some instances, automatically remove content that comes from a trusted flagger (Keller & Leerssen, 2020, p. 226). Trusted flaggers are an important part of the EU Code of Conduct, which stipulates that these companies will encourage the provision of notices and flagging of content that promotes incitement to violence and hateful conduct at scale by experts, particularly via partnerships with CSOs . . . and, where appropriate, to provide support and training to enable CSO partners to fulfil the role of a ‘trusted reporter’ or equivalent. (European Commission, 2016, p. 3) There are a wide range of other organisations involved in trusted flagging. Brian Fishman, former Director for Counterterrorism and Dangerous Organizations at Facebook (now Meta), notes that both YouTube and Facebook work with external teams, such as think tanks and NGOs, to develop sets of trusted flaggers that can assist in notifying platforms of extremist content (2019, p. 93). A report from Monika Bickert (Head of Global Policy Management) and Brian Fishman at Facebook identifies Flashpoint, the Middle East Media Research Institute (MEMRI), the SITE Intelligence group, and the University of Alabama at Birmingham’s Computer Forensics Research 502

Content moderation

Lab as some of the trusted flaggers that the company works with (2017). However, there is very little public information available, and while there are risks with making trusted flaggers’ identities publicly available, it is not clear whether platforms are selectively vetting and incorporating trusted flaggers and on what criteria. In light of staff layoffs at major tech platforms including Meta and Twitter (following Elon Musk’s acquisition of the latter), the extent to which platforms are ensuring the sustainability of these relationships is unclear. The EU reported in September 2019 that the companies had considerably expanded their network of ‘trusted flaggers’ since the establishment of the code, with 73 trusted flaggers at Twitter, 46 at YouTube, and 51 at Facebook (European Commission, 2019, p. 4). All three companies also organised training sessions to support these flaggers and provide them with details on the platforms’ community guidelines. A 2020 factsheet produced by the EU reports that 39 organisations sent 4,364 notifications to the IT companies, of which 1,851 were submitted through specific channels used by trusted flaggers between 4 November and 13 December 2019 (Reynders, 2020). In addition to trusted civil society, think tank, and NGO flaggers, Internet Referral Units (IRUs) are another set of trusted flaggers run by law enforcement. Europol’s IRU and the London Metropolitan Police Service’s CTIRU are the two that have been studied most extensively, though there are others in France, Belgium, and the Netherlands (Chang, 2017). IRUs are essentially bodies that actively monitor online spaces for illegal content and refer the content they find to social media platforms for review and enforcement. The EU IRU was established in July 2015 after the attack on Charlie Hebdo in Paris (Vieth, 2019, p. 329). The EU IRU’s mandate is to flag and refer terrorist and violent extremist content online to relevant partners (Chang, 2017, pp. 133–134). At the EU IRU, an analyst assesses the content that the EU IRU flags itself (Europol, 2020, p. 7), by EU Member States, or by third parties with an agreement with Europol before reporting it to the relevant social media platform (Chang, 2017, p. 135). Similarly, CTIRU, which has been active since 2010, actively searches for material using web crawlers and TweetDeck, passes on referrals from other police forces and third parties, and, unlike the EU IRU (Vieth, 2019, p. 329), also receives reports from the public (Reeve, 2019, p. 71). In 2019, the EU IRU reported referral of 25,453 instances of terrorist or violent extremist content (Europol, 2020, p. 6), while the UK government stated in 2017 that the CTIRU secured removal of 300,000 instances since February 2010 (HM Government, 2017). While platforms continue to use crowdsourced flagging mechanisms (Common, 2020, p. 128) and trusted flagging relationships (Keller & Leerssen, 2020), transparency reports show these mechanisms are outpaced by the volume of platforms’ own automated flagging. Only YouTube provides sufficiently granular data to explore the role of trusted flaggers.2 From October 2017 to March 2021, YouTube removed 116.9 million videos from its platform across the world. Of these, 101 million (86.4%) were flagged through automated tools. Just under ten million flags came from trusted flaggers, and just under six million flags came from users. In total, 156,754 videos flagged by NGOs and 555 videos flagged by governments were removed.3 It is not possible to ascertain how many of these videos were related to terrorism or violent extremism with the data provided by YouTube. While trusted flaggers are an important resource in flagging, at YouTube, the number of videos removed based on trusted flagging has decreased over time, from a peak of 2.2 million in 2018 to a steady level of around 200,000 since Q1 2020 which remains at a relatively steady level into 2023. This decrease is also apparent, for the most part, among NGOs.4 Consequently, the quantity of content from trusted flaggers is decreasing, in contrast to platforms’ own flagging processes. However, this data is not disaggregated by violation strand, so it is entirely possible that trusted flaggers’ prominence is increasing in the area of terrorism and extremism. There are a number of issues to raise in the context of trusted flagging. Groups focusing on freedom of expression have raised concerns about IRUs, noting their lack of transparency (Vieth, 503

Bharath Ganesh

2019). There is minimal information available on who trusted flaggers are and whether or not platforms have been selective in the flaggers they work with. YouTube states, Ideal candidates [for trusted flagger status] have identified expertise in at least one policy vertical [in the community guidelines], flag content frequently with a high rate of accuracy, and are open to ongoing discussion and feedback with YouTube about various content areas. (YouTube, 2021) The EU code of conduct stipulates that companies publish clear information on the conditions and criteria by which they select trusted flaggers, another area of lacking compliance (Castets-Renard, 2020, p. 319). Further, it is not clear how member states will incorporate communities harmed by hate speech and extremism in their national competent authorities or whether the EU’s selection of trusted flaggers has sufficiently included a diversity of voices; trusted flagging is a potential area for faith groups and racialised minorities to find pathways to make platforms more secure. As Brown indicates, the UK government has been rather selective in the organisations that have a voice in CVE (2019, p. 2), and it is possible that similar power dynamics are being repeated by platforms. These concerns about transparency and inclusion of a broad range of voices are all the more important given the increasing automation of content moderation, where decisions about what counts as terrorist, extremist, or radicalising content shape flagging, review, and enforcement at a massive scale.

Automated content moderation The vast majority of content reviewed and enforced is surfaced through automated detection, which involves keyword and image matching and AI classifiers (Macdonald et al., 2019). Keywords make it possible to flag posts contributed by users for containing terrorist, extremist, or hateful speech, making it possible to detect potential violations of community guidelines or the law and prevent content from being posted (Llansó et al., 2020). Similarly, it is also possible to produce digital fingerprints or ‘hashes’ of images that reduce an image to a string of characters and digits (Gorwa et al., 2020). By matching a hash from an image a user posts with hashes in a database of confirmed terrorist or extremist content, platforms can automate takedowns and filter uploads (Douek, 2020; see Llansó, 2020a). Such matching techniques do not involve artificial intelligence, and while they were useful in taking down duplicates of the Christchurch terrorist’s video or jihadist content (for example), they have significant limitations. Keyword lists require localised knowledge and expertise in specific languages to be effective in detecting extremist content. Further, the words that end up on keyword lists or image hash databases are affected by social and political pressure that affects content moderation and platform policies more broadly (Ganesh, 2020a; Gorwa et al., 2020). These limitations are based on what goes into the datasets for matching, which reproduces existing failures in countering extremism. In 2017, Microsoft, Twitter, Facebook, and YouTube set up the Global Internet Forum to Counter Terrorism (GIFCT), an inter-company institution that hosts a shared hash database of images and videos with terrorist content. As of August 2023, GIFCT has 24 members that can take advantage of GIFCT’s database to assist with automated content moderation. In its early transparency reports, GIFCT uses a taxonomy with four categories: ‘imminent credible threat’, ‘graphic violence against defenseless people’, ‘glorification of terrorist acts’, and ‘radicalization, recruitment, instruction’ (GIFCT, 2020). The institution uses the UN Security Council’s terrorism sanctions list, which essentially mandates that GIFCT’s hash database be entirely focused on jihadist entities 504

Content moderation

(Ganesh, 2021), a point noted by critical observers (Díaz & Hecht-Felella, 2021; Llansó, 2020b). While the institution brought forward recommendations to address this issue in 2021, its plans represent only incremental changes (see GIFCT, 2021, pp. 151–152). After the Christchurch attack in 2019, GIFCT set up what it refers to as a ‘Content Incident Protocol’ which the institution uses to add content to the hash database in response to attacks. GIFCT uses this for attacks pertaining to right-wing and male supremacist extremism in the taxonomy, titled ‘Christchurch, New Zealand attack and Content Incident Protocols’, that covers shootings at a synagogue in Halle, Germany, in 2019 and a male supremacist attack in Glendale, AZ (GIFCT, 2020). Currently, ‘GIFCT’s selective process of defining extremism prioritizes jihadism for the full force of pre-emptive content moderation at the same time that it exempts right-wing extremism from the power of its moderation technology’ (Ganesh, 2021). While matching can limit the circulation of extremist content (Macdonald et al., 2019), its efficacy and ability to counter radicalisation remain dependent on what is actually included in the databases. Here, industrial actors reproduce existing double standards noted for years in the counter-extremism space (Abu-Bakare, 2020; Kundnani & Hayes, 2018). Platforms also use AI for content moderation, primarily relying on supervised and unsupervised machine learning. In supervised machine learning, a classifier is trained based on data with annotations that indicate whether content violates a community guideline. Once analysts annotate a large corpus of data based on content flagged as violating a standard versus that which is not, a machine can be trained to predict whether or not new pieces of content should also be flagged, with varying levels of accuracy (Llansó et al., 2020). On the other hand, unsupervised machine learning learns patterns and associations from unlabelled data that can be clustered and categorised, which is useful to explore and categorise specific kinds of speech or types of users (Llansó et al., 2020). Classifiers of both the supervised and unsupervised variety power recommendation systems (Whittaker et al., 2021), search engine auto-fill recommendations (Noble, 2018), and other automated content distribution systems (Bucher, 2017). It is not clear what machine learning algorithms are being used to counter radicalisation online. Most of the accounts in the literature cited here are anecdotal. Gallacher (2019) discusses a tool built by ASI Data Science (now Faculty) that uses machine learning to detect jihadist content and Google Jigsaw’s Perspective API that identifies toxic speech. Gorwa et al.’s comprehensive review identifies that Facebook (in addition to an ISIS/Al-Qaeda-specific classifier) and YouTube use classifiers to enforce community guidelines on terrorism, violence, and toxic speech, while Twitter and Instagram use a quality filter on toxic speech (2020, p. 6). Macdonald et al. reference that Facebook analyses terrorist language and extremist networks on their platform and detects accounts from extremist users (2019, p. 184). Facebook (now Meta) and Google have divisions dedicated to cutting-edge machine learning with applications in and well beyond content moderation. For example, Facebook was active in developing multimodal approaches to analyse text and images together (as is necessary for memes, for example) with its Rosetta AI (Sivakumar et al., 2018). In 2020, Facebook, along with Driven Data, ran a hateful meme challenge that presented a multimodal classification problem to determine whether a meme is hateful. Facebook provided an annotated dataset they had constructed based on content on their platform (Kiela et al., 2021). The competition awarded the first prize of US$50,000 for a classifier that attained 84.5% accuracy, beating Facebook’s 64.7% baseline (Fitzpatrick, 2021). These are significant advancements in the algorithmic content moderation space and indicate that Facebook is advancing technology in this area to improve its moderation but also to lower costs. Automated content moderation presents significant threats to fundamental rights. Content moderation, often through automated means, has routinely taken down archival footage of human rights abuses (Banchik, 2021). Keyword and image filtering present serious concerns about the freedom 505

Bharath Ganesh

of expression across the world, which are exacerbated when judicial oversight, accountability, and transparency are lacking (Llansó, 2020a). In the context of the current regulatory environment surrounding content moderation against radicalisation, platforms will inevitably respond to short response windows with increased automation. The relationship trusted flaggers and IRUs have with platforms also raises questions about the potential for bias on the part of national competent authorities given the speed of automated takedown. Further, transparency and accountability of automated decisions – which are lacking, to say the least – are crucial to ensuring due process and clarity for users as well as researchers trying to understand how platforms are countering radicalisation (see Macdonald et al., 2019). However, it is important to stress that social media corporations’ enforcement makes use of a wide variety of responses and is not limited to simply taking content down or banning users. The next section briefly covers such alternative enforcement options.

Content moderation beyond take down Content moderation, as discussed so far, primarily relates to the ways in which platforms flag, review, and enforce their community guidelines. Platforms in fact have a wealth of options available to them outside taking down users or content that is already in use to counter terrorism, extremism, hate speech, and radicalisation (Jardine, 2019). Alternative enforcement options include tactics that limit terrorist, extremist, or radicalising content’s visibility on a given platform and reduce its searchability. Goldman (2021) develops a taxonomy of potential actions for content moderation, including content regulation, account regulation, visibility reductions, and monetary actions. For example, platforms add warning notes to disinformation, such as Donald Trump’s fabrications about the 2020 election (“Twitter Slaps Warning Label on Trump Tweet”, 2020). These warning notes might be found underneath a piece of content or as an ‘interstitial’ warning that requires reading a message and pressing a button before viewing it (Goldman, 2021). Platforms also take action on accounts that fall short of banning a user, such as temporarily removing their posting rights or removing credibility badges for repeated violations (Goldman, 2021, p. 38). In addition to content-level and account-level actions, platforms can reduce the visibility of users and content. For example, a shadowban is an action taken by a platform such that a user’s account appears to function normally to them, but in fact content is not visible to other users or blocked from pages that algorithmically aggregate content (Are, 2021; Goldman, 2021, p. 41; Myers West, 2018). Other visibility reduction actions include preventing content from surfacing through external or internal searches, preventing algorithmic promotion of a user’s content, and taking steps to reduce the virality of content. All of these are softer options than removing an account, but they can have an impact on reducing the possibility that audiences come across a particular user’s content (Goldman, 2021, pp. 42–47). Finally, platforms can diminish the ability of a user to earn revenues from their content (such as advertising revenue on YouTube). Having covered the regulatory context, flagging, automation, and alternative enforcement options, the final section of this chapter covers key normative issues in content moderation.

Key issues for research in content moderation and countering radicalisation With a thorough understanding of the changing regulatory environment, multistakeholder context, technology, and content moderation practices, this section moves on to review a few key issues for future research into content moderation and their impacts and possibilities in countering radicalisation. The remainder of the chapter explores debates on whether taking down extremist content is effective, the lack of transparency that impedes research, and inconsistencies in enforcement. 506

Content moderation

Deplatforming A key debate in content moderation is whether or not content should be taken down and whether users should be banned, often referred to as ‘deplatforming’. While there are many other enforcement options, scholars have begun to study whether or not deplatforming, particularly in the case of users, is worthy strategy in countering terrorism, extremism, and radicalisation. Deplatforming works in so far as it dramatically reduces audiences, disrupts extremist communities, and prevents them from benefiting from search, indexing, and algorithmic recommendation on major platforms. Three key studies make this point. From 2014 to 2017, Twitter became increasingly active in taking down pro-IS accounts active on the platform. Focusing on data from the time period between February and April 2017, Conway et al. find that deplatforming made ISIS’s presence on Twitter unsustainable. The loss of a foothold on Twitter meant that ISIS lost access to recommendations on ‘who to follow’ from the platform, the ability to use identifying images (due to moderation), and opportunities to use hashtags (Conway et al., 2019, pp. 150–151). Chandrasekharan et al. (2017), focusing on Reddit, find that banning subreddits significantly reduced hate speech behaviours on the site, undermined hateful communities, and reduced users’ use of hate speech after they migrated to other subreddits (2017, p. 17). Finally, Rogers’s (2020) study takes a cross-platform approach to explore the deplatforming of far-right users. In tracing deplatformed far-right celebrities to Telegram from other platforms, Rogers finds that the majority of these celebrities saw their audiences reduce dramatically and that their use of hateful language had mellowed (2020, p. 226). However, upon moving to Telegram, far-right celebrities remained highly active. These three studies indicate that deplatforming is effective in undermining extremist communities and thinning audiences, though it does not necessarily indicate that the activity of extremists online decreases after deplatforming. It is important to remember that this activity is on sites with much smaller audiences. On the other hand, whether deplatforming reinforces extremist identities is still an open ­question. Pearson (2018) indicates that deplatforming jihadists intensified their affective and emotional commitment to the extremist community. Similarly, Jasser et al. (2021) argue that users on Gab frequently invoke a victimhood narrative, which helps to construct a sense of community under attack from ‘Big Tech’ (2021, p. 7). However, victimhood narratives are endemic to the far right, weaponised by radical right politicians, alternative media networks, and far-right users and audiences on social media platforms and web forums (Deem, 2019; Ganesh, 2020b; Koster & Houtman, 2008). Thus, extremist communities facing deplatforming already make extensive use of victimhood narratives, but content moderation enables the construction of another alleged ‘oppressor’. There is no doubt that deplatforming disrupts extremists’ use of platforms and their digital community-building efforts. At the same time, it is clear that deplatforming risks reinforcing a sense of shared victimhood at the hands of ‘Big Tech’. However, the evidence does suggest that consistent deplatforming effectively degrades extremist communities, forces them to expend effort on returning to mainstream platforms, and reduces their audiences significantly, impeding their ability to do more than preach to the converted. In a media system driven by popularity and datafication (van Dijck & Poell, 2013), deplatforming should not be underestimated. Smaller platforms are precarious, have small audiences, have less advanced advertising and algorithmic recommendation infrastructure, and are technically vulnerable. Deplatforming works for specific goals: thinning audiences and marginalising extremist groups by relegating them to lower-quality platforms. A key objective for future research should be to examine whether or not, and under what conditions, deplatforming disrupts harmful networks and secures the dignity of those targeted 507

Bharath Ganesh

by extremists, and whether or not the threat of terrorism and extremism is actually increased as extremists migrate to smaller platforms that reinforce extremist communities.

Transparency The lack of transparency from governments that has long hampered empirical research into terrorism and extremism persists as platforms provide opaque reports and little data on advertising and activity to external researchers. Their highly curated transparency reports impede access to the data needed to understand content moderation and ensure accountability (Common, 2020; Gillespie et al., 2020, p. 24; Keller & Leerssen, 2020). There is little to no information on who trusted flaggers are or which NGOs are partnered with platforms, and as Suzor et al. (2019) argue, there is a lack of transparency in enforcement decisions provided to users and at the systemic level used by researchers. The lack of transparency limits primary sources and throttles research that could help to better estimate the impact of platform policies and new laws. This challenge is recognised in the Digital Services Act, which has specific provisions for researchers’ access to platform data and strict transparency requirements for platforms. Currently, however, investigations into content moderation activity are limited by the high-level data that is available through platforms’ transparency reporting. Creative workarounds are available in the literature. Suzor et al. (2019), for example, explore how transparency could be improved by surveying users affected by content moderation decisions as a way of providing more detail in this shrouded field. Researchers wishing to interrogate platform policies and their effect on extremist activities might benefit from working with civil society organisations active in reporting extremist activity in future research into these processes. Further, content takedown means that a large corpus of crucial historical importance is lost to researchers. While Twitter provided a specialised API with full historical data for researchers, since the company’s acquisition by Elon Musk, access to the APIs are only available at extremely high costs. Considering the extensive use of Twitter data in social media research, the shutdown of the Twitter API presents serious challenges for future research, and more importantly, demonstrates how contingent transparency is on the whims of corporations or even billionaires like Musk that can take control of them. In the context of deplatforming, as Rogers (2020) notes, and migration to alternative platforms (such as BitChute and Telegram) that are less ‘locked down’ than Twitter, YouTube, and Facebook, researchers might be able to find content removed from the mainstream platforms.

Inconsistent enforcement Platforms have been inconsistent in enforcement. It is clear that platforms have been quick to act in the case of jihadism and use a light touch in the case of hate speech and right-wing extremism (Díaz & Hecht-Felella, 2021; Ganesh, 2021; Siapera & Viejo-Otero, 2021). Part of this is due to the fact that the far right does not adhere to an identifiable organisational structure, operating more as a cultural scene, and very few right-wing groups are proscribed terrorist organisations (Conway, 2020). Yet vast amounts of right-wing extremist content evaded content moderation even on the grounds of hate speech rather than terrorism (Díaz & Hecht-Felella, 2021; Tech Transparency Project, 2020), despite research that demonstrates the extremity and hateful nature of far-right content on mainstream sites (Crosset et al., 2019; Ganesh, 2020b; Lewis, 2018). Further, the political context of these inconsistencies between far-right extremism and jihadism cannot be ignored: ‘heads of state, major political parties, some traditional media organisations 508

Content moderation

and broad swathes of Western publics’ identify with right-wing extremist content (Conway, 2020, p. 113). Facebook’s inconsistency with Britain First is a case in point. Despite numerous requests to take down Britain First’s page on Facebook – where it organised various ‘mosque invasions’ and constantly produced anti-Muslim content, higher-ups at Facebook ordered its Dublin moderators to give special protections for ‘political speech’ to both Tommy Robinson’s and Britain First’s pages (Thompson, 2019, p. 84). It was not until Britain First’s leaders were jailed that Facebook took the movement’s page down (for more examples, see Díaz & Hecht-Felella, 2021). Such inconsistency is not limited to instances of excepting specific groups from moderation; it also applies to other functional areas of platforms. A 2020 civil rights audit of Facebook found that the company continued to permit white supremacist content in advertising (Facebook’s Civil Rights Audit – Final Report, 2020). Recommendation algorithms on Facebook have, as recently as 2020, continued to serve recommendations from white supremacist groups (Tech Transparency Project, 2020). This is not limited to Facebook. In a late 2017 ‘purge’ of alt-right accounts, Twitter was highly inconsistent, with some prominent figures such as Richard Spencer staying online while others simply went to backup accounts (Ganesh, 2018). Further, employees at Facebook and YouTube raised concerns about their algorithms recommending far-right content, and these concerns were buried by Facebook (Horwitz & Seetharaman, 2020), while YouTube managers were reluctant to introduce changes that might detract from the considerable attention that this content received (Bergen, 2019). Counter-extremism policies are not consistent within companies, let alone between them. As Rogers (2020) finds, many of the users that were deplatformed from Instagram and Facebook remained on Twitter and YouTube and in fact benefitted from being deplatformed on Facebook’s platforms and not these others. Institutions like GIFCT present a welcome opportunity to coordinate these policies and develop consistency. However, as discussed earlier, there is little hope that GIFCT will enhance standards against far-right extremism without significant course correction. Generally, companies tend to wait for the identification of specific extremist groups by states or a public outcry before taking coordinated and consistent action in content moderation. By requiring companies owning very large online platforms to address “systemic risks”, in the EU the impementation Digital Services Act presents an important area for future on platform governance and counter-extremism.

Conclusion Content moderation is a key part of the fight against terrorist and extremist use of the web and social media platforms. While content moderation is largely a self-regulatory exercise, the EU is particularly active in developing standards for countering terrorist, extremist, hateful, and radicalising content and users. This policy is demonstrative of a push towards multistakeholder regulatory arrangements in which platforms retain authority over content, but governments increasingly have the ability to coerce compliance with standards on terrorist content and hate speech. The move towards trusted flagging and ‘competent authorities’ in EU legislation is a pathway for more civil society and government involvement in content moderation, though their efficacy, transparency, and inclusion remain either questionable or unclear. Further, increasing automation in content moderation stresses the importance of what counts in keyword lists and databases, and the presence of inscrutable algorithms and a lack of transparency raises serious risks for fundamental rights and accountability. As the inconsistencies in enforcement show, without significant reform, there is little hope that platforms’ light touch on the far right will do much to counter contemporary racist, xenophobic, and anti-democratic far-right movements. However, the evidence also suggests that deplatforming can play an important role in countering radicalisation online, making transparency, accountability, and consistency in content moderation ever more crucial for future research. 509

Bharath Ganesh

This mapping of actors, laws, technologies, and key issues stresses the multistakeholder nature of contemporary platform governance. This is not a guarantee of better governance outcomes. As Banchik (2021) argues, the opacity of contemporary content moderation scatters accountability as users make sense of inconsistent, unclear decisions. This is also true in the multistakeholder arrangements around content moderation to counter extremism: while platforms still retain the power to moderate content and make the decisions they wish, contemporary governance in this area scatters responsibility to flag and review across a range of institutions, stakeholders, and machines on a terrain that remains obdurately opaque to observers. Recent developments, such as the Digital Services Act, promise new avenues for regulation of online content, but major changes in the technology sector—including layoffs that have affected content moderation teams and the turn against transparency and open APIs for research—provide little hope that the key issues raised here will be fully addressed. While content moderation is tending towards governance arrangements that incorporate the private sector, government, and civil society, research exposing how content moderation decisions are made, evaluated, and enforced across a range of actors and sectors presents a crucial avenue for future understanding of counter-radicalisation online.

Notes 1 This is only an indicative list, based on what is available in the sources cited in this section, particularly the EU factsheets on the Code of Conduct. 2 Twitter and Facebook provide details on legal demands and NetzDG-specific takedowns, but these are not sufficient to be compared with the YouTube data on flagger types. 3 These figures are calculated manually by entering and tabulating data available from YouTube’s ‘Community Guidelines enforcement’ report (Google, 2021). As of July 2021, the data covers up to the period of March 2021. 4 Author’s calculation from Google (2021)

References Abu-Bakare, A. (2020). Counterterrorism and race. International Politics Reviews. https://doi.org/10.1057/ s41312-020-00074-x Archetti, C. (2019). The unbearable thinness of strategic communication. In C. Bjola & J. Pamment (Eds.), Countering online propaganda and extremism: The dark side of digital diplomacy (pp. 81–96). Routledge. Are, C. (2021). The shadowban cycle: An autoethnography of pole dancing, nudity and censorship on Instagram. Feminist Media Studies. https://doi.org/10.1080/14680777.2021.1928259 Banchik, A. V. (2021). Disappearing acts: Content moderation and emergent practices to preserve at-risk human rights – related content. New Media & Society, 23(6), 1527–1544. https://doi. org/10.1177/1461444820912724 Bergen, M. (2019, April 2). YouTube executives ignored warnings, letting toxic videos run rampant. Bloomberg. www.bloomberg.com/news/features/2019-04-02/youtube-executives-ignored-warnings-lettingtoxic-videos-run-rampant Bickert, M., & Fishman, B. (2017, November 28). Hard questions: Are we winning the war on terrorism online? About Facebook. https://about.fb.com/news/2017/11/hard-questions-are-we-winningthe-war-on-terrorism-online/ Borelli, M. (2021). Social media corporations as actors of counter-terrorism. New Media & Society. https:// doi.org/10.1177/14614448211035121 Brown, K. E. (2019). Gender, governance, and countering violent extremism (CVE) in the UK. International Journal of Law, Crime and Justice. https://doi.org/10.1016/j.ijlcj.2019.100371 Bucher, T. (2017). The algorithmic imaginary: Exploring the ordinary affects of Facebook algorithms. Information, Communication & Society, 20(1), 30–44. https://doi.org/10.1080/1369118X.2016.1154086 Castets-Renard, C. (2020). Algorithmic content moderation on social media in EU law: Illusion of perfect enforcement. University of Illinois Journal of Law, Technology & Policy, 2020(2), 283–324.

510

Content moderation Chandrasekharan, E., Pavalanathan, U., Srinivasan, A., Glynn, A., Eisenstein, J., & Gilbert, E. (2017). You can’t stay here: The efficacy of Reddit’s 2015 ban examined through hate speech. Proceedings of the ACM on Human-Computer Interaction, 1(CSCW), 31:1–31:22. https://doi.org/10.1145/3134666 Chang, B. (2017). From Internet referral units to international agreements; censorship of the Internet by the UK and EU. Columbia Human Rights Law Review, 49, 114. Citron, D. K. (2014). Hate crimes in cyberspace. Harvard University Press. Citron, D. K., & Franks, M. (2020). The Internet as a speech machine and other myths confounding section 230 reform. University of Chicago Legal Forum, 2020(1), 45–75. Common, M. F. (2020). Fear the reaper: How content moderation rules are enforced on social media. International Review of Law, Computers & Technology, 34(2), 126–152. https://doi.org/10.1080/13600869.2 020.1733762 Conway, M. (2020). Routing the extreme right. The RUSI Journal, 165(1), 108–113. https://doi.org/10.1080 /03071847.2020.1727157 Conway, M., Khawaja, M., Lakhani, S., Reffin, J., Robertson, A.,  & Weir, D. (2019). Disrupting Daesh: Measuring takedown of online terrorist material and its impacts. Studies in Conflict & Terrorism, 42(1–2), 141–160. https://doi.org/10.1080/1057610X.2018.1513984 Crawford, K., & Gillespie, T. (2016). What is a flag for? Social media reporting tools and the vocabulary of complaint. New Media & Society, 18(3), 410–428. https://doi.org/10.1177/1461444814543163 Crosset, V., Tanner, S., & Campana, A. (2019). Researching far right groups on Twitter: Methodological challenges 2.0. New Media & Society, 21(4), 939–961. https://doi.org/10.1177/1461444818817306 Deem, A. (2019). The digital traces of #whitegenocide and alt-right affective economies of transgression. International Journal of Communication, 13, 3183–3202. Díaz, Á., & Hecht-Felella, L. (2021). Double standards in social media content moderation (p. 40). Brennan Center for Justice. www.brennancenter.org/our-work/research-reports/double-standards-social-mediacontent-moderation Donovan, J. (2018). Parallel ports: Sociotechnical change from the alt-right to alt-tech. In M. Fielitz & N. Thurston (Eds.), Post-digital cultures of the far right (pp. 49–66). Transcript-Verlag. https://doi. org/10.14361/9783839446706-004 Douek, E. (2020). The rise of content cartels. Knight First Amendment Institute. https://knightcolumbia.org/ content/the-rise-of-content-cartels European Commission. (2016). Code of conduct on countering illegal hate speech online. https://ec.europa. eu/newsroom/just/document.cfm?doc_id=42985 European Commission. (2019). Assessment of the code of conduct on hate speech on line: State of play. Council of the European Union. https://ec.europa.eu/info/sites/default/files/aid_development_cooperation_fundamental_rights/assessment_of_the_code_of_conduct_on_hate_speech_on_line_-_state_of_play__0.pdf Europol. (2020). 2019 EU IRU transparency report. Europol. Facebook’s civil rights audit – final report. (2020). https://about.fb.com/wp-content/uploads/2020/07/ Civil-Rights-Audit-Final-Report.pdf Fishman, B. (2019). Crossroads: Counter-terrorism and the internet. Texas National Security Review, 2(2), 82–100. Fitzpatrick, C. (2021, January 7). Meet the winners of the hateful memes challenge. DrivenData Labs. www. drivendata.co/blog/hateful-memes-winners Gallacher, J. (2019). Automated detection of terrorist and extremist content. In B. Ganesh & J. Bright (Eds.), Extreme digital speech: Contexts, responses, and solutions. VOX-Pol. Ganesh, B. (2018). The ungovernability of digital hate culture. Journal of International Affairs, 71(2), 30–49. JSTOR. Ganesh, B. (2020a). Content moderation and censorship: Can we handle a double standard. In E. Leidig (Ed.), Mainstreaming the global radical right (pp. 280–284). Ibidem-Verlag. Ganesh, B. (2020b). Weaponizing white thymos: Flows of rage in the online audiences of the alt-right. Cultural Studies, 34(6), 892–924. https://doi.org/10.1080/09502386.2020.1714687 Ganesh, B. (2021, March 16). Platform racism: How minimizing racism privileges far right extremism. SSRC Items. https://items.ssrc.org/extremism-online/platform-racism-how-minimizing-racismprivileges-far-right-extremism/ Ganesh, B., & Bright, J. (2020). Countering extremists on social media: Challenges for strategic communication and content moderation. Policy & Internet, 12(1), 6–19. https://doi.org/10.1002/poi3.236 GIFCT. (2020). GIFCT transparency report. Global Internet Forum to Counter Terrorism.

511

Bharath Ganesh GIFCT. (2021). Broadening the GIFCT hash-sharing database taxonomy: An assessment and recommended next steps. GIFCT. Gillespie, T. (2018). Custodians of the Internet: Platforms, content moderation, and the hidden decisions that shape social media. Yale University Press. Gillespie, T., Aufderheide, P., Carmi, E., Gerrard, Y., Gorwa, R., Matamoros-Fernández, A., Roberts, S. T., Sinnreich, A., & West, S. M. (2020). Expanding the debate about content moderation: Scholarly research agendas for the coming policy debates. Internet Policy Review, 9(4). https://policyreview.info/articles/ analysis/expanding-debate-about-content-moderation-scholarly-research-agendas-coming-policy Ginosar, A. (2013). Media governance: A conceptual framework or merely a buzz word? Communication Theory, 23(4), 356–374. https://doi.org/10.1111/comt.12026 Goldman, E. (2021). Content moderation remedies [SSRN Scholarly Paper]. Social Science Research Network. https://doi.org/10.2139/ssrn.3810580 Google. (2021). YouTube community guidelines enforcement (flags). Google Transparency Report. https:// transparencyreport.google.com/youtube-policy/flags?hl=en Gorwa, R. (2019). What is platform governance? Information, Communication & Society, 22(6), 854–871. https://doi.org/10.1080/1369118X.2019.1573914 Gorwa, R., Binns, R., & Katzenbach, C. (2020). Algorithmic content moderation: Technical and political challenges in the automation of platform governance. Big Data & Society, 7(1). https://doi. org/10.1177/2053951719897945 Heldt, A. (2019). Reading between the lines and the numbers: An analysis of the first NetzDG reports. Internet Policy Review, 8(2). https://policyreview.info/articles/analysis/reading-between-lines-and-numbersanalysis-first-netzdg-reports HM Government. (2017). The government response to the fourteenth report from the home affairs select committee session 2016–17 HC 609: Hate crime: Abuse, hate and extremism online. Stationery Office. Horwitz, J.,  & Seetharaman, D. (2020, May  26). Facebook executives shut down efforts to make the site less divisive. Wall Street Journal. www.wsj.com/articles/facebook-knows-it-encourages-divisiontop-executives-nixed-solutions-11590507499 Human Rights Watch. (2021, March 25). Joint letter to EU parliament: Vote against proposed terrorist content online regulation. Human Rights Watch. www.hrw.org/news/2021/03/25/joint-letter-eu-parliamentvote-against-proposed-terrorist-content-online Internews. (2023). Safety at Stake: How to Save Meta’s Trusted Partner Program. Available: https://internews. org/wp-content/uploads/2023/07/SafetyAtStake_Internews.pdf Jardine, E. (2019). Online content moderation and the dark web: Policy responses to radicalizing hate speech and malicious content on the Darknet. First Monday. https://doi.org/10.5210/fm.v24i12.10266 Jasser, G., McSwiney, J., Pertwee, E., & Zannettou, S. (2021). “Welcome to #GabFam”: Far-right virtual community on gab. New Media & Society. https://doi.org/10.1177/14614448211024546 Keller, D. (2018). Observations on speech, danger, and money (No. 1807, Aegis Paper Series). Hoover Institution, Stanford University. www.hoover.org/research/internet-platforms-observations-speech-danger-and-money Keller, D. (2019, April 25). Regulating online terrorist content: A discussion with Stanford CIS experts about new EU proposals. Stanford Law School Blogs, Legal Aggregate. https://law.stanford.edu/2019/04/25/ regulating-online-terrorist-content-a-discussion-with-stanford-cis-directors-about-new-eu-proposals/ Keller, D. (2021, August 6). Amplification and its discontents. Knight First Amendment Institute at Columbia University. https://knightcolumbia.org/content/amplification-and-its-discontents Keller, D., & Leerssen, P. (2020). Facts and where to find them: Empirical research on Internet platforms and content moderation. In J. A. Tucker & N. Persily (Eds.), Social media and democracy: The state of the field, prospects for reform (pp. 220–251). Cambridge University Press. www.cambridge.org/core/books/ social-media-and-democracy/facts-and-where-to-find-them-empirical-research-on-internet-platformsand-content-moderation/78DE9202F2D00F2967EFC5CBDCE2CAF0 Kiela, D., Firooz, H., Mohan, A., Goswami, V., Singh, A., Ringshia, P., & Testuggine, D. (2021). The hateful memes challenge: Detecting hate speech in multimodal memes. ArXiv:2005.04790 [Cs]. http://arxiv.org/abs/2005.04790 Koster, W. D., & Houtman, D. (2008). Stormfront is like a second home to me. Information, Communication & Society, 11(8), 1155–1176. https://doi.org/10.1080/13691180802266665 Kundnani, A., & Hayes, B. (2018). The globalisation of countering violent extremism policies: Undermining human rights, instrumentalising civil society. The Transnational Institute. Latzer, M., Just, N., & Saurwein, F. (2013). Self- and co- regulation: Evidence, legitimacy and governance choice. In M. Price, S. Verhulst, & L. Morgan (Eds.), Routledge handbook of media law (pp. 373–397). Routledge.

512

Content moderation Lewis, R. (2018). Alternative influence: Broadcasting the reactionary right on YouTube. Data & Society Research Institute. https://datasociety.net/wp-content/uploads/2018/09/DS_Alternative_Influence.pdf Llansó, E. (2020a). No amount of “AI” in content moderation will solve filtering’s prior-restraint problem. Big Data & Society, 7(1). https://doi.org/10.1177/2053951720920686 Llansó, E. (2020b, July 30). Human rights NGOs in coalition letter to GIFCT. Center for Democracy and Technology. https://cdt.org/insights/human-rights-ngos-in-coalition-letter-to-gifct/ Llansó, E., van Hoboken, J., Leerssen, P., & Harambam, J. (2020). Artificial intelligence, content moderation, and freedom of expression. Transatlantic High Level Working Group, Annenberg Public Policy Center, University of Pennsylvania. Macdonald, S., Correia, S. G., & Watkin, A.-L. (2019). Regulating terrorist content on social media: Automation and the rule of law. International Journal of Law in Context, 15(2), 183–197. https://doi.org/10.1017/ S1744552319000119 Myers West, S. (2018). Censored, suspended, shadowbanned: User interpretations of content moderation on social media platforms. New Media & Society, 20(11), 4366–4383. https://doi.org/10.1177/1461444818773059 Noble, S. U. (2018). Algorithms of oppression: How search engines reinforce racism. New York University Press. Pearson, E. (2018). Online as the new frontline: Affect, gender, and ISIS-take-down on social media. Studies in Conflict & Terrorism, 41(11), 850–874. https://doi.org/10.1080/1057610X.2017.1352280 Puppis, M. (2010). Media governance: A new concept for the analysis of media policy and regulation. Communication, Culture & Critique, 3(2), 134–149. https://doi.org/10.1111/j.1753-9137.2010.01063.x Reeve, R. (2019). Human assessment and crowdsourced flagging. In B. Ganesh & J. Bright (Eds.), Extreme digital speech: Contexts, responses, and solutions. VOX-Pol. Regulation (EU) 2021/784 of the European Parliament and of the Council of April 29, 2021 on addressing the dissemination of terrorist content online, OJ L172/79. (2021). Reynders, D. (2020). 5th evaluation of the code of conduct factsheet. European Commission. https://ec.europa. eu/info/sites/default/files/codeofconduct_2020_factsheet_12.pdf Roberts, S. T. (2019). Behind the screen: Content moderation in the shadows of social media. Yale University Press. Rogers, R. (2020). Deplatforming: Following extreme Internet celebrities to Telegram and alternative social media. European Journal of Communication, 35(3), 213–229. https://doi.org/10.1177/0267323120922066 Siapera, E., & Viejo-Otero, P. (2021). Governing hate: Facebook and digital racism. Television & New Media, 22(2), 112–130. https://doi.org/10.1177/1527476420982232 Sivakumar, V., Gordo, A., & Paluri, M. (2018, September 11). Rosetta: Understanding text in images and videos with machine learning. Facebook Engineering. https://engineering.fb.com/2018/09/11/ai-research/ rosetta-understanding-text-in-images-and-videos-with-machine-learning/ Suzor, N. P., West, S. M., Quodling, A., & York, J. (2019). What do we mean when we talk about transparency? Toward meaningful transparency in commercial content moderation. International Journal of Communication, 13, 18. Tech Transparency Project. (2020). White supremacist groups are thriving on Facebook. Tech Transparency Project. www.techtransparencyproject.org/sites/default/files/Facebook-White-Supremacy-Report.pdf Thompson, P. A. (2019). Beware of geeks bearing gifts: Assessing the regulatory response to the Christchurch call. The Political Economy of Communication, 7(1), Article 1. http://polecom.org/index.php/polecom/ article/view/105 Twitter slaps warning label on Trump tweet. (2020, September 17). Reuters. www.reuters.com/article/ twitter-trump-idUKL4N2GE3KD Tworek, H., & Leerssen, P. (2019). An analysis of Germany’s NetzDG law. Transatlantic High Level Working Group, Annenberg Public Policy Center, University of Pennsylvania. van Dijck, J., & Poell, T. (2013). Understanding social media logic. Media and Communication, 1(1), 2–14. https://doi.org/10.17645/mac.v1i1.70 Vieth, K. (2019). Policing “online radicalization”: The framing of Europol’s Internet referral unit. In B. Wagner, M. Kettemann, & K. Vieth (Eds.), Research handbook on human rights and digital technology. Edward Elgar Publishing. www.elgaronline.com/view/edcoll/9781785367717/9781785367717.00025.xml Whittaker, J., Looney, S., Reed, A., & Votta, F. (2021). Recommender systems and the amplification of extremist content. Internet Policy Review, 10(2). https://policyreview.info/articles/analysis/ recommender-systems-and-amplification-extremist-content Wischmeyer, T. (2020). “What is illegal offline is also illegal online”: The German network enforcement act 2017. In B. Petkova & T. Ojanen (Eds.), Fundamental rights protection online: The future regulation of intermediaries (pp. 28–56). Edward Elgar Publishing. YouTube. (2021). YouTube trusted flagger program. YouTube Help. https://archive.is/zCBTq

513

33 RADICALISATION, COUNTER-RADICALISATION, AND GOVERNANCE IN FRAGILE STATES Oscar Gakuo Mwangi Introduction This chapter examines radicalisation, counter-radicalisation, and governance in fragile states. It pays attention to weak governance as a driver of radicalisation to violent extremism and as it affects the enactment of counter-radicalisation policy and practice. Various academic and policy studies indicate the ambiguity of the concept of state fragility and demonstrate the existence of a range of terminology that characterises the relative strength or weakness of states on a continuum from fragile states at one end to failed or collapsed states at the other. The ambiguity makes it difficult to apply such concepts to the analysis of political governance (Mcloughlin, 2012). Fragile states, in this chapter, are simply identified as those states that lack the capacity to effectively provide basic security to their populations as a result of weak governance institutions. Weak governance exacerbates security threats such as radicalisation to violent extremism. As a result, fragile states are increasingly prioritising counter-radicalisation so as to address the adverse effects of radicalisation on violent extremism. Counter-radicalisation in this chapter specifically refers to prevention and countering violent extremism (P/CVE). The P/CVE approach aims to prevent the rise of violent extremism through non-coercive methods. The chapter examines the appropriate academic and policy literature, which pays attention to the way that weak governance increases the threat of radicalisation to violent extremism and affects the implementation of counter-radicalisation policies and programmes. The chapter is divided into three sections. The first section surveys the literature regarding how weak governance makes radicalisation to violent extremism more likely. The second, citing examples from Africa and Asia, considers the way weak governance affects the development and enactment of counterradicalisation policies and programmes in fragile states. Here, the chapter draws on the concept of securitisation, derived from the Copenhagen School, to better understand how counterradicalisation plays out in fragile states. The concept explains how different aspects of life are accepted through the lens of ‘security’ (Breidlid, 2021; Buzan & Waever, 2009; Ciuta, 2009; Jackson, 2017; Pisoiu, 2012). The third section considers Kenya as a case study to illustrate the relationship between weak governance and the implementation of counter-radicalisation in fragile states.

DOI: 10.4324/9781003035848-37 514

Radicalisation, counter-radicalisation, and governance

Weak governance and radicalisation It has been put forward in the academic and policy literature on radicalisation that weak governance makes radicalisation to violent extremism more likely. The literature identifies three conditions of weak governance that can create favourable circumstances for the emergence of violent extremism. These are the exclusion and marginalisation of communities in political affairs, the inadequate provision of basic security, and external development and security interventions. This section examines the relationship between weak governance and radicalisation. The governance literature indicates that the exclusion and marginalisation of communities in political affairs in fragile states is a product of historical factors such as colonialism. Exclusion and marginalisation heighten various forms of ethnic, racial, gender, and religious intolerance conducive for radicalisation to violent extremism. Marginalisation refers to the way in which some communities in society feel less important than others due to the way they are treated or due to exclusion (El-Sayed, 2018; McCullough & Schomerus, 2017; Nasser-Eddine et al., 2011; Roberts, 2015; UNDP, 2016a). Marginalised communities feel excluded from meaningful political participation and become disillusioned by the inability of the state to provide for their basic needs. They also exploit historical and contemporary inter and intra-communal rivalries and tensions, further disrupting the existing fragile social fabric (Boukhars, 2012; Ilardo, 2020; USAID, 2020; World Bank, 2020). Under conditions of weak governance, decentralised political systems meant to mitigate historical and contemporary communal grievances and tensions further breed marginalisation and injustice conducive for radicalisation to violent extremism (Ilardo, 2020; UNDP, 2016a). This occurs when marginalisation and injustice are combined with a desire for political autonomy or self-determination. Political governance systems that aim to promote good governance through state building heighten rather than mitigate social tensions, thereby intensifying radicalisation. This is the case in Ethiopia, where ethno-linguistic federalism pursued to enhance state building has contributed to violent ethnic extremism (Adeto, 2020). A combination of weak governance and various forms of socio-political intolerance drives marginalised communities towards extremist organisations. In Central Asian states, the combination of weak governance and religious intolerance is identified as a driver of radicalisation (ICG, 2016; Ilardo, 2020). The governance literature establishes a direct relationship between the provision of basic security and radicalisation in fragile states. The literature emphasises that the lack of political will and/or administrative capacity to effectively control physical territory leads to violations of human rights and the rule of law, marginalisation and discrimination, and protracted conflict. The capacity to control territory and provide basic security is assessed in relation to the use of legitimate force (Boukhars, 2012; Nasser-Eddine et al., 2011; Osland & Erstad, 2020; Pasagic, 2020; UNDP, 2016b; USAID, 2020). The literature contends that the inability of fragile states to provide adequate security through legitimate force provides opportunities for such states to use illegitimate force. State security actors depict marginalised communities as existential threats to national security, thereby justifying the use of repressive measures (Ilardo, 2020; Osland & Erstad, 2020; Schomerus et al., 2017; United Kingdom Government Stabilisation Unit, 2018). The repression leads to collective reprisals and human rights abuses, further deteriorating levels of trust between communities and state security actors (Le Roux, 2019). Given that essential services are provided on the basis of political patronage, a combination of state repression and corruption reinforces radicalisation towards violent extremism. This provides extremist groups the opportunity to radicalise and recruit, thereby effecting changes in political leadership (Pasagic, 2020). State repression and corruption are drivers of radicalisation to violent extremism in the Sahel and Greater Horn of Africa regions (Babatunde et al., 2021; Boas, 2019; Kessels et al., 2016). 515

Oscar Gakuo Mwangi

The literature also emphasises that the inability of fragile states to control their territories leads to the creation of ungoverned spaces that facilitate radicalisation to violent extremism. Ungoverned spaces are also characterised by various forms of intrastate and interstate protracted social and political conflicts, such as ethnic conflict, rebellion, or civil war. These regions of insecurity are inhabited and used by actors who sometimes act to change their situation of insecurity and might make it more insecure by using violence (Witensburg & Zaal, 2012). Ungoverned spaces provide sanctuaries or safe havens for criminal actors and networks, terrorists and terrorism, posing a risk to national and regional security. Regions where ungoverned spaces are common include the Middle East, the Sahel, the Horn of Africa, and parts of Central Asia (Boas, 2019; Le Roux, 2019; Mwangi, 2017; Pasagic, 2020; Schomerus et al., 2017; UNDP, 2016b; Whelan, 2006). In the absence of the state, armed non-state actors step in and provide basic services. Hence armed non-state actors gain legitimacy in such spaces. This has been the case in Somalia, where violent non-state actors like the Union of Islamic Courts and AlShabaab simultaneously provide basic services and radicalise local-level communities. These violent non-state actors, for example, provide much-needed local-level security governance in the country (Mwangi, 2012, 2021a). The inability to control territories through the use of legitimate force also leads to porous borders. The governance literature asserts that porous borders facilitate the emergence of transnational security threats that are drivers of radicalisation into violent extremism. Transnational security threats emerge from migration and various types of transnational organised crime, such as illicit financial flows, drug, human, and weapon smuggling. There is a fundamental relationship between transnational security threats and violent extremism in fragile states (Babatunde et al., 2021; Cilliers, 2018; EC, 2015; McCullough & Schomerus, 2017; Osland & Erstad, 2020; Pasagic, 2020). The governance literature demonstrates that the rise of radicalisation to violent extremism is a function of exogenous factors. The factors largely emanate from the Global War on Terror. States with weak governance institutions have, since 11 September 2001, increasingly become a preoccupation for academics and policy makers including the international security and development community. These states are widely perceived as sources of violent extremism, hence threats to global security (Babatunde et al., 2021; Boas, 2017, 2019; Gelot & Hansen, 2019; Kundnani & Hayes, 2018; Nasser-Eddine et al., 2011; Pasagic, 2020). The literature identifies external security and development interventions, supposed to mitigate state fragility, as drivers of radicalisation. In the case of external development interventions, the relationship between development and governance needs in fragile states leads to what is known as the ‘fragility dilemma’ (Boas, 2019). The dilemma occurs when the unintended consequences of donor assistance, aimed at mitigating state fragility, augment authoritarian political leadership, thereby further enhancing state fragility. Political accountability is weakened because political leaders are conscious that external actors give priority to global security concerns (Boas, 2017, 2019; Osland & Erstad, 2020). The literature emphasises that fragile states take advantage of the much-needed resources availed to them by virtue of their good relations with external security and development partners. Hence, when the security and development interests of external actors are at stake, such actors tolerate and support governments that fail to comply with basic principles of good governance and development (Boas, 2019; Osland & Erstad, 2020). With regard to external security interventions, large-scale military deployment in fragile states also contributes to the rise of radicalisation to violent extremism (Boukhars, 2012; Osland & Erstad, 2020). The literature also emphasises that the deployment of external military forces to counter terrorism, radicalisation, and other forms of transnational organised crime in ungoverned spaces targets marginalised communities (Gelot & Hansen, 2019; Ilardo, 2020). The violence 516

Radicalisation, counter-radicalisation, and governance

arising from external interventions coupled with domestic state repression further contributes to radicalisation. The literature further demonstrates that while externally driven development and security interventions are intended to enhance public safety and security, they do not acknowledge the complex causalities of violence and radicalisation. As such, the interventions are drivers of radicalisation (Gelot & Hansen, 2019; Lewis, 2013). This is the case in Sahelian states such as Mali and Mauritania, where external security interventions are key drivers of radicalisation to violent extremism. These external interventions actually increase insecurity. They undermine governance systems, turning countries into zones of conflict that are then harder to govern effectively (Boukhars, 2012; Ilardo, 2020; Osland & Erstad, 2020). Other external drivers of radicalisation to violent extremism in fragile states identified in the literature include the rise of ideologies that advocate violence (Babatunde et al., 2021). The power vacuums in fragile states are at times filled by religious extremists who provide security and basic services so as to legitimise their ideological agendas (Devlin-Foltz, 2010). Hence, ideologies like Salafi Jihadism, which is driven by a series of global security factors, can survive in fragile states. In Africa and Asia, Salafi Jihadism is identified as a key driver of radicalisation to violent extremism (Babatunde et al., 2021; Devlin-Foltz, 2010; McCullough & Schomerus, 2017). The Afghan War of 1979–1989, for example, is an external factor that contributed to the rise of radicalisation to violent extremism in Somalia. The rise of Salafi Jihadism in Somalia can be attributed to persons who travelled to Afghanistan during the war and were crucial in the founding of Al-Shabaab (Mwangi, 2021a). These Somali individuals believed in violent jihad as a method of defending a Muslim country that had been attacked by infidels (Mwangi, 2012, 2021a). In essence, violent extremist groups are more likely to thrive in fragile states that prioritise regional and global security and development concerns rather than domestic ones (Babatunde et al., 2021; Gelot & Hansen, 2019; McCullough & Schomerus, 2017). Weak governance is likely to aggravate radicalisation to violent extremism in fragile states.

Weak governance, securitisation, and counter-radicalisation Weak governance also affects counter-radicalisation both in terms of the way policies and programmes are developed and in terms of how they are put into practice. This section surveys the literature, which pays attention to how weak governance affects the implementation of counterradicalisation in fragile states. In order to address security threats such as radicalisation to violent extremism and terrorism, fragile states are adopting whole-of-society approaches to counterradicalisation (OSCE, 2020). The whole-of-society approach creates prospects for fruitful engagement between state and non-state actors, thereby widening ownership and monitoring of counterradicalisation. It also considers local-level drivers of radicalisation to violent extremism (IIJ, 2021; OECD, 2006; OSCE, 2020). Hence, counter-radicalisation strategies are an integral component of national counterterrorism policy (Hardy, 2018; Rothermel, 2020). As already noted, counterradicalisation in this chapter refers to P/CVE. The governance literature identifies two significant factors that impact the implementation of counter-radicalisation in fragile states. The first is the discrepancy or lack of compatibility between global security and development priorities and national counter-radicalisation objectives. The second factor pays attention to the impact of counter-radicalisation on social categories, human rights, and the rule of law. Fragile states lack the institutional capacity and levels of trust between key state and nonstate actors required for meaningful policy implementation. This undermines their ability to effectively implement policies and programmes (Dews, 2021; IIJ, 2021; Mcloughlin, 2012; OECD, 2020; OSCE, 2020). A review of the governance literature indicates that the implementation of 517

Oscar Gakuo Mwangi

counter-radicalisation in fragile states is undermined by discrepancies between global security and development priorities and national counter-radicalisation objectives (Kundnani & Hayes, 2018; Jones, 2020). In order to promote good governance, fragile states adopt counter-radicalisation interventions, which aim to partly supplement the objectives of external development assistance and security provision. However, in fragile states, global actors, such as foreign states and international organisations, prioritise interventions that enhance traditional security and stability, whereas domestic non-state actors prefer interventions that pay attention to human security. States with weak governance therefore find themselves in a dilemma given that they too prioritise traditional security (Jones, 2020; Jones et al., 2021; Rrustemi, 2020). The dilemma provides opportunities for fragile states to securitise both counter-radicalisation practitioners and beneficiaries (Abu-Nimer, 2018a; El-Sayed, 2018; Boas, 2017, 2019; Kundnani & Hayes, 2018; McCullough & Schomerus, 2017; OSCE, 2020; Osland & Erstad, 2020; Thiessen, 2022). The governance literature points out that non-state actors, involved in counter-radicalisation programmes, in fragile states stress that external development actors’ priorities are given preference over the needs and demands of local communities (Abu-Nimer, 2018a). Beneficiaries also express concerns about the global security and development priorities of Western states taking precedence over local concerns (Abu-Nimer, 2018b). Non-state actors who implement counterradicalisation programmes thus find themselves in a moral dilemma given that they are part of the securitisation. Hence, the legitimacy and credibility of non-state actors are eroded when they participate in counter-radicalisation programmes (Abu-Nimer, 2018a; El-Sayed, 2018). From a development perspective, the proliferation of counter-radicalisation discourse has augmented development inequalities as donors prioritise fragile states that are highly prone to the risk of violent extremism. This is the case in the Sahelian states of Africa such as Mali and Niger (Boas, 2017, 2019; Briscoe  & Van Ginkel, 2013; McCullough & Schomerus, 2017; Osland & Erstad, 2020; Van Zyl, 2019). These practical and ethical concerns of counter-radicalisation practitioners and beneficiaries are evident in Africa. The literature indicates that in countries of Central and West Africa, such as Burkina Faso, Cameroon, Chad, Niger, and Nigeria, beneficiaries of counter-radicalisation programmes have also accused donor organisations of micromanaging local implementing organisations and of being insensitive to the confidentiality and privacy needs of local communities (Abu-Nimer, 2018a; Van Zyl & Frank, 2018). In Chad, for example, the role and credibility of nonstate actors in implementing the United States for International Development (USAID) counterradicalisation programmes have been questioned by beneficiaries. Beneficiaries point out that the programmes focus on hard rather than soft approaches when countering Al-Qaeda and Boko Haram risks in the Sahel region (Abu-Nimer, 2018a). The variance between external global priorities and domestic counter-radicalisation objectives is also evident in the Central Asia region. Despite stabilisation efforts aimed at promoting good governance, states in the region continue to face the challenges of counter-radicalisation (ICG, 2016). The governance literature illustrates that establishing administrative measures without institutional safeguards in counter-radicalisation is counter-productive because it securitises social categories such as ethnic, gender, racial, and religious groups, hence impacting human rights and the rule of law (Thiessen, 2022; Thiessen & Özerdem, 2019). Notably is the way counter-radicalisation is often used as a conceptual tool to legitimise and rationalise the war on terror (Stephens et al., 2021). States with weak governance often use the vague terminology of radicalisation and violent extremism to securitise political opposition, civil society organisations (CSOs), and ethnoreligious communities in order to contain political dissent through repressive counterterrorism measures. Repressive security practices that threaten human security in fragile states motivate 518

Radicalisation, counter-radicalisation, and governance

extremism and violence (Bak et al., 2019; Hardy, 2018; Jones et al., 2021; Mwangi, 2021b; Stephens & Sieckelinck, 2020; Stephens et al., 2021; UN, 2015). In the Philippines, for example, counter-radicalisation augments the securitisation of development, rights, and peace efforts. Counter-radicalisation measures obscure the state’s repression of individuals and organisations perceived to be sympathetic to political opposition. Through counter-radicalisation, the state has legitimised its pursuit of partisan political priorities (Arugay et al., 2021). Likewise, as a result of weak governance, the critical role of non-state actors in counterradicalisation is often hampered by the state through security laws related to counterterrorism (Bryden & Chappuis, 2016; Dews, 2021; OSCE, 2020; RAN, 2021). The literature underlines that counter-radicalisation not only targets non-state actors and local-level communities in order to contain political dissent but also to use them merely as sources of security intelligence information (OSCE, 2020). Although similar criticisms have been made of counter-radicalisation programmes in Western Europe, the practice is more prevalent in fragile states due to the lack of accountability and transparency regarding the way the information is gathered, why it is gathered, and how it is shared and used by state security actors. The lack of effective governance institutions that hold state security actors accountable for their actions exacerbates threats to counter-radicalisation practitioners and beneficiaries (OSCE, 2020). The securitisation of counter-radicalisation practitioners occurs in Central Asia, the Middle East, and North Africa. State supported counter-radicalisation CSOs prioritise campaigns based on traditional security approaches rather than a human security framework (Abu-Nimer, 2018a, 2018b; OSCE, 2020). In Kyrgyzstan, counter-radicalisation efforts focus on local-level resilience. However, counter-radicalisation strategies adopt a top-down approach. The approach promotes the partisan agendas of state and non-state actors rather than local-level communities (ICG, 2016; OSCE, 2020; Thiessen, 2022). Gender issues are also discussed in the governance literature. The literature is emphatic that the role of women in counter-radicalisation has also been weakened in fragile states. Due to weak governance institutions, women are largely excluded not only from the formulation and implementation of traditional security policies and strategies but also from counter-radicalisation (Ali & Mwamburi, 2021; Atela et al., 2021; Mesok, 2019; Rothermel, 2020). The literature indicates that counter-radicalisation policies and programmes in fragile states are often designed by a traditionally male-dominated state security sector, thereby exuding gender-specific issues (GAPS, 2018; Henty & Eggleston, 2018; Idris, 2020; LSE WB, 2017). The implementation of counter-radicalisation approaches can be counterproductive by exacerbating the risks women human rights defenders face (GAPS, 2018; Idris, 2020). Counter-radicalisation programmes do not reflect and account for the specific local-level threats that women confront (Mesok, 2019). Non-state actors that articulate gender issues emphasise the inclusion of a gender-perspective counter-radicalisation framework. They argue that this will desecuritise and enhance trust between state and non-state actors involved in countering radicalisation. It will also lead to the implementation of gender-sensitive counter-radicalisation to violent extremism (Mesok, 2019, 2022). The literature emphasises that the inclusion of women’s issues can create a non-securitised space and can be a foundational platform for creating and implementing effective counter-radicalisation strategies. Countries where the role of women in counterradicalisation has been undermined include Afghanistan, Bangladesh, Mozambique, Nigeria, and Tajikistan (Atela et al., 2021; GAPS, 2018; Henty & Eggleston, 2018; Mesok, 2019).

Radicalisation and counter-radicalisation in Kenya Kenya is used as a case study to illustrate the key arguments raised in the preceding sections, which focus on the conflict between donor priorities and domestic counter-radicalisation objectives, 519

Oscar Gakuo Mwangi

human rights, and the rule of law. Kenya is classified as a fragile state on the basis of several economic, political, and social indicators that undermine its governance institutions (FFP, 2021; OECD, 2020). Several studies explain that radicalisation to violent extremism in Kenya is motivated by several complex historical, political, economic, cultural, social, and psychological factors (Badurdeen, 2021; Chome, 2021; Mohamed, 2021; Papale, 2022; Rink & Sharma, 2018; Sahgal & Zeuthen, 2020; Wario, 2021). Radicalisation to violent extremism is prevalent in Kenya’s ungoverned spaces of the northeastern and coastal regions (Badurdeen, 2021; Wario, 2021). The north-eastern and coastal regions cover approximately 36% and contain slightly over 14% of Kenya’s territory and population, respectively. About 79% of Kenya’s 11.2% Muslims inhabit the regions. The regions, which border Somalia, experience high levels of poverty, unemployment, conflict, and violence (Republic of Kenya, 2019a, 2019b; Mwangi, 2017). The structural conditions of the north-eastern and coastal regions are a product of the country’s colonial and postcolonial policies. In particular, the marginalisation of ethno-religious communities enhances radicalisation to violent extremism (Badurdeen, 2021; Wario, 2021). The Somalia-based Al-Shabaab terror group is the most significant violent extremist group that operates in these regions. Al-Shabaab adopts Salafi-Jihadism as a political ideology to radicalise and recruit members so as to create an Islamic Caliphate in the Greater Horn of Africa (Botha, 2014; Bryden & Bahra, 2019; Mwangi, 2012; Hansen et al., 2019; Speckhard & Shajkovci, 2019). Over 90% of Al-Shabaab attacks in Kenya occur in the northeastern and coastal regions (GTD, 2022) Kenya’s counterterrorism measures are a political and security response to the frequency and intensity of Al-Shabaab operations in the country (Mwangi, 2021a). Counter terrorism hard approaches are conducted by the National Police Service (NPS), the National Intelligence Service (NIS) and the Kenya Defence Forces (KDF) (US, 2021). These hard approaches have been consistently criticised for violating human rights and augmenting violent extremism in the country. Hence, the state formulated a soft approach to address radicalisation to violent extremism and launched Kenya’s National Strategy to Counter Violent Extremism (NSCVE) in September 2016. The Strategy adopts a whole-of-society approach to counter-radicalisation (Crisman et al., 2020; EU, 2018; NCTC, 2022; Sharamo & Mohammed, 2020; Ogada, 2017; USAID, 2020). The NSCVE is the first approach to acknowledge radicalisation as a process and facilitate multisectoral holistic programmatic interventions to address the local-level underlying complex causes of violent extremism. It is decentralised through county action plans (CAPS). Emphasis is placed on local-level sound planning and effective implementation of counter-radicalisation (Crisman et al., 2020; Crosby & Pkalya, 2021; Republic of Kenya, 2016; Sharamo & Mohammed, 2020). Kenya’s National Counter Terrorism Centre (NCTC), a multi-agency state security sector body, is the lead agency responsible for coordinating state, non-state, bilateral, and multilateral actors involved in the implementation of counter-radicalisation. It also approves and monitors all locallevel, national, and international non-state actors engaged in counter-radicalisation efforts. All actors involved in counter-radicalisation are required to collaborate with state security agencies (Republic of Kenya, 2014, 2016). The predominant role of Kenya’s security agencies in the management of counter-radicalisation evolves from the nature of the country’s security governance. Despite the implementation of Kenya’s political devolution in 2010, and hence decentralisation of several state functions, security governance remains centralised under the national government (Crisman et al., 2020; Republic of Kenya, 2010). The NSCVE states that counter-radicalisation in Kenya will adopt a soft approach. However, at both the national and county levels, counter-radicalisation is managed by state actors with traditional notions of security. This raises fundamental issues of accountability and 520

Radicalisation, counter-radicalisation, and governance

transparency in the implementation process (Jones, 2020; Republic of Kenya, 2016; Musau, 2018; Ogada, 2017). At the national level, the political objective of the NCTC is to ensure that the security sector can closely monitor all state and non-state actors involved in counter-radicalisation initiatives (Crisman et al., 2020; Republic of Kenya, 2014; Ogada, 2017). At the local level, County Security and Intelligence Committees (CSIC) coordinate action with local-level non-state actors to ensure counter-radicalisation efforts conform to national security requirements. The CSICs handle all matters relating to security functions (Republic of Kenya, 2012, 2016). The pitfalls of implementing counter-radicalisation, under Kenya’s political devolution are reflected in the recurrent conflicts between the national and county governments regarding security matters. Counties have non-traditional security devolved responsibilities, which are included as work pillars in counter-radicalisation (Crosby & Pkalya, 2021). However, due to the centralisation of security governance, county governments feel alienated as they do not perceive themselves as part and parcel of counter-radicalisation. Hence, county governments are reluctant to incorporate the CAPs into other county integrated development plans essential for the implementation of counter-radicalisation (Crisman et al., 2020; Crosby & Pkalya, 2021; Jones, 2020; Mesok, 2019). Some counties in Kenya have successfully secured political support for their counter-radicalisation plans. The counties have assured state security actors that although traditional security is the responsibility of the national government, the responsibility of addressing violent extremism ideally belongs to county governments. (Crisman et al., 2020; Crosby & Pkalya, 2021; Ogada, 2017) There is a discrepancy between global priorities and Kenya’s counter-radicalisation objectives. County-based non-state actors refer to the counter-radicalisation agenda, supported by external development partners and enacted by the state, as irresponsive to the local-level basic need of communities. This emanates from the failure of the national action plan on counter-radicalisation to clearly specify the degree to which national and county governments are involved in implementation. This undermines the roles counter-radicalisation assigns to the counties. Counter-radicalisation is perceived as a top-down rather than a bottom-up approach that generates community-based solutions (Cleveland et al., 2020; Mesok, 2019). The implementation process is often criticised for prioritising international recognition and donor funding for counter-radicalisation instead of establishing meaningful consultative processes with local-level non-state actors (Crisman et al., 2020). The predominant role of state security actors in the implementation process adversely affects the role of counter-radicalisation practitioners in the country. Counter-radicalisation in Kenya complies with international law and policies requiring adherence to international human rights and humanitarian law. However, the state has not yet reoriented its security and security-related legislation and practices to adhere to counter-radicalisation (Republic of Kenya, 2016; Ogada, 2017; Freedom House, 2018; Mesok, 2019, 2022). These laws include the Proceeds of Crime and Anti-Money Laundering Act 2009, the Prevention of Organised Crime Act 2010, the Prevention of Terrorism Act 2012, and the Security Laws (Amendment) Act 2014. Counter-radicalisation efforts of non-state actors are curtailed by such laws, hence constraining their capacity to independently design and implement effective violence prevention programming (Mesok, 2019, 2022). The Prevention of Terrorism Act 2012, for example, was amended in July 2019 to require all non-state actors involved in counter-radicalisation to register and request approval for their programmes from the NCTC. The amendment was signed into law without extensive parliamentary debate. Under the new amendment, oversight of state agencies pertaining to counter-radicalisation is also further amalgamated under the NCTC (Crisman et al., 2020). The objective of collecting information on non-state actors and their activities has further centralised the role of the security agency in countering radicalisation (Crisman et al., 2020; Mesok, 2019, 2022). The amendment has decreased mutual trust between state security actors and non-state actors. Non-state actors 521

Oscar Gakuo Mwangi

emphasise that the registration requirement interferes with their ability to work in local-level communities that do not trust the national government (Crisman et al., 2020; Ramadhan & Ouma, 2020). The non-state actors emphasise that state monitoring of counter-radicalisation programmes will render such programmes counterproductive. The actors also emphasise that NCTC supervision amounts to securitising their counter-radicalisation efforts (Crisman et al., 2020; Ramadhan & Ouma, 2020). The deliberate use of non-state actors as an extension of the state’s intelligence gathering and surveillance has weakened counter-radicalisation efforts in Kenya (Jones et al., 2021; Ramadhan & Ouma, 2020; REINVENT Programme, 2021). The lack of essential information about counter-radicalisation risks and financing impacts counterradicalisation efforts in Kenya negatively (Ogada, 2017; Crisman et al., 2020). Non-state actors lack the basic security information required to monitor counter-radicalisation risks. Hence, tensions between state and non-state actors operating in a security-focused domain can exacerbate when the latter are not knowledgeable of security sector terminology or decision-making behaviours (Crisman et al., 2020; Freear & Glazzrard, 2020). Due to the lack of such basic information and knowledge regarding the sensitive nature of counter-radicalisation programming, local-level participants have been placed in harm’s way when non-state actors mishandle their personal data. Some counter-radicalisation programming has also been counterproductive by abetting Al-Shabaab messaging (Crisman et al., 2020). The lack of information on counter-radicalisation financing also raises accountability and transparency issues. The Ministry of Interior, whose portfolio includes internal security, is responsible for allocating money spent on counter-radicalisation. The NCTC oversees counter-radicalisation but does not control funding (Ogada, 2017). Like other security agencies, the NCTC has no publicly available budgets. All state security actors responsible for the implementation of counter-radicalisation are therefore not financially accountable for their actions (Ogada, 2017; Downie, 2019; Crisman et al., 2020). Counter-radicalisation in Kenya remains underfunded and heavily reliant on donor funding. This has impacted the ability of counter-radicalisation to be sustainable and operational. There has been yet no legislation enacted to provide for counterradicalisation funding. Counties, as integral components of the implementation process, are still expected to foot the bill for counter-radicalisation initiatives. This is despite the state commitment to support counter-radicalisation implementation (USAID, 2020; Crosby & Pkalya, 2021). The literature on Kenya’s counter-radicalisation efforts also pays attention to human rights and the rule of law. The literature emphasises that the implementation process negatively affects ethnic, gender, racial, and religious groups, hence impacting human rights and the rule of law (Freedom House, 2018; Musau, 2018; Mwangi, 2021b; Ogada, 2017; Papale, 2022). The counterradicalisation agenda depicts religion, in particular Islam, as a potential existential threat to national security, hence heightening existing fears that the war on terror is essentially a war on Islam (Mesok, 2019). The securitisation disproportionately targets non-state actors and ethno-religious minorities, particularly Muslims of Somali and Arab origin involved in counter-radicalisation. Non-state actors and local-level ethno-religious minorities are depicted as sympathisers or supporters of Al-Shabaab (Ali & Mwamburi, 2021; Kumah-Abiwu, 2022; Musau, 2018; Mwangi, 2019, 2021b; Mwangi & Mwangi, 2019; Papale, 2022). This hinders such local-level non-state actors from being effective partners in counter-radicalisation work (Hansen et al., 2019; Mesok, 2019; Ogada, 2017). Similarly, women in Kenya have largely been excluded from the formulation and implementation of counter-radicalisation policies (Mesok, 2019; Idris, 2020; Ndungú et al., 2017). Women community leaders in Kenya argue that the language of counter-radicalisation securitises gender and religion, particularly Islam. This has a negative impact on the rights of Muslim women in the country. Women community leaders emphasise that security policies must pay attention to the local-level threats that women confront (Idris, 2020; Mesok, 2019, 2022). 522

Radicalisation, counter-radicalisation, and governance

Conclusion This chapter draws on the concept of securitisation to better understand how counter-radicalisation plays out in fragile states. The chapter examines the governance literature on radicalisation, which pays attention to how weak governance increases the threat of radicalisation to violent extremism and affects the implementation of counter-radicalisation policies and programmes in such states. An examination of the challenges of implementing counter-radicalisation in fragile states contributes to a better understanding of radicalisation from a governance perspective. The literature demonstrates that weak governance exacerbates radicalisation to violent extremism in fragile states. This emanates from three key characteristics of fragile states: the exclusion and marginalisation of communities in political affairs, inadequate provision of basic security, and external development and security interventions. Exclusion and marginalisation enhance community grievances, which are conducive for radicalisation to violent extremism. The inability to provide adequate basic security leads to violation of human rights, the rule of law, and protracted conflict, thereby providing opportunities that are also conducive for radicalisation to violent extremism. External development and security interventions meant to mitigate state fragility further enhance fragility and violent extremism. The literature that examines the relationship between weak governance, securitisation, and radicalisation in fragile states pays attention to the discrepancy between global security and development priorities and national counter-radicalisation objectives. It also examines the impact of counter-radicalisation on social categories, human rights, and the rule of law in such states. The literature points out that the discrepancy between global priorities and local-level objectives in fragile states is to a large extent caused by weak institutional capacity and low levels of trust between key state and non-state actors required for meaningful counter-radicalisation policy implementation. The literature also points out that a lack of accountability and transparency in counter-radicalisation programmes in fragile states creates favourable circumstances for such states to securitise counter-radicalisation practitioners and beneficiaries. This adversely affects human security. Kenya is used as a case study to illustrate the key arguments raised in the chapter, which focus on the conflict between global priorities and domestic counter-radicalisation objectives, human rights, and the rule of law.

References Abu-Nimer, M. (2018a). Alternative approaches to transforming violent extremism: The case of Islamic peace and interreligious peacebuilding. In B. Austin & H. J. Giessman (Eds.), Transformative approaches to violent extremism (Berghof Handbook Dialogue Series No. 13). Berghof Foundation. Abu-Nimer, M. (2018b). De-Islamisation, de-securitisation and de-religionisation. In B. Austin & H. J. Giessman (Eds.), Transformative approaches to violent extremism (Berghof Handbook Dialogue Series No. 13). Berghof Foundation. Adeto, Y. A. (2020). Violent ethnic extremism in Ethiopia: Implications for the stability of the horn of Africa. African Journal on Conflict Resolution, 20(2), 10–36. Ali, F., & Mwamburi, D. (2021). Black Muslim women in security studies. Critical Studies in Security, 9(1), 31–35. https://doi.org/10.1080/21624887.2021.1904193 Arugay, A.A., Batac, M., & Street, J. (2021). An explosive cocktail-Counter-terrorism, militarization and authoritarianism in the Philippines. Saferworld. Atela, M., Ojebode, A., Makokha, R., Otieno, M., & Aina, T. (2021). Women organizing in fragility and conflict: Lessons from the #BringBackOurGirls movement, Nigeria. Gender & Development, 29(2–3), 313–334. https://doi.org/10.1080/13552074.2021.1979323 Babatunde, A., Adedimeji, M., Raji, S., Maweu, J., & Githigaro, J. (2021). Managing violent religious extremism in fragile states: Building institutional capacity in Nigeria and Kenya. Routledge.

523

Oscar Gakuo Mwangi Badurdeen, F. A. (2021, June). Study to examine the influence of contemporary Islamic ideologies in Kenya: Lamu and Tana River counties. REINVENT Programme. Bak, M., Tarp, K. N., & Liang, C. C. (2019, August 19). Defining the concept of “violent extremism”: Delineating the attributes and phenomenon of violent extremism (Geneva Paper 24). Centre for Security Policy. Boas, M. (2017). Fragile states as the new development agenda? Forum for Development Studies, 44(1), 149–154. https://doi.org/10.1080/08039410.2017.1264737 Boas, M. (2019). The Sahel crisis and the need for international support (Policy Dialogue No. 15). The Nordic Africa Institute Uppsala. Botha, A. (2014). Political socialization and terrorist radicalization among individuals who joined alShabaab in Kenya. Studies in Conflict  & Terrorism, 37(11), 895–919. https://doi.org/10.1080/10576 10X.2014.952511 Boukhars, A. (2012, April). The drivers of insecurity in Mauritania, Middle East (The Carnegie Papers). Carnegie Endowment for International Peace. Breidlid, T. (2021). Countering or contributing to radicalisation and violent extremism in Kenya? A critical case study. Critical Studies on Terrorism, 14(2), 225–246. https://doi.org/10.1080/17539153.2021.1902613 Briscoe, I.,  & Van Ginkel, B. (2013, March). The Nexus between development and security: Searching for common ground in countering terrorism (Policy Brief). International Centre for Counter-Terrorism (ICCT). Bryden, A., & Chappuis, F. (2016, April 26–27). Security sector governance and reform in Africa. Background Paper Developed for the Learning Lab on Security Sector Governance and Reform in Africa Coorganised by ASSN and DCAF in Partnership with OSF, Dakar, Senegal. Bryden, M., & Bahra, P. (2019). East Africa’s terrorist triple helix: The Dusit hotel attack and the historical evolution of the Jihadist threat. CTC Sentinel, 12(6), 1–11. Buzan, B., & Waever, O. (2009). Macrosecuritisation and security constellations: Reconsidering scale in securitisation theory. Review of International Studies, 35(2), 253–276. https://doi.org/10.1017/ S0260210511000726 Centre for Women, Peace and Security (LSE WB). (2017, August). Preventing/countering violent extremism and WPS: Concepts, practices and moving forward (Key Issues Paper). LSE WB. Chome, N. (2021, June). Kenya’s new violent extremism hotspots: As assessment of risk and vulnerability in Marsabit, the Rift Valley and Western Kenya. REINVENT Programme. Cilliers, J. (2018, August). Violence in Africa: Trends, drivers and prospects to 2023 (Institute for Security Studies (ISS) Africa Report 12). ISS. Ciuta, F. (2009). Security and the problem of context: A hermeneutical critique of securitisation theory. Review of International Studies, 35(2), 301–326. https://doi.org/10.1017/S0260210509008535. Cleveland, E., Glaeser, S., Jorgensen, L., Lewis, L. L., & Shropshire, A. (2020). Preventing and countering violent extremism: Localization in Kenya, Kosovo, and the Philippines, Robert M. La Follette school of public affairs. University of Wisconsin. Crisman, B., Yoon, C. S. A., Goos, C., Hull, D., Romano, E., Johnson, J., . . . Gungadurdoss, Y. (2020, January). Preventing violent extremism: Lessons from Kenya. Princeton University, Woodrow Wilson School of Public and International Affairs. Crosby, P., & Pkalya, D. (2021). Developing and implementing local action plans for preventing and countering violent extremism in East Africa: A strong cities toolkit with lessons from Kenya. Institute for Strategic Dialogue. Devlin-Foltz, Z. (2010, August). Africa’s fragile states: Empowering extremists, exporting terrorism. Africa security brief (No. 6). A Publication of the Africa Center for Strategic Studies. Dews, D. (2021, July). A “whole of society” approach? Exploring civil society inclusion in national frameworks to prevent and countering violent extremism (Policy Brief). Global Center on Cooperative Security. Downie, R. (2019). Kenya: Cooperation, co-optation and confrontation. In E. Malka (Ed.), Faith in the balance: Regulating religious affairs in Africa (pp. 80–111). Rowman & Littlefield. El-Sayed, R. (2018). Religious institutions and the challenges of extremism and terrorism. In B. Austin & H. J. Giessman (Eds.), Transformative approaches to violent extremism (Berghof Handbook Dialogue Series No. 13). Berghof Foundation. European Commission (EC). (2015). Operational guidelines on the preparation and implementation of EU financed actions specific to countering terrorism and violent extremism in third countries. European Commission. European Union (EU). (2018). Kenya-EU partnership for the implementation of the Kenya CVE strategy (TO5-EUTF-HOA-KE 34). EU.

524

Radicalisation, counter-radicalisation, and governance Freear, M., & Glazzrard, A. (2020). Prevention communication: Emerging lessons from participative approaches to countering violent extremism in Kenya. The RUSI Journal, 165(1), 96–106. https://doi.org/ 10.1080/03071847.2020.1734316 Freedom House. (2018, November). Kenya’s antiterrorism strategy should prioritize human rights, rule of law (Policy Brief). Freedom House. Fund For Peace (FFP). (2021). Fragile states index annual report 2021. The Fund for Peace. Gelot, L., & Hansen, S. J. (2019). They are from within us: CVE brokerage in south-central Somalia. Conflict, Security & Development. 19(6), 563–582. https://doi.org/10.1080/14678802.2019.1688961 Gender Action for Peace and Security (GAPS). (2018). Prioritise peace: Challenging approaches to preventing and countering violent extremism from a women, peace and security perspective. GAPS. Global Terrorism Data Base (GTD). (2022). Kenya. www.start.umd.edu/gtd Hansen, J. S., Lid, S., & Okwany, C. C. O. (2019, April). Countering violent extremism in Somalia and Kenya: Actors and approaches (Working Paper 16). Norwegian Institute for Urban and Regional Research. Hardy, K. (2018). Comparing themes of radicalisation with countering violent extremism policy. Journal for Deradicalization, 15, 76–110. Henty, P., & Eggleston, B. (2018). Mothers, mercenaries and mediators: Women providing answers to the questions we forgot to ask. Security Challenges, 14(2), 105–121. Idris, I. (2020). Gender and countering violent extremism (CVE) in the Kenya-Mozambique region. K4D Helpdesk Report. Ilardo, M. (2020). Violent extremism in the Sahel countering vulnerability to radical narratives: Towards a more pragmatic approach (Fokus 1). Austrian Institute for European and Security Policy (AEIS). International Crisis Group (ICG). (2016, October 3). Kyrgyzstan; state fragility and radicalisation (No. 83). Europe and Central Asia Briefing. International Institute for Justice and the Rule of Law (IIJ). (2021). Developing multi-actor P/CVE intervention programmes-implementing a whole of society approach, “do no harm” approach: IIJ training curriculum facilitator’s guide. https://www.theiij.org/wp-content/uploads/2021/11/1.-Facilitators-Guide-1.pdf Jackson, R. (2017). CTS, counter terrorism and non-violence. Critical Studies on Terrorism, 10(2), 357–369. Jones, M. (2020). A template for global South? Understanding promises and pitfall of preventing/countering violent extremism in Kenya. Emerging Insights, Royal United Services Institute for Defence and Security Studies. Jones, M., Wallner, C., & Winterbotham, E. (2021, September). Lessons for preventing and countering violent extremism (RUSI Occasional Paper). Royal United Services Institute for Defence and Security Studies. Kessels, E., Durner, T., & Schwartz, M. (2016, April). Violent extremism and instability in the greater horn of Africa: An examination of drivers and responses. Global Center on Cooperative Security. Kumah-Abiwu, F. (2022). Africa’s security landscape of security-development and human-rights issues. African Security Review, 31(2), 99–114. https://doi.org/10.108/102446029.2021.1980412 Kundnani, A., & Hayes, B. (2018). The globalisation of countering violent extremism policies: Undermining human rights, instrumentalising civil society. Transnational Institute. Le Roux, P. (2019, December). Responding to the rise in violent extremism in the Sahel (Africa Security Brief No. 36). https://africacenter.org/publication/responding-rise-violent-extremism-sahel/ Lewis, A. (2013). Violence in Yemen: Thinking about violence in fragile states beyond the confines of conflict and terrorism. Stability: International Journal of Security & Development, 2(11), 13, 1–22. https://doi. org/10.5334/sta.az McCullough, A., & Schomerus, M. (2017). What do we know about drivers of radicalization and violent extremism, globally and in Niger? Literature review (ODI Report). Overseas Development Institute. Mcloughlin, C. (2012). Topic guide on fragile states, governance and social development resource centre. University of Birmingham. Mesok, E. (2019). Women, peace and security and the prevention of violence: Reflections from civil society in the context of the fourth Swiss national action plan 1325. Swisspeace. Mesok, E. (2022). Counterinsurgency, community participation, and the preventing and countering violent extremism in Kenya. Small Wars & Insurgencies, 33(4–5), 720–741. https://doi.org/10.1080/09592318.2 022.2037908 Mohamed, M. H. (2021). Dangerous or political? Kenyan youth negotiating political agency in the age of “new terrorisms”. Media, War & Conflict, 14(3), 303–321. https://doi.org/101177/17506352211028406 Musau, S. (2018, January 21). Combating terrorism and upholding human rights in Kenya (APCOF Research Paper). APCOF.

525

Oscar Gakuo Mwangi Mwangi, O. G. (2012). State collapse, Al-Shabaab, Islamism and legitimacy in Somalia. Politics, Religion & Ideology, 13(4), 513–527. https://doi.org/10.1080/21567689.2012.725659 Mwangi, O. G. (2017). Statelessness, ungoverned spaces and security in Kenya. In T. Bloom, K. Tonkiss, & P. Cole (Eds.), Understanding statelessness (pp. 117–133). Routledge. Mwangi, O. G. (2019). The “Somalinisation” of terrorism and counterterrorism in Kenya: The case of refoulement. Critical Studies on Terrorism, 12(2), 298–316. https://doi.org/10.1080/17539153.2018.1498190 Mwangi, O. G. (2021a). Al-Shabaab. In The Oxford research encyclopedia of African history. Oxford University Press. https://doi.org/10.1093/acrefore/9780190277734.013.785 Mwangi, O. G. (2021b). The securitisation of civil society organisations, Islamism and counter-terrorism in Kenya: Case study of MUHURI and HAKI Africa. In S. N. Romaniuk & E. T. Njoku (Eds.), Counter-terrorism and civil society: Post-9/11 progress and challenges (pp. 204–215). Manchester University Press. https:// doi.org/10.7765/9781526157935.00022 Mwangi, O. G., & Mwangi, C. W. (2019). The securitization of political discourse in reinforcing regimes of power in Kenya. Language, Discourse & Society, Special Issue on Political Discourse and Historical, Colonial and Neo-Colonial Regimes of Power, 7(2), 55–71. Nasser-Eddine, M., Garnham, B., Agostino, K., & Caluya, G. (2011). Countering violent extremism (CVE) literature review. Defence Science and Technology Organisation/Commonwealth of Australia. National Counter Terrorism Centre (NCTC). (2022). NCTC director’s office. https://counterterrorism.go.ke/ nctc-directors-office/ Ndungú, I., Salifu, U., & Sigsworth, R. (2017, November). Violent extremism in Kenya: Why women are a priority (Monograph 197). Institute for Security Studies (ISS). Ogada, M. (2017, November). A policy content evaluation of Kenya’s national strategy to counter violent extremism (Policy Brief No. 9). Centre for Human Rights and Policy Studies. Organisation for Economic Co-Operation and Development (OECD). (2006). Whole of government approaches to fragile states. OECD Publishing. Organisation for Economic Co-Operation and Development (OECD). (2020). States of fragility 2020. OECD Publishing. https://doi.org/10.1787/ba7c22 Organization for Security and Co-Operation in Europe (OSCE). (2020). A whole-of-society approach to preventing and countering violent extremism and radicalization that lead to terrorism: A guidebook for Central Asia. OSCE. Osland, K. M., & Erstad, H. U. (2020). The fragility dilemma and divergent security complexes in the Sahel. Italian Journal of International Affairs, 55(4), 18–36. https://doi.org/10.1080/03932729.2020.1833474 Papale, S. (2022). Fuelling the fire: Al-Shabaab, counter-terrorism and radicalisation in Kenya. Critical Studies on Terrorism, 15(2), 356380. https://doi.org/10.1080/17539153.2021.2016091 Pasagic, A. (2020). Failed states and terrorism: Justifiability of transnational interventions from a counterterrorism perspective. Perspectives on Terrorism, 17(3), 19–28. Pisoiu, D. (2012). Pragmatic persuasion in counter-terrorism Critical Studies on Terrorism, 5(3), 297–317. https://doi.org/10.1080/17539153.2012.728051 Radicalisation Action Network (RAN). (2021, June 9). RAN activities on local coordination and the local approach to P/CVE (Special Overview Paper). RAN. Ramadhan, R., & Ouma, L. (2020, May). COVID-19 and countering violent extremism measures in Kenya (Briefing Note No. 1). Centre for Human Rights and Policy Studies. REINVENT Programme. (2021). Countering violent extremism through target support to Lamu County government: An integrated approach Kiunga Youth Bunge: Pilot intervention learning paper. REINVENT Programme. Republic of Kenya. (2010). The constitution of Kenya. Government Printer. Republic of Kenya. (2012, May 11). Appointment of county commissioners, gazette notice no. 6604 (The Kenya Gazette Vol. CXIV, No. 40). Government Printer. Republic of Kenya. (2014). The security laws (amendment) act, 2014 no. 19 of 2004: Kenya gazette supplement no. 167 (Acts No. 19). Government Printer. Republic of Kenya. (2016). National strategy to counter violent extremism. Government Printer. Republic of Kenya. (2019a, November). 2019 Kenya population and housing census. Vol. I: Population by country and sub-county. Kenya National Bureau of Statistics. Republic of Kenya. (2019b, December). 2019 Kenya population and housing census. Vol. IV: Distribution of population by socio-economic characteristics. Kenya National Bureau of Statistics.

526

Radicalisation, counter-radicalisation, and governance Rink, A., & Sharma, K. (2018). The determinants of religious radicalization: Evidence from Kenya. Journal of Conflict Resolution, 62(6), 1229–1261. https://doi.org/10.1177/0022002716678986 Roberts, A. (2015). Fragile states: A concept with a history. In P. Lewis & H. Wallace (Eds.), Rethinking state fragility. British Academy. Rothermel, A. (she/her/hers). (2020). Gender in the United Nations’ agenda in preventing and countering violent extremism. International Feminist Journal of Politics, 22(5), 720–741. https://doi.org/10.1080/1 4616742.2020.1827967 Rrustemi, A. (2020). The unintended consequences of peacebuilding interventions contributing to the rise of violent extremism. Security and Human Rights, 31, 102–120. https://doi.org/10.1163/18750230-BJA10004 Sahgal, G., & Zeuthen, M. (2020). The Nexus between crime and violent extremism in Kenya: A case study of the prison complexes. The RUSI Journal, 165(4), 54–67. https://doi.org/0.1080/03071847.2020.1793686 Schomerus, M., El Tarabousli-McCarthy, S., & Sandhar, J. (2017). Countering violent extremism (Topic Guide). GSDRC, University of Birmingham. Sharamo, R. D., & Mohammed, A. A. (2020, December). Countering violent extremism in Kenya: Expectations and innovations (Report No. 36). Institute for Security Studies East Africa. Speckhard, A., & Shajkovci, A. (2019). The Jihad in Kenya: Understanding Al-Shabaab recruitment and terrorist activity inside Kenya – in their own words. African Security, 12(1), 3–61. https://doi.org/10.1080/1 9392206.2019.1587142 Stephens, W., & Sieckelinck, S. (2020). Being resilient to radicalisation in PVE policy: A critical examination. Critical Studies on Terrorism. 13(1), 142–165. https://doi.org/10.1080/17539153.2019.1658415 Stephens, W., Sieckelinck, S., & Boutellier, H. (2021). Preventing violent extremism: A review of the literature. Studies in Conflict & Terrorism, 44(4), 346–361. https://doi.org/10.1080/1057610X.2018.1543144 Thiessen, C. (2022). The strategic ambiguity of the United Nations approach to preventing violent extremism. Studies in Conflict & Terrorism, 45(2), 111–132. https://doi.org/10.1080/1057610X.2019.1647685 Thiessen, C., & Özerdem, A. (2019). Turkey in Somalia: Challenging North/Western interventionism. Third World Quarterly, 40(11), 1976–1995. https://doi.org/10.1080/01436597.2019.1619074 United Kingdom Government Stabilisation Unit. (2018, December). Countering violent extremism in fragile and conflict-affected states. Stabilisation Unit Report. United Nations (UN). (2015, December 24). Plan of action to prevent violent extremism (A/70/674). UN. United Nations Development Programme (UNDP). (2016a). Preventing violent extremism through promoting inclusive development, tolerance and respect for diversity: A development response to addressing radicalization and violent extremism. United Nations Development Programme. United Nations Development Programme (UNDP). (2016b). Preventing and responding to violent extremism in Africa: A development approach. United Nations Development Programme. United States (US). (2021). Country reports on terrorism 2010. Department of State Publication. United States Agency for International Development (USAID). (2020). Policy for countering violent extremism through development assistance, NW. USAID. Van Zyl, I. (2019, September). Preventing violent extremism: Lessons from Africa (Africa Report No. 17). Institute for Security Studies (ISS). Van Zyl, I., & Frank, C. (2018, September). Preventing extremism in West and Central Africa: Lessons from Burkina Faso, Cameroon, Chad, Mali, Niger and Nigeria (West Africa Report 23). Institute for Security Studies (ISS). Wario, H. A. (2021, June). Study to examine the influence of contemporary Islamic ideologies in Kenya: Mandera, Garissa, Marsabit and Isiolo counties. REINVENT Programme. Whelan, T. (2006). Africa’s ungoverned space. Nacao & Defesa, 114(3), 61–73. Witensburg, K., & Zaal, F. (2012). Spaces of insecurity. In K. Witsenburg & F. Zaal (Eds.), Spaces of insecurity: Human agency in violent conflicts in Kenya (pp. 1–19). African Studies Centre. World Bank. (2020). World bank group strategy for fragility, conflict, and violence 2020–2025. World Bank.

527

34 LOCAL CONTOURS AND GLOBAL DISCOURSES IN COUNTERING VIOLENT RADICALISATION AND EXTREMISM A perspective from the global south Fathima Azmiya Badurdeen Introduction Africa is a key frontier in the Global War on Terror (GWOT) agenda of the West in countering terrorism and countering violent extremism (CVE) (Prestholdt, 2011; Herbst & Mills, 2003). The surge of models to counter terrorism (CT), violent extremism, and radicalisation in the East African region reveals the significance of the region as a playing field for models, approaches, and best practices in preventing and countering extremism (P/CVE). The global paradigm in the aftermath of 9/11 centred on American foreign policy towards Africa (Glickman, 2003), where strategies to curb terrorism aimed at cooperative energies via joint military collaborations with African states (Cilliers, 2006) to expand their agendas in the P/CVE endeavours (Kundnani & Hayes, 2018). Hard military measures to CT had limited results in deterring those involved and were incapable of dissuading new recruits or at-risk populations from joining violent armed movements (Masferrer & Walker, 2013). P/CVE became the panacea for the infamous hard military approaches and aimed at winning the hearts and minds of local populations in the war against terrorism (Williams, 2020). The UN Secretary-General ascertained this by calling all member states to adopt national strategies on P/CVE to facilitate a comprehensive implementation of the United Nations Global Counter-Terrorism Strategy (UN General Assembly, 2018). The plan of action promoted by the UN Secretary-General encouraged national plan of action to prevent violent extremism by addressing local drivers of violent extremism (UN General Assembly, 2015). Many countries in East Africa have developed or are developing and adopting strategies to prevent and counter violent extremism. It was envisioned that these strategies would provide coherence and coordination for national and international efforts on P/CVE. These strategies strengthened donor-led P/CVE efforts in the South to explore national needs and concerns in P/CVE interventions. The gradual evolution of P/CVE efforts reveals CVE as donor-led top-down efforts with less resonance in local contexts where global templates are used in P/CVE interventions. For others, P/CVE frameworks are state-centric elitist security frameworks that influence violent extremism definitions, conceptualisations, and interventions (Nzau & Guyo, 2018). Therefore, it is imperative DOI: 10.4324/9781003035848-38 528

Countering violent radicalisation and extremism

to review existing counter-radicalisation and CVE policy and programming to understand the P/ CVE framework and its applicability at the regional, national, and local levels in East Africa. The chapter acknowledges the ongoing definitional and conceptual debates in terms such as violent radicalisation, violent extremism, and other practices (CVE and PVE). These definitional and conceptual debates are well articulated in the previous chapters of this book. In this chapter, I have chosen to use the terms countering violent radicalisation and CVE, accommodating these debates. This chapter aims to provide a comprehensive understanding pertaining to preventing or countering violent radicalisation and extremism in the East African region. Following the introduction, the chapter will proceed to explain the historical context for the concept of violent extremism in East Africa, followed by a section on the evolution of P/CVE programming in East Africa. The chapter then focuses on the key challenges encountered by donors in navigating and balancing local meaning-making on security and global trends in P/CVE programming. The final section will conclude with an assessment of P/CVE in the East African region.

Historical context of violent extremism in East Africa Academic and practitioner-oriented interpretations of violent radicalisation and extremism in East Africa predominantly centre on Islamist extremism (Kessels et al., 2016). Specific countries in East Africa are marked for youth recruitment for violent radicalisation and extremist acts and have localised hubs for propagating Islamist extremist narratives (Davis, 2010). In recent years, the landscape of violent extremism has grown increasingly complex, with splintered fractions of terrorist movements and new waves of violent extremism (Rotberg, 2005). The events of 11 September 2001 acknowledged the international community’s anxiety over the issue of terrorism in Africa. Terrorism or its associated process of violent radicalisation is not a recent phenomenon in Africa or East African region. What was new in the African or the East African region was the influence of the 9/11 aftermath, which harped on an Islamist transnational terrorist wave, securing the security interests of the United States and the ‘West’ (Haynes, 2005). However, these African terrorism events in the 2000s rarely occurred in isolation. The rapidly escalating new threat of Islamist extremism represented an already prevailing risk of violent radicalisation in the region. Before the transpiring of the transnational violent extremism, Africa was affected by domestic terrorism or insurgency movements. Terror was used and justified as a deliberate strategy or an unintended consequence of insurgencies, liberation struggles, or secession movements (Mwakikagile, 2018). Terrorising populations by rulers and state actors marked the African historical debates on terrorism in the pre-independence and post-independence eras. Practices of counterterrorism became complex within these state-led forms of terrorism and violence. Scholars like Oladosu Ayinde have scrutinised African definitional debates on terrorism and emphasised the need for taking a historical perspective. The author identifies three distinct terrorlike phases in the pre-independence era, namely, the Afro-Oriental phase, the Afro-Occidental phase, and the Afro-Global phase. The Afro-Oriental phase marked a period where rulers practiced servitude and offered individuals as slaves for external actors. The period also marked Arab traders enslaving African residents. While these cases do not encompass a form of political violence, they do reveal practices where rulers instilled fear among the ruled and victimised the ruled to demarcate power. The Afro-Occidental phase in the 15th century highlighted the European colonialism era, marking the African enslavement period. The Afro-Global phase in the 19th century extended European exploitation of Africa through expansionist policies (Ayinde, 2016). Overall, different ideologies marked the pre-independence era, where instilling fear to control and victimise the powerless formed the dominant ideology of the rulers. 529

Fathima Azmiya Badurdeen

The region has histories of anarchism, liberation wars, counterinsurgency campaigns, and Cold War struggles, all of which included forms of terrorism and counterterrorism measures (Heing, 2017). Most liberation movements in Africa involved armed campaigns, guerrilla warfare, or terrorism, and the terms used to describe their actions depended on who formulated the definition of terrorism. Some of these armed campaigns include the ‘Movement party’ in Uganda, the Rwandan Patriotic Front, the Tigray People’s Liberation Front, the Eritrean People’s Liberation Front, the Mau Mau rebellion in Kenya, or the Zimbabwe African National Union. Following liberation struggles, the continent also faced domestic terrorism or insurgencies as part of the decolonisation package, reflecting failed governance systems and sustainable development under successive post-colonial rulers. Rooted in these violent struggles are identity politics, inequality, underdevelopment, poverty, and marginalisation from the centre. Some movements had widespread violence against civilians, employing terror, and depending on who and how these movements were analysed and labelled as terrorism, namely, the Lord’s Resistance Army (LRA) in northern Uganda, Mai Mai in the Eastern Democratic Republic of the Congo (DRC), violent movements by warlords in Somalia, and the Revolutionary United Front in Sierra Leone (Mwakikagile, 2018). In East Africa, contemporary waves of violent radicalisation and extremism can be classified into four types. The first type is the presence of international terrorist groups and their activities in the region. The geo-political landscape of Muslims in Africa has seen Arab-Israeli relationships play out in the region. During the Cold War era, Arab and Israeli strategists established alliances in the region, where Africa became a ‘new dimension of Arab-Israeli confrontation in the African political theatre’. The crisis of these relationships culminated in extremist acts in countries such as Sudan and Kenya in the 1960s (Kwarteng, 1996). Palestinian or pro-Palestinian armed groups carried out extremist acts from the 1960s to the 1980s. In 1973, the Palestinian terrorist group Black September carried out the assassination of the American Ambassador and Deputy Chief of Mission and a Belgian diplomat in Sudan. On 31 December 1980, pro-Palestinian militants bombed the Norfolk Hotel in Nairobi as a revenge act for Kenya’s assistance in Israel’s Operation Entebbe in 1976 (Muendo, 2016). In 1995, a failed attempt to assassinate Egyptian President Hosni Mubarak in Addis Ababa, Ethiopia, was conducted by the Egyptian terrorist group Al Gama’a al-Islamiyya (AP News, 1995). The Sunni-Shia rifts among the Muslims were relatively calmer, with periodic communal tensions in the East African region in the 1970s. However, the resultant extreme strands of Salafi-Wahhabi extremism to counter Shia influence have taken a toll since the mid-1980s, moulding the present waves of Islamist violent radicalisation and extremism in East Africa. The second type of violent extremists in East Africa are the affiliates of the two major international terrorist groups, Al-Qaeda and ISIS. East Africa has been a global Jihadi foothold from the early days of Al-Qaeda operations in the 1990s to the contemporary Al-Shabaab movement. Al-Qaeda claimed responsibility for a series of attacks in the region. This includes the concurrent bombing of American embassies in Nairobi and Dar es Salaam in 1998. In 2002, Al-Qaeda operatives were responsible for a car bomb attack on the Kikambala Paradise Resort in Mombasa, popular with Israeli tourists. Since 2007, Al-Qaeda in the Arabian Peninsula (AQAP) has operated in the East African region, mainly in Eritrea and Somalia. The AQAP operations were reinforced by the war in Yemen. AQAP and Al-Shabaab links were strengthened during this period. On the other end, ISIS pledged allegiance to existing terrorist groups and insurgencies such as Boko Haram and the Al-Shabaab splinter group ‘Jahba East Africa’ in an attempt to establish a global caliphate. Franchises of ISIS and Al-Qaeda in the region indicate that violent radicalisation and extremism are gaining momentum in East Africa (Githingu & Hamming, 2021). The enmity between these two transnational terrorist movements was evidenced in East Africa, where each movement competed for supporters and established their own cells in the region (Gardner, 2020). ISIS had their 530

Countering violent radicalisation and extremism

own radicalisation, recruitment and facilitated recruits to travel to Syria, Libya and Iraq or other areas where the movement was active. Subsequently, from 2016 to 2022, ISIS has launched its own attacks in East Africa, such as attacking Somali Government troops and bombing an AMISOM vehicle in 2016, and taking responsibility for a series of attacks in Uganda, Mozambique, and Tanzania. The third type of violent radicalisation and extremism in the East African region can be attributed to homegrown insurgency movements and terrorist networks with different motivations, strategies, and tactics. In Somalia, the Somali Islamic Brotherhood movement gave rise to the Somali Islamic Union, or Al-Itihaad al-Islamiya (AIAI), in 1984, which was a predecessor of the Al-Shabaab (International Crisis Group, 2014). Currently, Al-Shabaab movement in Somalia is at the forefront of Islamist extremism in East Africa. Al-Shabaab in Somalia evolved and was strengthened as a consequence of four major factors: Somalis with Afghan mujahideen training and ideological commitments formed the groundwork for the formation of Al-Shabaab (Ali, 2008). Further, the Al-Qaeda influence in 1991 with Osama bin Laden’s relocation and influence in Sudan strengthened regional Jihadi ties. Al-Qaeda’s 1998 attacks in Kenya and Tanzania invigorated Islamist endeavours, motivating Islamist movements (Vittori et al., 2009). Next, the collapse of the Siad Barre regime in Mogadishu in 1991 ascended Al-Shabaab’s prominence within warlord control of the country. Al-Shabaab, formed in 2006, was a military splinter group of the Islamic Courts Union (ICU) and later overthrew the ICU. Finally, the public resentment of Ethiopia’s incursion gave rise to the Al-Shabaab insurgency, where by 2008 the group had most of the territorial control of southern Somalia (Hansen, 2016). By 2012, Al-Shabaab had renewed links with AlQaeda. In 2021, Al-Shabaab revealed its military capability, directing attacks against the Somali state, international governments, and other international actors (Harrington & Thompson, 2021). Al-Shabaab terrorist activities have targeted countries contributing troops to the peacekeeping mission, AMISOM (African Union Mission in Somalia). These countries include Somalia, Kenya, Uganda, Djibouti, and Ethiopia. Many retaliatory attacks have targeted Kenya because the Kenya Defence Forces (KDF) have joined AMISOM. Terrorist attacks include the Westgate Mall siege, which killed 67 people in 2013, the attacks at Garissa University College, which killed 148 people, and the attack at the DUSSIT complex in 2019, which killed 21 people. Subsequent attacks have been conducted in Mandera, Garissa, Isiolo, and Lamu. Al-Shabaab also launched a suicide attack on a resort in Djibouti in 2014. In Ethiopia, there have been a series of failed attempts to bomb targets. Al-Shabaab established its own affiliates in Kenya, such as Al-Hijra and Jaysh al-Ayman to radicalise, recruit, and train recruits inside Kenya. In Tanzania, Al-Shabaab affiliates include the Ansar Muslim Youth Centre and Al-Muhajiroun (Githingu & Hamming, 2021). While Al-Shabaab has maintained links and affiliations with Al-Qaeda, splinter individuals and groups from Al-Shabaab have even pledged support to ISIS (Reva, 2017). In 2021, terrorists affiliated with the Central African branch of the Islamic State and a faction of the Allied Democratic Forces (ADF) claimed responsibility for two suicide bombings in Kampala, Uganda (Mahtani, 2021). This also includes the recent attack killing 40 students in western Uganda in June 2023 (BBC News, 2023). Since 2017, new franchises of ISIS have been developed with Mozambiques’s Ahlu Sunna Wal Jama, also known as Al-Shabab (Al-Shabab with one ‘a’ and not affiliated to the Somali Al-Shabaab movement). Al-Shabab had carried out many terrorist attacks in Mozambique’s Cabo Delgado province (Heyen-Dube & Rands, 2021). Interestingly, most of these transnational Islamist networks rely on domestic roots, injustices, and strife, which are politicised and framed within belief systems. Having a strong enemy, such as the United States or Israel, provides terrorist organisations with a sympathiser base for recruitment and tactical support. Hence, macroand micro-level nuances of local and global grievances with religiously embedded tenets sharpen and strengthen ideology formation for violent radicalisation (Badurdeen, 2019). 531

Fathima Azmiya Badurdeen

The fourth type of violent radicalisation and extremism consists of non-Islamist militant movements such as the LRA in Uganda. The LRA is a Christian fundamentalist movement conducting attacks in northern Uganda, the Central African Republic, South Sudan, and the Democratic Republic of the Congo. Similarly, the Allied Democratic Front (ADF) in Uganda is also labelled as a terrorist organisation (Romaniuk & Durner, 2018). Other groups labelled as extremists include secessionist organisations such as the Mombasa Republican Council (MRC), based in the coastal region of Kenya. It is alleged that the MRC had links with Al-Shabaab and its Kenyan affiliate, Al-Hijra (Megged, 2015).

Evolution of preventing and countering violent extremism programming in East Africa In East Africa, a surge of models to curb terrorism and violent extremism evolved along with Islamist extremism. CT strategies in the region followed the ‘global war on terror’ agenda developed partly to prevent the spread of radical Islamic militancy over the last decade (Cordesman, 2017). Schmidt (2018) posits that patronage to the West was based on the willingness of countries in the South to serve as allies in efforts for regional policing, provide military bases for western use, and curbing Islamist radical movements. Foreign development aid was used to incentivise counterterrorism, where international priorities were fulfilled by donor funding at the expense of local interests, with unintended implications such as stigmatising minority populations or contributing to local insecurities. States often followed policies or even used template legislations, such as the Kenyan ‘Suppression of the Terrorism Bill 2003’ and ‘Anti-terrorism Bill 2006’, to mitigate terrorism. In Kenya, these two bills were rejected by the parliament as they reflected haphazard planning to please western donors and political convenience (Oando & Achieng, 2021). It was comprehended that hard security measures reduced the military capacity of extremists and thereby helped in liberating communities from the impacts of the extremists. The use of military and militarised law enforcement efforts has diminished results in dissuading at-risk populations from joining violent extremist movements. In 2014, the Kenyan government launched Operation Usalama Watch in Nairobi and Mombasa neighbourhoods aimed at eliminating terrorism. The operation was prompted mainly by the Westgate Mall attack in Nairobi in 2013. The operation detected illegal immigrants, mainly of the Somali ethnic group, of whom some were arrested and others prosecuted on suspicion of involvement with terrorist activities. The operation met wide criticism as it was considered biased targeting specific ethnic and religious minorities (Lind et al., 2017). Similarly, Operation Linda Boni was launched in 2015 in Lamu County in the aftermath of the Mpeketoni attack, followed by subsequent attacks directed at the security forces as well as villages in Lamu. The operation was done in good faith to flush out the Al-Shabaab militants. The operation expanded for months in Lamu and fed on the already existing community grievances of the people whose livelihoods were affected and education disrupted. The local communities bore the brunt of attacks perpetrated by the Al-Shabaab as well as the government’s counter-military responses (Nyagah et al., 2017). In the face of these challenges encountered due to hard militarised responses, East African governments, donors, and civil society organisations are increasingly looking beyond military solutions towards a holistic approach to CVE. CVE and PVE floated by donors emphasised noncoercive interventions to mitigate the threat from segments of populations already engaged in violent extremism. To reduce risks, they focused on recommending innovative ways to identify and secure individuals, groups, or communities at risk of radicalisation, amplifying credible moderate voices, and increasing community cohesion (Khalil & Zeuthen, 2014). 532

Countering violent radicalisation and extremism

Kenya is a worthwhile consideration on the playing field for counterterrorism and CVE approaches against Islamist extremism. The government of Kenya launched the National Strategy for Countering Violent Extremism (NSCVE) 2016 (Beyond the Lines, 2016). The NSCVE defined violent extremism primarily as Islamist extremism in line with the GWOT and strengthened the priorities of the West (Mogire & Agade, 2011). In Kenya, terrorism is synonymous with Al-Shabaab and ISIS, in both policy and practice (Oando & Achieng, 2021). Hence, prevention and countering violent radicalisation and extremism are only focused on Al-Shabaab or ISIS forms of Islamist extremism. Martini and Njoku (2017) identify two factors that have influenced the conceptualisation of terrorism and violent extremism in Kenya. The first is the hasty development and dissemination of foreign policies on terrorism by Western countries and their acculturation in the south, setting an imperialistic approach to counterterrorism strategies against Islamist movements. The second is the effort made to distinguish terrorism from other acts of political violence. The focus on Islamist extremism has negated all other forms of violent radicalisation and extremism, and it has influenced the Kenyan political violence discourse as well as policies, programmes, and funded projects on P/CVE. Associating Islam with terrorism and violent extremism is problematic in a country that has a large Muslim minority, particularly around the coast and the north-east, where the unintended impacts of security measures have construed suspect communities along religious and ethnic enclaves (Badurdeen, 2018).

Donors, national agendas, and local meaning-making in P/CVE Softer approaches for mitigating violent extremism are embraced in many countries of the South that are grappling with terrorism. In Kenya, soft approaches are institutionalised and strategised under the National Counter Terrorism Centre (NCTC) and the NSCVE, which advocate for winning the hearts and minds of the populace against terrorism. The NSCVE marks the end of a strong security centred CT approach and paves the way for the Kenyan government to partner with local civil societies to prevent and counter Islamist violent radicalisation among at risk communities in the country (Life and Peace Institute, 2016). The cascading influence of NSCVE on the counties is visible via its county action plans (CAPs) on CVE (Badurdeen & Goldsmith, 2018). In Kenya, the Mombasa and Kwale CAPs are hailed as models that are emulated in other counties. The journey of the NSCVE and the CAPs garnered attention as templates and are flaunted as effective models to be referenced in East Africa. However, critiques argue that the nationally led CVE strategy in Kenya is an elitist venture, creating a hierarchical organisational approach to CVE implementation. Monitoring and regulating local CVE activities under the banner of national security constrains local-level initiatives in environments where communities are yet to build trust and alliances with the national government (Crisman et al., 2020). The NSCVE specifically emphasises the importance of communities that need to be consulted, hence strengthening the dialogue process between the government and the local communities (Cleveland et al., 2020). The CAPs envisaged the potential to bridge the broken state-citizen relationships in the peripheral communities and broker the strategic dialectic process of information flows to mitigate violent extremism. Nevertheless, CAPs are yet to reveal their significance at the county level, where consumers of the CAPs still feel the local voices of those affected are unheard, with local elites influencing the definition and implementation of violent extremism in the counties. Further, the strong reigns of the national strategy loom over the stringent regulations and implementation of the CAPs, reinforced via the Prevention of Terrorism Act 2012 and its Amendment in 2019. These regulations are considered a state-centric approach to 533

Fathima Azmiya Badurdeen

directing P/CVE efforts along state-defined lines of violent extremism and their ability to control alternative avenues of dissent. Most P/CVE projects are funded externally by donors, who are often engaged in defining and determining P/CVE interventions appropriate to their organisational mandates or agendas. This external dominance can overshadow local P/CVE interventions, as interventions are based on existing western templates. Since the inception of CVE projects, Kenya has strategised interventions mainly aligned to the Aarhus model of anti- and de-radicalisation strategies (Andersen & Wiuff Moe, 2015). This model was introduced to Kenya by the Danish Government in 2012. The collaboration between the Danish Security and Intelligence Service and the NCTC reflects the trendy use of exported models in countries of the global South. The Aarhus model is explicitly mentioned in the NSCVE and most of the County Action Plans (CAPs) in Kenya and is expected to be used during the implementation of these CAPs (Breidlid, 2021). The model was well received and contributed to the design of many CVE interventions in the region. The workability of imported CVE models is not without its problems. In Kenya, P/CVE interventions using the criteria of selecting individuals at risk or ‘at risk’ groups were considered discriminatory as they marginalised individuals based on religious or ethnic identities, influencing the creation of suspect communities among Muslim-dominated localities in Mombasa and Kwale. The desire to seek support for P/CVE efforts among Muslim communities has categorised individuals along the extremist versus moderate continuum (Badurdeen, 2018). Similarly, in Somalia, a report noted how the government misuses the term CVE or penetrates P/CVE initiatives where the police interrogate Muslim men in Mogadishu with or without justification, considering them at risk or labelling them as Al-Shabaab sympathisers. Most often, the criteria considered in these initiatives were ‘age’ and ‘religion’, where young Muslim men were considered susceptible to violence and radicalisation (Saferworld, 2020). Deliberate efforts are needed to broaden the ‘at risk’ category to reduce stigma and target of individuals for P/CVE programmes channelled along the criminology model of victim-focused policies (Bednarova, 2011). Similarly, Breidlid (2021) expressed concern about the use of theology and social-psychological radicalisation models to influence P/CVE in Kenya. The emphasis of such a model is on religion and individual social networks, which can be used by implementers to delegate surveillance on networks where law enforcement officials screw this towards specific ethnic and religious groups in the East African region. Governments do have the potential to control donors and civil society organisations via explicit or implicit policies and practices. Being in good books with the government is a necessity in donor operations, particularly in P/CVE programming. Countries can exhibit control via policy documents where P/CVE programming is favourable to the country, mainly with government priorities. Accordingly, this can influence the definitions and interpretations of violent extremism and P/CVE based on government priorities. Sometimes, CVE programming by donors may cater less to the impact of CT strategies such as arbitrary arrests or disappearances of individuals in CT, as it reduces the hassles of government confrontations in work operations.1 This securitisation of P/ CVE efforts restricts donors as well as civil society organisations from concentrating on the real trends in local violence and radicalisation (Crisman et al., 2020). Civil Society Organizations often sway between balancing donor and government agendas in P/CVE efforts versus local interests. Local priorities are among the lowest considerations where CSOs have to please donors and government agendas due to funding dependency. CSOs have to strike a balance between what they believe is good for the beneficiaries and what the government and donors believe is good for the beneficiaries. Donors, with their models, experiences, and funding, do shape how CSOs define and interpret the problem of violent extremism. 534

Countering violent radicalisation and extremism

External models and assumptions on what works well and what does not fail to consider local or indigenous knowledge in moulding the P/CVE project to local concerns. Despite donor’s good faith in funding particular CSOs and local communities with externally tested programmes, the missing local knowledge and perspectives affect or distort results. In some communities in Kenya, the local disputes of cattle rustling or clan-based conflicts are far more threatening than Islamist extremism. Hence, the focus on Islamist extremism overshadows local priorities that need immediate attention. Likewise, increased interventions to curb Islamist extremism can further exacerbate religious differences and create new religious-ethnic tensions, resulting in new forms of conflict. Being screwed towards Islamist extremism negates other forms of violent radicalisation and extremism, such as police brutality, in the name of counterterrorism as espoused by specific peripheral communities (Aroussi et al., 2020). Here, Kamau (2006) emphasises that the more the Kenyan government engages in counterterrorism, the more opportunities there are for mobilising for violent radicalisation and extremism. The author explains this as the yoyo phenomenon, which results in the proliferation of violent extremism in Kenya. Local definitions and understandings of terms and concepts such as violent radicalisation or extremism are contextual based on cultural, social, and political connotations (Ahmadi & Urwin, 2018). In a study conducted in three counties considered hotspots for violent extremism in Kenya, the local residents identified varied forms of violence as violent extremism apart from Islamist extremism. Residents did not use the term or concept ‘violent extremism’ in their everyday lives and conflated the term with other types of violence. Some explained gangs as violent extremism due to the increased gang activities in the counties studied. For others, election-related violence was considered violent extremism due to the extreme forms of political ideologies and violence used during the election cycles. Some expressed violent extremism as linked to terror attacks, Al-Shabaab recruitment, and distorted interpretations of religious texts used in the Al-Shabaab ideology. Women and girls identified domestic violence as a potential form of extremism, and few highlighted that abusers are violent and are likely to join or have associations with armed groups like gangs or violent extremist networks. However, this link warrants more research. The study emphasised that different perspectives were to be acknowledged when defining local terms for violent extremism, as local definitions are based on local drivers and manifestations. The concept of violent extremism was also time-bound as communities defined it with changing forms of violence. Immediately after a terror attack by the Al-Shabaab or the ISIS, people will talk about the Al-Shabaab or the ISIS. Close to political elections, people talk about political or election violence, and in gang prone areas, people identify gang violence as violent extremism. The study highlighted that violent extremism was seen as an outcome of a wider system of violence prevailing in particular contexts embedded in histories of violence, violent social movements, and historical violent events. Finally, the study acknowledged that constructions of violent extremism at the local level are shaped by lived experiences of insecurity and influenced by gender, ethnicity, social status, location, and interactions with the State (Badurdeen et al., 2022). Elitist perspectives in framing and mitigating violent extremism can suppress expressions of and resistance to local grievances by communities. In some communities, expressions of local grievances are termed as radicalisation or extremism, specifically when grievances represent government abuses, political and economic marginalisation, or religious discrimination. For example, in Kenya, the MRC is a secession movement, originally formed to express coastal grievances. For some, the movement was purely democratic, where the movement highlighted local concerns. Gradually, specific members of the MRC resorted to violence, and some were even alleged to have had links with Al-Shabaab. The government outlawed this group in 2010. Local community members wavered on characterising the group as violent, or violent extremists, as some members 535

Fathima Azmiya Badurdeen

felt the movement was an avenue to express the coastal grievances and marginalisation that have spanned decades due to lop-sided post-colonial development planning.2 This example shows how outlawing a group could overlook the difficult discussions that need to take place with those who are affected and are trying to convey their grievances via specific acts. To shun these movements from the wider discussions on local governance and development often paves the way for new forms of extremism against the state. These spaces also become bases for extremist networks such as Al-Shabaab and ISIS recruitment, where the state is often viewed as a perpetrator. Lacking space for dissent provides transnational extremist networks such as the Al-Shabaab in Kenya to fuel existing state-citizen deteriorating relationships for their recruitment drives. For example, the narratives of the two movements (MRC and Al-Shabaab) frame ‘the state’ as the perpetrator. MRC focuses on the ideology of state marginalising the coastal region of Kenya, while the Al-Shabaab focuses on the state marginalising and discriminating against Muslims in the Coast. Acknowledging these local-transnational extremism linkages is vital to P/CVE efforts (Badurdeen, 2019).

The way forward in P/CVE The flexibility of P/CVE projects and the ability to factor in changing environments and the threat of newer forms of VE are crucial. P/CVE projects should not confine themselves only to the Islamist extremist wave in the East African region but be open to prevent other forms of violent extremism trends in the region. Future research studies on violent extremism should recognise the presence of different forms of extremist movements in the same location or region and the fact that they affect and impact one another. Often, curtailing one extremist movement can reduce the other or prevent one movement from being co-opted into the other, as extremist groups may feed off each other’s narrative among the same base of populace. For example, violent radicalisation by the Al-Shabaab and the ISIS mobilises the same disgruntled populace in the peripheral communities of Kenya, often using each movement’s narratives linked to political and economic marginalisation, poverty, and religious discrimination. The mismatch between P/CVE interventions centred on the global or national agenda and the local civil society interpreting and integrating knowledge dissemination in prevention or countering does affect P/CVE effectiveness. The effectiveness of P/CVE is dependent on local and contextual knowledge of how locals define and propose solutions to what they consider violent extremism. P/ CVE projects often impose elite centred national and global perspectives to counter violent radicalisation and extremism. The top-down approaches sponsored by donors in collaboration with existing government strategies often lacked commitment by local communities, which confined themselves to participation on a project-to-project basis rather than seeing it as a long-term social change. Some civil society organisations incorporate CVE into their mandates, as CVE is an international buzz word and has the potential to attract donor funding. Hence, donors export their CVE concepts with no knowledge of the local context, local needs, or how the CVE intervention will benefit the local population.3 Lack of locally relevant evidence in P/CVE means that P/CVE will be confined to donor preferences rather than locally relevant forms of violent extremism (Badurdeen et al., 2022). Individual grievances such as social injustices, political marginalisation, poverty, and unemployment are important to take into consideration when attempting to understand the nature of violent radicalisation and extremism in the region (Khalil & Zeuthen, 2014; Badurdeen, 2019). Strategies in P/CVE need to engage with violent radicalisation as a wider societal issue rather than an obligation of particular communities. This entails accountability and responsibility on the part of the government and law enforcement authorities to make greater efforts to assess the problem of violent extremism as well as its solutions. It is often assumed that violent Islamist radicalisation is a Muslim 536

Countering violent radicalisation and extremism

problem and that the responsibility lies exclusively with Muslim communities. Stigmatising these communities has the potential to fuel another cycle of mobilisation for violence (Badurdeen, 2018). Do No Harm strategies remain key in P/CVE programming to minimise stigmatisation and labelling of individuals under the banners of ‘extremists’ or ‘moderates’ (Holmer et al., 2018). Evidence-based contextual programming to counter and keep up with existing trends in violent radicalisation and extremism in the East African region is steadily increasing. Locally relevant models to understand violent radicalisation processes and extremism are still work in progress, as most radicalisation and de-radicalisation models are co-opted from the West. Additionally, indigenous knowledge in programming is yet to be fully realised in P/CVE approaches, along with the necessity for sustainability (Oando & Achieng, 2021). Contemporarily, most P/CVE initiatives are linked to development agendas in programming as a positive response to address violent extremism in the region. Integrating P/CVE initiatives within a broader framework of development or peacebuilding is vital to bridge local grievances which are at the heart of the violent radicalisation processes. Romaniuk et al. (2018) suggests a four-part rubric for development actors who intend to work on CVE. First is the evidencebased nature of CVE interventions. In the East African region, not all countries are equipped with research to strengthen CVE initiatives, hence only countries with a strong evidence base can reap the benefits for an appropriate P/CVE strategy. Second, the need to consider the nature and type of development programming to address violent extremism. Do we find development actors better placed to address violent extremism, especially if violent extremism is influenced by state or law enforcement officials? Are we able to label different forms of violent extremism for intervention based of different local contexts? Third, the proportionality of development responses to prevent violent extremism in areas where there are other immediate priorities and security concerns, which may not fall under the label of violent extremism but may be contributory factors. Finally, it needs to be assessed whether we have a strong evidence base to gauge the effectiveness of development interventions in reducing or preventing violent extremism. For, evaluation of development-P/CVE programming is still at infancy, and the effectiveness is yet to be measured. Regional trends in P/CVE demand regional responses to prevent and counter violent radicalisation and extremism. Regional mechanisms have the potential to facilitate trans-boundary analysis of P/ CVE, nurture cross-border strategies in P/CVE, and influence partnerships to monitor state or locallevel strategies on P/CVE. Regional approaches towards P/CVE could be enhanced with South-South cooperation, mainly via bilateral or multilateral partnerships in P/CVE, where affected countries in the region can share best practices and learn from one another. The UN Counter-Terrorism Strategy has emphasised South-South cooperation in preventing terrorism. The Inter-Governmental Authority on Development (IGAD, 2018) has established a regional strategy for P/CVE, promoting regional P/ CVE practices in East Africa. The regional strategy strengthens the realisation of the African Union’s Agenda 2063 on socio-economic transformation for the long-term development agenda of the continent. However, funding for most of the regional activities mainly hails from the West, with their expertise, knowledge, and modelled templates. The success of these interventions is dependent on the workable nexuses of global, regional, and local perspectives on P/CVE in policy and practice rather than being influenced by external priorities of the West or elite-led decisions of the South.

Notes 1 Personal communication with NGO personnel, Mombasa, December 2017, during PhD fieldwork 2 Personal communication with religious leader, Kwale, June 2017, during PhD fieldwork 3 Personal communication with NGO personnel, Mombasa, July 2017, during PhD fieldwork

537

Fathima Azmiya Badurdeen

References Ahmadi, B., & Urwin, E. (2018). Measuring peace and violent extremism: Voices from the Afghan village. United States Institute of Peace. Ali, A. (2008, April 28–29). The Al-Shabaab Al-Mujahidin – a profile of the first Somali terrorist organization. Paper presented at the Joint Kenya-Uganda Border security and Management Workshop at IGAD Capacity Building Programme against Terrorism, Jinja, Uganda. Andersen, L. E.,  & Wiuff Moe, L. (2015). Responding to radicalization: Exporting the dilemmas. In N. Hvidt & H. Mouritzen (Eds.), Danish foreign policy yearbook 2015 (pp. 71–101). DIIS Danish Institute for International Studies. AP News. (1995, August 2). Three suspects arrested in failed attempts to assassinate Mubarak. AP News. https://apnews.com/article/6ecfb2c754f9cffd9e1fdcd7afa0b3e7 Aroussi, S., Jakala, M., Badurdeen, F. A., & Verhoest, X. (2020). Body maps of resistance understanding everyday resistance to violent extremism in Kenya. The University of Leeds. http://eprints.whiterose. ac.uk/165575/1/Body%20maps%20of%20resistance%20Book%20Final.pdf Ayinde, O. A. (2016). Beyond 9/11: Histories and spaces of terrorism in Africa. In M. Smith (Ed.), Securing Africa: Post 9/11 discourses on terrorism. Routledge. Badurdeen, F. A. (2018). Socio-legal implications of the laws to combat religious extremism in Kenya. In B. Green, J. Gunn, & M. Hill (Eds.), Religion, law and security in Africa (pp. 111–130). ACLARS. Badurdeen, F. A. (2019). Youth radicalization in the Coastal region of Kenya: Social movement theory approach [PhD Thesis, Department of Social Sciences, Technical University of Mombasa]. Badurdeen, F. A., Aroussi, S., & Jakala, M. (2022). Lived realities and local meaning-making in defining violent extremism in Kenya: Implications for preventing and countering violent extremism in policy and practice. Critical Studies on Terrorism, 15(5). Badurdeen, F. A., & Goldsmith, P. (2018). Initiatives and perceptions to counter violent extremism in the coastal region of Kenya. Journal for Deradicalization, 16. https://journals.sfu.ca/jd/index.php/jd/article/view/162 BBC News. (2023, June 17). Uganda school attack: Dozens of pupils killed by militants linked to Islamic State group. BBC News. https://www.bbc.com/news/world-africa-65937484 Bednarova, J. (2011). The heart of the criminal justice system: A critical analysis of the position of the victim. Internet Journal of Criminology. www.antoniocasella.eu/restorative/Bednarova_2011.pdf Beyond the Lines. (2016). The Kenya national strategy to counter violent extremism simple put! www.beyondthelines.org/simpliying-the-national-strategy-to-counter-violent-extremism/ Breidlid, T. (2021). Countering or contributing to radicalization and violent extremism in Kenya? A critical case study. Critical Studies on Terrorism, 14(2), 225–246. Cilliers, J. (2006). Africa, root causes and the “war on terror”. African Security Studies, 15(3), 57–71. Cleveland, E., Glaeser, S., Jorgensen, L., Negede, L., & Shropshire, A. (2020). Preventing and countering violent extremism: Localization in Kenya, Kosovo, and the Philippines. Robert M. La Follette School of Public Affairs, University of Wisconsin. Cordesman, A. (2017). Islam and the patterns in terrorism and violent extremism. Center for Strategic and International Studies. www.csis.org/analysis/islam-and-patterns-terrorism-and-violent-extremism Crisman, B., Yoon, C. S., Goos, C., Hull, D., Romano, E., & Johnson, J. (2020). Preventing violent extremism: Lessons from Kenya. Princeton University, Woodrow Wilson School of Public and International Affairs. Davis, J. (Ed.). (2010). Terrorism in Africa: The evolving front in the war on terror. Lexington Books. Gardner, F. (2020). Is Africa overtaking the Middle East as the new Jihadist battleground? BBC. www.bbc. com/news/world-africa-55147863 Githingu, B. M., & Hamming, T. R. (2021). The arc of Jihad: The ecosystem of militancy in East Africa. International Centre for the Study of Radicalisation. Glickman, H. (2003). Africa in the war on terrorism. Journal of Asian and African Studies, 38(2), 162–174. Hansen, S. J. (2016). Al-Shabaab in Somalia: The history and ideology of a militant Islamist group. Oxford University Press. Harrington, J., & Thompson, J. (2021). Examining extremism: Harakat al Shabaab al Mujahideen (al Shabaab). www.csis.org/blogs/examining-extremism/examining-extremism-harakat-al-shabaab-almujahideen-al-shabaab Haynes, J. (2005). Islamic militancy in East Africa. Third World Quarterly, 26(8), 1321–1339. Heing, B. (2017). War and terrorism in contemporary Africa. Cavendish Square Publishing.

538

Countering violent radicalisation and extremism Herbst, J., & Mills, G. (2003). Africa and the war on terror. The RUSI Journal, 148(5), 12–17. Heyen-Dube, T., & Rands, R. (2021). Evolving doctrine and modus operandi: Violet extremism in Cabo Delgado. Small Wars and Insurgencies. https://doi.org/10.1080/09592318.2021.1936956 Holmer, G., Bauman, P., & Aryaeinejad, K. (2018). Measuring up: Evaluating the impact of PCVE programs. United States Institute of Peace. IGAD. (2018). Regional strategy for preventing and countering violent extremism. https://cve.igad.int/wp-content/ uploads/2020/03/The-IGAD-Regional-Strategy-for-Preventing-and-Countering-Violent-Extremism.pdf International Crisis Group. (2014). Somalia: Al-Shabaab – it will be a long war (Africa Briefing No. 99). International Crisis Group. Kamau, W. C. (2006). Kenya and the war on terrorism. Review of African Political Economy, 33(107), 133–141. Kessels, E., Durner, T., & Schwartz, M. (2016). Violent extremism and instability in the greater horn of Africa: An examination of drivers and responses. Global Center on Cooperative Security. Khalil, J., & Zeuthen, M. (2014). A case study of countering violent extremism (CVE) programming: Lessons from OTI’s Kenya transition initiative. Stability: International Journal of Security & Development, 3(1). Kundnani, A., & Hayes, B. (2018). The globalisation of countering violent extremism policies undermining human rights, instrumentalising civil society. Societal Security Network of Excellence. www.tni.org/en/ publication/the-globalisation-of-countering-violent-extremism-policies Kwarteng, C. (1996). Africa and the PLO-Israeli peace initiative: Payoffs, pitfalls and paradoxes. International Journal on World Peace, 13(4), 17–45. Life and Peace Institute. (2016). Horn of Africa bulletin. Life and Peace Institute. Lind, J., Mutahi, P., & Oosterom, M. (2017). Killing a mosquito with a hammer: Al-Shabaab violence and state security responses in Kenya. Peacebuilding, 5(2), 118–135. Mahtani, D. (2021). The Kampala attacks and their regional implications. International Crisis Group. www. crisisgroup.org/africa/great-lakes/uganda/kampala-attacks-and-their-regional-implications Martini, A., & Njoku, E. (2017). The challenges of defining terrorism for counter-terrorism policy. In S. N. Romaniuk, D. Grice, D. Irrera, & S. Webb (Eds.), The Palgrave handbook of global counterterrorism policy (pp. 73–90). Palgrave Macmillan. Masferrer, A.,  & Walker, C. (2013). Counter-terrorism, human rights and the rule of law: Crossing legal boundaries in defence of the state. Edward Elgar Publishing. Megged, M. (2015). Al-Shabaab terrorists affiliate groups in East Africa. Strategic Intelligence. https://intelligencebriefs.com/al-shabaab-terrorists-affiliate-groups-in-east-africa/ Mogire, E., & Agade, K. M. (2011). Counter-terrorism in Kenya. Journal of Contemporary African Studies, 29(4), 473–491. Muendo, L. (2016, July 4). How Kenya played a vital, silent role in Entebbe, the most audacious hostage rescue in history. The Times of Israel. www.timesofisrael.com/kenya-a-silent-partner-in-the-mostaudacious-hostage-rescue-mission-in-history/ Mwakikagile, G. (2018). The African liberation struggle: Reflections. New Africa Press. Nyagah, T., Mwangi, J., & Attree, L. (2017). Inside Kenya’s war on terror: The case of Lamu. Saferworld. Nzau, M., & Guyo, M. (2018). The challenges of securing Kenya: Past experience, present challenges and future prospects. The Journal of Social Encounters, 2(1), 37–59. Oando, S., & Achieng’, S. (2021). An indigenous African framework for counterterrorism: Decolonising Kenya’s approach to countering “Al-Shabaab-ism”. Critical Studies on Terrorism, 14(3), 354–377. Prestholdt, J. (2011). Kenya, the United States and counter terrorism. Africa Today, 57(4), 3–27. Reva, D. (2017). How will ISIS setbacks impact Africa? Center for Strategic & International Studies. https:// issafrica.org/iss-today/how-will-isis-setbacks-impact-africa Romaniuk, P., & Durner, T. (2018). The politics of preventing violent extremism: The case of Uganda. Conflict, Security and Development, 18(2), 159–179. Romaniuk, P., Durner, T., Nonninger, L., & Schwartz, M. (2018). What drives violent extremism in East Africa and how should development actors respond? African Security, 11(2), 172. Rotberg, R. (Ed.). (2005). Battling terrorism in the horn of Africa. Brookings Institution Press. Saferworld. (2020). Thinking beyond “countering violent extremism” in Somalia: Local views on an international “project”. www.saferworld.org.uk/long-reads/thinking-beyond-acountering-violent-extremismain-somalia-local-views-on-an-international-aprojecta Schmidt, E. (2018). Foreign intervention in Africa after the cold war: Sovereignty, responsibility, and the war on terror. Ohio University Press.

539

Fathima Azmiya Badurdeen UN General Assembly. (2015). Plan of action to prevent violent extremism, report to the secretary general. www.un.org/counterterrorism/plan-of-action-to-prevent-violent-extremism UN General Assembly. (2018). The United Nations global counter-terrorism strategy review (A/RES/72/284). https://digitallibrary.un.org/record/1632770?ln=zh_CN Vittori, J., Bremer, K., & Vittori, P. (2009). Islam in Tanzania and Kenya: Ally or threat in the war on terror? Studies in Conflict and Terrorism, 32(12), 1075–1099. Williams, M. J. (2020). Preventing and countering violent extremism: Designing and evaluating evidencebased programs. Taylor and Francis.

540

INDEX

Note: Page numbers in italics indicate figures and page numbers in bold indicate tables. Aarhus model: crime prevention and 372; deradicalisation strategies and 534; multiagency (MA) approaches and 370, 372, 374, 376 – 377; P/CVE and 462; reintegration efforts and 454, 456; strengthening society in 373 Abas, N. 449 – 450 Abu Ghraib 442 Acheson, D. 356 Achieng’, S. 342 – 343 actuarial justice 403 – 404 actuarial risk assessment 476, 484, 490 Adnani, A. M. 222 AfD (Alternative for Germany) 221, 225 affect 4, 28 Afghanistan 234, 240 – 241, 321 Africa: counter radicalisation policy and 518; counterterrorism and 528; external security and development interventions 516 – 517; fragile states in 514 – 517; Global War on Terror (GWOT) and 528, 533; historical context of violent extremism in 529 – 531; homegrown insurgency movements in 530 – 531; liberation movements in 530; Muslim communities and 530; non-Islamist militant movements in 532; porous borders in 516; radicalisation to violent extremism in 514 – 518, 529 – 532; Sahel region 321, 515 – 518; securitisation of counterradicalisation in 519; see also East Africa; Kenya; Somalia African Union 537 Agenda 2063 (African Union) 537

Aggarwal, N. K. 360 Agnew, R. 118 Ahlu Sunna Wal Jama 531 Ahmad, F. 120, 124n6 Aitken, J. 58 – 59 Ajzen, I. 152 – 153 Alanen, V. 12 Alatas, S. H. 349 Aldrich, D. P. 326 Alexander II 60 Al Gama’a al-Islamiyya 530 Algemene Inlichtingen-en Veiligheidsdienst (AIVD) 35, 278 Algeria 277, 447 Al-Hijra 531 – 532 Alimi, E. Y. 108n2 Al-Itihaad al-Islamiya (AIAI) 531 al-Jamaah al-Islamiyah 447 Al-Khansaa Brigade 172 Alkheder, M. 442 Allely, C. 185 Alliance of Red Front Fighters 266 Allied Democratic Forces (ADF) 531 – 532 Al-Muhajiroun 87, 103, 167, 531 Al-Qaeda: Al-Shabaab and 531; counter narratives to 432, 434; East Africa and 530; foreign fighters and 233, 241; ideological radicalisation and 106; lone-actor terrorism and 218; narratives of 436; pre-crime offenses and 409; proscription of 407; recruitment by 35, 139; Salafi-Jihadi ideology and 71; state-community P/ CVE and 294; system level of analysis 77;

541

Index transnational networks and 79 – 80; UK influence 45, 400, 403 Al-Qaeda in the Arabian Peninsula (AQAP) 530 Al-Shabaab: Al-Qaeda and 531; counter radicalisation policy and 522; in East Africa 530; ISIS supporters and 531; Jahba East Africa group 530; in Kenya 106, 520, 535 – 536; military approach against 342; Mombasa Republican Council (MRC) and 532, 535; resistance to recruitment by 437; Salafi Jihadism and 517; Somalia and 516, 520, 531; terrorist attacks by 531 – 533; violent radicalisation and 536 Al-Shabab 531 alternative spirituality movements 140 Altier, M. B. 268, 448 Amarasingam, A. 10, 196, 238 American Revolution 58 – 59 American Society of Criminology (ASC) 113 Amirault, J. 120 anarchist radicalisation 61 – 64 Ancelovici, M. 108n1 Andersen, J. C. 106 Anderson, B. 197 – 198 Anderson, J. B. 120 Andersson Malmros, R. 358 – 359, 369, 378 – 379 Andrie, T. 454 – 455 anger: behavioural motivation and 158 – 159; extremists and 423; Muslim youth and 35 – 36, 39 – 40, 44 – 45, 134; physiological changes and 158; radicality and 427; terrorist messaging and 158 – 159 anomie theory 118 Ansar Muslim Youth Centre 531 anti-abortion movement 218 anti-cult movement 136, 140 Anti-Islamic Movement 106 anti-Semitism 93, 264 Anti-Terrorism and Rapid Response Service (ATPI) 278 Area Under the Curve (AUC) 478 Arielli, N. 233 – 235 Aristotle 197 Armee Islam du Salut 447 Aronoff, J. 136 Ashour, O. 447 Ashworth, A. 359, 408, 410 Asia 252, 339, 369, 514 – 515; see also Southeast Asia ASI Data Science 505 al-Assad, B. 205, 231 – 232 assassination: Catholic radicalisation 54 – 57; external affirmation and 56; personal crises and 55 – 56, 58; personal relationships and 56; self-sacrificial 55; solitary political 53 – 64; sponsorship of kings for 55, 57

Association for Citizenship Teaching (ACT) 285 at risk individuals: intervention evaluations 469 – 470; interventions and 461 – 462, 466 – 468; limitations of interventions 468 – 469; local level detection of 462 – 463, 467 – 468; mentoring and 467 – 469; militarised law enforcement efforts and 532; P/CVE training for 463, 467 – 468; pre-crime space and 461 – 462; public health model and 360, 461; risk assessment tools 464 – 465; structured professional judgement (SPJ) approach 464 – 465; terrorism and 24, 462; tertiary mentoring programmes and 461; UK Prevent strategy and 358, 360, 463, 476 – 477; see also risk assessment and management Attitudes-Behaviours Corrective (ABC) model 88, 267 audiences: counter narratives and 431, 434 – 435; extreme group messaging and 167; fear appeals and 159 – 160; hope and 161; narrative reasoning and 432; online communication and 224; persuasion and 433, 436 – 437; pride and 161 – 162; radicalisation process 150, 156, 159, 162 Augestad Knudsen, R. 462 Australia: anti-Muslim sentiment in 93; antiSemitism in 93; community-oriented P/CVE 293, 301, 303n1; Community Support Group programmes 295; counter radicalisation policy 257, 292 – 293, 325, 327; far-right extremism and 205; multiagency approaches in 369; multicultural state policies 295; P/CVE evaluation in 344 – 345, 345, 346, 347; public health model in 358, 364; white supremacist groups 80 Australian Strategic Policy Institute 71 Australia’s Counter-Terrorism Strategy 303n1 Austria 241 autism spectrum disorders 183 – 186 autonomisation 282 – 283 Ayinde, O. 529 Azov Battalion 80 Azzam, A. 240 – 241 Baaken, T. 271 Babchishin, K. M. 483 Badurdeen, F. A. 528 Baillet, S. 224 – 225 Bakardjieva, M. 198 Bakker, E. 328 Bali bombing (2002) 449, 453 Banchik, A. V. 510 Bandura, A. 26, 94 Banks, W. C. 120

542

Index Baqiya family 206 – 207 Baran, Z. 39 Barker, E. 139 Bartlett, J. 249 Baubérot, J. 288n1 Baudrillard, J. 341 Bauman, Z. 295 Baykal, A. 331, 333 Beelmann, A. 425 Begum, S. 280 behavioural motivation: action tendencies and 157; anger and 158 – 159; autonomy and 154; behavioural capacity and 154; discrete emotion theory 150 – 151, 157; emotionmotivated responses 157; fear appeals and 160; mental illness and 361; persuasion and 151; radical action and 43, 141; radical ideas and 253 – 254; reasoned action theory 150 – 156; research in 3 – 4; violence and 248 Behlendorf, B. 116 Belton, E. 325, 327 Benigni, M. C. 203 Benmelech, E. 135, 238 Bennett, N. 257 Benyettou, F. 284 Bergema, R. 135 Berger, J. M. 87, 263 – 265, 294, 302, 322 – 323 Berkman, A. 62 Berlin Christmas market attacks (2016) 113, 222 Berntzen, L. E. 106 Bersani, B. E. 120 Bertelsen, P. 373 Bhatt, A. 489 Bhui, K. S. 187, 356, 359, 361 Bickert, M. 502 Bigo, D. 282 – 283, 287 Binks, E. 138 Bin Laden, O. 449, 531 Bird, F. 139 Birdwell, J. 249 Birt, Y. 300 Bismarck, O. von 60 – 61 BitChute 508 Bjørgo, T. 371, 447 – 448 Black Lives Matter movement 427 Black Nationalism 218 Black September 530 Blair, T. 41, 403 Bliuc, A-M. 205 Bloom, H. 78 BNPT see Indonesian National Counter-Terrorism Agency (BNPT) Boko Haram 252, 518, 530 Bolin, T. A. 219 Bolstad, G. 395n1 Bonelli, L. 288n3

Borum, R. 39, 114, 141, 489 Bosi, L. 108n2 Boucek, C. 447 Boucher, J. 54 Bouhana, N. 5, 375 Bourdieu, P. 72, 282 – 284 Bouzar, D. 284 Bowman-Grieve, L. 204 boyd, d. 198 – 199 Braddock, K. 150, 255 – 256, 326 – 327, 360, 436 – 437, 439, 447 Brandon, J. 249, 375 Breidlid, T. 376, 534 Breivik, A. B. 106, 214, 219, 224 Briggs, R. 433, 439 Britain First 509 Brooke, N. 9 Brown, K. E. 504 Bruhn, J. G. 198 Bruno, F. 22 Bubolz, B. F. 185, 189n14 Buckingham, George Villiers 55 – 56, 65 Bugg, D. 203 Building Resilience Against Terrorism (Canada) 303n1 bunch of guys model 28, 40, 167, 249 Bundsgaard, J. 376 Burawoy, M. 314 – 315 Burschenschaft movement 60 Burton, R. 57 Bush administration 41, 45, 434 Busher, J. 1, 4, 320 – 321, 409 Byrd, S. C. 106 calculated propaganda 73 Cameron, D. 71, 92 Campbell, M. A., 481 – 482 Canada: criminalisation discourse in 341; CVE and 371; evaluation culture in 330; foreign fighters from 135; multi-agency approaches in 371; multicultural state policies 295; resilience and 303n1; risk assessment tools 464 – 465; terrorist sentencing in 121; violent radicalisation research from 338, 348 Canadian Practitioners Network for the Prevention of Radicalization and Extremist Violence (CPN-PREV) 330 Carlile, A. 406 Carlsbad Decrees 60 Carlsson, C. 119 Carson, J. V. 117 Cassam, Q. 248, 423 – 425, 427 Catesby, R. 56 Catholic radicalisation 54 – 57, 61 Cawley, O. 403 Center for the Analysis of Social Media 93

543

Index Central Directorate of Prevention Police (DCPP) 278 Central Unit for Investigations (NIC) 278 Centre for Research and Evidence on Security Threats 431 Chad 518 Challgren, J. 358, 361 Chandrasekharan, E. 507 Channel programme: criticism of 468; multi-agency approaches in 279, 370, 372, 374 – 375, 377; pre-crime offenses and 401 – 402, 409 – 410; radicalisation referral programme 92, 330, 359 – 360, 402, 410 – 411, 466 – 467 Chan platforms 223 – 224 charismatic groups 77 – 78, 80 Charlie Hebdo attacks (2015) 393 Chastel, J. 54 – 57 Chavis, D. 199 – 200 Cherney, A. 325, 327 Chernov Hwang, J. 446 – 448 Choudhury, T. 5, 399 Christakis, N. 76 Christchurch attacks (2019) 79, 82n2, 216, 219, 222 – 224, 504 Christodoulou, E. 416 Chubb, J. 315 Chukwuma, K. 247 civil resistance movements 4 Civil Society Against Violent Extremism (C-SAVE) 454 civil society organisations (CSOs): alienation of communities 300; deradicalisation and 454 – 455; funding for 300 – 301; as information conduits 299; P/CVE programmes and 293, 298 – 300, 302, 532, 534 – 535; securitisation of 518; statecommunity relationships 270 Clarke, R. V. 117 Clausewitz, C. von 80 – 82 Cleaver, E. 218 Clément, J. 54, 57 Clément, P.-A. 331 Clemmow, C. 120 Cloward, R. A. 100 Clubb, G. 106, 248 Coaffee, J. 297 cognitive drift 140 cognitive radicalisation: extremist beliefs and 248; group cohesion and 88; political ideology and 26 – 27; religious conversions and 140; research in 3 – 4; social relations and 28, 43 Cohen, L. E. 116 Coid, J. W. 187 Coke, E. 57 collective narratives 85 – 86 Colombia 255

colonialism: coloniality and 340; marginalised communities and 515, 520; medicalised language and 360 – 361; military occupation and 340; radicalisation and 42, 164; security strategies and 342 – 343; West and 42, 340 – 341 coloniality: P/CVE evaluation and 344 – 345, 345, 346, 347 – 349; production of knowledge and 349; PVR frameworks and 339 – 340; West and 340 coloniality of knowledge 340, 342 – 345, 347, 349 coloniality of power 339 – 340 Committee for Financial Security (CFS) 278 communication theories: counter narratives and 439, 441, 443; counter radicalisation and 162; discrete emotion theory 150, 152, 157; persuasive messaging and 155 – 156; radicalisation process and 162; reasoned action theory 150, 152 – 156; terrorist propaganda and 150 Communist International (Comintern) 240 community: conceptualisation of 197 – 198, 296; digital 296; imagined 197 – 198, 294 – 295; influence of 199; integration and fulfilment of needs 199; lived 295; membership in 199; migrant-background 295; multicultural state policies 295 – 296; Muslim engagement in 293 – 294; networks and 198; P/CVE partnership and 292 – 296, 300 – 302; project narrative of 26; psychological benefits of 199; radicalisation and 10; rationalisation and 198; resilience and 296 – 298, 303n1; responsibilisation of 294, 298 – 299; sense of 199 – 200, 204; shared emotional connections 199 – 200; social capital and 293; sociological approaches to 294 – 295; subcultures and 295; suspect 294, 299; symbolic attachment to 198; terrorism prevention and 298 – 299; see also online communities competitive escalation 102 complexity theory 78 – 79 Computer Forensics Research Lab (Univ. of Alabama) 502 – 503 Connell, R. W. 170 conspiracy theories 221 – 222, 266 Contentious Politics Paradigm 102, 104 content moderation: account-level actions 506; AI and 504 – 505; automated tools and 499, 501 – 505, 509; civil liberty concerns with 10, 501, 503 – 505; contentlevel actions 506; co-regulation and 500; counter radicalisation and 506; crowdsourced flagging and 498, 502 – 503; deplatforming and 507 – 508; enforcement and 498, 507 – 509; EU code of conduct

544

Index for 500 – 502; far-right extremism and 507 – 509; hate speech and 251, 498, 500 – 501, 505, 507; Internet Referral Units (IRUs) 503; machine learning algorithms 505; multistakeholder governance 500, 509 – 510; NetzDG and 500 – 501; platform governance 500 – 501; radicalisation and 499; regulatory context for 499 – 502, 506; researcher data and 509, 510; review and 498; self-regulation and 499 – 501, 509; shadowbans and 506; social media and 498; terrorist content and 498 – 501; transparency concerns 508 – 510; trusted flaggers and 502 – 504, 508, 510n1, 510n2; victimhood narratives and 507; visibility reduction actions 506 CONTEST strategy 250, 279, 293, 374, 401, 432, 466 Convention of the Prevention on Terrorism and a Counter- Terrorism Strategy 399 conveyor belt model 5, 39 – 40, 42 – 43, 91 – 92 Conway, M. 202 – 203, 223, 507 Coolsaet, R. 19, 34, 237 – 238 Copenhagen School 514 Corday, C. 58 – 59 Cordes, B. 25 Corman, S. 433, 442 Corner, E. 29, 120, 137, 180, 182 – 184, 186 Cornish, D. B. 117 Cottee, S. 118 Council of American Muslims for Understanding 432 Council of Australian Governments (COAG) 303n1 Council of Europe 396n8, 399 counterfactuals 8, 256, 462 Countering Violent Extremism (CVE): counter narratives and 359, 432, 437; ethical issues in 359 – 361; evaluation of 362; former extremists and 435; health promotion approach 356; information-gathering and sharing practices 377 – 378; lack of data for 362; multi-agency approaches to 361 – 363, 369 – 380; Muslims as suspect communities 359; non-prescriptive programmes for 426; public health model 355 – 365, 434; resources for 196; soft approaches in 355; strategies for 355; U.S. programme for 34, 45 – 46, 298, 300; see also counter radicalisation policy; preventing and countering violent extremism (P/CVE) counter messaging spectrum 433 counter narratives: audiences for 434 – 435, 441; citizen-led approaches 440; conceptualisation of 431 – 433, 442; content delivery 441; content of 436 – 437; counter messaging work 433; credibility and legitimacy in

434 – 435, 440; criticism of 432 – 433, 437 – 438, 442 – 443; democratisation and 442; diaspora community and 437; effectiveness of 6, 438 – 439, 443; evaluation of 438 – 439, 443; extremist propaganda and 256; former extremists and 326, 435, 437; gamification of 439; government-initiated 431, 441 – 442; ideology-oriented approaches 280; informal approaches 439 – 440; inoculation-style approaches 436 – 437, 439; Muslim campaigns 434; narrative theory and 432 – 433, 436; negative consequences 442; NGOs and 431, 435, 438, 442; normative critiques 440; online content and 207, 432; online/offline dynamics 441 – 442; P/CVE and 322, 359, 432; post-9/11 global focus on 431 – 432; presentation modes 437; as propaganda 440 – 441; theoretical base for 433, 437 – 438; UN discourse and 391 – 392; use of drama in 436; victims and survivors 435 counter radicalisation: blindspots around 9, 169, 173; Carlsbad Decrees and 60; communication and 162; counterterrorism and 5, 34, 45, 47, 384 – 385; defining 250; depluralisation of political concepts and values 267; discrete emotion theory 150; effectiveness of 9, 255 – 256; embodied practices in 4; empirical studies 7 – 9; EU strategies and 37 – 38, 45; gendered practices 164 – 165, 173; hope and 160 – 161; interventions and 4 – 7, 46; legitimacy of 256 – 257; mainstreaming and 385 – 395; methods and tools for 272; narrower approaches to 248 – 253, 255 – 256; pride and 160 – 162; reasoned action theory 150; relative understandings of radicalisation and 254 – 255; societal factors and 266; stakeholders in 5; targeting of ideas 256 – 257 counter radicalisation policy: approaches to 247 – 258, 276; autonomisation and 282 – 283; community-oriented approaches 279, 292 – 298; effectiveness of 247; European countries and 38, 46 – 47, 92, 276 – 286, 288n4, 400; evaluation of 320 – 321; focus of 7; fragile states and 514, 517 – 519; heteronomisation and 277, 282 – 286; ideology-focused 91 – 92, 279 – 280; impact of research on 314 – 317; interstitial and 283 – 284, 286; involvement in 276 – 281, 285 – 287; laïcité and 280 – 281, 284, 288n1; law-enforcement and 277 – 278; legitimisation of war on terror 518 – 519; mainstream approaches 302; multi-agency approaches to 281, 361 – 362, 369; multiculturalism and 295;

545

Index multi-level models for 92 – 93; non-state actors and 517 – 519; party politics and 282; pre-emptive criminal offenses and 400 – 401, 403 – 404; prison intelligence service and 285; probation officers and 285; public health model 5 – 6, 297, 355 – 357; PVE-E policies and 418 – 419; racialisation of Muslims and 409; radicalisation studies and 34; as reaction to attacks 281 – 282; risk assessment in 403 – 404; securitisation of 388 – 389, 519; security politics and 282 – 283; security professionals and 283 – 286; social problem models 277 – 279; state-community relationships 298 – 301; transparency concerns 503 – 504, 508; United Nations and 400; weak governance and 517 – 519; whole-of-society approach 517, 520; women and 519, 522; see also preventing and countering violent extremism (P/CVE) counter radicalisation research: comparative 263; data collection challenges 312 – 314; data production 313; deradicalisation professionals and 288; dissemination of 314 – 317; diversity in 262 – 263; ethical dilemmas and 313; evidence-based approaches 269; funding for 310 – 312; gender and 173, 311, 314; heterogeneous field of 308 – 310; non-academic stakeholders in 310; policy-making impact of 286, 314 – 317; practitioner networks in 310; reflexivity in 286 – 287, 314; researcher position in 306 – 309; research object in 307 – 309; state engagement and 315 – 317; technology and 313; undertheoretisation and 46; Western perspectives on 338, 342 counterterrorism: actuarial justice and 403 – 404; backlash outcomes 117 – 118; on causes of radicalisation 36 – 37; counter radicalisation and 5, 34, 45, 47; EU strategies and 37 – 38; gendered approaches to 165; impact of social media on 203; interdiction and 4; international cooperation in 389 – 390, 400; mainstreaming of counter radicalisation 384 – 395; multiple actors in 5; post-9/11 global focus on 24, 384, 387, 389 – 390, 395, 399; predictive element in 20; risk assessment in 393; soft/hard approaches in 355; South-South cooperation in 537; Western approaches to 533; see also Countering Violent Extremism (CVE); preventing and countering violent extremism (P/CVE) Cowan, D. E. 199 Cox, J. 75

Crawford, A. 370 Crawford, K. 502 Crenshaw, M. 19 – 20, 26 crime prevention: hard/soft approaches to 371; multi-agency approaches to 281, 369 – 372, 377; P/CVE strategies and 468; preemptive interventions and 467; problem of attribution in 324; radicalisation process and 281; situational 116 – 117; social 371; terrorism prevention and 404 crime-terror nexus 215, 238 criminal law: encouragement of terrorism 406 – 408; international cooperation in 400; preemptive/pre-crime offenses 399 – 401, 403 – 411, 411n1; preparation for terrorism 406; proscribed organisations 407, 410; terrorism offenses and 399, 403 – 411 criminal subculture theories 118 – 119 criminological theories: anomie theory 118; atheoretical analyses 120; criminal subculture perspectives 118; deprivation theories 118; learning perspectives 119; life-course perspectives 119, 122; opensource data and 122; psychopathology and 189n3; rational choice theory 117 – 118, 124n6; research in 114 – 115, 115, 120 – 122, 124n3, 124n4; routine activity theory 116 – 117; situational action theory 120; situational crime prevention (SCP) 117; situational perspectives 116, 122, 124n6; social control perspective 120, 122; social disorganisation theory 120; strain theory 118, 122; violent extremism and 113 – 122 critical discourse analysis 385 critical race theory 409 critical studies 339 – 341 Critical Terrorism Studies (CTS) 286 Crone, M. 253 – 254, 373, 376 Cronulla riots (2005) 205 Crosby, A. 341 Cross, R. 102 Cruijff, J. 235 C-SAVE 455 CTIRU (Counter Terrorism Internet Referral Unit/ London Metropolitan Police) 503 culture: collective programming and 72 – 73; defining 72; frames and 73; ideology and 72 – 74; Islamic 341; masculinity and 170 – 171; out-group prejudice and 73 CVE see Countering Violent Extremism (CVE) cycles of protest 101 Czolgosz, L. 63 – 64 Dalgaard-Nielsen, A. 167, 267 – 269, 371 Damiens, F. 57 – 58, 66 Daniels, J. 202

546

Index discrete emotion theory: action tendencies 158; affective-based perspectives and 150 – 152, 156 – 157; behavioural motivation and 150; emotions in 157 – 158, 158, 159 – 162 disengagement: behaviour and 446, 457n4; civil society organisations and 454 – 456; defining 446 – 447; deradicalisation and 446 – 456, 457n3; empowerment of women in 453 – 455; former extremists and 453 – 454, 456; government-initiated 449 – 452, 456; in Indonesia 446 – 456; mutual aid societies 453 – 454; political context and 448; professional development and 456; reintegration and 448 – 449, 454; Rusbult’s investment model 448; social networks and 448; state-civil society initiative 455 Dobbelaere, K. 139 Doherty, P. 476 domestic terrorism 263, 302, 355, 529 – 530 Do No Harm strategies 537 Doosje, B. 447 Driven Data 505 Dugan, L. 117 Duplex Theory of Hate 94 Duriesmith, D. 172 Dussel, E. D. 340 Duyvendak, J. W. 108n1

da Silva, R. 461 Davies, L. 421, 426 – 427 Dawson, L. L. 4, 75, 132, 138 – 139, 201 – 202, 204, 238 De Coensel, S. 489 decolonial studies: coloniality of knowledge and 349; coloniality of power and 340 – 341; criminalisation of Muslims and coloniality 341 – 342; critical studies and 340 – 341, 349n4; Islam associations with terrorism 341; North-to-South knowledge flow 341 – 344, 348; PVR frameworks and 340 – 341; security studies and 339 decolonial turn 340 Deikman, A. 76 De Koster, W. 204 Delanty, G. 197 – 198 delegitimation theory 27 Della Porta, D. 25 – 28, 101 – 102, 104 Demetriou, C. 108n2 Denmark: Aarhus model 372 – 374, 376 – 377, 454, 456, 462; counter radicalisation and 251, 326, 357; foreign fighters from 236, 240 – 241; interventions for at-risk individuals 467 – 469; interventions with former extremists 326, 435; ISIS reintegration strategies 448; lone-actor terrorists and 216; P/CVE and 462, 467; public health model in 357 – 358; social care logics 379; teacher reporting in 378 Densus 88 449 – 451, 453 Dentice, D. 203 Department of Information and Safety (RIS) 278 deprivation theories 118, 134 – 136 deradicalisation: belief systems 457n2; BNPT efforts 451 – 453; civil society organisations and 454 – 456; collective 446 – 447; conveyor belt model 43; defining 446 – 447; disengagement and 446 – 456, 457n3; effectiveness of 255 – 256; exit pathways 268 – 269; government-initiated 449 – 453, 456; ideological 446 – 447, 456n1; individual level 446 – 448; initiatives for 34, 46; neo-Nazis and 270; online communities and 208; policy and practice 91 – 93; positive personal relationships and 268; professional development and 456; push and pull factors 268, 448; reintegration programming and 4, 448, 450; state-civil society initiative 455; see also counter radicalisation Desmarais, S. 184 De Wolf, A. 447 Dharma, S. 449 – 450 Dhumad, S. 184 Dickens, G. L. 483

Earnest, J. T. 82n2 East Africa: historical context of violent extremism in 529 – 531; homegrown insurgency movements in 530 – 531; international terrorist groups in 530; Islamist extremists in 529; Jihadist groups and 530 – 531; nonIslamist militant movements in 532; P/CVE programmes in 321, 528 – 529, 532 – 537; radicalisation to violent extremism in 529 – 532, 537 Ebner, J. 74, 85, 89 Echevarria, A. J. 80 echo chamber effect 76 – 77 Economic and Social Research Council (ESRC) 431 education: exclusion and marginality in 423; extremism in 415 – 417; foreign fighters and 135; informal support and 427; multiperspectivity in 424 – 425; need for significance in 423 – 424, 426; P/CVE interventions and 269, 271, 321; prevention of extremism and 419 – 427; radicalisation in 417 – 418; radicalism and 423 – 424, 427; religious terrorism and 134; securitisation of 416 – 417, 427; social control perspective and 120; see also preventing violent extremism in education (PVE-E); schools

547

Index education for democratic citizenship (EDC) 421 Edwards, G. 108n1 Edwards, J. 408 Egypt 187, 447 – 448, 452, 530 Egyptian Islamic Jihad (EIJ) 447, 452 Eijkman, Q. A. 369 Eijkman, Q. A. M. 462 Ekland-Olson, S. 104 Elaboration Likelihood model 436 Elizabeth I 54, 56 Elser, G. 64 – 65 embodied practices 4, 107 emotions: action tendencies 157, 158; anger 158 – 159; behavioural motivation and 157; communication theories and 157; defining 157; fear 158 – 160; hope 160 – 161; manifest expressions and 157; neurological changes and 157; physiological changes and 157 – 158; pride 158, 160 – 161; radicalisation and 4, 99; subjective feeling and 157 empirical studies: atheoretical analyses 115; individual terrorists 22 – 24; language content 8; limitations of 7 – 10; NRMs and 140; open-source data and 7; psychopathy and personality traits 181; risk assessment and management 478; social movement theory and 167; terrorism and 150 equifinality 185 Erikson, E. 23 Eritrean People’s Liberation Front 530 ethics: dilemmas 426, 463, 518; ‘ethical deprivation’ 136; problems with CVE strategies 359 – 361; of research 6, 9, 10 – 11, 308, 311, 313, 326; of sharing information 323 Ethiopia 515, 531 Etzioni, A. 298 EU Counter-Terrorism Strategy 37 – 38 Europe: anarchist radicalisation in 61 – 64; counter radicalisation policy 38, 46 – 47, 92, 276 – 286; far-right terror attacks in 218; foreign fighters from 35, 231, 236 – 237, 241; jihadists in 132; national-separatist campaigns in 26; P/CVE evaluation in 329 – 330; political violence and 277; radicalisation in 53 – 57; terrorist attacks in 34, 36, 103, 113, 218; women’s journeys to join ISIS 165; youth anger and 35 – 36; see also European Union; Western Europe European Commission 36, 38 – 39, 41, 296, 369, 393, 419 European Council 38 European Counter Terrorism Centre (ECTC) 393 European Parliament 92

European Union: counter radicalisation and 34, 45 – 46, 400; counterterrorism strategies 37 – 38, 251, 392 – 394, 461; deradicalisation programmes 46; Digital Services Act 501, 508 – 509; Framework Decision on Combating Terrorism 399, 411n4; mainstreaming discourse and 389 – 390, 392 – 395; online content moderation and 499 – 504, 510; pre-emption and 399; radicalisation concept in 35 – 39, 41; radicalisation research and 92; root causes rhetoric 41 EU Security Union strategy 393 Euskadi Ta Askatasuna (ETA) 448 evaluation of P/CVE: access to data 323 – 324; at-risk interventions and 469 – 470; challenges of 321 – 332, 362; clarifying scope and objectives 321 – 323, 327; collaboration in 329 – 332; coloniality and 343 – 345; community-level approaches and 303; counter narratives and 438 – 439, 443; deradicalisation programmes and 255; effectiveness claims and 327 – 329; evidence-based policy agendas and 320; formative 321; grounded theory approach 325; linguistic turn in 347 – 348; measurement of negatives 324 – 325; multimethod approaches 325 – 326; policies and programmes 9 – 10, 273, 320 – 321; problem of attribution 324; quasi-/experimental designs in 326 – 327; randomised control trials (RCTs) 324, 326; realist framework 329; researchers and 317; self-evaluation and 329; social movement theory and 99; standardised assessment instruments for 322; theory-driven approaches 327 – 329, 332; Western-centric approach to 344 – 345, 345, 346, 347 – 349 Everitt, B. 187 Expert Group on Violent Radicalisation (EC) 38 – 39, 41 External Intelligence and Security Agency (AISE) 278 extreme right-wing (XRW): community and 292, 296, 298, 302; comparison with Islamic extremism 264, 267 – 268, 271; counter narratives and 440; foreign fighters and 235; online communities and 432; P/ CVE and 268 – 269, 271; radicalisation conceptualisations and 339; research on 8, 262, 267 – 268; state-community relationships 292, 294; see also far-right extremism Extremism Risk Guidance (ERG22+) 464, 477, 490 extremist groups: anger in messaging 159; disillusionment with 268; fear appeals and 160; frame alignment and 87; gender as

548

Index organising principle of 169 – 173; in-group bias and 263; ideological radicalisation and 76; life-course perspectives 119; mindset and 423 – 424, 427; out-group prejudice and 263 – 264; patriarchal structural norms in 172; pride in target audiences 161; relative psychic deprivation and 136; targeted messaging and 157 – 162; uncertaintyidentity theory and 87; see also terrorist groups Eysenck, H. J. 265

Fauzi, A. 453 – 454 Fawkes, G. 56 Fazel, S. 485 fear: avoidance-orientation 160; behavioural motivation and 160; counter-radicalisation initiatives and 257, 409, 417; hoaxes and 220; persuasive messaging and 159; physiological changes and 157; pre-emptive terrorism offences and 400; radicality and 427; susceptibility to harm 160; terrorist messaging and 158 – 160; weaponisation against enemies 160 fear appeals 159 – 160 Feddes, A. R. 326, 447 Feenberg, A. 198 Fein, M. 159 Felson, M. 116 Felton, J. 55 – 57, 65 female terrorism see women’s violence Ferguson, N. 138 Fernández de Mosteyrín, L. 306 Ferracuti, F. 20, 22, 118 Feve, S. 433, 439 fight and die scale 93 Fillieule, O. 105 Fishbein, M. 152 – 153, 155 – 156 Fisher, T. 320 – 321, 325, 328 Fishman, B. 502 Fitzgerald, J. 316 Flashpoint 502 foreign fighters: Al-Qaeda training and 233; characteristics of 236 – 237, 242; class background and 134 – 135, 236 – 237; country-specific factors 238 – 239; defensive frames 234; defining 232 – 233; deprivation theories 135 – 136; digital natives as 205; emergence of nation-states and 231; as foreign volunteers 233; ideology and 233, 235 – 238; Iraq war and 40, 134 – 135, 196, 236, 241; ISIS mobilisation 74, 135; Jihadist groups and 231, 234, 237, 240; linguistic and cultural homogeneity 135; Muslim youth and 135, 205 – 206; non-state aid 232 – 233; payment and 232 – 233; push and pull factors 237 – 239; radicalisation process 232, 234 – 235, 239 – 240; reasons for becoming 239; recruitment of 234 – 235, 240 – 241; reintegration research 357; research in 8, 40, 44, 236 – 239; risk factors and 357; search for meaning by 235 – 236; social media use 205 – 206; social ties and 240; socio-economic backgrounds 237 – 239; Syrian civil war and 40, 45, 74, 134 – 135, 196, 205 – 206, 231, 236, 241; terrorism and 233 – 234; threat of 232; transnational identities and 234 – 235;

Faccini, L. 185 Facebook: Britain First takedown requests 509; content moderation and 498 – 500, 502 – 504, 505; counter-extremism policies 508 – 509; counter-messaging on 438 – 439; counter radicalisation efforts 505; deplatforming and 508 – 509; EU code of conduct for 500; extremist information exchange on 201; far-right extremism and 509; foreign fighters and 205; GIFCT and 504; hate speech and 505; live-streaming of terrorist attacks 82n2; machine learning algorithms 505; suspension of extremist accounts 223, 434; trusted flaggers and 502, 510n2; white supremacist advertising on 509 Fahey, S. 116 – 117 Falls curfew 118 far left: Black Nationalism and 218; extremism and 218; mental disorders in terrorism 182; personality traits and 264; political attitudes and 265; terrorism and 20, 118, 270 far right: divisive identity politics and 47; gender roles in 166, 171; ideology and 71, 74; manhood and 171; online communities and 8, 203 – 205; in party politics 220 – 221; personality traits and 264; political attitudes and 265; quest for significance 74 – 75; radical discourse and 220 – 221; recruitment by 161; social movement theory and 107 far-right extremism: anti-abortion movement 218; antiscientific worldview 264; conspiracy theories and 221 – 222; counter narratives 440; deplatforming and 508; ideological radicalisation and 76, 79, 81 – 82, 86, 264; informal approaches to 440; isolated indoctrination and 76; leaderless resistance and 215; lone-actor terrorists and 213, 218, 224 – 225; mental disorders and 182; online content moderation and 508 – 510; personality traits and 264; protest masculinities in 172; radicalisation process 224, 267 – 268; radicalisation research and 8, 263, 339; transnational networks and 79; see also extreme right-wing (XRW)

549

Index Waffen-SS and 233; Western 22, 135, 231 – 232; women and 233 Forst, B. 120 four-stage radicalisation process 39, 87 Fowler, J. 76 fragile states: counter radicalisation implementation in 514, 517 – 518, 523; defining 514; domestic counter-radicalisation objectives 517 – 518; external security and development interventions 516 – 517; global security and development priorities 517 – 518; inadequate provision of basic security 515 – 516; porous borders in 516; radicalisation to violent extremism in 514 – 520, 523; repressive security practices in 517 – 519; rise of violent ideologies in 517; role of women in 519; securitisation in 523; social and political conflicts in 516; ungoverned spaces in 516; weak governance and 514 – 519, 523 frame alignment 87 frames 73 Framework Decision on Combating Terrorism 399, 411n4 framing theory 106 France: counter radicalisation policy 46, 280 – 284, 288n3, 288n4, 288n5; hijab bans 173; laïcité and 277, 280 – 281, 284, 288n1; pre-emptive offenses in 399 – 400; terrorist attacks in 280 – 281 Frankfurter, D. 64 Freedman, L. 80 free spaces 102 – 103, 203, 205 Freilich, J. D. 120 Freire, P. 424 – 425 French Revolution 58 – 59 Frijda, N. H. 157 fusion-plus-threat model 89, 90, 91 fusion to violence pathways 89, 91, 91 Futrell, R. 203 – 204

social construction of 164, 169 – 170; societal power and 164 – 165; terrorism policy and 165; violent group organisation and 169 – 172; see also masculinity gendered radicalisation: extremist groups and 169 – 173; grievances and 167; hierarchies of masculinity in 6, 165, 170 – 171; individuals and 166 – 167; intersectional approach to 166; male extremists 164, 167 – 168, 170; manhood in 171; men and women’s roles in 166, 171; misogyny and 165, 168; power and 170; protest masculinities 170 – 171; terrorism studies and 165 – 166, 168 – 169; women and 166, 168 – 169; see also women’s violence General Investigations and Special Operations Divisions (DIGOS) 278 Generation Identity 225 Gentry, C. E. 165, 173 Gérard, B. 54, 56 German Empire 61 Germany: assassination attempts in 61; counter radicalisation and 262 – 263, 269 – 272; far-left extremists in 270 – 271; far-right extremists in 23, 141, 213, 267 – 268, 270 – 271; foreign fighter recruitment and mobilisation 241; hijab bans 173; interventions for at-risk individuals 466; Islamist extremists in 267 – 268, 270 – 271; leftist movements in 101, 270; lone-actor terrorism in 216, 225; neoNazi deradicalisation in 270; Network Enforcement Act (NetzDG) 500 – 501; P/ CVE and 270 – 273; PVE-E policies in 419; terrorism related criminal offences in 400; Waffen-SS volunteers 234, 236, 240; white supremacist groups in 76 Ghez, J. 447 Gibraltar incident 118, 124n5 Giddens, A. 298 GIFCT see Global Internet Forum to Counter Terrorism (GIFCT) Gill, P. 29, 120, 183, 186, 188, 189n4, 216, 220, 375, 476, 485, 487 Gillespie, T. 502 Gillian, K. 108n1 Glazzard, A. 73 – 74, 329, 432 Global Community Engagement and Resilience Fund (GCERF) 296 Global Counter-Terrorism Forum (GCTF) 293, 296 Global Internet Forum to Counter Terrorism (GIFCT) 504, 509 Global North 3, 8, 46, 338 – 339 Global South 8, 46, 537 Global Terrorism Database (GTD) 344 Global War on Terror (GWOT) 45, 342, 528, 533

Gab 507 Galanter, M. 77 Gallacher, J. 505 Gambetta, D. 264 – 265 gamification 224, 439 Gamson, W. 203 Ganesh, B. 10, 498 Gaylin, W. 73 Gefen, D. 198, 201 gender: counter radicalisation research and 311, 314; counterterrorism approaches and 165; mainstreaming and 385 – 387; masculinist groups 165; masculinities and social relations 170; radicalisation policy and 6, 164 – 165; sexual violence tactics 165;

550

Index Gogh, T. van 36, 278 Goldenberg, A. 265 Goldman, E. 62, 506 Goodman, K. 476 Google 499, 505, 510n4 Gorwa, R. 505 Gøtzsche-Astrup, O. 188, 266 – 267 Greenbaum, Z. 426 Grever, M. 425 Grimes, J. D. 219 Grol-Prokopczyk, H. 120 Grossman, M. 292, 297 group-based violence 464 – 465 group norms 94 group radicalisation: echo chamber effect 76 – 77; hate and 76 – 77; identity fusion 77, 88 – 89, 91; ideology and 75 – 76; isolated indoctrination and 76 – 77; men’s friendship groups and 167; out-group prejudice and 76 – 77, 81, 87, 91; processes of 105; radical milieus and 104; as religious cult 76; research in 3; terrorism and 20 – 21; see also extremist groups; terrorist groups groups: charismatic 77 – 78, 80; defining 75 – 76; identity fusion 77, 88 – 89, 91; member relationships 76; social identity theory and 87 – 88; transnational networks 78; see also extremist groups; online communities Gruenewald, J. 120 Grynszpan, H. 64 Guantanamo Bay 442 Gunpowder Plot (1605) 54 – 57, 65 Gurr, T. D. 25, 30n1, 30n4 Gustloff, W. 64

hate speech: banning subreddits and 507; content moderation and 251, 498, 500 – 501, 505, 509; counter radicalisation and 251; far-right extremists and 509; social media and 506 Hayes, B. 338 – 339 Hayward, K. J. 119 health promotion approach 6, 356, 363 – 364 Heath-Kelly, C. 7, 358, 360, 402 Hegghammer, E. T. 171, 232, 237 Heldt, A. 500 Henry, É. 62 – 63 Henry III 54, 57 Henry IV 53 – 55, 57 Hertog, S. 264 – 265 heteronomisation 277, 282 – 286 Hezbollah 410 High-Level Commission Expert Group on Radicalisation (HLCEG-R) 419 Hillyard, P. 294, 299 Himmler, H. 240 Hindutva extremists 222 Hine, C. 199 Hitler, A. 64 – 65, 236, 240 Hizb ut-Tahrir 79, 284 Hödel, M. 62 Hoffman, B. 43 – 44, 240 Hofstede, G. 72 Holbrook, D. 75 Holman, T. 241 homegrown terrorism 139 – 140 Homel, R. 117 hope 160 – 161 Horgan, J. 5, 41, 75, 105, 136, 184, 249, 255 – 256, 320, 327, 447 – 448 Houtman, D. 204 Hsu, H. Y. 118 Hulusi, H. 248 human behaviour: cognition and 151; discrete emotion theory 150 – 151; emotion and 151; impact of Internet on 199; psychodynamic approaches 23; reasoned action theory 150; terrorist belief systems and 25; see also behavioural motivation Human Development Index (HDI) 238 Human Rights Council 92 human rights education (HRE) 421 Hunter, A. 198 Huq, A. 300 – 301

habitus 72 Hafez, M. 40 Hafez, M. M. 241 Hain, S. 24 al-Halabi, A. H. 452 Halle synagogue shooting (2019) 213, 223 – 224, 505 Hamas 159, 410 Hamm, M. S. 218 Hanau Germany attack (2020) 213 – 214, 221 Hanson, R. K. 481 – 482 Harbinson, H. J. 182 Hardy, K. 6, 248, 251, 355 Hare Krishna 139 Harris-Hogan, S. 358, 363 Hart, S. D. 465 Hasisi, B. 119 Hassan, G. 324 hate 76 – 77 Hate Mapper tool 93

Ibrahim, N. 452 Identifying Vulnerable People (IVP) 464, 477, 490 identity: biological kinship 89; cognitive drift and 140; compensatory 170; escalation and 102; online communities and 197; personally transformative experiences 89; personal self and 88 – 89; radicalisation and 86,

551

Index 102; resocialisation and deconditioning 139, 141; social group and 87 – 89, 102; transformation of 139 – 140; uncertainty and 87 identity crises 92, 137, 167 identity fusion: defining 77; echo chamber effect and 81; group identity and 77, 85, 88 – 89; linguistic markers for 93 – 94; pathways into 90, 91, 91; psychological risk factors for 93 – 94; self-sacrifice and 88 – 89, 90, 94; violent extremism and 91, 94 identity politics 47, 85 identity theory 23 ideological radicalisation: bridging narratives in 264 – 265; commitment to beliefs 151; deradicalisation and 446 – 447, 456n1; disengagement and 452; extremist milieus and 264 – 265; groups and 75 – 77, 81; ideological encapsulation and 102; individuals and 74 – 75, 81; isolated indoctrination and 76; organisational compartmentalisation and 102; P/CVE strategies and 470; personality traits and 264 – 265; political violence and 3, 65 – 66; role in violent extremism 71, 74, 85 – 86, 166, 263 – 265; system level 77 – 81; technology and 66; weaponization of cultural biases 73 – 74 ideology: calculated propaganda and 73; counternarratives and 280; counter radicalisation policy and 279 – 280; cultural context for 72 – 74; defining 72 – 73; doubt in 268; echo chamber effect 76 – 77; group radicalisation and 75 – 76; individual radicalisation and 37, 74 – 75; legitimizing 71, 74 – 75; master narratives and 73 – 74, 80; mobilisation of action and 73; persuasion and 151; quest for significance 74 – 75, 87 – 88; religious conversions and 138; social components of 75; strategic influence and 80; terrorist conceptions of 75; violent extremism and 71 Ilyas, M. 338 – 340, 342 – 343, 347 imagined communities 197 – 198, 294 – 295 immigrant communities 35 – 37, 39 IMPACT Europe project 330 Imron, A. 449 – 450 incel groups 170 – 172 INDEED project 330 Inderberg, A. M. S. 186 individual radicalisation: at-risk population and 463; availability and 104; deradicalisation and 447 – 448; emotional impact and 75, 81; four-stage ideological development 39; gender and 166 – 167; ideology and 37, 74; jihadist 103; legitimizing ideology and 71, 74 – 75; limited understanding of ideologies

74 – 75; mass publicity and 58; online communities and 196 – 197, 208; personal crises and 62, 103 – 104; personality traits and 187, 189n11, 189n12; personal trajectories and 36, 48, 58, 105 – 106; political assassination 53 – 63; pre-violence 186 – 187; psychopathology and 186; quest for significance 74 – 75; recruitment and 21, 36; religious conversions and 138 – 139, 186; research in 3 – 4; risk of apprehension 117; school-related violence and 415; search for hidden conspiracy 57 – 60, 63; social profiles of 62; social ties and 40, 103 – 104; terrorism studies and 20 – 21; unfreezing and 104; violent causes and 186, 267 Indonesia: civil society initiatives in 454 – 456; counter narratives and 439; democracy in 446; Densus 88 449 – 451, 453; deradicalisation and 255, 446 – 456; disengagement and 446, 448 – 456; government-initiated interventions 449; ISIS reintegration strategies 455; masculine radical identity in 172; military involvement in counterterrorism 451 – 453; as Muslim nation 446; NGOs and 343; police counter-terrorism 449 – 451, 453; prison radicalisation in 71; reintegration programming and 449 – 451; state-civil society initiative 455; Western knowledge in 343, 347 Indonesian Alliance for Peace (AIDA) 454 Indonesian National Counter-Terrorism Agency (BNPT) 451 – 453 Information Centre on Radicalisation (Amsterdam) 279 Ingram, H. J. 87 inoculation theory 436 – 437, 439 Instagram 505, 509 Institute for International Peace Building (YPP) 454 – 455 Institute for Strategic Dialogue 93, 432 institutional theory approach 370, 375 – 379 Inter-Governmental Authority on Development (IGAD) 537 Internal Intelligence and Security Agency (AISI) 278 International Centre for the Study of Radicalisation and Political Violence (ICSR) 37, 92 International Crisis Group 449 international relations (IRs) 385 – 386, 389 Internet: community and 10, 198, 200; extremist networking and 139, 204, 223; gender dynamics and 167; ideological conflicts and 66; impact on human behaviour 199; impact on terrorism 121, 196; mobilisation of foreign fighters and 241; radicalisation

552

Index research and 44; radical milieus and 202 – 203; social influence and 151; social interactions and 223; see also online communities Internet Referral Units (IRUs) 503 internment 118, 124n5 intersectionality 6, 166, 386 Iraq: American invasion of 35, 39 – 40; Jihadist groups and 231; mobilisation of foreign fighters for 40, 134 – 135, 196, 236, 241 Iraqi-Kurdish Peshmerga 233 Irish Republican Army (IRA) 30n3, 139 Irish Republican Movement 106 Iron March Forum 205 IS see Islamic State (IS) Islamic Courts Union (ICU) 531 Islamic State (IS): Al-Shabaab and 531; counter narrative campaigns against 442; East Africa and 530 – 531; female support for 233; foreign fighters and 74, 135, 196, 241; gender-based violence and 171; gender binary and 171 – 172; lone-actor terrorism and 218; masculinities and 171; mobilisation 46, 71; online communities and 202 – 203, 205 – 207, 432; pre-crime offenses and 409; propaganda 106, 135, 169, 196, 207, 241; proscription of 407; reintegration and 448, 455; social media use 203 – 204, 206 – 207; Syrian civil war and 231; system level of analysis 77; video propaganda and 341 – 342; violent radicalisation and 536; Western discourse on 341; Western European women and 165 Islamist terrorism: antiscientific worldview 264; East African 529; ideological radicalisation and 81 – 82, 85, 264; P/CVE programmes and 532 – 533; P/CVE research in 8, 267 – 268, 270 – 271; quest for significance 74 – 75; radicalisation and 43 – 44, 267 – 268; Salafi-Jihadi ideology and 71; transnational networks and 79 – 80; see also Jihadi terrorism Islamophobia 172, 300, 341, 421 Ismail, N. H. 172, 454 – 455 Israeli War of Independence (1947–9) 234 Italy: counter radicalisation policy 277 – 278, 280 – 281, 286, 288n2; far-left terrorist groups in 26, 277; far-right extremists in 218, 221, 277; foreign fighters from 277; leftist movements in 101; lone-actor terrorism in 216 Izard, C. E. 157

James I 56 James II 57 January 6, 2021 riot 441 Jarvis, L. 247 Jasper, J. M. 108n1 Jasser, G. 507 Jáuregui, J. 56 Jaysh al-Ayman 531 Jemaah Islamiyah (JI) 449 – 450 Jerusalem 116 – 117 Jihad Group 452 Jihadist extremist tendencies (EMI) 489 – 490 Jihadist groups: criminal subcultures and 118, 132 – 133; deplatforming and 507 – 508; East Africa and 530 – 531; foreign fighters and 231, 234, 237, 240; gender in 171 – 172; masculinities in 171; online content moderation and 507 – 508; protest masculinities in 170, 172; religious conversion and 133, 138 – 139, 141; social ties and 249; women and 172 Jihadi terrorism: academic research in 40; class background and 134 – 135; deprivation theories 134 – 135; foreign fighters and 40, 134 – 135; gender roles in 166, 171; as generational revolt 44; ideological radicalisation and 85 – 86; individual radicalisation and 103; invasion of Iraq and 35, 40; Islamist insurgency and 44; leaderless resistance and 215; lone-actor terrorists and 218; mobilising narratives and 40; moral outrage and 40; Muslim men and 164 – 165; online communities and 8, 203; personal discrimination and 40; radicalisation concept and 39; radicalisation process 92; religiosity and 133; social ties and 35, 40, 167; transnational networks and 103; warrior masculinities and 171 – 172; see also Islamist terrorism Johansen, M. L. 374, 376 Jones, D. A. 138 Jones, J. W. 73 Jonkman, H. 425 Joosse, P. 437 Jørgensen, M. 226n1 Juergensmeyer, M. 27, 72 – 73, 78, 254 Kaczynski, T. 214, 216 Kamau, W. C. 535 Karnavian, T. 449 – 450 Kautsky, K. 64 Kedward, R. 62 Kennedy, R. 64 Kenney, M. 447 Kenya: Aarhus model and 376, 534; Al-Shabaab and 106, 520, 522, 533, 535 – 536; counter

Jabarah, M. M. 76 Jabhat al-Nusra 231 Jabhat Fateh al-Sham 231 James, M. 268

553

Index radicalisation in 521 – 522, 532 – 536; counterterrorism and 520 – 522; county action plans (CAPs) 520, 533 – 534; County Security and Intelligence Committees (CSIC) 521; domestic counter-radicalisation objectives 519 – 522; exclusion of women from counter-radicalisation policies 522; as fragile state 520; global security and development priorities 521 – 522; indigenous PVR strategy in 343; Islamist extremists in 533, 535 – 536; local understandings of extremism in 535; Mau Mau rebellion in 342, 530; Mombasa Republican Council (MRC) 532, 535 – 536; National Counter Terrorism Centre (NCTC) 533; National Strategy to Counter Violent Extremism (NSCVE) 520 – 521, 533; non-state actors and 521 – 522; Operation Usalama Watch 532; Prevention of Terrorism Act 2012 521 – 522, 533; security governance in 520 – 522; security strategies in 342; soft approach to terrorism in 533; STRIVE II programme 325, 328; violent extremism in 520, 533, 535; weak governance and 514; Western approaches to counterterrorism in 533 Kepel, G. 44, 85 – 86 Kessler, R. 183 Khalil, J. 141, 267, 489 Khan, R. M. 341 Kidd, C. 53, 56 Kilcullen, D. 361 Kimmel, M. 170 King, E. 218 Kitt, B. 226n1 Klausen, J. 138, 167 Klor, E. F. 135, 238 Knight, S. 186 Knudsen, R. A. 384 Knutson, J. N. 22 Koehler, D. 46, 262, 264, 267, 269, 447 Koskenniemi, M. 387 Kotzebue, A. von 60 Kouachi brothers 284 Krueger, A. B. 134 Kruglanski, A. 5, 87, 94, 265, 423, 425 – 426, 447 Kühle, L. 257 Kuipers, P. 157 Kullmann, E. 61 Kundnani, A. 41, 257, 338 – 339 Kurtz, C. F. 78 Kutner, S. 324

Lazarus, B. N. 157 Lazarus, R. S. 157 League of Nations 395n5 learning perspectives 119 Lebanon 377, 462 Lee, B. 150, 431 Leerssen, P. 500 Leidig, E. 173 Levine, S. V. 136, 139 Lewis, J. 326 life-course research 119, 122, 186, 188 Lincoln, A. 60 Lincoln Brigade 236 Lindekilde, L. 108n3, 188, 213, 251, 257, 326, 373, 467 linkage concept 87 Lloyd, M. 484 Lofland, J. 138, 141 Logan, C. 484 Lombroso, C. 63 London 7/7 bombings (2005): EU counterterrorism and 392; immigrant communities and 36; international counterterrorism strategies 400; P/CVE and 262, 278; pre-crime offenses and 403; radicalisation concept and 37; radicalisation research and 113, 266; UK Prevent strategy and 401 London Metropolitan Police Service 503 Lone Actor Radicalisation and Terrorism (LART) dataset 214, 216, 218, 220 lone-actor terrorism: attack types 218 – 219, 219; bladed weapon attacks 222; characteristics of 213; chronological distribution 217; conspiracy theories and 221 – 222; copycatting attacks 222; empirical overview of 216; encouragement cues 221; far-right and 213, 218, 222, 224 – 225; geographic distribution 216, 226n2; ideological distribution 217, 218; ideological milieus 214; ideology and 66, 80; interrupted plots and 219 – 220; Islamist extremism and 218, 222; LART dataset and 214, 216, 218, 220, 226n3; legitimising of far-right discourse and 220 – 221; lifecourse perspectives 186; live-streaming and 223 – 224; mental illness and 136, 183, 215, 393; online communities and 216, 222 – 224; personality disorders and 184; political assassination and 53 – 64; political violence and 215 – 216; psychopathology and 184; radicalisation process 107, 108n3, 214 – 215, 222 – 225; radical milieus and 213 – 215; research in 215, 225; run-over attacks 222; self-narrativisation of attacks 224 – 225; social movement theory and 107; social ties and 214; trends in 220

LaFree, G. 113, 117, 119 – 120 laïcité 277, 280 – 281, 284, 288n1 Laqueur, W. 249 Lavin, T. 80

554

Index Lord’s Resistance Army (LRA) 530, 532 Loughall incidents 118, 124n5 Louis, W. R. 94 Louis XV 57 Luhrmann, T. 140 Lum, C. 120 Lydston, G. F. 64 Lygren, S. 106 Lyons, H. A. 182

plurality of 170; power and 170; protest 170 – 172; radicalisation process and 86; warrior 171 – 173; see also gender mass casualty attacks: anarchists and 61 – 62; 9/11 attacks 2; personality traits and 184; radicalised avengers and 64 – 65; terrorist psychopathology and 181 mass radicalisation 3 Mattheis, A. A. 172 Mau Mau rebellion 342, 530 Mauritania 517 Mauss, M. 254 Mazerolle, L. 370 – 371, 379 – 380 McAdam, D. 100 McCauley, C. 40, 74, 76, 88, 104 – 105, 141, 167, 264 McCulloch, J. 404 McDaid, S. 248 McDonald, K. 107 McDonald, S. 222 McDowall, D. 118 McKay, H. D. 120 McKinley, W. 63 McMillan, D. 199 – 200 Meiering, D. 264 Meleagrou-Hitchens, A. 135 Melucci, A. 103 Memmi, D. 200 – 201 mental illness: anti-cult movement and 136; behavioural radicalisation and 361; diagnostics and 183; ideological radicalisation and 75; life-course perspectives 188; lone-actor terrorists and 136, 183 – 184, 215, 393; NRMs and 136 – 137; prevalence and 182 – 183; terrorists and 29 – 30, 120, 180 – 185, 361 Merah, M. 280 Merari, A. 184 Merkl, P. H. 23 Merton, R. K. 118 Messerschmidt, J. W. 170 Metternich, Prince 60 micro-mobilisation 99, 101, 103 – 105 micro-radicalisation: anarchist radicalisation 61 – 64; assassination and 54 – 62; communication and 66; democratisation and 66; environment for 61; historical recurrence of 53, 65 – 66; ideology and 65 – 66; lone actors and 66; radicalised avengers and 64 – 65 Microsoft 500, 504 Middle East and North Africa (MENA) 344, 345, 346, 347 – 348, 516 Middle East Media Research Institute (MEMRI) 502 Mignolo, W. 340

Maaßen, H.-G. 213 Macdonald, C. 64 Macklin, G. 4 Macron, E. 221 Madriaza, P. 338 Madrid train bombings (2004) 36, 39, 113, 400 Maglia, C. 395n1 Mai Mai 530 MAIN (modality, agency, interactivity, and navigability) model 437 mainstreaming: conceptualisation of 385 – 388; of counter-terrorism through counterradicalisation 385 – 395; critical discourse analysis and 385, 389; elite co-option of 389; encroachment on individual liberties 388; EU discourse and 389 – 390, 392 – 395; funding for 388 – 389; gender and 385 – 387; international relations (IRs) and 385 – 386, 389; UK Prevent strategy and 401 – 402; UN discourse and 389 – 392, 394 – 395 Mair, T. 75 Majozi, N. 341 Malaysia 46, 448, 456 Malet, D. 231 – 232, 234 Mali 326, 517 Malkki, L. 1, 19 Maloney, P. 202 Malthaner, S. 29, 99, 104, 108n3, 167, 213 Maniscalco, M. L. 288n2 Mann, L. 447 Marat, J.-P. 59 Marchment, Z. 375, 476 Marion, R. 80 Marret, J. L. 249 Marsden, S. V. 1, 106, 327, 418 Martini, A. 306, 533 martyrdom 89, 91 masculinist groups: far-right and 166, 171; jihadi terrorism and 166, 171; misogyny and 170 – 171; power and 165; radicalisation process 165; violence and 170 masculinity: anti-colonialism and 172; emasculation and 170 – 171; gendered approaches to 164, 169; hegemonic 170 – 171; hierarchies of 165; IS and 171; manhood and 171;

555

Index Millennium Development Goals (MDGs) 391 Miller, C. 119 Miller, V. 119 Miller-Idriss, C. 80 Misiak, B. 265 misogyny 165, 168, 171 mobilisation: Al-Shabaab and 106; cycles of protest in 102, 537; defining 28; foreign fighters and 40, 44 – 46, 134 – 135, 234 – 235, 239 – 242; jihadists and 134, 196; micro-mobilisation 99, 101, 103 – 105; motivational factors for 92; political 102, 107; political violence and 10, 116; protest movements and 99 – 100; radicalisation process 165, 168; resource 100 – 101; rhetoric of 234; social 309; social ties and 40, 43; women and 165, 169; youth anger and 39 model of conversion 138 Moghaddam, F. M. 39, 87, 447, 489 Mombasa Republican Council (MRC) 532, 535 – 536 Monaghan, J. 120, 124n6 Moonshot 439 Moore, C. 233 Moral Disengagement Theory 94 Morin, D. 338 Morocco 448 Morris, N. A. 120 Morrison, J. 437 Morrison, J. F. 239 Morton-Bourgon, K. E. 481 – 482 Moskalenko, S. 40, 74, 76, 88, 104 – 105, 141, 167, 264 Mousnier, R. 54 Movement party 530 Mubarak, H. 530 Mujahidin KOMPAK 453 Müller, H. 65 Mullins, C. 40 multi-agency (MA) approaches: Aarhus model 372 – 374, 376 – 377; appropriateness of 379 – 380; Channel programme 92, 372, 374 – 375, 377; collaboration in 370, 376, 379; counter radicalisation policy and 281, 369; crime prevention and 281, 369 – 371; CVE and 361 – 363, 369 – 380; effectiveness of 371 – 372, 375; foundations of 370 – 371; hybridity and 378 – 379; institutional logics and 378 – 379; institutional theory approach to 370, 375 – 379; intelligence sharing and intervention 371 – 372, 374 – 375, 377 – 378; Kunskapshus (Knowledge-houses) 377; locally oriented public agencies and 369 – 370; objectives of 369, 371; organisational capacity and 371; police

and 371, 373; police power relations in 373, 375; public health model and 357; School, Social Services, and Police (SSP) model 372 – 373, 376; strengthening society 372 – 373; translation processes in 377 Multi-attribute Utility Technology (MAUT) 327 multifinality 185 Multi-Level Guidelines (MLG) 465, 477 – 478, 490 Murphy, C. 404 Musk, E. 503, 508 Muslim communities: Arab-Israeli relationships and 530; counter-narratives and 280; foreign fighters and 135, 205 – 206; global religious community and 198; hearts and minds strategies 279; identity politics and 47, 85; love-jihad conspiracies 222; moral outrage and 35, 40; P/CVE engagement with 293 – 294, 298 – 303, 534; police surveillance of 295, 299, 402, 463; racialisation of 409; radicalisation and 35, 45, 249, 279, 536 – 537; responsibilisation of 294, 298 – 299; stigmatising of 45, 164 – 165, 173, 205, 302, 339, 359, 375, 400, 537; SunniShia rifts 530; as suspect communities 294, 299, 359; UK Prevent strategy and 295, 298, 300 – 301, 400 – 401, 463 Muslim nations 446, 448 Muslim youth: alienation of 45, 338; anger and frustration of 35 – 36, 39 – 40; Cronulla riots and 205; jihadism and 134, 137; mobilisation as foreign fighters 39, 44 – 45, 205; securitisation of 375, 534; stigmatising of 375 Mwangi, O. G. 514 al Najjar, H. 452 narratives 432 – 433, 436, 441; see also counter narratives Nasser, K. 373, 376 National Anti-Mafia and Counter-terrorism Prosecution Office 278 National Health Service (NHS) 188 National Socialist Underground 213 National Strategy to Counter Violent Extremism (NSCVE) 520 – 521, 533 Nawaz, M. 284 Nazi stormtroopers 266 Neckbeard Deathcamp 440 Nesser, P. 74 Netherlands: counter radicalisation policy 277 – 282, 286, 288n4; foreign fighter recruitment and mobilisation 35, 135, 241; ideologyoriented approaches 280; interventions for at-risk individuals 466; lone-actor terrorism 216; multicultural state policies 295; P/ CVE evaluation in 330; PVE-E policies in

556

Index 419 – 420; reintegration programmes in 328; resilience training in 326; Salafi Jihadism in 279; society-oriented model 278; Spanish Civil War volunteers 235 – 236; Waffen-SS volunteers 240 Network Enforcement Act (NetzDG) 500 – 501 Neumann, P. R. 2, 37, 41, 92, 105, 168, 248, 250 Neville, E. 56 New Left Terrorism 8 – 9 new religious movements (NRMs): anti-cult movement 136, 140; class background and 134 – 135; converts to 133 – 143; deprivation theories 134 – 137; failure of prophecy and 133; foreign fighters 134 – 136; identity crises and 137; identity experimentation and 139; psychic deprivation and 136; psychological disorders and 136 – 137; radicalisation and 133; relative deprivation and 135 – 136; Salafi Jihadism and 133; social ties and 139; see also religious conversions NGOs: counter narratives and 280, 431, 435, 438, 440, 442; counter radicalisation and 281, 310, 421; dependence on Western funding 343; funding schemas and 311; P/CVE and 270; reintegration assistance and 455; statecivil society initiative and 455; as trusted flaggers 502 – 503, 508 Nice attacks (2016) 222 Nigeria 252 9/11 attacks: jihadist profiles and 249; protective measures after 117; radicalisation concept and 34; radicalisation research and 24, 113; root causes rhetoric 41; UN Security Council resolution and 390 Nivette, A. 187 Njoku, E. 533 Nobiling, C. 62 non-violent radicalism 249, 255, 266, 464 Norris, J. J. 120 Northern Ireland 117, 182, 255, 277, 299, 400 Norway 358, 379 NRMs see new religious movements (NRMs)

counter narratives and 207, 432; defining 198; democratic rationalization in 198; deplatforming and 507 – 508; embattled subgroups in 204; ethnography and 199; far-right and 8, 203 – 205, 432; free spaces and 203, 205; identity and 197; ideological indoctrination and 74; individual radicalisation and 10, 196 – 197, 208; information exchange 201; interactivity and 201, 208; Internet Referral Units (IRUs) 503; interpersonal trust and 197, 201; ISIS supporters and 202 – 203, 205 – 207; Jihadist 8, 203, 432; lone-actor terrorists and 216, 222 – 224; membership in 198 – 199; netizenship and social control 201 – 202; networked publics in 199; occurrence in a public space 201 – 202; offline events and 205, 222 – 223; personal concern 201 – 202, 208; policing of conduct 204; radical milieus and 104, 202 – 203, 224; research methodologies for 199, 205 – 208; sense of community in 199 – 200, 204, 207; social ties and 223; stability of membership 201 – 202; stable identities and 201 – 202; transnational power and 78; victimhood narratives and 507; violent extremism and 8, 196, 202 – 208; white supremacist groups and 203 – 204; women and 169; youth and 10; see also content moderation; social media online radicalisation (OVEST) 489 – 490 On War (Clausewitz) 81 Operation Entebbe 530 Operation Linda Boni 532 Operation Motorman 118, 124n5 Operation Usalama Watch 532 organisational radicalisation 309 Orientalism 339, 341, 409 OSCE (Organization for Security and Co-operation in Europe) 92 O’Shea, L. E. 483 Ottanelli, M. 62 Pakistan 448 – 449, 454, 456 Palestinians 530 Papacharissi, Z. 198 Papale, S. 106 Paris terrorist attacks (2015) 44, 113 Parker, D. 326, 378 Parnham, D. 220 Parry, W. 54, 56 Passanante, G. 62 – 63 Passenger Name Record (PNR) 393 Paty, S. 415 Pauwels, J. R. 120 Pawella, J. 288n5

Oando, S. 342 – 343 Obama administration 45 O’Connor, F. 108n3, 213 O’Donnell, A. 375 O’Dwyer, M. 64 Oliver, J. 442 Olver, M. E. 482 – 483 online communities: anti-Muslim sentiment and 205, 509; Baqiya family 206 – 207; boundary maintenance in 204, 206 – 207; characteristics of 200 – 201; collective identity and 205; complexity theory and 78;

557

Index Pearlstein, R. 24 Pearson, E. 164, 204, 507 Pedagogy of the Oppressed (Freire) 424 Percy, T. 65 Pernicone, N. 62 Perry, S. 116 – 117 personal crises: assassination attempts and 55 – 56, 58 – 60; individual radicalisation and 62, 103 – 104 personality: defining 189n1; diagnostics and 183 – 184, 189n6, 189n13; far right and 264; ideological radicalisation and 264 – 265; individual radicalisation and 186 – 187; political attitudes and 265; pro-violence and 187; radicalisation process and 182; violent extremism and 265 – 266 personality disorders 183 – 184, 189n1 persuasion: anger in 159; behavioural modification and 151; central routes and 436; counter narratives and 433, 436 – 437; defining 151; Elaboration Likelihood model 436; extremist narratives and 436; fear appeals and 160; hope and 161; inoculation theory and 436 – 437; peripheral processing and 436; reasoned action theory 155 – 156; social influence and 151 – 152 Petliura, S. 64 Pettinger, T. 402 Pettyjohn, S. 447 Philip II 54 – 55, 57 Philippines 519 Pickering, S. 404 Pilkington, H. 107 Pink, S. 199 Pisoiu, D. 4, 24, 141 Pistone, I. 371 Piven, F. F. 100 Pokalova, E. 238 police: community counterterrorism and 301; community policing and 300; detection of at-risk individuals 462 – 463; escalating policing 102; Indonesian counter-terrorism and 449 – 451; multi-agency approaches and 371, 373 – 375; power relations and 373, 375; social care logics 379; surveillance of Muslim communities 295, 299, 402; terrorist group confrontations with 24, 28 political assassination: anarchist radicalisation 61 – 63; historical moment and 65; lone actors 53 – 64; mass publicity and 56, 58, 66; micro-radicalisation 54 – 62; personal crises and 55 – 56, 58 – 60; radicalised avengers and 64 – 65; religious militancy and 54 – 55, 61; search for hidden conspiracy 57 – 60, 63; self-sacrificial 55, 59 – 60; see also assassination

political attitudes 265 political ideology: cognitive radicalisation and 26 – 27; far-right and 71; mass publicity and 62; misogyny as 165, 168, 171; radicalisation process and 37 – 38, 42 – 43; self-radicalisation and 37 – 39; terrorist groups and 26 – 27; violence and 65 – 66; see also ideological radicalisation Political Process approach 100 – 101, 104, 106 – 107 political violence: anger in 159; contextual factors for 100; gendered approaches to 171; ideological radicalisation and 3; lone-actor terrorists and 215 – 216; radicalisation and 276; radical milieus and 104; rational choice theory 117; relational dynamics and 105 – 106; research in 113 – 114; restraint in 8 – 9; social movement theory and 99 – 107; transnational networks and 103 Post, J. M. 23 postcolonial studies 339, 349n3 Postill, J. 199 Poway, California attack (2019) 79, 82n2, 223 Precht, T. 489 pre-crime offenses: as counter-law 408; counter radicalisation and 400 – 401; counterterrorism legislation and 5, 399; critical race theory and 409; criticism of 407 – 410; defining 411n1; encouragement of terrorism 406 – 408; EU countries and 399 – 400; expansion of 403 – 405; indoctrination and 400; infringement on human rights 408 – 409; legal regulation of terrorism and 399, 405, 410 – 411; possession of terrorist information 405 – 406; post-London bombings 403; preparation for terrorism 406; Prevent Duty and 409 – 410; proscribed organisations 407, 410; risk assessment and 461 – 462, 476 – 477; social transformation and 403; topography of 402; UK Prevent strategy and 401 – 402, 409 – 410 preventative justice 410 preventing and countering violent extremism (P/ CVE): backlash outcomes 117 – 118; campaigns against 299; civil society organisations (CSOs) and 532, 534 – 535; community-oriented approaches 292 – 303, 355; community resilience and 296 – 298, 303n1; counterterrorism policy and 262; development agendas and 537; Do No Harm strategies in 537; donor challenges in 529, 532 – 536; East African 321, 528 – 529, 532 – 537; effectiveness of 9, 47, 303, 320 – 323, 327 – 329; elitist perspectives in 528, 533, 535 – 536; environment for radicalisation and 34; European countries

558

Index and 92, 276 – 277, 279 – 282; exit pathways 268 – 269; far-left extremism and 279; far-right extremism and 279; foreign development aid and 532; fragile states and 514; gender mainstreaming and 386; Germany and 270 – 273; goals of 322; heteronomisation and 284 – 285; ideological radicalisation and 470; individual radicalisation and 3; international discourse and 387 – 388; leadership in 301; local contexts for 535 – 536; mainstreaming and 394; MAPEX project 271; means-based 300; methods and tools for 269 – 273; models for 267; multi-dimensional risk assessment 93; online content moderation 10; opportunity-reducing techniques 117; policy and practice 91 – 93, 277, 462 – 463; prediction models for 93; pre-emptive/ pre-crime offenses 399; public health model 5 – 6, 323; regional trends in 537; responsibilisation and 294, 298 – 299; societal factors and 266; state-community relationships 292, 298 – 299; training programmes 463; UN discourse and 391; values-based 300 – 301; see also Countering Violent Extremism (CVE); counter radicalisation policy; evaluation of P/CVE preventing violent extremism (PVE) 34, 45, 349n1, 415 preventing violent extremism in education (PVE-E): approaches to 416, 421; awareness-raising practices 420 – 421, 424; counter radicalisation policy 418 – 419; defining radicalisation in 418; history education and 424 – 425; individual practices 420 – 421; national strategies for 419 – 420; pedagogy of 415 – 418, 420, 423 – 427; primary/secondary target 422; teacher role in 421 – 422, 424 – 425; youth radicalisation and 417 – 418, 425 – 426 preventing violent radicalisation (PVR): coloniality and 339, 342 – 345, 345, 346, 347 – 349; conceptualisation of 338, 349n1; decolonial approach to 340 – 343; evaluation of 343 – 344; Global North perspectives in 338 – 339; indigenous strategies for 343, 348; Western funding for 343; Western production of knowledge 343 – 344, 347 – 348; see also preventing and countering violent extremism (P/CVE) preventive justice approach 359 – 360 pride 158, 160 – 162 prison 118 – 119, 285 Proactive Integrated Support Model (PRISM) 327 Profiles of Individual Radicalization in the United States (PIRUS) 268

propaganda: calculated 73; counter narratives and 440 – 441; ISIS and 106, 135, 169, 196, 207, 241, 341 – 342; radicalisation process 151; social influence and 150; strategic communication and 150; targeted messaging and 155; victimisation tactics 159 protest masculinities 170 – 172 protest movements: competitive escalation 102; conceptualisation of 100; cycles of protest in 101 – 102; escalating policing and 102; free/safe spaces 102 – 103; militant underground organisations and 99, 101 – 102, 104; movement entrepreneurs and 100; non-violent repertoires 99, 104; political context and 100 – 102; violence and 101 psychopathology: criminal behaviour and 189n3; defining 180; development of 185 – 186; diagnostics and 183 – 185, 189n6; extreme violence and 184, 186; life-course perspectives 186, 188; lone-actor terrorists and 184; prevalence and 184; radicalisation process and 137, 181 – 182, 185 – 188, 266; radicalisation studies and 44; taxonomic research and 184 – 185; terrorism studies and 21 – 22, 184; terrorists and 21 – 22, 180 – 189; traits of 182 public health model: at-risk population and 360, 461; benefits of 358 – 360, 365; categorisation of audiences 434; community resilience and 297; counter radicalisation and 5 – 6, 323; CVE framework for 355 – 365; CVE outcomes 361 – 362; discourse around disease in 356; ethical issues and 359 – 361; evaluation of 362; health promotion approach 6, 356, 363 – 364; medicalised language and 360 – 361; multi-agency approaches to 357; multidisciplinary approach and 357, 361 – 362; population surveillance in 360; prevention pyramid 357, 357, 358 – 359, 363 – 364; preventive justice approach 359 – 360; social determinants of extremism in 359 push and pull factors 86, 166, 237 – 238, 268, 448 Pyrooz, D. C. 120 Python, A. 93 QAnon 201, 205, 221 Quayle, E. 24 Quest for Significance Model 87 – 88, 94 Quijano, A. 339 – 340 Quilliam Foundation 284, 432 Qutb 171 Rabasa, A. 447 Rabbie, J. M. 30n2

559

Index radical action: behavioural changes and 141; embodied practices 107; free/safe spaces 102 – 103; historical moment and 48; radical ideas and 40, 43, 74, 152; terrorist activity and 88; Two Pyramids model and 88 radical ideology: conveyor belt model 39 – 40, 42 – 43; group mobilisation and 166; men’s friendship groups and 167; nonviolent principles and 4; personality traits and 265 – 266; policing of 256 – 257; psychopathology and 187; radical action and 40, 43, 74, 152, 249, 253 – 254; state intervention and 249 – 250; Two Pyramids model and 88; unwillingness to compromise and 423 – 424, 427; see also ideological radicalisation radicalisation: conceptualisation of 2, 5, 19 – 21, 29, 34 – 35, 47 – 48, 82n1; defining 41 – 42, 92, 124n2, 164, 266 – 267; as ideological problem 277, 279 – 280; laïcité and 277, 280 – 281, 288n1; as law-enforcement problem 277 – 278; models and metaphors 5, 39, 74, 87 – 88, 166, 267, 489 – 490; as political construct 34, 47; psychological studies on 85 – 88, 92; relational approach to 27 – 28; as social problem 277 – 279; see also gendered radicalisation; ideological radicalisation; individual radicalisation Radicalisation Awareness Network (RAN): education strategies and 419; EU and 46, 92, 296, 310, 318n1, 330, 392 – 393, 396n7; Inspiring Practices database 433 radicalisation policy: at-risk population and 462 – 463; conveyor belt model 91 – 92; fusion-plus-threat model 91; gender and 6, 164 – 165, 173 – 174; internment and 124n5; mainstream approaches 165, 173, 174n2; Muslim men securitisation and 164 – 165; Muslim women allies and 165; prevention and 45; public health model 6; situational crime theory 116 – 117; violent jihadism and 71 radicalisation process: behavioural changes and 141; bodily transformation 254; bunch of guys model 28, 40, 167, 249; conceptualisation of 248 – 249; contextual factors for 37 – 38, 42 – 43, 48, 92, 105 – 106, 272; equifinality and 185; foreign fighters and 232, 234 – 235, 239 – 240; group radicalisation 3, 105; ideology and 37 – 38, 42 – 43, 48; individual level 3, 103; mechanisms for 102, 108n2; mobilisation dynamics and 48; multifinality and 185; narrative justifications in 423; need for significance in 423, 426; pedagogical view of 425 – 426; personality traits and

182; persuasion and 151 – 152; phases of 92, 151; preventive work and 37 – 38; prison and 118 – 119; psychopathology and 137, 181 – 182, 185 – 188, 266; relative understandings of 254 – 255; religious conversions and 137 – 142; resocialisation and deconditioning 139, 141; role of communication in 437; social influence and 151, 293; socialisation in 141, 214, 224, 237, 267, 437, 489; social ties and 42 – 43, 48, 103 – 104, 167, 423; subcultures and 141 radicalisation research: cognitive versus behavioural 3 – 4; conceptual issues in 2; context versus ideology in 37 – 38, 42 – 43, 48; counterfactuals 8, 256, 462; counter radicalisation policies and 34; data collection challenges 312 – 313; empirical studies 7 – 10; ethical issues in 10 – 11; evidence-based approaches 86; far-right 8, 263, 339; foreign fighters and 40, 44; gendered approaches to 168, 173 – 174; individual trajectories and 42, 48; institutionalisation of 317; Islamist ideologies in 266 – 267; jihadi terrorism and 40, 44; meaningful comparison groups 8; multidisciplinarity of 1, 309; personal/ group identities 86; policy-making impact of 315 – 317; policy practitioners and 11; process of organisational radicalisation in 309; psychological mechanisms 266 – 267; re-pedagogization of 427; researcher position in 306 – 309; researcher safety and 11; research object in 307 – 309; social bonds and 42 – 43, 48; social movement theory and 103, 106 – 108; state engagement and 315 – 317; violent jihadist causes 85 – 86; Western-centric 338 – 339 Radicalise (game) 439 radical milieus: conspiracy theories and 221; disillusionment with 268; lone-actor terrorists and 213 – 215; online communities and 197, 202 – 203, 224; personality traits and 265; political violence and 104; shared ideology and 264; violent extremism and 167 Ragazzi, F. 9, 276, 288n1, 288n3, 288n4, 296, 299, 370, 418, 420 Ramakrishna, K. 71 Rambo, L. R. 138 Ramsay, G. 442 RAN (Radicalisation Awareness Network) 92 Ranstorp, M. 92 Rapanyane, M. B. 349n4 Rath, E. vom 64 rational choice theory 117 – 118, 124n6 Ravachol, F. 62 Ravaillac, F. 55 – 57, 66

560

Index Ravndal, J. A. 106 Razzaque, R. 76 – 77 Realist framework 329 reasoned action theory: background and contextual factors 152 – 153; behavioural beliefs and 154; behavioural intentions 155 – 156; behavioural motivation and 150 – 156; cognitive-based perspectives and 150 – 153; control beliefs and 154; defining 152; injunctive norms and 154; normative beliefs and perceived norms 154; predictors of behavioural intentions 153 – 155; system of relationships in 152, 153; targeted messaging and 155 – 156 recidivism: defining 487, 496; deradicalisation and 256, 324; life-course perspectives and 119; prediction instruments 478, 479 – 480, 481 – 483, 492n3; reintegration programming and 448; research in 496 – 497, 497; risk assessment and management 476, 492, 496; risk factors for 481 – 482, 485 – 488, 496; sexual violence and 488; terrorism rates 119 reciprocal intergroup radicalization (RIR) 106 reciprocal radicalisation 4 recruitment: Al-Qaeda 35, 139; Al-Shabaab and 106, 535 – 536; content moderation and 504; contextual factors for 37 – 38; far-right and 161; foreign fighters and 234 – 235, 240 – 241; gamification and 439; group radicalisation and 3; historical examples of 240 – 241; ideological radicalisation and 39; individual radicalisation and 21, 36; Jihadist 432, 442, 531, 536; social ties and 28, 101, 139; women and IS 165, 168 – 169 Red Army Faction 168 Reddit 507 Reed, M. S. 315 Reference Framework of Competences for Democratic Culture (RFCDC) 421 Reidy, K. 5 Reinares, F. 448 Reiner, B. 139 reintegration: deradicalisation and 4, 448, 450; disengagement and 448 – 450, 454; empowerment of women in 455; foreign fighter research and 357; governmentinitiated 270; ISIS returnees and 448, 455; NGOs and 455; professional development and 454, 456; programme theory in 328; recidivism prevention and 448 – 449; reducing idleness strategy 448; soft approach to 450 – 451 Rekawek, K. 235 Reker, H. 224 relational approach 27 – 28

religious conversions: alternative spirituality movements 140; homegrown terrorists and 139 – 140; identity experimentation and 139; identity transformation and 139 – 140; ideology and 138; individual radicalisation and 138 – 139, 186; jihadism and 133, 138 – 139, 141; model of conversion 138; NRMs and 133 – 143; process of 140; radicalisation process 137 – 142; as social construct 138; social networks and 139; verbal converts 141; youth and 137 – 138, 141 religious cults 76, 134, 139 Religious Rehabilitation Program 448 religious terrorism: assassination attempts 54 – 58; Catholic radicalisation 54 – 57; counter radicalisation and 252, 254; deprivation theories 134; deviation from righteous beliefs 46; differences in 44; mobilisation of foreign volunteers 44; NRMs and 142 – 143; pathways into 55; radicalisation and 133; tyrannicide and 54; value systems in 43; see also Islamist terrorism Research Center for Police Studies (PRIK) 454 Research Information and Communications Unit (RICU) 280, 435 resource mobilisation 100 – 101, 106 Resource Mobilisation Theory 100 responsibilisation 294, 298 – 299 Revolutionary United Front 530 Reynolds, S. C. 241 Rheingold, H. 198 Richards, J. 374 – 375 Richardson, J. T. 136 Richardson, L. 71, 74 RICU see Research Information and Communications Unit (RICU) Ridings, C. M. 198, 201 right-wing extremism see far-right extremism risk assessment and management: actuarial 476, 484, 490; Area Under the Curve (AUC) 478; challenges of 465; effectiveness of 477; effect size for violence perpetration 485 – 486, 486; frontline workers and 462 – 463, 467 – 468; implications for 486 – 488; insights from different instruments in 483, 489 – 490; interventions and 461 – 462, 466 – 468; moderation of predictive performance 481 – 483, 488 – 489; multivariate explanations in 485 – 486, 491; P/CVE training for 463; policing and 462 – 463, 476; pre-crime space and 461 – 462, 476 – 477; prediction instruments 478, 479 – 480, 481 – 484, 488 – 489, 489, 490 – 491; public health model 360, 461; recidivism literature 487, 487, 488, 492, 496; sexual violence and 481 – 483,

561

Index 488; structured professional judgement (SPJ) approach 464 – 465, 476, 484 – 486, 490 – 492; tools for 464 – 465, 477 – 478, 483, 485; unstructured 476, 484; see also at risk individuals Robinson, T. 284, 509 Rochford, E. B. 136 Rogers, R. 507 – 509 Romaniuk, P. 537 Rosato, V. 288n2 Ross, J. I. 30n2 routine activity theory 116 – 117 Roy, O. 44, 85 – 86, 167, 171 Royal United Services Institute (RUSI) 86 Roy van Zuijdewijn, J. de 231 Rushchenko, J. 120 Russo, A. 316 Rwandan Patriotic Front 530

school-related violence 215, 218, 326, 415; see also preventing violent extremism in education (PVE-E) schools: counter radicalisation in 281, 285, 326, 373, 388, 393, 415, 419 – 422; ethos of 426; Prevent Duty in 360, 409; prevention of extremism and 426 – 427; surveillance in 295; see also education; preventing violent extremism in education (PVE-E) School strike for the climate 427 Schroeder, R. 199 Schuurman, B. 9, 253, 263, 328 Schwalbe, C. S. 482 Schwartzband, S. 64 Scientology 139 Scott, J. W. 170 Seaward, A. 476 securitisation: civil society organisations (CSOs) and 518; counter radicalisation policy and 388 – 389, 519, 523; of education 416 – 417, 427; fragile states and 514; mainstreaming and 388; multi-agency (MA) approaches and 380; Muslim communities and 164 – 165, 294, 302, 375, 534; political dissent and 518 – 519 security studies: colonial-postcolonial-decolonial lens 339; gendered approaches to 165, 168, 173; radicalisation in 1; sociological approaches to 277; women in 165; women’s agency and 168 Sedgwick, M. 254 – 255 Selenica, E. 316 self-sacrifice: identity fusion and 88 – 89, 90, 94; motivation for 87 – 88; political assassination and 55, 59 – 60, 63; pro-group behaviours and 93; suicide terrorism and 89 Serge, V. 62 Sestoft, D. 372 Sethole, F. R. 349n4 Sewell, A. 248 sexual violence 165, 481 – 483, 488 Shah, A. 218 Shaw, C. R. 120 Shaw, E. D. 24 Sheikh, M. K. 254 Sherman, L. W. 116 Shortland, N. 359 – 360 Sieckelinck, S. 297 – 298, 371, 415 Sierra Leone 530 Sigerist, H. E. 363 significance quest approach 74 – 75, 87 – 88 Silber, M. D. 489 Silke, A. 74, 136 Simi, P. 119, 121, 185, 189n14, 203 – 204 Simon, J. 409

Sabaoon programme 454, 456 safe spaces 102 – 103, 106 Sageman, M. 11, 28, 40 – 41, 87, 92, 103, 139, 166 – 167, 171, 174, 240, 248 – 249, 312 Sahel 321 Said, E. W. 339 Salafi Jihadism: Al-Shabaab and 520; counter narrative campaigns against 432 – 434, 442; deradicalisation and 448; fragile states and 517; ideology and 71, 79; Islamic conversion and 133, 141; jihadi terrorism and 44; lone-actor terrorism and 218; opposition to civilian targets 452; P/CVE and 300 – 301; reintegration programming and 455 Saliba, J. A. 136 Saltman, E. 438 Salvini, M. 221 Sánchez-Cuenca, I. 8 Sand, C. L. 60 Sandberg, S. 106 Santa Fe Institute 78 Sarason, S. B. 199 Sarma, K. M. 465, 477 Sarwono, S. 454 Saudi Arabia 46, 255, 324, 448, 450, 456 Saul, B. 395n5 Savage, S. 76 Savoia, E. 326 Schmid, A. 47 Schmid, A. P. 30n1, 73, 255, 263, 395n2, 434 – 435 Schmidt, E. 532 Schmidtchen, G. 180 – 181 School, Social Services, and Police (SSP) model 372 – 373, 376

562

Index Singapore 46, 448, 450, 456 Singh, J. P. 482 – 483, 485 Singh, U. 64 Sirhan, S. B. 64 SITE Intelligence group 502 situational action theory 120 situational crime prevention (SCP) 117 situational crime theory 116, 124n6 Sivenbring, J. 369, 378 – 379 Skleparis, D. 462 Smith, J. 75 Smith, R. J. 218 Snow, D. A. 102, 104, 106 Snowden, D. 78 social construct 7 social control perspective 120, 122 social crime prevention 371 social disorganisation theory 120 social identity theory 87 – 88 social influence 139, 150 – 152; see also social ties social media: content moderation and 498 – 510; counter narratives and 434, 438 – 440; counter radicalisation efforts 499, 505; deplatforming and 507 – 508; extremist content and 196, 223 – 224, 499; farright platforms 223; foreign fighters and 205 – 206; GIFCT and 504; government regulation of 359; impact on counterterrorism 203; information exchange 201; ISIS supporters and 203 – 207; livestreaming of terrorist attacks 223 – 224; lone-actor terrorists and 223 – 224; platform governance 500; stability of membership 201 – 202; trusted flaggers and 502 – 503; Twitter suspensions 207 social movement organisations (SMOs) 100 social movement theory (SMT): contextual factors and 100 – 102, 108n1; framing theory 106; group messaging and 167; limitations of 28 – 29; NRMs and 139; Political Process approach 100 – 101, 104, 106 – 107; political violence and 99 – 107; processual perspective 100 – 102, 105, 108n1; protest studies in 99 – 102; radicalisation research and 103, 106 – 108; resource mobilisation 100 – 101, 106; safe spaces 106; social ties and 28, 43, 139; terrorism studies and 27 – 29 Social Network Analysis (SNA) 313 social ties: cognitive radicalisation and 28, 43; foreign fighters and 240; lone-actor terrorism and 214; radicalisation process and 42 – 43, 48; social movement studies and 28, 43; terrorist groups and 28 – 29, 30n3

Society for Terrorism Research 237 Sōka Gakkai 139 Soliman, A. 187 Somalia: Al-Qaeda and 530; Al-Shabaab and 516, 520; foreign fighters and 45, 205; homegrown insurgency movements in 531; P/CVE programmes in 534; radicalisation to violent extremism in 516 – 517, 530 – 531; Salafi Jihadism in 517; violent movements in 530 – 531 Somali Islamic Brotherhood 531 Somali Islamic Union 531 Somerville, J. 54 Southeast Asia: amok attacks in 225; Islamist terrorism in 71; P/CVE evaluation in 344, 345, 346, 347; terrorism and 446; see also Indonesia Soviet Union 240 Spaaij, R. 218 Spain 280, 286, 400 Spalek, B. 301 Spanish Civil War (1936–9) 234 – 236, 240 Special Investigative Department (ROS) 278 Spitzka, E. 63 – 64 Spivak, G. C. 348 Sprinzak, E. 21, 27 staircase to radicalisation model 39, 87, 489 Stankov, L. 187 Stark, R. 138, 141 Staub, E. 72 Steffen, F. 224 Stephens, W. 297 – 298, 371, 415 Sternberg, R. 94 Stevens, D. 253, 256 Stone, L. 183 Stormfront 203 – 205 Stormfront Downunder 205 strain theory 118, 122 Strategic Implementation Plan (SIP) 303n1 strategic influence 80 strategic narratives 80 Strindberg, A. 87 STRIVE II programme 325, 328 structured professional judgement (SPJ) approach: individual behaviour and 491; risk assessment and 464 – 465, 476, 484 – 486, 490 – 492 sub-cultural theory 141 sub-Saharan Africa 344 – 345, 347 – 348 substance abuse 183, 186 – 187 suicide terrorism: Al-Shabaab and 531; deprivation theories 134; foreign fighters and 45; identity fusion and 77, 89; ISIS supporters and 531; personality tests and 184; prevention of 117; prohibited encouragement of 406; self-sacrifice and 89

563

Index Sundar, S. S. 437 Sunstein, C. 76, 78 al-Suri, A. M. 452 Sutherland, E. H. 119 Suzor, N. P. 508 Swann, W. 93 Sweden 301, 358, 377 Swidler, A. 73 Switzerland 241 Syrian civil war: Jihadist groups and 231; mobilisation of foreign fighters for 40, 45, 74, 134 – 135, 196, 205 – 206, 231, 236, 241, 277; Shia fighters and 232 Syrian-Kurdish People’s Protection Units (YPG) 233 systems: complexity theory and 78 – 79; homeostasis and 78 – 79; ideology and 77 – 81; transnational networks and 79 – 81

263; individual-level data 122 – 123, 124n7; individual participation and 20 – 21; loneactor terrorism and 215; multi-level models for 27 – 28, 30n2; open-source data and 183; politically-motivated versus religious 43 – 44; psychological-structural model for 30n2; psychological studies 22 – 25, 29 – 30; psychopathology and 21 – 22, 184; push and pull factors in 238; radicalisation concept in 19 – 21, 28 – 30, 30n4, 41 – 42, 47 – 48; religious terrorism in 43; social movement studies 27 – 29; social networks and 240; social psychology model 30n2; soft/hard target characteristics 116 – 117; strain theory 118; system level of analysis 77; terrorist belief systems and 25 – 27; terrorist profiles in 21 – 25, 29, 93; terrorist testimonies in 25; women’s violence and 168 terrorist groups: belief systems 26 – 27; confrontations with police 24, 28; counterterrorism and 20; decision-making in 20; deradicalisation and 447; identity theory and 23; personal needs and 24; political and cultural context 26; political ideology and 26 – 27; psychopolitical process 27; recruitment and 240; sexual violence tactics 165; social ties and 28, 30n3; socio-economic backgrounds 22; terrorist testimonies and 25; see also extremist groups; group radicalisation terrorist manifestos: assassins and 65; far-right and 79; nationalist posturing 60 – 61; psychological patterns in 93 – 94 terrorist messaging: anger appeals 158 – 159; behavioural motivation and 155; counterpersuasion interventions 150, 155 – 156; discrete emotion theory 157; fear appeals 158 – 160; ideological change and 151, 153; pride appeals 158, 161 – 162; reasoned action theory 155 – 156; resonating with audience 167, 173 terrorist mindset model 489 Terrorist Radicalisation Assessment Protocol (TRAP-18) 465, 477 – 478, 490 terrorists: at-risk population 24, 462; belief systems 25 – 27; dehumanization of enemy 26; group membership and 24; identifying potential 93 – 94; identity theory 23; interpretation of ideology 75; mental health problems and 29 – 30, 120, 180 – 185, 361; mindset of 390, 423, 427, 489; narcissism and 23 – 24; personality traits and 182 – 183, 187, 189n11, 189n12; psychological studies on 22 – 25, 29 – 30, 93 – 94; psychopathology and 21 – 22, 180 – 189; radical beliefs and 249; recidivism and 119; role of early

Taliban 45 Tamil refugees 341 Tammikko, T. 376 Tannenbaum, M. B. 160 Tarrant, B. 82n2 Taylor, M. 24, 105, 253 technology: automated data collection and 313; content moderation and 499, 501 – 505, 510; ethnographic research and 313; ideological radicalisation and 66; online communities and 198; see also social media Telegram 507 – 508 Terror in the Mind of God (Juergensmeyer) 27 terrorism: conceptualisation of 48, 404 – 405; continuum of 216; deprivation theories 136; emergence of 27; entry processes in 266 – 270; EU criminalisation of 399; exit pathways 268 – 269; gamification of 224; gendered approaches to 165; as group activity 20 – 21; historical moment and 48; legitimizing ideology and 71, 74; new religious movements (NRMs) and 133; opportunity-reducing techniques 117; political and social grievances 25, 30n1; pre-crime offenses and 405 – 406; preparation for 406; radicalisation concept and 4 – 5, 34 – 35, 392; reintegration programming and 4; research in 19; root causes of 41; stages of 27; see also violent extremism Terrorism Act 2000 400 terrorism studies: data collection and 313; delegitimation theory 27; dispositional explanations and 105 – 106; domestic terrorism in 263; empirical 8, 22 – 24, 150; Eurocentric 22, 26; gendered approaches to 165 – 166, 168 – 169, 173 – 174; ideology in

564

Index socialisation 23 – 24; social ties and 28 – 29, 30n3; socio-economic backgrounds 22 Texas Revolution (1835–6) 234 Theory of Change approach 327 – 329 Think Again, Turn Away campaign 442 Thomas, P. 292, 299 Thornton, A. 375 Thoroughgood, C. 448 Tigray People’s Liberation Front 530 Tilly, C. 105 – 106 Tololyan, K. 26 Trump, D. 221 – 222, 441, 506 Tumelty, P. 233 Turkle, S. 197 Twitter: Baqiya family 206 – 207; content moderation and 499 – 500, 508; counterextremism policies and 509; countermessaging on 440; deplatforming and 507 – 508; EU code of conduct for 500; far-right extremism and 509; GIFCT and 504; hate speech and 505; ISIS supporters and 203 – 207; researcher data and 508; suspension of extremist accounts 207, 223; trusted flaggers and 502, 510n2 Two Pyramids model 88 Tworek, H. 500 tyrannicide 54 – 55

UN Counter-Terrorism Strategy 537 Understanding Terror Networks (Sageman) 28, 240 UNESCO 415 UNESCO Chair in Prevention of Radicalisation and Violent Extremism (UNESCO-PREV Chair) 330 UN Global Counter-Terrorism Strategy 396n6 Unification Church 136, 139 Union of Islamic Courts 516 United Kingdom: Al-Qaeda influence in 45; Commission for Countering Extremism 256; CONTEST strategy 250, 279, 293, 374, 401, 432, 466; counter narratives and 435; counter radicalisation policy 277 – 280, 282, 284, 286, 288n4, 293, 299, 302, 400, 504; counter-radicalisation programme 41, 45 – 46, 92, 250 – 251; evaluation of P/CVE 321; Extremism Risk Guidance (EGR22+) 464; Identifying Vulnerable People (IVP) system 464; ideology-oriented approaches 279 – 280; interventions for at-risk individuals 466 – 468; multi-agency approaches in 92, 359 – 360, 371; Muslim communities and 298 – 300; National Security Strategy 401; P/CVE and 298 – 299, 303n1, 321, 462, 466; P/CVE evaluation in 330; Prevent Duty in 280, 360, 374, 388, 400 – 404, 406, 409 – 411; RICU and 280, 435; risk assessment and 463, 476 – 477; suspect communities in 294, 299; targeting of ideas 257; teacher reporting in 378; terrorism related criminal offences in 400; Vulnerability Assessment Framework (VAF) 375, 464; see also Channel programme; UK Prevent strategy United Nations: counter radicalisation and 92, 252, 400; counterterrorism strategies 390 – 392; mainstreaming discourse and 389 – 392, 394 – 395; Preventing Violent Extremism (PVE) concept 34, 45; root causes rhetoric 41; Terrorism Prevention Branch 47 United Nations Global Counter-Terrorism Strategy 528 United Nations Security Council 165, 231 United Nations Sustainable Development Goals 364 United States: anti-abortion movement 218; community-oriented P/CVE 293; content moderation and 500; counter narratives and 432, 435, 442 – 443; counter radicalisation programmes 326; cult scare in 134, 139; CVE programme in 34, 45 – 46, 298, 300, 325, 357; far-right terrorism in 218; foreign fighters from 231; health promotion approach 364; invasion of Iraq 35, 39 – 40; medicalised language and 360 – 361; multi-agency approaches in 369;

Uganda 530 – 532 Uhl-Bien, M. 80 UK Prevent strategy: at-risk population and 358, 360, 463, 476 – 477; challenges to ideology in 401; Channel programme 374 – 375; counter radicalisation and 41, 280, 400 – 401; counterterrorism and 45, 250, 401 – 402, 404; critique of 339, 357, 360, 402 – 403; Department of Education and 285; European Union and 37 – 38; focus and scope of 401; focus on Islamist ideologies 45, 92, 257, 284, 360; focus on religious ideas 253; infringement on human rights 402 – 403; international interest in 400; mainstreaming and 388, 401 – 402; Muslim communities and 295, 298, 300 – 301, 400 – 401, 463; Muslim Contact Unit partnership 300; objectives of 250 – 251; party politics and 282; pre-crime offenses and 401 – 402, 409 – 410; review of 258n2; role of education in 419; surveillance in 402; WRAP training 463 Ukraine: anti-Jewish pogroms 64; foreign fighters in 232, 234 – 235; white supremacist groups in 76, 80 Ulsterisation 124n5 Umberto I 62 – 63 uncertainty-identity theory 87

565

Index P/CVE and 303n1; public health model 357; radicalisation and 45; radicalisation discourse in 257; risk assessment tools 465; root causes rhetoric 41; subcultures of violence in 118; terrorism prosecutions in 122; terrorism risk assessment 92; Think Again, Turn Away campaign 442 United States for International Development (USAID) 518 ‘Unite the Right’ rally (2017) 222 UN Security Council 386, 390, 399, 504 unstructured risk assessment 476, 484 U.S. Department of Homeland Security 492n1 US Patriot Act 342 Üstün, T. B. 183 Utøya massacre (2011) 219, 224

CVE frameworks for 296, 303n1; online communities and 8, 196, 202 – 208; outgroup prejudice and 263 – 264; personal/ group identities and 86; personality traits and 182 – 183, 187, 189n11, 189n12, 265 – 266; psychological elements of 263 – 264; psychopathology and 186; push and pull factors 448; radicalisation and 4 – 5, 19, 30n4; radical milieus and 263 – 264; reaction and 30n4; research in 113 – 115, 115, 116; routine activity theory 116 – 117; shared ideology and 264 – 265; situational perspectives 116 – 117; as social contagion 360; social influence and 151 – 152; social ties and 103 – 104; transnational networks and 79 – 80; unresolved grievances and 85 – 86; see also terrorism Violent Extremism Risk Assessment (VERA-2R) 464 – 465, 477, 490 violent radicalisation 37 – 40, 349n1 virtual communities see online communities Virtual Community, The (Rheingold) 198 Vries, G. de 37 Vulnerability Assessment Framework (VAF) 375, 464

Vaillant, A. 62 Vaisey, S. 72 van den Berg, J. W. 482 – 483 van Der Put, C. E. 481 – 482, 485 van de Weert, A. 249, 369, 462 Van San, M. 135 Venkatesh, V. 338, 341 Vergani, M. 86 Victoroff, J. 136, 187 Vidino, L. 82n1, 249, 288n2, 375 Viljoen, J. L. 483 violence: culture of 73; factors in escalating 91; group-based 464 – 465; ideology and 65 – 66; intersectional approach to 6; masculinity and 171 – 172; microcycles of 116; politicisation of 23; psychoticism and 187; terrorist belief systems and 26 – 27 violent extremism: at-risk population and 464; behavioural changes and 141; community resilience and 296 – 298; conceptualisation of 6, 48; conspiracy theories and 221 – 222, 266; countercultural networks and 28; criminal subculture theories 118; criminological theories and 113 – 122; defining 38, 124n1, 349n1; deprivation theories 118; deradicalisation and 447; disengagement and 447; entry processes in 266 – 270; EU strategies and 38; exit pathways 268 – 269; fragile states and 514 – 517; gendered approaches to 167; in-group bias and 263; group cohesion and 88; ideology and 71, 74, 79 – 81, 85 – 86, 263 – 264; illegality and 264 – 265; individual trajectories and 48; jihadi subculture and 118; life-course perspectives 119; local understandings of 535; medicalised language and 360 – 361; mindset for 390, 423 – 424, 427; motivation for 92; narratives and 436; national P/

Waffen-SS 233 – 234, 236, 240 Waldmann, P. 104, 167 Waldrop, M. M. 78 – 79 Waleciak, J. 271 Walker, C. 403, 406, 408 Walmsley, J. 9, 276, 418, 420 Walters, G. D. 482 Wansink, B. 425 Warfa, N. 187 Wayang Kulit shadow puppetry 452 weak governance: counter radicalisation policy and 517; external security and development interventions 516 – 517; fragile states and 514 – 519; inadequate provision of basic security 515 – 516; non-state actor roles 517 – 519; political marginalisation of communities 515; radicalisation and 514 – 515, 523; role of women under 519; securitisation of political opposition 518 – 519 Webber, D. 326 Weeks, D. 300 Weenink, A. W. 237 Weine, S. 328, 360 – 361 Wellman, B. 198 West: coercive infrastructure of 66; colonialism and 42, 339 – 341; coloniality of knowledge and 341 – 345, 347, 349; counter radicalisation and 251 – 252, 292, 302; evaluation of P/ CVE and 344 – 345, 345, 346, 347; foreign

566

Index fighters from 22, 135, 231 – 232; gendered dynamics in 173; Islamic threat and 35, 44, 47; production of PVR knowledge 343 – 344, 347 – 348; radicalisation and 53; revolutions in 58 Western Europe: emergence of radicalisation 53 – 54, 57; far-left terrorist groups in 26; foreign fighter recruitment and mobilisation 241; P/CVE evaluation in 344 – 345, 345, 346, 347; political assassination 53 – 62; see also Europe; European Union Westgate Mall attack (2013) 531 – 532 White, R. W. 30n3 Whitehouse, H. 85, 89 white supremacist groups: emotionally disturbed youth and 75; gender and 169; great replacement idea 79, 82n3; ideological radicalisation and 82; isolated indoctrination and 76; online communities and 203 – 204; persuasive narratives and 436; quest for significance 75; transnational networks and 80; violent extremism and 218 Whittaker, J. 222 whole-of-society approach 517, 520 Wiktorowicz, Q. 73, 87, 103 – 104, 167, 489 Wilhelm I 62 William I 54 – 57 William III 57 Williams, M. 77 Williams, M. J. 325 Wilson, B. R. 139 Wilson, J. 73 Wilson, T. 8, 53 Wimelius, M. 301 Windisch, S. 268 Winterbotham, E. 171, 173 Wischmeyer, T. 500 Wolfgang, M. 118 Wolfowicz, M. 120 women: agency and 168; far-right ideology and 74, 169; fragile states and 519, 522; gender as 165 – 166, 168 – 169; Jihadist gender

roles and 171 – 172; masculine identities and 170; motherhood and 171; role in counter-radicalisation 519, 522; support for Salafi-Jihadi groups 169, 172, 233; see also gender; women’s violence Women, Peace and Security (WPS) 165 women’s violence: agency and 168 – 169; gendered approaches to 165, 168 – 169; jihadism and 172; political ideology and 54, 168; recruitment to IS 165, 168 – 169; terrorism studies and 168; as transgressive 172 World Health Organization (WHO) 356 WRAP (Workshop to Raise Awareness of Prevent) training 463 Yamaguchi, A. 208 Yanez, Y. 113 Yemen 255, 448, 456 youth: immigrant community anger 35 – 37, 39; Jihadist radicalisation and 167, 417; Muslim community anger 35 – 36, 39 – 40, 44 – 45, 134; online communities and 10; radicalisation and 417 – 418, 425 – 426; religious conversions and 137 – 138, 141; search for heroism and brotherhood 85; white supremacy and 75, 417 YouTube: content moderation and 409 – 500, 502 – 503, 505; counter-extremism policies and 509; counter radicalisation efforts 505; deplatforming and 509; EU code of conduct for 500; far-right extremism and 509; GIFCT and 504; trusted flaggers and 502 – 503, 510n2 Yzer, M. C. 156 Zedner, L. 359, 408, 410 Zenker, E. 64 Ziamani, B. 220 Zickmund, S. 202 Zimbabwe African National Union 530 Zmigrod, L. 265 Zurcher, L. A. 104

567