“This impressive volume, the result of IDRC’s and DFID’s Safe and Inclusive Cities initiative and undertaken with researchers from cities in the Global South, is unique in its presentation of in-depth original research results within a cross-regional, cross-discipline holistic framework. In this way, it fills a gap in knowledge of the urban violence–poverty–inequality nexus. The book brings together a remarkable range of individual case studies, focusing on such issues as social cohesion, exclusion, and gendered violence, with provocative—at times counter-intuitive—policy solutions. In demonstrating that the challenges of ubiquitous urban violence transcend borders and showing that resistance and resilience are fundamental local coping strategies, this path-breaking book is essential reading for academics, policymakers, and practitioners alike.”

Caroline Moser, Emeritus Professor, University of Manchester, UK

“This second book based on research carried out under the Safe and Inclusive Cities initiative focuses its attention on interventions to address the extremely complex, diverse, layered, and often inter-connected forms of violence experienced in cities of the Global South. In a scenario where the state is often implicated as a perpetrator of violence and where social cohesion is frayed with groups pitched against each other, as the chapters illustrate, interventions are bound to be embedded in the complexity of the violence and hence ‘imperfect’. But, there is a lot to be learned from them as works in progress.

One point that is clearly made is that safety and inclusion cannot be treated as independent arenas; they are linked. The sheer diversity of forms of interventions discussed here is amazing, ranging from grassroots interventions where the stakes and the roles of women and children in peace-keeping are asserted, to state interventions such as livelihood and safety linking projects such as the Community Work Programme, community upgrading, and policing. Beyond the specifics of the interventions discussed, the value of the book is that it represents an analysis by researchers deeply engaged with their local contexts, and hence insights that are highly nuanced and appreciative of particularities, temporalities, and complexities. If the world must make a definitive movement towards attainment of the Sustainable Development Goals, listening to and reading these fascinating insights is a must for international, national, and local policymakers and all those interested in development dynamics.”

Amita Bhide, Professor and Dean, School of Habitat Studies, Tata Institute of Social Sciences, India

Reducing Urban Violence in the Global South: Towards Safe and Inclusive Cities, edited by Jennifer Erin Salahub, Markus Gottsbacher, John de Boer, and Mayssam D. Zaaroura, brings together nine chapters on problems and solutions to violence in some cities of the so-called Global South to understand their many similarities, but also their specific differences.

The world became urban in 2006 when, for the first time, more than half its population was living in cities. However, while this trend was not new for the developed world where urbanisation has now plateaued, Asia and Africa are urbanising at an impressive rate, as is Latin America. In line with this phenomenon, violence has urbanised or, in other words, it has a tendency to concentrate in cities. Typical urban violence has emerged, expressed through: occurrence in shared spaces (public and private); unequal access to services, equipment, and infrastructure; discrimination based on ethnicity, gender, age, or status as a migrant; and illegal markets such as money laundering or trade in illicit substances, among others.

The proposal to build safe cities has arisen to confront urban violence in its duality: its concentration in cities and its urban specificity. This book helps address this double objective, examining the problem through case studies from Latin America, Asia, and Africa. Undoubtedly, this is a book that contributes to our knowledge of the problem of urban violence, but also to building urban coexistence through structural responses to inequalities, violence prevention policies, and community participation.”

Fernando Carrión, Research Professor, FLACSO Ecuador

Reducing Urban Violence in the Global South

Reducing Urban Violence in the Global South seeks to identify the drivers of urban violence in the cities of the Global South and how they relate to and interact with poverty and inequalities. Drawing on the findings of an ambitious 5-year, 15-project research programme supported by Canada’s International Development Research Centre and the UK’s Department for International Development, the book explores what works, and what doesn’t, to prevent and reduce violence in urban centres.

Cities in developing countries are often seen as key drivers of economic growth, but they are often also the sites of extreme violence, poverty, and inequality. The research in this book was developed and conducted by researchers from the Global South, who work and live in the countries studied; it challenges many of the assumptions from the Global North about how poverty, violence, and inequalities interact in urban spaces. In so doing, the book demonstrates that accepted understandings of the causes of and solutions to urban violence developed in the Global North should not be imported into the Global South without careful consideration of local dynamics and contexts. Reducing Urban Violence in the Global South concludes by considering the broader implications for policy and practice, offering recommendations for improving interventions to make cities safer and more inclusive.

The fresh perspectives and insights offered by this book will be useful to scholars and students of development and urban violence, as well as to practitioners and policymakers working on urban violence reduction programmes.

Jennifer Erin Salahub is a Canadian public servant. She managed the Safe and Inclusive Cities initiative, a global research programme jointly funded by Canada’s International Development Research Centre and the United Kingdom’s Department for International Development.

Markus Gottsbacher is senior programme specialist with the Governance and Justice programme at the International Development Research Centre.

John de Boer is managing director of the SecDev Group.

Mayssam D. Zaaroura is the women’s rights knowledge specialist at Oxfam Canada.

Routledge Studies in Cities and Development

The series features innovative and original research on cities in the Global South, aiming to explore urban settings through the lens of international development. The series particularly promotes comparative and interdisciplinary research targeted at a global readership.

In terms of theory and method, rather than basing itself on any one orthodoxy, the series draws on a broad toolkit taken from across social sciences and built environment studies, emphasizing comparison, the analysis of the structure and processes, and the application of qualitative and quantitative methods.

The series welcomes submissions from established and junior authors on cutting-edge and high-level research on key topics that feature in global news and public debate.

The Politics of Slums in the Global South

Urban Informality in Brazil, India, South Africa and Peru

Edited by Véronique Dupont, David Jordhus-Lier, Catherine Sutherland and Einar Braathen

Social Theories of Urban Violence in the Global South

Towards Safe and Inclusive Cities

Edited by Jennifer Erin Salahub, Markus Gottsbacher, and John de Boer

Reducing Urban Violence in the Global South

Towards Safe and Inclusive Cities

Edited by Jennifer Erin Salahub, Markus Gottsbacher, John de Boer, and Mayssam D. Zaaroura

First published 2019

by Routledge

2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN

and by Routledge

52 Vanderbilt Avenue, New York, NY 10017

Routledge is an imprint of the Taylor & Francis Group, an informa business

Co-published with the

International Development Research Centre

PO Box 8500, Ottawa, ON K1G 3H9 Canada

info@idrc.ca / www.idrc.ca

The research presented in this publication was carried out with the financial assistance of Canada’s International Development Research Centre (IDRC) and the United Kingdom’s Department for International Development (DFID). The views expressed herein do not necessarily represent those of IDRC or its Board of Governors, or those of DFID.

© Contributors 2019. Licenced under the Creative Commons

Attribution 4.0 License. http://creativecommons.org/licenses/by/4.0

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: Salahub, Jennifer Erin, editor.

Title: Reducing urban violence in the global South : towards safe and inclusive cities / edited by Jennifer Erin Salahub [and three others]. Description: Abingdon, Oxon ; New York, NY : Routledge, 2019. | Series: Routledge studies in cities and development | Includes bibliographical references and index.

Identifiers: LCCN 2019002758 (print) | LCCN 2019005112 (ebook) | ISBN 9781351254649 (eBook) | ISBN 9780815368427 (hardback) Subjects: LCSH: Urban violence–Developing countries–Prevention. | Urban policy–Developing countries. | Sociology, Urban–Developing countries.

Classification: LCC HN981.V5 (ebook) | LCC HN981.V5 R43 2019 (print) | DDC 303.609173/2–dc23 LC record available at https://lccn.loc.gov/2019002758

ISBN: 978-0-8153-6842-7 (hbk)

ISBN: 978-1-351-25464-9 (ebk)

ISBN: 978-1-55250-598-4 (IDRC ebk)

Typeset in Sabon

by Wearset Ltd, Boldon, Tyne and Wear


List of figures and tables

Notes on contributors




Acronyms and abbreviations

Introduction: a fast-moving and dynamic urban landscape



West Africa

1Controlling violence: an economic integration skill in Côte d’Ivoire


2Opening up or closing off: urbanisation, violent crime, and the “poverty penalty” in Ghana’s four largest cities



Southern Africa

3Urban upgrading linked to positive social outcomes in Cape Town, South Africa


4Towards safer communities: the impact of the Community Work Programme on the prevention of urban violence in South Africa


5Is social cohesion the missing link in preventing violence? Case studies from South Africa and Brazil



Latin America

6Social exclusion, violences, and urban marginalisation in Central America: empirical evidence and consequences for public policy


7The contribution of informal institutionality to safe cities in Venezuela



South Asia

8Everyday violence in urban India: is planning the driver or mitigator?


9Pathways to inclusive urban development: contrasting experiences of relocated men and women in Colombo and Kochi


Conclusion: reflections on how to make cities safer and more inclusive



Figures and tables


I.1Urban populations in selected fragile countries, 1965–2030

1.1Map of the main informal transport hubs of Abidjan

1.2Organisational chart of an informal transport hub

1.3Evolution of violence in informal transport hubs

3.1Household survey points and proximity to VPUU intervention stratified by category of exposure

3.2Experience of severe violence in the last year by neighbourhood within Khayelitsha

3.3Experience of severe violence by proximity to VPUU intervention

3.4Symptoms of depression by neighbourhood within Khayelitsha

3.5Depression by proximity to VPUU intervention

3.6Satisfaction with infrastructure by neighbourhood within Khayelitsha

3.7Satisfaction with infrastructure by proximity to VPUU intervention

3.8Community participation by neighbourhood within Khayelitsha

3.9Community participation by proximity to VPUU intervention

3.10Social cohesion by neighbourhood within Khayelitsha

3.11Social cohesion by proximity to VPUU intervention

6.1Social clusters representing the territories studied in Costa Rica

6.2Social clusters representing the territories studied in El Salvador

8.1Study sites in Ahmedabad

8.2Informal markets studied in Guwahati

8.3Informal hill settlements studied in Guwahati 184

9.1Effects of the relocation on the MQHI scores of relocated households and control groups in the two cities—Colombo and Kochi

9.2Average happiness levels among DFDR and control groups in Colombo and Kochi

9.3Empirical characterisation of the impact of development-forced relocation on affected men and women in Colombo and Kochi


2.1Selected communities and their respective households/victims sampled

2.2Cross tabulation—type of violent crime and neighbourhood SES (from survey)

2.3Cross-tabulation—existence of neighbourhood watch committee and community cooperation and support against crime by neighbourhood SES (from survey)

2.4Cross tabulation—gender and types of violent crime among low-, middle- and high-SES respondents (from survey)

3.1Summary statistics for each outcome and covariates by exposure category

6.1Scales of social inclusion/exclusion

6.2Household clusters by group of territories (averages)

6.3Dynamics of violence by type of violence and by territory

6.4Initiatives to tackle violence in the territories

9.1Household survey: summary

9.2Logistic regressions to explain current happiness levels

9.3Household survey indicator groups and specific indicators 212

Notes on contributors


Robert Muggah specialises in cities, security, migration, and new technologies. He co-founded the Igarapé Institute—a think-and-do tank based in Brazil devoted to working on data-driven safety and justice in Latin America and Africa. He also co-founded the SecDev Group—focused on cyber-security and digital risk in Europe, the Middle East, and Asia. Robert is a senior adviser to the United Nations, the Inter-American Development Bank, the World Bank, and a number of companies ranging from Google to McKinsey’s. He is a frequent contributor to The Atlantic, BBC, CNN, El País, Folha de São Paulo, Foreign Affairs, Foreign Policy, Globe and Mail, the Guardian, Los Angeles Times, and New York Times, among others. He has delivered several TED talks (2017, 2016, and 2015), and supports social impact start-ups from Brazil to the US. He is a fellow or faculty member at the Chicago Council on Foreign Affairs, the Canadian Global Affairs Institute, Singularity University, the International Relations Department at the Pontifical Catholic University in Rio de Janeiro, and the World Economic Forum. Robert is the author of seven books and dozens of articles. He received his DPhil from the University of Oxford.

Lead editor (Conclusion)

Jennifer Erin Salahub is a Canadian public servant. Previously, she managed the Safe and Inclusive Cities initiative, a global research programme jointly funded by Canada’s International Development Research Centre (IDRC) and the United Kingdom’s Department for International Development. Before joining IDRC, Jennifer was senior researcher and team leader of the Fragile and Conflict-Affected States programme at the North–South Institute (NSI), where she researched topics such as peace-building, fragile states, and legislative effectiveness. Her specific focus was security and development: at NSI, she contributed to the fields of security sector reform, gender equality in fragile states, and gender-sensitive policing. Jennifer’s MA in Political Science (International Relations) is from McGill University, Canada.

Contributing editors (Introduction)

Markus Gottsbacher is a senior programme specialist with the Governance and Justice programme at IDRC. A political scientist and ethnologist by training, he develops and manages research projects in Latin America and the Caribbean to support efforts for peace-building, citizen security, and access to justice. Previously, Markus worked on an IDRC project in Honduras that focused on dealing with conflict over natural resources. He has almost a decade of experience with various UN agencies (mostly in Latin America), and has worked on issues ranging from drugs and crime to HIV/AIDS, and the rights of indigenous peoples and women. As a professor at several universities in Mexico, he focused on human security, international development and cooperation, and conflict prevention. Markus has worked for a number of NGOs and, while at the Austrian Study Centre for Peace and Conflict Resolution, he developed a training programme for peace-builders in conflict areas. Markus holds a PhD in Political Science from the University of Vienna.

John de Boer is managing director of The SecDev Group, an international risk consultancy firm; he is also an expert on development, humanitarian, and security challenges in situations of conflict and violence. He was previously senior policy advisor with the United Nations University’s Centre for Policy Research, where he researched vulnerability in urban contexts; specific challenges included urban violence and disaster, and organised crime. John was formerly programme leader for Governance, Security, and Justice at Canada’s IDRC. He has worked at the Canadian International Development Agency, where he served on the Afghanistan and Pakistan Task Force. John holds a PhD in Area Studies from the University of Tokyo.

Contributing editor (Conclusion)

Mayssam D. Zaaroura is Oxfam Canada’s Women’s Rights Knowledge Specialist working on feminist approaches to research on violence against women and girls and development interventions to end that violence. She is an expert on gender and violence in fragile and conflict-affected states, with over 15 years’ experience working on gender transformation, gender and forced migration, conflict and transition processes, women’s political empowerment in conflict and post-conflict settings, and with civil society organisations and political actors on women, peace, and security. Prior to joining Oxfam Canada, Mayssam was the programme officer with the Safe and Inclusive Cities Initiative in the Governance and Justice programme at IDRC, working on urban violence and on women’s empowerment in the Middle East, Africa, and Southeast Asia. As a communications expert, she was national desk editor and senior political journalist at the Daily Star newspaper in Lebanon, covering political issues in the Middle East and North Africa. She holds a Master’s degree in International Development and Global Studies from the University of Ottawa.

Chapter authors

Chapter 1

Kouamé Walter Kra is a senior lecturer in sociology in the Department of Anthropology and Sociology at Alassane Ouattara University in Bouaké (Côte d’Ivoire), where he has been teaching and researching since 2009. The holder of a doctorate in the Sociology of Development from the University of Cocody-Abidjan (2007), for the past ten years he has led research related to the development dynamics of Côte d’Ivoire. He is particularly interested in youth violence, community violence, and academic violence in the context of a Côte d’Ivoire that is struggling to reach its development milestones.

Chapter 2

Adobea Yaa Owusu is an associate professor at the Institute of Statistical, Social, and Economic Research at the University of Ghana. Her current research focuses on urban crime, access to health care, health and social development, HIV/AIDS, health services research, and programme monitoring and evaluation.

Martin Oteng-Ababio is an associate professor and coordinator of the Disaster Risk Reduction Programme in the Department of Geography and Resource Development at the University of Ghana. He is an urban environmental geographer with a special focus on sustainable environmental management, urban informality, urban crime dynamics and management, and disaster risk reduction.

George Owusu is a professor at the Institute of Statistical, Social, and Economic Research and director of the Centre for Urban Management Studies, both at the University of Ghana. His main areas of research include urban and regional development, crime and the built environment, youth entrepreneurship, and local governance and decentralisation.

Charlotte Wrigley-Asante is an associate professor in the Department of Geography and Resource Development at the University of Ghana. Her teaching and research areas include gender, poverty, and natural resources management; migration, women, and cross-border trade issues in Ghana; and gender and urban crime.

Martin Wiredu Agyekum is a PhD candidate at the Regional Institute for Population Studies at the University of Ghana. His main areas of research focus on population and health with a specific focus on maternal and child health, health care systems, gender and security, estimation and projection, and research methodology.

Chapter 3

Richard Matzopoulos is a chief specialist scientist and deputy director at the Medical Research Council’s Burden of Disease Research Unit and an honorary research associate at the University of Cape Town’s School of Public Health and Family Medicine, where he coordinates its Violence and Injury Research programme.

Kim Bloch is a Master of Public Health graduate and researcher at the University of Cape Town’s School of Public Health and Family Medicine, where she coordinated the community surveys that form the basis for Chapter 3 of this book. From 2015–2017, Kim worked for Violence Prevention through Urban Upgrading on a part-time basis with a specific focus on enhancing their fieldwork and research methods.

Sam Lloyd is a PhD student at the Institute of Science Technology and Policy at the Swiss Federal Institute of Technology in Zurich. His current research seeks to identify effective strategies for violence prevention in the urban settings of Bogotá, Colombia and Cape Town, South Africa. Formerly, he was a researcher at the Centre for Justice and Crime Prevention in Cape Town and at the University of Cape Town’s School of Public Health and Family Medicine. Sam completed his Master of Public Health at that university’s School of Public Health and Family Medicine, specialising in epidemiology and biostatistics.

Chris Berens is a geographic information scientist with more than 20 years’ experience in spatial data management. He currently works at Violence Prevention through Urban Upgrading, where he places special emphasis on the appropriate use of technology so as to embed projects and processes within community and organisational workflows.

Jonny Myers is an emeritus professor of Public Health Medicine at the Centre for Environmental and Occupational Health at the University of Cape Town’s School of Public Health and Family Medicine.

Mary Lou Thompson is a research professor in the Department of Biostatistics at the University of Washington and at the University of Cape Town’s School of Public Health and Family Medicine. She specialises in the statistical analysis of complex epidemiological studies.

Chapter 4

Malose Langa is a senior lecturer in psychology in the School of Community and Human Development at the University of the Witwatersrand and associate senior researcher at the Centre for the Study of Violence and Reconciliation in Johannesburg. His research interests include risk-taking behaviour among youth and their role in politics, the trauma of collective violence, and masculinity in post-apartheid South Africa.

Themba Masuku is an independent researcher with interests in social policy analysis, violence analysis, and research on policing. His experience in research and social policy in South Africa and the Southern African region spans more than 15 years. He has worked for several organisations in South Africa that focus on policing, violence, and human rights policy, and research. He has made specific contributions in research and policy analysis of policing and violence. In 2003, he was appointed to the technical team on Community Policing tasked with conceptualising Community Policing Forums in South Africa. He has provided technical support to the Lesotho Mounted Police Services in researching and developing their workplace policy on HIV/AIDS. At the time of writing Chapter 4 for this book, Themba was a research manager at the Centre for the Study of Violence and Reconciliation. More recently, following the Commission of Inquiry into the Marikana Tragedy in which the police killed 37 striking miners, Themba was appointed to the International Panel of Experts on policing that was tasked with investigating capacity and capability to perform public order policing within the South African Police Service. Themba holds a Bachelor’s degree in Law and a Master’s degree in Social Science.

Hugo van der Merwe is director of Research, Knowledge, and Learning at the Centre for the Study of Violence and Reconciliation (CSVR). Since joining CSVR in 1997, he has initiated and managed numerous research, advocacy, and intervention projects related to transitional justice, reconciliation, and peacebuilding in South Africa and the African continent. Hugo is the co-editor-in-chief of the International Journal of Transitional Justice. His publications include Advocating Transitional Justice in Africa: The Role of Civil Society (2018), Assessing the Impact of Transitional Justice (2009), Truth and Reconciliation in South Africa: Did the TRC Deliver? (2008), and Conflict Resolution Theory and Practice (1993). Hugo received his doctorate in Conflict Analysis and Resolution from George Mason University, US in 1999.

Chapter 5

Vanessa Barolsky is an associate research fellow at the Alfred Deakin Institute for Citizenship and Globalisation (ADI) working with the recently established Collaborative Centre of Excellence for Resilient Communities and Inclusive Societies. She works across several disciplinary areas, including sociology, anthropology, and criminology, to tackle questions related to social conflict and its transformation. This includes critical engagements with social cohesion, truth, and reconciliation, as well as political and criminal violence and questions of social regulation, law, and justice. Prior to her employment at ADI, Vanessa was a research specialist with the Democracy, Governance and Service Delivery Programme at the Human Sciences Research Council (HSRC) in South Africa, where she led numerous interdisciplinary studies on key social challenges such as access to justice, violent crime, and safety policy. Before joining the HSRC, she worked at the South African Truth and Reconciliation Commission and was one of the authors of the Commission’s final report.

Doriam Borges is an associate professor with the Graduate Program in Social Sciences at the State University of Rio de Janeiro and a researcher at the Laboratory of Analysis of Violence. He has research experience in the areas of violence, crime, and public security. He was the technical coordinator of the Victimization Survey in the Metropolitan Region of Rio de Janeiro that informs Chapter 5. Doriam is the author and co-author of various books and articles, including The Fear of Crime in the City of Rio de Janeiro: An Analysis of the Perspective of Beliefs of Danger (2012) and The Owners of the Hill: An Exploratory Analysis of the Impact of Pacifying Police Units in Rio de Janeiro (2012).

Chapter 6

Juan Pablo Pérez Sáinz has been a sociologist and a researcher at the Latin American Faculty of Social Sciences (FLACSO) since 1981. He currently resides in Costa Rica. Among his most recent publications are: A History of Inequality in Latin America: The Barbarism of the Markets, from the 19th Century until Today (Buenos Aires, Siglo XXI, 2016) and, as editor, Lives Besieged: Young People, Labor Exclusion and Urban Violence in Central America (San José, FLACSO, 2018).

Mario Zetino Duarte is research director of the José Simeón Cañas Central American University and associate researcher at FLACSO El Salvador. He has researched and published articles and books as an author, co-author, and translator on topics related to scientific research, youth culture, violence, criminology, citizen security, human mobility, and migration. Mario’s PhD in Sociology is from the University of Lund, Sweden.

Florencio Ceballos Schaulsohn is a sociologist with extensive experience working on citizen participation, access to information and communications technologies for poor and marginalised communities, and social and economic inclusion of youth affected by violence. From 2001 to 2005, he was director of Research at Ekhos I + C, a Santiago-based research and communications company, and taught at various universities in Chile and Costa Rica. He joined IDRC in 2005 as programme manager of the telecentre.org initiative and later became senior programme specialist for the Governance and Justice programme with a global portfolio. He holds a Master of Education Sciences from Université Paris 8 and pursued doctoral studies in Sociology at the École des Hautes Études en Sciences Sociales.

Chapter 7

Roberto Briceño-León is professor of sociology at the Central University of Venezuela and director of the Social Science Laboratory where, since 2005, he has coordinated the Venezuelan Violence Observatory. He has held positions at the Sorbonne in Paris and at the National Autonomous University of Mexico, and has been a research fellow at Oxford and at the Woodrow Wilson International Center for Scholars in Washington, DC. Roberto has served on the International Sociological Association Executive Committee, was global secretary of the International Forum for Social Science and Health, and directed the World Health Organization’s Small Grants Program for young researchers. He has published (or edited) 24 books in several languages and more than 200 scholarly articles.

Chapter 8

Darshini Mahadevia is a visiting professor at the School of Arts and Sciences, Ahmedabad University, Ahmedabad. She was formerly the director of the Centre for Urban Equity, CEPT University, from May 2016 to January 2019, and founder and coordinator of the Centre between 2008 and 2013. She served as the dean of the Faculty of Planning, CEPT University, from December 2012 to April 2016. Her research focuses on equity issues in urban areas, particularly housing, transport, employment, and land use planning. She holds a PhD from Jawaharlal Nehru University, a Master’s degree in Urban and Regional Planning from the School of Planning and Architecture, New Delhi, and a Diploma in Architecture from CEPT University. Darshini has more than 100 publications, books, book chapters, and journal articles to her credit. She is co-author of the Gujarat Human Development Report, 2004 and the Handbook of Urban Inequalities (New Delhi: Oxford University Press, 2012).

Renu Desai is an independent scholar based in Ahmedabad. She was senior research fellow at the Centre for Urban Equity, CEPT University, from 2013 to 2018. Her research examines urban transformation, urban informality, and housing in Indian cities, with a focus on questions of equitable development and urban citizenship. She holds a PhD in Architecture from the University of California, Berkeley. Renu has published numerous journal articles and book chapters, and is co-editor of Urbanising Citizenship: Contested Spaces in Indian Cities (New Delhi: SAGE, 2012).

Chapter 9

Rajith W. D. Lakshman is a research officer in the Cities Cluster at the Institute of Development Studies (IDS), UK. Before joining IDS, he was a senior lecturer in economics at the University of Colombo, Sri Lanka. He holds a PhD in Economics from the University of Melbourne, Australia, and has focused his research on migration (forced and voluntary), wellbeing, nutrition, and development. His current work looks at the livelihoods of internally displaced persons in Sri Lanka; wellbeing among urban informal settlers in Bangladesh and India; the modalities of reception in cities in Jordan and Lebanon, which affect the wellbeing of Syrian refugees and their host communities; and ways to work directly with municipal authorities participating in the Smart Cities Mission in India.

S. Irudaya Rajan is professor at the Centre for Development Studies (CDS), Thiruvananthapuram, India. At CDS (since 1998), he has coordinated eight major migration surveys in Kerala (with Professor K. C. Zachariah). He has further conducted or contributed to similar surveys in Goa, Tamil Nadu, Gujarat, and Punjab. He has published extensively in national and international journals on the social, economic, and demographic implications of international migration. He is currently engaged in several projects on international migration with the European Union, World Bank, and the UAE Exchange Centre, Abu Dhabi. He works closely with the Ministry of External Affairs, Government of India; the Department of Non-Resident Keralite Affairs, Government of Kerala; and the Kerala State Planning Board. He is editor of the annual series India Migration Report, published by Routledge, and editor-in-chief of the journal Migration and Development, published by Taylor & Francis.


Robert Muggah

The twenty-first century belongs to the city. The planet has undergone unprecedented urbanisation over the past half century. Today, well over half of the world’s population is urban, compared to less than 1 per cent in the seventeenth century. There are currently 31 mega cities of ten million people or more compared to just three in the 1950s. What is more, urbanisation is set to continue its relentless growth over the coming two decades. Future urban population expansion will be unevenly distributed: more than 90 per cent is predicted to occur in the so-called Global South, especially Africa and Asia.

The continued growth in the scale, number, and dispersion of cities is contributing to an “urban dilemma”. As I explained in a 2012 report for Canada’s International Development Research Centre (IDRC), the dilemma refers to the double-edged nature of rapid urbanisation. On the positive side of the ledger, urbanisation can be a powerful force for economic growth and pro-poor development. More negatively, urbanisation has the potential to increase protracted insecurity among the urban poor. This “darker side” of rapid and unregulated urbanisation could potentially disrupt its potential to stimulate growth, productivity, and wellbeing.

Not surprisingly, there is growing apprehension about the causes and consequences of the urban dilemma in multilateral and bilateral policy circles. United Nations agencies, the World Bank, and organisations ranging from the European Union to the Organisation for Economic Co-operation and Development have issued a number of reports on the implications of rapid urbanisation for stability and development. Many humanitarian and development aid agencies have elaborated policies and procedures to mitigate harms and harness the dividends. Military planners have also prepared new manuals to guide operations in urban warfare—considered by many to be the most likely battlefield of the future.

From 2012 to 2017, IDRC and the United Kingdom’s Department for International Development funded research to examine the urban dilemma, particularly in the Global South. The focus on cities in low-income and emerging economies was intentional. Indeed, there is comparatively limited awareness of the dynamics of urbanisation in the world’s fastest-growing cities. To begin filling this knowledge gap, the organisations commissioned dozens of researchers to undertake field-based and policy-oriented research on the intersections of urbanisation, poverty, inequality, and violence.

A key impetus guiding IDRC’s work on the urban dilemma was the desire to test novel theoretical approaches to appraising the key drivers of urban violence. Rather than relying exclusively on criminological and sociological models imported from the Global North—whether social disorganisation, institutional anomie, broken windows, or otherwise—IDRC sought to leverage new and alternative research methods to better understand the characteristics and dynamics of violence in cities of the Global South.

This ambitious edited volume by Jennifer Erin Salahub, Markus Gottsbacher, John de Boer, and Mayssam D. Zaaroura synthesises the outcomes of this IDRC-led process. It curates findings on the causes and consequences of urban dilemmas across more than 40 cities. It highlights the theoretical and methodological approaches adopted by a group of talented inter-disciplinary scholars heralding from universities and research institutes in the Americas, Africa, and Asia. Taken together, the collection lays out an innovative framework for scholars and, just as importantly, practitioners, to rethink how they conceive, diagnose, and respond to violence in cities of the South.

The edited volume adds to a growing body of scholarship produced by scholars and applied researchers in the Global South—especially Latin America. Indeed, there is a growing consensus on the core risks that give rise to urban violence, much of it based on quantitative and qualitative research pursued over the past decade. High on the list are social and income inequalities and concentrated poverty. Indeed, Latin America is home to ten of the world’s 15 most unequal countries (measured by their Gini scores) and 43 of the 50 most lethally violent cities in 2016. A group of Latin American researchers affiliated with the World Bank recently determined that a 1 per cent increase in teen pregnancies in the region is associated with at least 0.5 additional murders per 100,000 people.

Across Latin America, a context with which I am particularly familiar, scholars have also detected complex relationships between rapid and unregulated urbanisation and urban violence. Roughly 85 per cent of Latin Americans already live in cities, suburbs, and slums, some of them experiencing among the highest reported levels of criminal violence in the world. In Latin America, as in most places, homicidal violence tends to be hyper-concentrated spatially in peripheral urban areas that suffer from concentrated disadvantage. Cities, especially fast-growing ones, offer certain intrinsic opportunities for criminal activity (for instance, anonymity, prospective victims, and dilapidated infrastructure), compounded by economic neglect and scarce basic services.

In Latin American cities, violence can also be compounded by a higher density of actual and would-be offenders—particularly unemployed young men. At least 13 per cent of Latin America’s 108 million 15–24-year-olds are unemployed, which has encouraged a small number of them to commit “aspirational crime”. The available evidence, including that set out in this IDRC volume, suggests that Latin America’s murderers are, by and large, young, out of work, out of school, and out of options. In Brazil, for example, studies reveal that a 1 per cent rise in unemployment rates for men results in a 2.1 per cent spike in homicides.

Latin America’s alternately over-bearing and under-capacitated security and justice institutions are also responsible for aggravating urban violence. Law enforcement officers are frequently implicated in criminal activity, judicial systems are corrupt and over-burdened, and prisons are literally administered by gangs across parts of the northern triangle countries of El Salvador, Guatemala, and Honduras but also from Brazil and Colombia to Mexico and Venezuela. The mistrust in criminal justice institutions is profound, often in contrast to the perceived legitimacy of military entities and growing faith in authoritarian political figures.

The good news is that high-quality research—including the studies in this edited volume—can help identify entry-points to redress the urban dilemma and experiment with urban violence reduction. There are real and meaningful opportunities for South–South–North exchange and learning, particularly in relation to the common causes and consequences of violence in the city. The book foregrounds a range of exciting and intersectoral opportunities to improve safety and security in fragile cities. It is not so much a road map or prescriptive guide as a compendium of innovative and frequently counter-intuitive signposts.


Many hands make for light work, as the saying goes, and though the work has certainly not felt light at times, many hands have helped this book become a reality. We take this opportunity to thank the many supporters who have helped this culmination of the Safe and Inclusive Cities (SAIC) initiative come to fruition.

Beginning officially in 2012, SAIC passed through numerous hands during its five years of activity. In the years since, the editors and authors of this book and its sister volume, Social Theories of Urban Violence in the Global South: Towards Safe and Inclusive Cities (Routledge, 2018), have worked long hours, often while employed elsewhere, to ensure that the results of the initiative came together in a single reference. During that time, we have amassed a lengthy list of people to thank.

We acknowledge the unparalleled support of key colleagues associated with SAIC. Nola Haddadian, the International Development Research Centre’s publisher, pushed us forward, gave us encouragement, and spent weeks copy-editing chapters when many of us would have given up. Without her support, you would not be reading these pages. The Governance and Justice team, including Barbara Alves, Patricia Alves, Florencio Ceballos Schaulsohn, Kristin Corbett, Adrian Di Giovanni, Cam Do, Roula El-Rifai, Njeri Karuru, Veronique McKinnon, Martha Mutisi, Marie Renaud, Navsharan Singh, and Ramata Thioune, provided crucial support at key moments of SAIC’s history. We likewise acknowledge the support of colleagues at the United Kingdom’s Department for International Development: Iain King, Tom McEnroe, and Charlotte Morris, among others.

Providing intellectual guidance, offering words of wisdom, and challenging us to think critically along the way have been many colleagues across the globe. We thank, in particular, the many members of the 15 SAIC research teams, too numerous to name individually and scattered internationally in close to 20 countries; our co-authors rank chief among them. Professor Caroline Moser, SAIC’s strategic advisor, went above and beyond the call of her contract to mentor us and offer support, make connections, and inspire us to think creatively at key moments. Without Robert Muggah’s foundational literature review—our baseline—SAIC would not have flourished as it did. All research initiatives should have such a powerful springboard to start from.

Numerous colleagues, many of whom we now call friends, participated in SAIC workshops, outreach events, and the final SAIC conference, the SAIC Exchange 2017 (#SAICx17; www.safecitiesconference.com). They come from research, government, and policy backgrounds and are, regrettably, too many to name individually.

At Routledge, we thank Helena Hurd, Leila Walker, and their team for answering our many questions and for waiting so patiently for a manuscript. We thank several anonymous reviewers for their helpful feedback.

On a personal note, Jennifer—who has resolved never to try to write a book in her spare time again and so is unlikely to do so for quite some time, if ever—would like to thank Katy Nelson, for her unfailing support; her friends—her “urban family”—and extended family members for their patience and understanding; and, finally, James and Wynn McCrea, who make everything possible.

Markus is very grateful for all the encouragement and support from Jahel Itamar Garfias Jaramillo.

John dedicates this book to the incredible team at IDRC, with whom it was a privilege and honour to work.

This is Mayssam’s first book. She is eternally grateful for the support of her co-authors and that this milestone arrives at the same time as her first child, making this an exceptionally special year. She is very grateful to Jose Di Bella for his never-wavering love, encouragement, and support along both her professional and personal paths, and to Salwa Oneissi for always opening her world and pushing her forward to learn and experience more.

SAIC was made possible by the generous support of Canada’s International Development Research Centre (IDRC) and the United Kingdom’s Department for International Development (DFID). The opinions expressed in this work do not necessarily reflect those of IDRC or DFID. Responsibility for any errors or omissions of fact or judgement rests solely with us.

Jennifer Erin Salahub, Markus Gottsbacher, John de Boer, and Mayssam D. Zaaroura

Acronyms and abbreviations


African Development Bank


Agence d’Études et de Promotion de l’Emploi [Agency for the Study and Promotion of Employment] (Côte d’Ivoire)


Ahmedabad Municipal Corporation (India)


Ahmedabad Municipal Transport Services (India)


Nationalist Republican Alliance (El Salvador)


Batalhão de Operações Policiais Especiais [Special Police Operations Battalion of the military police of the State of Rio de Janeiro]


Bus Rapid Transit System (India)


Center for Epidemiological Studies-Depression scale


Department of Cooperative Governance and Traditional Affairs (South Africa)


Community Police Forum


Crime Prevention through Environmental Design


Centre for the Study of Violence and Reconciliation (South Africa)


Community Work Programme


development-forced displacement and relocation


Department for International Development (UK)


Democratic Republic of Congo


Expanded Public Works Programme (South Africa)


Revolutionary Armed Forces of Colombia


West African franc


focus group discussion


gross domestic product


Guwahati Municipal Corporation (India)


Ghana Statistical Service


internally displaced persons


International Development Research Centre


Instituto Nacional de Estadística [National Institute of Statistics] (Venezuela)


Institut national de la statistique [National Institute of Statistics] (Côte d’Ivoire)


Jawaharlal Nehru National Urban Renewal Mission (India)


Kreditanstalt für Wiederaufbau (German Development Bank)


Krishak Mukti Sangram Samiti (political organisation, India)


material quality of house index


non-governmental organisation


Safe and Inclusive Cities (programme)


Sustainable Development Goal


socioeconomic status


Socially and Economically Weaker Section (India)


Safe Node Area


Safe Node Area Committee


Syndicat national des transporteurs de marchandises en Côte d’Ivoire [National Union of Merchandise Transporters in Côte d’Ivoire]


United Nations Development Programme


United Nations Educational, Scientific and Cultural Organization


United Nations Office on Drugs and Crime


Pacifying Police Units (Brazil)


Victim Empowerment Programme (South Africa)


Violence Prevention through Urban Upgrading (programme, Cape Town, South Africa)


World Health Organization


A fast-moving and dynamic urban landscape

John de Boer and Markus Gottsbacher

Over three million people are moving into cities every week and municipal governments are underprepared for the coming population boom (World Bank 2018). The fastest growing metropolitan areas are in Asia and Africa. In fact, 90 per cent of all future population growth will occur in sprawling metropolises and small and medium-sized cities (Muggah 2017). According to the World Bank (2015), 60 per cent of the areas that are expected to be urban by 2030 have yet to be built and, by 2060, one billion new housing units will be needed to house the world’s growing population. By 2050, an estimated 86 of the world’s largest 100 megacities will be in Asia and Africa (Vidal 2018).

The sheer speed of urbanisation is outpacing the capacity of governments to keep up (Muggah 2017). Urbanisation is particularly rapid in fragile and conflict-affected countries. Over the past 40 years, the total urban population in fragile states has increased significantly (Bosetti, Cooper, and de Boer 2016). What were predominantly rural countries will become urban over the next 15 years—as illustrated in Figure I.1—and this has profound implications for development and security (see United Nations 2018a).

Cities such as Abidjan, Juba, Kabul, Kinshasa, Mogadishu, and Port au Prince are growing at unprecedented and unmanageable rates for what are already fragile settings. Experts fear that pressure from rapid urbanisation in these already stressed environments will severely destabilise countries, undermine state-building and reform efforts, and throw countries back into a spiral of violence (von Einsiedel 2017).

Risks facing twenty-first-century cities

What makes twenty-first-century cities unique compared with those of the past is that they are increasingly sites of converging opportunities and risks. Their hyperconnected and integrated systems enable new forms of finance, mobility, employment, and retail, health, and education services that open up the possibility of tackling some of our most pressing challenges. Yet, the interconnected and interdependent nature of our cities and their global reach also exposes them to a range of risks including cyber attacks, health pandemics, climate risks and natural disasters, extreme pollution, crime, and terrorist threats (Muggah 2016a, 2016b).

Figure I.1 Urban populations in selected fragile countries, 1965–2030.

Source: Bosetti, Cooper, and de Boer (2016).

Cities face a myriad of risks, with most of them combining both exogenous and endogenous characteristics (de Boer, Muggah, and Patel 2016). Internal risks include being situated in a floodplain or having access only to extremely poor or non-existent infrastructure. External risks include cyclones or conflict in neighbouring areas that can spill over in the form of unregulated migration and the associated political and social backlash. Importantly, city fragility is seldom the result of a monolithic root cause. Rather, the evidence suggests that the cumulative effects of overlapping political, social, economic, and environmental risks drive violence and victimisation in cities (Ambraseys and Bilham 2011). Moreover, recent research has demonstrated the importance of understanding how formal and informal institutions that govern decision-making in urban contexts influence the development trajectories of cities (Goodfellow 2014). Purely technical approaches that fail to take into account the political arrangements that determine who is able to access services, exercise their basic rights, benefit from the application of fundamental social norms, and prosper in a city will likely reinforce existing forms of exclusion and even exacerbate vulnerability.

Cities account for at least 70 per cent of total worldwide greenhouse gas emissions, are home to the majority of the world’s poor, and register higher homicide rates than rural areas. The ability of nation-states to tackle diverse global problems such as climate change, poverty, and fragility will depend on how cities are equipped to meet these challenges. Many mayors already recognise this reality and are teaming up to share innovative and data-driven solutions. Sustainable Development Goal 11—“Make cities and human settlements inclusive, safe, resilient and sustainable”—is emboldening city leaders to develop practical solutions to problems that so often divide and paralyse their federal counterparts (de Boer 2015).

One emerging risk is climate change, which is expected to drive human displacement to an unknown and unprecedented scale. Even the minimum projections in sea level rise put an estimated one billion people in Asia and Africa at risk. By some estimates, climate change will force up to 77 million urban residents into poverty over the next decade, reversing a long, hard-fought trend of global poverty reduction (de Boer and Patel 2016).

The rise in temperatures will also increase the habitable range of the disease-carrying Anopheles mosquito, creating greater risk that malaria and the Zika virus will spread to new regions. By 2050, the International Organization for Migration expects to see up to one billion people displaced from their homes by climate change. Complicating this further is the risk of pandemics, where cities could become centres of major outbreaks. We only need turn to the Zika and Ebola crises to see glimpses of what the future could hold, particularly when infrastructure and services are ill-equipped to cope with such outbreaks (de Boer and Patel 2016).

In cities affected by violence, fear has emerged as a key concern for many residents. Crime avoidance has become an important characteristic driving the behaviour of many individuals and communities in these cities. This has caused urban residents to modify how and when they move in the city: children have been prevented from attending school, those that can afford private security have built up veritable fortresses, and those who cannot afford to do so have developed coping strategies that include vigilantism and joining gangs or self-help community groups. Millions have also chosen, or been forced, to migrate elsewhere. The ultimate effect is the erosion of human capital and, in many cases, the decay of social cohesion within those cities.

The situation in some of the world’s most violent cities is stark. In Latin America, for example, levels of trust in one’s neighbours and institutions are shockingly low. Interpersonal trust averages just 17 per cent across the region and is as low as 3 per cent in Brazil (Latinobarometro 2016). Only two in ten Latin Americans have high levels of faith in their government. For Brazil, Chile, Colombia, Costa Rica, Paraguay, and Peru, the number is closer to one in ten (Latinobarometro 2016). Public opinion surveys consistently rank the capacity and integrity of judicial institutions, police, and public servants (municipal and national) as exceptionally low. Some 44 per cent of Latin Americans believe that the police are involved in crime (Muggah and Aguirre 2018). In Buenos Aires, less than 20 per cent of the population trust the police. In Lima and La Paz, that figure is less than 10 per cent and in Caracas it is 15 per cent (Ortega 2016).

A diverse array of structural, institutional, and proximate risk factors give rise to crime and victimisation in cities across the world. These factors can range from the abuse of alcohol to the availability of firearms and the abundance of drug trafficking networks, as well as systemic corruption and widespread inequality. They also include factors related to youth unemployment, gender discrimination and patriarchy, and the historical legacies of war and colonialism.

What makes cities unique is how the risks of violence and victimisation that they face tend to aggregate and converge, often with increasing intensity. Preventing risk aggregation requires a careful diagnostic of the particular factors that exacerbate insecurity and violence in a particular context combined with a plan to reduce these risks. Adopting a public health approach to understanding and preventing violence can often help. Whereas criminology tends to focus on the offender’s behaviour after a crime has been committed, public health approaches stress prevention and seek to minimise the mental and physical injuries that may result from a violent event. This means working with youth and other individuals at risk. It also requires focusing resources to address challenges in neighbourhoods where disadvantage is concentrated. Ultimately, it means identifying the risk factors facing residents in cities and working comprehensively with residents and authorities to take advantage of opportunities to address them.

Yet, if there is one central theme in this book, it is the importance of building social cohesion and trust within and between neighbourhoods. This finding adds to a chorus of recent research that stresses the same point: “social cohesion, trust, and a shared commitment to the community” are critical to violence reduction and social development in crime-ridden neighbourhoods, as experts such as Patrick Sharkey (2018) have documented with respect to violence reduction in New York City (cited in Gopnik 2018). Sharkey’s (2018) work demonstrates how “violence almost vanishes” in communities with high levels of social cohesion and trust. On the other hand, where there is concentrated poverty in addition to low levels of interpersonal and inter-communal trust, violent crime tends to be more rampant. Constructing and/or restoring the social fabric of a community is essential to reducing crime and violence. For this to happen, investments need to be made that enhance the ability of urban residents to build and rebuild the social support networks of their communities block by block.

Why is this book important?

Violence-reduction programmes in urban contexts are often imported from other contexts and implanted as one-size-fits-all models at the local level. Worse, this is done without taking into consideration the specificities of each host community. This book focuses on urban residents’ experience of initiatives designed and implemented in the Global South. Not only does it reflect on the importance of taking into account the local strategies of some of the people most vulnerable to multiple forms of violence in urban settings, it also highlights the importance of agency in contexts of vulnerability. Local residents are not passive victims; they are active participants in responding to violence. Resilience to urban violence can only be achieved when affected populations determine the policy agenda and interventions to be implemented in their communities. This volume offers a nuanced and deep understanding of local agency and why it is critical to sustained violence reduction.

As we noted earlier, cities have emerged as key economic, social, and cultural actors, yet continue to be mired in social segregation, inequality, and violence. Existing governance institutions are not always up to the task of delivering on their end of the social contract and, in many cases, lack the capacity and resources to plan over the long term or invest in programmes and infrastructure that build resilience. In many instances, decision-makers lack the political will to tackle the underlying structural causes of exclusion and violence.

In these circumstances, this book aims to provide answers to questions such as:

What alternative and innovative approaches to creating safe and inclusive cities in the Global South are policymakers experimenting with?

How are these new approaches altering the relationship between urban violence, poverty, and economic and gender inequality?

What strategies and initiatives to tackle these challenges are yielding results?

Are there lessons that can be learned for other contexts?

Answering these questions is particularly important in light of the Sustainable Development Goals (United Nations 2018b). As decision-makers move to implement these goals, they must be mindful of which policies and interventions work and which do not. The impact on the safety and security of large-scale programmes intended to create housing, improve access to services, and upgrade slums is not straightforward. In some cases, efforts to improve public housing and basic services in low-income settings have been captured by criminal markets. Gangs and militia have routinely influenced the management and delivery of such services, especially in informal settlements in Latin America and South Asia (Felbab-Brown 2013).

Nevertheless, there are also examples of how improvements in city housing, service delivery, and slum conditions yielded enhanced public security and urban safety (see Bauer 2010; USAID 2014; UNODC n.d.). Examples include Brazil’s celebrated public housing programme—Minha Casa Minha Vida (My Home My Life)—which aims to reach 6.75 million households, or 27 million people, by 2018.1 Mexico’s conditional cash transfer programme—Oportunidades—has generated important violence prevention effects, especially among intimate partners.2

The case of Medellin (Colombia) is another often cited example of urban upgrading leading to improvements in violence prevention and social cohesion. Under the rubric of “social urbanism”, Medellin’s plan integrated the poorest areas of the city with the centre and outlying areas. The city introduced a cable car and subway system, high-quality libraries and schools, and more participatory forms of engagement for the city’s 2.2 million residents. While replication of the Medellin experience in other cities is challenging, it does offer important and aspirational lessons to mayors, business people, and civil society leaders around the world.

Lessons from the failure of past policies have also been documented and need to be learned. This includes a continued tendency to implement mano dura policies, which, despite having produced little evidence of achieving their objectives (Muggah, Garzón, and Suárez 2018), are still being applied and reinforced. Part of this has to do with the fact that the majority of Latin Americans—61 per cent—favour a mano dura approach to public security.3 This viewpoint is consistent across the entire region regardless of whether the country suffers from high or low crime rates. This sentiment has been accompanied by a tendency toward the militarisation of police functions and repressive tactics across the region to restore public order. Honduras, El Salvador, Nicaragua, Colombia, and even Brazil have witnessed this first hand.

In El Salvador, the number of military personnel deployed in cities has increased from 876 to 13,827 over the past two decades. In El Salvador and other countries facing similar challenges relating to urban violence, the increasing militarisation of police functions has also led to a dramatic increase in the number of murders allegedly attributed to the police. Of the 4,004 homicides in El Salvador in 2010, 1 per cent (121) were alleged to have been committed by police officers; in 2017, that figure had risen to 10 per cent (412 of 3,954) (UNHRC 2018). In Brazil, police killings are also a significant problem with over 5,000 people killed by the police in 2017, up from approximately 4,000 in 2016 (Forum Brasileiro de Segurança Pública 2018).

Other important lessons include the failure of policies and programmes that apply a one-size-fits-all approach to crime and violence prevention. A key finding here is the importance of localising solutions in a way that is appropriate to the context and takes into account the political economy of violence. Initiatives are required that invite meaningful citizen participation in the design and implementation of violence reduction strategies. Additionally, the gendered dimensions of violence need to be considered fully. This book helps document these policy and practice failures and provides options for how societies and cities can escape the violence trap and build more inclusive cities that can prosper in the twenty-first century.

The origins and history of this book

From 2013 to 2017, 15 multidisciplinary teams of researchers in Latin America, South Asia, and sub-Saharan Africa worked in 16 countries and more than 40 cities as part of the Safe and Inclusive Cities (SAIC) initiative. Their common objective was to understand the drivers of urban violence, poverty, and inequalities as well as what works—and what doesn’t—in preventing violence in urban centres. With their research now complete, these teams have findings and recommendations that are relevant to cities of the Global North and the Global South, which include a variety of approaches to reducing violence. They share perspectives that emphasise public health, sociology, urban planning, and other approaches that aim to tackle structural and interpersonal violence and exclusion. In doing so, this book brings multiple perspectives and disciplines to bear in addressing a complex problem linked to a myriad of risk factors.

This book is the second and final volume resulting from a joint research programme funded by the Department for International Development (United Kingdom) and the International Development Research Centre (Canada). The first of the pair of books, Social Theories of Urban Violence in the Global South: Towards Safe and Inclusive Cities (Salahub, Gottsbacher, and de Boer 2018), sought to fill a theoretical and conceptual gap in the current literature on urban violence, which all but ignored contributions being made in the Global South.

This second volume focuses on practice. It describes and analyses the experience of researchers and practitioners applying these frameworks and approaches to better understand and respond to violence on the ground in their cities. In doing so, the book conveys new insights on the complexities of the linkages between inequality, exclusion, and violence, and represents an important addition to the contemporary literature on violence reduction.

Together, the two books serve to document and magnify the results of this global research programme. They aim to answer the following groups of questions:

1.On drivers of violence: What are the most important drivers of urban violence (social, cultural, political, economic, gender-based, etc.), and how are these both a cause and a consequence of poverty and inequalities? Why do communities that are comparable in terms of social and economic inequality and exclusion suffer different levels of violence?

2.On “what works and what doesn’t”: What are the most effective interventions—both formal and informal—for tackling the problems of urban violence, poverty, and inequalities, and why? How have these strategies improved the livelihoods and security of the most vulnerable and marginalised individuals and groups, in particular those of women, girls, and minorities? How have these strategies and interventions affected the legitimacy and accountability of public authorities?

3.On improving policy and practice: What conditions facilitate the development of effective policies and practices (both formal and informal) that promote security and protect the livelihoods of the most poor and vulnerable individuals and groups? How can incentives be framed to facilitate the development of effective policies and practices and to what extent—if at all—can these be replicated in other contexts?

In responding to these lines of inquiry, this book helps readers and researchers better understand key drivers of urban violence and how those drivers differ across continents and in cities as diverse as Rio de Janeiro and Jaffna. It also highlights promising interventions that can address everyday violence being experienced in Caracas, Colombo, and Cape Town. Finally, the book provides policymakers searching for ways to tackle some of their most pressing urban problems with a pathway towards more safe and inclusive cities.

What is the book about?

The chapters in this book contain both descriptive and analytical elements and examine what the authors’ analyses mean for understanding different manifestations of urban violence in the city and the specific cities studied. They provide new knowledge and insights on the types, drivers, and dynamics of urban violence, poverty, and inequalities within their research sites. In some cases, such as in Chapter 1 on Côte d’Ivoire, our co-authors investigate the links between these drivers for the first time. In this way, the book outlines critical policy and practice challenges that have worldwide implications. As the authors of the substantive chapters are based in the Global South, this represents a rare contribution to policy, practice, and academic discourse on urban violence.

The thematic scope of this book is far reaching. Without claiming to be exhaustive, we identify and tackle six major topics. They are:

the link between poverty, social inequality, and violence

the role of social programmes and urban upgrading in violence prevention

the link between the formal (official) and the informal (unofficial) in the city, giving particular weight to the role of the informal in reducing violence

the connection between social cohesion and violence prevention

a gendered perspective on the impact of policy and practice

the responses and agency of the most vulnerable populations affected by violence and exclusion.

In these ways, the research presented in this book places particular emphasis on approaches that: strengthen security and inclusion for the poorest groups; promote social empowerment at different life phases (youth, elderly persons, etc.); document the principal connections between violence, poverty, and inequality reduction; and, determine the varying impacts of violence on women and girls, not only as victims but also as agents for change. This new knowledge improves our understanding of the connections between violence, poverty, and inequalities and of efforts to reduce them in urban areas, in addition to helping to ensure that investments in development and security focus on, and are underpinned by, empirical evidence. Every chapter confronts several, if not all, of these elements.

Links between poverty, social inequality, and violence

In Chapter 2, Adobea Yaa Owusu, Martin Oteng-Ababio, George Owusu, Charlotte Wrigley-Asante, and Martin Wiredu Agyekum shed light on the important relationship between crime and poverty, using the examples of Ghana’s four largest cities. They question the universal applicability of policies and practices founded on studies of cities in the Global North that suggest neighbourhoods with higher levels of poverty experience more crime. Their research contributes to a more nuanced understanding of this relationship in the case of four major cities in Ghana: Accra, Kumasi, Sekondi-Takoradi, and Tamale.

In Chapter 6, drawing on research on municipalities in Costa Rica and El Salvador, Juan Pablo Pérez Sáinz, Mario Zetino Duarte, and Florencio Ceballos Schaulsohn criticise the somewhat homogeneous thinking that shapes national public policies as well as international practices and guidelines pursued by aid agencies. They discuss the need for a deeper understanding and characterisation of collective action, violence, and organised crime, along with differentiated interpretations of, and approaches to, overcoming the latter two.

Social programmes, urban upgrading, and violence prevention interventions

In Chapter 4, Malose Langa, Themba Masuku, and Hugo van der Merwe bring to our attention the ways in which criminal justice responses are insufficient in reducing violent crime. In so doing, they emphasise the importance of considering social violence prevention strategies. Their study of South Africa’s Community Work Programme brings important insights on the crucial role of such strategies, which have been underestimated with regard to their potential to curb urban violence and make communities safer.

In Chapter 3, Richard Matzopoulos, Kim Bloch, Sam Lloyd, Chris Berens, Jonny Myers, and Mary Lou Thompson focus on a well-known experience of urban upgrading in South Africa’s Cape Town, namely the Violence Prevention through Urban Upgrading (VPUU) initiative. The study finds that not only do interventions that build social and structural inclusion help to decrease crime and violence, but they also have desirable unintended outcomes such as improvements in mental health.

Connections between social cohesion and violence prevention

The authors of Chapter 5, Vanessa Barolsky and Doriam Borges, also examine the VPUU intervention and pair it with an initiative in Brazil to compare the influence of two major violence prevention interventions—in contexts of high levels of violence and inequality—on social cohesion and collective efficacy. Their research is based on long-term ethnographic studies of VPUU in two neighbourhoods of Khayelitsha township (Cape Town) and the Pacifying Police Units intervention in two favelas in Rio de Janeiro. Their findings suggest that, despite their successes in violence reduction, in some places or with some groups both interventions destabilised existing (informal) social networks that contribute to social cohesion, with negative results.

In Chapter 7, Roberto Briceño-León stresses the importance of informal institutionality as a way to curb violence. His findings, based on research in several Venezuelan cities, reveal the existence of a duality of norms resulting from formal and informal institutionality in zones of urbanisation. He argues that exclusion is not only spatial and territorial, but also social and regulatory. Informal institutionality is the practical response of urban communities to a lack of formal law and it can help to prevent or resolve conflicts outside the law.

Informal approaches to reducing violence and links to formality

Several chapters engage with the link between the formal and the informal in the city, placing particular emphasis on the role of the informal to reduce violence. For instance, Kouamé Walter Kra, the author of Chapter 1, highlights individual and collective solutions to overcome exclusion from the formal sector in the city. He looks at the role of the informal economy using the example of the life trajectory of a young man whose limited formal employment opportunities push him into a life of increasing violence centred on informal transportation hubs in Côte d’Ivoire. Kra thus sheds light on the links between the formal and the informal in urban contexts where a vast majority of urban dwellers are excluded from mainstream society. He also raises the important issue of criminalisation of informal strategies used by communities to cope with the daily challenges of poverty, exclusion, and violence.

Gendered perspectives

Many of our co-authors investigate the gendered dimensions of the impacts of urban policies and practices, in addition to the responses of the most vulnerable populations affected by violence and exclusion. In Chapter 8, Darshini Mahadevia and Renu Desai reflect on the impacts of urbanisation in India over the last two decades and how it has contributed to increasing marginalisation, conflict, and everyday violence. They describe the intrinsic competition for spaces in which the vulnerable—in particular, often-neglected women and girls—and the powerful negotiate their different needs and interests. Their chapter illustrates the challenges these populations face in their everyday lives, and the silence of policymakers vis-à-vis their rights to the city.4

Finally, in Chapter 9, Rajith W. D. Lakshman and S. Irudaya Rajan discuss the gendered impacts of forced relocation in cities. They study the relocation of residents of informal settlements in Colombo (Sri Lanka) and Kochi (India). While they show that improved housing is an important measure of an increased quality of life upon relocation, they argue that, on its own, it is not sufficient to address the negative consequences of development-forced displacement and relocation (DFDR).

What is the way forward?

The research findings from the multiple cities presented in this book make a crucial contribution to current debates on how to “make cities and human settlements inclusive, safe, resilient, and sustainable”, as articulated in Sustainable Development Goal 11 (United Nations 2018b). However, they are by no means all encompassing with regard to topics, territories, and populations.

There are several areas where deeper policy- and practice-relevant research would complement and strengthen the results presented here. We identify the following:

continuing to document and highlight the ever-changing needs, impacts, and responses of the most vulnerable vis-à-vis violence, exclusion, and poverty

defining innovative ways to link the so-called informal (based in community and society) to the formal (based in—mainly government—institutions) to create more inclusive cities

deepening research on the positive manifestations of social cohesion in urban contexts, and its relevance particularly for violence reduction and inclusion

supporting cutting-edge research-to-policy and -practice initiatives that contribute to the empowerment of the most vulnerable, especially on topics such as the gender transformative potential of cities (Moser 2016)

continuing to foster theoretical, conceptual, and methodological debates from and across the Global South and/or with the Global North on these topics.

We need to maintain strategies that reduce violence and inequalities. We need these strategies to complement each other, to take into account so-called formal and informal dimensions, and to take advantage of the best of both. We also need to heighten the global profile of existing and new ways to engage both formal and informal approaches to foster positive manifestations of social cohesion and enhance systemic approaches to tackle the manifold complexities of contemporary urban spaces.

This book brings together a global perspective and insights from researchers grounded in local urban contexts. It seeks to address key challenges to build safe and inclusive cities in practical ways. In so doing, it aims to overcome reliance on an abstract system of ideas that are admirable on paper, but found wanting when applied to the real-world challenges of cities faced with high levels of urban violence, poverty, and inequalities.


1See www.minhacasaminhavidainscricao.com/.

2See www.prospera.gob.mx/Portal/.

3A majority of the population in the following countries supports a mano dura approach: Costa Rica (78 per cent); Panama (77 per cent); Peru (77 per cent); Chile (75 per cent); Honduras (73 per cent); El Salvador (73 per cent); Uruguay (71 per cent); Guatemala (63 per cent). See Latinobarometro (2016) and UNHRC (2018).

4For a detailed discussion on the “Right to the City”, see Carrión and Erazo (2016).


Ambraseys, Nicholas, and Roger Bilham (2011). Corruption kills. Nature 469(7329): 153–155.

Bauer, Bettina (2010). Violence Prevention through Urban Upgrading: Experiences from financial cooperation. Frankfurt: KFW Entwicklungsbank. www.kfw-entwicklungsbank.de/Download-Center/PDF-Dokumente-Sektoren-Berichte/2010_03_Violence-Prevention_E.pdf.

Bosetti, Louise, Hannah Cooper, and John de Boer (2016). Peacekeeping in cities: Is the UN prepared? Tokyo: United Nations University Centre for Policy Research. https://cpr.unu.edu/peacekeeping-in-cities-is-the-un-prepared.html.

Carrión, Fernando, and Jaime Erazo (Eds.) (2016). El derecho a la ciudad en América Latina. Visiones desde la política [The right to the city in Latin America: Visions from politics]. Mexico City, Ottawa, and Buenos Aires: Universidad Nacional Autónoma de México, International Development Research Centre, and Consejo LatinoAmericano de Ciencias Sociales. www.flacso.edu.ec/portal/pnTemp/PageMaster/8mp0ap1b0ms2fzgx1gzumnbmie2cz1.pdf.

de Boer, John (2015). The sustainable development fight will be won or lost in our cities, 24 September. Geneva: World Economic Forum. www.weforum.org/agenda/2015/09/the-fight-for-sustainable-development-will-be-won-or-lost-in-our-cities/.

de Boer, John, and Ronak Patel (2016). Resilience in the future: 2025 and beyond. In International Federation of the Red Cross, World Disasters Report. Resilience: saving lives today, investing for tomorrow (190–221). Geneva: International Federation of the Red Cross. www.ifrc.org/Global/Documents/Secretariat/201610/WDR%202016-FINAL_web.pdf.

de Boer, John, Robert Muggah, and Ronak Patel (2016). Conceptualizing city fragility and resilience. Tokyo: United Nations University Centre for Policy Research. https://cpr.unu.edu/conceptualizing-city-fragility-and-resilience.html.

Felbab-Brown, Vanda (2013). Gangs, slums, megacities and the utility of population-centric COIN, 5 October. Washington, DC: The Brookings Institution. www.brookings.edu/research/interviews/2013/10/05-gangs-slums-megacities-utility-population-centric-coin-felbabbrown.

Forum Brasileiro de Segurança Pública [Brazilian Forum on Public Security] (2018). Segurança pública em números 2018 [Public security in numbers 2018]. São Paulo: Author. www.forumseguranca.org.br/wp-content/uploads/2018/08/FBSP_Anuario_Brasileiro_Seguranca_Publica_Infogr%C3%A1fico_2018.pdf.

Goodfellow, Tom (2014). Rwanda’s political settlement and the urban transition: Expropriation, construction and taxation in Kigali. Journal of Eastern African Studies 8(2): 311–329.

Gopnik, Adam (2018). “The great crime decline”. The New Yorker, 12 and 19 February. www.newyorker.com/magazine/2018/02/12/the-great-crime-decline.

Latinobarometro (2016). Informe 2016 [2016 report]. Santiago: Corporación Latinobarometro. www.latinobarometro.org/latNewsShow.jsp.

Moser, Caroline O. N. (2016). Gender transformation in a new global urban agenda: Challenges for Habitat III and beyond. Environment and Urbanization 29(1): 221–236.

Muggah, Robert (2016a). Cities are on the front-line of climate change. They must adapt or die, 27 October. Geneva: World Economic Forum. www.weforum.org/agenda/2016/10/cities-are-on-the-front-line-of-climate-change-they-must-adaptor-die/.

Muggah, Robert (2016b). Is urban terrorism the new normal? Probably, 17 January. Geneva: World Economic Forum. www.weforum.org/agenda/2016/01/is-urban-terrorism-is-the-new-normal-probably/.

Muggah, Robert (2017). These are the most fragile cities in the world – and this is what we’ve learned from them, 12 January. Geneva: World Economic Forum. www.weforum.org/agenda/2017/01/these-are-the-most-fragile-cities-in-the-world-and-this-is-what-we-ve-learned-from-them/.

Muggah, Robert, and Katherine Aguirre (2018). Which Latin American countries got safer in 2017? Americas Quarterly, 5 February. www.americasquarterly.org/content/which-latin-American-countries-got-safer-2017.

Muggah, Robert, Juan Carlos Garzón, and Manuela Suárez (2018). Mano Dura: The costs and benefits of repressive criminal justice for young people in Latin America. Rio de Janeiro: Igarapé Institute. https://igarape.org.br/wp-content/uploads/2018/04/Mano-Dura-The-costs-and-benefits-of-repressive-criminal-justice-for-young-people-in-Latin-America.pdf.

Ortega, Daniel E. (2016). Effectiveness versus legitimacy: Use of force and police training in Latin America, 5 January. Washington, DC: The Brookings Institution. www.brookings.edu/blog/up-front/2016/01/05/effectiveness-versus-legitimacy-use-of-force-and-police-training-in-latin-america/.

Salahub, Jennifer Erin, Markus Gottsbacher, and John de Boer (Eds.) (2018). Social theories of urban violence in the Global South: Towards safe and inclusive cities. London: Routledge.

Sharkey, Patrick (2018). Uneasy peace: The great crime decline, the renewal of city life, and the next war on violence. New York: Norton.

UNHRC: United Nations Human Rights Council (2018). Report of the Special Rapporteur on Extrajudicial, Summary or Arbitrary Executions on Her Mission to El Salvador. Advance Unedited Version, A/HRC/38/44/Add.2. www.ohchr.org/EN/HRBodies/HRC/RegularSessions/Session38/Documents/A_HRC_38_44_Add.2.docx.

United Nations (2018a). World urbanizaton prospects 2018. New York: Author. https://population.un.org/wup/.

United Nations (2018b). Sustainable Development Goals. https://sustainabledevelopment.un.org/?menu=1300.

UNODC: United Nations Office on Drugs and Crime (n.d.). A resource guide for municipalities: Community-based crime and violence prevention in urban Latin America (Draft 1). N.p.: Author. www.unodc.org/pdf/youthnet/tools_violence_prevention_handbook.pdf.

USAID: United States Agency for International Development (2014). Vanderbilt University examines impact of USAID’s crime and violence prevention efforts in Central America, 29 October. Washington, DC: Author. www.usaid.gov/documents/1862/vanderbilt-university-examines-impact-usaids-crime-violence-prevention-efforts.

Vidal, John (2018). The 100 million city: Is 21st century urbanisation out of control? Guardian, 19 March. www.theguardian.com/cities/2018/mar/19/urban-explosion-kinshasa-el-alto-growth-mexico-city-bangalore-lagos.

von Einsiedel, Sebastian, Louise Bosetti, Cale Salih, Wilfred Wan, and James Cockayne (2017). Civil war trends and the changing nature of armed conflict. Tokyo: United Nations University Centre for Policy Research. https://cpr.unu.edu/civil-war-trends-and-the-changing-nature-of-armed-conflict.html.

World Bank (2015). Investing in urban resilience: Protecting and promoting development in a changing world. Washington, DC: Author. https://openknowledge.worldbank.org/bitstream/handle/10986/25219/109431-WP-P158937-PUBLIC-ABSTRACT-SENT-INVESTINGINURBANRESILIENCEProtectingandPromotingDevelopmentinaChangingWorld.pdf?sequence=1&isAllowed=y.

World Bank (2018). Urban population (% of total). United Nations Population Division. World Urbanization Prospects: 2018 Revision. Washington, DC: Author. https://data.worldbank.org/indicator/SP.URB.TOTL.IN.ZS.

Part I

West Africa

1Controlling violence

An economic integration skill in Côte d’Ivoire

Kouamé Walter Kra


The results of the 2012 Agence d’Études et de Promotion de l’Emploi [Agency for the Study and Promotion of Employment] (AGEPE) employment survey estimate the employed population of Côte d’Ivoire at 9,492,150 persons or 65 per cent of the working-age population of 14,607,741, yielding an overall unemployment rate of 35 per cent. The unemployment rates for youth aged 14–24 and 14–35 are 13.8 per cent and 12.2 per cent, respectively. In addition, discouraged unemployed people1 represent 40.3 per cent of the total unemployed population. These figures demonstrate that unemployment—combined with the last two decades of sociopolitical instability (1990–2000, 2002–2011) and persistent poverty2 (INS 2015)—has become a complex reality and a structural problem for which employment policies have not been able to find a lasting solution. This may be because these policies only favour technocratic answers, such as the National Youth Fund, labour intensive works, etc., that are not based on a detailed analysis of socioeconomic integration for the unemployed.

In developing countries in general, and in West Africa in particular, the response to unemployment has been the subject of a literature structured around three axes:

1.the institutional response of governments to create the conditions for rapid growth, frequently with a strong focus on unskilled labour—a compromise that is only sometimes successful in the face of the scale and complexity of unemployment, as is the case in sub-Saharan Africa (Turnham 1993)—or the integration of the informal sector into the formal economy.

2.the collective response of the actors mobilising social, ethnic, and religious networks (e.g. the Yorubas of Benin/Nigeria and the Mourides of Senegal/Gambia) to conduct informal economic activities that generate employment (INS 2009a; Casson, Della Guista, and Kambhampati 2010; Benjamin et al. 2012).

3.the individual response: the work of individual actors who either leave the formal sector to create their own jobs—particularly in the informal sector where they are better off—or integrate into the informal economy (Maloney 2004).

Whether they take action individually or together, the actors experimenting with their own solutions to employment precarity or unemployment integrate into the informal sector in four main ways, particularly through the informal economy.

1.Some integrate themselves using informal financial borrowing from friends, relatives, or the tontine system, since bank loans are virtually inaccessible (Johnson 2004; Akoten, Sawada, and Otsuka 2006).

2.Some enter the informal sector through a mobile activity (street vendors and peddlers) or small trades (mechanics, carpenters, cabinet makers, sheet metal workers, etc.) practised mainly at home (Benjamin et al. 2012).

3.Some rely on the existence of social, ethnic and religious networks to acquire and exercise an economic activity (Benjamin et al. 2012).

4.Some refrain from registering with a state organisation, the tax administration, or other regulators or conceal part of their output although they are recognised by the administration (La Porta and Shleifer 2008).

The literature on methods of integration into the informal economy as an individual or collective response to unemployment or employment precarity has mainly focused on four forms of resources: borrowed financial resources, resources that stem from the nature of the activities in which they engage, sociocultural resources in terms of social and cultural capital, and diverted administrative resources. These resources have the common denominator of being external to the individual, who seizes them as opportunities for integration. The intrinsic means of the individual, not as an opportunity, but as autonomous resources mobilised as an individual response to the issue of unemployment or employment precarity, remain poorly documented. Research in this area could help in understanding individuals who enter the informal sector in response to unemployment as rational actors capable of invention or reinvention based on the autonomous resources at their disposal (Boudon 1979; Weber 2003). I seek to deepen this question by deciphering a new form of response to the endemic unemployment that Côte d’Ivoire is experiencing. In this chapter, I analyse violence as an economic integration skill through the experience of highly violent youth in the informal transport sector.

The concept of skill has been the subject of various approaches in the field of the sociology of work. Authors have analysed the relationship between skill and qualifications (Dubar 1996) and between skill and a job (Zarifian 1997). Others have critically questioned a sociological model of skill (Merchiers and Pharo 1992) or the widespread use of the concept (Dugué 1994). Although this analysis is inspired by the sociology of development, it tends to present skill as a catalyst for the formation—on a foundation of violence—of relationships of power or of intra- or inter-group domination for the control and management of an economic stake: informal transport hubs.3 From this point of view, I do not consider skill in the classical sense of a professional aptitude acquired through a process of instruction or qualifying training, which can be mobilised in the context of work in a company. Instead, I see it as a historically and socially constructed ability that the holder can use to respond to an individual or collective concern. This definition of skill leads us to the following question: how is violence established as a skill for these youth and how do they mobilise it to integrate into the informal transport sector?


The study was conducted in Abidjan, both the economic capital and largest city of Côte d’Ivoire, where the informal sector’s share of the local economic system was estimated at 76.7 per cent in 2002 (INS 2009b). The mobilisation of violence as an economic integration skill in this sector is observed mainly in greater Abidjan’s informal transport hubs. The city’s main informal transport hubs, shown in Figure 1.1, are:

Koumassi Grand Crossroads station

Siporex station in Yopougon

Abobo Roundabout station

Adjamé Liberté station

Bassam de Treichville station

Saint Jean de Cocody station.

Among these hubs, the Abobo Roundabout station was chosen as the primary study site because of the recurrence of violent clashes, characterised by higher intensity, compared to the brawls that take place at the other stations. Located north of Abidjan, about ten kilometres from the city centre, Abobo has more than a dozen precarious neighbourhoods and a shantytown that shelters about 60 per cent of the town’s population. Abobo is also a municipality subject to urban crime caused mainly by unemployment, poverty, lack of training for police personnel, proliferation of small arms, lack of material and financial means, and an insufficient number of law enforcement personnel, with only approximately 404 police officers for the entire municipality (UN-Habitat 2012). Finally, Abobo was one of the most violent social arenas noted during the post-electoral crisis of 2010–2011, being the site of both assassinations and massacres, in which youth took an active part (Varenne 2012).

The research team, of which I was part, used the biographical approach, scrutinising the trajectory of “Z” (the first letter of his nickname), a unionist4 chosen for his chaotic journey and his singular experience of crime in the informal transport hubs. Through lengthy non-directive interviews, the life story was used as a study technique to probe the different compartments of his life, focusing on the experiences of violence that marked his career. We also used the technique of the “commented photo album”. This consisted of making another appointment after the last interview, this time to go through his personal photo albums together and comment on them, emphasising images that may have a significant link with his economic integration through violence. For the purpose of triangulation, a snowball sampling approach was used to identify and interview other resources who had a good knowledge either of Z’s journey or the phenomenon of the criminalisation of informal transport hubs. The interviews were preceded by a session to provide information and an explanation of the research’s purpose and of the ethical precautions to preserve the anonymity and confidentiality of the informants, as the phenomenon being studied is particularly sensitive. For the data analysis, we used Tom Wengraf’s5 (Wengraf 2006) biographical narrative interpretative method, supplemented by an examination of the biographical logic and its modifications (Veith 2004) and by an understanding of the biographical course (Bertaux-Wiame 1991).

Figure 1.1 Map of the main informal transport hubs of Abidjan.

Source: Centre de cartographie et de télédetection/Bureau national d’études techniques et de développement [Centre for Cartography and Remote Sensing/National Bureau of Technical Studies and Development], 1996 and the research team’s own investigation, 2015. Design and Production Kouadio R. Oura and Kouadio A. Alla, 2015.

The results of our investigations are presented in this chapter in six sections. The first deals with Z’s social trajectory, emphasising the main stages that nourished his skill of violence. The second shows, through Z’s experience, the different forms of violence in informal transport hubs as well as the actors who perpetrate them. In the third section, I analyse how the context of unemployment and increasing poverty has favoured the use of violence as a means of economic integration, particularly among youth. The fourth section discusses the evolution of gnambro6 and unionist activities, showing how they have been criminalised through the emergence of a highly competitive game around the exclusive control of informal transport hubs. The fifth section analyses violence in the hubs, both as a source of fuel for other forms of violence building up in Ivoirian society and as an indicator of the reinvention of solidarity in the search for employment. Finally, the sixth section analyses the dynamics of perpetuation of the criminal economy at work in Abidjan’s informal transport hubs.

“Z”: from violence to integration in an informal transport hub

Born in Abobo, a working-class district of Abidjan, Z grew up with his parents in modest conditions. At the age of ten, he went to live with his uncle in a village in the sub-prefecture of Dabakala, in northern Côte d’Ivoire, where he carried out rough agricultural work. At the same time, he joined a group of peers, which he led and with whom he occasionally engaged in physical violence related to petty crime. But this propensity to violence increased, showing up in increasingly frequent fights with Z on the front line. This violence continued to the point where community members in his village discouraged their children from associating with him. The desire to earn money in order to acquire mystical powers and “to become even stronger” led Z, at the age of 25, to join the mass transport workforce, where he was hired and worked as a balanceur7 for eight years.

After this initiation stage, there were three high points in the dynamics of Z building his violence as a skill. The first concerns the acquisition of mystical resources (which he calls the code of “knowledge”) that granted him, and continue to grant him, a sense of power and invincibility. In this way, he acquired three types of “knowledge”: the mystical resources of protection against edged weapons (commonly called “anti-iron”), the mystical resources of protection against firearms (“anti-gun”), and the resources to defeat opponents in fist fights. He acquired them from his uncle in Dabakala, but also in Burkina Faso and Mali from other marabouts (Muslim hermits, or monks, known locally as the bouroula bara) to protect himself from the bandits who ruled in Abobo in the 1980s.

A second influential point relates to Z’s learning and practices of legal violence, through his experience as a security officer in a private security company in 1996. Among other topics, the training covered the handling of weapons (pump-action shotguns, revolvers firing rubber bullets, electric batons), the handling of edged weapons, techniques for combat, and activities such as guarding, bodyguarding, escorting, cash transportation, event security, etc. He then worked as a private security officer in Bouaké, the country’s second largest city, in a brewery and a cotton company.

But before reaching his position as a private security officer in Bouaké, Z joined the Black Tiger gang of Adjamé, a municipality of Abidjan, and, for seven months, had new experiences of criminal violence. These seven months corresponded to a period of unemployment preceding his new position in Bouaké. Joining the Black Tiger gang was the beginning of the third influential point in his trajectory. The gang’s mission was to regulate the loading and unloading of merchandise at the Gouro Market in Adjamé, a large commercial centre, mainly for the wholesale trade of food products from the interior of the country. The gang performed this activity for financial benefits, based on criminal violence. On this subject, Z told of a punitive expedition that the Black Tigers launched against the Abrassiens, a rival gang known to be violent, in retaliation for an attack against one of their own.

One day, the Abrassiens fell upon a friend like that, in Adjamé under the bridge. But when he escaped, he came and told us that they had attacked him there and he could not defend himself. At night, we robbed all the maquis [restaurant-bars] in Abrasse. We closed them all that day and emptied their cash boxes. Three of us entered a maquis, and one person remained at the door while the other two carried out the operation. We had no business with the ordinary customers; we were dealing with the Abrassiens. We made the Abrassiens leave the maquis. If they refused, we broke a bottle, slit their throat with it, and that was that.

Still during this third influential point, Z’s experience and acquisition of experiences of criminal violence with the Black Tigers were reinforced when he was recruited into the Forces Nouvelles rebellion in 2002. As part of this armed movement attempting to overthrow the regime in power, he said he participated in military operations in Abidjan, attacking military camps where he “slaughtered enemies” and took their weapons to strengthen the arsenal of his brothers-in-arms. Later in 2010, in the wake of the post-election crisis that turned into an armed conflict between pro-Ouattara and pro-Gbagbo forces, he was recruited once more as a member of the Invisible Commandos.8 During the post-electoral crisis, he also acted as a bodyguard for certain political figures and was mobilised from time to time to take part in strike attacks or counterattacks “sponsored by politicians”.

Z shifted into the activity of informal transport because of the precariousness of his guard work in Bouaké. He began his activities at the Abobo Roundabout station as a “trainee”, a kind of observer who came every day to immerse himself in the management and operation of the station. Then, one day, his skills in violence earned him the recognition of his union and allowed him to rise in rank to become an “element manager” with a “collection day” each week. Unions are mostly informal organisations that illegally control and exploit informal transport hubs (see Figure 1.2). Their members are called syndicats or syndicalistes [“unionists”] in the local slang. Z is a “route manager”, responsible for recovering from the “element manager” the illegal taxes taken from the drivers every evening. His ability to cause harm also led him to become a recruiter of fighters and leader of the theatre of operations during the attacks/counterattacks launched for the control of the hubs. He tells how he rose in the union:

At first, I observed the collection of taxes, until I earned my own collection day. It was during a clash between my group and another group of unionists. This enemy group wanted to coerce us. They came and told us, “You will not work!”. Yet we had occupied the space before them. They came and wanted to take this territory back from us. When I got the information, I told the children [alluding to the gnambros], “Tomorrow at this time of night, we must regroup at this place”. I employed my knowledge [mystical preparation] to give to the members of my group with whom I had to go to the site. I made them a potion in a small container, and everyone washed themselves with it. Guns will not penetrate that! You can carry your rifle, but we’ll tear it from you. We went to fight, and it was clean! [We were victorious]. When the leaders of my group saw me at work, they quickly gave me a day in the week for collection.

Between 2002 and 2011, many ex-combatants and fugitives from justice followed a trajectory closely similar to Z’s. Their presence and action in the informal transport hubs have made these places arenas for the production of criminal violence, the main manifestations of which are explained in the next section.

Demonstrations of violence in informal transport hubs

Z’s experience is indicative of a phenomenon that involves two categories of youth (unionists and gnambros) mobilising violence as a skill to integrate into and remain in the urban economic fabric. This is a recurring phenomenon that reflects the criminalisation of informal transport hubs in the city of Abidjan. During each episode of violence, control of these spaces or some of their sectors is the main reason for the clashes. Victory opens the way to capturing significant financial windfalls generated by the exploitation of the hubs. In the absence of official statistics and given the hidden nature of the management of this windfall and the discretion surrounding it, it is difficult to measure, even for only one hub.

In the informal transport hubs, three main actors interact on a daily basis: the gnambros, in charge of finding customers; the drivers, charged with driving the customers; and the unionists, who ensure that the drivers’ rights are respected. But the eruptions of violence generally pit the gnambros against one another, the unionists against one another, or both categories of actors against other joint forces of gnambros and unionists. In the management hierarchy of an informal transport hub, unionists are above gnambros (Figure 1.2). Thus, there is a union president responsible for dealing with public authorities in case of disputes, in order to maintain good relations for the survival of the organisation. Below the President is the boss or union manager who supervises the activities and serves as a link between the President and the General and the generals. The General is responsible for protecting the hub and organising its activities. He ensures that information is passed up and plans attacks to control new spaces or other hubs. The generals closely guard the General but have no decision-making power in the union organisation. Below the generals, are the route managers charged with recovering the funds collected in the territory each evening. The element managers are chosen by the route manager and have one day a week when they supervise the collection of the “taxes” and turn them over to the route manager. The elements and winzins, grouped under the name gnambros, along with the element managers, are not part of the union organisation. But, the gnambros are in charge of looking for customers and collecting “taxes” from drivers and giving them to the element managers. The generals mobilise them to lead raids of conquest, reconquest, or counterattack on other hubs.

It should be noted that women are not included in the groups of gnambros and unionists. Informal transport hubs are highly masculine arenas where men are able to control power through violence. Women in or around hubs are, as a rule, traders who suffer the effects of clashes (assault, theft, etc.). They are not directly involved in managing the hubs and even less in episodes of violence.

The main perpetrators of violence, gnambros and unionists, mainly consist of demobilised former combatants, fugitives from justice, former gang members, and homeless individuals commonly called bacrôman. Increasingly, elementary and high school students are joining these unions. To become a gnambro, one must fulfill basic criteria, the most important of which is the ability to be aggressive or warlike. With regard to the union organisation, membership is based on possessing a reputation for criminal exploits and an ability to mobilise combatants or lead expeditions for mass violence.

Figure 1.2 Organisational chart of an informal transport hub.

Source: Safe and Inclusive Cities programme survey data, 2015.


  Those who have decision-making power.

  Those who execute the orders of the decision-makers.

Violence in informal transport hubs usually manifests in two forms. The first concerns one-on-one fights between a unionist and a gnambro, or between two gnambros, usually with edged weapons (knives, machetes, etc.), which, in many cases, cause serious injuries. These fights often occur in the quest for customers (informal transport users), with each of the belligerents claiming “paternity” or ownership of the customer. In general, this type of violence lasts only a few minutes and ends with one of the fighters being victorious. However, it can sometimes degenerate into mass violence.

The second form of violence, the most significant, takes on a collective character and occurs either when one-on-one fights degenerate or are poorly controlled, or when a group of gnambros or unionists launch an attack on a hub to conquer or reconquer it. This second form of violence often pits groups of gnambros and/or groups of unionists against one other and is part of a dynamic of attack/counterattack. The attacks are aimed at specific targets, are organised, and revolve around four stages:

1.the recruitment of several fighters (gnambros and other categories of actors such as the “Microbes”9)

2.the grouping of combatants

3.the arming of the fighters

4.the raid.

The fighting can last for several days, alternating between continuous and sporadic clashes that ultimately result in very serious injuries and deaths. Mass fights are known for their extreme violence, which often results in heavy loss of life in the belligerent camps. In their dynamics, these clashes migrate from central hubs, which are the most economically profitable, to the less-profitable secondary hubs, while decreasing in intensity. But, it is the resistance capacity of the attacked group that ultimately determines the duration of the fighting.

In both one-on-one and mass conflicts, the belligerent actors mobilise four types of resources. First, there are the edged weapons, composed of forks (with sharpened and twisted teeth), clubs with spikes, knives, machetes (double-edged), babylones (heavy steel chain), small axes, grés (animal horns), brass knuckles, slingshots, and caiman teeth (a kind of stake). Then there are the firearms: Kalashnikovs, sawed-off shotguns, and automatic guns. They also use mystical weapons: rings, amulets, necklaces, bracelets, etc., that make them feel invulnerable to firearms, machetes, and knives, and as though they have an exceptional capacity to harm. Finally, they mobilise, either as mercenaries or in a spirit of solidarity, human resources outside the group of combatants who are designated in the community as reinforcements.10

By acting as the perpetrators of violence in informal transit hubs, unionists like Z create power and influence (De Latour 2001) which, in turn, maintain this violence. Their selection of gnambros to collect illegal taxes and enlist combatants for clashes maintains networks of actors engaged in the control of these disputed spaces. The union hierarchy strategically manipulates these networks of actors while channelling potential disputes, a prerequisite for keeping control of the spaces already conquered and potentially conquering others (Clegg 1989). Violence thus appears as a resource for the collective or individual acquisition of market share in Abidjan’s informal transport hubs. It also appears to be a necessity to preserve, maintain, and even increase their share of the market that generates substantial revenues for the actors involved in the conflicting game of organising and controlling the hubs. Ultimately, this violence is a sort of “employment insurance” that guarantees economic activity for everyone in a country where unemployment and poverty persist.

Violence as a guarantee of employment in an anomic context

The violent competitive game of the informal transit hubs, a source of economic integration and of substantial revenue, has emerged in a three-fold context of sociopolitical crisis, a drained economy, and deep poverty. On the political front, the decade of politico–military crisis (2002–2011) was marked by the use of auxiliaries by the belligerent military forces: the Forces Nouvelles, as the armed rebellion is known, and government forces for the 2002–2007 period, as well as the pro-Gbagbo forces and pro-Ouattara forces for the 2010–2011 period. However, the end of the hostilities and the economic recovery that followed11 did not offer real opportunities for economic integration for these ex-combatants. The mixed results of the various disarmament, demobilisation, and reintegration programmes confirm these findings. These ex-combatants have logically increased the size of the unemployed population with a particular expertise: the control of violence. Many of the unionists and gnambros, like Z, are ex-combatants from this decade of armed crisis. Faced with a labour market that offers this category of actors little or no real opportunity for integration, as well as the limits of the government’s ex-combatant reintegration policy, many of them have found an alternative to unemployment in the informal transport hubs. They seized this opportunity with a single essential skill: violence.

Economically, the decade from 2002 to 2010 exacerbated unemployment and poverty. Côte d’Ivoire experienced an energy crisis in the form of rolling blackouts and the decline of the production of petroleum products. The growth of the national economy slowed because of the post-election crisis of 2010. GDP decreased by 0.4 percentage points to 2.4 per cent at the end of 2010, compared to 3.8 per cent in 2009 (AfDB 2011). The main primary, secondary, and tertiary sectors have been affected by this crisis. In addition, insecurity in production areas, as well as dysfunctional distribution and marketing channels, have reduced all the gains accumulated since the signing of the 2007 Ouagadougou Political Agreement12 (AfDB 2011). At the end of February 2011, inflation had increased to 5.1 per cent (compared to –0.4 per cent in 2010), prices of major food products had increased by more than 25 per cent in some cases (fish by 26.5 per cent, oils and fats by 25.3 per cent) and government revenues decreased significantly, notably due to the ban on coffee and cocoa exports in the first four months of 2011 (AfDB 2011).

At the social level, the results of the Survey on Household Standard of Living (ENV2008) conducted by the Institut national de la statistique (INS) show an increase in income poverty from 38.4 per cent in 2002 to 48.9 per cent in 2008 (INS 2009a). Although the poverty rate at the national level dropped to 46.3 per cent in 2015 (INS 2015), it still remains high. This reflects the fact that nearly one in two inhabitants of Côte d’Ivoire is poor. These results also indicate that 10 per cent of these poor people are considered to be the poorest, with an extreme poverty line of FCFA101,826 per capita per year (or about US$171). The average household size is 4.7 people, with an average of six people in poor house-holds. The city of Abidjan is by far the area where poverty has increased the most, with 21.0 per cent living in poverty in 2008 compared to 14.9 per cent in 2002, an increase of 6.1 percentage points. Abidjan greatly contributes to a national poverty rate of 8.9 per cent.

In this worrying context, informal transport hubs offered ex-combatants like Z a social arena for reinventing the activities of gnambros and unionists by giving them a more violent, and especially criminal, character. This is an alienating situation in which these actors innovate to meet a need for economic integration through the use of illicit means (Merton 1997). Unemployment in Côte d’Ivoire has led these categories of actors to enter the informal transport market, where they have mobilised their ability to commit crimes as a skill to ensure and sustain their integration. The violence produced to guarantee employment in these economic spaces in conflict is therefore not a spontaneous phenomenon. It was built on the framework of the history of unions in this sector of activity. One of the key steps in this story is the multiplication of unions, which opened the era of confrontational and violent competition for control of the hubs.

From the plurality of unions to the advent of bloody competition

The phenomenon of gnambros and unions emerged in the 1990s, in a context of sociopolitical change characterised by the birth of the multi-party system that led to union liberalisation. Alongside the single transport union, the Syndicat national des transporteurs de marchandises en Côte d’Ivoire (SNTM-CI, National Union of Merchandise Transporters in Côte d’Ivoire), new unions emerged giving rise to a competitive game centred on the collection of daily boarding taxes in the informal transport hubs. The gnambros and unionists were then mobilised by these organisations to collect and guarantee their share of taxes, with each union organisation having its own “collection day” at a given hub. At that time, up to the beginning of the 2000s, violence already existed in the informal transport hubs. But it was expressed mainly in the form of social tensions between the actors exploiting them (Steck 2004) and focused on the issue of negotiating or renegotiating the spaces. It also manifested through acts of verbal violence and petty crime: pickpocketing of transport users, small fights between balanceurs targeting the same passengers, etc.

However, the desire for more “collection days” introduced a logic of power relations into the competitive game. These power relations resulted in recurring clashes between gnambros and/or unionists, no longer just idle or delinquent youth, but including new actors: ex-combatants, habitual offenders, and ex-gang members. The expression of these clashes has thus shifted over the years, from simple fights to particularly bloody brawls resulting in serious injuries and deaths. These criminal clashes are guided by a new issue: control of the informal transport hubs and the financial income generated by their exploitation (see Figure 1.3).

Figure 1.3 Evolution of violence in informal transport hubs.

Source: Safe and Inclusive Cities programme survey data, 2015.

This shift from “simple violence” to criminal violence in the hubs occurred in the favourable context of the decade of armed crisis. This period generated non-reintegrated ex-combatants and promoted the escape of many inmates from several prisons, while also aggravating the country’s economic situation—already bleak in the aftermath of the 1999 coup d’état—and accentuating youth unemployment and poverty among the population. Z’s experience is indicative of this criminalisation of the informal transport hubs, which has spread over the years to several municipalities in the city of Abidjan (see Figure 1.1).

The role of unionists and gnambros like Z as social actors bringing about change has been instrumental in this evolution. The contextual elements described above are similar to social forces that have altered the life trajectories and behaviours of these two categories of actors. Directly or indirectly, their everyday experiences and perceptions have been shaped (Scott 2008), eventually leading them to define criminal violence as a response to unemployment. In other words, unionists and gnambros are social actors who, like hundreds of thousands of other young Ivoirians, are confronted with structural unemployment. But they processed the information received from the sluggish socioeconomic context differently and, consequently, favoured a response based on the use of violence, their only skill. They challenge, from below, the Ivoirian system of employment.

Moreover, in a context of habitus (Bourdieu 1980), resorting to crime in response to unemployment reflects the dispositions instilled in these youth (i.e. unionists and gnambros) through their experiences of violence. For example, Z had these experiences as a part of the Black Tiger gang, the 2002 armed rebellion, and the Invisible Commandos in 2010. Through a specific process of desocialisation/resocialisation to which they were exposed during these experiences, they internalised violence as a form of preferred response, which they tend to reproduce by adapting it to their unemployed status. Here, violence is the result of a model of behaviour transposed to the field of informal transport. It is a model, moreover, that shows these youth to be competent and capable social actors (Giddens 1986). They are, in short, individual or collective actors endowed with expertise in violence and possessing the capacity to embody their experiences to solve the problem of economic integration that they face.

The evolution of the competitive game between gnambros, between unionists, and between the two categories of actors, from a stage of simple tension to one of criminalisation in the span of a decade, also took place in a context of strong, uncontrolled urbanisation. The socioeconomic infrastructure construction process and urban development actions in the city of Abidjan were accompanied by the “abandonment” of several spaces that initially remained “empty” and were later privatised to serve as informal transport hubs. Z, like many other youth with the same profile, seized the opportunity offered by the lawlessness of these spaces that had become residues of uncontrolled urbanisation to achieve his economic integration. This dynamic reveals, moreover, another facet of these hubs: privatised and coveted public spaces that, through the violence occurring there, have not only become arenas of reinvention for the social codes of informal transport and employment, but also laboratories for the reinforcement of other forms of violence that are shaking Ivoirian society.

Informal transport hubs: a breeding ground for multi-faceted violence and social reinvention

The criminalisation of informal transport hubs has created a breeding ground for multi-faceted violence. In other words, it serves as a springboard for other forms of violence in Ivoirian society. These acts of violence include political violence, a form of violence perpetrated in the political arena centred on victory in municipal, legislative, regional, or presidential elections or, more broadly, in power relations between opposing sides, to conquer a political power (Crettiez 2008). According to our interviews with several sources, candidates in the 2011 legislative elections and the 2012 municipal elections mobilised groups of gnambros or unionists to wreak havoc at their opponents’ campaign headquarters. In addition to paying those involved in such expeditions, the candidates promised to repay them with direct support to ensure the continuation of their activities in the mass transport hubs. In some parts of his life story, Z claims to have been repeatedly asked to engage “his men” in this type of political violence.

The second form of violence fuelled by the criminalisation of informal transport hubs is the phenomenon of the Microbes. Field data show that clashes between gnambros, between unionists, or between the two categories of actors constitute opportunities for these youth to acquire new experiences of crime and reinforce their violent skills. Recruited at times as reinforcements, the Microbes participate actively in bloody clashes and “pay themselves on the ground” by taking advantage of the chaos to physically attack and rob users, residents, or drivers of mass transport vehicles. These experiences, renewed during the frequent clashes between groups at the hubs, contribute to increasing the capacity for harm of the Microbes and thus perpetuate this phenomenon, which persists in Abidjan.

Beyond multi-faceted violence, the reinvention of the response to the issue of unemployment has generated a new form of social organisation in the informal transport hubs. The regulation of the transport activity is supported and supervised, not by legal structures, but by actors with a high capacity to cause harm. Abidjan’s informal transport hubs are now regulated by criminal violence, through which the strongest or most harmful conquers the hub, dictates its rules, and controls its management. This form of social organisation—a source of insecurity for the general public—perpetuates itself by feeding on this regulatory violence, the incapacity of public authorities (the Ministries of Transport and Interior and municipalities) to stem it sustainably, and, especially, the income captured daily by the actors involved. Although it generates significant resources for the state, this entire sector of the informal economy escapes its control through the phenomenon of gnambros. Ultimately, the action of unionists and gnambros to solve their unemployment problem has paradoxically brought about another phenomenon detrimental to society: the insecurity of users, residents, and drivers of transport vehicles in mass transport hubs, who are often victims of bloody clashes.

In addition, with the phenomenon of mobilising reinforcements, the criminalisation of informal transport hubs has become an indicator of the reinvention of solidarity in the process of economic integration of youth. Traditionally, in the Ivoirian context, this solidarity consisted of elders (father, mother, sister, brother, aunt, or uncle) using their network in the public or private sector to place their son, daughter, brother, sister, niece, or nephew in a job. But alongside this family/community solidarity manifested by means of relational resources, another form of solidarity has developed, guided by the principle of mutual aid in and through violence. In the process of looking for work, solidarity in its traditional sense was already symptomatic of a reversal of value. Getting a job for an unemployed person was more a matter of relational capacity than of merit, hence the concepts of “séphonisme” or more recently “rattrapage” (both Ivoirian slang for finding work or appointing people from one’s family, community, village, or ethnic group). With solidarity in and through violence as practised in mass transit hubs, there is another facet of this reversal of value. In such a social arena, obtaining a job is not about relational capacity, let alone merit, but rather about the capacity for collective harm, the aggregated criminal potential.

Thus, employment in this industry is regulated by violence. It offers itself to actors capable of imposing themselves by force, to those with a greater capacity for harm than other actors in search of economic integration. Once conquered, employment is also maintained by force, giving rise to a reinvention of the response to unemployment. In this respect, our results differ from those of our predecessors on the question of strategies or policies for the integration of youth in general, and troubled youth in particular. For youth in general, their economic integration is made possible through a personal downgrading process (having an a priori level of training higher than that required for the job held) that gives them access to employment and mobility in the labour market (Forgeot and Gautié 1997), through associative programmes based on the principle of mutual aid (Comeau 2000) or an intermediate strategy of institutional arrangements put in place by public authorities (Lefresne 2003). For troubled youth, including migrants, delinquents, and youth in conflict with the law, there are two approaches. The first approach specifically affects migrants who can use their ability to adapt over time (Renaud, Piché, and Godin 2003) or family, ethnic, or national social links as strategies/resources to access employment (Öztürk 2006). In the second approach, that relating to youth in conflict with the law, economic integration can be achieved with the help of a constructive pedagogy that allows youth to build a life project and take control of their destiny (Association jeunesse et droit 2002), or a sponsorship system that involves entrepreneurs and elites in the integration process (Clément and des Sceaux 2006).

Cullen, Williams, and Wright (1997) cautioned against any action programme that sees employment as a panacea for delinquency. I join them in stating that employment can even be a source of violence in some cases. This is reflected in the criminalisation, through the game of conflictual economic integration, of the informal transport hubs and the social transformations it has brought about in Côte d’Ivoire. We are now faced with a dynamic and complex phenomenon, which also seems to be establishing itself for the long term.

Towards the sustainability of the criminal economy in informal transport hubs

The phenomenon of the criminalisation of Abidjan’s informal transport hubs shows a mobilisation of violence as a guarantee of employment and an instrumentalisation of this violence in the political game, all of which contribute to its perpetuation.

Indeed, since the two categories of actors (gnambros and unionists) successfully mobilise violence, a bid for economic integration, they will tend to maintain this violence by acting on the factors that can promote its expression. The practice of paying bribes (“contributions” in the jargon of the informal transport hubs) to the police to avoid or delay their interposition between belligerents during a confrontation is one of the factors. These also include the use of reinforcements and the recruitment of the Microbes, a growing criminal category, to amplify each side’s capacity for harm. To maintain violence in hubs, gnambros and unionists will also tend to act on material or immaterial resources that contribute to this capacity for harm. Z and the other gnambros and unionists we spoke to place great importance on these resources in the informal transport environment. Above all others are the mystical resources that these actors use, both in Côte d’Ivoire and in the countries of the sub-region (Burkina Faso, Mali, Niger, etc.). These resources also consist of the invention of edged weapons, often “with supernatural powers” or a “rare or non-existent antidote”. Several examples were discussed earlier, but there is another, recently discovered and known to be easily accessible, particularly effective, and without antidote: the broken tonic bottle (named after a refreshing drink), which can be acquired for FCFA200 (less than one dollar).

The analysis also shows that political actors are exploiting the criminal potential of Abidjan’s informal transport hubs for electoral purposes or in power struggles. Logically, these political actors, or political leaders in general, will seek to accommodate the status quo by avoiding attempts to permanently stem the phenomenon of the criminalisation of informal transport hubs. Here, the notion of status quo integrates the permanent tension prevailing in the hubs with a high risk of confrontation versus the absence of a real policy to stop this phenomenon. But it also incorporates occasional interventions by the police between the belligerents in the case of clashes, the purpose of which seems to be closer to political marketing than a real desire to stem the phenomenon. This accommodation embraces a form of paradox. Leaders or public authorities give a speech on resolving the phenomenon without force, for which they claim not to have the means: “It is not by force that we will solve this problem. We are not ready to go after the gnambros”, as the Minister of Transport told the National Union of Journalists of Côte d’Ivoire (30 April 2015). Instead, they favour a reintegration policy for gnambros and unionists to “gradually eliminate” the phenomenon. Paradoxically, it persists to the point that almost not a month goes by without deadly clashes erupting in informal transport hubs. For example, a series of clashes that resulted in injuries, and sometimes deaths, broke out between gnambros and/or unionists after the Minister’s media appearance. These attacks were followed by major clashes: 30 September 2015, in Treichville; 4 November 2015, in Koumassi; 2 September 2016, in Adjamé; 8 November 2016, in Abobo; and 2 December and 24 December 2016, in Treichville again.

The logical consequence of the mobilisation of violence as a guarantee of employment and the instrumentalisation of this violence in politics is the perpetuation of the phenomenon of the criminalisation of informal transport hubs. This perpetuation is inexorably accompanied by two major problems: the insecurity of the vast majority of mass transport users and the recurrent compromise of their mobility. Informal mass transport serves the majority of mass transport users in Abidjan, far ahead of SOTRA, a state-owned urban transport company which is finding it increasingly difficult to meet the demand for collective mobility. The weakness of its fleet of buses and water taxis has considerably reduced its public transport capacity over the years, due to a lack of renewal and insufficient maintenance. In addition, SOTRA is often criticised for the irregularity and overloading of its vehicles, recurring accidents, and breakdowns in transit. This public transport supply deficit has legitimised the alternative mobility offering structured around informal transport hubs. Informal modes of transport, relatively accessible to the less affluent populations, grew from 6 per cent to 17 per cent of the urban transport market share between 1988 and 2002 (Kassi 2007: 3). If crime in the hubs is perpetual, then the physical integrity of the users and even of local residents will be constantly threatened by thefts, rapes, and assaults, of which they are often the victims or “collateral damage”. Similarly, each outbreak of violence would result in complete or partial paralysis of mass transport in Abidjan, as is often already the case.


This study aimed to analyse the violence mobilised by Z—a unionist working in an informal transport hub in Abobo—as a skill to achieve economic integration in the face of endemic unemployment in Ivoirian society. A biographical approach based on the life story allowed us to scrutinise the trajectory of Z, whose skill in violence was built through lived experiences, particularly his membership in Adjamé’s Black Tiger gang, his integration into the armed rebellion of 2002, and his participation in the urban guerrilla activities of the Invisible Commandos in the post-election crisis of 2010.

Z’s experience shows that actors in search of work resort to differential responses in a country where unemployment remains a structural problem. This unemployment is characterised by both a persistent mismatch between supply and demand for work and the absence of a real policy designed to absorb graduates, dropouts, or people without education and, most importantly, ex-combatants and habitual criminals who aspire to integrate into the economic fabric. Above all, Z’s experience shows that he and many other unionists and gnambros have come to define violence as a response to unemployment. This has led to the criminalisation of informal transport hubs across Abidjan. In its dynamics, this criminalisation has generated social transformations in the management of these disputed spaces, paving the way for a possible perpetuation of the phenomenon.

Previous research on the means of economic integration of youth highlight four types of resources: institutional resources (public integration policies); associative, community, family, or ethnic resources; pedagogical or educational resources; and personal or intrinsic resources in terms of the adaptability of the actor searching for economic integration. Our results differ from the first three types of resources. They are similar to the fourth (personal or intrinsic resources), but are distinguished by the form of the economic integration resource analysed: violence as an integration skill. This result reflects the emergence of another finding on the issue of the economic integration of the unemployed. It reveals that in addition to conventionally known resources, there is another form of intrinsic, self-generated resources that allows this category of actors to integrate into the economic fabric: the capacity for harm and the violence acquired as a skill for accessing employment.

Under these conditions, it can be assumed that, in the long term, the informal transport sector will become the preserve of one or a few groups made up of actors with a profile or trajectory similar to that of Z. This law of the strongest or the most violent would paradoxically make this sector not an economic space for the absorption of youth in search of an income-generating activity, but an arena where a natural selection takes place that limits access to employment. These groups could, moreover, become an obstacle to modernisation of the informal transport sector if this modernisation puts their interests at risk. Therefore, further research may be needed to deepen the analysis of the potential risks associated with the modernisation of informal transport, with the understanding that the actors or groups of actors who—like Z—are flourishing economically will likely have interests that are incompatible with the concept of modernisation.

Union organisations that are illegally constituted, and legitimised by violence and their ability to force themselves on disputed spaces that are economically profitable, raise the problem of insecurity in the informal transport sector in Abidjan. Whether experienced or felt, insecurity is a real challenge to ensuring the long-term mobility of users. To meet this challenge, we recommend that decision-makers delegitimise, above all, these groupings of unions and gnambros. This delegitimisation will have to occur in three main stages. First, the authorities must regain control of the hubs, based on the absolute need to no longer make this sector an issue of political power. This requires the public authorities to abandon the laissez-faire attitude that they seem to have adopted so far in the face of the phenomenon and bring, instead, an adequate and lasting response. Second, there must be a movement toward the necessary modernisation of the urban informal transport sector. The idea is to restructure mass transit hubs into economic assets that are free of violence and provide jobs. This modernisation will have to be done in a balanced way, preserving the gains in terms of the cost of transport access for the less well-off, the safety of the users, and—to the extent possible—the security of the pool of youth who live on the income from this sector. Finally, the public transport service offered by the state must be developed to make it both a factor in improving the mobility of users and a sector that absorbs former unionists and erstwhile gnambros.


This research was co-funded by Canada’s International Development Research Centre (IDRC) and the UK’s Department for International Development (DFID) within the framework of the international programme of research on urban violence, poverty and inequality, called Safe and Inclusive Cities. It also benefitted from the active partnership of the University Alassane Ouattara in Côte d’Ivoire and the UNESCO Chair in Bioethics attached to this university. The author would like to thank the principal investigator of the Ivoirian team, Professor Francis Akindès, for his scientific supervision; Jennifer Salahub, principal administrator of the programme; and Ramata Molo Thioune, senior programme specialist, for their collaboration and commitment, as well as Dorgelès Gbèkè and Gervais Coffi for the collection of field data.


1The discouraged unemployed are those who wish to work and are available to do so but who report that they are no longer looking for work because the prospects of finding it seem too low.

2According to the results of the Enquête sur le niveau de vie des ménages en Côte d’Ivoire (ENV 2015)—Profil de la pauvreté [Survey on the standard of living of households in Côte d’Ivoire (ENV 2015)—Poverty profile] conducted by the Institut national de la statistique (INS 2015), the poverty rate is estimated at 46.3 per cent, compared to 48.9 per cent in 2008. This shows that although poverty is certainly declining, it does persist. Abidjan is the leading contributor to poverty with 9.3 per cent of poor people, compared with 7.5 per cent for Haut Sassandra, 5.7 per cent for Tonkpi, and 5.3 per cent for Gbêkê. The analysis of employment in relation to poverty status indicates that more than a third of the employed population is poor (35.5 per cent). These data are as recent as possible in a country where national statistics on socioeconomic sectors (poverty, unemployment, employment, health, etc.) are not systematically updated every year, or even every two or three years. See www.ins.ci/n/templates/docss/env2015.pdf.

3These are physical spaces that, in most cases, are in the public domain, but that are forcibly occupied by groups of youth to turn them into economic hubs where private mass transport vehicles load and unload users in an informal transport system.

4In their own parlance, unionists are those in charge of the management and illegal control of informal transport hubs.

5See Tom Wengraf (2004). The Biographic-Narrative Interpretive Method—Short-guide. NCRM Working Paper (unpublished). http://eprints.ncrm.ac.uk/30/.

6At informal transport hubs, gnambros are responsible for illegally collecting loading fees from the drivers and handing them over to the managers. They are also mobilised for clashes between unions where, taking advantage of the disorder, they assault and rob users.

7A balanceur is a young man who escorts passengers on a mass transport bus, under the authority of the driver. His role is to search for passengers, get them aboard, and collect transit fees.

8An armed group that harassed pro-Gbagbo forces in Abobo and gradually deployed to other municipalities to take part in the “Battle of Abidjan”, which led to the arrest of Laurent Gbagbo on 11 April 2011.

9“Microbes” are children and youth between the ages of ten and 20 who, in groups of four to more than 20, commit acts of physical aggression, often fatal, on ordinary people on the streets or in homes for the purpose of stealing money or material goods (smartphones, cell phones, jewellery, etc.). See Akindès (2018).

10Reinforcements are recruited from among ex-gangsters or ex-combatants from the various armed crises that the country has experienced. They are endowed with dexterity in combat, a capacity for handling weapons, and mystical powers. They are financially supported by the category of actors (gnambros or unionists) soliciting them in case of clashes. They are mobilised among groups of gnambros or unionists operating in a hub other than that which is the object of the clashes. In this case, they come to assist their “friends” in danger. At times, the Microbes are found among them.

11The economy improved considerably after each period of crisis with growth rates between 2 and 3 per cent from 2008 to 2010 and between 8 and 9 per cent from 2012 to 2014 (World Bank 2015).

12The peace agreement signed between the Ivoirian Government and the rebel Forces Nouvelles to end the 2002 military–political crisis.


AfDB: African Development Bank (2011). The ADB and Côte d’Ivoire: Enhanced cooperation to end the crisis. Abidjan: Author. www.afdb.org/fileadmin/uploads/afdb/Documents/Publications/C%C3%B4te_Ivoire_anglais.pdf.

AGEPE: Agence d’Etudes et de Promotion de l’Emploi [Agency for Study and Promotion of Employment, Côte d’Ivoire]—Ministère en charge de l’emplois (2016). Côte d’Ivoire—Enquete nationale sur la situation de l’emploi 2012. Abidjan: Author. www.ins.ci/n/nada/index.php/catalog/78.

Akindès, Francis (2018). Understanding Côte d’Ivoire’s “Microbes”: The political economy of a youth gang. In Jennifer Erin Salahub, Markus Gottsbacher, and John de Boer (Eds.), Social theories of urban violence in the Global South: Towards safe and inclusive cities (161–182). London and Ottawa: Routledge and International Development Research Centre (IDRC).

Akoten, John E., Yasuyuki Sawada, and Keijiro Otsuka (2006). The determinants of credit access and its impacts on micro and small enterprises: The case of garment producers in Kenya. Economic Development and Cultural Change 54(4): 927–944.

Association jeunesse et droit (2002). Le succès de l’insertion des jeunes en difficulté [The successful integration of youth in difficulty]. Journal du droit des jeunes 3(213): 8–11. www.cairn.info/revue-journal-du-droit-des-jeunes-2002-3.htm.

Benjamin, Nancy, Ahmadou Aly Mbaye, Ibrahima Thione Diop, Stephen S. Golub, Dominique Haughton, and Birahim Bouna Niang (2012). The informal sector in Francophone Africa: Firm size, productivity and institutions. Washington, DC: Agence Française de Développement and World Bank. http://documents.world-bank.org/curated/en/737371468202815116/pdf/699350PUB0Publ067869B09780821395370.pdf.

Bertaux-Wiame, Isabelle (1991). Analyse du récit de vie et paradigme indiciaire [Analysis of the life story and the index paradigm]. Etudes et Séminaires 8: 84–105.

Boudon, Raymond (1979). La logique du social [The logic of the social]. Paris: Hachette.

Bourdieu, Pierre (1980). Le sens pratique [Practical sense]. Paris: Minuit.

Casson, Mark C., Marina Della Giusta, and Uma S. Kambhampati (2010). Formal and informal institutions in development. World Development 38(2): 137–141.

Clegg, Stewart R. (1989). Frameworks of power. London: SAGE Publications.

Clément, Pascal, and Garde des Sceaux (2006). Le parrainage, une nouvelle dynamique pour l’insertion des jeunes sous main de justice [Sponsorship, a new dynamic for the integration of youth under the hand of justice]. Les Cahiers Dynamiques 37(1): 4–7.

Comeau, Yvan (2000). L’insertion de jeunes adultes par l’économique: les expérimentations du mouvement associatif [The insertion of young adults by the economic: the experiments of the associative movement]. In Geneviève Fournier and Marcel Monette (Eds.), L’insertion socioprofesionnelle. Un jeu de stratégie ou un jeu de hasard? [Socioprofessional insertion. A game of strategy or a game of chance?] (125–142). Quebec: Les Presses de l’Université Laval.

Crettiez, Xavier (2008). Les forms de la violence [The forms of violence]. Paris: La Découverte.

Cullen, Francis T., Nicolas Williams, and John Paul Wright (1997). Work conditions and juvenile delinquency: Is youth employment criminogenic? Criminal Justice Policy Review 8(2–3): 119–143.

De Latour, Élaine (2001). Métaphores sociales dans les ghettos de Côte-d’lvoire [Social metaphors in the ghettos of Côte-d’lvoire]. Presses de Sciences Po 2: 151–167.

Dubar, C. (1996). La sociologie du travail face à la qualification et à la compétence [The sociology of work in the face of qualification and skill]. Sociologie du travail 2(96): 56–73.

Dugué, Élisabeth (1994). La gestion des compétences: Les savoirs dévalués, le pouvoir occulté [Skills management: Devalued knowledge, hidden power]. Sociologie du travail 3(1): 273–292.

Forgeot, Gérard, and Jérôme Gautié (1997). Insertion professionnelle des jeunes et processus de déclassement [Professional integration of youth and downgrading process]. Économie et statistique 304–305: 53–74.

Giddens, Anthony (1986). The constitution of society: Outline of the theory of structuration. Cambridge: Polity Press.

INS: Institut national de la statistique (2009a). Enquête sur le niveau de vie des ménages—Profil de la pauvreté (ENV2008) [Survey on household standard of living in Côte d’Ivoire—Poverty profile (ENV2008)]. Abidjan: Author. www.ins.ci/n/templates/docss/env2008.pdf.

INS: Institut national de la statistique (2009b). Étude sur le secteur informel à Abidjan en 2008 [Survey on the informal sector in Abidjan in 2008]. Abstract. Abidjan: Author. www.ins.ci/nada1/ddibrowser/?id=9/.

INS: Institut national de la statistique (2015). Enquête sur le niveau de vie des ménages en Côte d’Ivoire (ENV 2015)—Profil de la pauvreté [Survey on household standard of living in Côte d’Ivoire (ENV 2015)—Poverty profile]. Abidjan: Author. www.ins.ci/n/templates/docss/env2015.pdf.

Johnson, Susan (2004). Gender norms in financial markets: Evidence from Kenya. World Development 32(8): 1355–1374.

Kassi, Irène (2007). Régulations des transports populaires et recomposition du territoire urbain d’Abidjan [Regulation of popular transport and recomposition of the urban territory of Abidjan]. PhD thesis in Geography, Université Bordeaux 3.

La Porta, Rafael, and Andrei Shleifer (2008). The unofficial economy and economic development. Brookings Papers on Economic Activity 2: 275–364.

Lefresne, Florence (2003). Les dispositifs d’insertion professionnelle des jeunes [Methods for the professional integration of youth]. In Florence Lefresne (Ed.), Les jeunes et l’emplois (86–108). Paris: La Découverte.

Maloney, William F. (2004). Informality revisited. World Development 32(7): 1159–1178.

Merchiers, Jacques, and Patrick Pharo (1992). Eléments pour un modèle sociologique de la compétence d’expert [Elements for a sociological model of expert skill]. Sociologie du travail 34(1): 47–63.

Merton, Robert King (1997). Structure sociale, anomie et deviance [Social structure, anomie and deviance]. In R. Merton (Ed.), Eléments de théorie et de méthode sociologique (163–187). Paris: Armand Colin.

Öztürk, Ceren (2006). Les stratégies d’insertion des “jeunes turcs” de France [Integration strategies for the “Young Turks” of France]. Revue européenne des migrations internationales 22(3): 179–201.

Renaud, Jean, Victor Piché, and Jean-François Godin (2003). L’origine nationale et l’insertion économique des immigrants au cours de leurs dix premières années au Québec [The national origin and economic integration of immigrants during their first ten years in Quebec]. Sociologie et sociétés 35(1): 165–184.

Scott, James C. (2008). Weapons of the weak: Everyday forms of peasant resistance. New Haven and London: Yale University Press.

Steck, Jean-Fabien (2004). Requiem pour une gare routière: succès et fragilités d’un territoire informel à Yamoussoukro [Requiem for a bus station: Success and fragility of an informal territory in Yamoussoukro]. Autrepart 4(32): 95–114.

Turnham, David (1993). Employment and development: A new review of evidence. Paris: OECD Development Centre Studies.

UN-Habitat (2012). Côte d’Ivoire: Profil urbain d’Abobo [Côte d’Ivoire: Urban profile of Abobo]. Nairobi: Author. https://unhabitat.org/books/cote-divoire-profil-urbain-dabobo-french/.

Varenne, Leslie (2012). Abobo-la-guerre: Côte d’Ivoire: Terrain de Jeu de la France et de l’ONU [Abobo-of-War: Ivory Coast: Playground of France and the UN]. Paris: Broché.

Veith, Blandine (2004). De la portée des récits de vie dans l’analyse des processus globaux [The significance of life stories in the analysis of global processes]. Bulletin de Méthodologie Sociologique 84(1): 49–61.

Weber, Max (2003). Economie et société [Economy and society]. Paris: Pocket.

Wengraf, Tom (2006). Interviewing for life-histories, lived situations and personal experience: The Biographic-Narrative Interpretive Method (BNIM) on its own and as part of a multi-method full spectrum psycho-societal methodology. Unpublished. www.uel.ac.uk/wwwmedia/microsites/cnr/documents/Wengraf06.rtf.

World Bank (2015). Performances économiques Ivoiriennes depuis la fin de la crise post-électorale [Ivorian economic performance since the end of the post-electoral crisis]. Abidjan: Author. http://documents.banquemondiale.org/curated/fr/132981468230123662/pdf/956010FRENCH0W0te0March020150Final0.pdf.

Zarifian, Philippe (1997). La compétence, une approche sociologique [Skill: A sociological approach]. L’orientation scolaire et professionnelle 26(3): 429–444.

2Opening up or closing off

Urbanisation, violent crime, and the “poverty penalty” in Ghana’s four largest cities

Adobea Yaa Owusu, Martin Oteng-Ababio, George Owusu, Charlotte Wrigley-Asante, and Martin Wiredu Agyekum


Ghana has experienced unprecedented levels of urbanisation over the last half century. Between 1960 and 2010, the country’s urban population more than doubled, from 23.1 per cent to 50.9 per cent (GSS 2012; World Bank 2014). This urban population growth is estimated to increase to 65 per cent by 2030 (World Bank 2014). Crossing the 50.0 per cent urban population growth mark in 2010 was unprecedented, presenting several socioeconomic implications (Songsore 2009; Obeng-Odoom 2013; Songsore et al. 2014). Among these, crime is rising in tandem with urbanisation (Oteng-Ababio et al. 2016a), without the governmental apparatus having the commensurate ability to adequately provide the necessary protection for citizens and property. As a result, an intense fear of criminal victimisation has arisen in urban Ghana (GSS 2010; Songsore et al. 2014; Frimpong 2016). On the other hand, the World Bank (2014) has highlighted a positive aspect of rapid urbanisation in Ghana—the reduction of poverty. Unfortunately, this urbanisation-driven poverty reduction also shows a trend of continuing the sharply skewed income disparities/inequities sociospatially—between households, neighbourhoods, and rural and urban areas (UNDP 2007; Aryeetey, Owusu, and Mensah 2009; World Bank 2014). Several authors have therefore drawn a correlation between urbanisation, poverty, and crime victimisation in Ghana (Owusu et al. 2015; Oteng-Ababio et al. 2016a).

In this context, we report findings based on the primary data collected by the Safe and Inclusive Cities (SAIC) team based in Ghana for the “Exploring Crime and Poverty Nexus in Urban Ghana” project. Principally, the team studied crime in Ghana’s four largest cities (Accra, Kumasi, Sekondi-Takoradi, and Tamale), which constitute about 40 per cent of the nation’s population (GSS 2012), and tested whether the predominant theories of crime, mostly from the Global North, apply in the Ghanaian context. Based on the three different socioeconomic neighbourhoods studied in each of these cities, the project primarily examined the spatial relationship(s) between poverty and the incidence of crime (Owusu et al. 2015: 10).

Our findings mostly confirm the “poverty penalty” regarding violent crime victimisation in the cities and communities studied. We found that our low and, particularly, middle socioeconomic status (SES) neighbourhoods experienced more sophisticated and more successful violent crimes such as armed robbery, shootings, strangulation, rape, and murder, compared to the high-SES communities. Conversely, respondents in the high-SES neighbourhoods had the highest police presence. They also seemed to have the greatest ability to muster and deploy personalised/individualised sophisticated crime prevention and mitigation apparatus, including paying bribes to the local police service for increased surveillance in their communities.

In fact, respondents in the middle-SES communities seemed to have experienced more dangerous crimes, compared to those in the low-SES areas. Understandably, respondents in the middle-SES communities were found to be the most vulnerable to violent crime as they lacked both the individual and communal ability as well as the apparatus to prevent/fight crime, compared to those in the high- and low-SES communities. Our findings showed that respondents in the middle-SES communities do not have the more sophisticated crime-fighting apparatus that their high-SES counterparts employ. Nor do they have the community-initiated attempts to ward off crime that the low-SES areas utilise—for example, community watch committees that keep an eye on the neighbourhood and report any suspicious behaviour to the authorities are rare. Middle-SES community respondents lack the sophisticated crime-fighting apparatus of the high-SES areas and community-initiated attempts to ward off crime, such as community watch committees that keep an eye on the neighbourhood and report any suspicious behaviour to the authorities, are rare, unlike in low-SES areas. Middle-SES communities also lack social cohesion, which is known to help prevent crime (Owusu et al. 2016; Gafa 2017).

Additionally, a relationship between SES and reporting crime to law enforcement authorities was discovered. Respondents from high-SES areas reported crime to the police the most often, those in the low-SES areas showed the second highest level of reporting, and those in middle-SES communities reported crime the least. While there was a general perception of low confidence in the police, respondents in the middle-SES communities may be exposing themselves to more violent crime by reporting crime to the police the least. They are possibly giving the erroneous impression that they are safer and need less police surveillance, while having a higher appeal to criminals compared to low-SES communities.

We also found gender and city-level differences in the types of violent crime experienced, and in the reporting of criminal victimisation to the police and other community-level leaders. Furthermore, our findings showed that other indigenous arrangements for resolving crime-related victimisation exist in the communities studied.


Our operational definition of crime is a breach of an established penal code or, in the absence of linking it to a well-established penal system, a contravention of a defined community’s previously agreed/established mores. Underpinning this understanding is a definition of crime as “an action or omission which constitutes an offence and is punishable by law” and/or “an action or activity considered to be evil, shameful, or wrong” (Simpson and Weiner 1989). We also use the World Health Organization (WHO 2002: 4) definition of violence:

… the intentional use of physical force or power, threatened or actual, against oneself, another person, or against a group or community, that either results in or has a high likelihood of resulting in injury, death, psychological harm, mal-development or deprivation.

Caldeira (2012), Muggah (2012), and Oteng-Ababio et al. (under review) have noted that this definition of violence overlaps the definition of crime. The US Department of Justice’s Federal Bureau of Investigation (FBI 2011: 1) defined violent crime as “offenses which involve force or the threat of force”. Examples of violent crimes, as defined by the FBI (2011: 1), are “murder and non-negligent manslaughter, forcible rape, robbery, and aggravated assault, followed by the property crimes of burglary, larceny-theft, and motor vehicle theft, but … not [including] arson”. It is important to note that some property crimes are considered violent crimes.

Violent crimes were operationally defined as those that involve the use or threat of force against another individual, group, or property and/or such attacks that inflict physical and psychological/emotional harm on an individual or a group of persons and that involve using force. For example, stealing a parked motorbike was considered non-violent, while snatching a motorbike from the owner was considered a violent crime.

Ecological concentration of crime and the nexus between crime, poverty, and urbanisation

Several authors have observed the ecological/geographical concentration of crime, i.e. that a relationship exists between the size of a city and crime rates (Owusu et al. 2015; Oteng-Ababio et al. 2016a; Rosenfeld, Vogel, and McCuddy 2018). The literature has even established that certain types of houses are unique to particular neighbourhoods and that certain housing-neighbourhood types attract some crimes disproportionately (Cozens, Hillier, and Prescott 2001: 225–226; Sohn 2016; Gafa 2017). For instance, Gafa’s (2017) study in Accra concluded that compound houses/shared homes, common to low-SES neighbourhoods, are typically more criminogenic than other housing types (apartments, detached homes and multi-storey buildings) usually found in middle- and high-SES neighbourhoods in Ghana. Owusu et al. (2015) and Wrigley-Asante et al. (2016) found that certain types of crimes occur more frequently in neighbourhoods of a particular status than in others in Ghana. Patino et al. (2014) found similar results in Medellin, Colombia. Similarly, Sohn (2016) concluded that median income levels for neighbourhoods correlated with crime.

Such studies have also often concluded that urban crime is coterminous with poverty. Hipp and Yates (2011), Taniguchi and Salvatore (2012), Sohn (2016), and Oliveira, Bastos-Filho, and Menezes (2017) have drawn a positive correlation between poverty/low family SES or poor residential neighbourhoods and crime. McIlwaine and Moser (2003) linked youth violence to chronic and vicious poverty. Wirth (1938) is arguably the earliest known researcher to have drawn a connection between urbanisation, poverty, and crime, arguing that urbanisation erodes the familial bonds and the concept of neighbourliness. In the Ghanaian context, studies by both Oteng-Ababio et al. (under review) and Frimpong (2016: 97) concluded that, in the urban communities of Tamale and New Takoradi, poverty and unemployment were important underlying factors in youth “disorder” in these cities. Nevertheless, a direct, causal crime–poverty link is not straightforward, because not all poor individuals and communities are criminogenic (Fafchamps, Gabre-Madhin, and Minten 2005).

The crime prevention through environmental design (CPTED) planning strategy has often been used both theoretically and empirically to explain the differential criminogenic levels and crime prevention potentials of urban neighbourhoods, particularly based on their SES, design and land use in the correlation between neighbourhoods, crime prevention, and crime levels (Grohe 2011; Cassidy et al. 2014; Patino et al. 2014; Sohn 2016). Cassidy et al. (2014) explain how an improvement in facilities, such as maintenance of neighbourhoods, including CPTED measures and lighting, helps to decrease crime. Sohn (2016) studied defensible and permeable design of urban spaces and their effect on crime in Seattle, concluding that density and diversity of neighbourhood land use such as bus stops, average number of storeyed buildings, and providing more streets for better linkages, had an adverse effect on neighbourhood crime levels.

Advancing the “broken windows theory”, a variant of CPTED that argues “that visible cues of physical and social disorder in a neighbourhood can lead to an increase in more serious crime”, Patino et al. (2014: 48) note that the outward appearance of a neighbourhood mirrors its socioeconomic status in Medellín, Colombia. They found that urban layouts that were more disordered—crowded and cluttered, with small houses, narrow roads, and minimal green spaces—had higher rates of homicide.

Although UN-Habitat (2012: 69) acknowledges that a correlation exists between poverty/deprivation and crime, Patino et al. (2014) mention that cities such as Cairo, Hong Kong, New York, and Singapore are exceptions to this association. Instead, the authors blame poor urban planning, design, and management for increased criminal tendencies. They emphasise that relative income inequality has a greater influence on crime and violence than poverty per se, since inequality leads to exclusion from education, employment, health, and basic infrastructure.

The general idea behind CPTED theory and crime is that richer neighbourhoods are more often able to defend against crime while poorer ones are forced to use permeable designs which provide little or no protection (Owusu et al. 2015, 2016). On the other hand, unlike richer neighbourhoods, poor neighbourhoods tend to have strong social cohesion, which is a factor in preventing and reducing crime (Owusu et al. 2016; Gafa 2017).

In tandem with these findings, Banfield (1968: 180) states, “Theories about the causes and cures of crime tend to be variations of ones about the causes and cures of hard-core poverty.” This established and prevalent association between economic deprivation/poverty and violent crime has considerable support (Gennetian et al. 2012; White 2012; Cassidy et al. 2014; Dubow et al. 2016; McAra and McVie 2016). The same applies to the association between poverty and youth violent crime victimisation (White 2012) and young peoples’ engagement in crime and violence (McIlwaine and Moser 2003; Cassidy et al. 2014; McAra and McVie 2016); however, Kim et al.’s (2016) study found no difference regarding the effects of SES and youth violence in Seattle.

Moreover, violence and criminogenic levels in neighbourhoods are highly correlated with the kind of neighbourhood in question (Gennetian et al. 2012; Taniguchi and Salvatore 2012; Cassidy et al. 2014; Oliveira, Bastos-Filho, and Menezes 2017). Neighbourhoods with high levels of poverty tend to see high levels of violence (Gennetian et al. 2012; Cassidy et al. 2014), and the reverse is true (Cassidy et al. 2014). This association is what Flango and Sherbenou (1976: 331) termed the “situational determinants of crime”, or what Cassidy et al. (2014: 79) call “place effects” of violence. These include neighbourhood factors such as poverty (Gennetian et al. 2012; White 2012) and lower median incomes (Sohn 2016). Even family structure (White 2012) and relational quality in a household (Dubow et al. 2016; McAra and McVie 2016), city size (Rosenfeld, Vogel, and McCuddy 2018), and national inflation trends in relation to monetary crime (Rosenfeld, Vogel, and McCuddy 2018) were found to be determinants of violence.

Using longitudinal studies in the UK and the US, Oliveira, Bastos-Filho, and Menezes (2017) conclude that crime is unequally distributed in cities and different crimes are unevenly concentrated. Cassidy et al. (2014: 78) contend that the concentration of poverty in a neighbourhood has a more entrenched effect on crime and crime victimisation than the effect of poverty alone because it encapsulates the combined effects of factors such as poor housing, unemployment, limited social integration, and a lack of social mobility (see also McAra and McVie 2016; Taniguchi and Salvatore 2012). Cassidy et al. (2014: 78–79) maintain that poverty concentration correlates with high rates of violence and crime through interactive mechanisms such as the “contagion effects of bad behaviour” (for example, peer influence on vandalism among youth).


We triangulate both qualitative and quantitative data for the findings in this chapter. Our quantitative work centred on a survey of adult household heads of 2,745 households in purposively selected communities of differing SES in each city. A combined clustered, multi-stage probability proportionate-to-size sampling was used to select survey respondents beginning with the use of enumeration areas provided by the Ghana Statistical Service (GSS) from the 2010 Housing and Population Census. Nearly 11 per cent of the respondents were in high-SES communities, 41.0 per cent were in middle-SES neighbourhoods, and 48.1 per cent lived in low-SES communities (see Table 2.1). The data were analysed using SPSS version 20.0.

Drawing on Songsore (2003) and Owusu, Agyei-Mensah, and Lund (2008), we describe low-SES neighbourhoods as sharing the following characteristics. They are typically unplanned, with poor infrastructure maintenance. Dwellings house multiple households and amenities are shared among the households in a compound. Compounds typically do not have an exterior wall. Overcrowding is common. Community members usually have income-earning and social activities in and around the houses. In such circumstances, good social bonding is maintained and residents are often available in and around their homes to provide a presence, which potentially deters intruders (Owusu et al. 2016; Gafa 2017).

Middle-SES communities display a vast relative improvement in nearly all respects, including usually having enclosed walls, less clutter, better maintenance, fewer residents per house, and better availability and less sharing of amenities. High-SES communities are well planned. They often have well-kept surroundings, good access, individualised walled family/household housing units, and good infrastructure (water, electricity, sanitation, and drainage/gutters). Residents typically work outside their homes, leaving the environs quiet, individualised, and prone to intrusion. Target hardening gadgets, alarms, private security, and trained dogs are often used to ward off potential intruders (Owusu et al. 2015; Bagson 2017). Such neighbourhood classifications are often congruent with the SES of the inhabitants (Patino et al. 2014).

Table 2.1 Selected communities and their respective households/victims sampled



Community SES

Victims sampled qualitatively

Households sampled for survey

Proportion per SES group for survey: n (%)


Airport West




High SES 300
























Middle SES 1,125



















Chapel Hill






Anaji Estate




Low SES 1,320


New Takoradi






Russian Bungalow






















2,745 (100.0%)

Source: Authors.

Our qualitative data comprised in-depth semi-structured interviews with 40 victims of crime—14 from Tamale, ten each from Kumasi and Accra, and six from Sekondi-Takoradi. The respondents, 22 men and 18 women, were randomly chosen from within the survey households to represent different experiences of crime. Most were individuals who had experienced violent crime and were available to be interviewed. In the few instances where the victims were unavailable or were minors, heads of households or their next closest representatives were interviewed. Respondents were asked a series of questions about the types of crime they had experienced. Responses were analysed using thematic content analysis (Lieblich, Tuval-Mashiach, and Zilber 1998).

This analysis, however, has some limitations, necessitating careful interpretation of the findings. First, this is a retrospective study that allows room for recall bias. Second, we cannot claim causality from the study because it is cross-sectional. Finally, regarding the gendered analysis, the survey respondents were heads of households, who may have been reporting for household members of a different sex. This may, for instance, confound the gender-based analysis.


Our findings are grouped in four thematic areas:

1.types of violence experienced in the neighbourhoods

2.how communities respond to or cope with the violence

3.the role of the police and other state institutions

4.gendered and neighbourhood socioeconomic dimensions of the findings.

Types of violent crimes being experienced in the neighbourhoods

Thirty-nine per cent (1,076) of survey respondents said they or a member of their household had been a victim of crime.

The survey respondents mentioned experiencing a series of violent crimes 467 times; these were grouped into nine main types (Table 2.2). Among these, the most common crime was burglary with entry (34.5 per cent), followed by theft of livestock (10.9 per cent) and, almost equally, by attempted burglary (8.8 per cent) and car vandalism/theft of car parts (8.4 per cent). “Other” violent crimes made up 17.1 per cent. This group included rape, defilement (sex with a minor), violent threats, kidnapping, and assault. Assault was reported by 4.3 per cent (n = 20), and rape was reported by 1.1 per cent (n = 5) of survey respondents. Analysis by type of violent crime and neighbourhood SES showed a statistically significant relationship between the two variables (Kendal tau p = 0.002, Pearson chi square [χ2] = 31.427, p value = 0.026). This means that there were significant differences in the type of violent crime committed, given a particular neighbourhood’s SES.

Table 2.2 Cross-tabulation—type of violent crime and neighbourhood SES (from survey)


Type of violent crime victimisation



Theft of car/car hijacking

Theft from car

Car vandalism/theft of car parts

Theft of motorcycle

Theft of bicycle

Theft of livestock

Burglary with entry

Attempted burglary














































































Kendal tau p = 0.002 Pearson chi square =

31.427) (p value = 0.026)








Source: Authors.

In fact, results from cross-tabulation of proportions of crime mainly showed a negative correlation between neighbourhood SES and violent crimes. Generally, the lower the SES, the higher the number of reported violent crimes, and vice versa. Respondents in the low-SES neighbourhood reported 49.5 per cent of the violent crimes, the middle-SES followed closely with 45.6 per cent, and those in the high-SES neighbourhood reported only 4.9 per cent of the violent crimes studied.

Of our interviewees, 85 per cent (16 females and 18 males) reported violent crimes. These included one reported murder, four attempted murders, four rapes, and an attempted rape. Three shootings, two gun assaults, two motorbike robberies, and several forced entry/break-ins into homes/rooms with the theft of property were also reported. Several of these crimes included the use of weapons by the perpetrators such as guns, machetes, and knives. An uncle of a 16-year-old girl reported that she was forced to undress, ganged raped by “area boys”, her body slashed with a knife, and a black powdery substance put into the cut skin as a way of shaming and silencing a girl who “is too known [proud] and disrespectful … she walks as if she is more beautiful”. Other crimes that included weapons were armed robbery, attempted armed robbery, highway robbery, theft from car/vandalism of car/theft of car parts, and domestic physical abuse. The remainder included theft of livestock, burglary with entry, attempted burglary, and assaults/threats to kill. Two people also mentioned that a band of thieves, who steal motorbikes, operate late at night in the high-SES neighbourhood studied in Tamale.

… It is a very quiet area. The residential areas like RCC [Regional Coordinating Council] road and when you take this road to the new stadium between the sub-station to Nyohini, in the night you can’t use a motorbike to pass there.

(46-year-old male victim, high-SES neighbourhood, victim of theft of a motorbike, Tamale)

How communities respond to or cope with violent crime

It emerged from the interviews that a number of community leaders are usually involved in handling crime victimisation cases. These include chiefs, local level political leaders (known as Assembly members), spiritual leaders (Christian, Muslim, and a variety of traditional ones), family elders, and even parents. In addition, a few respondents mentioned praying on their own or seeking the intervention of God regarding their victimisation.

Our findings showed that reporting or the suggestion to report crime victimisation to chiefs for assistance was peculiar to Tamale, and specifically to that city’s low- and middle-SES communities. The respondents who reported that chiefs helped in resolving crime-related problems were divided in their opinion about the effectiveness of their reporting. In the low- and middle-SES communities in Tamale (particularly the former), it was mostly felt that the chiefs and the police aligned themselves with the ruling government (the National Democratic Congress) and, if you were not a supporter of the government, the chiefs and police would not follow up on cases reported to them.

The chiefs and the police are almost the same. They don’t also exhibit fairness on any matter that is brought before them.… The Imams too, because they are now also putting their hands into politics, they don’t anymore exhibit fairness to their people. Because of politics, chiefs, Imams and the police are all no more executing fairness or justice among their people.

(22-year-old male, low-SES neighbourhood, victim of gun assault, Tamale)

Truly speaking, we did not report it [murder case] to the chiefs because chiefs in this Tamale are not on their skin1 to deliver justice to the people of Aboabo.… We also realised that the chiefs had then taken sides.

(25-year-old male, low-SES neighbourhood, cousin of a murder victim, Tamale)

About 30 per cent of the survey respondents said that their neighbourhoods have community watch committees. Respondents from low-SES, followed by those from middle-SES neighbourhoods, were more likely to respond that their neighbourhoods had a watch committee compared to those in the high-SES neighbourhood. This relationship is very stringently important statistically (Pearson Chi-square [χ2] = 100.805; p value = 0.000).

Furthermore, the kind of neighbourhood is also very important for the existence of community cooperation and support against crime victimisation. Communities with higher SES were consistently more likely to self-report individualism/weak community bonding, support, and cooperation against crime: 62.2 per cent for high-SES, 47.6 per cent for middle-level, and 36.7 per cent for low-SES neighbourhoods. Again, this correlation was stringently statistically significant (Pearson Chi-square [χ2] = 123.953; p value = 0.000) (see Table 2.3).

The qualitative findings corroborated these conclusions. As a crime victim from the high-SES community in Tamale whose motorbike was violently taken from him around 10 p.m. at a traffic light affirmed: “I told you, this area, everybody is for himself”. Interviewees from only three communities (two middle-SES, one low-SES) mentioned that they have neighbourhood watch committees:

Table 2.3 Cross-tabulation—existence of neighbourhood watch committee and community cooperation and support against crime by neighbourhood SES (from survey)

Existence of neighbourhood watch and neighbourhood SES

Neighbourhood SES

Neighbourhood has crime watch committee





Don’t know













































Pearson Chi-square (χ2 = 100.805) (p value = 0.000)

Community cooperation and support against crime and neighbourhood SES

Neighbourhood SES

Community cooperation or support




Strong community bonding (SCB)

Individualism/weak community bonding

Both (SCB and individualism)













































Pearson Chi-square (χ2 = 123.953) (p value = 0.000)

Source: Authors.


Figures may not total 100% due to rounding errors.

Though the police are working [elsewhere], if they can patrol our community, it can help. Though we have a “watchdog committee”, their work is not up to satisfaction. They only come out when it is 12 a.m. and blow whistles.

(69-year-old male, victim of forced entry for theft at night with attempted strangulation, middle-SES neighbourhood, Kumasi)

The interviewees showed patterns of reporting crime based on city, SES of their community, and gender. About one-third (32.3 per cent) did not report the crime to either the police or local community leaders. Those who did not report were most likely to be from Tamale (63.6 per cent of non-reporting interviewees). Interviewees in Accra were most likely to have reported their victimisation to an authority. About 12 per cent of interviewees, all female, reported the crime only to community or religious leaders. These differences are further discussed at the end of the chapter in the Conclusions.

Survey respondents generally preferred to solve their own problems and tighten their security, without necessarily informing the police. The vast majority of them did not inform the police of their victimisation. Neither did they consider going through the court system to seek arbitration for their crime victimisation. Victims from low-SES neighbourhoods were least likely to report to the police, followed by those from middle-SES areas. The interviews showed some deviation from the survey respondents, as middle-SES interviewees were the least likely to contact the police or another authority to report a crime.

Our findings show that the rather low level of reporting of violent crime to the police was mostly based on a lack of confidence in them. Respondents trust the courts a little more than the police, although most of them said they had not had any experience with the courts. Generally, it was felt that contacting the police would not be beneficial. The police would not follow up on the case (particularly felt by residents in the low- and middle-SES neighbourhoods, and in Tamale), would take your money, and/or waste your time delaying the case. Others did not report because they felt they lacked the required evidence, the crime was insignificant, or they did not have the time to follow it up.

The vast majority of respondents also did not report the crime victimisation to their community leaders—chiefs, and political and religious leaders. This was particularly so in Tamale. Respondents in Kumasi were more likely to have reported the crime to community leaders.

Some interviewees reported that community members encouraged them to report the crime to community leaders, including religious/spiritual leaders who reportedly use divination to identify the perpetrator or the whereabouts of lost items, to predict the return of stolen property, or to spiritually influence the perpetrators to make restitution or confess. Religious leaders will also pray for the victim and use these cases as examples in their preaching to admonish potential criminals.

Also, in Tamale, two high-SES victims reported that there exists a “Youth Chief” in the Jakara Yili suburb of Kukuo to whom people report crime, particularly those involving youth. However, the feeling was that a bribe would be required or the case would amount to nothing. As one victim from Tamale whose motorbike was stolen told us, “he, too, will be demanding some money”. Like the Youth Chief, respondents also mentioned a “Chief of Thieves” in Tamale to whom one can report a crime. This person is purported to find the perpetrator and resolve the issue for a fee:

You pay money to the Chief and they can retrieve a motorbike for you? But doing that, you are encouraging indiscipline. You are encouraging an unlawful way of doing things.

(46-year-old male, high-SES neighbourhood, victim of motorbike theft, Tamale)

Consequently, particularly in Tamale and Kumasi, low- and middle-SES respondents were likely to feel that their crime victimisation was a matter for the divine and often left it “in the hands of God”, believing He would reveal and punish the perpetrators and restore what was lost.

While low- and middle-SES respondents put their trust in God, those in high-SES and some middle-SES neighbourhoods, particularly in Accra, take matters into their own hands. They exchange telephone numbers and use social platforms to network across the community. They also use sirens, alarms, private security guards, and sophisticated security systems involving barbed wire, security dogs, and even electrified fences to improve household security. As a result, communal security was said to be improving in these areas.

These notwithstanding, some of the relatively rich respondents felt that a return to traditional methods of social bonding and networking are needed, where neighbours look out for and care for each other. A respondent in the high-SES neighbourhood in Tamale also mentioned collaboration between the Christian Council and Muslim leaders in organising workshops on crime prevention. In the low-SES neighbourhood in Tamale, it was reported that community members assist the police when arresting criminals.

Role of the police and other state institutions

For the most part, information on the role of the state was gathered through in-depth interviews. A slim majority (55.9 per cent) of the interviewees reported their victimisation to the police. Victims from Accra were most likely to have reported the crime to the police (90.0 per cent), while those in Tamale were least likely to have done so (36.4 per cent). Only 50 per cent of violent crime victims interviewed in both Sekondi-Takoradi and Kumasi reported their victimisation to the police.

Generally, survey respondents perceived the role played by state criminal justice institutions, especially the police, negatively, with the courts being viewed as somewhat better. The respondents saw interaction with the police as a waste of time and money. That said, the state is seen as having only a minimal role in the threats of crime reported to the authors of this chapter. In other words, the state is seen to be absent in responding to crime. This was most pronounced in the low-SES neighbourhoods, where there are particularly few police stations and a limited police presence in general.

No, we don’t have a police station and that is also another big problem. The police station is far, and we have always insisted that they should provide us with one here, but this has fallen on deaf ears.

(68-year-old male, low-SES neighbourhood, victim of forced entry with property theft, Sekondi-Takoradi)

The same applies to police patrols, which interviewees from Kumasi, Tamale, and Sekondi-Takoradi said were sometimes brief, especially in the low-SES, followed by the middle-SES communities. Conversely, victim-interviewees in high-SES neighbourhoods in Accra and Tamale reported a high frequency of police patrols in their communities. Unique to Sekondi-Takoradi was mention that both the police and military frequently patrol the low-SES neighbourhood studied (New Takoradi), mainly due to the existence of nearby places of national security interest and in pursuit of suspects.

This place, sometimes we have soldiers patrolling these areas, because of the oil pipeline and the extension work going on at the harbour. But I think the police can also increase their patrol activities.

(22-year-old female, victim of forced entry with theft, low-SES neighbourhood)

Yet again, two respondents from neighbourhoods in Tamale (Aboabo, low SES) and Kumasi (Anloga, middle SES) reported involving their local community-level political representatives, popularly called “Assembly members”, in their case. However, the Assembly members’ roles in dealing with these cases were conflicting. In Aboabo, it was reported that an Assembly member in the community had called for the arrest of someone who was accused of murder. Conversely, in Anloga, it was reported that the Assembly member did not permit the victim to report a case of break and enter with theft of money and other property to the police. The Assembly member managed the case himself by asking the perpetrator to refund the stolen money, which his mother did. The victim felt that it was not appropriate for the Assembly member to have handled the case informally since it did not serve as a deterrent to the perpetrator who, shortly afterwards, committed a far more violent crime.

Gendered and neighbourhood dimensions of the findings

Gender was statistically very important to the types of violent crime experienced by respondents (Pearson Chi-square [χ2] = 33.369; p value = 0.000). Based on the household survey, females—particularly those living alone—were far more likely to have been victims of burglary with entry (39.9 per cent of females and 28.8 per cent of males surveyed). Additionally, three women, but no men, reported being kidnapped and more females reported other unspecified violent crimes (21.8 per cent of females, 12.4 per cent of males). Yet, interestingly and quite unheard of in Ghana, three of the five victims of reported rape cases in the survey were male.

The study findings also reflected a gendered accumulation of assets as men ordinarily have greater access to capital. Males were more likely to have reported theft of a bicycle (11.1 per cent of males, 3.4 per cent of females), theft of motorbikes (9.3 per cent of males, 3.4 per cent of females), theft from a car (8.0 per cent of males, 4.2 per cent of females), and theft of livestock (12.8 per cent of males, 9.2 per cent of females). Both males and females were almost equally likely to experience crimes such as attempted burglary, car vandalism/theft of car parts, and theft of car/car hijacking.

Cross-tabulation by gender, type of violent crime, and inter-SES neighbourhoods (see Table 2.4) generally follows a similar trend, with women, particularly those within middle-SES neighbourhoods, reporting burglary with forced entry more frequently than men. The same applies to the victims of car vandalism/theft of car parts. It is important to note that 60 per cent of victims of assault in the survey were female; and all cases of assault were reported exclusively in low-SES neighbourhoods.

Results from interviews corroborated some of the observed gendered differences in the survey. Women reported certain types of violent attacks more frequently (such as rape, attempted rape, assault, and kidnapping). This may be due to the general lesser relative strength of females when facing male perpetrators. It could also be explained by perpetrators taking advantage of the generally higher vulnerability of women. For instance, the interviews showed a trend where female victims of burglary/forced entry into homes, accompanied with other crimes such as rape, attempted rape, attempted murder, etc., were more likely to be alone at the time of victimisation. As the example below from Kumasi shows, perpetrators may have prior information about the increased vulnerability of their female victims, which contributes to the incidents.

Around 2 a.m., I heard that someone has broken the padlock to my house. When I woke up, I saw three people and one of them was holding a knife [and said] that if I scream he will kill me.… My husband works in the night and so does not sleep in the house with us … I heard that it is someone who gave them that tip.… And after they have robbed you, they force to sleep with you.

(29-year-old female, middle-SES neighbourhood, victim of forced entry with theft and attempted murder, Kumasi)

Other violent crimes such as murder, shootings, and property thefts were more likely to have male victims.

Table 2.4 Cross-tabulation—gender and types of violent crime among low-, middle-, and high-SES respondents (from survey)


MIDDLE SES—Ever been a victim of any crime



Theft of car/car hijacking

Theft from car

Car vandalism/theft of car parts

Theft of motorcycle

Theft of bicycle

Theft of livestock

Burglary with entry

Attempted burglary













































































LOW SES—Ever been a victim of any crime




Theft from car

Car vandalism/theft of car parts

Theft of motorcycle

Theft of bicycle

Theft of livestock

Burglary with entry

Attempted burglary













































































HIGH SES—Ever been a victim of any crime





Car vandalism/theft of car parts


Theft of bicycle

Theft of livestock

Burglary with entry

Attempted burglary












































































Source: Authors.


Figures may not total 100% due to rounding errors.

I was sleeping in the night around 10:30 to 11:00 thereabouts.… Suddenly, I saw that someone was strangling my neck, trying to kill me. I struggled with him and overpowered him.

(69-year-old male, middle-SES neighbourhood, victim of attempted property theft with strangulation, Kumasi)

The property-related crimes exemplify that males have a higher degree of property ownership and operation/management of items such as cars, motorcycles, bicycles, and livestock in Ghana. This argument is strengthened by the fact that both males and females reported attempted burglary or car vandalism/theft in similar proportions. Notably, these are instances of unsuccessful violent crimes that did not often involve resistance from the victim, or were typically carried out in the absence of the victim.

The interviews further showed neighbourhood effects within these gendered differences. It seemed that males and females within the low-SES neighbourhoods reported more heinous crimes more often. Victims of gun-related offences such as murder, attempted murder, and gun assaults were all male, almost all of whom lived in the low-SES neighbourhoods.

On 2nd January 2013, around 3 a.m. … I was walking with my girlfriend when we saw two gentlemen coming towards us … one of them told me to give him my car keys and the money I have [or] else he will shoot me.… I resisted … the guy shot me but luckily for me, the bullets fell on my side. When I fell down, they took my GHc 3502 and my phone. I have over eight bullets in my side now but I don’t have any help to go for operation for them to remove the bullets.

(35-year-old male, low-SES neighbourhood, victim of shooting and theft, Kumasi)

These results are borne out by the survey data (see Table 2.4). Moreover, all cases of rape and attempted rape reported by interviewees were from low-SES communities. From the narratives of the victims, Aboabo in Tamale, for instance, seemed like a melting pot of violent crime where suspicion alone could lead to one being murdered and where some residents were depicted as unbridled gangsters who could own/wield guns, shoot others at will, and go scot-free. On the other hand, the stealing of motorbikes and car parts, and one attempted armed robbery, were the only violent crimes reported in the high-SES communities.

We were here in the night and some guys came in.… The gates have been locked and I saw the guys come in, some scaled the wall into our house. And then all that I saw was they had cutlasses and started breaking the gates and so forth … before they [the police] came, they had looted all they could. (72-year-old male, low-SES neighbourhood, victim of forced entry and armed robbery, Accra)

Findings from the interviews also showed a gender- and SES-related trend in reporting violent crime. Males (66.7 per cent) were more likely to have reported victimisation to the police than females (43.8 per cent). Of all the violent crime victims who reported to the police, only 36.8 per cent were female. Conversely, females (25.0 per cent) were more likely to report their violent crime victimisation to others such as community leaders.

For the interviewees, reporting the violent crime victimisation to someone in authority showed an inverse relationship with the SES status of their neighbourhoods. Seventy-five per cent of respondents in the high-SES communities reported the crime to the police/someone in authority as opposed to 58.3 per cent of those in the middle and 68.4 per cent of those in the low-SES communities. Among the interviewees who reported the crime to someone in authority, those in the high-SES communities were most likely to have done so (75.0 per cent), followed by those in the low-SES communities (61.1 per cent), and finally those in the middle-SES communities (41.7 per cent).

Interviewees from the high-SES communities corroborated our findings from the survey: they reported using target hardening and other sophisticated means such as private security services, electronic alarms, security doors, barbed wire, and private police patrols (paid) to ward off crime.

I tapped the alarm signal first … we blew the alarm and he ran away.… Admittedly, the doors to our inner rooms are tough security doors. We made a report to the estate’s security … and they also quickly brought us security men.… We also told the police so we had police patrol … from time to time at night. My husband made some investigations and we realised other estates around had security—police protection and surveillance.… So together with the other residents we decided to make arrangements for it but this is not for free.…

(39-year-old female, high-SES neighbourhood, victim of attempted gang armed robbery, Accra)

Some respondents in the middle-SES communities also reported that, when in danger, they sometimes call for police patrols and receive a response. Furthermore, interviewees in most low- and a few middle-SES neighbourhoods mentioned having citizens’ watch committees.

Conclusions and implications for policy and practice

In sum, our findings corroborate previous literature regarding the ecology of crime that argues that crime is not randomly distributed in geographical space (Gennetian et al. 2012; White 2012; McAra and McVie 2016; Oteng-Ababio et al. 2016a, 2016b; Sohn 2016, etc.). We found important links between urbanisation, poverty, and violent crime in Ghana, and confirmed the importance of the “situational determinants” of crime (Flango and Sherbenou 1976), as well as the “place effects” (Cassidy et al. 2014: 79) of urban crime and violence. Specifically, we generally found a negative correlation between the SES of the neighbourhood and the amount of crime reported as well as the seriousness of violent crimes. That is, these tended to decrease/be less serious as the SES of the communities studied increased. Also, the lower the SES of the neighbourhood, the higher the propensity for the crimes reported to be physical violations, while the high-SES areas were more associated with property-related violent crimes.

Some inter-city differences in the types of crimes reported also emerged. For instance, more serious crimes such as murder, shootings, and rape were reported both in Kumasi and, particularly, Tamale than in Sekondi-Takoradi, especially, and Accra. It is clear that we cannot discount “the cultural settings” (Flango and Sherbenou 1976: 331) of these neighbourhoods, and even those of the cities studied. Owusu et al. (in press) note that each of the cities studied is unique.

There were also inter-city differences in the reporting of violent crime to the police and to community leaders. Qualitative respondents in Accra reported crime to the police the most often, with those in Tamale doing so the least. This may be due to the higher presence of police in Accra (Abudu, Nuhu, and Nkuah 2013; Bagson and Owusu 2016; Oteng-Ababio et al. 2016a), compared to other cities. Perceptions about the compromised loyalty of the police were most prevalent in Tamale, possibly explaining why victims of violent crime there reported to the police the least. Qualitative respondents in Kumasi were also most likely to have reported the crime victimisation to local, non-religious leaders, possibly reflecting the acclaimed status of Kumasi as “the cultural centre of Ghana”.

The finding that female interviewees were more likely to report their crime victimisation to informal community leaders may reflect the entrenched patriarchy in Ghana. Those women may identify more with such informal community leaders, all of whom happened to be men, as “heads” and authority figures within their communities rather than the police, who are formal and typically distant from the communities. Difficulties with accessing the police, such as the need to travel to distant police stations, offer bribes, or provide evidence that a crime has been committed were said to unduly delay the investigation of cases reported to them. Overall, the cost in terms of time, transport, and lost wages in addition to a perception of partiality on the part of the police, contribute to women’s reporting choices. Compared to men in Ghana, women are less likely to have the resources, such as money for transportation. Furthermore, based on gendered norms filtered through patriarchy, the female victims may have felt more comfortable dealing with informal types of social regulation and authority than formal ones.

Boateng (2018) documented that in Ghana, informal control systems are powerful in shaping crime victims’ reporting behaviour. He demonstrates a rational choice approach underlying the decision about how and to whom to report a crime. It involves a complex process of evaluating both the costs and benefits of reporting (Bowles et al. 2009), particularly to the police (Kaukinen 2002). Consequently, victims report crime to the police only if they perceive the benefits to outweigh the costs (Tarling and Morris 2010). In light of this, Boateng (2018) suggests that when it comes to reporting crime, victims give high priority to informal systems of support rather than formal ones. While noting that this is problematic, Boateng posits that the most plausible explanation for victims preferring to report crimes to the informal types of social regulation and authority is that they are more likely to receive a favourable reaction from informal groups than from formal ones. Boateng (2018) articulates further that this may be true in all social contexts, and particularly more prevalent within African societies. Crime victims’ sociodemographic characteristics influence their reporting to the police, and this is true regarding crime victims in Ghana (Boateng 2018).

Furthermore, the qualitative interviews affirm findings by previous authors that compared to the low-SES communities, those in the high- and middle-SES neighbourhoods have a higher police presence and more patrols (Abudu, Nuhu, and Nkuah 2013; Bagson and Owusu 2016). Conversely, our findings affirm previous research that residents of low-SES neighbourhoods are most likely to team up for neighbourhood watch activities in response to crime, while those in the high-SES communities use modern, expensive target hardening/CPTED devices and private security mechanisms to protect themselves from crime victimisation (Owusu et al. 2015).

Our study has several implications for policy and practice. First, crimes were usually not reported due to very low confidence in the police, in particular, but also in the court systems. It is clear that there is a need for significant reform of the criminal justice system to establish, as well as restore, confidence in it. Second, we suggest that the police service and the local government ministry make attempts to intensify efforts to encourage citizens to do more crime reporting. This will help the police to improve and strategise more thoroughly regarding crime control efforts. Third, we recommend that the police and National Bureau of Investigation work on ridding communities of dangerous informal forms of crime control such as the “Chief of Thieves”.

Fourth, this study has unearthed data showing that middle-SES communities may be particularly vulnerable to violent crime. Middle-SES neighbourhoods have a lower police presence than the high-SES communities and fewer personalised/individualised target hardening/CPTED mechanisms. At the same time, residents in the middle-SES neighbourhoods studied revealed the fewest community-based communal attempts to ward off crime, such as having neighbourhood watch committees. This may also be due to the fact that, as became clear in this study, these middle-SES neighbourhoods were the least likely to report crimes to the police, creating what our study has shown to be an erroneous impression that they may be safer, and require less of a police presence. However, this study found that they are the most prone to violent crime.

The finding that the selected middle-SES neighbourhoods attract more violent crime complicates the analysis from mainstream literature on “the poverty trap and crime victimisation” otherwise referred to as the “concentric zone theory” of crime and urbanisation as originally proposed by Shaw and McKay (1942). This theory states that there is a consistent inverse relationship between the SES of neighbourhoods and crime victimisation. We recommend that the middle-SES communities studied in this research improve community collaboration to enhance their security. Residents of the middle-SES communities studied are also encouraged to step up their reporting of violent crime to the police. Broadly, however, our study has affirmed most of the previous findings in the mainstream literature about the link between crime victimisation and poverty.

Our study also found that victims were unlikely to report crime to community leaders. Based on the significant legitimacy and reputation of such leaders and the fact that nearly all communities in Ghana have a variety of these informal leaders, the Local Government Ministry should strategise to find ways of orientating these leaders (Chiefs, Assembly members, religious leaders) to help mitigate violent crime, particularly in the urban communities studied. Given the financial and time costs involved in using official arbitration systems, it would be appropriate to strengthen alternative dispute resolution mechanisms such as the indigenous ways of handling criminal victimisation revealed in mainly low-income neighbourhoods. Additionally, the findings provide insight into neighbourhood SES and security adaptation measures. More security networking against crime in these cities is also recommended. We suggest a combined police and Local Government Ministry effort to encourage more proactive community policing such as setting up community crime-stopper activities, including supporting neighbourhoods to better organise watch committees and sharing intelligence.

Furthermore, our study found a gendered aspect related to the type of violent crime victimisation. Females were disproportionately victimised in contexts where they would be challenged due to their lower physical strength. For instance, data triangulation showed that most of the cases of assault, rape, and attempted rape happened to women. There were more break-ins and accompanying crimes (rape, attempted rape, petty property thefts) against women. Interestingly, the qualitative data showed that this was particularly true when the women were alone, without a male family member or partner/husband/boyfriend. The perpetrators were also said to be neighbours who had foreknowledge that the women were alone. We recommend increased efforts to prevent these violent crimes; for example, by working against gender stereotypes that underpin the prevalence of rape and assault against women.

Another gendered aspect of the crime victimisation we uncovered was that more gun-related violence, including gang-style armed robbery and property theft in general, was targeted at males. Violent property crime against men underscores the greater opportunity men have to own and control social resources, which makes them a target for this type of crime. It is possible that perpetrators use weapons, such as guns, against male targets because they are more likely to be able to fight back.

Finally, we propose that to help curb violent crime in the studied communities, additional short-, medium-, and long-term efforts toward poverty alleviation be undertaken by policymakers. These should particularly target residents in the urban slums we studied as our results support previous findings (for instance, Patino et al. 2014: 48) showing that neighbourhoods tend to reflect the SES of their residents. The short-term measures should strengthen and possibly widen current social protection efforts against poverty. Medium- and longer-term strategies such as ensuring better/higher education, employable skill development, and higher youth employment avenues are proposed.


1A “skin”, usually the full skin of an animal, for instance a goat or sheep, which has been cleaned and dried, is a symbol of the authority of a chief in the three northern regions of Ghana. In such communities, chiefs sit on a skin to receive homage or handle chieftaincy-related matters.

2Approximately US$75.


Abudu, Amadu M., Yusif Nuhu, and Joseph K. Nkuah (2013). Bridging the security gap in Ghana: The role of private security actors. Developing Country Studies 3(10): 11–30.

Aryeetey, Ernest, George Owusu, and Emmanuel Joseph Mensah (2009). An analysis of poverty and regional inequalities in Ghana. Global Development Network (GDN), Working Paper No. 27. New Delhi: GDN.

Bagson, Ernest (2017). The place and space of informal crime prevention strategies in urban Ghana. PhD thesis submitted to the Institute of Statistical, Social and Economic Research (ISSER), University of Ghana, Accra.

Bagson, Ernest, and Adobea Y. Owusu (2016). Securing the urban space: On who’s terms? Insights from poverty and crime baseline survey in Tamale, Ghana. Ghana Journal of Geography 8(1): 124–147.

Banfield, Edward C. (1968). The unheavenly city: The nature and future of our urban crisis. Boston, MA: Little, Brown and Company.

Boateng, Francis D. (2018). Crime reporting behavior: Do attitudes toward the police matter? Journal of Interpersonal Violence 33(18): 2891–2916; Epub 24 February 2016. doi:10.1177/0886260516632356.

Bowles, Roger, Maria Garcia Reyes, and Nuno Garoupa (2009). Crime reporting decisions and the costs of crime. European Journal on Criminal Policy and Research 15(4): 365–377.

Caldeira, Teresa P.R. (2012). Fortified enclaves: the new urban segregation. In Jan Lin and Christopher Mele (Eds.), The urban sociology reader (419–427). London: Routledge.

Cassidy, Tali, Gabrielle Inglis, Charles Wiysonge, and Richard Matzopoulos (2014). A systematic review of the effects of poverty deconcentration and urban upgrading on youth violence. Health & Place 26: 78–87.

Cozens, Paul, David Hillier, and Gwyn Prescott (2001). Crime and the design of residential property—exploring the perceptions of planning professionals, burglars and other users: Part 2. Property Management 19(4): 222–248. doi.10.1108/EUM0000000005784.

Dubow, Eric F., L. Rowell Huesmann, Paul Boxer, and Cathy Smith (2016). Childhood and adolescent risk and protective factors for violence in adulthood. Journal of Criminal Justice 45: 26–31.

Fafchamps, Marcel, Eleni Gabre-Madhin, and Bart Minten (2005). Increasing returns and market efficiency in agricultural trade. Journal of Development Economics 78(2): 406–442.

FBI (2011). Crime in the United States, 2010: Violent crime. Federal Bureau of Investigation, US Department of Justice. https://ucr.fbi.gov/crime-in-the-u.s/2010/crime-in-the-u.s.-2010/violent-crime.

Flango, Victor Eugene, and Edgar L. Sherbenou (1976). Poverty, urbanisation, and crime. Criminology 14(3): 331–346.

Frimpong, Louis K. (2016). Geography of fear of crime: Examining intra-urban differentials in Sekondi-Takoradi Metropolis, Ghana. Ghana Journal of Geography 8(1): 79–102.

Gafa, B. (2017). The upsurge of Crime Prevention through Environmental Design (CPTED) measures in urban Ghana: Perceptions, motivations and challenges. MPhil thesis submitted to the Department of Geography and Resource Development, University of Ghana, Accra.

Gennetian, Lisa A., Matthew Sciandra, Lisa Sanbonmatsu, Jens Ludwig, Lawrence F. Katz, Greg J. Duncan, Jeffrey R. Kling, and Ronald C. Kessler (2012). The long-term effects of moving to opportunity on youth outcomes. Cityscape: Journal of Policy Development and Research 14(2): 137–167.

Grohe, Bonnie (2011). Measuring residents’ perceptions of defensible space compared to incidence of crime. Risk Management 13(1/2): 43–61.

GSS: Ghana Statistical Service (2010). Victimisation survey in Ghana. Accra: GSS/UN Office on Drugs and Crime (UNODC).

GSS: Ghana Statistical Service (2012). 2010 population and housing census. Summary report of final results. Accra: Author.

Hipp, John R., and Daniel K. Yates (2011). Ghettos, thresholds, and crime: Does concentrated poverty really have an accelerating increasing effect on crime? Criminology 49(4): 955–990.

Kaukinen, Catherine (2002). The help-seeking decisions of violent crime victims: An examination of the direct and conditional effects of gender and the victim–offender relationship. Journal of Interpersonal Violence 17(4): 432–456.

Kim, Elizabeth B.K., Amanda B. Gilman, Karl G. Hill, and J. David Hawkins (2016). Examining protective factors against violence among high-risk youth: Findings from the Seattle Social Development Project. Journal of Criminal Justice 45: 19–25.

Lieblich, Amia, Rivka Tuval-Mashiach, and Tamar Zilber (1998). Narrative research: Reading, analysis, and interpretation. Applied Social Research Methods series, Vol. 47. Thousand Oaks, CA: SAGE Publications.

McAra, Lesley, and Susan McVie (2016). Understanding youth violence: The mediating effects of gender, poverty and vulnerability. Journal of Criminal Justice 45: 71–77.

McIlwaine, Cathy, and Caroline Moser (2003). Poverty, violence and livelihood security in urban Colombia and Guatemala. Progress in Development Studies 3(2): 113–130.

Muggah, Robert (2012). Researching the urban dilemma: Urbanization, poverty and violence. Ottawa: International Development Research Centre (IDRC). www.idrc.ca/EN/PublishingImages/Researching-the-Urban-Dilemma-Baseline-study.pdf.

Obeng-Odoom, Franklin (2013). Governance for pro-poor urban development: Lessons from Ghana. London: Routledge.

Oliveira, Marcos, Carmelo Bastos-Filho, and Ronaldo Menezes (2017). The scaling of crime concentration in cities. PLoS One 12(8): e0183110. doi:10.1371/journal.pone.0183110.

Oteng-Ababio, Martin, George Owusu, Charlotte Wrigley-Asante, and Adobea Y. Owusu (2016a). Longitudinal analysis of trends and patterns of crime in Ghana (1980–2010): A new perspective. African Geographical Review 35(3): 193–211.

Oteng-Ababio, Martin, Adobea Y. Owusu, George Owusu, and Charlotte Wrigley-Asante (2016b). Geographies of crime and collective efficacy in urban Ghana. Territory, Politics, Governance 5(4): 459–477.

Oteng-Ababio, Martin, George Owusu, Charlotte Wrigley-Asante, Adobea Y. Owusu, Ernest Bagson, and Louis Kusi Frimpong (under review). “Shifting position with the shade”: A diagnosis of youth networks, urbanism and violence in Tamale, Ghana. Security Journal.

Owusu, Adobea Y., Charlotte Wrigley-Asante, George Owusu, Martin Oteng-Ababio, Ernest Bagson, and Martin W. Agyekum (in press). “Who the cap fits …”: Profile of crime victims in Ghana’s four biggest cities. Victims & Offenders.

Owusu, George, Samuel Agyei-Mensah, and Ragnhild Lund (2008). Slums of hope and slums of despair: Mobility and livelihoods in Nima, Accra. Norsk Geografisk Tidsskrift [Norwegian Journal of Geography] 62(3): 180–190.

Owusu, George, Charlotte Wrigley-Asante, Martin Oteng-Ababio, and Adobea Y. Owusu (2015). Crime prevention through environmental design (CPTED) and built-environmental manifestations in Accra and Kumasi, Ghana. Crime Prevention & Community Safety 17(4): 249–269.

Owusu, George, Martin Oteng-Ababio, Adobea Y. Owusu, and Charlotte Wrigley-Asante (2016). Can poor neighbourhoods be correlated with crime? Evidence from urban Ghana. Ghana Journal of Geography 8(1): 11–31.

Patino, Jorge E., Juan C. Duque, Josep E. Pardo-Pascual, and Luis A. Ruiz (2014). Using remote sensing to assess the relationship between crime and the urban layout. Applied Geography 55: 48–60.

Rosenfeld, Richard, Matt Vogel, and Timothy McCuddy (2018). Crime and inflation in U.S. cities. Journal of Quantitative Criminology 1–16. doi:10.1007/s10940-018-9377-x.

Shaw, Clifford R., and Henry D. McKay (1942). Juvenile delinquency and urban areas: A study of rates of delinquents in relation to differential characteristics of local communities in American cities. Chicago, IL: University of Chicago Press.

Simpson, J., and Edmund Weiner (Eds.) (1989). Definition of crime. English Oxford Living Dictionaries [online]. Oxford: Oxford University Press. https://en.oxforddictionaries.com/definition/crime.

Sohn, Dong-Wook (2016). Residential crimes and neighbourhood built environment: Assessing the effectiveness of crime prevention through environmental design (CPTED). Cities 52: 86–93.

Songsore, Jacob (2003). The urban housing crisis in Ghana: Capital, the state versus the people. Ghana Social Science Journal 2(1): 1–31.

Songsore, Jacob (2009). The urban transition in Ghana: Urbanisation, national development and poverty reduction. Legon-Accra: University of Ghana.

Songsore, Jacob, Osman Alhassan, Sebastian K. Avle, Paulina Amponsah, Marshall Kala, and Mary A. Chama (2014). Environmental health and disaster risks, livelihoods and ecology within the Korle-Lagoon in Accra, Ghana. Accra: Ghana Universities Press.

Taniguchi, Travis A., and Christopher Salvatore (2012). Exploring the relationship between drug and alcohol treatment facilities and violent and property crime: A socioeconomic contingent relationship. Security Journal 25(2): 95–115.

Tarling, Roger, and Katie Morris (2010). Reporting crime to the police. British Journal of Criminology 50(3): 474–490.

UNDP: United Nations Development Programme (2007). Ghana human development report 2007: Towards a more inclusive society. Accra: Author.

UN-Habitat: United Nations Human Settlements Programme (2012). Global report on human settlements 2007: Enhancing urban safety and security. London/Sterling, VA: Author. www.un.org/ruleoflaw/files/urbansafetyandsecurity.pdf.

White, Nicole (2012). The economy–crime relationship revisited: The significance of recent macroeconomic conditions and social policy changes for child poverty and youth violence trends. PhD dissertation, Department of Criminology and Criminal Justice, University of Missouri-St. Louis, MO.

WHO: World Health Organization (2002). World report on violence and health. Geneva: Author. www.who.int/violence_injury_prevention/violence/world_report/en/.

Wirth, Louis (1938). Urbanism as a way of life. American Journal of Sociology 44(1): 1–24.

World Bank (2014). Rising through cities in Ghana: Ghana urbanisation review— Overview report. Washington, DC: World Bank.

Wrigley-Asante, Charlotte, George Owusu, Martin Oteng-Ababio, and Adobea Y. Owusu (2016). Poverty and crime: Uncovering the hidden face of sexual crimes in urban low-income communities in Ghana. Ghana Journal of Geography 8(1): 32–50.

Part II

Southern Africa

3Urban upgrading linked to positive social outcomes in Cape Town, South Africa

Richard Matzopoulos, Kim Bloch, Sam Lloyd, Chris Berens, Jonny Myers, and Mary Lou Thompson


Among the interventions to reduce crime and violence in South Africa, there is growing interest in those that address the social and structural environments that entrench exclusion and are considered important antecedents to a range of deleterious health and well-being outcomes. In Cape Town, the Violence Prevention through Urban Upgrading (VPUU) programme applies an integrated approach, with a strong focus on social interventions and community participation, alongside urban upgrading. In this chapter, we used data from three community-based surveys, conducted in Khayelitsha in 2013, 2014, and 2015, to explore the association between the infrastructure developed by VPUU and several social outcomes: (1) experiencing interpersonal violence; (2) signs and symptoms of depression; (3) satisfaction with neighbourhood infrastructure; (4) involvement in community activities; and (5) social cohesion.

The initial findings were positive. Respondents who lived within 2 km of VPUU infrastructure (i.e. ‘the intervention area’) experienced less violence, showed fewer signs of depression, and were more satisfied with the infrastructure in their neighbourhood than those who lived further than 2 km from VPUU infrastructure. There were similar levels of social cohesion among residents exposed to the intervention and those who were not exposed, but residents living outside the intervention area were more likely to be active in their communities.

These findings support recent policy developments that foreground safety through environmental design. For example, the Western Cape Province’s 2013 Integrated Provincial Violence Prevention Policy Framework highlights the importance of the modification of urban environments and public spaces (Western Cape Government 2013). At a national level, the 2016 White Paper on Safety and Security includes the integration of safety, crime, and violence prevention principles into urban and rural planning and design to promote perceptions of and actual safety as a key theme (Civilian Secretariat for Police Service 2016). There are several other research questions that should be answered before these interventions are scaled-up nationally: Are there other factors that might partially explain the positive findings related to urban upgrading and improved social outcomes? Are there unintended consequences of urban upgrading such as displacement of violence to other areas? Are urban upgrading interventions such as the VPUU programme sustainable? Until such time as these questions are addressed, the urban upgrading intervention implemented by VPUU should be considered a promising intervention.


South Africa has high levels of interpersonal violence—a major contributor to the burden of disease (Pillay-van Wyk et al. 2016)—and one of the world’s most unequal income distributions (World Bank 2018), both of which undermine quality of life and development progress. The South African National Development Plan acknowledges the relationship between poverty, inequality, and crime, and there is growing interest in interventions that address the social and structural environments, which have entrenched exclusion and inequality arising from the country’s Apartheid past (National Planning Commission 2012). For example, the 2016 White Paper on Safety and Security foregrounds “safety through environmental design” as one of six key themes for crime and violence prevention alongside an effective criminal justice system; early interventions directed at children; comprehensive and high-quality victim support; effective and integrated service delivery; and active public and community participation. The White Paper calls for “the integration of safety, crime and violence prevention principles into urban and rural planning and design that promotes safety and facilitates feeling safe” (Civilian Secretariat for Police Service 2016: 18).

Cape Town, the capital city of South Africa’s Western Cape Province, has typically had the country’s highest or second highest provincial homicide rate (Prinsloo et al. 2016) Here, the municipal government co-funded an urban upgrading programme that addresses violence called VPUU. VPUU adopts an integrated community participative approach premised on situational, social, and institutional crime prevention that also includes crime prevention through environmental design principles (Lloyd and Matzopoulos 2019). The approach encompasses building interventions across four categories: buildings, sports facilities, landscaping, and surfacing. In conjunction, social interventions, such as a neighbourhood watch programme, activate these spaces. The programme attempts to offer redress for the deep inequality of Apartheid-era segregated urban development and has broad appeal among different tiers of government. It has been highlighted in the Western Cape Provincial Government’s Integrated Violence Prevention Policy Framework as exemplifying a “whole-of-society” approach to preventing violence. VPUU is currently being implemented in several high-violence urban communities in Cape Town that feature prominently in the national and provincial rankings of police precincts with the highest number of murders per capita, as well as one semi-rural site and one rural site in Western Cape Province. In Cape Town, VPUU operates in two designated Safe Node Areas (SNAs1) within Khayelitsha: the Harare SNA and the Monwabisi Park SNA.

Evaluation research has shown that interventions aimed at urban upgrading show promise, but there is a dearth of studies from low- and middle-income countries (Cassidy et al. 2014). There is also a pressing need for empirical evidence of the effectiveness of modifying the built environment in conjunction with social interventions as a violence prevention intervention in the South African setting.

In this chapter, we use data from three community-based surveys conducted in Khayelitsha, Cape Town in 2013, 2014, and 2015 to provide a preliminary analysis of the associations between living in proximity to the VPUU intervention area and five outcomes: (1) experiences of severe interpersonal violence; (2) showing signs and symptoms of depression; (3) satisfaction with neighbourhood infrastructure; (4) involvement in community activities; and (5) social cohesion. We considered that these outcomes were implicitly addressed in the design of the intervention and would be measurably better among the exposed population if the intervention was successful.


Study design

The study was based on a repeated cross-sectional design with surveys conducted annually in 2013, 2014, and 2015.

Study population and sampling procedure

The sampling area covered the suburb of Khayelitsha, Cape Town, which is approximately 30 km from Cape Town’s city centre and, at the 2011 census, had a population of 391,749 (City of Cape Town 2013). A total of 1200 households were randomly selected for the first survey in 2013 using the 2012 version of geographical information systems data for all formal dwellings in Cape Town (i.e. registered serviced households) originally sourced from the City of Cape Town’s address database (City of Cape Town Open Data Portal 2015). Informal dwellings (i.e. houses that were not registered with and serviced by the municipality at the time of their construction) were mapped using the 2011 version of the Eskom SPOT 5 Building Count dataset (de la Rey 2008). The 2014 and 2015 surveys were conducted at the same approximate geographic locations selected in 2013, but previously interviewed households were substituted with neighbouring dwellings. This provided three annual cross-sections from the same 1200 locations. Fieldwork took place from July to November each year.

For each survey, the ‘main household’ at each property was defined as the residence of the property’s owner. Where the owner was not resident, the household with the longest residing tenant was considered as the main household. Surveys were completed by the female head of the main household (i.e. the adult female identified as the primary caretaker of the house and household members). A male adult resident was selected if there was no female adult resident or if the main household adult female was unable to respond due to psychological or hearing impairment. Comparisons in strata restricted to males were limited by smaller sample sizes. Our rationale for selecting female respondents was twofold. First, the questionnaire included a range of topics relating to the household and the neighbourhood that might be reported differentially by gender and we required a structured sampling process that would limit potential bias if fieldworkers had the discretion to select respondents from within a household. Second, our understanding, based on pilot studies and consultation with community members, indicated that female respondents were likely to have a more complete understanding of household and neighbourhood dynamics because they had more restricted mobility than their male counterparts in a high violence community like Khayelitsha.

Households were substituted after three unsuccessful visits, if the property was vacant, if residents refused to be interviewed, if the resident had been interviewed in a previous survey, or if there was no one aged 18 or older in residence. Substitutions were made from a second random sample drawn immediately after the initial phase of fieldwork, after which fieldworkers systematically visited households to the left and then to the right of the households selected for the second random sample. The study was approved by the Human Research Ethics Committee of the Faculty of Health Sciences, University of Cape Town (HREC REF 637/2013) and all residents provided written informed consent.

Instruments and procedures

Trained fieldworkers collected information on sociodemographics, experiences of interpersonal violence, signs and symptoms of depression, satisfaction with neighbourhood infrastructure, participation in community activities, and social cohesion using a structured questionnaire.


For interpersonal violence, we constructed a binary measure for experience of severe interpersonal violence if the respondent reported any attempted or actual rape or murder of someone well known to the respondent, assault with a weapon, or two or more of the following incident types affecting the respondent personally: assault (with or without a weapon), rape, robbery, petty theft,2 arson, car hijacking, victim or witness of another’s experience of discrimination-based violence,3 drug-related violence, gang-related violence, victim or witness of another’s experience of targeted or threatened community justice, or domestic violence.

To measure signs and symptoms of depression, we used the 10-question version of the Center for Epidemiological Studies-Depression scale (CES-D-10) adapted from the original 20-question version developed as a screening tool for depression in the general population and suitable for use in Black and White English-speaking American populations of both sexes across a wide age range and socioeconomic strata (Radloff 1977; Andresen et al. 1994). Each CES-D-10 question was measured on a 4-point scale (0 to 3), which produced a composite score between 0 and 30. A CES-D-10 cut-off of 13 was used to indicate the presence of significant signs and symptoms of depression, which has been validated among isiXhosa-speaking South Africans, the predominant language group in Khayelitsha (Baron, Davies, and Lund 2017).

To measure satisfaction with neighbourhood infrastructure, respondents were asked to report their current satisfaction (yes/no) with the conditions and maintenance of the following infrastructure and services provided in whole or in part by VPUU: roads or pathways, storm water drainage, streetlights, high mast lights, electrical supply, refuse removal, green space, public open space, sports and recreational facilities, water taps (in areas with informal houses), nearest waste collection point, nearest pension pay-out point, nearest clinic, nearest library, and nearest sports facility. Each response was scored on a 0/1 scale, which produced a composite score ranging from 0 (i.e. not satisfied with any aspect of infrastructure) to 15 (satisfied with all aspects of infrastructure). We used a cut-off of 75 per cent (i.e. satisfied with more than 11 of the 15 services) to indicate sufficient neighbourhood infrastructure satisfaction.

For community participation, respondents were asked whether they were actively involved in any of five different types of community groups that we were concerned with: (1) neighbourhood services or problems; (2) social, religious, or hobby groups; (3) formal or informal sports groups; (4) voluntary security-related organisations; or (5) other voluntary organisations. These categories were derived from the work of De Silva et al. (2007). Residents that were actively involved in two or more groups were considered as being ‘active in the community’.

The 2015 survey measured social cohesion by asking residents the following questions: (1) How common is it that neighbours help each other out?; (2) How common is it that neighbours do things together?; (3) How common is it that people in your neighbourhood are aggressive?; and (4) Can the majority of people in your neighbourhood be trusted? Each question was rated on a 5-point Likert scale (reversed for question 3 and with missing values recorded as zero), providing a cumulative score ranging from 0 to 20.

Defining the intervention area

Exposure to the intervention was defined as the household’s proximity to the VPUU built environment. Households situated within 2 km—equivalent to a 20 minute walk—were considered exposed to the intervention (i.e. intervention group). According to urban fabric theory, this measure is consistent with the limit of the walking city (Newman, Kosonen, and Kenworthy 2016). Households located outside this zone were considered to be unexposed to the intervention (i.e. the control group). Figure 3.1 illustrates the spatial distribution of exposed and unexposed households included in the study.

Statistical analysis

Survey results were pooled from each of the annual cross-sections and prevalence was estimated for each of the four binary outcomes. Exposed and unexposed households were compared using the Pearson’s Chi-squared test for the binary outcomes and Student’s t-test to compare the mean scores for social cohesion. Summaries were stratified by gender and formal/informal area to take into account possible effect modification. Acknowledging the multiple comparisons made in this study, associations were assessed using a significance level of alpha = 0.01 rather than the conventional 0.05. The analysis was conducted using Stata version 13.1 (Stata Corporation, USA).

Figure 3.1 Household survey points and proximity to VPUU intervention stratified by category of exposure.


The sample consisted of 3703 participants, most of whom were women (82 per cent). A similar number of houses were sampled in formal and informal areas and for each year from 2013 to 2015. The stratified findings for each outcome that follow were not adjusted for potential confounders and may be reflective of other differences across the exposure groups (see Table 3.1).

Table 3.1 Summary statistics for each outcome and covariates by exposure category


Exposure to VPUU intervention*

All areas
(n = 3703)*


(n = 2138)

(n = 1565)


n (%)

n (%)

n (%)




Area type



  348 (17)

1645 (82)


  952 (45)

1182 (55)

  282 (19)

1164 (80)


1017 (65)

  547 (35)

  630 (18)

2809 (82)


1969 (53)

1729 (47)






  694 (32)

  741 (35)

  703 (33)


  499 (32)

  516 (33)

  550 (35)


1193 (32)

1257 (34)

1253 (34)


Experience of violence

Signs and symptoms of depression

Satisfaction with neighbourhood infrastructure

Community participation


Social cohesion (n = 1192)


  482 (23)

  327 (16)

  422 (19)

1041 (51)

Mean (SD)

     11.4 (3.4)


  216 (15)

  152 (11)

  408 (27)

  591 (40)

Mean (SD)

   12.0 (3.4)


  698 (20)

  479 (14)

  830 (22)

1640 (46)

Mean (SD)

     11.7 (3.2)


* Exposed households were situated less than 2 km from the infrastructural intervention.


The survey confirmed the considerable burden of violence experienced by residents of Khayelitsha. Overall, 19.7 per cent of residents were considered to have been exposed to severe interpersonal violence within the last year. There was considerable variation in the experience of violence by neighbourhood within Khayelitsha (see Figure 3.2).4 Site C (34.0 per cent), Site B (22.2 per cent), and Harare (19.4 per cent) were the areas that experienced the highest levels of violence.

In the intervention area, 14.7 per cent of households experienced severe interpersonal violence, which was significantly less than the 23.3 per cent of households outside the intervention area (p < 0.001) (see Figure 3.3).

Gender was an important risk factor for violence. Men were more likely to have experienced severe interpersonal violence within the last year than women—23.2 per cent compared to 18.9 per cent (p = 0.016). However, both men (p = 0.001) and women (p < 0.001) living in the intervention area were significantly less likely to experience severe interpersonal violence than those living outside the intervention area.

Figure 3.2 Experience of severe violence in the last year by neighbourhood within Khayelitsha.

Figure 3.3 Experience of severe violence by proximity to VPUU intervention.

The type of area where residents lived also influenced the risk of experiencing severe interpersonal violence. Experience of severe interpersonal violence was significantly higher among residents living in informal areas (22.8 per cent) compared to residents living in formal areas (16.8 per cent) (p < 0.001). Although the VPUU intervention was more concentrated in formal rather than informal areas, the significantly lower prevalence of severe interpersonal violence among residents exposed to the intervention was observed in both formal and informal areas. Residents in informal areas were at significantly lower risk of severe interpersonal violence if they were exposed to the intervention than if they lived outside the intervention area (16.6 per cent vs 25.6 per cent; p < 0.001). Similarly, in formal areas, people living within the intervention area experienced less severe interpersonal violence than those outside the area (13.5 per cent vs 20.2 per cent; p < 0.001).

Signs and symptoms of depression

Overall, 13.9 per cent of residents were considered to be exhibiting signs and symptoms of depression. Outside the intervention area, 16.3 per cent of respondents reported signs and symptoms of depression, which was significantly higher than the 10.6 per cent reported by the exposed population (p < 0.001) (see Figures 3.4 and 3.5). Prevalence of depression was similar among men and women (13.1 per cent and 14.0 per cent, respectively). Among women, there was a significantly lower prevalence of depression in the intervention area compared to outside the area (10.9 per cent vs 16.2 per cent; p < 0.001), whereas the lower prevalence of depression among men living in the intervention area compared to outside the area was not statistically significant (11.0 per cent vs 14.8 per cent; p = 0.165).

Prevalence of depression was similar in formal and informal areas: 15.0 per cent of residents living in informal areas and 12.9 per cent of residents living in formal areas showed signs and symptoms of depression (p = 0.068). In formal intervention areas, residents were significantly less likely to show signs and symptoms of depression (9.9 per cent) than residents living in formal, control areas (16.0 per cent) (p < 0.001). In informal areas, residents residing within the intervention area also had a lower prevalence of depression (11.8 per cent) than residents living outside the intervention area (16.5 per cent) (p = 0.014).

Figure 3.4 Symptoms of depression by neighbourhood within Khayelitsha.

Figure 3.5 Depression by proximity to VPUU intervention.

Satisfaction with neighbourhood infrastructure

Overall, just 22.4 per cent of residents were satisfied with the range of infrastructure and services in their neighbourhoods, but residents were significantly more satisfied if they lived within the intervention area (27.0 per cent vs 19.1 per cent; p < 0.001). Gender was not identified as an independent determinant of satisfaction with infrastructure; approximately 21 per cent of both men and women reported satisfaction with services (p = 0.92). Among both male and female respondents there was significantly more satisfaction with services within the VPUU intervention area compared to those outside it (24.8 per cent vs 17.5 per cent of men and 24.7 per cent vs 17.8 per cent of women, respectively; p < 0.001 for all comparisons).

Overall, living in formal areas was significantly associated with satisfaction with neighbourhood infrastructure (see Figures 3.6 and 3.7): residents living in informal areas were significantly less likely to be satisfied with infrastructure than those living in formal areas (10.1 per cent vs 33.2 per cent; p < 0.001). Within formal areas, residents were significantly more likely to report satisfaction if they lived within the intervention area (36.8 per cent vs 29.4 per cent; p = 0.001). However, there was no statistical difference in the level of satisfaction among residents living in informal areas when comparing those in and outside the intervention area (8.6 per cent vs 10.7 per cent; p = 0.167). It should be noted that because VPUU was primarily involved in the upgrading of infrastructure in the public space, it had considerably less influence on overall infrastructure in informal settlements, which were mainly residential and where the local municipality was the primary service delivery agent.

Figure 3.6 Satisfaction with infrastructure by neighbourhood within Khayelitsha.

Community participation

Overall, 46.3 per cent of residents were considered ‘active’ participants in their community (i.e. were actively involved in two or more community groups). Residents outside the intervention area were significantly more likely to be active participants than residents within the intervention area (50.7 per cent vs 40.0 per cent; p < 0.001). Women tended to be more involved in community activities than men, but not significantly so (46.5 per cent vs 43.8 per cent; p = 0.224) and they were significantly more likely to participate in community activities if they lived outside the intervention area (50.9 per cent vs 40.1 per cent; p < 0.001). Men were also more likely to participate if they lived outside the intervention area (47.8 per cent vs 38.5 per cent; p < 0.021).

Figure 3.7 Satisfaction with infrastructure by proximity to VPUU intervention.

Levels of community participation were generally higher in informal areas than in formal areas (48.6 per cent vs 44.0 per cent; p = 0.006). The higher participation levels observed outside the intervention area were consistent in both informal areas (40.9 per cent vs 52.2 per cent; p < 0.001) and formal areas (48.8 per cent vs 39.5 per cent; p < 0.001) (see Figures 3.8 and 3.9).

In interpreting the community participation findings, it is important to note that the measure of community participation in this analysis is a composite of several different activities. The rationale for participation in voluntary organisations (security-related or otherwise) vs private interest groups (such as sports organisations) could be very different, and conflating these activities into a single participation measure might not be appropriate.

Figure 3.8 Community participation by neighbourhood within Khayelitsha.

It could be argued that VPUU would impact all forms of participation (e.g. sports clubs through improved infrastructure and church or book clubs by improving mobility and strengthening opportunities for interactions and social networks). However, it is more likely that any community participation effect from VPUU would be concentrated within the designated SNAs and would be focused on participation in voluntary organisations and voluntary security-related activities, particularly neighbourhood patrols.

The 2 km distance-based exposure used in the current analysis did not take into account the differential exposure of households within the SNA to these VPUU activities, which may have diluted any possible effect. Figure 3.9 is instructive in that within the area exposed to the VPUU infrastructure (as defined by the 2 km boundary), the areas within the demarcated SNAs appear to have relatively higher community participation scores.

Figure 3.9 Community participation by proximity to VPUU intervention.

In addition, there is emerging evidence that, in low-income contexts, community participation may be negatively associated with mental health, particularly for those from a relatively privileged socioeconomic position (Lloyd 2017). In this volume, Barolsky and Borges (Chapter 5) allude to the association between certain negative communitarian forms of community participation that are associated with interpersonal violence. This does seem evident when we compare areas with high levels of community participation (Figure 3.9) and those that experience high levels of interpersonal violence (Figure 3.2). Further analysis is required to understand the effect of the intervention on different forms of participation separately, and which forms of participation are in fact desirable in terms of strengthening communities and building social cohesion.

Social cohesion

Social cohesion was the only non-binary outcome and was recorded only in the 2015 survey. We used a narrow measure of social cohesion that evaluated four separate questions pertaining to interpersonal connections that provided a continuous score from 0 to 20. The mean score was 11.7. In an ethnographic study of the intervention area, Barolsky (2016) cautioned that the VPUU intervention’s attempts to formalise the urban space, while beneficial, might disrupt complex social networks by not fully utilising community resources and conceptions of social solidarity.

Using our measure of social cohesion, we could not find any empirical evidence to support this hypothesis in the overall sample, nor among male or female heads of households, nor in formal or informal areas, according to our measure (see Figures 3.10 and 3.11). We found, overall, that residents living within the intervention area had higher social cohesion scores than residents living outside the intervention area (average scores 12.0 vs 11.4; p = 0.004). Among females, there was a positive, but again not significant, association between social cohesion and proximity to the intervention compared to other areas (average scores 11.9 vs 11.4; p = 0.017), with similar levels recorded among males (average scores 12.0 vs 11.3; p = 0.117). There were no significant differences in average social cohesion scores when comparing female with male respondents (average scores 11.7 vs 11.6; p = 0.765).

Figure 3.10 Social cohesion by neighbourhood within Khayelitsha.

Figure 3.11 Social cohesion by proximity to VPUU intervention.

Residents living in formal areas had significantly, but modestly, higher average social cohesion scores than residents living in informal areas (average scores 11.9 vs 11.4; p = 0.002). A modest positive association between the intervention and higher levels of social cohesion was also found in the informal areas. Residents in informal areas living within the intervention area had higher social cohesion scores than residents of informal areas living outside the intervention area (average scores 12.0 vs 11.1; p < 0.001). In formal areas, social cohesion scores were similar between residents that lived close to the intervention and those that lived further away (average scores 12.0 vs 11.9; p = 0.762) (see Figures 3.10 and 3.11).

It may well be that the intervention could be improved through a more nuanced understanding of the pathways to enhance social cohesion. However, this initial analysis indicates that, far from the VPUU intervention having a negative impact because of its disruption of pre-existing non-Western communitarian forms of social cohesion, VPUU is either a beneficial addition to social cohesion or it is at least unlikely to be part of a zero-sum context between indigenous and colonially imposed dimensions of social cohesion.

One further aspect in relation to Barolsky’s (2016) work that we were able to address was the relationship between social cohesion and violence, in respect of which there is limited empirical research in South Africa. In our 2015 survey, residents that reported experiencing severe interpersonal violence had significantly lower social cohesion scores than those who did not experience severe interpersonal violence (average scores 11.0 vs 11.9; p < 0.001). This association leads to the thesis that increased social cohesion may reduce violence, or vice versa. What remains to be tested is whether the lower prevalence of exposure to severe violence in the intervention area results from the effect of the intervention on interpersonal violence directly or through the mechanism of improved social cohesion.


Two programmatic interventions in South Africa have already been shown to be effective in reducing violence through strong study designs: Stepping Stones, an HIV prevention programme promoting safer sexual behaviour among youth (Jewkes et al. 2006), and IMAGE, a microfinance intervention for the prevention of intimate-partner violence (Pronyk et al. 2006). Notwithstanding these positive developments, programmatic interventions face challenges with replication and scale-up. There remains a need to develop the evidence base for more complex interventions and passive structural interventions that address the urban space, given the prominence of such initiatives in policy documents.

In this study, comparison between households exposed versus those not exposed to the VPUU intervention area suggests a positive association between living in proximity (within 2 km) of the VPUU intervention and three of the five social outcomes: experience of violence, signs and symptoms of depression, and satisfaction with neighbourhood infrastructure. For social cohesion, scores were, on average, similar in intervention and non-intervention households, although slightly higher in the intervention area. There was a significant negative association between the intervention and community participation, but the nature of this participation and whether, as defined, it can be considered a characteristic with positive community impact is unclear. The findings were broadly consistent for both genders and for households in formal and informal areas.

In interpreting the statistical significance of comparisons between groups, the magnitude of the differences in outcomes must be considered. For most comparisons, these were modest. For violence and depression, the difference in prevalence between exposed versus unexposed households was approximately 5 per cent. For community participation, the difference in prevalence between exposed and unexposed households was also approximately 5 per cent, but in the opposite direction. For infrastructure satisfaction, the difference in prevalence between exposed and unexposed households was approximately 9 per cent. For social cohesion (measured on a scale of 1 to 20), the difference in mean scores between households inside versus outside the intervention area was 0.5.

These initial findings corroborate the positive effects of infrastructural and urban upgrading interventions such as VPUU and provide evidence to support recent policy developments that foreground environmental interventions. However, it should be noted that, except for the stratification by gender and formal versus informal areas, we have presented an unadjusted analysis that does not account for other independent factors known to influence these outcomes and which may also be associated with VPUU exposure. The findings could well be reflective of other differences across the exposure groups. Further analyses should also consider and adjust for potential confounding.

We also recognise the limitations of some of the simple binary measures and scores that we used to parsimoniously describe multiple outcomes and which may have diluted the estimate of the intervention effect. For example, our use of a binary exposure based on proximity to the intervention does not take into account more complex urban geography. Residents considered to be within the area of influence of the intervention may in fact have limited access due to the nature of road, rail, and pedestrian infrastructure. Investigation of the impact of VPUU could be enhanced through further analysis using more complex geoinformatic methods. In addition, the intervention area was defined based on proximity to the VPUU built environment and did not take exposure to VPUU social programming into account, which may also influence outcomes. We did not consider the cost implications of the intervention in this analysis, but note that replicating and scaling-up programmatic interventions in under-resourced contexts is also challenging and costly.

All of these aspects can be addressed through more complex multivariable analyses of each outcome. However, the study consists of a series of post-VPUU intervention cross-sections and therefore cannot address temporal aspects of cause and effect. For example, areas in Khayelitsha within 2 km of the intervention that have ostensibly benefitted from the intervention may always have been safer and more habitable. Alternatively, increased population density in the intervention area may also have caused a reduction in crime. Both these hypotheses would need to be tested through the analysis of longitudinal data that investigates risk prior to and after the intervention and measures possible displacement of violence to other areas.

The approach utilised in this chapter to evaluate the VPUU intervention could be applied to assess whether other large-scale infrastructure projects have a positive effect on social and health outcomes. This would provide further insight into the potential causal mechanisms through which interventions such as VPUU impact violent crime and other outcomes, whether any effect is due to the explicit inclusion of crime prevention through environmental design principles, or whether it is the infrastructure expenditure alone that matters. If there is indeed a benefit for multiple outcomes at a population level, then it would seem the approach adopted by VPUU is worthy of consideration.


1The VPUU methodology emphasises community participation and ownership. Once the boundaries of a SNA have been defined, all developments are preceded by a social compact, developed in consultation with community representatives and an elected representative body.

2Fieldworker quality control revealed inconsistent conceptualisation by fieldworkers and residents with the result that some muggings, in which people were threatened but not injured, were incorrectly recorded as “petty theft” instead of robbery.

3Examples include xenophobic or homophobic assaults.

4We also defined a measure for “any violence” reflecting a resident’s experience of one or more of the following incidents in the previous year: assault (without a weapon), murder, rape, robbery, arson, car hijacking, discrimination-based violence, drug-related violence, gang-related violence, community justice, domestic violence, and petty theft. Overall 32.5 per cent of residents experienced some form of violence, and the findings were broadly consistent irrespective of our choice of definition for experience of violence.


Andresen, Elena M., Judith A. Malmgren, William B. Carter, and Donald L. Patrick (1994). Screening for depression in well older adults: Evaluation of a short form of the CES-D (Center for Epidemiologic Studies Depression Scale). American Journal of Preventive Medicine 10(2): 77–84.

Barolsky, Vanessa (2016). Is social cohesion relevant to a city in the global South? A case study of Khayelitsha township. South African Crime Quarterly 55: 17–30.

Baron, Emily C., Thandi Davies, and Crick Lund (2017). Validation of the 10-item Centre for Epidemiological Studies Depression Scale (CES-D-10) in Zulu, Xhosa and Afrikaans populations in South Africa. BMC Psychiatry 17(6): 1–14.

Cassidy, Tali, Gabrielle Inglis, Charles Wiysonge, and Richard Matzopoulos (2014). A systematic review of the effects of poverty deconcentration and urban upgrading on youth violence. Health & Place 26: 78–87. www.ncbi.nlm.nih.gov/pubmed/24412655.

City of Cape Town (2013). City of Cape Town—2011 Census Suburb Khayelitsha. Cape Town: City of Cape Town http://resource.capetown.gov.za/documentcentre/Documents/Maps%20and%20statistics/2011_Census_CT_Suburb_Khayelitsha_Profile.pdf.

City of Cape Town Open Data Portal (2015). http://web1.capetown.gov.za/web1/opendataportal/default.

Civilian Secretariat for Police Service (2016). White paper on safety and security. Pretoria: Author. www.saferspaces.org.za/uploads/files/2016-WPSS.pdf.

de la Rey, Adri (2008). Enabling decision making with the SPOT 5 Building Count. PositionIT July/August: 33–38. www.ee.co.za/wp-content/uploads/legacy/Enabling%20decision%20making1.pdf.

De Silva, Mary J., Sharon R. Huttly, Turdy Harpham, and Michael G. Kenward (2007). Social capital and mental health: A comparative analysis of four low income countries. Social Science & Medicine 64(1): 5–20.

Jewkes, Rachel, Mzikazi Nduna, Jonathan Levin, Nwabisa Jama, Kristin L. Dunkle, N. Khuzwayo, Mary Koss, Adrian Puren, Kate Wood, and Nata Duvvury (2006). A cluster randomized-controlled trial to determine the effectiveness of Stepping Stones in preventing HIV infections and promoting safer sexual behaviour amongst youth in the rural Eastern Cape, South Africa: Trial design, methods and baseline findings. Tropical Medicine & International Health 11(1): 3–16.

Lloyd, Sam (2017). The association between community participation and mental health in two low-income communities in Cape Town, South Africa. Unpublished paper. Cape Town: University of Cape Town.

Lloyd, Sam, and Richard Matzopoulos (2018). Preventing violence in Cape Town: The public health evaluation approach. In Jennifer E. Salahub, Markus Gottsbacher, John de Boer, and Mayssam D. Zaaroura (Eds.), Social theories of urban violence in the Global South: Towards safe and inclusive cities (183–207). London and Ottawa: Routledge and International Development Research Centre (IDRC).

National Planning Commission (2012). National Development Plan 2030: Our future—make it work. www.gov.za/documents/national-development-plan-2030-our-future-make-it-work.

Newman, Peter, Leo Kosonen, and Jeff Kenworthy (2016). Theory of urban fabrics: Planning the walking, transit/public transport and automobile/motor car cities for reduced car dependency. Town Planning Review 87(4): 429–458. http://online.liverpooluniversitypress.co.uk/doi/10.3828/tpr.2016.28.

Pillay-van Wyk, Victoria, William Msemburi, Ria Laubscher, Rob E. Dorrington, Pam Groenewald, Tracy Glass, Beatrice Nojilana, Jané D. Joubert, Richard Matzopoulos, Megan Prinsloo, Nadine Nannan, Nomonde Gwebushe, Theo Vos, Nontuthuzelo Somdyala, Nomfuneko Sithole, Ian Neethling, Edward Nicol, Anastasia Rossouw, and Debbie Bradshaw (2016). Mortality trends and differentials in South Africa from 1997 to 2012: Second National Burden of Disease Study. Lancet Global Health 4(9): e642–e653. http://dx.doi.org/10.1016/S2214-109X(16)30113-9.

Prinsloo, Megan, Richard Matzopoulos, Ria Laubscher, Jonny Myers, and Debbie Bradshaw (2016). Validating homicide rates in the Western Cape Province, South Africa: Findings from the 2009 injury mortality survey. South African Medical Journal 106(2): 193–195.

Pronyk, Paul M., James R. Hargreaves, Julia C. Kim, Linda A. Morison, Godfrey Phetla, Charlotte Watts, Joanna Busza, and John D.H. Porter (2006). Effect of a structural intervention for the prevention of intimate-partner violence and HIV in rural South Africa: A cluster randomised trial. The Lancet 368(9551): 1973–1983. www.ncbi.nlm.nih.gov/pubmed/17141704.

Radloff, Lenore Sawyer (1977). The CES-D scale: A self-report depression scale for research in the general population. Applied Psychological Measurement 1(3): 385–401.

Western Cape Government (2013). Integrated Provincial Violence Prevention Policy Framework. Cape Town: Author.

World Bank (2018). World Bank GINI index. Geneva: Author. http://data.worldbank.org/indicator/SI.POV.GINI/.

4Towards safer communities

The impact of the Community Work Programme on the prevention of urban violence in South Africa

Malose Langa, Themba Masuku, and Hugo van der Merwe


Despite investing heavily in strengthening the criminal justice system over the last decade, levels of crime and violence are still high in South Africa. While extensive attention has been given to the role of police and the courts in curbing violent crime, social violence prevention strategies that show great potential, such as the Community Work Programme (CWP), have not been acknowledged or prioritised in addressing this problem.

The rates of crime and violence in South Africa are among the highest in the world. For example, the latest 2016–2017 crime statistics show an increase in the most significant categories of violent crime in South Africa: the murder rate has shown a steady increase over the last five years from a low of 15,554 murders in 2011–2012 to 19,016 in 2016–2017. This means the murder rate in South Africa is 34.1 murders per 100,000 people (Africa Check 2017), about five times the global average.

Other categories of crime and violence in South Africa, such as rape and violent public protests, are also significantly higher than in countries across the globe. This general high crime rate is taking place within the context of increased political and social tensions, and is manifest in attacks against foreign nationals and protests against poor delivery of basic services in low-income communities.

Against this backdrop, the Centre for the Study of Violence and Reconciliation (CSVR) undertook a study in six urban areas in the vicinities of Johannesburg and Cape Town to examine the impact of the CWP in preventing violence. The CWP is a government initiative under the aegis of the Department of Cooperative Governance and Traditional Affairs (COGTA), which was designed to provide “an employment safety net” to unemployed people by offering them regular work two days each week, totalling 100 days per year. In 2015, at the time of the study, general CWP participants were earning R700 (approximately US$50) per month and their coordinators were earning R2,000 (US$140) per month. In terms of the programme’s scope, any unemployed South African citizen over the age of 18 is eligible to join the CWP.

The research consisted mainly of individual and focus group interviews with CWP participants and coordinators, police, councillors, and other local stakeholders regarding the impact of the CWP in preventing urban violence. The study engaged intensively with the CWP in six diverse communities, namely, Ivory Park, Orange Farm, and Kagiso (situated in Gauteng Province); Bokfontein (North West Province); and Grabouw and Mannenberg (Western Cape).

The initial part of the chapter explains the inception and implementation of the CWP and provides a practical illustration of how the programme operates. It then reflects on the various avenues through which these activities may directly and indirectly impact on the present high levels of violence and crime experienced by these communities. We argue that the CWP, as a state-sponsored employment programme, has the (partially realised) potential to contribute to the prevention of urban violence in South Africa, and that this experience holds important lessons for other countries.

The inception and development of CWP within the South African context

The CWP emerged against the backdrop of a number of social problems, which include high levels of poverty and lack of employment opportunities. The official unemployment rate in South Africa is estimated to be at 27.6 per cent (Statistics South Africa 2017). The expanded unemployment rate which includes those who wanted to work but did not look for work rose to 36.8 per cent in 2017 (Reuters 2017).

Urban townships and informal settlements are particularly prone to high unemployment rates. Urban poverty, combined with extreme inequality, a legacy of apartheid political repression, and many other factors have given rise to various social ills, including HIV/AIDS, violent crime, alcoholism, gang violence, family violence, etc. (Seedat et al. 2009; Ratele 2013). Against this backdrop, in 2006–2007, the Office of the Presidency initiated various programmes aimed at creating job opportunities especially for people who were less likely to be employed in the formal sector, which seek people with specific skills and qualifications. According to one of the key CWP architects, the strategies were part of a programme to “address structural inequality, to restructure the economy and put South Africa on a new path” (Philip 2007: 8).

As a result of the high rate of unemployment, various government programmes were implemented to promote skills development and create work opportunities. The Expanded Public Works Programme (EPWP) was launched as a poverty-alleviation strategy by providing temporary employment through involvement in infrastructural and technical projects such as building roads, dams, sewerage systems, storm water drains, and bridges (Nzimakwe 2008). EPWP participants were provided with 100 days of full-time work within major infrastructure projects run by various government departments, mainly the Department of Public Works. The main aim of the EPWP was to provide participants with technical skills and training in the hope that they would become permanently employed once they had completed their involvement (Nzimakwe 2008). Major criticisms made against the EPWP were that it provided too few days of work for people to learn skills that would sufficiently equip them for permanent formal employment, and that the projects were mainly state-driven rather than community-driven and -oriented (Nzimakwe 2008).

In response to these criticisms, the CWP was established in 2007 as a pilot project. Like the EPWP, it offers 100 days of part-time work, spread over two days a week throughout the year, but the CWP generally extends over multiple years, compared to the short-term contracts offered by the EPWP. The work to be done through the CWP is selected through a participatory process that identifies “useful work” that the community feels will contribute to the public good and improve the community’s quality of life, whereas EPWP projects are identified directly by the state.

Philip (2013) provides positive indications of the effectiveness of the CWP in alleviating poverty and unemployment in communities: by 2015, 186 sites had been established, operating in 140 municipalities across the country, with approximately 1,000 participants per site (COGTA 2015: 4). In his State of the Nation Address in January 2016, President Jacob Zuma asserted that the state aims to increase CWP participants to one million. The CWP is not intended as a provider of permanent employment but rather as an “employment safety net” while people are still searching for permanent job opportunities (Philip 2013). However, the reality is that many stay in the programme over multiple years.

Participants work at a community level to provide a wide range of social services, which vary from one site to the next. They include: early childhood support, maintaining community gardens that grow food for indigent families, care for the elderly and the sick, and street cleaning. Additionally, some of the projects are directly focused on addressing local safety concerns. For instance:

conducting community patrols of crime hotspots

cutting long grass to create safer living environments

recruiting ex-offenders into the project as part of the reintegration strategy

organising campaigns against crime

providing education and support for substance abuse

assisting victims of domestic violence.

The CWP was not developed as a violence prevention programme but, as our research demonstrates, it presents a viable strategy to prevent violence in communities through addressing the underlying and direct causes of violence. This was confirmed in the individual and focus group interviews that CSVR researchers conducted with ward councillors, police officers, community police forum leaders, police officials, school principals, NGO representatives, and CWP representatives (participants, coordinators, facilitators, and site staff members). Interviews with all these participants were wide-ranging, covering the history of the CWP, its recruitment processes/identification of potential participants, and the community projects undertaken to prevent violence.

This chapter seeks to illustrate how various initiatives taken by CWP participants are contributing to violence prevention. It will also serve to illuminate how the violence prevention potential of these projects could be upscaled. The limitations of these initiatives will also be discussed, particularly the gender dynamics that impact on their effectiveness.

Impact of CWP in violence prevention initiatives

As discussed earlier, crime and violence are major concerns for many South African communities. So it is not surprising that some of the projects initiated by the CWP participants were aimed at reducing the crime and violence in their communities. Examples of various crime and violence prevention initiatives through CWP are outlined below.

Cutting grass and community patrols

Crime surveys in South Africa show that common street robberies, which involve personal possessions being taken through the threat or use of force, are the most frequent everyday experience of crime in many urban settings that shape people’s sense of safety (Bruce 2010). These robberies are influenced by environmental factors such as streetlights or long grass where perpetrators can easily hide. Given these concerns about common street robberies, one of the key activities in which CWP participants engaged in many urban settings involved cutting the long grass in “crime hot spots” where people have been attacked (Bruce 2015b). According to Bruce (2015b), this is an example of “environmental” work done by CWP participants that has a safety focus; it aims to create a safe environment so people can walk freely in open spaces that are generally considered to be unsafe and dangerous. Cleaning the environment and engaging with local governmental authorities to fix street lights created some safety within these spaces. Local residents interviewed asserted that the cutting of long grass has significantly increased their safety.

This corner used to have a lot of trees and very long grass. You see, criminals will hide there at night and take people’s bags when they come from work, and then run into those bushes. We now, as CWP, decided to clean that rubbish. The area is clean and there is no more crime. We cleaned that place as CWP.

(Individual interview with CWP participant)

We clean parks.… We cleaned this other bridge where street boys used to stay and rob people. We have patrollers in that place when people come from the train so that no one takes their bags.

(Individual interview with CWP participant)

In addition to cleaning the environment, CWP participants were also involved in patrols to make streets safer, especially at night and in the early hours of the morning when many common street robberies occur. It was reported in the interviews that the presence of community patrollers on the street discouraged people from participating in crime.

People used to get mugged in the mornings. And then we’d patrol there … at 6:00. [From 7:00 to 8:00] we would go to the schools to escort children.… We would go back at 12:00 until 15:00 when children go back home.… They’d fight each other, but when they see us, they would not fight.… Their mothers were happy when they see us standing guard there to make sure the children are safe.… So people were happy, even my neighbours were happy. And when we were not there they would complain.

(Group interview with CWP patrollers)

People rely on community patrollers for their own safety in this community. Criminals know us that we patrol the streets to stop crime.

(Individual interview with a CWP patroller)

In some communities, CWP patrollers also accompany children on their way to and from school to protect them from being victims of crime or car accidents (Bruce 2015a). In areas where there is a clash between gangs, the CWP members ensure that children are taken off the streets. CWP participants played this role to ensure the safety of children in Manenberg, a township in Cape Town characterised by high levels of gang violence. After an intensification of gang violence in Manenberg in 2014, the CWP site management initiated a campaign against it. Several marches took place under the banner “Take Back Our Streets” (Mullagee and Bruce 2015). Although gang violence remains a critical problem in this community, many informants interviewed asserted that these public campaigns help in raising awareness about crime and violence in Manenberg.

At the moment the campaign is only between 5:00 and 6:00 p.m. We want to increase our time on the road, but people are scared of what might happen to them if they stay out late.

(Group interview with CWP participants)

The visibility of the CWP-initiated “Take Back Our Streets” campaign increased both inside and outside the community as it gained momentum. Every Wednesday evening, there were marches against gang violence.

(Group interview with CWP participants)

The CWP participants often work closely with the police and other law enforcement officials, including community policing forums. In the Western Cape, some of them receive training provided by the provincial Department of Community Safety. In Orange Farm, a township on the outskirts of Johannesburg, CPW participants worked closely with the police to address the problem of domestic violence. In the context of increasing levels of crime and violence in South Africa, police stations were mandated to establish a Victim Empowerment Programme (VEP) to support victims of crime and violence in avoiding secondary victimisation. The CWP participants’ work on this project involves providing basic counselling to survivors of domestic violence. They also help survivors obtain protection orders by assisting them in completing the necessary paperwork and accompanying them to court to file their cases. CWP participants also work with local organisations that assist abused women. Their long-term goal is to ensure that their VEP service is provided 24 hours a day, especially on the weekends, as many cases of domestic violence occur due to heavy drinking during this period.

In addition to the direct assistance provided to victims, public campaigns are held by CWP participants involved in the VEP to raise awareness about domestic violence in Orange Farm. These public campaigns are conducted in collaboration with the police and other local organisations. The campaigns acknowledge that although women are more often victims of domestic violence, men are also encouraged to seek help if they are in abusive relationships. This is in line with the emerging literature that it is important for campaigns opposing domestic violence to include men as victims, rather than to simply treat them as potential perpetrators of domestic violence (Sathiparsad 2008).

Ex-offenders and anti-crime initiatives

In South Africa, it is estimated that close to 6,000 offenders are released from various correctional facilities each month, either through parole or completion of their sentences (Muntingh 2008). It is hoped that upon their release many will no longer engage in criminal activities. However, due to the lack of dedicated interventions aimed at facilitating their reintegration, many ex-offenders re-offend and return to prison. Currently, there are no reliable statistics on recidivism in South Africa, but it is estimated to be high (Gaum, Hoffman, and Venter 2006). Many ex-offenders face challenges in finding employment, increasing their risk of re-offending and returning to jail (Muntingh 2008). At some sites, such as Manenberg, Orange Farm, and Ivory Park, ex-offenders were recruited to participate in the CWP. Being able to join the CWP facilitated their reintegration into the community and decreased their risk of re-offending.

Going [to] prison was [easier] than coming out of prison. In prison there are many programmes that are provided and I am glad that I received a lot of training while in prison. I am now a qualified chef and can prepare a variety of dishes and menus. The biggest problem is finding a job with a criminal record; it’s not easy, my brother … I am lucky to be in the CWP soon after I came out of prison.

(Individual interview with an ex-offender working in the CWP)

Believe me, my friend, I would not be here if it was not for CWP. I would be back at jail but I can say CWP has saved [me].

(Individual interview with an ex-offender working in the CWP)

Ex-offenders in Orange Farm and Ivory Park have also been involved in activities with a specific crime prevention focus such as going into local schools to encourage young people to stay away from criminal involvement. They use their own life stories as examples to others (especially youth) that “crime is not good” and that “crime does not pay”.

We want to spread a message that crime is not good.… We have public anti-crime campaigns.… We tell [young people] as ex-offenders because we know that crime is not good.

(Group interview with CWP ex-offenders)

With crime prevention programmes in CWP, we have a project whereby we motivate young people [in and out of school] not to do crime and drugs.

(Group interview with CWP ex-offenders)

Uggen and Staff (2001) argue that the involvement of ex-offenders in various community projects can be a “turning point” in their lives, motivating them not to reoffend. Bruce (2015c) contends that involvement in community work hastens ex-offenders’ reintegration into society and enables them to resist returning to a life of crime. It was also evident from the interviews that ex-offenders saw their involvement in the CWP as positive, giving meaning to their lives through the work they were doing in schools and the wider community. Other campaigns by ex-offenders included public campaigns aimed at preventing substance abuse among youth.

However, other studies (Petrosino, Turpin-Petrosino, and Finckenauer 2000; Petrosino, Turpin-Petrosino, and Buehler 2013) have questioned the effectiveness of awareness raising campaigns about the risks associated with crime. These studies argue that campaigns, such as Scared Straight in the USA, have been found to be ineffective in deterring young people from involvement in criminal activity. Despite reviews questioning their effectiveness, these campaigns still remain popular. Follow-up interviews and detailed studies still need to be done on the effectiveness of the CWP and its prevention of violence, but preliminary findings show that it is effective in providing ex-offenders with an opportunity to reintegrate into society and play a constructive role in assisting other youth to avoid risk-taking behaviour.

Gender dynamics in the CWP

The majority of participants at all CWP sites are women. From April 2014 to March 2015, there were 202,599 participants at 186 CWP sites across South Africa. Of this total, 75 per cent were women, and 42 per cent of those were women over 35 years of age. Women represent the majority of coordinators, office staff, and participants at all the CWP sites (Brankovic 2016). Men represented 25 per cent of CWP participants during the same period, with 13 per cent of total participants being men over the age of 35 (Brankovic 2016).

A number of reasons were advanced to explain why there are fewer men in the CWP than women. These include the issue of male pride and differences in the approaches that men and women have to the low wages paid by the CWP.

A lot of men refuse to join the CWP because they say that the money is too little. However, for most women even if the money is little we are able to persevere and make it work. Other men are full of pride and refuse to work for little money or be associated with a poverty relief project.

(Focus group interview with CWP participants)

This is a very difficult job for many men to do. It is not easy for a man to be seen working in a poverty relief project and earn the kind of money that we earn.… Men would rather try other means like looking for piece jobs and marketing themselves elsewhere in order to get better-paying jobs. Just look at my group here—I have only 3 men out of 30 people. As you can see, the men that are working are very old and no longer have stamina to take on difficult jobs.… It’s difficult for unemployed younger men who still have school-going children to enrol in the programme also because money will definitely not be enough to sustain the family.

(Focus group interview with CWP participants)

Interviews with male participants provided a different set of reasons as to why so few men join the CWP programme, despite the fact that many are unemployed. The most common reason given was the lack of financial security in the CWP.

It’s difficult for unemployed young men to join the CWP because the money we get is pathetic. Not many men who have huge responsibilities of raising children would be keen to work for R71 [about US$5.75] per day. Most men believe that working for CWP is a waste of their time because they can make more money per day than CWP participants make in a month. I mean it’s easier to say men don’t want to work hard but the reality is that the CWP income is very little if you have [the] responsibilities of raising children.

(Focus group interview with CWP participants)

Another male respondent concurred:

I personally think that it is not that men are lazy or don’t want to work hard, it is because the money is very little. Remember that when this project was introduced, we were the first people to join the CWP and there were many men who joined the programme but left because we were promised permanent jobs but that never happened.…

(Individual interview with a male CWP participant)

The reasons for women joining the CWP were generally different from those of men, although both emphasised the issue of providing food security to their families. Many women who were interviewed had never worked before and had opened a bank account for the first time when they joined the CWP. For these women, being part of the CWP gives them an income, ensuring that they are able to raise their children and be financially independent of their partners. Many women in the CWP have young children and some receive a child support grant from the government. As men are not eligible for such grants, they have fewer options to supplement their CWP income. As a result, men’s income is relatively low compared to that of some of the women they work with in the CWP. It is possible that this impacts men’s sense of masculinity, as men traditionally associate work with a particular social status and the ability to support their family by being the primary breadwinner (Hunter 2006).

Some CWP participants suggested that men fear they will be ridiculed or lose social status if they participate in the programme. This is partly because some men feel the CWP is not a “real” job, as men tend to associate work with going to the factories five days per week and working under highly demanding conditions. This point is well illustrated by Catherine Campbell (1992) in her work on men in the mines who link their work underground with constructions of hegemonic masculinity. According to Campbell (1992), it is in these diamond and gold mines that “masculinity emerged as a master narrative”. The masculine identity was then developed as an enduring symbol of bravery, strength, and hard work. It is possible that men may consider CWP work to be too feminine.

Men stay at home—they feel “what are people going to say?” if they work here. But women do not care.

(Individual interview with CWP participant)

Men think a real job is about going to the firms and working the whole week.

(Individual interview with CWP participant)

One CWP senior representative suggested that opportunities should be created within the programme for men to learn new technical skills, such as welding and plumbing. It was implied that learning these skills would make men more interested in the programme as it would facilitate their employability beyond the CWP, and that such technical skills would connect with their sense of masculinity. It is also possible that some men are uncomfortable with the idea of working in the CWP as they regard work involving vegetable gardening, home-based care, and early childhood projects as “women’s work”. In terms of gender roles, these activities are traditionally reserved for women and considered emasculating for men. It is therefore important that men’s lack of participation in the CWP is seen within the context of the kind of work that the CWP does. This raises the question of whether men would participate in the CWP in greater numbers if the nature of the work changed.

On the other hand, the CWP does empower women economically and socially. There is, however, emerging evidence that some men feel disem-powered when their female partners are the main breadwinners in the household while they are not able to contribute anything financially due to their unemployed status. It has been suggested that this may increase the risk of domestic violence (Bruce 2015a; Brankovic 2016). However, the benefits of women being part of the CWP outweigh these risks. Female participants value their participation, the support they receive through the CWP, and the new skills that they acquire in the programme (Brankovic 2016).

Policy implications of CWP in preventing violence

There are many policy implications for the CWP. The impact of the CWP on violence can be examined in relation to the various underlying and direct factors that the different projects seek to address. These include:

addressing underlying causes, such as poverty

direct factors, such as the integration of ex-offenders

long-term prevention, such as early childhood development

short-term measures, such as crime patrols.

In addition, the way that the CWP is implemented—through community participation in its design, collaborative inclusive implementation across divisions in a community, and through building agency in a local community—serves as a key contributor to building community cohesion and local capacity to address social problems. A strong example of the success that can result from a CWP comes from Bokfontein, a community created when the citizens of two communities were evicted and relocated in 2005–2006. The formation of a CWP in Bokfontein brought harmony and peace to the relationships between the citizens of the original two communities, including the absence of xenophobic violence (Langa and von Holdt 2011).

A common CWP initiative involves the provision of early childhood services. The existing literature shows that early childhood programmes can serve as a violence prevention mechanism. For example, a study conducted by Berk (2001) proved that children who have attended early childhood centres have higher self-esteem. This self-confidence helps children to resist peer pressure to engage in risk-taking behaviours such as crime and violence (Berk 2001). Peer pressure works easily on children who are in dire need of love and approval from friends or others. Research in this area shows that early childhood programmes can play an important role in preventing violence. At a policy level, it is therefore important that early childhood services are made compulsory for all children in South Africa as this may help children to develop cognitively, socially, and psychologically.

For many participants, the CWP contributes to economic inclusion. Ordinarily many people would not be doing useful community work if it was not for CWP. The CWP provides participants with some training/skills and facilitates their economic inclusion, despite the fact that the money is modest. Some participants asserted that they are able to use these meagre resources to cover basic needs. At a policy level, attempts need to be made to try and recruit young men to social programmes such as CWP. This may help to reduce crime and violence. Many research studies (Newburn and Stanko 1994; Newburn and Shiner 2006) have shown that young people are more likely to be involved in criminal activities. The CWP may also be used as a space in which to provide mentorship about what it means to be a young male in the new South Africa, and to identify opportunities that these young men may explore rather than resorting to crime and using drugs.

The CWP facilitates the development of social cohesion through various processes, which in turn play a role in violence prevention. Before the inception of the CWP in the six communities studied, many community members did not know one another despite living in the same neighbourhood. The inception of the CWP facilitated the process of participants getting to know one another, which served as the first step towards them working together to assist their community members. The spirit of togetherness and oneness is a key element in the CWP’s work. For example, CWP participants meet once or twice a week to discuss work to be undertaken in the community. These weekly meetings enhance social relations and the spirit of collegiality among community members. In addition, CWP participants are typically organised to work in teams. Generally, the CWP appears to foster a high level of cohesion among community members, drawing participants from different neighbourhoods to work together for the betterment of their communities. At a policy level, the CWP may assist in bringing social cohesion, peace, and reconciliation to communities if implemented in an inclusive and participatory manner.

Concluding remarks

It is important to recall that the CWP was not specifically intended to be a crime- and violence-prevention programme when the government conceived it. However, given the ways that safety concerns have been prioritised by local communities, the CWP has framed local projects so as to directly address the causes of violence and it has the potential to play an expanded role in this arena. In communities that have been torn apart by violence and conflict (both political and interpersonal), the CWP offers a model for restoring agency to a community to address its priority concerns in a way that builds social cohesion and restores the capacity to act collectively to address violence.

It is, however, important to acknowledge that the effectiveness of the CWP differs from one community to another, depending on how it is implemented. In the sites studied, the programme’s success depended entirely on the participants’ commitment to implementing violence prevention initiatives. Some policy impact has been made through this work and the government now acknowledges that the CWP does have the potential to prevent violence in communities, as shown through our work and the policy briefs we have written (see Bruce 2015b, 2015c; Brankovic 2016; Langa 2017).


Africa Check (2017). FACTSHEET: South Africa’s crime statistics for 2016/17. https://africacheck.org/factsheets/south-africas-crime-statistics-2016/17.

Berk, Laura E. (2001). Development through the lifespan, second edition. Needham Heights, MA: Allyn and Bacon.

Brankovic, Jasmina (2016). Women’s contribution to social cohesion and violence prevention through the Community Work Programme. CWP Policy Brief 3. Johannesburg: Centre for the Study of Violence and Reconciliation.

Bruce, David (2010). The ones in the pile were the ones going down: The reliability of violent crime statistics. South African Crime Quarterly 31: 9–17.

Bruce, David (2015a). Preventing crime through work and wages: The impact of the Community Work Programme. South African Crime Quarterly 52: 25–37.

Bruce, David (2015b). The good, the better, and the best: How the Community Work Programme can reach its full potential as an instrument of community development in South Africa. CWP Policy Brief 1. Johannesburg: Centre for the Study of Violence and Reconciliation.

Bruce, David (2015c). Working for safety: The Community Work Programme as a tool for preventing violence and building safer communities. CWP Policy Brief 2. Johannesburg: Centre for the Study of Violence and Reconciliation.

Campbell, Catherine (1992). Learning to kill? Masculinity, the family and violence in Natal. Journal of Southern African Studies 18(3): 614–628.

COGTA: Department of Cooperative Governance and Traditional Affairs (2015). Community Work Programme: Implementation manual. Pretoria: Author.

Gaum, Goudine, Sandy Hoffman, and Jan H. Venter (2006). Factors that influence adult recidivism: An exploratory study in Pollsmoor prison. South African Journal of Psychology 36(2): 407–424.

Hunter, Mark (2006). Fathers without Amandla: Zulu-speaking men and father-hood. In Linda Richter and Robert Morrell (Eds.), Baba: Men and fatherhood in South Africa (99–107). Cape Town: HSRC Press.

Langa, Malose (2017). Reintegration of ex-offenders: Opportunities for the Community Work Programme to assist in the community reintegration of ex-offenders. CWP Policy Brief 4. Johannesburg: Centre for the Study of Violence and Reconciliation.

Langa, Malose, and Karl von Holdt (2011). Bokfontein amazes the nations: Community Work Programme (CWP) heals a traumatised community. In John Daniel, Prishani Naidoo, Devan Pillay, and Roger Southall (Eds.), New South African review 2: New paths, old compromises? (256–274). Johannesburg: Wits University Press.

Mullagee, Fairuz, and David Bruce (2015). Building a good nation in Manenberg: A case study of the Manenberg Community Work Programme. Johannesburg: Centre for the Study of Violence and Reconciliation.

Muntingh, Lukas M. (2008). Prisoner re-entry in Cape Town—An exploratory study. Civil Society Prison Reform Initiative (CSPRI) Research Paper No. 14. Cape Town: CSPRI.

Newburn, Tim, and Michael Shiner (2006). Young people, mentoring and social inclusion. Youth Justice 6(1): 23–41.

Newburn, Tim, and Elizabeth A. Stanko (1994). Just boys doing business: Men, masculinities and crime. London: Routledge.

Nzimakwe, Thokozani Ian (2008). Addressing unemployment and poverty through public works programmes in South Africa. International NGO Journal 3(12): 207–212.

Petrosino, Anthony, Carolyn Turpin-Petrosino, and James O. Finckenauer (2000). Well-meaning programs can have harmful effects! Lessons from experiments in Scared Straight and other like programs. Crime & Delinquency 46(3): 354–379.

Petrosino, Anthony, Carolyn Turpin-Petrosino, and John Buehler (2013). Scared Straight and other juvenile awareness programs for preventing juvenile delinquency: A systematic review of the randomized experimental evidence. The Annals of the American Academy 589: 41–62.

Philip, Kate (2013). Transformative potential of public employment programmes. Graduate School of Development Policy and Practice Occasional Paper Series No. 1/2013. Cape Town: University of Cape Town.

Philip, T.K. (2007). Enterprise on the margins: Making markets work for the poor? PhD thesis, University of the Witwatersrand, Johannesburg.

Ratele, Kopano (2013). Subordinate Black South African men without fear. Cahiers d’Études africaines 53(209–210): 247–268.

Reuters (2017). South Africa’s unemployment rate steady at 27.7 percent. www.reuters.com/article/safrica-unemployment/south-africas-unemployment-rate-unchanged-at-27-7-pct-in-q3-idUSJ8N1MO005.

Sathiparsad, Reshma (2008). Developing alternative masculinities as a strategy to address gender-based violence. International Social Work 51(3): 348–359.

Seedat, Mohamed, Ashley Van Niekerk, Rachel Jewkes, Shahnaaz Suffla, and Kopano Ratele (2009). Violence and injuries in South Africa: Prioritising an agenda for prevention. The Lancet 374: 1011–1022.

Statistics South Africa (2017). Quarterly Labour Force Survey. Pretoria: Author. www.statssa.gov.za/?p=10658.

Uggen, Christopher, and Jeremy Staff (2001). Work as a turning point for criminal offenders. Corrections Management Quarterly 5(4): 1–16.

5Is social cohesion the missing link in preventing violence?

Case studies from South Africa and Brazil

Vanessa Barolsky and Doriam Borges


This chapter constitutes a comparative investigation of the role of two major violence prevention interventions in South Africa and Brazil in building social cohesion and collective efficacy based on ethnographic studies of these programmes. Violence Prevention through Urban Upgrading (VPUU) was implemented in the township of Khayelitsha in Cape Town, South Africa, while Pacifying Police Units (known by its Portuguese acronym, UPP) was set up in the favelas of Rio de Janeiro, Brazil.

Social cohesion is a broad concept that has been widely deployed in global policy discourse since the 1990s. The South African Presidency has described social cohesion as “the factors that hold a society together” (The Presidency, Republic of South Africa 2004: iv). These “factors” refer to a broad array of social characteristics perceived to contribute to connectedness and solidarity within society, including the ties that bind citizens to each other and to the state. They encompass questions related to a sense of belonging and trust, to the form and extent of cooperation between citizens, as well as to shared inclusion in economic and political life.

Social disorganisation theory posits that local social cohesion or connectivity between citizens is critical to mediating the link between urban disadvantage and crime. It argues that social ties, shared norms, and social interaction are important because they create informal social controls which inhibit “delinquent” behaviour if residents are prepared to collectively intervene for the “common good” (Shaw and McKay 1942; Sampson, Raudenbush, and Earls 1997). Collective efficacy investigates how social ties can prevent violence when they are translated into collective action at a neighbourhood level through informal social controls. Sampson, Raudenbush, and Earls (1997: 918) define collective efficacy as “social cohesion among neighbours combined with their willingness to intervene on behalf of the common good”.

Thus far, however, there has been little empirical research or theoretical engagement with social cohesion and collective efficacy and their relationship to violence in the specific conditions of the Global South, and even less research about how violence prevention interventions mobilise social cohesion in these contexts. Both interventions in this study were broadly influenced by theories around social disorganisation which link the spatial distribution of crime to concentrated disadvantage in particular urban neighbourhoods. Addressing crime is therefore envisaged as involving spatially targeted interventions that also try to change social relations in the neighbourhoods where they are implemented.

Although both programmes were broadly informed by social disorganisation theory, VPUU and UPP implemented what Muggah (2012: ix) has categorised as “soft” (voluntary) and “hard” (coercive) approaches, respectively, that attempted to reclaim urban space in different ways. UPP seeks to forcibly re-establish state authority in favelas to integrate them into the wider city and to reinstate a normative social order without drug traffickers. “Softer” approaches such as that used by VPUU seek to restore “order” to the urban environment through a variety of urban renewal and upgrading initiatives combined with social interventions aimed at changing social and economic relations.

Shaw and McKay’s original Community Action Programme in the suburbs of Chicago, based on social disorganisation theory, has been described as “a conceptual framework meant to encourage and facilitate indigenous social invention” (Schlossman and Sedlak 1983: 428). In their subsequent work, building on Shaw and McKay’s original project, Sampson, Raudenbush, and Earls (1997: 918) sought to explore this “indigenous” capacity by investigating “the effectiveness of informal mechanisms by which residents themselves achieve public order”. However, as Warner, Beck, and Ohmer (2010: 355) point out, “the humanistic philosophy underlying social disorganization theory” that “encouraged the community to develop their own solutions to problems and were based primarily on providing social support” has been subsequently undermined by interpretations of disorganisation theory that focused on increasing surveillance by communities of anti-social conduct and reporting to police. This trend was reflected in the interventions that took place in South Africa and Brazil and both struggled to make the most effective use of “indigenous social invention”.

The fieldwork we conducted in South Africa and Brazil revealed the diverse roles that violence prevention interventions can play in building the type of social cohesion that can have a long-term violence prevention effect.

Our chapter explores how VPUU, a violence prevention intervention co-funded by the German Development Bank that seeks to address conditions of violence and poverty in Khayelitsha, South Africa, interacted with existing forms of social relation and organisation. Drawing on international development policy discourse, we argue that the intervention was informed by a Euro-American imaginary of a managed society characterised by ordered and economised social relations founded on a normative conception of a formalised city and the economic-rational actor. This conception of the city has been critiqued by a growing body of urban scholars who have sought to understand cities in the Global South on their own terms rather than evaluating their “failure” against the ideal of the Western city (Nuttall and Mbembe 2008; Myers 2011; Pieterse 2011; Roy 2011; Parnell and Robinson 2012; Pieterse and Simone 2013). VPUU’s conception of the city was profoundly contested by the material reality of informal norms of regulation and control in the fluid space of Khayelitsha where “everyday” networks often have significant social and symbolic resonance (Tshehla 2002; Shearing and Wood 2003; Steinberg 2012; Gillespie 2014; Super 2016). As a result, the role of the intervention in building social cohesion and collective efficacy was ambiguous.

Ethnographic research in two favelas in Rio de Janeiro, Tabajaras/Cabritos and Cidade de Deus, where the UPP intervention was implemented, similarly showed complex effects in relation to social cohesion. We found that, paradoxically, where the intervention was most successful in reducing violence it had the most negative effect on social cohesion. In the community of Tabajaras, the reduction of violence brought about by the intervention led to a process of gentrification with new residents moving into the area, resulting in the decline of existing networks of cohesion. In the community of Cidade de Deus, the UPP was not as successful in reducing violence as in Tabajaras, while also complicating existing forms of cohesion and governance as the police failed to displace the drug dealers who were becoming a competing source of authority. At the same time, the UPP was perceived to have had a negative effect on existing social organisations in the favela as they took over the roles of community organisations and appropriated funding previously directed to them.


An ethnographic methodology was used in this research to help elucidate how social relations and cohesion were understood—and produced—by social actors themselves. Ethnography is a widely used qualitative methodology that seeks to interpret the meanings located in social and cultural systems, as well as the way in which our understanding of the world is socially constructed within these contexts (Fetterman 2010; O’Reilly 2012).1 The study therefore sought to understand how shared “webs of significance” (Geertz 1973: 5) or perceptions of “reality” in communities shaped conceptions of social relations and social norms and the impact this had on the manner in which individuals understood and interacted with the VPUU and UPP interventions. As a result, this investigation did not attempt to carry out a quantitative measurement of the impact of the interventions on violence or social cohesion that could provide a conclusive finding on whether the programmes decreased violence or increased social cohesion. Instead, it sought to analyse what meanings social actors attributed to the interventions, including how they were understood and experienced within the context of existing social relations and cultural norms.

Ethnography is valuable when conducting evaluation work as it examines human behaviour in relatively natural conditions, which allows for a deeper understanding of the social context in which a programme takes place. Most importantly, ethnographic approaches to evaluation emphasise the need to critically engage with the perspective of all stakeholders (Youker 2005), including those who may be less empowered (Lay and Papadopoulos 2007). In line with the phenomenological approach of ethnography, which recognises that different groups and individuals may have varying perceptions of reality and that these conceptions of reality inform their actions and beliefs, the central research goal of ethnographic approaches to evaluation is to “ ‘get inside’ the world of program participants and to understand, describe, and explain the program and its effects from participants’ points of view” (Dorr-Bremme 1985: 65), as well as “its consequences in terms of participants’ realities and meaning systems” (Dorr-Bremme 1985: 66). The analysis of a programme in terms of the “reality” of programme participants is important because “it is in terms of those (and not the evaluator’s concepts of reality and meaning) that program participants routinely take action” (Dorr-Bremme 1985: 67). Understanding the perspective of those who engage with programmes therefore helps unpack their effects and outcomes.

Like quantitative researchers, qualitative social science researchers need to establish the credibility and trustworthiness of their research. While quantitative researchers use the concepts of “validity” and “reliability” to measure the credibility of their research, evaluating the trustworthiness of qualitative research relies on different criteria and standards.2 Ethnographic researchers use techniques such as prolonged engagement,3 thematic analysis,4 and “peer debriefing”5 to ensure their work meets accepted standards of academic rigour.

Ethnography in general tends to emphasise an inductive approach to data collection and analysis, where the researcher goes to the field with an “open” mind, which allows for the discovery of unanticipated “serendipity” (Bryant and Charmaz 2007) in the data (i.e. rich new data that were not necessarily included in the initial research design). At the same time, the research was clearly iterative. No researcher goes into the field without being informed by certain ideas, concepts, and theories (O’Reilly 2012). Nevertheless, these concepts and theories were “played against systematically gathered data” (Strauss and Corbin 1994: 277) in a continuous “conversation” (Strauss and Corbin 1994: 280) between extant theory and newly collected information. This approach of carefully grounding the analysis in the data supports what Lincoln and Guba (1986) have identified as another important criterion of trustworthiness in qualitative research, namely, “confirmability”, which relates to demonstrating that the researcher’s findings and interpretations are clearly derived from the data and are not distorted by bias. This requires the researcher to “demonstrate how conclusions and interpretations have been reached” (Nowell et al. 2017: 3).

In developing the analysis, efforts were made to triangulate data from multiple sources, although this was inhibited in some instances by limited access to internal documentation from either of the interventions in Brazil or South Africa. For this reason, the study included a review of both academic and grey literature on the programmes and related issues. It also solicited the viewpoints of community and government stakeholders and professional practitioners.6 Transferability7 and dependability8 are two further criteria for establishing the trustworthiness of qualitative research that we have built into our approach.

Violence prevention through urban upgrading in South Africa

The VPUU intervention was investigated as a socially and culturally embedded undertaking through an analysis of empirical material relating to three key contexts where contention emerged around the programme:

the consultative processes in which it engaged

its efforts to formalise informal trading

its efforts to formalise community patrolling.

Spaces of contention provide rich data for understanding the interaction between VPUU and Khayelitsha residents, in addition to the effects of the intervention in relation to social cohesion.

The research exposed some of the difficulties of translating models of social order and citizenship that inform violence prevention interventions formulated in the Global North to an urban space—such as Khayelitsha—characterised by deep informality, an ethics of communitarianism, and local conceptions of community justice that involve violence. Tensions around these contending rationalities are not unique to the VPUU intervention. As Watson (2003: 401) argues,

a vast gap exists between the notion of “proper” communities held by most planners and administrators (grounded in the rationality of Western modernity and development), and the rationality which informs the strategies and tactics of those who are attempting to survive, materially and politically, in the harsh environment of Africa’s cities.

While Southern cities are often viewed as places “bursting at the seams, overtaken by their own fate of poverty, disease, violence, and toxicity”, Roy (2009: 820) argues that they are also places where those living in abject poverty apply their resourcefulness and innovation to survive and thrive. This is not to idealise these urban environments but to argue for an analysis that attends to the way in which their current conditions and modes of organisation are the result of a complex legacy of colonialism, capitalism, globalisation, and extraordinary rates of recent urbanisation. In addition, inadequate policy and planning has left many of these urban centres struggling to cope with the multiple social challenges created by this complex legacy. In a context like this, “illegality and informality tug at the normative roots of the state leading to an arena charged with the violence of and toward the governed” (Rao 2006: 229).

Nevertheless, the imaginary of an “ordered” city, founded on the model of European and American urbanisation, remains the primary conceptual framing for development interventions such as those funded/co-funded by the German Development Bank and implemented by VPUU in Khayelitsha. In this vision of the city, urban planners favour formality, order, and modernisation to promote an international urbanism associated with a vision of the modern city as “hygienic” (Steck et al. 2013), sanitary, and “respectable”. In this context, informality can be misunderstood and misrecognised as a result of normative notions of “rational” economic behaviour and values (Neves and du Toit 2012).

These differences in the “webs of significance” in which actors are entangled lead to what can be called a problem of incommensurability. “Incommensurability” refers to something more profound than disagreement or contradiction—both of these assume a common language, a common frame of reference (Pellizoni 2003: 209). Instead, incommensurability refers to a condition where there is no shared language or conceptual framework: “What is lacking is a single description and connection of the facts, a shared vision of the meanings of concepts and principles” (Pellizoni 2003: 206).

In these contexts, common ground may be hard to find. As a Khayelitsha trader explained, expressing the sense of incomprehension several interviewees articulated, “We don’t understand what VPUU is saying” (Informal group interview, Ntlazane traders association, Khayelitsha, February 2014). In this milieu, a number of respondents argued that they found it hard to be heard in relation to the intervention: “They don’t sit down and talk to the people they found here to find solutions” (Focus group, informal traders, October 2014). And: “They just come to us only to tell us what they have decided to do” (Informal group interview, Ntlazane traders association, February 2014).

Local methods

The fieldwork in Khayelitsha was carried out by research team member Ncedo Mngqibisa over a period of ten months. He immersed himself in the communities living in the Harare and Kuyasa sections of Khayelitsha by conducting daily field visits that allowed him to produce a “thick description” of the “way of being” (Geertz 1973: 5) of these communities through semi-structured interviews, focus groups, informal conversations, and ongoing systematic observations, which were recorded in field notes.

All interactions probed questions about the way in which people do and do not cooperate in Khayelitsha, forms of social and other organisation, the degree of sociality between neighbours, experiences and norms around violence, and local responses to violence, including both formal interventions such as VPUU and informal activities such as community patrols and vigilante action.

Interviewees were identified through a “snowball sampling” methodology that gave the researcher deepening access to different components of the community. Snowball sampling is particularly useful for accessing “hidden” (Salganik and Heckathorn 2004: 193) or more “vulnerable” and “impenetrable” (Atkinson and Flint 2001: 1) social groups. The research began with a process of community profiling that involved identifying and interviewing key community leaders from local government, civil society, schools, and non-governmental organisations. While every effort was made to speak to a range of role players with different perspectives, a snowball sampling approach does introduce the possibility of bias as a result of the fact that the methodology depends on referral from one interviewee to another, individuals that are almost inevitably linked within social or other networks. Nevertheless, the researchers sought to engage with a wide range of stakeholders from different constituencies.

The research on VPUU was constrained by the fact that the lead research institute—the Human Sciences Research Council—was unable to secure formal cooperation with the VPUU intervention, although this was the initial intention of the research project and significant efforts were made to establish this cooperation. Therefore, this study relies on the perspectives of those who interacted with the intervention and the publicly available documentation we could obtain.

Through a process of engagement with the Khayelitsha community in Harare and Kuyasa, informal traders emerged as a group who had a significant level of engagement with, and a stake in, the VPUU intervention and hence were interviewed systematically, both individually and in a focus group. Another focus group was held with informal traders who are foreign nationals to gain their perspective on informal trading in the township. In addition, focus groups were held with beneficiaries of the VPUU social development programme who had received funding from the organisation for community-based projects, as well as with young entrepreneurs who had been using VPUU facilities such as the “Hub” business development space. Finally, focus groups were held with young men and young women, respectively, to draw out the gendered dimensions of violence in Khayelitsha.

Background of the VPUU programme

The VPUU programme was initiated in 2004 as a partnership between the City of Cape Town and the German government that contributed funds via the German Development Bank (Kreditanstalt für Wiederaufbau, KfW). This partnership was both financial and ideological in the sense that the German Development Bank provided finance to implement a model of “integrated violence prevention” through urban upgrading that it had already applied in two other developing world contexts characterised by high levels of violence, namely, El Salvador and Colombia. It now sought to execute a similar model in South Africa based on the perceived success of these interventions.

The German Development Bank’s model of violence prevention seeks to address violence by using “conventional urban planning tools” (Bauer 2010: 5) and “coherent and integrated town planning” (Bauer 2010: 6) to create an ordered and managed urban environment that bridges “the divide between the formal and the informal city” (Bauer 2010: 3) and “stabilise[s] the social environment” (Bauer 2010: 5). Speaking on behalf of VPUU, Chris Giles (2014: 2728) argued, “What the programme would like to show over time is the increase in what we would call managed urban areas.” Urban upgrading, from this perspective, creates the foundation for new forms of citizenship based on physical and symbolic ownership of space. “For inhabitants, having an official address means formally being [a] resident of a city” (Bauer 2010: 12). This ordered urban environment can ostensibly prevent violence.

VPUU was therefore established as an urban upgrading programme drawing on the principles of Crime Prevention through Environmental Design (CPTED), which posits that changes in the built environment can both deter offender behaviour and increase feelings of safety among residents. The intervention locates upgrading within a model of integrated violence prevention with three key pillars:

1.situational crime prevention (urban upgrading, construction of public facilities)

2.social crime prevention (funding for community-based organisations and educational support, e.g. early childhood development)

3.institutional crime prevention (promoting local economic development through skills training and “empowering” communities to run facilities built through urban upgrading).

In the context of Khayelitsha, the challenge was, however, not simply the unidirectional “transfer of knowledge” (problematic as this construction itself is), but a task of translation between two different knowledge systems—that of high modernity, which values rationality and autonomy in all its forms, and another knowledge system founded on what has been called in the South African context, Ubuntu, an ethos of community and reciprocity.

Ubuntu, a Nguni word, signifies an extremely complex concept that is not easily translated into English but nevertheless has a profound impact on African belief systems across the continent. It can be described as evoking a “communal” or communitarian ethic (Bohler-Muller 2005: 4) in which the very meaning of being human is located in one’s relation to others. The essence of Ubuntu is that an individual owes his or her existence to the existence of others: “ ‘I am’ because ‘you are’ and you are because ‘I am’ ”. Unlike the discourse of rights, which emphasises individual rights, Ubuntu tends to emphasise mutual obligation and duty to support the welfare of the community.

Community participation

The concept of community participation at both the planning and implementation stages is a central organising principle in the formal design of the VPUU intervention. Drawing on the methodology outlined by the KfW, which sees such community participation as critical to the success of violence prevention interventions, VPUU argues that it engaged in a participatory methodology that “strives for negotiated solutions in cooperation with communities” (VPUU 2013: 2). Community participation in, and ownership of, the intervention are seen as crucial mechanisms through which it contributes to building social cohesion.

However, Piper (2012) has called the form of community consultation that VPUU engages in, and the forums it creates, “designed” in ways that allow for a very limited form of direct citizen participation in democratic decision-making. While the VPUU model allows for community participation in identifying the key social issues to be addressed in an area, those who participate in this process are carefully selected through a prior process in which VPUU audits and chooses the organisations that will take part in its key decision-making structure, the Safe Node Area Committee (SNAC). The process of identifying the critical issues to be addressed, and the means to address them, is carefully supervised by VPUU professional staff to ensure that “unrealistic” expectations are managed.

Consultation regarding the intervention concerns the technicalities of how the intervention will be implemented, rather than an engagement about its substantive grounds. Broader community meetings are held, largely as forums for the endorsement of decisions already taken, rather than contexts where fundamental alterations to the intervention could be implemented. For example, a consultative community meeting is held once a community action plan has been formulated in response to the process of identifying local needs; however, the plan is launched and signed by the Mayor at the end of the “consultative” meeting, “suggesting that the process is largely a symbolic one” (Piper 2012: 7). This managed process of consultation is purported to be necessary to protect the new SNAC leadership from local, particularly political, elites who might attempt to capture the development process.

Therefore, while there are elements of representative democracy in the design of the VPUU model, it is not participatory in the sense of creating the space for the type of extended deliberation where even if “ ‘citizens or their representatives disagree morally’, they continue to reason together to reach mutually acceptable decisions” (Roux 2006: 17). Rather, the objective of the consultative processes that do take place appears to be largely instrumental and designed to ensure the efficient implementation of the intervention through the selection and socialisation of a cohort of “responsible” leaders who are tutored, through training, in the practices, norms, and ethics of the economic rational actor. In addition, they are deliberately shielded from the vagaries of public engagement, which could subject them to alternative claims from the local community and hence slow the execution of the intervention. Inevitably, the SNAC forum was not completely immune to political contestation and patronage claims. A former VPUU reference group member argued, “Due to politics we were removed from that [SNAC]. People wanted to bring in their people and so forth” (Personal interview, former VPUU reference group member, February 2014).

The potential limitations of this model of engagement, both on pragmatic grounds and in relation to building the kind of meaningful consensus that could foster social cohesion, were evidenced by a variety of sites of tension regarding VPUU’s implementation, where the substantive grounds and rationale for the programme were contested by specific community stakeholders.

It was in the intervention’s deeply contractarian imaginary of state and society that the most fundamental tensions emerged between it and the communities with which it engaged. VPUU envisaged a world in which “rational”, autonomous agents freely enter administrative or legal contracts, which specify the terms of the relationship between agents. This was disputed by a local imaginary of social relations founded on reciprocity or Ubuntu, which normatively structures social life in the township. Contemporary modernity in South Africa is in fact a product of the intersection of these imaginaries of social order; however, in its forceful articulation of a contractarian vision the VPUU intervention became a flashpoint (detailed below) for a display of the tensions between these worldviews. Most importantly, inseparably tied to these social imaginaries were contending visions of moral order. In terms of Ubuntu, ethical personhood, as opposed to mere existence, is realised through the collective and by means of actively carrying out duties and obligations to kin and community. On the other hand, in the contractarian imaginary, moral norms derive their normative force from the idea of contract or mutual agreement between individuals who act to rationally maximise their self-interest (Cudd and Eftekhari 2017).

Informal traders

A pivotal site where tensions emerged between the formality that VPUU sought to create and local practices of informality related to disputes around trading kiosks VPUU built as part of its upgrading initiative. While these kiosks provided important services that had not been available to traders previously, such as access to water, electricity and shelter from the weather, these objective conditions were subjectively construed by some traders as a symbolic capture of local space by “outsiders” that undermined pre-existing relationships of sociality and reciprocity. One trader asserted, “They treat us as Germans while we are in South Africa” (Personal interview, trader, February 2014).

Lefebvre (1991) has argued that space is socially produced and it is within physical space that social identities and social relations are embodied and enacted. VPUU seeks to establish new forms of ownership of space, in line with classic CPTED principles that theorise the link between urban space and crime in terms of the notion of “territoriality”, i.e. the concept that a sense of ownership of and responsibility for space can help reduce crime. CPTED is fundamentally about redesigning and re-signifying space to create a new spatial symbolism. What used to be “bad” and “dangerous” space is reconstituted as “safe”, “aesthetic”, and orderly.

However, it is exactly this ownership of space and the re-signification of its meaning that was challenged by traders who disputed whether VPUU had the right to displace, physically and symbolically, the informal trading spaces they had occupied on “our land” (Personal interview, trader, February 2014) and replace them with formalised space that traders now had to pay for. “Now we pay but we were occupying and using that area when there was just sand there. They moved us and build premises there and told us to pay” (Focus group, informal traders, October 2014).

The amount of rent charged by VPUU (R900 per month at the time of the research) was a deep source of discontent among traders in a context where the average monthly income in Khayelitsha is R2000 per month (Seekings 2013). Several traders appear to have struggled to meet their financial obligations and were indebted to VPUU. The intervention was therefore seen as reconstituting available land into inaccessible and expensive space. While traders’ real administrative “ownership” of the land on which they previously traded is questionable, the process of upgrading was experienced as an alienation from this land, which made them junior participants in an upgrading initiative rather than autonomous economic agents.

It was in VPUU’s attempts to establish formal contractual relationships with traders that some of the key disputes regarding the meaning of the space of Khayelitsha emerged. For traders, these contracts signified a surrender to unequal terms of engagement that required them to sign “on the dotted line” to secure access to newly constituted trading spaces from which they could be subsequently evicted and which imposed on them a new and foreign regulatory regime. “Keep in mind that they build in your spot with your business not registered” (Focus group, informal traders, October 2014). One trader explained, “You know how we work.… We don’t even have business accounts. We work to feed ourselves. They need to understand that this is not the same as town” (Personal interview, trader, February 2014).

At the centre of the VPUU imaginary was what has been called the “economic man” or homo economicus, the classic Western “entrepreneur”—a self-interested, utility-maximising individual whose major rationale is to generate profit by extracting surplus value. Through its contractual agreements with traders, VPUU actively sought to constitute this homo economicus during training that endeavoured to re-socialise traders into this normative paradigm as well as by enforcing regulatory “business” practices such as “auditing” of traders “books”. One trader commented, “What they do is send an auditor. Who are they sending the auditor to? Whose money are they auditing? They did not give you money” (Personal interview, trader, February 2014).

VPUU’s enforcement of the contractual agreements it entered into with the traders, including eviction if rent was not paid, was interpreted by traders in terms of the norms of reciprocity and informality that allow survivalist businesses to endure through community and kin networks, even in an environment where there is little immediate income or surplus value created. Therefore, VPUU’s warnings of eviction to traders who did not pay rent were perceived as a betrayal of these relationships of reciprocity, a vision of reciprocity which VPUU representatives reportedly caricatured as charity. “They say, ‘This is not charity, vacate if you cannot afford. We have a list of people that want to move in’ ” (Personal interview, trader, March 2014). “You will leave and go stay at the location and what will you eat?” (Informal group interview, Ntlazane traders association, February 2014). These testimonies echo sentiments we heard expressed by many traders. Myers (2011) argues in this vein that the integration of social networks and patterns of sociality into structured formal forms of urban development can be a poor substitute for previous forms of economic reciprocity and sociality. In this context, formality can constitute a threat to social networks and patterns of sociality, and can render the benefits that may be derived from it “doubtful or uncertain” (Neves and du Toit 2012: 142).

Community patrols

Another site of tension concerned the way VPUU tried to assist in fostering police–community relationships through support of Community Police Forums (CPFs) and the neighbourhood watches or community patrols that functioned under them. In a context where there is a “pluralisation of the governance of security” (Shearing and Wood 2003: 403) in the township, these community patrols were only one initiative within a wide variety of informal patrols that would arise sporadically in response to problems of crime and violence, often conducted by men and informed by a communitarian ethos which could include violence against residents allegedly involved in crime or simply breaking informally imposed “curfews”. This violence was performed as a public spectacle to police the boundaries of an often authoritarian conception of moral community. This was the normative framework in which the neighbourhood watches, established under the CPFs, also operated. CPFs are constitutionally mandated local forums operating at the level of the police station, intended to foster a cooperative relationship between the South African Police Service and the community in which they function. There is widespread acknowledgement, however, that these forums have not been functioning effectively in Khayelitsha. An investigation by the police’s inspectorate in 2012 found that “generally” the contribution that the CPFs are making to improve police–community relationships in the Khayelitsha area was “questionable” (SAPS 2014).

VPUU provided support to the CPFs through contractual relationships which sought to establish “standards” for neighbourhood watch conduct in legalistic and procedural terms, drawing on Western Cape government regulations for CPFs. Funding was contingent on CPFs meeting these terms, which included “accurate” documentation of neighbourhood watch membership and activities as well as “development contracts” with individual volunteers (Giles 2014). While VPUU wanted to regularise patrols in ways that would mitigate violence and make them sustainable, this procedural methodology exposed the tensions between the approach it attempted to introduce and those that framed the activities of existing patrols. As a former patroller argued, VPUU “did their work in English” (Focus group, Older men, October 2014), i.e. in a “foreign” manner.

The VPUU intervention into community police patrols was an important attempt to counter the violence committed by some of these groups and was lauded by a number of residents, one of whom argued, for example, that, “On the topic of patrolling, VPUU helped us a lot” (Focus group, Older women, October 2014). However, the programme did not take into account the informal patrols that were genuine attempts by citizens to take responsibility for their own safety in the context of the allegedly inadequate and corrupt policing in Khayelitsha. These informal patrols are also, as Gillespie (2014) points out, part of a network of informal community regulation that has existed in many townships for decades; they represent a legitimate form of self-governance that needs to be recognised when designing interventions so that the most positive aspects of these community initiatives can be harnessed and built upon.

Through its efforts to establish contractual relationships with CPFs in Khayelitsha, VPUU attempted to reform neighbourhood watches in terms of a Western conception of voluntarism as the product of individual altruism, rather than an assumed part of the relationality of daily life. A VPUU volunteer safety coordinator articulated it as: “in terms of the understanding within our communities” (Mtwana 2014: 159). However, a Western concept of voluntarism pertinent to voluntary work as a form of middle-class self-sacrifice was anachronistic in a context of extreme poverty, which was already characterised by reciprocity as both an ethos of living and a way of surviving poverty.

In accordance with its contractual agreements with the CPFs, VPUU therefore specified a new administrative process for the regularisation of patrollers. This involved a formal “registration process” and a “contractual obligation” in terms of which individuals acknowledged the “voluntary” nature of their role (i.e. they would not receive a salary) and stipulated exactly what activities they would undertake. In return, volunteers would be offered opportunities for “development” to be provided by VPUU (Mtwana 2014: 165). While this agreement was meant to clarify the terms of the relationship between patrollers and VPUU, these legal terms complicated the ethics of reciprocity that underpinned informal patrols in which no material reward was expected but, equally, no accountability could be enforced.

In this paradigm, residents (particularly men) would engage in patrols as a perceived need arose to defend the interests of an often ambiguously defined “community”. These activities were informed by a communitarian ethic in which identity is realised through actions that support community and kin. This was different from the legal and individualised contractual relationship that VPUU sought to institute where obligations and duties are defined by a contract, which autonomous agents freely choose to enter. While existing patrols were sometimes carried out under the auspices of locally constituted organisations, this hierarchical relationship was often weak and, in many instances, patrols were largely spontaneous responses to contexts of severe violence. In this environment, then, there was very little of the sort of accountability that VPUU wanted to introduce through its contractual relationship with patrollers which, in some cases, led to a proliferation of violence by the patrollers.

By introducing a different approach into the ongoing practices of patrolling, VPUU disrupted existing ethics of reciprocity in which individuals had patrolled with no expectation of material reward or individual accountability. Patrollers who entered contractual relationships with VPUU assumed this was a commercial relationship and they would be rewarded in monetary terms, either immediately or in the future. The “currency” that VPUU offered, i.e. “development” opportunities that could potentially better equip patrollers to enter the labour market, was viewed with scepticism by some in the light of pervasive, structural unemployment. The coffee that VPUU offered volunteers became a metaphor for the scarcity of the reward they received: “Other people are demotivated because they patrol for years and don’t get anything. They just get coffee with milk only. You patrol with bare hands” (Personal interview, March 2014).

The interaction between VPUU’s aspiration to regularise CPF patrolling in terms of a normative understanding of their role and function, and the unintended impacts this produced, illustrates the complexity of introducing Western models of community policing to environments such as Khayelitsha, without significantly adapting to or engaging with local forms of regulation and social policing. A number of studies of policing in South Africa have noted the difficulties and unpredictability of transferring international “models” of crime prevention, such as community policing, to contexts outside of the environments in which they were initially formulated (Brogden and Shearing 1993; Van Der Spuy 2000; Brogden 2002, 2004; Pelser and Louw 2002; Steinberg 2011). Brogden (2002: 92) argues that the dissemination of the concept of community policing in South Africa was part of a wider process of “commodification” of community policing as a result of processes of globalisation. However, as Steinberg (2011: 350) notes, “the results of policy transfer, especially to societies in the process of a major transition, are often unknowable”.

The UPPs in Rio de Janeiro, Brazil

In 2015, fieldwork was carried out in two favelas9 in Rio de Janeiro: Cidade de Deus,10 where the UPP was established in February 2009, and Tabajaras/Cabritos,11 where the intervention was launched in January 2010. Forty semi-structured interviews were conducted with residents and leaders of the two localities, 20 interviews in each. The focus of this work was to understand sociability, social cohesion, and the relationship of UPP police officers with residents.

Local methods

The concept of social cohesion that the Laboratory for the Analysis of Violence at Rio de Janeiro State University worked with was analysed at the community level and operationalised according to six dimensions:

identification with the area and sense of belonging

mutual trust among neighbours

mutual cooperation among neighbours

intensity and quality of social interactions among neighbours

micro-regulation of social deviance, also conceived as social efficacy, i.e. the belief that one can and should intervene when individuals break basic social norms in order to restrain that behaviour

participation in associations and institutions at the local level.

Ethnographic work took place over seven months in the community of Cidade de Deus and over five months in Tabajaras. Interviews were recorded and transcribed, though seven interviewees did not agree to being recorded. In these cases, the analysis relied on notes taken by the researchers. Apart from the formal interviews, the researchers participated in many informal conversations and conducted systematic observation with weekly visits to the communities. Unlike in South Africa, focus groups were not carried out as the level of conflict and social tension in these favelas meant that residents felt that their safety could be compromised by expressing their views openly in a focus group. All the material was recorded in field notebooks, which was later coded as:

statements by members of the community

verbatim statements

behavioural observations

interpretations of the researchers.

Recorded interviews were also coded according to a list of categories that included topics outlined in the interview questionnaire and emergent themes that appeared during the research.

Background of UPP

The absence of the state in favelas is one of the ways that public security managers in Rio de Janeiro explain the control of these territories by criminal groups. Based on this assumption, a series of public policies and programmes aimed at reducing violence and controlling drug trafficking have been implemented over the last 30 years.

In 2008, the government of the State of Rio de Janeiro initiated the UPP programme. Its objective was to regain state control of the favelas and reduce armed confrontations, using an approach focused on long-term prevention. One of the main pillars of this programme was the application of a concept of “Proximity Police” in order to change the model of police action away from sporadic, reactive, temporary interventions and lethal operations. The philosophy and strategy of Proximity Policing focuses on prevention, with the citizen becoming the centre of police attention. Success in arresting offenders is seen to be the result of a close relationship between the police and society, which makes it possible to create a highly developed network of solidarity that provides protection for citizens.

Inaugurated in December 2008, the UPP was created as an alternative to the community policing approach previously tried in the favelas of Rio de Janeiro. The “pacification” process applied by the UPPs follows four predefined steps:

1.The Batalhão de Operações Policiais Especiais (BOPE), the Special Police Operations Battalion of the military police of the State of Rio de Janeiro, occupies the territory in a favela through armed intervention in the locality. This is the phase with the greatest potential for violent conflict as the BOPE essentially invades a territory controlled by armed groups. However, the state government makes public announcements about these BOPE operations in advance, reducing the potential for conflict to some degree. In general, members of criminal groups choose to leave the favela before the BOPE operation begins.

2.Once the initial territorial acquisition has been carried out, the BOPE identifies and removes sources of resistance and those identified as “criminals”. Once the situation is more stable, the third phase of occupation is initiated.

3.Police officers from the UPP enter the favela and begin operations to establish “law and order”.

4.The UPP work to establish a positive relationship with residents based on trust and peacekeeping.

Although the time and labour required for each stage varies, depending on the specific situation, the overall process remains the same.

The policing programme is set up in terms of a decree12 that provides a very basic outline for its implementation. The decree defines the criteria for selection of the locations where UPP will be implemented, outlines project objectives, and the organisational aspects of the programme. Subsequently, there is a process of formalisation and systematisation to develop a model of policing for the favela based on its specific conditions.

An intervention like the UPP seeks to replace the previous tactic of police invasions in the favelas, which had resulted in armed confrontations, with the ongoing presence of a fixed group of police officers trained in a more community-based approach. According to the international literature on community policing in these types of programmes, the police cultivate relationships with the population of mostly poor urban communities (Kelling and Coles 1996; Piquero et al. 1998; Moraes and Cano 2007). The success of police efforts to reduce crime often depends on community involvement in the programme.

The UPP intervention has been criticised as having sought to effect change through significant militarisation of the daily life of favela inhabitants as a new instrument of coercion of the working classes and as a strategy “to make noble”, i.e. upgrade areas that are of interest for tourism and major sporting events (Batista 2011; Fleury 2012; Freeman 2012). The distribution of UPPs across the city would seem to support this assertion as it was not implemented in the most violent western region of Rio de Janeiro but instead focused on the southern, central, and northern areas of the city. What has emerged are real gaps between coercive and punitive practices and the stated objectives of the programme: guaranteeing rights of citizenship and meeting the social needs of the population. It also does not appear to have met the objective of integrating favelas into the mainstream dynamics of the city through the expansion of public services (garbage collection, basic sanitation, transportation, etc.) and private services (regularisation of the connection of electricity and cable TV, gas distribution, etc.) (FGV 2012). In some cases, when these services have been expanded this has led to a process of gentrification that has forced favela residents out of their homes.

What has changed with the UPP?

Despite criticisms, it is well recognised that the UPP has reduced violence in the favelas where it has been implemented. The most widely acknowledged change was the reduction of armed confrontations between groups of traffickers and police officers (see, for example, Cano, Borges, and Ribeiro 2014). A report by the Stanford Center for International Development found that, in areas where the UPP intervention was implemented, police killings would have been 60 per cent higher without the intervention (Magaloni, Franco, and Melo 2015). Various researchers have found the results listed below regarding the UPP (IBPS 2009; CECIP 2010; de Souza e Silva 2010; Burgos et al. 2011; Oliveira and Abramovay 2012; Rodrigues and Siqueira 2012; Serrano-Berthet 2013; Cano, Borges, and Ribeiro 2014):

a suspension or significant reduction in violent shootings and police incursions

the virtual elimination of the overt carrying of weapons by actors other than the police

greater freedom of inhabitants to come and go

reduced stigma of living in some favelas

greater internal and external social control over corruption and abuse of power by police.

The terms used to describe this change by residents are “war” and “peace”, respectively, which indicates a major improvement in their sense of personal security from the insecurity and fear residents reported in the period before the UPP’s entry into their communities.

On the other hand, public security experts have identified several problems in the programme that may affect its sustainability and legitimacy. Among them are the absence of social investments; difficulties in the relationship between police officers and residents of the favelas; the perception that there is still authoritarian control over communities; the low institutionalisation of so-called proximity policing; and the channels of dialogue between the population and police, with the consequent tendency to transform police commanders into the “favela’s manager” or “new owners of the hill” (Cano, Borges, and Ribeiro 2014).

The impact of the UPP on social cohesion

The analysis of the impact of the UPP on social cohesion in the two favelas of Rio de Janeiro where the research was conducted was based on the perception of residents and people working in both territories regarding the effect of this policing model on the dynamics of community relations.

The analysis revealed the specificity of patterns of social cohesion in both sites, related to their particular histories and current forms of social organisation. The research explored social cohesion at the community level in terms of:

feelings of belonging to the community and identification with a territory

the degree of mutual trust and cooperation between neighbours

the intensity and quality of interaction between neighbours

micro-regulation of social deviance

participation in associations and institutions at the local level.

In terms of mutual trust and cooperation between neighbours, many residents of the two favelas mentioned that they are willing to help their neighbours, but they do not usually ask favours of them. That is, they are open to providing some kind of help and even do it with some frequency, but they emphasise that there is no reciprocity in this relationship. Nevertheless, bonds of cooperation seem to be strong in times of personal crisis and collective tragedies caused by natural disasters, diseases, fires, etc.

The research revealed that interaction among residents is strongly associated with age. Generally, older people stay at home and attend few social gatherings, while their younger counterparts participate in community events more frequently. However, regardless of age, residents of both favelas were open to receiving visits from inside and outside the community.

In Tabajaras/Cabritos, the intervention had an ambiguous impact on social cohesion. The implementation of the UPPs led to a reduction of violence and increased perceptions of safety. This encouraged the entry of new residents to the favela, which is located in a privileged position in the city within an important tourist region, near major commercial centres, and with good transport connections and services. Since the implementation of the UPPs, properties in the area have increased enormously in value and, therefore, many people from different neighbourhoods, other municipalities, and even from other countries have been attracted to live in the slums of the southern zone of Rio de Janeiro. This change of profile of the favela led to an increase in rental prices and the legalisation and regularisation of some public services such as electricity and water, which residents had previously accessed illegally but without cost. In light of this, some residents, usually the poorest, had to leave these favelas because they could not afford the increased cost of living in the area. In this context, the ethnography revealed that it has become commonplace for original residents to comment about “new people” on the streets that they do not recognise. This generates a sense of estrangement in an environment where previously residents had been closely tied together.

In this context, residents who had recently moved to the favela did not know the local rules and, according to some interviewees, generated “disorder” and disrespect with regard to community norms. The greater turnover of residents, the changing profile of people living in the favela, and the increase in the price of services have transformed interaction, cooperation, and social control in the area. Some informants reported that they did not feel comfortable asking their neighbours for help because they often did not know them. If there was a problem, the UPP would be the first to be contacted rather than asking for the help of a neighbour or any other known organisation in the favela. Thus, the UPP seemed to loosen community bonds because it caused a greater transit of outsiders who knew nothing of the local reality, the dynamics and routine of the favela.

In Cidade de Deus, the UPP reportedly led to a decline in local organisations. Before the UPP was implemented, local community groups organised many activities such as karate, dancing, or other cultural activities. However, according to some leaders in the area, after the UPP was implemented, many parents withdrew their children from social projects as they no longer saw the need to prevent them from playing in the streets; the favela was perceived to have become “less violent” because of the UPP, reducing the need to enrol children in social projects. Thus, one of the unexpected effects of UPP was the loss of investments in local organisations that had enabled social projects in the favela. The lower number of participants in attendance meant that some organisations had to limit their activities. However, this favela still needs a substantial amount of social investment. Its residents, especially children and adolescents, remain in a situation of social vulnerability and would likely benefit from continuing to participate in social programmes.

With regard to micro-regulation, when respondents were asked who or what institution would be called in situations of conflict between neighbours or even a crime, Cidade de Deus residents generally mentioned the “traffic” (drug traffickers). Under “favela laws” that still structure life in Cidade de Deus, the police should not be called since the territory remains under the control of drug traffickers, even after the implementation of the UPP. In this context, it would be the traffickers who would solve the favela’s problems, be it a conflict between neighbours or a crime such as robbery or rape. The police could only be called in limited situations, such as the removal of the body of a homicide victim. With two systems of authority now in place, there is tension in the relations of the inhabitants of Cidade de Deus. Residents have to show respect for the UPP at the same time as obeying the rules of the traffickers, without allowing the police to know about this. According to interviewees, life was relatively “easy” before the UPP because the favela was unambiguously “ruled” and controlled by the traffickers alone.


VPUU and UPP were quite different interventions in many ways, but both sought to reconstitute social relations, establish new norms, and stabilise the social fabric within the communities in which they were located as part of long-term efforts to counter violence. While they adopted different methodologies, both were informed by a desire to “recapture” the space of the favela or township from the informal sources of authority and regulation that had governed it and which were seen as contributing to violence.

In Rio de Janeiro, these “formalisation” processes took place through attempts to forcibly evict drug traffickers and replace them with state police in order to create normative modes of social regulation. In contrast, VPUU tried to recapture areas of the township experiencing high levels of violence through urban upgrading initiatives that were seen as a way of facilitating local ownership and investment in this space. The strong emphasis on formalisation by VPUU, drawn from international models of urban upgrading, created tensions with some local residents, such as informal traders and community patrollers, regarding aspects of its implementation. While the UPP was an entirely Brazilian driven initiative, it too was influenced by international crime prevention models and also struggled to build sustainable relationships with citizens in the favelas in which it was implemented. The UPP intervention was able to demonstrate more dramatic reductions in violence than VPUU, particularly in its initial period of implementation. However, it is unclear how sustainable these reductions are and whether either of the interventions have had an impact on cohesion and collective efficacy in ways that will help reduce violence in the long term.

While theories of social disorganisation posit a “breakdown” in social relations and norms as a result of urbanisation, which they argue contributes to violence, the research showed that there was not an absence of organisation or collective norms in the favelas of Rio de Janeiro or the neighbourhoods of Khayelitsha, but a complex network of informal authority, regulation, and organisation. Neither intervention appears to have been able to harness these local resources, which were instead seen as an impediment to social control. While slum upgrading interventions may seek to “gentrify” areas in order to reduce violence, the research in Tabajaras/Cabritos showed the potentially negative impact of these processes on the poor, who may be driven out of upgraded areas, thus exacerbating and displacing inequality while existing social bonds in a neighbourhood are undermined.

In Khayelitsha, urban upgrading did not lead to significant reductions in violence that would fundamentally change the character of the area or the profile of its residents. What the research revealed, however, was that despite the fragmentation and division in the community, there was a strong ethos of reciprocity, expressed in the adage, “I am human because you are”, which led residents to help and even risk their lives for each other. Sometimes, collective action in support of neighbours could lead to violent retribution. In this context, close social ties could have a negative impact on violence. VPUU struggled to mobilise the positive aspects of this ethic of reciprocity and build cohesion as it limited the participatory aspects of the programme and pursued an approach that privileged Western norms and social relations. In Cidade de Deus, the UPP reduced violence, but complicated existing systems of social authority, thereby creating a dual authority structure that included both the police and drug traffickers. At the same time, local social organisations suffered as a result of the diversion of funds away from them to the state.

Therefore, despite the fact that these interventions were in fact premised on the idea that “disorganisation” of the social fabric contributes to violence, paradoxically, the ways in which they intervened often exacerbated this problem and may well influence their long-term impact.

Findings from this research project suggest the following changes would lead to improvements in policy and programming to prevent violence:

1.Investigate how social cohesion works in developing countries so that assumptions based on how it operates in the Northern context do not negatively influence developing world programming.

2.Unpack social cohesion as a concept when developing policy. Break the idea down to identify violent and non-violent forms of cohesion so that specific elements can be targeted when designing interventions.

3.Take informal mechanisms of social regulation into account when designing interventions. Informal mechanisms, both positive and negative, such as volunteer security guards and vigilante groups, are critical for understanding how social cohesion and collective efficacy can prevent violence or may, in some contexts, exacerbate it. As Shaw and McKay (1942) argued, this type of knowledge helps promote successful interventions that are relevant to the local context and encourage “indigenous social invention” (Schlossman and Sedlak 1983).

4.Involve communities democratically in violence prevention initiatives. Recognise and seek to strengthen, rather than bypass, local institutions and local traditions wherever possible. This includes integrating communities in the design and conceptualisation of interventions as partners so that programmes resonate with local norms and systems of regulation. International models of violence prevention should be drawn on as a resource rather than uncritically implemented. These approaches may need to be revisited, revised, or reformulated in the light of local conditions. Interventions may be useful for increasing security and reducing violence, but without the democratic participation of the people affected they risk being unsustainable and potentially causing harm.

5.Learn more about local operating environments before interventions are implemented. Conducting contextualised qualitative and quantitative analyses in every community before interventions—such as VPUU in South Africa or Brazil’s UPP—are rolled out would contribute to these programmes’ success. These investigations should include analysis of demographic and social conditions in addition to informal networks, mechanisms of social regulation, and formal and informal power structures. This will help to ensure interventions are appropriately attuned to the environment in which they are to be implemented.


1Since the late 1970s, ethnography has increasingly been recognised as having an important role in evaluating social programmes in fields such as education and health (Guba 1978; Guba and Lincoln 1981; Dorr-Bremme 1985; Fetterman and Pitman 1986; Patton 1990; Schensul and Schensul 1990; Nastasi and Berg 1999; Hopson 2002; Morgan-Trimmer and Wood 2016). For example, Schensul and Schensul (1990: 51) note that, “Early in the AIDS epidemic, evaluators and intervention researchers recognized that experimental design and other traditional approaches to research and evaluation could not provide sufficient information to program developers and policy makers to guide interventions in communities.” In this context, ethnography provided a crucial means to collect data on issues such as “life-styles, social networks, needle-sharing activities, sexual patterns, and decision-making processes of drug users” (Schensul and Schensul 1990: 51). This study has drawn on some of the insights that have emerged from ethnographic approaches to evaluation.

2Lincoln and Guba (1986) developed four key criteria that are now widely used to establish the trustworthiness of qualitative research: credibility, transferability, dependability, and confirmability. There are recognised techniques for ascertaining the extent to which each of these criteria has been applied in a research study. For example, credibility concerns the researcher’s confidence in the “truth” or validity of their findings. Techniques identified by Lincoln and Guba (1986) for establishing credibility that were used in this study include “prolonged engagement”, “peer de-briefing”, “member-checks”, and “triangulation” (Lincoln and Guba 1986: 77).

3Prolonged engagement concerns the researcher or researchers immersing themselves in the field long enough to understand the social context of a particular phenomenon, including speaking with a range of relevant actors. In this study, extended ethnographic fieldwork of six to ten months at each research site included engagement with a wide range of community and government actors at the field sites and at the city level in both countries through interviews, focus groups, meetings, and informal conversations. Following the fieldwork, prolonged engagement with the ethnographic material continued as researchers immersed themselves in the data, including reading and re-reading them during a systematic thematic coding process.

4Thematic analysis has been described as a “method for identifying, analyzing, organizing, describing, and reporting themes found within a data set” (Braun and Clarke 2006: 79). A theme captures and unifies fragments of information and ideas into a coherent concept that “tells a story” about the data, for example, what they say about “gender”, “violence”, or “community”.

5“Peer de-briefing” was used during the coding process to come to a consensus about the meaning of the data; it is an important technique for establishing the credibility of research. Peer de-briefing refers to a process of systematic engagement between researchers where each researcher exposes their ideas and analysis to a peer or colleague in order to bring to light “aspects of the inquiry” that may otherwise have remained implicit (Lincoln and Guba 1986: 308). In this study, the data were analysed through a process of systematic engagement between field researchers at the sites of data collection and broader research teams in both South Africa and Brazil, as well as through a cross-country review of data.

6The research team endeavoured to implement a consultative approach to the research process through “member checks” and “analyst triangulation” by testing interpretations and conclusions from the research with members of the groups from which the data were obtained. Lincoln and Guba (1986) have emphasised the importance of this technique for establishing credibility. Analyst triangulation occurs when different perspectives on the findings of the research are elicited from a panel of engaged stakeholders. In this study, consultative workshops and reference group meetings with stakeholders from community, government, and professional practitioners were held at critical points in the research in both countries to facilitate “member checking” and “analyst triangulation”. This included engagement around the initial research plan, during the course of the research project to present preliminary findings, and in report back meetings at the conclusion of the research.

7Transferability is another criterion for establishing the trustworthiness of qualitative research and relates to the extent to which the research is applicable in contexts other than where it was conducted. This can sometimes be a complex requirement in qualitative research, which generally focuses on in-depth interpretative analysis of a particular case that is not easily generalisable across different contexts. Here, “The researcher cannot know the sites that may wish to transfer the findings” (Nowell et al. 2017: 3).

Nevertheless, the transferability of qualitative research can be facilitated by what anthropologist Geertz (1973: 5) has called “thick description”. This refers to a detailed account of field experiences in which the researcher makes explicit patterns of cultural and social relationships and puts them in context (Holloway 1997). These “thick descriptions” make it possible for researchers reading the analysis to make an informed assessment as to whether the findings and analysis, developed in one context, could be applicable in another context. In this study, “thick description” of the data is provided in the case studies of the field contexts in South Africa and Brazil.

8Dependability is a further criterion for ascertaining the trustworthiness of qualitative research and relates to the replicability of the research process. It asks whether another researcher would be able to replicate the study based on the information provided by a research study. Dependability is primarily established through making the research processes transparent in a “dense description” of research methods (Krefting 1991). This makes the research “auditable” in the sense that researchers reading the analysis are able to understand how research conclusions were arrived at. In this study, the extended methodology section in this chapter is an attempt to ensure that the research methods and process of this study are explicit and transparent.

9A favela is a low-income historically informal urban area in Brazil.

10Cidade de Deus, popularly known as CDD, is a neighbourhood in the West Zone of the city of Rio de Janeiro, Brazil. It has long been considered one of the most dangerous regions of the city.

11Tabajaras/Cabritos is a favela between the neighbourhoods of Copacabana, a popular tourist destination, and Botafogo, one of the wealthy neighbourhoods of the South Zone of Rio de Janeiro.

12Decree-law 42,787 of 6 January 2011.


Atkinson, Rowland, and John Flint (2001). Accessing hidden and hard-to-reach populations: Snowball research strategies. University of Surrey Social Research Update 33: 1–4.

Batista, Vera Malaguti (2011). O Alemão é muito mais complexo [German is much more complex]. Paper presented at the 17th International Seminar on Criminal Sciences, São Paulo, Brazil, August 2011.

Bauer, Bettina (2010). Violence Prevention through Urban Upgrading: Experiences from financial cooperation. Frankfurt: KFW Entwicklungsbank. www.kfw-entwicklungsbank.de/Download-Center/PDF-Dokumente-Sektoren-Berichte/2010_03_Violence-Prevention_E.pdf.

Bohler-Muller, Narnia (2005). The story of an African value. Focus: Ten years after Makwanyane. SA Public Law 20(2): 266–280.

Braun, Virginia, and Victoria Clarke (2006). Using thematic analysis in psychology. Qualitative Research in Psychology 3(2): 77–101.

Brogden, Mike (2002). Implanting community policing in South Africa: A failure of history, of context, and of theory. Liverpool Law Review 24(3): 157–179.

Brogden, Mike (2004). Commentary: Community policing: A panacea from the West. African Affairs 103(413): 635–649.

Brogden, Mike, and Clifford Shearing (1993). Policing for a new South Africa. London: Routledge.

Bryant, Anthony, and Kathy Charmaz (2007). The Sage handbook of grounded theory. London: SAGE Publications.

Burgos, Marcelo Baumann, Luiz Fernando Almeida Pereira, Mariana Cavalcanti, Mario Brum, and Mauro Amoroso (2011). O efeito UPP na percepção dos moradores das favelas [The effect of UPP on the perception of favela residents]. Desigualdade e Diversidade – Revista de Ciências Sociais da PUC-Rio 11: 49–98.

Cano, Ignacio, Doriam Borges, and Eduardo Ribeiro (2014). Os Donos do Morro. Uma Avaliação Exploratória do Impacto das Unidades de Polícia Pacificadora (UPPs) no Rio de Janeiro [The Hill Owners. An Exploratory Evaluation of the Impact of the Pacifying Police Units (UPPs) in Rio de Janeiro]. São Paulo and Rio de Janeiro: Fórum Brasileiro de Segurança Pública/LAV/UERJ.

CECIP: Centro de Criação de Imagem Popular (2010). O impacto sobre a primeira infância das políticas de segurança pública e iniciativas comunitárias em comunidades urbanas de baixa renda [The impact on early childhood of public safety policies and community initiatives in low-income urban communities]. Rio de Janeiro: Author.

Cudd, Ann, and Seena Eftekhari (2017). Contractarianism. In Edward N. Zalta (Ed.), The Stanford Encyclopedia of Philosophy [Online, Spring 2017 Edition]. Stanford, CA: Metaphysics Research Lab, Stanford University. https://plato.stanford.edu/archives/spr2017/entries/contractarianism/.

de Souza e Silva, Jailson (2010). As Unidades de Polícia Pacificadora e os novos desafios para as favelas cariocas [The Pacifying Police Units and the new challenges for the Rio de Janeiro favelas]. Seminário Aspectos Humanos da Favela Carioca. Rio de Janeiro: Laboratório de Etnografia Metropolitana (Le Metro)/IFCS/UFRJ.

Dorr-Bremme, Donald W. (1985). Ethnographic evaluation: A theory and method. Educational Evaluation and Policy Analysis 7(1): 65–83.

Fetterman, David M. (2010). Ethnography: Step by step, Third Edition (Applied Social Research Methods Series). Thousand Oaks, CA: SAGE Publications.

Fetterman, David M., and Mary Anne Pitman (Eds.) (1986). Educational evaluation: Ethnography in theory, practice, and politics. Newbury Park, CA: SAGE Publications.

FGV: Fundação Getulio Vargas Projetos (2012). Indicadores socioeconômicos nas UPPs do estado do Rio de Janeiro [Socioeconomic indicators in the UPPs of the state of Rio de Janeiro]. Rio de Janeiro: Author.

Fleury, Sonia (2012). Militarização do social como estratégia de integração: o caso da UPP do Santa Marta [Social militarisation as an integration strategy: The case of the UPP of Santa Marta]. Sociologias, Porto Alegre 14(30): 194–222.

Freeman, James (2012). Neoliberal accumulation strategies and the visible hand of police pacification in Rio de Janeiro. Revista de Estudos Universitários 38(1): 95–126.

Geertz, Clifford (1973). The interpretation of cultures: Selected essays. New York: Basic Books.

Giles, Chris (2014). Commission of Inquiry into Allegations of Police Inefficiency in Khayelitsha and a Breakdown in Relations between the Community and Police in Khayelitsha. 12 February 2014. Commission Transcript, 2723–2764.

Gillespie, Kelly (2014). The context and meaning of “mob justice” in Khayelitsha. Report for the Khayelitsha Commission: Commission of Inquiry into Allegations of Police Inefficiency in Khayelitsha and a Breakdown in Relations between the Community and Police in Khayelitsha.

Guba, Egon. G. (1978). Toward a methodology of naturalistic inquiry in educational evaluation (CSE Monograph Series in Evaluation No. 8). Los Angeles, CA: Center for the Study of Evaluation, University of California.

Guba, Egon. G., and Yvonna S. Lincoln (1981). Effective evaluation: Improving the usefulness of evaluation results through responsive and naturalistic approaches. San Francisco, CA: Jossey-Bass.

Holloway, Immy (1997). Basic concepts for qualitative research. London: Wiley-Blackwell.

Hopson, Rodney K. (2002). Making (more) room at the evaluation table for ethnography: Contributions to the responsive constructivist generation. In Katherine E. Ryan and Thomas A. Schwandt (Eds.), Exploring evaluator role and identity: A volume in evaluation and society. Greenwich, CT: Information Age Publishing.

IBPS: Instituto Brasileiro de Pesquisa Social (2009). O impacto das Unidades de Polícia Pacificadora nas favelas da cidade do Rio de Janeiro [The impact of the Pacifying Police Units in the favelas of the city of Rio de Janeiro]. Rio de Janeiro: Author.

Kelling, George L., and Catherine M. Coles (1996). Fixing broken windows: Restoring order and reducing crime in American cities. New York: The Free Press.

Krefting, Laura (1991). Rigor in qualitative research: The assessment of trustworthiness. American Journal of Occupational Therapy 45: 214–222.

Lay, Margaret, and Irena Papadopoulos (2007). An exploration of fourth generation evaluation in practice. Evaluation 13(4): 486–495.

Lefebvre, Henri (1991). The production of space. Translated by Donald Nicholson-Smith. Oxford: Blackwell Publishing.

Lincoln, Yvonna S., and Egon G. Guba (1986). But is it rigorous? Trustworthiness and authenticity in naturalistic evaluation. The Evaluation Marketplace: Exploring the Evaluation Industry 30: 73–84.

Magaloni, Beatriz, Edgar Franco, and Vanessa Melo (2015). Killing in the slums: An impact evaluation of police reform in Rio de Janeiro. Working Paper No. 556. Stanford, CA: Stanford Center for International Development.

Moraes Dias da Silva, Graziella, and Ignacio Cano (2007). Between damage reduction and community policing: The case of Pavāo-Pavāozinho-Cantagalo. In Tom R. Tyler (Ed.), Legitimacy and criminal justice: An international perspective (186–214). New York: Russel Sage Foundation.

Morgan-Trimmer, Sarah, and Fiona Wood (2016). Ethnographic methods for process evaluations of complex health behaviour interventions. Trials 17: 232–242.

Mtwana, Nontutuzelo T. (2014). Commission of Inquiry into Allegations of Police Inefficiency in Khayelitsha and a Breakdown in Relations between the Community and Police in Khayelitsha, 12 February 2014. Commission Transcript.

Muggah, Robert (2012). Researching the urban dilemma: urbanization, poverty and violence. Ottawa: International Development Research Centre (IDRC). www.idrc.ca/sites/default/files/sp/Images/Researching-the-Urban-Dilemma-Baseline-study.pdf.

Myers, Garth (2011). African cities: Alternative visions of urban theory and practice. London: Zed Books.

Nastasi, B. K., and Marlene J. Berg (1999). Using ethnography to strengthen and evaluate intervention programs. In Jean J. Schensul, Margaret D. LeCompte, G. Alfred Hess, Bonnie K. Nastasi, Marlene J. Berg, Lynne Williamson, Jeremy Brecher, and Ruth Glasser (Eds.), Ethnographers toolkit. Walnut Creek, CA: AltaMira Press.

Neves, David, and Andre du Toit (2012). Money and sociality in South Africa’s informal economy. Africa 82(1): 131–149. https://core.ac.uk/download/pdf/62634217.pdf.

Nowell, Lorelli S., Jill M. Norris, Deborah E. White, and Nancy J. Moules (2017). Thematic analysis: Striving to meet the trustworthiness criteria. International Journal of Qualitative Methods 16: 1–13.

Nuttall, Sarah, and Achille Mbembe (Eds.) (2008). Johannesburg: The elusive metropolis. Durham, NC: Duke University Press.

Oliveira, Fabiana Luci de, and Pedro Vieira Abramovay (2012). As UPPs e o longo caminho para a cidadania nas favelas do Rio de Janeiro [The UPPs and the long road to citizenship in the favelas of Rio de Janeiro]. In Fabiana Luci de Oliveira, Maria Tereza Aina Sadek, Pedro Vieira Abramovay, and Izabel Nuñez (Eds.), UPPs, direitos e justiça: Um estudo de caso das favelas do Vidigal e do Cantagalo [UPPs, rights and justice: A case study of the Vidigal and Cantagalo favelas] (123–147). Rio de Janeiro: Fundação Getulio Vargas.

O’Reilly, Karen (2012). Ethnographic methods, Second Edition. Abingdon, UK: Routledge.

Parnell, Susan, and Jennifer Robinson (2012). (Re)theorizing cities from the Global South: looking beyond neoliberalism. Urban Geography 33(4): 593–617.

Patton, Michael Q. (1990). Qualitative evaluation and research methods, Second Edition. Newbury Park, CA: SAGE Publications.

Pellizoni, Luigi (2003). Uncertainty and participatory democracy. Environmental Values 12(2): 195–224.

Pelser, Eric, and Antoinette Louw (2002). Where did we go wrong? A critical assessment of crime prevention. SA Crime Quarterly 2: 1–4.

Pieterse, Edgar (2011). Introduction: Rogue urbanisms. Social Dynamics: A journal of African studies 37(1): 1–4.

Pieterse, Edgar, and AbdouMaliq Simone (Eds.) (2013). Rogue urbanism: Emergent African cities. Johannesburg: Jacana Media.

Piper, Laurence (2012). Development trustees not rent-seeking deployees: The designed meaning of community participation in the Violence Prevention through Urban Upgrading Project (VPUU) in Cape Town, South Africa. Unpublished paper.

Piquero, Alex, Jack Greene, James Fyfe, Robert Kane, and Peter Collins (1998). Implementing community policing in public housing developments in Philadelphia: Some early results. In Geoffrey Alpert and Alex Piquero (Eds.), Community policing: Contemporary reading. Prospect Heights, IL: Waveland Press.

Rao, Vyjayanthi (2006). Slum as theory: The South/Asian city and globalization. International Journal of Urban and Regional Research 30(1): 225–232.

Rodrigues, André, and Raíza Siqueira (2012). As Unidades de Polícia Pacificadora e a segurança pública no Rio de Janeiro [The Pacifying Police Units and public security in Rio de Janeiro]. In André Rodrigues, Raíza Siqueira, and Mauricio Lissovsky (Eds.), Unidades de Polícia Pacificadora: Debates e reflexões [Pacifying Police Units: Debates and reflections]. Comunicações do Iser (Instituto de Estudos da Religião, Rio de Janeiro) 31(67): 9–52.

Roux, Theunis (2006). Democracy. In Stuart Woolman, Michael Bishop, and Jason Brickhill (Eds.), Constitutional Law of South Africa, Second Edition. Cape Town: Juta.

Roy, Ananya (2009). The 21st-century metropolis: New geographies of theory. Regional Studies 43(6): 819–830.

Roy, Ananya (2011). Slumdog cities: Rethinking subaltern urbanism. International Journal of Urban and Regional Research 35(2): 223–238.

Salganik, Matthew J., and Douglas Heckathorn (2004). Sampling and estimation in hidden populations using respondent-driven sampling. Sociological Methodology 34(1): 193–239.

Sampson, Robert J., Stephen W. Raudenbush, and Felton Earls (1997). Neighborhoods and violent crime: A multilevel study of collective efficacy. Science 277(5328): 918–924.

SAPS: South African Police Service (2014). Opening Statement. Khayelitsha Commission of Inquiry into Allegations of Police Inefficiency in Khayelitsha and a Breakdown in Relations between the Community and the Police in Khayelitsha. www.khayelitshacommission.org.za/2014-01-29-12-13-54/23-january-2014-opening-statements/179-south-aftican-police-service-opening-statement.html.

Schensul, Jean J., and Stephen L. Schensul (1990). Ethnographic evaluation of AIDS prevention programs: Better data for better programs. The Evaluation Marketplace: Exploring the Evaluation Industry 46: 51–62.

Schlossman, Steven, and Michael Sedlak (1983). The Chicago Area Project revisited. Crime and Delinquency 29(3): 398–462.

Seekings, Jeremy (2013). Economy, society and municipal services in Khayelitsha. Report for the Commission of Inquiry into Allegations of Police Inefficiency in Khayelitsha and a Breakdown in Relations between the Community and the Police in Khayelitsha. Rondesbosch: Centre for Social Science Research, University of Cape Town.

Serrano-Berthet, Rodrigo (Ed.) (2013). O retorno do Estado às favelas do Rio de Janeiro: Uma análise da transformação do dia a dia das comunidades após o processo de pacificação das UPPs [Bringing the state back into the favelas of Rio de Janeiro: Understanding changes in community life after the UPP pacification process]. Rio de Janeiro: Fundação Getulio Vargas/World Bank.

Shaw, Clifford R., and Henry D. McKay (1942). Juvenile delinquency and urban areas. Chicago, IL: The University of Chicago Press.

Shearing, Clifford, and Jennifer Wood (2003). Nodal governance, democracy, and the new ‘denizens’. Journal of Law and Society 30(3): 400–419.

Steck, Jean-Fabian, Sophie Didier, Marianne Morange, and Margot Rubin (2013). Informality, public space and urban governance: An approach through street trading (Abidjan, Cape Town, Johannesburg, Lome and Nairobi). In Simon Bekker and Laurent Fourchard (Eds.), Politics and policies: Governing cities in Africa (145–168). Cape Town: HSRC Press.

Steinberg, Jonny (2011). Crime prevention goes abroad: Policy transfer and policing in post-apartheid South Africa. Theoretical Criminology 15(4): 349–364.

Steinberg, Jonny (2012). Establishing police authority and civilian compliance in post-apartheid Johannesburg: An argument from the work of Egon Bittner. Policing and Society: An International Journal of Research and Policy 22(4): 481–495.

Strauss, Anselm, and Juliet M. Corbin (1994). Grounded theory methodology: An overview. In Norman K. Denzin and Yvonna S. Lincoln (Eds.), Handbook of qualitative research, First Edition (273–285). Thousand Oaks, CA: SAGE Publications.

Super, Gail (2016). Volatile sovereignty: Governing crime through the community in Khayelitsha. Law and Society Review 50(2): 450–483.

The Presidency, Republic of South Africa (2004). Social cohesion and social justice in South Africa. Pretoria: Policy Co-ordination and Advisory Services.

Tshehla, Boyane (2002). Non-state justice in post apartheid South Africa—A scan of Khayelitsha. African Sociological Review 6(2): 47–70.

Van Der Spuy, Elrena (2000). Foreign donor assistance and policing reform in South Africa. Policing and Society: An International Journal of Research and Policy 10(4): 343–366.

VPUU: Violence Prevention through Urban Upgrading (2013). Semi-annual progress report No. 4., January to June 2013, VPUU Phase 3.

Warner, Barbara D., Elizabeth Beck, and Mary L. Ohmer (2010). Linking informal social control and restorative justice: Moving social disorganization theory beyond community policing. Contemporary Justice Review 13(4): 355–369.

Watson, Vanessa (2003). Conflicting rationalities: Implications for planning theory and ethics. Planning Theory & Practice 4(4): 395–407.

Youker, Brandon. W. (2005). Ethnography and evaluation: Their relationship and three anthropological models of evaluation. Journal of MultiDisciplinary Evaluation 2(3): 113–142.

Part III

Latin America

6Social exclusion, violences, and urban marginalisation in Central America

Empirical evidence and consequences for public policy

Juan Pablo Pérez Sáinz, Mario Zetino Duarte, and Florencio Ceballos Schaulsohn

This chapter has three objectives. First, we attempt to identify the dynamics of violence in urban territories in Central America where social exclusion is a significant problem.1 Second, we consider the responses that the residents of these territories have attempted to implement to deal with these dynamics. And finally, based on this analysis, we reflect on issues that emerge in terms of public policy and that point to problems extending beyond the territories studied.

With these aims in mind, the chapter is divided into four sections. The first contextualises the study sites in terms of the issue of social exclusion. This is followed by the main section, which examines the dynamics associated with the types of violence identified. The third section presents a critical analysis of a series of initiatives aimed at dealing with violence in the territories studied, highlighting their achievements and challenges. Finally, the conclusion offers three reflections that aim to define a series of questions and challenges from the perspective of public policy.

The territories: an overview from the perspective of social exclusion

The sites studied were located in Costa Rica (Concepción Arriba, Pacuare, and Cariari) and El Salvador (Quebrada Honda, Palomar, and Los Ranchos in Santa Tecla and El Puente, San Simón, and El Cocotal in Sonsonante). Concepción Arriba is located in the San José metropolitan area. It is the result of public housing projects sponsored by the Costa Rican government in the late 1980s to accommodate rural migrants. Pacuare is the product of subdivision and urbanisation processes promoted by public institutions, and is populated by people from different parts of the city of Limón. And Cariari, located in the province of Limón, was originally a rural settlement that has been subject to urbanisation, attracting migrants from different parts of the country.

The three territories of Quebrada Honda, Palomar, and Los Ranchos,2 all located in Santa Tecla (the metropolitan area of San Salvador), share the characteristic of being spontaneous urban developments located in environmentally vulnerable zones. Like every marginalised zone of Santa Tecla, there is a presence of gangs connected with Mara Salvatrucha, one of the two big maras in the country and the one that controls this locality. El Puente is an urban settlement located in the intermediate city of Sonsonate, first established some 50 years ago as the result of a subdivision process. San Simón is a precarious settlement that developed spontaneously as a result of the occupation of disused urban spaces. Finally, the third site located in Sonsonante is El Cocotal, the product of both subdivisions and informal settlements. This neighbourhood was still a disputed territory between the Mara Salvatrucha and the 18th Street gang at the time this study was conducted.3

As noted above, these are territories marked by the prevalence of social exclusion. Throughout this chapter, we will compare the group of Costa Rican study sites with the Salvadoran group.

Our focus on social exclusion requires the consideration of two dimensions. The first is primary exclusion, reflecting extreme disempowerment either in terms of employment precarity or the inability to ensure accumulation and growth by means of self-generated economic activity. The second dimension, which may consolidate primary exclusion, relates to state neglect reflected in a lack of access to social citizenship in terms of education and/or social security (i.e. to a social protection floor that makes equality of opportunity a viable possibility) (Pérez Sáinz and Mora Salas 2007; Pérez Sáinz 2012).

The analysis proposed here has been operationalised through the use of three scales (see Table 6.1), framed in terms of the situations that reflect the most extreme levels of inclusion and exclusion.4

The scores on these three scales, although calculated on an individual level, are aggregated according to the household to which the individual belongs, using a standardised range of values from 0 to 10.5 Thus, on each scale a score of 0 reflects a situation of maximum exclusion for the house-hold in question: it contains only long-term unemployed persons (more than one year without work) and/or unpaid workers, none of the house-hold members has any schooling, and no one has access to social security. Conversely, a score of 10 reflects maximum inclusion: household members in the labour market are paid employees who enjoy all available employment benefits and/or run medium-sized or large enterprises, all have completed higher education, and all have access to social security.

Based on these scales, the households in the territories of each country have been grouped into clusters.6 The averages for each cluster on these scales and their respective sizes are detailed in Table 6.2.

The precarity of these clusters has been charted on social maps using a system of coordinates where the y-axis corresponds to the labour market insertion scale, and the x-axis shows a simple average of the other two scales (education and social security). The y-axis reflects the dynamics of inclusion/exclusion generated by the market, while the x-axis shows the same type of dynamics sponsored by the state through aspects of social citizenship. Figures 6.1 and 6.2 represent the social maps for the two groups of territories.

Table 6.1 Scales of social inclusion/exclusion

Extremes of the scale

Labour market insertion scale

Education level scale

Social security scale

High inclusion

Non-precarious paid employment

Higher education

Direct coverage


Ownership of large or medium-sized enterprises



High exclusion

Extremely precarious paid employment

No schooling

No direct or indirect coverage


Self-generated subsistence activity




Unemployment Unpaid work



Table 6.2 Household clusters by group of territories (averages)

Group of territories and clusters

Labour market insertion scale

Education level scale

Social security coverage scale

Cluster size (%)

Costa Rican






























Source: Pérez Sáinz, Calderón Umaña, and Brenes Camacho (2016).

Both figures, or “social maps”, which offer a more immediate picture, as well as Table 6.2, reveal similarities and differences between the two groups of territories.

There are two similarities. The first is that both groups of territories are characterised by social dualisation, as there are two clusters of households with very different probabilities of social exclusion/inclusion.7 The second similarity is that the averages on the education level scales are very similar for the two groups of sites: 3.03 for the Costa Rican territories and 2.92 for the Salvadorans. Two dynamics lie behind this similarity: on the one hand, the increasing deterioration of public education in Costa Rica and, on the other, the “human capital” initiative implemented by four Nationalist Republican Alliance (ARENA) governments in El Salvador, with a clear (neo)liberal orientation.

Figure 6.1 Social clusters representing the territories studied in Costa Rica.

Figure 6.2 Social clusters representing the territories studied in El Salvador.

The differences, however, are many. The first is related to the weight of the cluster with the highest probability of social exclusion. In the Salvadoran group, this includes nearly two-thirds of all households; however, for the Costa Rican group, it contains only slightly more than half. As might be expected, there is a higher rate of social exclusion in the Salvadoran sites. Second, as can be observed on the “social maps”, the social inclusion cluster in the Costa Rican territories is located high up on the map, suggesting a strong level of inclusion. Conversely, in the case of the Salvadoran group, this cluster is closer to the middle, suggesting that inclusion is weaker. Differences between the social security coverage scales and, especially, the labour market scales can explain this contrast. A third difference, which can also be seen in the maps, is that the exclusion cluster in the Salvadoran group is located in a clear exclusion zone, close to the origin of the coordinates. On the other hand, in the Costa Rican group this cluster is further to the right. This difference is explained mainly by the impact of social security. The score on this scale for the exclusion cluster in the Costa Rican group is 5.56, reflecting the fact that the old welfare state is holding steady and mitigating the effects of primary exclusion from basic markets.8 Conversely, the score on this scale is only 0.24 for the Salvadoran sites, revealing the virtual absence of this component of social citizenship.

Consequently, although social exclusion is prevalent in both the Costa Rican and Salvadoran territories, it differs in intensity and form. These differences are important to understanding the types of violence that characterise these sites, as will be explained below.

The dynamics of violence in the territories

Our focus on social exclusion is based on the central role it plays in power relations, insofar as this phenomenon is understood as the result of extreme disempowerment occurring either in terms of precarity of paid employment or of exclusion from opportunities to accumulate wealth.9 In this sense, social exclusion is an expression of extreme inequalities (Pérez Sáinz 2016). When the state is not able to neutralise this disempowerment through social citizenship, the excluded members of society, exposed to a specific situation of marginalisation, are able to empower themselves through violence.10 The disempowerment expressed in the mechanisms of social exclusion can be understood in terms of objective disempowerment, which in turn gives rise to forms of subjective disempowerment. Such forms refer to how exclusion is accepted subjectively in the relationships that people establish with their peers and with society; that is, by constructing a subjective view through which they interpret their own condition in relation to society and give meaning to alternative actions to attempt to seize some kind of control (empowerment) over their lives and the objective conditions they face (Zetino Duarte 2003, 2006). This is the key factor linking inequality, social exclusion, and certain types of violence.

The dynamics of the different types of violence that are considered in this study have been analysed with a special focus on their relational nature. Emphasis is placed on the central role of power as a mechanism for social relations and for violence. This mechanism refers to the relational use of the various resources to which individuals and social groups have different degrees of access, and through which they establish social relations. Such resources position individuals and social groups in the structures and dynamics of their relationships and, at the same time, orient them subjectively to those relationships. Within these contexts and structures, the use of resources configures power relations and gives content and meaning to the unique dynamic of each type or form of violence in a given context. This is expressed in the capacity to direct the course of both the relationship dynamics and the resulting outcomes. The resources of individuals or social groups, whether material or symbolic, are configured and re-configured historically—objectively and subjectively—in the context of the conditioning factors imposed on them by social structures and in accordance with their need to confront reality. Power, however, is not synonymous with violence. When the relational dynamic of power or the configuration of power structures in different contexts generates effects which, intentionally or not, cause harm, threaten lives or personal security, or the conditions leading to these developments, violence occurs (Zetino Duarte 1997, 2003, 2006; Zetino Duarte, Brioso, and Montoya 2015).

Understanding the dynamics of violence in these terms, one can see that the central role of power as a mechanism is not located on a specific level of reality (macrostructure, social context, community, family, or individual) as a determining factor of the others; rather, it is present on any and all of these levels, depending on the particular dynamic of the level on which the violence is expressed and of the relationships conditioning the other levels in relation to that level. The focus of this study has therefore been on the configuration of different types of violence on relational levels, in the context of the dynamics produced on those levels by social exclusion in the territories.

This study initially considered two types of violence, both instrumental but with different purposes. Social violence pursues esteem, honour, and prestige, while profit-seeking violence aims for the enhancement of pleasure and material benefit (Calderón Umaña 2013). However, this research has demonstrated that one of these forms takes on a central role and becomes contextual violence, which is effectively a third category. It is precisely in relation to this third category that the link between social exclusion and the different forms of violence can be seen most clearly. This observation leads us to the dynamics of violence that have been identified in the study sites, as shown in Table 6.3.

The first reflection we can draw from Table 6.3 is that the manifestation of contextual violence differs depending on the country. In the case of Costa Rica, it is through drug micro-markets, while, in El Salvador, it is the control of the territories by the maras. This difference is not unrelated to the expressions of social exclusion. In the Costa Rican sites, it is associated with primary disempowerment, while in the Salvadoran group, it is linked to state neglect.

Table 6.3 Dynamics of violence by type of violence and by territory

Types of violence

Costa Rican territories

Salvadoran territories


Drug micro-markets

Generation of imaginary of fear

Dispute and control of territory by maras

Stigmatisation of territory


Theft and robbery associated with drug use

The maras engage in extortion but also offer protection against external profit-seeking violence


Violence between couples and between neighbours

Resolution of extreme family and neighbourhood conflicts by the maras


Victim/perpetrator binary in neighbourhood violence is diminished These forms of violence are naturalised and rendered invisible

Source: Pérez Sáinz (2015: Chapters II and III).

Indeed, in the Costa Rican territories there was already a demand for drugs, but the current supply is more diversified and has been boosted by the redefinition of international drug trade routes between South and North America, making Central America a bridge. Payment in kind11 (in drugs) for logistical services rendered in the region to support the drug trade, in addition to the growing problem of money laundering, has led to this increase in drug use in the local territories studied. This is a side effect of the trade and these micro-markets, although they form part of the global drug network, are not located on the main route. This does not mean that their external definition will not change in the future if local drug consumption in transit countries becomes key to the profitability of this activity. The local drug market is characterised by three elements: it is based on trust and betrayal brings severe reprisals; it creates specific territorialities, privileging certain points in the territory because of its operations; and its activity is based on violence. Due to the final element, this activity generates fear and insecurity among residents of the territories, leading to their social isolation (Calderón Umaña and Salazar Sánchez 2015).

On this point, it is important to highlight that locals turn to drug dealing because it offers a much better income than that offered by the occupations available to people constrained by exclusion and marginalisation. In other words, a connection is established between primary exclusion and this type of violence. This dilemma is significant in the case of youth,12 a fact that supports Valenzuela’s (2012) observation that the “narco-world” has become an important space for the definition of life projects among young people.13

In the case of the Salvadoran territories, contextual violence is structured around juvenile gangs or maras.14 These groups do not merely control specific points within the territory, but the whole territory. This control is essentially the result of the absence of the state, having effectively given up its monopoly on coercion in these areas and leaving other actors to fight over it. In this case, we find another link between social exclusion and contextual violence, not in relation to primary exclusion but to state neglect.

This control is often the subject of dispute between gangs; indeed, at the time this research was conducted, a dispute was ongoing in one of the sites in Sonsonante. The violent activity of the maras has multiple dimensions, but in this study there are three that stood out. The first is the recruitment of minors. Second, the widespread use of violence generates isolation and silence among local residents, restricting the spaces for social interaction in the territories. Only in evangelical churches, which proliferate in these contexts, is a certain degree of social interaction possible, albeit in the context of religious practice. Third, the violent actions of the maras lead to the stigmatising of the territories by other sectors of society, with consequent problems for local residents when they interact with people outside their own territory (Zetino Duarte, Brioso, and Montoya 2015). Indeed, this stigmatisation is the result of a metonymic relationship between the criminals themselves and the territory where they reside. In prevailing perceptions, especially in the media, the territory is socially homogenised and all its residents are deemed criminals (Cornejo 2012).15 This metonymy is clear in the case of young people not only in El Salvador but also in Guatemala and Honduras, where they are automatically viewed as gang members (Aguilar Umaña 2016).

Theft and assault are the main manifestations of the second type of violence considered in this study—profit-seeking violence—in the Costa Rican sites. In most cases, residents of these territories tend to associate this criminal activity with local drug consumption. This means that, although the study has identified this type of violence as a sporadic phenomenon, it constitutes the main source of fear for residents of these territories, who respond to these threats in isolated and individual ways (Calderón Umaña and Salazar Sánchez 2015).

Profit-seeking violence also exists in the Salvadoran territories studied, but it is engaged in exclusively by the maras. The victims are the local residents, who are subjected to different forms of extortion, especially in the case of small businesses.16 However, this monopoly on violence includes protection against profit-seeking violence of external origin. This points to the ambiguous role of these gangs, making it impossible to deal with them through a simple analysis or facile measures (Zetino Duarte, Brioso, and Montoya 2015).

This same ambiguous role is evident in relation to the third form of violence: social violence. In the Salvadoran territories, the maras control and restrict extreme expressions both of family conflicts (in which the victims are mostly women and children) and neighbourhood conflicts, as they aim to prevent external intervention in their resolution.17 This type of control mechanism does not exist in the Costa Rican territories, where both violence between couples (in which women are the main victims) and between neighbours is commonplace but is not viewed by residents as being as threatening as profit-seeking violence.

There are two phenomena in relation to social violence that are common to all the territories in both countries. One is violence that tends to be rendered invisible and naturalised. This is mainly due to the significance taken on by violence that becomes contextual because it is around such violence that the local imaginaries of fear are constructed. Second, in neighbourhood conflicts, the importance of the victim/perpetrator binary is diminished because the roles are often exchanged. Indeed, this phenomenon points to the danger of essentialising this dichotomy by assigning fixed identities that end up stigmatising the people concerned (Aguilar Umaña 2016).

Initiatives against violence in the territories

In this study, three initiatives in the territories to prevent and/or confront violence were identified. Two are in the Costa Rican territories (Cariari and Concepción Arriba), while the third has been applied in most of the Salvadoran territories analysed.

The main characteristics of these initiatives are outlined in Table 6.4.

Collective actors have been identified in the territories, some of which are articulated around particular or general interests or needs. Notable in the case of general interests are development associations. However, in all the study sites, resident participation in these kinds of organisations is extremely rare, and thus their potential impact is very limited.18 It is in this sense that we argue that the territories have residents but no community. The absence of community means that of the three initiatives against violence identified in the territories, two were of external origin.

The only relative exception is Cariari because the implementation of the Territorios Seguros [Secure Territories] initiative depends on actions by the residents themselves. It is a recent initiative that began in a nearby community but has since expanded—initially on the Atlantic coast—and is now nationwide. Through charismatic regional social leadership, a plan has been established to recover citizenship, especially social citizenship, for subjugated sectors, in which women have played a leading role. Its work focuses on three lines of action: citizen security, resolution of conflicts between citizens and institutions, and training the individuals participating in the process, with the aim of consolidating skills so that they can create their own productive choices. Indeed, Territorios Seguros is self-described as a “concept of life”19 and its activities focus on engaging public institutions so they have a presence in the territories where they have been organised.

Table 6.4 Initiatives to tackle violence in the territories


Costa Rican territories

Salvadoran territories


Internal: Territorios Seguros

External: Community Security Programme

External: Municipal Citizen Security Plans (United Nations Development Programme), focusing on territories or implemented through municipal networks


Strengthening of social citizenship

Connecting residents with police service

Context of truce

Strengthening of community organisations and of institutional networks in the territories

Redefinition of power relations between community organisations and the maras


Addressing social violence (Territorios Seguros)

Sustainability of the initiative (Community Security)

Sustainability of the initiative after end of truce


Development of community actors in the territories

Source: Pérez Sáinz (2015: Chapters IV and V).

In Concepción Arriba, we studied experiences in the context of the Community Security Programme implemented by the Ministry of Public Security. This involves training local residents, through different modules, in the prevention of violence in conjunction with the police service. The experiences analysed in our study suggest that implementation is uneven. There are three factors that could explain this variation: the period of residence of the inhabitants, which is generally associated with home ownership; the delimitation of the territory; and the operation of the resultant committee. Specifically, a committee composed of long-term residents, overseeing a small and clearly delimited territory and engaging in other activities in addition to security initiatives, is the type of committee with the best chance of success (Salazar Sánchez and Pérez Sáinz 2015).

In the case of the Salvadoran territories, the experiences studied relate to the Municipal Citizen Security Plans in which the role of the United Nations Development Programme (UNDP) has been crucial. Two types of implementation have been identified. The first is more focused and limited to a closed territory, as in the case of Palomares in Santa Tecla. The second is more open and operates through municipal networks, as happens in Sonsonante. These two types of implementation have had different effects on the prospect of turning the populations of the territories into social actors capable of tackling the dynamics of violence. The most notable aspect of the focused approach in Santa Tecla is that it has reinforced the organisational structures in the settlements. It has strengthened residents’ capacity for management and independent action without depending on the dynamics of local government, generating new leadership structures and rebuilding the social fabric. Conversely, in Sonsonante, new leadership structures have not emerged. The community association is not autonomous and does not have a capacity for engagement. All actions are dependent on the local government. The mechanisms for management, planning, decision-making, and execution of the work are based outside the community, under the control of the local government (Brioso, Zetino, and Montoya 2015).

These programmes have been assessed in terms of their achievements, as well as the challenges they face. Thus, in the case of Territorios Seguros, there has been some very significant specific success in the fight against the local drug trade due to the collaboration of this organisation with the judiciary. It is also important to highlight its clear commitment to social citizenship. Its purpose involves engaging with the state in defence of the public good by asserting citizens’ rights, and there can be no doubt that it has helped members of the organisation to develop civic awareness. However, there is no similar initiative to deal with the issue of primary exclusion. Territorios Seguros also faces significant challenges in tackling the two types of social violence considered in this study. Finally, it remains to be seen whether the organisation’s actions, under its organisational model, will crystallise in the emergence of a community actor defined in accordance with the interests of the local residents.

In Concepción Arriba, the experiences considered reveal that the Community Security Programme has resulted in engagement and connection between local residents and the police service. An analysis of this success confirms the recovery of community organisation. We have identified three challenges in this territory in relation to this type of experience. First, there is a danger that the demands of residents may be limited to the issue of security. Second, as a corollary to the first challenge, if the threats associated with violence are dealt with, the committee will cease to operate, which indicates a sustainability problem. And third, as a result of the previous two factors, the construction of community would be at risk (Salazar Sánchez and Pérez Sáinz 2015).

The Salvadoran cases are likewise characterised by achievements and challenges. Notable among the achievements under the Municipal Citizen Security Plans are the following: strengthening of local leadership; “social capital” has been stimulated in terms of trust and cohesion among residents; and greater confidence in institutions has developed. There are also three challenges that these initiatives face. The first is related to the inevitable tension between the two logics of public security: the logic of repression of crimes and that of social reintegration of offenders. Unsurprisingly, these types of projects are favoured at times when the second logic prevails and, conversely, they are threatened when the logic of repression predominates. Second, it is important to take into account the sustainability of the projects, especially the economic initiatives, which can generate secure sources of income that serve to reduce the effects of social exclusion. These two challenges affect the gangs and, in this respect, we need to consider a third factor, related to the change of perspective among the leaders of the maras concerning whether these types of interventions represent a transformation. Due to the mode of implementation of the Municipal Citizen Security Plan in Santa Tecla, there is a certain degree of specificity in the case of the achievements in Palomares. There has been a strengthening of community organisation that has resulted in a reconfiguration of the power relations between the mara and local residents (Brioso, Zetino, and Montoya 2015).

However, these achievements are overshadowed by the aforementioned tension between the two logics of public security—the key issue for the viability of these kinds of initiatives. Indeed, the fieldwork for this study in the Salvadoran territories was only possible because it coincided with a truce between gangs. Nevertheless, there have been situations of repressive intervention on the part of law enforcement agencies that have undermined the processes encouraged by this type of initiative. The truce is now over and at the time of writing (2016–2017) the Municipal Citizen Security Plans have been subordinated to the logic of the special nationwide security measures approved by the Salvadoran Legislative Assembly in April 2016. The preventive component of the original plans has been completely superseded by the repressive component. As a result, the achievements indicated above for the territories included in this study are in jeopardy, and the dominant presence of the repressive violence of the state is being consolidated to confront the violence of the maras. In this way, the contextual violence in these territories is mutually conditioned.

Conclusions: consequences for public policy

Public policy on a national scale—and the advice and recommendations provided by multilateral bodies offering technical support and cooperation that aim to influence that policy—is based on the assumption that certain criteria and findings are true. In the case of the reduction and prevention of urban violence in Latin America, this advice and the discourse associated with it involves mapping a characterisation of the violence, reinforcing community strategies in response to it, and developing a one-dimensional definition of what organised crime represents in the local space.

The research presented in this chapter aims to appeal to the prevailing common sense in the world of policies that address the different types of violence in cities. Three conclusions drawn from our findings are proposed that question and complicate the following three aspects: collective action, characterisation of the violence, and representations of organised crime.

The first conclusion relates to the difficulties associated with an a priori assumption of the existence of a community actor that would suggest the establishment of community within a given territory. The second is that violence cannot be conceived of in the singular, but needs to be approached as a diverse phenomenon that requires us to consider “violences”, assessing and characterising each instance of violence in order to understand how to prioritise it from the perspective of public policy. And third—specifically relevant to the Salvadoran case—is the need to understand that the phenomenon of the maras reflects an ambiguous reality inherent in criminal organisations that have evolved into quasi-official agencies in contexts of state neglect, in a way that extends beyond their simple characterisation as an agent of violence. Each of these conclusions is discussed in more detail in the following paragraphs.

First, there is a large body of literature that analyses the difficulties and limitations associated with the development of active communities in urban territories marked by exclusion and violence. The case studies included in this research have made it possible to identify three factors that complicate collective action and organisation: territorial factors, social factors, and factors associated with violence.

With respect to territorial factors, we need to recognise first these study sites are shaped by different migration flows. The multiple origins of their residents hinder the establishment of a collective identity and the mechanisms of social cohesion associated with such an identity. The relatively recent establishment of these settlements makes it difficult for residents to identify with their territory, as this is something normally associated with long-term coexistence. Added to this are problems associated with home ownership by the residents, which contributes to a weakening of the sense of belonging to the territory. Another important factor is the role and presence of the state in these communities. Inadequate recognition of these settlements by the city and its authorities, reflected, for example, in limited access to basic services, sends a message of exclusion or of not belonging to the city.

The establishment of communities in these kinds of territories is also complicated by the social heterogeneity of their composition, which is aggravated by the impact of social exclusion. The social maps examined in the study are characterised by a tendency towards social dualisation. Not all residents are similar from a socioeconomic perspective: one cluster of households is irrevocably trapped in exclusion while another manages, with varying degrees of success, to distance itself from that exclusion.

On the other hand, the study has revealed that contextual and profit-seeking forms of violence, as generators of fear and insecurity, provoke spirals of isolation and silence in the territories. In this context, the evidence appears to indicate that only religious practice and association with religious institutions present in the territory offer a space for the re-establishment of social interaction, although their effects are limited.

In short, we need to view community actors as pre-existing realities for which the establishment of public policy entails risks and has consequences. Community organisations and actors exist in the territory, but their capacity is limited as they are characterised by contradictions and a complex anthropology. In terms of policy recommendations, this means that organisations providing external interventions should understand and acknowledge—at the same time—that they are not acting in a vacuum and that part of their task is precisely to strengthen and contribute to the generation of a social fabric that will facilitate positive social cohesion. In this endeavour, a central element of any policy should be to incorporate a gender perspective that clarifies how women and men are differentiated as social actors in the territory, and their different definitions of the roles that such collective social actors can play in response to violence.

The second idea associated with the implications of public policy that emerges from this study is related to the conceptual pluralisation of violence in the territory. With respect to contextual violence, the study demonstrates the impact of causal factors external to the territories, associated mainly with social exclusion.

Social exclusion is expressed in terms of disempowerment in basic markets and as the absence of the state in the territories. In this sense, it is not enough for public policy to focus only on the state component; it should also take into account the basic markets associated with the problem: unemployment, paid employment without protection and rights, and engagement in activities that are self-generated but trapped in survival logic. It is insufficient to influence residents through employability and/or entrepreneurial initiatives if these are offered in isolation, on the basis of limited engagement; they need to be associated with changes to the socio-economic structure in which the residents exist and which, to a large extent, defines and restricts their empowerment through employment.

In the definition of public policy priorities and angles of intervention, it is important to reconsider the tensions and representations associated with profit-seeking and social violence. We believe that the Costa Rican case can offer clues in this respect. In the imaginaries of residents, but also of the authorities, expressed in surveys and interviews, it is profit-seeking violence that emerges as the most threatening and which therefore should be prioritised—and, indeed, it is here where the public authorities tend to focus their attention. However, according to the testimonies of the residents, profit-seeking violence tends to be sporadic, while social violence, expressed in conflicts between couples or between neighbours, is commonplace and constant, and its consequences can be just as severe as those of profit-seeking violence. Nevertheless, social violence tends to be naturalised and rendered invisible, both from the perspective of government action and from the actions of quasi-official actors such as the Salvadoran gangs.

Social violence associated with couples or families, and its invisibility, reveals the need to incorporate a greater emphasis from a gender perspective into any analysis leading to the definition of public policy priorities. The invisibility of social violence is, to a large extent, a consequence of the naturalisation—as distinct from and stronger than the normalisation—of gender violence that affects women and girls. To the extent that it is not recognised as such, or that its importance is dismissed, social violence tends to be given a lower priority than profit-seeking violence. The proposal drawn from this study is that social violence should be made visible and given the priority it deserves. In other words, there needs to be a re-balancing of the importance of these two types of social violence (between couples or between neighbours) in relation to their context.

A third and final consideration for public policy, limited in this case to the Salvadoran context, relates to the maras and their inherent ambiguity from the perspective of the territory and the actors who coexist in it. This study confirms something that research into organised crime and quasi-official organisations has shown repeatedly in other parts of the world: on the one hand, these organisations hold a monopoly on violence, generating fear and terror among local residents; on the other, they fill a vacuum left by the state, offering and mediating access to essential goods, including control of violence in the local territory. In addition to committing violence, the maras operate as regulators and, in a complex way, as guarantors of protection against profit-seeking violence coming from outside the territories (committed by “other” maras). They also act as mediators in domestic and neighbourhood conflicts as a way of keeping the state out of their territories. If we fail to understand these opposing (and, at first glance, contradictory) perspectives that inhabit the imaginaries in the local space, we cannot hope to generate effective community responses to mitigate violence.

We believe that from the perspective of effective and inclusive public policies to counteract violence and exclusion, there needs to be a political commitment to the reintegration of young people, involving them in the establishment of a community actor. This response, however, poses as many risks as it does opportunities. In terms of opportunities, it would be feasible to implement strategies of this kind only when the logic of recomposition of the social fabric, as a response to public security, predominates over the logic of repression and exclusion. Moreover, these strategies are possible only insofar as the presence of the institutional network of the state is reintroduced to the territories, filling the institutional vacuum that has contributed to the rise of the gangs. If these initiatives are successful in strengthening community organisation (rather than assuming it to be pre-existing), the power of the maras could be reduced. Finally, economic projects that attempt to deal with primary exclusion need to be successful and sustainable, generating lasting sources of income which, while they may not be similar to profits from criminal activity, at least allow individuals to lead a decent life and re-establish public recognition of the value of labour.

Regarding the risks, the first is related precisely to the lack of sustainability of the aforementioned economic projects that aim to respond to exclusion, feeding young people’s argument that society does not provide them with the means to achieve their aspirations of wealth, power, and prestige. The second risk has to do with the imposition of the logic of repression. The logic of reintegration does not always predominate, and the cycles of “peace-making” or truce have their own dynamics that may also conceal a hidden agenda among gang leaders—to use such moments of calm as an opportunity to reinforce their power and control over the territories.

These are the clearest conclusions from the three basic ideas that emerge from this research. We conclude by reiterating those ideas, as they form the core of our message: we should not assume the existence of a community actor in a territory that would reflect the existence of community; we should not speak of violence in the singular, but of “violences”, and assess each form of violence to determine how to prioritise it; and, specifically relevant to the Salvadoran cases, the phenomenon of the maras reveals an ambiguous reality that extends beyond their mere characterisation as agents of violence, suggesting that the analysis cannot be simple, and the solutions will not be easy.


1This chapter summarises the main empirical findings, as well as their consequences in relation to public policy, of the research project carried out in Costa Rica and El Salvador between April 2013 and June 2015, in the context of the Safe and Inclusive Cities programme developed by Canada’s International Development Research Centre (IDRC) and the United Kingdom’s Department for International Development (DFID). The project involved a survey of house-holds, the main results of which are presented in terms of social exclusion, followed by a stage of qualitative analysis related to different forms of violence and to initiatives in response to them. This research focused on nine study sites: three in Costa Rica and six in El Salvador. These sites are urban territories located in metropolitan areas and in intermediate cities in both countries, all characterised by a prevalence of social exclusion.

2The names of the Salvadoran territories are fictitious to protect key informants from these locations.

3A more detailed description of these territories can be found in Pérez Sáinz (2015: Annex A), available online (in Spanish) at www.flacso.or.cr/images/flippingbook/pdfs/libros/exclusion_social_violencia.pdf.

4The development of these scales is discussed in Pérez Sáinz (2015: Annex B).

5For households where none of the members are employed or unemployed, the score on the labour scale has been estimated based on criteria related to the quality of the source of non-labour income transferred to this type of household.

6The procedure used was two-step cluster analysis.

7This phenomenon of dualisation is equally clear in each of the sites considered separately. The only exception is Concepción Arriba, where households are broken into four clusters. This is a site located in Costa Rica’s largest metropolitan area, where the diversity of options in the labour market would appear to generate greater social heterogeneity.

8This refers to markets in which a society’s basic resources are exchanged: labour, capital, land, and knowledge. These markets are asymmetrical power environments that generate different types of inequalities, making the generation, circulation, and appropriation of surpluses possible under capitalism (Pérez Sáinz 2016).

9This proposition is in contrast to current approaches to poverty in Latin America, where deprivation is understood in non-relational terms and defined according to standards established by experts. Such approaches conceal any reference to power and conflict.

10There are other responses to marginalisation: immigration, religious activity, or collective action through social movements. In other words, not every excluded person resorts to criminal activity.

11This is a phenomenon that has been identified in other regions since the 1980s (Briceño-León 2007).

12The dilemma can be resolved through various types of mediation: socialisation processes (especially at the family level), interactions with public institutions, couples/relationship therapy, capacity for agency, etc. (Espíndola Ferrer 2013). This means that not every young person growing up in a marginalised area is inevitably doomed to a life of crime.

13Nevertheless, it is also important to note that young people may become involved in crime due to debts contracted with dealers rather than out of a quest for profit (Zamudio Angles 2013).

14For obvious security reasons, this research has made the maras an object of study. However, when discussing the different dimensions considered, these gangs emerged as a key explanatory element. On this point, see the interpretative diagram of the Salvadoran territories (Pérez Sáinz 2015: 251).

15Thus, Manzano Chávez (2009), in her study of “critical neighbourhoods” in Santiago de Chile, concludes that most of the population does not completely lose their ties to the social system.

16The so-called “renteo” is key to the economic survival of the maras. A study conducted in El Salvador on the extortion of small businesses identified three types: systematic “symbiotic” extortion, offering protection in exchange for payment; systematic “non-symbiotic” extortion, merely taking money without offering any service in return, which has a predatory effect because it ultimately destroys the businesses it targets; and opportunistic extortion, engaged in by independent individuals who take advantage of the existence of the first two types (Ponce et al. 2016).

17On this point, it has been noted that agents of violence who control territories abandoned by the state can provide four types of public goods essential to local life: protection against external attacks; intermediation in conflicts within the community, whether domestic or between neighbours; mediation of community activity by filtering and permitting certain actions by agents and institutions from outside the community; and the celebration of festive activities that entertain the community and that are highly valued by local residents (Perea et al. 2014).

18In the survey of households, the highest participation was in Cariari, where only 8.9 per cent of households had at least one member participating in community organisations.

19This conception is systematised in terms of progressive processes of empowerment of citizens in relation to the state (Salazar Sánchez and Pérez Sáinz 2015: 148–155).


Aguilar Umaña, Isabel (2016). Victimarios y víctimas de la violencia: de nexos invisibilizados y falsas dicotomías en el Triángulo Norte de Centroamérica [Victims of violence: Invisible links and false dichotomies in the Northern Triangle of Central America]. In Markus Gottsbacher and John de Boer (Eds.), Vulnerabilidad y violencia en América Latina y el Caribe [Vulnerability and violence in Latin America and the Caribbean] (73–89). Mexico City: Siglo XXI/Proyectos Estratégicos Consultoría.

Briceño-León, Roberto (2007). Sociología de la violencia en América Latina [Sociology of violence in Latin America]. Quito: FLACSO.

Brioso, Larissa, Mario Zetino, and Margarita Montoya (2015). La acción comunitaria frente a la exclusión y la violencia. Configurando potencialidades en asentamientos de El Salvador [Community action in the face of exclusion and violence: Setting potentialities in settlements in El Salvador]. In Juan Pablo Pérez Sáinz, Larissa Brioso, Rodolfo Calderón Umaña, Margarita Montoya, Karla Salazar Sánchez, and Mario Zetino (Eds.), Exclusión social, violencia y respuestas comunitarias en ciudades CentroAmericanas: explicando variaciones para orientar políticas [Social exclusion, violence and community responses in Central American cities: Explaining variations to guide policies] (132–161). San José, Costa Rica and San Salvador: FLACSO.

Calderón Umaña, Rodolfo (2013). Proposiciones analíticas para el estudio de la violencia en Centroamérica: una mirada desde la exclusión social [Analytical proposals for the study of violence in Central America: A perspective related to social exclusión]. Revista Digital de la Maestría en Ciencias Penales [Master of Criminal Sciences Digital Journal] 5: 199–226.

Calderón Umaña, Rodolfo, and Karlr Salazar Sánchez (2015). Dinámicas de violencia en las comunidades costarricenses [Dynamics of violence in Costa Rican communities]. In Juan Pablo Pérez Sáinz, Larissa Brioso, Rodolfo Calderón Umaña, Margarita Montoya, Karla Salazar Sánchez, and Mario Zetino (Eds.), Exclusión social, violencia y respuestas comunitarias en ciudades CentroAmericanas: explicando variaciones para orientar políticas [Social exclusion, violence and community responses in Central American cities: Explaining variations to guide policies] (41–65). San José, Costa Rica and San Salvador: FLACSO.

Cornejo, Catalina Andrea (2012). Estigma territorial como forma de violencia barrial. El caso del sector de El Castillo [Territorial stigmatisation as a form of neighborhood violence: The case of El Castillo]. Revista INVI 27(76): 177–200.

Espíndola Ferrer, Fabiana (2013). “Grietas” en el tejido social. Experiencias biográficas de jóvenes montevideanos des los “lugares” del espacio social [“Cracks” in the social fabric. Biographical experiences of young people from Montevideo from the “places” of the social space]. PhD thesis in Social Sciences with specialisation in Sociology submitted to El Colegio de México, Mexico City.

Manzano Chávez, Liliana (2009). Violencia en barrios críticos. Explicaciones teóricas y estrategias de intervención basadas en el papel de la comunidad [Violence in critical neighbourhoods: Theoretical explanations and intervention strategies based on the role of the community]. Santiago: RiL Editores/CESC.

Perea, Carlos Mario, Ana María Jaramillo, Andrés Rincón Morera, Michel Misse, César Alarcón, and Max Yuri Gil (2014). Introducción. La paradoja latino-Americana. Las ciudades en perspectiva comparada [Introduction. The Latin American paradox. Cities in comparative perspective]. In Ana María Jaramillo and Carlos Mario Perea (Eds.), Ciudades en la encrucijada: Violencia y poder criminal en Río de Janeiro, Medellín, Bogotá y Ciudad Juárez [Cities at the crossroads: Violence and criminal power in Rio de Janeiro, Medellín, Bogotá and Ciudad Juárez] (9–48). Medellín: Corporación Región and Instituto de Estudios Políticos y Relaciones Internacionales (EPRI)/Universidad Nacional de Colombia.

Pérez Sáinz, Juan Pablo (2012). Exclusión social. Una propuesta crítica para abordar las carencias materiales en América Latina [Social exclusion: A critical proposal to address the material deficiencies in Latin America]. In Juan Pablo Pérez Sáinz (Ed.), Sociedades fracturadas. La exclusión social en Centroamérica [Broken societies: Social exclusion in Central America] (11–48). San José, Costa Rica: FLACSO.

Pérez Sáinz, Juan Pablo (Ed.) (2015). Exclusión social y violencias en territorios urbanos centroAmericanos [Social exclusion and violence in Central American urban territories]. San José, Costa Rica: FLACSO.

Pérez Sáinz, Juan Pablo (2016). Una historia de la desigualdad en América Latina: La barbarie de los mercados, desde el siglo XIX hasta hoy [A History of inequality in Latin America: The barbarism of the markets, from the 19th century until today]. Buenos Aires: Siglo XXI.

Pérez Sáinz, Juan Pablo, and Minor Mora Salas (2007). La persistencia de la miseria en Centroamérica. Una mirada desde la exclusión social [The persistence of misery in Central America: A perspective related to social exclusion]. San José, Costa Rica: Fundación Carolina/FLACSO.

Pérez Sáinz, Juan Pablo, Rodolfo Calderón Umaña, and Gilbert Brenes Camacho (2016). Exclusión social, violencia y ámbito doméstico. Evidencia y reflexiones desde Centroamérica [Social exclusion, violence and the domestic environment. Evidence and reflections from Central America]. Papeles de Población 22(87): 9–41.

Ponce, Carlos, Margarita Beneke de Sanfeliú, Mario Chávez, and Mauricio Shi (2016). Extorsiones a la micro y pequeña mmpresa de El Salvador, documento Proyecto Solucion [Micro and small business extortion in El Salvador]. San Salvador: FUSADES.

Salazar Sánchez, Karla, and Juan Pablo Pérez Sáinz (2015). Estrategias de confrontación ante algunas formas de violencia. La experiencia de dos territorios costarricenses [Strategies to confront some forms of violence. The experience of two Costa Rican territories]. In Juan Pablo Pérez Sáinz, Larissa Brioso, Rodolfo Calderón Umaña, Margarita Montoya, Karla Salazar Sánchez, and Mario Zetino (Eds.), Exclusión social, violencia y respuestas comunitarias en ciudades CentroAmericanas: explicando variaciones para orientar políticas [Social exclusion, violence and community responses in Central American cities: Explaining variations to guide policies] (98–131). San José, Costa Rica and San Salvador: FLACSO.

Valenzuela, José Manuel (2012). Narcocultura, violencia y ciencias socio-antropológicas [Narcoculture, violence and socioanthropological sciences]. Desacatos 38: 95–102.

Zamudio Angles, Carlos Alberto (2013). Jóvenes en el narcomenudeo: el caso Ciudad de México [Young people in the drug trade: The case of Mexico City]. URVIO 13: 111–123.

Zetino Duarte, Mario (1997). Att förstå Brottsligheten bland ungdomar med invandrarbakgrund: En Tolkning av två levnadsberättelser. [Understanding delinquency in youth with immigration backgrounds: An interpretation of two life stories]. Master’s of Social Science thesis (No. 23) submitted to Linnaeus University, Växjö.

Zetino Duarte, Mario (2003). Vi kanske kommer igen om det låser sig [We may come back if we get stuck]. PhD dissertation submitted to Lund University, Lund.

Zetino Duarte, Mario (2006). Empoderamiento y prevención [Empowerment and prevention]. Estudios CentroAmericanos ECA 61(693–694): 701–727.

Zetino Duarte, Mario, Larissa Brioso, and Margarita Montoya (2015). Dinámicas de violencia en los territorios salvadoreños [Dynamics of violence in the Salvadoran territories]. In Juan Pablo Pérez Sáinz (Ed.), Exclusión social y violencias en territorios urbanos centroAmericanos [Social exclusion and violence in Central American Urban Territories] (99–144). San José, Costa Rica: FLACSO.

7The contribution of informal institutionality to safe cities in Venezuela

Roberto Briceño-León

“After they killed my oldest son, I was left thinking about what I could do.” This is how Maria started to explain to me and my research team how her struggle for peace began. With her face still wracked with pain, she tells us that he was her oldest child. “I can’t tell you that he was the one I loved most, because you love all your children the same.” But she recognises that she had a very special fondness for him, perhaps due to how painful and complicated his birth was. “It was like the boy didn’t want to be born.”

Vengeance was the norm in the barrio; the “snake” had to be killed to pay back the blood debt. And that debt had to be collected by the brother, the father, the cousin, or any man in the area where they lived. Aggression is regarded as a collective affront, suffered by everyone: the family, the neighbourhood, and the barrio.

The war between the gangs from the high part of the mountain and those from the low part of the mountain had been going on for many years. An invisible border was established as there was a danger of being attacked by the gang from the other territory: they are both a neighbour and an enemy. Economic pressures gave rise to feuds: control of the trade in drugs and stolen goods justified control of the territory. But then, over time and with the dead, that had been forgotten; honour and pride had taken over.

“I said I didn’t want more dead, I didn’t want revenge, this slaughter had to stop …” With her gesture, Maria, as a mother, renounced her right to revenge, to demand punishment exacted by the hands of her family members and neighbours. Her gesture surprised both friends and enemies—especially the men—because it changed the rules that had driven the conflicts for years: the endless chains of blood debts that were impossible to collect without starting a new cycle of pain and death. “But I didn’t do it out of goodness, I did it for my own interests because I had another son and I didn’t want them to kill him too.…”

And so the peace process began in the district. Perhaps this is also how, many centuries ago and in many places, a process of civilisation began that led to the building of a social contract and the rule of law.

The city, pacts, and violence

Cities are houses, streets, neighbourhoods, industries, businesses, roads, parks, but above all they are norms. The city exists in the rules that govern daily coexistence, agreements that ensure the mechanisms for access to resources and for resolving disagreements and conflicts. That is why the city is the “polis”: the centre of politics, understood as the way of producing arrangements for cooperation and competition, and regulating and legitimising domination and subjugation. This is what reduces the use of violence and makes it possible for people to live together without wars and deaths.

Perhaps due to their omnipresence, the rules of the city have become invisible. This is likely their strength; we come to live with the laws without needing to be reminded of them all the time. And a society is at peace when those rules are internalised (Habermas 1996).

The reduction of interpersonal violence in Europe was accompanied by a long process of building awareness that rejected violence. In Paris, the Feast of St. John used to end with dropping cats into the flames of a bonfire; then, awareness transformed into rules and laws, and today animal protection societies have even been able to stop the use of animals for drug testing and medical studies. This historic process of heightening awareness and personal containment, described by Elias (1987), has controlled violence. Executions and corporal punishment, once on display, were completely prohibited or, at least, hidden from view (Foucault 2000). Awareness prevailed and created the social rule, and then came law.

Societal life requires agreements and regulations, but they are needed even more in the city due to its population density (Wirth 1938). In the city, people are crowded together, and both the human closeness and the response to urban anonymity needed to protect individual privacy require more rules for coexistence. Personal ties and trust are no longer enough to drive social relations; general and abstract laws are required.

Urban rules support coexistence because they pacify society by making the behaviour of others predictable (North, Wallis, and Weingast 2009). Rules regulate the reproductive mechanisms of society, through both alliances and kinship, as well as through the mechanisms for the creation and distribution of wealth and for resolution of conflicts. Society can reproduce itself due to material conditions, but the trust that arises from the fact that each person knows what to expect from another, in a way that is understandable and reliable, is also needed.

Stabilisation of the expectations of society comes about by imposing law or formal institutionality (Luhmann 2005), and the same goal can also be achieved through rules that are neither written, nor systematised, which I call informal institutionality. By stabilising expectations, people know what they should do and what they can expect from the actions of others. Informal institutionality therefore consists of the norms shared by a population that were created, exist, and must be complied with in that territory, separate from what is established by formal law (Briceño-León 2016a).

Although informal institutionality is outside the dominant, formal legal system, this does not necessarily mean that it is against or in competition with the law, but rather that it can perform functions that complement, substitute, or accommodate formal institutions (Helmke and Levitsky 2006). However, informal institutionality can also be controlled by criminal gangs or illegal armed groups (guerrillas or paramilitary groups), which replace the state and control territories to extract profit.

The strength of informal institutionality lies in the recognition that it is carried out by the stakeholders involved and their capacity to effectively guide the behaviour of the people, both through their voluntary willingness to comply as well as their expectation that not doing so would lead to the application of penalties for offenders (O’Donnell 2006).

Both types of institutionality can have the same social force in steering behaviour towards that which is prescribed—the right thing to do—as well as to accept any negative consequences—the punishment—that a prohibited behaviour may have (Durkheim 1996; Messner, Rosenfeld, and Karstedt 2013). What I highlight in this chapter is that informal institutionality exists and is very strong in Latin American cities. It fills an incapacity or void in formal law to regulate urban development and coexistence. Informal institutionality, I argue, should not necessarily be understood as a negative reality, nor as a shortcoming of society that must be resisted; rather, to the contrary, it can represent a contribution to the building of safe cities.

The urban structure of the city and institutionality

Urban growth has led our research team to identify five zones in Venezuelan cities:

1.Historic City Centre

2.Private Formal Residential Development

3.Public Formal Residential Development

4.Consolidated Informal Urban Development

5.Unconsolidated Informal Urban Development.

These zones differ from each other in their physical aspect, their built environment, and their social and normative characterisation. Analysis of urban zones has a long-standing tradition in sociology, from analysis of the internal organisation of American cities conducted using concentric theories (Burgess and Locke 1953) to the theses on urban structures for Latin America (Yujnovsky 1971; Camacho 2016). Our five zones draw on this literature and are a tool for data analysis.

Formal institutionality is predominant in three of the five zones. The first is the Historic City Centre, which has a varied urban volume and social diversity. There, we find houses up against each other, with colonial-style interior patios, beside buildings ranging in height from medium to very tall. The checkerboard layout established by order of King Phillip II of Spain continues to be in use.

In the twentieth century, under the influence of city garden concepts from England and following the advent of the private automobile and the design of North American suburbs, extensive areas of Private Formal Residential Developments, the second zone, arose and attracted the urban elite (Amato 1970). This type of residential development was the result of activity by private property developers and the existing urban regulations were adapted to these designs. Over time, and with greater demand for housing, these zones were densified and replaced by multi-family buildings. Consequently, the elite moved to zones further away.

Heavy demand for housing from low-income residents, who could not access the private formal market, led governments to establish a third zone: Public Formal Residential Development. In these areas, mass-market housing production policies were developed to serve this population. The programmes were of different scopes, depending on the resources available from governments, and had very disparate impacts on the cities. Unlike the first two zones, where there is private ownership of houses and land, ownership is very heterogeneous in Public Formal Residential Development zones. In some cases, the government granted ownership of the dwelling to residents, but not the land on which it sits. In other cases, it granted ownership of both; in still other cases, it granted ownership of neither.

The other two urban zones consist of informal urban development. The difference between the two is that one has consolidated both physically and in its social cohesion and rules, even if in an informal manner, while the other is still in the process of consolidation. Both were built spontaneously by people, who invaded public or private land and developed it as they wished. Very often this urban development process occurred without planning or design, except during the construction itself and according to needs, technological capacity to adapt to the land, and the agreements or disputes that occurred during the occupation and densification process (Ziccardi 2008; Bolívar 2011, 2016).

In the Consolidated Informal Urban Development zones, inhabitants have known each other for decades and now the second and third generations of families live there. The grandparents worked the land; the parents were working class; and their children or grandchildren focused on education, with many graduating from university. In the newer generations, some are employed, but others are unable to enter the workforce and are caught in networks of crime and drugs. Although there are exceptions, it is most common for them to own their homes but not have ownership of the land.

Because they were established more recently or due to their financial or physical difficulties, other urban zones are known as Unconsolidated Informal Urban Developments. In many cities, the creation of new informal urban development zones has been a continuous process of expansion, whether a result of the arrival of new inhabitants or of the housing requirements of the second or third generations (Cravino 2006). In these zones, there is not only a lack of consolidation of the urban structure and housing, but also of social relationships, and neither the rules nor the power or leadership of any group has been established. The manner of social integration is different because, unlike the consolidated zones, inhabitants do not come from a rural area or small cities, but rather were born and have grown up in big cities.

In these zones of urban development and informal institutionality, there is a regulatory duality that results from the time they were created as their existence was not recognised by the state (Calderón 2005). For decades, the plans for Venezuelan cities showed these vast areas in green, like parks. Since the zones were illegal, without the necessary permits, they could not be recognised as residential and business areas. Frequently they did not comply with formal urban development requirements, such as the required slope of the land, location of buildings on the land, distance from other buildings, or compliance with sanitary regulations for drinking and waste-water. Since they did not meet the necessary standards, they were ignored and considered vacant forest land or city parks.

This caused ambivalence in the city, where the political authorities do not recognise informal urban development; they do not create rules and laws, nor adapt existing ones, to regulate the urban and social existence of these areas; nor are they capable of imposing the rules of the formal city on those urban areas (Briceño-León 2016b).

In these two excluded zones of the city, violence and insecurity prevail. For example, 84 per cent of homicides occur in these areas (INE 2010). The victims are poor people who die at the hands of other poor people (Pedrazzini 2005). The population of these areas therefore faces double exclusion: first, physical exclusion due to the lack of integration into the city and, second, exclusion from regulatory integration and security.

Informal institutionality to contain violence

In Venezuela’s barrios, both consolidated and unconsolidated, people live in the shadow of the law. Social life in those parts of the city is not outside the law, nor is it in harmony with or governed by the rule of law. Certainly, barrios develop “outside” the law: they are built on land that has been invaded without the construction permits required in other urban zones. Houses were built on spaces that are not authorised for development and on land belonging to others (the property of the state or private owners). As a result, the people of the barrios have survived for many years under the threat of eviction and with the fear of losing the money, effort, and dreams they invested in them (Bolívar 2016).

As such, the institutionality that is experienced in the barrios does not fully conform to the legislation of the country, nor to the urban norms of the city. But the barrios are there, they exist and prosper, and, in many cases, the authorities have supported their residents in building an urban life by giving them services, water, power, sewers, schools, and medical centres. Some national or local authorities have even granted residents ownership of the land. But they are still not integrated into the city in a spatial sense, nor in a regulatory and legal manner, and this is a source of conflict and violence. Exclusion is not only spatial and territorial, it is also social and regulatory. However, it cannot be affirmed that these are spaces where anomie rules. There is not a regulatory or institutional void, but what does exist is a distinct form of regulation, an informal institutionality that is in the shadow of the law.

Informal institutionality is the practical response of urban communities to a lack of formal law that is applicable to and works in their social contexts: it ensures individual rights and constitutes a guide for preventing or resolving conflicts outside the law. Informal institutionality is not necessarily illegal or criminal, although it can be. We can say that it is paralegal because it is outside, but in the shadow of formal law. It intends to copy both the law and the formal practices of the rule of law. This paralegalism can be illegal because it realises acts that are not considered or proscribed by law; however, it is not criminal, as it identifies with the ultimate purposes of formal institutionality, but not with its methods. In this sense, paralegalism acts as a substitute for law (Helmke and Levitsky 2006). Since there is an alignment with the purposes of regulating social relationships and stabilising expectations in order to reduce conflicts, I maintain that it operates in the shadow of formal law. For this very reason, local authorities often grant recognition to the rites of informal institutionality (acting, for example, as witnesses to the signing of lease or sale agreements) despite their illegality.

However, there is no state that enforces compliance with informal institutionality, nor punishment for those who do not comply. It is force of habit: the autoritas of respected and respectable individuals, the general will of reason, as Locke (2007) said. When this consensus and reason fail, and in the absence of established mechanisms for punishment, it opens the way for citizens to take justice into their own hands or for criminal gangs to take retribution. These credible threats, often more credible that those of the punishment provided by the formal criminal justice system, combined with consensus and habit, lend recognition and functionality to informal institutionality.

In our research, we found many experiences that demonstrate the functioning of informal institutionality and its contribution to the containment of violence. Below are some examples.

The invisible borders of crime

“I can’t accompany you any further”, the community leader exclaimed suddenly, guiding her down the steps to the ravine and stopping abruptly on the step. “And why are you stopping here, Juan?”, replied the researcher in surprise. There was nothing obvious to prevent him from continuing, no sign, no tree, not even a landing on the steps. “Because from here on I cannot guarantee your safety”, he responded regretfully, but confident in what he was saying. Then he explained that this area was controlled by “another” gang with whom they had no agreements.

(A mountainous barrio to the east of Caracas, 2014)

Cities have invisible borders. People are afraid of being victims of crime. The national survey used to inform this chapter took place in 2015. Residents of 1,500 homes, selected at random, across the country were interviewed. Its results demonstrated that the quality of life had diminished since the 2014 survey, and showed how living with violence and insecurity has affected the population. For example, 58 per cent were afraid of being victims of crime in their neighbourhood (Briceño-León and Camardiel 2016a). These invisible borders mark the space in which criminal actors deal in illegal goods and protection; they are territories gained over years of bloody battles and building bonds of solidarity or fear with local inhabitants and of respect by others. In the impoverished barrios, these borders are accentuated, given that the presence of law enforcement and the rule of law is scarce or non-existent, so security and order must be provided by the inhabitants themselves (Caldeira 2000).

The borders in the barrios, like those between countries, are a source of permanent tension, as any incursion could be considered a threat and violently repelled. It is a case of protecting both the market and honour. In one of the research zones, aggression could take the symbolic form of some gang members shining spotlights on the territories of another gang. In the darkness of night, the interjection of a ray of light from a spotlight told the rival gang: “I’m watching you, I can reach you with my light and also with my bullets”. And because of that, the response from the other gang often came in the form of bullets. And that is how the fight began.

The response from community organisations was to develop agreements to first respect, and then overcome, these invisible borders. It was not an easy task. There was dialogue with gang members and with individuals from different areas, in an attempt to reinstate the right of free movement, but it was not easy as there were misgivings on both sides. The first thing that was done was to formalise the borders, so that no one was in doubt about where they could and could not go. Then some rules of mutual respect were established; one of them was as trivial as not using spotlights to threaten each other. Next, residents established a means by which another area could be notified (and, in some way, request permission) to inform the opposing side that some people were going to pass through their territory, but that they were doing so in peace, not aggression, to reduce the risk of conflict (Zubillaga, Llorens, and Souto 2016).

Subsequently, in some barrios of Caracas and Ciudad Guayana, residents were able to establish a “common zone”—a territory not controlled by any of the gangs—where community activities could take place. One of the activities conducted by female teachers was to organise games for children on both sides of the border, making them equals through play. The intent was to create a situation in which they could grow up as friends rather than rivals, so that empathy could arise in daily life, and they could see each other as children just like themselves and not a threatening enemy (Pernalete 2016). It is a strategy similar to the school lunch programme in India, in which members of different castes were seated at the table, including, especially, “Dalits” or untouchables, to create a space of equality and freedom (Sen 2009).

A third step, more difficult and with achievements that are still precarious, is to tear down the borders and build a space of citizenship in which all have full rights to the city.

Cease-fire agreements

“No more dead!” the mother said, shrouded in grief; but, how can this be achieved? The aim of the mother was to break the chain of retaliation and revenge that had led to the violence, rather than reproducing it and making it persistent.

(Near central Caracas, 2014)

Cease-fire agreements have been used in various places in Latin America, most notably perhaps in the case of El Salvador, where an agreement was reached between the two major organised gangs and the national government in an effort to reduce homicides. In Venezuela, the so-called “peace zones” were established—although not in an openly official manner as an explanatory plan was never published—in which the national government gave control of a territory to criminal gangs so that they would exercise control and reduce violence. In order to do so, the state restricted or prevented the access of law enforcement personnel to these zones. This, in effect, saw the state give up its sovereignty over the zones, as had been the case in the demilitarised zone established between the Colombian government and the Revolutionary Armed Forces of Colombia (FARC) guerrillas in the area of San Vicente del Caguán, from 1999 to 2002, before it was retaken by the army.

The aim of the cease-fire agreements that were in operation in an area of the Libertador municipality of Caracas was to minimise the damage caused by armed violence. The intention was not for youth to give up gangs or crime, much less punish offenders or killers, but rather to simply establish a truce, prevent the violence, and, through this action, to rebuild a social bond and bring peace. The interesting thing about this proposal, which followed years of social work by community groups from universities and the Catholic church, was that it intended to change the situation through words and a social norm—an agreement—rather than dissuasion or repression. In this sense, it meant building an institutional framework based on some rules of the game being backed by the moral authority of the mothers of the youth, along with representatives of the church.

These rules of the game of informal institutionality had two significant traits. First, they did not represent a threat to other forms of crime because although the agreement was not authorised, it did not seek to eliminate other crimes such as the sale of drugs. The agreement was confined to resolving conflicts by methods other than weapons and gunshots. But, second, the agreement meant that if the youths did not comply with it, offenders would be reported to the police. In other words, there was a latent threat of formal law enforcement. These ambiguities are what allow me to assert that informal institutionality acts outside, but in the shadow of, formal law.

Social pacts rewritten

“And how am I going to live if it is prohibited, if I raised all of my children and even put them through university by selling beer in the barrio?”, Dominga protested in front of the small window of her house, from which, for decades, every afternoon and evening she had offered “very cold” beer to her customers. Venezuelan law prohibits the sale of beer in places that do not have the required permits. To receive a permit, the sales cannot be conducted from a house or occur close to a school. “But here everything is close to everything”, she said.

(A barrio in Caracas, 2014)

The sale of alcohol in the barrios is a permanent source of conflict as it makes friendly coexistence difficult. In these spaces, drinkers hang around in the street talking and listening to music until late at night, while others have to work early the next day. If bottles break, glass is left in the streets creating a hazard for children playing there the next day; and, since there are no toilets available and beer is a diuretic, consumers end up urinating anywhere, resulting in foul odours in the neighbourhood. Finally, between complaints and celebrations, the sale of beer sometimes ends in brawls and people being injured, some seriously.

The response was to apply formal law and prohibit the sale of beer. This was tried in a barrio of the Chacao municipality of Caracas. To prevent sales, the beer distributors’ trucks were prohibited from entering the area and the police stationed themselves at the barrio’s only entrance. Nonetheless, the measure turned out to be futile. Bottles entered in a variety of ways: via the ravine that crossed the barrio, through holes that were opened in the walls of workshops adjacent to the area, and even hidden in baby carriages. The conflicts between neighbours persisted and beer was still sold at night. The only thing that was achieved was an increase in its price.

It was then that a different process began. Officials from City Hall and local law enforcement talked to sellers, distributors, non-drinking neighbours, and consumers of alcohol, and they began to establish agreements that would reconcile the parties in conflict (Carrillo Peraza and Espinoza 2008). In the end, they negotiated some agreements that were voted on and signed by all; it was their “social contract”. The rules agreed upon established limited hours for beer sales and restricted the volume of music; beer could not be sold in bottles, but must be served in plastic cups; and each family that wanted to sell beer had to allow their customers to use the toilet in their house.

Those who agreed to the established rules could continue the illegal-legal sale of beer. Failure to comply with the rules could lead to penalties and the tolerance on the part of formal institutions toward its illegality could be lost. A new institutionality had been constructed, this time written, approved, and signed by the people; a social contract by consensus was created, but it was “paralegal”.

From a legal point of view, it was a bold measure because something that had been prohibited was being endorsed. Like the entire barrio itself, with its houses, streets, families, shops, and schools, something illegal was being tolerated.

The grey zone of institutionality

“I only rent the second floor of my house under a written, signed contract”, said Julián emphatically at the start of the meeting. His house had been built years ago halfway up the steps going up the mountain. It wasn’t the most coveted or expensive area, like the lower part, next to the main street along the public transport route; nor was it the cheapest, like the part high up, more recently built and more difficult to reach. Because of this, his rent was modest.

(A mountainous barrio near Caracas, 2014)

The problem Julián faces is that under Venezuelan law none of the dwellings located in informal urban development zones, consolidated or not, can be rented; this business is explicitly prohibited and punishable. Nonetheless, the largest and most active rental market that exists in Venezuela is in the informal barrios, and half of those dwellings have been rented under written contracts that in themselves represent confession of a crime by the parties to the contract (Nikken et al. 2006; Briceño-León 2008).

In an article on the state and citizenship, O’Donnell (1993) imagines the maps of countries painted in three colours: blue areas where there is presence of the state with homogeneous territorial and functional control, for example, Norway; green areas where there is territorial but not functional control, as occurs in the southern United States; and brown areas where there is scant territorial and functional control, as occurs in large areas of the countries of Latin America. His metaphor was highly successful, as it showed the limitations of formal institutionality. However, in his analysis, O’Donnell omitted the substitute forms of rule of law that arise in Latin America. In those areas, there is not a regulatory void or transgression, but some social regulations that may be somewhat fragile in their legality yet efficient at enabling the coexistence of citizens in addition to preventing conflicts and reducing violence. In this sense, the rentals in the barrios demonstrate very clearly that there is an informal alternative of substituting and reproducing the law, with some written words that formalise agreements and establish rules and penalties to which the parties are subject. The grey zone of informality is that informal institutionality: it is the area of social life regulated by rules that cannot be said to be either legal or criminal.

Measures taken by governments, both on the left and on the right, have resulted in repression of the informal mechanisms for coexistence and conflict resolution. What has been achieved, as occurred with the rentals, is greater social exclusion, placing individuals outside the law, blocking their access to the formal mechanisms for conflict resolution and to justice, and pushing them toward violence and justice at their own hands. Informal institutionality fills this void and converts it into a force to contain violence, but this can only be achieved—to a certain extent—when there is an absence of the legitimate force of the state to enforce compliance with agreements or to punish offenders. If this is not the case, the “natural” force—as described by Hobbes (1980)—of other citizens or criminals that substitute for the state appears to enforce compliance with agreements and punish offenders.

The symbolic rescue of public space

“The basketball court was built to prevent boys from going down the wrong path, and now it turns out that the punks are selling drugs there”, said the teacher from the barrio. The measure, she recalls, had the support of everyone: “the barrio needed a place where youths could practise sports rather than hanging out on the corner.” However, the physical construction was not accompanied by social control of the territory, so the gangs took over the place and the police could not or would not do anything to get rid of them.

(A barrio in Ciudad Guayana, 2014)

The public space of cities is a privileged place to exercise freedom and equality: the privileged place of social interaction, it is the territory of the market, celebration, and politics. But in situations of violence, when security is reduced and fear increased, people abandon public spaces and hide away in their houses. In our 2015 survey using national random sampling, we found that 76 per cent of people were afraid of being a victim in the centre of their city and 82 per cent were afraid of being attacked or robbed on public transport. As a result, 63 per cent had restricted the places and times that they did their shopping and 65 per cent had restricted their sports or leisure activities (Briceño-León and Camardiel 2016a). And if citizens abandon public space out of fear, the opportunity for cultural and material exchange, or of goods and ideas, is lost among the different and unequal citizens who make up the essence of the city.

In the contemporary city, the street and the square have lost their strength as public spaces, and the substitutes that imitate them in the private spaces of shopping centres have gained force. Contemporary urban residents may escape to shopping centres for many reasons: the centrality of the space, an optimised artificial climate, and, without a doubt, among these is seeking the safety that the street no longer offers.

Rescue of the public space is also a rescue of freedom and its value as being common and shared. In our research, different experiences were studied in which communities were competing with criminal gangs for control of public space in order to make them safer and more inclusive. Since they could not do so through force—the responsibility of the state and its police—they did it through community activities. In one barrio in Ciudad Guayana, mothers formed an alliance with the Catholic parish and decided to celebrate religious holidays in the street. The “processions”, a stroll with prayers and religious figures that had been suspended in the past or had avoided certain places, were now designed to occupy these territories. In a barrio in Caracas, at a place where the street widens and transforms into a kind of square, drug sellers had set up their sales operations. Since families steered clear of that area, the solitude of the space made business easier for the dealers. So, the community, with support from City Hall, established a programme of cultural activities and youth parties to actively occupy the space and take it back for the community, displacing the delinquents.

In another barrio in Caracas, established on a thin strip of land located between an area of middle-class houses and the highway that crosses the city from east to west, two rows of houses formed a street. Over the years, what started out as substandard housing, transformed into two- and three- storey buildings. Some people were even able to purchase a modest auto-mobile, for work or family enjoyment. But insecurity and conflicts emerged, so the community organisation decided to intervene and restrict access to the area. The residents’ association conducted a census of owners and tenants and gave identification cards to each resident of the area. Then, they decided to assign a parking space on the street for each family’s car and thus prevent abuse and conflict over parking spots (Hernández and Chacón 2016). All these measures reinforced security and improved co existence in the barrio. Paradoxically, however, although all the measures were outside formal law, they represent a valuable paralegal culture (Pérez Perdomo and Friedman 2003). This is the kind of innovative contribution that informal institutionality can provide to make a city safer, and it cannot be overlooked.

Gender as a tool of peace

In Venezuela, over 90 per cent of homicide victims are men, similar to statistics for the rest of Latin America (INE 2010; UNODC 2013; Muggah and Alvarado 2016). An even greater percentage of murderers are also men. Gender studies have asked why there is such a huge gap when there are equal numbers of men and women. The response given by the men in the study is very enlightening: It is because women are cowards. They run away. They don’t confront. For those who understand violence to be a man’s affair, avoiding violence is a woman’s affair. In the poor and violent areas of the city there exists a macho culture that privileges a violent masculinity: men who flee violence are scorned; they are treated as women.

Use of the gender perspective in many studies has focused on showing, reasonably, the processes involved in the misogynistic victimisation of women, and the strategy of domination involved in this behaviour. However, in this investigation, we encounter an additional dimension and it is that this supposed vulnerability and cowardice of women can become an important tool for peace. The underestimated female values and behaviours are an affront to violent male culture, and represent a countercultural proposal (Roszak 1969) that subverts the dominant patterns by ignoring or confronting them.

Female avoidance as a counterculture of violence

The culture of avoidance by women represents an opportunity for peace and the containment of violence. The scornful attitude of men toward women for avoiding violence constitutes a powerful cultural force that can be used to prevent violence and educate men on safe behaviour.

This is not easy to achieve. On one hand, men, especially young men, consider that adopting behaviour to avoid violence can cause them problems, since they would be assuming a “wrong” gender role, putting the construct of their identity in jeopardy. On the other hand, as with any cultural gender trait, dominant values and behaviours are shared by both sexes. Thus, machismo is not only a male thing, and some mothers push their boys to fight, so that they can learn “how to be men”; and some wives demand that their husbands involve themselves in violence and act “as men”.

However, the high rate of victimisation has led many mothers, wives, and girlfriends to become agents of peace. Mothers have played the most relevant role because they have the moral strength to stop the violence. In that gender pattern, the mother is outside the violence. She cannot be considered a threat, nor can her actions be considered aggressive. Men can inflict violence on other men, but, in their supposed masculine superiority, they are cautious about inflicting it on women, and much less on mothers. That is the reason why programmes such as the “peace-building mothers” of Ciudad Guayana can have more impact and operate more audaciously than if they were carried out by men (Pernalete 2016). The weakness of the gender role attributed to women has been turned into a strength.

The advantages of female teachers

As an institution, the school represents the most effective mechanism for socialisation and social control. But in the barrios of Venezuela, schools have been besieged by violence. Youths have brought violence from the streets into the classrooms. Conflicts that once were settled with fights after school became confrontations inside the school; and when teachers attempted to discipline these youths, they threatened their teachers with their violent relatives or with weapons they were able to smuggle into class.

In that context, gender roles began to take on relevance. Male teachers represented a challenge to violent youths because they are seen first as a man who was punishing them, not as a teacher. This led the youths to understand they had to “act as a man” when interacting with another man. The male teachers knew this and stopped punishing their students; they were cautious and afraid of violence. Female teachers had to come to their aid. A female teacher, as a woman, had more protection than a man: they could talk and reprimand with a mother’s voice.

In the face of the dominant values in the streets of the barrio, of violence, manliness, and the arbitrary power of weapons, female teachers represent the greatest resistance. The school represents a world governed by norms and subject to authority: it represents a counterculture of formal and informal institutionality combined, and female teachers become the preferred and dominant instruments of change.

The symbolic force of religion

Many of the female teachers are nuns who are part of Catholic educational communities. In this case, the role of woman and teacher is made even more powerful by the religious factor. These women are not mothers, nor can they be since they have taken Catholic vows of celibacy, but in common parlance they are called “mothers”. “Mother Teresa”, for example, is a nun who runs a school in a barrio. She has her role as a common educator, but represents much more because, as a woman, she has the maternal force that identifies her with the mother of Jesus Christ who weeps for her murdered son. All of this symbolism makes the nuns a cultural force for the containment of violence.

In the case of the Evangelists, the religious factor plays a different role as it affords them a gender significance that is different from peaceful or non-violent behaviour. As mentioned above, among the young men of the barrio, the attitude of avoiding violence and confrontation is associated with traditional female behaviour, as is the habit of not consuming alcohol. However, when the reason behind this behaviour takes on a religious tone, it loses its female nature. Since the majority of Evangelical preachers are men, their preaching of non-violence and social decency is no longer considered feminine, but religious. As a result, for many pacifist young men who want to avoid getting involved in violence, or others who want to get out of the criminal world in the barrios or in prisons, converting to the Evangelist religion is a way of protecting themselves while continuing to be considered “men”.


Cities of the future will be those that prove themselves to be safe and inclusive (Sassen 2011; de Boer 2015). In order to achieve this, significant productive transformations are required to ensure their sustainability, the cultural strength of their citizens that allows those who are different and unequal to live together, and the realistic and shared rules of the game that make the above possible. This study shows that informal institutionality can be a very powerful tool for establishing this culture of citizenship and the rules of coexistence as an effective means of containing violence and developing the well-being of cities.

In programmes for improving the quality of life in cities, both in the physical aspect as well as in respect of safety, public policies have dismissed informal institutionality. Sometimes they have ignored it and left it out; at other times, they have persecuted and punished it. That is what was done for decades with the barrios and informal entrepreneurs (De Soto 1987). Their existence was not accepted; they were excluded, persecuted, and threatened with eviction. The dominant social representation was that a better city was only possible if the barrios disappeared.

Then, that changed and the barrios were recognised by society, governments, and international organisations. It was accepted that the poor had to have a place to live in the city and, in view of the lack of formal and legal opportunities to find one, millions of poor people had developed a solution themselves (Bolívar 2011). Acknowledging them was not only accepting that they existed as a social group and constructed space, but also recognising that there was another way of making a city.

Something similar has occurred with the regulatory systems that govern the barrios. The absence of formal laws that take into account their urban and social singularity, and the inability to enforce formal laws and guarantee the rule of law in these areas, has excluded them from regulations. In the absence of formal laws that bring order to their coexistence, the inhabitants of the barrios have created their own rules, their own informal institutionality.

Returning to the five urban zones that can be identified in the cities, in the two most populous zones—Consolidated Informal Urban Development and Unconsolidated Informal Urban Development—informal institutionality is the true regulator of social processes. And, its efficiency is greater when cooperation and complementarity have been possible between formal and informal regulations. When the state retreats and gives up its capacity for real control in some territories, then informal institutionality may fall into the hands of criminal groups and be perverse. However, in cooperation with the state and local authorities, it can be a valuable tool for building coexistence and citizenship.

This regulatory reality must not just be accepted as a rarity, an extravagance of legal pluralism. On the contrary, it is about promoting the diversity and wealth provided by informal institutionality to make it more feasible to establish the rule of law and for citizens to exercise their rights (Briceño-León and Camardiel 2016b). The study conducted allows us to conclude that informal institutionality can be a powerful tool for increasing inclusion and security in the city.

To stop the violence and build urban security in the Global South, it is necessary to walk in new ways. It is necessary to recognise that formal institutionality, as it has been applied, does not offer solutions to an important part of the urban population. Nor have the “hard-line” mano dura policies improved the situation. It is necessary to innovate and allow a new social pact to be built between legal formalism, anomie and the violation of human rights that cannot be exclusively formal, nor exclusively informal. Institutionality that can respond to social and urban diversity must be a combination, a mixture, a mestizo response, as these societies are mestizos. This new institutional framework should allow for the contributions that the communities have made with their informal regulations, accept the multiple ways of regulating daily life that these societies have developed, and allow the existence of multiple mechanisms to prevent and resolve conflicts. Safe and inclusive cities, where the unequal and different coexist, require a lot of awareness, some rules, and little force. This is how they will become cities for citizens.


Amato, Peter W. (1970). Papel de la élite y patrones de asentamiento en la ciudad LatinoAmericana [The role of the elite and settlement patterns in the Latin American city]. Revista de la Sociedad InterAmericana de Planificación [Journal of the Inter-American Planning Society] 4(13–14): 22–34.

Bolívar, Teolinda (2011). Desde adentro: viviendo la construcción de las ciudades con su gente [From within: People’s experience during the construction of cities]. Quito: OLACCHI/Municipio Metropolitano de Quito.

Bolívar, Teolinda (2016). Los barrios en las ciudades Venezolanas [The neighborhoods in Venezuelan cities]. In Roberto Briceño-León (Ed.), Ciudades de vida y muerte. La cuidad y el pacto social para la contención de la violencia [Cities of life and death: The city and the social pact for the containment of violence] (59–86). Caracas: Editorial Alfa.

Briceño-León, Roberto (2008). Libertad para alquilar. El mercado informal de vivienda en Caracas [Freedom to rent: The informal housing market in Caracas] Territorios. Revista de Estudios Urbano Regionales [Territories: Journal of Regional Urban Studies] (18–19): 103–127.

Briceño-León, Roberto (2016a). La teoria sociológica de la institucionalidad y el pacto social [The sociological theory of institutionality and the social pact]. In Roberto Briceño-León (Ed.), Ciudades de vida y muerte. La ciudad y el pacto social para la contención de la violencia [Cities of life and death: The city and the social pact for the containment of violence] (117–139). Caracas: Editorial Alfa.

Briceño-León, Roberto (2016b). Las ciudades: escenario o causa de la violencia? [The cities: Scene or cause of violence?]. In Roberto Briceño-León (Ed.), Ciudades de vida y muerte. La ciudad y el pacto social para la contención de la violencia [Cities of life and death: The city and the social pact for the containment of violence](15–35). Caracas: Editorial Alfa.

Briceño-León, Roberto, and Alberto Camardiel (2016a). La inseguridad como pérdida de la libertad y del derecho a la cuidad [Insecurity and the loss of liberty and rights in the city]. In Anitza Freitez (Ed.), Venezuela: vivir a medias [Venezuela: half life] (55–70). Caracas: Universidad Católica Andres Bello.

Briceño-León, Roberto, and Alberto Camardiel (2016b). Ciudades de ciudadanos [Cities of citizens]. In Roberto Briceño-León (Ed.), Ciudades de vida y muerte. La ciudad y el pacto social para la contención de la violencia [Cities of life and death: The city and the social pact for the containment of violence] (373–393). Caracas: Editorial Alfa.

Burgess, Ernest W., and Harvey J. Locke (1953). The Family: From institution to companionship. New York: American Book Company.

Caldeira, Teresa P. (2000). Cidade de muros. Crime, segregação e cidadania em São Paulo [City of walls: Crime, segregation and citizenship in São Paulo]. São Paulo: EDUSP.

Calderón, Julio (2005). La ciudad ilegal, Lima del Siglo XX [The illegal city: Twentieth century Lima]. Lima: Universidad Nacional Mayor de San Marcos.

Camacho, Oscar Olinto (2016). Las ciudades Venezolanas y la exclusión social [Venezuelan cities and social exclusión]. In Roberto Briceño-León (Ed.), Ciudades de vida y muerte. La ciudad y el pacto social para la contención de la violencia [Cities of life and death: The city and the social pact for the containment of violence] (39–57). Caracas: Editorial Alfa.

Carrillo Peraza, M.R., and L.M. Espinoza (2008). La Victoria del Ciudadano. Un modelo de acceso a la justicia a nivel local [The victory of the citizen: A model of access to justice at the local level]. Caracas: Alcaldía de Chacao.

Cravino, María Cristina (2006). Las villas de la ciudad. Mercado e informalidad urbana [The villas of the city: Market and urban informality]. Buenos Aires: Universidad Nacional General Sarmiento.

de Boer, John (2015). Resilience and the fragile city. Stability: International Journal of Security and Development 4(1): 17.

De Soto, Hernando (1987). El Otro Sendero. La revolución informal [The Other Path: The informal revolution]. Bogotá: Editorial Oveja Negra.

Durkheim, Émile (1996). Sociologie et philosophie [Sociology and philosophy]. Paris: Quadrige/Press Universitaires de France.

Elias, Norbert (1987). El proceso de la civilización. Investigaciones sociogenéticas y psicogenéticas [The process of civilization: Sociogenetic and psychogenetic investigations]. Mexico City: Fondo de Cultura Económica.

Foucault, Michel (2000). Surveiller et punir: naissance de la prison [Discipline and punish: The birth of the prison]. Paris: Gallimard.

Habermas, Jürgen (1996). Between facts and norms. Contribution to a discourse theory of law and democracy. Cambridge, MA: The MIT Press.

Helmke, Gretchen, and Steven Levitsky (2006). Introduction. In Gretchen Helmke and Steven Levitsky (Eds.), Informal institutions and democracy: Lessons from Latin America (1–30). Baltimore, MD: Johns Hopkins University Press.

Hernández, M., and A. Chacón (2016). Dinámicas municipales contra la violencia: el caso Chacao [Municipal dynamics against violence: The Chacao case]. In Roberto Briceño-León (Ed.), Ciudades de vida y muerte. La ciudad y el pacto social para la contención de la violencia [Cities of life and death: The city and the social pact for the containment of violence] (255–281). Caracas: Editorial Alfa.

Hobbes, Thomas. (1980). Leviathán: o la materia, forma y poder de una república, eclesiástica y civil [Leviathan or the matter, form and power of an ecclestical and civil republic]. Mexico: Fondo de Cultura Económica.

INE: Instituto Nacional de Estadística [National Institute of Statistics] (2010). Encuesta Nacional de Victimización y Percepción de Seguridad Ciudadana 2009 [National Survey of Victimization and Perceptions of Citizen Security 2009]. Vice-Presidencia de la República. Caracas: Author.

Locke, John (2007). La ley de la naturaleza [The law of nature]. Madrid: Tecnos.

Luhmann, Niklas (2005). El derecho de la sociedad [The law of society]. Barcelona: Herder.

Messner, Steven F., Richard Rosenfeld, and Susanne Karstedt (2013). Social institutions and crime. In Francis T. Cullen and Pamela Wilcox (Eds.), Criminological theory (405–423). Oxford: Oxford University Press.

Muggah, Robert, and Nathalie Alvarado (2016). Los homicidios en América Latina: Seamos ambiciosos [Homicides in Latin America: Let’s be ambitious]. El País, 29 September. http://internacional.elpais.com/internacional/2016/09/29/america/1475159591_036416.html.

Nikken, Pedro, Rogelio Pérez Perdomo, Marcos Vilera, and Elisabeth Fassano (2006). Estudio de una Mediación Oficial en Conflictos relacionados con vivienda en los Barrios de Caracas [Study of an Official Conflict Mediation related to housing in the Barrios of Caracas]. In Cátedra de Jurídica [Chair of the Sociology of Law], Sociología Jurídica [Legal Sociology] (325–343). Caracas: Universidad Central de Venezuela.

North, Douglass C., John Joseph Wallis, and Barry R. Weingast (2009). Violence and social orders: A conceptual framework for interpreting recorded human history. Cambridge: Cambridge University Press.

O’Donnell, Guillermo (1993). Estado, democratización y ciudadanía [State, democratization and citizenship]. Nueva Sociedad [New Society] 128: 62–87.

O’Donnell, Guillermo (2006). On informal institutions, once again. In Gretchen Helmke and Steven Levitsky (Eds.), Informal institutions and democracy. Lessons from Latin America (285–289). Baltimore, MD: Johns Hopkins University Press.

Pedrazzini, Yves (2005). La violence des villes [The violence of cities]. Paris: Enjeux Planète.

Pérez Perdomo, Rogelio, and Lawrence M. Friedman (2003). Legal culture in the age of globalization. In Lawrence M. Friedman and Rogelio Pérez Perdomo (Eds.), Legal culture in the age of globalization: Latin America and Latin Europe. Stanford, CA: Stanford University Press.

Pernalete, L. (2016). Mujeres de mano extendida: un muro de contención de la violencia en comunidades populares de ciudad Guayana [Women with an extended hand: A barrier to contain violence in disadvantaged communities of Guayana]. In Roberto Briceño-León (Ed.), Ciudades de vida y muerte. La ciudad y el pacto social para la contencion de la violencia [Cities of life and death: The city and the social pact for the containment of violence] (317–345). Caracas: Editorial Alfa.

Roszak, Theodore (1969). The making of a counter culture. Berkeley, CA: University of California Press.

Sassen, Saskia (2011). Cities in a world economy. Thousand Oaks, CA: Pine Forge Press.

Sen, Amartya (2009). The idea of justice. Cambridge, MA: The Belnap Press of Harvard University Press.

UNODC: United Nations Office on Drugs and Crime (2013). Global study on homicide. Trends, context, data. New York: Author.

Wirth, Louis (1938). Urbanism as a way of life. Americal Journal of Sociology 44(1): 1–24.

Yujnovsky, Oscar (1971). La estructura interna de la ciudad. El caso LatinoAmericano [The internal structure of the city: The Latin American case]. Buenos Aires: SIAP.

Ziccardi, Alicia (2008). Pobreza y exclusión social en las ciudades del siglo XXI [Poverty and social exclusion in 21st century cities]. In Alicia Ziccardi (Ed.), Procesos de urbanización de la pobreza y nuevas formas de exclusión social [Processes of urbanization of poverty and new forms of social exclusion] (9–33). Bogotá: Siglo del Hombre Editores CLACSO-CROP.

Zubillaga, V., M. Llorens, and J. Souto (2016). Una tregua es posible: la violencia y el pacto del cese al fuego entre mujeres y jovenes armados [A truce is possible: Violence and the cease-fire pact between women and armed youth]. In Roberto Briceño-León (Ed.), Ciudades de vida y muerte. La ciudad y el pacto social para la contencion de la violencia [Cities of life and death: The city and the social pact for the containment of violence] (225–253). Caracas: Editorial Alfa.

Part IV

South Asia

8Everyday violence in urban India

Is planning the driver or mitigator?

Darshini Mahadevia and Renu Desai


Urban development and planning, two important pillars of India’s economic growth and development pathway,1 were given importance in the ‘New Urban Agenda’ adopted at the Habitat III conference in 2016 to help achieve Sustainable Development Goal (SDG) 11—safe, inclusive, resilient, and sustainable cities. This chapter is located within the urban development dynamics of India, a country where, according to the 2011 census, 32 per cent of the population live in urban areas, and where the urban population is expected to increase to 40 per cent by 2030.2 India’s urban development journey over the last two decades has coincided with increasing marginalisation, exclusion, conflict, and everyday violence in the cities. This violence has gone unnoticed and unanalysed in the urban planning and policy-making world.

Cities are inherently conflictual spaces due to their heterogeneity and the resulting different interests operating in the city space (Wirth 1938, cited in Rodgers 2010). Moser and Horn (2011: 8) define conflict as “situations where individuals and groups have incongruent interests that are contradictory and potentially mutually exclusive”. They argue that conflicts tip over into physical violence when they remain unresolved. In scholarly and policy debates, the definition of violence has been expanded beyond physical acts that cause injury to include the exercise of power for specific gain in ways that lead to psychological harm, material deprivation, and symbolic disadvantage. The World Health Organization (WHO) defines violence as

the intentional use of physical force or power, threatened or actual, against oneself, another person, or against a group or community, that either results in or has a high likelihood of resulting in injury, death, psychological harm, maldevelopment, or deprivation.

(WHO 2002: 5, cited in Moser 2004: 4)

This is closer to Galtung’s (1969) definition of indirect violence or structural violence as a form of violence through which a social structure or institution harms people by preventing access to their basic needs. More recently, Rylko-Bauer and Farmer (2016: 1) refer to structural violence as “the violence of injustice and inequity” which is “embedded in ubiquitous social structures [and] normalised by stable institutions and regular experience”.

This chapter extends the notion of structural violence, building on both Galtung (1969) and Rylko-Bauer and Farmer (2016), to include the dominant paradigms and practices of urban planning and governance which negatively impact the well-being, agency, and capabilities of certain groups in the city on one hand, and create injustice and inequity, on the other. We examine land management, housing provision, water supply systems, and public transport systems within urban planning, and their related governance processes, as drivers of structural and everyday violence, and also explore instances where conflicts tip over into physical violence.

Urban planning, governance, and violence

Neoliberal globalisation in the Global South has “rendered [cities] vulnerable to new and growing forms of specifically urban risk” (Beall 2007: 5) that include inadequate infrastructure and housing, poor service delivery, and deficient and/or corrupt local government. It has put many low-income communities at risk of social and physical threats emanating from unregulated globalised industrialisation as well as climate change (Bayat and Biekart 2009: 822). However, the use of the term neoliberalism indicates a previously existing social-democratic welfare-distributing policy regime; whereas, in India, in spite of the welfare pronouncements by and accommodativeness of the state, there was no radical change in inequality and power relations after Independence (Frankel 2006).

The economic reforms of 1991 increased urban inequalities (Mahadevia and Sarkar 2012), which have worsened through inequitable urban planning (Mahadevia 2011, 2014). By urban planning, we mean land management tools and practices, including the provision of infrastructure, such as water, sanitation, and transport; supply of housing, including its location in the city; and land use planning through the instrument of the Master Plan. City branding as “global” or “world-class” has led to the criminalisation of the poor, and large-scale evictions and displacements (Baviskar 2003; Mahadevia 2011). As an instrument of planning, the Master Plan has also deemed many areas of the city to be illegal and, as a result, the Indian state has engineered the “elite capture” of urban spaces. It has also subverted pro-poor provisions of Master Plans (see Verma 2002; Roy 2009). The poor thus find spaces in the cities through “occupancy urbanism”, a gradual process of informal land occupation under political patronage (Benjamin 2008), while living in fear of the constant threat of eviction. By being deemed “illegal”, the informal urban living spaces created by the poor themselves are denied the provision of basic services. Using the conceptualisations of Galtung (1969) and Rylko-Bauer and Farmer (2016), the urban planning paradigm has institutionalised structural violence within India’s cities.

Governance is not just about what the state does in implementing plans, but also what the state does not do. A good example relates to the provision of basic services. The denial of essential urban services to the poor—because governments lack the political will, capability, or capacity to meet the needs of low-income communities—has also led to the emergence of non-state providers operating as mafias (Zaidi 1998; Erman and Eken 2004 writing about Istanbul). These informal sector entrepreneurs and middlemen—often with links to government functionaries and the police—use the vulnerability and unmet needs of the poor to manipulate them. As Hasan (1996: 56) notes,

the inability of the State to provide services, employment, and access to the corridors of power for the poor also leads to a failure to provide justice and protection. This failure introduces and sustains a system of violence, coercion, and extortion in urban areas.

Such processes emanating from urban governance are not only forms of structural violence, but also cause non-state actors to threaten or use physical violence. The fact that the means of violence are no longer monopolised by the state and that non-state actors either collude or conflict with the state is interlinked to urban planning and governance processes as well.

Local design issues can also create opportunities for perpetrators of violence, conditions for tensions to escalate to violence, or an environment that invokes fear. Paved streets with streetlights and multiple activities throughout the day can create safe environments. In a culture such as India’s, where machismo is valued, everyday conflicts can escalate into violent confrontations. Geographic concentrations of low-income populations can create ghettos where criminals can find shelter (World Bank 2010).

Economic globalisation has brought back the state in a big way. We call this neo-statism, which has resulted in the elite control of ideology and resources, causing massive disruption in the lives of the poor as they are evicted from their housing and lose their livelihoods in the name of economic growth and the building of world-class cities (Citizens’ Groups 2009; Mahadevia 2011).

In this chapter, we conceptualise links from urban planning and governance to violence in three ways:

deprivation and exclusion created through urban planning and governance as structural violence

conflicts embedded in existing urban planning paradigms, mechanisms, legislation and policies, institutions, and governance processes as structural violence

structural violence resulting from urban planning and governance (as outlined above) leading to physical violence as well as the threat or fear of physical violence.

The study took place in Ahmedabad and Guwahati, India. Ahmedabad is the brand ambassador for the Gujarat Development Model of high growth accompanied by low human development (Hirway and Mahadevia 2005; Hirway 2014), entrenched communalised polity, and social exclusion (Shah 2014). Guwahati is the gateway to Northeast India and a city and state that expresses ethnic conflicts and conflicts with immigrants from Bangladesh. In 2011, Ahmedabad had a population of 6.5 million while Guwahati had reached one million.

Everyday violence in Ahmedabad and Guwahati


Methodologically, we looked at the unfolding of structural violence in its everyday form in the two cities. The research is based on qualitative methods: individual interviews, focus group discussions, and key informant interviews. We identified the research sites through secondary and media reports and stakeholder consultations. We identified six drivers related to urban planning and governance that have led to structural and sometimes physical violence in both cities. These are:

1.land policies and processes

2.housing policies

3.infrastructure services: their denial, irregularity, and privatisation

4.urban design

5.elite capture of governance/decision-making

6.elitist discourse on topics such as granting of individual property rights, bourgeois urban environmentalism, and water as a consumer good.

We did not study all of the drivers in detail in both cities due to time and resource constraints as well as the interests of local stakeholders.

In Ahmedabad, the focus areas and case-study locales were:

1.conflicts linked to land, housing, and basic services in urban informal peripheral localities, which are also religious (Muslim) ghettos (focusing on one such locality populated by more than 25,000 families)

2.conflicts linked to land, housing, and basic services in slum resettlement sites constructed using Jawaharlal Nehru National Urban Renewal Mission (JNNURM) funds (focusing on three slum resettlement sites on the urban periphery) (see Figure 8.1)

3.women’s safety in traversing public spaces and in accessing and using public transport (focusing on two low-income localities).

Figure 8.1 Study sites in Ahmedabad: Bombay Hotel and Vatwa Basic Services to the Urban Poor (BSUP) resettlement sites, within the boundaries of the Ahmedabad Municipal Corporation (AMC)

In Guwahati, the focus areas and case-study locales were:

1.hill settlements that are experiencing conflict because of the demand for land rights (focusing on several neighbourhoods on two hills, which have larger numbers of low-income dwellers)

2.street vendors’ markets that witness conflict over space and management (focusing on one lease market and one informal market) (see Figure 8.2)

3.women’s safety in accessing and using public transport (focusing on six low-income localities) (see Figure 8.3).

Pathways to everyday violence

Land as a driver of structural violence

In Ahmedabad’s Bombay Hotel area, located in the informal urban periphery, a decade went by with the state being completely absent and residents having no political voice in the city due to the combined effects of living in an informal development and their identity as Muslims in a city and state dominated by a right-wing Hindu political party. Informal developments have grown up on agricultural land through informal land transactions and devious behaviour by land developers without development permissions. Bombay Hotel’s development into a large and dense Muslim neighbourhood has a violent history. Muslims displaced due to state-engineered communal riots in 2002 (Concerned Citizens Tribunal – Gujarat 2002)3 either purchased a plot/house in the locality or were resettled there by charity organisations. The nuclearisation of Muslim families living in the old city areas also led to the search for affordable homes to purchase, which they found in Bombay Hotel as builders floated low-instalment-based housing schemes. All the land transactions are on sale agreements made on stamp papers, which are quasi-legal documents that record the monetary exchange but not the transfer of ownership. The legal ownership of the lands thus remains with the original farmers. Benjamin (2008) calls this “occupancy urbanism”.

In 2013, planning came to this locality through the Town Planning Scheme, which is a land pooling and readjustment mechanism. The mechanism mandates that up to 50 per cent of the original land be vested with the planning authority for uses categorised as ‘public purposes’ such as roads, water, and sanitation infrastructure; education and health facilities; and Socially and Economically Weaker Section (SEWS) (low-income) housing known in the global literature as social housing. The implementation of two Town Planning Schemes as an urban planning tool would have demolished about 10 per cent (2,200) of the houses in the area. Since the residents do not hold legal land rights, they do not qualify for compensation. This led to tensions, followed by mobilisation and protests.

Figure 8.2 Informal markets studied in Guwahati.

Figure 8.3 Informal hill settlements studied in Guwahati, within the boundaries of the Guwahati Municipal Corporation (GMC).

This lane is supposed to be demolished under [the] TP [Town Planning] Scheme. We don’t know whether we will get a house or not. We have invested all our life savings in the house and have just completed paying our instalments and now this fear lingers above us.

(Resident of Bombay Hotel Area, Ahmedabad)

Urban planning as a driver of structural violence would have led to a further spiral of counter-violence and state violence, as we see later in the Guwahati case. But, instead, the cycle of violence was prevented through the creation of a political space for the Muslim residents. In recent years, slight shifts in politics in the city and the state saw Ahmedabad’s right-wing Hindu political party attempting to woo voters from among the Muslim community. Demolishing such a large Muslim neighbourhood would have attracted bad press and therefore the demolitions under the Town Planning Scheme were put on hold.4 Further, the boundaries of electoral constituencies were changed during the 2010–2012 period, bringing a change in local elected representatives. Centrist party (Indian National Congress) candidates won the Ahmedabad Municipal Corporation and state assembly elections for this locality, leading to pressure from local leaders and residents on the elected representatives to halt demolitions and extend basic services to the locality. The residents were successful on both counts. Significantly, the conflict over eviction was mitigated when the elected representatives intervened, causing the local government to modify the implementation of its Town Planning Scheme to address people’s concerns (Mahadevia, Pai, and Mahendra 2018). However, at the time of writing the process is incomplete, and conflict over eviction could arise again as the Town Planning Scheme implementation progresses. In the meantime, municipal services, such as drainage and water supply, are being extended to the locality. This has the potential to alleviate some of the conflicts and violence emerging from the informal provision of services as well as structural violence resulting from deprivation (discussed later).

In Guwahati’s hill settlements, conflicts between the hill settlers and local government related to the illegal occupation of land, particularly the forest areas, have become endemic. While evictions had stopped by early 2015 on account of massive political mobilisation and land rights becoming a major political issue in the region, the threat from the government persists and occupants remain fearful. Guwahati has a land regime where access to land is through ‘illegal occupation’, locally called dakhal, which has to be legalised through an application to the state for lease rights. This process is both discriminatory, which itself is a form of structural violence as we have defined it, and expensive, creating the potential for conflict. There are also conflicts about land access among various ethnic groups and people with allegiances to different political parties. Bengali Muslims, in particular, face discrimination. They are perceived as immigrants from Bangladesh and their practice of accessing land is politically charged. Their encroaching hill settlements do not have access to basic services, which again, following our definition, is a form of structural violence. Provision of basic services is arranged by the local collectives called Unnayan Samitis [development committees]. The state has informally passed on the responsibility for provision of services to the collectives and there are potential conflict points regarding access and the collection of charges for managing these services; these conflicts can become violent at any point in time.

In Guwahati, encroachments on eco-sensitive zones, such as hills and wetlands, have led the state government to demolish settlements. Residents of these areas have now organised as Krishak Mukti Sangram Samiti (KMSS) to resist the demolition. KMSS’ resistance has occasionally led to counter-violence against the state, which has responded with lathi-charges (baton-wielding) and firing weapons, leading to deaths and subsequent grievances against the state.

Now [the] government is claiming that due to encroachment on [the] hills, [a] flood situation emerges in the city. But the truth is that there is no scientific drainage system in the city. [The] drains are too narrow with respect to the rainwater that [the] city gets. But government is only blaming [the] hill dwellers for this.

(Hill dweller in Guwahati)

When the residents of the hill settlements and wetlands realised that the state government was legalising encroachments by commercial establishments while denying them land rights, the fight became even more bitter (Rajkhowa 2014). In one such episode in 2014, the KMSS called for self-immolation as a method of protest, leading one man to burn himself to death. The cycle of violence ended with this self-inflicted brutality; the KMSS leadership was arrested and agitation was temporarily suspended, but without an end to the conflict. The battle lines between the state and the land rights movement were drawn at the time we completed our fieldwork (Mahadevia, Mishra, and Joseph 2017). In the Guwahati case, we see the existence of discrimination—and thus structural violence—in the application of rules, leading to inequality of rights. Resistance against such discrimination has had the potential to turn violent. When this happened, the counter-violence by residents of informal settlements triggered violence by the state, and finally self-inflicted violence, creating a web of violence.

Land access is also a driver of conflict in Guwahati’s street markets. The national Street Vendors (Protection of Livelihood and Regulation of Street Vending) Act 2014 mandates that cities provide about 2 per cent of their land in the Master Plan and in any local plans for vending purposes. As a result, the “natural markets” do not get disturbed. Natural markets are those which spring up on their own due to demand from consumers. In spite of this legislation, Guwahati Municipal Corporation (GMC) had not allocated land for street markets. Instead, the GMC changed its governance system and leased the markets to certain bidders for an annual rent, allowing the lessee to recover this rent from the occupying vendors. However, the lessees recover the rent, but do not organise space for vendors, leading to multiple conflicts. There is competition and conflict among vendors to occupy space on a first-come, first-served basis. The conflict is also gendered, as the male vendors tend to occupy more space than their female counterparts. Ethnic and religious dimensions also exist. In particular, Muslim vendors face the ire of tribal vendors due to historic conflicts between them over land issues (Mahadevia, Mishra, and Joseph 2017).

As the lessee collects rents from the vendors without the oversight of the GMC, they often extract arbitrary and exorbitant rents, frequently through the use of musclemen who resort to verbal abuse and the threat and even use of physical violence. The vendors have no recourse in the face of these threats. The peculiarity of this system of governance, euphemistically called a ‘PPP’ or public–private partnership, is nothing less than the withdrawal of the state by outsourcing its responsibilities to generally unaccountable private sector actors. Lastly, there are local musclemen and youth club members who also extort money from vendors at informal markets. In these situations, the vendors have no avenues of redress from the state. Together, these dynamics feed a pervasive sense of fear in the two markets we studied, Beltola and Ulubari.

Once, we reached Beltola market at 2 a.m. and were unloading our goods from the truck. Some drunken local goons [thugs] asked for INR [Indian rupee] 1,000. When we refused to pay, they beat us. Only [the] police can help us in this kind of situation. In Beltola market, [the] lessee and local goons beat us; no one comes forward to rescue us.

(Male vendor in Beltola market, Guwahati)

In the case of street vendors in Guwahati, we see a case of structural violence emerging from non-implementation of existing legislation and the denial of rights to access land for the purpose of earning a livelihood. This form of structural violence emanates from an urban planning mechanism, which then transforms into multiple points of conflict some of which result in interpersonal violence. A situation in which governance functions are handed over to non-state actors has created the conditions for everyday threats and violence by those given the power to act on the state’s behalf. Finally, the lack of any mediation in these multiple points of conflict perpetuates inequalities and a cycle of structural violence that is converted into everyday violence.

Housing as a driver of structural violence

The lack of an affordable housing policy and related schemes led to the development of informal housing. Low-income households prefer to live in locations where they can easily access a livelihood and do not have to incur commuting costs. If the formal housing is expensive in such locations, they tend to live in informal housing. In Ahmedabad, for example, informal housing emerged on lands in central areas of the city close to opportunities for work, such as on the banks of the Sabarmati River that divides Ahmedabad. The riverfront development project displaced between 12,000 and 15,000 households of which approximately 11,000 were resettled at various JNNURM housing sites (Mahadevia 2014). Many were resettled at the Vatwa resettlement sites on the urban periphery, along with evictees from other infrastructure projects. The Vatwa sites comprise 9,200 dwelling units across seven sites that were selected due to their low land prices. We selected three of these sites, Sadbhavnanagar, Kusha Bhau Thakre (KBT) nagar,5 and Vasant Gajendra Gadkar (VGG) nagar for our research. Here, we found that urban planning has been used as a tool to displace low-income households from core areas of the city to the underdeveloped periphery, rather than bringing them into the urban mainstream and providing them with a wider set of opportunities. This is a typical case of reproducing and, in many cases, deepening inequalities through the urban planning process and further entrenching structural violence.

The constrained mobility and stressed livelihoods faced as a result of relocation have deepened the structural violence in the lives of the majority of residents. The resettled residents, uprooted from their former homes from which they could walk or cycle to work, are now forced to use motorised transport and pay for their travel to work. Specifically, women’s livelihoods were extremely constrained due to the fear of harassment and violence in commuting longer distances. Many dropped out of the labour market or began to work from their homes. At home, they produce goods on a piece-rate basis, resulting in a decline in their income (see Desai, Parmar, and Mahadevia 2017 for details). All of these factors together have increased housing and transport costs for the residents and have pushed them below the poverty line. Residents reported spending between one-third and one-quarter of their income on commuting, while also indicating increased housing costs.

This deepening of structural violence, through induced poverty, has led to thefts of private and common property. Robberies and burglaries became widespread at the Vatwa resettlement sites. Petrol from bikes; the tyres of rickshaws, motorbikes, and even bicycles; lids of overhead water tanks; tubing covering electric wires; and water pipes have all been stolen. Many residents have been robbed, sometimes at knife-point, while moving in and around the sites. Residential burglaries have taken place where cash, jewellery, mobile phones, and other items have been stolen. Residents try not to leave their houses unoccupied for long.

Residents described how they felt insecure; women, in particular, did not step out of their homes after dark, which was not the case where they used to live. Communities that used to be close knit have been dispersed across different sites and randomly situated within those sites due to the house allotment process. This haphazard approach to resettlement has led to social disruption and resulted in the loss of moral authority that local leaders, elders, and residents in general were able to exercise previously. Overall, internal informal social control is now lacking at the resettlement sites, creating a “mahol” or environment in which crime is committed with impunity and youth, in particular, easily stray towards theft, gambling, and illicit activities such as selling/consuming drugs. This latter situation is exacerbated by the absence of feasible livelihood options.

When there are no jobs, the youngsters get spoilt. They get into wrong activities [sic]. They do not have money for the transport fare … when a person goes hungry then he will steal; he will get into bad businesses.

(Female resident at the Vatwa resettlement site)

If [a child’s] mother is not at home the whole day and they are hungry then they might steal. If I leave my shop unattended just now and if a child who has not eaten since morning comes by, he might pick up something.… Today he might pick up something costing INR5; tomorrow he will steal something more.

(Female resident at the Vatwa resettlement site)

The built environment at the resettlement sites has also enhanced crime. Some stretches of the main road have few activities. The lack of ‘eyes on the street’ has created opportunities for robberies and the harassment of women. Within the sites, large sections do not have functioning streetlights and, where they are present, those who engage in crime break the lights whenever they are repaired. The common passageways in most buildings do not have functioning lights due to disputes about electricity bill payments. Overall, the physical environment is intimidating, especially for women. Thus, we see structural violence leading to increased crime. Emanating from it is everyday fear of crime and violence among the residents. Lack of proper policing, due to the overall failure of local security governance at the site, has deepened these fears among residents, which have a strong gender dimension.

Water supply systems as a driver of structural violence

Conflicts related to water are primarily due to the lack of a formal water supply in informal settlements. This is an urban planning issue. However, conflicts have emerged: in Bombay Hotel, these are due to the supply by informal water providers; in the Vatwa resettlement sites, they are due to the local government’s approach to the design and governance of water infrastructure.

In Bombay Hotel, the absence of a municipal water supply has led to the emergence of many different kinds of informal water supply arrangements, such as builders or better-off residents providing water from private bore-wells and residents fetching water from surrounding factories. Some of these arrangements mitigate extreme deprivation and conflicts around water, but many lead to a variety of conflicts:

among residents of a neighbourhood

between residents of different neighbourhoods

between residents and water suppliers

between residents and the local government (see Desai and Sanghvi 2017 for details).

The conflicts often involve verbal and low-intensity physical fights on a daily basis. The informal water suppliers are motivated purely by profit; their approach to supplying water reflects this, with fixed territories of supply to protect profits and coercive practices. They collect monthly charges whether or not they are able to supply adequate water. Residents are unable to oppose this due to a lack of alternative sources and the high-handedness and threats from the suppliers. In fact, residents cannot even complain about these operators without raising their ire. Multiple types of everyday conflicts emerge from the coercive management of water supply systems by these non-state actors. All conflicts have the potential to result in violence between residents and the water suppliers as well as among the residents themselves.

In one of the societies [a community], a local leader complained to a politician about the poor quality of water being supplied by the bore-well operator. This angered the operator who then stopped supplying water to the residents which in turn led to an argument between the residents and [the] local leader as the former felt that the latter should not have complained to the politician as this had totally cut off their access to water.

(Resident of the Bombay Hotel area)

Municipal officials have been aware of unregulated groundwater extraction, its sale in the area, and the consumption of this contaminated water by residents. These facts point to the local government’s complicity in the situation. In recent years, the local government has started to send water tankers into the locality. However, this method of supply is wholly inadequate, leading to violence at the tankers, frequently among women who usually bear the family responsibility for water collection:

Sometimes, these fights are bad. A few days ago, two women physically attacked each other and pulled each other’s hair. We had to call the police. One woman was sent to the hospital and the police arrested the other woman. Women fight with each other because only one tanker comes here for so many people and we cannot be certain that each of us will get water.

(Woman from Bombay Hotel)

In a couple of rare instances, residents have managed to collectively dig bore-wells to make arrangements for water, freeing them from the water suppliers and associated conflicts. This collective effort is an important bottom-up practice, but it still does not ensure good quality water and does not totally address the issue of deprivation and structural violence.

At the Vatwa resettlement sites, water provision was arranged without sufficiently resolving outstanding governance questions. This led to a lack of potable water and inadequate running water due to leaking and blocked water pipes. The structural violence created by resettlement on the urban periphery through socially disruptive processes has made the possibility of residents managing, maintaining, and sustaining the water supply extremely rare, thereby perpetuating structural violence. Municipal officials sometimes intervene, either out of benevolence or due to political pressure, but this generally results in uneven and inadequate interventions, furthering micro-local inequalities. Some officials are also unsympathetic towards residents because they consider resettlement to be an act of charity. As a consequence of insulting them in this way, residents have retaliated by vandalising public property at the municipal office, exhibiting counter-violence against the state. Here, despite residents’ protests, the local government has remained unresponsive and has withdrawn the low level support that it was providing to keep the water supply system working. Residents have almost completely stopped making demands on the government in the face of its increasingly uncompromising stance. But with access to water not resolved, the point of conflict has shifted to conflicts among the marginalised: among residents and between residents and water operators over maintenance and repairs (see Desai 2018 for details). Furthermore, inequities are also reproduced as women and children regularly bear the brunt of gathering adequate amounts of water for the household.

Public transport as a driver of women’s insecurity

Ahmedabad has good public transport coverage through buses operated by Ahmedabad Municipal Transport Services (AMTS) and the Bus Rapid Transit System (BRTS). However, there are issues related to frequency, connectivity, and affordability of public transport in Bombay Hotel and the Vatwa resettlement sites. Women were forced to commute using para-transit vehicles such as ‘shuttles’ (three-wheeled auto-rickshaws that operate on a shared basis). Many women perceived the shuttles to be unsafe when sharing them with male passengers or when they had to take them alone because they feared the male driver.

She used to go from here to the BRTS road in a shuttle and then from there to Dani Limda in another shuttle and then walk from there to school. The driver would keep a watch on her and would not take any other passenger when she was in his rickshaw; he would tease her and take her through different routes every day. Out of fear, she stopped going to school.

(Resident of Bombay Hotel regarding her niece’s experience)

In Bombay Hotel and Vatwa, walking to public transport stops was also challenging for women. They often face insecurity and harassment in public spaces. There are multiple drivers of this insecurity, many of which stem from the built environment: poor lighting; uneven roads and the absence of footpaths; and vacant spaces and structures along access roads that can be used by goons for illicit activities such as manufacturing, selling, and/or using alcohol and drugs or gambling. Multiple causes, including a lack of good employment opportunities, draw many male youths into illicit activities, which then lead them to harass girls and young women. As such, women in Bombay Hotel and Vatwa will only undertake a trip if it is a necessity.

Our research on public transport and women’s safety in Guwahati reveals that the city does not have adequate public transport systems to facilitate women’s mobility. Women are forced to depend on para-transit vehicles called ‘trekkers’ (open jeeps), where they sit in close proximity to men and are often sexually harassed. The virtual absence of the state from the provision of public transport, and the resulting over-dependence on para-transit, was found to both create and aggravate conditions for conflicts between various actors, including women and transport operators. As a result, women were not able to access or engage in the public realm in a manner equivalent to their male counterparts. Furthermore, there are other conditions that contribute to women’s insecurity: poor lighting on the streets; a lack of footpaths or sufficiently wide footpaths; unsafe areas such as alcohol dens on the roadsides; and the absence of assistance from law enforcement agencies for women in distress. The women living on the hills do not come out after dark because there are no streetlights on the paths leading to and from the hills. These paths are also slippery, particularly during the monsoon season, which creates further safety concerns.

Since women are nowadays advancing so much in every field, even surpassing men in some, the men probably feel that the patriarchal system of the society needs to be maintained under any circumstances and, hence, to balance the power equation—or rather to assert their own power—they try to dominate women in every possible way, one of which is crime against women … scaring the women into restricting their mobility so that they stay put in their “traditional” place in their homes and [do] not venture into the world outside.

(Elder woman leader of a women’s committee during a focus group discussion, Guwahati)

On the periphery of the city, women’s safety and security is a structural issue that has emerged either through land markets that have forced low-income households to seek housing on low-priced land or through urban planning that has created resettlement housing. In both situations, land markets have influenced location choices, as peripheral locations mean low land values that are affordable for low-income households. They purchase housing directly from the market or the state. The other structural issue is the lack of affordable public transport facilities and, worse still, the scarcity of last-leg connectivity from public transport to the users’ housing locations. The last-leg connectivity to and from public transport is provided by para-transit operators, who are entirely male and completely unregulated. This creates safety and security concerns. The lack of a built environment that establishes safe conditions further exacerbates fear. Both urban planning and the lack of urban governance have created conditions of structural violence that results in an everyday fear of harassment and violence against women.

Discussion and conclusions

These case studies, with the exception of the Vatwa resettlement site, narrate the everyday experiences of living in informal spaces, working in informal spaces, and travelling using informal transport modes. The informal sector has provided space and opportunities to cope with urban living and improve the lives of the urban poor. However, dependence on the informal sector also means that there is inherent conflict with the state’s urban planning legislation and rules, leading to state crackdowns through evictions and demolitions in the name of ‘legality’ and ‘urban planning’. For situations such as those in Bombay Hotel (Ahmedabad) and the hill settlements (Guwahati), the existing urban planning paradigm, tools, and practices offer, respectively, a limited solution or do not provide a solution. Also, the planning approach does not create affordable space allocations for petty commercial activities such as street vending. Street vendors then end up as part of the informal sector, with a solution proposed in the form of the national legislation enacted in 2014, as mentioned above, but which is not adequately implemented. Thus, the main driver of structural violence emanates from the urban planning paradigm and related tools in Indian cities.

In the case of Bombay Hotel, we believe that urban planning has provided a limited solution to its challenges because the process of implementing the Town Planning Scheme has resulted in partially halting evictions and in providing basic services. The political negotiations, through the mediation of elected representatives, have resulted in fewer conflicts; fewer conflicts resulting in violence (as we see in Guwahati); even less deprivation; and, therefore, less structural violence. The negotiations, predictably, have resulted in a series of trade-offs that provided some improvement in local living conditions. But, given the nature of planning regimes that are intolerant of informality, the threat of eviction may return.

Connected to these dynamics is the lack of a housing policy for the urban poor. They cannot afford housing in the legal market and are forced to seek solutions in the informal market. Housing policy, therefore, is the second driver of structural violence in the cities we studied. Attempts to address informal settlements through urban planning demonstrate insensitivity—or even hostility—toward slum dwellers. This takes the form of: (1) considering them to be encroachers and hence illegitimate occupiers of urban land, leading to evictions; and/or (2) considering slums/informal settlements as unsuitable to occupy precious urban land, offering their residents housing only on low-value property. Both approaches have the potential to cause displacement, which, if contested, has further potential to generate violence and counter-violence as seen in Guwahati.

Further, the local governments also tend to depend on the logic of commercial markets to find land on which to construct public housing for the urban poor. This results in low-income residents in the city centre being pushed to its periphery. This is the case in Ahmedabad, where the displaced families’ opportunities for a livelihood declined and transportation to and from work was compromised (particularly for women), pushing some people (mainly young men) to crime. The resettlement sites then become sites of violence and perpetual fear that threatens to engulf women and reduce their opportunities. Private land ownership regimes may hold a limited solution to the need to find land for the urban poor. Through mechanisms such as a Town Planning Scheme—like that in practice in Ahmedabad (see Mahadevia, Pai, and Mahendra 2018)—or the former Urban Land Ceiling and Regulation Act 1976 (now abolished), private lands could potentially be acquired for public purposes such as social housing.

Deprivation of public services also leads to structural violence. For example, a deficit in the public water supply and weak public governance allows non-state actors to manage the water supply, charge for poor quality services, and extort payments, all with the implicit support of municipal leaders. In resettlement sites, the approach toward the design and governance of water infrastructure, without giving thought to the socioeconomic background and capacities of the residents, is layered on top of the social and economic tensions caused by resettlement. This results in water deprivation as well as everyday conflict among residents and between residents and water supply operators over issues of repair and maintenance.

In order to have access to equal opportunities, public transport is extremely important. But in Ahmedabad, the BRTS has been unaffordable for low-income households and particularly the women among them (Mahadevia, Joshi, and Datey 2012), while at the same time the AMTS bus services are not dependable. In Guwahati, a public transport system within the city is virtually absent. Wherever public transport is available, women require safe access free of sexual harassment or the threat thereof. Women forced to use shared and congested para-transit, such as shuttles and trekkers, face harassment from male co-passengers and sometimes from the drivers; this must also be rectified.

The absence of the state, evidenced by a lack of planning and the provision of public goods, as well as the related creation of an institutional void, has resulted in the involvement of various profit-oriented and unregulated non-state actors, often creating fear and situations of conflict and violence. In addition, the absence of state or institutional mechanisms to mediate potential conflict-causing conditions results in everyday conflicts and occasional violence. Many of these non-state actors have political links and operate with the support of local ‘goondas’ (thugs/musclemen). These muscular service providers govern through the threat of violence and abuse.

In India, low-income groups have been left on their own to find housing and employment, negotiate their citizenship rights, and incrementally make their lives (Mahadevia 2015). The state enters the picture later through various patronage linkages, accommodating residents’ needs only under the compulsion of electoral politics. Low-income households are able to gradually improve their lives, but this process is not sustainable nor transformative, and the poor tend to live in a perpetual state of vulnerability where access to their rights depends on the situation. The state exists in an ‘instrumental form’, supporting capital accumulation by the people with means; but, it has left the poor to gradually improve their lives largely unassisted. This neo-statism, experienced mainly in large cities, has resulted in urban planning becoming a tool, a decision-making process controlled by the elite, to evict the poor from urban spaces. The resulting conflicts can lead to violence and/or crime. There are multiple routes through which this happens as demonstrated in the case studies.

This elite capture of land has brought protests by those whose life spaces are eroded by real estate-and big finance-led urban development. These protests have the potential to turn violent, inviting the state’s wrath and crackdown, leading to counter-violence from those adversely affected by this development model. The stage is set for perpetual conflict, violence, and crime. This is particularly so in the localities inhabited by the poor, where they become both the victims and the perpetrators of violence and crime.

India is on the cusp of rapid urbanisation. But, unlike the cities in Latin America and Africa, national statistics do not show a high incidence of crime and violence. But, structural violence as well as everyday, low-key, or petty violence is experienced in most cities as illustrated through our case studies in Ahmedabad and Guwahati. Urban planning and governance have become tools for one group to oppress another through the mediation of, or the deliberate avoidance of mediation by, the state. At the same time, accommodative state and flexible planning, under the pressure of electoral politics, creates spaces for negotiation as seen in Bombay Hotel. But, overall, violence is embedded within the mainstream and technocratic urban planning paradigm and its practices.

How will there be redemption from the structural violence that emanates from structural inequalities, which are unfortunately on the rise, in India’s cities? This chapter points to a need to change the approach to bring about the democracy and equity voiced in the protests, and evidenced by the struggles, of low-income urban communities in a broad sense. More specifically, it calls for urban planning that comes from the bottom up and is based on consultative processes, recognising peoples’ own efforts in creating their lives in cities. Urbanisation is a continuous process without end; therefore, urban planning must also be a continuous process with open-ended outcomes. Urban planning needs to be more participatory, not a product bound to legalities. But, for this to happen, urban development pathways must be created, through planning and governance, that prevent webs of violence, structural violence slipping into everyday violence, and conflicts tipping into episodes of violence, as well as pathways of negotiation that reduce violence and conflict.


This chapter presents the summary findings of the research project titled ‘Dynamics of Poverty, Inequality and Violence in Indian Cities: Towards Inclusive Policies and Planning’, based at the Centre for Urban Equity, CEPT University, Ahmedabad. The research was carried out by Aseem Mishra, Suchita Vyas, Shachi Sanghvi, Yogi Joseph, and our research partner organisations: Centre for Development, Ahmedabad; North-East Network; and the Society for Social Transformation and Environmental Protection in Guwahati. The financial support for this project was provided by IDRC (Canada) and DFID (UK).


1The importance of this is expressed in numerous programmes, including Pradhan Mantri Awas Yojana-Housing for All (Urban) (PMAY), Smart Cities Mission, and Atal Mission for Rejuvenation and Urban Transformation (AMRUT).

2Calculated from the McKinsey report on India that projects that the country’s urban population will be 590 million in 2030 (MGI 2010).

3In 2002, starting from the end February continuing through to May, Hindu right-wing groups attacked Muslim neighbourhoods and economic activities in the state of Gujarat. Ahmedabad was the most affected city. Muslim households from mixed neighbourhoods migrated to Muslim-dominated neighbourhoods such as Bombay Hotel for safety in numbers.

4Authors’ discussion with the Chief Town Planner of Ahmedabad.

5Nagar is a local term for a housing settlement.


Baviskar, Amita (2003). Between violence and desire: Space, power, and identity in the making of metropolitan Delhi. International Social Science Journal 55(175): 89–98.

Bayat, Asef, and Kees Biekart (2009). Cities of extremes. Development and Change 40(5): 815–825.

Beall, Jo (2007). Cities, terrorism and urban wars in the 21st century. Working Paper No. 9, London School of Economics, Crisis States Research Centre, London.

Benjamin, Solomon (2008). Occupancy urbanism: Radicalizing politics and economy beyond policy and programs. International Journal of Urban and Regional Research 32(3): 719–729.

Citizens’ Groups (2009). Half time or time out: Citizens’ review—Jawaharlal Nehru National Urban Renewal Mission: A collective view from below. New Delhi: Author.

Concerned Citizens Tribunal—Gujarat 2002 (2002). Crime against humanity. An inquiry into the carnage in Gujarat—Volume I: List of incidents and evidence. Mumbai: Citizens for Peace and Justice.

Desai, Renu (2018). Urban planning, water provisioning and infrastructural violence at public housing resettlement sites in Ahmedabad, India. Water Alternatives 11(1): 86–105.

Desai, Renu, and Shachi Sanghvi (2017). Water and conflict in Bombay Hotel, Ahmedabad: Urban planning, governance and infrastructural violence. Economic and Political Weekly 52(7): 49–57.

Desai, Renu, Vaishali Parmar, and Darshini Mahadevia (2017). Resettlement, mobility and women’s safety in cities. IIC Quarterly 43(3–4): 65–76.

Erman, Tahire, and Aslýhan Eken (2004). The “Other of the Other” and “unregulated territories” in the urban periphery: Gecekondu violence in the 2000s with a focus on the Esenler case, Istanbul. Cities 21(1): 57–68.

Frankel, Francine R. (2006). India’s political economy 1947–2004: The gradual revolution, Second Edition. New Delhi: Oxford University Press.

Galtung, Johan (1969). Violence, peace and peace research. Journal of Peace Research 6(3): 167–191.

Hasan, Arif (1996). Raising the curtain on the urban drama: The need for a new approach. In United Nations Economic and Social Commission for Asia and the Pacific (UNESCAP), Living in Asian cities: The impending crisis, causes, consequences and alternatives for the future. Report of the Second Asia-Pacific Urban Forum. New York: United Nations.

Hirway, Indira (2014). Assessing the inclusiveness of growth in Gujarat. In Indira Hirway, Amita Shah, and Ghanshyam Shah (Eds.), Growth or development: Which way is Gujarat going? (83–138). New Delhi: Oxford University Press.

Hirway, Indira, and Darshini Mahadevia (2005). The Gujarat human development report, 2004. Ahmedabad: Mahatma Gandhi Labour Institute.

Mahadevia, Darshini (2011). Branded and renewed? Policies, politics and processes of urban development in the reform era. Economic and Political Weekly 46(31): 56–64.

Mahadevia, Darshini (2014). Institutionalising spaces for negotiations for the urban poor. In O. P. Mathur (Ed.), Inclusive urban planning: State of the urban poor report, 2013 (148–166). New Delhi: Ministry of Housing and Urban Poverty Alleviation, Government of India and Oxford University Press.

Mahadevia, Darshini (Ed.) (2015). Shelter security in urban India: Pathways, barriers and outcomes. New Delhi: Concept.

Mahadevia, Darshini, and Sandip Sarkar (2012). Handbook of urban inequalities. New Delhi: Oxford University Press.

Mahadevia, Darshini, Rutul Joshi, and Abhijit Datey (2012). Accessibility and sustainability of bus rapid transit in India. UNEP Risoe Centre on Energy, Climate and Sustainable Development, Technical University of Denmark, Roskilde.

Mahadevia, Darshini, Aseem Mishra, and Yogi Joseph (2017). Ecology vs housing and the land rights movement in Guwahati. Economic and Political Weekly 52(7): 58–65.

Mahadevia, Darshini, Madhav Pai, and Anjali Mahendra (2018). Ahmedabad: Town planning schemes for equitable development—Glass half full or half empty? World Resources Report Case Study. Washington, DC: World Resources Institute. https://wriorg.s3.amazonaws.com/s3fs-public/wrr-case-study-ahmedabad.pdf.

MGI: McKinsey Global Institute (2010). India’s urban awakening: Building inclusive cities, sustaining economic growth. Delhi: McKinsey & Company. www.mckinsey.com/~/media/McKinsey/Featured%20Insights/Urbanization/Urban%20awakening%20in%20India/MGI_Indias_urban_awakening_full_report.ashx.

Moser, Caroline (2004). Urban violence and insecurity: An introductory roadmap. Environment and Urbanization 16(2): 3–16.

Moser, Caroline, and Philipp Horn (2011). Understanding the tipping point of urban conflict: Conceptual framework paper. Working Paper No. 1, University of Manchester. https://assets.publishing.service.gov.uk/media/57a08ae040f0b64974000812/60712_Moser_understanding2.pdf.

Rajkhowa, J. P. (2014). Conduct eviction of encroachers without discrimination. The Sentinel, 6 July.

Rodgers, Dennis (2010). Urban violence is not (necessarily) a way of life: Towards a political economy of conflict in cities. WIDER Working Paper 2010/020. Helsinki: UNU-WIDER. www.wider.unu.edu/sites/default/files/wp2010-20.pdf.

Roy, Ananya (2009). Why India cannot plan its cities: Informality, insurgence and the idiom of urbanization. Planning Theory 8(1): 76–87.

Rylko-Bauer, Barbara, and Paul Farmer (2016). Structural violence, poverty and social suffering. In David Brady and Linda M. Burton (Eds.), The Oxford handbook of the social science of poverty. New York: Oxford University Press.

Shah, Ghanshyam (2014). Governance of Gujarat: Good governance for whom and for what? In Indira Hirway, Amita Shah, and Ghanshyam Shah (Eds.), Growth or development: Which way is Gujarat going? (517–556). New Delhi: Oxford University Press.

Verma, Gita Dewan (2002). Slumming India. New Delhi: Penguin.

WHO: World Health Organization (2002). World report on violence and health. Geneva: Author.

Wirth, Louis (1938). Urbanism as a way of life. American Journal of Sociology 44(1): 1–24.

World Bank (2010). Violence in the city: Understanding and supporting community responses to urban violence. Washington, DC: Author. http://siteresources.worldbank.org/EXTSOCIALDEVELOPMENT/Resources/244362-1164107274725/Violence_in_the_City.pdf.

Zaidi, S. Akbar. (1998). Urban safety and crime prevention. Paper presented at UNCHS (Habitat) Regional Symposium on Urban Poverty in Asia, Fukuoka, Japan, 27–29 October 1998. www.fukuoka.unhabitat.org/docs/occasional_papers/project_a/05/crime-zaidi-e.html.

9Pathways to inclusive urban development

Contrasting experiences of relocated men and women in Colombo and Kochi

Rajith W. D. Lakshman and S. Irudaya Rajan


Planned urban growth processes often forcibly displace and relocate people within cities and towns, affecting their livelihoods, access to jobs, and social integration. They pose a major challenge to achieving the inclusion in cities that is emphasised in the New Urban Agenda (United Nations 2017). Known as the silent companion of urban growth (Cernea 1993), development-forced displacement and relocation (DFDR) can become a violent process subjecting the affected urban neighbourhoods to abuses of their rights at different points of the displacement–relocation cycle. This is particularly true for DFDR communities in low or lower-middle income groups in rapidly urbanising cities (Jayatilaka, Lakshman, and Lakshman 2018). Moreover, it is generally accepted that the burden of displacement is not equal among all members of a family: it is far greater for vulnerable members who are often women, children, the old, and the sick (Carpenter 2005, 2016). This chapter recognises DFDR in cities as a form of urban violence and seeks to measure and quantify some of its effects and costs on those groups that are affected through a gendered lens. It adopts a comparative approach covering two South Asian cities: Colombo in post-war Sri Lanka and Kochi in South India.

Recent debates on the link between rapid urbanisation and violence in cities commonly conceptualise urban violence as criminal in nature and measure it through national crime records (Bhide, Maringanti, and Mahadevia 2017). These “direct” forms of violence, resulting in physical and psychological harm, are aptly covered in the World Health Organization’s (2002: 5) definition of violence, which sees the issue as a public health concern:

The intentional use of physical force or power, threatened or actual, against oneself, another person, or against a group or community, that either results in or has a high likelihood of resulting in injury, death, psychological harm, maldevelopment or deprivation.

Other definitions of violence are much broader and more specifically include material deprivations as a form of “structural violence” (Galtung 1969) or symbolic disadvantage (Schröder and Schmidt 2001; Fox 2015). As Kasper et al. (2017) point out, urban violence can have a range of motivations (e.g. economic, political, or criminal), victims (gender-based, youth, etc.), and perpetrators (e.g. a mob, or even the state, as in this chapter). These broader, more nuanced definitions can be particularly useful for studying low-intensity violence such as reported in the South Asia region (Muggah 2012: 46; Bhide, Maringanti, and Mahadevia 2017). Following this, we see benefit in using a broad definition such as that proposed by Fox (2015: 31): “force, coercion, or psychological manipulation of any kind, used in a harmful or destructive way against some being that has an interest in not being harmed or destroyed.” Doing so helps to articulate the “direct” violence involved in the forced displacement phase of DFDR processes while also accommodating the more nuanced experiences of structural violence as encountered by folk on the margins of the city (Winton 2004).

Violence and exclusion are closely linked in urban settings across the world (Winton 2004; Moser and McIlwaine 2006; Umaña 2018). This is one reason why the ideal of the inclusive city seeks more equal and fair treatment for marginalised groups in cities, such as migrants, children, women, minority groups, and the urban poor (United Nations 2015, 2017). The growing inequality in income and wealth experienced in urban areas across the globe is a major barrier to addressing urban inequality in specific places. State-sanctioned DFDR processes in cities, as we examine, directly speak to these topics. For example, Kasper et al.’s (2017: 11) proposed measures to strengthen inclusiveness in cities include several that are routinely discussed in DFDR contexts: promoting access to urban services such as housing and education; increasing urban accessibility; increasing government accountability; and ensuring basic land tenure (Brown 2017; Herath, Lakshman, and Ekanayaka 2017). DFDR in cities—which is often (but not always) part of a government-led process of urban planning—can either exacerbate the divisions within cities or act as a catalyst through which exclusion and violence can be addressed (Brown 2017).

This chapter engages with the notion of inclusive cities by differentiating between the experiences of DFDR-affected women and men. This focus is important for three reasons. First, DFDR-affected women are known to have particular protection and assistance needs that are different from those of men (Thukral 1996; Ahmad and Lahiri-Dutt 2006). The studies that examine this phenomenon comprise a subset of the broader literature that highlights men and women as having differentiated experiences in situations of conflict, upheaval, and displacement (Alsaba and Kapilashrami 2016; te Lintelo, Lakshman, et al. 2018). Second, the evidence shows that displaced women can more effectively assist in households’ coping and recovery compared to men (Momsen 2004: 130). It is critical to note that even when replacement housing is ensured—which is often not the case, although it is true for the DFDR cases we examine—household-level coping remains an important element in how those affected reinforce their livelihood strategies. Third, the gendered characteristics of DFDR—the distinctiveness of women’s needs and the importance of their coping mechanisms—have critical policy implications for relocation (Desmond 2012; UN-Habitat 2013).

The differentiated experience of women affected by DFDR in cities is better understood by using gender theory to frame the subject. For example, at the broadest level, the use of a gender and development approach is critical as it examines how women’s socially constructed roles, in relation to those of men, determine their experiences of systematic subordination (Moser 2012). Despite evidence to the contrary, cities in general provide women with better opportunities to escape some of these effects reinforced by gender (Chant and McIlwaine 2016). Though urbanisation is often linked to women having better access to services and employment opportunities, in addition to lower fertility rates, these positive aspects of city living often fail to reach the urban poor. Migrant women in cities are particularly disadvantaged in this regard as migration depletes their social networks and isolates them (Hilfinger Messias 2011). DFDR as a form of (forced) migration has similar effects. Anwar et al.’s (2014) take on the mobility–gender intersection and urban violence also helps to shed light on our findings.

Our research sought to compare and contrast displacement and resettlement experiences in a post-war context compared with a regular city: Colombo, Sri Lanka and Kochi, India, respectively. Both are examples of South Asian cities that are planning to achieve high rates of economic growth under conditions of rapid urban population growth. Though the two cities are in different countries, they are quite similar in their history and socioeconomic profiles. For example, post-independence Kerala state and Sri Lanka have travelled similar socioeconomic trajectories with social development given precedence over economic growth (Bhat and Rajan 1990; Zachariah and Rajan 1997; Lakshman and Tisdell 2000; Sen 2013). As a result, both cities currently share similar levels of human development as measured by the United Nations Development Programme’s (UNDP) Human Development Index (see Kundu 2015; UNDP 2015). In fact, all relocation projects examined here (in both cities), with their high-quality replacement housing and/or attractive compensation packages, are testimony to this.

We contend that the main distinction between the two cities arises from Colombo’s post-war legacy. The choice of two case-study cities offers a chance to compare how post-war (Colombo) and more typical (Kochi) governance structures negotiate development-driven relocation in cities. Yet, a more interesting anchor for this comparative study is the contrast between the natures of the DFDR projects considered in these two cities.

In Colombo, the projects sought to relocate urban dwellers from multiple informal settlements into high-rise flats.1 In Kochi, the projects sought to relocate people from the then periphery of the city (that has since been engulfed by the sprawling city) to areas further away from the city centre. In Colombo, the specific development activity that spurred relocation is either city beautification or slum upgrading, whereas in Kochi the drivers of relocation were specific urban development (construction) activities.

The chapter is organised as follows: the next section briefly outlines the methodology used. The following two sections present the findings of a cross-sectional analysis of the household survey. The penultimate section cross-examines the findings in relation to those of other researchers. The final section provides some concluding remarks.

Methodology and data

This chapter is based on data from a three-year (2013–2016) research study.2 While the researchers have collected other data including key informant interviews, community profiles, and in-depth household interviews, the main focus of this chapter is the second wave of a longitudinal household survey which was conducted in 2015 and 2016. The research locations of relocated and resettled communities within each city were selected purposively. While our research interests guided the selection, we also had to take certain practical limitations into consideration. For example, in Colombo, at the time of the research, more recent relocation sites were very difficult to access for research purposes. In addition to urban populations affected by DFDR, comparable control groups who had not experienced relocation were also included in the survey. The control groups and sites were selected based on how relevant they were as control cases for the selected resettlement and relocation sites. From within each site, individual households were randomly selected for inclusion in the survey.3 Pre-interview screening questions were used to ensure that the households matched the categories of interest. For instance, some house-holds had moved into the relocation sites as private buyers and had not, therefore, experienced relocation; these households were excluded.

In each of the two cities, a team of enumerators collected data under the supervision of researchers. The enumerators visited households in gender-balanced pairs to ensure that women were always interviewed in the presence of a female enumerator. In Colombo, the DFDR group consisted of households relocated from an informal settlement into high-rise buildings and the control group consisted of households who continued to live in similar informal settlements. The DFDR households from Kochi were surveyed in their replacement houses in relocation sites. The control group in Kochi consisted of households from settlements adjacent to the ones that were displaced. In Kochi, the control groups were, to a certain degree, affected by the development projects that displaced their erstwhile neighbours: the development of neighbouring land had increased the value of their land. In this way, the group was a control in the sense that they were never relocated, but not in the sense that they were not affected by the development projects which led to the relocation of their neighbours.

The study uses a material quality of house index (MQHI) that combines 11 variables related to the built quality of a house.4 The MQHI has a minimum possible value of 0 and maximum of 11; a value of 0 indicates the worst quality of housing and 11, the best.

Table 9.1 outlines summary statistics for the survey data that includes 1,217 households (totalling 5,570 individual residents) from the two cities combined. While the sampling was done at the household level, we collected some data about all the individuals in the surveyed households. For example, demographic details for all individuals were collected from the main respondent. The household level statistics in Table 9.1 show that a larger proportion of main respondents in Colombo were female (approximately 70 per cent) while in Kochi, there was a greater balance between male and female respondents. A higher proportion of respondents in Colombo identified theirs as female-headed households (approximately 30 per cent in Colombo compared with about 12 per cent in Kochi).

The survey employed a five-point Likert Scale—response options of very happy, happy, neither happy nor unhappy, unhappy, and very unhappy—to collect self-reported data on happiness from all the individuals in a house during the time enumerators visited them. Additionally, individuals from the DFDR groups were asked to recall their happiness “before relocation”; the individuals among the control groups, of course, had no relocation experience, so only reported on their current happiness. This approach meant that only those individuals present at the time enumerators visited the house were able to respond to the question about happiness. Table 9.1 shows that 642 male respondents (41 per cent of total respondents to the happiness question) and 917 female respondents (59 per cent of total respondents) were questioned regarding happiness. A higher proportion of female respondents answered these questions in Colombo than in Kochi.

Quality of replacement housing: the perspective of relocated men and women

A strong argument in favour of DFDR is that the relocated or resettled people will live in a better house. Figure 9.1 plots the distribution of MQHI scores according to city—Colombo or Kochi—DFDR or control group, and whether the MQHI measurement relates to the current period (i.e. the time of the survey, the 2015–2016 period) or to a period before relocation/resettlement. The current/before division is relevant only to the DFDR group as the control group was not relocated/resettled.

Figure 9.1 captures the effect that the relocation had on the MQHI scores of the relocated DFDR households in the two cities. Clearly, the distribution of MQHI scores had shifted rightwards—i.e. indicating an improvement—after the relocation; this shift is statistically significant. This suggests that the MQHI scores improved unequivocally following relocation, which is consistent with the aim of DFDR programmes to improve the quality of housing of the relocated households in both cities. Figure 9.1 also suggests that the DFDR group in Colombo has better quality housing than the control group. In contrast, the average “current” MQHI score for the relocated households in Kochi is similar to that of the control group. This can be interpreted as evidence that the MQHI scores of households that were not relocated also improved after the projects were implemented.5 Such an improvement may well be a reflection of the ex-post real estate price increases enjoyed by the control group in Kochi. This is unequivocal evidence that resettled/relocated households did receive better quality housing compared to what they had before. But, was that a sufficient condition to ensure their well-being and happiness?

Table 9.1 Household survey: summary



Colombo DFDR

Colombo Control

Kochi DFDR

Kochi Control




   721 (51%)

   641 (49%)

   564 (49%)

   838 (49%)

2,764 (50%)



   690 (49%)

   666 (51%)

   595 (51%)

   855 (51%)

2,806 (50%)



   277 (20%)

   293 (22%)

   226 (19%)

   278 (16%)

1,074 (19%)



   284 (20%)

   293 (22%)

   202 (17%)

   295 (17%)

1,074 (19%)



   283 (20%)

   250 (19%)

   239 (21%)

   352 (21%)

1,124 (20%)



   273 (19%)

   264 (20%)

   245 (21%)

   329 (19%)

1,111 (20%)



   294 (21%)

   207 (16%)

   247 (21%)

   439 (26%)

1,187 (21%)

Responded to happiness question


   134 (34%)

   133 (35%)

   159 (50%)

   216 (46%)

   642 (41%)


   255 (66%)

   249 (65%)

   162 (50%)

   251 (54%)

   917 (59%)

Total individuals







Household headship


     95 (34%)

     85 (29%)

     34 (13%)

     47 (12%)

   261 (21%)



   183 (66%)

   212 (71%)

   228 (87%)

   333 (88%)

   956 (79%)

Gender of household’s main respondent













Total households







Source: Authors’ data and calculations.

Figure 9.1 Effects of the relocation on the MQHI scores of relocated households and control groups in the two cities—Colombo and Kochi.

Source: Authors’ data and calculations.


The top two rows compare the “before” displacement and “current” distributions of MQHI scores of the sample of DFDR communities from the two cities while the last row shows the distribution of MQHI scores for the control groups.

Figure 9.2 presents an analysis of respondents’ self-reported “current” and “before” happiness data, where the height of a bar corresponds with the average happiness score for a given group. The figure uses three panels to illustrate the results for male respondents, for female respondents, and for all respondents. The “current” and “before” bars for the DFDR community in Colombo indicate that both male and female respondents have marginally improved happiness levels following relocation. In Kochi, the same comparison yields clear evidence of the collapse of happiness after relocation. One might be tempted to attribute this to nostalgia, which could potentially inflate the self-assessed “before” happiness levels. While nostalgia could certainly be an element, the control group responses give us faith in the recalled happiness data: recalled happiness levels (grey bars in Figure 9.2) relating to the time the DFDR populations were in informal settlements look strikingly similar to the current happiness levels of the control group. The control group’s assessment of their current level of happiness is not subject to recall bias or nostalgia.

Figure 9.2 also reveals preliminary findings regarding the gendered well-being effects of development-driven relocation in cities. For example, gender differences can be noted when comparing happiness outcomes between the DFDR and control groups in Kochi: while female respondents in the control group in Kochi are happier (average happiness score of 4 vs. 3.5 for male respondents), the difference is negligible in the DFDR group. Our point is that, while the gender-neutral collapse of happiness in Kochi following relocation attracts much attention—as it should—these gendered impacts also need scrutiny. Male/female comparison for Colombo in Figure 9.2 suggests that the relocation experience in that city is ostensibly without any gender undertones: the average male and female happiness levels are all but identical in Colombo. The visible absence of gendered effects in happiness data is itself an interesting finding worthy of further examination.

Figure 9.2 Average happiness levels among DFDR and control groups in Colombo and Kochi.

Source: Authors’ data and calculations.


The relevant average happiness scores are provided inside the bars.

Table 9.2 Logistic regressions to explain current happiness levels




Male-headed households (Reference: Female-headed households)





Head/spouse (Reference: Another household member)





Respondent is female (Reference: Respondent is male)





Education > Grade 9 (Reference: Education is </= Grade 9)





Current MQHI





Control group (Reference: DFDR group)













Pseudo R2



Prob > chi2



Source: Authors’ data and calculations.


Coefficients are proportional odds ratios and figures in parentheses refer to standard errors.

* p < 0;

** p < 0.05;

*** p < 0.01.

To further examine and perhaps corroborate or refute the findings presented in Figure 9.2, we must use multivariate techniques; we chose a regression approach. Given our main interest is in measuring the gendered impacts on the probability of being “happy” or “very happy” with relocation/resettlement, we recoded the 5 point Likert scale as a binary scale (1 if “happy” or “very happy” and 0 otherwise). We then applied a binary logistic statistical regression to these data. The regression results give a more accurate assessment of the effect of relocation/resettlement on happiness because they are able to control for multiple confounding effects that the graphical approach could not achieve. The results are presented in Table 9.2, separately for each city. Each model predicts the probability of being happy in the “current” period.6 They measure the effect that relocation may have had on the happiness of individuals from each city while also emphasising the impact of:


being a member of a female-headed household

the individual’s education level

“current” level of household MQHI score

whether the individual belonged to the DFDR or the control group.

By including information on household headship and on individuals’ education levels, the regressions add further layers of complexity to the picture in Figure 9.2. The regression results broadly confirm the results presented in Figure 9.2, and more clearly quantify gender nuances.

In Table 9.2, the results are presented as odds ratios. For example, in both models “Male-headed household” estimators suggest that, if all other variables in the models are held constant, individuals in male-headed households (compared to those in female-headed households), are likely to be happier compared to individuals in female-headed households. Both these parameters are significant at the 10 per cent level. The analysis suggests that an individual from a male-headed household in Colombo is 1.465 times (or 46.5 per cent) more likely to be happy than an individual from a female-headed household in the same city. In Kochi, an individual from a male-headed household is 51.2 per cent more likely to be happy than an individual from a female-headed household. This is not news as female-headed households tend to have lower incomes than their male-headed counterparts; it is proof that the regression findings are congruent with the wider literature. Though female-headed households are faring poorly compared to the male-headed ones, women seem to be happier or more content with life than the menfolk, irrespective of the headship status of their households: in Colombo women are 37.5 per cent more likely to be happy compared to men, and in Kochi the figure is 43.5 per cent.

In addition to information on headship and gender, we also included “current” MQHI levels in the analyses to test whether the material quality of the house mattered to current happiness. Even though the estimated values of these parameters suggest that improved MQHI scores are associated with higher levels of happiness, these estimates are not statistically significant. Thus, in confirmation of what was suggested in the earlier graphical analysis, the regression results also affirm that higher material quality of housing is not a significant factor in predicting happiness. Related to this is the parameter estimate of being in the “Control group”. The estimated value from Colombo suggests that individuals from the control group are less (0.827 times or 17.3 per cent) likely to be happy compared to the DFDR group but, again, this value is not statistically significant. However, in Kochi, the result is strongly significant at the 1 per cent level and suggests that an individual from the control group is 6.248 times more likely to be happy compared to someone who was relocated.

These results are intriguing precisely because the MQHI—a prime focus of experts involved in relocation—seems unable to explain self-reported happiness. The next section attempts to shed more light on this by looking at other impacts that relocation has had on the affected urban dwellers.

It is not only about shelter: sociocultural impact of relocation within cities

In this section, we look at some of the ways the relocation may have changed a given population’s sociocultural parameters. Doing this helps to check whether these changes could explain some of the peculiar happiness results/outcomes revealed earlier, particularly the result where the MQHI scores (a key focus of best case DFDR practitioners) were not a significant predictor of happiness. We do this by comparing an assessment of socio-cultural parameters according to whether the assessment was made by a male respondent or by a female respondent. The comparison is also dis-aggregated by city7 and the results are summarised in Figure 9.3.

Following te Lintelo and Lakshman (2015), Figure 9.3 compares the average levels of 25 sociocultural indicators between the two comparator groups (DFDR and control) with a view to exposing the different effects the relocation/resettlement has had on men and women. The 25 indicators are explained in Table 9.3 and are divided into four groups: perceived security (seven questions), freedoms (four questions), civic activity (six questions), and social networks (eight questions). The first column in Figure 9.3, for example, contrasts the average score assigned to each of these indicators by male respondents in the Colombo DFDR group with those assigned by male respondents in the Colombo control group.8 The two averages are linked with an arrow; arrowheads that point to the left suggest that the average score for the DFDR group was higher than that for the control group and arrow length indicates the difference between the two averages.9 Right pointing arrows, on the other hand, indicate that the average score in the control group is higher: for example, in the first column, this may be interpreted as evidence that men in the Colombo control group are inclined to think that their sociocultural environment is better than their counterparts in the Colombo DFDR group. The second column in Figure 9.3 shows the same analysis using the male respondents from Kochi while the third and fourth columns analyse responses from female respondents. The black, grey, and white dots in each row indicate whether the difference captured in a left- or right-pointing arrow is statistically significant.

Figure 9.3 Empirical characterisation of the impact of development-forced relocation on affected men and women in Colombo and Kochi.

Source: Authors’ data and calculations.


Average scores for 25 indicators are compared using arrows where the arrowheads point from average “current” score for the DFDR group (baseline) to average “current” score for the control group. In addition to comparing the averages (using arrows) the figure also illustrates whether the estimated differences in average scores are statistically significant. Significant independent sample T-test results are indicated by dots against the relevant question. The dots are uniformly colour coded where white (○) indicates significance at 10 per cent level; grey (●) at 5 per cent level; and black (●) at 1 per cent level.

Table 9.3 Household survey indicator groups and specific indicators

Indicator group


Perceived security

Perceive neighbourhood to be safe in general


Perceive neighbourhood to be safe at night


Neighbourhood squabbles/tensions exist


Women can move around freely in this neighbourhood after 8 p.m.


Sexual harassment exists in the neighbourhood


Publicly visible social miscreants (alcoholism/gambling/drugs) in the neighbourhood


Domestic violence (physical) exists


Organising and/or following religious activities


Participating in community groups


Membership in political parties



Civic activity

Self-help groups


Clubs (arts, sports, etc.)


Religious organisations


Farming/fishermen groups


School/hospital development committees, PTAs, etc.


Other groups

Social networks

Do you think people here would contribute to the safety of the locality by alerting others if any suspicious activity is observed?


Do you have neighbours who may help you in an emergency like sickness, death, etc?


Do you have neighbours who may help you in a financial emergency by offering a loan?


Do you have neighbours who may help if you or a family member wants to find a job?


Do you have neighbours with whom your children can stay while you attend to an emergency?


Do you have a neighbour to look after your property while you are away for a week?


Do you think most people in your locality can be trusted?


Do you think most people in your locality contribute to keeping the environment clean?

The analysis presented in Figure 9.3 can be interpreted as evidence that sociocultural conditions in the DFDR sites are weaker than those in the control sites: 61/100 comparisons in Figure 9.3 (where the arrowheads point to right) support this assertion while 39/100 comparisons (with left pointing arrows) contradict it. Out of these, only a few (23/100) are significant at the 10 per cent level (see T-tests reported by dots in Figure 9.3). The bulk of these significant T-test results (16/23) suggest that the control group is faring better in comparison to the DFDR group. In contrast, only seven left pointing arrows are significant. Thus, the smaller set of statistically significant differences in Figure 9.3 also confirms that the sociocultural environment is a weaker influence in the DFDR sites.

While the situation that the contrast between DFDR and control groups brought out in Figure 9.3 is useful, what is perhaps more relevant for the present chapter is the gendered view about the respective situations that seems to jump out: men and women seem to have strikingly contrasting views. As noted above, there are 23 instances of statistically significant differences highlighted in Figure 9.3. Thirteen of these compare the views of men in the DFDR and control groups; ten compare those of women. What is interesting is that in nine out of these ten issues/outcomes are identified as significantly different by women, suggesting that the situation in the control group is better for women, particularly in Kochi. The analysis of responses by men contrasts with the above because results from the men do not clearly favour the control group. While men rank the control case better than the DFDR case on eight indicators, they also rank the DFDR case better than the control case on six indicators.

The results suggest that “perceived security” is the thematic area in Figure 9.3 that reports the highest number (13) of significant differences between the DFDR and control groups, while most of these (11 out of 13) suggest that perceived security is higher in the control group. For example, the two indicators on safety within the neighbourhood (in general and at night) confirm that safety is significantly lower in the DFDR setting. Both men and women in Kochi are of this opinion, while only women in Colombo seem to have this perception. Moreover, while the assessment by DFDR men in Colombo on whether women can move about freely after 8 p.m. is statistically higher than that of the men from the control group, women from the same locations in Colombo did not see it this way: the difference between DFDR and control group women in Colombo is not statistically significant. In Kochi, women seem to think that sexual harassment in public spaces and domestic violence is higher in DFDR settings vis-à-vis the control setting, though men’s experiences do not seem to reflect this. In addition to the DFDR/control comparison, Figure 9.3 also suggests strong differences between Colombo and Kochi: most arrows for Kochi are placed close to 1, which is the best outcome, while corresponding arrows for Colombo are further away from 1. In fact, in some cases, the arrow is quite close to 0, which is the worst possible outcome. Take for example, the indicator of domestic violence, which is assessed very close to 1 by both men and women in Kochi but as less than 0.5 in Colombo.10

Other thematic areas covered in Figure 9.3 do not seem to suggest as strong a distinction between DFDR and control settings, at least not as poignantly as in the case of “perceived security”. For example, men in both cities seem to think that few indicators of “social networks” are significantly different between the DFDR and control settings. According to the views of the men, the control setting in Colombo had better social networks than the DFDR setting—see indicators on being alert to safety threats and the availability of financial help (Table 9.3)—while the reverse was true in Kochi.

The above results can be summarised thus:

Overall, the evidence is strongly consistent with a scenario in which the sociocultural environment of the DFDR respondents had deteriorated following relocation.

The instances of evidence to the contrary are sparse and visible only when men’s perceptions are considered.

An inferior sociocultural environment in the DFDR sites vis-à-vis the control sites seems to explain much of the puzzle relating to why ensuring the quality of replacement housing given to the those relocated/resettled had not led to a significant improvement in the happiness of the DFDR-affected communities in Colombo and Kochi.

Indicators of “perceived security” are the worst performing indicators, which may have contributed to the poor happiness scores among the DFDR-affected men and women, especially the latter.

Discussion and conclusion

The case studies from Colombo and Kochi suggest that the provision of better quality housing for the urban dwellers faced with DFDR, while crucial, is unlikely on its own to ensure that the well-being of the affected communities/neighbourhoods is safeguarded. The evidence presented here confirms that, despite ensuring a higher material quality of shelter—which was significantly better than the quality of housing in informal settlements in post-war Colombo or in low-cost formal settlements in Kochi—the DFDR populations studied reported lower levels of happiness compared to the levels reported in the control groups. While, in this chapter, we focused exclusively on survey data and quantitative analysis, qualitative data from the same research project complements these findings: for example, Collyer, Amirthalingam, and Jayatilaka (2017); Herath, Lakshman, and Ekanayaka (2017); and Lakshman, Herath et al. (2016) use qualitative data to outline the contours of community-level social impacts that have impacted men and women differently. George and Rajan (2015) is a similar study focusing on DFDR experiences in Kochi.

The results are compatible with the findings of other studies that have looked at relocation in cities where improvement in housing conditions failed to improve relocatee well-being. Bender (2008: 37), in relation to relocations in London to make way for the 2012 Olympics, points out that 73 per cent of the affected people were happy about the new houses, yet 51 per cent reported that their circumstances had not improved following the relocation.11 The author shows that poor well-being outcomes came about because the city developers had subordinated the relocatees’ rights to information and the cost of housing had increased or became more regularised for most relocatees. Perlman (2004) contrasts the conditions in the favelas of Rio de Janeiro between 1969 and 2002 to show that, while there have been considerable physical improvements, other dimensions of life in the favela, including security and health, have drastically worsened.

The gendered perspectives on relocation/resettlement examined here resonate well with the broader literature on women’s experiences post-relocation. It is the gendered lens that explains why housing—while a major component—is insufficient, in particular for women, due to specific factors such as safety issues, social network disruptions, etc. These non-material elements are crucial to well-being and happiness in urban settings (te Lintelo, Gupte et al. 2018). A community’s social capital is closely linked to its members’ perceptions of liveability, their sense of belonging, and the trust they place in their neighbours (Hutchinson et al. 2009). Women in particular, but also other vulnerable groups, benefit from such social connections (Meleis, Birch, and Wachter 2011). In fact, the depletion of perceived safety by women is largely to do with a loss of community after relocation/resettlement and an inability to rebuild the same feeling amid the strangers with whom they are forced to live (George and Rajan 2015; Lakshman, Ekanayaka, and Lakshman 2016).

It is possible that planned DFDR in cities, as profiled here, also transforms the role of the urban neighbourhood, as documented in other cities (Forrest and Kearns 2001). The neighbourhood of informal settlements plays a primary role as an arena for what Forrest and Kearns (2001) describe as “extended domestic activities”. For example, it is common for domestic activities, such as accessing water (shared public tap), clothes washing, and even using the (shared) toilet, to happen in the public space of the neighbourhood in informal settlements.12 When the capacity to perform these activities moved from the “neighbourhood space” of informal settlements to the “home space” of relocated/regenerated formal settlements, the need to do them in public became redundant. Forrest and Kearns (2001) also characterise a distinction between “neighbourhood” and “neighbouring”: in disadvantaged neighbourhoods, the quality of “neighbouring” is an important element in peoples’ ability to cope, whereas in more affluent areas the “neighbourhood” may be more important than interacting with one’s neighbours to create a community.

While we are not suggesting that the disadvantaged/marginalised communities that were relocated had become affluent after the relocation, it is quite possible that the substantial increase in the quality of housing caused a change in the role of the neighbourhood from the perspective of the affected households. Perhaps this explains why women perceived this change in ways that differed from the men. Some of the stresses experienced by urban women (see Meleis, Birch, and Wachter 2011), such as changed work status, safety concerns, lack of convenient transportation, etc., are likely to multiply following forced resettlement. In Colombo, for example, women were disproportionately affected by loss of livelihood and constraints on children’s education due to resettlement and associated transportation problems and disruption of social networks (Lakshman, Ekanayaka, and Lakshman 2016), something that is made all the more challenging by the fact that women in urban areas frequently work in informal sector enterprises connected to the neighbourhood (Moser 2012). In Kochi, women were additionally stressed by having to enter into new waged labour and by the threat of state violence (George and Rajan 2015).

This chapter looked at household-level impacts of development relocation in two medium-sized, yet fast-growing cities in South Asia, to ascertain the burden of relocation from the perspective of the affected women and men. It turned out that the projects we studied had—to a large extent—secured materially improved post-relocation housing outcomes for the relocated populations, a possibility that had to a large part, if not completely, been ignored in the literature. This study was therefore able to identify sociocultural concerns that could not be addressed by looking at urban relocation purely as a problem of housing, though clearly the latter is indeed a hugely important element. Our research findings lead us to recommend that those designing DFDR policies consider adopting a more holistic view of housing, that can provide much more than higher material quality in replacement housing.13 Moreover, policymakers who adopt a gender-mainstreamed approach to DFDR policy are likely to make powerful inroads into addressing some of the neglected areas of DFDR, such as safety issues and social network disruptions, thus making DFDR a more positive experience for relocated women and men.


This research was funded under the IDRC- and DFID-funded Safe and Inclusive Cities initiative. The authors acknowledge help from Kopalapillai Amirthalingam, Danesh Jayathilaka, Dhammika Herath, Iresha Lakshman, Mike Collyer, Indu Kalamani, Mohammed Alikhan, and Sunethra Perera who were involved in the research, but not in writing the chapter. The International Centre for Ethnic Studies, Sri Lanka and the Centre for Development Studies, Kerala implemented the research project in Colombo and Kochi, respectively.


1The high-rise flats in Colombo were built on the site of a former informal settlement. So, there was an element of in situ resettlement, at least for the original dwellers from that site.

2The study also included Jaffna in Sri Lanka; data from that site are not used in this chapter.

3On-site randomisation was used.

4The variables were min-max normalised and additively combined to construct the index.

5Here we assume that the MQHI of the control group about the time relocation happened was similar to that of the relocated group.

6The data on the status “before” relocation are not included in the regression because those data are only available for the DFDR group.

7The assessment scores were normalised using the min-max method before averaging across various comparator categories.

8The respondents referred to here are the main respondents who answered the survey questions and who comprise a sub-sample of the individuals who answered the happiness questions. Table 9.1 outlines the male/female breakdown of respondents. For example, the main respondents in the survey of the DFDR group in Colombo were 69 per cent female and 31 per cent male.

9Some of the differences are negligible, in which case the body of the arrow is not visible at the present scale of the figure. In these cases, only the head (tip) of the arrows is visible.

10Our coding is as follows: 1 means no domestic violence exists; 0.5 means domestic violence exists “to some degree”; and 0 means domestic violence “definitely” exists.

1139 per cent felt worse off and 12 per cent saw no difference.

12The data show that dwellers in the informal settlements studied here have reported a significantly higher proportion of residents using shared water and/or sanitation facilities.

13See Collyer, Amirthalingam, and Jayatilaka (2017) for a discussion of how the International Covenant on Economic, Social and Cultural Rights (1966), for example, provides a definition of “adequate housing” that describes such a holistic view.


Ahmad, Nesar, and Kuntala Lahiri-Dutt (2006). Engendering mining communities: Examining the missing gender concerns in coal mining displacement and rehabilitation in India. Gender, Technology and Development 10(3): 313–339.

Alsaba, Khuloud, and Anuj Kapilashrami (2016). Understanding women’s experience of violence and the political economy of gender in conflict: The case of Syria. Reproductive Health Matters 24(47): 5–17. www.sciencedirect.com/science/article/pii/S0968808016300106.

Anwar, Nausheen H., Daanish Mustafa, Sarwat Viqar, Amiera Sawas, and Humeira Iqtidar (2014). Urbanization, gender and violence in millennial Karachi: A scoping study. Environment, Politics and Development Working Paper Series #66, Department of Geography. London: King’s College. www.academia.edu/6599931/Urbanization_Gender_and_Violence_in_Millennial_Karachi_A_Scoping_Study.

Bender, C. (2008). London 2012—Displaced by the Olympics. Trialog 98(3): 34–39.

Bhat, P. N. Mari, and S. Irudaya Rajan (1990). Demographic transition in Kerala revisited. Economic and Political Weekly 25(35 and 36): 1957–1980.

Bhide, Amita, Anant Maringanti, and Darshini Mahadevia (2017). Safe and inclusive cities: Pointers towards a research agenda. Economic and Political Weekly 52(7): 39–40.

Brown, Allison (2017). Urban planning and violence: Cause or catalyst for change. Economic and Political Weekly 52(7): 83–90.

Carpenter, R. Charli (2005). “Women, children and other vulnerable groups”: Gender, strategic frames and the protection of civilians as a transnational issue. International Studies Quarterly 49(2): 295–334.

Carpenter, R. Charli (2016). “Innocent women and children”: Gender, norms and the protection of civilians. London: Routledge.

Cernea, Michael M. (1993). The urban environment and population relocation. World Bank Discussion Papers No. 152. Washington, DC: World Bank.

Chant, Sylvia, and Cathy McIlwaine (2016). Cities, slums and gender in the Global South: Towards a feminised urban future. London: Routledge.

Collyer, Michael, Kopalapillai Amirthalingam, and Danesh Jayatilaka (2017). The right to adequate housing following forced evictions in post-conflict Colombo, Sri Lanka. In Katherine Brickell, Melissa Fernández Arrigoitia, and Alexander Vasudevan (Eds.), Geographies of forced eviction (47–69). London: Palgrave Macmillan.

Desmond, Matthew (2012). Eviction and the reproduction of urban poverty. American Journal of Sociology 118(1): 88–133. https://scholar.harvard.edu/files/mdesmond/files/desmond.evictionpoverty.ajs2012.pdf.

Forrest, Ray, and Ade Kearns (2001). Social cohesion, social capital and the neighbourhood. Urban Studies 38(12): 2125–2143.

Fox, Michael A. (2015). Reflections on violence. In A. Fiala (Ed.), The peace of nature and the nature of peace: Essays on ecology, nature, nonviolence, and peace (31–40). Boston, MA: Brill Rodopi.

Galtung, Johan (1969). Violence, peace, and peace research. Journal of Peace Research 6(3): 167–191. www2.kobe-u.ac.jp/~alexroni/IPD%202015%20readings/IPD%202015_7/Galtung_Violence,%20Peace,%20and%20Peace%20Research.pdf.

George, Ann, and Irudaya Rajan (2015). Changing cities and changing lives: Development induced displacement in Kochi, Kerala. Cities People Places: An International Journal on Urban Environments 1(1): 1–18. https://cpp.sljol.info/articles/abstract/10.4038/cpp.v1i1.5/.

Herath, Dhammika, Rajith W. D. Lakshman, and Asela Ekanayaka (2017). Urban re-settlement in Colombo from a wellbeing perspective: Does development-forced resettlement lead to improved wellbeing? Journal of Refugee Studies 30(4): 554–579. https://doi.org/10.1093/jrs/few043.

Hilfinger Messias, DeAnne K. (2011). The health and well-being of immigrant women in urban areas. In Afaf Ibrahim Meleis, Eugenie L. Birch, and Susan M. Wachter (Eds.), Women’s health and the world’s cities. Philadelphia, PA: University of Pennsylvania Press.

Hutchinson, Rebbeca N., Mary A. Putt, Lorraine T. Dean, Judith A. Long, Chantal A. Montagnet, and Katrina Armstrong (2009). Neighborhood racial composition, social capital and black all-cause mortality in Philadelphia. Social Science and Medicine 68(10): 1859–1865. www.ncbi.nlm.nih.gov/pmc/articles/PMC5472423/.

Jayatilaka, Danesh, Rajith W. D. Lakshman, and Iresha M. Lakshman (2018). Urban community profiles: Safe relocation and resettlement in post-war Sri Lanka. In Jennifer E. Salahub, Markus Gottsbacher, and John de Boer (Eds.), Social theories of urban violence in the Global South: Towards safe and inclusive cities (79–95). London and Ottawa: Routledge and International Development Research Centre (IDRC). https://idl-bnc-idrc.dspacedirect.org/bitstream/handle/10625/56926/IDL-56926.pdf.

Kasper, Eric, Gordon McGranahan, Dolf J. H. te Lintelo, Jaideep Gupte, Jean-Pierre Tranchant, Rajith W. D. Lakshman, and Zahrah Nesbitt-Ahmed (2017). Inclusive urbanisation and cities in the twenty-first century. IDS Evidence Report No. 220. Brighton, UK: Institute of Development Studies. https://opendocs.ids.ac.uk/opendocs/handle/123456789/12849.

Kundu, Tadit (2015). Why Kerala is like Maldives and Uttar Pradesh, Pakistan. Live Mint, 17 December. www.livemint.com/Politics/3KhGMVXGxXcGYBRMsmDCFO/Why-Kerala-is-like-Maldives-and-Uttar-Pradesh-Pakistan.html.

Lakshman, Iresha M., Asela Ekanayaka, and Rajith W. D. Lakshman (2016). Going to school from a relocated urban community: Struggling for education within imposed walls. Migration and Development 6(2): 198–214.

Lakshman, Iresha M., Dhammika Herath, Asela Ekanayaka, and M. M. Alikhan (2016). Experiences of a relocated community in Colombo: A community profile of Sinhapura, Wanathamulla. Research Paper. Colombo: International Centre for Ethnic Studies (ICES). http://ices.lk/publications/6002/.

Lakshman, W. D., and Clement A. Tisdell (2000). Sri Lanka’s development since independence: Socio-economic perspectives and analyses. New York: Nova Science Publishers.

Meleis, Afaf Ibrahim, Eugenie L. Birch, and Susan M. Wachter (Eds.) (2011). Women’s health and the world’s cities. Philadelphia, PA: University of Pennsylvania Press.

Momsen, Janet H. (2004). Gender and development. London: Routledge.

Moser, Caroline O. N. (2012). Gender planning and development: Theory, practice and training. London: Routledge.

Moser, Caroline O. N., and Cathy McIlwaine (2006). Latin American urban violence as a development concern: Towards a framework for violence reduction. World Development 34(1): 89–112. www.brookings.edu/wp-content/uploads/2016/06/200601moser.pdf.

Muggah, Robert (2012). Researching the urban dilemma: Urbanization, poverty and violence. Ottawa: International Development Research Centre (IDRC). www.idrc.ca/EN/PublishingImages/Researching-the-Urban-Dilemma-Baseline-study.pdf.

Perlman, Janice E. (2004). Marginality: From myth to reality in the favelas of Rio de Janeiro, 1969–2002. In Ananya Roy and Nezar AlSayyad (Eds.), Urban informality: Transnational perspectives from the Middle East, Latin America, and South Asia (105–146). Lanham, MD: Lexington Books.

Schröder, Ingo W., and Bettina E. Schmidt (2001). Introduction: Violent imaginaries and violent practices. In B. E. Schmidt and I. W. Schröder (Eds.), Anthropology of violence and conflict (6–24). London: Routledge.

Sen, Amartya (2013). The ends and means of sustainability. Journal of Human Development and Capabilities 14(1): 6–20.

te Lintelo, Dolf J. H., and Rajith W. D. Lakshman (2015). Equate and conflate: Political commitment to hunger and undernutrition reduction in five high-burden countries. World Development 76: 280–292. https://doi.org/10.1016/j.world-dev.2015.07.013.

te Lintelo, Dolf J. H., Jaideep Gupte, J. Allister McGregor, Rajith Lakshman, and Ferdous Jahan (2018). Wellbeing and urban governance: Who fails, survives or thrives in informal settlements in Bangladeshi cities? Cities 72: 391–402. https://doi.org/10.1016/j.cities.2017.10.002.

te Lintelo, Dolf J. H., Rajith W. D. Lakshman, Wissam Mansour, Emma Soye, Teo Ficcarelli, and Will Woodward (2018). Wellbeing and protracted urban displacement: Refugees and hosts in Jordan and Lebanon. Brighton, UK; Beirut; and Geneva: Institute of Development Studies, ACTED Lebanon, and IMPACT Initiatives. https://opendocs.ids.ac.uk/opendocs/ds2/stream/?#/documents/3637225/page/1.

Thukral, Enakshi Ganguly (1996). Development, displacement and rehabilitation: Locating gender. Economic and Political Weekly 31(24): 1500–1503.

Umaña, Rodolfo Calderón (2018). Violence and social exclusion in urban contexts in Central America. In Jennifer E. Salahub, Markus Gottsbacher, and John de Boer (Eds.), Social theories of urban violence in the Global South: Towards safe and inclusive cities (99–120). London and Ottawa: Routledge and International Development Research Centre (IDRC). https://idl-bnc-idrc.dspacedirect.org/bitstream/handle/10625/56926/IDL-56926.pdf.

United Nations (2015). Transforming our world: The 2030 Agenda for Sustainable Development. New York: Author. https://sustainabledevelopment.un.org/post2015/transformingourworld/publication.

United Nations (2017). The New Urban Agenda. Quito: Author. http://habitat3.org/the-new-urban-agenda/.

UNDP: United Nations Development Programme (2015). Human development report 2015: Work for human development. New York: Author. http://hdr.undp.org/sites/default/files/2015_human_development_report.pdf.

UN-Habitat (2013). State of women in cities 2012–2013: Gender and the prosperity of cities. Nairobi: Author.

Winton, Ailsa (2004). Urban violence: A guide to the literature. Environment and Urbanization 16(2): 165–184. https://doi.org/10.1177/095624780401600208.

World Health Organization (2002). World report on violence and health. Geneva: Author. http://apps.who.int/iris/bitstream/handle/10665/42495/9241545615_eng.pdf;jsessionid=A2E41214C4AECDAE975F20E0674A07C1?sequence=1.

Zachariah, K. C., and S. Irudaya Rajan (Eds.) (1997). Kerala’s demographic transition: Determinants and consequences. New Delhi: SAGE Publications.


Reflections on how to make cities safer and more inclusive

Jennifer Erin Salahub and Mayssam D. Zaaroura


The violence that exists in and is catalysed by cities comes in myriad forms that are created and recreated in urban spaces across the globe, as the chapters in this volume document. These different types of violence can be public or private, structural, environmental, physical, or psychological. They can disproportionately affect women or men, ethnic groups, the poor, or the middle class. They can be catastrophic such as the demolition of informal housing or commonplace like the street harassment women in every city face daily. They can be interwoven with poverty, social exclusion, and the failure of the state to provide sufficient or appropriate services, including safety and security.

While the chapters in this book are grouped geographically, the challenges our co-authors describe often transcend borders. Yet as multiple forms of violence are ubiquitous, so is resistance and resilience to them. Our co-authors have documented both the types of violence city dwellers are facing and the strategies they implement in coping with them, in addition to the actions urban residents take to respond to violence and prevent its resurgence. Specifically, our co-authors have addressed:

gendered acts of violence

social cohesion and social exclusion

manifestations of structural and infrastructural violence, including forced displacement

interpersonal violence.

In this chapter, we review the different types of violence documented through the Safe and Inclusive Cities (SAIC) initiative, including those written about in this volume and its sister book, Social Theories of Urban Violence in the Global South: Towards Safe and Inclusive Cities (Salahub, Gottsbacher, and de Boer 2018), as well as those identified by SAIC research documented in other works. Through that discussion, we identify interventions and approaches that have worked to reduce violence and consider how they could be applied to different contexts. We focus throughout on the agency of individuals and communities as they demonstrate their resilience in coping with and responding to the varied manifestations of violence in their urban environments.

Forms of violence and resilience

Manifestations of gendered violence

We challenged our co-authors and other SAIC researchers1 to consider and make visible the gendered dimensions of their investigations. The results were frequently compelling and sometimes unexpected. They showed the importance of an intersectional gender analysis, as rarely were women’s experiences homogenous, be it within a neighbourhood, community, city, or country. More often, we saw similarities among women’s and men’s experiences based on shared socioeconomic characteristics rather than their gendered differences, though there are important exceptions.

Young men working their way up through the hierarchy of Abidjan’s informal transport hubs (Kouamé Walter Kra, Chapter 1) share experiences of violence with their counterparts in South Africa’s townships (Richard Matzopoulos, Kim Bloch, Sam Lloyd, Chris Berens, Jonny Myers, and Mary Lou Thompson, Chapter 3) and Brazil’s favelas (Taylor et al. 2016), and—in fact—look to their Brazilian brothers as role models (Akindès 2017). Women fighting against discrimination and (infra)structural violence in India (Darshini Mahadevia and Renu Desai, Chapter 8) share experiences of victimisation with their sisters in Zimbabwe (Stewart, Katsande, and Chisango 2016; Stewart et al. 2018), but their coping mechanisms and approaches to resistance differ.

Experiences of harassment of women in public spaces and physical violence in domestic spheres were—unsurprisingly—near ubiquitous across the cities we studied. While the evidence from all cases confirms the need for further action and feminist activism, it also reveals inspiring and compelling stories of individual and community agency in the face of discouraging odds. Take, for example, the solidarity and resistance against violence exhibited by the members of Territorios Seguros (Secure Territories) in Costa Rica, as described by Juan Pablo Pérez Sáinz, Mario Zetino Duarte, and Florencio Ceballos Schaulsohn in Chapter 6. These community organisers, led by women, refused to allow their social citizenship to be winnowed away by social violence (e.g. conflict among neighbours) or profit-seeking violence (e.g. narco-trafficking). Instead, they worked with their neighbours to build and secure their homes and communities, and promote capacity building through non-violent conflict resolution.

This grassroots-led initiative in Central America contrasts with the success that a state-led programme in South Africa is having in empowering women and reducing violence. The Community Work Programme (CWP) is a community-centred poverty-reduction initiative that is having—mainly positive—gendered impacts on the communities in which it operates (Malose Langa, Themba Masuku, and Hugo van der Merwe, Chapter 4). Most CWP employees are women over 35 years of age and the modest income they receive through the programme, combined with other government grants, is helping lift them and their families above the poverty line. The work they do is largely unskilled, but is helping to make their communities safer, especially when work such as safety patrols and clearing vegetation that impedes night-time visibility on footpaths is prioritised. The study found that there is a risk of increased domestic tension, potentially spilling over into violence, as women gain more relative power in the household via their increased income. However, the authors argue that the benefits outweigh the risks and other programmes in South Africa are working to reduce domestic violence writ large.

This opens the issue to a broader discussion on the need for initiatives that focus on economic empowerment of women and that can work more holistically by combining them with efforts to end violence against women. Whether focusing on the gendered dimensions of this issue or when working to prevent youth violence, economic incentives and empowerment initiatives have to address immaterial components such as social inclusion and efforts to look at social norms that prevent or emerge as a result. The reverse is also true: violence reduction initiatives need to address the challenges of economic exclusion, something that a more holistic approach to these issues would incorporate.

Other countries could learn from these examples. In Ghana, the SAIC research team documented the disproportionate burden of violent crime borne by women, but noted that reporting to state authorities was low among female survivors, likely due to a lack of trust that they will see justice (Adobea Yaa Owusu, Martin Oteng-Ababio, George Owusu, Charlotte Wrigley-Asante, and Martin Wiredu Agyekum, Chapter 2). Owusu and colleagues call for reform within the police service to help women (and men) have more confidence in the police and the wider criminal justice system. We echo this recommendation and point out that approaches such as those in Costa Rica and South Africa, tailored for the Ghanaian context, could yield positive results.

Optimistically, these successes—or pieces of these successful programmes—could even be applied in the most challenging situations. Reducing poverty through gainful and meaningful employment, or creating a modicum of security in concert with neighbours, could go a long way towards addressing the multifaceted challenges that youth in Côte d’Ivoire (Kra, Chapter 1; Akindès 2018) and their counterparts in the Democratic Republic of Congo (DRC) (IDRC 2016) face. That said, more holistic approaches will also be needed to address the interlinking political crisis, poverty, social exclusion, addiction, violence, psychological trauma, and toxic masculinities that characterise their precarious lives.

Social cohesion and social exclusion

Several of the SAIC research teams set out to test the validity and utility of violence reduction theories, developed in the Global North and applied in the Global South.2 Others approached their research questions via different entry points—evaluating an intervention or unpacking a situation that goes against the trend, for example—and identified social cohesion as a key factor in understanding the phenomenon they were studying. Regardless of how they approached the question of social inclusion, exclusion, or cohesion, all of our co-authors pushed our understanding of the nature of social cohesion and exclusion a bit further and helped all of us to better understand how the ties that bind communities together influence the violence and exclusion that different segments of those communities experience. Moreover, they did this using a variety of methods—qualitative, quantitative, participatory, and ethnographic—that complement each other, providing a more robust picture of the situations they investigated.

In South Africa, three research teams came to the idea of social cohesion from very different perspectives. Two of those teams examined the same intervention, Violence Prevention through Urban Upgrading (VPUU).3 One team took a public health approach to understanding the impact of the intervention on violence and other factors, including social cohesion in a suburb of Cape Town. The other saw this as an opportunity to look at how different violence prevention/reduction interventions impact social cohesion and collective efficacy4 in communities by comparing VPUU with Brazil’s Police Pacifying Units (UPP, using its Portuguese acronym) in Rio de Janeiro. Using different approaches to researching the question, Matzopoulos et al. (Chapter 3) and Vanessa Barolsky and Doriam Borges (Chapter 5) come to different conclusions about how VPUU influences social cohesion in Khayelitsha.

From our perspective, these are complementary rather than contradictory results. Matzopoulos et al. (Chapter 3) take a narrow definition of social cohesion and statistically measure it across a random and representative sample, showing that VPUU has a slightly positive impact on those measures of social cohesion. In short, their results tell us that, overall, VPUU is supporting social cohesion. Barolsky and Borges (Chapter 5), on the other hand, approach the question from an ethnographer’s lens to unpack the social networks of the community, including key individuals and groups, in the context of post-colonial, post-Apartheid South Africa. They find that VPUU has not been universally positive when it comes to community experiences of social cohesion and offer suggestions for how it could better integrate into existing social structures to improve its impact and broaden its positive influence.

But VPUU is not the only programme in South Africa influencing social cohesion in urban communities. The CWP also seems to be supporting positive manifestations thereof, for example by helping participants get to know other neighbourhood residents and build trust, work in teams, and meet weekly to discuss priorities for upcoming work (Langa, Masuku, and van der Merwe, Chapter 4). With this newfound camaraderie, CWP participants seem to demonstrate high levels of commitment to improving their communities. Our co-authors also identify the potential for programmes similar to CWP to foster social cohesion in other communities through an inclusive and participatory approach.

While results from SAIC research in South Africa are uplifting, those further north on the continent are less so. In Côte d’Ivoire and the DRC, social exclusion is driving many youths to find the sense of community every human needs in pernicious gangs fuelled by competition, drug abuse, and physical violence (Kra, Chapter 1; IDRC 2016). As Kra traces the trajectory of a young man known as “Z” through the structure that controls Abidjan’s informal transport hub, we gain crucial insight into his unenviable position characterised by limited options for formal or licit employment, minimal schooling, and a trajectory of increasingly violent behaviour.

Outside of the licit market, Z’s skills help him advance in the transport hubs network in and around Cote d’Ivoire’s capital, extorting money from those who transit through it and battling with rival groups. Rather than the profit-seeking violence that Pérez Sáinz, Zetino Duarte, and Ceballos Schaulsohn (Chapter 6) identify in Costa Rica and El Salvador, Z’s experience is presented as a coping mechanism against the structural violence, lack of options, and abandonment by the state that he experiences: this is how he makes a living, in the absence of other options and a social safety net (Kra, Chapter 1). Similarly, Congolese youth have been observed to form family structures within youth gangs in an apparent attempt to regain the social connections they have lost (IDRC 2016).

Quick fixes to reduce these different manifestations of violence and reintegrate youth into safer, healthier communities are few and far between. As the root of Z’s path to violence is structural—a lack of employment opportunities linked to poverty, limited educational opportunities, and a weak social safety net—so must be the cure: functional national and local governance institutions, job creation, education, psychosocial support, and the like. All of these take considerable time, money, political will, and community buy-in to effect change. Researching these topics in locations where research is difficult to conduct—such as these studies in Côte d’Ivoire and DRC—are important first steps to developing durable solutions.

On a slightly more promising note, there are encouraging signs from SAIC research in Latin America, notwithstanding the democratic deficit that Venezuelans are facing at the time of writing (2018–2019). In six urban spaces in Costa Rica and El Salvador where social exclusion is high, three initiatives are helping to build the community and, in turn, reduce violence and insecurity. In Costa Rica, Territorios Seguros and a state-led Community Security Programme are having positive impacts, with some variation in outcomes in the locations where the latter is implemented. In El Salvador, a programme led by the United Nations Development Programme, the Municipal Citizen Security Plans, is yielding promising results in some parts of the country—building the capacity of the community to tackle the dynamics of violence—whereas in others, it is having less of an impact (Pérez Sáinz, Zetino Duarte, and Ceballos Schaulsohn, Chapter 6).

Given the crisis in Venezuela at the time of writing, there is little hope of positive state intervention to reduce violence; indeed, reports point to the state as the perpetrator of considerable violence against its people (see, for example, Amnesty International 2017). However, grassroots activism, particularly that led by women, is helping to pacify even the most violent urban barrios of the country (Roberto Briceño-León, Chapter 7). Through the establishment of norms, pacts, and other types of social institutions, mothers, nuns, female teachers, and others are helping to end the violence that has so negatively impacted them and their families. As the Venezuelan state continues to weaken, these forms of informal social institutionality will become increasingly important as sites of resistance, agency, activism, and hope.

Manifestations of structural and infrastructural violence, including forced displacement

Whether it is due to development or conflict, being forced to leave one’s home brings with it moments of violence. In many cities, efforts to apply a layer of urban planning over the existing organic, but informal, growth of the city can put residents in conflict with the state and with developers. These processes can exhibit both structural and physical violence, with different results depending on how they are implemented. In Kochi, India and Colombo, Sri Lanka (Rajith W. D. Lakshman and S. Irudaya Rajan, Chapter 9), we see gendered differences in satisfaction among those relocated from informal urban settlements—slums—to apartment towers. Generally, women seem to be more content with their new situation. At the same time, both men and women express frustration with where and how they have been moved, including grievances related to a decline in the quality of their social and cultural lives that is not mitigated by an increase in the material quality of their new housing. These findings suggest that governments and developers need to think as carefully about the social implications of urban design as they do about engineering; in other words, they need to consider the immaterial to the same extent as the material when it comes to reducing urban violence. Moving people from single-storey dwellings with permeable lines between public and private spaces into tower blocks of single-unit apartments where neighbours rarely cross paths and social interaction is mainly transactional can backfire, even if the new accommodation represents a move up the property ladder.

Elsewhere in India, Mahadevia and Desai (Chapter 8) explored how the capture of the state by private interests is leaving residents displaced, disappointed, and in danger. Informality is a means for coping with the pressures of urban life: distances travelled for work or school, a lack of affordable housing, and pressures from developers and the state to gentrify parts of the city deemed undesirable. People living in informal settlements, often the result of being displaced—or “invisiblized”, as other Safe and Inclusive Cities researchers in India characterise it (Burte, Kamath, and Joshi 2016: 48)—by state policies that prioritise the profits of developers and fail to provide basic public services such as public lighting, potable water supply, and reliable public transport. Already living on the margins of the city, these people are pushed to the margins of society to eke out their livelihoods. This situation often puts them in competition with other city dwellers and in conflict with the law. Though these conflicts rarely turn physically violent, this structural violence grinds away at individuals and communities and demonstrates a failure of the state to fulfil its obligations, whether in national law or as part of the social contract. The solution to these challenges lies in local action leading to improved governance: it is not that the Indian state or local municipalities lack the capacity to deliver services and treat their citizens humanely; it is that they lack the political will to do so.

This situation is not unique to South Asia. In Zimbabwe (Stewart, Katsande, and Chisango 2016; Stewart et al. 2018), residents displaced from the immediate periphery of the city to suburbs much further away by politically motivated evictions such as Operation Murambatsvina, and sold on the idea of free land in exchange for owner-led development, are learning the hard way that structural violence is iterative and can endure for decades. These mostly female residents, once centrally located, albeit in accommodation designed for single men rather than families, find themselves living on the outskirts of Harare. In their new location, they face rising debt for municipal services they cannot access, legal action for not paying their bills, and threats of eviction because they have not been able to develop an appropriate structure on the land they have been granted. The result is a catastrophic failure of the state to appropriately and humanely implement the country’s basic law. This negligence puts women and the families they often head in precarious living conditions without access to the rudimentary, but reliable, municipal services they previously had. In this case, only governance reform and a political recommitment to apply the resources of the state to the service of its citizens, and with respect for their constitutional rights, will bring about wholesale change. In the meantime, legal professionals and activist groups do what they can to minimise the impact on their clients.

Interpersonal violence

The constellation of categories of violence that we and our co-authors present intersect with one another in dynamic and often catalytic ways. But what about the type of violence that typically comes to mind when we think of “violence”? So far, we have focused on types of violence that connect with urbanisation and the city. Interpersonal violence is one of those and, while it is not uniquely urban, the ways in which it is linked to other urban forms of violence makes it, we think, different from interpersonal violence in rural settings. Rare are the cases where structural violence in cities does not boil over, at some point, into physical violence between individuals. Equally rare are the moments when domestic violence exists in an urban space free from other types of violence. We see interpersonal violence in, for example:

the crime tracked and analysed in Ghana and South Africa: assaults, rapes, and homicides (Owusu et al., Chapter 2; Matzopoulos et al., Chapter 3)

the fights between neighbours in communities in Costa Rica (Pérez Sáinz, Zetino Duarte, and Ceballos Schaulsohn, Chapter 6)

the tactics of control, intimidation, and extortion used by gangs and narco-traffickers in the urban spaces of El Salvador (Pérez Sáinz, Zetino Duarte, and Ceballos Schaulsohn, Chapter 6), Venezuela (Briceño-León, Chapter 7), and Brazil (Taylor et al. 2016)

the attacks and counter-attacks of the unionist and gnambro youth who control Abidjan’s informal transport hubs (Kra, Chapter 1)

the violent crime, including sex crimes, committed by Shegué and Kuluna youth gangs in the slums of Kinshasa, DRC (IDRC 2016)

the repressive approach of the state in Venezuela (Briceño-León, Chapter 7; Briceño-León 2018) and the militarised approach to pacification in Brazil (Barolsky and Borges, Chapter 5)

the gendered violence that sees women disproportionately victimised in their homes and by their intimate partners, and men disproportionately the victims of homicide (see, for example, Anwar et al. 2016; Taylor et al. 2016).

These examples are joined by multiple instances of violence around the world that urban dwellers cope with on a daily basis and that have become so commonplace as to be banal. We argue that because this interpersonal physical violence is inextricable from the other types of urban manifestations of violence described in this volume, the solution lies in addressing the root causes that push conflict over into physical violence, as identified above. At the same time, investments in more immediate solutions are necessary to prevent people from getting hurt. Examples are urban upgrading, poverty-reduction initiatives such as the CWP, non-violent conflict resolution, feminist activism leading to the full empowerment of women and other marginalised groups, and other immediate, on-the-ground interventions that support people in finding alternatives to paths that will lead them to violence.

Conclusions and policy recommendations

Resolving the urban dilemma?

In his literature review for the SAIC initiative, Muggah (2012) identifies robust separate literature on three topics—urbanisation, poverty, and violence—but a paucity of analysis at the nexus of them. He identifies an “urban dilemma”: the challenge of rapid urbanisation fuelling economic growth while at the same time creating sprawling cities, sites of different types of violence and deprivation as governance structures are unable to keep pace. As the International Development Research Centre (Canada) and the Department for International Development (UK) were developing the SAIC initiative, Muggah’s (2012) report reinforced the need for a cross-regional, cross-disciplinary approach to researching—and, hopefully, resolving—the urban dilemma.

Bringing together teams working within and across countries in Latin America, South Asia, and sub-Saharan Africa, the initiative took an interdisciplinary approach to unpacking the linked challenges of rapid urbanisation, poverty, and violence. Looking across the deep dives that each research team took into the dynamics of the cases they studied, we see a number of points that stretch across many, if not all, cases.

First, social cohesion and social inclusion are crucial pieces of the inequality–poverty–urban violence nexus. Violence reduction programmes in the Global South, imported from the North, that fail to take local cultural specificities into consideration miss opportunities to address the roots of violence and different types of inequalities. Crime prevention through environmental design or that seeks to alter physical spaces without also addressing how people, as social beings, interact with that space and with each other in that space likewise offer important but only superficial solutions. SAIC research in West Africa (IDRC 2016; Akindès 2018; Kra, Chapter 1) demonstrates that social exclusion and marginalisation—more than economic poverty—are driving an increase in violent youth gangs, many of which see themselves as “socially dead” (IDRC 2016: 2). As Muggah (2012: 57) notes, policymakers tend to assume “that urban violence is in some ways a failure of youth integration into labor markets and persistent unemployment”. SAIC research tells us that employment opportunities are important, but that the social aspects of feeling emotionally attached to, included in, and having a stake in the future of society are equally significant.

Beyond youth gangs, the importance of social cohesion is echoed in places such as Ghana, where Owusu et al. (Chapter 2) show that the strong social ties within communities of low socioeconomic status provided a protective effect against crime, including violent crime, when compared with those of middle socioeconomic status. (High status communities, though low on social cohesion are able to afford physical protection in ways that their low- and middle-status counterparts cannot.) Similarly, Langa, Masuku, and van der Merwe (Chapter 4) discuss the ways in which the CWP facilitates the development of (positive manifestations of) social cohesion among participants in South Africa and thereby helps to prevent violence. In the same country, Barolsky and Borges (Chapter 5) discuss how interventions can disrupt existing social networks, leaving some community members, who previously felt included, feeling like they are outside of the community. In other work, Barolsky (2016: 21–22) explains the negative manifestations of social cohesion: public violence, gang-imposed social order, and—we would suggest—the performance of toxic masculinity.

Likewise, in Costa Rica, the experience of the members of Territorios Seguros shows the profound effect of communities coming together to work toward a shared objective (Salazar Sanchez and Pérez Sáinz 2015). Elsewhere in Latin America, Briceño-Léon (2018 and Chapter 7, this volume) shows how the institutionalisation of social norms is helping reduce violence in the barrios of Caracas and other Venezuelan cities. Meanwhile, in Bogotá, Lima, and Santiago, Arias and Tocornal Montt (2018) show that differing migration patterns impact community cohesion and, consequently, tension, conflict, and violence in these cities.

Second, the role of eviction, forced displacement, and physical exclusion as drivers of urban violence cannot be underestimated. As Lakshman and Rajan (Chapter 9) and their co-investigators, Jayatilaka, Lakshman, and Lakshman (2018), describe, forced resettlement, whether due to war or urban development, has lasting consequences for urban residents. In some cases, these can be positive: many development-displaced residents in Colombo report being happier with their current accommodation than they were before relocation (Lakshman and Rajan, Chapter 9). However, the stresses on social ties sometimes outweigh the material benefits of improved housing. These feelings are reflected by the residents of Rio de Janeiro, displaced by the development of their favela to create facilities for the 2016 Olympic Games (IDRC 2017a).

Experiences in Zimbabwe present even poorer outcomes, particularly for women displaced by the state from the peri-urban area to much further suburbs (Stewart, Katsande, and Chisango 2016; Stewart et al. 2018). Findings in India connect with experiences in Sri Lanka and Zimbabwe. In Mumbai, Ahmedabad, and Guwahati, those excluded from the urban centre by the collusion of the state and private interests find themselves disconnected from crucial social networks, far from their places of employment or education without access to reliable and efficient public transport, and often in spaces that are poorly lit and maintained. These locations create opportunities for miscreants to prey upon the vulnerable, particularly women and girls (Burte, Kamath, and Joshi 2016; Mahadevia and Desai, Chapter 8).

Finally, SAIC research has made major contributions to understanding gendered violence in cities of the Global South, building on the ground-breaking work of Chant and McIlwaine (2016), identifying it as an additional risk factor for urban violence in addition to those that Muggah (2012: 39) identifies across the literature: “urbanization, urban density, poverty and inequality, and youth bulges”. Three SAIC projects focused specifically on the gendered dimensions of urban violence: in Rio de Janeiro (Brazil) and Maputo (Mozambique); Karachi and Rawalpindi-Islamabad (Pakistan); and Harare, Bulawayo, and Kadoma (Zimbabwe). Yet, many more have incorporated gendered analyses in their research, yielding important insights into how women’s and men’s experiences of urban violence, poverty, and inequality differ.

Unpacking the pervasive and violently toxic masculinities that contribute to the high rate of male homicides in Rio de Janeiro and Maputo, Taylor et al. (2016) and Mariano, Slegh, and Roque (2018) describe, for the first time, the specific pressures that low-income men and boys in particular experience in a context of pervasive urban violence. In Rio de Janeiro, the state, non-state armed groups, and drug traffickers perpetuate this violence. In post-conflict Maputo, men and boys face violence where employment opportunities are scarce and the social safety net is weak. These findings suggest that programmes to help men and women to better understand and deconstruct unhealthy masculinities and femininities can enable them to rebuild healthy ideas of what it means to be a man or woman in all their social roles—protector, breadwinner, caregiver, loving partner. They can also assist in breaking the cycles of violence that see men and women reinforce these unhealthy images and, ultimately, die, albeit in different ways.

In Pakistan, SAIC researchers focused on the access that women and some, largely marginalised, men have to public spaces and their mobility in that space, particularly via public transport. Their research provides new insights into the spaces women can and cannot occupy freely and how control of women in the public and private spheres blurs because of prevailing gender roles in Pakistan (Anwar et al. 2016; Anwar, Viqar, and Mustafa 2018). Similar constraints on women’s and girls’ movements are documented in India by Mahadevia et al. (2014). There, as in Pakistan, women are often subject to the profit-maximising incentives of minibus and auto-rickshaw drivers who offer a public service but work in the private sector. In both cases, insufficient and/or poorly managed publicly run transport, combined with women’s need to move about the city, results in women falling victim to harassment, groping, and worse as they move around the urban space.

In Zimbabwe (Stewart, Katsande, and Chisango 2016; Stewart et al. 2018), as in Sri Lanka, India, and many other contexts, SAIC researchers describe the gendered nature of exclusion, deprivation, predation by the state, and displacement, as described above. Their contributions provide unique insights into the gendered micro-dynamics of exclusion as different agents of the Zimbabwean state—municipal and national, and mainly men—use women’s lives and shelters as pawns in a political power game at the macro level. The state’s failure to fulfil its constitutional duty to the residents of Hatcliffe and Mbare (suburbs of Harare from which residents were displaced) could be attributed to institutional weakness. However, the fact that these women continue to be charged for municipal services that they do not receive and are pursued through legal means for their debts speaks to a more targeted and conscious use of them and their situation to achieve a political end—regardless of the state’s failure to uphold its end of the bargain or its constitutional obligations to protect them.

Options for policy and practice

While each SAIC research team sought to document, understand, and explain the manifestations of violence and inequalities around them, they also aimed to identify ways to prevent them and strategies to end the violence and exclusion that already exist. Four key entry points emerged that could have positive impacts on very different places (IDRC 2017c). These promising practices are:

Community-owned public employment programmes: As Langa, Masuku, and van der Merwe (Chapter 4) discuss, the CWP has enormous potential to leverage a cash-for-work poverty-reduction scheme into an initiative that will build positive manifestations of social cohesion, provide important community services to prevent violence, and address financial stresses that can boil over into interpersonal violence.

Urban upgrading that targets the root causes of violence: Improved public spaces, transit routes, and community infrastructure, as described by Matzopoulos et al. (Chapter 3), can have a significant positive impact on urban violence. But such initiatives must also take into consideration the impact that urban upgrading, and the processes surrounding it, can have on existing social networks and social cohesion (Barolsky and Borges, Chapter 5; Barolsky 2016). Equally important to note is that research is ongoing as to whether the positive impact of VPUU, in this case, is net or if some or all of the reduction in Khayelitsha’s violence has been the result of displacement to surrounding neighbourhoods.

Fostering non-violent masculinities that will make both men and women safer in their urban environments: Through programmes that break down the oppressive gender roles that put both men and women at risk, and using participation to recast masculinities and femininities in ways that are healthier for all (Stewart, Katsande, and Chisango 2016; Taylor et al. 2016; IDRC 2017b; Anwar, Viqar, and Mustafa 2018; Mariano, Slegh, and Roque 2018; Stewart et al. 2018).

Pursuing research and programming that can be gender transformative: Moser (2017) identifies urban safety in public spaces as one of several promising practices that can create space for a renegotiation of gendered power relations, through which women can claim more power or create new power of their own. Drawing on the SAIC research presented above, our co-authors and partners have identified multiple entry points to initiate the structural changes that are needed within society to bring about a transformation in gender relations, resulting in true equality and the end of patriarchal social structures. These include opportunities to recast violent masculinities, initiatives such as Territorios Seguros in Costa Rica, women’s agency in creating safe spaces in Venezuela, and initiatives in Zimbabwe to support suburban women in claiming their constitutional rights and holding a government in transition to account.

Additionally, at least three themes surfaced where the evidence suggests that more work needs to be done to understand how we can help make cities safer. Areas where more evidence and consideration are needed include:

Socially inclusive urban planning: Insight into how to create inclusion through urban planning processes. This would, we believe, require a move beyond consultation to involve marginalised groups unfamiliar with planning processes in decision-making about how their cities are built and rebuilt.

Education, employment, and civic engagement for youth: Youth, both young men and young women, present particular challenges for the urban violence–poverty–inequality nexus, especially in countries experiencing a youth bulge. We see combined efforts to improve education, create meaningful employment opportunities, and show youth that they have a stake in society’s future as crucial to developing the social cohesion and inclusion that SAIC research suggests will prevent violence in cities (IDRC 2017d).

How urbanisation and efforts to prevent and reduce urban violence can contribute to gender transformation: We should be able to leverage the rapid pace of change in urban centres—including social change—to create gender transformative spaces. Research is needed to identify examples of where this has already happened and what lessons can be learned to help initiate/replicate it in other spaces and more broadly.

Although the SAIC initiative has concluded, our co-authors and other research partners continue to study these issues, along with many counterparts. With their dedication, knowledge, and global network, we are confident that the contribution of SAIC research will help overcome the urban dilemma, making cities safer and more inclusive for all.


1SAIC brought together a network of more than 75 researchers based in 19 countries, who investigated different aspects of urban violence, poverty, and inequalities in more than 40 cities in 16 countries. For more information on the initiative, specific research projects, research teams, and the outputs and outcomes of their research, visit www.idrc.ca/cities.

2For further insight into the theories tested and refined by SAIC researchers, see Salahub, Gottsbacher, and de Boer (2018).

3Violence Prevention through Urban Upgrading is an urban community development programme, based in South Africa, that seeks to increase safety through a variety of physical improvements and social interventions.

4Collective efficacy is a community’s social cohesion combined with its willingness to take action for the common good (see, for instance, Barolsky 2016 or Arias and Tocornal Montt 2018).


Akindès, Francis (2017). “Faire l’incroyable”: Paroles aux enfants dits Microbes. [“Do the unbelieveable”: In the words of the children called Microbes] Documentary film produced by Francis Akindès. Abidjan: YouTube. www.youtube.com/watch?v=c8YQZnmmm1U.

Akindès, Francis (2018). Understanding Cote d’Ivoire’s “Microbes”: The political economy of a youth gang. In Jennifer Erin Salahub, Markus Gottsbacher, and John de Boer (Eds.), Social theories of urban violence in the Global South: Towards safe and inclusive cities (161–182). London: Routledge.

Amnesty International (2017). Venezuela: Lethal violence, a state policy to strangle dissent, 10 July. London: Author. www.amnesty.org/en/latest/news/2017/07/venezuela-violencia-letal-una-politica-de-estado-para-asfixiar-a-la-disidencia/.

Anwar, Nausheen H., Daanish Mustafa, Amiera Sawas, and Sharmeen Malik (2016). Gender and violence in urban Pakistan. Karachi: Institute of Business Adminstration.

Anwar, Nausheen H., Sarwat Viqar, and Daanish Mustafa (2018). Intersections of gender, mobility, and violence in urban Pakistan. In Jennifer Erin Salahub, Markus Gottsbacher, and John de Boer (Eds.), Social theories of urban violence in the Global South: Towards safe and inclusive cities (15–31). London: Routledge.

Arias, Enrique Desmond, and Ximena Tocornal Montt (2018). Social disorganization and neighbourhood effects in Latin America. In Jennifer Erin Salahub, Markus Gottsbacher, and John de Boer (Eds.), Social theories of urban violence in the Global South: Towards safe and inclusive cities (121–138). London: Routledge.

Barolsky, Vanessa (2016). Is social cohesion relevant to a city in the global South? A case study of Khayelitsha township. South African Crime Quarterly 55: 17–30.

Briceño-León, Roberto (2018). Urban poverty and institutions in Venezuela. In Jennifer Erin Salahub, Markus Gottsbacher, and John de Boer (Eds.), Social theories of urban violence in the Global South: Towards safe and inclusive cities (139–157). London: Routledge.

Burte, Himanshu, Lalitha Kamath, and Nirali Joshi (2016). Global narrative report. People, places and infrastructure: Countering urban violence and promoting justice in Mumbai, Rio, and Durban. Mumbai: Centre for Urban Policy and Governance. https://idl-bnc-idrc.dspacedirect.org/bitstream/handle/10625/56495/IDL-56495.pdf.

Chant, Sylvia, and Cathy McIlwaine (2016). Cities, slums and gender in the Global South: Towards a feminized urban future. London: Routledge.

IDRC: International Development Research Centre (2016). Urban violence and exclusion in the Democratic Republic of Congo. SAIC Research Results. Ottawa: Author.

IDRC: International Development Research Centre (2017a). Evicted: Forced displacement, exclusion, and violence in the city. Ottawa: Author.

IDRC: International Development Research Centre (2017b). Gender and violence in cities: Rapid growth, entrenched gender identities drive gendered violence in urban spaces. Ottawa: Author.

IDRC: International Development Research Centre (2017c). What works to prevent urban violence: Solutions from Safe and Inclusive Cities. Ottawa: Author.

IDRC: International Development Research Centre (2017d). When community ties lead to violent cities: Exploring social cohesion and mistrust of the state as drivers of urban violence. Ottawa: Author.

Jayatilaka, Danesh, Rajith W. D. Lakshman, and Iresha M. Lakshman (2018). Urban community profiles: Safe relocation and resettlement in post-war Sri Lanka. In Jennifer Erin Salahub, Markus Gottsbacher, and John de Boer (Eds.), Social theories of urban violence in the Global South: Towards safe and inclusive cities (79–95). London: Routledge.

Mahadevia, Darshini, Aseem Mishra, Anurita Hazarika, Yogi Joseph, and Tinam Borah (2014). Women’s safety and transport. Guwahati Policy Brief 1. Ahmedabad: Centre for Urban Equity.

Mariano, Esmeralda, Henny Slegh, and Sílvia Roque (2018). Men in the city: Changing gender relations and masculinities in Maputo, Mozambique. In Jennifer Erin Salahub, Markus Gottsbacher, and John de Boer (Eds.), Social theories of urban violence in the Global South: Towards safe and inclusive cities (32–48). London: Routledge.

Moser, Caroline O. N. (2017). Gender transformation in a new global urban agenda: Challenges for Habitat III and beyond. Environment and Urbanization 29(1): 221–236.

Muggah, Robert (2012). Researching the urban dilemma: Urbanization, poverty and violence. Ottawa: International Development Research Centre (IDRC). www.idrc.ca/sites/default/files/sp/Images/Researching-the-Urban-Dilemma-Baseline-study.pdf.

Salahub, Jennifer Erin, Markus Gottsbacher, and John de Boer (Eds.) (2018). Social theories of urban violence in the Global South: Towards safe and inclusive cities. London: Routledge.

Salazar Sanchez, Karla, and Juan Pablo Pérez Sáinz (2015). Estrategias de Confrontación ante Algunas Formas de Violencia: La experiencia de dos territorios Costarricenses [Confrontation strategies in the face of some forms of violence: The experience of two Costa Rican communities]. In Juan Pablo Pérez Sáinz (Ed.), Exclusión social y violencias en territorios urbanos CentroAmericanos [Social exclusion and forms of violence in Central American urban communities] (145–190). San José: FLACSO.

Stewart, Julie E., Rosalie K. Katsande, and Olga Chisango (2016). Shelter, a home a house or housing? Working paper. Harare: Southern and Eastern African Regional Centre for Women’s Law. www.searcwl.ac.zw/downloads/shelter-home-house.pdf.

Stewart, Julie, and Rosalie Katsande, with Olga Chisango and Sian Maseko (2018). “We don’t know when the trucks will come”: The quest for safe and inclusive cities in Zimbabwe. In Jennifer Erin Salahub, Markus Gottsbacher, and John de Boer (Eds.), Social theories of urban violence in the Global South: Towards safe and inclusive cities (49–62). London: Routledge.

Taylor, Alice Y., Tatiana Moura, J. L. Scabio, E. Borde, J. S. Alfonso, and Gary Barker (2016). This isn’t the life for you: Masculinities and nonviolence in Rio de Janeiro, Brazil. Results from the International Men and Gender Equality Survey (IMAGES) with a focus on urban violence. Washington, DC and Rio de Janeiro: Promundo.


Page numbers in bold denote tables, those in italics denote figures.

18th Street gang 136

Abidjan (Côte d’Ivoire), informal transport hubs 10, 18, 1937, 222, 225; criminalisation of 2833; demonstration of violence in 2326; ex-combatants employed in 3, 28; gnambros 23, 24, 25, 27, 28, 29, 30, 31, 32, 33, 35, 367n8, 37n10, 228; Microbes 3031, 33, 37n9; modernisation of 35; multi-faceted violence and 3032; mystical resources of actors 2122, 33; political actors and 30, 32, 33; power relations 28; solidarity manifested in 31; study methods 1921; threats to users security and mobility 3334, 35; unionists 23, 2425, 27, 2830, 31, 32, 33, 35, 37n10, 228; weapons use 26, 33; women and 24

Accra, Ghana 9, 40, 43, 46, 47, 52, 53, 54, 57, 59

Afghanistan, urban population 2

agency, local 5

Ahmedabad, India 180, 181, 182, 185, 188192, 193, 194195, 196, 230

alcohol 4, 163164

Amirthalingam, Kopalapillai 215

Anwar, Nausheen H. 202

assault 42, 47, 49, 55, 61, 228

Banfield, Edward C. 44

Barolsky, Vanessa 84, 86

Bauer, Bettina 111

Beall, J. 178

Beck, Elizabeth 105

Bender, C. 215

Benin 17

Berk, Laura E. 100

Boateng, Francis D. 5960

Bogotá, Colombia 230

Brazil 222; police killings 6; police, militarisation of 6; public housing programme 6; trust levels 3; see also Rio de Janeiro, Pacifying Police Units (UPP)

Brogden, Mike 118

broken windows theory 4344

Bruce, David 93, 94, 96

Buenos Aires, trust in police 3

Bulawayo, Zimbabwe 231

burglary 42, 47, 48, 49, 54, 56, 57, 188, 189

Campbell, Catherine 98

Cape Town, Violence Prevention through Urban Upgrading (VPUU) programme 10, 6989, 104, 105106, 108118, 124125, 224, 232; background to 111112; community consultation 112113; and community participation 69, 71, 73, 8083, 81, 82, 83, 86

community patrols 115118; contractual relationships 113, 114115, 116, 117; and depression signs and symptoms 69, 71, 73, 7778, 78, 79, 86; and experience of severe interpersonal violence 69, 71, 7273, 76, 7677, 77, 86; formal versus informal areas 77, 78, 7980, 81, 8485, 87; gender factors 76, 78, 8081, 84, 87; homo economicus and 115; and informal traders 114115; and neighbourhood infrastructure satisfaction 69, 71, 73, 7980, 80, 81, 86; and police—community relationships 115118; and Safe Node Area Committee (SNAC) forum 112, 113; and social cohesion 69, 71, 7374, 8386, 84, 85, 8687, 224; study methods 7175, 109110; and violence reduction 7677

capitalism 109

car crime 42, 47, 48, 49, 55, 56, 57

Caracas, Venezuela, trust in police 4

Cassidy, Tali 43, 4445

cease-fire agreements between government and gangs, El Salvador 162163

Center for Epidemiological Studies-Depression Scale (CES-D-10) 73

Centre for the Study of Violence and Reconciliation (CSVR) 90

Chad, urban population 2

Chant, Sylvia 230231

Chicago, Community Action Programme 105

children 3, 100

Chile 3

citizenship, social 136, 138, 139, 145

civic engagement 233

climate change 3, 178

climate risks 2

collective action 9, 147

collective efficacy 104105, 224, 234n4

collective identity 147

Collyer, Michael 215

Colombia 3, 111, 162; police 6; see also Medellin

Colombo, Sri Lanka, gendered characteristics of development-forced displacement and relocation (DFDR) 11, 200, 202, 203, 204, 216, 226, 230; happiness and well-being levels 207, 208, 209, 210, 215; material quality of house index (MQHI) scores 206, 207; sociocultural impact 211, 212214, 216

colonialism, historical legacies of 4, 109

common good 104

community, commitment to 4

community consultation/involvement 125; Cape Town, VPUU programme 112113

community groups 3

community participation, and Violence Prevention through Urban Upgrading 69, 71, 73, 8083, 81, 82, 83, 86

community policing: South Africa 115118; Western models of 118; see also Rio de Janeiro, Pacifying Police Units (UPP)

community responses, Ghana 41, 4953, 54, 60, 61

Community Security Programme, Costa Rica 144, 145, 225226

Community Work Programme (CWP), South Africa 910, 90103, 222223, 224225, 230, 232; cutting grass and community patrols 9395; ex-offenders and anti-crime initiatives 9597; gender dynamics 9799; policy implications 99101; and social cohesion 100101; “Take Back Our Streets” campaign 9495; and Victim Empowerment Programme (VEP) 95

commuting 188

concentric zone theory 61

contagion effects 45

contextual violence 140141, 142, 143, 146, 147, 148

Corbin, Juliet M. 107

corruption 4

Costa Rica 3, 9, 225226, 228

Costa Rica, social exclusion and violence (s) 135, 139; Community Security Programme 144, 145, 225226; dynamics of violence 141142, 143, 148; and education 137, 138; initiatives against violence 143144, 145; and labour market participation 137, 138, 139; and social security 137, 139; Territorios Seguros initiative 143144, 145, 222, 225, 230, 233

Côte d’Ivoire 223; economy 27, 37n11; energy crisis 27; Forces Nouvelles 22, 27; Ouagadougou Political Agreement (2007) 27; poverty 2728, 36n2; Survey on Household Standard of Living 27, 36n2; unemployment 17, 27, 28, 2930, 34; urban population 2; see also Abidjan (Côte d’Ivoire), informal transport hubs

crime 2, 4; avoidance of 3; definitions of 42; ecological/geographical concentration of 4245; housing types and 4243; inequality and 44; neighbourhood factors 44; organised 9, 147; poverty and 9, 4041, 43, 44, 45, 62; reporting of, Ghana 4142, 4950, 5153, 58, 5960, 6061, 223; situational determinants of 44, 58; see also youth crime Crime

Prevention through Environmental Design (CPTED) 4344, 60, 111, 114, 229

crime victimisation see Ghana, crime victimisation

criminalisation, of Abidjan informal transport hubs 2833

cyber attacks 2

Democratic Republic of Congo (DRC) 223, 225, 228; urban population 2

Department for International Development (UK) 7, 229

development-forced displacement and relocation (DFDR), gendered perspective on 200220; happiness and well-being levels 204, 207, 208, 209, 209210, 215; material quality of house index (MQHI) scores 204, 206, 207, 210; methodology and data 203204; perceived security 212, 213214, 215; social networks 212, 214, 215; sociocultural impact 211214, 216

direct violence 200, 201

disempowerment 136; social exclusion and 139140, 148

Dorr-Bremme, Donald W. 107

drug micro markets, Costa Rica 141142

drug trafficking 4

Earls, Felton 104, 105

Ebola crisis 3

economic integration of unemployed 1718; see also Abidjan (Côte d’Ivoire), informal transport hubs

economic risk 2

economy: Côte d’Ivoire 27, 37n11; see also informal economy

education 136, 137, 138, 201, 225, 233

Ekanayaka, Asela 215

El Salvador 9, 111, 225, 228; cease-fire agreements between government and gangs 162; Nationalist Republican Alliance (ARENA) 138; police killings 6; police, militarisation of 6

El Salvador, social exclusion and violence (s) 135, 136, 139; and drug micro markets 140141; dynamics of violence 141, 142143; and education 137, 138; initiatives against violence 144145, 145146; and labour market participation 137, 138, 139; maras (gangs) 136, 141, 142, 143, 147, 148, 149, 150; Municipal Citizen Security Plans 144145, 146, 226; and social security 137, 139

Elias, Norbert 156

employment 223, 225, 229, 233; empowerment through 148; precarity of 136; see also labour market participation

employment programmes 232; see also Community Work Programme (CWP); Expanded Public Works Programme (EPWP)

employment protection and rights 148

empowerment 139140; employment and 148; of women 222223, 228

energy crisis, Côte d’Ivoire 27

ethnography 106108, 126n1

ex-offenders, and Community Work Programme (CWP), South Africa 9596

Expanded Public Works Programme (EPWP), South Africa 9192

family structure 44

FARC (Revolutionary Armed Forces of Colombia) 162

Farmer, Paul 178, 179

favelas see Rio de Janeiro, Pacifying Police Units (UPP)

fear 3

Federal Bureau of Investigation (FBI) 42

feminist activism 222, 228

firearms: availability of 4; see also gun-related violence

Flango, Victor Eugene 44

forced displacement 230; see also development-forced displacement and relocation (DFDR)

Forces Nouvelles (Côte d’Ivoire) 22, 27

formal institutionality 156, 160

Forrest, Ray 215216

Foucault, Michel 156

Fox, Michael A. 201

fragile states 1

Franco, Edgar 121

Frimpong, Louis K. 43

Gafa, B. 43

Galtung, Johan 177, 178, 179, 201

Gambia 17

gangs 3; cease-fire agreements with El Salvador government 162163; services management and delivery 5; see also maras; Microbes; youth gangs

gender 6, 148, 149, 167, 222223, 230231, 231232; and Community Work Programme (CWP), South Africa 9799; and crime victimisation, Ghana 4142, 5455, 56, 57, 58, 59, 6162; and development-forced displacement and relocation see development-forced displacement and relocation (DFDR); discrimination 4; and mobility 202, 231; and reporting of crime 4142, 58, 223; as tool for peace, Venezuela 167169; and VPUU programme, South Africa 76, 78, 8081, 84, 87; see also women

gender and development 202

gender relations, transformation of 232233

George, Ann 215

German Development Bank 105, 109, 111

Ghana 228; urbanisation 40

Ghana, crime victimisation and socioeconomic status (SES) 4042, 43, 4562, 229; community responses 41, 4953, 54, 60, 61; gender dimensions 4142, 5455, 56, 57, 58, 59, 6162; implications for policy and practice 6062; neighbourhood effects 5758; police and 41, 50, 52, 5354, 59, 60; poverty penalty 9, 4041; and reporting crime 4142, 4950, 5153, 58, 5960, 6061, 223; study methods 4547; and types of crime experienced 4142, 47, 48, 49, 5455, 56, 57

Giles, Chris 111

globalisation 109, 178, 179

gnambros (Côte d’Ivoire) 3, 24, 25, 27, 28, 29, 30, 31, 32, 33, 35, 367n8, 37n10, 228

governance 179; see also India, urban planning, governance and structural violence

greenhouse gas emissions 2

Guatemala 142

Guba, Egon G. 107, 126n2, 127n5, 127n6

gun-related violence 49, 57, 62

Guwahati, India 182, 183, 184, 185188, 192193, 194, 195, 196, 230

Guwahati Municipal Corporation (GMC) 187

Habitat III conference (2016) 177

Haiti, urban population 2

Harare, Zimbabwe 231

“hard” (coercive) approaches to security 105

Hasan, Arif 179

health risks 2, 3

Herath, Dhammika 215

Hipp, John R. 43

HIV prevention programmes 86

Hobbes, Thomas 165

homicide 228; police 6

homicide rates 3; Rio de Janeiro 231; South Africa 90

homo economicus 115

Honduras 6, 142

Hong Kong 44

Horn, Philipp 177

housing 201; crime and types of 4243; as driver of structural violence 188189, 193, 194; see also informal housing/settlements; public housing

Human Development Index 202

identity, collective 147

IMAGE (intimate-partner violence intervention) 86

inclusive cities 201

income inequality 44, 70, 201

incommensurability, problem of 109

India 231; Street Vendors (Protection of Livelihood and Regulation of Street Vending) Act (2014) 186187; Urban Land Ceiling and Regulation Act (1976) 194

India, urban planning, governance and structural violence 11, 179197; housing policies 188189, 193, 194; land policies and processes 182, 185188, 194, 195196; public transport 192193, 195; street vending 186187, 194; urban planning Master Plans 178; water supply systems 190191, 194195; women and 189, 192193

inequality 4, 9, 229; and crime 44; income 44, 70, 201; and social exclusion 139, 140; wealth 201; see also gender

inflation trends 44

informal approaches: to violence reduction 1011; of social regulation 125

informal economy 10, 1718; unemployment and 1718; see also Abidjan (Côte d’Ivoire), informal transport hubs

informal housing/settlements 186, 188, 190, 194, 207, 215216, 221

informal institutionality 156157; see also Venezuela, informal institutionality

infrastructural violence 226227; see also development-forced displacement and relocation (DFDR)

infrastructure: inadequate 2, 3; satisfaction with, and VPUU 69, 71, 73, 7980, 80, 81, 86

institutional crime prevention 111

institutionality: formal 156, 160; see also informal institutionality institutions, trust in 34

interconnectedness of cities 12

interdependence of cities 12

International Development Research Centre (Canada) 7, 229

International Organization for Migration 3

interpersonal violence 69, 71, 7273, 76, 7677, 77, 86, 227228; see also assault; homicide; intimate-partner violence; rape

intimate-partner violence 86, 228

Iraq, urban population 2

Islamabad, Pakistan 231

Jawaharlal Nehru National Urban Renewal Mission (JNNURM) 180

Jayatilaka, Danesh 215

Kadoma, Zimbabwe 231

Karachi, Pakistan 231

Kasper, Eric 201

Kearns, Ade 215216

Khayelitsha township see Cape Town, Violence Prevention through Urban Upgrading (VPUU) programme

kidnapping 47, 48, 55, 56

Kinshasa, DRC 228

Kochi, India, gendered characteristics of development-forced displacement and relocation (DFDR) 11, 200, 202, 203204, 216, 226; happiness and well-being levels 207, 208, 209, 210, 215; material quality of house index (MQHI) scores 206, 207; sociocultural impact 211, 213, 214, 216

Krishak Mukti Sangram Samiti (KMSS) 186

Kumasi, Ghana 9, 40, 46, 47, 52, 53, 54, 55, 57

La Paz, Bolivia, trust in police 4

labour market participation 136137, 138, 139

Lakshman, Rajith W.D. 215

land policies and processes, and structural violence, India 182, 185188, 194, 195196

land tenure 201

Latin America: mano dura approach 6, 170; trust levels 34; see also individual countries

Lefebvre, Henri 114

Lima, Peru 230; trust in police 4

Lincoln, Yvonna S. 107, 126n2, 127n5, 127n6

machismo/macho culture 167, 179

McIlwaine, Cathy 43, 230231

McKay, Henry D. 104, 125

Madagascar, urban population 2

Magaloni, Beatriz 121

Mahadevia, Darshini 231

malaria 3

Mali, urban population 2

mano dura approach 6, 170

Maputo, Mozambique 231

Mara Salvatrucha 136

maras (gangs), El Salvador 136, 141, 142, 143, 147, 148, 149, 150

Mariano, Esmeralda 231

masculinities: non-violent 232; toxic 223, 231; see also machismo/macho culture

Medellin, Colombia 43, 59; social urbanism plan 6

Melo, Vanessa 121

Mexico, Oportunidades (conditional cash transfer programme) 6

Microbes 3031, 33, 37n9

migration 2, 3, 202; see also development-forced displacement and relocation (DFDR)

militarisation of police 6

militia 5

mobility—gender intersection 202, 231

Moser, Caroline 43, 177, 232233

Mourides 17

Mozambique 231

MtWana, Nontutuzelo T. 117

Muggah, Robert 105, 229, 231

Municipal Citizen Security Plans, El Salvador 144145, 146, 226

murder see homicide Myanmar, urban population 2

Myers, Garth 115

Nationalist Republican Alliance (ARENA), El Salvador 138

natural disasters 2

neighbourhood factors 44, 216

neighbourhood watch programmes 5051, 60, 70, 115118

neighbouring 216

neo-statism 179, 195

neoliberalism 178

New Takoradi, Ghana 43

New Urban Agenda 177, 200

New York City 44

Nicaragua 6

Nigeria 17

norms 2, 104, 106, 107, 113, 118, 124, 156157, 223, 226, 230

nuns 168169, 226

occupancy urbanism 178, 182

O’Donnell, Guillermo 165

offenders, reintegration of 146, 149, 150

Ohmer, Mary L. 105

Oliveira, Marcos 43, 44

Operation Murambatsvina, Zimbabwe 227

Oportunidades (conditional cash transfer programme), Mexico 6

organised crime 9, 147

Oteng-Ababio, Martin 43

Owusu, George 43

Pacifying Police Units see Rio de Janeiro, Pacifying Police Units (UPP)

Pakistan 231

pandemics 2, 3

Paraguay 3

Patino, Jorge E. 4344

patriarchy 4, 59, 233

Pellizoni, Luigi 109

Perlman, Janice E. 215

Peru 3; see also Lima

Philip, Kate 92

Piper, Laurence 112

place effects 44, 59

police: Cape Town, and VPUU programme 115118; Ghana 41, 50, 52, 5354, 59, 60, 223; killings 6; militarisation of 6; proximity 119; trust in 34; see also community policing; Rio de Janeiro, Pacifying Police Units (UPP)

political economy of violence 6

political risk 2

political violence, Abidjan, informal transport hubs 30

pollution 2

population growth 1

poverty 3, 4, 9, 229; concentration 45; Côte d’Ivoire 2728, 36n2; and crime 9, 4041, 43, 44, 45, 62

power relations 28, 139140

private security 3, 45, 58

profit-seeking violence 140, 141, 142, 147, 148, 149, 222

property crime 42, 49, 62

Proximity Policing 119

public health approaches 4, 224

public housing 5; Brazil 6; India 194

public lighting 192193, 227, 230

public security, logics of 145146

public space: access to 231; safety in 232233; symbolic rescue of 165167

public transport 192193, 195, 227, 230, 231

Rajan, Irudaya 215

Rao, Vyjayanthi 109

rape 42, 47, 49, 55, 57, 61, 228

rattrapage 31

Raudenbush, Stephen W. 104, 105

reciprocity, ethic of 125

reintegration of offenders 146, 149, 150

religion 147148, 168169

reporting of crime, Ghana 4142, 4950, 5153, 58, 5960, 6061, 223

repression, logic of 145146, 149, 150

Revolutionary Armed Forces of Colombia (FARC) 162

Rio de Janeiro 215, 230, 231; homicide 231; Special Police Operations Battalion (BOPE) 120

Rio de Janeiro, Pacifying Police Units (UPP) 10, 104, 105, 106, 118124, 125, 224; background of 119121; changes as result of 121; impact on social cohesion 122124; study methods 118119

risks facing twenty-first century cities 14

robbery 42, 49, 62, 188, 189

Roque, Silvia 231

Roy, Ananya 108109

rule of law 155, 160, 161, 170

rules of the city 156

Rylko-Bauer, Barbara 178, 179

Safe and Inclusive Cities (SAIC) initiative 7, 229

Salvatore, Christopher 43

Sampson, Robert J. 104, 105

Santiago, Chile 230

Scared Straight campaign, USA 9697

Seattle 43, 44

security: private 3, 45, 58; public, logics of 145146

Sekondi-Takoradi, Ghana 9, 40, 46, 47, 53, 54, 59

self-help community groups 3

Senegal 17

séphonisme 31

services: access to 5; gangs/militia and delivery of 5

Sharkey, Patrick 4

Shaw, Clifford R. 104, 105, 125

Sherbenou, Edgar L. 44

Simpson, J. 42

Singapore 44

situational crime prevention 111

situational determinants of crime 44, 58

size of cities 44

skill: sociological model of 1819; violence as 19

Slegh, Henny 231

social citizenship 136, 138, 139, 145

social cohesion 3, 4, 44, 104105, 118, 125, 148, 224226, 229230, 232; and Community Work Programme (CWP), South Africa 100101; and Pacifying Police Units (UPP), Rio de Janeiro 122124; and Violence Prevention through Urban Upgrading (VPUU) 69, 71, 7374, 8386, 84, 85, 8687, 224

social crime prevention 111

social disorganisation theory 104, 105, 124

social exclusion 136, 224226, 229; and (dis)empowerment 139140; and inequality 139, 140; see also Costa Rica, social exclusion; El Salvador, social exclusion

social inclusion/exclusion, scales of 136, 137

social networks 212, 214, 215, 230, 232

social security 136, 137, 139

social urbanism plan, Medellin, Colombia 6

social violence 140, 141, 143, 148149, 222

socioeconomic status (SES) see Ghana, crime victimisation and socioeconomic status (SES)

“soft” (voluntary) approaches to security 105

Sohn, Dong-Wook 43

Somalia, urban population 2

SOTRA 3334

South Africa 222, 228; community policing 115118; crime and violence levels 70, 90; Department of Cooperative Governance and Traditional Affairs (COGTA) 90; Expanded Public Works Programme (EPWP) 9192; IMAGE (intimate-partner violence intervention) 86; income inequality 70; murder rate 90; National Development Plan 70; Stepping Stones (HIV prevention programme) 86; unemployment 91; Victim Empowerment Programme (VEP) 95; White Paper on Safety and Security (2016) 69, 70; see also Cape Town, Violence Prevention through Urban Upgrading (VPUU) programme; Community Work Programme (CWP); Western Cape Province

space: ownership of 114; as socially produced 114; see also public space

Sri Lanka see Colombo, Sri Lanka

Staff, Jeremy 96

Stanford Center for International Development 121

state violence 146, 226

state-building 1

Steinberg, Jonny 118

Stepping Stones (HIV intervention programme) 86

Strauss, Anselm 107

street lighting 192193, 227

street vendors, India 186187, 194

Street Vendors (Protection of Livelihood and Regulation of Street Vending) Act (2014), India 186187

structural violence 177178, 201, 225, 226227; see also development-forced displacement and relocation (DFDR); India, urban planning, governance and structural violence

Sudan, urban population 2

Sustainable Development Goals (SDGs) 3, 5, 177

Syndicat national des transporteurs de marchandises en Côte d’Ivoire (SNTM-CI) 28

Tamale, Ghana 9, 40, 46, 47, 4950, 5253, 54, 57, 59

Taniguchi, Travis A. 43

target hardening 45, 58, 60

Taylor, Alice Y. 231

teachers, female 168, 226

territoriality 114

Territorios Seguros initiative, Costa Rica 143144, 145, 222, 225, 230, 233

terrorist threats 2

theft 47, 48, 55, 56

toxic masculinities 223, 231

transport sector see Abidjan (Côte d’Ivoire), informal transport hubs; public transport

trust: in institutions 34; intercommunal 4; interpersonal 3, 4

Ubuntu 112, 113

Uggen, Christopher 96

UN-Habitat 44

unemployment 1718, 148, 229; Côte d’Ivoire 17, 27, 2930, 34; informal economy and 1718; responses to 1718; South Africa 91; youth 4, 17; see also Abidjan (Côte d’Ivoire), informal transport hubs; Community Work Programme (CWP); labour market participation

United Nations Development Programme (UNDP) 144, 202, 226

United States (US), Scared Straight campaign 9697

Unnayan Samitis 186

urban dilemma 229

urban fabric theory 74

Urban Land Ceiling and Regulation Act (1976), India 194

urban planning 226227, 233; see also India, urban planning, governance and structural violence

urban upgrading see Violence

Prevention through Urban Upgrading (VPUU)

urbanisation 1, 43, 109, 124, 229; Ghana 40

Valenzuela, José Manuel 142

Venezuela 230, 233; gender dynamics as a tool for peace 167169; macho culture 167; state violence 226; trust in police 4; urban structure of the city and institutionality 157159

Venezuela, informal institutionality 10, 159167, 169170, 226; cease-fire agreements 162163; grey zone of institutionality 164165; invisible borders of crime 161162; sale of alcohol 163164; symbolic rescue of public space 165167

Victim Empowerment Programme (VEP), South Africa 95

victimisation 4; of women 167; see also Ghana, crime victimisation

vigilantism 3, 116

violence: contextual 140141, 142, 143, 146, 147, 148; definitions of 42, 177, 200201; direct 200, 201; and power 139140; profit-seeking 140, 141, 142, 147, 148, 149, 222; social 140, 141, 143, 148149, 222; state 146, 226; see also infrastructural violence; interpersonal violence; structural violence

Violence Prevention through Urban Upgrading (VPUU) see Cape Town, Violence Prevention through Urban Upgrading (VPUU) programme

voluntarism 117

war, historical legacies of 4

Warner, Barbara D. 105

water supply 190191, 194195, 227

Watson, Vanessa 108

Weiner, E. 42

Western Cape Province, Integrated Provincial Violence Prevention Policy Framework 69, 70

Wirth, Louis 43, 156, 177

women 222223, 230; avoidance of violence 167168; and Community Work Programme (CWP), South Africa 97, 98, 99; development-forced displacement and relocation (DFDR)-affected 201202, 205, 207, 208, 210, 211, 213214, 216, 226; empowerment of 222223, 228; and informal transport hubs, Abidjan 24; insecurity of, Indian cities 11, 189, 192193; migrant 202; mobility 231; teachers 168, 226; Territorios Seguros intitiative, Costa Rica 143, 222, 233; as agents of peace, Venezuela 167169, 226; victimisation of 167, 228 (Ghana 5455, 56, 57, 58, 59, 61); see also gender

World Health Organization (WHO) 42, 177, 200

Wrigley-Asante, Charlotte 43

Yates, Daniel K. 43

Yemen, urban population 2

Yorubas 17

youth 233

youth crime 43, 44, 45, 52

youth gangs 225, 228, 229; cease-fire agreements 162163; see also maras; Microbes

youth unemployment 4, 17

Zika virus 3

Zimbabwe 227, 230, 231, 233; gendered dynamics of exclusion 231232; Operation Murambatsvina 227

Zuma, Jacob 92