Conflict and Fragility: Investing in Security: A GLOBAL ASSESSMENT OF ARMED VIOLENCE REDUCTION INITIATIVES
Conflict and Fragility
Investing in Security
A GLOBAL ASSESSMENT OF ARMED VIOLENCE
REDUCTION INITIATIVES
GENEVA
DECLARATION
This work is published on the responsibility of the Secretary-General of the
OECD. The opinions expressed and arguments employed herein do not
necessarily reflect the official views of the Organisation or of the governments of
its member countries ; or those of the Geneva Declaration on Armed Violence
and Development or the United Nations Development Programme (UNDP).
ISBN 978-92-64-12453-0 (print)
ISBN 978-92-64-12454-7 (PDF)
Series: Conflict and Fragility
ISSN 2074-3645 (print)
ISSN 2074-3637 (online)
Photo credits: Cover © Franco Bosetti/Dreamstime.com.
© OECD 2011
You can copy, download or print OECD content for your own use, and you can include excerpts from OECD
publications, databases and multimedia products in your own documents, presentations, blogs, websites and
teaching materials, provided that suitable acknowledgment of OECD as source and copyright owner is given.
All requests for public or commercial use and translation rights should be submitted to
rights@oecd.org.
Requests for permission to photocopy portions of this material for public or commercial use shall be addressed
directly to the Copyright Clearance Center (CCC) at
info@copyright.com or the Centre français d'exploitation du
Please cite this publication as:
OECD (2011), Investing in Security: A Global Assessment of Armed Violence Reduction Initiatives,
Conflict and Fragility, OECD Publishing.
INVESTING IN SECURITY: A GLOBAL ASSESSMENT OF ARMED VIOLENCE REDUCTION AND PREVENTION INITIATIVES – © OECD 2011
ACKNOWLEDGEMENTS – 3
Acknowledgements
This report was edited by Robert Muggah and Achim Wenmann
and reflects the inputs of many researchers and research teams. Authors
responsible for country mappings include: Fernando Malta, Robert Muggah,
Alessandra Oberling, Ilona Szabo de Carvalho, and Monica Viceconti for
Brazil; Julie Abbass for Burundi; Katherine Aguirre Tobón, David Correal,
Pamela Góngora Salazar, Benedict Hayes, Juan Masullo, Santiago Millán,
Claudia Navas, Jorge A. Restrepo, Miguel Ángel Rincón, and Alonso Tobón
García for Colombia; Johenneso Dahn, Dariusz Dziewanski, Jonathan Fayiah,
Ezekiel Freeman, Obi Joe, Zoema Kargbo, Christine Lang, and Sebastian
Taylor for Liberia; David Bruce, Adèle Kirsten, Themba Masuku for South
Africa; and Philipp Stucki for Timor-Leste.
The final programme database, analysis and accompanying figures were
developed by the Conflict Analysis Resource Centre (CERAC) with support
from the Small Arms Survey (SAS). In addition to inputs provided by core
research team members, comments were gratefully received from Erwin
van Veen (OECD DAC), Paul Eavis (UNDP BCPR), Zachary Taylor (UNDP
BCPR), Jaco Beerends (Ministry of Foreign Affairs, the Netherlands) and
the members of the OECD DAC Armed Violence Advisory Group. Credit is
also due to member states of the Geneva Declaration on Armed Violence and
Development – specifically the core group – for their substantive contribution.
Finally, this publication would not have been possible without the generous
financing of the UNDP and the Governments of Norway, Switzerland and
the UK.
INVESTING IN SECURITY: A GLOBAL ASSESSMENT OF ARMED VIOLENCE REDUCTION AND PREVENTION INITIATIVES – © OECD 2011
TABLE OF CONTENTS – 5
Table of contents
Executive summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .11
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .17
Chapter 1. Conceptualising armed violence reduction and prevention . . . . . . 21
Conceptual framework. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
Introducing the typology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
Promising AVRP initiatives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
Chapter 2. Mapping armed violence reduction and prevention
programming trends . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .31
Chapter 3. Case study summaries. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
Brazil . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
Burundi . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
Colombia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
Liberia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
South Africa . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
Timor-Leste . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
Conclusion. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75
INVESTING IN SECURITY: A GLOBAL ASSESSMENT OF ARMED VIOLENCE REDUCTION AND PREVENTION INITIATIVES – © OECD 2011
6 – TABLE OF CONTENTS
Figures
Figure 1.1 Categorising AVRP activities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
Figure 2.1 "Direct" and "indirect" programming in six case studies, 1990-2010 . . 33
Figure 2.2 Evolution of programming in selected cases, 1990-2012 . . . . . . . . . . 34
Figure 2.4 Budget ranges of AVRP programming . . . . . . . . . . . . . . . . . . . . . . . . 36
Figure 2.3 Time horizons of AVRP programming . . . . . . . . . . . . . . . . . . . . . . . . 36
Figure 2.5 Gender dimensions of all AVRP programmes . . . . . . . . . . . . . . . . . . 38
Figure 2.6 AVRP monitoring and evaluation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
Figure 3.1 Types of implementing agencies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
Figure 3.2 Most prominent risk factors addressed through "indirect"
programming . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
Figure 3.3 Types of armed actors targeted by "direct" programming . . . . . . . . . 52
Figure 3.4 Main risk factors addressed by "indirect" programming . . . . . . . . . . 53
Figure 3.5 Specific instruments of armed violence . . . . . . . . . . . . . . . . . . . . . . . 57
Figure 3.6 "Direct" intervention strategies in relation to institutions . . . . . . . . . 57
Figure 3.7 Types of proximate risk factors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
Figure 3.8 Types of risk factors targeted by "indirect" programmes . . . . . . . . . . 61
Figure 3.9 Liberia: Budgets for interventions . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
Figure 3.10 Types of funders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
Figure 3.11 South Africa: Main risk factors addressed through "indirect"
programming . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
Figure 3.12 Types of risk factors addressed by "indirect" programming . . . . . . . 69
Tables
Table 1.1 AVRP programming typology. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
Table 2.1 Most common types of armed violence addressed across all cases . . 35
Table 2.2 Direct AVRP interventions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
Table 2.3 Most frequently cited risk factors . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
Table 2.4 "Indirect" AVRP programmes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
Table 3.1 What types of armed violence do your programmes address? . . . . . . 47
Table 3.2 Most common "direct" and "indirect" AVRP interventions in Brazil. . . 47
Table 3.3 Most common "direct" and "indirect" AVRP interventions in
Burundi . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
Table 3.4 Most common "direct" and "indirect" AVRP interventions in
Colombia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
Table 3.5 Most common "direct" and "indirect" AVRP interventions in Liberia . . 61
Table 3.6 Most common "direct" and "indirect" AVRP interventions in South
Africa . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
Table 3.7 South Africa: Annual budget range of programmes (USD) . . . . . . . . 66
Table 3.8 Most common "direct" and "indirect" interventions in Timor-Leste . . 68
Table 3.9 Types of funders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
INVESTING IN SECURITY: A GLOBAL ASSESSMENT OF ARMED VIOLENCE REDUCTION AND PREVENTION INITIATIVES – © OECD 2011
TABLE OF CONTENTS – 7
Boxes
Box 0.1 What is armed violence? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .18
Box 1.1 Promising practice in Burundi. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
Box 1.2 Promising practice in Brazil . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
Box 1.3 Promising practice in South Africa . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
Box 3.1 Denouncing crime in Brazil . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
Box 3.2 Pacification police in Rio de Janeiro . . . . . . . . . . . . . . . . . . . . . . . . . . 45
Box 3.3 Youth AVRP in Brazil . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
Box 3.4 Addressing youth violence before it happens in Colombia . . . . . . . . . 55
Box 3.5 Ensuring adequate reintegration as part of DDRR . . . . . . . . . . . . . . . 59
Box 3.6 Armed violence prevention through employment . . . . . . . . . . . . . . . . 59
Box 3.7 Addressing reintegration of IDPs for peace. . . . . . . . . . . . . . . . . . . . . 70
INVESTING IN SECURITY: A GLOBAL ASSESSMENT OF ARMED VIOLENCE REDUCTION AND PREVENTION INITIATIVES – © OECD 2011
ABBREVIATIONS – 9
Abbreviations
AFL Armed Forces of Liberia
AVRP Armed violence reduction and prevention
BCPR Bureau for Crisis Prevention and Recovery
CAVR Comisção Acolhimento, Verdade e Reconcil
(Commission for Reception, Truth and Reconciliation)
CTDT Commission technique de désarmement des civils et de lutte
contre la prolifération des armes légères et de petit calibre
(Technical Commission for Civilian Disarmament and the
Fight Against the Proliferation of Small Arms and Light
Weapons)
DAC Development Assistance Committee
DDR Disarmament, Demobilisation and Reintegration
F-FDTL Falintil-Forças de Defesa de Timor-Leste
(Timor-Leste Defence Force)
GDP Gross domestic product
IADB Inter-American Development Bank
IDP Internally displaced person
INCAF International Network on Conflict and Fragility
JSSR Justice and security sector reform
LNP Liberian National Police
M&E Monitoring and evaluation
NGO Non-governmental organisation
OECD Organisation for Economic Co-operation and Development
OMC One Man Can Campaign
INVESTING IN SECURITY: A GLOBAL ASSESSMENT OF ARMED VIOLENCE REDUCTION AND PREVENTION INITIATIVES – © OECD 2011
10 – ABBREVIATIONS
PRONASCI Programa Nacional de Segurança Pública com Cidadania
(National Programme for Public Citizen Security)
PRSP Poverty reduction strategy paper
SGBV Sexual and gender-based violence
SSR Security system reform
TRC Truth and reconciliation commission
UNDP United Nations Development Programme
UNICEF United Nations Children's Fund
UNMIT United Nations Mission in Timor-Leste
UNOB United Nations Mission in Burundi
UNODC United Nations Office for Drugs and Crime
UPP Unidade de Polícia Pacificadora
(Pacification Police Units)
WHO World Health Organization
INVESTING IN SECURITY: A GLOBAL ASSESSMENT OF ARMED VIOLENCE REDUCTION AND PREVENTION INITIATIVES – © OECD 2011
EXECUTIVE SUMMARY – 11
Executive summary
Conservative estimates indicate that at least 740 000 men, women, youth
and children die each year as a result of armed violence, most of them in
low- and medium-income settings (Krause, Muggah, Wenmann 2008). The
majority of these deaths occur in situations other than war, though armed
conflicts continue to generate a high incidence of casualties. Approaches
to preventing and reducing these deaths and related suffering are becoming
increasingly important on the international agenda. The United Nations (UN)
Secretary General (2009) and UN General Assembly (2008) highlighted the
relationships between armed violence and under-development and various
high-level diplomatic processes are drawing more attention to promising
solutions. In spite of the global preoccupation with the costs and consequences
of armed violence, comparatively little evidence exists about how to stem its
risks and effects. Virtually no information is available on armed violence
reduction and prevention (AVRP) interventions, much less their effectiveness.
This report aims to fill this gap. It seeks to generate more understanding
of what works and what does not when it comes to armed violence reduction
and prevention (AVRP), to stimulate further evaluation and to contribute to
more effective and efficient policies and programmes. The report is based on a
large-scale mapping of AVRP activities around the world, focusing primarily
on programming trends in six countries – Brazil, Burundi, Colombia, Liberia,
South Africa and Timor-Leste. These countries represent the very different
programming contexts – from high rates of urban criminal violence to
protracted post-conflict insecurity – in which development practitioners are
currently engaged. While offering new data and analysis, this assessment builds
directly on the report Armed Violence Reduction – Enabling Development
produced by the Organisation for Economic Co-operation and Development's
International Network on Conflict and Fragility (OECD, 2009a).
An important evolution of AVRP programming in all six countries over
the past decade was detected. Approximately two-thirds of all armed violence
prevention and reduction activities reviewed in Brazil occurred between 2005
and 2010. Likewise, in Burundi, Colombia, Liberia, and Timor-Leste, nearly all
initiatives began after 2005. Not only does the report highlight the importance
INVESTING IN SECURITY: A GLOBAL ASSESSMENT OF ARMED VIOLENCE REDUCTION AND PREVENTION INITIATIVES – © OECD 2011
12 – EXECUTIVE SUMMARY
of internationally-mediated peace processes and security promotion efforts as
important entry points for preventing and reducing violence, it highlights the
significant investments made by national governments and non-governmental
organisations (NGOs) in more developmental approaches to AVRP over the
past decade.
This report draws attention to the experimentation and innovation of
AVRP initiatives. It finds that many actors are already actively engaged in
"direct" and "indirect" AVRP activities, even if they label their initiatives
by a different name. Many different practical approaches are used in AVRP
activities to achieve the common objective of improving safety and security.
Not only are the defence, police and justice sectors involved, but also specialists
involved in urban planning, population health, tertiary and secondary education
and youth programming. What many have in common is the experience of
pursuing common comprehensive interventions to improve safety and security.
Implementing agencies are similarly varied, ranging from multilateral and
bilateral agencies to governments, NGOs and private organisations engaged
in relief, development and social entrepreneurship. The most promising AVRP
activities are forged on the basis of inter-sectoral partnerships and evidencebased
approaches, and operate simultaneously at the local and national levels.
Key findings
The report offers a rich, empirical overview of the diversity and scope
of armed violence reduction and prevention efforts. Specific observations
include:
Considerable variation in the types of violence addressed by AVRP
interventions: AVRP activities are not restricted to preventing and reducing
violence associated with armed conflict or crime alone. Overall, the global
mapping registered more than 20 separate categories of armed violence in
which actors were involved, with some interventions focused on more types
of violence than others. This reflects both the dynamic nature of armed
violence in low- and medium-income countries, and also the diverse range of
programming options on the ground.
The "armed violence" label is not always recognised nor uniformly
applied by practitioners in low- and medium-income settings: "Direct" and
"indirect" AVRP interventions range from public and citizen security and crime
prevention to conflict prevention, peacebuilding, pacification and community
policing. This linguistic and programmatic diversity is to be encouraged since
it reflects local histories, cultures, and social realities. Multilateral and bilateral
policy makers and practitioners must therefore be attentive to the semantics of
armed violence when designing their interventions with local partners.
INVESTING IN SECURITY: A GLOBAL ASSESSMENT OF ARMED VIOLENCE REDUCTION AND PREVENTION INITIATIVES – © OECD 2011
EXECUTIVE SUMMARY – 13
There is a considerable overlap between "direct" and "indirect" AVRP
programming: Most organisations involved in AVRP claim to be pursuing
predominantly "indirect" programming, focused on mitigating proximate and
structural risk factors through education, employment and targeted development
programming at "at-risk" groups. A smaller proportion claim to be addressing
the instruments, actors or enabling institutions of armed violence "directly", via
legislative initiatives to regulate and control firearms, working with gangs and
collecting weapons from former combatants and civilians. Many organisations
blend the two programming approaches.
A significant proportion of AVRP interventions seek to prevent and
reduce collective and inter-personal violence, particularly violence against
women: Both the assessment of large, development agency AVRP databases
and the findings generated from the survey highlight the importance attached
by programmes to reducing sexual and gender-based violence. Interventions
address "at-risk" male youth and perpetrators through a combination of
activities emphasising education, strengthening social and family networks,
and employment. A comparatively smaller range of activities target female
victims and survivors.
The global AVRP agenda is biased in favour of actions endorsed and
supported by international agencies and national governments: A review
of existing inventories of AVRP activities reinforces the incorrect perception
that most activity is supported by international actors, public authorities, and
non-governmental organisations, or takes place exclusively in upper-income
settings. The persistent bias in the mainstream literature on armed violence
prevention and reduction underlines a gap in the identification, analysis and
evaluation of cross-border, sub-national, metropolitan, community-based
and grass-roots activities, especially in lower- and medium-income contexts.
Multilateral and bilateral support for AVRP programmes appears to be
most common in low-income, post-conflict contexts, while national, public
authority-led and NGO efforts are more common in medium-income, crimeaffected
settings: The report detects more international and donor government
agencies operating in post-conflict settings as compared to other non-war
environments. Important exceptions are the Inter-American Development Bank
(IADB) and the United Nations Development Programme (UNDP). Both support
integrated, citizen-security approaches in countries affected by high homicide
and victimisation rates, but not necessarily armed conflict.
Recent, high-level policy engagement in armed violence is supported by
two decades of relevant AVRP programming experience: There is nothing new
about addressing AVRP as part of wider development aid programmes. Although
not necessarily described as armed violence prevention or reduction per se, a vast
array of interventions has emphasised conflict prevention, peacebuilding and
wider security and safety priorities since the early 1990s.
INVESTING IN SECURITY: A GLOBAL ASSESSMENT OF ARMED VIOLENCE REDUCTION AND PREVENTION INITIATIVES – © OECD 2011
14 – EXECUTIVE SUMMARY
Moving forward
The report sets out a baseline against which international and national
development agencies can begin thinking through prospects for AVRP. For
many organisations the language of armed violence may seem new and
unfamiliar. For most practitioners, however, the importance of preventing and
reducing armed violence to allow for investments to proceed is beyond question.
The report sets out a number of practical suggestions to help the development
sector move forward on this critical agenda. To this end, development agencies
can:
Undertake AVRP "audits" or "inventory" initiatives: By taking stock
of their portfolios, multilateral, bilateral and non-governmental agencies
can determine where they have strengths and weaknesses. It would also
allow organisations to begin assessing their own profile and direction with
respect to AVRP more generally. OECD member countries could begin by
inventorying their own activities in this regard.
Identify and reinforce the goals, indicators and promising AVRP
practices: Rather than continuing to debate over definitions of armed violence,
a key priority should be to ensure that stakeholders analyse their common
problems and support comprehensive responses. To do this, development
agencies will need to establish clear and achievable goals, methodologies
for quantifying results and appropriate indicators, to design, implement and
monitor interventions and their outcomes.
Adopt integrated and evidence-based approaches to preventing and
reducing armed violence: The report demonstrates how the most effective
"direct" and "indirect" interventions are multi-sector, operate at multiple
levels, and rely on extensive partnerships among many actors. Such activities
should promote both security and wider development outcomes, with the two
being mutually reinforcing. For interventions to be sustainable and ultimately
scaled-up, these kinds of integrated initiatives are imperative.
Document good or promising practice with reliable evaluations:
Effective AVRP interventions are overwhelmingly based on high-quality
evidence and routine baseline assessments. It is critical that development
agencies document evidence of what works and what does not. While
circumstances shape the form and function of AVRP interventions, development
agencies need to assess the outputs and outcomes of such activities in both
lower- and middle-income settings.
Link the AVRP agenda to the promotion of peacebuilding and
statebuilding: Evidence has shown that promoting the capacity of public and
civil society to document, prevent and reduce armed violence, strengthens state
authority and legitimacy. Indeed, from a development practitioner perspective,
INVESTING IN SECURITY: A GLOBAL ASSESSMENT OF ARMED VIOLENCE REDUCTION AND PREVENTION INITIATIVES – © OECD 2011
EXECUTIVE SUMMARY – 15
a more explicit focus on AVRP in existing peacebuilding and statebuilding
strategies (OECD, 2010a) could produce significant benefits for local safety and
security. The combination of "direct" and "indirect" AVRP interventions at
various levels, focusing on the instruments and perpetrators of armed violence
and positively manipulating the broader enabling environment could also
generate important outcomes.
INVESTING IN SECURITY: A GLOBAL ASSESSMENT OF ARMED VIOLENCE REDUCTION AND PREVENTION INITIATIVES – © OECD 2011
INTRODUCTION – 17
Introduction
There are literally thousands of armed violence reduction and prevention
(AVRP) interventions underway around the world. Some regions – North
America, Western Europe, South-Eastern Europe, Latin America and the
Caribbean – seem to have more programming experience than others.
Indeed, North and sub-Saharan Africa, South, South-East, and Central Asia
and Central and Eastern Europe exhibit less activity (though not necessarily
indicating low levels of actual programming experience).
As a social phenomenon, armed violence is multi-faceted and defies
simple or ready-made solutions (Krause, Muggah and Wenmann, 2008).
Any effort to make a meaningful dent must be backed up with a robust
evidence base, strong inter-sector partnerships, and a comprehensive package
of activities. The evidence presented in this report shows that AVRP is
not only possible, it is already well underway (WHO, 2009). It singles out
the innovative strategies and approaches undertaken by numerous public
authorities, private sector entities and civil society organisations working
on the frontlines, to contribute to safety and security and enable meaningful
development opportunities to proceed.
The report underlines how targeted and appropriately-tailored interventions
– including those that combine both "direct" and "indirect" measures to prevent
and reduce armed violence – can lead to measurable improvements in security,
whether recorded as declining homicides, violent assault, rape or domestic
abuse or improved perceptions of security, mobility and wellbeing. However,
the evidence base identifying what is going on, where, and supported by whom,
remains comparatively thin, especially in low- and medium-income settings.
The development sector has an extremely important role to play in
supporting AVRP activities, particularly in countries affected by or emerging
from armed conflict or experiencing acute criminal violence. Most frontline
aid agencies have already assumed this responsibility. Indeed, the World
Bank, the Inter-American Development Bank (IADB) and the United Nations
Development Programme (UNDP) are heavily involved in cutting-edge
AVRP programming, and have been for almost two decades. As growing
numbers of other multilateral, bilateral and national entities become more
INVESTING IN SECURITY: A GLOBAL ASSESSMENT OF ARMED VIOLENCE REDUCTION AND PREVENTION INITIATIVES – © OECD 2011
18 – INTRODUCTION
involved in AVRP interventions, the underlying institutional architecture
shaping these activities will need to be clearly defined and understood.
To accelerate the process, the Organisation for Economic Coordination and
Development (OECD) and the UNDP – as part of the International Network on
Conflict and Fragility (INCAF) – initiated a scoping effort to map out AVRP
activities around the world (OECD, 2010a, 2010b). The intention was to develop
preliminary evidence of the diverse policy and programming experiences,
extract key trends and patterns, and ultimately identify promising AVRP
interventions for comprehensive evaluation. To make the process manageable,
the assessment focused on a selection of country case studies to highlight the
different contexts in which AVRP activities are underway.
This report is nested in a wider debate on the issues of armed violence
and development. It has been written in response to the language and
recommendations of the United Nations Secretary General's Report (UNSG,
Box 0.1. What is armed violence?
Armed violence is difficult to define but easy to recognise. Most attempts to
define violence tend to focus on settings, tools and outcomes. For example,
the World Health Organization (WHO) highlights the ways violence occurs in
multiple environments, and includes a range of vectors and the causes of physical
and psychological harm (WHO, 2002). Likewise, the OECD DAC (2009a)
sets out some general parameters: "… armed violence is the intentional use of
force (actual or threatened) with arms or explosives, against a person, group,
community or state, that undermines people-centred security and/or sustainable
development". This working definition covers armed violence perpetrated in
both armed conflict and non-conflict settings.*
Key risk factors associated with armed violence can be divided into at least
four categories. These include i) individual (e.g. youth, male, poor behaviour
control, history of aggressive behaviour, low education achievement, substance
abuse, exposure to violence); ii) relationship (e.g. poor family supervision,
exposure to punishment, low family attachment, low socio-economic status,
association with delinquents); iii) community (e.g. low social capital, high levels
of unemployment, gangs, guns and narcotics, access to alcohol); and iv) societal
(e.g. quality of governance, laws on social protection, income inequality, urban
growth and cultures sanctioning violence).
* The definition of armed violence that is used for data collection from the various
sources in this report does not distinguish between legitimate and illegitimate uses of
force. It also presumes that resorting to violence can be legitimate in some circumstances
in accordance with relevant international and national law.
INVESTING IN SECURITY: A GLOBAL ASSESSMENT OF ARMED VIOLENCE REDUCTION AND PREVENTION INITIATIVES – © OECD 2011
INTRODUCTION – 19
2009), UN General Assembly Resolutions (2008), the Geneva Declaration
on Armed Violence and Development (2006) and the Oslo Commitments on
Armed Violence (2010). The report has also utilised information from the
UN-led Armed Violence Prevention Programme (AVPP), and, in particular,
the extensive activities of the public health community on violence and injury
prevention. Programmatically, the report builds on the ongoing efforts of
the World Bank, IADB, UNDP Bureau for Crisis Prevention and Recovery
(BCPR), UN Office for Drugs and Crime (UNODC), WHO and others.
Report structure
The report charts out a basic roadmap to guide prospective efforts to
document and evaluate AVRP programmes worldwide. It targets development
practitioners and policy makers in multilateral and bilateral agencies,
international organisations and community-based associations. In featuring the
findings of a review of global experience and laying out preliminary findings
from mappings undertaken in six settings – Brazil, Burundi, Colombia, Liberia,
South Africa, and Timor-Leste1 – the report offers the first comparative
inventory of AVRP ever undertaken. The findings are not exhaustive: only those
programmes and projects that i) directly or indirectly targeted armed violence or
ii) applied a "diagnosis-treatment-results" framework were selected.
The survey of the six countries provides new and original insights
into the AVRP activities taking place in each of the countries. Following
consultations with numerous agencies and individuals, a shortlist of 570
AVRP initiatives was entered into a database for statistical analysis.2 Case
information was collected through a combination of desk and field research
(including key informant interviews, site visits and an on-line survey). While
focused predominantly on six lower- and middle-income contexts, the mapping
methodology illustrates the types of data that can be collected through a
systematic, yet decentralised research effort.
The report is divided into four main sections. The first chapter sets out a
conceptual framework and typology to assist development decision-makers
and practitioners to acquire a better understanding of the different categories of
AVRP programming. Chapter 2 offers a review of existing "global" inventories,
designed to collect experiences associated with violence prevention. Chapter 3
synthesises the findings from the six selected case studies. The final section
provides conclusions and recommendations and highlights some key trends.
INVESTING IN SECURITY: A GLOBAL ASSESSMENT OF ARMED VIOLENCE REDUCTION AND PREVENTION INITIATIVES – © OECD 2011
20 – INTRODUCTION
Notes
1. These cases have been selected because they feature sufficient data and evidence
of AVRP activities; offer promising future evaluations or political commitment
to AVRP; are geographically representative; and cover different contexts in
which AVR programming takes place.
2. These included 179 programmes in Brazil, 45 in Burundi, 219 in Colombia, 44 in
Liberia, 58 in South Africa, and 25 in Timor-Leste.
INVESTING IN SECURITY: A GLOBAL ASSESSMENT OF ARMED VIOLENCE REDUCTION AND PREVENTION INITIATIVES – © OECD 2011
1. CONCEPTUALISING ARMED VIOLENCE REDUCTION AND PREVENTION – 21
Chapter 1
Conceptualising armed violence reduction and prevention
This chapter sets out a basic typology of different Armed violence
reduction and prevention (AVRP) programmes and highlights
emerging promising practices. It "sets the scene" for the empirical
assessment featured in subsequent chapters. Armed violence reduction
and prevention interventions can be direct, indirect or components of
wider development schemes. Direct interventions aim to influence the
instruments, actors and institutional environments that enable armed
violence. Indirect interventions counter the proximate and structural
risk factors that shape armed violence onset and intensity. Broader
development schemes may not have armed violence reduction and
prevention as their primary aim but can nonetheless contribute to
reductions in insecurity over time.
INVESTING IN SECURITY: A GLOBAL ASSESSMENT OF ARMED VIOLENCE REDUCTION AND PREVENTION INITIATIVES – © OECD 2011
22 – 1. CONCEPTUALISING ARMED VIOLENCE REDUCTION AND PREVENTION
It is essential for development practitioners to acquire a common and shared
understanding of what is, and what is not, an armed violence reduction and prevention
(AVRP) intervention. To support this goal, the report presents a preliminary
AVRP typology – assembled on the basis of extensive surveys and expert interviews
– to allow interventions from around the world to be tracked and compared.
This chapter introduces a concise, conceptual framework for tracking AVRP programmes
and concludes with a reflection on "best practice" AVRP cases.
Conceptual framework
Any conceptual framework should be guided by an overall classification
scheme that allows for spatial, temporal, and programmatic comparison. This
report draws explicitly from the Organisation for Economic Co-ordination and
Development (OECD) armed violence "lens" that distinguishes interventions
according to whether they are:
Direct programmes that seek to address the instruments, actors and
institutional environments enabling armed violence,
Indirect programmes that address proximate and structural risk
factors giving rise to armed violence;1 and
Broader development programming that, while not having
prevention and reduction of armed violence as a key objective, can
nevertheless produce additional benefits
Drawing from existing typologies (WHO, 2004; Marc, 2009; IADB, 2003;
McLean and Blake Lobban, forthcoming) and new verifiable findings, the conceptual
framework highlights a wide spectrum of possible programming entry
points. It also considers a wide range of disciplinary perspectives, such as public
health and epidemiology, crime prevention and justice, conflict prevention and
peacebuilding. It also draws attention to distinct intervention types to assist development
practitioners in their design, implementation, and monitoring efforts. It
is important to stress that the focus of this report is not to evaluate good practice,
but rather to document the range and types of current experiences.
Figure 1.1 provides an illustration of the ways in which "direct", "indirect",
and broader development initiatives can be distinguished. These three
categories are not necessarily pursued in isolation. Indeed, many cutting-edge
AVRP programmes intentionally blur "direct" and "indirect" approaches – for
example focusing simultaneously on reducing firearms availability and working
with "at-risk" male youth, while seeking to mitigate the likelihood of misuse
through targeted employment schemes, after-school education programmes,
psychological support and even family planning activities. Large-scale
development programmes can also positively address relevant proximate and
structural risk factors associated with armed violence prevalence, enhancing the
value of such investments beyond their primary developmental aims.
INVESTING IN SECURITY: A GLOBAL ASSESSMENT OF ARMED VIOLENCE REDUCTION AND PREVENTION INITIATIVES – © OECD 2011
1. CONCEPTUALISING ARMED VIOLENCE REDUCTION AND PREVENTION – 23
As noted in the OECD report and repeatedly acknowledged by the
practitioners surveyed as part of this report, any AVRP intervention should be
premised on a solid evidence base and an understanding of the local and regional
context (OECD, 2009a). This requires carefully administered conflict and
political economy analysis, as well as survey and surveillance-based assessments
to ensure that activities build on local perceptions and actual experiences, as
well as relevant capacities and capabilities. In best-case scenarios, affected
communities may also participate in the elaboration of assessments, design and
implementation of interventions and monitoring and evaluation of activities.
There are many AVRP activities; however, no blueprint or simple template
of an AVRP programme exists. Indeed, AVRP programmes are often referred to
by practitioners as initiatives, schemes, or projects and may not easily conform
to conventional programming logic that sets out a "diagnostic-treatment-results"
model.2 In order to capture the full range of AVRP efforts underway, a more
flexible accounting approach should be developed.
Figure 1.1. Categorising AVRP activities
Direct AVRP
arms collection,
management and destruction,
gang mentorship activities and
legislative changes to
national/municipal
firearms regulation
Indirect AVRP
Proximate and structural risks
targeted employment and education
schemes for 'at-risk' youth, street lighting
and targeted development in
violence-affected areas, strengthening
access to justice
Programming on
broader development issues
large scale urban renewal schemes,
public transport systems, population
health monitoring, environmental
resource governance
INVESTING IN SECURITY: A GLOBAL ASSESSMENT OF ARMED VIOLENCE REDUCTION AND PREVENTION INITIATIVES – © OECD 2011
24 – 1. CONCEPTUALISING ARMED VIOLENCE REDUCTION AND PREVENTION
Introducing the typology
Past efforts to establish clear categories for violence prevention and
reduction have struggled to capture their heterogeneous and multi-dimensional
characteristics (Bellis et al., 2010). Any typology must therefore avoid being
overly deterministic or prescriptive, while simultaneously allowing for
sufficiently broad categories, so as to capture multi-phased interventions. In order
to map out programme experiences in the six country contexts, Table 1.1 applies
the conceptual framework and the OECD armed violence "lens", noted above,
together with distinct programmatic approaches.3 The typology then provides
examples from the six review countries.
"Direct" AVRP programmes include those focused predominantly on
the instruments, actors and institutions that enable armed violence. Activities
are wide-ranging and include efforts to seize, collect, buy back, promote
amnesties, and destroy small arms and light weapons, ammunition and bladed
and blunt instruments. Other efforts focus on "at-risk" children and youth,
male and female perpetrators, gangs and criminal groups and even non-state
armed groups and terrorists. Interventions focused on institutions range from
informal mediation and neighbourhood watch associations, to checkpoints
and search and seizure activities, to the reform of law enforcement agencies.
Table 1.1. AVRP programming typology
Programme priorities
Examples
from case
studies Programming approaches
Examples
from case
studies
Direct programming
Instruments Small arms and light weapons 72 Weapons collection and destruction 47
Ammunition 43 Weapons seizures 26
Conventional military equipment 25 Voluntary gun-free zones 15
Explosive remnants of war and
unexploded ordinance
31 Securing armouries and managing stocks 6
Perpetrators Age profile Informal mediation and local dispute resolution 56
Children 122 Checkpoints and stop/searches 14
Youth 100 Neighbourhood watch activities 12
Adults 57 Local militias and home guards 6
Gender profile Private security actors 7
Both male and female 128 Formal or track 1/1.5 negotiation 2
Male only 11
Female only 4
INVESTING IN SECURITY: A GLOBAL ASSESSMENT OF ARMED VIOLENCE REDUCTION AND PREVENTION INITIATIVES – © OECD 2011
1. CONCEPTUALISING ARMED VIOLENCE REDUCTION AND PREVENTION – 25
Programme priorities
Examples
from case
studies Programming approaches
Examples
from case
studies
Perpetrators Perpetrator profile
(cont.) Active armed groups 57
Gangs and youth groups 55
Organised crime groups 47
Community groups 33
Armed forces and police 30
Former combatant groups 12
Individual delinquents 11
Vigilante groups 3
Militia or paramilitary groups 5
Institutions Local and municipal authorities 85 Investments in local/urban/ national governance 77
Police and law enforcement 79 Training and monitoring of enforcement 67
Military and paramilitary 25 Promotion of justice and security system reform 54
Social welfare 24 Strategies to enhance community policing 46
Public health 21 Investment in local or traditional courts and
strategies to resolve disputes
33
Justice and transitional justice 6 Large-scale public administration reform 17
Indirect programming
Table 1.1. AVRP programming typology (continued)
Risk factors (selected)
Examples
from case
studies Programming approaches
Examples
from case
studies
Legacies of violence 295 Youth programming activities 233
Marginalised youth 245 Media and civil awareness campaigns 207
Gender-based discrimination 164 Skills development programmes 184
Rising inequality 142 Targeted education interventions 180
Presence of armed groups 130 Community empowerment 112
Availability of weapons 90 After-school activities 110
Psychological trauma 75 Home visits, care groups and social
service delivery
107
Economic deprivation 77 Targeted employment schemes 103
Family problems 32 Interventions to prevent income inequality
and social marginalisation
98
INVESTING IN SECURITY: A GLOBAL ASSESSMENT OF ARMED VIOLENCE REDUCTION AND PREVENTION INITIATIVES – © OECD 2011
26 – 1. CONCEPTUALISING ARMED VIOLENCE REDUCTION AND PREVENTION
"Indirect" AVRP programmes address a wide variety of risk factors.
The most frequently cited risk factors in the six case studies included the
presence of armed groups, legacies of violence, marginalised youth, genderbased
discrimination, and rising income inequality. Interventions range from
voluntary to enforcement-based activities. Concurrently, the most common
"indirect" programming approaches introduced to mitigate these risk factors
included youth programming schemes, media and civil awareness campaigns
(formal and informal), skills development programmes, targeted education
interventions, and urban renewal/environmental design activities. In order to
ensure a wider collection of "indirect" programmes, the report also included
the option "other" on the online survey.
Promising AVRP initiatives
A number of challenges arise when documenting and tracking AVRP
activities. First, it is difficult to distinguish between what can be classified
as "direct" or "indirect" AVRP programming, or a combination of the two.
Risk factors (selected)
Examples
from case
studies Programming approaches
Examples
from case
studies
Cross-border trafficking 31 Treatment and rehabilitation of individuals 95
Exposure to recent violent events 27 Job creation and employment 94
Exposure to violence
representations
19 Group therapy and treatment 82
Forced recruitment 15 Public or private health interventions 69
Demand or supply of drugs 8 Community and individually-targeted DDR 68
Environmental and urban design (including
lighting)
64
Urban/slum upgrading and renewal 41
Better security monitoring and surveillance,
including "hotspot mapping"
101
Justice and penal reform, including
increased penalties
42
Reductions in the availability and selling of
alcohol, particularly for minors
38
Community prohibitions and ordinances 14
Table 1.1. AVRP programming typology (continued)
INVESTING IN SECURITY: A GLOBAL ASSESSMENT OF ARMED VIOLENCE REDUCTION AND PREVENTION INITIATIVES – © OECD 2011
1. CONCEPTUALISING ARMED VIOLENCE REDUCTION AND PREVENTION – 27
Second, it is difficult to know whether specific interventions actually work
or can be considered effective in the absence of an evaluation.4 This report
does not determine the success of specific AVRP activities, however, several
AVRP "promising practice" examples can be found in the selected country
settings.5
The selected "promising practice" examples were drawn from the AVRP
programming database established by the case study mapping teams. It
should be noted that their selection was not made on the basis of a formal
programme evaluation. Instead, selection was determined on the basis of a
series of straightforward questions in the on-line survey undertaken by the
authors of the report.6 Future mapping and evaluation exercises undertaken
by the OECD and its partners can refine these best practices by determining
selection criteria from the outset.
Box 1.1. Promising practice in Burundi
Since 2007, the Mine Advisory Group (MAG) has been involved in managing
leftover mines and weapons stockpiles in Burundi. Specifically, it has supported
the weapons destruction workshop in Bujumbura where more than 8 000
weapons have been destroyed. MAG has partnered directly with the national
armed forces in order to destroy 312 man-portable air defence systems. MAG
also implements a comprehensive Physical Security and Stockpile Management
(PSSM) project with the national police, to destroy unsecured small arms and
light weapons stockpiles held at police stations following civilian disarmament
campaigns. The agency also seeks to improve the security of police armouries
and to provide armourers with training in safe storage and disposal.
Box 1.2. Promising practice in Brazil
Extensive efforts are underway in Brazil to reduce gang violence in urban areas. In
2003, the state government, state prosecutor's office and mayor's office formulated
a programme entitled Fica Vivo (Stay Alive) to reduce the homicide rate of young
people aged 15-19. The initiative aims to improve the quality of life in "at-risk"
communities, to minimise the likelihood of young men resorting to armed violence.
Alongside specific recreation and cultural activities, it features a systematic
monitoring system to ensure that youth do not turn to gangs. The programme
is administered by 27 community centres in metropolitan areas and, since its
inception, has resulted in a 50% reduction in homicide in the targeted areas.
INVESTING IN SECURITY: A GLOBAL ASSESSMENT OF ARMED VIOLENCE REDUCTION AND PREVENTION INITIATIVES – © OECD 2011
28 – 1. CONCEPTUALISING ARMED VIOLENCE REDUCTION AND PREVENTION
One way that development policy makers and practitioners distinguish
between good and bad practice is by determining if interventions have been
designed as programmes with clear results-based frameworks.7 Of course
many AVRP activities tend to be more project-oriented, and could benefit
from adopting a more coherent and strategic framework. Another way of
determining whether a given AVRP programme is a best practice case is
whether a "theory of change" has been incorporated.8 Theory of change
categories are now included in monitoring and evaluation of conflict prevention
and peacebuilding activities and are featured in OECD guidance documents
(OECD, 2009b).
Successful AVRP interventions are not only those that have a resultsbased
framework or clearly articulated theory of change. Relying exclusively
on proof of "good practice" may unintentionally result in selection bias
and the exclusion of a wide range of innovative, ongoing activities. It could
result in only counting those interventions already supported by donors,
who themselves structure assistance according to the presence of a theory of
change or results-based framework.
Box 1.3. Promising practice in South Africa
South Africa has one of the highest rates of sexual violence in the world. The
One Man Can Campaign (OMC) aims to transform the attitudes and behaviour
of men. Specifically, it encourages men and boys to advocate for gender equality,
to promote and sustain change in their personal lives, and to change the gender
norms driving the rapid spread of HIV-AIDS. The OMC campaign is conducted
in all of South Africa's provinces and in countries across Southern Africa and
each year reaches between approximately 3 000 and 5 000 men and boys from
all walks of life.
INVESTING IN SECURITY: A GLOBAL ASSESSMENT OF ARMED VIOLENCE REDUCTION AND PREVENTION INITIATIVES – © OECD 2011
1. CONCEPTUALISING ARMED VIOLENCE REDUCTION AND PREVENTION – 29
Notes
1. Proximate risk factors include inter alia the presence of alcohol, narcotics and
weapons or even gangs, while structural risk factors refer to economic crises,
income inequality, marginalised youth, gender-based discrimination, and
legacies of violence. There is no universally-agreed list of risk factors but the
factors mentioned are widely recognised to account for the core armed violence
risk factors (OECD, 2010a, 2010b).
2. Just focusing on a strictly defined AVRP programmes would significantly
under-report the breadth and depth of activities currently pursued by states, nongovernmental
agencies and private sector actors around the world.
3. Some categories may overlap and can be defined in subsequent, more detailed
evaluations of AVRP programmes.
4. Key questions include for example: how can outsiders know whether an
intervention managed to successfully prevent or reduce armed violence? What
are the benchmarks, methodologies for quantifying results and indicators of
successful reduction or – often more challenging to demonstrate – prevention?
5. With more focused evaluation in specific programmatic interventions, the
category of promising practice can be refined to identify true "best practice"
models of intervention.
6. These questions include: has the programme been underway for more than 2
years? Does the programme feature a monitoring and evaluation system? Is the
programme multi-sector and multidimensional in approach? Does the programme
include elements of "direct" AVRP programming? Does the programme have any
supportive information highlighting outcomes? Only respondents that were able
to respond affirmatively to these five questions were included.
7. The defining feature of programming approaches is that they are embedded in a
results-oriented process including four main components: i) a clearly articulated
problem statement; ii) a diagnostics-treatment-results framework, including
the definition of targets, success criteria, and measurement indicators; iii) the
implementation and monitoring of the treatment of the problem; and iv) a pre/
post-intervention analysis and impact review.
8. A "theory of change" defines the steps to be followed from an initial situation to
the achievement of a specific goal. It clearly articulates the underlying assumptions
shaping the current and future situations. It requires implementing partners to
clarify long-term goals, identify measurable indicators of success, and formulate
relevant actions to achieve these goals. It also forms the basis for strategic
planning, on-going decision-making and evaluation (Act Knowledge, 2009).
INVESTING IN SECURITY: A GLOBAL ASSESSMENT OF ARMED VIOLENCE REDUCTION AND PREVENTION INITIATIVES – © OECD 2011
2. MAPPING AVRP PROGRAMMING TRENDS – 31
Chapter 2
Mapping Armed violence reduction and prevention
programming trends
This chapter considers the general characteristics of armed violence
reduction and prevention (AVRP) activites in Brazil, Burundi,
Colombia, Liberia, South Africa and Timor-Leste. It detects a surge in
policies and programmes over the past five years and some innovative
shifts in programming theory and practice. It features a comparative
analysis of direct, indirect and broader AVRP activities in each setting,
the types of armed violence specific interventions aim to redress, their
gender dimensions, their timelines, approaches to monitoring and
evaluation, and budgets.
INVESTING IN SECURITY: A GLOBAL ASSESSMENT OF ARMED VIOLENCE REDUCTION AND PREVENTION INITIATIVES – © OECD 2011
32 – 2. MAPPING AVRP PROGRAMMING TRENDS
Between March and November 2010, Six case studies were generated
between March and November 2010 based on an intensive online survey
and key informant interviews. The surveys were adminsitered in multiple
languages (English, French, Portuguese and Spanish). A total of 570 armed
violence reduction and prevention (AVRP) interventions were short-listed,
including 179 initiatives in Brazil, 45 from Burundi, 219 in Colombia, 44 in
Liberia, 58 in South Africa, and 25 from Timor-Leste.
The report revealed that there are more programmes targeting risks
that give rise to armed violence ("indirect") than those tackling firearms,
armed perpetrators or enabling institutions ("direct"). Figure 2.1 presents
the relative distribution of programming types between 1990 and 2010. It
shows how many implementing agencies are adopting integrated approaches
– combining "direct" and "indirect" activities. Indeed, OECD members and
partners would do well to acknowledge (and further support) the prominence
of comprehensive approaches that sequence direct AVRP interventions with
medium- and longer-term indirect components.
AVRP initiatives have been ongoing in Brazil, Burundi, Colombia, and South
Africa since the early 1990s. Programmes were introduced to address the escalating
rates of violence and widespread insecurity, especially in the rapidly urbanising
cities of Bujumbura, Bogota, Cali, Medellin, Rio de Janeiro, São Paolo, Cape
Town and Johannesburg. Many of these interventions combined enforcement with
conflict prevention, peacebuilding, crime reduction, and citizen security priorities.
A significant increase in AVRP programming was detected in all six
case studies since the mid-2000s, with roughly two-thirds occurring between
2006 and 2010 (Figure 2.2). More than three quarters of all AVRP activities in
Brazil occurred during the past five years. Likewise, in Burundi, Colombia,
Liberia, and Timor-Leste, almost all registered programmes were initiated
after 2005. This recent surge in AVRP activity could have various explanations.
For example, in Brazil, the national public and citizen security initiative
(PRONASCI) and the upcoming World Cup (2014) and the Olympics (2016),
could have influenced the increase.
The scope and scale of AVRP programming appears to be changing over
time. In Brazil, for example, the first AVRP activities were initiated at the
time of a military dictatorship and focused on ensuring national security
through enforcement and repression (the national slogan since the 1960s has
been "order and progress"). However, over the past few decades the domestic
agenda has shifted from national to municipal public safety. By 2000, public
security had become a central component of presidential campaigns and
by the end of the decade, public safety policies emphasised participatory
approaches and citizen or civic safety.
INVESTING IN SECURITY: A GLOBAL ASSESSMENT OF ARMED VIOLENCE REDUCTION AND PREVENTION INITIATIVES – © OECD 2011
2. MAPPING AVRP PROGRAMMING TRENDS – 33
Figure 2.1. "Direct" and "indirect" programming in six case studies, 1990-2010
Colombia (n=219)
Programmes
on broader
issues
21%
16%
11%
42%
1%
3%
7%
Indirect AVRP
(risk factors)
Direct AVRP
Liberia (n=44)
Programmes
on broader
issues
11%
14%
11%
36%
25%
Indirect AVRP
(risk factors)
Direct AVRP
South Africa (n=58)
Programmes
on broader
issues
5%
2% 2%
83%
7%
Indirect AVRP
(risk factors)
Direct AVRP
Timor-Leste (n=25)
Programmes
on broader
issues
28% 32%
24%
16%
Indirect AVRP
(risk factors)
Direct AVRP
Programmes
on broader
issues
23%
39%
16%
10%
4%
2%
2%
Indirect AVRP
(risk factors)
Brazil (n=179) Burundi (n=45)
Programmes
on broader
issues
24%
31%
38%
4%
2%
Indirect AVRP
(risk factors)
Direct AVRP
INVESTING IN SECURITY: A GLOBAL ASSESSMENT OF ARMED VIOLENCE REDUCTION AND PREVENTION INITIATIVES – © OECD 2011
34 – 2. MAPPING AVRP PROGRAMMING TRENDS
In Colombia, pre-1990s programmes were heavily influenced by national
security considerations, particularly in light of the ongoing war against guerrillas
and paramilitaries. By the mid-1990s, interventions were influenced by the
government-led decentralisation process as well as increases in armed violence
across most of Colombia's major cities. Since 2003, however, activity has increased
tremendously, partly resulting from the disarmament and demobilisation of
paramilitaries and growing civil society engagement. This is mirrored somewhat
in Liberia and Timor-Leste where a post-conflict disarmament and demobilisation
focus has expanded to a wider consideration of security sector institutions.
The case studies revealed the relationship between donor investment
and geographic location. AVRP interventions in Latin America (Brazil and
Colombia) tend to feature more public sector involvement, particularly at the
Figure 2.2. Evolution of programming in selected cases, 1990-2012
Brazil (n=179)
1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012
Burundi (n=45)
1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012
Colombia (n=219)
1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012
Liberia (n=44)
1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012
South Africa (n=58)
1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012
Timor-Leste (n=25)
1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012
INVESTING IN SECURITY: A GLOBAL ASSESSMENT OF ARMED VIOLENCE REDUCTION AND PREVENTION INITIATIVES – © OECD 2011
2. MAPPING AVRP PROGRAMMING TRENDS – 35
municipal level with metropolitan authorities (including mayors and associated
state-level policing authorities). Thus, bilateral donors and international
development agencies appear to be comparatively smaller players in the
region.1 In Africa, however (Burundi, Liberia and South Africa), the key actors
remain international and local non-governmental agencies.
In sub-Saharan Africa, many AVRP programmes are typically financed
and administered by bilateral donors, international agencies, non-governmental
and community-based agencies, or private organisations, with less engagement
by national and municipal public sector counterparts. This partially reflects
the role of the state in promoting public security and the relative capacities of
governmental institutions.
The case studies noted several overlapping trends regarding the direction
and objectives of AVRP programming. For example, the most common
categories of armed violence addressed by all 570 interventions are youth,
domestic, interpersonal, urban and sexual violence (Table 2.1). Armed violence
occurring within or between communities or in the household is given high
priority, whereas violence generated by security forces, insurgent groups or
organised crime receives less attention.
Table 2.1. Most common types of armed violence addressed across all cases
Number of responses
Type of armed violence Brazil Burundi Colombia Liberia South Africa Timor-Leste
Total
responses
Youth violence 113 29 58 27 38 14 279
Domestic violence 92 28 46 29 43 5 243
Interpersonal violence 72 34 48 23 42 13 232
Urban violence 99 11 77 6 27 2 222
Sexual violence 81 30 13 32 45 5 206
Gang violence 51 12 37 10 34 11 155
School violence 76 2 18 11 34 2 143
Physical and sexual violence
against children and adolescents
77 8 8 1 43 4 141
Intra-state armed conflict 1 22 106 6 0 5 140
Violent organised crime 45 11 27 6 24 1 114
INVESTING IN SECURITY: A GLOBAL ASSESSMENT OF ARMED VIOLENCE REDUCTION AND PREVENTION INITIATIVES – © OECD 2011
36 – 2. MAPPING AVRP PROGRAMMING TRENDS
In all case studies, the AVRP programmes ran for three years, with a few
extending beyond that (Figure 2.3). Annual budgets for AVRP programmes
appear to fall into two general categories. In Brazil and Colombia, for example,
budgets are between USD 100 000-500 000 per year. However, Burundi, Liberia,
and South Africa have a significant number of large-scale AVRP programmes
(more than USD 1 million) but also smaller scale programmes (less than
USD 100 000) (Figure 2.4).
Figure 2.3. Time horizons of AVRP programming
Number of responses
76
26
73
27
12
14 16
3
31
3 5 3
0
10
20
30
40
50
60
70
80
Brazil Burundi Colombia Liberia South Africa Timor-Leste
0-3 years 3-5 years 5-7 years 7-9 years
Figure 2.4. Budget ranges of AVRP programming
Number of responses
0
5
10
15
20
25
30
35
Brazil Burundi Colombia Liberia South Africa Timor–Leste
< USD 10 000
USD 10 000–25 000
USD 25 000–50 000
USD 50 000–100 000
USD 100 000–250 000
USD 250 000–500 000
USD 500 000–1 000 000
USD 1 000 000–2 000 000
> USD 2 000 000
INVESTING IN SECURITY: A GLOBAL ASSESSMENT OF ARMED VIOLENCE REDUCTION AND PREVENTION INITIATIVES – © OECD 2011
2. MAPPING AVRP PROGRAMMING TRENDS – 37
The studies also revealed that direct AVRP interventions targeted a
combination of instruments, perpetrators or associated institutions, though
no clear trend emerged (Table 2.2).2 For example, the most common efforts
targeting instruments are firearms collection/destruction and weapons seizures,
whilst the most common activities targeting perpetrators are those involving
informal mediation and education, which focus on "at-risk" youth.3 Meanwhile,
Table 2.2. Direct AVRP interventions
Number of responses
Brazil Burundi Colombia Liberia
South
Africa
Timor-
Leste
Total
response
Interventions targeting instruments
Weapons collection and destruction 18 13 9 1 3 3 47
Weapons seizure 12 11 0 0 3 0 26
Weapons amnesties and buyback programmes 7 7 3 1 3 0 21
Voluntary gun-free zones 4 0 6 3 2 0 15
Securing armouries 0 6 0 0 0 0 6
Armourer training 0 3 0 0 0 0 3
Law enforcement 0 0 0 0 3 0 3
Interventions targeting perpetrators
Informal mediation 10 13 17 11 4 1 56
Education 0 13 0 6 2 0 21
Checkpoints 6 0 5 1 0 2 14
Neighbourhood watch 0 0 5 2 1 4 12
Private security actors 3 0 3 0 0 1 7
Local militias or home guards units 5 0 0 0 0 1 6
Peer pressure 0 3 0 0 0 0 3
Prosecution of perpetrators 0 2 0 0 0 0 2
Formal mediation 0 0 2 0 0 0 2
Interventions targeting institutions
Improved local/urban/national governance 24 16 24 7 1 5 77
Better law enforcement 27 12 10 8 5 5 67
Justice and security sector reform (JSSR) 31 10 0 5 5 3 54
Community policing 28 4 7 2 3 2 46
Local or traditional courts and dispute
resolution mechanisms 5 8 10 6 3 1 33
Public administration reform 10 1 1 0 2 3 17
Education interventions 0 3 0 1 1 0 5
More professional military and police 0 3 0 0 0 0 3
Community structures 0 0 0 1 0 0 1
Crime prevention information 0 0 0 0 1 0 1
INVESTING IN SECURITY: A GLOBAL ASSESSMENT OF ARMED VIOLENCE REDUCTION AND PREVENTION INITIATIVES – © OECD 2011
38 – 2. MAPPING AVRP PROGRAMMING TRENDS
strategies designed to promote the rule of law and access to justice were
commonly used with institutions.
"Direct" programming tended to be sensitive to gender-related issues
(Figure 2.5), with the bulk of interventions showing no discrimination between
the sexes. This is interesting in South Africa and Liberia, where one would
expect a series of initiatives focusing on male perpetrators and female victims,
Figure 2.5. Gender dimensions of all AVRP programmes
Percentage of responses
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Brazil Burundi Colombia Liberia South Africa Timor-Leste
Children Youth Adults No response
Table 2.3. Most frequently cited risk factors
Number of responses
Type of proximate risk factor Brazil Burundi Colombia Liberia South Africa Timor-Leste
Total
responses
Legacies of violence 101 40 96 18 27 13 295
Marginalised youth 108 19 47 25 35 11 245
Gender-based discrimination 74 22 0 28 35 5 164
Rising inequality 91 14 0 14 17 6 142
Presence of armed groups 45 5 62 5 11 2 130
Availability of weapons 40 12 17 6 14 1 90
Economic crises 36 6 0 17 10 8 77
Trauma 16 15 0 21 23 0 75
Family problems 0 0 26 0 6 0 32
Cross-border trafficking 19 0 0 11 1 0 31
Exposure to recent violent events 0 0 24 0 3 0 27
Forced recruitment 0 0 15 0 0 0 15
INVESTING IN SECURITY: A GLOBAL ASSESSMENT OF ARMED VIOLENCE REDUCTION AND PREVENTION INITIATIVES – © OECD 2011
2. MAPPING AVRP PROGRAMMING TRENDS – 39
given the rates of sexual violence.4 Meanwhile, in Brazil, many "direct" AVRP
interventions focusing on males also register substantial female participation,
suggesting a broadening of selection criteria.
Table 2.4. "Indirect" AVRP programmes
Number of responses
Brazil Burundi Colombia Liberia
South
Africa
Timor-
Leste
Total
response
Informal/formal voluntary strategies
Youth programming activities 108 13 64 16 26 6 233
Media and civil awareness campaigns 90 31 36 28 20 2 207
Skills development programmes 85 13 35 14 34 3 184
Education interventions 7 12 101 19 35 6 180
Community empowerment interventions 13 13 46 22 17 1 112
After-school activities 74 0 9 4 19 4 110
Home visits, care groups and social service
delivery 78 2 13 9 5 0 107
Targeted employment schemes 65 7 14 7 7 3 103
Interventions designed to address income
inequality and social marginalisation 61 12 11 7 6 1 98
Treatment and rehabilitation of individuals 19 11 35 10 19 1 95
Job creation and employment programmes 43 5 20 10 13 3 94
Group therapy and treatment 28 4 24 6 20 0 82
Public or private health interventions 39 4 7 7 11 1 69
Incentive-based DDR 31 9 23 3 0 2 68
Environmental or urban design 36 1 0 3 6 1 47
Urban/slum upgrading and renewal 30 1 4 1 5 0 41
Research 0 2 21 0 1 0 24
Informal/formal enforced interventions
Better security monitoring 59 11 15 7 5 4 101
Justice and penal reform 5 11 3 7 10 6 42
Reducing the availability and consumption of
alcohol 15 0 7 4 12 0 38
Community prohibitions and ordinances 3 2 2 4 2 1 14
Mine action 2 2 7 1 0 0 12
Strengthening formal institutions 0 8 0 0 2 0 10
INVESTING IN SECURITY: A GLOBAL ASSESSMENT OF ARMED VIOLENCE REDUCTION AND PREVENTION INITIATIVES – © OECD 2011
40 – 2. MAPPING AVRP PROGRAMMING TRENDS
In comparison, "indirect" AVRP programmes tended to be extremely
diverse (Table 2.3). While all six countries focus on a wide range of proximate
and structural risk factors, they additionally concentrate on legacies of
armed violence, marginalised youth, gender-based discrimination, and rising
inequality. Several important risk factors, such as family challenges, exposure
to recent violent events, unemployment, and lack of education, are less well
represented, however, this may be at least partly attributable to the design and
translation of the survey.5 Future surveys could also focus on risk factors at the
individual, relational, communal, and societal levels.
The vast majority of "indirect" AVRP programmes under review also
showed a tendency towards voluntary (rather than enforcement-based)
approaches.6 Table 2.4 illustrates a series of different types of interventions
led by "at-risk" youth programming, media and civil awareness campaigns,
skills and livelihood development programmes, and educational interventions.
Strategies that drew on enforcement tactics emphasised enhanced
crime and "hot spot" monitoring and reforms to the justice and penal sectors,
including increasing penalties and incarceration periods.
The review also established that AVRP monitoring and evaluation is not
consistent (Figure 2.6). This may be because many interventions are short-term
while outcomes and impacts are long-term, making a systematic assessment
within existing project-cycles difficult. Additionally, to determine programme
effectiveness, routine monitoring requires good surveillance, analysis capacities
and evidence, which may not be available in every intervention. Even in countries
with robust public-surveillance capacities, such as Brazil and Colombia, 47% and
70% of respective responses indicated that AVRP activities claimed not to have
had monitoring and evaluation capacities.
Figure 2.6. AVRP monitoring and evaluation
Percentage of responses
47%
20%
70%
23%
3%
60%
53%
80%
30%
77%
97%
40%
0%
20%
40%
60%
80%
100%
120%
Brazil Burundi Colombia Liberia South Africa Timor-Leste
Does your programme include monitoring activities?
No Yes
INVESTING IN SECURITY: A GLOBAL ASSESSMENT OF ARMED VIOLENCE REDUCTION AND PREVENTION INITIATIVES – © OECD 2011
2. MAPPING AVRP PROGRAMMING TRENDS – 41
Notes
1. This stands in contrast to the "global review" earlier in the paper that highlighted
the relatively significant role of the IADB and World Bank in financing AVRP
activities across Latin America. These findings may reveal an underlying bias
in reporting – most international financing is nevertheless channelled to public
institutions (and not NGOs) suggesting that there may in fact have been a lagged
effect of their investment.
2. This is in part due to a high non-response rate for this question.
3. A review of all programming contexts suggests that "direct" AVRP programmes
principally target children and youth. It was not possible to identify clear
trends in relation to how "direct" programmes addressed specific categories of
"armed groups" because the perceptions of armed violence "types" and related
"perpetrators" varied widely between the six cases.
4. It is possible that different trends may emerge if the caseload of respondents is
expanded in future rounds of this mapping.
5. The options on the questionnaire were a choice of known risk factors, as well as
an "other" option for additional categories. The risk factors included departed
from the notion that, from the perspective of a state, there are external and internal
risk factors. External risk factors include, inter alia, economic and environmental
crises, cross-border trafficking, external interference; internal risk factors include
rising economic inequality, marginalised youth, gender based discrimination,
legacies of violence, presence of armed groups, availability of weapons, and
trauma. (OECD DAC/INCAF, 2010b).
6. As with the review of direct AVRP programming approaches, responses to
indirect AVRP efforts also yielded a high no-response rate.
INVESTING IN SECURITY: A GLOBAL ASSESSMENT OF ARMED VIOLENCE REDUCTION AND PREVENTION INITIATIVES – © OECD 2011
3. CASE STUDY SUMMARIES – 43
Chapter 3
Case study summaries
This chapter considers the wide range of armed violence reduction
and prevention (AVRP) activities in Brazil, Burundi, Colombia,
Liberia, South Africa and Liberia. It reviews the historical and
social factors giving rise to specific forms of AVRP, but also profiles
the policies and activities in each context. In states affected by and
emerging from armed conflict, approaches may be more direct
and include controlling the tools of violence or demobilising and
reintegrating combatants. In states experiencing acute rates of violent
crime, interventions may be more indirect and emphasise recurring
risks such as chronic youth unemployment and extreme inequality.
INVESTING IN SECURITY: A GLOBAL ASSESSMENT OF ARMED VIOLENCE REDUCTION AND PREVENTION INITIATIVES – © OECD 2011
44 – 3. CASE STUDY SUMMARIES
This chapter highlights the main findings from the six country mappings.
Each of the cases reported below i) summarises the wider political, economic
and historical dynamics of armed violence, ii) considers the basic characteristics
of armed violence reduction and prevention (AVRP) interventions, and iii) highlights
a number of relevant qualitative findings. To allow for comparison, each
focuses on the key programming characteristics, including the relationships
between "direct" and "indirect" programming, the structure of funding and
donor support, intervention targets, programming types, key risk factors and
monitoring and evaluation capabilities.
Brazil
Brazil has one of the highest homicide rates in the world and, although this
has decreased in recent years, the national rate is still 25 per 100 000 (Waiselfisz,
M. and J. Jacobo, 2010). Violence is concentrated among young people,
especially young black males. Indeed, the juvenile homicide rate jumped from 30
per 100 000 in 1980 to 50.1 per 100 000 in 2007, while for black youths it reached
66 per 100 000 (Waiselfisz, M. and J. Jacobo, 2010). Homicide rates for other
population groups declined from 21.2 to 19.6 per 100 000 over the same period.
At the same time, there are numerous public, private and non-governmental
led efforts to prevent and reduce armed violence, commonly referred to as
"public safety" or "public security" initiatives. Many of these developed in
the wake of drug-related violence during the 1990s and the opening up of
democratic space. Civil society supported emerging campaign agendas, which
linked violence to social justice, police aggression, impunity, and even small
arms availability and misuse.
Before the 1990s, unrest and delinquency was met almost exclusively with
a heavy fist. Likewise, domestic civil society and faith-based groups tended to
focus more on poverty alleviation and welfare promotion – an ethos that persists
today. Over the past decade and a half, however, violence – including armed
violence – became categorised as a social problem of the country. The public
security agenda was also used as a means to justify all manner of investment
across disparate sectors. Indeed, many public and private institutions, civic action
groups, faith-based associations and community-based groups began capitalising
on violence in order to raise funds, pass white papers and drum-up votes.
Brazilian political and public authorities have also initiated numerous
activities to prevent and reduce armed violence in all its major cities and
amongst specific "at-risk" groups. Launched in 2007, PRONASCI appears to
have contributed to an increase in both stability and social welfare. Creative
interventions focused on encouraging the public to report on crime are
proving successful (Box 3.1). Meanwhile, the deployment of pacification
police units (UPP) to the "hot" zones of selected cities, beginning with Rio de
INVESTING IN SECURITY: A GLOBAL ASSESSMENT OF ARMED VIOLENCE REDUCTION AND PREVENTION INITIATIVES – © OECD 2011
3. CASE STUDY SUMMARIES – 45
Janeiro, is benefiting the security situation, though there are early indications
that violence may be spreading to adjoining municipalities (Box 3.2).
To better understand the characteristics and dynamics of "direct" and
"indirect" AVRP interventions, a Brazil-based mapping team interviewed
more than 400 specialists in the police, justice, penal, crime prevention,
social welfare and development and public health sectors. In the process,
analysis was conducted for 179 programmes and projects by telephone, online
and face-to-face interviews.
Overall, the study found that the expression "armed violence" is not
widely applied in Brazil, even though many people are engaged in violence
prevention and reduction. Preferred concepts include "public safety" and
"citizen security", and to a lesser extent "public order" and "pacification".
Nevertheless, historically, there has been considerable focus on combining
security and development activities. Today, many public entities and nongovernmental
organisations opportunistically use concepts as a means to
advance a wide range of projects.
Box 3.1. Denouncing crime in Brazil
In order to expand the surveillance and response to criminal violence in Brazil,
the government launched Dial Denounce. Dial Denounce aims to increase
reporting on crime and through the active involvement of community members
as "advocates" and "denouncers". The project operates through a 24-hour
call-centre that forwards denunciations to the police branch responsible for the
investigation. As part of the country's National Programme to Combat Sexual
Violence Against Children, the government also set up "Dial 100" to promote
the denunciation of actual and would-be perpetrators. It has registered and
responded to more than 130 000 separate claims.
Box 3.2. Pacification police in Rio de Janeiro
The Pacification Police Unit (UPP) intervention was launched in 2008 to transform
the police model in Brazil. Beginning in the state of Rio de Janeiro, the UPP
first reclaims territory, by force if necessary, focusing on shanty-town neighbourhoods
formerly controlled by narco-traffickers and private militia. The UPP then
deploys male and female community police to improve the services provided by
the police and equally the public perception of the police. To date, some 18 favelas
including more than 44 communities (240 000 people) have been "pacified", violent
crime has dropped dramatically and property values have increased.
INVESTING IN SECURITY: A GLOBAL ASSESSMENT OF ARMED VIOLENCE REDUCTION AND PREVENTION INITIATIVES – © OECD 2011
46 – 3. CASE STUDY SUMMARIES
AVRP programming trends in Brazil
Most of Brazil's 179 AVRP activities are "indirect" (60%) as compared to
"direct" (39%). The vast majority of support for these activities comes from
national public authorities (25%) and local government/mayoral representatives
(24%). The private sector also plays an important role (15%) followed by
national NGOs and international NGOs. Bilateral donors provided support in
just 3.3% of cases, multilateral donors in 7.3% and international NGOs in 8.5%
(Figure 3.1).
Forty-eight percent of "direct" AVRP interventions targeted domestic
violence, with 31% targeting youth, gang and school violence. 16% of
interventions focused on interpersonal violence, while 9% addressed sexual
violence and 11% focused on "other" categories. Overall, more than one-third
of all "direct" programmes focused on children and youth; almost one-fifth
focused on youth and adults, while just over one-tenth focused exclusively on
adults (Table 3.1).
Brazil has a wide range of "direct" and "indirect" intervention types
(Table 3.2), though most are centred on voluntary "indirect" interventions,
which promote youth programming, media and civil awareness campaigns
and skills development. The majority of "direct" interventions relate to
justice and security system reform, community policing and improved law
enforcement.
Figure 3.1. Types of implementing agencies
Aggregate responses
85
25
25
17
13
8
8
5
3
0 10 20 30 40 50 60 70 80 90
Local government/town hall
National company
National government
International NGO
National NGO
Bilateral donor agency
Local community organisation
International company
Regional organisation
INVESTING IN SECURITY: A GLOBAL ASSESSMENT OF ARMED VIOLENCE REDUCTION AND PREVENTION INITIATIVES – © OECD 2011
3. CASE STUDY SUMMARIES – 47
Table 3.1. What types of armed violence do your programmes address?
Frequency Percent Valid percent Cumulative percent
Domestic violence 88 48.1 48.1 48.1
Interpersonal violence 16 8.7 8.7 56.8
Gang violence 1 0.5 0.5 57.4
School violence 7 3.8 3.8 61.2
Sexual violence 9 4.9 4.9 66.1
Urban violence 11 6.0 6.0 72.1
Youth violence 23 12.6 12.6 84.7
Other 20 10.9 10.9 95.6
No response 8 4.4 4.4 100.0
Total 183 100.0 100.0
Table 3.2. Most common "direct" and "indirect" AVRP interventions in Brazil
Direct interventions Responses Indirect interventions Responses
Instruments Informal/formal voluntary
Weapons collection and destruction 18 Youth programming activities 108
Weapons seizures 12 Media and civil awareness campaigns 90
Weapons amnesties and buyback campaigns 7 Skills development programmes 85
Voluntary gun-free zones 4 Home visits, care groups, and social services delivery 78
Perpetrators After school activities 74
Informal mediation 10 Targeted employment schemes 65
Checkpoints 6 Informal/formal enforced
Local militias and home guard units 5 Better security monitoring 59
Private security actors 3 Reducing availability and consumption of alcohol 15
Institutions Justice and penal reform 5
JSSR 31 Community prohibition and ordinances 3
Community policing 28 Mine action 3
Enhanced law enforcement 27
Improved local/urban/national governance 24
INVESTING IN SECURITY: A GLOBAL ASSESSMENT OF ARMED VIOLENCE REDUCTION AND PREVENTION INITIATIVES – © OECD 2011
48 – 3. CASE STUDY SUMMARIES
There is generally widespread agreement that "unattached" youth in
Brazil – young males and females who are disconnected from stable familial,
societal, educational and welfare systems – are most susceptible to engaging
in armed violence. Thus, the principal risk factor addressed through "indirect"
programming was marginalised youth (20%). Other risk factors included
legacies of violence (19%), inequality (17%) and gender-based discrimination
(14%). Only 7% of respondents identified the availability of weapons as an
important risk factor (Figure 3.2).
As Figure 3.2 illustrates, the majority of formal "indirect" AVRP interventions
focused on youth programming (25%), with activities also focusing on afterschool
activities (17%), redressing income inequality (14%) and job creation
schemes for "at-risk" youth (10%). More informal "indirect" programming
ranged from civic awareness campaigns to promote violence awareness (21%)
and targeted skills development (20%) to specialised care groups, enhanced
social services delivery (19%), direct home visits (14%) and wide-ranging
employment schemes (15%).
A major challenge in Brazil, as elsewhere, is determining what kinds
of interventions work and which do not. There is a growing emphasis in
the security and development sectors on the importance of monitoring and
evaluating interventions.1 Just over half (55%) of the programmes mapped
had adopted some form of monitoring and evaluation capacity, which is
surprising given the growing emphasis in Brazil on results- and evidencebased
approaches.
Figure 3.2. Most prominent risk factors addressed through "indirect" programming
Percentage of responses
20%
19%
17%
14%
8%
7%
7%
4%
3%
1%
0% 5% 10% 15% 20% 25% 30% 35%
Marginalised youth
Legacies of violence
Rising inequality
Gender-based discrimination
Presence of armed groups
Availability of weapons
Economic crises
Cross-border trafficking
Trauma
No response
INVESTING IN SECURITY: A GLOBAL ASSESSMENT OF ARMED VIOLENCE REDUCTION AND PREVENTION INITIATIVES – © OECD 2011
3. CASE STUDY SUMMARIES – 49
A large number of qualitative insights emerged from the Brazil case
that might help shape emerging best practices on AVRP programming. For
example, there is a general sense among programme implementers that
prevention – focusing on early interventions to address key risk factors – is
highly effective in mitigating armed violence. Virtually all respondents
noted that activities that mobilise education and vocational alternatives (for
both "at-risk" adolescents and youth), recreation and sporting activities, and
cultural investments, play a key role in deterring youth from violent behaviour.
Most respondents also highlighted the fundamental importance of
adopting comprehensive and integrated interventions. Virtually all respondents
emphasised the need for wide-ranging and full-spectrum approaches – i.e. early
prevention together with enforcement. Most activities made reference to the
importance of not only engaging "at-risk" youth, but also promoting civilian
protection, community policing and human rights advocacy, together with
wider social programming.
Though only half the respondents claimed to apply strict monitoring and
evaluation practices, most highlighted the need for evidence when amending
or restructuring priorities and activities. Many stressed the importance of
documenting key opportunities and constraints, as well as publicising successes.
More practically, justice- and police-led activities appeared to privilege the
critical role of data and evidence in shaping interventions, including mapping
out trends in order to target and respond to crime "hot spots".
Box 3.3. Youth AVRP in Brazil
Young males are the most common perpetrators and victims of armed
violence in Brazil. Many interventions are designed to directly and indirectly
promote armed violence prevention and reduction to minimise associated
risks for youths. For example, Programme H aims to engage young men and
their communities in discussions on gender relations and male-on-female
violence. It supports educational activities, community campaigns, and
an innovative evaluation module for assessing the programme's impact on
attitudes. Meanwhile, Peace Squares SulAmerica focuses on preventing
violence among adolescents and youth. As of 2010, five neighbourhood centres/
plazas had been renovated and given to the community. The aim is to alter the
risk factors shaping armed violence by changing the urban environment through
the provision of sports, leisure and cultural alternatives.
INVESTING IN SECURITY: A GLOBAL ASSESSMENT OF ARMED VIOLENCE REDUCTION AND PREVENTION INITIATIVES – © OECD 2011
50 – 3. CASE STUDY SUMMARIES
Burundi
The international aid sector – and to some extent Burundians themselves
– have typically adopted peacebuilding and conflict-management strategies
to reduce and prevent armed violence. Since the end of armed conflict in
2000, various direct AVRP activities have been implemented including
disarmament, demobilisation and reintegration (DDR), civilian disarmament,
security system reform, transitional justice for former perpetrators of armed
violence, and the promotion of non-violent elections.
While the incidence and intensity of armed violence has tapered off in
the past decade, armed violence still claims thousands of lives every year.
As recently as 2008, Burundi reportedly suffered 1 049 violent deaths, 1 262
injuries, and a firearm homicide rate of 12.3 per 100 000 people – above the
global average of 7.6 per 100 000.2
Opportunities for candid dialogue on difficult topics now exist, despite the
post-conflict legacies of revenge and impunity prevailing since independence.
The successful integration of Hutu and Tutsi former combatants into both
the national police force and military, as well as the emergence of multiple,
independent media outlets, were crucial to the success of the initial programmes
addressing armed violence issues and laid the foundations for the 2005 elections.
A series of internationally-sanctioned and sponsored AVRP interventions
has been credited with promoting Burundi's post-conflict security. For
example, the Demobilisation, Reinsertion and Reintegration Project (PNDRR),
was approved and funded in 2004 via the Multi-Donor Demobilisation and
Reintegration Programme (USD 41.8 million) and a grant (USD 36 million),
with the World Bank and the Burundian Government as partners in the
projects' implementation. By June 2006, 29 000 ex-Gardiens de la Paix
and combattants militants were reinserted into civil society and received
reintegration support. More than 23 000 adult ex-combatants and over 3 261
former child soldiers had been demobilised and had received (re)insertion
support by April 2008.
However, reintegration fell behind as the overall development programming
to support the demobilised failed to emerge.3 In the meantime, the early stages
of disarmament, including that launched by the Commission technique de
désarmement des civils et de lutte contre la prolifération des armes légères et
de petit calibre (CTDT) had disappointing results. Indeed, from 2006 until the
end of 2009, only 40 000 of the estimated 100 000 – 300 000 small arms in the
country were handed over (Pezard, S. and S. de Tessières, 2009).
Notwithstanding the DDR and civilian disarmament efforts, armed
violence remains a significant problem. The key insurgent, Palepehutu-FNL,
continued fighting despite signing cease-fire agreements in 2006 and 2008.
INVESTING IN SECURITY: A GLOBAL ASSESSMENT OF ARMED VIOLENCE REDUCTION AND PREVENTION INITIATIVES – © OECD 2011
3. CASE STUDY SUMMARIES – 51
In 2009 the group agreed to formally disarm and transition to a political
party. However, armed violence remained a challenge as the population
was still not secure. Targeted assassinations and the use of grenades and
armed youth gangs to advance party interests became increasingly prevalent.
Banditry and arms proliferation is also considered to be a serious concern
(Human Rights Watch, 2009; Lemarchand, R., 2004; Uvin, P., 2009).
The current security situation is marked by revenge and a high level
of impunity. Banditry is the largest source of armed violence around the
country, followed by conflicts over land and access to property, domestic
violence, sexual violence and political violence. There are also numerous
violent deaths every month, with neither the cause nor the perpetrators being
identified by police. There has therefore been considerable focus in Burundi
on promoting peacebuilding through awareness building, dispute resolution,
and enhanced policing.
AVRP programming trends in Burundi
The Burundi survey analysed 45 programmes across five provinces,
including Ngozi, Kirundo, Ruyigi, Makamba, and Mwaro. Overall, the mapping
confirms the general finding that "armed violence" is not a category or label
widely used by national or local practitioners. Indeed, despite acknowledgement
among respondents that the reduction and prevention of armed violence are clear
aims, AVRP programming itself does not exist. Instead, the primary focus is on
peacebuilding, conflict management and security system reform.
AVRP interventions are grouped into two financial categories. The
largest category includes more than two-thirds of all cases, with budgets of
more than USD 500 000. The second group receives funding of less then
USD 25 000 – with local-level organisations being the primary recipients. In
contrast to Brazil and Colombia, more than two-thirds of all reported funding
comes from international donor organisations or NGOs.
Burundi has a comparatively high proportion of respondents reporting
"indirect" AVRP activities – 64% said their activities were "direct" programming
while 97% said their activities were also "indirect" programming. Table 3.3 details
the most common programmatic "direct" and "indirect" AVRP interventions in
Burundi. Responses are comparatively balanced between "direct" and "indirect"
interventions, with most "direct" interventions focused on the instruments of
violence – notably the control, collection and destruction of small arms.
These interventions also focus on the perpetrators – through informal
mediation and education activities – and on key institutions, especially the
military and policing sectors. Meanwhile "indirect" interventions focus
primarily on media and civil awareness campaigns, though youth programming
and skills development are common, especially among "at-risk" groups.
INVESTING IN SECURITY: A GLOBAL ASSESSMENT OF ARMED VIOLENCE REDUCTION AND PREVENTION INITIATIVES – © OECD 2011
52 – 3. CASE STUDY SUMMARIES
Table 3.3. Most common "direct" and "indirect" AVRP interventions in Burundi
Direct interventions Responses Indirect interventions Responses
Instruments Informal/formal voluntary
Weapons collection and destruction 13 Media and civil awareness campaigns 31
Weapons seizure 11 Youth programming activities 13
Weapons amnesties and buyback 7 Skills development programmes 13
Securing armouries 6 Community empowerment interventions 13
Perpetrators Education interventions 12
Informal mediation 13 Interventions against income inequality and social
marginalisation
12
Education 13 Informal/formal enforced
Peer pressure 3 Better security monitoring 11
Prosecution 2 Justice and penal reform 11
Institutions Strengthening formal institutions 8
Improved local/urban/national governance 16 Mine action 2
Enhanced law enforcement 12 Community prohibition and ordinances 2
Justice and security system reform 10
Local or traditional dispute resolution/courts 8
Figure 3.3. Types of armed actors targeted by "direct" programming
Percentage of responses
21%
21%
18%
17%
14%
5%
2%
2%
2%
0% 5% 10% 15% 20% 25%
State armies
Non-state armed groups
Ex-combatants
Individuals
Community group
Criminal groups
Vigilante groups
State militias
Gangs
INVESTING IN SECURITY: A GLOBAL ASSESSMENT OF ARMED VIOLENCE REDUCTION AND PREVENTION INITIATIVES – © OECD 2011
3. CASE STUDY SUMMARIES – 53
It is worth noting that what might be categorised as "direct" AVRP programmes
target the individual rather than the community.4 "Direct" programming
overwhelmingly targets adult and youth perpetrators, and many of the 29
interventions directly addressing armed violence focus on the military/police and
former combatants (51% and 41% respectively). Moreover, 74.9% of all "direct"
interventions also address the reform of the military and/or the police, including
laws, directives and policies. Finally, the majority of respondents felt that DDR
and SSR programming constituted "direct" armed violence reduction (Figure 3.3).
Many "indirect" AVRP programmes encourage sensitisation or awarenessbuilding.
The critical risk factors for armed violence in Burundi are legacies
of violence (22%), gender-based discrimination (12%), marginalised youth
(10%) and trauma (9%) (Figure 3.4). 45% of all respondents highlighted the
importance of independent radio as an effective tool to prevent armed violence.
This is no doubt a reflection of the role played by the media in encouraging
ethnic violence in the 1980s and 1990s. Given low literacy rates in the country,
radio broadcasts also seem more effective for sensitisation than print media.
Measuring outcomes is difficult in Burundi, given the poor standard of
surveillance and information collection found in both the security sector and the
conventional development community. Despite the requirements of multilateral
and bilateral donor partners, few interventions report monitoring and evaluation
capacities. Indeed, 77% of respondents had no monitoring and evaluation
Figure 3.4. Main risk factors addressed by "indirect" programming
Percentage of responses
22%
12%
10%
9%
8%
7%
3%
3%
3%
3%
3%
3%
0% 5% 10% 15% 20% 25%
Legacies of violence
Gender-based discrimination
Marginalised youth
Trauma
Rising inequality
Availability of weapons
Large numbers of ex-combatants
and returnees
Economic crises
Land disputes
Elections
Human rights abuse
Presence of armed groups
INVESTING IN SECURITY: A GLOBAL ASSESSMENT OF ARMED VIOLENCE REDUCTION AND PREVENTION INITIATIVES – © OECD 2011
54 – 3. CASE STUDY SUMMARIES
mechanisms whatsoever. Where monitoring was reported, the outputs were
quarterly or annual reports, documenting outputs and financial information.
From a more qualitative perspective, a large number of Burundian
respondents emphasised the challenge of international, regional and domestic
arms flows. Indeed, many actors felt powerless to engage with regional efforts
to control arms trafficking. They highlighted the challenges of policing
the country's porous borders and regular flow of handguns and automatic
rifles from one Great Lakes country to another. Local actors are aware that
investments have been made to enhance border controls and forensics, but
argue that concrete steps beyond simply marking weapons must be adopted
Other respondents emphasised the importance of AVRP. Some suggested
that the recent shift in donor perspective from peacebuilding to poverty
reduction may be premature. Indeed, armed banditry is on the rise and, by
all accounts, includes local police and administrators. The potential threat of
routine criminal violence to wider national security is considered to be very
real. However, most felt that if the formal economy improved, this would be
beneficial for armed violence prevention and reduction.
Colombia
Colombia's ongoing, armed conflict is now accompanied with staggering
levels of organised and petty crime organised crime. Thus the level of armed
violence and insecurity remains well above the international average. Though
the intensity and diversity of armed violence in Colombia5 is difficult to
explain, it obviously affects the development of the country.
Despite the reported declines during 2002-05, armed violence in
Colombia is still very high and has been increasing since 2005. Non-lethal,
inter-personal armed violence is also on the rise, although this is concentrated
in cities. The political and instrumental use of violence, including targeted
assassinations and intimidation of witnesses to ongong judicial cases, also
appears to be increasing in certain areas.
There are complex relationships between armed groups and armed
violence in Colombia. For example, homicidal violence directly attributable to
the armed conflict is now relatively minor and is confined to rural areas.
Relationships also exist between conflict and non-conflict violence. For
example, growing criminal violence is increasingly being linked to former
combatants and the availability of surplus military weaponry, despite the
demobilisation of more than 30 000 paramilitaries since 2003.
Colombia has been relatively slow to undertake systematic strategies
to reduce and prevent armed violence. "Direct" and "indirect" AVRP
programmes have only formally appeared within the last three decades, with
INVESTING IN SECURITY: A GLOBAL ASSESSMENT OF ARMED VIOLENCE REDUCTION AND PREVENTION INITIATIVES – © OECD 2011
3. CASE STUDY SUMMARIES – 55
interventions emerging between 1990 and 2005, and a consolidation and
increase in AVRP initiatives occurring since 2006.6
AVRP programming trends in Colombia
Sixty-two organisations and institutions were involved in interventions to
prevent and reduce armed violence among local and national governments,
organisations of the civil society, international organisations, foundations and
others. Overall, these organisations and institutions administered some 219
AVRP programme initiatives.
"Indirect" AVRP programmes are, as in other cases, more common than
"direct" AVRP programmes in Colombia. More than half (54%) the interventions
focused on mitigating the proximate risk factors of armed violence while 20%
addressed armed violence directly by tackling the instruments, perpetrators and/
or institutions. Some 16% of the interventions are fully integrated, addressing
armed violence both "directly" and "indirectly". One category of programming
where "direct" and "indirect" interventions are often combined is in relation to
youth (Box 3.4)
Public authorities and civil society are the primary actors administering
AVRP interventions across Colombia. These range from periodic neighbourhood
safety promotion initiatives, to structured, far-reaching, multi-sector programmes
emphasising metropolitan or national security. Most reported activities (38%) are
run by local governments, 21% by the national government7 and 14% by NGOs.
AVRP interventions principally target conflict-related violence. However,
they must also address the variety of violence types found in Colombia, such
Box 3.4. Addressing youth violence before it happens in Colombia
The Golazo project is being implemented in what are widely considered to be the
most "at-risk" areas for armed violence. The project objective is to strengthen social
development and reduce incentives to become involved in armed violence by promoting
sports activities. Children and youth are encouraged to participate in a range
of different after-school activities and parents are invited to support their children.
The Jóvenes a lo Bien initiative seeks to reduce juvenile and gang violence in major
cities, predominantly through disarmament, mediation and business-sponsored
vocational study schemes. Building on these and other activities, the national
Programme for Inclusion, Violence Prevention and Youth Employment also seeks
to reduce risk factors associated with violence by creating educational and professional
opportunities. Both private and public sector companies have joined this
latter initiative, providing jobs for approximately 200 youths, aged 18 to 29.
INVESTING IN SECURITY: A GLOBAL ASSESSMENT OF ARMED VIOLENCE REDUCTION AND PREVENTION INITIATIVES – © OECD 2011
56 – 3. CASE STUDY SUMMARIES
as protracted armed conflict, interpersonal crime, narco-trafficking, gangs
and gender. Current AVRP efforts tend to be reactive and focus on reduction
rather than prevention. They are only put into practise once the incidence of
armed violence has been identified and emphasised as a problem by public
figures and institutions.
Table 3.4 provides an overview of the "direct" and "indirect" AVRP
interventions in Colombia. It notes the emphasis on "indirect" activities:
educational interventions, youth programming, community empowerment,
media and civil awareness campaigning, skills development and treatment
and rehabilitation. It also reveals the transformation in the institutions
shaping the onset of violence, including local governance and improved
law enforcement and notes the importance of informal mediation among
perpetrators, and the role of weapons collection and destruction.
Over half of the documented "direct" interventions in Colombia focused
on controlling instruments, actors and institutions, targeting explosives,
remnants of war and small arms and light weapons (Figure 3.5). The remaining
48% "direct" interventions targeted youth. Other interventions sought to
reform the institutional environment shaping armed violence by enhancing
law enforcement, reinforcing traditional courts and dispute mechanisms, and
promoting community policing strategies (Figure 3.6).
Table 3.4. Most common "direct" and "indirect" AVRP interventions in Colombia
Direct interventions Responses Indirect interventions Responses
Instruments Informal/formal voluntary
Weapons collection and destruction 9 Education interventions 101
Voluntary gun-free zones 6 Youth programming activities 64
Weapons amnesties and buyback 3 Community empowerment interventions 46
Perpetrators Media and civil awareness campaigns 36
Informal mediation 17 Skills development programmes 35
Checkpoints 5 Treatment and rehabilitation of individuals 35
Neighbourhood watch 5 Informal/formal enforced
Private security actors 3 Better security monitoring 15
Institutions Mine action 7
Improved local/urban/national governance 24 Reducing availability and consumption of alcohol 7
Better law enforcement 10 Justice and penal reform 3
Local or traditional dispute resolution/courts 10 Community prohibition and ordinances 2
Community policing 7
INVESTING IN SECURITY: A GLOBAL ASSESSMENT OF ARMED VIOLENCE REDUCTION AND PREVENTION INITIATIVES – © OECD 2011
3. CASE STUDY SUMMARIES – 57
Colombian organisations addressing armed violence have to take account
of a wide range of risk factors. The most frequently addressed risk factor
relates to legacies of violence, followed by the presence of armed groups
and marginalised youth (Figure 3.7). These findings are consistent with the
historical patterns of armed conflict in the country, but also demonstrate that
youth are the primary perpetrators and victims of armed violence.
Monitoring and evaluation of AVRP activities is, again, substandard
with only 30% of respondents acknowledging that M&E took place. Of those
responding, 13% tracked intentional murder rates at the local level, 10%
monitored violent victimisation and 6% observed the rate of landmine and
unexploded ordnance victims. In addition, 11% assessed local perceptions of
security, another 11% monitored reported crime rates and 10% assessed other
socio-economic indices.
Figure 3.5. Specific instruments of armed violence
Percentage of responses
25%
20%
16%
14%
10%
10%
4%
0% 5% 10% 15% 20% 25% 30%
No response
No specific weapons
Explosive remnants
Ammunition
Small arms and light weapons
Knives and other bladed or
sharp instruments
Conventional military equipment
Figure 3.6. "Direct" intervention strategies in relation to institutions
Percentage of responses
Public administration reform
Community policing
No specific strategy selected
Local or traditional courts and dispute
resolution mechanisms
Better law enforcement
No response
Improved local/urban/national
governance
1%
10%
14%
14%
14%
15%
33%
INVESTING IN SECURITY: A GLOBAL ASSESSMENT OF ARMED VIOLENCE REDUCTION AND PREVENTION INITIATIVES – © OECD 2011
58 – 3. CASE STUDY SUMMARIES
The analysis of Colombia observes how both conflict and non-conflict
forms of armed violence are being addressed. While political attention has
been devoted to the DDR process, public, private and non-governmental
actors have sought to simultaneously engage with escalating urban violence.
Despite a massive expansion in AVRP activities, Colombian experts
acknowledge that greater attention is needed to enhance the monitoring and
evaluation of locally-organised interventions.
Liberia
Seven years after the Comprehensive Peace Agreement ended Liberia's
14-year civil war, the country is still unstable. Significant progress has been
made towards post-conflict reconstruction, through programmatic efforts
in Disarmament, Demobilisation, Resettlement and Reintegration (DDRR),
SSR, and the Truth and Reconciliation Commission (TRC) (Box 3.5).
Nevertheless, violence continues to occur and, in some cases, is increasing.
Violent crime is increasing in many communities (Amnesty International,
2009) with minor disputes deteriorating into assault and fighting, especially
during holiday periods when alcohol consumption increases. Armed robbery is
fuelled by high unemployment, lack of policing, and a willingness to use violent
means for economic ends – a mentality reinforced among many young Liberians
Figure 3.7. Types of proximate risk factors
Percentage of responses
0% 5% 10% 15% 20% 25% 30% 35%
Legacies of violence
Presence of armed groups
Marginalised youth
Family problems
Exposure to recent violent events
Exposure to violence
(representations)
Availability of weapons
Forced recruitment
Demand or supply of drugs
Poor performance at school
30%
20%
15%
8%
8%
6%
5%
5%
3%
1%
INVESTING IN SECURITY: A GLOBAL ASSESSMENT OF ARMED VIOLENCE REDUCTION AND PREVENTION INITIATIVES – © OECD 2011
3. CASE STUDY SUMMARIES – 59
during the war (Gompert and Stearns, 2006). Most robberies occur in the home
and increase during the rainy season.
Sexual and gender-based violence (SGBV), including rape and domestic
violence, was widely used as a weapon of war and still remains a significant
threat to women and girls (Republic of Liberia, 2008). Vigilantism, or so-called
"mob justice", is highly publicised and is perceived to be a significant threat
to individual and community security (Republic of Liberia, 2008). It is
disconcerting to note that relatively small-scale incidents can quickly escalate
into major, destabilising events.8
As a country emerging from conflict and still facing instability, armed
violence reduction is a central component of the recovery and reconstruction
process. Specifically, "peaceful" economic growth is incorporated into the
Liberian Poverty Reduction Strategy Paper (PRSP), however two of four
strategic pillars in the PRSP (e.g. economic viability and access to basic
infrastructure and services) could prioritise AVRP. Most observers agree
that a key area for enhancing armed violence prevention is employment,
especially for un/under-employed youth (Box 3.6).
Box 3.5. Ensuring adequate reintegration as part of DDRR
The DDRR Agricultural Training Programme focuses on training, resettlement
and reintegration of ex-combatants and war-affected community members. The
programme identifies ex-combatants and other "at-risk" households/groups,
enrols them in a sustained, residential agricultural training curriculum, based
on the self-identified needs of the participants, and reintegrates graduates into
communities of their choice. A programme evaluation measures the programme's
success according to the self-reported rates of economic and social reintegration
among participants and among the members of the resettlement community.
Box 3.6. Armed violence prevention through employment
The Emergency Employment Programme was designed to reintegrate thousands of
war-affected people by providing employment as an alternative to the existing war
economy. While seeking to provide an income supplement and livelihood support
for affected populations, the overall goal of the programme was to sustain the
peace process in Liberia. A key criticism, however, is that short-term employment
interventions need to be complemented by the creation and expansion of more
sustained employment opportunities or they result in disillusionment.
INVESTING IN SECURITY: A GLOBAL ASSESSMENT OF ARMED VIOLENCE REDUCTION AND PREVENTION INITIATIVES – © OECD 2011
60 – 3. CASE STUDY SUMMARIES
Official, media, and popular perspectives differ substantially on the way
armed violence is defined and perceived as a social problem. The official
view is that much investment and energy is committed to reforming the
security sector, primarily the national police and armed forces. However, the
results of these efforts are more at the central municipal level, rather than in
peri-urban and rural areas.
Media reporting on armed violence focuses on the problem of collective
communal violence (often described as "mob" or "vigilante" justice). This
may reflect the concern that sectoral reform and the strengthening of security
and judicial processes at the community level are moving too slowly, and that
dissatisfied local community groups may re-form as rival factions. Evidence has
shown that armed violence consists predominantly of assault and armed robbery/
theft. This, in turn, seems to explain why many observers believe that economic
difficulties are themselves frequently described as the cause of criminal activity.
AVRP programming trends in Liberia
Thirty-eight organisations were identified for the Liberian survey and 44
separate AVRP initiatives were analysed. The majority of these programmes
are implemented with non-governmental partners, predominantly international
and national NGOs. Eighty separate partners were featured in total. While
most organisations did not describe their activities as specifically targeting
AVRP, they endorsed the distinctions of "direct" and "indirect" interventions
and the definition of armed violence noted in the introduction.
Both SSR and DDR are acknowledged in Liberia as constituting
"direct" AVRP programming.9 SSR appears to be more common. SSR has
focused more on strengthening the Liberian National Police (LNP) and the
Armed Forces of Liberia (AFL), rather than security and justice capacity
and function at the community level. More enforced programming seems
relatively limited in the post-conflict Liberian context.
Table 3.5 highlights "direct" and "indirect" AVRP interventions in Liberia.
While "direct" AVRP interventions tend to focus on perpetrators (mediation)
and improved law enforcement, numerous "indirect" AVRP activities focus on
media and civil awareness, community empowerment, education, and youth
programming. While the list of interventions is not exhaustive, it does highlight
the heavy focus on "indirect" efforts designed to minimise the risk of armed
violence.
A considerable amount of "indirect" AVRP programming focuses on
preventing sexual violence (SGBV) and also targets "at risk" youth (Figure 3.8).
These are acknowledged priorities at the highest levels in Liberia and are relatively
consistent with the demographic profile of violence in Liberia. However, there
is less engagement with escalating banditry, theft, robbery and economically
INVESTING IN SECURITY: A GLOBAL ASSESSMENT OF ARMED VIOLENCE REDUCTION AND PREVENTION INITIATIVES – © OECD 2011
3. CASE STUDY SUMMARIES – 61
motivated criminal violence. Some respondents noted the importance of
enhancing the policing, judicial and penal capacity on the one hand, and
supporting community-based work on reducing criminality on the other.
Table 3.5. Most common "direct" and "indirect" AVRP interventions in Liberia
Direct interventions Responses Indirect interventions Responses
Instruments Informal/formal voluntary
Voluntary gun-free zones 3 Media and civil awareness campaigns 28
Weapons amnesties and buyback 1 Community empowerment interventions 22
Weapons collection and destruction 1 Education interventions 19
Perpetrators Youth programming activities 16
Informal mediation 11 Skills development programmes 14
Education 6 Job creation and employment programmes 10
Neighbourhood watch 2 Informal/formal enforced
Checkpoints 1 Better security monitoring 7
Institutions Justice and penal reform 7
Better law enforcement 8 Reducing availability and consumption of alcohol 4
Improved local/urban/national governance 7 Community prohibition and ordinances 4
Local or traditional dispute resolution/courts 6 Mine action 1
Justice and Security Sector Reform (JSSR) 5
Figure 3.8. Types of risk factors targeted by "indirect" programmes
Percentage of responses
18%
16%
14%
12%
11%
9%
7%
4%
3%
0% 5% 10% 15% 20% 25%
Gender-based discrimination
Trauma
Legacies of violence
Economic crises
Rising inequality
Cross-border trafficking
Availability of weapons
Presence of armed groups
Marginalised youth
INVESTING IN SECURITY: A GLOBAL ASSESSMENT OF ARMED VIOLENCE REDUCTION AND PREVENTION INITIATIVES – © OECD 2011
62 – 3. CASE STUDY SUMMARIES
The estimated budget reported for AVRP programming in Liberia is
approximately USD 76 million, of which USD 13.5 million is allocated to the
UNMIL component of AVRP. The average annual budget for AVRP programmes
operated by other organisations is approximately USD 910 000 (Figure 3.9),
with almost half having annual budgets of USD 250 000 or less. Most of the
programmes are planned for 2006 to 2012.
Figure 3.9. Liberia: Budgets for interventions
Percentage of responses
0% 5% 10% 15% 20% 25%
USD 100 000–250 000
USD 1 000 000–2 000 000
USD 50 000–100 000
No response
USD 25 000–50 000
> USD 2 000 000
USD 500 000–1 000 000
USD 250 000–500 000
< USD 10 000
Figure 3.10. Types of funders
Percentage of responses
31%
28%
13%
12%
6%
5%
3%
1%
1%
0% 5% 10% 15% 20% 25% 30% 35%
Multilateral donor agency
International NGO
Bilateral donor agency
National government
Regional organisation
Local community organisation
No response
National NGO
ICRC
INVESTING IN SECURITY: A GLOBAL ASSESSMENT OF ARMED VIOLENCE REDUCTION AND PREVENTION INITIATIVES – © OECD 2011
3. CASE STUDY SUMMARIES – 63
As the surveillance and data analysis capacities are weak, so too is
the monitoring and evaluation of the AVRP programmes. Given the high
proportion of non-governmental agencies involved, the absence of dedicated
assessments is alarming. Of the few AVRP programmes that claim to be
analysing trends, the indicators focus on mortality/morbidity and incidentreporting
data, but also use a number of qualitative and poorly-defined
measurements.
Overall, programmes focusing on the problem of armed violence seem
to be decreasing in Liberia. Notwithstanding the evidence generated from
the mapping assessment, a number of respondents claim that there is a
shortfall in longer-term programmatic support to ex-combatants and affected
communities (e.g. sustainable approaches to employment generation and
household income generating capacity). What is more, there appears to be
a limited emphasis on using AVRP as an explicit objective of community
level recovery and development project work. These reductions could be
considered premature.
South Africa
High levels of violence – including armed violence – have been a
prominent feature of South African society for almost two decades. While
political violence captured public attention during the late 1980s and 1990s,
following the end of apartheid and the historic 1994 elections, the public, the
public became increasingly concerned about the high levels of violent crime,
which are among the highest in the world. Thus numerous initiatives are now
being implemented to address crime and criminal violence.
AVRP programming in South Africa can be divided into three categories.
The first is increased Government investment in the criminal justice system
(most notably the police) and improved legislation. Beyond increasing the
numbers of criminal justice personnel this initiative has a rather weak
strategic focus, though this may have been remedied slightly by a recent
Criminal Justice Review.
The second category includes amending legislation to incorporate changes
lobbied for by civil society. In addition to ensuring the new legislation is
approved, civil society is also involved in the implementation of certain
elements of the new laws. The third category comprises the implementation of
numerous civil society initiatives, focusing on violence against women, firearm
violence, violent organised crime or responses to violence. These initiatives are
defined as social crime prevention, victim empowerment, restorative justice
and population health.
INVESTING IN SECURITY: A GLOBAL ASSESSMENT OF ARMED VIOLENCE REDUCTION AND PREVENTION INITIATIVES – © OECD 2011
64 – 3. CASE STUDY SUMMARIES
One category that has not appeared in contemporary analyses of South
African crime and violence over the past two decades is the investment in
private security by middle-class civilians and the formal business sector.
This may be because it is difficult to analyse in terms of programming, even
though those involved understand how to address the problem of crime.
The work undertaken by neighbourhood watches and vigilante groups is
also not classified as programming, although those governed by legislation
and/or integrated into local-level crime prevention activities are more easily
analysed.
AVRP programming trends in South Africa
The South African mapping process documented 58 programmes that qualified
as addressing AVR. Of these, 43 were administered by NGOs (9 community-
based and 34 with a broader focus) and 13 by government agencies at the
national, provincial and local levels (including inter-governmental departments
and criminal justice agencies).
Many South African programmes are preoccupied with crime or violent
crime rather than armed violence per se. Activities that do address armed violence
focus on violence against women (domestic violence, intimate partner violence or
sexual violence) and the victimisation of children, particularly in schools.
Table 3.6. Most common "direct" and "indirect" AVRP interventions in South Africa
Direct interventions Responses Indirect interventions Responses
Instruments Informal/formal voluntary
Weapons collection and destruction 3 Education interventions 35
Weapons seizure 3 Skills development programmes 34
Weapons amnesties and buyback 3 Youth programming activities 26
Law enforcement 3 Media and civil awareness campaigns 20
Perpetrators Group therapy and treatment 20
Informal mediation 4 After school activities 19
Education 2 Informal/formal enforced
Neighbourhood watch 1 Reducing availability and consumption of alcohol 12
Institutions Justice and penal reform 10
Better law enforcement 5 Better security monitoring 5
Justice and Security Sector Reform (JSSR) 5 Community prohibition and ordinances 2
Community policing 3 Strengthening formal institutions 2
Local or traditional dispute resolution/courts 3
INVESTING IN SECURITY: A GLOBAL ASSESSMENT OF ARMED VIOLENCE REDUCTION AND PREVENTION INITIATIVES – © OECD 2011
3. CASE STUDY SUMMARIES – 65
Most AVRP programming in South Africa is "indirect". A few address
armed violence directly, while five combine both "direct" and "indirect"
programming. Several targeted small armed and light weapons, with one
targeting knives and other bladed instruments. As illustrated in Table 3.6,
most reported activities in South Africa include educational interventions,
skills development, youth programming activities, media and civil awareness
campaigns, and group therapy and treatment.
The incidence of sexual violence in South Africa is amongst the highest
in the world. Likewise, particularly since the end of Apartheid, the country
has been plagued by collective violence in major cities – especially street
gangs. "Indirect" AVRP programming has therefore included gender-based
discrimination, marginalised youth and other related issues. Many respondents
are also involved in programmes addressing legacies of violence, trauma,
rising income inequality and availability of weapons, which were also cited as
key risk factors (Figure 3.11).
Figure 3.11. South Africa: Main risk factors addressed through "indirect" programming
Percentage of responses
17%
17%
13%
11%
8%
7%
5%
5%
4%
3%
2%
2%
1%
1%
1%
0% 5% 10% 15% 20% 25% 30% 35%
Gender-based discrimination
Marginalised youth
Legacies of violence
Trauma
Rising inequality
Availability of weapons
Presence of armed groups
Economic crises
Family problems
Other options
Availability of alcohol
Single and young parents
Poor performance at school
Race based discrimination
Exposure to recent violent events
INVESTING IN SECURITY: A GLOBAL ASSESSMENT OF ARMED VIOLENCE REDUCTION AND PREVENTION INITIATIVES – © OECD 2011
66 – 3. CASE STUDY SUMMARIES
Most AVRP interventions feature budgets of less than USD 250 000 per
annum. Out of 53 respondents, 17% claimed an annual budget of less than
USD 50 000. Meanwhile, 57% reported an annual budget range of between
USD 50 000 and USD 250 000. Finally, a further 26% noted budgets ranging
between USD 250 000 and USD 2 000 000 per year (Table 3.7).
Many of the AVRP programmes receive funding from a variety of sources.
On average, more than two different funding sources were recorded for each
programme. Less than half (41%) are funded by international NGOs, bilateral
donor agencies and multilateral donors agencies with 30% receiving support from
public agencies in South Africa, at the national, provincial or local level. Finally,
37% also claimed investment from companies (predominantly South African).
In contrast to virtually all other cases, nearly all respondents conduct
monitoring and evaluation activities. However, the Monitoring and Evaluation
(M&E) conducted is not necessarily formal or of a high standard. Many
specifically use indicators related to the number of incidents of violent
victimisation (28%) or injury (17%).
Timor-Leste
Over the past years Timor-Leste has experienced two major outbreaks
of violence: the struggle for independence (1975-99) and the internal armed
clashes in the security forces in 2006. The Timorese society also faces the
more chronic phenomena of violence, namely domestic violence against
women and youth violence, which is often related to long-standing grievances
among Timorese communities.10
Table 3.7. South Africa: Annual budget range of programmes (USD)
Number of responses Percent
USD 10 000–25 000 3 5%
USD 25 000–50 000 7 12%
USD 50 000–100 000 12 21%
USD 100 000–250 000 14 24%
USD 250 000–500 000 7 12%
USD 500 000–1 000 000 1 2%
USD 1 000 000–2 000 000 1 2%
> USD 2 000 000 8 14%
No response 5 9%
Total 58 100%
INVESTING IN SECURITY: A GLOBAL ASSESSMENT OF ARMED VIOLENCE REDUCTION AND PREVENTION INITIATIVES – © OECD 2011
3. CASE STUDY SUMMARIES – 67
The most severe episode of collective violence occurred during the
struggle for Timorese independence. According to the final report of the
Commission for Reception, Truth and Reconciliation (CAVR) between 1974
and 1999 almost one-quarter (102 800 to 183 000 people) of the Timorese
population died as a result of armed violence.
The more recent outbreaks of violence in 2006, 2007 and 2008 are
comparatively modest. In 2006, for example, a group of F-FDTL soldiers
deserted and initiated a dispute within the security forces, including the
police. During the course of the tensions, some 38 people died and many
more were injured (United Nations, 2002). More significant was the number
of internally displaced persons that reached 150 000, most of whom were
clustered in the capital city, Dili.
Several categories of armed violence have been identified in Timor-Leste
in the post-independence period. These are typically associated with gangs,
martial arts groups, and domestic violence. While the former groups were
forged and fostered during Indonesian occupation and played a role in the
independence movement, gangs are a more recent phenomenon. Concern
escalated during the 2006 outbreak of violence as some gangs appeared to
have connections to and be manipulated by political elites.
It is important that youth gangs in Timor-Leste be studied separately.
Some of these groups are classified as grass-roots social movements defending
the interests of their communities, while others are more closely identified
with criminal networks (Scambary, J., 2006; Scambary, J., 2009). Others,
particularly martial arts groups, maintain closer relations with Timorese
security forces, which is an obstacle to establishing a comprehensive strategy.
Most analysts in Timor-Leste claim that sexual and gender-based violence
– especially domestic violence – is the main category of armed violence.
A 2008 opinion poll by the Asia Foundation showed that 15% of Timorese
families had experienced domestic violence during the previous two years,
while only 7% were assaulted by unknown individuals.11 Other analysts
note that property disputes and the possibility of armed violence between
rival groups and families over property and title issues, have become more
common.
AVRP programming trends in Timor-Leste
The mapping process in Timor-Leste focused on a smaller selection of
"direct" and "indirect" AVRP programmes. As in other cases, armed violence
(and AVRP) is not a common concept in Timor-Leste. Moreover, interventions
designed to address armed violence in Timor-Leste tend to focus on specific
sectors or actors rather than on adopting integrated or comprehensive
approaches.
INVESTING IN SECURITY: A GLOBAL ASSESSMENT OF ARMED VIOLENCE REDUCTION AND PREVENTION INITIATIVES – © OECD 2011
68 – 3. CASE STUDY SUMMARIES
"Direct" programming in Timor-Leste was comparatively limited. A few
interventions focused on the instruments,12 while others focused on actors
and institutions. Legislative reform in the security sector, police training
and improvements to law enforcement, and strengthening conflict resolution
mechanisms among youth groups, gangs, and martial arts groups were most
common.
Most AVRP interventions in Timor-Leste are "indirect" (84%) (Table 3.8).
Approximately 20% claim to undertake a combination of both "direct" and
"indirect" activities. Most "indirect" interventions target legacies of violence
and marginalised youth, in addition to grievances arising from economic
crises and inequality (Table 3.8).
Both "direct" and "indirect" AVRP interventions are predominantly
funded, designed and implemented, by multilateral and bilateral donor
agencies and their partners. This tends to be a common feature of postconflict
societies, particularly in developing countries. This is more evident
in Timor-Leste because the United Nations accompanied the country through
independence and its transitional administration. Several interviewees
commented that the UN system needs Timor-Leste to be a success story and
therefore commits a high amount of human and financial resources for the
stabilisation and socio-economic development of this country.
Approximately 36% of AVRP activities in Timor-Leste underwent some
kind of monitoring. Examples of indicators used include displacement and
resettlement rates of IDPs and perception of security. Core indicators of
Table 3.8. Most common "direct" and "indirect" interventions in Timor-Leste
Direct interventions Responses Indirect interventions Responses
Against instruments Informal/formal voluntary
Weapons collection and destruction 3 Youth programming activities 6
Against perpetrators Education interventions 6
Neighbourhood watch 4 After school activities 4
Checkpoints 2 Skills development programmes 3
Informal mediation 1 Targeted employment schemes 3
Private security actors 1 Incentive-based DDR 2
Against institutions Informal/formal enforced
Improved local/urban/national governance 5 Justice and penal reform 6
Better law enforcement 5 Better security monitoring 4
JSSR 3 Community prohibition and ordinances 1
Public administration reform 3
INVESTING IN SECURITY: A GLOBAL ASSESSMENT OF ARMED VIOLENCE REDUCTION AND PREVENTION INITIATIVES – © OECD 2011
3. CASE STUDY SUMMARIES – 69
armed violence, like the homicide rate, direct and indirect battle deaths and
violent victimisation, are not necessarily monitored. This could be because
most programmes are "indirect".13
Respondents noted that AVRP programming in Timor-Leste is undergoing
a period of transition. Most of the operational initiatives were designed in direct
response to the crisis in 2006/07. Having addressed the more imminent security
threats created by this crisis, programmes then focused on the recovery of
the society in Timor-Leste (e.g. the return of IDPs) (Box 3.7). Many of these
programmes, including the interventions by United Nations Mission in Timor-
Leste (UNMIT) have nearly finished or need extending. The long-term planning
of major actors, like United Nations Development Programme (UNDP), is
gradually changing to more conventional development assistance.
Figure 3.12. Types of risk factors addressed by "indirect" programming
Percentage of responses
27%
22%
16%
12%
10%
4%
4%
2%
2%
0% 5% 10% 15% 20% 25% 30% 35%
Legacies of violence
Marginalised youth
Economic crises
Rising inequality
Gender-based discrimination
No response
Presence of armed groups
Availability of weapons
Land and property issues
Table 3.9. Types of funders
Responses
N Percent
Bilateral donor agency 14 41.2
Multilateral donor agency 13 38.2
National government 3 8.8
No response 4 11.8
Total 34 100.0
INVESTING IN SECURITY: A GLOBAL ASSESSMENT OF ARMED VIOLENCE REDUCTION AND PREVENTION INITIATIVES – © OECD 2011
70 – 3. CASE STUDY SUMMARIES
In the meantime, other international actors are starting their planned exit
from Timor-Leste (e.g. the Norwegian Refugee Council), and local NGOs will
be expected to fill these gaps. The multiplicity and diversity of the local NGOs
is such that they are quite capable of doing so. Additionally, international
actors have invested a lot of resources to strengthen human capacity in the
local civil society for planning and implementing programmes.
However, the experience of 2006 demonstrates that an early withdrawal
without sustainable reforms can result in the re-eruption of violence. The
two most pressing issues, comprehensive SSR and land law reform, seem
to be very difficult to implement in view of the lack of committed partners
within the government. Innovative programming is needed which offers
suitable incentives to governmental actors. This task can only be undertaken
by international actors and their ability to deliver will affect the long-term
stability and development of Timor-Leste.
Notes
1. In both Rio de Janeiro and São Paolo, local governments are tracking the
relationships between armed violence and MDG achievement. One group, the
Public Security Forum, is using the Survey of Living Conditions conducted by
the Seade Foundation in 2006 to assess the attainment of MDG indicators related
to income, employment, sanitation and housing conditions. This consists of a
household survey of approximately 20 000 households, including 5 500 respondents
in the Metropolitan Region of São Paulo. (Muggah and Restrepo, 2011).
Box 3.7. Addressing reintegration of IDPs for peace
The Strengthening Early Recovery for Comprehensive and Sustainable Reintegration
of IDPs Project initially offered financial and food assistance, as well as
transportation for the return of IDPs. So-called "social mobilisers" were trained
to engage with community level councils in order to reach more inclusive and
participatory planning, involving local stakeholders. Particular focus was placed
on support for infrastructure projects to increase social and economic interaction
and foster bonds between IDPs and receiving/host communities.
INVESTING IN SECURITY: A GLOBAL ASSESSMENT OF ARMED VIOLENCE REDUCTION AND PREVENTION INITIATIVES – © OECD 2011
3. CASE STUDY SUMMARIES – 71
2. Figures based on the Armed Violence Observatory of Burundi (Pézard and de
Tessières, 2009).
3. For a review of the evidence of DDR interventions, see (Muggah [ed.] 2009)
4. In Burundi, AVRP interventions target individuals (35% of "direct" interventions)
more than communities (14% of "direct" interventions).
5. For a review of trends and dynamics of armed violence in Colombia since the
1950s, see, for example, Small Arms Survey (2006).
6. The first period witnessed a shift of focus away from national security towards
interventions targeting the prevention and reduction of urban violence. The second
period saw a consolidation and increase in the number of AVRP programmes
throughout the country, predominantly associated with the DDR process.
7. From a budgetary perspective, however, it is the national government that tends
to be the principal funding body.
8. In February, at least four people were reportedly killed when Muslims and
Christians clashed in the northern city of Voinjama. United Nations peacekeepers
and local security forces were able to intervene, but there were fears that violent
retributions could destabilise Monrovia (United Nations, 2009).
9. An emphasis in "direct" programming on small arms and light weapons does
not seem in accord with the types of weapons recorded in most cases of (armed)
violence.
10. For an examination of urban violence and household survey findings in Timor-
Leste see Muggah (ed.), 2010.
11. The same opinion poll also showed that over the past five years the situation
for women in Timor-Leste has deteriorated. In 2004, 19% of the interviewees
responded that men have the right to hit their wives. But in 2008, 21% were of
the same opinion. Meanwhile, in 2004, 75% of respondents rejected the right of
husbands to beat their wives while in 2008 just 34% agreed with this statement.
Some 44% argued that the right of the husband needed to be assessed on a caseby-
case level.
12. Addressing the instruments of violence in the form of weapons collection was the
focal point of only two initiatives run by the Timorese security forces, namely
operation Halibur and operation Kilat.
13. There are two noteworthy exceptions (run by BELUN) that monitor violence
more directly, but in these programmes monitoring is not used in order to assess
the impact of the programmes, but is itself the goal of these initiatives.
INVESTING IN SECURITY: A GLOBAL ASSESSMENT OF ARMED VIOLENCE REDUCTION AND PREVENTION INITIATIVES – © OECD 2011
CONCLUSION – 73
Conclusion
This report represents a groundbreaking effort to map out armed violence
reduction and prevention (AVRP) programmes around the world. It introduces an
innovative conceptual framework and survey methodology, and new empirical
material. It is not a "how to" guide to programming, but rather a descriptive
overview of the state of AVRP programming. While not exhaustive – the focus
was primarily on mapping six settings and 570 initiatives – it is substantial. The
findings are illustrative of the many thousands of programmes being advanced to
prevent armed violence in lower- and middle-income contexts.
First, the assessment revealed the enormous number of activities being
undertaken with respect to the prevention and reduction of armed violence.
Many of these are focused "directly" on controlling and reducing access
to weapons, engaging perpetrators and reforming legislation and security
practices. However, the majority of interventions are pursued "indirectly"
– seeking to manipulate and diminish the proximate and structural risks of
armed violence at their source. At the forefront of AVRP are those interventions
combining both "direct" and "indirect" approaches – targeting both the risks
and symptoms – many of which are pursued at the municipal level. Accordingly,
development agencies should ensure their support focuses on comprehensive
and community-focused interventions.
Second, despite the scale and scope of AVRP, the report finds that the
descriptive label "armed violence" is not commonly applied in practice.
For example, in South Africa the focus amongst public authorities and nongovernmental
organisations tends to be on preventing and reducing criminal,
domestic and youth violence. In Colombia and Brazil (and indeed across
Latin America and the Caribbean), citizen and community security are used
as synonyms for armed violence prevention and reduction. OECD donors and
international agencies will need to adapt their terminology to local contexts
if they are to advance wider AVRP priorities in the future.
Third, the report highlights the fundamental role of development actors
– from multilateral and bilateral donors to non-governmental organisations
(NGOs) and civic action groups – in promoting AVRP. Across all cases, interventions
highlighted the ways in which organisations could tighten their focus
INVESTING IN SECURITY: A GLOBAL ASSESSMENT OF ARMED VIOLENCE REDUCTION AND PREVENTION INITIATIVES – © OECD 2011
74 – CONCLUSION
on recurring development challenges (poverty and income inequality in high
risk areas, youth unemployment and literacy, youth recreation and cultural
activities, family planning and early childhood development, etc.,) to prevent
and reduce armed violence. Equally, it is critical for investments to be made
into strengthening the capacities of local partners and partnerships across sectors
to monitor and measure performance.
While the wealth of small-scale and innovative programmes reflects the
dynamism and social entrepreneurship that exists in this field, future successes
will require in-depth evaluations, investments to scale-up activities and the
development of long-term programming interventions. All six case studies have
demonstrated the fundamental importance of evidence-generation and the key
role of innovative partnerships – particularly between public authorities, local
civil society actors and the private sector – with international agencies playing
a facilitating and supportive role. Encouraging a partnership-driven approach
can also enhance the legitimacy and capacity of actors in affected areas.
Ultimately, while small, short-lived initiatives are often essential for catalysing
action and generating demonstration effects, they are not a long-term solution.
INVESTING IN SECURITY: A GLOBAL ASSESSMENT OF ARMED VIOLENCE REDUCTION AND PREVENTION INITIATIVES – © OECD 2011
BIBLIOGRAPHY – 75
Bibliography
Act Knowledge (2009), Theory of Change; Theory of What? www.
Amnesty International (2009), Human Rights in Republic of Liberia, Human
report-2009.
Bellis, M.A., et al. (2010), Preventing and Reducing Armed Violence:
What Works?, United Nations Development Programme and Norwegian
Ministry of Foreign Affairs, New York and Oslo, www.preventviolence.
info/showResourcespdf.aspx?id=18a25404-4a63-4353-8055-cbb3ceb116ff.
Geneva Declaration on Armed Violence and Development (2006), www.
Gompert, D.C. and B. Stearns (2006), "Comprehensive Security Sector
Reform: making Liberia safe", Africa Policy Journal, Fall, Vol. 2, Harvard
University, Boston.
GTZ (Deutsche Gesellschaft für Technische Zusammenarbeit) (2007),
Securing a Better Future: Martial Arts Groups and Their Potential to
Become Part of a Peace Constituency in Timor-Leste, GTZ, Dili.
Human Rights Watch (HRW) (2009), Pursuit of Power: Political Violence
reports/2009/06/03/pursuit-power.
IADB (Inter-American Development Bank) (2003), Guidelines for the Design
of Violence Reduction Projects, IADB, Washington, DC.
Krause, K., R. Muggah and A. Wenmann (2008), Global Burden of Armed
Violence, Geneva Declaration Secretariat, Geneva.
Krause, K., R. Muggah and E. Gilgen (2011), Global Burden of Armed
Violence, Cambridge University Press, Cambridge.
INVESTING IN SECURITY: A GLOBAL ASSESSMENT OF ARMED VIOLENCE REDUCTION AND PREVENTION INITIATIVES – © OECD 2011
76 – BIBLIOGRAPHY
Lemarchand, R. (1997), "The Burundi Genocide", in S. Totten, W. Parsons
and I. Charny (eds.), Century of Genocide: Eyewitness Accounts and
Critical Views, Garland Publishing, New York., pp. 317-333.
Marc, A. (2009), Making Societies More Resilient to Violence: A Conceptual
Framework for the Conflict, Crime, and Violence Agenda. Conflict,
Crime and Violence Paper, World Bank, Washington, DC.
McLean, A. and S. Blake Lobban, Assessment of Community Security and
Transformation Programmes in Jamaica, forthcoming.
Muggah, R. (2005), "No Magic Bullet: A Critical Perspective on
Disarmament, Demobilization and Reintegration (DDR) and Weapons
Reduction in Post-conflict Contexts", The Round Table, Vol. 94, Iss. 379,
pp. 239-252.
Muggah, R. (ed.) (2009) Security and Post-Conflict Reconstruction: Dealing
with Fighters After War, Routledge, London
Muggah, R. (ed.) (2010), Urban Violence in an Urban Village: Case Study of
Dili, Geneva Declaration Secretariat, Geneva.
Muggah, R. and Restrepo, J. (2011), "More Violence Less Development", in
Global Burden of Armed Violence, Cambridge University Press, Cambridge.
National Programme for Public Citizen Security (PRONASCI),
http://portal.
OECD (Organisation for Economic Co-operation and Development) (2009a),
Armed Violence Reduction: Enabling Development, OECD, Paris.
OECD (2009b), Guidance on Evaluating Conflict Prevention and Peacebuilding
Activities, OECD, Paris.
OECD (2010a), International Support to Statebuilding in Situations of
Fragility and Conflict, OECD, Paris.
OECD (2010b), Accelerating Progress Towards the Millennium Development
Goals in Countries Affected by Conflict, Violence and Insecurity,
Document DCD/DAC/INCAF(2010)3, OECD, Paris.
Oslo Commitments on Armed Violence (2010), www.osloconferencearmedviolence.
no/.
Pézard, S, and S. de Tessières (2009), Insecurity is Also a War: An Assessment
of Armed Violence in Burundi, Geneva Declaration Secretariat, Geneva.
Public Security Forum (2010), Anuário do Fórum Brasileiro de Segurança
Pública. Brazilian Public Security Forum, São Paulo.
INVESTING IN SECURITY: A GLOBAL ASSESSMENT OF ARMED VIOLENCE REDUCTION AND PREVENTION INITIATIVES – © OECD 2011
BIBLIOGRAPHY – 77
Republic of Liberia (2008), Poverty Reduction Strategy Paper, IMF Country
Scambary, J. (2006), "A Survey on Gangs and Youth Groups in Dili, Timor-
Youth_Gangs_in_Dili.pdf.
Scambary, J. (2009), "Groups, gangs, and armed violence in Timor-Leste",
Leste-Violence-IB2-ENGLISH.pdf
theories_of _change.
Small Arms Survey (2006), Small Arms Survey 2006: Unfinished Business,
Oxford University Press, Oxford.
violence_injury_ prevention/violence/activities/en/avpp_overview.
UN (2006), Report of the Independent Special Commission of Inquiry for
Timor-Leste, United Nations, Geneva.
UN (2009), Nineteenth progress report of the Secretary-General on the
United Nations Mission in Liberia, Report to the Security Council
S/2009/411, United Nations, New York, p. 4.
Uvin, P. (2009), Life After Violence: A People's Story of Burundi, Tufts,
Boston.
UNSG (United Nations Secretary-General) (2009), Promoting Development
through the Reduction and Prevention of Armed Violence, Document
A/64/228, United Nations, New York.
Waiselfisz, M. and J. Jacobo (2010), Mapa da violência 2010: Uma Anatomia
dos Municípios Brasileiros, Instituto Sangari, São Paulo.
WHO (World Health Organization) (2002), World Report on Violence and
Health, WHO, Geneva.
WHO (2004), Handbook for the Documentation of Interpersonal Violence
Prevention Programmes, WHO, Geneva.
WHO (2009), Violence Prevention: The Evidence – Overview, WHO,
Geneva.
ORGANISATION FOR ECONOMIC CO-OPERATION
AND DEVELOPMENT
The OECD is a unique forum where governments work together to address the economic,
social and environmental challenges of globalisation. The OECD is also at the forefront of efforts
to understand and to help governments respond to new developments and concerns, such as
corporate governance, the information economy and the challenges of an ageing population.
The Organisation provides a setting where governments can compare policy experiences, seek
answers to common problems, identify good practice and work to co-ordinate domestic and
international policies.
The OECD member countries are: Australia, Austria, Belgium, Canada, Chile, the
Czech Republic, Denmark, Estonia, Finland, France, Germany, Greece, Hungary, Iceland, Ireland,
Israel, Italy, Japan, Korea, Luxembourg, Mexico, the Netherlands, New Zealand, Norway, Poland,
Portugal, the Slovak Republic, Slovenia, Spain, Sweden, Switzerland, Turkey, the United Kingdom
and the United States. The European Union takes part in the work of the OECD.
OECD Publishing disseminates widely the results of the Organisation's statistics gathering
and research on economic, social and environmental issues, as well as the conventions, guidelines
and standards agreed by its members.
THE GENEVA DECLARATION ON ARMED VIOLENCE AND DEVELOPMENT
The Geneva Declaration on Armed Violence and Development, endorsed by more than 100
countries, commits signatories to supporting initiatives intended to measure the human, social,
and economic costs of armed violence, to assess risks and vulnerabilities, to evaluate the
effectiveness of armed violence reduction programmes, and to disseminate knowledge of best
practices. The Declaration calls upon states to achieve measurable reductions in the global burden
of armed violence and tangible improvements in human security by 2015. Core group members
include Brazil, Colombia, Finland, Guatemala, Indonesia, Kenya, Morocco, the Netherlands,
Norway, the Philippines, Spain, Switzerland, Thailand, and the United Kingdom. Affiliated
organizations include the Bureau of Crisis Prevention and Recovery (BCPR) and the United Nations
Development Programme (UNDP), the Development Assistance Committee (DAC) of the
Organisation for Economic Co-operation and Development (OECD), and the Quaker United Nations
Office (QUNO). For more information about the Geneva Declaration, related activities, and
THE UNITED NATIONS DEVELOPMENT PROGRAMME (UNDP)
The United Nations Development Programme (UNDP) is the UN's global development network,
advocating for change and connecting countries to knowledge, experience and resources to help
people build a better life.
OECD PUBLISHING, 2, rue André-Pascal, 75775 PARIS CEDEX 16
(43 2011 23 1 P) ISBN 978-92-64-12453-0 – No. 59035 2011
Please cite this publication as:
OECD (2011), Investing in Security: A Global Assessment of Armed Violence Reduction
Initiatives, Conflict and Fragility, OECD Publishing.
This work is published on the OECD iLibrary, which gathers all OECD books, periodicals and
information.
-:HSTCQE=VWYZXU: ISBN 978-92-64-12453-0
43 2011 23 1 P
Conflict and Fragility
Investing in Security
A GLOBAL ASSESSMENT OF ARMED VIOLENCE REDUCTION
INITIATIVES
Contents
Chapter 1. Conceptualising armed violence reduction and prevention
• Conceptual framework
• Introducing the typology
• Promising AVRP initiatives
Chapter 2. Mapping armed violence reduction and prevention programming trends
Chapter 3. Case study summaries
• Brazil
• Burundi
• Colombia
• Liberia
• South Africa
• Timor-Leste
