Security Sector Reform in Central Asia: Exploring the Policy-Practice GAP of Police Reforms and the Civil Society Factor in Kazakhstan and Kyrgyzstan

Security Sector Reform in Central Asia: Exploring the Policy-

Practice Gap of Police Reforms and the Civil Society Factor in

Kazakhstan and Kyrgyzstan

Master thesis

Olivier Korthals Altes

Radboud University Nijmegen – School of Management

MSc Conflicts, Territories and Identities Master thesis supervisor: Dr. H.W. Bomert Second reader: Dr. M. van Leeuwen Version August 14, 2018

i S4810007 O.P. Korthals Altes

Security Sector Reform in Central Asia: Exploring the Policy-

Practice Gap of Police Reforms and the Civil Society Factor

in Kazakhstan and Kyrgyzstan

Olivier Korthals Altes

For the cover page, images have been used from the following sources, clockwise fro m top left: Yaroslavskaya (2011), Abiba kir (2017), The Prime M inister of Ka zakhstan Website (2016) and Flagpedia. The p icture on the bottom le ft is taken by myself at the OSCE Conference on Security Sector Governance and Reform, Bishkek, December 19, 2017.

ii S4810007 O.P. Korthals Altes

Table of contents

1. Introduction………...………..1

1.2 Research objective and research questions………....4

1.3 Scientific and societal relevance………....5

1.3.1 Scientific relevance………....5

1.3.2 Societal relevance………...6

1.4 Research methodology and data collection ……….………...8

1.5 Structure………....10

2. Literature review and conceptual framework………..………...12

2.1 The concepts of Security Sector Reform, democratic governance, police reforms and civil society...…12

2.1.1 Security Sector Reform………...11

2.1.2 Democratic governance of the security sector..………...16

2.1.3 Police reforms………..…………18

2.1.4 Civil society………..20

2.2 Bridging the policy-practice gap………..……….23

2.3 SSR and state-building………….……….25

2.4 Towards the second generation of SSR……….27

2.5 Conceptual framework………..29

2.5.1 Contextual analysis………29

2.5.2 Policy formulations………...32

2.5.3 Measuring the progress of SSR and introducing the indicators ………..……….33

2.6 Summary………37

3. Kazakhstan and Kyrgyzstan in context……….……….38

3.1 Regional context………38

3.2 Kazakhstan……….41

iii S4810007 O.P. Korthals Altes

3.2.2 Conflict, security and the police……….43

3.2.3 Civil society………...45

3.3 Kyrgyzstan……….49

3.3.1 State, polity and society……….49

3.3.2 Conflict, security and the police……….……50

3.3.3 Civil society……….…..52

3.4 Summary………55

4. The Kazakh and Kyrgyz policies on police reforms, and the role of international

organizations………..………57

4.1 Kazakhstan……….59

4.1.1 The role of international organizations………59

4.2 Kyrgyzstan ………63

4.2.1 The role of international organizations……….65

4.3 Summary ………...69

5. The Kazakh and Kyrgyz practices of police reforms, and the role of civil society…….…71

5.1 Kazakhstan……….71

Indicator #1: Public oversight and monitoring………..………71

Indicator #2: Integrity…..………..73

Indicator #3: Participation……..………...…………76

Indicator #4: Transparency……..………..78

Indicator #5: Upholding the rule of law and human rights……….………..79

Indicator #6: Responsiveness………80

5.2 Kyrgyzstan………..………...81

Indicator #1: Public oversight and monitoring………..………...81

Indicator #2: Integrity…..………..83

Indicator #3: Participation…..………...…………85

Indicator #4: Transparency……..………..87

Indicator #5: Upholding the rule of law and human rights……….………..88

Indicator #6: Responsiveness….………...89

iv S4810007 O.P. Korthals Altes

6. Conclusions………92

6.1 Answers to research questions………...92

6.2 Recommendations………..94

6.2.1 Policy implications……….94

6.2.2 Suggested research directions………95

6.3 Reflections and limitations……….95

Bibliography……….97

Appendixes………..104

Appendix I CMR Index………104

Appendix II Overview of SSR related indexes and datasets………105

Appendix III Transcriptions of conducted interviews and surveys..……….………..107

v S4810007 O.P. Korthals Altes

Executive summary

Security Sector Reform (SSR) as a current dominant paradigm in security assistance programmes aims at (re)building and professionalising security institutions while creating and consolidating democratic control over these institutions. Democratically governed security institutions are accountable and transparent for public oversight and must guarantee citizens’ rights and freedoms while executing its primary tasks. While Security Sector Reform has enjoyed great successes in its conceptualisation and adoption in key policy documents of most international organisations working on security and democratisation, its impact remains limited and its empirical base rather weak. In many SSR programmes, a policy-practice gap can be identified, or a discrepancy between prescribed strategies and policies and the progress of SSR over time.

While all-encompassing SSR fieldwork studies are usually conducted over an extended period of time, I have decided to focus on the policy-practice gap of democratising the police forces in Kazakhstan and Kyrgyzstan. The police forces are the most profound security institutions in Central Asia, originating from the Soviet period and hard to change. Without paying attention to the dimension of democratic governance, external security assistance to police forces might lead to tolerating politicised and corrupt security forces, which negatively affects the long-term political and economic stability. SSR programs need to be implemented carefully in the political and social context. Where institutional and legal frameworks to democratically govern the security forces ( in this case, the police) could be installed in time, changing the actual culture and behavioural attitude towards democratic governance turns out to be more difficult, as human nature is impossible to manage according to fixed programme schedules.

Based on the existing literature, a conceptual framework has been designed to analyse the regional and national context of Kazakhstan and Kyrgyzstan, their policy formulations, and progress or practice towards police reforms based six indicators: #1 Public oversight and monitoring; #2 Integrity; #3 Participation of citizens and civil society; #4 Transparency; #5 Upholding the rule of law and human rights; and #6 Responsiveness to local citizens. Furthermore, key literature on SSR, development and democratisation points at the important role of civil society, that has the necessary knowledge and skills to strengthen state-citizens relationships as well as guaranteeing local ownership. To test the important factor of civil society, I have identified a number of key CSOs involved in police reforms in both countries.

Overall, one must be critical when analysing strategic and policy documents that refer to notions such as trust building, enhancing transparency or respecting the freedoms and rights of citizens. Such written commitments or statements are not always consistent with actual effor ts, especially when analysed within an authoritarian or flawed democratic setting like the one in Kazakhstan or Kyrgyzstan. Rather, these documents must be read next to sources of relevant online databases, CSOs, (international) NGOs and research institutes that

vi S4810007 O.P. Korthals Altes

approach the actual practice of police reforms on the ground more completely. Empowering civil society alone will not be enough, as long as the attitude of the government, the Ministry of Internal Affairs and the broader public (especially in Kazakhstan) towards democratic reforms will not change. Based on this research, civil society could therefore be seen as an important but not the only essential factor in bridging the policy-practice gap of police reforms.

vii S4810007 O.P. Korthals Altes

Preface

One and a half year ago, I didn’t have any idea how I would conduct my master thesis research on security issues in Central Asia, leave alone within the framework of what internship as obligatory part of the Master’s specialization of Conflicts, Territories and Identities at Radboud University Nijmegen. It was out of historic interest that I wanted to pick out one or more Central Asian states, sharing the same Soviet past, as case studies for my master thesis research. In February 2017, the International O ffice suggested me to apply for a Central Asian European Universities (CASEU) scholarship, as Radboud University is linked in this exchange network between a number of European and Central Asian universities. In May 2017, I was gladly told to be selected for this scholarship to spend six months as an Erasmus Mundus exchange student at the Al-Farabi Kazakh National University (KazNU) in Almaty, Kazakhstan, from September 2017 until March 2018.

Next to working on my master thesis, I followed courses for one semester at the local Department of international Relations to keep my scholarship. My journey to speak to experts on my research topic of SSR brought me to multiple institutions that include KIMEP University in Almaty, Nazarbayev University in Astana, and the OSCE Academy and the American University of Central As ia in Bishkek. Moreover, I became associate research fellow at the OSCE Academy to make use of the academy’s expertise and facilities and presented the preliminary research findings on February 23, 2018. During my exchange period, I was glad to have met and exchanged views with so many international and regional students.

First of all, I would like to thank my supervisor Bert Bomert for guiding me throughout the whole writing process of my thesis that had lasted over one and a half a year, and Mathijs van Leeuwen for grading my thesis as second reader. Furthermore I want to thank Marijke Delamarre for her support and enthusiasm to undertake this step of engaging in the CASEU exchange programme, and Evert van der Zweerde for sharing his experiences at KazNu before my departure. From KazNU, I want to thank my contact person Nazym Makhmutova, who always helped me out with all practical issues, and Zhulduz Baizakova and Fatima Kukeyeva, for their teachings and perspectives on my research topic. From the OSCE Academy I want to thank Alexander Wolters and Indira Saturkulova for offering me the associate research fellowship position, and Payam Foroughi for the discussions on my master thesis research. I also want to thank Nargiz Kassenova and Meruert Makhmutova fro m Kazakhstan, and Aida Alymbayeva and Raushan Bolataliyeva from Kyrgyzstan, for their recommendations on my research topic and indicating the civil society organizations engaged in SSR and police reforms.

Olivier Korthals Altes Rotterdam, August 2018

viii S4810007 O.P. Korthals Altes

List of abbreviations

ADB Asian Development Bank

AIPA Asian Institute of Poverty Alleviation ASEAN Association of Southeast Asian Nations

AU African Union

BOMCA Border Management in Central Asia BTI Bertelsmann Transformation Index CACO Central Asian Cooperation Organization CADAP Central Asia Drug Action Programme CAT Coalition Against Torture

CEE Central and Eastern Europe CiB [OSCE] Centre in Bishkek

CIS Commonwealth of Independent States CMR Civil-Military Relations

CPI Corruption Perceptions Index CSI Civil Society Index

CSI Community Security Initiative CSO Civil Society Organization

CSTO Collective Security Treaty Organization DCAF Democratic Control of Armed Forces DCI Development Cooperation Instrument DFID Department for International Development EAEU Eurasian Economic Union

ECOWAS Economic Community of West African States

EIDHR European Instrument for Democracy and Human Rights EPCA Enhanced Partnership Cooperation Agreement

EU European Union

GDP Gross Domestic Product

KIBHR Kazakhstan International Bureau for Human Rights and Rule of Law ICG International Crisis Group

IDB Inter-American Development Bank IPAP Individual Partnership Action Plan LCG Local Self-Governance

LCPC Local Crime Prevention Centre LPRC Legal Policy Research Centre MIA Ministry of Internal Affairs

NATO North Atlantic Treaty Organization NGO Non-Governmental Organization OAS Organization of American States

OECD Organization for Economic Cooperation and Development OECD ACN OECD Anti-Corruption Network

OECD DAC OECD Development Assistance Committee

OSCE Organization for Security and Cooperation Europe PAC Public Advisory Council

PAP Police Assistance Programme PARP Planning and Reform Process PCA Partnership Cooperation Agreement

ix S4810007 O.P. Korthals Altes

PfP Partnership for Peace PMO Prime Minister’s Office PMP Police Matters Programme PRI Penal Reform International PRP Police Reform Programme

SCO Shanghai Cooperation Organization SSR Security Sector Reform

SSRI Security Sector Reform Index

UN United Nations

UNDP United Nations Development Programme

UN DPKO United Nations Department of Peacekeeping Operations VERLT Violent Extremism and Radicalization that Lead to Terrorism WGI Worldwide Governance Indicator

x S4810007 O.P. Korthals Altes

List of figures

Table 2.1 Transitional category (Stojanovic 2009, p. 80)………..31

Table 2.2 Contextual Analysis of Central Asia, Kazakhstan and Kyrgyzstan………...32

Table 2.3 Policy formulations towards police reforms in Kazakhstan and Kyrgyzstan………33

Table 2.4 Measuring the progress of democratic governance of police forces in Kazakhstan and Kyrgyzstan……36

Table 3.1 Civil society actors engaged in police reforms in Kazakhstan……….……….48

Table 3.2 Civil society actors engaged in police reforms in Kyrgyzstan………..54

Table 4.1 Kazakhstan’s policies towards police reforms………...62

Table 4.2 Kyrgyzstan’s policies towards police reforms………...69

Figure 5.1 Bertelsmann Transformation Index of Kazakhstan between 2006-2018……….73

Figure 5.2 Corruption Perceptions Index of Kazakhstan between 2003-2011………..75

Figure 5.3 Corruption Perceptions Index of Kazakhstan between 2011-2017………..75

Figure 5.4 WGI Voice and Accountability of Kazakhstan between 2003-2016………...……77

Figure 5.5 Bertelsmann Transformation Index of Kyrgyzstan between 2006-2018………..82

Figure 5.6 Corruption Perceptions Index of Kyrgyzstan between 2003-2011………...84

Figure 5.7 Corruption Perceptions Index of Kyrgyzstan between 2011-2017………...84

1 S4810007 O.P. Korthals Altes

1. Introduction

1.1 Background

Law enforcement agencies in Central Asia are in urgent need of reform within the context of a dynamic, contrasting and changing environment. Since their independence from the Soviet Union in 1991, the five Central Asian states – Kazakhstan, Kyrgyzstan, Tajikistan, Turkmenistan and Uzbekistan – have been confronted with political upheavals and security issues in response to the rapid transformations of their economies and political systems, from an authoritarian to a more democratic rule, and from a closed society towards further integration into the global economy. Simultaneously, these relatively young countries also have to deal with outdated and bureaucratic institutions, a lack of capital and knowledge, and laggard economies, due to their overdependence on natural resources or conditions of underdevelopment. Furthermore, important security challenges for all Central Asian countries are authoritarian governance, deeply-rooted state corruption, tensions over water and energy resources, and, to a certain extent, Islamist and regionalist insurgencies. In the meantime, Central Asian countries might be sensitive to security threats that spill over as a result of the ongoing war in Afghanistan, such as international terrorism, refugee waves and drug trafficking (Boonstra, Laruelle & Peyrouse 2014).

While the military plays a more limited role in Central Asia, the police forces are at the forefront in tackling these (transnational) security threats and providing a secure environment, necessary for peaceful and stable political and economic development. Experts argue that Central Asian states could benefit from police reforms within a broader agenda on improving internal security and development, in line with the comprehensive approach of Security Sector Reform (SSR) (Faltas 2010, pp. 16-17). This concept of SSR, originating from the international fields of conflict, security, and development, generally entails the (re)building of state-centred security services, controlled by democratic governance. It is a process that should lead to more effective and accountable security forces while at the same time raising public trust, accountability, and legitimacy among the local population. Consequently, SSR could help Central Asian states in reaching their overarching targets of gaining and maintaining political and societal stability, developing and modernising its economy, and realising democratic reforms. Regarding stability, SSR supports modernising and professionalising the police forces, making them more effective in overcoming security challenges. In light of the stimulation of the economy, SSR promotes transparent and accountable governance of the security sector, which in turn encourages and attracts domestic and foreign investments. Finally, SSR instigates broader reform to improve democratic governance of the security sector. Compared to the initiatives of other major actors in the

2 S4810007 O.P. Korthals Altes

Central Asian region – in particular Russia, China and the United States, that all mainly focus on ‘hard dimensions’ of security, including the transfer of equipment and joint military training and exercises – EU and OSCE programmes tend to incorporate these softer elements or developmental dimensions of security, for instance general support for the rule of law and civilian institutes controlling and monitoring security services (Albrecht, Stepputat & Andersen 2010, p. 80; Boonstra, Marat & Axynova 2013, p. 6).

SSR is considered to be the dominant paradigm of external security assistance worldwide. Many regional and international organisations and institutions – for instance, the United Nations (UN), the European Union (EU), the Organisation for Security and Cooperation in Europe (OSCE), and the Development Assistance Committee of the Organisation for Economic Cooperation and Development (DAC OECD), later followed by the African Union (AU) and the Economic Community of West African States (ECOWAS) – have adopted, used and mainstreamed SSR in their security assistance and development programmes. It is widely recognised that if security assistance is not directly linked to human security, (liberal) peacebuilding, development, and sustainability, it will most likely fail in the longer term (Edmunds 2004; Ball 2007; EU 2005; OECD DAC 2005, 2007; UN 2008). While external actors should limit themselves to a facilitator’s role, providing financial support, sharing knowledge and organising activities, SSR is mainly an internal process inherently linked to the principle of ownership. The sustainability of SSR depends on the willingness of major stakeholders and the active engagement of other (non-)state actors such as the judiciary, criminal investigation and prosecution services, parliament and civil society. Especially civil society actors – consisting of among others community representatives, research institutes, human rights groups and activists, as well as local offices of international organisations – are seen as essential actors in ensuring local ownership of SSR processes, due to their expertise, legitimacy and advocacy of local voices and needs (Caparini 2004, 2010; Gordon 2014; Mobekk 2010; Van Leeuwen & Verkoren 2015). As a result of its comprehensive approach, SSR impacts are beyond the security forces and its management, touching upon building relationships between people and the state.

Although in policy papers and academic articles the notion of SSR is broadly formulated, pointing out the (alleged) comprehensive links between security, development, the rule of law and democracy, the impact of SSR programmes in Central Asia (and in a number of other regions and countries as well) remains rather limited. This discrepancy between the advocated policies on the one hand, and the actual outcomes on the other is referred to as the so-called ‘policy-practice’ or ‘concept-context’ gap. In much of the literature and policy documents dealing with SSR from an outsider’s perspective, “there is a tendency to approach particular settings of SSR assistance with a set of predetermined ideas

3 S4810007 O.P. Korthals Altes

about what the security situation is and what needs to be fixed” (Ball & Hendrickson 2009, p. 39). Top-down approaches that mainly involve national stakeholders often overrule or ignore other voices, concerns, and needs, and might exclude local ownership during phases of policymaking, implementation, oversight, and evaluation of SSR programmes. Even though democratic control of the security sector brings stability and prosperity in the long run, one must be careful with being too optimistic about short-term prospects for change. In many cases, more lasting solutions for (re)building state and security structures emerge as (unintended) consequences of social processes of negotiation, contestation, and adaptation, rather than as a causal or direct outcome of clear inputs as formulated in policy briefs and strategy documents (Egnell & Haldén 2009; Andersen 2011; Jackson 2011). While institutional and legal frameworks for SSR can be created according to time schedules, changing the culture and behavioural attitudes towards security and people-state relations cannot be perfectly steered or managed given its highly political nature.

This brings me to the focus of my thesis research, namely reforming the police forces within the dimension of democratic governance in Kazakhstan and Kyrgyzstan. According to the principles of democratic policing, law enforcement agencies must perform their primary tasks, that include maintaining public order, preventing and combating crime, arresting and detaining criminals and suspects, and providing assistance and services to the public, while upholding the individual’s freedoms and rights and be being accountable and transparent (OSCE 2008b, p. 9). It is also agreed among (international) policymakers and experts that community-based policing could foster better relationships between police forces and the local people, so as to better address their concerns and cooperate in reducing crime and other public issues.

However, the police institutions in Kazakhstan and Kyrgyzstan remain reluctant to any change, frequently accused of human rights abuses, unlawful killing of protestors or arresting and detaining suspects, and regarded as heavily corrupt and associated with organised crime (Lewis 2011a, p. 9). Apart from deficient funding and operational capacities, a lack of democratic governance results into police forces that are not held accountable for their actions, and often used as government’s instruments to unlawfully monitor, harass and arrest political opponents or targeted groups (International Crisis Group 2002, pp. i-ii). Critics, policymakers as well as researchers, therefore question whether SSR programmes and donor policies directed at police reforms in Kazakhstan and Kyrgyzstan tend to strengthen the authoritarian regime’s oppressive capacities, rather than opening up chances for gradual, democratic reforms of security services. Compared with other SSR dimensions that include effectiveness and efficiency (Stojanovic 2009, p. 91), democratic governance requires deep, structural changes of policing culture and behaviour that should act on behalf of people’s

4 S4810007 O.P. Korthals Altes

needs instead of regime security. Such structural changes mostly take place over a longer period of continuous contention and friction, while quick fixes may undermine the diversity of opinions and perspectives that could otherwise have offered alternative or more tailored solutions (Marat 2018).

Therefore, this research analyses the policy formulation and progress (or ‘practice’) of democratic governance of the police forces in Kazakhstan and Kyrgyzstan over the past two decades. Within the region of Central Asia, these two countries are the most deeply engaged in cooperative projects and activities with external actors on reforming their police forces, or, at least, on familiarizing them with principles of democratic policing. Moreover, through my study exchange to the Al-Farabi Kazakh National University in Almaty, Kazakhstan, and research fellowship at the OSCE Academy in Bishkek, Kyrgyzstan, during the period of September 2017 until March 2018, I was able to consult and exchange perspectives with local academia and experts on these two case studies. Civil society actors that are (more or less) engaged in police reforms will be identified and included in this analysis as well. While I analyse the Kazakh and Kyrgyz contexts, policies and practices of police reforms separately, I provide space within my thesis to compare the different outcomes of police reforms and the salient factor of civil society. I will argue that genuine and strategic approaches towards comprehensive police reforms from above, as well as a vibrant and well-organised civil society are essential factors for change.

1.2 Research objective and research questions

The main research objective of this master thesis is to analyse the policy-practice gap regarding police reforms in Kazakhstan and Kyrgyzstan, based on an in-depth analysis of the dimension of democratic governance and the involvement of civil society actors. The central research question of my master thesis is:

To what extent is there a policy-practice gap of democratic governance of the police forces in Kazakhstan and Kyrgyzstan, and what role could the civil society play in bridging this gap?

In order to be able to answer this central question, some sub-questions have been formulated:

- What is the essence of the (theoretical) notions of Security Sector Reform, democratic governance, police reforms and civil society?

5 S4810007 O.P. Korthals Altes

- What conceptual framework can be designed for the analysis of the progress of democratic governance of the police forces and the involvement of civil society? - In what context are Kazakhstan and Kyrgyzstan formulating and implementing their

policies of reform of their respective police forces?

- What policies are being formulated by the governments of Kazakhstan and Kyrgyzstan as well as external facilitators regarding reforms of its police forces?

- To what extent is progress being made of democratic governance of the police forces in Kazakhstan and Kyrgyzstan?

- How crucial is the role of civil society in democratising the police forces?

1.3 Scientific and societal relevance

1.3.1 Scientific relevance

This master thesis research, focusing on democratic governance of the police forces and emphasising the role of civil society, might contribute new and interesting perspectives to scientific debates on SSR for a couple of reasons. First, while the literature on SSR has a normative emphasis, the empirical base remains rather limited. There are only a few systematic research publications available, where processes of SSR are embedded within a clear framework, with benchmarks or indicators to (qualitatively) measure the progress of SSR over time. However, such universal indicators or benchmarks have not been created yet. As a result, there is a lack of knowledge regarding whether or not SSR programmes have a long-term impact on peace, stability, and democratisation within recipient countries, including the question of whether or not international interventions have made a positive impact on (post-conflict) reconstruction processes. Policymakers might still overestimate the transformative abilities of external actors acting in very different socio-political structures (Etzioni, in Jackson 2011, p. 1813). There are a number of all-encompassing, extensive SSR field research projects going on, but it can take years before the necessary data for all indicators, actors, and contextual factors have been collected and processed. On the other side, there are many case studies of SSR, but they mostly lack a systematic approach. Overall, there have only been very few analyses looking beyond the immediate evaluation of existing security sector or police reform programmes.

Second, the context-specific cultural and practical dimensions in particular settings where SSR programmes have been or are implemented, have made it difficult to harmonise international policies in the security domain or to synchronise these with the local needs and priorities. There are many different understandings and priorities of SSR by international organisations and local actors, especially since SSR is implemented in a broad range of

6 S4810007 O.P. Korthals Altes

contexts, spread over various continents, ranging from states in transition to failed states. Just defining or assessing SSR policies on the basis of its performance in the most complex and conflict-ridden contexts provides an incomplete representation (Sedra 2010, p. 17). Conflict-ridden societies – for instance Afghanistan or the ones spread over the African continent, the region with the largest concentration of fragile, failed and post-conflict states with complex hybrid systems – get much attention of SSR policy analyses and discussions. On the other hand, Central Asian states are characterised by a strong, unitary, post-Soviet, and authoritarian state apparatus, which brings about other challenges in terms of transitions to democratic governance. It is wise to explore the particular socio-political setting of Kazakhstan and Kyrgyzstan, as the successful transformations of security sectors in other former Soviet countries, for example the Baltic and Central and East European (CEE) states, can give false or unrealistic expectations of SSR policies and police reforms in Central Asia. State, society and polity are important contextual factors to be taken into account in planning and executing SSR activities.

Third, studies regarding SSR programmes that integrate the perspectives from both academic researchers and local advocacy groups are lacking, which slows down the efforts to collect detailed research findings that bridge the gap between theory and practice, so as to better inform and influence (international) policy debates. In other words, there are significant discrepancies between the objectives of SSR as articulated by international organisations and foreign donors on the one hand, and the needs or perspectives of receiving countries on the other. From the perspective of civil society, there is some specific data available on the progress of democratising the police forces in Kazakhstan and Kyrgyzstan; for example, on the transparency of police reports or citizens’ surveys on local policing. However, such data hasn’t been marshalled within the conceptual framework of SSR, or democratic governance more specifically, yet, and subsequently connected to broader scientific debates on SSR.

Therefore, a specific SSR research focus on democratic governance, police forces, and civil society allows to go more in-depth and collecting more detailed information from local practitioners and experts, than it would be possible in extensive research with a broader focus on SSR. Using selected indicators makes it possible to measure actual progress over a longer period of time, all against the backdrop of the particular set of these two Central Asian case studies.

1.3.2 Societal relevance

Perspectives and data from this research could also be applicable to discussions within the spheres of policymaking and society. First, no noticeable progress in police reforms will be made if the political sensitivity around democratic policing isn’t actively being addressed.

7 S4810007 O.P. Korthals Altes

Law enforcement agencies that aren’t accountable and controllable, continue to be used as repressive instruments by government officials or commit illegitimate and unlawful acts with impunity, which might have serious societal consequences for Kazakhstan and Kyrgyzstan. As public trust of the police forces will not improve, local people will be more reluctant to cooperate with the police to resolve main security issues together. Corrupt police officers will keep accepting bribes or perform extortion practices, which in turn severely damages a secure and integer environment, suitable for economic development and business.

Second, while concepts of SSR and democratic governance are adopted in key documents of international and regional organisations and have some successes in its applications in other regions and countries, these concepts are hardly known in Central Asia, where security is understood in protecting the regime interests rather than the people’s interests. There is hardly any independent research or analytical centre in Kazakhstan or Kyrgyzstan, specialised in the field of SSR, and, consequently, there are practically no institutions capable of providing decision makers with expertise on security policies and police reforms (Musabayeva 2013, p. 86). Although some individual experts are involved in specific issues regarding police reforms and civil society, with access to the necessary information, there are no local, institutional frameworks to bring forward significant problems in SSR or police reforms to the majority of citizens. Such institutional frameworks could be local media, or professional groups like associations of journalists or academic communities, to carry out reporting, investigations or research on SSR.

Moreover, due to its historic, geographic position in bringing Europe and Asia together, Central Asia has sparked interest from neighbouring and international countries in cooperating within many policy areas that include security, economy and sustainable development. Secure environments, ensured by well-functioning and democratically governed police forces, are necessary to further improve these cooperation agreements. Political and economic developments in Kazakhstan and Kyrgyzstan, as well as transregional challenges, directly or indirectly impact interests of other countries and regions; see, for example, the dependency on the energy market or international terrorist threats. In addition to Chinese investments, Russian security cooperation and (recently decreasing) American geostrategic interests, the OSCE and EU are actors promoting good governance, rule of law, democratisation and human rights, including Security Sector Reform. Although Central Asian governments are initially averse to deep reforms, states like Kazakhstan and Kyrgyzstan have expressed interests in working together with the EU and the OSCE in areas such as development, education, and pressing security threats. It is clear that a long-term presence of the EU and OSCE in the region provides some opportunity to bring gradual changes in the security sector.

8 S4810007 O.P. Korthals Altes

Therefore, research on SSR and on democratising police reforms in particular, contributes to a higher amount of established (academic) literature for the research community, and offers useful perspectives for policymakers to tackle societal issues, necessary to build up and maintain a well-functioning society that is also attractive to foreign countries to invest or work together with. Moreover, an in-depth analysis of the course of police reforms in Kazakhstan and Kyrgyzstan brings up more realistic perspectives on the impact of SSR programmes in the foreseeable future.

1.4 Research methodology and data collection

SSR is a complicated concept, encompassing a wide range of actors, policy areas, and dimensions, being implemented in many different socio-political contexts. Therefore, it has not been easy to delimit this research into a suitable master thesis subject and develop an appropriate conceptual framework, in order to formulate what objectives need to be analysed, which actors to select, what the ‘practice’ of SSR policies is, which indicators to select to measure progress, and, not the least important, where and how to retrieve the data to provide such an in-depth analysis. Clearly, I mostly rely on various qualitative research methods, including a review of the relevant academic and policy-oriented literature, documents, and reports, (semi-structured) interviews and surveys (through Skype or Google Forms), and consulting online available databases related to SSR. In line with the phrase ‘less is more’, I use a magnifying glass on police reforms within SSR, with civil society as the important factor. With respect to police reforms, I have decided not to analyse whether or not law enforcement agencies enforce their assigned tasks or are well equipped, trained or funded, but rather how law enforcement agencies are democratically governed, held responsible and are responsive to local people’s needs. Given this focus on police reforms, it is also relevant to sketch the broader political and regional contexts, how they affect police reforms.

The secondary sources include academic studies on SSR and police reforms, official policy and legal documents of the governments of Kazakhstan and Kyrgyzstan, as well as media and online reports and policy briefs by international nongovernmental organisations (NGOs) and civil society organisations (CSOs). A literature review gives insight into the theory and practice of SSR, but also in how SSR relates to theories of state building, and therefore the consequences for planning and implementing SSR programmes. From a number of assessment frameworks – providing a broad range of questions regarding political, societal and economic issues –I have selected the questions to develop and adjust the conceptual framework for the contextual analysis of Kazakhstan and Kyrgyzstan (Section 2.5). A concise analysis of the historical development of state structures and society of Kazakhstan and

9 S4810007 O.P. Korthals Altes

Kyrgyzstan gives insight in how state and security structures of these two states actually function. Moreover, over the years, from 2003 on, the OSCE Annual Report of the Secretary

General on Police-Related Activities provides a good overview of the programmes and

activities related to police reforms in Kazakhstan and Kyrgyzstan. However, there are less detailed sources or systematic data available on the actual effects of police reforms on the ground, or on which civil society actors are exactly involved in these processes. Personal communication with experts from local institutes – including KIMEP University and the Legal Policy Research Centre (LPRC) in Kazakhstan, and the OSCE Academy and the International University of Central Asia in Kyrgyzstan – helped in identifying civil society actors engaged in police reforms. These civil society actors include academics, NGO workers, human rights advocates and media experts.

To measure the progress of democratic governance of the police forces, I had to rely on qualitative data, derived from reports, interviews, and surveys with local civil society actors. Unlike other areas –such as human rights, democratisation or economic development – there are no general measuring methods established for SSR or police reforms. Based on personal communication with local experts and the main principles of democratic policing, as also put forward in the OSCE Guidebook on Democratic Policing (2008b), I have selected six indicators to measure the democratic governance of police forces in Kazakhstan and Kyrgyzstan, respectively:

#1 Public oversight and monitoring, concerning the oversight institutions and mechanisms of citizens and civil society to monitor the police and to cooperate with them;

#2 Integrity, or battling corruption within law enforcement;

#3 Participation of citizens and civil society in policymaking, implementing and evaluating police reform programmes (generally described as ‘local ownership’); #4 Transparency, or the degree of access to police reports and statistics;

#5 Upholding the rule of law and human rights during policing, but also the freedom of press to report illicit police performances; and

#6 Responsiveness and trust from local people.

There are a few reports and policy briefs from international organisations or civil society actors available to fill in the data for the respective indicators. Moreover, although not applied to police forces specifically, online databases such as the Transparency Corruption Perceptions Index or Voice and Accountability of Worldwide Governance Indicators (unfortunately, the Civil Society Index database could not be accessed) give some useful

10 S4810007 O.P. Korthals Altes

information on more general, longer-term developments of democratic governance in Kazakhstan and Kyrgyzstan. Yet, such indicators measuring general patterns of democratic governance need to be complemented with other, more specific data to provide a better overview of the progress of police reforms. In Section 2.5, the selected questions and theoretical discussions regarding the indexes for SSR in the conceptual framework are addressed. For more information on the assessment frameworks consulted for this thesis, see the appendixes.

Based on this research methodology, I hope to provide more specific data on democratic governance of police forces. On the other hand, I have not included other actors such as the judiciary, other security actors or political groups. Moreover, neither have I included other main issues with police reforms, such as their effectiveness in reducing crime, anti-drug smuggling activities or arresting suspects.

1.5 Structure

After discussing the thesis subject, the research questions, the scientific and societal relevance and methodology, the next step is of course an extensive discussion of the relevant academic and policy-related literature. This issue is addressed in Chapter 2, which answers the first two sub-questions. The rise of the (comprehensive) SSR concept from the late 1990s within international development circles is discussed, as well as the essence of the dimension of democratic governance within SSR, followed up by the concepts of police reforms and civil society. I will also illustrate theoretic debates on overcoming the policy-practice gap and the essence of linking SSR to state building. At the end of this chapter, I present the conceptual framework, refined for this research.

Chapter 3 sketches the regional, political, and security context of Kazakhstan and Kyrgyzstan. Civil society actors involved in police reforms will be identified in particular as well as the legal and political framework in which they operate.

Chapter 4 analyses key documents of Kazakh and Kyrgyz policies towards police reforms and civil society, as no kind of reform can be initiated without the consent of local authorities. This chapter also includes, based on the OSCE Annual Report of the Secretary

General on Police-Related Activities (2003 through 2016), an overview of police reform

programmes and related activities and events organised for government officials, police officers and civil society actors.

Chapter 5 discusses the actual progress of SSR, using the six indicators within the dimension of democratic governance. I will compare the results of these indicators as well explain the results in Kazakhstan and Kyrgyzstan, based on the main factor of civil society.

11 S4810007 O.P. Korthals Altes

Finally, Chapter 6 will provide the answers to the research questions and make recommendations for both policymakers and academics. I will also reflect on the process of my master thesis, highlighting the limitations and points of improvement.

12 S4810007 O.P. Korthals Altes

2. Literature review and conceptual framework

2.1 The concepts of Security Sector Reform, democratic governance, police

reforms and civil society

In analysing theories of Security Sector Reform, one has to start with addressing the main (contested) concepts. In the context of this research, these are ‘Security Sector Reform’ as such, and ‘democratic governance’, ‘police reforms’ and ‘civil society’, respectively.

2.1.1 Security Sector Reform

The concept of Security Sector Reform came prominently on the international policymaking agenda after the Cold War, during the 1990s, and is linked to debates on poverty reduction, human security, sustainable development, good governance, and democratisation. The UK Department for International Development (DFID 1998) was the first to develop a holistic understanding of SSR, aimed at providing and protecting a safe and secure environment for the local poor. This constituted a significant break from the Cold War era security assistance programmes by the major powers in both East and West, that rather aimed to protect their own ideological interests through regime security of allied and friendly countries. The end of the bipolar division of the world paved the way for development and security assistance agendas of the OECD and other international organisations. In addition, successful examples of reforming and democratising the military and other security forces in Central and Eastern Europe (CEE), Latin America and Southern Africa, further stimulated ways of rethinking security assistance (Ball 2007, p. 87; Ball & Hendrickson 2009, pp. 9-12; Hendrickson & Karkoszka 2009, pp. 21-22; Sedra 2010, p. 16).

This new conceptualisation of security assistance implied that the effectiveness and professionalism of the security sector should not only be defined on the basis of its capacities (defence, police, border management, intelligence services, etc.), but also on how the security sector is controlled, managed, and held accountable by a well-functioning system of checks and balances. The various SSR concepts as defined by research institutions, donors, and international organisations during the late 1990s and early 2000s, focusing on establishing appropriate structures and mechanisms for democratic control over the security sector, are generally referred to as being the ‘first generation’.

Despite a broad consensus among major international donor organisations, albeit essential for building the foundations of more coherent and integrated approaches to support countries in reforming their security sector, there is yet no universal definition of Security Sector Reform. The OECD Development Assistance Committee (2005, p. 20) and the EU (2005, p. 9) describe SSR as “the transformation of the ‘security system’, which includes all

13 S4810007 O.P. Korthals Altes

the actors, their roles, responsibilities and actions, working together to manage and operate the system in a manner that is more consistent with democratic norms and sound principles of good governance, and thus contributes to a well-functioning security framework”. The UN Secretary-General (2008, p. 6), in a comprehensive report on international security, defines SSR as “a process of assessment, review and implementation as well as monitoring and evaluation led by national authorities that has as its goal the enhancement of effective and accountable security for the state and its peoples without discrimination and with full respect for human rights and the rule of law”. In addition, in its partnership programmes with friendly countries, NATO (2005, p. 24) focusses on (re)building effective and efficient state defence institutions under democratic control to meet domestic needs and international commitments, while the OSCE (2016c, pp. 13-14) strives to enhance security sector governance and reform through a cross-dimensional approach, by identifying and supporting links between the political and military, the economic and environmental, as well as the human dimensions.

Moreover, these organisations also formulate the main goals of SSR. The OECD DAC (2007, p. 21) names establishing effective governance, oversight and accountability in the security system, improving and sustaining the delivery of security and justice services, and developing local leadership and ownership of the reform process. The UN (2008, p. 13) advocates the development of effective, inclusive and accountable security institutions so as to contribute to international peace and security, sustainable development and the enjoyment of human rights by all. Finally, Ball (2007, pp. 96-97) calls for strengthening the professionalism of security services, fostering capable and responsible civil authorities and civil society, according a high priority to the rule of law, and developing regional approaches to security problems.

In spite of these differences, definitions of SSR contain at least two key normative elements. First, democratisation and civilian control are considered essential in any process of SSR. Second, the importance of effectiveness and efficiency in SSR are emphasised in all conceptualisations. While related, these two elements of SSR can have their own particular demands, rewards and problems (Edmunds 2004, pp. 65-66). It is also important to realise that SSR activities can focus on just but also on multiple institutions (in this research, the police forces) or policy areas (for example, counterterrorism or border management). Nevertheless, such focused activities shouldn’t neglect the overall SSR framework and must be conducted as part of a wider cooperation between different stakeholders and donors.

The next important question is what groups are exactly the targets of processes of reforming the security sector. Clearly, not only forces delivering direct security such as the police, the armed forces, and intelligence services are involved, but also oversight, judiciary

14 S4810007 O.P. Korthals Altes

and civil society actors should be engaged in SSR programmes. OECD DAC (2005, pp. 20-21) distinguishes a middle range of four groups of actors involved in Security Sector Reform:

(1) Core security actors: armed forces; police; gendarmeries; paramilitary forces; presidential guards, intelligence and security services (both military and civilian); coast guards; border guards; customs authorities; reserve or local security units (civil defence forces, national guards, militias).

(2) Security management and oversight bodies: the Executive; national security advisory bodies; legislature and legislative select committees; ministries of Defence, Internal Affairs, Foreign Affairs; customary and traditional authorities; financial management bodies (Finance ministries, budget offices, financial audit and planning units); and civil society organisations (civilian review boards and public complaints commissions).

(3) Justice and law enforcement institutions: judiciary; Justice ministries; prisons; criminal investigation and prosecution services; human rights commissions and ombudsmen; customary and traditional justice systems.

(4) Non-statutory security forces, with whom donors rarely engage: liberation armies; guerrilla groups; private bodyguard units; private security companies; political party militias.

Some scholars offer a rather minimalist approach, that only involves the armed forces, intelligence services, police, judiciary, paramilitary forces, border and customs officials (Fitz-Gerald, Yusufi & Greene, in Hernandez & Reyes 2011, p. 6), while others (Ball, Bouta & Van de Goor 2003, p. 32; Ball 2007, p. 91) argue, in line with the above distinction, that there are five major categories of local actors that influence the quality of democratic Security Sector Governance: (1) bodies mandated to use force; (2) justice and public security bodies; (3) civil management and oversight bodies; (4) non-state bodies; and (5) non-statutory civil society bodies. The first three groups constitute what is commonly called the security sector, while the last two categories refer to civil authorities that are required to play an active role in managing and monitoring this security sector. Furthermore, external assistance meant to strengthen democratic governance of the security sector will be more effective if external actors work coherently towards a common framework (Ball 2007; EU 2005; Hendrickson & Karkoszka 2007; OECD DAC 2005; UN 2008). This, therefore, requires building partnerships across government departments and foreign agencies. But above all, most experts stress the need for and importance of ownership of local actors; external actors and donors should be involved in facilitating the SSR process by delivering guidance and resources.

15 S4810007 O.P. Korthals Altes

It is also acknowledged that SSR programmes have by no means a ‘one size fits all’ model, and therefore have to be carefully implemented, depending on the specific contexts of recipient countries. The main criteria used for assessing the context for implementing SSR concern the type of political system (ranging from democracy, through a system in transition, to autocracy); the level of (socio)economic development; and the security situation. For example, Wulf (2011, p. 346) provides a table of various contexts, including categories of ‘developing’, ‘post-authoritarian’ or ‘transitional’, and ‘post-conflict’, each with its own specific issues and key priorities. Countries characterised by underdevelopment and weak governance more often than not have excessive military spending, poor management, and a general waste or lack of resources, while authoritarian states are characterised by democratic deficits, legacies of violation of human rights, and weak civil societies. Consequently, SSR programmes in developing countries should focus more on economic development in order to secure stable conditions for reform. It should be noted that post-conflict states suffer from legacies of violence, a level of mistrust among the major parties, and a perceived risk of renewed conflicts.

While most SSR scholars acknowledge the importance of context in designing individual country programmes, in reality, donors tend to employ the same kind of tools, strategies, and ideas, often implemented by the same officials or contractors, moving from a previous setting to the next (Sedra 2010, p. 103). This lack of innovation and contextualisation in donor or external assistance approaches can partly be blamed for the problems the implementation of SSR confronts. Yet, one must be also careful in categorising recipient countries in too simplistic overviews. According to Ball (2004), countries can’t always be simply categorised along the lines of traditional groups, since every state has its own historical legacy of conflict and share of obstacles to reform. Therefore, she states that a contextual approach should focus on additional criteria, including political, psychological, normative, economic, institutional, societal and geopolitical dimensions. By understanding the political relationships between key actors and looking beyond formal legislation and organisational structures, the incentives for (changes of) SSR can be addressed. Hereby, Luckham (2000, pp. 17-18) argues that instead of one unified and coherent body, security sectors should rather be seen as a terrain of shifting security coalitions or factions, each with its own organisational culture and competing interests.

Finally, fundamental principles have been set up regarding how SSR goals should be reached. In its handbook, Security System Reform and Governance (2005, pp. 22-24), OECD DAC has formulated some fundamental principles for SSR, including policy coherence by closer cooperation between donor and partner countries, and the so-called whole-of-government or holistic approach. In general, processes of SSR should be people-centred,

16 S4810007 O.P. Korthals Altes

locally owned and based on democratic norms and internationally accepted ideas of human rights principles and the rule of law. Consequently, the three core values of a well-functioning security system are: (1) a nationally owned concept of security; while establishing (2) well-defined policies and strengthening governance of the security institutions; and (3) building institutional mechanisms for implementation and capacity throughout the entire security system. In accordance with this line of thinking, the UN (2008, pp. 11-12) formulates some important lessons learned from previous SSR experiences, dealing with how the transformation of the security sector is linked to national goals and relationships between various groups and institutions, while acknowledging that it is inherently a political process, based on realistic considerations of what is financially, operationally and logistically viable. Therefore, SSR should not just be a mere technical matter of making security forces more efficient, but it should be regarded as an integral part of making a society secure as well as democratically accountable.

2.1.2 Democratic governance of the security sector

In policy documents, numerous overlapping definitions and descriptions of ‘Security Sector Reform’, ‘Security Sector Governance’, ‘civilian control of the security sector’ and ‘democratic governance of the security sector’ are used. While states have traditionally governed their security sectors in many different ways, SSR processes have always been linked to the concept of governance, which “denotes the overall setting, application, and enforcement of rules that guide the exercise of political authority. It is the process by which political decisions are made and implemented” (Schroeder 2010, p. 10). Governance of the security sector thus specifies the ways of governing the state’s monopoly on the legitimate use of force from the perspective of both state and non-state actors. According to Schroeder (2010, p. 11), good governance refers to democratic norms of accountability, transparent decision-making processes and a security apparatus that is completely subordinated to the control of civil authorities. For political reasons, the term Security Sector Governance (SSG) is sometimes used instead of SSR. This is mainly the case in societies where the leadership remains highly sensitive to the concept of ‘reform’. Governance refers to an entire system of structures, actors, processes, and functions by which a particular unit, such as a government agency or the state itself, is managed. ‘Reform’, on the other hand, refers to the process of changing an existing reality (Hernandez & Reyes 2011, p. 13). Generally, democratic governance could be seen as one of the three main dimensions inherent to SSR, next to efficiency (how well is the security sector financed and managed?) and effectiveness (how well are the security actors performing in their allocated functions such as reducing crime, bringing about a secure environment, decrease smuggling, etc.?).

17 S4810007 O.P. Korthals Altes

The emerging concept of democratic governance of the security sector can be linked to the post-Cold War paradigm of international security and global governance, encompassing international cooperation, democracy, human rights, and development. Yet, the 1993 UN Vienna Declaration, or any other similar document calling for the consolidation of democracy and human rights, made little or no specific references to the democratic governance of the security sector in particular (Hänggi 2003, pp. 12-13). Standards on democratic governance of the security sector were later adopted at conferences such as the 2000 Warsaw Conference of the Community of Democracies or the Club of Madrid (2001). The Geneva Centre for the Democratic Governance of Armed Forces (DCAF), the leading international think tank specialised in SSR/SSG, has supported policy development and guidance for regional and international organisations through a mapping study of its approaches to SSR since 2000. It has sustained and expanded its cooperation on research, policies, and practices of SSR with regional and other multilateral organisations, such as the EU, NATO, OSCE, OECD, but also the AU and ECOWAS in Africa. DCAF also conducted research on the UN’s approach to defence sector reform and finalised an integrated technical guidance note on monitoring and evaluation of SSR support. As a result, the adoption of the first stand-alone resolution on SSR by the UN was embedded in Security Council Resolution 2151 (2014), marking a watershed moment in international approaches to SSR (DCAF 2016, p. 34). It will also explore potential partnerships with the Organisation of American States (OAS) and the Inter-American Development Bank (IDB) in Latin America and the Caribbean, and with the Association of Southeast Asian Nations (ASEAN) and the Asian Institute of Poverty Alleviation (AIPA) in Southeast Asia. It has become clear from these trends, that SSR and democratic governance are getting a central place in policies and practices of more multilateral actors and regional organisations around the world.

There are good explanations for the prevalence of democratic governance in policies above other forms of security. In fact, it is acknowledged that democratic governance is crucial for the success of democratic consolidation and sustainable economic and social development. When it comes to the security sector, good governance includes the effective, efficient, participatory, accountable and transparent functioning of state institutions that have a monopoly on the legitimate use of force. If people and states are not secure from fear of violence at a local, national or international level, development will not be sustainable (Ball, Bouta & Van de Goor 2003, p. 30; Born, Caparini & Fluri 2000, p. 8; DFID 1998, p. 1). According to Ball (2007, pp. 85-86), lacking democratic governance of the security sector not only leads to tolerance of politicised security forces, but also to war as a tool of resolving disputes, to disrespect for the rule of law by security forces and political elites alike, to serious human rights violations, to budget allocations skewed towards the security forces, or to a

18 S4810007 O.P. Korthals Altes

diminished capacity of security forces to carry out their constitution-mandated tasks of protecting the people and communities.

Finally, experts prefer the term ‘democratic governance’ over ‘civilian control’ of the security sector. The notion of democratic control makes more sense, as some rightly point out that civilian governments are not always democratic. Luckham (2000, p. 17) warns against stereotyping military and security bureaucracies as sole obstacles to Security Sector Reform, that “must be controlled by civilian structures”. Sometimes civilian administrations form an obstacle to democratic governance of the security sector when their professional interests lie more in partisan politics or corruption. A focus on civilian control might significantly impede understanding the larger and more complex relationships concerning democracy and security forces. Democratic control of armed forces depends on the broader societal quality of democracy itself, that is shaped by institutions, national politics, and civil society. It also must give way to creating mechanisms and systems that enable voices from diverse actors to be heard to inform the process (Gordon 2014, p. 6). Scholars like Gordon acknowledge that the complexity of democratic governance of the security sector should be recognised, and that ignoring alternative voices or compromising sustainability in favour of quick results might be short-sighted, might pave the way for future failures or neglect dimensions that could otherwise lead to success.

2.1.3 Police reforms

From all security institutions, the police are often the most visible and have the most impact on people’s daily lives. They are responsible for maintaining public order and are the guardians of the criminal justice system. According to the OECD DAC (2007, p. 163), “the definition of ‘police’ applies to all law enforcement agencies, such as the police and gendarmerie like constabulary forces, that exercise police powers, especially the powers of arrest and detention”. Where over the past 50 years traditional security literature has focused on civil-military relations (CMR) and civil control of the armed forces in the context of national defence, the recent emergence of SSR in academic and policy circles gave more attention to a broader range of security actors, such as intelligence agencies, border guards and, in this case, police forces (Bruneau & Matei 2008; Lewis 2011a). The growing literature on SSR argues that a neutral and effective police force, under democratic oversight, is an essential element in processes of peacebuilding, democratisation, and state-building.

In one of the first extensive literature reviews on experiences of foreign assistance to police reform and human rights organisations’ reports on police abuses, Bayley (2000, p. 75) puts forward some important lessons learned about the process of democratising the police. He argues that democratising the police abroad is a four-step process, namely (1) design of