BUILDING STABLE GOVERNMENTS IN POST-ETHNIC CIVIL WAR SOCIETIES: THE IMPORTANCE OF COMMUNITY POLICING by SHAWNA MEECHAN A DISSERTATION Presented to the Department of Political Science and the Graduate School of the University of Oregon in partial fulfillment of the requirements for the degree of Doctor of Philosophy September 2018 ii DISSERTATION APPROVAL PAGE Student: Shawna Meechan Title: Building Stable Governments in Post-Ethnic Civil War Societies: The Importance of Community Policing This dissertation has been accepted and approved in partial fulfillment of the requirements for the Doctor of Philosophy degree in the Department of Political Science by: Dr. Craig Parsons Chairperson Dr. Craig Kauffman Core Member Dr. Karrie Koesel Core Member Dr. Shaul Cohen Institutional Representative and Janet Woodruff-Borden Vice Provost and Dean of the Graduate School Original approval signatures are on file with the University of Oregon Graduate School. Degree awarded September 2018 iii © 2018 Shawna Meechan iv DISSERTATION ABSTRACT Shawna Meechan Doctor of Philosophy Department of Political Science September 2018 Title: Building Stable Governments in Post-Ethnic Civil War Societies: The Importance of Community Policing States recovering from violent ethnic conflict face many challenges in trying to rebuild. Primary among these challenges is how to best provide for the security of the people. In states that choose democratic regimes, security must be provided by a civilian police force. One major challenge in the process of post-ethnic conflict rebuilding is addressing any conceptions within the community that the police are biased against marginalized groups in society. This is often particularly difficult in post-ethnic conflict states where the police have been associated, fairly or not, with one side of the conflict. In such situations, how do the police gain the confidence of the community so that policing is seen as legitimate and fair? This dissertation combines the literature on post-ethnic conflict rebuilding with the considerable literature on community policing to argue that implementing community policing may be the solution to this problem. Broadly speaking, community policing embraces the concept that the work of the police is not separate from the community but should be approached as a collaborative partnership between the police and the community they serve. This requires community input and active participation in policing at every stage from problem identification to idea formation and implementation to v evaluation and reform. This style of policing builds relationships between the police and the community and works to develop community confidence in the police. This argument is evaluated through one in-depth study and one initial assessment of states that have attempted to implement a form of community policing after ethnic conflict: Northern Ireland following their ‘Troubles’ and Kenya after their 2007 post- election violence respectively. Two specific conclusions arise from this study. First, the adaptive nature of community policing enables the police to build community confidence through policing practice that is responsive to the needs of the specific communities being served. Second, the symbolism surrounding the police has a strong effect on individuals’ abilities to acknowledge or accept police reform in post-ethnic conflict states. vi CURRICULUM VITAE NAME OF AUTHOR: Shawna Meechan GRADUATE AND UNDERGRADUATE SCHOOLS ATTENDED: University of Oregon, Eugene, Oregon Queen’s University Belfast, Belfast, Northern Ireland, United Kingdom New York University, New York, New York DEGREES AWARDED: Doctor of Philosophy, Political Science, 2018, University of Oregon Master of Science, Political Science, 2015, University of Oregon Master of Arts, Comparative Ethnic Conflict, 2007, Queen’s University Belfast Bachelor of Arts, 2006, New York University AREAS OF SPECIAL INTEREST: Comparative Ethnic Conflict Conflict Resolution Public Policy PROFESSIONAL EXPERIENCE: Graduate Teaching Fellow, University of Oregon Department of Political Science, Sept 2012 – June 2016, Sept 2017 – June 2018 GRANTS, AWARDS, AND HONORS: Graduate Fellowship for International Research, Oregon Sasakawa Young Leaders' Fellowship Fund (SYLFF), 2016 - 2017 General University Scholarship, University of Oregon, 2016 - 2017 Everett D. Monte Scholarship, University of Oregon College of Arts and Science, 2016-2017 W.C. Mitchell Graduate Summer Research Scholarship, University of Oregon Department of Political Science, 2016 Institutional Representative, Institute for Qualitative and Multi-Method Research, 2015 vii Neil David Blackman Memorial Scholarship, University of Oregon Department of Political Science, 2014-2015 Risa Palm Graduate Fellowship, University of Oregon College of Arts and Science, 2014-2015 viii ACKNOWLEDGMENTS I wish to express sincere appreciation to Professor Craig Parsons for being a dedicated and supportive mentor and adviser. I would also like to thank Professors Craig Kauffman, Karrie Koesel, and Shaul Cohen for reading dozens of drafts and always providing excellent feedback for this and many other projects. In addition, special thanks are due to the Police Service of Northern Ireland and the many officers and staff who offered their time and insights for this dissertation. I am sincerely grateful to Dr. Jonny Byrne of the University of Ulster whose insights were vital in helping me define the outcome of community confidence in the police. I am eternally grateful to my writing partner and dear friend, Zeina Salame, without whom I’m not sure I would have ever finished writing. I would like to thank the Institute for Qualitative and Multi-Method Research for honing my research skills. Portions of this research were supported by the W.C. Mitchell Graduate Summer Research Scholarship from the University of Oregon Department of Political Science. This dissertation was supported in large part by the Oregon Sasakawa Young Leaders' Fellowship Fund (SYLFF), a community to whom I am truly grateful and honored to be a part of. Finally, I am deeply grateful to my mom, dad, and brother for sticking with me through this arduous journey. Your love and support has never wavered, and I absolutely could not have done this without you. ix For every teacher who has ever inspired me, especially Leonard Lippi, Steve Workman, and Professor Sanford Gordon. x TABLE OF CONTENTS Chapter Page I. INTRODUCTION AND LITERATURE REVIEW ................................................ 1 Introduction ............................................................................................................ 1 Important Concepts ................................................................................................ 6 State Stability ................................................................................................... 6 Why Ethnic Violence? ..................................................................................... 7 Review of the Literature on Post-Ethnic Conflict Rebuilding ............................... 9 Community Reconciliation .............................................................................. 10 Ethnic Representation in Government ............................................................. 12 Security ............................................................................................................ 13 Democracy and Civilian Policing .......................................................................... 20 Community Policing .............................................................................................. 24 Defining Community Policing ......................................................................... 25 Theoretical Underpinnings............................................................................... 31 Pragmatism ................................................................................................ 32 Practice Theory .......................................................................................... 33 Symbolism ................................................................................................. 36 Expectations: What Community Policing Looks Like .................................... 39 Symbolism ................................................................................................. 39 Practice ....................................................................................................... 41 Community Policing Theory Simplified .......................................................... 43 Methods.................................................................................................................. 44 Process Tracing ................................................................................................ 46 Case Selection .................................................................................................. 47 Shadow Case Selection .................................................................................... 51 Data ........................................................................................................................ 55 Interviews ......................................................................................................... 55 Conclusion ............................................................................................................. 58 II. ETHNICITY AND POLICING IN NORTHERN IRELAND ................................ 61 Introduction ............................................................................................................ 61 Historical Setting ................................................................................................... 61 The Troubles .......................................................................................................... 65 Policing in Northern Ireland during the Troubles .................................................. 70 Peace Agreement ................................................................................................... 76 xi Chapter Page Results .................................................................................................................... 77 Conclusion ............................................................................................................. 82 III. IMPLEMENTING COMMUNITY POLICING: SYMBOLISM ......................... 84 Introduction ............................................................................................................ 84 The Patten Report .................................................................................................. 86 Naming the Problem .............................................................................................. 88 From Patten Report to Implementation .................................................................. 96 Making the Leap .................................................................................................... 100 Badges and Emblems ....................................................................................... 101 Representation.................................................................................................. 104 Physical Barriers .............................................................................................. 106 Community Reaction ............................................................................................. 108 Continuing Baby Steps .......................................................................................... 117 Conclusion ............................................................................................................. 122 IV. IMPLEMENTING COMMUNITY POLICING: PRACTICE .............................. 124 Introduction ........................................................................................................... 124 The Patten Report .................................................................................................. 125 Retraining the Police ........................................................................................ 126 Ongoing Community Consultation .................................................................. 129 Decentralizing Decision Making ..................................................................... 131 Implementation ...................................................................................................... 135 Results: Successes and Failures ............................................................................. 143 Tempering Expectations ........................................................................................ 152 Conclusion ............................................................................................................. 157 V. KENYA ................................................................................................................... 160 Introduction ............................................................................................................ 160 Historical Setting ................................................................................................... 161 Election Violence ............................................................................................. 169 Peace Agreement ............................................................................................. 173 Policing in Kenya ................................................................................................... 174 National Taskforce on Police Reforms .................................................................. 180 Restructuring and Decentralizing the Police .................................................... 184 Police Accountability and Transparency ......................................................... 187 xii Chapter Page Police Working Conditions and Training ........................................................ 189 Police-Community Relations ........................................................................... 191 Implementation ...................................................................................................... 192 Results .................................................................................................................... 196 Conclusion ............................................................................................................. 206 VI. CONCLUSION...................................................................................................... 209 Introduction ............................................................................................................ 209 Extensions and Practical Implications ................................................................... 211 Limitations ............................................................................................................. 219 Important Caveats .................................................................................................. 221 Corruption .............................................................................................................. 224 Future Research ..................................................................................................... 226 Conclusion ............................................................................................................. 228 REFERENCES CITED ................................................................................................ 230 xiii LIST OF FIGURES Figure Page 1. RUC emblem. ........................................................................................................ 90 2. Weapons Used Against the Police during the Troubles. ........................................ 92 3. An armed officer polices a Loyalist parade ........................................................... 93 4. PSNI Land Rover ................................................................................................... 94 5. Strand Road Station, Derry .................................................................................... 95 6. Donegall Pass Station, Belfast ............................................................................... 95 7. PSNI emblem submissions .................................................................................... 102 8. PSNI emblem as adopted ....................................................................................... 104 9. Coleraine Police Station ......................................................................................... 107 10. Musgrave Police Station ........................................................................................ 107 11. Police keep the peace at a Loyalist parade ............................................................. 108 12. Recommended Command Structure ...................................................................... 134 13. Peace Wall in Belfast ............................................................................................. 147 xiv LIST OF TABLES Table Page 1. Community Attitudes toward the RUC, 1998........................................................ 79 2. Community Attitudes on RUC Reform, 1998 ....................................................... 80 3. Community Confidence in the Police, longitudinal ............................................... 81 1 CHAPTER I INTRODUCTION AND LITERATURE REVIEW Introduction How do states recover from violent ethnic conflict? What role does security play in this process? In post-ethnic conflict democracies, how do civilian police forces gain the confidence of the community? These questions guide this dissertation project. The literature on post-conflict rebuilding acknowledges the importance of personal security but has a distinct lack of attention to domestic policing and how it is implemented. That is, this literature emphasizes the need for individuals in society to feel secure for the state to rebuild internal legitimacy and stability after violent ethnic conflict, but the discussion of how to provide such security tends to be limited to the make-up of the security: is it best provided by international peace-keepers? The military? Civilian police?1 This dissertation focuses on post-ethnic conflict democracies where the civilian police will ultimately be responsible for ensuring security. This need for civilian policing is justified in the literature, as discussed below, but the post-conflict literature tends to stop the conversation here. Once the decision has been made as to who will provide security, there is little discussion as to how. This leaves important questions unaddressed. Why do 1 Charles T. Call, "Competing Donor Approaches to Post-Conflict Police Reform," Conflict, Security & Development 2, no. 01 (2002); Charles T. Call and William D. Stanley, "Protecting the People: Public Security Choices after Civil Wars," Global Governance 7, no. 2 (2001); Paul Collier, "Ethnic Civil Wars: Securing the Post-Conflict Peace," Harvard International Review 28, no. 4 (Winter) (2007); Chaim Kaufmann, "Possible and Impossible Solutions to Ethnic Civil Wars," International Security 20, no. 4 (1996); Katherine Glassmyer and Nicholas Sambanis, "Rebel: Military Integration and Civil War Termination," Journal of Peace Research 45, no. 3 (2008); William D. Stanley and Charles T. Call, "Military and Police Reform after Civil War," Contemporary Peacemaking: Conflict, Peace Processes and Post-war Reconstruction (2008). 2 certain reforms succeed in building community confidence in the police when others fail? Why do certain reforms work in certain post-ethnic conflict cases and not in others? The literature on policing addresses these questions but has not previously been connected to the literature on ethnic conflict. There is general agreement in this literature that policing reform should be democratic and include a respect for fundamental human rights.2 David Bayley even argues that “the elements of democratic police reform are no longer problematic.” He elaborates, “Few people would disagree with accountability, protection of human rights, and transparency, and most would accept the importance of developing a service orientation.”3 However, “outlining the contours of what democratic policing is supposed to look like is one thing; translating these into practice on the ground in another matter entirely.”4 One subset of this literature suggests that one way to translate democratic policing into practice – building community confidence in the police, and therefore legitimizing the police and through them, the state – is through community policing.5 Community policing can be broadly understood as a philosophy of 2 David H Bayley, Democratizing the Police Abroad: What to Do and How to Do It (Washington, DC: U.S. Department of Justice, Office of Justice Programs, National Institute of Justice, 2001); Changing the Guard: Developing Democratic Police Abroad (Oxford, UK: Oxford University Press, 2006); Marina Caparini and Otwin Marenin, "Police Transformation in Central and Eastern Europe: The Challenge of Change," in Transforming Police in Central and Eastern Europe: Process and Progress, ed. Marina Caparini and Otwin Marenin (Geneva, Switzerland: Geneva Centre for the Democratic Control of Armed Forces, 2004); Philip C Stenning and Clifford D Shearing, "Reforming Police: Opportunities, Drivers, and Challenges," Australian and New Zealand Journal of Criminology 38, no. 2 (2005). 3 Bayley, Democratizing the Police Abroad: What to Do and How to Do It, 76. 4 Graham Ellison, "A Blueprint for Democratic Policing Anywhere in the World: Police Reform, Political Transition, and Conflict Resolution in Northern Ireland," Police Quarterly 10, no. 3 (2007): 244. 5 Policing in Israel: Studying Crime Control, Community Policing, and Counterterrorism, (Boca Raton, FL: CRC Press, 2016); Linda S Miller and Karen M Hess, Community Policing: Theory and Practice (Minneapolis: West Publishing Co, 1994); Robert C Trojanowicz and Bonnie Bucqueroux, Community Policing: How to Get Started (Cincinnati, Ohio: Anderson Publishing Co, 1994); Andre Normandeau, "Community Policing in Canada: A Review of Some Recent Studies," American Journal of Police 12, no. 1 3 policing “that emphasizes working proactively with citizens to solve crime related problems and prevent crime.”6 In this model of policing, citizens are viewed not as passive consumers of public safety, but as active participants in helping to provide public safety.7 Importantly, community policing is contextual and practical in its application. That is, while there are guiding principles and expectations as to what community policing looks like, there is no specific formula that can be applied everywhere to establish effective democratic policing.8 The theory of community policing is purposefully intended to be adaptive to the needs of the specific communities in which it is implemented and therefore may provide more direct guidance than the general ideals of democratic policing provide. Specifically, as community policing emphasizes the active participation of the community in the process of policing, we would expect to see community consultation and input at all levels of the policy making process. Community consultation should be seen in the problem identification, policy development, decision- making, and implementation stages of policing. This, it is argued, will build relationships between the police and the community that foster confidence in the police.9 This dissertation combines these two literatures, bringing the logic of community policing to the discussion of post-violent ethnic conflict state rebuilding. Specifically, I (1993); Dennis Rosenbaum and Arthur J Lurigio, "An inside Look at Community Policing Reform: Definitions, Organizational Changes, and Evaluation Findings," Crime & Delinquency 40, no. 3 (1994). 6 Miller and Hess, 16. 7 Trojanowicz and Bucqueroux. 8 Ellison. 9 Bayley, Democratizing the Police Abroad: What to Do and How to Do It; David H Bayley and Robert Perito, "Police Corruption: What Past Scandals Teach About Current Challenges," in Special Report (Washington, DC: United States Institue of Peace, 2011). 4 perform one in-depth case study and one initial assessment of states that have attempted to implement a version of community policing after violent ethnic conflict: Northern Ireland and Kenya respectively. This study leads to two important conclusions. First, the adaptive nature of community policing enables the police to build community confidence through policing practice that is responsive to the needs of the specific communities being served. At its most basic level, community policing encourages community engagement at all stages of the process of police reform and implementation. As such, if done well, community policing will lead to new relationships between police and the communities they serve. This is how communities gain confidence in the police – through the everyday lived experience of responsive, respectful, adaptive police practice encouraged by community policing at its best. The literature on post-conflict security has largely ignored issues of implementation of security, and therefore does not do enough to highlight the role of the civilian police in post-conflict rebuilding efforts. There is an underlying assumption that police practice is important in this literature, but little on what good police practice looks like. Therefore, in my case studies, I highlight the degree to which the police: (1) consulted with the community at different stages of the reform process; (2) identified community needs and developed practices that responded to those needs; and (3) identified community concerns and developed practices that respected those concerns. To the extent police do all three of those things, they can be considered to have ‘good community policing practices,’ and we should expect to see improved levels of community confidence in the police sometime thereafter. Second, the symbolism surrounding the police has a strong effect on individuals’ abilities to acknowledge or accept police reform in post-ethnic conflict states. Symbols 5 themselves do not make people trust the police, but they can often be barriers to trusting the police if the symbols surrounding the police have become associated with discriminatory or biased behavior. This is especially true in ethnically divided societies where symbols are often associated with one culture within society over another. Therefore, careful and considered reform of policing symbols is important to open the possibility for policing reform to be accepted by disenfranchised communities in post- ethnic conflict states. In this dissertation, I attempt to highlight the kinds of symbolic changes that are likely to be most impactful in post-ethnic conflict states which, combined with good community policing practices should lead to improved levels of community confidence in the police. The remainder of this chapter will lay out my argument in more detail and explore how it fits into the larger literature. First, I explore several important concepts that help to clarify the scope of this dissertation including how I define state stability and why I specifically focus on ethnically based conflict. I then perform a very brief review of the literature on post-conflict rebuilding and highlight where I believe this dissertation can provide new and important insights. I briefly explain the specific focus this dissertation has on policing in post-conflict states that pursue democratic forms of governance before reviewing the literature on community policing, identifying both the theoretical underpinnings and the expectations of successful community policing. After situating this dissertation in the larger literature on post-ethnic conflict rebuilding and community policing, the last three sections of this chapter outline the methodology used and the data collected to examine my expectations in the cases of 6 Northern Ireland and Kenya. The conclusion of this chapter lays out the structure of the remainder of this dissertation. Important Concepts As implied above, the motivation for this dissertation comes from the post-ethnic conflict literature on building state stability. In this section, I explain how I define state stability in this project. Additionally, I briefly justify my choice to focus on ethnic conflict as a category distinct from other kinds of conflict. State Stability For my purposes, stability can be defined as the general acceptance of the legitimacy of the state by the populace. That is, the populace generally accepts the state in its current form without violent rebellion10 and regardless of their personal feelings about the effectiveness of the current government. Put another way, individuals may dislike the people or policies of the government but are not generally violently protesting the state borders or structure of government. This is a purposefully minimalist definition of stability, or what some scholars might call a ‘negative peace.’11 While the absence of violence is certainly not a measure of the presence of peace and harmony, this dissertation’s primary focus is on how states achieve the conditions under which they might pursue other important objectives, like a positive peace or the provision of certain state services and functions. 12 Therefore, I am bracketing the initial concept of stability 10 Max Weber, "Objectivity in Social Science and Social Policy," in The Methodology of the Social Sciences, ed. Edward A Shils and Henry A Finch (New York: The Free Press, 1949); Kenneth E. Boulding, Stable Peace (Austin: University of Texas Press, 1978); TheGlobalEconomy.com, "Political Stability - Country Rankings," https://www.theglobaleconomy.com/rankings/wb_political_stability/. 11 Johan Galtung, "Violence, Peace, and Peace Research," Journal of Peace Research 6, no. 3 (1969). 12 See Melani Cammett and Edmund Malesky, "Power Sharing in Postconflict Societies: Implications for Peace and Governance," The Journal of Conflict Resolution 56, no. 6 (2012).; Collier.; and Caroline 7 at a minimalist place where there is not widespread violent contestation of the state’s legitimacy. I focus on stability over a ten-year period in this dissertation for several reasons. First, while I agree with those who caution against using too long a length of time to define stability because of the likelihood of new and unrelated variables complicating the analysis, 13 I also agree with those who argue that state stability requires more than a minimal period of cessation of violence. 14 Additionally, it seems reasonable to assume that stability may take some time to ‘set in.’ I also agree with Samuel Huntington15 that there is importance associated with elections in democracies as the peaceful transition of power is a key marker for stability. Therefore, I want a period that will include at least one national election. As most national elections around the world take place somewhere on a 3 to 7-year cycle, 10 years should incorporate at least one election. Why ethnic violence? I focus on conflicts that are specifically delineated along ethnic lines. While most political scientists, anthropologists, sociologists, and historians agree that ethnicity is a construct based on imagined ties and shared histories,16 most also argue that those Hartzell, "Explaining the Stability of Negotiated Settlements to Interstate Wars," Journal of Conflict Resolutions 43, no. 3 (1999). 13 Robert Mandel, Global Security Upheaval: Armed Nonstate Groups Usurping State Stability Functions (Stanford, California: Stanford Security Studies, and imprint of Stanford University Press, 2013); Boulding. 14 See Caroline Hartzell, Matthew Hoddie, and Donald Rothchild, "Stabilizing the Peace after Civil War: An Investigation of Some Key Variables," International Organization 55, no. 1 (2001): 184.; Karl DeRouen Jr., Jenna Lea, and Peter Wallensteen, "The Duration of Civil War Peace Agreements," Conflict Management and Peace Science 26, no. 4 (2009). 15 Samuel Huntington, The Third Wave: Democratization in the Late Twentieth Century (Norman: University of Oklahoma Press, 1993), 267. 16 See Joane Nagel, "Constructing Ethnicity: Creating and Recreating Ethnic Identity and Culture," Social Problems 41, no. 1 (1994). This concept is also explained well, though contested as a valuable approach, in 8 identities, imagined or not, have real consequences for the ways in which humans interact with the world and each other. Conflicts in which ethnicity is used to identify and internally unify opposing sides present special problems that deserve special attention. Several critics of ethnic conflict as a category claim that most if not all ‘ethnic conflicts’ are based in economic or other complaints and that ethnicity is simply incidental to the true causes of the conflict.17 Others claim that the framework of ethnic conflict is simply less useful than other frameworks and should therefore be rejected as a primary tool of analysis in evaluating conflict.18 Despite the arguments against the usefulness of ethnic conflict as its own category, a growing number of scholars recognize and use ethnic conflict as defining a distinct set of conflicts. David Horowitz argues that ethnic conflicts “possess certain baseline characteristics in common.”19 These include societies that are deeply divided along ethnic lines, with relatively large ethnic groups that operate at the center of politics rather than on the periphery, and groups are defined by ascriptive differences.20 Scholars Chris Smaje, "Not Just a Social Construct: Theorizing Race and Ethnicity," Sociology 31, no. 2 (1997); Benedict Anderson, Imagined Communities: Reflections on the Origin and Spread of Nationalism (New York, New York: Verso, 1983). 17 See Beverly Crawford and Ronnie D. Lipschutz, eds., The Myth of "Ethnic Conflict": Politics, Economics, and "Cultural" Violence (Berkeley: International and Area Studies, University of California Berkeley, 1998).; John Mueller, "The Banality of "Ethnic War"," International Security 25, no. 1 (2000).; Nicholas Sambanis, "Do Ethnic and Nonethnic Civil Wars Have the Same Causes?: A Theoretical and Empirical Inquiry," Journal of Conflict Resolutions 45, no. 3 (June) (2001).; Victor Le Vine, "Conceptualizing "Ethnicity" and "Ethnic Conflict": A Controversy Revisited," Studies in Comparative International Development 32, no. 2 (1997).; and Charles King, "The Myth of Ethnic Warfare," Foreign Affairs 80, no. 6 (2001). 18 Bruce Gilley, "Against the Concept of Ethnic Conflict," Third World Quarterly 25, no. 6 (2004). 19 Donald Horowitz, Ethnic Groups in Conflict, 2nd ed. (Berkeley, California: University of California Press, 2000; repr., 2nd), 17. 20 Ibid. 9 such as Clifford Geertz argue that ethnicity and ethnic conflict are not only important, but also inevitable as group identification is strong and only becomes reinforced over time.21 Other scholars such as Anthony Smith have focused on the mobilization of ethnicity as a tool for conflict. He argues that ethnic identities in themselves do not create or necessitate conflict, but that elites can and have mobilized them to achieve political ends.22 Dozens of scholars have contributed to the study of ethnic conflict: its incidence, duration, and role in post-ethnic conflict rebuilding.23 Additionally, research centers have been established all over the world to further the study of ethnic conflict.24 While there are legitimate debates about when and where ethnic tensions should be understood as the main motivation in certain conflicts, most experts agree that many important conflicts are substantially rooted in ethnic tension and therefore are worth studying as a distinct category of conflicts. Review of the Literature on Post-Ethnic Conflict Rebuilding Since the end of the Cold War, many cases of intrastate violence have resulted in the maintenance of existing state borders with a variety of strategies for solving internal 21 Clifford Geertz, Old Societies and New States: The Quest for Modernity in Asia and Africa (London: Free Press of Glencoe, 1963). 22 Anthony D. Smith, Nationalism: Theory, Ideology, History (Malden, Mass.: Polity Press, 2001), 54-55. 23 See Ravi Bhavnani and Dan Miodownik, "Ethnic Polarization, Ethnic Salience, and Civil War," Journal of Conflict Resolutions 53, no. 1 (2009).; Collier.; DeRouen Jr., Lea, and Wallensteen.; James D. Fearon and David D. Laitin, "Ethnicity, Insurgency, and Civil War," American Political Science Review 97, no. 1 (2003).; Glassmyer and Sambanis.; Caroline Hartzell and Matthew Hoddie, "Institutionalizing Peace: Power Sharing and Post-Civil War Conflict Management," American Journal of Political Science 47, no. 2 (2003).; Stuart J. Kaufman, "Spiraling to Ethnic War: Elites, Masses, and Moscow in Moldova's Civil War," International Security 21, no. 2 (1996).; William J. Long and Peter Brecke, War and Reconciliation: Reason and Emotion in Conflict Resolution (Cambridge, Mass: MIT Press, 2003).; Jose G. Montalvo and Marta Reynal-Querol, "Ethnic Polarization, Potential Conflict, and Civil Wars," The American Economic Review 95, no. 3 (2005).; among many others. 24 Some examples include those at the University of Notre Dame, Queen’s University Belfast, Columbia University, and the University of Washington to name a few. 10 grievances. This is a significant change from states gaining independence through the end of colonialism or the break-up of ethnically diverse states into sovereign nation-states as was common in the period between the end of World War II and the end of the Cold War. The literature on how these states rebuild after conflict broadly sorts into three categories of strategies for rebuilding: ethnic representation in government, community reconciliation, and personal security. Each of these contribute to post-ethnic conflict state stability in important ways. Community Reconciliation Many scholars, and a great number of practitioners, emphasize the importance of community reconciliation after violent conflict. They argue that ethnic conflict destroys relationships between neighbors and creates barriers to trust that must be overcome before communities can participate in and support the state. Simply put, if old wounds are not addressed and remedied, individuals will be unable to move on and old tensions will rise to the surface quickly.25 Within this group, there is debate over what strategies for 25 See Susanne Buckley-Zistel, "We Are Pretending Peace: Local Memory and the Absence of Social Transformation and Reconciliation in Rwanda," in After Genocide: Transitional Justice, Post-Conflict Reconstruction and Reconciliation in Rwanda and Beyond, ed. Philip Clark and Zachary Kaufman (New York, New York: Columbia University Press, 2009).; John Ishiyama and Jeremy Backstrom, "Reconciliation and Conceptual Complexity: The Case of Post Conflict Kenya," African and Asian Studies 10, no. 4 (2011).; J. Nasongo, J. Achoka, and L. Wamocha, "Is Forgiveness and Amnesty a Panacea to Kenya's Post-Conflict Crisis?," African Journal of Political Science and International Relations 3, no. 4 (2009).; Priscilla Hayner, Unspeakable Truths: Transitional Justice and the Challenge of Truth Commissions (New York, NY: Routledge, 2011).; and J. Sarkin, "Promoting Justice, Truth, and Reconciliation in Transitional Societies: Evaluation Rwanda's Approach in the New Millennium of Using Community Based Gacaca Tribunals to Deal with the Past," International Law FORUM du droit international 2, no. 2 (2000). 11 community reconciliation are the most useful 26 and who should be the primary targets of reconciliation projects.27 The extent to which community reconciliation can provide post-conflict stability on its own is questioned by many of its most prominent practitioners.28 Some scholars imply the importance of other post-conflict rebuilding strategies working in concert with reconciliation, even if those interactions are not explicit in the study.29 Community reconciliation is perhaps one of the most thoroughly explored categories of explanatory emphasis in post-conflict rebuilding. Repairing old and building new relationships after conflict is very difficult and therefore very interesting both theoretically and empirically. However, it does not seem to offer enough in terms of state stability to be a plausible explanation on its own. Individuals who still fear for their safety are not likely to participate in community reconciliation procedures fully or willingly, undermining the effectiveness of such programs. Security needs to come first. 26 Ifat Moaz, "An Experiment in Peace: Reconciliation-Aimed Workshops of Jewish-Israeli and Palestinian Youth," Journal of Peace Research 37, no. 6 (November) (2000). Antonia Chayes and Martha Minow, eds., Imagine Coexistence: Restoring Humanity after Violent Ethnic Conflict (San Francisco: Jossey-Bass, 2003), 7. Andrew Rigby, Justice and Reconciliation: After the Violence (Boulder, CO: Lynne Rienner Publishers, 2001). Gary Bass, Stay the Hand of Vengeance: The Politics of War Crimes Tribunals (Princeton, New Jersey: Princeton University Press, 2000).; Charles T. Call, "Is Transitional Justice Really Just?," The Brown Journal of World Affairs 11, no. 1 (2004). 27 John Paul Lederach, Building Peace: Sustainable Reconciliation in Divided Societies (Washington, DC: U.S. Institute of Peace Press, 2010), 41-42. Moaz. Chayes and Minow, 7. 28 ; Call, "Is Transitional Justice Really Just?."; Mohammed Abu-Nimer, ed. Reconciliation, Justice, and Coexistence: Theory & Practice (Lanham, MD: Lexington Books, 2001). Aneelah Afzali and Laura Colleton, "Constructing Coexistence: A Survey of Coexistence Projects in Areas of Ethnic Conflict," in Imagine Coexistence: Restoring Humanity after Violent Ethnic Conflict, ed. Antonia Chayes and Martha Minow (San Francisco: Jossey-Bass, 2003); Diana Chigas and Brian Ganson, "Grand Visions and Small Projects: Coexistence Efforts in Southeastern Europe," ibid., ed. Antonia Chayes and Martha Minow; Moaz. Naomi Roht-Arriaza, "The New Landscape of Transitional Justice," in Transitional Justice in the Twenty-First Century: Beyond Truth Versus Justice, ed. Naomi Roht-Arriaza and Javier Mariezcurrena (New York, NY: Cambridge University Press, 2006), 3. 29 Lederach. 12 Ethnic Representation in Government Scholars who focus on ethnic representation in government broadly argue that, after violent ethnic conflict, minority groups are concerned about having a voice in the decisions and direction of government. Therefore, representation must be ensured in some way. Representation can be understood as either direct representation, with some number of politicians in government coming from minority ethnic groups to ensure the physical presence of ethnic diversity, often in the form of power-sharing agreements30 – what many scholars call consociationalism.31 Alternatively, some have argued for the requirement of representation in votes. That is, executives must receive votes from a certain percentage of each ethnic group within a state to be eligible for office – what can be called centripetalism.32 This kind of representation, it is argued, incentivizes moderation in politicians, as they are required to appeal to multiple groups to be elected, rather than catering to one group’s preferences. Even some of those who seem to take a side in the argument over consociationalism versus centripetalism have ultimately argued that neither solution is perfect and may not provide, on its own, stable states in the aftermath of violent ethnic 30 Hartzell, Hoddie, and Rothchild.; Hartzell and Hoddie.; DeRouen Jr., Lea, and Wallensteen.; Bumba Mukherjee, "Why Political Power-Sharing Agreements Lead to Enduring Peaceful Resolution of Some Civil Wars, but Not Others?," International Studies Quarterly 50, no. 2 (2006). 31 Arend Lijphart, "Constitutional Design for Divided Societies," Journal of Democracy 15, no. 2 (2004); Donald Horowitz, "Ethnic Power Sharing: Three Big Problems," ibid.25 (2014); R.B. Andeweg, "Consociationalism," International Encyclopedia of Social & Behavioral Sciences (Elsevier Ltd., 2001); Joel Selway and Kharis Templeman, "The Myth of Consociationalism? Conflict Reduction in Divided Societies," Comparative Political Studies 45, no. 12 (2012). 32 Donald Horowitz, "Democracy in Divided Societies," Journal of Democracy 4, no. 4 (1993); Benjamin Reilly, "Institutional Designs for Diverse Democracies: Consociationalism, Centripetalism, and Communalism Compared," European Political Science 11, no. 2 (2012); Allison McCulloch, "Does Moderation Pay? Centripetalism in Deeply Divided Societies," Ethnopolitics 12, no. 2 (2013). 13 conflict. Horowitz, for example, has promoted centripetalism at various times throughout his career33 and in 2014 he explicitly argued that there is no perfect solution.34 Arend Lijphart and Adrian Guelke both emphasize political will as crucial to the success of consociationalism, implying that these kinds of arrangements often fail.35 What these authors seem to argue is that ethnic representation in government cannot explain, in itself, post-ethnic conflict state stability. Security, it seems logical to suggest, may be the missing link. Security The final explanatory emphasis in the post-conflict rebuilding literature centers on the importance of personal security. This group of scholars argues the primary concern of the average person following violent conflict is their safety.36 Without feelings of personal security, it will be impossible for individuals to comfortably participate in community reconciliation activities or feel that political representation is meaningful. There are two main approaches to ensuring security in a post-conflict setting. The first is a militaristic approach that indicates that peace in a post-civil war society can be maintained only through the presence of a strong national army or outside military 33 Horowitz, Ethnic Groups in Conflict; "Democracy in Divided Societies." 34 "Ethnic Power Sharing: Three Big Problems." 35 Arend Lijphart, "Review Article: The Northern Ireland Problem; Cases, Theories, and Solutions," British Journal of Political Science 5, no. 1 (1975); Democracy in Plural Societies: A Comparative Exploration (New Haven: Yale University Press, 1977); Arend Lijphart and Don Aitkin, Electoral Systems and Party Systems: A Study of Twenty-Seven Democracies, 1945-1990 (New York: Oxford University Press, 1994). Adrian Guelke, Politics in Deeply Divided Societies (Cambridge, UK: Polity, 2012), 132. 36 See Kaufman.; Collier; Glassmyer and Sambanis; Rama Mani, Beyond Retribution: Seeking Justice in the Shadows of War (Cambridge, Mass: Polity Press, 2002). 14 intervention.37 Chaim Kaufmann takes the rather extreme position that there are only three possible solutions to ethnic civil war: “with complete victory of one side; by temporary suppression of the conflict by third party military occupation; or by the self- governance of separate communities.”38 Other authors find there are other possible solutions to peace-keeping dilemmas in post-civil war societies but agree that militaristic peace-keeping forces are the most likely to be successful.39 Still others focus on a military peacekeeping force, but with an emphasis on state armies that incorporate rebel forces, which they argue should create stability.40 The other approach to security focuses on civilian police forces. Military peacekeeping may be sustainable in the immediate aftermath of a civil war, but there must come a time when peacekeeping responsibilities are devolved to the civilian police force. If settlements do not address issues of community trust in policing in the long term, there is a chance that violence will erupt again when this step is taken. Rama Mani argues that when actors in a post-conflict society focus too narrowly on enforcing the ‘rule of law’ through a military force, they do so in the interest of short-term peace to the detriment of long-term stability.41 Therefore, one important step in reforming the police is to de-link it from the military in societies where the two have become intertwined 37 Kaufman; Collier; Glassmyer and Sambanis. 38 Kaufmann, 139. 39 Collier. 40 Glassmyer and Sambanis. 41 Mani, 53. 15 during conflict.42 Many practical problems arise in this process, such as who should be recruited to the new police force and how. For instance, if the military is downsized, civilian police forces are a natural fit for former members of the military. Should these individuals be allowed to join the police force? Mani demonstrates through a thorough examination of the El Salvador and Haitian demilitarization of police that while these questions are difficult, addressing them at the start of a new government is the best way to ensure stable, trusted policing in post-war societies.43 Similarly, Charles Call and William Stanley argue that policing reform after civil war is important and should be addressed in peace agreements to better ensure they are carried out in the long term.44 In his many writings on policing in post-civil war societies, Call focuses primarily on the structural issues of police reform – whether to integrate former combatants, a focus on training and professionalization, and/or the integration of human rights – but does not dig much below the surface of these overarching ideas.45 This is likely because his primary focus is on what, if anything, the international community could and should do to help post-conflict states rebuild.46 He concludes, in large part, that police reform is important and useful in building medium- and long-term security,47 but again he does not address the specific reforms that are most useful. Several 42 Ibid., 59. 43 Ibid., 59-63. 44 Call and Stanley. 45 Stanley and Call. 46 Call and Stanley. 47 Call, "Competing Donor Approaches to Post-Conflict Police Reform."a 16 other scholars have performed case studies looking at police reform in Bosnia and Herzegovina,48 South Africa,49 and Nigeria50 among others that reach much the same conclusion. In recent years, there has been a larger focus on the role of civilian police forces in post-ethnic conflict societies.51 These studies analyze the experiences of policing in different ethnically divided societies and highlight the difficulties and some of the practical solutions employed in these states. They focus primarily on the widespread issue of police corruption.52 These authors argue that the issue of police corruption, though problematic everywhere, is a particular challenge in countries emerging from violent ethnic conflict.53 In analyzing this problem, they also examine the steps that have been taken to combat this issue in different countries around the world and evaluate the success or failure of these steps.54 While the issue of police corruption is undoubtedly a major obstacle to communities developing confidence in the police, and therefore developing a sense of 48 Gemma Collantes Celador, "Police Reform: Peacebuilding through 'Democratic Policing'?," International Peacekeeping 12, no. 3 (2005). 49 Janine Rauch, "Police Reform Anbd South Africa's Transition," in Crime and Policing in Transitional Societies (Jan Smuts House, University of the Witwatersrand, Johannesburg2000). 50 Innocent Chukwuma, "Police Transformation in Nigeria: Problems and Prospects,"ibid. 51 Aogan Mulcahy, "Policing in Divided Societies," International Encyclopedia of the Social and Behavioral Sciences (Elsevier Ltd, 2015); Graham Ellison and Jim Smyth, The Crowned Harp - Policing Northern Ireland (London: Pluto Press, 2015); Policing in Israel: Studying Crime Control, Community Policing, and Counterterrorism.; among others. 52 Bayley and Perito; Kempe Ronald Hope, Sr., ed. Police Corruption and Police Reforms in Developing Societies (Boca Raton: CRC Press, 2016); Rick Sarre, Dilip K. Das, and H.J. Albrecht, eds., Policing Corruption: International Perspectives (Lanham: Lexington Books, 2005). 53 Bayley and Perito. 54 Hope. 17 personal security, these studies do not do enough to highlight the many other obstacles to community confidence in the police that exist in post-ethnic conflict societies such as the perception of ethnic favoritism or discrimination,55 histories of police violence,56 and the reliance of some communities on paramilitary groups.57 For example, by focusing on corruption, these authors emphasize the absence of negative behaviors over the presence of positive behaviors in the police and therefore miss an important part of the solution. That is, an individual in society is likely to be glad if they no longer experience illegal shakedowns by members of the police, but the absence of that obvious corruption will not create a sense of confidence in the capability of the police.58 There needs to be an equal emphasis on creating new, positive police behavior. Similarly, the focus on corruption creates a false narrative of everyone distrusting the police equally. However, in ethnically divided societies, it is often the case that the police are associated with the socially advantaged group and therefore disadvantaged groups have an additional grievance 55 David H. Bayley, "Post-Conflict Police Reform: Is Northern Ireland a Model?," Policing: A Journal of Policy and Practice 2, no. 2 (2008); Ronald John Weitzer, Policing under Fire: Ethnic Conflict and Police- Community Relations in Northern Ireland (Albany: State University of New York Press, 1995). 56 Hope; Edwin A. Gimode, "The Role of the Police in Kenya's Democratisation Process," in Kenya: The Struggle for Democracy, ed. Godwin R. Murunga and Shadrack Wanjala Nasong'o (New York: Palgrave Macmillan, 2007); Ahmet H. Aydin, "The Future of Policing in Turkey: Current and Proposed Organizational Changes in Public Policing," Police Studies 19, no. 4 (1996). 57 Bayley, Changing the Guard: Developing Democratic Police Abroad; Caparini and Marenin; Chukwuma; Arrick Jackson and Alynna Lyon, "Policing after Ethnic Conflict - Culture, Democratic Policing, Politics and the Public," International Police Strategy and Management 25, no. 2 (2002). 58 This assertion is backed up in the social psychology literature on gains and losses. Some sources include: Lorraine Chen Idson, Nira Liberman, and E. Tory Higgins, "Distinguishig Gains from Nonlosses and Losses from Nongains: A Regulatory Focus Perspective on Hedonis Intensity," Journal of Experimental Social Psycology 36, no. 3 (2000); Jehan Sparks and Alison Ledgerwood, "When Good Is Stickier Than Bad: Understanding Gain/Loss Asymmetries in Sequential Framing Effects," Journal of Ecperimental Psychology: General 146, no. 8 (2017); Charles S Areni and Mitchell Burger, "Memoried of "Bad" Days Are More Biased Than Memories of "Good" Days: Past Saturdays Vary, but Past Mondays Are Always Blue," Journal of Applied Social Psychology 38, no. 6 (2008). 18 against the police that is not likely to be alleviated with addressing corruption. These authors allow corruption to do too much of the work for community distrust of the police to the detriment of other important factors specifically present in ethnically divided societies. In their assessment of policing after ethnic conflict, Arrick Jackson and Alynna Lyon make explicit what many of the authors reviewed so far have been arguing implicitly: namely that the police after ethnic conflict require the confidence of the community to have legitimacy.59 Police legitimacy, as defined by Ronald Weitzer, is “acceptance of the moral authority of a police force and its rights to enforce laws and issue commands.”60 Jackson and Lyon make the compelling argument that the police face a special challenge in convincing the community that they are trustworthy. However, they do not suggest how this is to be done. Instead, they take a rather fatalistic view that, especially when large portions of the police force carry over from conflict to post-conflict policing, the police will be unable to “overcome their negative past.”61 They argue that the police must “generate a viable constituency and prove to the population that they are trustworthy and legitimate in upholding the rule of law.”62 However, once again, they do not say how this is to be done. I argue that none of these studies has dug beneath the surface enough to identify not just whether police reform works, but how and why it succeeds or fails. As 59 Jackson and Lyon, 234. 60 Weitzer, 85. 61 Jackson and Lyon, 234. 62 Ibid. 19 mentioned, Call argues that the inclusion of a focus on police reform in peace agreements increases the likelihood of their long-term success, but he does not go into detail about what specific categories of reform should receive focus.63 Jackson and Lyon argue that community buy-in is necessary for the police to have legitimacy, but do not explore how such buy-in is achieved.64 In her article on police reform in Afghanistan following the U.S. intervention in 2001, Tonita Murray specifically identifies what she calls the “implementation gap” which “is concerned with such problems as lack of attention to the practicalities of implementation, lack of a common view, coordination and the translation of policy into practice.”65 She argues that without careful planning of how the Afghanistan police were going to put policy into practice, many of the reforms, such as training, failed to produce intended results. Murray argues that the international groups who helped to set up the new Afghanistan police service did not thoughtfully plan how their policy goals would be implemented. I am building on this insight – that it is not just written policy, but the practice of policing that makes good policing – and arguing further that community policing provides the best bridge to span her implementation gap in post-ethnic conflict societies. I argue that it is both the advanced planning and the ongoing learning and modifying of the implementation of policy that has made the reforms to policing in Northern Ireland successful in building community confidence in the police. In other words, I agree with 63 Call, "Competing Donor Approaches to Post-Conflict Police Reform." 64 Jackson and Lyon. 65 Tonita Murray, "Police-Building in Afghanistan: A Case Study of Civil Security Reform," International Peacekeeping 14, no. 1 (2007): 120. 20 Murray that there needs to be advanced, thoughtful planning to make implementation of policy successful, but it is also important to continually evaluate and modify police practice to ensure that it is translating policy into practice in the best way. In terms of the three primary identifiers of good community policing I mentioned above, this means that it is not enough to see the police engage with the community at all stages of police reform once, but that consultation and engagement is institutionalized and ongoing. In other words, community policing implies ongoing relationships with between the police and the community, not a simple one-time consultation. Democracy and Civilian Policing One possible reason for the limited literature about civilian policing in the aftermath of violent ethnic conflict may be that, in an effort to speak to a broader number of cases which include a significant number of authoritarian and hybrid regimes, scholars have treated post-conflict security as one broad category as opposed to breaking it down into its various forms more specifically. As mentioned above, some scholars have argued that personal security is the most important guarantee for individuals in the aftermath of violent conflict.66 For some of these scholars, any competent force including the military or an existing police force can provide security, or, in some cases, it is argued that outside forces such as United Nations Peacekeepers would be the preferred force to ensure unbiased protection of all citizens.67 These arguments are compelling especially when applied to specific cases where the history of violence in the state has been extreme, such 66 Collier; Kaufmann; Mani; Glassmyer and Sambanis. 67 Donna Pankhurst, "Issues of Justice and Reconciliation in Complex Political Emergencies: Conceptualizing Reconciliation, Justice and Peace," Third World Quarterly 20, no. 1 (1999): 250. 21 as in Rwanda following the 1994 genocide or Columbia following their decades-long violent civil war. Another reason these arguments may be so prominent is that the literature on post-conflict rebuilding would mostly be categorized as international relations literature. This field of political science tends to focus on the proximate causes and solutions to conflict, as opposed to the long-term solutions. Scholars who emphasize the need for long-term solutions to post-conflict security often also acknowledge that the immediate need for security may be met in a number of different ways as laid out in the literature.68 However, in the context of states that choose to pursue democratic forms of government in the aftermath of violent ethnic conflict it is important to examine what these long-term solutions look like. In authoritarian states, domestic security can be, and often is, provided by a military security force. Such forces do not require the confidence of the communities they serve to be effective. That is, with the authority and backing of the government, military police forces can use any means required to maintain order, with or without the consent or approval of the community. However, most scholars agree that democracy requires the guarantee of certain civil rights and liberties.69 This requires democratic states to develop policing that is necessarily more restrained than in authoritarian regimes. That is, if citizens in democratic states are entitled to certain limitations on government interference, then governments are not able to force compliance through military rule. Therefore, as mentioned above, there must come a time 68 See Mani; Jackson and Lyon; Murray. 69 Michael Coppedge et al., "Conceptualizing and Measuing Democracy: A New Approach," Perspectives on Politics 9, no. 2 (2011). 22 after conflict when the responsibilities of policing devolve to a civilian police force. If issues of policing are not addressed at the time of a peace agreement or other end to violence, it is likely that violence will resume when such transitions do take place.70 Thus, it is important to not only understand the role of security in post-ethnic conflict states, but to specifically address how security will be provided by a civilian police force and what reforms should be addressed in this process. It should be noted that the definition of democratic governance used here leans toward the more liberal or substantive definitions of democracy as opposed to limited or procedural definitions. The debate over what the definition of democracy is and should be is ongoing in the field and is much too rich to be adequately reviewed here.71 However, it is worth noting that not all scholars agree on the necessary attributes a democratic regime must have. Some scholars argue for a minimalistic definition of democracy in which there are regular free and fair elections, most individuals are allowed to vote, and there are some guarantees for civil liberties such as freedom of expression and access to sources of information (such as freedom of the press).72 There are many reasons to use this procedural definition of democracy in political science. Maybe the most important reason is that broader definitions increase the number of cases that fit the description of democracy and therefore provide a larger sample size when making comparisons and drawing conclusions. 70 Mani, 58; Call, "Competing Donor Approaches to Post-Conflict Police Reform." 71 For a good review of the history of this debate see Coppedge et al. 72 Robert A. Dahl, A Preface to Democratic Theory (Chicago: University of Chicago Press, 1956); Adam Przeworski et al., Democracy and Development: Political Institutions and Well-Being in the World, 1950- 1990 (Cambridge: Cambridge University Press, 2000); Joseph A. Schumpeter, Capitalism, Socialism, and Democracy (New York: Harper & Bros, 1950). 23 On the other hand, some scholars choose to use more liberal or substantive definitions of democracy. Such definitions imply that democracy is not just about certain procedures and rights, but that there is a depth or strength to such rights that can be measured and should be considered when determining how democratic a state is. Scholars that use substantive definitions tend to argue that democracies can always deepen or become more democratic. Substantive definitions of democracy often include elements such as the level of participation in civil society, measures of equity and equality among minority populations, and accountability of government institutions among other traits.73 For the purposes of this dissertation, I adopt a definition of democracy that might be called ‘procedural plus.’ This falls somewhat short of what Coppedge et al define as ‘liberal democracy’ which seems to imply expansive and well-protected civil liberties.74 Because I am focused on the state immediately post-ethnic conflict, I do not want to severely limit the range of relevant cases by only looking at cases of deep democracies with extensive civil rights and liberties. However, it is important that there be at least one additional criteria on top of the procedural definition of democracy for security to be able to contribute to state stability post-conflict. Specifically, democracy in this case requires legal or administrative accountability. This is to say that there is legal recourse for individuals in the community to report bad behavior, in this case of police officers, and reasonably expect that the complaint will be taken seriously, adjudicated fairly, and 73 See Dahl. on 'Madisonian Democracy.'; Scott Gordon, Controlling the State: Constitutionalism from Ancient Athens to Today (Cambridge, MA: Harvard University Press, 1999); David Held, Models of Democracy, 3rd ed. (Stanford, CA: Stanford University Press, 2006). 74 Coppedge et al., 253. 24 appropriate action taken where justified. The justification for this additional requirement is simple: for the community to trust that security provided by the police after conflict will be fair and unbiased, they must also be confident that the state has the mechanisms to address and correct bad behavior when it arises. In other words, marginalized communities are unlikely to accept that reform of how policing is delivered is indicative of real rather than superficial change unless they also trust that deviations from reform will be addressed and corrected. Without making any normative judgements as to what is the ‘best’ form of government in a post-ethnic conflict state, this dissertation concentrates on the special problems associated with providing security through a civilian police force in post-ethnic conflict states pursuing democratic forms of governance using a ‘procedural plus’ definition of democracy. In these cases, how do police earn the confidence of the community following violent ethnic conflict? Community Policing Previous sections identified where the literature on post-ethnic conflict rebuilding has fallen short in identifying how security is provided by the civilian police. There is a separate literature, however, that addresses the issue of building community confidence in the police through specific reforms: the literature on community policing. Community policing can generally be understood as an institutional change in the delivery of policing that works to build positive relationships between the community and the police, changing negative perceptions and increasing police legitimacy through the integration of 25 community input into policing at every stage.75 Using the literature on community policing, I am able to develop a set of expectations that I can use to evaluate how (or how much) community policing was implemented in my cases, and whether practices associated with community policing increased levels of trust with citizens, thereby enhancing state stability. In this section, I will briefly review the literature on community policing to identify expectations regarding what kinds of police practices we should see in places where community policing has been fully implemented. Defining Community Policing Community policing was largely developed theoretically and practically in the late 1980s through the 1990s. The purpose of community policing, or almost any policing reform, is to improve the quality and efficiency of the police. Community policing, in particular, does this in part through a focus on building better relationships between the police and the communities they serve.76 This is particularly important after ethnic conflict. As Gemma Celador argues, “In most, if not all, war-torn societies (from intra- state conflict) the police were, prior to and/or during the conflict period, politically biased, militarized, corrupt, ethnically (or group) divided, disrespectful of human rights 75 Herman Goldstein, "Toward Community-Oriented Policing: Potential, Basic Requirements, and Threshold Questions," Crime & Delinquency 33, no. 1 (1987); Gary Cordner, "Community Policing: Elements and Effects," Police Forum 5, no. 3 (1995); Ellison; Joanne Ziembo-Vogl and Woods Devere, Jr., "Defining Community Policing: Practice Versus Paradigm," Police Studies 19, no. 3 (1996); Brian N. Williams, Ralph S. Brower, and W. Earle Klay, "Community-Centered Police Professionalism: A Template for Reflective Professionals and Learnign Organisationa with Implications for the Co-Production of Public Safety and Public Order," The Police Journal: Theory, Practice and Principles 89, no. 2 (2016). 76 Goldstein; Ellison; Robert M. Lombardo, David Olson, and Monte Staton, "The Chicago Alternative Policing Strategy: A Reassessment of the Caps Program," Policing: An International Journal of Police Strategies & Management 33, no. 4 (2010); Elinor Ostrom and Gordon Whitaker, "Does Local Community Control of Police Make a Difference? Some Preliminary Findings," American Journal of Political Science 17, no. 1 (1973); Williams, Brower, and Klay. Among others. 26 and inefficient at ensuring the security of all citizens.”77 This creates a condition in which post-conflict policing specifically must address issues of community confidence in the post-conflict police as socially disadvantaged communities are likely to be skeptical of the police in general and socially advantaged groups will be distrustful of changes to the police they see as capitulating to the demands of former combatants.78 Importantly, the specific practices the police employ when they interact with the community, often in moments of extreme emotion and with physical interventions, have concrete importance in whether the community accepts the police as legitimate. In the context of this dissertation, this relates directly to the post-conflict literature’s focus on security as important for post-conflict stability. As Robert Gimode highlights, “the police force is the most apparent manifestation of the state,”79 and therefore community interactions with the police after conflict will affect how communities interpret the policies and prejudices of a post-conflict state. The ways in which police policy translates into policing procedure matter significantly in how the community interprets the role and intentions of the police. This is to say that the lived experience of interactions with the police are how communities learn to interpret the police, and, especially after violent ethnic conflict these interactions are how people learn to change their perceptions of the police. When community policing is implemented well, we would expect to see changes in public attitudes toward the police. This may be measured differently in different societies 77 Celador, 364. 78 R. Allen Hayes, "Policing in Northern Ireland: Community Control, Community Policing, and the Search for Legitimacy," Urban Affairs Review 49, no. 4 (2012). 79 Gimode, 228. 27 depending on the resources available. For example, in Northern Ireland they have been conducting at least yearly public surveys on perceptions of the police that we can use to track changes in public confidence in the police. To get to the outcome of increased community confidence in the police, we must also understand how community policing is practiced and what it looks like. In their article on the definition of community policing, Joanne Ziembo-Vogl and Devere Woods, Jr. provide a thorough history of the different definitions that emerged in academia and in practice through this period. They present these ideas as competing definitions; however, we can view these differing definitions as emphasizing different aspects of a set of related practices to suit the authors’ immediate purposes. In this way, we can see this conversation as developing a set of expectations for community policing as opposed to definitions in conflict. I draw on all these authors to develop some specific expectations about both what community policing looks like in practice, and the outcomes we would expect to see if community policing is implemented well. The academic definitions of community policing Ziembo-Vogl and Woods discuss include those that focus on action – “proactive consultation”80 and “cooperation.”81 Other definitions focus on community policing as a philosophy of proactive, as opposed to reactive, policing.82 These definitions assert that community policing, then, will include institutionalized ways for the community to interact with the 80 Normandeau. 81 Helen Greene, "Community Oriented Policing in Florida," ibid., no. 3. 82 David L. Carter and Allen D. Sapp, "Issues and Perspectives of Law Enforcement Accreditation: A National Study of Police Chiefs," Journal of Criminal Justice 22, no. 3 (1994). 28 police in a forum that is intended to produce conversation and cooperation between the police and the community. This might include regular meetings in which the community and police identify issues of primary concerns for the community and discuss solutions, as happened in Chicago with the Chicago Alternative Policing Strategy (CAPS) program.83 Alternatively, this could look like more official groups where the community selects representatives to meet with the police regularly for consultation like the District Policing Partnerships (DPPs) in Northern Ireland. Some scholars instead identify common themes of community policing such as an emphasis on improving and increasing police-community contacts and decentralization of the police,84 or simply stating the overarching premise of better connecting with the community.85 Working from these definitions, community policing should include some institutional reorganization of the police that decentralizes decision making powers, allowing officers in the field to have some discretion in how best to implement police policy. It should also include specific efforts to build new relationships between officers and the community. This might look like a reorganization of the police to focus around community policing teams with other forms of policing supporting the community teams, as happened in Northern Ireland. Or this might include the physical relocation of police officer housing to the neighborhoods where the officers serve so that they are physically a part of the community they work in, as was recommended in Kenya. 83 Lombardo, Olson, and Staton. 84 Dennis Rosenbaum, "Community Crime Prevention: A Review and Synthesis of the Literature," Justice Quarterly 5, no. 3 (1988); Rosenbaum and Lurigio. 85 Herman Goldstein, "Toward Community-Oriented Policing: Potential, Basic Requirements, and Threshold Questions," ibid.33, no. 1 (1987); Problem-Oriented Policing (Philadelphia: Temple University Press, 1990). 29 Practitioners, Ziembo-Vogl and Woods note, have had an equally difficult time finding consensus on what community policing is exactly.86 These definitions vary from a focus on the geographically-based method of policing,87 to a philosophy of community input in policing,88 to simply policing in the public eye.89 These definitions argue that community policing teams will have specified, geographically bound areas where they work to build relationships with the community through input like that described above. Ziembo-Vogl and Woods choose to focus on Robert Trojanowicz and the work he began before his death in surveying community police officers to define what they thought they were doing when implementing community policing.90 With this project in mind, they begin from his more encompassing definition of community policing as “…a philosophy of full-service, personalized policing where the same office patrols and work in the same area on a permanent basis, from a decentralized place, working in a proactive partnership with citizens to identify and solve problems.”91 This conversation about community policing has continued for more than 20 years following the height of academic interrogation of the subject. Though these and many other authors emphasize different features of community policing to some degree, they all share a central thought that community policing is collaborative, adaptive, and 86 Ziembo-Vogl and Devere. 87 Mary Ann Wycoff and Timothy N. Oettmeier, Evaluating Patrol Officer Performanc under Community Policing: The Houston Experience (National Institute of Justice, 1994). 88 Thomas Joseph, "Walking the Minefields of Community-Oriented Policing," FBI Law Enforcement Bulletin 63, no. 9 (1994). 89 R.V. Fulton, "Community Policing on Horseback," Law Enforcement Technology 20, no. 5 (1993). 90 Ziembo-Vogl and Devere. 91 Trojanowicz and Bucqueroux, 6. 30 emphasizes specialized knowledge of the communities the police serve gained through experience and cooperation. Together this literature has produced a widely-recognized set of interrelated guidelines for how to build effective trust-based policing. One such guideline is that policing must be context specific.92 That is, there is an underlying assumption that different communities have different needs and expectations of the police. To deliver on good community policing, then, the police must not only consult with the community, but invest time and resources in developing experiential knowledge of how people in different communities receive or interpret police behavior. This requires police decentralization, community cooperation, and generally more contact between the police and the communities they serve.93 Some authors, specifically Trojanowicz, have given very specific prescriptions for how community policing should best be delivered. He argues that specific officers should be stationed permanently in the same positions in the same neighborhoods to develop the necessary relationships needed for good community policing.94 While this might produce the desired results of officers having deep knowledge of the community, it does not seem a realistic recommendation for all police services and may in fact lead to some undesired outcomes. For example, what happens if one or a few officers develop poor relationships with the community? Keeping them permanently in their positions may work to entrench negative perceptions of the police rather than improving community confidence. 92 Bayley, Democratizing the Police Abroad: What to Do and How to Do It. 93 Aydin; Rachel Neild, "Democratic Police Reforms in War-Torn Societies," Conflict, Security & Development 1, no. 01 (2001); Christopher E. Stone and Heather H. Ward, "Democratic Policing: A Framework for Action," Policing and Society: An International Journal 10, no. 1 (2000). 94 Trojanowicz and Bucqueroux; Community Policing: A Contemporary Perspective (Cincinatti: Anderson Publishing 1990). 31 Similarly, this recommendation also does not account for the natural rotation that will exist in the police. By this I mean that police officers, like any other person, will want to move up in their jobs, will physically move occasionally, or may seek change in their work life for many other reasons. The stipulation that an officer stay permanently in their job is an unrealistic expectation to put on officers that is not put on many, if any, other professionals. What Trojanowicz is implying, though, is that permanence of officers is one way to operationalize police knowing the context of and the people in their community. Drawing on all these authors, I emphasize an understanding of community policing that focuses on community consultation and engagement in every stage of policing. Specifically, I agree with Gary Cordner that since “community policing means many things to many people,” it is most useful to focus on the elements of community policing in “the Philosophical Dimension, the Strategic Dimension, and the Programmatic Dimension.”95 In other words, police organizations that engage in community policing will include community engagement in the development of their overarching policing philosophy, in the development of their policing strategies, in their implementation of their policing programs, and in the ongoing evaluation and alteration of policing. Theoretical Underpinnings There are three literatures that can be identified as contributing the theoretical underpinnings of community policing: pragmatism, practice theory, and political 95 Cordner, 1. 32 symbolism. In each of these literatures, one can identify important aspects of community policing which help to develop the expectations of how community policing should be practiced. Pragmatism To develop expectations about what kinds of community policing reforms and practices are more likely to increase community confidence in the police, I draw insights from pragmatist theories of institutional change. In the pragmatist literature, John Dewey develops his concept of ‘habit.’ Habit, he argues, is an “acquired predisposition to ways or modes of response.”96 Habit, he further explains, is informed by prior experiences and activities, but is also dynamic and changing. In other words, while habit is in part a result of past experiences, it is also ‘projective’ or anticipatory of future behavior. Put another way, habit is lived and changing, not static. He emphasizes that habits are not routine by definition, but instead are always slightly different based on the combination of situation, skill, and rule.97 That is, routine or the exact response to every similar situation is not habit, but the breakdown of habit that has been separated from learning and skill. As Charles Sabel put it, “Dewey stressed… the mutual dependence of individuals and institutions, and the way the stunting of the one impoverished the other. Institutions and individuals thus had to change together or not at all.”98 Gerald Berk and Dennis Galvan use Dewey’s ‘habit’ to develop what they call ‘creative syncretism,’ an explanation of 96 John Dewey, Human Nature and Conduct: An Introduction to Social Psycology (Amherst, N.Y.: Prometheus Books, 2002), 40. 97 Ibid. 98 Charles Sabel, "Dewey, Democracy, and Democratic Experimentalism," Contemporary Pragmatism 9, no. 2 (2012). 33 institutional change that emphasizes the agency of actors within institutions in forming and reforming the rules of said institutions through habit.99 Dewey’s idea of ‘habit’ can help to understand some of the principles of community policing. In community policing, there is an understanding that the institution of policing can and should be dynamic and changing in response to the needs of the community. As with Dewey’s ‘habit,’ this implies that the development of routine in policing without dynamic learning and changing is not a good institutional development, but a breakdown in habit. In other words, community policing understood through the literature on pragmatism should always be dynamic and responsive while also drawing on experience to inform best practice. It should show us learning and continual adjustment based on experience and learned will. It should be slightly different in its application in every situation and will lead to different practices in different contexts. Deweyian pragmatism argues that this is the best possible outcome. Institutions, like the police, will function best when designed in such a way to “reduce the chance of organizational habits congealing into limited routines” and to “detect and dis-entrench routines that have become obstructive.”100 Community policing is built to do just that. Practice Theory Another literature that provides some theoretical underpinnings for community policing, and which is related to but distinct from pragmatism, is practice theory. Bourdieu develops his theory of practice using the concept of habitus – the embodied, 99 Gerald Berk and Dennis Galvan, "How People Experience and Change Institutions: A Field Guide to Creatuve Syncetism," Theory and Society 38, no. 6 (2009). 100 Sabel, 36. 34 internalized history – that “is the active presence of the whole past of which it is the product. ”101 He argues that Practices cannot be deduced either from the present conditions which may seem to have provoked them or from the past conditions which have produced the habitus, the durable principle of their production. They can therefore only be accounted for by relating the social conditions in which the habitus that generated them was constituted, to the social conditions in which it is implemented.102 In other words, Bourdieu is arguing that practice is not just the result of social conditioning and historical experience, but the combination of that history with the context of the specific social conditions in which that practice takes place. The habitus both informs and is shaped by practice. He says, “Understanding ritual practice is not a question of decoding the internal logic of a symbolism but of restoring its practical necessity by relating it to the real conditions of its genesis, that is, to the conditions in which it functions.”103 He uses this logic to argue that we must look beyond the simple logic of structures, and instead focus on the power relations present at the origin of the structure and therefore implicit in their design. This is a subtle, though important difference from Dewey’s ‘habit.’ Where habit for Dewey is a combination of prior experience and intentional creative action in response to context, habitus for Bourdieu is more an explanation of the internalization of 101 Pierre Bourdieu, The Logic of Practice, trans. Richard Nice (Stanford, CA: Stanford University Press, 1980), 56. 102 Ibid. 103 Outline of a Theory of Practice, trans. Richard Nice (Cambridge, United Kingdom: Cambridge University Press, 1977), 114. 35 adjustments in action in response to context. In other words, habit is intentionally creative where habitus is responsive to creative changes.104 Many other political scientists employ the theory of practice in their work. Vincent Pouliot uses Bourdieu’s theory of practice in his discussion of international security, arguing that interstate peace becomes possible in and through the practice of diplomacy. That is, “the contingent alignment between the practitioners’ dispositions (the stock of background knowledge accumulated through experience) and their positions in the field of international security (defined by evolving rules of the game and stocks of valued resources” is what makes possible the idea of diplomacy as the commonsensical practice for international peace.105 James Scott makes an argument for the importance of practice in his description of metis or specialized knowledge gained through experience.106 He argues that the activities of the state often tend to ignore metis in an effort to make generalized rules. While generalized rules have value in a society, they fail to recognize the importance of experientially acquired, specialty knowledge. This concept of specialized knowledge leading to ideal practice in localized contexts is particularly relevant to the argument made in this dissertation. Practice theory helps to explain how police gain specialized knowledge through community policing. That is, when decision-making is decentralized, officers build relationships and communicate regularly with the community, and those communications 104 Berk and Galvan, 551, footnote 3; J. Butler, "Performativity's Social Magic," in Bourdieu: A Critical Reader, ed. R. Schusterman (Oxford: Blackwell, 1999). 105 Vincent Pouliot, International Security in Practice: The Politics of Nato-Russia Diplomacy (New York: Cambridge University Press, 2010), 1. 106 James C. Scott, Seeing Like a State: How Certain Schemes to Improve the Human Condition Have Failed (New Haven: Yale University Press, 1998). 36 inform future behavior, community police officers develop the specialized knowledge required to perform good community policing. This is particularly important in moments of high tension where there may be a tendency among both police and communities to revert to ‘muscle memory’ in their handling of situations.107 Practice theory helps us to understand how police learn to combat the urge to revert to ‘old’ tactics that were the source of some portion of the distrust in the police during conflict through community policing. Instead, the police actively engage with Dewey’s pragmatic notion of ‘habit,’ employing new strategies to build confidence in the police. Symbolism The literature on political symbolism is also useful in understanding community policing. According to Mayer Zald, symbols use in politics is usually understood in one of two ways: as a part of the history of ideas and “the interaction of social and cultural systems,”108 or as a reflection of the social arrangement used to justify and reify that arrangement.109 Rozann Rothman makes the argument that humans have a natural desire to order the world around them and they do so, in part, through creating symbols and imbuing them with meaning. She argues, “symbols function as vehicles of communication and the means to strengthen group cohesion.”110 Susanne Langer argues that symbols are “vehicles for the conception of objects” and it is the conceptions, not the 107 Adrian McAllister, (paper presented at the Reflecting Back, Moving Forward: Fifteen Years Since the Establishment of the PSNI, University of Ulster, Belfast, United Kingdom, 9 November 2016). 108 Mayer N. Zald, "Politics and Symbols: A Review Article," Sociological Quarterly 7, no. 1 (1966): 85. 109 Ibid., 85-86. 110 Rozann Rothman, "Political Symbolism," in The Handbook of Political Behavior, ed. S.L. Long (Boston, MA: Springer, 1981), 285. 37 symbols themselves that evoke meaning.111 Symbols are both flexible and consistent in their meaning, possessing the same level of contradiction as the political world in which they operate.112 Therefore, symbols can be manipulated and change in their meaning over time. One obvious example of such manipulation is the swastika. The swastika was first a symbol of peace and harmony in many ancient cultures around the world before it was coopted by Adolf Hitler and the Nazi Party.113 Now, for most people in the modern world, the swastika is now a symbol of hate, violence, and genocide. The manipulation of this symbol is one example of the dual flexibility and consistency of symbols. The difficulty in developing an explicit model of symbols and their importance to politics is that the meaning of a symbol is implicit in nature. As Dan Sperber argues, “symbolic interpretation is not a matter of decoding, but an improvisation that rests on an implicit knowledge and obeys unconscious rules.”114 In this way, it is only possible to know that a symbol is important to a community, but not necessarily what that specific important is, because that symbol is likely to carry multiple, implicit meanings. Community policing, which as has been asserted is necessarily context specific and reflective of both the Dewey’s habit and Bourdeau’s habitus, is a system in which we would expect symbolic change to take into consideration the multiple meanings of 111 S. K. Langer, Philosophy in a New Key (New York: Mentor Books, 1948), 61. 112 Rothman, 286. 113John Black, "The Symbol of the Swastika and Its 12,000-Year-Old History," Ancient Origins: Reconstructing the Story of Humanity's Past, http://www.ancient-origins.net/myths-legends/symbol- swastika-and-its-12000-year-old-history-001312?nopaging=1. 114 Dan Sperber, Rethinking Symbolism (New York: Cambridge University Press, 1975), xi. 38 symbols and to result in new police symbols that are inclusive of all of society, or at least culturally neutral. The symbolism surrounding the police after ethnic conflict strongly affects the ability for people and communities to change their perceptions of the police in both positive and negative ways. The police are associated with a great deal of symbolism from their uniforms, badges, and emblems to the equipment they carry to the buildings in which they are located. Such symbols help to give the police a sense of belonging and purpose, and work to distinguish the police from the rest of the community in which they serve when they are on duty. As members of society who straddle the divide between civilians and the government,115 the symbolism of the police helps to distinguish the special role they play – for them and for the community. When police symbolism becomes associated with perceived bias or bigotry, the symbolism of the police becomes a barrier to some or all portions of the community. This is not to say that simply changing the name, uniform, badge, or other symbols associated with the police will automatically create community confidence in the police, but that such symbolic change opens the intellectual space to see the police as a new or reformed institution and gives the changes in police practice described below a chance to be effective. Importantly, any change to police symbols must also be mindful of communities who aligned themselves with the police during conflict. In some cases, one or more groups within a society see the police as ‘on their side’ or simply as allies. For these groups and for many members of the police force itself, changes in the symbolism of the 115 Belfast Civilian 4, interview by Shawna Meechan, 1 November, 2016, Belfast, United Kingdom. 39 police may be perceived as disrespectful of those who served and sacrificed during conflict. Further, making even symbolic changes to the police can be perceived as placating or capitulating to the demands of those recently (and often currently) seen as enemies. It is necessary to consider the concerns of these groups when making changes to police symbols after violent ethnic conflict to prevent unintentional barriers being formed between the reformed police and these groups. Expectations: What Community Policing Looks Like The understanding of community policing presented here, and its theoretical underpinnings lead me to two sets of expectations that we should see when community policing is applied in post-ethnic conflict societies. First, symbolism will be extremely important in opening the possibility for disadvantaged communities to build confidence in the police after ethnic conflict. Second, we should see concrete changes in police institutions and practice that incorporate community input throughout the process of policy development and implementation. Symbolism My first set of expectations is based on the idea that symbolism is important to the success of community policing. More concretely, with respect to symbolic change, community policing implies three specific outcomes. First, because of the consultative aspect of community policing, we would expect to see evaluation of the symbols of the police with community input. Such evaluations will help to identify the extent to which certain symbols need to be changed. For example, does the design of the police badge contain symbols that are associated with one part of the community and not the other? Would the community, in general, prefer a new completely culturally neutral symbol? Or 40 would they prefer a more inclusive design with multiple ethnic groups represented? The context specific nature of symbolic importance combined with the ethos of community participation in community policing requires that we see some sort of formal community input in symbolic evaluation. Second, we would expect to see movement toward symbolic representations of openness in the police. That is, since community policing requires contact and relationship building with the community, we should see some of the physical barriers to connection represented in the police to be reduced. For example, during conflict it is common for police to militarize in appearance, wearing body armor, driving reinforced vehicles, and building high walls or adding barbed wire to police stations. The implementation of community policing implies that these physical barriers between the police and the community will be reduced as quickly as can be done safely. More specifically, we should see patrol officers shifting to less bulky and militarized uniforms, see a shift from visually armored cars to less conspicuous vehicles, and see a shift in the construction and retrofitting of police buildings to be more integrated in the community and less foreboding for those who need to enter them. Importantly, these are all high-cost and long-term changes. Therefore, while we likely will not see immediate symbolic changes of this kind, we would expect to see an official plan for such changes and incremental steps in that direction over time. Third, employing community policing in ethnically divided societies also implies addressing issues of representation in the police. That is, if one portion of the community literally cannot see themselves represented in the police force, it will be difficult to accept any other symbolic changes as genuine. Therefore, in addition to the symbolic changes 41 discussed above, community policing in post-ethnic conflict societies will involve a scheme for increasing inclusiveness in the make-up of the police. This could be seen in a recruitment scheme meant to advantage recruits from underrepresented groups, a quota system in new hires, encouraged early retirement of officers, and/or programs for additional mentoring for existing minority-representing police officers so they can move up in the organizational structure of the police. Some or all of these elements should be present to increase the symbolic representation of the police. Practice The second set of expectations that community policing presents revolve around the changes to the institutions and practices of policing. Community policing posits that perceptions of the police will change positively if there is a shift toward a police- community partnership at all stages of the policy creation and implementation process. There are three broad areas of change that we would expect to see when community policing is enacted. First, we would expect to see some sort of training or reorientation of the police toward this new philosophy of community policing. Simply rearranging the organizational structure of the police to reflect community policing ideals will not, in itself, change policing behaviors. There needs to be buy-in from the police themselves if they are going to properly execute the ideals of community policing. Therefore, we would expect to see some sort of training of the police to orient themselves to community policing. This could be new training program for all police or an overhaul of the initial cadet training programs that includes an emphasis on human rights in policing and 42 building positive relationships with the community. A lack of such training will likely result in no or minimal increased community confidence in the police. Second, we would expect to see the involvement of members of the community in each stage of the policy making process. This includes the early stages of identifying problems, to developing ideas for solutions, to deciding which ideas to try, to the implementation of those ideas. Importantly, we would also expect to see continuing conversation and consultation as ideas are tried and revised to reflect the needs of the community and effective policing. We should expect to see this orientation toward partnership in both the institutions of the police and in the day-to-day practices of individual police officers on the ground. More concretely, this means that community policing requires institutionalized community input and participation in the initial evaluation of the police (in Northern Ireland through the Independent Commission on Policing or in Kenya through the National Taskforce on Police Reforms). We should also see a program or institution set up to facilitate ongoing community input in policing like was established in the Chicago CAPS program. Finally, we should be able to identify individual officers who have developed good relationships with individuals in the community shown through regular contact, whether by phone or email or in person. Third, we would expect to see decentralization of the police, giving community officers more autonomy in determining how best to handle policing in their communities. More importantly, though, we would see that autonomy lead to changes in police practice given the differing needs of different communities. That is, even in a relatively small society, different neighborhoods will have different dynamics and needs. We would expect to see different communities develop slightly different ways of policing in 43 reaction to the needs of the community. For example, if there is a certain festival or holiday that is celebrate each year, the celebrations may be more or less contentious in different neighborhoods and the police, in consultation with the community, will develop different ways of addressing the issues. An example we will look at more closely in later chapters involves the different ways the police in Northern Ireland have developed for policing the contentious parades season in different neighborhoods. Community policing implies we should find several examples like this where the police have developed strategies with the community to reduce tension and crime that is slightly different than communities with different needs or concerns. Community Policing Theory Simplified All the expectations I have laid out so far address what sorts of changes in practice and symbols community policing encourages. However, the primary purpose of community policing is that it is intended to create community confidence in the police through the changes explained above. Therefore, the most crucial expectation generated by the community policing literature is that we should see a positive change in community attitudes toward the police as community policing is implemented. Put another way, the changes to symbols and police practice laid out above may be considered my independent variables and community confidence in the police is my dependent variable. I am examining the plausibility of this theory using case comparisons to see if implementing community policing correctly actually does increase community confidence in the police. This requires some minor qualifications, however. Specifically, understanding that it takes time for communities to build confidence in the police, especially after 44 violent ethnic conflict, changes in the public’s level of confidence in the police would likely be slow, taking several years, and subject to fits and starts in reaction to important events in the community. That is, while this study attempts to isolate the important effects of community policing in post-ethnic conflict communities, it is important to acknowledge that these changes do not happen in isolation and that changes in community confidence in the police (my dependent variable) are likely to be affected by outside forces, otherwise unaccounted for in the theory of community policing. Such forces might include important political events that may, at least temporarily, result in large shifts in community confidence in the police. Ultimately, however, if community policing theory is correct, then we should see an increase in community confidence in the police over time. As already discussed above, I focus on the 10-year period following the implementation of community policing in my cases studies. Methods To test these expectations, I perform one in-depth case study and one initial assessment of a second case, or shadow case. The primary case I focus on is the experience of Northern Ireland following the period of prolonged ethnic violence known as the ‘Troubles.’ I follow the underlying ethno-cultural tensions that led to the outbreak of violence and the extremely painful 30-plus years of fighting of the Troubles. I then scrutinize the implementation of the 1998 Belfast/Good Friday peace agreement (the Agreement) and show the special role played by the police in rebuilding the Northern Irish state over the next thirteen years. I end my initial examination in 2011, the ten-year anniversary of the implementation of the re-vamped Police Service of Northern Ireland (PSNI). I show how the nearly polar opposite perceptions of the police as protectors in 45 the Protestant/Unionist/Loyalist (PUL) community and aggressors in the Catholic/Nationalist/Republican (CNR) community were at the forefront of the initial outbreak of violence during the Troubles. These tensions were therefore necessary to address in the Agreement and became one of the most important tests for the relative success of the Agreement and acceptance of the Northern Irish state after the Troubles, especially in the CNR community. In this longitudinal study, I use historical accounts, official state data and statistics, survey data, journalistic accounts, and interviews to develop the complicated story of the police in Northern Ireland. I highlight survey data on community perceptions of the police to track changes in those perceptions from various demographic constituencies within Northern Ireland from the last years of the Troubles beginning in 1997 through 2011. In addition, I use interviews conducted with police officers and civilians in Northern Ireland to supplement these statistics with explanations of why these changes have occurred. The second case, or shadow case, is an initial investigation of similar dynamics in the very different context of Kenya in the aftermath of the post-2007 election violence. Following this election, more than 1,000 people were killed, tens of thousands were injured, and hundreds of thousands were displaced.116 In the Kenyan case, the violence led to a peace agreement in which a temporary power-sharing government was installed until a new constitution could be written. Like Northern Ireland, the Kenyan agreement prioritized the reformation of police institutions and led to major changes to the structure and functioning of the police under the new 2010 constitution. While further intensive 116 Human Rights Watch, Ben Rawlence, and Chris Albin-Lackey, Ballots to Bullets: Organized Political Violence and Kenya's Crisis of Governance (New York: Human Rights Watch, 2008). 46 field research is necessary for a full comparative analysis of these cases, the initial examination of the Kenyan case presents important and interesting insights that help to define the scope conditions of my argument. Specifically, while the process of evaluation and implementation of police reform in Kenya looked remarkably like that in Northern Ireland on the surface, closer examination shows that Kenya did not implement community policing as claimed. A first look at why this happened helps to develop additional qualifications that better define where community policing can effectively be employed to build community confidence in the police. The remainder of this section identifies and justifies the primary methodological tools used in this dissertation. I also justify my case selection of Northern Ireland and Kenya. Process Tracing I am interested in the causal mechanisms at work in developing community confidence in the police after violent ethnic conflict. As discussed above, many scholars have identified post-conflict security as a necessary element of post-conflict state stability, but little has been done to explain the mechanisms through which this is done. I utilize process tracing to examine the mechanisms at work in implementing community policing in Northern Ireland and Kenya. George and Bennett argue, “Process tracing is an indispensable tool for theory testing and theory developments not only because it generates numerous observations within a case, but because these observations must be linked in particular ways to constitute and explanations of the case.”117 Process tracing is 117 Alexander L. George and Andrew Bennett, Case Studies and Theory Development in the Social Sciences (Cambridge, Mass.: MIT Press, 2005), 207. 47 defined as “the analysis of evidence on processes, sequences, and conjunctures of events within a case for the purposes of either developing or testing hypotheses about causal mechanisms that might causally explain the case.”118 It does this by looking at things such as the “…intentions, expectations, information, small-group and bureaucratic decision-making dynamics, coalition dynamics, [and] strategic interaction” that make up the complex processes that lead to observed outcomes.119 Case Selection A case is understood in this dissertation in the sense of George and Bennett where “a case is an instance of a class of events,”120 as opposed to a “phenomenon for which we report and interpret only a single measure on any pertinent variable.”121 For this dissertation, a ‘class of events’ can be understood as “a phenomenon of scientific interest… that the investigator chooses to study with the aim of developing theory … regarding the causes of similarities or differences among” cases.122 In this way, case studies may be understood as either a comparison of observations within a single case or a comparison across a small number of cases. I therefore term my study of Northern Ireland as a single case study containing multiple within-case observations. 118 Andrew Bennett and Jeffrey T. Checkel, eds., Process Tracing: From Metaphor to Analytic Tool (Cambridge: Cambridge University Press, 2015), 7. 119 Andrew Bennett and Alexander L. George, "Process Tracing in Case Study Research," in MacArthur Foundation Workship on Case Study Methods (Belfer Center for Science and International Affairs (BCSIA), Harvard University1997). 120 George and Bennett, 17.Emphasis mine. 121 Harry Eckstein, "Case Studies and Theory in Political Science," in Handbook of Political Science, ed. Fred Greenstein and Nelson Polsby (Reading, Mass.: Addison-Wesley, 1975), 85. 122 George and Bennett, 17-18. 48 My primary focus on Northern Ireland has several justifications. First, Northern Ireland is both academically123 and practically124 held up as an exemplar case in the study of post-ethnic conflict rebuilding. This type of justification may be termed ‘idiographic.’125 This is to say that some cases are intrinsically important for their historical significance or their personal importance for the group or actors involved. Though these types of cases are not often chosen for the purpose of creating generalizable conclusions, they can provide insights of broader significance.126 Northern Ireland has been held up as an important case for several reasons. As a part of the Agreement that officially ended the Troubles of Northern Ireland, the parties agreed that issues of policing reform needed to be addressed, though the specifics of those reforms were left up to future discussion. It was recommended that a working group should be formed to examine the police and make recommendations for future reform.127 This group, called the Independent Commission on Policing for Northern Ireland and composed of experts from all over the world including at least two notable experts in community policing, produced a report which included many observations of the position and perception of the 123 Robert Perry, "The Devolution of Policing in Northern Ireland: Politics and Reform," Politics 31, no. 3 (2011); Hayes; M. Brunger, "Governance, Accountability and Neighborhood Policing in Northern Ireland: Analysing the Role of Public Meetings," Crime Law and Social Change 55, no. 2-3 (2011); Guelke; Bayley. Among others. 124 Branka Marijan and Sean Brennan, "Policing the Past and Present in Northern Ireland," (2014), http://secgovcentre.org/2014/09/policing-the-past-and-present-in-northern-ireland/; Derry Officer 2, interview by Shawna Meechan, 4 October, 2016. 125 Eckstein; Jack S Levy, "Case Studies: Types, Designs, and Logics of Inference," Conflict Management and Peace Science 25, no. 1 (2008); Arend Lijphart, "Comparative Politics and the Comparative Method," American Political Science Review 65, no. 3 (1971). 126 John Gerring, Case Study Research: Principles and Practices, 2nd ed. (Draft2016), 39-40. 127 "Belfast (Good Friday) Agreement," (United Kingdom: Northern Ireland Office, 1998). 49 police in Northern Ireland and made 175 recommendations for reform, almost all of which were implemented in whole or in part.128 This model for assessing the police in the aftermath of ethnic violence has been implemented by many other countries following their own experiences with ethnic conflict.129 Additionally, individuals involved in the Northern Ireland transition have been invited to share their experiences and advice in many parts of the world including Chicago, Illinois, USA,130 Israel,131 and Kenya132 among many others. That Northern Ireland has been held up as an example of how to successfully reform the police in ethnically divided societies makes the examination and deeper understanding of the mechanisms at work there valuable in its own right. It is also important to ensure that when actors from Northern Ireland pass on their own experiences, there is an understanding of how the specific successes of Northern Ireland can translate into general advice for other societies. Second, Northern Ireland is a state that has experienced ample external pressure and assistance to succeed in their transition from violence. Understanding how and where they struggled and succeeded in their police reforms under a ‘best case’ scenario of economic prosperity and considerable investment from the United Kingdom, the United States, and the European Union is useful in highlighting the most important and most 128 Desmond Rea and Robin Masefield, Policing in Northern Ireland: Delivering the New Beginning? (Liverpool: Liverpool University Press, 2014). 129 Examples include Kenya, Spanish Basque country, and the United States among many others. 130 Forum for Cities in Transition, "Meeting," (Derry, Northern Ireland2015). 131 Derry Officer 2. 132 Philip N. Waki et al., "Report of the Commission of Inquiry into Post-Election Violence (the Waki Report)," (Nairobi, Kenya2008); Peter Gastrow, "The Kenya National Task Force on Police Reform: Some Key Recommendations Summarised," https://www.ipinst.org/wp- content/uploads/2010/01/pdfs_summary_policereformreport.pdf. 50 difficult aspects of reform. Other states that experience far less or no international support and pressure to succeed will have all the more trouble successfully creating post-conflict stability. In these states, it will be all the more necessary to reform the police and to concentrate on those areas of policing that are most helpful in building community confidence in the police if they are to have any chance of success. That is, if Northern Ireland struggled to develop community confidence in the police even in a situation where they had external support and external pressure to succeed, then we are likely to identify areas of reform that will be universally (or nearly so) difficult. Third, Northern Ireland has the ability to provide ample new evidence that is not readily available elsewhere. Northern Ireland had more than ten years of experience under the reformed police force when I began this project. This longevity provided a sufficient amount of data to make some reasonable conclusions in the development of the arguments forwarded in this project. Other states that have had similar lengths of time to implement police reform are slightly more difficult to obtain information on. For example, in South Africa, the reforms to the police after the end of Apartheid have been monitored, but as of 2014, they had yet to implement a measurement for ‘customer satisfaction.’133 So while there is some data available on efficiency measures within the police, they do not have ample historical data on community perceptions of the police. Also, because Northern Ireland has been held up as an example for many years, the police service is open to and comfortable with outside scholars’ inquiries and was therefore more than willing to cooperate in my research. I was able to get access to 133 Andrew Faull, "Performance Measurement in Police Agencies: A Report for the Khayelitsha Commission of Inquiry," (2014). 51 interview police officers at varying levels of power in two different cities and multiple different station locations. I was also able to speak with community members with varying levels of involvement with the police over the years. This access is much more difficult to get in other countries where they are still struggling to reform the police after conflict or live in fear that reforms already enacted will be endangered if researchers probe them too soon. This was certainly the case in Kenya when I attempted to arrange approved research of the police and the reforms enacted since the end of the 2007 election violence in late 2016 and early 2017. The attitude among those I was able to speak with was that the government was likely to be reticent to grant approval for my research in fear that the reforms would not hold through the (then) upcoming election. As John Gerring points out, within-case evidence is important not just for its abundance, but for its usefulness for the project.134 While I might be able to obtain some interviews in other similar cases, I have reason to believe the quality of the evidence that I was able to obtain in Northern Ireland is high. Shadow Case Selection In addition to the in-depth examination of Northern Ireland, I employ Kenya as a shadow case. It is difficult to find a definition of ‘shadow cases’ in the political science literature but is generally seen as a case that is used in a more informal fashion to compare to the primary cases.135 Many scholars use shadow cases throughout their case studies to provide brief evidence of similar or opposing dynamics to the case under examination. For example, in Theda Skocpol’s seminal work on social revolutions, she 134 Gerring, 40. 135 Ibid., 96. 52 uses shadow cases to highlight the special case of successful social revolutions and how they differ from other similar cases of failed social revolutions. Specifically, she discusses how Prussia faced a societal political conflict, but adapted to the pressures rather than revolting against them.136 She uses this shadow case to identify and separate the specific features of the set of cases her theory applies to and where it might fail in its explanatory power. She also argues her theory applies to states that are (and have always been) independent of colonial rule and so do not apply in other cases of revolution such as Mexico, Bolivia, and Cuba.137 The use of shadow cases provides important insights in many cases studies, even though they are rarely specifically acknowledged for their importance.138 Like other scholars, I make reference to several ‘shadow cases’ throughout this project. However, I also deliberately explore one such shadow case in more detail – Kenya. If, as Gerring and Lee Cojocaru (2016) suggest, shadow cases are only useful to “provide brief points of comparison for the case(s) of primary interest,”139 then Kenya falls somewhere between a ‘shadow case’ and a direct comparison. However, for the purposes of this paper, I call Kenya a shadow case to clarify that I do not intend to present it as a direct comparison but want to highlight for the reader the importance of a future in-depth comparison with such an unlikely candidate. 136 Skocpol. Theda, States and Social Revolutions: A Comparative Analysis of France, Russia, & China (Cambridge, UK: Cambridge University Press 1979), 110. 137 George and Bennett, 120; David Collier and James Mahoney, "Research Note - Insights and Pitfalls: Selection Bias in Qualitative Research," World Politics 49, no. 1 (1996): 81. 138 Hillel Soifer, "Shadow Cases in Comparative Politics," (Temple University, 2017). 139 John Gerring and Lee Cojocaru, "Selecting Cases for Intensive Analysis: A Diversity of Goals and Methods," Sociological Methods & Research 45, no. 3 (2016): 407. 53 I chose Kenya to fill this role for several reasons. At first glance, Kenya makes a good candidate for a ‘most different systems’ design, also referred to as Mill’s ‘method of agreement.’140 Northern Ireland and Kenya are two very different states that have suffered violent ethnic civil conflict. Northern Ireland is economically advanced with high levels of foreign investment, while Kenya is working to develop its economy. They have very different political, geographic, historical, and demographic conditions. Northern Ireland is a part of the United Kingdom, while Kenya is an independent former colony of Great Britain. Northern Ireland is located in the geographical north and is a part of Europe, while Kenya is located south of the equator in continental Africa. While both countries have diverse populations, Northern Ireland has two primary ethnic affiliations and Kenya has 23 officially recognized tribes with as many as 70 separately identifiable groups. The duration of conflict was also different, with Northern Ireland’s Troubles lasting approximately thirty years and Kenya’s 2008 election violence only several months. Additionally, these periods of violence ended more than ten years apart. Northern Ireland’s Belfast/Good Friday Agreement was signed in 1998 and Kenya’s violence ended in early 2009. Despite these differences, Northern Ireland and Kenya share two important similarities that make their comparison informative. First, both chose to implement democratically based power-sharing governments following their respective periods of violence. Second, both Northern Ireland and Kenya implemented policing reforms which were promotes as forms of community policing. These similar outcomes under the 140 John Stuart Mill, A System of Logic, Ratiocinative and Inductive, Being a Connected View of the Principle of Evidence and the Methods of Scientific Investigation, 10th ed. (London: Longmans, Green, and Co., 1879). 54 seemingly completely different conditions of Northern Ireland and Kenya made Kenya a great candidate for initial comparison. In addition to presenting as a good ‘most different systems’ case, Kenya provides an interesting comparison because Kenyans explicitly looked to Northern Ireland with respect to police reform to inform their own transition. Both states specifically called on experts in community policing to help with the evaluation of the police, and the many recommendations for change look remarkably similar in both cases. This not only highlights the importance of Northern Ireland and an exemplar case, but also allows for an examination of how the lessons of Northern Ireland have been translated to other societies in similar periods of transition. By examining how Kenya followed and/or veered away from the Northern Ireland example, we get a first look at the policy implications of the Northern Ireland experience for other post-ethnic conflict states. It is important to highlight, however, the infeasibility of a true comparative analysis at this time. The intensive fieldwork required to do a direct comparison with Northern Ireland was not possible for me in the timeframe of my dissertation project. However, a solid understanding of a second case helps to highlight where the specificity and generalizability of the Northern Ireland case separate. In other words, examining the Kenya case may also help to define the scope conditions of my argument. If Kenya was unsuccessful in implementing community policing given the explicit reference to Northern Ireland’s process in their own reforms, what conditions in Kenya may have made it more difficult for community policing to be implemented? A first look at the Kenyan case may help to identify these conditions, and therefore clarify the applicability of my argument. 55 Data Gerring argues, “if a case is to add to our knowledge of a subject it must provide new evidence – evidence that is presumably not available – or not easily available.”141 It is important, therefore, to discuss the nature of the new evidence provided in this study. Much of the evidence I provide in the later chapters of this project comes from historical texts, journalistic accounts, and publicly available survey data. However, I provide new and important pieces of evidence from original interviews conducted over the course of three years and three different trips to Northern Ireland. Interviews Interviews allow scholars an unequaled ability to “achieve a level of dynamism or fluidity of interaction with those whose behavior they hope to understand.”142 Interviews allow us to develop a deeper understanding of process, context, competing interpretations of events, and produce more complete accounts to explain and understand events.143 Interviews can be structured in many different ways, but there are two prevailing approaches that political scientists seem to take. More positivist scholars attempt to ask impartial questions looking for unbiased data that can use as evidence in an analysis. Interpretivist scholars, on the other hand, tend to emphasize the subjective nature of an interview and use the technique to identify possibly competing formulations of the 141 Gerring, 40. 142 Diana Kapiszewski, Lauren M. MacLean, and Benjamin Lelan Read, Field Research in Political Science: Practices and Principles (Cambridge, United Kingdom: Cambridge University Press, 2015), 190. 143 Robert Stuart Weiss, Learning from Strangers: The Art and Method of Qualitative Interview Studies (New York: Free Press; Maxwell Macmillan International, 1994), 9-11. 56 ‘truth.’144 Like many other scholars, my interviews fall somewhere between these two extremes. I conducted semi-structured interviews of police officers and civilians, asking a general set of questions aimed at understanding how attitudes toward the PSNI changed over the first ten years of its existence. In this way, while the guiding questions were always the same, I allowed the interviewees to guide me to what they felt was the most important information. Ontologically, I argue this is the best way to obtain the information I am seeking because I cannot assume that I will know a priori what the important data will be, but I can guide the conversation toward the topics I am most interested in unpacking. My sampling strategy was guided in part by the policies and procedures of conducting research with the PSNI. Namely, I was required to obtain permission from the Evidence Based Policing Group panel of the PSNI to conduct my interviews with officers. From there, the PSNI contacted several officers in charge of neighborhood policing districts throughout the two main cities in which I performed my research – Derry and Belfast. These officers were mostly very generous with their time and with assisting to set up additional interviews with officers under their command. This process could easily have lent itself to bias on behalf of the officers interviewed. Naturally, the PSNI would want to send me to interview the ‘best’ officers who are least likely to have negative views of their work or the PSNI in general. However, I found that I still managed to get a representative sample of officers by conducting the additional interviews with the officers under the command of those to whom I was originally sent. 144 Kapiszewski, MacLean, and Read, 191. 57 Of the 20 interviews I was able to conduct with the police, 9 were from Derry and 11 from Belfast, 14 were men and 6 women, 10 identified as PUL and 7 as CNR while 3 chose not to identify, and years of service in the police varied with 2 officers under 10 years of service, 9 between 10 and 19 years, and 9 with more than 20 years with the police in Northern Ireland. The police were asked a set of specific questions about if, how, and when they experienced changes in attitude from the public and what effect, if any, did specific events such as changes in police policy and important political moments had on their relationships with the community. In conducting interviews with community members, I was also constrained in my ability to use random sampling techniques for two reasons. First, Northern Ireland is a culture in which news largely travels through social interaction. Therefore, a fair number of the community interviews I was able to conduct were with people who had heard about the work I was doing from friends or connections in the community. The nature of this kind of communication often meant that I was asked to interview specific people rather than my asking them. Second, the sensitive nature of discussing the police and police reform also put subjects in danger of revealing information that could be considered criminal. Because of this I was careful to not specifically seek out individuals known or suspected to be criminals for fear their discussions with me could put them in legal or physical jeopardy. That said, I was able to conduct a small number of interviews with community members who were happy to share their thoughts with me. My questions for this group focused on the experiences of these individuals and their understanding of their communities’ attitudes toward the police over time. Again, I specifically asked questions about if, how, and when their attitudes toward the police changed over time. In 58 most cases, my questions led the interviewees to take the conversation in their own direction. In every case, I made sure all my base questions were answered, but allowed the interviewee to guide the conversation where they felt appropriate in each case. This semi-structured form of interviewing provided a lot of information that was comparable across interviewees, but also allowed for natural variation in experience and interpretation of events. This process provided the bulk of my new evidence. They also were the primary information source for the development of my arguments. Conclusion In this chapter I justified my focus on community policing in post ethnic conflict societies by identifying what I see as a hole in the literature on post-ethnic conflict governance. The current literature tends to focus on larger questions of providing security and whether that security should be provided by military forces or civilian police. Even those who have looked specifically at policing have primarily concentrated on what international organizations and donor countries can or should so to assist in building police forces. There is little in this literature that explores the difficult questions of how civilian police forces gain the confidence of the communities they serve after violent ethnic conflict. The literature on community policing, however, does address the difficult question on how to provide policing in ethnically divided societies. This dissertation attempts to bring these two literatures together to explore the how community policing might work to engender community confidence in the police in post-violent ethnic conflict societies. I will do this through an in-depth case study of community policing in Northern Ireland following the ‘Troubles’ and a second, brief case study of community policing in Kenya after their 2007 election violence. 59 The remainder of my dissertation is organized as follows. In Chapter II, I provide the reader with the minimal historical background necessary to understanding the conflict in Northern Ireland and the deep ethnic entrenchment associated with the organizing structures of society, and especially the police, in the years leading up to and including the Troubles. In this chapter, I also describe the level and prevalence of the violence experienced during the Troubles and where things stood in 2011 to provide a base understanding for how significantly things have changed since the end of the Troubles. In Chapter III, I introduce the Independent Commission on Policing for Northern Ireland and their report, which included 175 recommendations for reforming the police. In this chapter, I focus on how community policing was implemented in Northern Ireland through symbolic reform. This chapter highlights that community policing in post-ethnic conflict societies requires thinking broadly about the police and the many ways in which the police are presented in society. While reforming practice is vital, symbolism carries a lot of weight and requires attention to ensure people will give practical reforms a chance. In Chapter IV, I focus on how community policing was implemented in Northern Ireland through police practice. Specifically, I identify how the PSNI embraced the consultative aspect of community policing by continuously engaging the community in forming police policy and its implementation. I also identify how the changes in practice that resulted from this process helped to build community confidence in the police. Having thoroughly explored these insights in Chapters III and IV, Chapter V is where I perform an initial exploration of the case of Kenya and develop the additional criteria for the usefulness of the insights gleaned from the Northern Ireland case. Finally, in Chapter VI, I explore the 60 strengths and weaknesses of these insights and discuss other cases where they may be applied and refined in future research. 61 CHAPTER II ETHNICITY AND POLICING IN NORTHERN IRELAND Introduction Northern Ireland, like many other deeply divided societies, has a long history of ethnic entrenchment and accusations of biased policing. The historical make-up and association of the police with the state created specific issues that made reform of the police particularly important in the process of recovery from violent ethnic conflict. In this chapter I give a brief overview of the history of ethnic division in Northern Irish society and the police to provide the reader with a contextual understanding of where the police stood in regard to the community up to, during, and just after the Troubles. I give special attention to the level of violence during the Troubles, the role of the police during that time, and the state of community confidence in the police at the end of the Troubles. It is important to note that I do not intend, nor could I reasonably fit in this dissertation a complete history of Northern Ireland. Instead, my intention is to provide the reader with the minimal knowledge necessary to understanding the special role that the police play in Northern Ireland and the difficult task that was building community confidence in the police after the Troubles. Historical setting The development of sectarian tensions in Ireland can be traced back to the 12th century Norman invasion of the island that led to more than 800 years of English rule. Throughout this period, there was ever-present tension between the majority ethnically Irish population and the minority ethnically British settlers who had displaced the native Irish that manifest in numerous rebellions. Ireland has demonstrated a deeply entrenched 62 reluctance to accept the authority of the British state and a greater willingness historically to challenge that authority with force when necessary. Possibly due in part to this continuous state of violent rebellion, several attempts were made to enact Home Rule in Ireland in the late 19th and early 20th centuries. Home Rule in Ireland was intended to allow for the local Irish gentry (at that time primarily the descendants of Protestant, British transplants to the island) to form an independent Irish Parliament within the United Kingdom of Great Britain and Ireland. While many thought this a sufficient solution to the problem of English interference in Ireland’s affairs, many others insisted on total independence from Great Britain. In 1914, the UK Parliament passed the Third Home Rule Act, but suspended its implementation due to the onset of the First World War. This was a time of great political turmoil in Ireland with political alliances fracturing over the role of the Irish in the war effort and the morality of continuing to fight for independence from Britain during the war. During this time, a small group of rebels led the famous 1916 Easter Rising in Dublin. The rebels were soundly defeated by the British army and were later tried and executed for treason. However, the force and brutality used by the British forces to suppress the rebellion, as well as the executions backfired and led to an increased support for the rebels and their political agenda. In the general elections of December 1918, Sinn Féin, the political party associated with the rebels, won three-quarters of all parliamentary seats in Ireland and in January of 1919 declared Ireland’s independence from Great Britain. However, while Sinn Féin won the majority of seats, the Unionists had won a majority of seats in four of the nine northern counties of Ulster. When Sinn Féin declared an all independent Ireland, the Unionists in Ulster declared their continued allegiance to Britain. This led to the 63 outbreak of the Irish War of Independence that ended with the Anglo-Irish Treaty in December 1921. The Treaty codified the partitioning of Ireland with six of the counties of Ulster in the north becoming Northern Ireland and remaining a part of the United Kingdom and the 26 remaining counties becoming the Irish Free State, eventually the Republic of Ireland. The Anglo-Irish Treaty had several contentious points that caused a rift in the fighting force of the Irish Free State, the Irish Republican Army (IRA). First, the codification of the six northern counties as a separate entity from the rest of Ireland with the ability to opt-in to remain in the United Kingdom (which it did immediately) was an unacceptable compromise to many who insisted on a whole free Ireland. Secondly, the Treaty did not grant true and full independence in Ireland, but rather set up the Irish Free State as an autonomous dominion of the British Empire with the monarch as the official head of state, in the same vein as modern-day Canada and Australia. This arrangement would require members of the new Irish parliament to take a loyalty oath to the British Empire and would allow the British to maintain certain strategic ports in Ireland with the Royal Navy. These and other issues led to a split in the IRA, with the ‘republicans’ under the leadership of Michael Collins arguing that this was the best deal that the Irish could hope for and, while imperfect, would allow for the possibility of true independence in time. The ‘nationalists’ led by Éamon de Valera found these compromises unacceptable and wanted to continuing fighting until there was true independence and freedom for the whole island of Ireland as one nation. These disagreements led to the outbreak of the Irish Civil War in June 1922. 64 In addition to, and separate from, the Civil War, there was intercommunal violence and fighting in Northern Ireland and significant violence against the police in Belfast. The partition of the island was completed with the delineation of the border by a border commission and an amendment to the Anglo-Irish Treaty in 1925. The partition left a majority Unionist community, mostly Protestant and of British descent, who strongly preferred to remain a part of the United Kingdom with a significant Nationalist minority, mostly Catholic of Irish descent, (approximately 33.5%)145 who strongly desired a united Ireland inclusive of the six counties in Northern Ireland. While the legal status of Northern Ireland remains as a part of the United Kingdom, the Irish Constitution states that “the national territory consists of the whole island of Ireland”146 and reunification has remained a stated goal of the Republic of Ireland. The politics of 1925 through to the start of the Troubles in the 1960s were relatively stable, if deeply discriminatory. It is widely accepted, with a few notable objections, that the Unionist/Protestant majority systematically discriminated against Nationalists/Catholics.147 This discrimination came in the form of employment practices, political disenfranchisement, and gerrymandering. The Protestant majority controlled several of the most profitable industries in Northern Ireland and practiced employment 145 "Census - 1926 - Northern Ireland," (Belfast, United Kingdom: The Government of Northern Ireland`, 1926). 146 "Constitution of Ireland," (Dublin, Republic of Ireland: The Stationery Office, 2015). 147 Until just before his death in 2014, the Reverend Ian Paisley Sr., one of the most prominent opponents of the Catholic Civil Rights Movement, refused to acknowledge such discrimination or its extent. In the end he admitted that the political discrimination that prevented many Catholics from voting was indeed discrimination. Today, most in the unionist/Protestant community will readily admit that Catholic living conditions and political disenfranchisement in Northern Ireland had been historically terrible. (Rev. Ian Paisely, Sr., interview by Eamonn Mallie, 13 January, 2014.) 65 discrimination, keeping Catholics (even Unionist or non-politically affiliated Catholics) out of certain industries and/or unable to advance within those industries. Additionally, housing laws kept Catholics out of certain neighborhoods and unable to purchase their own property. Further, enfranchisement was limited to those who were property owners or primary tenants, a practice abolished in England in the 1940s, but which continued in Northern Ireland until 1969. Finally, for those few Catholics who were able to get around these major roadblocks, gerrymandering and the suspension of proportional representation in 1929 meant that Unionists were overrepresented in political office and Nationalists/Catholics were unable to use the political system to address any of the real or perceived prejudices they suffered. While these tensions were never far from the surface, Northern Ireland remained relatively peaceful in the interim between 1925 and the 1960s with a few bursts of relatively contained and short-lived violence. The Troubles A combination of the above-mentioned discrimination with the economic decline of the 1950s and 1960s and the influence of other civil rights movements around the world led to organized resistance by many Catholics in Northern Ireland in which they sought “more equitable access to political power, social provision and cultural recognition.”148 The civil rights movement in Northern Ireland was modeled on the nonviolent tactics used by the African American civil rights movement in the United States and the black rights movement in South Africa. In the early years of the 148 John Darby, "Northern Ireland: The Background to the Peace Process," Conflict Archive on the Internet (CAIN), Ulster University. 66 movement, leaders focused on non-violent resistance mostly in the form of letter writing and lobbying, followed later by parades and marches. Many assign the official start of the Troubles to the deployment of British soldiers to Northern Ireland to help restore order in Derry/Londonderry in August 1969 during the ‘Battle of the Bogside,’149 though this start date is disputed. While originally seen by many Catholics as an acknowledgement of the inadequate policing providing by the Royal Ulster Constabulary (RUC), the use of British soldiers to control the Catholic community came to be seen as a continuation of centuries of Irish oppression at the hands of the British. The leaders of the Provisional Irish Republican Army (PIRA) capitalized on this shift in public perceptions, co-opting the non-violent civil rights movement and capitalizing on its well-organized structure to turn the attention toward the violent struggle for reunification of Northern Ireland with the Republic as the only solution to the problems of Catholics in the north. Over the nearly thirty years of the Troubles, there were many notable incidents in which large numbers of individuals were killed or injured, or in which prominent figures were hurt or killed. The above-mentioned ‘Battle of the Bogside’ resulted in at least eight deaths and many hundreds of injuries.150 The violent events of ‘Bloody Sunday’ in January 1972, where the British Army opened fire on unarmed civilians protesting internment and led to the deaths of 14 and many more injuries, have been well- documented in history and remain a difficult subject in Northern Ireland nearly fifty years 149 Sean McMahon, Battles Fought on Irish Soil: A Complete Account (Dublin, Ireland: Londubh Books, 2010). 150 Martin Melaugh, "A Chronology of the Conflict - 1969," CAIN Web Service, http://cain.ulst.ac.uk/othelem/chron/ch69.htm. 67 later. In July of that same year, the PIRA escalated their ongoing bombing campaign, setting off 22 bombs outside shops in the center of Belfast, killing nine and injuring more than 100 on what has come to be known as ‘Bloody Friday.’151 While there are many more well publicized individual incidents of violence and clashes between the Catholic/Nationalist/Republican (CNR) community and the Protestant/Unionist/Loyalist (PUL) community and/or the police, it is more important for the purposes of this project to understand the day-to-day experiences of those living in Northern Ireland during the Troubles. In the remainder of this section, I will give a general overview of the state of the violence. In the next section, I will specifically look at the role of the police both in who they targeted with violence and who targeted them. From 1969 through to the signing of the Belfast/Good Friday Agreement (‘the Agreement’) in 1998, more than 3,400 people were killed specifically as a result of the conflict while many thousands more were injured. Some violence continued after the signing of the Agreement, however. Therefore, I include in my assessment of the violence the 3,623 deaths from 2771 individual incidents of violence from 1969 through 2001.152 For the number and type of incidents, as well as the number of injuries that did not result in death, I am relying on data provided by the Police Service of Northern Ireland (PSNI), and some data found on the Conflict Archive on the Internet (CAIN), a comprehensive collection of data and writings about the conflict in Northern Ireland. 151 "'Bloody Friday' - Summary of Main Events," CAIN Web Service, http://cain.ulst.ac.uk/events/bfriday/sum.htm. 152 Michael McKeown, "Post-Mortem: An Examination of the Patterns of Politically Associated Violence in Northern Ireland During the Years 1969-2001 as Reflected in the Fatality Figures for Those Years," (Belfast, United Kingdom2009; revised 2009), 6. 68 The levels of violence varied through the Troubles with the fewest deaths occurring in 1995 (nine deaths) and the most in 1972 (480 deaths).153 The number of deaths is itself horrific, but it is also important to understand the dynamics of who was killed and who was doing the killing. Over the course of this period, 1586 Catholics and 1299 Protestants who lived in Northern Ireland154 were killed.155 While the largest majority of fatalities were classified as ‘other non-combatants’ or people who did not fit into the many specific categories of non-combatants specified in the database (more than 1720), servicemen in the British Army and RUC had a significant amount of deaths (817), as did individuals affiliated with paramilitary groups both Republican and Loyalist (670).156 On the other hand, it was primarily the paramilitary groups responsible for those deaths with 2013 deaths attributed to Republican paramilitary groups and 1018 deaths attributed to Loyalist paramilitary groups. The security services, however, were also responsible for a significant number of deaths at 378.157 In addition to the many deaths reported, many thousands of people were injured in the many incidents of violence during the Troubles. While some data is missing from the early years of the conflict, particularly on civilian injuries, more than 46,000 people are 153 "Database of Deaths Associates with Violence in Northern Ireland 1969-2001," (2009). 154 The religious affiliation of non-Northern Irish residents was not documented in this database because, it was argued, religious affiliation was not relevant for non-residents. 155 McKeown, "Post-Mortem: An Examination of the Patterns of Politically Associated Violence in Northern Ireland During the Years 1969-2001 as Reflected in the Fatality Figures for Those Years," 19. 156 Ibid., 18. 157 Ibid., 14. 69 known to have been injured as a result of the conflict.158 Considering the number of well- known violent clashes that occurred in the first few years of the conflict where civilian injury data is missing, we can comfortably assume that this number underestimates the total of injuries by at least several hundred. These deaths and injuries were the result of more than 36,000 shootings, 15,925 bombs, and 2,200 incendiary devices between 1969 and 2001.159 It is also important to note that the daily experiences of violence in Northern Ireland varied largely based on the socio-economic status of individuals. While anyone could be affected by a random bomb placed in or near a shop in a city center, the experiences of those who were wealthier and lived in nicer neighborhoods tended to be very separate and far less violent than those who were working class and lived in working class neighborhoods.160 For example, one civilian I interviewed who was born in the late 1970s and therefore was raised entirely during the Troubles explained that, as his family was relatively wealthy and lived in an upper-class Protestant neighborhood, he had very little memory of being affected by the Troubles or the resulting violence.161 Conversely, another citizen of about the same age stated that his childhood had been entirely shaped by the violence of the Troubles because he had grown up in a working-class Catholic neighborhood in the town of Portadown, where sectarian violence was an everyday 158 CAIN, "Persons Injured (Number) Due to the Security Situation Sin Northern Ireland (Only), 1969 to 2003," ed. Police Service Northern Ireland (Belfast2003). 159 Police Service Northern Ireland, "Police Recorded Security Situation Statistics Northern Ireland: Security Related Incidents, 1969-2017," (Belfast2017). 160 Marie-Therese Fay, Mike Morrissey, and Marie Smyth, Northern Ireland's Troubles: The Human Cost (Sterling, VA: Pluto Press in association with the Cost of the Troubles Study, 1999). 161 Belfast Civilian 9, interview by Shawna Meechan, 3 September, 2015, Belfast. 70 occurrence.162 These varied experiences were echoed in many of the interviews I conducted. Those who were relatively well off, were less affected by the Troubles and tend to have fewer specific memories of violence or interactions with the police during the Troubles while those who were poorer or working-class had many and visceral memories of their experiences. The number of deaths attributed to the Troubles also clearly demonstrate this extreme difference in experience based on physical location. Belfast, where the significant majority of deaths in Northern Ireland occurred, can be broadly divided into four areas. In North and West Belfast there were a combined 1282 deaths between 1969 and 2001, where the more affluent neighborhood of South Belfast only had 189 total deaths and East Belfast counted only 139 deaths. Importantly, what these disparities tell us is that different neighborhoods have different experiences with and needs from the police, even in a geographically small country like Northern Ireland. Policing in Northern Ireland before and during the Troubles It is important to understand the history of the relationship between the police and the community they serve in Northern Ireland to fully understand why any changes in that relationship would have a profound effect on the success or failure of post-conflict rebuilding. The legal partitioning of the island effectively disbanded the previous police force in Ireland and created two separate forces, the Garda Síochána in the south and the RUC in the north. The initial force of the RUC, which formally came into being in June of 1922, was significantly different from the other police forces in the United Kingdom. The RUC was “insulated from British traditions of minimum fore, ‘policing by consent’ of the 162 Belfast Civilian 10, interview by Shawna Meechan, 3 September, 2015. 71 public, and political independence.”163 The new forced was armed, large (having far more police per capita than any other part of the United Kingdom), and notoriously biased against the Catholic community. While Catholic/Nationalist opinions of the police varied over time from supportive, to full on contempt, extreme Republicans always refused to accept the RUC as legitimate.164 One of the many complaints the CNR community had about the RUC was the drastic underrepresentation of Catholics on the force. In 1922, the force consisted of approximately 2,000 men, of which over 500 were Roman Catholic by 1925. While this is a significant portion, it was still less than the proportion of Roman Catholics living in Northern Ireland at the time and was a forbearer of the continued problems the RUC would have with recruiting from that community.165 This problem was exacerbated by the fact that Republican leaders discouraged young Catholic men from entering the force and the IRA targeted those who disregarded these urgings.166 This disparity in representation within the police remained an issue throughout the history of the RUC despite numerous reports throughout the years that indicated that the RUC needed more officers that were Catholic. Many of these reports implied, or in some cases outright stated, that the Catholic community itself was to blame for this lack of representation due 163 Weitzer, 32. 164 Ibid., 34-35. 165 Rea and Masefield, 8-9. 166 Weitzer, 35. 72 to the extreme social pressure within that community to stay away from the policing profession.167 Undoubtedly, this was a significant deterrent to young Catholic officers joining the force, but it should also be noted that the culture and make-up of the force was unwelcoming to the minority Catholic community. For example, in 1923, a special Orange Lodge (a branch of the Masonic-style brotherhood committed to promoting Protestant dominance in all areas) was formed for constables in the RUC and the Home Affairs minister attended the first meeting.168 This was a significant break in tradition from the police in Ireland in the 19th Century, which had prohibited police membership in the Orange Order or the Catholic Association.169 The RUC was also supported by the Ulster Special Constabulary (also referred to as the “B-Specials), a part-time armed corps comprised almost exclusively of Ulster Protestants. This open embracement of the concept of Protestant supremacism served as evidence for the CNR community that the RUC did not intend to serve them and was also a deterrent to joining the force. The events of the Troubles created even more animosity between the CNR community and the RUC. Within a few months after the ‘Battle of the Bogside’ and the introduction of the British Army on the streets of Northern Ireland to assist the police, the then head of the RUC resigned and was replaced by the former Commissioner of the City of London Police, Sir Arthur Young. Within a year of this change, the size of the RUC 167 Rea and Masefield, 12. 168 Weitzer, 45. 169 Stanley H. Palmer, Police and Protest in England and Ireland, 1780-1850 (Cambridge: Cambridge University Press, 1988), 274. 73 had also changed drastically, increasing to a force of 4,000 regular officers and a new reserve of 1,500.170 In 1976, the RUC re-established the policy of police as primary enforcers of the law with the British Army only present for additional support and back up. In practice this meant the police took the lead role in all patrols with army personnel coming in behind, rather than taking the lead, for example. This also meant, however, that the RUC were regularly outfitted with weapons and re-enforced vehicles that appeared very militaristic.171 So while the goal may have been to reduce the need for military assistance in policing, the result was that the RUC themselves became more militarized and arguably drove a bigger wedge between the police and the CNR community. The Battle of the Bogside and many of the other famous incidents of violence between Catholic and Protestant communities also demonstrate what the CNR community saw as unfair and biased policing practices. During the early civil rights marches of the late 1960s, the RUC was often unresponsive to requests from the marchers for police protection. The RUC also often refused to interfere when Loyalist counter-demonstrators became violent toward the protestors, and were even accused of inciting violence against demonstrators themselves.172 The many incidents of clashes between Republican and Loyalist demonstrators in the late 1960s along with the perceived bias of the police, which varied from indifference to accusations of collusion 170 Rea and Masefield, 10. 171 Ibid., 12. 172 Weitzer, 59-60. 74 with Loyalists, led to a deep deterioration in the relationship between the RUC and the CNR community.173 The wedge between the RUC and the CNR community was mirrored by the sense of pride and ownership in the RUC felt by the PUL community. Many in this community saw the RUC as ‘our police,’ and many working-class Protestants saw a career in the RUC as both honorable and a rite of passage for young Protestant men. Many in the PUL community praised the RUC and blamed the need to resort to violence to control riots on a lack of officers, as opposed to any bias in the police.174 The direct contrast in how the two communities felt about the police made it extremely difficult to even discuss the reforms that some Nationalists insisted were necessary for the RUC to build a better relationship with the CNR community. An investigation in 1969 led to the production of a document deeply critical of the RUC and its structure claiming that “any police force, military in appearance and equipment, is less acceptable to minority and moderate opinion than if it is clearly civilian in character.”175 The report, known as the Hunt Report, made a series of recommendations for reforming the RUC including demilitarizing the police and working to improve community relations. While some of these recommendations were adopted in theory, little changed in the execution of policing in practice except that the force doubled in size thanks to the creation of a new 173 Ibid., 61. 174 Cameron Commission, "Disturbances in Northern Ireland," ed. Chairman Lord Cameron (Belfast: HMSO, 1969); Scarman Tribunal, "Violence and Civil Disturbances in Northern Ireland in 1969," ed. Chairman Justice Scarman (Belfast: HMSO, 1972). 175 Hunt Committee, "Report of the Advisory Committee on Police in Northern Ireland," ed. Chairman Lord Hunt (Belfast: HMSO, 1969), 21. 75 reserve force.176 Importantly, much of the PUL community saw any changes to the RUC as a rebuke of the police, decreasing the morale of officers and jeopardizing the security of the PUL community simply to appease the CNR community.177 The reaction to these suggested changes toward the beginning of the Troubles would be the precursor of reactions to come in the reform of the RUC after the Troubles. Despite the differing opinions between the PUL and CNR communities about why it was so, there was a clear pattern, especially at the beginning of the Troubles, of police and security forces using more deadly force against Catholics. In the first documented year of the conflict, 1969, of the 10 Catholics killed, seven of them were killed by the RUC whereas of the seven Protestants killed, only two were killed by the British Army and none by the RUC.178 This trend continued into the early 1970s where substantially more Catholics were killed than Protestants, and the majority of Catholics killed were victims of government security forces (usually the British Army or the RUC). By 2001, the security forces had together killed 317 Catholics compared to 45 Protestants.179 Not surprisingly, then, in 1970 Republican paramilitary groups (primarily consisting of the PIRA, but others as well), developed a strategy for targeted violence. They chose to specifically target violence against members of the security forces and economic centers seen to support the oppressive policies of the state.180 176 Weitzer, 64. 177 Ibid., 65. 178 McKeown, "Database of Deaths Associates with Violence in Northern Ireland 1969-2001." 179 "Post-Mortem: An Examination of the Patterns of Politically Associated Violence in Northern Ireland During the Years 1969-2001 as Reflected in the Fatality Figures for Those Years," 26. 180 Ibid., 7. 76 Peace Agreement The official end date of the Troubles is disputed, but for the most part it is agreed that the signing of the Belfast/Good Friday Agreement (‘Agreement’)181 in 1998 constitutes the official end despite periodic acts of violence that have continued ever since.182 The Agreement was divided into three ‘strands’ dealing with different aspects of the conflict. Strand One addressed the democratic institutions of Northern Ireland. Within this Strand some of the subsections addressed the issues of rights safeguards, equality of opportunity, decommissioning, security, policing and justice, and prisoners in addition to specifically addressing issues with the assembly and political representation. Strand Two discussed the North/South Ministerial Council and addressed the issues of necessary cooperation between the institutions of Northern Ireland and the Republic of Ireland. Finally, Strand Three addressed the British-Irish Council and the British-Irish Intergovernmental Conference to institutionalize continued conversation and cooperation between the British and Irish governments over the future of Northern Ireland. In accordance with the provisions in Strand One, the Independent Commission on Policing for Northern Ireland (‘Commission’) was formed with Conservative politician Chris Patten as its head. The Commission was to “inquire into policing in Northern Ireland, consult widely, and make proposals for future policing structures and 181 The name of the Agreement is also the source of some debate between the PUL and CNR communities. Generally, members of the PUL community prefer the name ‘Belfast Agreement’ as this is keeping with the tradition of naming peace accords for the city in which they are signed. Members of the CNR community tend to use ‘Good Friday Agreement’ as the deal was settled on the Catholic holy day of Good Friday 1998. In an attempt to maintain as much neutrality as possible and to show respect for the valid arguments on both sides, I opt not to take sides and call it the ‘Belfast/Good Friday Agreement’ in full, or the ‘Agreement’, in this dissertation. 182 Arthur Aughey, The Politics of Northern Ireland: Beyond the Belfast Agreement (Abingdon: Routledge, 2005). 77 arrangements, including the police force composition, recruitment, training, culture, ethos and symbols.”183 The proposals made by the Commission aimed “to create a police service that would be effective, operate in partnership with the community, cooperate with the Garda Síochána” – the police force of the Republic of Ireland – “and other police forces, and be accountable both to the law and the community which it was to serve.”184 On September 9, 1999, the Commission released its official report (usually called the ‘Patten Report’) and made several important recommendations. These recommendations included but were not limited to renaming the RUC the Police Service of Northern Ireland (PSNI), creating a new policing board and district policing boards to ensure community input and police accountability, creating a Police Ombudsman and Complaints Tribunal, changing the symbols of the police force, instituting a 50/50 recruitment policy for new officers from the Catholic and Protestant communities, instituting a new code of ethics and oath of office which would include a strong emphasis on human rights, and an emphasis on community policing, among many others.185 Most, though not all, of the recommendations of the Patten Report were adopted and the PSNI officially replaced the RUC on 3 November 2001. Results The specifics of the process of implementation of the Patten Report’s many recommendations will be addressed in detail in the next two chapters. Before exploring 183 "Belfast (Good Friday) Agreement." Annex A. 184 The Independent Commission on Policing for Northern Ireland (Great Britain) and Chris Patten, "A New Beginning: Policing in Northern Ireland: The Report of the Independent Commission on Policing for Northern Ireland," (Norwich, England1999). 185 Ibid. 78 how this was done and why implementation succeeded or failed, it is worth briefly exploring the ‘outcome’ that has made Northern Ireland a model for other post-conflict societies. Specifically, Northern Ireland has been successful at building community confidence in the new PSNI according to survey data that has been regularly collected for many years. The data from these surveys has been used as evidence of the success of policing reform in Northern Ireland and, therefore deserves some attention here. In 2001, the Northern Ireland Policing Board was established as a result of the recommendations of the Patten Report. As a part of their statutory duties, the Policing Board began to collect information on, among other things, public attitudes toward the police and specific police institutions. The Northern Ireland Statistics and Research Agency conducts several nationwide omnibus surveys annually to collect data for different clients including the Policing Board and provides it to the Policing Board to collate and distribute. As a result, we have consistent data starting from October of 2001, six months after the PSNI officially replaced the RUC, on public attitudes toward the police broken down by age, sex, and religious affiliation. Prior to 2001, the Northern Ireland Life and Times Survey provides similar though slightly different data on public attitudes toward the police that is helpful in providing a baseline for how attitudes toward the police varied among the communities of Northern Ireland at the end of the Troubles. According to the 1998 Northern Ireland life and Times Survey, there was a significant difference in how Catholics and Protestants in Northern Ireland felt about the police. In response to the question, “In general, how satisfied or dissatisfied are you with the way police in Northern Ireland do their job?” the responses broke down as shown in Table 1. 79 Table 1. Community Attitudes toward the RUC, 1998 186 While the responses to this question show that most Catholics were at least indifferent, and at most deeply dissatisfied with the RUC in 1998, the results for a different question provide an even starker picture of the state of policing at the end of the Troubles. Illustrated in Table 2, the same survey also asked individual’s opinions about what should happen to the RUC. Specifically, the survey asked, “There has been a lot of discussion recently about reforming the RUC. Do you yourself think that the RUC should: Be disbanded, Be reformed but not disbanded, Should stay just as it is now, Don’t know, or Refuse/unwilling to answer.” This data shows the deep community divide over the state of the police in Northern Ireland. More than 80 percent of Catholics thought the RUC need to be disbanded or reformed while 60 percent of Protestants wanted the RUC to remain unchanged. 186 ARK, "Northern Ireland Life and Times Survey, 1998," ed. ARK (www.ark.ac.uk/nilt, 1999). 80 Table 2. Community Attitudes on RUC Reform, 1998 187 With the implementation of most of the recommendations of the Patten Report, those in favor of disbandment and reform won the debate. Beginning with the first Omnibus Survey after the implementation of the new PSNI, the questions became slightly more specific and targeted to get an understanding of how the communities of Northern Ireland felt about the PSNI as a whole, the specific officers working in their neighborhoods, and certain institutions of the PSNI. Again, later chapters will address some of the specific changes in attitude over time, but it is worth looking at the general attitude toward the police at three points in time: October 2001, January 2012, and April 2017. I include this last date to give a brief preview of how policing in Northern Ireland has gone in the first five years after the official end of this study in 2011, a point I will expand on in the conclusion of this dissertation. Table 3 give the responses to the question, “Do you think that the Police Service of Northern Ireland does a good job or a poor job in Northern Ireland as a whole?” for the surveys conducted in October 2001, January 2012, and April 2017. This table shows us a 187 Ibid. 81 significant shift in the Catholic community perceptions of the police in the first ten years of the service. We can see that Catholics who rate the service as very/fairly good jumped from 52% in 2001 to 63% in 2012. Maybe even more interesting is the drop in Catholics who rated the police very/fairly poor from 25% in 2001 to 12% in 2012. Yet, while we see this large shift in the Catholic community’s perception of the police, we do not see a corresponding negative shift in the Protestant community’s perceptions of the police as one might expect given the history described above. It is also worth a brief mention that in the five years after the initial investigation period, both Catholic and Protestant community perceptions of the police increaed dramatically taking the overall approval rating of the PSNI from a fairly consistent 69% in the first ten years to a striking 78% in its fifteenth year. I will explore a few possible reasons for this additional shift in positive perceptions in the conclusion chapter. Table 3. Community Confidence in the Police, longitudinal 188 Northern Ireland Statistics and Research Agency and Northern Ireland Policing Board, "April 2003 Omnibus Survey: Northern Ireland Policing Board Module," (Belfast2003). 189 "Public Perceptions of the Police, Dpps, and the Northern Ireland Policing Board January 2012 Omnibus Survey," (Belfast2012). 190 "Public Perceptions of the Police, Pcsps, and the Northern Ireland Policing Board," (Belfast2017). Rating (%) Survey Year (Month) 2001 (Oct)188 2012 (Jan)189 2017 (Apr)190 C P Tot C P Tot C P Tot Very/fairly good 52 72 69 63 73 69 76 84 78 Neither good nor bad 20 10 14 23 16 19 15 11 14 Very/fairly poor 25 7 15 12 11 11 9 4 6 No answer/DK/refusal 2 1 2 1 0 1 0 1 1 82 Conclusion This chapter has provided a brief overview of a small number of important moments in Northern Irish history that contributed to the period of extreme violence this dissertation is examining. It is important to note that for the sake of space and clarity, I have not included many other important events that some may argue contributed just as much, if not more to the ending of the Troubles and the ultimate focus on police reform. Specifically, I did not make reference here to the St. Andrew’s Agreement which not only identified policing reform as important for peace in Northern Ireland, but also led to the production of a report on policing in Northern Ireland that was referenced by the Independent Commission on Policing Reform when they performed their analysis of policing in Northern Ireland. I chose to keep the historical background in this chapter minimal for several reasons. First, there is simply too much information that could be included, and any selection of some events will inevitably be criticized by some who think I have left out an important event. For this reason, I have included the least historical background information possible to make this dissertation comprehensible to those with little or no previous knowledge of the Northern Ireland situation without including every single event that affected the outcome of the Troubles. Second, though St. Andrew’s absolutely shaped what the final Belfast/Good Friday Agreement would look like, including the focus on policing reform, it was not the final document that was used to guide the post- Troubles rebuilding process. It is interesting, but not absolutely necessary to know the history of St. Andrew’s to understand how the Agreement was implemented. Finally, this dissertation is focused on the process of police reform after the signing of the Agreement. 83 So, while including more background information would help to highlight how Northern Ireland came to focus on community policing, it does not tell us about how it was implemented and why it worked. This chapter has also demonstrated the integral role that the police force has played in Northern Ireland throughout their history and shown how the police have been directly linked to the legitimacy of the state in the eyes of its citizens. In a democratic state, the police function to provide security for all regardless of ethnic affiliation, de- escalate volatile situations, deter crime, and protect civil rights. When these duties are performed to the best of each officer’s abilities, citizens “are ready to identify and to support the police because the institution has both attained legitimacy in the eyes of the public and won its confidence.”191 However, when the police fail in these duties, they lose the confidence of the public and jeopardize the legitimacy of the state. When this happens, it is important to identify how to move forward in the aftermath of ethnic conflict to build community confidence in the police and therefore acceptance of the legitimacy of the state. As was discussed in the previous chapter, Northern Ireland did this through the implementation of community policing. In the following chapters, I will detail the specific ways in which Northern Ireland implemented community policing and why it was successful in building community confidence in the police. 191 Gimode, 228. 84 CHAPTER III IMPLEMENTING COMMUNITY POLICING: SYMBOLISM Introduction The conventional account of how Northern Ireland recovered from the Troubles following the signing of the Belfast/Good Friday Agreement (‘the Agreement’) goes something like the following. Politically, a parliament was established with a power- sharing executive with a First Minister from the majority ethno-political identity group and a Deputy First Minister from the dominant minority ethno-political group. In this arrangement, the executive cabinet positions would be distributed based on the proportionate share of votes each political party received and the political determination of the First and Deputy First Ministers as to which positions went to each party. In addition to these political arrangements, a commission was established to evaluate the role of the police in Northern Ireland and make suggestions for improving community confidence in the police. Finally, several initiatives aimed at community reconciliation and cross-community relationship building received state and external funding. The combination of these changes over time, common wisdom says, created a relatively stable and prosperous Northern Ireland. The common wisdom described here is not wrong, but it is incomplete. Like a Cliffs Notes version of a complicated novel, this story gives you the major plot points without the nuanced details of how events unfolded and all the ways in which each step of the process might have gone wrong, and sometimes did. This story does not capture the importance of specific changes made to policing that were a part of the larger process of political and social reform and were vital to the success of the political transition. The common wisdom laid out above misses the details of the process of rebuilding and leaves 85 out the fact that, even with these big ‘plot points’ things could have turned out differently if it were not for the specific focus on community policing that guided police reform. In this and the following chapter, I focus on how community policing was implemented in Northern Ireland and how the successes and failures of community policing translated in community confidence in the police. Specifically, I highlight the reforms enacted after the evaluation of the police conducted by the Independent Commission on Policing for Northern Ireland (the ‘Commission’) and the publication of their subsequent report (the ‘Patten Report’). The implementation of community policing in the context of the Patten Report indeed did produce the outcomes I argued in Chapter I of this dissertation would be present in the case of good community policing: symbolic change and changes in police practice. In this chapter, I will focus on the changes to police symbolism that were made and how the process and outcomes of those changes were consistent with the principles of community policing. As discussed in Chapter I of this dissertation, the principles of community policing lead us to several expectations regarding symbolism that we should see when community policing is implemented well. Specifically, community policing implies that the community will be engaged in the evaluation of police symbols to determine what, if any, change will be necessary to build community confidence in the police. We should also see the reduction of physical barriers between the police and the community including demilitarized uniforms, vehicles, and buildings. The implementation of community policing in ethnically divided societies also requires attention to the representativeness of the police. We would expect to see efforts to ensure that members of minority and marginalized communities are represented in the police. If all of this is 86 done well, then the theory of community policing predicts that we will see an increase in community confidence in the police over time. This chapter will discuss how symbolic changes to the police in response to the recommendations of the Patten Report were undertaken and how, though often left out of the explicit conversation about community policing, the process and outcome of symbolic changes in Northern Ireland engaged the principles of community policing as much as did changes to police practice. The Patten Report The Belfast/Good Friday Agreement acknowledged that a new beginning for the police was necessary if peace was to be maintained in Northern Ireland.192 Even this statement was a major concession for Unionists who felt the RUC was the only thing that kept the country from complete chaos and destruction during the Troubles. This was also a concession for Nationalists who felt that the RUC had been at best complicit and at worst a willing accomplice in the discrimination and violence the Catholic community experienced during and before the Troubles. However, the parties to the Agreement were able to compile a list of necessary attributes the police would need if they were to achieve widespread cross-community support. This list included, Structures and arrangements … such that the police service is professional, effective and efficient, fair and impartial, free from partisan control; accountable both under the law for its actions and to the community it serves; representative of the society it polices, and operates within a coherent and cooperative criminal justice system which conforms with human rights norms.193 192 Brice Dickson, "New Beginnings? Policing and Human Rights after the Conflict," ed. Michael Cox, Adrian Guelke, and Fiona Stephen (Manchester: Manchester University Press, 2006), 170. 193 "Belfast (Good Friday) Agreement." 87 Despite the specificity of these, and other, requirements from the Agreement, the participants delegated the job of developing the explicit recommendations that gave the means through which these goals were to be achieved to the commission that was to be created as a result of the Agreement – the Independent Commission on Policing for Northern Ireland. The Commission was set up shortly after the signing of the Agreement on 3 June 1998, was chaired by the Right Honorable Christopher Patten, and consisted of seven additional members with a variety of expertise in policing, knowledge of Northern Ireland, and academic understanding of important concepts such as policing with the community and administrative issues of policing.194 The members of the Commission, though small in number themselves, set the initial tone for the turn toward community policing in Northern Ireland. Dr. Gerald W. Lynch, for example, was president of John Jay College of Criminal Justice in New York where he created a course called ‘Human Dignity and the Police.’ This course, which was taught to police officers in more than 31 different countries in its first four years, focused on engaging the community and bettering relationships between the police and the communities they serve.195 Another member, Professor Clifford Shearing, is considered a pioneer in community policing, developing the idea of ‘nodal policing’ where the police work with many partners to ensure a community conception of policing as a joint activity between the police and the 194 Rea and Masefield, 29, 44. 195 Gerald W. Lynch, "Police Training in Human Dignity," in Policing in Centrral Europe: Comparing Firsthand Knowledge with Experience from the West, ed. Milan Pagon (Ljubljana, Slovenia: College of Police and Security Studies, 1996). 88 community.196 In addition to reaching out to and visiting other police departments throughout the world, the Commission also solicited and accepted public comments from anyone with an interest in the reform of the police in Northern Ireland and held open forums in each of the 26 District Council areas of Northern Ireland. In all, these meetings were attended by more than 10,000 people and the Commission accepted approximately 2,500 written submissions.197 The Commission engaged the principles of community policing from the beginning, not just by soliciting input and participation from the community in these first stages of problem identification and idea formation, but in its continued reference to what Professor Shearing called the ‘golden thread’ of their mission – the concept that policing came from not just the formal institution of the police, but from “the mobilization of a network of capacities and knowledges located within [a] variety of institutional nodes.”198 The final report produced by the Commission, commonly referred to as the Patten Report in reference to the esteemed chair, included 175 specific recommendations and had significant sections on human rights, accountability, and policing with the community. Naming the Problem As one government official described it, many people in the PUL community see everything that has happened since the passage of the Agreement entirely as a loss. According to this thought process, any gain for the CNR community must, by its very 196 Jennifer D. Wood, "Pioneers in Policing: Cliffor Shearing," Police Practice and Research 15, no. 6 (2014): 524. 197 Rea and Masefield, 29. 198 Clifford D Shearing, "A Nodal Conception of Governance: Thoughts on a Policing Commission," Policing and Society 11 (2001): 261. 89 nature, mean a loss for the PUL community.199 Nowhere was this attitude more apparent than in the changes made to the police in Northern Ireland. As previous chapters have laid out, the RUC was perceived to be a police force of and for the PUL community in Northern Ireland. That is to say that the large majority of police officers in the RUC were Protestant (upwards of 97% at times),200 members of the PUL community felt that the police were ‘their police’ and often expected the RUC to be a given career path for PUL boys.201 The CNR community, on the other hand, felt unfairly targeted and discriminated against by the RUC.202 This perception was amplified by the symbolism associated with the RUC. The name Royal Ulster Constabulary invoked the monarchy of Great Britain, emphasizing ties to Britain and the Crown that were at best irritants and at worst direct insults to the Irish-identifying CNR community. In addition to the name of the police, the uniform, emblem, and flag also carried many symbols that associated the RUC with Britain. The emblem of the RUC, as depicted in Figure 1 below, was composed of a crown on top of a harp. For many in the PUL community, this symbolized continuity and sacrifice. However, many in the CNR community viewed this symbol as a representation of the Crown asserting its authority over Ireland and created a barrier to both direct interactions with the police and recruitment to their ranks.203 199 Belfast Civilian 11, interview by Shawna Meechan, 18 October, 2016. 200 Jack Holland, "Ruc: The Politics of Policing " The Irish Echo (2011), https://www.irishecho.com/2011/02/ruc-the-politics-of-policing-2/. 201 Derry Civilian 2, interview by Shawna Meechan, 16 September, 2015; The Independent Commission on Policing for Northern Ireland (Great Britain) and Patten. 202 Weitzer, 84-87. 203 Rea and Masefield, 165. 90 Figure 1. RUC Emblem According to the Patten Report, the name, symbols, and uniform of the police became so politicized over time that the CNR community took “the position that the name is symbolic of a relationship between the police and unionism and the British state.”204 Further, the Patten Report stated, a “new beginning cannot be achieved unless the reality that part of the community feels unable to identify with the present name and symbols associated with the police is addressed.”205 Yet, even with a general understanding that a large minority of the community could not or would not ever be comfortable with the police under the name and symbols of the RUC, the specific recommendations206 to change these polarizing symbols were extremely difficult to implement. 204 The Independent Commission on Policing for Northern Ireland (Great Britain) and Patten. Paragraph 17.4. 205 Ibid. Paragraph 17.6. 206 Ibid. Paragraph 17.6-17.7. 91 The Patten Report also made recommendations to change the composition of the police, stating, “the parties to the Agreement believe it essential that the police service should be ‘representative of the society it polices.’”207 The Patten Report further elucidated the ways in which the RUC was not representative of the community. For example, approximately 8% of RUC officers were Roman Catholic while more than 40% of Northern Ireland identified as Roman Catholic.208 The Patten Report went on to state that “real community policing is impossible if the composition of the police service bears little relationship to the composition of the community as a whole.”209 It is important to acknowledge that while the Patten Report notes other underrepresented groups in the officer corps including women, ethnic minorities, gay and lesbian officers, and people from ‘working class’ backgrounds, the recommendations of the Report focused primarily on correcting the imbalance between the PUL and CNR officers as this problem was seen as ‘a case apart’210 that could and should be addressed through recruitment and composition goals. The goal of altering the composition of the police was not to create a system where “Catholics police Catholics and Protestants police Protestants,” but rather to create a system that was more “efficient and effective” – a result that required a sufficiently representative force that was able to garner strong ties between the police and the communities they serve.211 207 Ibid. Paragraph 14.1. 208 Ibid. Paragraph 14.1. 209 Ibid. Paragraph 14.2. 210 Ibid. Paragraph 14.9. 211 Ibid. Paragraph 14.2. 92 The last set of symbolic changes suggested in the Patten Report addressed the physical barriers that existed between the police and the community. The ‘security situation’ created by the Troubles had led to a militarization of the RUC in both action and appearance. For example, by the end of the Troubles it was standard policy for officers to wear ‘bomb jackets’ when performing public order policing to protect them from the real possibility of explosive devices being used against them. Figure 2 below identifies a list of weapons regularly employed against the police in public order situations according to the research conducted by the Commission and reported in the Patten Report. 212 Figure 2: Weapons Used Against the Police during the Troubles 212 Ibid. Paragraph 9.4. 93 Not surprisingly, bomb jackets were large and bulky and themselves put a large barrier between the police and he community. The police were also armed with guns, a situation unique to Northern Ireland in the United Kingdom at the time. Figure 3 below provides an example of how officers appeared in public order situations. Acknowledging that the ‘peaceful’ situation that was promoted by the Agreement had not yet been achieved at the time of the writing of the Patten Report, the Commission recommended that, as soon as was safe and practicable, the police should disarm.213 214 Figure 3. An armed officer polices a Loyalist parade Similarly, it became standard practice in some areas for the police to use armored land rovers like the one shown in Figure 4 not just for public order policing, but also for regular patrols. These vehicles look more like army tanks than police vans and reinforced 213 Ibid. Paragraph 8.19. 214 Colin Brennan, "Seventy One Police Officers Injured after Fourth Night of Violence in the North," Irish Mirror (2013), https://www.irishmirror.ie/news/irish-news/northern-ireland-riots-71-police-2056588. 94 the idea that the police were an extension of the military as opposed to a separate, civilian force. Therefore, the Patten Report recommended limiting the use of these vehicles to high danger policing situations and eventually moving the land rovers out of regular stations entirely to be stored for use in extremely dangerous situations only.215 216 Figure 4. PSNI Land Rover In addition to these mobile barriers, police stations were also reinforced and barricaded during the Troubles in response to the specific targeting of the police by (primarily republican) paramilitary groups.217 These buildings did not appear inviting to the community, but rather presented themselves as more like military barracks with some combination of high walls, barred or no windows, and barbed wire as can be seen in 215 The Independent Commission on Policing for Northern Ireland (Great Britain) and Patten. Paragraph 8.8 - 8.9. 216 News Highland, "Detectives Investigation Paramiliatry Style Shooting in Derry," Highland Radio (2017), http://www.highlandradio.com/2017/11/09/detectives-investigating-paramilitary-style-shooting-in- derry/. 217 Fay, Morrissey, and Smyth. 95 Figure 5 and Figure 6 below. The intention of these architectural changes was to keep the people inside the stations safe, but also resulted in keeping the community out. 218 Figure 5. Strand Road Station, Derry 219 Figure 6. Donegall Pass Station, Belfast 218 Garrett Hargan, "Almost 250 Detained at the Strand Road Police Station in the Past Year Had to Be Taken to Hospital, New Figures Reveal," Derry Now (2017), https://www.derrynow.com/news/almost- 250-people-detained-strand-road-police-station-past-year-taken-hospital-new-figures-reveal/178445. 219 Ainsworth. Paul, "Closure 'Could Damage Relations'," South Belfast News (2012), http://belfastmediagroup.com/closure-%E2%80%98could-damage-relations%E2%80%99/. 96 In the Patten Report, the Commission argued, “A new policing style based on community partnerships calls for a new style of police station, accessible and welcoming to the public.”220 While acknowledging that it would be foolish to assume that the police were no longer under threat of attack after the signing of the Agreement, the Commission argued that it was possible to provide safe and secure police stations that did not look like fortresses. Therefore, they recommended that new police buildings be built to look more like normal buildings. They should be approachable and visible from the street, though they should also have enhanced security features that could be activated if and when necessary.221 In addition, they recommended that as existing police stations were updated, they should prioritize making the stations more welcoming to visitors.222 From Patten Report to Implementation The many recommendations of the Commission presented in Patten Report fell into a variety of categories including human rights, police accountability, policing with the community, policing in a peaceful society, public order policing, management and personnel in the police, restructuring and reducing the police force, composition and recruitment of the police, training and education, culture and symbols, and cooperation with other police services. Predictably, the Patten Report was controversial in its findings and provoked a variety of responses. The British government generally supported the recommendations of the report and committed to working toward their implementation. The CNR community felt the recommendations did not go far enough in assessing the 220 The Independent Commission on Policing for Northern Ireland (Great Britain) and Patten. Paragraph 8.5. 221 Ibid. Paragraph 8.6. 222 Ibid. Paragraph 8.7. 97 actions of the RUC during and before the Troubles and that the Commission should have drawn explicit conclusions about the rights and wrongs committed by the RUC during that time. The PUL community, on the other hand, was disappointed and offended by many of the recommendations claiming that the Patten Report should have explicitly honored the 301 RUC members who had been killed during the Troubles and that it specifically dishonored this sacrifice by suggesting that the symbols and name of the police should be changed.223 Despite the strength and diversity of opinion about the Patten Report, the British government undertook to pass a law in Parliament that would implement the Patten Report’s recommendations for the police in Northern Ireland. The Police (Northern Ireland) Act 2000 was adopted on 23 November 2000.224 Among the Patten Report’s many recommendations that addressed symbolic or demographic issues, some were direct and explicit while others remained vague, to be worked out by others after the publication of the Patten Report. Some of the specific suggestions for changing the symbols associated with the police included changing the name from the Royal Ulster Constabulary to the Northern Ireland Police Service (eventually changed to Police Service of Northern Ireland so as not to be confused with the Northern Ireland Prison System), discontinuing the tradition of flying the British Union Jack flag over police buildings,225 and instituting a new recruitment system designed to increase the representativeness of the police.226 Some of the vague directions 223 Rea and Masefield, 46-47. 224 Police (Northern Ireland) Act 2000. 225 The Independent Commission on Policing for Northern Ireland (Great Britain) and Patten. Paragraph 17.6.-17.7. 226 Ibid. Paragraph 14.16-15.18. 98 that came without specific suggestions for implementation included adopting “a new badge and symbols which are entirely free from any association with either the British or Irish states” and adopting “a new, more practical style of uniform” for police officers.227 While the more direct recommendations surrounding the demographic composition of the police carried their own significant political hurdles to implementation, the suggestions to develop new symbols and uniforms required extensive consultation with all parts of the community on top of political difficulties. This required the active engagement of political actors from across the spectrum, many of whom had political incentives to refuse to participate in the process of redesigning the ‘new’ police service. Before discussing the process by which some of the important decisions were made around the symbols of the PSNI, it is worth digressing a little to acknowledge that these negotiations and changes were happening in the larger context of implementing the whole of the Agreement. While debates over if and how to implement the recommendations from the Patten Report were going on, the same people were also debating the progress of decommissioning the weapons of paramilitaries and setting up the devolved Northern Irish government as well as the cross-border political arrangements. All of this was happening while these participants were trying to maintain the support of their constituents and address the needs of the victims of the Troubles. It is not surprising in this context that the attempt to de-politicize the police was itself a political minefield where those involved had to navigate the many different views and deeply held beliefs of the communities in Northern Ireland. 227 Ibid. Paragraph 17.6. 99 One example which helps to illustrate this complicated dynamic can be found in the talks held between the British and Irish governments and three major political parties including the UUP (the more moderate major PUL representing party), the SDLP (the moderate primarily CNR representing party), and Sinn Féin (the extremist CNR party) in the summer of 2001. These talks, held over several months at Weston Park in England resulted in some agreement between the parties on the issues of “policing, normalisation, the stability of institutions and decommissioning.”228 While these talks were considered successful and produced some agreement on several fronts, the British and Irish governments were forced to accept that despite many hours of negotiations, Sinn Féin would not be able to endorse the new police service at that stage.229 Despite the disappointment surrounding Sinn Féin’s inability to commit to supporting the new policing arrangements, the Weston Park talks did result in the promise of political representatives from the SDLP on the new Policing Board (which will be discussed in greater detail below) as well as the encouragement for representatives of the CNR community to apply for independent positions on the board.230 The Weston Park talks illustrate an important point about the restructuring of the police not taking place in a political vacuum, but in the complicated and muddy terrain of trying to rebuild a functioning devolved government with positive relations with both Britain and Ireland. In attempting to highlight the role that the police played in helping to 228 "Weston Park Agreement," ed. Northern Ireland Office and Department of Foreign Affairs (Belfast, United Kingdom: Northern Ireland Office Department of Foreign Affairs,, 2001). 229 Rea and Masefield, 54-55. 230 "Weston Park Agreement." 100 build state stability after the Troubles, I never want to ignore the greater context in which police reform was occurring. Police reforms were developed and implemented simultaneously with other important changes to the political arrangements of Northern Ireland, which makes it entirely likely that these other events shaded the community confidence in the police in ways that are hard to separate from the police reforms themselves. It is also important, however, to acknowledge the reverse: that the importance of the police is woven through many of the other threads of the peace process and the success (or failure) at each step of the process of implementation of the new PSNI was a reflection of the broader process as a whole. Yet, unlike the implementation of other parts of the Agreement (specifically the political elements), the reform of policing managed to maintain a mostly positive trajectory throughout the transition from the Troubles to peace. This makes the exploration of the success of police reform in Northern Ireland unique among the other post-conflict projects. Making the Leap Despite the many obstacles and through incredibly hard political work, the first steps toward building the new police service were taken with the establishment of the Northern Ireland Policing Board (the ‘Board’) in 2001. This Board was composed of nine independent (non-political) members from the community and ten members nominated by political parties. The first composition of the board did not include any members from Sinn Féin as they had refused to formally endorse the new policing scheme at that point. The Northern Ireland Secretary of State selected the independent members of the board from applicants from across the community, deliberately chosen to be diverse in both experience and demographics. 101 Badges and Emblems The Board held its first meeting on 7 November 2001. Among many other duties, the Board was tasked with developing the new flag and emblem for the PSNI. The issue of the name change had been settled through legislation prior to the establishment of the Board, but the Police (Northern Ireland) Act 2000 left the regulation of the emblem and flag to the Secretary of State who was required to consult with the Board before making such regulations.231 In effect, this placed the onus on the Board to develop the new emblem and flag since they were, at the time, the most representative body with the authority to undertake this task. The process of developing the new PSNI emblem and flag got a jump-start before the Board was first assembled. Given that the members of the Board were announced in late September 2001 and the official title change from RUC to the PSNI was to occur in November, government Ministers had secretly commissioned preliminary submissions for the emblem redesign for the Board to be able to work with immediately in the hopes that a decision might be reached quickly.232 Some of the submissions were published by the BBC and are shown in Figure 7 below. 231 Police (Northern Ireland) Act 2000. 232 Rea and Masefield, 169-70. 102 233 Figure 7: PSNI emblem submissions Unfortunately, the intentional neutrality of all the proposed designs that had been developed before the Board assembled were met with reactions ranging from extreme indifference to strong criticism. Ian Paisley Jr., a political member of the Board representing the Democratic Unionist Party (DUP), argued in a BBC interview that the directive from the Patten Report that the new emblem refrain from any overtly Irish or British symbols resulted in a meaningless design to which no one in the community could relate.234 At the same time, Gerry Kelly of Sinn Féin complained that the government should have already dealt with this issue when it decided on the name change. At the time the proposed new emblems were released to the public in mid-November 2001, PSNI officers were serving in RUC uniforms and the new class of recruits had already been admitted, also without new uniforms. Kelly argued that this issue should have already 233 "Unionists Reject Police Badge Design," BBC News World Edition (2001), http://news.bbc.co.uk/2/hi/uk_news/northern_ireland/1664681.stm. 234 Ibid. 103 been dealt with and showed impatience for the implementation of the Patten Report recommendations in whole.235 The Board responded to these reactions by forming an ad hoc committee tasked with addressing the new emblem and flag as quickly as possible. Within two weeks of the formation of the ad hoc committee, they presented a new proposal to the full Board, which included a new emblem design that: Explores the notion of inclusiveness and parity with the simple stylistic representation of a variety of symbols that reflect our diversity, our aspirations, and our desire to mutually respect and protect difference through policing.236 The design, as shown in Figure 8 below, was that of a sunburst with a star in the middle, which provided space for six equally spaced, and equally sized symbols between the points of the star. These symbols included an olive branch, a harp, a torch, the scales of justice, a shamrock leaf, and a crown. The center of the star was filled in with the cross of St. Patrick.237 The emblem presented by the ad hoc committee technically violated the recommendation of the Patten Report to remove all symbols associated with either the British or Irish communities. However, it embraced the spirit of giving equal prominence to the cultural heritage of both the CNR and PUL communities in Northern Ireland while also emphasizing the ideas of peace and justice that the PSNI were meant to ensure. Many members of the Board said the design would not have been their first choice, but according the Board Chairman Desmond Rea, they all agreed that it was an acceptable 235 Ibid. 236 Rea and Masefield, 172. 237 Ibid., 172-73. 104 compromise and unanimously recommend the emblem to the Secretary of State.238 The progress on the emblem also accelerated progress on the new uniforms for the PSNI, which would be ready by the time the first group of PSNI recruits finished their training and entered the force in spring of 2002. 239 Figure 8: PSNI emblem as adopted Representation To address the issues of demographic composition in the police, the Patten Report made several recommendations with the explicit goal of reaching approximately 30% CNR representation in the police within 10 years, a balance which the Commission had been advised was necessary “to ensure that a minority does not find itself submerged within a majority organizational culture.”240 First, acknowledging the anticipated 238 Ibid., 173. 239 Police Service of Northern Ireland, "Www.Psni.Police.Uk," https://www.psni.police.uk/. 240 The Independent Commission on Policing for Northern Ireland (Great Britain) and Patten. Paragraph 14.9-14.10. 105 transition to a peaceful society in Northern Ireland, the Patten Report recommended a considerable reduction in the total number of officers in the new police force over the course of approximately ten years.241 This reduction in numbers was to coincide with an increase in representation from the CNR community. To forward these goals simultaneously, a voluntary early retirement scheme was suggested to allow those RUC officers who so wished the opportunity to leave the service with generous benefits, which would happen in concert with the institution of a 50% Protestant and 50% Catholic recruitment scheme for new police cadets (‘50:50 recruitment’).242 It is important to note that the burden of changing the culture of the police was not placed entirely on the shoulders of the PUL community, though it was interpreted that way by many in Northern Ireland at the time. The Patten Report included several recommendations that instructed the CNR community to lift the social sanctions around Catholics becoming police officers. For example, the Patten Report recommended that the Gaelic Athletic Association (GAA) repeal its Rule 21, which prohibited members of the police from being members of the Association.243 The GAA was an important social organization within the CNR community and exclusion from this group would amount to ostracism from the whole community. The Patten Report also called on community leaders to “remove all discouragements to members of their communities applying to join the police and make it a priority to encourage them to apply.”244 Since few social barriers 241 Ibid. Paragraph 13.9. 242 Ibid. Sections 13-14. 243 Ibid. Paragraph 15.2. 244 Ibid. Paragraph 15.2. 106 existed to joining the police in PUL communities, and many young PUL boys were groomed from an early age to assume they would join the RUC, these instructions were specifically for the CNR community, which had a history of intimidating any Catholic who joined the RUC. Despite these pointed instructions to the CNR community, the changes to the emblem and recruitment strategies in the new police service were far more visible and therefore garnered the bulk of the attention in the community during the transition from the RUC to the PSNI. Physical Barriers While much of the other symbolic change had deadlines for completion, the changes to the arms, uniforms, vehicles, and buildings of the police were slower for several reasons. First, the capital investment in changing buildings, vehicles, and uniforms for thousands of officers is high and money had to be prioritized. For example, changes and updates to police stations were addressed as money came available and physical updates were necessary. Similarly, new patrol cars were purchased as old cars needed replacing, but the PSNI did not junk perfectly well functioning vehicles. Understandably, these updates were slow and gradual over the years, yet there is certainly evidence that the changes that have been made have followed the recommendations of the Patten Report. The photograph below of police stations in Coleraine (Figure 9) and Belfast (Figure 10) are examples of the new, more approachable buildings in which the PSNI has invested. In these photos one can see the clear emblem of the PSNI with the name of the station. Additionally, both stations have pedestrian entrances easily visible from the street and neither has obvious signs of reinforcement such as barbed wire or barred windows. 107 245 Figure 9. Coleraine Police Station 246 Figure 10. Musgrave Police Station, Belfast 245 "Portrush: 15 Arrested over Drugs Offences," BBC News (2014), https://www.bbc.com/news/uk- northern-ireland-30616424. 246 108 In addition to the changes to the buildings, the PSNI also made moves to reduce the militarized nature of their public order uniforms as seen in Figure 11 below. Whenever possible, they wear their regular uniforms with ‘stab vests’ which are a less bulky protective vest that allow officers more freedom of movement and are also less imposing for the public. One officer said that the switch from bulletproof bomb vests to the stab vests was a step toward what he called “community standard policing.” He felt this signaled the demilitarization of the police to the community and the police alike.247 248 Figure 11. Police keep the peace at a Loyalist parade Community Reaction As discussed above, the PUL community had a strong affinity for and attachment to the RUC. Before and during the Troubles, the RUC was seen as a Protestant force for 247 Derry Officer 1, interview by Shawna Meechan, 28 September, 2016. 248 Agence France Presse, "N.Ireland Unionist Parades Pass Off Peacefully," (2014), https://www.yahoo.com/news/n-ireland-unionist-parades-pass-off-peacefully-003702341.html. 109 the Protestant people,249 but according to a PUL community leader, “it’s not a Protestant service anymore.”250 As a result of the Patten Report recommendations and the changes that were made to the service since the end of the Troubles, there developed an attitude within the PUL community that the police are no longer ‘their’ service, according to one prominent PUL community leader.251 This transition began with the changes to the name, emblem, flag, and uniforms of the police. A large and vocal portion of the PUL community argued that changing these symbols was a sign of disrespect to those RUC officers who had fought and lost their lives in the course of the Troubles.252 Others felt that, by changing the symbols associated with the RUC and downplaying the connection to Britain, the CNR community had ‘won’ and that they were being rewarded for years of violence against the police.253 Despite these strong objections, political leaders in the PUL community actively participated in the processes described above, and so had a stake in the new symbols that were adopted for the police. The Commission was deeply aware of the possibility that this would be the reaction of the PUL community and acknowledged as much in the Patten Report.254 However, they were also insistent that to move forward and transform the RUC into a police service that the whole community could support, their recommendations should be 249 Derry Civilian 1, interview by Shawna Meechan, 12 September, 2016. 250 Derry Civilian 2. 251 Ibid. 252 Jim Smyth, "Symbolic Power and Police Legitimacy: The Royal Ulster Constabulary," Crime Law and Social Change 38, no. 3 (2002). 253 Derry Civilian 2. 254 The Independent Commission on Policing for Northern Ireland (Great Britain) and Patten. Paragraph 17.7 - 17.8. 110 adopted. In other words, the primary target of these changes was not the PUL community, but the CNR community. As was explicitly stated in the Patten Report, the RUC was not representative of the community it served, and it was specifically under- representative of the CNR community. To some extent, then, the changes to the symbols of the police were not aimed at appeasing the PUL community, but at convincing the CNR community that the police service and policing in Northern Ireland under the PSNI was going to be different that it had been under the RUC. These symbolic changes were a first step in that direction. The question is, then, how successful these changes were in convincing the CNR community to give the PSNI a chance. How would we know? Many have highlighted survey data as evidence of increased openness to the police in the CNR community. As mentioned in Chapter 2, more than 80% of Catholics thought the RUC should be reformed or disbanded in 1998, and in October of 2001, shortly after the PSNI officially replaced the RUC, 52% of Catholics thought the PSNI was doing a fairly/very good job. Others have looked to politician and party leadership for evidence. Without doubt, Sinn Féin remained publicly skeptical, refusing to actively participate in the Policing Board or any of the political institutions around the police. However, the SDLP also represented part of the CNR community, though a much smaller portion than Sinn Féin, and did actively participate in policing reform. Others still have referenced community participation in the process in the police reform process. Members of the CNR community acting as individuals did participate in the process as members of the Policing Board and in the newly formed District Policing Partnerships (which will be discussed in greater detail in the next chapter). 111 Beyond participation on the Board and other community-based initiatives pursued by the PSNI, the CNR community showed signs of a new willingness to engage with the police that had not been present under the RUC. The Police (Northern Ireland) Act 1998 instructed the Police Authority to obtain views of the community on policing. This duty was passed onto the Police Board when it was formed. Unfortunately, the surveys that have been used to assess community views on the police are only available, then, from October of 2001. However, as was discussed in the previous chapter, some older surveys give us similar information that can be used to measure changes in community confidence in the police since the end of the Troubles. For example, the 1998 Northern Ireland Life and Times Survey asked, “In general, how satisfied or dissatisfied are you with the way the police in Northern Ireland do their job?” The disparity in the responses from Catholic and Protestant respondents were stark. While 77 percent of Protestant respondents were either ‘quite’ or ‘very’ satisfied, only 42 percent of Catholics agreed. Similarly, 28 percent of Catholics were ‘not very’ or ‘not at all’ satisfied while only 10 percent of Protestants agreed.255 In the April 2002 Northern Ireland Omnibus survey, conducted shortly after the official adoption of the new symbols and uniforms, the most similar question, “Do you think that the Police Service of Northern Ireland does a good job or a poor job in Northern Ireland as a whole?” was answered very differently. A similar 74 percent of Protestants thought the PSNI did a ‘very/fairly good’ job while 54 percent of Catholics agreed – a 12 percent jump from 1998 and a 2 percent jump from the same survey question sixth months earlier in October 2001. Similarly, while 10 percent 255 http://www.ark.ac.uk/nilt/1998/Crime_and_Fear_of_Crime/PCSAT.html 112 of Protestants found the PSNI to do a ‘very/fairly poor’ job, only 20 percent of Catholics agreed – an 8 percent drop from 1998 and a 5 percent drop from October 2001.256 While the results seen in these surveys cannot in whole be attributed to the specific success of symbolic changes, they are one indication that members of the CNR community reacted to the changes with more open attitudes and a willingness to give the PSNI a chance. There had not been time to implement in full all the changes in police practice that would come with the new PSNI before the October 2001 survey was conducted, but there was a clear shift in the CNR community’s attitudes toward the police. That the CNR community specifically saw the symbolic changes as important was reflected in the interviews I conducted with both community members and police officers. One officer pointed out that the new PSNI uniforms were less militarized than the RUC uniforms and that, he felt, made it more possible to interact with the community ‘on a human level.’ A he put it, we were able to de-militarize. We were able to do away with the army. We were able to do away with armored cars... People were using not bullet-proof vests, but they were wearing stab vests. And so, there was a, a reduction in … the armed aspect of it. More to, more community standard policing. And that was making real inroads, that people we saw … where the real focus was on about servicing the community. 257 His experience was that the community found him more approachable and that, even when a community member approached him in anger, they were able to have a conversation rather than a physical confrontation, which the more militarized RUC uniform seemed to encourage. 256 Northern Ireland Statistics and Research Agency and Northern Ireland Policing Board, "April 2003 Omnibus Survey: Northern Ireland Policing Board Module." 257 Derry Officer 1. 113 Similarly, a different officer stated that, even though he was personally against the name change, which he felt was disrespectful to those RUC officers who had come before him, it was clearly important. As a Catholic, he saw how the attitude among his own friends and family changed with the adoption of the PSNI name and that it was worth ‘swallowing the pill’ of bad feelings among the police to get the buy-in from the CNR community.258 One community leader who specifically works with current and former dissidents mentioned that the name change was vitally important for those members of the CNR community who had been arrested or harassed by the RUC.259 Another community leader who has worked with many former prisoners agreed that the symbolic changes, and specifically the reduction of explicitly British symbolism in the police, had created the opening she needed to be able to get those former prisoners to be open to the possibility of a new start with the PSNI.260 Interestingly, the issue of representation in the police service was probably the most frustrating for the police officers and community members with which I spoke. While each of them acknowledged that 50:50 recruitment has been successful in changing the face of the police service, almost all said something to the effect of ‘it could have been done better.’ For officers, there was a frustration with the idea that it was even necessary. Many officers did not see themselves as either part of the PUL or CNR community during the Troubles, but as a part of a separate community with its own 258 Belfast Officer 2, interview by Shawna Meechan, 3 November, 2016. 259 Belfast Civilian 5, interview by Shawna Meechan, 10 November, 2016. 260 Belfast Civilian 4. 114 identity and its own history: that of the police.261 This was echoed by one civilian I spoke with who is related to an officer who was active in the RUC during the Troubles who felt that she too had a stronger affiliation with other police families than she had with her religious or political community.262 That said, despite the feelings of the officers themselves, the community in general did and still does see real or perceived community affiliations of officers and those perceptions influence how open the community is to accept the police service as a whole. For example, one officer who did not want to record his interview discussed the role of representation in the force rather reluctantly. In his mind, it does not and should not matter how each individual officer identifies, but it is still important to have representation on the force. In further conversation it became clear that he resented that some people in the community treat him differently depending on which community they assume he identifies with, but he also understood the need for the community to see people ‘like themselves’ in the force. He acknowledged this by specifically pointing out how proud he was that the PSNI had significantly increased female representation in the service.263 Similarly, one community member I spoke with was particularly critical of the idea of ‘diversity for diversity’s sake’ but also forced himself to admit that the homogeneity of the RUC had been a barrier for him and other Catholics he knew to interacting with the police. So, while he was not personally a fan of 50:50 recruitment as 261 Belfast Officer 1, interview by Shawna Meechan, 19 October, 2016; Belfast Officer 8, interview by Shawna Meechan, 10 November, 2016; Derry Officer 8, interview by Shawna Meechan, 6 October, 2016. 262 Derry Civilian 3, interview by Shawna Meechan, 21 September, 2015. 263 Belfast Officer 8. 115 a policy, he agreed that the better representativeness of the PSNI was helpful for changing the attitudes of many in the CNR community toward the police.264 This point of view was reflected again and again with the officers and civilians I interviewed. Most officers, including those who readily admitted that they had been accepted into the police under and possibly as a result of the 50:50 recruitment scheme, felt that merit should be the only consideration for acceptance or promotion within the police.265 Members of the CNR community argued that 50:50 would not have been necessary if the RUC had made an effort all along to not be exclusionary and sought out qualified CNR cadets. Members of the PUL community thought 50:50 recruitment amounted to discrimination against qualified PUL applicants to the police. In all, almost no one I spoke with liked 50:50 recruitment as a concept, and yet all accepted that it had been successful in achieving the aim of 30 percent CNR representation within the PSNI in the goal of 10 years. As one government official said, the reality was that from his point of view, 50:50 recruitment “had a profound effect on how policing was received” within the communities of Northern Ireland.266 One additional point on how the community reacted to all the symbolic changes to the police instituted in 2001 is that the most extreme reactions may have only occurred in areas with greater diversity. Some of the officers were serving in more rural and more homogenous (usually favoring PUL) areas during this time. In one officer’s experience, he said the community he served barely registered the changes. He was serving in a 264 Belfast Civilian 2, interview by Shawna Meechan, 26 October, 2016. 265 Belfast Officer 3, interview by Shawna Meechan, 3 November, 2016. 266 Belfast Civilian 7, interview by Shawna Meechan, 25 October, 2016. 116 predominately PUL community in 2001 and mentioned that he did not notice any change in the attitudes of the community nor their relationships with the police.267 The same was true for another officer who was a civilian at the time of the change from RUC to PSNI and lived in a different, primarily PUL community outside of Derry city. He also said that the changes were not noticeable in his area and the community barely reacted.268 This isolation may account for the negligible change in the PUL community’s attitudes toward the police mentioned at the beginning of this section. According to census data from 2001, 66 percent of people in Northern Ireland lived in areas that were either 99 percent Protestant or 99 percent Catholic.269 It seems reasonable that members of the PUL community who lived primarily among other PUL and who did not regularly interact with many members of the CNR community could ignore what seemed like huge changes to the police because they did not feel the effects of those changes. It is also worth exploring why there was not a more negative reaction to these changes from the PUL community. In areas where the police were only called on to manage normal police duties, as opposed to those associated with security during the Troubles, it seems likely that the community members would have less emotional reactions to the police altogether. That is, these PUL community members likely had weaker ties to the police both during and after the Troubles than those in neighborhoods where they regularly called on the police for protection. Therefore, the majority of the 267 Belfast Officer 4, interview by Shawna Meechan, 3 November, 2016. 268 Derry Officer 6, interview by Shawna Meechan, 6 October, 2016. 269 Lee A. Smithey, Unionists, Loyalists, and Conflict Transformation in Northern Ireland (New York: Oxford University Press, 2011), 16. 117 PUL community living in nearly exclusively PUL neighborhoods had muted if any reaction at all to the symbolic changes that were so deeply important to the CNR community. This is not to say that some members of the PUL community were not deeply frustrated with these changes, and some did take them very personally. However, the weaker overall PUL reaction to the symbolic changes of the police make sense in this context. Continuing Baby Steps The story of the symbolic changes to the police in Northern Ireland, though highly concentrated in the early years following the Belfast/Good Friday Agreement, continued throughout the first ten years of the new PSNI. The slow de-militarization of the uniforms and equipment of the PSNI continually altered the relationships it had with the community. Additionally, and importantly, political changes had a large impact on the success of the PSNI breaking through to larger segments of the CNR community, those that had stubbornly refused to participate in the new policing process and those who had been too afraid to support it openly. As discussed in earlier, the de-militarization of police uniforms played a significant role in opening the space for dialogue, especially in CNR communities. This change was significant because, during the Troubles, the police and the military were seen as one entity. As one officer explained, I first came here in 1990, so I’ve experienced a different society, a different brand of policing, a different way of policing and it was purely focused on security. … Our stations were all mortar-proofed, the vehicles we drove about were armored, the patrol tactics we were using was probably preventing police officers being attacked. We were patrolling with military.270 270 Derry Officer 1. 118 Every visual representation of the police was associated with war. Even when the uniforms were initially changed to differentiate from the old militarized RUC, much of the armored equipment and gear remained. In some areas, the transition away from ballistic vests and armored land rovers was much slower, but in many areas, this began to relax as the years passed and good relationships developed between the police and the communities they served.271 As one officer put it, “Places where we couldn’t have taken armored cars in at a time without coming under attack. These guys are walking. They’re on bikes.”272 The importance of maintaining this separation from the military continued to influence the success of gaining the confidence of the community. One civilian in Belfast noted that, though his heavily CNR community has made huge strides in learning to trust the police, a police initiative in the neighborhood several years ago was given the title ‘Operation’ which immediately put up barriers for those who thought it sounded more like a military endeavor than a police initiative.273 The purpose of the exercise, he said, was lost to the immediate wall erected when members of his CNR community began to associate it with the military. The success of the continued changes to the uniform and equipment of the PSNI after the first big change in 2001 is difficult to measure through anything other than interviews. While much progress has been made, it has been scattered and uneven in its distribution. For example, I visited one police station in a part of Belfast that had some 271 Belfast Officer 2. 272 Derry Officer 1. 273 Belfast Civilian 3, interview by Shawna Meechan, 28 October, 2016. 119 success in following the Patten Report recommendation to “have the appearance of ordinary buildings”274 while a different station across town still had the appearance of a fortress with high walls and barbed wire. One station in Derry city had a parking lot full of armored vehicles that resemble what people in the United States think of as S.W.A.T. vehicles with protective grills and covered wheels and another station had what looked like normal sedans. The disparity in when certain communities experience the changes in police appearance means that it is difficult to capture the success or failure of these changes in anything like a national survey. However, it seems clear from the interviews I conducted that both the police and the communities they serve agree that the fewer physical barriers between police and civilians, the more likely it is that dialogue can replace confrontation. In addition to the slow trickle of de-militarized equipment and uniforms, a highly important event also contributed to a major shift in the CNR community’s attitude toward the PSNI. At its Ard Fheis (the term used for the annual political party conference where they decide their platform for the year) in January of 2007, Sinn Féin decided to back the PSNI and participate fully in policing in Northern Ireland. While this move was more than just symbolic in that Sinn Féin took positions on the Policing Board and actively promoted the PSNI within the CNR community, the symbolism was hugely important as well. As one civilian interviewee said, it was not just that Sinn Féin’s support for policing gave people permission to trust the police, but it also took away a barrier for those who wanted to participate with policing. They were no longer going against their own 274 The Independent Commission on Policing for Northern Ireland (Great Britain) and Patten. Paragraph 8.5. 120 community to do it.275 More than this, it gave people permission to have relationships with police officers that before had been taboo. One officer told this story about a Catholic friend of his. I had a friend who was raised in a really, really hard part of the city; a really hard line Nationalist, Republican community. His family were very, very Republican. And we had to keep our friendship a secret. You know, unfortunately he died, but at the same time, I know that after Sinn Féin agreed with the policing process his family were much more open towards me and much more welcoming towards me. They were very, very open.276 The police I interviewed with one notable exception all praised Sinn Féin signing onto policing as a major breakthrough in working with the community. One officer was told by multiple community members with whom he interacted for the first time “Sinn Féin has told me it’s okay to call the police now.”277 Another said that without a doubt Sinn Féin signing onto policing and the Policing Board made the biggest difference for his relationship with the community.278 One officer even commented that now police are judged by the community on their performance, and they only got this chance because of Sinn Féin signing on to policing.279 Only one officer I spoke with disagreed with the symbolic importance of this political move by Sinn Féin. He contended that the move was essentially superficial as, in his experience, Sinn Féin had already been supporting the police in practice for several years.280 However, the community members I spoke 275 Belfast Civilian 2. 276 Derry Officer 2. 277 Ibid. 278 Belfast Officer 4. 279 Belfast Officer 8. 280 Belfast Officer 1. 121 with, once again, were nearly universal in their assessment that Sinn Féin’s endorsement of the police made a huge difference in building community confidence in the police. This seems to be reflected in the survey data once again. As of October 2006, the last Omnibus survey conducted before the Sinn Féin endorsement, 51 percent of Catholics surveyed rated the overall performance of the PSNI as ‘very/fairly good’ while 21 percent rated it ‘very/fairly poor.’ Just six months later and after the endorsement, 61 percent rated the performance of the PSNI as ‘very/fairly good’ and only 13 percent as ‘very/fairly poor.’281 This significant change in attitude speaks to the weight of this move within the CNR community. The importance of Sinn Féin’s endorsement was mentioned repeatedly in interviews. One interviewee claimed Sinn Féin signing on to policing was the single most important step for the post-Troubles rebuilding process. According to her, up until Sinn Féin signed on, there was not and would not be full community support for the police. Sinn Féin’s endorsement also gave support to the accountability mechanisms that had been set up as a result of the Patten Report. Maybe most importantly, senior politicians who had been openly and loudly skeptical of the PSNI began to tell people to report crimes to the police.282 For one community advocate who works to build relationships between the police and the community, the move by Sinn Féin opened doors and was a huge step forward in community relations,283 and for another individual, his support for 281 Northern Ireland Statistics and Research Agency and Northern Ireland Policing Board, "April 2003 Omnibus Survey: Northern Ireland Policing Board Module." 282 Belfast Civilian 6, interview by Shawna Meechan, 15 November, 2016. 283 Belfast Civilian 5. 122 both the PSNI and Sinn Féin were linked. He chose to join the party and support the PSNI when Sinn Féin got on board.284 Conclusion The tightrope walk associated with the symbolic changes to the police in the first ten years of the PSNI has been unfortunately underplayed in the general story that is told about Northern Ireland’s post-Troubles rebuilding. The difficulty in creating the space for members of the CNR community to be open to a ‘new’ police while not completely alienating the PUL community has been under-appreciated for the impact that the success of these changes had. The Independent Commission on Policing for Northern Ireland payed special attention to the history and cultural diversity of Northern Ireland and recommended changes that were aimed at removing any barriers to community confidence in the police. The symbolic changes to the name, uniform, flag and emblem, and representation in the police were all difficult, contested, and ultimately successful. In the first ten years of the new Police Service of Northern Ireland, these changes resulted in an increase in officers from the CNR community from 8 percent of the service in 2001 to 30 percent of the service in 2011.285 Additionally, there was a considerable drop in the number of CNR identifying people who thought the PSNI was doing a ‘very/fairly’ poor job around the two times highlighted here when major symbolic changes were instituted – from 25 percent in October of 2001 to 20 percent in April 2002 after the adoption of the changes to the name, uniform, flag and emblem; and from 21 percent in October of 2006 284 Belfast Civilian 1, interview by Shawna Meechan, 25 October, 2016. 285 Deborah McAleese, "Psni Orders Investigation into Low Number Od Catholic Recruits," Belfast Telegraph (2016), https://www.belfasttelegraph.co.uk/news/northern-ireland/psni-orders-investigation-into- low-number-of-catholic-recruits-35023903.html. 123 down to 13 percent in April 2007 following Sinn Féin’s endorsement of policing.286 Importantly, this data shows that there was growing CNR confidence in the PSNI that happened without a corresponding fall in PUL support. This shows that, despite concerns at the outset of symbolic police reform, changes in community confidence in the police were not a ‘zero sum game.’ The symbolic rebranding that came with the implementation of community policing in Northern Ireland worked to increase CNR community confidence without alienating the PUL community. While these symbolic changes did not happen in a vacuum, but in concert with other important changes in Northern Ireland, this chapter has demonstrated the specific importance of the symbols of the police in either turning off portions of the community or opening up the space for creating new relationships community members that felt alienated from or persecuted by the police during the course of ethnic conflict. Had post- Troubles Northern Ireland not taken the steps to overhaul the police and their image within the CNR community, it is difficult to imagine that an un-altered RUC would have been successful in obtaining the confidence of the CNR community. Though many in both the police and the larger community remained skeptical of the changes, the general consensus I found was that the symbolic changes were important and necessary to make room for the changes of perception that followed. 286 Northern Ireland Statistics and Research Agency and Northern Ireland Policing Board, "Public Perceptions of the Police, Dpps, and the Northern Ireland Policing Board," (Belfast, United Kingdom2007). 124 CHAPTER IV IMPLEMENTING COMMUNITY POLICING: PRACTICE “Policies themselves probably wouldn’t be different; maybe how they’re interpreted, maybe.”287 It’s always a “wait and see” approach to deciding to trust. The implementation of any policy doesn’t engender trust, it’s watching and experiencing how these policies are executed.288 Introduction At the official launch of the new PSNI emblem on 26 March 2002, Assistant Chief Constable Alan McQuillan stated that, while the new uniforms and emblem that had been adopted for the PSNI would be the most visual change to the police, they had also been working for more than two years on the underlying structure and make-up of the police. These less visible changes intended to address police accountability and refocus on the role of the police in protecting and promoting human rights were just as important as the visible change in uniform, if not more so.289 As discussed in Chapter I of this dissertation, the principles of community policing lead us to a set of expectations of what changes in police practice we should see when community policing is implemented well. Specifically, community policing implies that the community will be engaged in all stages of police reform from problem identification to idea formation to implementation and review. We should also see some sort of training or reorientation of police officers toward the new concepts and principles of community policing. Additionally, the implementation of community policing should 287 Derry Officer 5, interview by Shawna Meechan, 6 October, 2016. 288 Belfast Civilian 3. 289 Rea and Masefield, 175. 125 lead to decentralized decision-making systems that give more autonomy to individual officers to adapt policing in certain neighborhoods to the needs of those communities. This should produce policing that is consistent in its overarching policy of working with the community, but with slight variations in how policing is practiced based on the differing needs of the community. Finally, if all of this is done well, then the theory of community policing also predicts that we will see an increase in community confidence in the police over time. In this chapter, I will focus on the changes in police policy and practice that were made in reaction to the Patten Report and how the process and outcomes of those changes reflected the principles of community policing. I will highlight the difficulties faced in this process and how ongoing community input and engagement in the process of reform led to the eventual positive outcomes seen in Northern Ireland. The Patten Report As outlined in the previous chapter, the Independent Commission on Policing for Northern Ireland was established after the signing of the Belfast/Good Friday Agreement to evaluate the RUC and develop a set of recommendations for the reform of the police in Northern Ireland. The Commission produced the Patten Report with 175 recommendations for police reform covering a wide range of topics. The last chapter focused on those recommendations that addressed what I have termed ‘symbolic change,’ or those changes that could be seen in the physical representation of the police from their ethnic representation, to their badges and uniforms, to the vehicles and equipment they use, and the physical buildings in which they work. 126 In addition to these changes, the Patten Report also made many recommendations about how policing should be delivered in Northern Ireland. Specifically, the Commission made explicit that the PSNI should be focused on delivering policing with an emphasis on human rights, accountability, and policing with the community. Proponents of community policing would argue that these are three dimensions of community policing as opposed to separate areas of focus. In other words, community policing done well requires a respect for human rights, accountability of the police, and a focus on community-police partnerships. The following sections break these recommendations up based on the three expectations of community policing outlined above. Retraining the Police In the significant section on human rights, the Commission stated, “We cannot emphasize too strongly that human rights are not an impediment to effective policing but, on the contrary, vital to its achievement.”290 Without accusing the RUC of failing to consider human rights in the course of policing, the Patten Report emphasized the need to make human rights to basis of the new PSNI and made seven recommendations specifically aimed at this goal. One suggestion was to replace the controversial oath sworn by officers of the RUC, which included swearing allegiance to the Queen, to a more neutral oath that explicitly referenced the importance of human rights in policing. The old oath was, I... swear by Almighty God that I will well and truly serve our Sovereign Lady the Queen in the office of (rank) without favour of affection, malice or ill-will; that I 290 The Independent Commission on Policing for Northern Ireland (Great Britain) and Patten. Paragraph 4.3. 127 will to the best of my power cause the peace to be kept and preserved and that I will prevent to the best of my power all offences against the same; and that, while I shall continue to hold the said office, I will faithfully, according to law, to the best of my skill and knowledge, discharge all the duties of the said office and all such duties as may be attached to said office by law and that I do not now belong to and that I will not, while I shall hold the said office, belong to any association, society, or confederacy formed for or engaged in any seditious purpose, or any purpose tending to disturb the public peace, or in any way disloyal to our Sovereign Lady the Queen and that I will not, while I shall hold the said office, engage or take part in the furthering of any such purpose, or take or administer, or assist or be present at or consent to the administering of, any oath or engagement binding myself or any other person to engage in any such purpose.291 The suggested text in the Patten Report read, I hereby do solemnly and sincerely and truly declare and affirm that I will faithfully discharge the duties of the office of constable, and that in so doing I will act with fairness, integrity, diligence and impartiality, uphold fundamental human rights and accord equal respect to all individuals and their traditions and beliefs.292 The Commission recommended this oath be taken by all new and existing officers to affirm that “the importance of human rights as the very purpose of policing … be instilled in every officer from the start.”293 The removal of the language around officer involvement in “any association, society, or confederacy formed for or engaged in any seditious purpose, or any purpose tending to disturb the public peace, or in any way disloyal to our Sovereign Lady the Queen” was also an important change. This language was problematic most obviously because many potential Catholic officers would be unwilling to pledge an oath of loyalty to the Queen. Removing this language was a 291 Emphasis mine. 292 The Independent Commission on Policing for Northern Ireland (Great Britain) and Patten. Paragraph 4.7. 293 Ibid. Paragraph 4.7. 128 subtle, but important way to retrain officers to see their job as serving the community as opposed to serving the state or the Queen. The recommendations of the Patten Report also included incorporating the European Convention on Human rights into a new Code of Ethics,294 including an assessment of awareness of human rights in execution of duties in officer evaluations,295 and appointing a human rights attorney to the staff of the police legal services296 among others. In addition to this reorientation toward a focus on human rights in policing, the Patten Report dedicated an entire chapter to retraining the police. The Commission did not lay out a detailed new training program, but rather recommended that such a program should be created and centered on the goals and recommendations of the Patten Report. They recommended that there be considerable community input in the final training for new recruits,297 that problem-solving and partnership with the community be central to training,298 that community awareness training be incorporated into every aspect of police training and that representatives from all cultural and religious traditions be consulted in that training,299 and that the curricula for training be publicly available and easily accessible.300 So while the specific curricula of the training was left to further 294 Ibid. Paragraph 4.8. 295 Ibid. Paragraph 4.10. 296 Ibid. Paragraph 4.11. 297 Ibid. Paragraph 16.10. 298 Ibid. Paragraph 16.16. 299 Ibid. Paragraph 16.17. 300 Ibid. Paragraph 16.25. 129 consultation, the Commission made clear that the training should be focused on community policing and human rights. Ongoing Community Consultation In addition to the suggested new oath and reorientation of policing around the idea of human rights and community policing, the Patten Report also included recommendations for ensuring that human rights violations by the police be addressed, corrected, and punished if necessary. These recommendations can generally be understood under the heading of accountability. Accountability of the police was given two chapters in the Patten Report – one examining the state of and arrangements for accountability under the RUC and one on recommendations for the future. The Commission argued that democratic policing must be based on consent from the community if it is to be effective, and consent depends on accountability.301 They examined several different kinds of accountability including democratic accountability, transparency, legal accountability, financial accountability, and internal accountability.302 The way in which the Commission envisioned the delivery of this accountability was focused around the ideals of community policing. Whenever possible, accountability would come in the form of community consultation and participation. First among these recommendations was the creation of the new Policing Board, discussed in the previous chapter for their role in selecting the new emblem of the PSNI, whose primary function would be to “hold the Chief Constable and the police service 301 Ibid. Paragraph 5.2. 302 Ibid. Paragraph 5.2. 130 publicly to account.”303 Among its many functions, the Board would be responsible for setting medium-term (3-5 year) goals for the PSNI and monitoring police performance with an eye to whether the PSNI is achieving those goals. The Commission made clear that the Board should be the primary institution of accountability, but that, “an important theme of this Report is that policing should be decentralized, and that there should be constant dialogue at local levels between the police and the community.”304 Therefore, in addition to the Board, the Patten Report recommended creating District Policing Partnership Boards (DPPBs) that would facilitate communication between the police and the communities they serve directly. These would provide a space for the community to voice concerns and for the police to present information about their activities and take questions from the community. The DPPBs would “represent the consumer, voice the concerns of citizens, and monitor the performance of the police in their districts.”305 The Patten Report recommended that both the Board and the various DPPBs should meet in public once a month to encourage transparency.306 To address legal accountability of the police, the Patten Report recommended the creation of a Police Ombudsman whose role would be to both hear and respond to complaints about the police as well as to seek out and investigate problems in policing when there had been no formal complaint.307 303 Ibid. Paragraph 6.3. 304 Ibid. Paragraph 6.25. 305 Ibid. Paragraph 6.29. 306 Ibid. Paragraph 6.36-6.37. 307 Ibid. Paragraph 6.41. 131 Again, the set of recommendations surrounding accountability harken back to the principles of community policing discussed in Chapter I of this dissertation. As the Commission directly stated in the Patten Report, good policing relies on the consent of the community and consent requires accountability. As I argued, this is a phenomenon specific to policing in democratic societies. Nevertheless, taking for granted that Northern Ireland sought democratic political solutions to the conflict, this is still important in that the Commission emphasized that accountability comes, at least in part, from community participation in the policing process. The recommendation to institute DPPBs and for their activities to be ongoing on an at least monthly basis is a clear example of the implementation of ongoing community input in policing. The make-up of these accountability institutions and their mandates directly reflect community policing principles. Decentralizing Decision Making The most overt reference to community policing in the Patten Report falls under the heading of ‘policing with the community.’ This set of recommendations was prefaced in the Patten Report with the stipulation that, while many of the recommendations could and should be implemented immediately, the implementation of many of the other recommendations would be dependent on the ‘security situation.’308 In other words, the level of violence experienced in the community and by the police would necessarily set limits on what kinds of changes could be implemented and when. At the same time, the recommended changes were intended to improve the security situation, so there was a 308 Ibid. Paragraph 7.1. 132 degree to which some changes would be made on faith that the community would respond positively, making the security situation more stable, and enabling further change. The phrase ‘policing with the community’ came out of the desire to clarify the often called for, but at this point ill-defined concept of community policing. The Commission believed that what most proponents of community policing were calling for was a system in which the police participate with and respond to the needs of the community and where the community actively communicates with and supports the police. Or, as they put it, What we do mean is: the police working in partnership with the community; the community thereby participating in its own policing; and the two working together, mobilizing resources to solve problems affecting public safety over the longer term rather than the police, alone, reacting short term to incidents as they occur.309 To facilitate this style of policing, the Patten Report made several recommendations regarding decentralized policing including the institution of the DPPBs mentioned in the previous section. In addition, the Patten Report recommended that neighborhood policing be the core of police work.310 In their investigation of the RUC, the Commission found a consistent complaint that the hierarchy and bureaucracy of the RUC made it nearly impossible for officers to get their work done in the most efficient manner. The Commission found that there was little to no consultation of on-the-ground officers when 309 Ibid. Paragraph 7.3. 310 Ibid. Paragraph 7.8. 133 it came to policy changes and implementation. The structure of police was more in the style of ‘command’ than ‘management.’311 A Superintendent commanding a sub-division answers to the Chief Superintendent commanding a division, who have another Superintendent as a Deputy Divisional Commander. Above him there is a regional headquarters, with an Assistant Chief Constable and another Chief Superintendent as Deputy Chief Constable. Only at this point for the line of management lead to central police headquarters.312 Importantly, this led to the basic recommendation that managers be allowed to manage. The Commission found that this level of hierarchy made it impossible for those with experiential knowledge of policing on the ground to make any decisions that might improve the delivery of policing.313 The Patten Report took this information and developed the neighborhood policing system in which managers would be allowed to manage their teams, control their budgets to an extent, and allow for the knowledge of those closest to the community to inform minor changes in policing practice on the ground. In this system, the neighborhood teams work directly with the community daily, forming relationships and building trust. Police are known to the community by face and name and are considered a part of those communities. To that end, the Patten Report recommended that police should patrol on foot whenever possible in terms of the security situation.314 Through these methods, the neighborhood officers become much more intimately familiar with both the problems they see in the neighborhood and those issues that are most on the minds of the people in 311 Ibid. Paragraph 10.2. 312 Ibid. Paragraph 10.5. 313 Ibid. Paragraph 10.7. 314 Ibid. Paragraph 7.13. 134 those communities. They should then use this knowledge to make their own suggestions and set their own policing priorities in their neighborhoods.315 These teams then would be supported by all other specialized teams within the police, rather than being a minor unit working independent from the other units of the police.316 To this end, the Patten Report included a graphic, depicted in Figure 12 below, of how they thought the organizational chart of the new police would look. This structure emphasized the concept that those closest to the community and in the most direct contact with citizens would be considered the highest authority on what the community needed from the police (as opposed to the traditional structure with the Chief as the head and highest authority in the police). 317 Figure 12. Recommended Command Structure 315 Ibid. Paragraph 7.14. 316 Ibid. Paragraph 7.8. 317 Ibid. Paragraph 7.8. 135 Implementation The brief analysis above highlights some of the important ways in which the Commission embraced the principles of community policing and worked to incorporate those principles into the 175 recommendations of the Patten Report. Their recommendations focused on changing the culture of the police themselves to be oriented toward working with and respecting the community, creating accountability mechanisms that directly engaged the public, and reorganizing the command structure of the police to allowed decentralized decision-making. However, the success of community policing relies not just on the initial assessment of the police, but in the ongoing community consultation and engagement with the policing process. In other words, the Patten Report was a set of non-binding recommendations for reforming the police in Northern Ireland and while within the first two to three years of the PSNI, “almost every last word of what Patten recommended [had] been implemented,”318 this did not necessarily mean that community policing was embraced fully or was immediately successful. The culture of human rights was embraced by the leadership of the PSNI and was becoming more and more embedded in the language and culture of the PSNI. Accountability mechanisms were adopted and immediately put into practice. The DPPBs (adopted under the slightly less cumbersome name of District Policing Partnerships – DPPs), were formed and began regularly meeting and consulting with the community. The Policing Board and the Police Ombudsman’s office were also instituted and immediately began working. Policing with the community became the mantra of the PSNI and neighborhood policing teams became the focus of interaction with the community. The transition was not seamless, however, 318 Dickson, 170. 136 and the institution of the policies associated with the guiding principles often met bumps on the ground. This section explores some of the specific policies that were implemented with the new PSNI and how those policies were experienced by the community. Embedding the culture of human rights within the police force became an embraced policy. In my interviews, almost all the officers I spoke with mentioned the words ‘human rights’ at least once during the interview. As one officer I spoke with put it, “Human rights is… I mean everything we do is looking first and foremost at people’s protection of Article II right of life, privacy. And all our actions that we take, that forms the basis of it.”319 Later in the interview, he asserted that, while it was a difficult concept to grasp at the beginning, as time went on the concept of human rights became fundamental to the thought process of policing. In every decision, the first thought became “How do I protect the human rights of the people I am interacting with?” Their thought process became centered on seeing other people’s points of view and doing their best to protect those rights.320 Certainly, the idea and words of ‘human rights’ seem to be everywhere in the PSNI. For example, the oath taken by officers adopted whole the language recommended in the Patten Report, adding an additional line, “and that while I continue to hold the said office I will to the best of my skill and knowledge discharge all the duties thereof according to law.”321 The PSNI also produces an annual ‘Human Rights Programme of Action,’ a document that describes the ways in which the PSNI is currently or plans to 319 Derry Officer 1. 320 Ibid. 321 Police (Northern Ireland) Act 2000. 137 better address human rights within the work of the PSNI each year, with regard to the Policing Board’s annual Human Rights Report (produced yearly since 2005). Accountability was also a concept that was embraced in the implementation of the PSNI. Specifically, the Ombudsman’s office and the DPPs were set up, funded with the start of the PSNI, and made great efforts to reach out to the community to ensure knowledge of these organisms of accountability and how to access and participate in them. For example, the Ombudsman’s office began compiling regular publicly available reports with information about the public perception of the police, community awareness of the Police Ombudsman, community perception of the independence of the Ombudsman, and knowledge of how to make complaints about the police among other information.322 These reports, which the office began publishing in late 2000, provide information to the public about the complaints that have been made against the police and how those complaints have been resolved through the office. It can be argued that this office has been successful in making the community aware of its presence and function with 84% of respondents saying they had heard of the Police Ombudsman’s office in October 2011323 compared to only 57% of respondents in 2001.324 Importantly, the public has been consistently aware that the Ombudsman is independent of the PSNI with 81% and 83% of respondents saying they thought the Ombudsman to be independent in 2000 and 2011 respectively. 322 The Police Ombudsman for Northern Ireland, "Public Awareness of the Northern Ireland Complaints System," (Belfast, United Kingdom2001). 323 The Police Ombudsman for Northern Ireland, "Annual Report on Public Awareness Od the System for Complaints Againts the Police in Northern Ireland, 2011," (Belfast, United Kingdom: Office of the Police Ombudsman fro Northern Ireland, 2011). 324 The Police Ombudsman for Northern Ireland. 138 The Police Ombudsman’s office is another accountability mechanism that embraces the principles of community policing. While there is no specific provision in its mandate requiring consultation with the community or the police in how they investigate complaints, the whole purpose of the Ombudsman’s office is to be responsive to the needs and specifically the complaints of the community. In this way, the Ombudsman’s office is the arm of policing in Northern Ireland that is most directly responsive to the community. At a conference on policing in Northern Ireland held in November 2016, the then sitting Police Ombudsman, Chief Executive Adrian McAllister, argued that the reason the Police Ombudsman Northern Ireland office has been successful is due to the policies surrounding the authority of the office.325 The Ombudsman is responsible for every complaint made against a police officer from ‘this officer was rude to me’ to in- custody deaths. To accomplish proper investigations for all complaints, they also have free and total access to all PSNI records. The PSNI may not withhold records requests from the Ombudsman’s office. Further, the Ombudsman is not limited to waiting on complaints from the public to pursue investigations. They have the power to ‘call themselves in’ if they think there is a matter worth investigating. In this way, they can act on behalf of the community even without specific community complaint. All these policies, in addition to the transparency provided by the regular reports mentioned above, have resulted in community confidence in the Police Ombudsman, and therefore confidence in the accountability of the police themselves.326 325 McAllister. 326 Ibid. 139 The DPPs created a new space for the community to communicate directly with the officers that serve them and for police to hear the specific concerns and needs of the community. The implementation of the DPPs was dictated in the Police (Northern Ireland) Act 2000 which stated the role of the DPPs to obtain and present the views of the community on policing to the District Commander and to provide a general forum for discussion and consultation on policing matters in the district, among other duties.327 In all, 26 DPPs were established, corresponding with the already existing District Councils. The different DPPs had varying levels of membership, seven had the minimum 15 people, five had 17 representatives, and 14 had the maximum of 19 members. All DPPs membership had a simple majority of elected officials, with the remainder Independent Representatives selected through an application process to be broadly representative of the community. The Policing Board had final say on appointments. When the first set of appointees were announced, they ranged in age from 22 to 75, came from all walks of life, and were gender and culturally representative.328 The DPPs were very quickly successful in engaging the community in the policing process. The first DPPs began meeting in April 2003,329 and the first Omnibus survey conducted after this in October 2003 showed that 65% of the population (63% of Catholics and 69% of Protestants) had heard of the DPPs and 64% of the population (57% of Catholics and 68% of Protestants) said they were prepared to contact their local 327 Police (Northern Ireland) Act 2000. Section 16. 328 Rea and Masefield, 184. 329 Ibid., 190. 140 DPP to raise issues/questions about the police.330 I spoke with several individuals who have served or currently serve on their local DPPs (now called Policing and Community Safety Partnerships – PCSPs) as well as several officers who all stated that the communication facilitated by the DPP/PCSPs was an important step in building community confidence in the police. One interviewee who joined his local DPP after the tragic death of a friend stated that he had no relationship with the police, RUC or PSNI, until that time. Because of his involvement, he told me, he was able to not only communicate with the police about the issues he and his community were facing, but was also able to hear, learn about, and comment on the strategies the PSNI were using to try to combat the issues he raised. As a staunch Republican from a working-class neighborhood, this was the first time he was able to develop confidence in, if not the entire PSNI, the individual officers he met and worked with through the DPP.331 Similarly, many officers and other community leaders stated that the DPPs shortened the pipeline of information between the community and the police. For example, one community leader said that prior to the institution of the DPPs, if a crime happened that needed to be investigated by the police someone within the area affected would call a community leader, such as this individual, who would then call the police to let them know what needed doing and where. Then the community leaders would arrange for protection for the police within that space for the time they needed it. With the introduction of the DPPs, community activists within those same communities had direct 330 Northern Ireland Policing Board and Northern Ireland Statistics and Research Agency, "Public Perceptions of the Police and the Northern Ireland Policing Board October 2003," (Belfast2003). 331 Belfast Civilian 1. 141 contact with the police and were able to ensure safety without having to go through an intermediary.332 The DPPs are one of the most obvious examples of community policing at work in Northern Ireland. The organizations provide forums for community members and the police to speak directly to one another in a setting that is intended to facilitate such conversation. The institution of neighborhood policing teams was the way in which the PSNI worked to build relationships with the community in a less formal and more everyday way than the DPP/PCSPs could provide. Following very closely the recommendations of the Patten Report, the PSNI invested heavily in and arranged the work of the police primarily around neighborhood teams who worked and patrolled in relatively small, defined areas. These officers worked to know the people in the neighborhoods they were assigned to, developed relationships, and acted as intermediaries whenever other branches of the PSNI needed to operate within their assigned neighborhoods. The personal relationships these officers were able to foster humanized the police in many areas where there had been little to no interaction with the police or at least little positive interaction. As one officer demonstrated in the following anecdote, one major obstacle to the community having confidence in the police was the belief that the officers were not real people, but just arms of the state. I remember standing on the city walls here, in a blue suit, which was public order. And I was a normal cop, but I, [had on] some protective gear, and [a] young guy came up. And he started giving a bit of abuse. So, I was chatting to him … and after a while he sort of settled down. [He says] ‘I, I hate all Protestants. You know they taught me that in St. Joe’s’ which is a local school. I says ‘I’m sure they didn’t teach you that. You may have learnt that when you were there, but I’m sure St. Joe’s did not teach you to hate Catholics, or to hate Protestants.’ And he 332 Derry Civilian 1, interview by Shawna Meechan, 18 September, 2015. 142 says, ‘I hate all Protestants, I hate all cops.’ I said, well do you hate me? ‘Oh no, you’re all right.’ But I was all right because he’d engaged with me and I’d talked to him and spoke with him.333 This officer found that simply interacting with the boy in this situation tore down a wall that had been built through social conditioning and isolation from the police. Through this one interaction, the officer may not have wholly changed the boy’s mind, but he was able to point out to him that his ‘hatred of all Protestants and cops’ was not absolute. This was the theory behind the PSNI’s focus on neighborhood policing in action. A different officer highlighted how the various arms of community policing discussed in this dissertation did not work in isolation from one another, but instead mutually reinforced the work each was doing. For example, he said that in the early years of the PSNI new recruits were required to first train with the neighborhood teams for 3-4 weeks before going out into the field. This, he argued, led to better interactions between ‘regular’ (or non-neighborhood) officers and the community. He argued that because the neighborhood teams have specific skills when it comes to talking and de-escalation, they had the ability to impart at least the basics of these skills in new recruits to help things go more smoothly in other divisions.334 For example, he said that most new recruits go into Response, the teams assigned to be the first to arrive at the scene of a crime. Because Response officers are the first to interact with citizens who are often in distress, in these situations, the skills they learned from their training with the neighborhood teams helps them to be more sensitive and communicate better with the civilians with which they interact. Even though they are not able to spend as much time with each victim as a 333 Derry Officer 1. 334 Belfast Officer 1. 143 neighborhood officer would given the parameters of their job, they are able to utilize their additional training to improve their interactions with the community. Results: Successes and Failures In the highly charged atmosphere of post-Troubles Northern Ireland, there were essentially two forms of policing the community monitored closely: public-order policing at events such as demonstrations or parades, and day-to-day policing including the handling of common crimes such as murder and burglary on down to vandalism and other anti-social behavior. The ways in which the PSNI handled policing in these situations has largely shaped how the community views not just the competence of the police, but also the level of bias in policing. Public order policing was often the source of much scrutiny due in part to its very public performance in highly publicized and political atmospheres. In particular, the policing of parades during the summer ‘parade season’ was and remains highly politically charged. Parading has been a large part of the culture of Northern Ireland since at least the eighteenth century. Many consider these annual parades a display of culture and celebration of history. However, of the 2722 parades in 2001, 2218 of those were labeled loyalist while only 133 of them were nationalist (with the remaining categorized as ‘other’),335 so it is understandable that the parades season is often seen as a time for loyalist celebrations, upsetting the nationalist community. In that same year, 29 of the parades experienced disorder. While the statistics do not provide specific information on what the nature of the disorder was or who, if anyone, was arrested, there has been a 335 Police Service of Northern Ireland (PSNI), "Parade Statistics 2000-2002," (Belfast, United Kingdom: Police Service Northern Ireland, 2003). 144 history of clashes over the routes the parades take, some of them marching directly through the middle of primarily CNR communities antagonizing the residents.336 With the history of the parades season as a catalyst for clashes between the CNR and PUL community, it is not surprising that the PSNI’s approach to policing these events was under the microscope from the beginning. The Public Processions (Northern Ireland) Act of 1998 created a new body that would remove the responsibility of deciding the legality of certain parades and their approved routes from the police.337 Instead, the Parades Commission made these decisions and the police were charged with enforcing their rulings. This provided some minimal cushion in how police were perceived when policing these events. However, even though the Parades Commission was helpful in separating the police from decisions about parading in the eyes of the community,338 “they had to give advice on the impact of a parade and to give effect to any determination by the Parades Commission which had the force of law.”339 In other words, as one community member presented it, even with the advent of the Parades Commission, if the police are enforcing decisions that are considered illegitimate by a portion of the community, they will be blamed.340 Therefore, it became that much more important to communicate and work with the community on how policing would be done at parades. 336 Barbara Zedler, "Parades," Northern Ireland Foundation, https://northernireland.foundation/projects/sharedfuture/parades/. 337 Public Processions (Northern Ireland) Act 338 Belfast Civilian 5. 339 Rea and Masefield, 307. 340 Belfast Civilian 4. 145 One interviewee told me the following story to demonstrate how important ongoing communication between the police and the community is to gaining community confidence in the police over time. The PSNI used to police the annual parades with their backs to the people marching (usually Protestants) and facing the protesters (usually Catholics objecting to the routes or time of the parade). The CNR community interpreted this body language as a sign that the police assumed Catholics would be the source of any trouble and the police were there to protect the Protestant marchers. When clashes between the protesters and marchers occurred in which the police arrested Catholics but not Protestants, the Catholic community interpreted this as discrimination by the police. After a series of violent clashes in the early 2000s, and with consultation with community leaders, the police changed their methods so that when lining up to police a parade route, every other officer faced the paraders and the protesters/observers in turn. This simple change in policing procedure made a huge difference in how the community in that neighborhood interpreted the role of the police. They were there to maintain order, de-escalate problems when they arose, and protect whomever needed protecting from whomever became an aggressor.341 Due to the success of this change in tactics, the interviewee claimed, the PSNI were able to make further inroads in some communities that had been historically difficult to police during parading. For example, certain community leaders would come to the parades and be on hand to calm members of their communities in moments of high tension. An officer could flag down a known leader to point out a community member who was causing trouble and de-escalate the situation before police intervention and/or an arrest would be necessary.342 In a separate interview, another community leader also independently mentioned this same example as one in which the lived experiences of 341 Belfast Civilian 2. 342 Ibid. 146 those who were present at the parades and saw the change in police practice led to internalized positive feelings about the PSNI.343 This example of change in parading policing tactics was the direct result of experience and communication between the police and the community. It is also a demonstration of the decentralized decision-making that the introduction of neighborhood policing teams allowed in the PSNI. This change was localized and specific to certain areas with highly contentious parades. In areas where the parades were smaller or not as contentious, the officers have not necessarily made these same changes. For example, in one station in Belfast where I conducted a series of interviews, the officers were consistent in stating that the parades that happen within their jurisdiction are not particularly contentious because the traditional routes do not run through neighborhoods or in areas where communities would clash. For these officers, parades have not historically needed special policing. Instead, the neighborhood in East Belfast that is in their jurisdiction has one specific interface area (a street where a heavily or exclusively CNR neighborhood borders a predominately or exclusively PUL neighborhood) where they face the most difficult challenges to community policing.344 It is easy to understand why interface areas present difficulties for the police in Northern Ireland. Many of these interface areas have what have been termed ‘peace walls,’ like the one shown in Figure 13, erected to separate the two communities and reduce violence. 343 Belfast Civilian 5. 344 Belfast Officer 9, interview by Shawna Meechan, 8 September, 2017; Belfast Officer 10, interview by Shawna Meechan, 8 September, 2017; Belfast Officer 11, interview by Shawna Meechan, 8 September, 2017. 147 345 Figure 13. Peace Wall in Belfast However, these walls also work to retrench the divisions in the community and can be difficult to remove once erected. As one government official stated, the places where peace walls have been successfully removed since the end of the Troubles are the areas where the community had already developed good relationships with the PSNI officers covering their area.346 Again, this demonstrates the importance of community policing both in developing neighborhood specific policies, but also generally building relationships between the community and the neighborhood officers in an area. The success of eliminating sometimes 90-foot walls with barbed wire can depend on how much the community trusts that the police will be present and fair if and when contentious moments arise. 345 Shawna Meechan, "Peace Wall in Belfast," (2017). 346 Belfast Civilian 11. 148 While public order policing is by its nature very public, and therefore easily monitored and publicized, most people still have their only interactions with the police in the course of day-to-day policing of common crime. Therefore, it was just as important for the PSNI to listen to the community and work with them to improve relationships through positive interactions in these situations as in public order policing situations, if not more so. Like with parading, the most important advancements in this area also came from listening to and learning from the reactions of the community. For example, the community can generally agree that there will be times when the police need to search homes of suspected criminals. Even during the Troubles, there were times when the CNR community accepted the role of the RUC in performing searches. One officer gave the example of a child-abuse investigation in the heart of a CNR estate in Derry city during the Troubles. Sinn Féin worked with the RUC to ensure the community was willing to let the police into the estate “unfettered and deal with it” because the issue of child abuse within their community was so toxic to the political work they were doing, it was better for them to allow the RUC to take care of the situation.347 The way in which searches are performed makes a huge difference in whether the community finds a particular search legitimate, however. One officer explained that during the Troubles, when he was sent to search one or two homes in a CNR estate, the entire estate would be shut down, blocked at every entrance and exit by armed officers with military back up. Understandably, he said, the community was turned off by this practice and blamed the police for being placed essentially under siege. 347 Derry Officer 3, interview by Shawna Meechan, 5 October, 2016. 149 Everybody was basically being tarred with the same stick because you maybe were not involved at all, but you’re still getting caught maybe taking the kids to school. You’re being stopped at a vehicle check-point. There’s military standing about. There’s police standing about. We’re all armed to the teeth.348 The same officer said that when the PSNI needs to perform a search now We avoid around school time. We avoid around tea time. We avoid around when school’s getting out. We avoid doing early house searches, like 5 or 6 o’clock in the morning, unless there is a critical reason as to why we need to do that. If there’s children in the house, we will try and do the search after the kids go to school, and so on.349 In addition, when it is possible and appropriate, they will call community representatives ahead of time to let them know that a search is going to happen and why, and then will let them know when the police have cleared the area. In this way, the community feels they are being respected and not randomly targeted.350 Over time, this policy has also allowed the police to perform surprise searches with the general support of the community. Where in the past community members would throw rocks and other projectiles at the police when they performed a search,351 now the odd surprise search is seen as more legitimate because the PSNI have been so thorough about communication. The community is more willing to accept the necessity of a surprise search when they are warned about most searches.352 On an even more individual level, the practices of neighborhood policing actively changed individual people’s minds about the PSNI. As one community member put it, 348 Ibid. 349 Ibid. 350 Derry Civilian 1, "Interview 1." 351 Derry Civilian 3; Derry Civilian 2. 352 Derry Officer 3. 150 neighborhood teams were the medium through which the community experienced the police ‘delivering on policing.’ “These teams led to personal relationships, one-on-one conversations, and personal contacts with the police.”353 The policy of developing relationships with activists and members of the community was truly taken to heart by every neighborhood officer I interviewed; and, conversely, every community activist and most other individuals I interviewed had the phone number of at least one officer they could contact personally. All five of the neighborhood team leaders I interviewed had multiple mobile phones and my interviews were interrupted more than once to answer calls coming in from the community. This level of engagement makes it that much easier for officers to build relationships in the community. As one officer put it, I do think people like the fact that you have a mobile telephone and they can ring you. And if they do have a concern, they’re not going through to the switchboard, they’re ringing the officer. Even though they know you’re not always on duty, but they’ll leave a message now and they’ll know, ‘Oh, aye, they will ring me back.’ And I think all that has helped. Whereas before, you know, there was no personal contact like that. They were just ringing in and possibly getting anybody.354 When I asked the officer if this meant they are never really off duty, she responded that it is not policy for officers to take their work mobile phones home with them, but many do because, Sometimes you’re wondering, you know, depending on who, who is doing the ringing, you think ‘well, maybe if they ring me and I’m… and they don’t get answer, they’ll be vulnerable.’ So, you take your phone just in case. You know you find yourself, maybe, doing that more and more. But then, I think that also helps because people then know you sort of care. So, I think that helps, you know, gain 353 Belfast Civilian 3. 354 Derry Officer 7, interview by Shawna Meechan, 6 October, 2016. 151 the trust too. You know, because we’re human too, so of course you care. You know you want the best for people.355 This level of engagement and commitment from neighborhood officers has built relationships that would have seemed impossible during the Troubles. One officer told a story of a friend of his, who had won ‘officer of the year’ shortly after the institution of the PSNI saying, Now he was a guy who was in the RUC, but the community knew him, they trusted him, they built that relationship with him, and that I think has been sort of the greatest success. That engagement with people. That genuine engagement with people. That honesty with people has been the change. Policies are good, you know, but it comes down to personal attitude and people’s willingness to engage on a real genuine level.356 Another pointed out that he had received an email addressed ‘A Chara’, which is Irish for ‘Dear.’357 That an individual in the community would feel comfortable sending an email with that greeting to an officer demonstrates a level of comfort not experienced under the RUC. The community members I interviewed were also insistent that the individual interactions with the police were the most important factors in gaining confidence in the PSNI as a whole. As one community member stated, “Experience is what makes individuals come to trust the police; day to day living.”358 Yet another stated, “Trust comes from the policies plus the experience while waiting to see outcomes.”359 “In the 355 Ibid. 356 Derry Officer 1. 357 Derry Officer 2. 358 Belfast Civilian 2. 359 Belfast Civilian 1. 152 end,” another community interviewee stated, “good policing is what determines legitimacy. We should always be questioning and improving. This is a burden for the police, but it’s what they signed up for.”360 Tempering Expectations The information provided here seems to provide good evidence that community policing has been implemented as close to in line with the principles discussed in Chapter I of this dissertation as possible. The results discussed in Chapter II also seem to tell us that community policing in Northern Ireland has succeeded in its goal of building community confidence in the police. However, it is important not to be overly starry-eyed about the successes of community policing in Northern Ireland. It was not an automatic nor easy switch from the CNR community deeply distrusting the RUC to embracing the PSNI. Similarly, it was not an easy transition for the members of the police to make the switch from the RUC to the PSNI. In Chapter II, I demonstrated that there was a significant increase in community confidence in the police in the first ten years of the PSNI, specifically within the CNR community who had been deeply distrustful of the RUC. However, what those statistics do not demonstrate well is that this shift was not linear. For example, in the first 5 years of the PSNI from 2001-2006, the omnibus survey was conducted twice a year, usually in April and September. In that time, Catholic community confidence in the police seemed to consistently waiver with the percentage of Catholics who rated the overall performance of the police as ‘very/fairly good’ tending to slowly increase from around 50% to 55% in 360 Belfast Civilian 4. 153 April, but always dropping to below or about 50% in September.361 This pattern indicates that the summer months in Northern Ireland tended to create more tension between the Catholic community and the police. This is not surprising given that the marching season described above begins in July and runs through September. Even while issues with policing parades were worked on regularly over time, this season clearly continued to create hard feelings between the Catholic community and the PSNI in its early years. However, beginning with the survey conducted in October 2007, notably the first survey after Sinn Féin endorsed the PSNI, Catholic community confidence jumped to 62% and actually increased over the next several surveys coming in at 65% in April 2008, 66% in September 2008, and a high of 67% in April 2009.362 However, following some particularly violent parades protests in July 2009 where the police used water cannons and baton rounds to disperse rioters,363 overall confidence in the police dropped a few points in the Catholic community to 61% in September 2009 and stayed fairly level at 62% in September 2010 and 63% in January 2012.364 Importantly, what these last series of numbers show is that the level of confidence the Catholic community has in the police 361 Northern Ireland Statistics and Research Agency and Northern Ireland Policing Board, "April 2003 Omnibus Survey: Northern Ireland Policing Board Module."; "Public Perceptions of the Police, Dpps, and the Northern Ireland Policing Board September 2005," (Belfast2005); Northern Ireland Statistics and Research Agency and Northern Ireland Policing Board, "Public Perceptions of the Police, Dpps, and the Northern Ireland Policing Board October 2006," (Belfast2006). 362 Northern Ireland Policing Board and Northern Ireland Statistics and Research Agency, "Public Perceptions of the Police, Dpps, and the Northern Ireland Policing Board April 2009 Omnibus Survey," (Belfast2009). 363 "Violence at Belfast Parade Route," BBC News Channel (2009), http://news.bbc.co.uk/2/hi/uk_news/northern_ireland/8146817.stm. 364 Northern Ireland Statistics and Research Agency and Northern Ireland Policing Board, "Public Perceptions of the Police, Dpps, and the Northern Ireland Policing Board January 2012 Omnibus Survey." 154 may be affected by one particularly negative incident, but it tends to stay relatively constant, even with some small gains, otherwise. The pattern of gains and losses in the CNR community’s confidence in the police described above is worth emphasizing. The large dip in the CNR community’s confidence in the police following the 2009 parade season reflects the importance of individual interactions with the police as highlighted in many of the interviews I conducted and brought up throughout this dissertation. One community leader stated, “Trust still comes down to individual actions within the police. Good cops lead to good interactions lead to trust.” That same community leader went on to say that interactions in any situation, whether high tension like parading or low tension like traffic stops, will affect how the community views the whole of the PSNI. He said, “[The police] could explain themselves better, be more personable. One bad interaction with the police can ruin the community’s view of the whole PSNI.”365 The incremental gains in community confidence made from 2001-2008 were damaged, at least temporarily, because of the perceived harsh tactics used by the police during the 2009 parading season. This is the burden of the police, to balance the need for public order and safety with the rights of those protesting. As another community leader put it, “In the end, good policing is what determines legitimacy. We should always be questioning and improving. This is a burden for the police, but it’s what they signed up for.”366 In addition to the early cycle of making some gains in Catholic community confidence in the PSNI only to lose those gains every summer, the police themselves had 365 Belfast Civilian 5. 366 Belfast Civilian 4. 155 difficulties with the transition. I spoke with one family member of an RUC officer who said that while she and her family embraced the peace process, it was a difficult transition to become a part of the community again. She explained that, especially during the Troubles, RUC officers and their families often felt like a community apart from the rest of Northern Ireland, including separate from other non-police Protestants. For her family, the transition from isolationism within the world of the police and police families to becoming a part of the larger community was extremely difficult.367 This feeling of separation is still felt by some officers today. When asked if he was comfortable sharing his ethnic/religious affiliation for an interview, one officer answered that he was Protestant but would consider himself an officer if discussing community divisions. He said that his experience had been that the conflict had three sides: Catholics, Protestants, and the police. He did not feel he was a ‘part’ of the Protestant community, but rather a separate community of officers.368 Possibly, it was the lack of understanding of this dynamic among those making recommendations for police reform that resulted in what some saw as inadequate consideration of the police themselves in the reform process. Several of the officers I interviewed who had been a part of both the RUC and PSNI and had lived through the transition expressed frustration at what they saw as a lack of consultation with officers in how the transition should proceed. One officer articulated it this way, The way the British government handled it in terms of the police was, ‘We’re your employer. Like it or lump it.’ And that caused untold psychological hurt and damage to the officers. And it was so disjointed and so confused and so frustrating. Em, for example, I can remember that they issued this decree that 367 Derry Civilian 3. 368 Belfast Officer 1. 156 said, ‘Nowhere in any police station will there be anything relating to the Royal Ulster Constabulary. It will not be.’ Police officers who had been through so much, who had, who had lost so much. Who had lost colleagues, who had had horrendous psychological experiences were told, “Shut up. Shh. Don’t. Don’t talk. You’ll offend somebody.”369 This seeming lack of care or attention to the needs of the officers staying on from the RUC to feel like they had a say in how the transition moved forward may also be responsible in part for why the transition was slow and trust between the police and the CNR community was difficult to achieve. If the police do not fully buy into the ethos of community policing, it will not work. However, it is tough to get real buy in from officers if they are simply told what to do instead of feeling like they were consulted and considered in reform. The slow buildup of confidence between the police and the community is reflected in the stories relayed above. The negative reactions to police tactics at parades, for example, were not easily or quickly solved. Instead, the ultimate change to the new tactics described in the anecdote above was the result of multiple meetings between the police and the community, innumerable bad feelings, and a deep struggle to overcome the distrust between the police and those members of the CNR community who had their misgivings about the PSNI cemented by early bad experiences.370 However, what these statistics backed up by the stories relayed above do show is that the principle of ongoing communication and collaboration between the community and the police was embraced and maintained even through serious struggles in Northern Ireland and, in the end 369 Derry Officer 2. 370 Belfast Civilian 2. 157 produced the kind of community confidence in the police promised by proponents of community policing. Conclusion Police reform in Northern Ireland after the Troubles was largely shaped by the principles of community policing as laid out in Chapter I of this dissertation. The Independent Commission on Policing for Northern Ireland emphasized the importance of community confidence for the police to have legitimacy in the Patten Report. Their many recommendations for reforming the police in Northern Ireland focused on the goal of gaining legitimacy through community confidence in the police and sought to do this through embracing community policing. In this chapter, I have demonstrated how well police reform in Northern Ireland mapped onto the second set of expectations laid out in Chapter I of this dissertation regarding the structures of institutions and practices of community policing. First, there were specific changes to the training of the police in Northern Ireland meant to refocus policing philosophy around the concept of policing with the community. This was demonstrated through the changed to the oath of office for police in the new PSNI as well as the changes in training that required new officers to spend time training with neighborhood teams. Second, there was clear involvement of the community at every stage of the reform process, and importantly, there exist permanent institutions and mechanisms for the ongoing input of the community in policing. The Police Ombudsman’s Office, the Policing Board, and the DPP/PCSPs provide avenues for the community to not just complain about the police, but to also provide innovative ideas and solutions for their own community’s problems. Third, police reform in Northern Ireland 158 involved a decentralization in some decision-making so that neighborhood teams were able to creatively solve the problems they faced through working with the communities they serve. In other words, officers are able to work within the general rules of the police while still allowing the localized knowledge they gain through regular interactions with the public to guide final decision making. The examples given of different neighborhood teams focusing on different issues such as parading or interface areas demonstrates this flexibility. Importantly, I do not want to paint the picture that this transition was easy or perfectly executed. There were several bumps in the road to implementing community policing in Northern Ireland after the Troubles. There could have been more outreach done to the members of the police who chose to stay on during the transition from the RUC to the PSNI. There were also several specific incidents over the years that inflamed tensions and, at least temporarily, diminished community confidence in the police. However, the success of the community policing initiative in Northern Ireland can be seen in the significant increase in community confidence in the PSNI over its first ten years of existence. These changes did not happen in a vacuum. However, this and the previous chapter have demonstrated how community policing was implemented in the context of post-ethnic civil conflict Northern Ireland. This process involved both symbolic and practical changes to the structure and delivery of policing. In the next chapter, I will explore how Kenya attempted to follow a very similar model for implementing police reform following their extremely violent 2007 national elections. I will lay out the brief history of ethnic conflict in Kenya and how it erupted in 2007. Then I will then look at 159 the specific reforms they undertook and how, despite consultation with many proponents of community policing and specific reference to Northern Ireland’s process, Kenya seems to have failed to implement community policing. 160 CHAPTER V KENYA Introduction The story of how Northern Ireland implemented community policing and moved from a 30-plus year violent ethnic conflict to state-stability and high levels of community confidence in the police is important in its own right. However, it is also important to examine the degree to which the lessons of Northern Ireland’s approach to community policing can be applied to other instances of state rebuilding after violent ethnic conflict. Northern Ireland is often held up as an example for other ethnically divided communities of how to reform policing and build community confidence in the police.371 Several of the officers and community members I interviewed had been invited to speak in other regions of the world, including Israel and Palestine and South Africa, on their experiences with police reform.372 One group I met was organizing training on policing in ethnically divided communities that they were planning to export to police around the world, starting with the Chicago Police Department.373 One country that explicitly examined the experiences of Northern Ireland when reforming their own police was Kenya in the aftermath of their 2007 election violence. Following a particularly ethnically divisive presidential race in 2007, Kenya experienced an outbreak of violence that left 1,133 dead and 350,000 internally displaced people, according to the Report of the Commission of Inquiry into Post-Election 371 Perry; Brunger; Guelke; Bayley; Marijan and Brennan. 372 Belfast Civilian 2; Derry Officer 2. 373 Forum for Cities in Transition. 161 Violence (commonly referred to as the Waki Report for its chairperson, Justice Philip Waki).374 In this chapter, I will first briefly explore the ethnic tensions that led to the outbreak of violence prior to, during, and following the 2007 presidential election and will then look at how Kenya responded to this violence. Like Northern Ireland, Kenya has a history of distrust between the police and community and therefore there was a specific focus on reforming the police in the aftermath of violence. I discuss the shape these reforms took, noting the specific influence of the experiences of Northern Ireland, and discuss how their implementation was approached. Finally, I provide an initial discussion of how the Kenyan community has responded to these reforms and make some suggestions for why this response has been so different from that in Northern Ireland. Historical Setting Kenya, much like Northern Ireland, has a long history in which interactions between the authorities and the community have been fraught. It makes the most sense to start the story of Kenya’s history of ethnic tensions and distrust of the police with colonization. In 1895, the British government claimed the land of modern-day Kenya as a colony and a protectorate and set up a formal British administration.375 Much like colonialism in other parts of Africa, the Kenyan colonial experience was shaped largely by policies of racism. For example, large portions of land that cut through the modern- day provinces of the Rift Valley, Nyanza, Western, and Central were reserved for white European settlers and became known as the ‘White Highlands.’376 Other policies were 374 Waki et al., 305. 375 Kenya Embassy, "About Kenya History," http://www.kenyaembassydc.org/aboutkenyahistory.html. 376 James Edelman and Elise Bant, Unjust Enrichment in Australia (Oxford: Oxford University Press, 2006), 1. 162 also put in place to force Africans to work for low wages on public works projects or on settler farms.377 It was during the colonial period that the British also used tribal distinctions among the Africans to divide and control the local population. The Kenyan people represent a wide variety of ethnicities that could be divided into more than 70 distinct groups. The largest ethnic group, the Kikuyu, make up about 20% of the overall population (about 7 million people) with the next largest group (the Luo) making up just under 14%, while the smallest ethnic groups such as the El Molo account for only 500 people. The five largest ethnic groups in Kenya include the Kikuyu, Luo, Luhya, Kamba, and Kalenjin and together account for about 70% of the Kenyan population. There are many other smaller and well-known groups including the Maasai, and many groups have shared linguistic traits such as those who speak variations of Bantu or Niolitic languages.378 While there were several tribal identities in Kenya at the time of colonization, they had regularly intermixed and intermarried.379 Under British colonial rule, these identities were solidified and politicized. For example, the Kikuyu people were forced to become tenant farmers while the Maasai were defined as herders by the British and were confined to specific areas. The Kikuyu and Maasai were not allowed to enter each other’s 377 Brian Van Arkadie, "Reflections on Land Policy and the Independence Settlement in Kenya," Review of African Political Economy 43, no. sup 1: Special Issue: Land, Liberation and Democracy: A Tribute to Lionel (2016): 61-63. 378 Claire Robertson, "Beyond 'Tribes': Violence and Politics in Kenya," Origins: Current Events in Historical Perspective 1, no. 7 (2008); University of Pennsylvania African Studies Center, "Kenya - Ethnic Groups," https://www.africa.upenn.edu/NEH/kethnic.htm. 379 Robertson. 295; Roxana Gutierrez-Romero, "To What Extent Did Ethnicity and Economic Issues Matter in the 2007 Disputed Kenyan Elections?," Development Policy Review 31, no. 3 (2013). 163 territory, and so the ethnic divisions that had previously been very fluid became fixed.380 These divisions were enforced for all the ethnic groups in Kenya.381 The influence of colonialism on the importance and prominence of ethnicity cannot be overstated. The lasting effects of colonialism in Kenya can be seen in the form of government and structure of the constitution, the laws surrounding the allocation of land, and in the continued political relevance of ethnicity and ethnic identification. At the time of independence, the form and structure of Kenya’s government was strongly debated. One side of the debate, a political coalition of two of the largest ethnic groups, the Kikuyu and Luo, preferred a strong central government with large powers resting with the president.382 This essentially would be a copy of the colonial government, with Africans at the head instead of the British, and would create a weak legislature and judicial branch that had little independence from the executive and, therefore, no real checks on the executive’s powers.383 A coalition of smaller ethnic groups opposed this structure of government, fearing that the rights of smaller ethnic groups (specifically land rights) would be overrun by such a strong central government and instead preferred a kind of ethnic federalism (majimboism in Swahili) where semi- autonomous regions based on ethnicity would have more control supplemented by a 380 Robertson. 381 John Lonsdale, "Kenya: Ethnicity, Tribe, and State," openDemocracy (2008), https://www.opendemocracy.net/article/kenya_ethnicity_tribe_and_state. 382 Peter Kagwanja and Willy Mutunga, "Is Majimbo Federalism? Constitutional Debate in a Tribal Shark- Tank," Opinion, Daily Nation (2001), https://www.nation.co.ke/oped/1192-340744-mbwm4j/index.html. 383 H. Kwasi Prempeh, "Constitutionalism, Ethnicity and Minority Rights in Africa: A Reply to Jeremie Gilbert," International Journal of Constitutional Law 11, no. 2 (2013). 164 much weaker central government.384 Opponents of majimboism claimed this kind of decentralized and tribalistic government structure would be destabilizing to the fledgling independent Kenya.385 When Jomo Kenyatta (an ethnic Kikuyu) and the strong central government coalition won the first election in 1964, most regional powers were abolished and Kenya’s first constitution had a strong central government.386 In addition to winning the argument over the form of government, Kenyatta also convinced his main political rival, Daniel arap Moi (an ethnic Kalenjin), that it was better for the country to present a united government and Moi agreed to merge his Kenya African Democratic Union (KANU) party with Kenyatta’s Kenya African National Union (KANU). The remaining major opposition party, the Kenya People’s Party (KPU) was banned,387 and Kenya became a de facto one-party state with no effective opposition parties.388 Some have argued that this was necessary to suppress the more extreme political elements in the newly independent Kenya, emphasizing peace and stability in order to address economic and political legacies of colonialism.389 However, it is important to note that this was a step away from the democratic institutions initially promised for independent Kenya. 384 David Anderson and Emma Lochery, "Violence and Exodus in Kenya's Rift Valley, 2008: Predictable and Preventable?," Journal of Eastern African Studies 2, no. 2 (2008): 330. 385 Kenya Embassy; David Anderson, "Majimboism: The Tourbled History of an Idea," in Our Turn to Eat! Politics in Kenya since 1950, ed. Daniel Branch and Nic Cheeseman (Berlin: Lit Verlag, 2008). 386 Colin Leys, Underdevelopment in Kenya, the Political Economy of Neo-Colonialism (London: Heinemann Ltd, 1975), 212. 387 Kenya Embassy. 388 Human Rights Watch, Rawlence, and Albin-Lackey, 30. 389 Kenya Embassy. 165 President Moi succeeded President Kenyatta upon his death and in 1982 made Kenya an official one-party state, amending the constitution to reflect this.390 However, under intense international pressures to democratize, multi-party elections were reintroduced in 1991. The re-introduction of multi-party elections naturally led to a resurgence of the debate over majimboism.391 Because Moi was Kalenjin, one of the smaller of the major ethnic groups with just under 11% of the population,392 he was able to use the resurgence of the majimboism debate to his advantage and managed to win his elections in 1992 and 1997. However, both elections were marked by extreme violence and ethnic tensions.393 For example, to shore up his support in the Rift Valley among ethnic minorities, President Moi tried to incite hatred toward the Kikuyu to drive them out of the area. He was successful in promoting clashes that led to up to 1,500 people dead and 300,000 displaced.394 When the hybrid party of the National Rainbow Coalition and their presidential candidate Mwai Kibaki (an ethnic Kikuyu) ran for office in 2002, they promised a new constitution within 100 days that would end corruption in government.395 The various parties that made up the Coalition signed a Memorandum of Understanding in which they 390 Ibid. 391 Human Rights Watch, Rawlence, and Albin-Lackey, 30. 392 University of Pennsylvania African Studies Center. 393 Human Rights Watch, Rawlence, and Albin-Lackey. 394 Human Rights Watch, "Failing the Internally Displaced: The Undp Displaced Persons Program in Kenya," (New Yoek: Human Rights Watch, 1997), 39. 395 National Rainbow Coalition (Kenya), "Democracy and Empowerment: Manifesto of the National Rainbow Coalition (Narc)," (Nairobi: National Rainbow Coalition, 2002). 166 agreed to a general formation of government under the unified coalition.396 Unfortunately, when they won, the Coalition almost immediately fractured over how to reform the constitution with particular contention over the powers of the president and many of the promised reforms were lost or watered down in the process of negotiations. When this weak reform proposal was put to a national vote, the factions within the National Rainbow Coalition split with several groups supporting the “no” campaign. The referendum ultimately failed with 57% of Kenyans voting “no”,397 leaving Kenya with the same constitutional issues from before.398 Issues of land are intimately related to the constitutional issues Kenya has experienced. When Kenya became independent in 1963, the British handed over most of the approximately 20 percent of Kenyan land they had claimed for themselves during colonialism.399 However, rather than handing the land over to the people or tribes from whom the land had been taken, it was given to the new Kenyan government.400 The new government, as already noted, had been formed with the help of the British as a reflection of the colonial government before it and therefore did not include any provisions for collective land rights. Instead, the new government maintained the system of private ownership of land, which upset many of the indigenous people who had maintained agreements for the collective use of land under colonialism. While under colonialism, 396 Gilbert M Khadiagala, "Political Movements and Coalition Politics in Kenya: Entrenching Ethnicity," South African Journal of International Affairs 17, no. 1 (2010): 75. 397 Ibid., 77. 398 International Crisis Group, "Kenya in Crisis," (Nairobi; Brussels: International Crisis Group, 2008). 399 Human Rights Watch, Rawlence, and Albin-Lackey, 26. 400 Van Arkadie, 64-65. 167 they had not been allowed to own the land (it was held in ‘trust’ for them by the government), they had been allowed to use it according to their traditional communal practices.401 In the new independent Kenya, President Kenyatta did not acknowledge traditional land use and instead sold the land as private property to any willing buyer with the ability to pay. This policy led to much of the land being sold to members of Kenyatta’s own Kikuyu ethnic group instead of to those who had traditionally used the land.402 The policy of selling historically common-use land to individuals continued under Kenyatta’s successor, President Moi. Moi publicly villainized the ‘land grabs’ of the Kikuyu while continuing the sales quietly, arguably because it was the Kikuyu who had the money to purchase the land. These actions throughout the Moi presidency from 1979-2002 increased the amount of contested land as well as the animosity of other minority ethnic groups toward the Kikuyu.403 The 2002 campaign of Kibaki’s National Rainbow Coalition focused heavily on the unjust distribution of land in Kenya and promised to rectify the disparities that had been introduced under colonialism and reinforced under the administrations of Presidents Kenyatta and Moi.404 When he was elected, Kibaki created the Ndung’u Commission to examine the history and current patterns of corruption and favoritism in land allocation. Ultimately, the report produced 401 Human Rights Watch, Rawlence, and Albin-Lackey, 27. 402 Ibid. 403 Judi Wakhungu, Elvin Nyukuri, and Chris Huggins, "Land Tenure and Violent Conflict in Kenya," ed. African Centre for Technology Studies (ACTS) (Nairobi: African Centre for Technology Studies (ACTS), 2008). 404 HRW Kenya’s Unfinished democracy, 2002 168 by the Commission was deemed too controversial and the Kibaki government chose not the implement its recommendations.405 To reiterate, ethnicity is tied closely to both the political and land debates in Kenya. As such, ethnicity has been a common tool used to manipulate the electorate and garner support for politicians. As mentioned earlier, the debates over constitutional form and the election of Jomo Kenyatta were organized along ethnic lines with the larger ethnic groups of the Kikuyu and Luo supporting Kenyatta and a strong centralized government and most minority ethnic groups preferring federalism based on majimboism. With the reintroduction of a multi-party system under President Moi, ethnicity was again activated to garner support for Moi while creating an ‘enemy’ in the Kikuyu against which smaller ethnic groups in Kenya could direct their frustrations.406 Ethnic tensions and violent clashes were associated with the 1997 elections and several investigations concluded that the violence was incited and organized by politicians.407 Despite these accusations and some examples of specific evidence against individuals, no one has been prosecuted for these crimes.408 Similarly, in the 2002 elections, even though Kibaki himself was Kikuyu, his support of a new constitution and the possibility of the institutionalization of majimboism was a large reason for his successful run.409 405 Human Rights Watch, Rawlence, and Albin-Lackey, 27. 406 Anderson and Lochery, 329. 407 Human Rights Watch; Anderson and Lochery. 408 Human Rights Watch, Rawlence, and Albin-Lackey, 33. 409 Ibid. 169 Election Violence The preceding sections highlight the continuously shifting alliances among ethnic groups and political parties throughout Kenya’s independent history. After the National Rainbow Coalition fell apart over the constitutional debates of 2002, Kibaki dissolved the cabinet and reformed the government under the new heading of the Party of National Unity (PNU) which included those factions of the National Rainbow Coalition which had supported his “yes” campaign for constitutional reform.410 The 2007 Kenyan elections were marked by much of the same ethnically divisive rhetoric and ethnically based violence of elections past. The candidates all had substantive platforms, but also leaned heavily on ethnic divisions in their campaigning.411 The PNU was accused by the opposition of ruling in favor of President Kibaki’s ethnic group, the Kikuyu. The opposition party, the Orange Democratic Movement (ODM), was led by Raila Odinga, an ethnic Luo, and was therefore seen to be associated with the Luo. This was reinforced by the PNU’s focus on Luo traditions, claiming that an uncircumcised man could not be president.412 This rhetoric led to at least 41 election-linked deaths prior to the official election in December.413 However, while it is true that this violence had been building for 410 Khadiagala, 77. 411 Clark C. Gibson and James D. Long, "The Presidential and Parliamentary Elections in Kenya, December 2007," Electoral Studies 28, no. 3 (2009). 412 Human Rights Watch, Rawlence, and Albin-Lackey, 8. 413 Gutierrez-Romero, 295. 170 some time, the immediate trigger of the violence of the election was the perceived rigging of the election.414 Voting during the 2007 election proceeded relatively smoothly and without incident with a record number of Kenyans registering and turning out to vote.415 The problems began to arise two days later when the results started to come in and despite major losses to the ruling PNU party in the legislature, the incumbent Kibaki was pulling ahead in presidential votes. Even before the official vote count was announced, violence broke out all over the country in reaction to this perceived election rigging.416 The following day, 30 December, Kibaki was officially announced the winner of the presidential election.417 According to Human Rights Watch, in the last few hours of tabulating votes, a more than one million vote lead for opposition candidate Raila Odinga disappeared and resulted in a slim victory for Kibaki, a result made all the more unbelievable by the parliamentary vote in which Mr. Odinga’s ODM won 99 seats to the PNU’s 43.418 These discrepancies in addition to the months of heightened ethnic rhetoric and increasing inter-ethnic violence led to the start of one of the most violent periods of ethnic violence in Kenya’s independent history. 414 Opolot Okia, "The Role of the Police in the Post Election Violence in Kenya 2007/8," Journal of Third World Studies 28, no. 2 (2011): 259. 415 Jeffrey Gettleman, "Disputed Vote Plunges Kenya into Bloodshed," The New York Times, 31 December 2007. 416 Human Rights Watch, Rawlence, and Albin-Lackey, 8. 417 Nic Cheeseman, "The Kenyan Elections of 2007: An Introduction," Journal of Eastern African Studies 2, no. 2 (2008): 166. 418 Human Rights Watch, Rawlence, and Albin-Lackey, 36. 171 The election-triggered violence was primarily concentrated in the Rift Valley where both the PNU and ODM had large numbers of supporters, though incidents of violence were reported from all parts of the country.419 In the immediate aftermath of the election, violence was mostly targeted against members of President Kibaki’s ethnic group, the Kikuyu. However, this quickly led to reprisal attacks by Kikuyu attacking other ethnic minorities who were (or were assumed to be) supportive of the opposition. In all, more than 1,000 people were killed and over 500,000 displaced during two months of violence.420 The violence in most cities and towns in the Rift Valley began under similar circumstances. Ethnically-based campaigning had convinced many minority ethnic groups that a victory for the opposition ODM party would result in the redistribution of Kikuyu land to them. On top of this, they were told that if ODM were to lose the election, that in itself would be proof of election tampering. Therefore, when Human Rights Watch (HRW) interviewed residents in the areas with the worst violence, they were told several times that the non-Kikuyus were “merely doing by force what they had been denied a chance to do at the ballot box.”421 In several areas, meetings were called in the days prior to the election. At these meetings, leaders riled up the communities, especially the young men, and prepared them for violence in the event of an ODM loss. Community members 419 Okia; Human Rights Watch, Rawlence, and Albin-Lackey; Anderson and Lochery. 420 Human Rights Watch, Rawlence, and Albin-Lackey, 49. Other sources have said that it was more than 1,400 killed and more than 600,000 displaced (James Brownsell, "Kenya: What Went Wrong in 2007?," Al Jazeera (2013), https://www.aljazeera.com/indepth/features/2013/03/201333123153703492.html.). I will use Human Rights Watch’s number as I have confidence in their sources and am unclear where the numbers others have put forward come from. 421 Human Rights Watch, Rawlence, and Albin-Lackey, 50. 172 who were reluctant to attend such meetings were met with threats of violence and property destruction.422 Faced with this level of violence targeting Kikuyu communities, it is easy to understand how Kikuyu could justify reprisal attacks throughout the country. However, the meetings that were held in ethnic minority communities were not unique to them. Several Kikuyu interviewed by HRW stated that there were similar meetings in other places where Kikuyu leaders incited the Kikuyu youth to participate in violence against any ethnic minority they assumed supported the PNU.423 This manipulation of ethnic identity to justify violence against others was the hallmark of the post 2007 election violence. There are several well-publicized individual incidents of extreme violence including a horrifying attack on a church in Kimbaa where many Kikuyu were seeking refuge. Attackers barricaded the church with petrol-soaked mattresses and set them on fire. In total, more than 30 people were killed in the attack, including children and a handicapped woman in a wheelchair.424 However, with a few notable exceptions, most of the violence was random and unpredictable. Individuals might be killed by a gang of young boys in the street or by a stray bullet shot by the police. In some areas, regular citizens turned on their neighbors and killed, mutilated, or burned the homes of those they 422 Ibid., 52. 423 Ibid., 60. 424 Ibid., 55. 173 had interacted with regularly before. In other areas, it was claimed that ‘people with money’ came in and paid the jobless to kill or harm others.425 Peace Agreement On 28 February 2008 President Kibaki and the ODM’s Odinga signed the National Accord and Reconciliation Act (the Accord).426 The Accord would create a new position of Prime Minister who would be an elected member of parliament and the head of the majority party in the National Assembly and would have the power to “coordinate and supervise the execution of government functions.”427 The Accord would also provide for the division of several important executive cabinet positions based on the relative strengths of the parties in the National Assembly. Kibaki convened the National Assembly on 6 March to make the necessary constitutional changes to implement the agreement and they were passed in the National Accord and Reconciliation Act.428 It took several weeks to implement the Accord in full, as several details had not been worked out in the initial agreement such as the number of cabinet positions and the process for the writing of a new constitution.429 However, on 17 April 2008, Odinga and the Cabinet were sworn in. The new coalition government eventually produced a new draft constitution, which was passed in a national referendum by 67 percent on 4 August 425 Ibid., 60-62. 426 Okia. 427 "Key Points: Kenya Power-Sharing Deal," BBC News World Edition (2008), http://news.bbc.co.uk/2/hi/africa/7269476.stm. 428 National Accord and Reconciliation Act, (28 February 2008). 429 Ibid. 174 2010, more than two years after the initial Peace Accord was signed.430 The new constitution addressed issues of human rights, a level of federalism in power sharing between the national and county governments, the creation of a National Land Commission, and reforms to the police. It also created greater checks on the power of the president, allowing impeachment for the first time and eliminated the position of Prime Minister.431 Policing in Kenya The origins of the police in Kenya stem, once again, from colonialism. As in other places in the world, colonial rule in Kenya “was illegitimate, and being so, it relied on oppression, repression and brutal force to sustain itself.”432 The police were the enforcers of the policy of conquer and control, and therefore developed relationships with the Kenyan people that were violent, aggressive, and oppressive. This history has also strongly shaped the way the police in independent Kenya have operated. The Kenyatta and Moi governments used the institution of the police to shore up political support and harass anyone who was suspected or known to oppose the regime.433 The police were first established in Kenya in 1887 as an instrument of colonial power to assist in the accomplishment of colonial aims.434 From the beginning, there was 430 Mariama Diallo, "New Kenyan Constitution Ratified," Voice of America (2010), https://www.voanews.com/a/kenyas-new-constitution-ratified-100158209/123395.html. 431 Xan Rice, "Kenya Votes for New Constitution," The Guardian (2010), https://www.theguardian.com/world/2010/aug/05/kenya-votes-for-new-constitution; The Constitution of Kenya, (27 August 2010). 432 Gimode, 227. 433 Ibid. 434 Kenneth Omeje and John Mwangi Githigaro, "The Challenges of State Policing in Kenya," Peace and Conflict Review 7, no. 1 (2012): 65. 175 an ethnic element to the police force. Much like they had labeled the Maasai as herders and the Kikuyu as tenant farmers, the British believed that the Kamba and Kalenjin ethnic groups had the most soldierly qualities.435 The force was primarily used to protect the white areas of Kenya and to suppress protests and rebellions by Africans.436 The mobilization of the police against the Africans in Kenya in the name of ‘upholding the law’ (regardless of how undemocratic those laws), came to mark a constant in the behavior of the police. Rather than serving the community, the police were acting to serve the political elite and their supporters. As David Bayley argues, when the police use their “authority primarily to serve the interests of the government, they belie the democratic promise of government for the people. The most dramatic contribution police can make to democracy is to become responsive to the needs of individual citizens.”437 In the 1940s and 1950s, African soldiers returning from World War 2 started demanding more democratic rights and the police were enlisted to vet public gatherings and issue permits for events that met the requirements for ‘public order,’ a move which further politicized the police.438 The police became even more militarized in response to the Mau Mau Rebellion from 1952-1964, in which a large group of Africans (largely of Kikuyu ethnicity) instituted a guerilla war against the colonial occupiers of Kenya in an attempt to expel 435 David Throup, "Crime, Politics, and the Police in Colonial Kenya, 1939-1963," in Policing and Decolonization: Politics, Nationalism, and the Police, 1917-1965, ed. David Anderson and David Killingray (New York: Manchester University Press, 1992). 436 Gimode, 230. 437 Bayley, Democratizing the Police Abroad: What to Do and How to Do It, 13. 438 Gimode, 231. 176 colonizers and take back the land for native Kenyans.439 In response to the rebellion, emergency powers were invoked and the police were given powers to arrest anyone without cause, declare certain areas under quarantine, and to shoot to kill anyone they deemed to violate these rules.440 However, after the rebellion and with independence seemingly on the horizon, the first attempts to address the ethnic composition of the police to bring it in line with that of the general population were made.441 Under the first independent Kenyan regime of Jomo Kenyatta, promises to deal with issues surrounding the oppressive police force went unfulfilled. Instead of pursuing democratic institutions, Kenyatta’s government concentrated almost all power in the president, which allowed him to manipulate the police to control the people and violently suppress political opposition.442 Kenyatta, who relied heavily on ethnic loyalties and patronage within all aspects of government, appointed ethnic Kikuyu to nearly every leadership position in government, including the many branches of the police.443 It is widely accepted that Kenyatta then commissioned the police to arrest and carry out assassinations of some political opponents and to illegally detain others for many years.444 439 Lonsdale. 440 Gimode, 232. 441 Anthony Clayton, "Kenya," in Monographs in International Studies. Africa Series (Athens, Ohio: Ohio University Center for International Studies, 1989). 442 Annie Mageka, "Police Reform in Kenya: Challenges and Opportunities," Centre for Security Governance 2018, no. 9 October (2015), http://secgovcentre.org/2015/10/police-reform-in-kenya- challenges-and-opportunities/. 443 Charles Hornsby, Kenya: A History since Independence (London: I.B. Tauris, 2012). 444 Gimode, 235-38. 177 Upon the death of Kenyatta and the subsequent ascension of Moi, Moi’s co- ethnics, the Kalenjin, were appointed to many of these same positions and oppression and abuses by the police in Kenya were taken to new heights.445 As Edwin Gimode states, “Moi governed through violence, harassment, intimidation and increased use of organs of state security for self-preservation.”446 He vigorously pursued any person who dared to speak against him including students and professors, lawyers, journalists, and any political opposition. He also strongly defended the ‘right’ of the police to use force to maintain order and quell dissent.447 The use of the police to repress political opposition, and therefore to protect the administration rather than citizens, was the context in which the 2007 election violence erupted. It is not surprising that some felt the response of the police during the violence to be unevenly distributed and simultaneously inadequate and excessive in the use of force. Unfortunately, there is not data available on the ethnic makeup of the police during this time, but there was a general understanding that the police were not representative of the community in ethnic or gender composition.448 When Kibaki declared victory in the election, he also issued a ban on all public demonstrations, which he justified as necessary to prevent violence.449 As they had in the past, the police enforced this ban vigorously in opposition areas, and less so in pro- 445 Hornsby. 446 Gimode, 239. 447 Omeje and Mwangi Githigaro, 65. 448 The World Bank, "Kenya Economic Development, Police Oversight, and Accountability," (2009), 16 and 33. 449 Human Rights Watch, Rawlence, and Albin-Lackey. 178 government areas, which resulted in many hundreds of Kenyans dead at the hands of the police. For example, in Nairobi, several protests erupted in the slums where ODM had high levels of support. The police used live ammunition to quell protests which evolved into riots and to prevent new demonstrators in the slums from joining those who had made it to city center. HRW conducted interviews with several witnesses to and victims of the police violence in the slums and heard stories of police shooting to kill any person they saw on the streets, regardless of whether that individual was participating in protests. They also recounted innocent bystanders being shot by stray bullets on the tightly packed, narrow streets of the slums.450 Similar scenes were reported in Kisumu and Eldoret. In Kisumu, the local police admitted to using live ammunition, but claimed it was necessary to prevent looting, even though the protesters were unarmed and the majority of the victims of police shootings were in the slums where there were no shops to loot.451 This led to the deaths of 50 people by the police in the initial round of violence in Kisumu, with an additional 25 dead in the second day. At least 30 of the victims were shot in the back, according to post-mortem reports.452 A second major grievance arose from these police shootings, namely that the police seemed to resort to deadly force only in areas where pro-opposition protests were occurring such as the slums of Nairobi and Kisumu where officers were ordered to use live ammunition and some officers felt there was an ‘unofficial shoot-to kill policy’ in 450 Ibid., 46-48. 451 Ibid., 41-43. 452 Okia, 262. 179 place.453 Alternatively, when pro-government mobs were burning homes and killing minority ethnics in Naivasha and Nakuru, the police did little to nothing to intervene.454 In other areas such as Eldoret and Molo/Kuresi, the police were accused of siding with Kalenjin militias, allowing unacceptable violence against Kikuyu.455 This difference in application of the law may be attributed to several factors. First, the legacy of colonial policing included a distribution of officers that was drastically uneven throughout Kenya. Under colonialism, the police were primarily located in urban commercial areas and traditional European residential areas while rural Kenya remains drastically underserved.456 Therefore, the more rural Naivasha and Nakuru had fewer officers who may have felt incapable of successfully intervening to stop violence given their relatively few numbers and equipment. Second, a similar legacy exists surrounding where police are stationed. Regulations from 1957 stated that not more than 45% of officers could be from the area where they serve, meaning that primarily co-ethnics are not policing one another.457 This also may account for why the police were seen as simultaneously backing the Kikuyu in some areas while also backing Kalenjin militias in others. The drastically different responses to the violence in different areas led to a host of questions about the police’s capacity to protect the people, the political nature of the 453 Human Rights Watch, Rawlence, and Albin-Lackey, 39. 454 Ibid. 455 Ibid. 456 Oliver Vanden Eynde, Patrick M. Kuhn, and Alexander Moradi, "Ethnic Politics and Job Performance in the Kenyan Police 1957-1970," (2015). 457 Ibid. 180 orders meted out in certain areas, and the bias of the police themselves.458 As the police are the most direct arm of the state with which the public interacts,459 the brutal repression of Kenyans in mostly opposition areas coupled with the lack of adequate protection elsewhere functioned to cement many Kenyans’ lack of confidence in the police. National Taskforce on Police Reforms It was in the context of all this history that the Accord was negotiated. Among other priorities including addressing ethnic representation in government, the Accord prioritized the reformation of law enforcement institutions. In response to this, the government set up the National Taskforce on Police Reforms (NTPR) in May of 2009, headed by retired Judge Philip Ransley.460 In the course of its work, the NTPR reviewed research and police-related academic literature, received submissions from stakeholders, visited the eight provinces of Kenya conducting interviews and listening sessions, and visited several other countries that had engaged in police reform, including Northern Ireland.461 However, unlike the Independent Commission on Policing for Northern Ireland, the NTPR consisted entirely of Kenyans and did not include any person with specific expertise in community policing. In October 2009, the NTPR released a report of its findings, commonly referred to as the Ransley Report in reference to the Taskforce Chair and included 200 458 Okia. 459 Monique Marks, "Policing for Democracy? The Case of the Public Order Police Unit in Durban," Africa Development 24, no. 1/2 (1999): 234. 460 Mageka. 461 Gastrow, 3. 181 recommendations for police reform. The Ransley Report identified several areas of concern for citizens, the police, and government. Specifically, they found there was a perception that the two police forces in Kenya (the Kenya Police and the Administration Police) were competing and had overlapping duties. There was also a perception of an increase in crime and complicity of the police in these crimes, as well as a perception of police corruption, a clear lack of resources for the police to perform their duties, low police morale, and an unacceptably high tolerance for corruption among the police.462 To address these issues, the recommendations of the Ransley Report focused on the concept of community policing in Kenya.463 The Police Reform Implementation Committee (PRIC) was established to coordinate and monitor the implementation of the Ransley Report’s recommendations and was composed of policing experts, government officials, members of the police, and civilians.464 The recommendations could be broken down into roughly four categories, all touching on aspects of community policing discussed in Chapter I of this dissertation – restructuring and decentralizing the police, focusing on police accountability and transparency, improving working conditions and training of the police, and improving police-community relations. Some of the specific recommendations included merging the Kenya Police with the Administration Police under one new National Police Service (NPS), creating a new National Police Service Commission tasked with developing 462 Ibid., 3-4. 463 Omeje and Mwangi Githigaro, 65-66; Charles Otieno et al., "Joint Response to the Cic on the National Police Service Proposed Regulations," (London: Usalama Reforms Forum, 2015). 464 Gastrow, 15. 182 training policies among other duties, and establishing a body charged with dealing with complaints against the new NPS.465 It is worth briefly highlighting the parallels between the process of police reform in Northern Ireland and Kenya up to and including the issuing of recommendations for reform. In both states, the approach to reform was participatory in nature. While there were relatively small teams of experts in the respective groups responsible for assessing the police prior to reform, in both states the community was actively engaged and encouraged to participate in the initial assessment. The Northern Ireland Commission included specialists in community policing who clearly influenced the specific recommendations made to meet the ideals of community policing. Kenya’s Taskforce did not have the same level of expertise, but they did specifically consult with Northern Ireland and reference their reforms in the analysis of Kenya’s police. And the recommendations themselves fall into similar categories that address those areas we would expect in community policing – retraining the police around policing with the community, devolving decision-making powers as much as is practicable, and community involvement in policing. In addition to these parallels, it is also important to highlight the differences between the circumstances of Northern Ireland and Kenya up to the point of police reform. Specifically, the tone of perceived police bias was different across these contexts. In Northern Ireland, the police were seen as biased in favor of the PUL community as demonstrated through the extreme imbalance between PUL and CNR police officers and 465 Mageka. 183 through the enforcement of discriminatory laws. In Kenya, the police were seen as corrupt, unprofessional, and biased in favor of the state as opposed to any specific ethnic group. The difference here is subtle, but important for police-community relations. In Kenya, the police have been an organization associated with the ruling entity of the state and enforcing the corrupt intent of those rulers. Since independence, this has translated into the police working to enforce the will of the government, regardless of the person in charge. In practice, this has led to a general bias in favor of the Kikuyu, though during Moi’s presidency there was also favoritism toward Kalenjin. This has led to ethnically biased policing in Kenya. While the police in Kenya and Northern Ireland were both accused of ethnically biased policing, the bias in Northern Ireland was consistently one against the Catholic minority. In Kenya the bias was more fluid and shifted depending on who was in power, though in practice this was usually the Kikuyu. This subtle difference in perceived bias may be the result of the historical structure of government, or regime type, in Northern Ireland and Kenya. Northern Ireland has consistently had a democratic form of government since its formation, even if that democracy was flawed. Kenya, on the other hand, could not really claim to be democratic until at least 1991 with the reintroduction of multi-party elections. Even then, the Kenyan government continued to be rated as a hybrid regime by the Economist Intelligence Unit’s Democracy Index. The form of government strongly affects the role of the police, as discussed in Chapter I of this dissertation. Importantly, in democracies, the legitimacy of the police is dependent upon community confidence in the police. In Northern Ireland, the PUL community had strong confidence in the police and saw the police as working for the people. In Kenya, where the police were created and developed under a colonial 184 then authoritarian regime, the police never gained legitimacy in the eyes of the people. Therefore, the police were not seen as serving the people, but the whims of the government. The police were not professionalized with clear duties and rules, but rather random enforcers of the governments’ wishes. In Northern Ireland, while the CNR community viewed the police to be biased arms of a biased state, the police were more professional in their duties. While both these situations called for reform of the government and the police to stop the ethnic violence, the possibility of reform without the need to address the deep corruption of the state and the police in Northern Ireland was far more plausible than in Kenya. That is, both states needed to address ethnic bias in government and in policing, but Kenya had the additional task of addressing the other forms of pervasive corruption that made the police essentially play things of whomever was in power. Restructuring and Decentralizing the Police The original recommendation to merge the two police forces of Kenya, the Kenya Police Force (Kenya Police) and the Administration Police, came from the Waki Commission Report on the investigation into the post-election violence. The report stated, “the Commission believes that developing a world class police service for Kenya would be best achieved through the creation of a single police agency. This means integrating the current two police organs.”466 The NTPR considered this recommendation, but also conducted a large number of interviews with various stakeholders including citizens and police officers who strongly opposed the complete 466 Waki et al., 436. 185 merger of the two forces. They ultimately recommended maintaining the two-police force structure under a single organizational umbrella called the National Police Service (NPS) because they found that the Administration Police had built relationships and were trusted particularly in rural communities, and the NTPR did not want to alienate these areas in the restructuring.467 They found many people they interviewed thought the Administration Police were “friendly, less bureaucratic in dealing with complaints, more accessible to the people, among other positive attributes.”468 This, it was argued, made the Administration Police more able to ‘police with the community.’ Logistically, the Ransley Report argued, it simply did not make sense to merge fully the two police forces as they had different mandates, training, and specialties and needed different reforms. For example, they were established under different laws: the Kenya Police Force and Kenya Police Reserve were created by the Police Act, CAP 84 Laws of Kenya in 1961 and the Administrative Police by the Administrative Police Act, CAP 85 in 1958. Both institutions had also been addressed in various constitutional reforms over the years, and each time the reforms maintained the two separate forces.469 It did not make sense, they argued, to create a brand new police service, or to merge the two, when their structures and officers had such disparate reform needs.470 Instead of a full merger, the Ransley Report recommended passing a new policing bill that would clearly demarcate the duties 467 Gastrow, 4. 468 National Task Force on Police Reforms (Kenya), "Report of the National Task Force on Police Reforms," (Nairobi, Kenya2009), 37. 469 Ibid., 13-23. 470 Ibid., 38. 186 of the two police services to avoid overlap, and create a new National Policing Council to oversee and coordinate both services.471 The Ransley Report also recommended the creation of a Police Service Commission whose duties would include recruitment, appointment, and promotion of officers, reviewing standards and qualifications for all members of both forces, and maintaining disciplinary control of the police, among others.472 This new Commission would help to create and maintain standards of professionalism that had been lacking in the police in Kenya as well as boost police morale. This proposed independent commission would be separate from the police and composed of persons of ‘high moral character’ including civilians, retired officers, and government officials. Additionally, the NTPR recommended that the hierarchy of the police be decentralized. They found that the highly centralized nature of the Kenya Police had led to a system of minimal accountability that also eroded the powers of provincial and local police.473 Other systems of policing the NTPR observed in their work had more decentralized systems, a change they argued Kenya needed to make if there was to be any improvement in the service of the police.474 One recommendation was the institution of Provincial Policing Authorities (PPAs) in each province to create support and trust from the people the police serve directly, as well as to increase accountability.475 These 471 Ibid. 472 Ibid., 43-44. 473 Gastrow, 6. 474 National Task Force on Police Reforms (Kenya), 46. 475 Ibid., 47. 187 civilian groups would create a space for the police to consult with the members of the PPA on provincial policing priorities476 as well as to monitor the progress on achievement of set goals and compliance with policing standards.477 The PPAs would also provide some legitimacy for the police through direct consultation with members of the community about community needs and concerns. Police Accountability and Transparency Accountability necessarily includes a lack of corruption. Issues of corruption and misuse of police authority have been historical problems for the police in Kenya. Police corruption can be understood, narrowly, as police “who use their position and authority for personal rather than public benefit. More broadly, corruption applies to any violation of rules even when there is no personal gain, as in perjury, physical abuse of prisoners, sexual misconduct, robbery, and racial profiling.”478 According to Kempe Hope, Sr., “by any measure or indicator, Kenya is regarded as one of the most corrupt countries in the world.”479 It is not surprising then that the Ransley Report made several recommendations aimed at reducing the presence and appearance of corruption in the police. The first of these was the recommendation to develop a new code of ethics that would highlight the ‘key values and principles’ of the police.480 This new code of ethics 476 Gastrow, 7. 477 National Task Force on Police Reforms (Kenya), 48. 478 Bayley and Perito, 3. 479 Hope, 89. 480 National Task Force on Police Reforms (Kenya), 69. 188 should center around the principles of Justice, Equality, Accountability, and Efficiency. While the NTPR left the specifics of the code of ethics up to the two police services to create, they named several key components including professional conduct, self-control, and avoidance of conflicts of interest.481 Many traffic officers, for example, owned or had stakes in matatus (small buses and minivans used for public transportation) which constitutes a clear conflict of interest.482 The Report specifically highlighted this common conflict of interest and recommended that all police should have to declare and have approved all business interests outside of policing and specifically that they should be prohibited from having any share of interest in public transport, towing services, private security, and weigh bridges.483 To address external oversight of the police, the NTPR recommended the creation of an Independent Policing Oversight Authority (IPOA) composed of civilians who had the power to monitor and investigate police activities.484 The IPOA would have far- reaching authority to investigate reports of police misconduct and to compel attendance at hearings. This arrangement, the NTPR argued, would create public confidence that police misconduct would be investigated impartially and that guilty parties would be held to account – building public confidence in the police in the long term.485 Specifically, the IPOA would have the authority to investigate allegations of misconduct by individual 481 Ibid., 69-71. 482 Gastrow, 8. 483 National Task Force on Police Reforms (Kenya), 71. 484 Ibid., 82. 485 Gastrow, 7. 189 officers or the police as a whole, taking complaints from the public or officers themselves, or they could initiate an investigation without an initial complaint if they saw fit.486 Though, importantly, they were only empowered to recommend punishment but not enforce these suggestions. Police Working Conditions and Training It was clear to the NTPR that a major issue with policing in Kenya was low morale among officers. The police themselves had stated in various reports and surveys that they wanted to work for a police service they could be proud of, that cares about them, that rewards good work and promotes people fairly and impartially, and provides adequate pay and equipment to do the job well.487 The NTPR, therefore, made recommendations focused on balancing these requests from the police with the public’s expectation that the police would be impartial, friendly, and respectful. One issue they focused on was the distribution of resources within the police. The Ransley Report noted that there was an unnecessary number of officers in the combined forces of the Kenya Police and Administration Police. In 2009, this number was at 73,956 according to the Kenya Police and Administration Police.488 This is a ratio of 1 officer to every 522 Kenyans.489 At the time, neighboring Tanzania had 1 officer for every 1,298 486 National Task Force on Police Reforms (Kenya), 83-84. 487 Ibid., 89. 488 Ibid., 99. 489 Africa Check, "Kenyatta's Numbers on Kenyan Police to Civilian Ratio Still Don't Add Up," https://africacheck.org/spot-check/kenyattas-numbers-kenyan-police-civilian-ratio-still-dont-add/. 190 citizens.490 In addition to the significantly larger force in Kenya, the police were also severely underequipped. Therefore, the Ransley Report recommended a temporary moratorium on police recruiting and that funds allocated for recruitment be redistributed to cover other important problems. They also recommended, when recruiting resumed, the standards should be raised for admittance to the police to include a requirement for more education and that training time should be increased in length from the then 9- month standard to a 15-month process with an end-of-training evaluation before admittance.491 Finally, lower ranking police in Kenya are required to reside in specified police housing in the station grounds for ease of deployment, while officers of higher ranks may receive a stipend to rent outside the stations. However, there is not sufficient housing provided for the police and their families, the housing provided is dilapidated and unsanitary, and the housing allowance provided to the higher-ranking officers is insufficient to rent decent accommodations.492 Therefore, it was recommended that the government prioritize improved housing for the police and increase housing allowances for the higher-ranking officers so they can afford to live in the communities they are serving.493 In addition to these recommendations, the NTPR recommended providing 490 Mary Kimani, "Security for the Highest Bidder," Africa Renewal Online, no. October 2009 (2009), https://www.un.org/africarenewal/magazine/october-2009/security-highest-bidder. 491 National Task Force on Police Reforms (Kenya), 100-01. 492 Ibid., 114. 493 Ibid., 117-18. 191 adequate life and medical insurance, developing a system for leave accrual and use, and increased salaries, among other improvements.494 Police-Community Relations The final category of recommendations dealt with improving police-community relations. While the above recommendations all serve to improve service and reduce barriers between the police and the community, the NTPR made recommendations related to the perceptions of the police “by many Kenyans as a coercive instrument of the state that serves the narrow interest of the elite.”495 Specifically, the NTPR focused on the idea of community policing as presented by Amnesty International, where community policing is defined as A collaborative effort by the police and the community that identifies problems of crime and disorder and involves all elements of the community in the search for a solution of these problems. This approach to policing is based on the assumption that the police alone cannot control crime, but require the active support of the community, to prevent, detect crime, reduce fear and crime and improve communication between the community and the police.496 Keeping this definition in mind, the Ransley Report recommended the development of Community Police Committees (CPCs) at several different levels of the community from district level down to the block level of 5-10 households.497 The need to reform the police was not a new idea in reaction to the post-election violence, but something that had been discussed for some time. The results of some of these 494 Ibid., 118-39. 495 Ibid., 171. 496 Anneke Osse, "Understanding Policing: A Resource for Human Rights Activists," (Amsterdam: Amnesty International Nederland, 2006). 497 National Task Force on Police Reforms (Kenya), 174. 192 discussions was a push to experiment with the concept of community policing and the first experiment in community policing was launched in Ruai in 2005. Since that initial experiment in April 2005, the Kenyan government had been committed to developing better policing through community involvement.498 The Ransley report, noting this commitment, recommended that the National Policy on Community Policing be fast- tracked and include the community in its development process.499 Implementation The Ransley Report, much like the Patten Report in Northern Ireland, was the primary driver of police reforms in Kenya. Like the Patten Report, it also did not have the weight or force of law and required legislative action to ensure the recommendations were adopted in part or in whole. The 2010 Constitution laid out the legal framework for the restructuring of the police. Specifically, Article 243 of the 2010 Constitution established the new National Police Service (NPS), which included the Kenya Police Service (KPS) and Administrative Police Service (APS) under the authority of one Inspector-General (IG) and two Deputy Inspector-Generals responsible for the day-to-day operations of the two services. In addition to the 2010 Constitution, the National Police Service Act, 2011 (NPS Act 2011) effectively gave the power to the new NPS and the IG to implement the new structure and rules of the NPS. Under the NPS Act 2011, the IG was given a variety of powers and responsibilities. These included the responsibility to implement policy decisions and to coordinate and audit all police operations. The IG was also required to keep the 498 Ibid., 173. 499 Ibid., 182. 193 government informed on policing matters and provide advice when called for. The IG was responsible for the organization and maintenance of police stations as well as the structure and distribution of the police force. The IG was tasked with implementing and monitoring the success of community policing initiatives as well as implementing the decisions of the IPOA, including compensating victims of police misconduct when appropriate. The IG was to establish the internal affairs unit to investigate internal matters of police misconduct. In addition to these and many other duties, the IG was also responsible for compiling data and statistics relevant to the police including statistics on crime, public perceptions of the police, and demographic statistics.500 It is worth highlighting the creation of the internal affairs unit (IAU) under the IG’s authority. The IAU was required to be separated physically from the rest of the NPS in different buildings. They had the authority to investigate complaints received from members of the public of from police officers. They were also to conduct investigations at the request of a senior officer, the IG, or at the recommendation of the IPOA. It has the power to recommend disciplinary actions from reprimands, to suspensions, to firings to the IG. They are also tasked with maintaining records of all complaints received.501 In addition to this internal, though independent, oversight group, the 2010 Constitution also created the IPOA, which was effected through the Independent Police Oversight Act, 2011.502 The IPOA was intended as a mechanism to hold the police 500 Republic of Kenya, "National Police Service Act, 2011," Republic of Kenya (Nairobi, Kenya: Republic of Kenya, 2011). 501 Ibid. 502 "Independent Policing Oversight Authority Act, 2011," (Nairobi, Kenya: Republic of Kenya, 2011). 194 publicly accountable as well as create transparency.503 Essentially, this agency was created to ensure that the work of the IAU was transparent to the public, and not just the NPS. This is why the IPOA has the power to conduct its own investigations, whether as the result of a specific complaint received by the IPOA or through their own initiative. They are also empowered to make disciplinary recommendations based on their investigations and to make both those recommendations and their responses public. They are also empowered to maintain their own records of any investigations that IAU undertakes as well as make recommendations to other government agencies, including the judiciary.504 The NPS Act 2011 also specified the distinct roles of the KPS and the APS. These distinctions are minor, but important. Like the differences between federal and local police in the United States, both forces have the responsibility to serve and protect the public, but they have slightly different emphasis in how they do this. The KPS was assigned the duties of protection of the public, prevention and investigation of crime, general maintenance of law and order, and the apprehension of suspects. The APS – akin to a United States federal police force – while also tasked with the general maintenance of law and order and protection of the public, was given the additional duties of border patrol and security, protection of government property, special stock theft protection, and coordination with other government agencies in conflict-management.505 These duties, while defined as distinct by the language of the Act, do overlap somewhat in reality. In 503 Ibid. Section 5. 504 Ibid. Section 6. 505 "National Police Service Act, 2011." 195 general, any officer whether NPS or APS is expected to help maintain the general sense of law and order. However, the distinction between the two is similar to that of the difference between federal and local police in the United States. The APS is tasked with protecting the border, protecting government property, supporting government agencies in the enforcement of regulations, and coordinating with other government agencies in peace building and conflict management. The KPS, on the other hand, is more locally focused, with their duties concentrated on preventing and prosecuting regular crimes within the populace.506 To address some of the morale issues highlighted in the Ransley Report, the 2010 Constitution also created the structure of the National Police Service Commission (NPSC), officially given power through the National Police Service Commission Act, 2011.507 The NPSC is charged with developing the basic standards for evaluating police officers, evaluating and determining promotions within the NPS, and recruitment of new officers.508 It is important to note that while the NPS Act 2011 created and laid out the specific organizational structure and duties of the PPAs, eventually renamed County Policing Authorities (CPAs), these were never formed. These structures, arguably the reform that most specifically adhered to the ideals of community policing through 506 Kempe Ronald Hope, Sr., "In Pursuit of Democratic Policing: An Analytical Review and Assessment of Police Reforms in Kenya," International Journal of Police Science & Management 17, no. 2 (2015): 93. 507 Republic of Kenya, "The National Police Service Commission Act, 2011," (Nairobi, Kenya: Republic of Kenya, 2011). 508 Ibid. 196 community input, were never embraced, and therefore never fully understood by the community. This will be discussed further below. Results An initial assessment of the state of policing in Kenya shows that the reforms implemented in the past few years have not been successful in engendering a greater degree of community confidence in the police. This is not to say that they have been completely unsuccessful in accomplishing some of the other goals highlighted in the Ransley Report. In the first five years since the implementation of the NPS Act 2011, the NPS structure was established. This included many important updates such as creating a centralized set of emergency numbers for the public to contact the police, the development of strategic plans for the many different structures in the NPS, and the development of a better communications strategy.509 Theoretically, these updates should have improved community relations as they were built to make communication between the police and the community easier and more transparent. Other changes included several new policies including some relating to gender and respect for human rights in policing. The Police Code of Conduct and Ethics was also revised and included several clauses on professionalism and impartiality in policing. Specifically, police are barred from showing their political affiliation through actions in favor of a specific candidate or political party.510 Similarly, there was a specific focus on conflicts of interest and a series of rules were developed to prevent police from owning side-businesses and accepting 509 Hope, Police Corruption and Police Reforms in Developing Societies, 100. 510 Cyrus Ombati, "Don't Express Your Political Affiliations, Police Now Told in New Code of Conduct," Standard Digital (2016), https://www.standardmedia.co.ke/article/2000192134/don-t-express-your- political-affiliations-police-now-told-in-new-code-of-conduct. 197 graft.511 Police training has been updated to include anticorruption and ethics, and the welfare of officers has been improved through new housing and comprehensive life and health insurance programs.512 However, despite these achievements, according to the 2014/2015 Afrobarometer survey only 36% of Kenyan’s said they trust the police ‘a lot’ or ‘somewhat,’ an assessment of the police which is not universal in East Africa where on average 57% of East Africans say they trust the police ‘a lot’ or ‘somewhat.’513 Why do Kenyans have a particularly low opinion of the police, especially given the amount of time, money, and effort that was expended on reforming the police? Comparing the police reform implementation process in Northern Ireland and Kenya suggests several possible answers. The Kenya case highlights the importance of police professionalism, and specifically addressing police corruption. The history of the deep corruption and lack of professionalism of the police in Kenya made it particularly difficult for the people of Kenya to be convinced that police reforms would lead to real changes in the competence and trustworthiness of the police. Harkening back to the discussion of the definition of democracy in Chapter I of this dissertation, I asserted that democracy as it is defined for the purposes of this dissertation includes the procedural definition of democracy 511 Ibid.; Hope, Police Corruption and Police Reforms in Developing Societies, 101; Elvis Ondieki, "Police to Sign New Rules to Guide Their Conduct on Duty," Daily Nation (2017), https://www.nation.co.ke/news/New-rules-to-usher-in-reforms-in-police-force/1056-3846142- jlp2rd/index.html. 512 Hope, Police Corruption and Police Reforms in Developing Societies, 100-01. 513 Stephen Buchanan-Clarke and Rorisang Lekalake, "Violent Extremism in Africa: Public Opinion from the Sahel, Lake Chad, and the Horn," ed. Afrobarometer (2016), 7. 198 (elections, suffrage, and some civil rights) plus accountability. While the CNR community in Northern Ireland prior to and during the Troubles would likely have argued that the police were biased in their application of the law, the United Kingdom and Northern Irish governments were not accused of the more blatant forms of corruption such as bribery and graft. The Kenyan government has a history of and reputation for extreme corruption including favoritism and patronage in government as well as a long history of graft and extortion by the police. Therefore, Kenya needs to demonstrate accountability and reduced or eliminated corruption before community policing can be effective. This will be discussed further in the conclusion chapter of this dissertation. I would also argue that Kenya struggled to make the kind of gains seen in Northern Ireland’s community confidence in the police because they did not actually implement community policing. As the Usalama Reforms Forum noted in 2015, “a major problem with the reform process is that none of the strategies employed so far have aimed at critically transforming or altering the fundamental principles of policing in Kenya.”514 In other words, the police experimented with the concept of community policing but never fully implemented it. There was no change to the structure and form of the police to reflect a community policing service. This can best be understood by looking at the expectations of community policing laid out in Chapter I of this dissertation. First, symbolic change has been at best non-existent, and at worst a cause for distrust rather than an opening for new relationships. While the Kenya Police and 514 Otieno et al., 2. 199 Administration Police were nominally merged under the new NPS, the people of Kenya, and the police themselves, have maintained the old names and have not adopted this new name in their day-to-day interactions. Further, the uniforms of the KPS and APS have remained the same since the change. Additionally, there was no change in the people who were either KPS officers or APS officers. The names, appearance, and individuals involved in the police all remained the same. This may be part of the reason why the “service has struggled to transform its image and overcome the factors that constrain its improvement and development.”515 There was no visible signal that things ever changed for the average Kenyan. There are other examples to suggest the lack of symbolic change in the police has amplified the negative perception the community has of the police. In their interviews with the public, the NTPR identified a widespread perception, and much evidence, of corruption of the police. Specifically, it was clear that the police used roadblocks to extort bribes from motorists and the citizens of Kenya. During their investigative stage, the NPTR issued an initial report that suggested all roadblocks in Kenya be removed unless they served a specific safety or traffic control purpose.516 The police did remove a significant number of those roadblocks by the time the final report was issued.517 However, when I traveled to Kenya in winter of 2016/2017, I observed many roadblocks all over the capital city of Nairobi and was told in casual conversation with the locals that the police regularly used these roadblocks to extort money from the people. The 515 Ibid., 4. 516 Gastrow, 8. 517 Ibid. 200 resurgence of roadblocks as a source of extortion by the police would certainly signal to the community that the ‘new’ police is no different from the old. Further, police officers owning matatus continues to be an issue, at least within Nairobi. When I was riding these matatus to get around the city, I had several locals advise me to ride certain matatus over others. They told me that certain companies were owned in whole or in part by members of the police and those matatus were guaranteed uninterrupted passage through the city. On the other matatus, the drivers had to pay off the police early in the morning, an important ritual to ensure that the matatus could run the remainder of the day without the police boarding and extorting money from the passengers. The second set of expectations around community policing involve the changes to the institutions and practices of policing. Many of the reforms suggested in the Ransley Report for Kenya that seem to parallel those in Northern Ireland lack important components that make them community policing. For example, the CPAs in Kenya seem very similar to the DPPs in Northern Ireland in form. CPAs are supposed to have several police representatives and at least six citizens appointed by the Governor of the County. These CPAs are supposed to be tasked with monitoring the local police and providing quarterly reports to the IG on the compliance of the local police with stated goals and police rules.518 However, as mentioned above, even though CPAs were created by the NPS Act 2011, they were never enacted. In 2016, President Uhuru Kenyatta, Kenya’s fourth president and son of the first President Jomo Kenyatta, announced that he wanted to ‘activate’ the CPAs to help assess and address local issues of security, but nothing has 518 Republic of Kenya, "National Police Service Act, 2011." Section 41. 201 yet happened.519 The idea of CPAs as forums for community input into policing also seems to have been lost in the years since the Ransley Report was issued. Many communities and officers are afraid that the CPAs would overlap duties with the police and interfere in the provision of security.520 If the original purpose of CPAs was community engagement and oversight of the police, it has been lost in translation. Similarly, the oversight mechanisms of the NPS are independent in name, but not in function. The IPOA is empowered to investigate the police and make recommendations for discipline, but those recommendations do not have any force and are made to the IG. In other words, while the investigations may be independent, the ultimate responsibility for enforcing discipline remains internal to the NPS. Further, legislative updates to the NPS Act 2011 have worked to weaken the independence of the IG, who is now considerably more dependent on the executive for his recruitment and retention. This, as well as actions by acting IGs have led to very low community confidence in the IG. Only 12% of respondents in a March 2014 survey said they had confidence in the IG.521 In addition to the lack of confidence in the independence of the IPOA and the IG, the community has also seen little to no improvement in the day-to-day actions of the police themselves. As Anneke Osse stated, “police are still ineffective, unjust in their 519 Cyrus Ombati and Roselyne Obala, "Governors Demand Operationalisation of County Policing Authority as Police Bosses "Resist"," Standard Digital (2018), https://www.standardmedia.co.ke/article/2001264810/governors-demand-operationalisation-of-county- policing-authority-as-police-bosses-resist. 520 Ibid. 521 Osse, 913. 202 operations, actions and decisions, and lacking in accountability.”522 Despite the overtures to community policing, the NPS is still a reactive force, focused on punishing crime rather than working with the community to prevent it. Crime is underreported and there remains a general lack of confidence in the police from the community.523 Police continue to be accused of actively committing (rather than preventing) crime, using excessive force, and lacking general competence in providing security.524 There are several other reasons that police reforms have not taken hold or translated into community confidence in Kenya that are unrelated to the lessons from Northern Ireland. In the same Afrobarometer survey from 2014/2015, 86 percent of Kenyans said they trust the armed forces ‘a lot’ or ‘somewhat.’525 So there is not a general distrust of the government or security services, but something specific about the police that Kenyan’s do not trust. Why? The most glaring answer seems to be corruption. In the same survey, 69 percent of Kenyans said they thought ‘most’ or ‘all’ of the police were corrupt.526 From the above examples of extortion of motorists and matatu operations, this explanation seems compelling. While there was a general distrust of the police in Northern Ireland by the CNR community, the corruption allegations primarily stemmed from accusations of colluding with or covering up for loyalist paramilitary actions, and not from accusations of extortion. The Kenyan police faced similar 522 Ibid., 916. 523 Otieno et al., 5. 524 Osse, 916. 525 Buchanan-Clarke and Lekalake, 9. 526 Ibid., 31. 203 accusations of bias and even collusion surrounding the 2007 post-election violence, but this was compounded with the additional history of petty corruption in the form of extortion. It seems that addressing additional forms of corruption outside assumed ethnic bias is an important factor in building community confidence after ethnic conflict. It was apparent in my limited conversations with Kenyans and through my own observations that there is a deep mistrust of the police in Kenya and that many locals do their best to avoid any and all interactions with the police whenever possible. When interactions with the police cannot be avoided, the default position is to expect extortion and hope that it is affordable. According to Transparency International Kenya, from 2010-2014 the police were the most bribery prone institution in Kenya.527 Corruption in the police is a huge problem that may be the biggest barrier to building community confidence in the police in Kenya. Hope has done considerable and convincing work on the role and pervasiveness of corruption in the police in Kenya. He argues that, though the reforms that have been put in place have been successful in many respects (the formation of NPS created a centralized command for the police, new codes of conduct were developed, an anti-corruption strategy has been developed, community policing strategies were developed, conditions of police officers have improved, and there is greater representation of women in the police, among many other reforms),528 corruption 527 Transparency International Kenya, "The East African Bribery Index Trends Analysis (2010-2014)," ed. Transparency International (Kenya2015), 23. 528 Hope, "In Pursuit of Democratic Policing: An Analytical Review and Assessment of Police Reforms in Kenya," 95. 204 remains one of, if not the biggest impediment to building community confidence in the police.529 There are several other reasons that may explain why the police in Kenya have been unsuccessful in engendering public trust. It is important to acknowledge the variety of other variables that make the situations in Northern Ireland and Kenya drastically different. First, Northern Ireland, while a separate and unique political entity, is still a part of the United Kingdom and therefore has the pressures and resources associated with the wealthy nation. Kenya, as an independent former colony of the United Kingdom, has many of the same political institutions of Northern Ireland, but does not have the economic resources or same political pressures as Northern Ireland. Which is not to say that they do not face strong political pressure from other nations in Africa, especially its neighbors and partners in the East African Community – just that those pressures are different. There are a great many other differences between the countries including ethnic diversity. Northern Ireland has two primary ethnic groups and a small number of other minorities whereas Kenya has 23 acknowledged ethnic groups and up to 47 more that people identify with in the country. In his seminal work on ethnic conflict, Donald Horowitz argues that highly homogenous and highly heterogeneous societies are less violent, whereas states experience more violence when large ethnic minorities compete with an ethnic majority.530 From this basic assertion, Jose Montavalo and Marta Reynal- 529 Hope, Police Corruption and Police Reforms in Developing Societies. 530 Horowitz, Ethnic Groups in Conflict. 205 Querol (2005) and Reynal-Querol (2002)531 develop a measure of ethnic polarization which they use to measure the ethnic distribution in a society – focusing not just on the number of ethnic groups represented in a state, but their relative size within society. They find this measure important in determining both civil war onset532 and civil war duration.533 It is reasonable to extrapolate from this that the effectiveness of post-war settlements may also be affected by ethnic polarization. It may be that the highly polarized Northern Ireland is more likely to be successful in establishing post-war stability because the negotiating parties represent such a large portion of the people of the country where in the highly fractionalized Kenya, negotiators will be representing a far smaller percentage of the ethnic minorities in the country. It could be that the demographic make-up of Kenya is such that implementation of any changes is going to be more difficult because of the need to navigate the different needs and perceptions of a far greater number of relatively small ethnic groups than most other countries recovering from ethnically based violent conflict. Another factor may be the level of commitment the government has made to support the changes to the police. In Northern Ireland, there has been consistent and increasing support for the changes to the police shown through legislation and participation in policing institutions by all the major political parties. In Kenya, however, there has been some erosion of the support for the changes. For example, in 2014 and 531 Montalvo and Reynal-Querol; Marta Reynal-Querol, "Ethnicity, Political Systems, and Civil Wars," Journal of Conflict Resolutions 46, no. 1 (2002). 532 Montalvo and Reynal-Querol. 533 "Ethnic Polarization and the Duration of Civil Wars," Economics of Governance 11, no. 2 (2010). 206 2015 several amendments were made to key pieces of legislation including the National Police Service Amendment Act (2014), the National Police Service Commission Amendment Act (2014), and the miscellaneous Amendment Bill 2015 which reduced the independence of the IPOA, made appointments dependent on the president, and eroded the independence of other portions of the police service.534 When the changes were given little time or support to take effect and genuinely alter the relationship between the community and the police, it is not surprising that they failed. Further research is necessary to understand the relationships between the factors described here. Developing a deeper understanding of the lack of progress in building community confidence in the police in Kenya would require extensive interviews with the people and police officers of Kenya. This is a project I hope to attend to in the near future. Conclusion Northern Ireland’s implementation of community policing has been held up as an example of how cities and countries experiencing violent ethnic conflict can build community confidence in the police. Kenya following their 2007 post-election violence is an example of a country that specifically looked to the experiences of Northern Ireland for inspiration and guidance in their own police reforms. Despite the many similarities in the structure and form of the police reforms between Northern Ireland and Kenya and the explicit focus on community policing in both cases, the success of those changes has been drastically different. Northern Ireland significantly enhanced the level of community confidence in the police overall and specifically within the 534 Kenya, 24. 207 Catholic/Nationalist/Republican community while Kenya has one of the lowest levels of confidence in the police of any African nation, and the absolute lowest in East Africa.535 A closer examination of the changes that were recommended and enacted in Kenya reveals some significant differences from Northern Ireland. Though the Kenyan reforms were explicit in their intent to engage community policing, they did not do this. There was little attention paid to issues of the symbolism of the Kenyan police. This meant that the people of Kenya did not have any visual clues that the police before and after the changes were any different. The same people were officers, wearing the same uniforms and badges, with a name so insignificantly different from before that most people including the police themselves never adopted the new name in everyday conversation. The structure of the police and the practice of policing in Kenya also did not see significant changes. The reorganization of the police created some superficial places for citizen input and oversight, but it appears these changes were not seen or felt by the average Kenyan citizen. The CPAs were never enacted, and the people who were recommended to serve on the IPOA were all people of prominence in the community, immediately excluding access to average citizens. Additionally, while Northern Ireland took the extra step of ensuring public access to some portion of these groups’ meetings, the same step was not taken in Kenya. So rather than inspiring confidence that the IPOA was promoting the interests of the community and holding deviant officers to account, there is the appearance of yet another group that can be corrupted since their operations are not open to public observation. Maybe the biggest difference between Northern 535 Buchanan-Clarke and Lekalake. 208 Ireland and Kenya is the widespread perception of the police as a corrupt institution with corrupt people at every level. If the institution and the people in it are all irredeemably corrupt, it is not hard to understand why the Kenyan people would be unable to learn to trust the police. 209 CHAPTER VI CONCLUSION Introduction This dissertation has combined the literature on post-ethnic conflict state rebuilding with the literature on community policing to assess how community policing can work to build community confidence in the police in post-ethnic conflict societies. In bringing the logic of community policing to post-ethnic conflict rebuilding, I highlighted the ideals of community policing and how these can be especially useful in building relationships between previously marginalized communities and the police after conflict. I examined the implementation of community policing in Northern Ireland after the Troubles and showed how the commitment to the ideals of community policing led to significant increases in community confidence in the police. I also highlighted the importance of implementing community policing in a holistic manner through a first-look analysis of Kenya after the 2007 post-election violence where community policing was not implemented in full. The post-conflict literature often identifies personal safety in the aftermath of ethnic conflict as the primary concern for individuals. This security can be provided in several ways: through international peacekeeping forces, domestic militaries, or civilian police. Those that focus specifically on post-ethnic conflict security tend to emphasize the importance of democratic policing and building community confidence in the police, but they do not tell us how this is accomplished. Policing literature, however, suggests that community policing is designed specifically to build relationships between the police and the communities they serve. This method of policing, then, may be the specific solution that is missing in the post-ethnic conflict rebuilding literature. 210 After their more than 30-year violent ethnic conflict known as the Troubles, Northern Ireland focused on police reform as an important step in their rebuilding process. The Belfast/Good Friday Agreement included a provision that created the Independent Commission on Policing for Northern Ireland, a group assembled to assess the state of the police in Northern Ireland and make recommendations for reform. The Commission included two prominent experts on community policing who clearly had a strong influence on the shape of police reform. The recommendations focused on several areas of reform that we expect in community policing: reorganizing the police to allow more decentralized decision-making processes; introduction of structures to encourage community involvement in identifying and deterring problems; and changes to the symbolism and make-up of the police force among others. These changes and the continued commitment to the ideals of community policing in Northern Ireland led to a significant increase in marginalized community confidence in the police in the first ten years of the new service that have continued to improve into the second decade of the service. The success of community policing in Northern Ireland has encouraged other states to embrace community policing after violent ethnic conflict. Kenya specifically examined the experiences of Northern Ireland when looking to reform their own police after the 2007 post-election violence. On the surface, at least, Kenya’s process of reform looked remarkably like Northern Ireland’s process. They commissioned a National Taskforce on Police Reforms like Northern Ireland’s Commission and studied the police and made recommendations for reform which emphasized community policing. However, despite the explicit focus in the recommendations, Kenya did not fully 211 implement community policing and was unable to do so because of the deep history of corruption in the police. I agree with Hope, Sr. and others that Kenya needs to address this underlying problem of corruption and lack of accountability in the police before community policing efforts can be successful in engendering community confidence. Examining this additional hypothesis will require further research. The remainder of this chapter explores the usefulness of my argument. I discuss some possible cases where community policing might be applied to build community confidence in the police in ethnically divided societies, including those where ethnic-civil war has not yet arisen. I also address the limitations of community policing in solving the problems of ethnic tensions. Finally, I discuss my agenda for further research and next steps for this project. Extensions and Practical Implications In Chapter I, I explained that I specifically address policing reform as a way to build confidence in the state in in post-ethnic conflict societies that choose democratic forms of government where democracy is defined as ‘procedural plus accountability.’ In other words, the scope of my argument is limited to post-conflict democracies. There are two primary reasons to focus on such democratic states. First, authoritarian states have other options for ensuring safety and security of its citizens after ethnic conflict outside of a civilian police force. Many states, including the Democratic Republic of Congo, have relied on some combination of state military and international peacekeeping forces in establishing and maintaining peace.536 In these situations, it is not necessary for 536 Ambassade de la Republique Democratique du Congo, "Overview of Security Efforts by the Democratic Republic of Congo," (Washington, DC). 212 individuals and civilians to have ‘confidence’ in the capabilities and lack of bias in the police; it is enough that people believe the threat of force credible to prevent new outbreaks of violence. Democracies that at least meet the minimal definition of ‘procedural plus’ used in this dissertation do not have this option. Instead, they face the difficult work of developing systems of policing that require community confidence to function effectively. Groups and individuals in society must believe in the ability of the police to apply the law fairly and without bias across ethnic groups if the police are to be effective. This may be an important, and to this point not fully explored, explanation for the lack of success of community policing in Kenya. As discussed in the results section of the Chapter V of this dissertation, Kenya did not really enact community policing. They did not engage with symbolic reform and did little to change police practice to conform to the ideals of community policing. Community confidence in the police also requires accountability and a lack of corruption. Meeting the definition of ‘procedural plus’ democracy used here would then be an additional necessary condition for the success of community confidence in the police. Many scholars would argue that Kenya should be defined as either a hybrid regime, along the lines of a competitive authoritarian regime or a clientalistic regime. The examination of Kenya provided in this dissertation seems to show that, at the very least, it does not meet the definition of ‘procedural plus’ democracy as I have explained it. Kenya does not have a history of accountability and much of that was left unaddressed in the reforms of the 2010 Constitution. Kenya has been marked by clientalistic behavior since independence. The use of land sales to purchase the political loyalty of wealthy Kikuyu has been a trademark of every government in Kenya since 213 independence. In this setting, the police were charged with protecting the property rights of the wealthy and did not have nor need the confidence of the community to do this effectively. Further, at the time of the writing of this dissertation, some have accused the current government of Kenya of sliding back in to authoritarianism due to crackdowns on the media and judiciary.537 As I have emphasized several times in this dissertation, one key component to the success of community policing is time for the changes to take effect and be experienced by the community. If community policing efforts are stalled or reversed because of regime instability, then they are likely to fail. This, however, does not make the examination of Kenya less important, but rather better demonstrates why democracy may be a precondition for the success of community policing, and therefore appropriately narrows the scope of my argument here. In Chapter I, I argued that I did not want to overly limit the scope of my argument so that there were too few cases to which it applied. However, the first-glance at Kenya given in Chapter V makes the case that my scope needs to be more limited. It is not enough to be a procedural democracy with elections and some civil rights. The Kenyan case, instead, suggests that there are some additional criteria that imply a deeper democracy than is present in Kenya for community policing to succeed. The second reason to focus on democratic states is that there is a normative preference for democracy in the international community. Diplomatic and military pressure has regularly been employed, specifically by western powers, to encourage or 537 "Is Democracy in Kenya at Risk?," The Week (2018), http://www.theweek.co.uk/92642/is-democracy- in-kenya-at-risk; Neha Wadekar, "The Future of Kenya's Democracy Is Hanging in the Balance," The Atlantic (2017), https://www.theatlantic.com/international/archive/2017/11/kenya-election-kenyatta- odinga-violence/545747/; Kevin J. Kelley, "Democracy on Decline in Kenya, Ngo Finds," Daily Nation (2018), https://www.nation.co.ke/news/politics/Democracy-in-Kenya/1064-4266504-btxrt6z/index.html. 214 demand democratic reforms in non-democratic states. While attempting to stay out of the debate on the normative preference for democracy, this trend indicates that democracies are likely to be attempted in states that are reorganizing post-ethnic conflict. If this is the case, then it is important to understand how these states can and should reform policing to gain community confidence and function effectively in a democracy. Community policing, then, should work to build community confidence in the police in other cases where states are attempting to rebuild democracies after violent ethnic conflict. For example, Sri Lanka experienced a long and violent civil war from 1983 until the ceasefire in 2009 between the Tamil minority population and the Sinhalese majority population.538 Similar to the case of Northern Ireland, this conflict ended with no changes to the borders of Sri Lanka and a democratic government.539 The grievances of the Tamil also sound similar to those of the CNR community in Northern Ireland prior to the outbreak of war. There was a dispute over the legal status of Indian Tamils in Sri Lanka as well as grievances over Sinhala being declared the official language of independent Sri Lanka, leaving Tamil-speaking minorities out.540 Other policies that were considered discriminatory against the Tamil population including some relating to access to education all contributed to the outbreak of civil war.541 During the war, the military 538 Jayshree Bajoria, "The Sri Lankan Conflict; Backgrounder," (2009), https://www.cfr.org/backgrounder/sri-lankan-conflict. 539 Central Intelligence Agency, "The World Factbook: Sri Lanka," https://www.cia.gov/library/publications/the-world-factbook/geos/ce.html. 540 Bajoria. 541 Ibid. 215 was accused of participating in extra-legal violence against the Tamils,542 which led to great distrust of the security forces in the aftermath of war.543 The similarity of these circumstances to those of Northern Ireland make Sri Lanka an excellent case to test my arguments about community policing further, especially as we approach the ten-year anniversary of the ceasefire that ended the violence. The most important implication of this study and the arguments developed here are the practical applications of the lessons learned. If good and successful police reform comes from a focus on community policing, then individuals charged with developing plans for post-conflict state rebuilding will know where to focus their efforts in police reform. Similarly, international sponsors of reform and international organizations that lend expertise to such efforts will have something to guide their efforts as well. For example, Cyprus has made recent strides in working toward a peace process and a unified government between Greek-Cypriots who currently hold power in the southern part of the island and Turkish-Cypriots who declared independence in 1983 (though it has only ever been acknowledged as independent by Turkey). In the past few years, several rounds of negotiations have been held to attempt to politically unify the island.544 If the Cypriots succeed in forming a new, unified and internationally 542 Amnesty International USA, "Sri Lanka - the Northeast: Human Rights Violations in a Context of Armed Conflict," Amnesty International USA, https://web.archive.org/web/20090214094900/http://www.amnestyusa.org/document.php?lang=e&id=5E7 F89A231872062802569A600601598. 543 Johann Rebert, "Community Policing in Sri Lanka: A Foundation for Wider Police Reform?," (2017), https://eu-civcap.net/2017/03/01/community-policing-in-sri-lanka-a-foundation-for-wider-police-reform/. 544 "Cyprus Talks End without a Peace and Reunifications Deal," BBC News (2017), https://www.bbc.com/news/world-europe-40530370; Staff Reporter, "After Macedonia Deal, Greece and Russia Call for Cyprus Solution," Cyprus Mail Online (2018), https://cyprus-mail.com/2018/06/13/after- macedonia-deal-greece-and-russia-call-for-cyprus-solution/. 216 recognized government, police reform will surely be a priority for the new state. Reform should focus on community policing which would include creating new police symbols that are culturally sensitive and inclusive of the major ethnic groups on the island, developing programs for community involvement in the police, and decentralizing police command structures to allow for local expertise in decision-making. Since all democracies require community confidence in the police for the police to maintain legitimacy and function effectively, it follows that any democracy experiencing a lack of community confidence in the police could benefit from embracing community policing – even without extreme violent ethnic conflict. For example, in the past few years, several police forces across the United States have experienced a severe lack of community confidence, especially among the African American population.545 In cities like Chicago, Baltimore, and St. Louis where there have been several high-profile incidents of the police killing African American men, communities have protested and made clear that there is deep distrust of the police by the community. This is also evidenced by the Black Lives Matter movement, which began in 2013 in response to the acquittal of George Zimmerman in the murder of Trayvon Martin. This movement seeks to raise awareness of the disproportionate number of African American people who are pulled over, arrested, and prosecuted by the police.546 This situation clearly fits a similar model to the one we have seen in Northern Ireland. The police are perceived as biased against one community (African Americans), 545 Rich Morin and Renee Stepler, "The Racial Confidence Gap in Police Performance," (2016), http://www.pewsocialtrends.org/2016/09/29/the-racial-confidence-gap-in-police-performance/. 546 "Herstory," Black Lives Matter, https://blacklivesmatter.com/about/herstory/. 217 a different portion of the community sees the police as ‘protecting’ them (see the Blue Lives Matter Movement),547 and the result is a portion of the community that has lost confidence in the police. The police face a difficult situation in that it is not possible to be truly effective if a portion of the community does not have confidence in you. They will be less likely to call the police when issues arise in the neighborhood and may even turn to extralegal ways of addressing crime. In Northern Ireland, the CNR community would often turn to paramilitary groups to handle issues of crime rather than the police. It is possible that people in the African American community in Chicago, Baltimore, and St. Louis would feel more comfortable turning to local gangs to handle issues. Even if they do not take this extreme step, the police are unable to do their job when the community does not feel comfortable calling for help when they otherwise would. The circumstances of these communities seem like a good fit for implementing community policing. And, indeed, many of the original experiments in community policing began in places like Chicago. However, those experiments were small and many lost funding after a short time. The success of a long-term commitment to community policing in Northern Ireland should encourage new and larger community policing initiatives elsewhere. It is reasonable to think that reforms to the Chicago and St. Louis police forces that focus on developing better relationships with the community could make major inroads into building African American community confidence in the police as they have in the past. 547 Blues Lives Matter is a countermovement to the Black Lives Matter movement, in which the members seek, among other objectives, to highlight murders of police and first responders. More information can be found on their website, https://www.themaven.net/bluelivesmatter/. 218 The specifics of what those reforms look like will need to be tailored to the experiences and needs of those communities, but a focus on ideals of community policing should open the space for the community to be willing to accept reform as genuine. For example, it may not be necessary to change the badge of the Chicago police, but it may be worth exploring the level of militarization of their uniforms or vehicles. Would it be possible to wear gear that is less militaristic while also maintaining officer safety? When performing police activities in African American neighborhoods, could they ensure that code names are less militaristic? One civilian in Belfast, Northern Ireland specifically mentioned the alienation he experienced when he heard an officer refer to a set of wellness checks in his neighborhood as an ‘Operation.’ He thought the term ‘Operation X’ sounded like a military endeavor, not a police initiative.548 There are any number of symbolic changes that might be made in these U.S. cities that could signal to the African American community that reform is real. Undoubtedly, there are also police structures and practices that would help to build community confidence if enacted. Again, without presuming to prescribe the necessary changes without expertise in the area in question, one could imagine that regular and open forums for the community to present their problems and help develop solutions in conjunction with the officers in their neighborhoods could help to build community confidence in the police. When the community feels heard and sees the police attempting to work with them to solve problems in their neighborhoods, they are more likely to develop confidence in the police overall. 548 Belfast Civilian 3. 219 Limitations While there are many situations where my arguments about the effectiveness of community policing building confidence in the police and the state would be useful, it is also important to note their limitations. As should be clear from the preceding section, for instance, states with authoritarian or hybrid governments would not provide good candidates for my argument. These states do not face the same constraints as democratic states and, therefore do not require community confidence for successful policing, only the credible fear of the security force or continued appeasement of the police force to continue its support and protection of the regime. However, there are other, non- authoritarian situations where my arguments may not be as useful. Community policing may be less effective in building state legitimacy after ethnic conflict when two circumstances exist: ethnic homogeneity in a geographic region and devolved policing. This set of circumstances exists in certain countries facing separatist movements like that in the Spanish Basque Country. In this situation, the minority ethnic group, the Basque, lives almost entirely separately from the rest of the population in the country and has its own independent police force, the Ertzaintza. The ethnic conflict is between the Basque on the Spanish side of the border (a portion of the Basque country also exists in France) and the larger Spanish government from whom they desire complete independence. This case has a distinct feature that separates it from other incidents of ethnic conflict. Namely, the police are not considered biased because they are independent from the rest of the Spanish police. If the police are not considered to be a part of the problem during conflict (a rare, but not unique situation), then reform of the police will likely not affect community perception of the legitimacy of the state. In the 220 case of separatist movements that wish to siphon off an ethnically distinct portion of the community into an independent state, police reform may do little to address the underlying issues of ethnic conflict, and therefore post-conflict rebuilding. In fact, reforming an existing independent police force into an integrated state police force might lead to more tensions and new episodes of violence rather than new confidence in the police. Scotland is another example of a separatist movement that fits into this category. In this case, an identity group which is relative homogeneous and geographically distinct from the rest of the existing state wishes to be siphoned off and declared independent and sovereign. Working under the premise that I began this dissertation with, the difficulty of establishing and having new states recognized internationally, failed separatist movements will require state rebuilding. In this situation, again, the police in Scotland are already devolved, and are not considered biased against an ethnic community in Scotland. Community policing may be enacted and received well in Scotland, but if one goal of police reform is to translate community confidence in the police into community confidence in the state overall, it does not seem likely that community policing will have this effect in this situation. It should be noted that some may argue that Northern Ireland’s conflict was, in essence, similar to that of Basque and Catalonia. Northern Ireland, as has been well established, had its own independent police force. However, and importantly, the police were perceived by both communities to side with the PUL community and actively to discriminate against the CNR community during the Troubles. This put the police directly in the middle of the conflict in a way that the Ertzaintza and the Scottish police 221 were not. If the police are seen as impartial or uninvolved in conflict, then employing community policing might very well build good relationships between the community and the police after conflict characterized like that in Basque and Scotland, but it is not likely to affect confidence in the state. Important Caveats In addition to the limitations mentioned above, it is important to note there are some other important variables that may affect policing and post-conflict state stability that I did not incorporate into my analysis. These include, but may not be limited to economic status, geopolitical importance and international pressures, and post-colonial legacies. Each of these variables may influence both the direction of post-ethnic conflict state rebuilding efforts and their success. The economic status of post-ethnic conflict states seeking to implement police reform is possibly one of the most important factors not included in this study. A state’s Gross Domestic Product (GDP) has been included in dozens of studies as a statistically significant predictor of the probability of the onset,549 duration,550 and long-term effects of civil war.551 The logic here is easy to understand – states with more resources will have more incentives to prevent war, as they have more to lose through war. Moreover, war will be more likely to end in a decisive military victory because the resources of the 549 Bruce Bueno de Mesquita, Predicting Politics (Columbus, OH: Ohio State University Press, 2002); James D. Fearon, Kimuli Kasara, and David D. Laitin, "Ethnic Minority Rule and Civil War Onset," American Political Science Review 101, no. 1 (2007); Helge Holtermann, "Explaining the Development- Civil War Relationship," Conflict Management and Peace Science 29, no. 1 (2012). 550 Montalvo and Reynal-Querol, "Ethnic Polarization and the Duration of Civil Wars." 551 Paul Collier, "On the Economic Consequences of Civil War," Oxford Economic Papers 51, no. 1 (1999). 222 state make winning more probable.552 Wealth is an important factor in determining both the incentives and capabilities of combatants and the state in a conflict, and therefore is important to consider. However, in this case, my question does not address why a conflict starts, lasts, or ends, but rather what reform of the police should look like after conflict. That said, however, it is important to note that reform, and especially good reform, takes time and capital. Therefore, in future analysis of this question it will be important for me to include the resources available to states after conflict as one important factor in the success of police reform. In my research in Northern Ireland, one major theme that continued to arise in interviews with police and civilians alike was the reduction in financial support for police reforms after the initial 10-years of the PSNI. Many interviewees opined that the greatest threat to the long-term success of police reform programs was the reduction in funds and officers for community policing initiatives. Many feared that the progress achieved in building good relationships between the community and the police will fade as community officers spend less and less time on the streets in regular patrol and are forced to limit their time to responding to calls.553 Another variable which I did not give a lot of attention was the idea of the geo- political importance of the state and international pressures. Inevitably, some states hold great importance in the world while the international community has sadly ignored others. The importance of a state in the eyes of the international community may vary through the years and may increase based on the type or intensity of conflict experienced 552 Kaufmann. 553 Belfast Officer 1; Belfast Officer 2; Belfast Officer 3; Belfast Officer 7, interview by Shawna Meechan, 3 November, 2016; Belfast Officer 8; Derry Officer 1; Derry Officer 5. 223 (i.e. Rwanda gained considerable international attention after the 1994 genocide due to the extremity and severity of the violence). However, states like Northern Ireland have received much more international attention than many conflicts in sub-Saharan Africa due largely to their geographic placement in the global west, diaspora of citizens to other western countries, and its position within the United Kingdom. Northern Ireland experienced great pressure from the outside world to end the conflict, and subsequently received enormous support in its recovery efforts from states like the United States and Canada that have large numbers of Irish immigrants and descendants. Conversely, while Kenya holds considerable importance in East Africa and on the continent in general, they received far less international pressure to settle grievances from the global west, as well as less assistance after the violence. Many African countries offered assistance, but with far fewer resources than were made available to Northern Ireland. These differences in pressure and support from the international community clearly influence the type and success of reforms after conflict. While Kenya undoubtedly borrowed from Northern Ireland’s experiences and reform efforts, they did not have the continued international scrutiny and support that Northern Ireland experienced. Small but important legislative changes to the NPS Act 2011 in the years after its initial passage went largely unnoticed in the international community. One could easily assume that similar changes to Northern Ireland’s reform efforts would have seen much more scrutiny. In my continued work on this subject, this issue should be studied more and emphasized as an important factor in the overall success of post-conflict police reform and state stability. 224 Finally, the issue of post-colonial legacies is one that I briefly mentioned in the assessment of Kenya but deserves further attention in later research. Colonialism created special issues, especially in Africa and South America, which western countries do not have to deal with on the same level. Everything from the organization of the government, to the physical borders of states, to the languages spoken and the crops grown in these states can be tied back to the legacies of colonialism. Future research should include an analysis of how these legacies shape the current politics of states with ethnic conflict and the likelihood of reform success. All the variables mentioned here, and several more, may influence the success of community policing and rebuilding in post-ethnic conflict states. Undoubtedly, more research needs to be done that incorporates these factors and accounts for the effects they have on success. Corruption In Chapter I of this dissertation I asserted that democracy for the purposes of this dissertation includes the minimal procedural aspects of democracy plus accountability. Accountability includes the absence of corruption. Corruption is often associated with illiberal democracies, but it is separable and so deserves special attention. For example, Singapore is consistently ranked in the top 10 least corrupt countries in the world according to Transparency International’s Corruptions Perceptions Index (ahead of many famous liberal democracies including the United States, United Kingdom, and Australia),554 but is considered an illiberal democracy because its media is not free and 554 "Singapore 6th-Least Corrupt Country in the World: Annual Index," (2018), https://www.channelnewsasia.com/news/singapore/singapore-6th-least-corrupt-country-in-the-world- annual-index-9979996. 225 elections are severely restricted.555 Police corruption is therefore important to highlight as an important piece of accountability. Police forces with deep histories of corruption and bribery may implement community policing programs all they like but will likely fail to gain community confidence if corruption is not addressed directly. The International Police Executive Symposium (IPES) held in Poland in 2001 had the theme “Corruption: A Threat to World Order.”556 In composing a book based on the presentations at the conference, the authors of Policing Corruption: International Perspectives place the pervasive problem of corruption in the police in perspective. This collection of works addresses issues of corruption in the police from some of the world’s least corrupt countries such as the United Kingdom (ranked 2) to the most corrupt country in the world, Nigeria (ranked 168) according to 2001 world rankings.557 In the book the authors discuss different levels of corruption – from low-level bribes of food and small amounts of money to grand-scale million dollar pay offs. They also describe other forms of corruption such as tampering with crime scenes and the use of excessive force. The authors discuss the different consequences of police corruption for society. Police corruption makes the job of policing difficult because victims of and witnesses to crimes will be less willing to cooperate with the police if they suspect that the police may be corrupt. Good police cannot rely on the participation of the community in the pursuit 555 556 Sarre, Das, and Albrecht, 1. 557 George T. Kurian, The Illustrated Book of World Rankings, 8th ed. (Armonk, NY: M.E. Sharpe, Inc, 2001), 67-68. 226 of justice under the cloud of corruption. Corruption in the police leads to a lack of community confidence in the police and an erosion of the moral stature of the police.558 The pervasiveness of police corruption is something that needs to be addressed in concert with, or likely before the implementation of, community policing in the aftermath of violent ethnic conflict. In ethnically divided societies, the corruption of the police may primarily be seen as bias in application and enforcement of the law, as was largely the case in Northern Ireland. However, if other forms of corruption such as bribery, excessive force, or other forms of police misconduct are also associated with the police in a given state, post-conflict reform will need to include addressing issues of corruption directly to be successful. Clearly, this needs further research. In future work, I plan to incorporate a specific analysis of police corruption into my arguments through the deeper analysis of the Kenyan police, a re-evaluation of the Police Service of Northern Ireland, and additional police forces not included in this project. Future Research This project has made some progress evaluating the dynamics of policing in post- ethnic conflict societies and its influence on community confidence in the police. I would like to expand this project further by performing in-depth interviews with police and civilians in Kenya – like those performed in Northern Ireland. Additional fieldwork and interviews will provide a better understanding of why community policing failed in Kenya. Following on the previous section, corruption may be a serious component of 558 Sarre, Das, and Albrecht, 8-9. 227 what is missing in my current analysis and more work in Kenya could help to better operationalize this. This project originated in the desire to analyze how policing interacts and sequences with other strategies for post-ethnic conflict state stability – namely community reconciliation and ethnic representation in government. Northern Ireland has had a volatile history with establishing and maintaining a functioning parliament. In the first ten years following the signing of the Belfast/Good Friday Agreement, the Northern Ireland Assembly (commonly called Stormont) was established and suspended four times before a relatively stable Assembly was formed in 2007. At the time of this writing, Stormont has been suspended once again for more than a year due to an inability to form a coalition government. An analysis of how the forming and suspension of Stormont corresponded with changes and reforms in policing and with major community reconciliation initiatives could provide important information about how societies recover from violent ethnic conflict. I plan to begin work on such a project in the near future. This dissertation has also sparked a deep interest for me in the special role that civilian police forces play in democratic societies. Officers are, by definition, both citizens of the state and an enforcement arm of the state. Straddling the divide between civilian and government official is unique to this kind of work and creates unique problems. I plan to continue my study of civilian police forces in ethnically divided societies at different levels. This study has concentrated on police forces organized at the national level. I would like to further this study by looking at more federalized forms of police forces, such as city and county police in the United States. How does working 228 separate from, but to a degree in concert with, other police forces change how the police approach their work in deeply divided societies, if at all? Finally, one more extension of this project I would like to pursue is an analysis of how the PSNI work with the Garda Síochána of the Republic of Ireland in policing the border between Northern Ireland and the Republic. How do different state police forces work together to police international borders in the aftermath of violent conflict? How do issues of border policing affect matters of internal policing in one or more of the bordering states? I would like to work to identify the set of issues specific to policing borders between states where ethno-nationalist ties cross political boundaries by examining policing the border between Northern Ireland and the Republic of Ireland. Specifically, I would like to examine how the two police forces approached coordination after the end of the Troubles and focus on how the extensive policing reforms in Northern Ireland affected the relationship with the Garda Síochána of the Republic of Ireland. I would also like to include an exploration of the negotiations related to the impending exit of the United Kingdom from the European Union and how this move will affect issues of policing the border in the future. Conclusion After violent ethnic conflict, personal security is vital for citizens of a state to move forward and build confidence in the state. In democratic societies, security is provided by a civilian police force. Unfortunately, the police are often perceived to have taken sides during ethnic conflict such that marginalized groups feel alienated and unprotected by the police. For policing to be successful, it requires the confidence of the community that the police will competently and fairly apply and enforce the law. 229 Building that confidence following ethnic conflict requires reform that is tailored to the context in which it is implemented. Policing literature suggests that implementation of community policing will achieve this goal. Community policing is designed to build community confidence in the police through involving the community in policing at all stages including problem identification, idea formation, implementation, and evaluation. Community policing also emphasizes decentralized decision making in the police to encourage those with specialized knowledge to help develop creative solutions to community problems. When implemented well, these principles lead us to expect decentralization in the organizational structure of the police, retraining of the police to be oriented to working with the community, and institutionalized forums for community engagement with policing. Additionally, community policing implies there should be special attention to symbolic changes in the police from badges and emblems, to the make-up of the police force, to the physical appearance of police buildings and vehicles that will work to open up the police to positive interactions with the community. When states commit to fully embracing community policing in the aftermath of ethnic conflict, they should see increased community confidence in the police. 230 REFERENCES CITED (PSNI), Police Service of Northern Ireland. "Parade Statistics 2000-2002." Belfast, United Kingdom: Police Service Northern Ireland, 2003. Abu-Nimer, Mohammed, ed. Reconciliation, Justice, and Coexistence: Theory & Practice. Lanham, MD: Lexington Books, 2001. Africa Check. "Kenyatta's Numbers on Kenyan Police to Civilian Ratio Still Don't Add Up." https://africacheck.org/spot-check/kenyattas-numbers-kenyan-police- civilian-ratio-still-dont-add/. Afzali, Aneelah, and Laura Colleton. "Constructing Coexistence: A Survey of Coexistence Projects in Areas of Ethnic Conflict." In Imagine Coexistence: Restoring Humanity after Violent Ethnic Conflict, edited by Antonia Chayes and Martha Minow, 3-20. San Francisco: Jossey-Bass, 2003. Agence France Presse. "N.Ireland Unionist Parades Pass Off Peacefully." (2014). Published electronically 12 July. https://www.yahoo.com/news/n-ireland- unionist-parades-pass-off-peacefully-003702341.html. Agency, Northern Ireland Statistics and Research, and Northern Ireland Policing Board. "Public Perceptions of the Police, Dpps, and the Northern Ireland Policing Board." Belfast, United Kingdom, 2007. Ambassade de la Republique Democratique du Congo. "Overview of Security Efforts by the Democratic Republic of Congo." Washington, DC. Amnesty International USA. "Sri Lanka - the Northeast: Human Rights Violations in a Context of Armed Conflict." Amnesty International USA, https://web.archive.org/web/20090214094900/http://www.amnestyusa.org/docum ent.php?lang=e&id=5E7F89A231872062802569A600601598. Anderson, Benedict. Imagined Communities: Reflections on the Origin and Spread of Nationalism. New York, New York: Verso, 1983. Anderson, David. "Majimboism: The Tourbled History of an Idea." In Our Turn to Eat! Politics in Kenya since 1950, edited by Daniel Branch and Nic Cheeseman. Berlin: Lit Verlag, 2008. Anderson, David, and Emma Lochery. "Violence and Exodus in Kenya's Rift Valley, 2008: Predictable and Preventable?". Journal of Eastern African Studies 2, no. 2 (July 2008): 328-43. Andeweg, R.B. "Consociationalism." In International Encyclopedia of Social & Behavioral SciencesElsevier Ltd., 2001. 231 Areni, Charles S, and Mitchell Burger. "Memoried of "Bad" Days Are More Biased Than Memories of "Good" Days: Past Saturdays Vary, but Past Mondays Are Always Blue." Journal of Applied Social Psychology 38, no. 6 (2008): 1395-415. ARK. "Northern Ireland Life and Times Survey, 1998." edited by ARK: www.ark.ac.uk/nilt, 1999. Aughey, Arthur. The Politics of Northern Ireland: Beyond the Belfast Agreement. Abingdon: Routledge, 2005. Aydin, Ahmet H. "The Future of Policing in Turkey: Current and Proposed Organizational Changes in Public Policing." Police Studies 19, no. 4 (1996): 39- 56. Bajoria, Jayshree. "The Sri Lankan Conflict; Backgrounder." (2009). Published electronically 18 May. https://www.cfr.org/backgrounder/sri-lankan-conflict. Bass, Gary. Stay the Hand of Vengeance: The Politics of War Crimes Tribunals. Princeton, New Jersey: Princeton University Press, 2000. Bayley, David H. Changing the Guard: Developing Democratic Police Abroad. Oxford, UK: Oxford University Press, 2006. ———. Democratizing the Police Abroad: What to Do and How to Do It. Washington, DC: U.S. Department of Justice, Office of Justice Programs, National Institute of Justice, 2001. Bayley, David H, and Robert Perito. "Police Corruption: What Past Scandals Teach About Current Challenges." In Special Report. Washington, DC: United States Institue of Peace, 2011. Bayley, David H. "Post-Conflict Police Reform: Is Northern Ireland a Model?". Policing: A Journal of Policy and Practice 2, no. 2 (1 January 2008): 233-40. "Belfast (Good Friday) Agreement." United Kingdom: Northern Ireland Office, 1998. Belfast Civilian 1. "Interview." By Shawna Meechan (25 October 2016). Belfast Civilian 2. "Interview." By Shawna Meechan (26 October 2016). Belfast Civilian 3. "Interview." By Shawna Meechan (28 October 2016). Belfast Civilian 4. "Interview." By Shawna Meechan (1 November 2016). Belfast Civilian 5. "Interview." By Shawna Meechan (10 November 2016). Belfast Civilian 6. "Interview." By Shawna Meechan (15 November 2016). 232 Belfast Civilian 7. "Interview." By Shawna Meechan (25 October 2016). Belfast Civilian 9. "Interview." By Shawna Meechan (3 September 2015). Belfast Civilian 10. "Interview." By Shawna Meechan (3 September 2015). Belfast Civilian 11. "Interview." By Shawna Meechan (18 October 2016). Belfast Officer 1. "Interview." By Shawna Meechan (19 October 2016). Belfast Officer 2. "Interview." By Shawna Meechan (3 November 2016). Belfast Officer 3. "Interview." By Shawna Meechan (3 November 2016). Belfast Officer 4. "Interview." By Shawna Meechan (3 November 2016). Belfast Officer 7. "Interview." By Shawna Meechan (3 November 2016). Belfast Officer 8. "Interview." By Shawna Meechan (10 November 2016). Belfast Officer 9. "Interview." By Shawna Meechan (8 September 2017). Belfast Officer 10. "Interview." By Shawna Meechan (8 September 2017). Belfast Officer 11. "Interview." By Shawna Meechan (8 September 2017). Bennett, Andrew, and Jeffrey T. Checkel, eds. Process Tracing: From Metaphor to Analytic Tool. Cambridge: Cambridge University Press, 2015. Bennett, Andrew, and Alexander L. George. "Process Tracing in Case Study Research." In MacArthur Foundation Workship on Case Study Methods. Belfer Center for Science and International Affairs (BCSIA), Harvard University, 1997. Berk, Gerald, and Dennis Galvan. "How People Experience and Change Institutions: A Field Guide to Creatuve Syncetism." Theory and Society 38, no. 6 (November 2009): 543-80. Bhavnani, Ravi, and Dan Miodownik. "Ethnic Polarization, Ethnic Salience, and Civil War." Journal of Conflict Resolutions 53, no. 1 (2009): 30-49. Black, John. "The Symbol of the Swastika and Its 12,000-Year-Old History." Ancient Origins: Reconstructing the Story of Humanity's Past, http://www.ancient- origins.net/myths-legends/symbol-swastika-and-its-12000-year-old-history- 001312?nopaging=1. Board, Northern Ireland Policing, and Northern Ireland Statistics and Research Agency. "Public Perceptions of the Police and the Northern Ireland Policing Board October 2003." Belfast, 2003. 233 ———. "Public Perceptions of the Police, Dpps, and the Northern Ireland Policing Board April 2009 Omnibus Survey." Belfast, 2009. Boulding, Kenneth E. Stable Peace. Austin: University of Texas Press, 1978. Bourdieu, Pierre. The Logic of Practice. Translated by Richard Nice. Stanford, CA: Stanford University Press, 1980. Les Editions de Minuit. ———. Outline of a Theory of Practice. Translated by Richard Nice. Cambridge, United Kingdom: Cambridge University Press, 1977. Brennan, Colin. "Seventy One Police Officers Injured after Fourth Night of Violence in the North." Irish Mirror (2013). Published electronically 16 July 2013. https://www.irishmirror.ie/news/irish-news/northern-ireland-riots-71-police- 2056588. Brownsell, James. "Kenya: What Went Wrong in 2007?" Al Jazeera (2013). Published electronically 3 March. https://www.aljazeera.com/indepth/features/2013/03/201333123153703492.html. Brunger, M. "Governance, Accountability and Neighborhood Policing in Northern Ireland: Analysing the Role of Public Meetings." Crime Law and Social Change 55, no. 2-3 (April 2011): 105-20. Buchanan-Clarke, Stephen, and Rorisang Lekalake. "Violent Extremism in Africa: Public Opinion from the Sahel, Lake Chad, and the Horn." edited by Afrobarometer, 2016. Buckley-Zistel, Susanne. "We Are Pretending Peace: Local Memory and the Absence of Social Transformation and Reconciliation in Rwanda." In After Genocide: Transitional Justice, Post-Conflict Reconstruction and Reconciliation in Rwanda and Beyond, edited by Philip Clark and Zachary Kaufman, 125-44. New York, New York: Columbia University Press, 2009. Bueno de Mesquita, Bruce. Predicting Politics. Columbus, OH: Ohio State University Press, 2002. Butler, J. "Performativity's Social Magic." In Bourdieu: A Critical Reader, edited by R. Schusterman, 113-28. Oxford: Blackwell, 1999. CAIN. "Persons Injured (Number) Due to the Security Situation Sin Northern Ireland (Only), 1969 to 2003." edited by Police Service Northern Ireland. Belfast, 2003. Call, Charles T. "Competing Donor Approaches to Post-Conflict Police Reform." Conflict, Security & Development 2, no. 01 (2002): 99-120. 234 ———. "Is Transitional Justice Really Just?". The Brown Journal of World Affairs 11, no. 1 (1 July 2004): 101-13. Call, Charles T., and William D. Stanley. "Protecting the People: Public Security Choices after Civil Wars." Global Governance 7, no. 2 (Apr-June 2001): 151-72. Cameron Commission. "Disturbances in Northern Ireland." edited by Chairman Lord Cameron. Belfast: HMSO, 1969. Cammett, Melani, and Edmund Malesky. "Power Sharing in Postconflict Societies: Implications for Peace and Governance." The Journal of Conflict Resolution 56, no. 6 (2012): 982-1016. Caparini, Marina, and Otwin Marenin. "Police Transformation in Central and Eastern Europe: The Challenge of Change." In Transforming Police in Central and Eastern Europe: Process and Progress, edited by Marina Caparini and Otwin Marenin, 3-20. Geneva, Switzerland: Geneva Centre for the Democratic Control of Armed Forces, 2004. Carter, David L., and Allen D. Sapp. "Issues and Perspectives of Law Enforcement Accreditation: A National Study of Police Chiefs." Journal of Criminal Justice 22, no. 3 (1994): 195-204. Celador, Gemma Collantes. "Police Reform: Peacebuilding through 'Democratic Policing'?". International Peacekeeping 12, no. 3 (2005): 364-76. "Census - 1926 - Northern Ireland." Belfast, United Kingdom: The Government of Northern Ireland`, 1926. Central Intelligence Agency. "The World Factbook: Sri Lanka." https://www.cia.gov/library/publications/the-world-factbook/geos/ce.html. Chayes, Antonia, and Martha Minow, eds. Imagine Coexistence: Restoring Humanity after Violent Ethnic Conflict. San Francisco: Jossey-Bass, 2003. Cheeseman, Nic. "The Kenyan Elections of 2007: An Introduction." Journal of Eastern African Studies 2, no. 2 (2008): 166-84. Chigas, Diana, and Brian Ganson. "Grand Visions and Small Projects: Coexistence Efforts in Southeastern Europe." In Imagine Coexistence: Restoring Humanity after Violent Ethnic Conflict, edited by Antonia Chayes and Martha Minow. San Francisco: Jossey-Bass, 2003. Chukwuma, Innocent. "Police Transformation in Nigeria: Problems and Prospects." In Crime and Policing in Transitional Societies. Jan Smuts House, University of the Witwatersrand, Johannesburg, 2000. 235 Clayton, Anthony. "Kenya." In Monographs in International Studies. Africa Series, 109- 41. Athens, Ohio: Ohio University Center for International Studies, 1989. Collier, David, and James Mahoney. "Research Note - Insights and Pitfalls: Selection Bias in Qualitative Research." World Politics 49, no. 1 (1996): 56-91. Collier, Paul. "Ethnic Civil Wars: Securing the Post-Conflict Peace." Harvard International Review 28, no. 4 (Winter) (2007): 56-60. ———. "On the Economic Consequences of Civil War." Oxford Economic Papers 51, no. 1 (1999): 168-83. "Constitution of Ireland." Dublin, Republic of Ireland: The Stationery Office, 2015. Coppedge, Michael, John Gerring, David Altman, Michael Bernhard, Steven Fish, Allen Hicken, Matthew Kroenig, et al. "Conceptualizing and Measuing Democracy: A New Approach." Perspectives on Politics 9, no. 2 (June 2011): 247-67. Cordner, Gary. "Community Policing: Elements and Effects." Police Forum 5, no. 3 (July 1995): 1-8. Crawford, Beverly, and Ronnie D. Lipschutz, eds. The Myth of "Ethnic Conflict": Politics, Economics, and "Cultural" Violence. Berkeley: International and Area Studies, University of California Berkeley, 1998. "Cyprus Talks End without a Peace and Reunifications Deal." BBC News (2017). Published electronically 7 July. https://www.bbc.com/news/world-europe- 40530370. Dahl, Robert A. A Preface to Democratic Theory. Chicago: University of Chicago Press, 1956. Darby, John. "Northern Ireland: The Background to the Peace Process." Conflict Archive on the Internet (CAIN), Ulster University. DeRouen Jr., Karl , Jenna Lea, and Peter Wallensteen. "The Duration of Civil War Peace Agreements." Conflict Management and Peace Science 26, no. 4 (2009): 367-87. Derry Civilian 1. "Interview 1." By Shawna Meechan (18 September 2015). ———. "Interview 2." By Shawna Meechan (12 September 2016). Derry Civilian 2. "Interview." By Shawna Meechan (16 September 2015). Derry Civilian 3. "Interview." By Shawna Meechan (21 September 2015). Derry Officer 1. "Interview." By Shawna Meechan (28 September 2016). 236 Derry Officer 2. "Interview." By Shawna Meechan (4 October 2016). Derry Officer 3. "Interview." By Shawna Meechan (5 October 2016). Derry Officer 5. "Interview." By Shawna Meechan (6 October 2016). Derry Officer 6. "Interview." By Shawna Meechan (6 October 2016). Derry Officer 7. "Interview." By Shawna Meechan (6 October 2016). Derry Officer 8. "Interview." By Shawna Meechan (6 October 2016). Dewey, John. Human Nature and Conduct: An Introduction to Social Psycology. Amherst, N.Y.: Prometheus Books, 2002. New York: Henry Holt and Co., 1922. Diallo, Mariama. "New Kenyan Constitution Ratified." Voice of America (2010). Published electronically 5 August. https://www.voanews.com/a/kenyas-new- constitution-ratified-100158209/123395.html. Dickson, Brice. "New Beginnings? Policing and Human Rights after the Conflict." edited by Michael Cox, Adrian Guelke and Fiona Stephen. Manchester: Manchester University Press, 2006. Eckstein, Harry. "Case Studies and Theory in Political Science." In Handbook of Political Science, edited by Fred Greenstein and Nelson Polsby. Reading, Mass.: Addison-Wesley, 1975. Edelman, James, and Elise Bant. Unjust Enrichment in Australia. Oxford: Oxford University Press, 2006. Ellison, Graham. "A Blueprint for Democratic Policing Anywhere in the World: Police Reform, Political Transition, and Conflict Resolution in Northern Ireland." Police Quarterly 10, no. 3 (2007): 243-69. Ellison, Graham, and Jim Smyth. The Crowned Harp - Policing Northern Ireland. London: Pluto Press, 2015. Eynde, Oliver Vanden, Patrick M. Kuhn, and Alexander Moradi. "Ethnic Politics and Job Performance in the Kenyan Police 1957-1970." 2015. Faull, Andrew. "Performance Measurement in Police Agencies: A Report for the Khayelitsha Commission of Inquiry." 2014. Fay, Marie-Therese, Mike Morrissey, and Marie Smyth. Northern Ireland's Troubles: The Human Cost. Sterling, VA: Pluto Press in association with the Cost of the Troubles Study, 1999. 237 Fearon, James D., Kimuli Kasara, and David D. Laitin. "Ethnic Minority Rule and Civil War Onset." American Political Science Review 101, no. 1 (2007): 187-93. Fearon, James D., and David D. Laitin. "Ethnicity, Insurgency, and Civil War." American Political Science Review 97, no. 1 (2003): 75-90. Forum for Cities in Transition. "Meeting." Derry, Northern Ireland, 2015. Fulton, R.V. "Community Policing on Horseback." Law Enforcement Technology 20, no. 5 (May 1993): 32-35. Galtung, Johan. "Violence, Peace, and Peace Research." Journal of Peace Research 6, no. 3 (1969): 167-91. Gastrow, Peter. "The Kenya National Task Force on Police Reform: Some Key Recommendations Summarised." https://www.ipinst.org/wp- content/uploads/2010/01/pdfs_summary_policereformreport.pdf. Geertz, Clifford. Old Societies and New States: The Quest for Modernity in Asia and Africa. London: Free Press of Glencoe, 1963. George, Alexander L., and Andrew Bennett. Case Studies and Theory Development in the Social Sciences Cambridge, Mass.: MIT Press, 2005. Gerring, John. Case Study Research: Principles and Practices. 2nd ed. Draft2016. Gerring, John, and Lee Cojocaru. "Selecting Cases for Intensive Analysis: A Diversity of Goals and Methods." Sociological Methods & Research 45, no. 3 (2016): 392- 423. Gettleman, Jeffrey. "Disputed Vote Plunges Kenya into Bloodshed." The New York Times, 31 December 2007. Gibson, Clark C., and James D. Long. "The Presidential and Parliamentary Elections in Kenya, December 2007." Electoral Studies 28, no. 3 (2009): 497-502. Gilley, Bruce. "Against the Concept of Ethnic Conflict." Third World Quarterly 25, no. 6 (2004): 1155-66. Gimode, Edwin A. "The Role of the Police in Kenya's Democratisation Process." In Kenya: The Struggle for Democracy, edited by Godwin R. Murunga and Shadrack Wanjala Nasong'o. New York: Palgrave Macmillan, 2007. Glassmyer, Katherine, and Nicholas Sambanis. "Rebel: Military Integration and Civil War Termination." Journal of Peace Research 45, no. 3 (2008): 365-84. Goldstein, Herman. Problem-Oriented Policing. Philadelphia: Temple University Press, 1990. 238 ———. "Toward Community-Oriented Policing: Potential, Basic Requirements, and Threshold Questions." Crime & Delinquency 33, no. 1 (1987): 6-30. Gordon, Scott. Controlling the State: Constitutionalism from Ancient Athens to Today. Cambridge, MA: Harvard University Press, 1999. Greene, Helen. "Community Oriented Policing in Florida." American Journal of Police 12, no. 3 (Fall 1993): 141-55. Guelke, Adrian. Politics in Deeply Divided Societies. Cambridge, UK: Polity, 2012. Gutierrez-Romero, Roxana. "To What Extent Did Ethnicity and Economic Issues Matter in the 2007 Disputed Kenyan Elections?". Development Policy Review 31, no. 3 (2013): 291-320. Hargan, Garrett. "Almost 250 Detained at the Strand Road Police Station in the Past Year Had to Be Taken to Hospital, New Figures Reveal." Derry Now (2017). Published electronically 8 August 2017. https://www.derrynow.com/news/almost- 250-people-detained-strand-road-police-station-past-year-taken-hospital-new- figures-reveal/178445. Hartzell, Caroline. "Explaining the Stability of Negotiated Settlements to Interstate Wars." Journal of Conflict Resolutions 43, no. 3 (1999): 3-22. Hartzell, Caroline, and Matthew Hoddie. "Institutionalizing Peace: Power Sharing and Post-Civil War Conflict Management." American Journal of Political Science 47, no. 2 (2003): 318-32. Hartzell, Caroline, Matthew Hoddie, and Donald Rothchild. "Stabilizing the Peace after Civil War: An Investigation of Some Key Variables." International Organization 55, no. 1 (2001): 183-208. Hayes, R. Allen. "Policing in Northern Ireland: Community Control, Community Policing, and the Search for Legitimacy." Urban Affairs Review 49, no. 4 (2012): 557-92. Hayner, Priscilla. Unspeakable Truths: Transitional Justice and the Challenge of Truth Commissions. New York, NY: Routledge, 2011. Held, David. Models of Democracy. 3rd ed. Stanford, CA: Stanford University Press, 2006. "Herstory." Black Lives Matter, https://blacklivesmatter.com/about/herstory/. Holland, Jack. "Ruc: The Politics of Policing " The Irish Echo (2011). https://www.irishecho.com/2011/02/ruc-the-politics-of-policing-2/. 239 Holtermann, Helge. "Explaining the Development-Civil War Relationship." Conflict Management and Peace Science 29, no. 1 (2012): 56-78. Hope, Kempe Ronald, Sr. "In Pursuit of Democratic Policing: An Analytical Review and Assessment of Police Reforms in Kenya." International Journal of Police Science & Management 17, no. 2 (2015): 91-97. ———, ed. Police Corruption and Police Reforms in Developing Societies. Boca Raton: CRC Press, 2016. Hornsby, Charles. Kenya: A History since Independence. London: I.B. Tauris, 2012. Horowitz, Donald. "Democracy in Divided Societies." Journal of Democracy 4, no. 4 (October 1993): 18-38. ———. Ethnic Groups in Conflict. 2nd ed. Berkeley, California: University of California Press, 2000. 2nd. ———. "Ethnic Power Sharing: Three Big Problems." Journal of Democracy 25, no. 2 (2014): 5-20. Human Rights Watch. "Failing the Internally Displaced: The Undp Displaced Persons Program in Kenya." New Yoek: Human Rights Watch, 1997. Human Rights Watch, Ben Rawlence, and Chris Albin-Lackey. Ballots to Bullets: Organized Political Violence and Kenya's Crisis of Governance. New York: Human Rights Watch, 2008. Hunt Committee. "Report of the Advisory Committee on Police in Northern Ireland." edited by Chairman Lord Hunt. Belfast: HMSO, 1969. Huntington, Samuel. The Third Wave: Democratization in the Late Twentieth Century. Norman: University of Oklahoma Press, 1993. Idson, Lorraine Chen, Nira Liberman, and E. Tory Higgins. "Distinguishig Gains from Nonlosses and Losses from Nongains: A Regulatory Focus Perspective on Hedonis Intensity." Journal of Experimental Social Psycology 36, no. 3 (May 2000): 252-74. International Crisis Group. "Kenya in Crisis." Nairobi; Brussels: International Crisis Group, 2008. Ireland, The Police Ombudsman for Northern. "Annual Report on Public Awareness Od the System for Complaints Againts the Police in Northern Ireland, 2011." Belfast, United Kingdom: Office of the Police Ombudsman fro Northern Ireland, 2011. "Is Democracy in Kenya at Risk?". The Week (2018). Published electronically 29 March 2018. http://www.theweek.co.uk/92642/is-democracy-in-kenya-at-risk. 240 Ishiyama, John, and Jeremy Backstrom. "Reconciliation and Conceptual Complexity: The Case of Post Conflict Kenya." African and Asian Studies 10, no. 4 (2011): 366-86. Jackson, Arrick, and Alynna Lyon. "Policing after Ethnic Conflict - Culture, Democratic Policing, Politics and the Public." International Police Strategy and Management 25, no. 2 (2002): 221-41. Joseph, Thomas. "Walking the Minefields of Community-Oriented Policing." FBI Law Enforcement Bulletin 63, no. 9 (1994): 8. Kagwanja, Peter, and Willy Mutunga. "Is Majimbo Federalism? Constitutional Debate in a Tribal Shark-Tank." Opinion, Daily Nation (2001). Published electronically 20 May 2001. https://www.nation.co.ke/oped/1192-340744-mbwm4j/index.html. Kapiszewski, Diana, Lauren M. MacLean, and Benjamin Lelan Read. Field Research in Political Science: Practices and Principles. Cambridge, United Kingdom: Cambridge University Press, 2015. Kaufman, Stuart J. "Spiraling to Ethnic War: Elites, Masses, and Moscow in Moldova's Civil War." International Security 21, no. 2 (1996): 108-38. Kaufmann, Chaim. "Possible and Impossible Solutions to Ethnic Civil Wars." International Security 20, no. 4 (1 April 1996): 136-75. Kelley, Kevin J. "Democracy on Decline in Kenya, Ngo Finds." Daily Nation (2018). Published electronically 17 January 2018. https://www.nation.co.ke/news/politics/Democracy-in-Kenya/1064-4266504- btxrt6z/index.html. Kenya Embassy. "About Kenya History." http://www.kenyaembassydc.org/aboutkenyahistory.html. Kenya, Transparency International. "The East African Bribery Index Trends Analysis (2010-2014)." edited by Transparency International. Kenya, 2015. "Key Points: Kenya Power-Sharing Deal." BBC News World Edition (2008). Published electronically 38 February. http://news.bbc.co.uk/2/hi/africa/7269476.stm. Khadiagala, Gilbert M. "Political Movements and Coalition Politics in Kenya: Entrenching Ethnicity." South African Journal of International Affairs 17, no. 1 (2010): 65-84. Kimani, Mary. "Security for the Highest Bidder." Africa Renewal Online, no. October 2009 (2009). https://www.un.org/africarenewal/magazine/october-2009/security- highest-bidder. 241 King, Charles. "The Myth of Ethnic Warfare." Foreign Affairs 80, no. 6 (2001): 165-70. Kurian, George T. The Illustrated Book of World Rankings. 8th ed. Armonk, NY: M.E. Sharpe, Inc, 2001. Langer, S. K. Philosophy in a New Key. New York: Mentor Books, 1948. Le Vine, Victor. "Conceptualizing "Ethnicity" and "Ethnic Conflict": A Controversy Revisited." Studies in Comparative International Development 32, no. 2 (1997): 45-75. Lederach, John Paul Building Peace: Sustainable Reconciliation in Divided Societies. Washington, DC: U.S. Institute of Peace Press, 2010. Levy, Jack S. "Case Studies: Types, Designs, and Logics of Inference." Conflict Management and Peace Science 25, no. 1 (2008): 1-18. Leys, Colin. Underdevelopment in Kenya, the Political Economy of Neo-Colonialism. London: Heinemann Ltd, 1975. Lijphart, Arend. "Comparative Politics and the Comparative Method." American Political Science Review 65, no. 3 (September 1971): 682-93. ———. "Constitutional Design for Divided Societies." Journal of Democracy 15, no. 2 (Apr 2004): 96-109. ———. Democracy in Plural Societies: A Comparative Exploration. New Haven: Yale University Press, 1977. ———. "Review Article: The Northern Ireland Problem; Cases, Theories, and Solutions." British Journal of Political Science 5, no. 1 (1975): 83-106. Lijphart, Arend, and Don Aitkin. Electoral Systems and Party Systems: A Study of Twenty-Seven Democracies, 1945-1990. New York: Oxford University Press, 1994. Lombardo, Robert M., David Olson, and Monte Staton. "The Chicago Alternative Policing Strategy: A Reassessment of the Caps Program." Policing: An International Journal of Police Strategies & Management 33, no. 4 (2010): 586- 606. Long, William J., and Peter Brecke. War and Reconciliation: Reason and Emotion in Conflict Resolution. Cambridge, Mass: MIT Press, 2003. Lonsdale, John. "Kenya: Ethnicity, Tribe, and State." openDemocracy (2008). Published electronically 17 January. https://www.opendemocracy.net/article/kenya_ethnicity_tribe_and_state. 242 Lynch, Gerald W. "Police Training in Human Dignity." In Policing in Centrral Europe: Comparing Firsthand Knowledge with Experience from the West, edited by Milan Pagon. Ljubljana, Slovenia: College of Police and Security Studies, 1996. Mageka, Annie. "Police Reform in Kenya: Challenges and Opportunities." Centre for Security Governance 2018, no. 9 October (2015). http://secgovcentre.org/2015/10/police-reform-in-kenya-challenges-and- opportunities/. Mandel, Robert. Global Security Upheaval: Armed Nonstate Groups Usurping State Stability Functions. Stanford, California: Stanford Security Studies, and imprint of Stanford University Press, 2013. Mani, Rama. Beyond Retribution: Seeking Justice in the Shadows of War. Cambridge, Mass: Polity Press, 2002. Marijan, Branka, and Sean Brennan. "Policing the Past and Present in Northern Ireland." (2014). Published electronically Sep 18. http://secgovcentre.org/2014/09/policing- the-past-and-present-in-northern-ireland/. Marks, Monique. "Policing for Democracy? The Case of the Public Order Police Unit in Durban." Africa Development 24, no. 1/2 (1 January 1999): 221-68. McAleese, Deborah. "Psni Orders Investigation into Low Number Od Catholic Recruits." Belfast Telegraph (2016). Published electronically 6 September. https://www.belfasttelegraph.co.uk/news/northern-ireland/psni-orders- investigation-into-low-number-of-catholic-recruits-35023903.html. McAllister, Adrian. Paper presented at the Reflecting Back, Moving Forward: Fifteen Years Since the Establishment of the PSNI, University of Ulster, Belfast, United Kingdom, 9 November 2016. McCulloch, Allison. "Does Moderation Pay? Centripetalism in Deeply Divided Societies." Ethnopolitics 12, no. 2 (June 2013): 111-32. McKeown, Michael. "Database of Deaths Associates with Violence in Northern Ireland 1969-2001." 2009. ———. "Post-Mortem: An Examination of the Patterns of Politically Associated Violence in Northern Ireland During the Years 1969-2001 as Reflected in the Fatality Figures for Those Years." Belfast, United Kingdom, 2009; revised 2009. McMahon, Sean. Battles Fought on Irish Soil: A Complete Account. Dublin, Ireland: Londubh Books, 2010. Meechan, Shawna. "Peace Wall in Belfast." 2017. 243 Melaugh, Martin. "'Bloody Friday' - Summary of Main Events." CAIN Web Service, http://cain.ulst.ac.uk/events/bfriday/sum.htm. ———. "A Chronology of the Conflict - 1969." CAIN Web Service, http://cain.ulst.ac.uk/othelem/chron/ch69.htm. Mill, John Stuart. A System of Logic, Ratiocinative and Inductive, Being a Connected View of the Principle of Evidence and the Methods of Scientific Investigation. 10th ed. London: Longmans, Green, and Co., 1879. Miller, Linda S, and Karen M Hess. Community Policing: Theory and Practice. Minneapolis: West Publishing Co, 1994. Moaz, Ifat. "An Experiment in Peace: Reconciliation-Aimed Workshops of Jewish-Israeli and Palestinian Youth." Journal of Peace Research 37, no. 6 (November) (2000): 721-36. Montalvo, Jose G., and Marta Reynal-Querol. "Ethnic Polarization and the Duration of Civil Wars." Economics of Governance 11, no. 2 (2010): 123-43. ———. "Ethnic Polarization, Potential Conflict, and Civil Wars." The American Economic Review 95, no. 3 (1 June 2005): 796-816. Morin, Rich, and Renee Stepler. "The Racial Confidence Gap in Police Performance." (2016). Published electronically 29 September. http://www.pewsocialtrends.org/2016/09/29/the-racial-confidence-gap-in-police- performance/. Mueller, John. "The Banality of "Ethnic War"." International Security 25, no. 1 (2000): 42-70. Mukherjee, Bumba. "Why Political Power-Sharing Agreements Lead to Enduring Peaceful Resolution of Some Civil Wars, but Not Others?". International Studies Quarterly 50, no. 2 (2006): 479-504. Mulcahy, Aogan. "Policing in Divided Societies." In International Encyclopedia of the Social and Behavioral SciencesElsevier Ltd, 2015. Murray, Tonita. "Police-Building in Afghanistan: A Case Study of Civil Security Reform." International Peacekeeping 14, no. 1 (2007): 108-26. Nagel, Joane. "Constructing Ethnicity: Creating and Recreating Ethnic Identity and Culture." Social Problems 41, no. 1 (1994): 152-76. Nasongo, J., J. Achoka, and L. Wamocha. "Is Forgiveness and Amnesty a Panacea to Kenya's Post-Conflict Crisis?". African Journal of Political Science and International Relations 3, no. 4 (April 2009): 122-32. 244 National Accord and Reconciliation Act. 28 February 2008. National Rainbow Coalition (Kenya). "Democracy and Empowerment: Manifesto of the National Rainbow Coalition (Narc)." Nairobi: National Rainbow Coalition, 2002. National Task Force on Police Reforms (Kenya). "Report of the National Task Force on Police Reforms." Nairobi, Kenya, 2009. Neild, Rachel. "Democratic Police Reforms in War-Torn Societies." Conflict, Security & Development 1, no. 01 (2001): 21-43. News Highland. "Detectives Investigation Paramiliatry Style Shooting in Derry." Highland Radio (2017). Published electronically 9 November 2017. http://www.highlandradio.com/2017/11/09/detectives-investigating-paramilitary- style-shooting-in-derry/. Normandeau, Andre. "Community Policing in Canada: A Review of Some Recent Studies." American Journal of Police 12, no. 1 (Spring 1993): 57-73. Northern Ireland Statistics and Research Agency, and Northern Ireland Policing Board. "Public Perceptions of the Police, Dpps, and the Northern Ireland Policing Board October 2006." Belfast, 2006. Northern Ireland Statistics and Research Agency, and Northern Ireland Policing Board. "April 2003 Omnibus Survey: Northern Ireland Policing Board Module." Belfast, 2003. ———. "Public Perceptions of the Police, Dpps, and the Northern Ireland Policing Board January 2012 Omnibus Survey." Belfast, 2012. ———. "Public Perceptions of the Police, Dpps, and the Northern Ireland Policing Board September 2005." Belfast, 2005. ———. "Public Perceptions of the Police, Pcsps, and the Northern Ireland Policing Board." Belfast, 2017. Okia, Opolot. "The Role of the Police in the Post Election Violence in Kenya 2007/8." Journal of Third World Studies 28, no. 2 (October 2011): 259-75. Ombati, Cyrus. "Don't Express Your Political Affiliations, Police Now Told in New Code of Conduct." Standard Digital (2016). Published electronically 18 February 2016. https://www.standardmedia.co.ke/article/2000192134/don-t-express-your- political-affiliations-police-now-told-in-new-code-of-conduct. Ombati, Cyrus, and Roselyne Obala. "Governors Demand Operationalisation of County Policing Authority as Police Bosses "Resist"." Standard Digital (2018). Published electronically 3 January. 245 https://www.standardmedia.co.ke/article/2001264810/governors-demand- operationalisation-of-county-policing-authority-as-police-bosses-resist. Omeje, Kenneth, and John Mwangi Githigaro. "The Challenges of State Policing in Kenya." Peace and Conflict Review 7, no. 1 (Fall 2012): 64-87. Ondieki, Elvis. "Police to Sign New Rules to Guide Their Conduct on Duty." Daily Nation (2017). Published electronically 12 March 2017. https://www.nation.co.ke/news/New-rules-to-usher-in-reforms-in-police- force/1056-3846142-jlp2rd/index.html. Osse, Anneke. "Understanding Policing: A Resource for Human Rights Activists." Amsterdam: Amnesty International Nederland, 2006. Ostrom, Elinor, and Gordon Whitaker. "Does Local Community Control of Police Make a Difference? Some Preliminary Findings." American Journal of Political Science 17, no. 1 (Feb 1973): 48-76. Otieno, Charles, Linda Omanya, Jacob Atiang', Catherine Chegero, Eric Atinga, Alfred Shereta, and Rapudo Hawi. "Joint Response to the Cic on the National Police Service Proposed Regulations." London: Usalama Reforms Forum, 2015. Paisely, Rev. Ian, Sr. "Paisley: Genesis to Revelation - Face to Face with Eamonn Mallie." By Eamonn Mallie (13 January 2014). Palmer, Stanley H. Police and Protest in England and Ireland, 1780-1850. Cambridge: Cambridge University Press, 1988. Pankhurst, Donna. "Issues of Justice and Reconciliation in Complex Political Emergencies: Conceptualizing Reconciliation, Justice and Peace." Third World Quarterly 20, no. 1 (February 1999): 239-56. Paul, Ainsworth. "Closure 'Could Damage Relations'." South Belfast News (2012). Published electronically 10 March 2012. http://belfastmediagroup.com/closure- %E2%80%98could-damage-relations%E2%80%99/. Perry, Robert. "The Devolution of Policing in Northern Ireland: Politics and Reform." Politics 31, no. 3 (2 September 2011): 167-78. Police (Northern Ireland) Act 2000. Police Service Northern Ireland. "Police Recorded Security Situation Statistics Northern Ireland: Security Related Incidents, 1969-2017." Belfast, 2017. Police Service of Northern Ireland. "Www.Psni.Police.Uk." https://www.psni.police.uk/. Policing in Israel: Studying Crime Control, Community Policing, and Counterterrorism. Boca Raton, FL: CRC Press, 2016. 246 "Portrush: 15 Arrested over Drugs Offences." BBC News (2014). Published electronically 28 December 2014. https://www.bbc.com/news/uk-northern- ireland-30616424. Pouliot, Vincent. International Security in Practice: The Politics of Nato-Russia Diplomacy. New York: Cambridge University Press, 2010. Prempeh, H. Kwasi. "Constitutionalism, Ethnicity and Minority Rights in Africa: A Reply to Jeremie Gilbert." International Journal of Constitutional Law 11, no. 2 (1 April 2013): 438-43. Przeworski, Adam, Michael E. Alvarez, Jose Antonio Cheibub, and Fernando Limongi. Democracy and Development: Political Institutions and Well-Being in the World, 1950-1990. Cambridge: Cambridge University Press, 2000. UK Parliment. Public Processions (Northern Ireland) Act Rauch, Janine. "Police Reform Anbd South Africa's Transition." In Crime and Policing in Transitional Societies. Jan Smuts House, University of the Witwatersrand, Johannesburg, 2000. Rea, Desmond, and Robin Masefield. Policing in Northern Ireland: Delivering the New Beginning? Liverpool: Liverpool University Press, 2014. Rebert, Johann. "Community Policing in Sri Lanka: A Foundation for Wider Police Reform?" (2017). Published electronically 1 March. https://eu- civcap.net/2017/03/01/community-policing-in-sri-lanka-a-foundation-for-wider- police-reform/. Reilly, Benjamin. "Institutional Designs for Diverse Democracies: Consociationalism, Centripetalism, and Communalism Compared." European Political Science 11, no. 2 (June 2012): 259-70. The Constitution of Kenya. 27 August 2010. Republic of Kenya. "Independent Policing Oversight Authority Act, 2011." Nairobi, Kenya: Republic of Kenya, 2011. ———. "National Police Service Act, 2011." Nairobi, Kenya: Republic of Kenya, 2011. ———. "The National Police Service Commission Act, 2011." Nairobi, Kenya: Republic of Kenya, 2011. Reynal-Querol, Marta. "Ethnicity, Political Systems, and Civil Wars." Journal of Conflict Resolutions 46, no. 1 (2002): 29-54. 247 Rice, Xan. "Kenya Votes for New Constitution." The Guardian (2010). Published electronically 5 August. https://www.theguardian.com/world/2010/aug/05/kenya- votes-for-new-constitution. Rigby, Andrew. Justice and Reconciliation: After the Violence. Boulder, CO: Lynne Rienner Publishers, 2001. Robertson, Claire. "Beyond 'Tribes': Violence and Politics in Kenya." Origins: Current Events in Historical Perspective 1, no. 7 (2008). Published electronically April. Roht-Arriaza, Naomi. "The New Landscape of Transitional Justice." In Transitional Justice in the Twenty-First Century: Beyond Truth Versus Justice, edited by Naomi Roht-Arriaza and Javier Mariezcurrena. New York, NY: Cambridge University Press, 2006. Rosenbaum, Dennis. "Community Crime Prevention: A Review and Synthesis of the Literature." Justice Quarterly 5, no. 3 (September 1988): 323-95. Rosenbaum, Dennis, and Arthur J Lurigio. "An inside Look at Community Policing Reform: Definitions, Organizational Changes, and Evaluation Findings." Crime & Delinquency 40, no. 3 (1994): 299-314. Rothman, Rozann. "Political Symbolism." In The Handbook of Political Behavior, edited by S.L. Long. Boston, MA: Springer, 1981. Sabel, Charles. "Dewey, Democracy, and Democratic Experimentalism." Contemporary Pragmatism 9, no. 2 (December 2012): 35-55. Sambanis, Nicholas. "Do Ethnic and Nonethnic Civil Wars Have the Same Causes?: A Theoretical and Empirical Inquiry." Journal of Conflict Resolutions 45, no. 3 (June) (2001): 259-82. Sarkin, J. "Promoting Justice, Truth, and Reconciliation in Transitional Societies: Evaluation Rwanda's Approach in the New Millennium of Using Community Based Gacaca Tribunals to Deal with the Past." International Law FORUM du droit international 2, no. 2 (1 May 2000): 112-21. Sarre, Rick, Dilip K. Das, and H.J. Albrecht, eds. Policing Corruption: International Perspectives. Lanham: Lexington Books, 2005. Scarman Tribunal. "Violence and Civil Disturbances in Northern Ireland in 1969." edited by Chairman Justice Scarman. Belfast: HMSO, 1972. Schumpeter, Joseph A. Capitalism, Socialism, and Democracy. New York: Harper & Bros, 1950. 1942. 248 Scott, James C. Seeing Like a State: How Certain Schemes to Improve the Human Condition Have Failed. New Haven: Yale University Press, 1998. Selway, Joel, and Kharis Templeman. "The Myth of Consociationalism? Conflict Reduction in Divided Societies." Comparative Political Studies 45, no. 12 (2012): 1542-71. Shearing, Clifford D. "A Nodal Conception of Governance: Thoughts on a Policing Commission." Policing and Society 11 (2001): 259-72. "Singapore 6th-Least Corrupt Country in the World: Annual Index." (2018). Published electronically 22 February 2018. https://www.channelnewsasia.com/news/singapore/singapore-6th-least-corrupt- country-in-the-world-annual-index-9979996. Smaje, Chris. "Not Just a Social Construct: Theorizing Race and Ethnicity." Sociology 31, no. 2 (1997): 307-27. Smith, Anthony D. Nationalism: Theory, Ideology, History. Malden, Mass.: Polity Press, 2001. Smithey, Lee A. Unionists, Loyalists, and Conflict Transformation in Northern Ireland. New York: Oxford University Press, 2011. Smyth, Jim. "Symbolic Power and Police Legitimacy: The Royal Ulster Constabulary." Crime Law and Social Change 38, no. 3 (October 2002): 295-310. Soifer, Hillel. "Shadow Cases in Comparative Politics." Temple University, 2017. Sparks, Jehan, and Alison Ledgerwood. "When Good Is Stickier Than Bad: Understanding Gain/Loss Asymmetries in Sequential Framing Effects." Journal of Ecperimental Psychology: General 146, no. 8 (2017): 1086-105. Sperber, Dan. Rethinking Symbolism. New York: Cambridge University Press, 1975. Staff Reporter. "After Macedonia Deal, Greece and Russia Call for Cyprus Solution." Cyprus Mail Online (2018). Published electronically 13 June. https://cyprus- mail.com/2018/06/13/after-macedonia-deal-greece-and-russia-call-for-cyprus- solution/. Stanley, William D., and Charles T. Call. "Military and Police Reform after Civil War." In Contemporary Peacemaking: Conflict, Peace Processes and Post-war Reconstruction2008. Stenning, Philip C, and Clifford D Shearing. "Reforming Police: Opportunities, Drivers, and Challenges." Australian and New Zealand Journal of Criminology 38, no. 2 (2005): 167-80. 249 Stone, Christopher E., and Heather H. Ward. "Democratic Policing: A Framework for Action." Policing and Society: An International Journal 10, no. 1 (2000): 11-45. The Independent Commission on Policing for Northern Ireland (Great Britain), and Chris Patten. "A New Beginning: Policing in Northern Ireland: The Report of the Independent Commission on Policing for Northern Ireland." Norwich, England, 1999. The Police Ombudsman for Northern Ireland. "Public Awareness of the Northern Ireland Complaints System." Belfast, United Kingdom, 2001. The World Bank. "Kenya Economic Development, Police Oversight, and Accountability." 2009. Theda, Skocpol. States and Social Revolutions: A Comparative Analysis of France, Russia, & China. Cambridge, UK: Cambridge University Press 1979. TheGlobalEconomy.com. "Political Stability - Country Rankings." https://www.theglobaleconomy.com/rankings/wb_political_stability/. Throup, David. "Crime, Politics, and the Police in Colonial Kenya, 1939-1963." In Policing and Decolonization: Politics, Nationalism, and the Police, 1917-1965, edited by David Anderson and David Killingray, 127-57. New York: Manchester University Press, 1992. Trojanowicz, Robert C, and Bonnie Bucqueroux. Community Policing: A Contemporary Perspective. Cincinatti: Anderson Publishing 1990. ———. Community Policing: How to Get Started. Cincinnati, Ohio: Anderson Publishing Co, 1994. "Unionists Reject Police Badge Design." BBC News World Edition (2001). Published electronically 19 November. http://news.bbc.co.uk/2/hi/uk_news/northern_ireland/1664681.stm. University of Pennsylvania African Studies Center. "Kenya - Ethnic Groups." https://www.africa.upenn.edu/NEH/kethnic.htm. Van Arkadie, Brian. "Reflections on Land Policy and the Independence Settlement in Kenya." Review of African Political Economy 43, no. sup 1: Special Issue: Land, Liberation and Democracy: A Tribute to Lionel (2016): 60-68. "Violence at Belfast Parade Route." BBC News Channel (2009). Published electronically 13 July 2009. http://news.bbc.co.uk/2/hi/uk_news/northern_ireland/8146817.stm. Wadekar, Neha. "The Future of Kenya's Democracy Is Hanging in the Balance." The Atlantic (2017). Published electronically 14 Novemeber 2017. 250 https://www.theatlantic.com/international/archive/2017/11/kenya-election- kenyatta-odinga-violence/545747/. Wakhungu, Judi, Elvin Nyukuri, and Chris Huggins. "Land Tenure and Violent Conflict in Kenya." edited by African Centre for Technology Studies (ACTS). Nairobi: African Centre for Technology Studies (ACTS), 2008. Waki, Philip N., G. McFadyen, P. Kambale, G.M. Kegoro, and D.S. Manjanja. "Report of the Commission of Inquiry into Post-Election Violence (the Waki Report)." Nairobi, Kenya, 2008. Weber, Max. "Objectivity in Social Science and Social Policy." In The Methodology of the Social Sciences, edited by Edward A Shils and Henry A Finch, 49-112. New York: The Free Press, 1949. Weiss, Robert Stuart. Learning from Strangers: The Art and Method of Qualitative Interview Studies. New York: Free Press; Maxwell Macmillan International, 1994. Weitzer, Ronald John. Policing under Fire: Ethnic Conflict and Police-Community Relations in Northern Ireland. Albany: State University of New York Press, 1995. "Weston Park Agreement." edited by Northern Ireland Office and Department of Foreign Affairs. Belfast, United Kingdom: Northern Ireland Office Department of Foreign Affairs,, 2001. Williams, Brian N., Ralph S. Brower, and W. Earle Klay. "Community-Centered Police Professionalism: A Template for Reflective Professionals and Learnign Organisationa with Implications for the Co-Production of Public Safety and Public Order." The Police Journal: Theory, Practice and Principles 89, no. 2 (2016): 151-73. Wood, Jennifer D. "Pioneers in Policing: Cliffor Shearing." Police Practice and Research 15, no. 6 (2014): 519-32. Wycoff, Mary Ann, and Timothy N. Oettmeier. Evaluating Patrol Officer Performanc under Community Policing: The Houston Experience. National Institute of Justice, 1994. Zald, Mayer N. "Politics and Symbols: A Review Article." Sociological Quarterly 7, no. 1 (1966): 85-91. Zedler, Barbara. "Parades." Northern Ireland Foundation, https://northernireland.foundation/projects/sharedfuture/parades/. 251 Ziembo-Vogl, Joanne, and Woods Devere, Jr. "Defining Community Policing: Practice Versus Paradigm." Police Studies 19, no. 3 (Fall 1996): 33-50.