From War to Democracy: Dilemmas of Peacebuilding

From War to Democracy

Attempts to introduce democracy in the wake of civil war face a critical problem: how can war-torn societies move toward peace and democracy when competitive politics and hard-fought elections exacerbate social and political conflict? Through a study of six themes (peacekeeping, management of violence, power sharing, political party transformation, elections, civil society, and international reactions to democratization crises) this volume considers the dilemmas that arise in pursuing peace after civil war through processes of democratization. The contributors’ research highlights the complex relationship between democratization, which is competitive, and peacebuilding or efforts to achieve reconciliation. The book offers insights into more effective action in peacebuilding in light of the short-term negative effects that democratization can introduce. It is a thought-provoking work that seeks both to advance theory and to provide policy-relevant findings to facilitate more effective and durable transitions from war to democracy. ANNA K. JARSTAD

is Assistant Professor and Coordinator for the Conflict and Democracy Program in the Department of Peace and Conflict Research at Uppsala University, Sweden.

TIMOTHY D. SISK

is Associate Professor in the Graduate School of International Studies at the University of Denver.

From War to Democracy Dilemmas of Peacebuilding

Edited by ANNA K. JARSTAD

and TIMOTHY D. SISK

CAMBRIDGE UNIVERSITY PRESS

Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, São Paulo Cambridge University Press The Edinburgh Building, Cambridge CB2 8RU, UK Published in the United States of America by Cambridge University Press, New York www.cambridge.org Information on this title: www.cambridge.org/9780521885669 © Cambridge University Press 2008 This publication is in copyright. Subject to statutory exception and to the provision of relevant collective licensing agreements, no reproduction of any part may take place without the written permission of Cambridge University Press. First published in print format 2008

ISBN-13 978-0-511-39492-8

eBook (NetLibrary)

ISBN-13 978-0-521-88566-9

hardback

ISBN-13 978-0-521-71327-6

paperback

Cambridge University Press has no responsibility for the persistence or accuracy of urls for external or third-party internet websites referred to in this publication, and does not guarantee that any content on such websites is, or will remain, accurate or appropriate.

Contents

List of tables

page vii

About the authors

viii

List of acronyms

xiii

Acknowledgments

xvi

Introduction ANNA K. JARSTAD AND TIMOTHY D. SISK Part I 1

2

The perils of war-to-democracy transitions

17

Part II

37

The security context

Peacekeeping and democratization

Part III

5

The political process

Power sharing: former enemies in joint government ANNA K. JARSTAD

80 103 105

When rebels change their stripes: armed insurgents in post-war politics ¨ DERBERG KOVACS MIMMI SO

6

39

Violence in war-to-democracy transitions ¨ GLUND KRISTINE HO

4

15

Dilemmas of war-to-democracy transitions: theories and concepts ANNA K. JARSTAD

VIRGINIA PAGE FORTNA

3

1

134

Post-war elections: uncertain turning points of transition BENJAMIN REILLY

157

v

vi

7

8

9

Contents

Civil society in war-to-democracy transitions ROBERTO BELLONI

182

Part IV

211

International engagement

International responses to crises of democratization in war-torn societies PETER WALLENSTEEN

213

Peacebuilding as democratization: findings and recommendations TIMOTHY D. SISK

239

References

260

Index

284

Tables

1.1 Recent cases of peacebuilding after civil war

page 33

2.1 Effects of peacekeeping on the duration of peace

53

2.2 Mean democratization scores (Polity), by peacekeeping

55

2.3 Democratization one year out

58

2.4 Democratization two years out

59

2.5 Democratization five years out

60

2.6 Democratization one year out, by peacekeeping mission type

66

2.7 Democratization two years out, by peacekeeping mission type

67

2.8 Democratization five years out, by peacekeeping mission type

68

2.9 UN vs. non-UN peacekeeping

72

Appendix: List of cases

77

4.1 Examples of recent power-sharing accords

112

vii

About the authors

is Assistant Professor at the School of Politics, International Studies and Philosophy at Queen’s University, Belfast. He was most recently a Fellow with the International Security Program and the WPF Program on Intrastate Conflict at Harvard University. He received his Ph.D. from the Graduate School of International Studies at the University of Denver (2003). An Italian national, Belloni has extensive work experience in Southeast Europe, in particular BosniaHerzegovina, and also Kosovo and Croatia. His research focuses on peacebuilding and post-war democratization, with particular emphasis on the Balkans. He has published in the Review of International Studies, International Peacekeeping, the Journal of Peace Research, Ethnopolitics, Civil Wars, the Whitehead Journal of Diplomacy and International Relations, International Studies Perspectives and in edited volumes. His recent work includes the volume State Building and International Intervention in Bosnia (2007).

ROBERTO BELLONI

is Associate Professor at the Political Science Department and the Saltzman Institute of War and Peace Studies at Columbia University. Her research focuses on the durability of peace in the aftermath of both civil and interstate wars. She is the author of Does Peacekeeping Work? Shaping Belligerents’ Choices after Civil War (2008) and Peace Time: Cease-Fire Agreements and the Durability of Peace (2004) and has published articles in International Organization, International Studies Quarterly, International Studies Review, and World Politics. She is currently working on a research project on longterm historical trends in war termination. Fortna has been a Fellow at the Hoover Institution at Stanford University (2004–2005) and a Visiting Fellow at the American Academy of Arts and Sciences in Cambridge, MA (2002–2003). She

VIRGINIA PAGE FORTNA

viii

About the authors

ix

was a pre-doctoral and then a postdoctoral fellow at the Center for International Security and Cooperation at Stanford University (1997–1999). Her graduate work was done at the Government Department at Harvard University (Ph.D. 1998). Before graduate school, she worked at the Henry L. Stimson Center, a think tank in Washington, DC. She is a graduate of Wesleyan University. is Assistant Professor at the Department of Peace and Conflict Research, Uppsala University, Sweden. She has specialized in the inter-linkages between conflict resolution and violence. Her research has covered issues such as the dilemmas of democratization in countries emerging from violent conflict, the importance of trust in peace negotiation processes, and the role of international actors in dealing with crises in war-torn societies. Cases analyzed include Guatemala, Northern Ireland, South Africa, Kosovo, and Sri Lanka. Ho¨glund has been a pre-doctoral visiting fellow at the Conflict Management Program, ‘‘Nitze School of Advanced International Studies’’ (SAIS), Johns Hopkins University, Washington, DC, and at the International Institute for Applied Systems Analysis (IIASA), Laxemburg, Austria. She holds a Ph.D. in peace and conflict research from Uppsala University. Recent publications are ‘‘‘Sticking One’s Neck Out’: Reducing Mistrust in Sri Lanka’s Peace Negotiations’’ (2006), with Isak Svensson, in the Negotiation Journal; ‘‘Violence by the State: Official Spoilers and Their Allies’’ (2006), with I. William Zartman, in John Darby (ed.), Violence and Reconstruction; and ‘‘Violence and the Peace Process in Sri Lanka’’ (2005), in the journal Civil Wars.

¨ GLUND KRISTINE HO

ANNA K. JARSTAD

is Assistant Professor and Coordinator for the Conflict and Democracy Program at the Department of Peace and Conflict Research at Uppsala University, Sweden. She specializes in power sharing as a form of conflict management, in large-N studies, and in in-depth studies, for instance in Cyprus, New Zealand, Kosovo, Bosnia and Herzegovina, and Macedonia. Her work also includes research on peacebuilding, democratization, and conflict management in ethnically divided societies. A prior research project focused on dilemmas in nation building, political integration, and ethnic identities in Cyprus. That project was supported by the Swedish Institute of International Affairs and focused on the consequences for conflict resolution given

x

About the authors

the EU accession of the Republic of Cyprus. She holds a Ph.D. in peace and conflict research from Uppsala University. Her doctoral dissertation, Changing the Game: Consociational Theory and Ethnic Quotas in Cyprus and New Zealand (published 2001) investigated how ethnic quotas can contribute to changes in the actor’s ranking order of preferences by upgrading the value of cooperation, thereby contributing to viable peace. Recent publications are ‘‘To Share or to Divide? Negotiating the Future of Kosovo,’’ Civil Wars 9 (3), 2007, and ‘‘Peace by Pact: Data On the Implementation of Peace Agreements,’’ with Ralph Sundberg, in Ashok Swain, Ramses Amer, and Joakim ¨ jendal (eds.), Globalization and Challenges to Building Peace (2007). O is Director of the Center for Democratic Institutions at the Crawford School of Economics and Government, Australian National University. His work focuses on democratization, political institutions, and conflict management, and he has advised numerous governments and international organizations on these issues. He has held visiting appointments at Oxford, Canterbury, and Harvard universities, and his work has received financial support from the Carnegie Corporation of New York, the United States Institute of Peace, the East–West Centre, and the Australian Research Council. His publications include Democracy and Diversity: Political Engineering in the Asia-Pacific (2006), Democracy in Divided Societies: Electoral Engineering for Conflict Management (2001), Electoral Systems and Conflict in Divided Societies (1999), Democracy and Deep Rooted Conflict: Options for Negotiators (1998), and The International IDEA Handbook of Electoral System Design (1997). He has also written for academic journals such as Comparative Political Studies, the Journal of Democracy, International Security, The National Interest, Party Politics, Electoral Studies, the International Political Science Review, the Australian Journal of International Affairs, International Peacekeeping and the Asian Survey. He holds a Ph.D. in political science from the Australian National University. BENJAMIN REILLY

T I M O T H Y D . S I S K is Associate Dean and Associate Professor at the Graduate School of International Studies (GSIS), University of Denver, and Director of the Center for 21st Century Global Governance, a research and policy development institute at GSIS. He also serves as an Associate Fellow of the Geneva Centre for Security Policy in Geneva,

About the authors

xi

Switzerland. Sisk specializes in peace processes and international conflict resolution, especially negotiation, mediation, and international intervention in contemporary wars. His recent research has focused on systematic approaches to conflict assessment and institutions and processes for conflict management. He is currently finishing a book titled Bargaining with Bullets. Violence, Negotiation and International Mediation in Civil Wars. A former Program Officer and Research Scholar at the federally chartered United States Institute of Peace in Washington, DC, Sisk was a Washington-based scholar and analyst of international relations and US foreign policy for fifteen years. He is the author of five books and many articles, including Democratization in South Africa (1995) and Power Sharing and International Mediation in Ethnic Conflicts (1995). Sisk earned a Ph.D. ‘‘with distinction’’ in political science (comparative politics, research methods) from the George Washington University, in 1992, and an MA in International Journalism (1984) and a BA in Foreign Service and German (1982) from Baylor University. is Assistant Professor at the Department of Peace and Conflict Research at Uppsala University in Sweden. She holds a M.Sc. in Peace and Conflict Studies from Uppsala University (1999) and a M.Sc. in International Relations from the London School of Economics and Political Science (2000). She has been a pre-doctoral fellow with the Belfer Center for Science and International Affairs at Harvard University. Her research interests include non-state actors in civil wars, conflict resolution processes, and post-war democratization. In her doctoral thesis she examined why some rebel groups but not others successfully transform to viable political parties following peace agreements in intrastate armed conflicts, with a focus on cases from Cambodia, El Salvador, Mozambique, and Sierra Leone. She has written the research report Democratization and Armed Conflicts in Weak States for the Swedish International Development Cooperation Agency (Sida) and has published in the journal Civil Wars.

M I M M I S O¨ D E R B E R G K O V A C S

P E T E R W A L L E N S T E E N holds the Dag Hammarskjo ¨ ld Chair in Peace and Conflict Research at Uppsala University, Sweden (since 1985) and is the Richard G. Starmann Sr. Research Professor of Peace Studies at the University of Notre Dame, Indiana, USA (since 2006). The second, updated edition of his book Understanding Conflict Resolution: War,

xii

About the authors

Peace and the Global System was published in 2007. Wallensteen directs the Uppsala Conflict Data Program (www.ucdp.uu.se) which publishes annual updates on political conflict in the SIPRI Yearbooks, the Journal of Peace Research, and the Human Security Reports. Recently data on conflict prevention in low-intensity conflicts have been added. Wallensteen also follows UN affairs on issues relating to prevention, negotiation, and resolution of conflict. An assignment is to lead the Special Program on the Implementation of Targeted Sanctions which analyzes concrete experiences of UN and EU sanctions (www.smartsanctions.se). A report was presented to the UN Security Council in 2003 (Making Targeted Sanctions Effective: Guidelines for the Implementation of UN Policy Options, with Carina Staibano and Mikael Eriksson). He is co-editor of International Sanctions: Between Words and Wars in the Global System (2005).

Acronyms

ACS ANC ASEAN AU CNDD-FDD

CODES CPP DDR DPA DPI DRC ECOMOG ECOWAS EPL EPLF EPRDF EU FARC FMLN

FRELIMO FRETILIN

Assembly for Civil Society (Guatemala) African National Congress (South Africa) Association of Southeast Asian Nations African Union National Council for the Defense of DemocracyForces for the Defense of Democracy (Conseil National pour la De´fense de la De´mocratieForces pour la De´fense de la De´mocratie, Burundi) Convention for a Democratic South Africa Cambodian People’s Party disarmament, demobilization, and reintegration Dayton Peace Agreement Department of Public Information (United Nations) Democratic Republic of the Congo Economic Community of West African States Monitoring Group Economic Community of West African States Popular Liberation Army (Eje´rcito Popular de Liberacio´n, Colombia) Eritrean People’s Liberation Front Ethiopian People’s Revolutionary Democratic Front European Union Revolutionary Armed Forces of Colombia (Fuerzas Armadas Revolucionarias de Colombia) Farabundo Martı´ National Liberation Front (Frente Farabundo Martı´ para la Liberacio´n Nacional, El Salvador) Liberation Front of Mozambique (Frente de Libertac¸a˜o de Moc¸ambique) Revolutionary Front of Independent East Timor (Frente Revoluciona´ria de Timor-Leste Independente) xiii

xiv

List of acronyms

FUNCINPEC National United Front for an Independent, Neutral, Peaceful, and Cooperative Cambodia (Front Uni National pour un Cambodge Inde´pendant, Neutre, Pacifique, et Coope´ratif) GAM Free Aceh Movement (Gerakan Aceh Merdeka, Indonesia) GDP gross domestic product HDZ Croatian Democratic Union (Hrvatska demokratska zajednica) ICG International Crisis Group ICTY International Criminal Tribunal for the Former Yugoslavia IDEA International Institute for Democracy and Electoral Assistance IFOR Implementation Force (Bosnia and Herzegovina) IRA Irish Republican Army (Northern Ireland) KFOR Kosovo Force KPU National Elections Commission (Indonesia) LRA Lord’s Resistance Army (Uganda) LTTE Liberation Tigers of Tamil Eelam (Sri Lanka) LURD Liberians United for Reconciliation and Democracy MINURSO United Nations Mission for the Referendum in Western Sahara MINUGUA United Nations Verification Mission in Guatemala MINUSTAH United Nations Stabilization in Haiti MODEL Movement for Democracy in Liberia MPLA Popular Movement for the Liberation of Angola (Movimento Popular de Libertac¸a˜o de Angola) NATO North Atlantic Treaty Organization NGOs non-governmental organizations NPFL National Patriotic Front of Liberia NPP National Patriotic Party (Liberia) OAS Organization of American States OECD/DAC Organization for Economic Cooperation and Development / Development Assistance Committee OHR Office of the High Representative OLF Oromo Liberation Front (Ethiopia) ONUMOZ United Nations Operations in Mozambique ONUSAL United Nations Observer Mission in El Salvador

List of acronyms

OSCE Palipehutu– FNL PDK PR PSA RENAMO RUC RUF SDA SDS Sida SFOR SLMM SNTV SPLM SRSG SSR SWAPO UCK UN UNAMSIL UNAVEM UNDP UNITA UNMIK UNTAES UNTAET UNTAG URNG

xv

Organization for Security and Cooperation in Europe Party for the Liberation of the Hutu People–Forces for the National Liberation (Parti pour la libe´ration du peuple Hutu – Forces nationales de libe´ration) Democratic Party of Kosovo (Partia Demokratike e Kosove¨s) proportional representation power-sharing accord Mozambic National Resistance (Resisteˆncia Nacional Moc¸ambicana) Royal Ulster Constabulary (Northern Ireland) Revolutionary United Front (Sierra Leone) Party of Democratic Action (Stranka Demokratske Akcije, Bosnia and Herzegovina) Serb Democratic Party (Srpska Demokratska Stranka, Bosnia and Herzegovina) Swedish International Development Cooperation Agency Stabilization Force (Bosnia and Herzogovina) Sri Lanka Monitoring Mission single non-transferable vote Sudan People’s Liberation Movement Special Representative of the Secretary-General (United Nations) security sector reform South West Africa People’s Organization (Namibia) Kosovo Liberation Army (Ushtria C¸livimtarc e Kosove¨s) United Nations United Nations Mission in Sierra Leone United Nations Angola Verification Mission (I, II, III) United Nations Development Program National Union for the Total Independence of Angola (Unia˜o Nacional para a Independeˆncia Total de Angola) United Nations Mission in Kosovo United Nations Transitional Administration for Eastern Slavonia United Nations Transitional Administration in East Timor United Nations Transition Assistance Group (Namibia) Guatemalan National Revolutionary Unity (Unidad Revolucionaria Nacional Guatemalteca)

Acknowledgments

This project has been a truly collective endeavor where all the authors have commented and contributed to each other’s work. The book’s editors, Anna Jarstad and Timothy D. Sisk, are deeply grateful to all authors in the project for making it such a rewarding task. Our first meeting was held in April 2005 in Sandhamn, Sweden, where we discussed the theoretical framework and each contribution to the project. The project was generously supported by research awards from the Swedish Vetenskapsra˚det (Research Council) and the Grant Program of the United States Institute of Peace, and the editors thank these organizations for their kind contributions. The project participants would like to acknowledge a number of scholars who have contributed comments and insights on the project papers, including panel chairs and discussants at the 2006 Annual Meeting of the International Studies Association, namely Larry Diamond, Roland Paris, and Donald Rothchild. Thania Paffenholz also deserves a special thanks for her comments on the chapter on civil society. We are grateful for the recommendations of the two anonymous reviewers who read the draft manuscript for Cambridge University Press. Additionally, International IDEA (the Institute for Democracy and Electoral Assistance) and the University of Denver’s Graduate School of International Studies also graciously supported the March 26–27, 2006, Vail Symposium ‘‘Dilemmas of Democratization in War-Torn Societies’’ at which the papers in this project were presented. Judith Large at IDEA is thanked especially for her ongoing support of the project and for comments and insights at an early design phase of the research. The authors are grateful to the participants at Vail for their especially valuable contributions to the project’s designs and especially its framing and conclusions. They are: Pauline Baker President, the Fund for xvi

Acknowledgments

xvii

Peace; Elizabeth Cousens, Vice President International Peace Academy; Tom Farer, Dean, Graduate School of International Studies, University of Denver; Karen Feste, Professor, Graduate School of International Studies and Director, Master of Arts Program in Conflict Resolution, University of Denver; Matthew Hodes, Director, Conflict Resolution Program, the Carter Center; Carlos Juarez Dean, College of International Studies, Hawai’i Pacific University; Sakuntala KardirgamarRajasingham, Head, South Asia Program, International IDEA; Brett Lacy, Former Officer in Charge, Liberia, the Carter Center; Johanna Mendelson Forman, Senior Program Officer, Peace, Security and Human Rights, United Nations Foundation; Hiroko Miyamura, Senior Political Officer, Electoral Assistance Division, United Nations; Lawrence Robertson, Conflict Specialist, Office of Conflict Management and Mitigation, US Agency for International Development; Ambassador Gordon Smith, Director, Centre for Global Studies, University of Victoria; Massimo Tommasoli, Director of Operations, International IDEA. Participants of the research seminar at the Department of Peace and Conflict Research, Uppsala University, especially Cecilia Albin, Thomas Ohlson and Mats Hammarstro¨m, have contributed with valuable comments on the theory chapter, and Ralph Sundberg has compiled data on peace agreements. Additionally, several graduate students at GSIS have made especially helpful comments on the book: they are Kris Bauman, Andrew Barwig, Sumani Dash, and Arturo Lopez-Levy. The editors and authors would also like to pay tribute to the late Donald Rothchild, who passed away in February 2007 following a brief but aggressive illness. Don read and commented on several of the chapters in this book, in his usual collegial style, and his prolific work on peacebuilding in divided societies was formative in our collective learning. His life as a scholar will continue to be one that we aspire to emulate.

Introduction ANNA K. JARSTAD AND TIMOTHY D. SISK

Introducing democracy in the wake of civil war raises a stark question: How can societies shattered by war, with all the deep social enmity, personal suffering, and economic devastation that war brings, simultaneously move toward peace and democracy when competitive politics and hard-fought elections exacerbate social and political conflict? This book explores this question from two somewhat disparate strands of scholarly research: democratic transition theory and practice, which emphasizes, in the move from authoritarian rule to more democratic politics, elite-negotiated democratization pacts, popular mobilization, political party transformation, constitution making, electoral design, and resurrection of civil society; and theory and practice of post-war peacebuilding, with its emphasis on the elite and public negotiation of comprehensive peace agreements, the search for security through ceasefires, demobilization of armed forces, inclusion and reconciliation, external security guarantees, and long-term conflict transformation. Introducing democracy in the wake of war has become a standard practice: since the 1990s, democratization is an integral part of international peacebuilding missions in the wake of civil war. Democracy and peace – two often-desired goals – are promoted in war-torn societies shattered by war. However, today’s headlines – from Afghanistan, Kosovo, Nepal, and the Democratic Republic of Congo (DRC) to name but a few – reflect the evident dilemmas of war-to-democracy transitions. In these and other cases, democracy and peace do not always move forward hand in hand: sometimes, advances in democratization threaten peace, and the compromises necessary for peace restrict or defer democratization. This book is about the dilemmas that arise in pursuing peace after civil war through processes of democratization. We seek to identify and evaluate the core dilemma of peacebuilding versus democratization, and several manifestations of this dilemma, in six issue areas: peacekeeping, management of violence, power sharing, political 1

2

Anna K. Jarstad and Timothy D. Sisk

party transformation, elections, civil society, and international reactions to democratization crises. This research effort enhances the understanding of the complex relationship between the two processes, with an eye toward more effective action in peacebuilding. The book also outlines ways to accommodate the negative effects that occur when the processes of democratization and peacebuilding clash. It proposes designs of peace missions that avoid creating dilemmas, but also identifies situations where dilemmas cannot be overcome and where a choice has to be made between efforts to promote peace or democracy. In this way, the research presented in this volume seeks both to advance theory and to provide policy-relevant findings to facilitate more effective and durable war-to-democracy transitions.

War-to-democracy transitions: patterns and rationales In bringing armed conflicts in the 1990s and early 2000s to an end, a critical question emerged for policymakers seeking to secure sustainable peace: How can the international community assist societies wracked by internal war to transform in a way that deep-rooted social conflicts can be ameliorated through non-violent means? Two contending answers to this question have been put forward: one is to separate warring parties by creating new sovereign states, especially if the war has been fought among territorially distinct ethnic, linguistic, or national groups (Kaufmann 1997). The other is to end the war by encouraging the parties to negotiate a settlement and to undergo a warto-democracy transition within an existing state, in which conflicts on the battlefield or the street are ended through the sequenced introduction of democracy: elections, parliamentary politics by political parties, independent judicial institutions, and resuscitation of civil society all underpinned by a basic floor of human rights usually enshrined in newly negotiated constitutions.1 No post-Cold War civil war has been terminated by a peace agreement stipulating partition, in line with the first plausible outcome. Indeed, of the instances of partition of existing states since the end of the Cold War, none were the outcome of a negotiated peace agreement between the government and the armed opposition group (although 1

Recent evidence suggests that today’s wars are much more likely to end at the peace table than on the battlefield; see Eriksson and Wallensteen (2004).

Introduction

3

some settlements involve extensive decentralization that approaches internal partition). The breakup of the former Soviet Union was a disintegrating empire, and it occurred not as a result of a civil war. The breakup of Yugoslavia was a result neither of a peace deal, nor of a civil war. On the contrary, the wars in former Yugoslavia occurred after the declaration of independence by the respective former republics. The independence of Eritrea (de facto in 1991 and de jure in 1993) was never agreed in a peace agreement, but proclaimed after the military victory of the opposition, and a subsequent referendum. Timor Leste (formerly East Timor) achieved its independence after an agreement between Portugal and Indonesia; no negotiations took place between the Revolutionary Front of Independent East Timor (FRETILIN) and Indonesia. Also in this case, independence was approved in a referendum (UCDP 2007). Indeed, many of the internal armed conflicts involving territorial claims have been ‘‘terminated’’ by cease-fire agreements but have not advanced to comprehensive peace settlements in part because the international bias against the creation of new states as an outcome of civil war usually takes partition off the table: Cyprus, Azerbaijan (Karabagh), and Georgia are all examples. The second alternative, to encourage the warring parties to reach a comprehensive, negotiated settlement featuring a transition to democracy, is essentially the default approach of the international community in its response to end contemporary wars. The bias against partition of existing states in the international system is one reason, but it is not the only one. The other is that partition in itself does not solve the problem of contending social groups living together; it only rearranges territorial borders but does not solve the problem of managing social conflict (Chesterman, Farer, and Sisk 2003). Thus, the move toward democracy after war is the imperative even in those instances where prior historical legacies have led to newly independent states such as Bosnia and Herzegovina or Timor Leste. War termination today is principally about building anew or to rebuilding functioning, secure, stable, and democratic (or ‘‘republican’’) states (Barnett 2006). Data generated from the Uppsala Conflict Data Program demonstrate that the most common provision in peace accords for resolving conflicts over government is the holding of elections, while peace accords after conflicts over territory often establish local governance over the disputed territory. After a peak during 1991–1992 when fiftyone armed conflicts were active, the number of conflicts has for the last

4

Anna K. Jarstad and Timothy D. Sisk

couple of years decreased to a level equivalent to the levels of the 1970s (around thirty armed conflicts). Since 2004, all conflicts have been recorded as some type of intrastate conflict.2 Peace agreements were concluded in one-third of armed conflicts that have been active since the end of World War II, and more conflicts are being settled than new ones are emerging (Harbom and Wallensteen 2005; Harbom, Ho¨gbladh, and Wallensteen 2006). It is not surprising, then, that comprehensive peace agreements in civil wars today – from Namibia in 1990 to more recent cases of Afghanistan, Liberia, or the Ivory Coast – generally envisage democracy as the end-state of a peacebuilding transition process, replete with promises for the full protection of human rights, for electoral processes in lieu of battlefield encounters, for transitional justice mechanisms often lenient to those who have waged violence, and for the promised arbitration of disputes through law instead of the rule of the gun.3

Dilemmas: international community perspectives The actors involved in war-to-democracy transitions view the dilemma of peacebuilding versus democratization from different perspectives. For whom is this dilemma experienced, and how? The external motives for post-war democratization are compelling. In the cold reality of negotiated peace agreements following civil war today, where the international community’s normative and material levers of inducement are ubiquitously brought to bear, the war-termination choice for a process of democratization is today a preferred choice. For the international community democratization is a process by which the root causes and articulated grievances of the parties can be 2

3

According to Uppsala Conflict Data Program (UCDP), the conflict between Iraq and the coalition of the United States, the United Kingdom, and Australia was coded as interstate in 2003 and as internationalized internal armed conflict for the years 2004, 2005, and 2006. An armed conflict is defined by UCDP as a contested incompatibility that concerns government or territory or both, where the use of armed force between two parties results in at least twenty-five battle-related deaths. Of these two parties, at least one has to be the government of a state. For detailed definitions of the different categories of armed conflicts, see www.ucdp.uu.se. In this book, the term civil war is used more broadly than the UCDP definition of intrastate war. Here it also refers to minor internal armed conflicts that do not meet the UCDP criteria. For an evaluation of post-war peacebuilding, see Paris 2004.

Introduction

5

negotiated without recourse to bloodshed and, ideally, consistent with the norms and principles outlined in international law (Franck 1992). Likewise, democratization is increasingly linked to state building, for without an electoral process there is no mechanism for generating internal legitimacy for peace agreements. As Benjamin Reilly appropriately observes, ‘‘In any transition from conflict to peace, the creation or restoration of some form of legitimate authority is paramount . . . the support of the citizenry must be tested and obtained’’ (Reilly 2003a: 174). The faith-like belief in an ‘‘internal’’ democratic peace in the postCold War era is as strong as international liberalism’s devotion to an international democratic peace. Kofi Annan, the seventh SecretaryGeneral of the United Nations (UN), succinctly described the connection between democracy and peace: At the center of virtually every civil conflict is the issue of the State and its power – who controls it, and how it is used. No conflict can be resolved without answering those questions, and nowadays the answers almost always have to be democratic ones, at least in form . . . Democracy is practised in many ways, and none of them is perfect. But at its best it provides a method for managing and resolving disputes peacefully, in an atmosphere of mutual trust.4

For the international community, a war-to-democracy transition has a certain undeniable appeal: the alternatives of authoritarian control or partition are most often shelved as untenable outcomes for the international community. But at the same time, democratization and peacebuilding introduce acute dilemmas for external actors. Pauline Baker insightfully summarizes the inherent tensions in international action in war termination, stemming from the countervailing pressures within the international community (and, conceivably, within individuals such as policymakers who are internally weighing alternative approaches to war termination): [c]onflict managers tend to concentrate on short-term solutions that address the precipitous events that sparked the conflict; above all, they seek a swift and expedient end to the violence. Democratizers tend to concentrate on longer-term solutions that address the root causes of the conflict; they search for enduring democratic stability. The former see peace as a precondition for

4

UN Secretary-General Kofi Annan, ‘‘Why Democracy is an International Issue,’’ Cyril Foster Lecture, Oxford University (UK), June 19, 2001 (available at www.un.org/News/ossg/sg/stories/statements_search_full.asp?statID=11).

6

Anna K. Jarstad and Timothy D. Sisk

democracy; the latter see democracy as a precondition for peace. (Baker 2001: 760)

External actors face perplexing problems, for example whether to include perpetrators of violence in power sharing, to hold elections despite insufficient security (with the hope that violence will ebb), to bargain mostly with elites or to try to engage a wider group of parties (such as political parties or civil societies), or whether to engage rebel forces with a view toward their transformation into political parties. These issues also arise when international actors are considering the extent of their involvement in civil war termination efforts, in how to engage (such as helping parties design the course of events in a war-todemocracy transition, or ‘‘sequencing,’’ and when to leave). Peacebuilders in war-torn societies face the difficult challenges of providing security, fostering resuscitation of civil society, transforming armed actors into human-rights-abiding democrats, providing basic humanitarian relief and ‘‘peace-divided’’ development, and breaking the rent-seeking ties of political economy that fueled the war for states and rebel forces alike (Collier et al. 2003). Perhaps the most difficult dilemma faced by international actors, particularly in UN peace operations, are challenging questions over the use of coercive measures such as force. Use of force by peace operations to buttress a negotiated settlement, especially when the legitimacy of action by the international community is disputed or resisted (see the respective chapters by Virginia Page Fortna, Kristine Ho¨glund, Roberto Belloni, and Peter Wallensteen in this volume), is risky and prone to backfire. Should the UN be in the business of ensuring democracy at the barrel of a gun?5 When international actors engage, their interests may be insufficiently aligned causing a disconnection among the various types of international actors who – generally with good intent – engage in the efforts to bring peace to war-shattered states. Roberto Belloni shows that coordination problems among international actors have been central barriers to the deepening of peace in efforts to engage civil 5

Joanne Mariner of Human Rights Watch argues that in the case of Haiti, for instance, it was important for the UN to use its military clout to prevent violent disruption of the country’s elections. She writes that ‘‘It is crucial for the elections to be credible in the eyes of the Haitian people. Otherwise, instead of advancing much-needed stability they could trigger yet another crisis.’’ See Haiti: Secure and Credible Elections Crucial for Stability, available at http://hrw.org/english/docs/ 2006/02/06/haiti12611.htm.

Introduction

7

society in the war-to-democracy transition process in Bosnia and Herzegovina. Peter Wallensteen shows that international responses are episodic in the face of crises in war-to-democracy processes, and that problems of coordination and will inherently limit the ability of the international community to help parties negotiate successfully to overcome the turbulence of such transitions. Moreover, they face difficult choices over the instruments of support and coercion that could potentially be brought to bear. Finally, there is increasingly concern about the issue of authority in war-to-democracy transitions. The international community has assumed a more authoritative role through extensive international oversight or even transitional administration in cases either where the local authorities fail to prevent crimes against humanity and mass violence (e.g., Sierra Leone, Timor Leste), or when the state itself has been culpable in committing atrocities (e.g., in Cambodia and Kosovo, where government authorities were accused of genocidal crimes). A difficult challenge arises as a result of the need for firm international control of the situation to manage problems of spoiler violence, or to organize elections, while engaging in such a way that empowers local actors (such as electoral management bodies), affirms state sovereignty, and leads to a capable, functioning state when the international community’s oversight ends. Simon Chesterman has shown that this challenge is inherent in transitional administrations and that the United Nations, especially, faces the problem of building democracy in war-torn societies in ways that are fundamentally undemocratic (Chesterman 2004).

Dilemmas: protagonist perspectives The endogenous motives for civil war protagonists is simple: democratization provides a set of rules under which conflict can continue to be waged through formal, rule-oriented institutions such as electoral and parliamentary processes that offer a fundamental floor of human rights in the event one party or another finds itself on the losing side of collective decision-making processes. In John Rawls’ classic book A Theory of Justice (1971), he postulates hypothetical negotiations among individuals seeking to establish anew a political community while ignorant about their future positions and status. In peace talks, the protagonists negotiate the future through what is essentially a new Rawlsian social contract, albeit without a fully obscured ‘‘veil of

8

Anna K. Jarstad and Timothy D. Sisk

ignorance.’’6 As Viktor Vanberg and James Buchanan (1989: 61) have argued: ‘‘Cooperation can replace conflict only if the differing interests, held with different intensities by different persons, can be traded-off or compromised, actually or symbolically, in a social contract.’’ Civil wars by definition feature factions that have some degree of coherence; indeed, the coherence of a protagonist group, such as a rebel force that seeks to represent an ethnic interest, is a key variable in explaining the likelihood of negotiated settlements to civil wars in the first place. Civil war protagonists view the conclusion of peace agreements through democratization as attractive, but risky; likewise, they view peacebuilding approaches such as power-sharing pacts as less risky, but unattractive. As Mimmi So¨derberg Kovacs and Anna K. Jarstad claim in their respective chapters of this volume, protagonists’ aversion to democratization and peacebuilding poses severe obstacles for a war-to-democracy transition. Rebels are not always interested and able to transform into political parties, because if they emerge as a political party, they risk losing the election. Power sharing can provide guaranteed inclusion – and thereby an incentive for such transformation – but how does inclusion of former warring parties affect the quality of democracy that emerges? In turning to the war-to-democracy formula for war termination, protagonists in civil wars face difficult challenges: because the international system fails to adequately and consistently provide for external security guarantees, protagonists face difficult dilemmas of uncertainty. Comprehensive peace agreements do not end conflict (or even violence, as Kristine Ho¨glund demonstrates); they simply set up processes that give peace a chance to unfold over time. As civil war negotiations ensue, state incumbents or rebel forces find obstacles of democratization due to grave uncertainty for the future: they have an insufficient capacity to determine whether through democratic processes – notably but not only elections – they will be enabled to protect their vital interests into the future. The turbulence of war does not offer a safe place from which to make judgments about whether peace agreement guarantees, constitutional guarantees, laws and institutions will be sufficient protection over time. While there may 6

In this sense, comprehensive peace agreements can be considered incipient social contracts, based on the principal of reciprocity that links the pursuit of justice and fairness to the establishment of political institutions; see Rawls (1971: 99).

Introduction

9

be interest in escaping a conflict trap through a democratization formula,7 it is a shaky accord upon which to base future prospects. In sum, as several scholars have artfully shown, the long shadow of future competition in elections creates a classic security dilemma for civil war protagonists over time, one that grips them in a thick pall of uncertainty. On the other hand, peace agreements that limit uncertainty in democratization processes – such as power-sharing pacts – also contain challenges for protagonists in terms of their strategies. Just as during the war parties faced choices over whether to talk or fight, a negotiated peace pact does not alleviate trade-offs related to strategy, it only changes them. That is, in post-war transitions, especially as elections loom and mobilization of constituencies heightens, protagonists must choose strategies that simultaneously can maximize their vote share – often, by emphasizing lines of conflict and difference – while needing to conciliate with opponents in implementation of a peace pact (such as disarmament). Protagonists in war-to-democracy dilemmas face these challenges on a daily basis: cooperation and conflict go side by side as bargainers in implementing peace agreements pursue countervailing pressures of mobilization and conciliation.8 The issues of timing and sequencing are both sources of the dilemma between efforts to promote democracy and peace – and key to a possible way to a synchronized war-to-democracy transition. As the chapter by Virginia Page Fortna in this volume indicates, here protagonists respond to the putative assurances of external parties that – through 7

8

Formulas are broad principles framed to narrow the parameters of a conflict’s outcome; a formula defines an overarching concept that frames the parameters of the solution and defines the terms of trade or establishes a principle under which the conflict can be cooperatively managed. In economist’s terms, it defines the contract zone. In order for parties to accept a formula, often but not always proposed by a mediator, it must be seen as just and satisfactory; cover all major issues; incorporate all sides’ demands; and contain a basic vision of post-war arrangements. A formula is not a settlement, but rather a statement of the scope of the conflict’s outcomes and the general procedures to get the parties to settlement: a formula is an agreement on certain basic conceptual issues needed to be resolved before the bargaining on details can begin in earnest, for example a general declaration of principles or framework agreement. See Druckman (1986). For a review of the international community’s experience in post-war governance, see ‘‘Governance in Post-Conflict Situations: Lessons Learned,’’ United Nations Development Program and the Christian Michelsen Institute, May 2004, available at www.cmi.no/events/?undp-2004-governance-in-post-conflict-situations.

10

Anna K. Jarstad and Timothy D. Sisk

the deployment of a peace operation – some of their problems of uncertainty and strategy can be mitigated by the security presence and skillful mediation diplomacy of outside actors, augmented as it normally is through humanitarian relief and pledges of long-term development aid. While engagement with external, international actors by civil war protagonists raises a number of problems and obstacles – the primary one is certainly security. Are the external guarantees for ensuring compliance sufficient to allay protagonists’ fears while choosing a peace-through-democratization formula? The commitment problem is especially acute in the long term: today, protagonists well know that peacekeeping missions do not last forever, and indeed there are pressures on the United Nations, for example, to manage a transition quickly so as to move resources on to the next crisis (today, shifting from Liberia to Sudan). Thus, external capacities to resolve protagonists’ commitment problems are temporary: over time, this issue, too, cannot be avoided. Settlements in civil wars reflect the convergence point of the parties’ preferences over new rules structures, or institutions, for the state once arms have been laid down. Waterman (1993: 292) argues that ‘‘civil wars are conflicts over political order,’’ and settlements in them entail the ‘‘re-creation of the conditions for a viable, common political order.’’ Importantly, settlements do not end conflicts: they are simply agreements to continue bargaining under consensually defined rules of interaction. Not surprisingly, settlements in internal conflicts often take the form of new constitutions or significant packages of amendments to existing constitutions. In the course of formal substantive negotiations, parties formulate their positions based on their expectations of how the structure of the new institutions will serve their interests; they exercise ‘‘analytical imagination’’ about the costs and benefits of alternative institutions, such as the electoral system (Sisk 1995). Therefore, settlements do not definitively end civil wars, but instead they are promises to end conflicts by creating new democratic rules of the game to which all parties at the table can agree.

Exploring the dilemmas This book investigates the dilemmas of democratization in war-torn societies. In the first chapter (Part I), Jarstad investigates the tensions between peacebuilding imperatives and democratization more fully;

Introduction

11

presenting in depth and in reference to the existing literature a typology of dilemmas that captures their multidimensional nature. The four types of dilemmas that may arise when the processes of democratization and peacebuilding have adverse effects on each other are subcategories of the overarching dilemma and pertain to trade-offs between efforts to promote peace or democracy. The first type of dilemma, the horizontal dilemma, regards the relation between the elites of warring parties and democratic political parties that experience a trade-off between inclusion and exclusion. The vertical dilemma entails the difficult choice between efficacy and legitimacy. It regards the relation between the elite and mass politics. For the sake of legitimacy, involvement of the people in all phases of the peace process is desirable. But for the sake of efficacy, certain negotiations need to be held in secret. The systemic dilemma refers to the issue of ownership, of local versus international control of the processes of democratization and peacebuilding. International involvement may be necessary to end violence and to facilitate negotiations. But, democratization as well as peacebuilding needs to be driven by local motives and actions. The citizens of the countries that receive support to democratization and peacebuilding must feel that they own the processes, and that democracy is not imposed from the outside. The temporal dilemma occurs when there are trade-offs between short-term and long-term effects on democratization and peacebuilding. Efforts to support democratization may in the short run increase the risk of violence, and thereby in the long run undermine the chances for democracy to take root. Peacebuilding may involve restrictions on democratic freedoms such as freedom of the press and mass demonstrations. In the long run, such constraints may cause unrest and not only have a negative effect on democratization, but also turn into an obstacle for implementation of the peace agreement. Part II of the volume addresses the key security issues that arise in a war-to-democracy transition, how they are experienced, and how they are addressed. The management of security raises a whole host of potential problems, challenges, and obstacles to democracy: the demobilization, return, and reintegration of former combatants, lingering threats of political violence, the need for election-related dispute resolution, and solving the ‘‘credible commitment’’ problem through the deployment of peace operations. In Chapter 2, Fortna explores the dilemma that arises when peacekeepers help provide the security and

12

Anna K. Jarstad and Timothy D. Sisk

stability necessary for initial steps toward democracy, while the presence of such foreign actors can undermine the long-term development. She also investigates the possibility that precisely by providing peace and security, peacekeepers may undermine some of the incentives that the pressure of war-making provides to processes of state building and democratization. The third chapter, by Ho¨glund, analyses the complex relationship between democratization and violence. Democratization creates new opportunities and motives for violence. And measures to combat violence – for instance through the use of coercive measures and the inclusion of spoilers in politics – can undermine the democratization process. She argues that new forms of violence emerge as the organized political conflict subsides, rendering old ways of dealing with violence ineffective. Part III is about political transitions. Because protagonists in today’s war rarely lose everything on the battlefield – instead, the war-to-democracy transition is negotiated – they usually end up sharing power either in temporary, transitional arrangements or in permanent, power-sharing constitutions. As Jarstad writes in Chapter 4, power sharing is essentially a coalescence of elites at the top, and one of the enduring concerns about power-sharing solutions are their top-down orientation, their perceived inflexibility, and the allegations that in many instances they reinforce ethnic, nationalist, or extremist tendencies. When rebels lay down their arms, they often do so in the expectation that they might win at the ballot box what they fought for on the battlefield: they seek to transform into political parties. So¨derberg Kovacs investigates this issue in Chapter 5 on rebel-to-political party transformations, illustrating the conditions under which some forces can adapt and evolve, whereas others simply cannot. Likewise, during the war-to-democracy transition critical choices are made over electoral processes: the sequencing of elections, the electoral system formula, the nature of elections (e.g., to a legislature, constituent assembly, or both), and other critical election-related issues such as application of citizenship laws. What are the principal findings from research and practical lessons learned on sequencing, shaping, and managing electoral processes in war-torn societies? Reilly’s contribution in Chapter 6 evaluates the track record on the pivotal issue of elections, and in particular the conditions under which electoral processes may promote democratization but undermine peace, when

Introduction

13

peacebuilding undermines democratization, and the conditions under which electoral processes may contribute to both goals. In Chapter 7, Roberto Belloni focuses on civil society; he argues that there is an urgent need to move beyond the platitudinous endorsement of ‘‘civil society’’ as essential element of peacebuilding by asking the tough questions about which elements of society may be civil, and worthy of inclusion, and those that are not. Part IV evaluates the role of the international community and, in conclusion, provides the policy-relevant findings that emerge from this research. In Chapter 8, Wallensteen explains the lateness of international responses to democratization crises in war-torn societies, arguing that much earlier action is required if such crises are to be avoided and the dual objectives of peacebuilding and democratization are to simultaneously advance. In the book’s concluding chapter, Timothy D. Sisk returns to the question of ‘‘dilemmas for whom’’ and pulls together the key findings of the foregoing contributions on how international actors and protagonists must face up to the reality that there are inherent trade-offs and difficulties in war-to-democracy transitions. The clear message of this research is that in some instances, the dilemmas of democratization simply cannot be avoided: either the imperatives of peacebuilding and conflict management, or the imperatives of democratization, must be singularly chosen in a particular situation as the right path to peace. Generally, the choice is made for peace over democratization (by delaying or restricting elections, for example); this choice is understandable, but it is a myopic one. Rather than foregoing democratization, this research reveals that it must be done more effectively, and cunningly. In some instances, the dilemmas can be avoided: through ingenuity, policymakers and protagonists in conflict may be able to reconcile the divergent choices they face and design sequences of change that can simultaneously advance war-torn societies on the path to peace and toward the best hope for peacebuilding in the long run: democracy.

PART I

The perils of war-to-democracy transitions

1

Dilemmas of war-to-democracy transitions: theories and concepts ANNA K. JARSTAD

War-torn societies entering the path toward democracy and peace face the challenge that efforts to achieve one of these desirable goals can have negative effects on the other. This chapter puts forward the concept of war-to-democracy transitions to capture the dynamic between the two interacting processes of democratization and peacebuilding. Democratization refers to the process of opening up political space, including improvements regarding contestation, participation, and human rights. Peacebuilding regards the post-settlement period and includes the implementation of the peace agreement.1 The focus is on the dilemmas that arise when the two processes have adverse effects on each other. A dilemma is a trade-off situation, where usually the choice is between two bad things. However, here the dilemmas regard two goals widely held to be mutually reinforcing, namely democracy and peace. Such dilemmas for post-war transitions occur when actions taken in the name of democratization have negative effects on the peace process. Dilemmas are also activated when actions taken in the name of 1

A broad understanding is that peacebuilding denotes the various efforts in support of political, institutional, and social transformation necessary to bring about lasting peace (Bertram 1995). Besides the commonly used definition of peacebuilding as employed in the Agenda for Peace, there are at least two other interpretations of peacebuilding. Drawing on Johan Galtung, peacebuilding has been used to refer to non-elite processes, beyond and below the state. Other scholars use the term peacebuilding broadly to refer to peacemaking, peacekeeping, and conflict prevention (Call and Cook 2003). Such efforts may involve the local population and local elites as both initiators and recipients of assistance, as well as intergovernmental organizations and non-governmental organizations (NGOs). For our analysis of war-to-democracy transitions, peacebuilding begins with a peace accord that settles at least one of the incompatibilities at stake in the conflict, such as control over territory or government. Peace agreements do not always put an end to violence. In other cases, the fighting ends but an agreement regulating the incompatibilities is not reached until years later. The focus here is on post-war peacebuilding, i.e., the prevention of a relapse into conflict. Peacebuilding is thus defined as efforts to implement and consolidate violent peace agreements.

17

18

Anna K. Jarstad

peace have negative effects on democratization. In addition, the dilemmas are often interacting: there is seldom one dilemma at a time, and one dilemma may make another dilemma even more critical. Such dilemmas pose severe challenges for both local and international actors engaged in peacebuilding and democratization. The chapter discusses four types of trade-off situations where the choice is between reforms to promote democracy versus efforts to secure peace. These dilemmas are here referred to as the horizontal dilemma (i.e., inclusion versus exclusion), the vertical dilemma (i.e., legitimacy versus efficacy), the systemic dilemma (i.e., local versus international ownership of the processes), and the temporal dilemma (i.e., long-term versus short-term efforts). Failure to deal with such dilemmas can result in a return to war (e.g., Angola 1992 and Liberia 2000). Alternatively it can result in backsliding to authoritarianism as in, for example, Haiti 1994–2005 and Ethiopia after the elections in 2005. It is suggested that a theoretical explanation of why democratization in war-torn societies succeeds or fails needs to include a simultaneous analysis of these four dilemmas. The overarching purpose of this chapter is to enhance our understanding of why efforts to promote democracy and peace do not always go together. This book builds on research stemming from previously separate discourses on democratization, peacebuilding, and conflict theory to construct a framework for the analysis of simultaneous democratization and peacebuilding. The concept of war-to-democracy transitions is developed and explored. This chapter discusses why the combination of a legacy of war, reforms to democratize, and efforts to build peace often result in dilemmas where peacebuilding and democratization have adverse effects on each other. The failure to deal with such dilemmas can have devastating effects, thus undermining both long-term democratization and peace. A broad conclusion is that when the choice is between securing the peace and promoting democracy, peace should be given priority. To date, discourses on democratization and armed conflict have not been integrated, leaving a gap in our understanding of potential tradeoffs between peace versus long-term democratization for societies shattered by conflict. Democracy is commonly understood as a system where diverse interests are managed through ongoing negotiations and accommodated by accountable and legitimate institutions. Although conflicts are seldom fully resolved, democracy supposedly manages

Dilemmas of war-to-democracy transitions

19

them by peaceful means (Commission on Global Governance 1995; Przeworski 1991). In this way democracy and peace reinforce each other. Democracy is consolidated when peaceful means of conflict management are accepted as ‘‘the only game in town’’ (Linz and Stepan 1996: 5). Sustainable peace presupposes a system of governance where diverse interests and grievances are accommodated by negotiations and compromises (Licklider 2005: 35; Wallensteen 2002: 139–144). Democratization and peacebuilding have also been wishfully thought of as parallel and mutually beneficial processes. However, researchers and policymakers have identified an apparent paradox: while democracy as a political system is associated with peaceful conflict management both within and between states, the road to democracy is often conflict-ridden.2 The conditions typical for war-torn societies, as well as the dynamics of and interplay between the two processes of peacebuilding and democratization, contribute to this contradiction. While non-violent conflicts are healthy features of any democracy, violent conflicts undermine the foundations for a functioning democracy. Electoral violence, political assassinations, violent riots, and extreme levels of crime are threats to the new political order and to basic civilian security. Failure to deal with violence can lead to escalation of violence and a vicious circle of retribution and violations of human rights. After a peace deal is reached, the legacies of war tend to linger. Insecurity and unsolved grievances mean that political elites, as well as civil society, remain polarized and that the basis for inclusive ideologies is weak. In combination with a shattered infrastructure, and an economy structured on the spoils of war, this polarization implies that democratization faces particular challenges in post-war societies. This is why the core elements of democracy, such as popular participation, mobilization of interest groups, and open competition between political parties, increase the risk of violent conflict in societies entering a democratization process. In addition, efforts in support of peace deals constrain the process of democratization. For example, the inclusion of former rebels in government for the sake of peace may undermine democratic legitimacy and long-term stability. When this is the case, 2

See, e.g., Brass 1991; Brown, Lynn-Jones, and Miller 1996; Gleditsch and Hegre 1997; Mann 2005; Mansfield and Snyder 1995; Maoz and Russett 1993; Snyder 2000.

20

Anna K. Jarstad

the simultaneous processes of peacebuilding and democratization have adverse effects on each other.

War-to-democracy transition Many contemporary democratization processes take place in societies shattered by war. Previous research has often focused either on democratization (often analyzing transitions taking place in societies that have not experienced armed conflict) or peacebuilding after intrastate armed conflicts without any analysis of democratization. A common assumption is that democracy implies peace and, vice versa, that peace implies democracy. Multidimensional peace operations set out to achieve both peace and democracy. The expectation is that post-war3 transitions result in both peace and democracy. However, the obstacles facing many societies undergoing such transitions suggest a need to combine the experiences of war-shattered societies with an integrated theoretical framework on the processes of democratization and peacebuilding. The transition from relatively stable authoritarianism in Yugoslavia to conflict-ridden democratization in Kosovo and the resumption of war after elections in Angola 1992 give an indication of the broad scope of cases undergoing simultaneous peacebuilding and democratization. These transitions vary a great deal, for example in terms of starting point (e.g., previous history of governance and type of warfare), conduct of international engagement, and progress toward democracy and peace. The plentitude of cases and efforts intended to promote both democratization and peacebuilding gives rise to the need for a

3

The phenomenon of war-to-democracy transition takes place in societies shattered by violent conflict. Such societies are often referred to as ‘‘post-conflict’’ cases. However, the term ‘‘post-conflict’’ invites the interpretation that (1) there has been a violent conflict and (2) the conflict is now solved and violence has ceased. This term is actually a misnomer. By convention, ‘‘post-conflict’’ usually denotes societies affected by armed conflicts, where only parts of the conflicts are solved and where some organized violent behavior still lingers on. Moreover, it is problematic to use the term ‘‘post-conflict’’ when, in fact, non-violent conflicts are part of all societies – also so-called post-conflict societies. Thus, the term ‘‘conflict’’ fails to distinguish those societies where conflicts are settled by peaceful means from societies where violent relations prevail. The term ‘‘post-war’’ is used here, for want of a better one, to refer to situations where the major warfare has ceased, but where some incompatible issues may remain unsolved.

Dilemmas of war-to-democracy transitions

21

comprehensive understanding of post-war transitions where democratization and peacebuilding are treated as separate phenomena, which might or might not reinforce each other. The core question is how a war-to-democracy transition can be achieved peacefully. The challenges of democratization in post-war societies can be studied by simultaneously looking at the peace process and the democratization process: how do these processes develop? when do they reinforce each other? and when and why do they clash? The two intertwined processes, from violent conflict to peace on the one hand, and from authoritarian rule to democracy on the other hand, are here labeled a war-to-democracy transition. The two processes are treated as separate, but related, and the dynamic within and between these processes is in focus. This means that while in practice democratization and peacebuilding often overlap, they are two analytically different processes. The perception of democratization and peacebuilding as two logically separate processes facilitates our understanding of the conditions under which efforts to promote democracy and peace clash. This conceptual framework can be used to analyze the effects of such dilemmas for war-to-democracy transitions and enhance our understanding of how such transformations can be facilitated.

Dilemmas: horizontal, vertical, systemic, and temporal Four types of dilemmas may arise when the processes of democratization and peacebuilding have adverse effects on each other: the horizontal, the vertical, the systemic and the temporal.4 Firstly, the horizontal dilemma concerns the issue of which groups should be represented in the processes of peace and democratization. This decision regards the horizontal relation between the elites of warring parties and of democratic political parties. A selected group of elites may more easily commit to difficult compromises, while comprehensive peace negotiations may result in more lasting agreements by involving 4

The labeling of these four types of dilemmas is my own but builds on previous research. Such research identifies obstacles and dilemmas related to democratization in war-torn societies, specifically to what is here labeled the temporal dilemma (see, e.g., Cousens 2001a; de Zeeuw 2005); the systemic dilemma (see, e.g., Burnell 2005; Chandler 1999; Chandler 2004; Knaus and Martin 2003); the horizontal dilemma (see, e.g., Stedman 1997), and what is here labeled the vertical dilemma (see, e.g., Cousens 2001a; Holsti 1996; Paris 2004).

22

Anna K. Jarstad

all parties with a stake in post-war developments. Broad inclusion is also in line with democratic theory on power sharing, which suggests that the more groups represented in the process, the more democratic it is. Some groups also have legitimate reasons to demand political power after years of oppression and discrimination. Research furthermore suggests that warring parties are more likely to sign a peace deal if they are guaranteed a share in the future government (Lijphart 1977; Walter 2002). However, when broad inclusion is extended to violent parties, it may have negative effects on democratization. Such inclusion can be seen as a reward for violence and thereby contradict the democratic principle of non-violence. This is particularly true when inclusion implies amnesty for persons who have committed human rights violations during the war. Thus, the horizontal dilemma involves a trade-off between inclusion (e.g., for the sake of reaching a peace deal or broad representation) and exclusion (e.g., for the sake of reaching a compromise solution and perhaps also for long-term democratization). This dilemma also affects the prospects for peace. A rebel group that expects to be excluded from future governments and control over part of the territory may find peace too costly. For this reason, a peace deal often stipulates inclusion of the main warring parties in the political process. Peace agreements providing for guaranteed positions in government have been reached in cases such as Burundi 2000, Cambodia 1991, and the Democratic Republic of the Congo 2002. However, it may be difficult to identify which groups need to be included for the sake of peace. If the peace agreement grants some warring parties seats in the government, this may provide an incentive for other groups to use violence to gain the same political status. For example, the power-sharing agreement for Sudan 2004 did not include the rebels in Darfur and fighting continues. Not even the May 2006 agreement included all rebels and these groups demand concessions before laying down their arms. Moreover, inclusion may not end violence: some groups pursue a dual strategy of violence and politics. For democracy to take root, actors mobilized for war have to abandon military methods for negotiations and compromises. These leaders also have to convince their followers that they should demobilize and be prepared for concessions. Ideally, they should also abandon excluding ideologies, such as ethnonationalism, and strive for broad-based democratic support. But also parties commonly labeled as terrorist organizations, such as Hamas, can gain democratic legitimacy via parliamentary elections. Although

Dilemmas of war-to-democracy transitions

23

the hope is that this will lead to a change in Hamas’ politics, democratic institutions do not always produce peaceful democrats. Exclusion of potential spoilers is an alternative strategy for promoting peaceful democratization. It rests on the notion of excluding nationalists and authoritarian actors for the sake of only allowing democratic movements to develop into political parties and compete for power. This strategy more clearly opens up for new actors. However, research has demonstrated that excluded groups to a greater extent return to violent tactics (Gurr 2000a; Stedman 1997). Civil society is often excluded from power-sharing deals. The exclusion of such segments of society leads to an uneven start for parties in a democratization process. One possibility to overcome this negative effect of power sharing is to include a broad range of actors in the peace negotiations and also in the future government. This was done, for example, in the Democratic Republic of the Congo, where the 2002 peace accord included a wide spectrum of society. Other cases include the Ivory Coast 2003 peace deal, which in addition to the warring parties included the main political parties, and the Liberia 2003 agreement, which included not only all warring parties to the conflict, but also representatives from the civil society (Nilsson 2006: 110; Nilsson and So¨derberg Kovacs 2005). Secondly, the vertical dilemma entails the difficult choice between efficacy and legitimacy. It pertains to the relation between elite and mass politics. On the one hand, legitimacy is expected to increase when the people are involved in all phases of the peace process, and also have a chance of influencing the crafting of a new constitution. On the other hand, the elites often have an interest in a non-public process. They want to signal resolve – that they are not prepared to make concessions – in order to get the best deal they can at the negotiation table. After a peace agreement, however, elites are expected to be conciliatory toward former foes and also urge their followers to demobilize and accept concessions. At the same time, some elites use the demands of extremist groups to push for additional concessions from the other parties to the conflict also after a peace deal. An alternative approach, or addition to broad inclusion during peace negotiations, is to strive for public support after a peace deal or democratic constitution is drafted. One way is to hold a referendum to try to ensure such legitimacy. The constitution of Iraq was approved despite the vast Sunni boycott of the October 2005 referendum. However, if

24

Anna K. Jarstad

the new constitution or peace deal does not receive sufficient popular support, the whole process is delayed. This was the case, when the Greek Cypriot majority voted against United Nations SecretaryGeneral Kofi Annan’s plan to unite Cyprus in April 2004. The majority rule typically used in referenda may in divided societies exacerbate polarization. Under such conditions, it might be necessary for new leaders to emerge for negotiations to restart. Thirdly, the systemic dilemma refers to the issue of ownership, that is of international versus local control of the processes of democratization and peacebuilding. Third-party engagement in peacebuilding might generate a dilemma of peace versus democracy. On the one hand, international involvement may be necessary to end violence and to facilitate negotiations. Support for democratic developments aims to promote stability and institutions for conflict management. But on the other hand, both sustainable peacebuilding and democratization depend on the commitment of local people and elites. Contemporary peacebuilding sometimes includes temporary external control over political processes. These structures are not formally accountable to the citizens in these states. In such cases local ownership is weak, thus risking to halt or reverse the process and even to alienate people from democracy as an ideal. For example, when the Office of the High Representative (OHR) stepped in to dismiss elected officials in Bosnia and Herzegovina, some people lost trust in the legitimacy of the system. Such international intervention risks increasing support for extremism and ethnochauvinism. At the same time, implementation of peace agreements includes politically risky steps that local politicians cannot always take, for fear of alienating their own constituencies. But in a post-war context, moderate political parties tend to be lacking or marginalized. For this reason, international support can be pivotal to strengthen the capacity of local moderate groups and thereby facilitate democratization. Ideally, democracy promotion entails support to a locally driven democratization process. It is often the case that at least some local actors demand democratization. Today it would be difficult to withhold from people the right to choose a government – even in cases with unclear status of the state, such as Kosovo and Palestine. At the same time, international actors are reluctant to provide funds without conditions. Assistance is often earmarked for specific purposes, which do not always correspond to the most pressing local needs. Also, international engagements are often

Dilemmas of war-to-democracy transitions

25

short-term, thereby prompting donors to try to speed up the process of democratization. Fourthly, the temporal dilemma regards trade-offs concerning shortterm versus long-term effects on democratization and peacebuilding. Efforts to support democratization may in the short run increase the risk of violence, and thereby in the long run undermine the chances for democracy to take root. Likewise, peacebuilding may involve restrictions on democratic freedoms such as freedom of press and mass demonstrations. In the long run, such constraints may cause unrest and turn into an obstacle for the implementation of the peace agreement. The timing of elections also activates the temporal dilemma. Democracy means rule by the people, and it is difficult to think of another way to ensure democratic legitimacy than through elections. After a war, a democratic election also serves the purpose of bringing a decisive end to the war and of sealing the peace deal. Consequently, elections have come to be seen as the crowning event of the peacebuilding phase and an ‘exit strategy’ for organizations engaged in international peace missions. However, the first post-war election is often riddled with violence and flawed election outcomes. One reason is that some actors expect to lose political power or control over valuable resources as a consequence of the peace and democratization processes. By threats and intimidation, these actors may seek to disrupt the transition, overthrow the election results, or prevent election campaigns or voters from going to the polls. In the worst cases, elections trigger violent conflict and the process of democratization is halted or reversed. For this reason, it has been suggested that elections should be postponed until the conditions are stable and democratic institutions are in place (Cousens 2001b; Mansfield and Snyder 2002a; Paris 2004). The sequencing of peacebuilding has caused dilemmas for Kosovo, for example, where democratic institutions were introduced before its international status was settled, and before reconciliation between the warring groups and democratic norms have taken root. It remains to be seen whether this method to promote democratization will contribute to a successful war-to-democracy transition. Not dealing with these four types of dilemmas can have devastating effects. The Freedom House ranking is a crude, and some would even say deceptive, indicator of democracy, but nevertheless gives us an idea

26

Anna K. Jarstad

about which countries are far from democratic, or ‘‘not free’’ as it is termed by this organization. A quick look at the Freedom House ranking 2005 for the fourteen conflict locations where major peacebuilding missions were launched between 1989 and 1999 (Paris 2004) shows that very few have achieved what Freedom House calls ‘‘freedom.’’ Only Namibia, El Salvador, and Croatia are ranked ‘‘Free.’’ Angola, Cambodia, Kosovo, and Rwanda are ranked ‘‘Not Free,’’ and the rest ‘‘Partly Free.’’5 The lessons learned from missions aiming for democratization and peacebuilding in, for example, Liberia and Haiti demonstrate that when there is a choice between promoting democracy and peace, securing the peace is pivotal. To understand why dilemmas of war-todemocracy transitions occur, and why they can have such devastating effects, it is suggested that theoretical explanations can be found in three areas of research: the efforts to promote peacebuilding, the conflictual character of democratization processes, and the legacy of war.

Potentially negative effects of peacebuilding The ending of war does not always mean the end of violence. In fact, peacebuilding can in extreme cases make things worse. Recently, the notion that negotiated settlements are the best way to end civil wars have been challenged. According to Monica Toft, civil wars that end in a military victory, in particular those that end in a rebel victory, are associated with higher levels of democracy in the longer perspective than are wars that end in a cease-fire or peace agreement (Toft 2003).6 However, contemporary conflicts are increasingly ended by negotiated deals – rather than on the battlefield – and it is therefore important to analyze the obstacles involved in the implementation of peace settlements. Several aspects of peacebuilding may give rise to dilemmas between efforts to promote peace and support to democratization. This is the case when peacebuilding includes multiple tasks with

5

6

The ‘Partly Free’ 2005 cases that are included also in Roland Paris’ analysis of major peace missions 1989–1999 are Nicaragua, Mozambique, Liberia, Bosnia and Herzegovina, Guatemala, East Timor, Sierra Leone (Freedom House 2006; Paris 2004). For definitions and coding according to Freedom House, refer to www.freedomhouse.org/template.cfm?page=15year=2005. See more on this issue in Fortna’s chapter in this volume.

Dilemmas of war-to-democracy transitions

27

competing objectives, when the coordination between a multitude of actors falter, when project-oriented and short-termed missions fail to mitigate negative long-term effects, and when international engagement makes the host society dependent on external support. For peace to become viable, it is not only the conditions that generated the conflict that need to be addressed. In addition, peacebuilding should be designed in such a way that the above-mentioned dysfunctions can be avoided. Furthermore, it is suggested that peacebuilding needs to focus on security and the designing of self-sustaining institutions. Otherwise there is a risk that peacebuilding gives rise to new conflicts or that it undermines democratization. Since the 1990s, democratization has become an integral part of the conflict-prevention agenda. Multidimensional peace operations have become the model for contemporary peace promotion.7 Such missions seek not only to prevent violence, but also to address the root causes of conflict. In An Agenda for Peace (1992), former UN Secretary General Boutros Boutros-Ghali suggested that post-war peacebuilding was a form of conflict prevention via social and political reconstruction. In 1996, the Agenda for Peace was supplemented by the Agenda for Democratization, and in the Framework for Cooperation in Peacebuilding (2001) UN Secretary-General Kofi Annan further developed the notion that democratization is part of peacebuilding (United Nations 2001: Annex 1:1). In December 2005, the UN General Assembly and the Security Council jointly agreed to set up a Peacebuilding Commission to help countries emerging from conflict manage the transition to stability and development. War is costly, but much of the costs of war occur after it is over. Economic reconstruction is often vital for sustainable peace, but peacebuilding missions are often insufficiently funded (Collier et al. 2003; Paris 2004; Woodward 2002). Efforts to ‘‘demilitarize politics’’ are also expensive parts of peacebuilding. This includes disarmament and demobilization of previously warring parties, destruction of weapons, reformation of the security sector, and issues related to

7

The first generation peacekeeping was mainly a military exercise aimed at upholding cease-fires through the method of separating warring parties and thereby providing opportunities for negotiations. The second generation peacekeeping, also labeled multidimensional peace operations, includes both military and civilian components.

28

Anna K. Jarstad

democratization, such as the transformation of rebel groups into political parties (Lyons 2005; Spear 2002). Democratization in war-torn societies also includes support for constitutional and legal reforms, the establishment of election administration, training of election staff and media professionals, political party assistance, international election and human rights monitoring and civil society aid. Experiences from countries such as Cambodia, Ethiopia, Uganda, and Guatemala demonstrate that peacebuilding does not always move societies toward democracy. In these cases, malfunction of institutions is often seen as a key problem: these institutions are largely non-transparent and unaccountable to members or society in general, politically biased, and financially unsustainable. The impact of international engagements may be obstructed by, for instance, interagency rivalries or donors’ ambition to demonstrate short-term results. International support can also foster a ‘‘culture of dependence.’’ This is especially problematic for support for peacebuilding and democratization, as these processes need to be based on local needs and driven by the people in the recipient society (Chandler 2004; de Zeeuw 2005; Diamond 1999b; Paris 2004).

The conflictual elements of democratization Successful democratization requires a minimum level of security and consensus on which territory and people constitute the state: without a defined demos, how can you have democracy? While this stance was advocated already thirty years ago (Rustow 1970), contemporary democratization is nevertheless promoted where these conditions are lacking.8 It is suggested that this old recommendation needs to be taken seriously. However, the necessity of ensuring that these preconditions are in place before embarking on democratization seemed to be forgotten as more and more countries began to hold regular elections. Research commonly assumed that democratic transitions developed gradually, from political liberalization toward a consolidated democracy. Although some violence occurred in earlier democratization waves, it was not seen as a serious threat to peace. On the contrary,

8

This has been noted also by, for example, Juan Linz and Alfred Stepan (1992) and Mansfield and Snyder (2005).

Dilemmas of war-to-democracy transitions

29

the notion of ‘‘democratic peace’’ – that democracies virtually never go to war with each other – prompted both researchers and policymakers to conclude that the expansion of the democratic zone would reduce the risk of armed conflicts (e.g., Dahl 1971; Diamond 1997a; Diamond et al. 1997; Huntington 1991; Linz and Stepan 1996; O’Donnell, Schmitter, and Whitehead 1986). At the same time, research has shown that democracy as well as democratization both contain conflictual elements (e.g., Hegre et al. 2001; MacMillan 2003; Mann 2005; Mansfield and Snyder 1995a and b; Mansfield and Snyder 2002b; Mansfield and Snyder 2005; Maoz and Russett 1993; Oneal, Russett, and Berbaum 2003; Rosato 2003; Thompson and Richard 1997). Not only may the move toward democracy fail; in addition, democratization can exacerbate violent conflict. It is proposed that two aspects of democratization can activate dilemmas in war-torn societies: the essence of the stipulated goal of democratization, namely democracy itself; and the shifts involved in the democratic transitions. Firstly, democracy by definition includes conflictual elements that can have severe effects in societies polarized by violence. In line with Robert Dahl’s conception of polyarchy, key elements of democracy include contestation (including elections), participation, and basic human rights (Dahl 1971). Although democracy ideally stipulates conflict regulation through norms and institutions, it also induces conflict via increased contestation and polarization. Public contestation provides opportunities for replacement of elected officials. As the political candidates compete for votes, they emphasize their differences rather than their common ground. Likewise, during an election campaign the constituencies are mobilized, thus oftentimes enhancing already high levels of polarization. Hence, democracy by definition implies opposition and mobilization along distinctive lines and a certain degree of polarization. This means that democracy provides both opportunities and incentives for conflict. Whereas conflicts are most often managed peacefully in consolidated democracies, this feature can have severe effects in a typical post-war society where weapons abound and people remain polarized. Secondly, the process of democratization entails particular features which increase the risk of violence (Gleditsch and Hegre 1997; Mansfield and Snyder 1995a and b). The movement toward democracy entails changes and shifts of power. This increases the risk of use of violence by those who lose or fear losing power and by those that feel

30

Anna K. Jarstad

that they should gain more. In accordance with the definition of democracy above, democratization refers here to improvements in contestation (such as more political parties and candidates), participation (broadening of the electorate) and human rights (for example, freedom of speech and freedom to organize demonstrations). An emerging field of research specifies the conditions and types of political transformations highly correlated with violent conflict.9 One factor that contributes to the high levels of violence during the beginning of a democratization process is that public expectations tend to be high. People often demand rapid and real improvement in the quality of life. But democracy does not automatically result in other desirable goals such as economic development and equality. Many of the obstacles embedded in the democratization process also relate to the fact that the different democratization components cannot be achieved all at once. Initially there are typically great discrepancies between public demands and the institutional capacity to deliver, as well as between public loyalty to the state and the state’s capacity to control undemocratic elements and make legitimate political decisions. When reforms do not work in concert, risk of violence increases. Given the conflictual elements of democratization, basic consensus – concerning the legitimacy of the state, its territory and its citizens – is necessary to prevent democratization from turning violent. However, it is often disagreement on these specific issues which caused violent conflict to erupt in the first place. Unfortunately, democracy does not solve the issue of what constitutes the state; which territory should belong to it and which people should be citizens of the state. Even after a peace deal many actors continue to challenge the state. Also when former warring parties become part of the government, some continue to use violent tactics with the ambition to change the character of the 9

Some findings suggest that states that are becoming more democratic reduce the risk of interstate war by half. However, especially rocky and rapid transitions or reversals are associated with an increased risk of war (Ward and Gleditsch 1998). Other scholars suggest that because each move toward democracy or authoritarianism entails a risk of violent conflict, and because democracy is the most stable regime type, rapid democratization is less risky than a gradual process. Furthermore, not only transitions, but also certain phases are associated with conflict. The initial phase of democratization as well as consolidated semidemocracies – cases where no significant political change has occurred for some time – are more war-prone than consolidated democracies and autocracies (Hegre et al. 2001).

Dilemmas of war-to-democracy transitions

31

state. Even in instances where the elites have decided to disarm, people may remain polarized. This means that war-shattered societies are particularly vulnerable to the risks of democratization.

The legacy of violent conflict After a violent intrastate conflict, conditions conducive to democratization are typically absent and the legacies of conflict tend to linger.10 Arms are widely available and often used, even after a peace deal has been signed. Owing to psychological trauma of violence and fear of renewed violence, mass mobilization along extremist lines remains. Political ideologies are based on exclusive group-based interest rather than on universal, society-wide interests. Threats and violence prevent political candidates from running for office and hinder voters from going to the polling stations. Political trust is low, which hinders cooperation across subcultures. Whereas many civil society organizations play a pivotal role in humanitarian assistance and reconstruction after war, there are examples where voluntary organizations foment intergroup violence. This was the case during the civil war in Lebanon 1975–1989. In the worst case, such activities can even contribute to genocidal violence. In Rwanda, radio broadcasting was used to instigate the genocide of Tutsis in 1994. During and after ethnic wars, people tend to seek protection in areas where the majority population belong to their own ethnic group and thus become displaced within their own country (Posen 1993). In such contexts, proponents of a moderate ideology face a high risk of becoming targets both of extremist violence by people belonging to the same ethnic kin, and members of other ethnic groups who portray all nonmembers as enemies. To escape violence, moderates often have to leave for other states and become refugees. Competition for votes based on increasingly extremist rhetoric, so-called politics of outbidding, can also enhance polarization and foster ethnic tension (Rabushka and Shepsle 1972: 187). Such outbidding has taken place in, for example, post-war Bosnia and Herzegovina, where nationalist parties have cemented their early grip of power in successive

10

For a discussion on conditions conducive to democracy and democratization, see, for example, Robert Dahl (1971) and Samuel Huntington (1991: 37–38).

32

Anna K. Jarstad

elections. Additionally, in many post-war societies the political party organization is absent, weak, or fragmented. The 2005 Afghan parliamentary elections demonstrate this point. The over 5,800 candidates that ran for office were formally part of different political parties, but personalistic attributes and clan politics largely substituted the role of ideologies.11 In addition to the initially contested issues, such as control over government or control of territory, new issues emerge during the conflict. At the same time, there are often fewer resources to share or divide after a conflict. Typically, the economy is weak, the level of unemployment high, the infrastructure shattered, and natural resources destroyed or inaccessible, for example due to land mines. Refugees and internally displaced persons are often prevented from returning to their pre-war homes because of new occupants or destroyed houses, insecurity, and lack of economic resources. Thus, an agreement seldom means that a conflict is resolved. Conflicting attitudes, behavior, and issues remain to be transformed after the fighting has stopped (Lederach 1997).

Simultaneous peacebuilding and democratization The cases in focus are war-torn societies that undergo simultaneous peacebuilding and democratization. For our purposes, it is important to include both cases where a war-to-democracy transition has taken place and cases where such transition is impeded by difficult dilemmas. To recall, peacebuilding is here seen as the implementation of a peace agreement. This means that the pertinent cases for our analyses are post-settlement cases.12 Democratization refers here to improvements in contestation, participation, and human rights. This minimal definition allows us to analyze also cases where only minor moves toward democracy have occurred. To give a snapshot picture of pertinent cases, the list below (see Table 1.1) indicates forty-two conflict locations where peace

11 12

See Reilly’s chapter in this volume. In addition, some of the chapters make reference also to cases where no peace agreement has been reached, in order to illustrate and compare processes that increase our understanding of democratization and peacebuilding in war-torn societies.

Dilemmas of war-to-democracy transitions

33

Table 1.1 Recent cases of peacebuilding after civil war Conflict location

Name and year of latest peace agreement or process

Afghanistan Angola Bangladesh (Chittagong Hill) Bosnia and Herzegovina Burundi Cambodia Chad

Post-Bonn process (2001) Lusaka process (1994–2002) Chittagong Hill Tracts Peace Accord (1997)

Colombia Comoros (Anjouan) Congo Croatia DRC Djibouti El Salvador Georgia (Abkhazia) Guatemala Guinea-Bissau Haiti India (Bodoland) India (Tripura) Indonesia (Aceh) Israel (Palestine) Ivory Coast Liberia Macedonia Mali (Azawad) Mexico Moldova (Dniestr) Mozambique Nepal Niger (Air and Azawad)

Dayton Peace Agreement (1995) Arusha process (2000–2003) Paris Agreement (1991) Various peace and reconciliation processes (1989–2006) EPL-government peace process (1991) and Common Agenda process (1999–2002) Federalization process (2000–2003) National Dialogue process (1999–2001) Erdut Agreement (1995) Inter-Congolese Dialogue process (1999–2003) Accord de reforme et concorde civile 2001 Geneva process (1990–1992) Declaration on measures for a political settlement (1994) Esquipulas/Oslo/UN processes (1987–1996) Abuja Peace Agreement (1998) MINUSTAH peacebuilding mission (from 2004) Bodoland Autonomous Council Act (1993) Memorandum of Settlement (1993) Memorandum of understanding, Indonesia and GAM (2005) Oslo process (1993–2000) Accra process (2003–2006) Accra Comprehensive Peace Agreement (2003) Ohrid Agreement (2001) Pacte National (1992) San Andre´s Accords (1996) Memorandum on the Basis for Normalization of Relations (1997) Rome process (1990–1992) Comprehensive Peace Agreement (2006) Agreement on a Definite Peace (1995)

34

Anna K. Jarstad

Table 1.1 (cont.) Conflict location

Name and year of latest peace agreement or process

Papua New Guinea (Bougainville) Philippines (Mindanao) Philippines Rwanda Senegal (Casamance) Sierra Leone Somalia Sudan (Southern Sudan) Sudan (Darfur) Tajikistan Uganda UK (Northern Ireland) Yugoslavia (Kosovo) Yugoslavia (Slovenia)

Bougainville Peace Agreement (2001) Mindanao Final Agreement (1996) General Agreement for Peace (1995) Arusha process (1991–1993) General Accord between Senegal and MFDC (2004) Abuja Ceasefire Agreement (2000) Cairo Declaration on Somalia (1997) Machakos process (2002–2005) Darfur Peace Agreement (2006) Moscow Declaration (1997) Yumbe Peace Agreement (2002) Good Friday Agreement (1998) Rambouillet Agreement (1999) Brioni Agreement (1991)

Source: Uppsala Conflict Data Program, March 2007. Selection made with assistance of Ralph Sundberg. The name of the contested territory is indicated in parentheses after each conflict location. Conflicts without parentheses concern contest over government.

agreements have been signed during the post-Cold War era.13 The conflict location for civil armed conflicts, and the latest peace accord for each conflict, is listed above. In most of these cases, there have also been improvements in contestation and human rights after the conflict

13

The processes of democratization and peacebuilding are ‘‘moving targets’’ and there is no existing database that captures these phenomena. The list included here is based on the Uppsala Conflict Data Program (UCDP). This database has the benefits of including all armed conflicts with at least twenty-five battlerelated deaths and is a systematic and continuous collection endeavor. However, it does not capture all peace processes. The cases in the list are coded by UCDP as conflict locations with peace agreements. In addition, they fulfill the following criteria: they are intrastate peace agreements (interstate peace agreements are excluded) and the peace agreement was signed during the period 1989–2006. Cases not included in the list, according to the criteria above, include, for example, Iraq 2003 (not coded by UCDP as a peace agreement) and South Africa 1994 (not coded as an armed conflict after 1989), East Timor/Indonesia, Ethiopia, Lebanon, Nicaragua, Namibia, and Cyprus. Although these cases do

Dilemmas of war-to-democracy transitions

35

was formally ended. Where this is the case, a war-to-democracy transition can be considered under way. A brief look at the list gives us an idea of the many war-torn societies where challenges face the people that live in these locations, and the international community organizations which aim to support the processes of democratization and peacebuilding. There is no easy way to achieve peace and democracy simultaneously after civil war. Not only are the conditions for the initiation of a democratization process unfavorable after intrastate conflict, in addition, the opening up of political space aggravates these conditions. The two processes of democratization and peacebuilding may clash, thus leading to negative effects on each other. Such potential quandaries reflect the inherent conflictual nature of democracy and democratization as well as the difficult process toward peace. Appreciation of the particular dilemmas that arise in each post-war situation is necessary for the design of proper means to advance synchronized democratization and peacebuilding. Ideally, such analysis makes it possible to avoid several of the dilemmas and to properly design means to support peace and democracy simultaneously. However, at particular points in time the inevitable choice arises between promoting efforts to democracy or peace. Without a minimum level of peace it is impossible to achieve free political contestation, popular participation, and human rights. This is why this book suggests that the sequencing, timing, and design of peace missions are vital for international support for war-to-democracy transitions. In the long run, the central issue is not choosing between peace or democracy, but rather what steps toward peace and democracy should be taken when, and how are they best timed, sequenced, and combined? This book suggests that a minimal level of security is important before elections take place. Violence needs to be reduced to permit elections, if legitimate government is to result. In this way, securing a minimal level of peace is a first necessary step for successful war-to-democracy transitions. At the same time it is important not to postpone elections for too long, as elections performs various functions after war. The book also

not fulfill the criteria above, some of the chapters in this book make reference to these and other cases, as they prove valuable for understanding the broader issue of war-to-democracy transitions. For UCDP data and definitions, refer to www.pcr.uu.se/research/UCDP/index.htm.

36

Anna K. Jarstad

outlines ways to promote democratization short of elections, by opening up space for actors outside the formal peace process. Whereas the process of democratization is very vulnerable to a breach of peace, peacebuilding often progresses despite setbacks in democratization. To recall, efforts to promote security is only one element of peacebuilding. Successful war-to-democracy transitions involve political, institutional, and social transformation to bring conditions that enable lasting peace and democracy.

PART II

The security context

2

Peacekeeping and democratization VIRGINIA PAGE FORTNA

The international community has endeavored to maintain peace and foster stable, democratic polities in a host of war-torn countries since the end of the Cold War. A primary tool in this effort has been the practice of international peacekeeping – the deployment of international personnel, generally under the auspices of the United Nations or regional organizations, to oversee the transition from war to peace. Maintaining peace in the aftermath of civil war is difficult enough, but the added task of shepherding a country toward democracy is fraught with challenges and dilemmas. States recently torn asunder by warfare do not provide particularly fertile ground for the growth of democracy. Nor is peacekeeping, on the face of it, an especially powerful tool. Peacekeeping missions are habitually under-funded, under-equipped, and understaffed. They are cobbled together, often at the last minute, with begged-for personnel (among whom there may be no common language). They are often given unrealistically short mandates and overly ambitious time-lines and benchmarks. And they are asked to perform miracles, turning countries in which political institutions, economic infrastructure, and the very fabric of society have all been devastated by civil war into stable, functioning, democratic states. This chapter assesses how well they do at accomplishing this virtually impossible task. More specifically, what role does international peacekeeping play in the democratization of post-war societies? It shows that peacekeeping has neither a clear positive nor a clear negative effect on democratization. Rather, positive and negative effects appear to cancel each other out, reflecting inherent dilemmas in the attempt to foster both stable peace and democracy in the aftermath of civil war. Peacekeeping enhances stability, that is, it helps to prevent the recurrence of war. Peacekeepers might usefully be thought of as a modernday equivalent of the podesta´, hired by warring clans in medieval Genoa to administer the city and keep the peace, too weak to take 39

40

Virginia Page Fortna

power themselves, but strong enough to tip the balance against either side should it attack the other (Greif 1998). But as this collaborative project on the ‘‘dilemmas of democratization’’ emphasizes, not all good things necessarily go together. That which promotes stability may not promote, and may in fact undermine, democracy. Second only to stable peace, democratization is a core goal of the international community when it undertakes peacekeeping missions.1 Since the mission in 1989 in Namibia, which marked a turning point in peacekeeping with the end of the Cold War, most peacekeeping missions involve significant democratization components, including monitoring or even running elections, civil rights monitoring and training, fostering civil society and democratic political institutions. But the two main goals of peacekeeping, maintaining peace and stability on the one hand, and fostering democracy on the other, may often be in direct conflict. The effectiveness of peacekeeping on democratization is open to debate. Doyle and Sambanis (2006), Heldt (2007), and Pickering and Peceny (2006) find that intervention fosters democratization. Similarly, Wantchekon (2004) argues that impartial peacekeepers provide one of the conditions for democracy to emerge from civil war. However, Bueno de Mesquita and Downs (2006) find no positive effect of intervention. Weinstein (2005) argues that outsiders’ attempts at state- and democracy building can impede the development of strong and democratic political and economic institutions, that in some cases at least, post-war societies would be better off left to their own devices in a process of ‘‘autonomous recovery.’’2 Similarly, Marten (2004: 155) proposes that peacekeepers should limit their goals to providing stability, and not try to transform societies. ‘‘The notion of imposing liberal democracy abroad is a pipedream,’’ she argues. The widespread use of peacekeeping in civil wars since the end of the Cold War has made it possible to end the fighting with a truce or 1

2

Paris (2004) argues that all of the fourteen major peacebuilding missions between 1989 and 1999 have included democratization, along with marketization, as a strategy for consolidating peace. See also Andersson (2000). Wantchekon and Neeman (2002) propose a model by which democracy can emerge in war-torn states without the intervention of outsiders, though they acknowledge that some peacekeeping functions, such as monitoring and trust building, might encourage democratization – a notion that Wantchekon (2004) develops further.

Peacekeeping and democratization

41

settlement in a number of cases that, were peacekeepers not available, would likely have dragged on for some time until a clear victor eventually emerged. Peacekeeping has thus curtailed the violence but has also led to less decisive outcomes (Fortna 2005). These less decisive outcomes may also make democracy less likely to emerge.3 The question of whether peacekeeping furthers democratization in war-torn states or hinders it is thus an open one. This chapter aims to explore, both theoretically and empirically, the relationship between international peacekeeping and domestic processes of democratization in post-war societies. I define peacekeeping as the deployment of international personnel to help maintain peace and security in the aftermath of war.4 All peacekeeping missions involve military personnel, though they may or may not be armed, and many missions include substantial civilian components as well. This definition includes both operations based on the traditional principles of peacekeeping, specifically the consent of the belligerents themselves and the passive use of force, as well as peace enforcement missions that relax these conditions considerably. However, I also distinguish among types of missions, for the effects of enforcement missions may differ from those of consent-based peacekeeping. And perhaps most important for a study of peacekeeping and democratization, the effects of multidimensional missions that include large civilian components engaged in election monitoring, human rights training and monitoring, police reform, institution building, economic development, and so on may be quite different from the effects of more traditional types of observer or interpositional peacekeeping missions. Peacekeeping is thus broken into the following four categories:

3

4

Toft (2003) argues that civil wars that end in a victory for one side, especially those that end in rebel victory, are followed by higher levels of democracy over the long term than are wars that end in a truce or peace settlement. She suggests that this is because clear military winners are better able to consolidate political institutions. Her empirical analysis on this includes only a bivariate relationship, and may be picking up regression toward the mean in democracy scores after a civil war. (See below for further empirical investigation of the relationship between war outcome and democratization.) For a fuller discussion of the definition and types of peacekeeping, see Fortna (2008).

42

Virginia Page Fortna

 Observation missions are small unarmed deployments of military and sometimes civilian observers to monitor a cease-fire, the withdrawal or cantonment of troops, or other terms of an agreement, such as elections. Their main tasks are simply to watch and report on what they see. Examples include the peacekeeping missions in Angola in 1991 (UNAVEM II) and in the Western Sahara (MINURSO), and the New Zealand and then Australian-led missions in Papua New Guinea in 1997–1998 (the Truce Monitoring Group and Peace Monitoring Group, respectively).  Interpositional missions (also known as traditional peacekeeping missions) are deployments of lightly armed troops. Like observer missions, they monitor and report on compliance with an agreement, but they also often serve to separate forces or to help demobilize and disarm military factions. The UN missions in Angola in 1994 (UNAVEM III) and in Guatemala in 1996 (MINUGUA) are examples.  Multidimensional missions include both military and civilian components helping to implement a comprehensive peace settlement. In addition to the roles played by observer or interpositional missions, they perform tasks such as organizing elections,5 human rights training and monitoring, police reform, institution building, economic development, and so on. Examples include the missions in El Salvador (ONUSAL), Mozambique (ONUMOZ), and Namibia (UNTAG).  Peace enforcement missions are mandated to use force for purposes other than self-defense and involve substantial military contingents to provide security and ensure compliance with a cease-fire. Some enforcement missions are also multidimensional, consisting of large civilian components as well as relatively robust military forces. The West African and United Nations missions in Sierra Leone in 1999 (ECOMOG and UNAMSIL) and NATO missions in Bosnia (IFOR and SFOR) fall in this category. The first three categories together encompass what are often referred to as Chapter VI peacekeeping, or consent-based missions, while the fourth category is sometimes referred to as Chapter VII peacekeeping. Chapter VI missions depend on the consent of the belligerents themselves and are mandated to use force only in self-defense, while 5

Note that I distinguish between organizing or running elections and election observation, which is a task often mandated to observational or interpositional missions.

Peacekeeping and democratization

43

Chapter VII missions do not necessarily require consent and, as noted above, rely more centrally on the use (or potential to use) force. This breakdown of mission types does not distinguish missions that temporarily take over the administration of the country, as in Cambodia, Kosovo, or East Timor, from other types of peacekeeping. These transitional administration missions may have rather different effects than missions that oversee national administration of the state only during the transition to peace. There are too few such missions (only these three) in the data examined here to evaluate their separate effects (they are included in the quantitative analysis as multidimensional enforcement missions, as appropriate).6 I do, however, include a brief discussion of democratization in these three cases below. This study includes both peacekeeping undertaken by the United Nations (the majority of peacekeeping missions) and by regional organizations or ad hoc groups of states, although I draw distinctions between UN and non-UN missions in some of the analyses below.7 Of course, to evaluate the effects of peacekeeping on democratization, we must compare post-war societies that received peacekeeping with those that did not, where belligerents were left to their own devices after the war. And because peacekeepers are not sent to 6

7

Another case of transitional administration, though not of an entire country, is the UN’s mission in Eastern Slavonia (UNTAES), a region of Croatia. The peacekeeping missions in Bosnia, by various parties including the UN, NATO, the EU, and the OSCE, had some aspects of a transitional administration but are not generally included in that category. For a qualitative assessment of the three cases mentioned above, as well as others that took over state administration to a lesser degree, see Chesterman (2004). Transitional administrations may be becoming more common over time, such that a quantitative assessment of their effects will become possible. In some cases, UN and non-UN missions work simultaneously or sequentially. For example, there have been both NATO and UN peacekeepers in Bosnia, US-led and UN peacekeepers in Haiti, and Economic Community of West African States (ECOWAS) and UN peacekeepers in Sierra Leone and Liberia. There are not enough of these ‘‘hybrid’’ cases to evaluate their effects separately, though they may be becoming more frequent. I also do not distinguish ‘‘Integrated Missions,’’ a term that has recently been used to refer to the need for integration among various bodies within the UN that provide peacekeeping, peacebuilding, humanitarian assistance, and development aid, as well as integration between the UN and other peacekeeping organizations. As a recent report on the topic puts it, ‘‘There is no adequate definition of an integrated mission. Nor is there an example of an integrated mission that serves as a model for what an integrated mission should be’’ (Eide et al. 2005, 9). I therefore code missions according to the more accepted typology described here.

44

Virginia Page Fortna

conflicts at random, we also need to take into account other factors that affect both where peacekeepers go and democratization, including a country’s previous experience (if any) of democracy, economic conditions, and so on.8

Two related dilemmas Two fundamental dilemmas of security and democracy face would-be peacekeepers and the societies they aim to help. The first is that outside assistance may be needed to achieve stable peace, but this external assistance may thwart domestic processes of democratization. The second emerges from the paradoxical relationship between war and democracy – democracy requires peace and stability, but war and the pressures of war-making on the state can provide incentives for democratization.

Internal vs. external control and legitimacy As Jarstad notes in her introduction to this book, the emergence of democracy requires low levels of violence, and low levels of mistrust, conditions that are unlikely in the immediate aftermath of civil war. International peacekeepers can help reduce violence – both the threat of a full-scale return to war, and lower levels of day-to-day violence. International monitors can also increase levels of trust between former belligerents and competing political forces. In these ways, peacekeepers can help foster the conditions for successful democratization. Peacekeepers can also foster the initial establishment of democratic institutions by helping to set up and monitor elections. On the other hand, by taking a large measure of control over political processes in the transition from war to peace, outside peacekeepers can undermine the principles of accountability and domestic legitimacy that are bedrocks of democratization. In the short term, democracy has no chance of taking root if a society is plagued by high levels of political violence, or worse yet, if fullscale war resumes. Democracy requires peace and stability, and wartorn societies often require outside help in maintaining peace. War is 8

For systematic studies of where peacekeepers tend to deploy, see Gilligan and Stedman (2003); and Fortna (2008, Chapters 2 and 3).

Peacekeeping and democratization

45

much more likely to resume, all else equal, if belligerents are left to their own devices (Doyle and Sambanis 2000; Fortna 2004, 2008). However, the very thing that can help ensure lasting peace, outside intervention, often reduces the political space available for the emergence of home-grown, domestically legitimate and accountable political institutions. As Chesterman (2004: 1) argues, there are fundamental tensions and contradictions involved in attempting ‘‘to establish the conditions for legitimate and sustainable national governance through a period of benevolent foreign autocracy.’’ Peacekeeping may thus help foster conditions for the initial emergence of democracy, but at the same time undermine conditions for it to thrive over the longer term. The arrival of foreign troops as peacekeepers by definition undermines state sovereignty. It also introduces a temporary and potentially powerful actor into the domestic political scene, one whose presence affects the incentives and behavior of indigenous political, economic, and military actors. Particularly with the advent of large multidimensional peacekeeping missions, these outsiders play a large role in the transitional administration of the state (at the extreme, taking over state administration entirely, as in Kosovo or East Timor). Ideally, when the peacekeepers depart, they leave behind them at least nascent democratic institutions. Peacekeeping missions can help hold the government to promises of political reform and can assist rebels in the transition from military organizations to political ones. They can provide security during the tense periods of election campaigning and balloting. And they can help reform armies and police forces and provide human rights training, protecting political rights that are fundamental in a functioning democracy. In other words, they can help start a war-torn society on the road to democracy. But the large footprint of peacekeeping missions may trample and crowd out local democratization efforts as much as empower them. Peacekeeping creates a precedent of control by actors who are not domestically accountable, or perhaps accountable at all.9 They can create an artificial political structure that collapses when they leave, and they can create a situation where the state’s reliance on

9

Problems of peacekeepers’ accountability (or lack thereof) have been most noticeable in the area of prostitution and sexual abuse (Mendelson 2005).

46

Virginia Page Fortna

international assistance interferes with its accountability to local populations for support and funds (Weinstein 2005: 27). In short, while peacekeepers can help provide security and incentives for initial steps toward democracy, the presence of foreign actors can undermine the long-run development of domestically legitimate democratic institutions. In the terminology used elsewhere in this volume, this is a systemic dilemma which can cause a temporal dilemma.

Democracy through war or peace? A more fundamental dilemma concerns the relationship between democracy on the one hand, and war or peace, on the other. For democracy to emerge there must be some semblance of peace and security. Political violence and the atrocities of civil war are fundamentally antithetical to the norms of cooperation, non-violent resolution of political conflict, and basic trust inherent in a functioning democratic system. There is good reason for the conventional wisdom that democracy cannot be established without peace. On the other hand, there is reason to suspect that war itself can create the conditions for democratization. It was the need to raise funds and manpower for war making that led to political reform, and therefore democratization, in the process of European state making (Tilly 1985, 1990). States and subjects traded political rights for the taxes and soldiers required for war. Weinstein argues that this same logic applies to contemporary civil wars – that ‘‘war itself generates strong incentives for rulers to secure the consent of governed and build representative institutions.’’ This creates a policy dilemma between stopping mass killing in the short term and allowing ‘‘the processes of internal, institutional change that warfare reflects’’ (Weinstein 2005: 4, 12). While Weinstein focuses more on effective state institutions than on democratic ones per se, his argument nonetheless suggests that ‘‘autonomous recovery’’ may be better for democratization than ‘‘aided recovery’’ through peacekeeping and external financial aid.10

10

Weinstein explores this argument in three cases: (1) Uganda, which consolidated state institutions and developed a ‘‘no-party democracy’’ after Museveni’s rebel victory in 1986; (2) Eritrea, where ‘‘a democratic decentralization of political power, a participatory constitution-making exercise, and the building of national political institutions reflective of a new national identity were all

Peacekeeping and democratization

47

Wantchekon (Wantchekon and Neeman 2002; Wantchekon 2004) argues that democracy can emerge directly from civil war when warlords have economic incentives to create peace (so as to extract resources more efficiently from society).11 They will prefer empowering the citizenry (i.e., democratizing) to inviting in an external enforcer because the latter cannot credibly commit to neutrality. Wantchekon draws a distinction between a neutral (and relatively weak) arbitrator such as a UN peacekeeping mission, and a powerful external enforcer. The former, he argues, can assist in the transition to democracy, while the latter is a substitute for democracy. But this suggests that the moves away from strict neutrality and toward more militarily robust peacekeeping, undertaken in the late 1990s, may be detrimental to democratization. If Wantchekon’s reasoning is correct, there may thus be something of a dilemma between militarily effective peacekeepers, better able to maintain stability, and those of the relatively weak and neutral variety that can assist democracy rather than impede it in his model. While Wantchekon contends that limited peacekeeping is good for democracy, the implication of his argument points to a similar dilemma as that suggested by Weinstein. It is the cost of war (in this case the economic inefficiencies of war) that makes militarily powerful actors (warlords) democratize. If peacekeeping is too successful in its primary task of maintaining peace, these actors may not need to democratize to reap the benefits of peace. Peacekeeping and the stable peace it helps produce can thus become a substitute for democracy rather than an aid to it.

Empirical implications These dilemmas suggest that the empirical relationship between peacekeeping and democratization will not be a straightforward positive

11

remnants (and products) of the EPLF’s [Eritrean People’s Liberation Front] warmaking effort’’ (p. 18); and (3) Somalia, where Weinstein argues effective substate political structures were rebuilt ‘‘after the UN pulled out’’ (p. 14). While none of these countries is a paragon of successful democratization, Weinstein would argue that they provide examples of the development of effective and relatively representative state institutions. He does not, however, compare these cases to democratization processes in cases where the international community has intervened. This will not be the case when warlords can finance their operations and line their pockets with proceeds from drugs or easily extractable mineral wealth such as diamonds.

48

Virginia Page Fortna

one. On the one hand, if stable peace is necessary for democracy, then since peacekeepers improve the chances for lasting peace, the presence of peacekeepers should lead to more democratic outcomes, all else being equal. Furthermore, peacekeeping can minimize day-to-day violence, build trust between former enemies, oversee initial elections, monitor human rights, and generally help put war-torn societies on the road toward democracy. On the other hand, the second dilemma explored here implies that we might see mixed or even negative effects of peacekeeping. Meanwhile, the first dilemma suggests that we need to distinguish between the short-term and long-term effects of peacekeeping on processes of democratization. Specifically, it suggests that the presence of peacekeepers should enhance democratization in the short term, but undermine it in the long term. The remainder of this chapter consists of a quantitative empirical exploration of these hypotheses. Of the civil wars ending since the end of the Cold War, how have the democratization trajectories of those with peacekeepers compared to those without? Has peacekeeping led to higher or lower levels of democracy in the short term and in the long term? Have different types of peacekeeping had different effects on democracy?

The data In order to examine the effects of peacekeeping on democratization, we need to examine both cases in which peacekeepers were deployed and those where belligerents were left to their own devices in the aftermath of fighting. I use a data set created for a project evaluating the effects of peacekeeping on the stability of peace (Fortna 2008). The data consist of ninety-five cease-fires or breaks in the fighting from 1989 through 1999 in almost sixty civil wars. The data build on those compiled by Doyle and Sambanis (2000; 2006), but I have added a number of short-lived cease-fires not included in their data or in other data on civil wars. Inclusion of these ultimately unsuccessful attempts to maintain peace is particularly important for a study of post-war stability and democratization as their omission would truncate variation in the dependent variables and introduce selection bias. The cases are listed in the appendix to this chapter. I restrict the analysis to wars ending between 1989 and 1999 for several reasons. I examine only the post-Cold War era because

Peacekeeping and democratization

49

peacekeeping was only very rarely used in civil wars during the Cold War. Moreover, before 1989, its purpose was to contain conflicts from drawing in the superpowers more than to maintain peace within the conflict itself. More critical for this study, it was rarely used in an attempt to bring democracy to war-torn societies until after 1989. I include cases only through the end of 1999 for more practical reasons. Data for some important control variables are unavailable after that time. More important, this allows me to observe both whether peace lasts and trajectories of democratization for at least five years after the point of a cease-fire for all of my cases. So while to be included in this study, a break in the fighting must occur before the end of 1999, observation of the main dependent variable – post-war democratization – continues through 2004. Ideally, to study long-term effects on democratization we would have more than five years of information since the most recent cases. But without a crystal ball, that is not possible.12 Data on democratization are taken from two sources: the Polity IV data set (Marshall and Jaggers 2002) and Freedom House (2005). The Polity data rate countries on an annual basis on a 20-point scale from full autocracy (10) at one end to full democracy ( þ10) at the other, while Freedom House ranks countries, on two 7-point scales, in terms of their civil liberties and political rights.13 Both of these measures allow us to discern fairly fine-grain moves toward (or away from) democracy and ‘‘freedom’’ in post-war societies. While there is significant overlap in these two measures (as one would hope), they emphasize different features of democracy. The Polity data focus relatively more on the institutions and procedures of democracy, including elections, institutional constraints on the executive, and forms of competitiveness in political participation. The Freedom House data rate countries based on a checklist of political rights (including those concerning the electoral process, political pluralism and participation, government functioning and corruption), and civil liberties (including 12

13

Similarly, while it will be important to evaluate the effects of current peacekeeping cases, such as those in the Ivory Coast, the Sudan, and Liberia, it is too early to judge their effects on democratization. Here I use an aggregate of these two Freedom House scores, rather than that organization’s designations of ‘‘free,’’ ‘‘partly free,’’ and ‘‘not free’’ since collapsing aggregate scores into only three categories entails a loss of information.

50

Virginia Page Fortna

freedom of expression and belief, freedom of association, rule of law, and personal autonomy and individual rights).14 Using both the Polity and Freedom House data allows me to check the robustness of findings about peacekeeping and democratization across these two measures. For each data source, I measure democratization one, two, and five years after the fighting stops. Because I am interested in movement along the continuum from autocracy to democracy, I focus on change in democracy scores rather than their absolute level. Democratization one, two, and five years out is therefore measured as the difference between a country’s Polity or Freedom House score in the year before the fighting stops and its score one, two, and five years after the war ends.15 I invert the Freedom House scores so that higher numbers represent greater moves toward ‘‘freedom’’ so as to match the direction of the Polity measure of democratization. Studying democratization in the immediate aftermath of war presents some complications because war often resumes. Because I include even very short-lived cease-fires in these data, there are cases in which war breaks out anew before our measures of democratization are taken. To be more specific, war resumes in less than one year in thirty-one cases (about a third of the total), within two years in another eight, and within five years in a further thirteen (that is, peace lasts fewer than five years in over half of the cases). It is not obvious how these cases should be treated. On the one hand, we might drop these cases from the analysis, focusing rather on democratization in those 14

15

The Freedom House data have come under more criticism than the Polity data for being somewhat arbitrary, and for entailing a political bias (for example, in favor of US allies and of free market economic policies). To the extent that the international community has shared this bias, as Paris (2004) argues it has, this may provide a closer measure of what it is trying to achieve in post-war transitions. I use the year before the fighting stops as the baseline, rather than the year the fighting stops because the annual Polity and Freedom House scores may reflect moves along the democracy continuum that occur between the date of a cease-fire and the end of the calendar year. Lagging the baseline by one year ensures that I am not capturing post-war levels in this baseline. I could also use democracy scores from before the war as a baseline, but since in some cases war has been raging for many years and democracy levels often fluctuate over these spans, I would be capturing changes in democracy that occur during the war as part of my measure of post-war democratization. Furthermore, I include democracy at the start of the war as a control variable to capture effects of prior history with democracy.

Peacekeeping and democratization

51

cases where peace is holding.16 On the other hand, these cases are substantively interesting for a project on the relationship between post-war democratization and post-war stability. In the empirical analyses that follow I take both tacks, presenting findings both for the full set of cases, and for the set restricted to cases in which peace is holding one, two, and five years out. This allows me, to an extent, to distinguish the direct effects that peacekeeping has on democratization from the indirect effects that it has because it makes peace more stable. Many post-war societies do experience democratization. In the full set of cases examined here (that is, including those in which war has since resumed), the average democratization score one year after the fighting stops is a hair over one point along the twenty-point Polity scale. While most cases experience no change in their level of democracy this soon after the war, and a few experience moves toward autocracy, about a third see positive developments in terms of democratization. Many of these are modest, one or two point increases in Polity scores (e.g., Namibia and Djibouti), but others were more dramatic, including Mali’s fourteen-point jump, despite another round of war. Not surprisingly, the democratization story is rosier if we restrict the analysis to only those sixty-four cases in which peace held for the first year. The one year democratization average for these cases is 1.35 using the Polity measure. The trends are similar in the Freedom House data. Two years out, we see even greater strides toward democracy. This is not surprising as many democratic institutions need time to begin to take root.17 Examining all cases, including those where peace falters before two years are up, the average level of democratization is 1.42 points on the Polity scale with close to 40 percent of post-war countries making at least some moves toward greater democracy. As before, those cases where peace has held fare even better. Their average democratization after two years is 1.67 Polity points, with fewer backsliders 16

17

In some cases, war resumes but ends relatively quickly so that our measure of democratization is taken when the country is back at peace. Because this second cease-fire is also included in the data, these cases will enter the analyses again, with democratization scores taken one, two, and five years after the second break in the fighting. In a small number of cases (3), the second cease-fire occurs in the same year as the first so that democratization scores are duplicated. These cases are Guinea–Bissau 1998, India–Assam 1991, and Russia–Chechnya 1996. Also, because states at war tend to have low levels of democracy, a statistical regression toward the mean would suggest some positive moves toward democracy as we get farther in time from the fighting.

52

Virginia Page Fortna

and larger gains in democracy. This improvement over time is as visible in the Freedom House scores as it is in the Polity scores. Five years out, the picture is remarkably close to that only two years out – if we look at all cases. The average democratization over five years is 1.46 points in Polity, only 0.04 higher than after two years. But the discrepancy between cases in which war resumes and those where it lasts is much larger five years out than in shorter-run examinations. Of cases where peace has held, the five year democratization average is 2.78, dramatically higher than after just one or two years. In only two cases in which peace has lasted at least five years are democracy scores lower than they were during the war: Azerbaijan and Mali, which had managed to democratize significantly in the early 1990s despite ongoing fighting (see above), but lost ground, only slightly, afterward. Meanwhile almost half of the cases moved away from autocracy and toward democracy, including quite dramatic gains in Ethiopia, Croatia, and Mozambique. The Freedom House scores show even more improvement over five years. By this time, almost half the cases had moved toward ‘‘freedom’’ relative to the year before the war ended (including very large gains in places like Mali and South Africa), while the rest are about evenly divided between no change and some movement away from ‘‘freedom.’’ Of the restricted set of cases that saw no new war, 65 percent made gains in political rights and civil liberties, with only a handful moving away from ‘‘freedom.’’ Clearly then, democratization often happens in the aftermath of civil war. But does peacekeeping by the international community foster this democratization, or is it just as likely to hinder it? Stable peace is better for the growth of democracy than renewed warfare; that is apparent in the comparison of the full set of cases with those in which peace lasts. And we know from previous research that peacekeeping helps sustain stable peace. The risk of war recurring is much lower when peacekeepers are present than when belligerents are left to their own devices (Fortna 2004, 2008). Table 2.1 shows the results of hazard analysis of the duration of peace after civil war. The hazard ratios in this table indicate the effect of variables, including peacekeeping and a series of control variables, on the relative ‘‘hazard’’ of war resumption.18 Hazard 18

Results are from Weibull regressions. Results are even stronger if a Cox proportional hazards model is used. Robust standard errors are calculated

Peacekeeping and democratization

53

Table 2.1 Effects of peacekeeping on the duration of peace Time-varying measure

All peacekeeping Victory Treaty Identity war Cost of war Factions Democracy Infant mortality Failed past agreement Government army size Contraband financing Mountainous terrain Perm-5 contiguous Perm-5 former colony Duration of war Shape parameter P No. of subjects No. of observations Log pseudolikelihood

Time-constant measure

Hazard ratio (RSE)

P > |Z|

Hazard ratio (RSE)

P > |Z|

0.43 (0.15) 0.12 (0.07) 0.25 (0.11) 1.34 (0.51) 1.19 (0.10) 0.56 (0.17) 0.97 (0.03) 1.00 (0.00) 1.19 (0.49) 1.00 (0.00) 2.42 (0.83)

0.02 0.00 0.00 0.45 0.04 0.05 0.35 0.52 0.68 0.53 0.01

0.19 (0.09) 0.08 (0.05) 0.30 (0.13) 1.05 (0.35) 1.20 (0.09) 0.70 (0.21) 0.99 (0.03) 1.00 (0.00) 1.21 (0.51) 1.00 (0.00) 2.83 (1.04)

0.00 0.00 0.01 0.89 0.02 0.24 0.61 0.45 0.65 0.63 0.01

1.05 (0.17) 0.58 (0.33) 1.27 (0.51) 0.92 (0.03) 0.61 (0.05)

0.76 0.34 0.60 0.01 0.00

1.21 (0.20) 0.35 (0.22) 0.94 (0.37) 0.92 (0.03) 0.66 (0.06)

0.26 0.09 0.88 0.01 0.00

93 129 142.77

93 129 137.83

Weibull Regressions. Cases clustered by country for calculation of robust standard errors.

ratios are interpreted relative to 1.0, such that a hazard ratio of 2.0 would indicate a doubling of the risk of war, all else equal; a hazard ratio of 0.6 would indicate a 40 percent reduction in the risk of war. Effects are shown for two versions of the peacekeeping variable. Column 1 shows the effect of whether peacekeepers are currently present, while column 2 shows the effects of having had a with cases clustered so that the model assumes independence between cases in different countries but not necessarily between cases in the same country.

54

Virginia Page Fortna

peacekeeping mission, even if it has since departed.19 The former is a much more conservative estimate of peacekeeping effects (because peacekeepers are not given credit for peace that lasts after they depart), while the latter represents a more realistic assessment of the ‘‘success’’ of peacekeeping. But in either case, it is clear that peacekeeping has a large and significant effect on stability. The hazard ratio of 0.41 for the measure in column 1 indicates that peacekeeping reduces the risk of war by 59 percent, all else equal. The measure of peacekeeping used in column 2 indicates an 82 percent decline in the risk of another war when peacekeepers have deployed, relative to when belligerents are left to keep peace on their own.20 But the key question for this project is whether peacekeeping promotes democracy in post-war societies, or whether democracy is better served in those cases where belligerents manage without outside interference.

Peacekeeping and democratization: findings Table 2.2 shows the mean democratization levels for cases with no peacekeepers, cases where Chapter VI consent-based peacekeeping missions were deployed, and cases where more robust Chapter VII enforcement missions were deployed. The table also describes the results of tests of the statistical significance of these differences. On the face of it, democratization appears to be aided by the presence of peacekeepers. Cases with consent-based peacekeeping experience, on average, more post-war democratization than cases with no peacekeeping, and cases with enforcement peacekeeping missions experience higher democratization still.21 These differences are generally not 19

20

21

In the duration analysis jargon, these correspond to ‘‘time-varying covariates’’ and ‘‘time-constant covariates’’ measures, respectively. For explanation of the statistical method, and much more detailed analysis of the effects of peacekeeping on the duration of peace, see Fortna (2008, especially Chapter 5). Results shown here use the Polity measure of democratization. The pattern is much the same when the Freedom House measure is used, with two exceptions: both one and five years out, in the restricted set of cases where peace has lasted to date, democratization is slightly lower on average for Chapter VI missions than when no peacekeepers deploy. For this measure at least, the effect of consentbased peacekeeping on democracy is largely about maintaining stability rather than a direct effect on civil liberties and political rights.

Peacekeeping and democratization

55

Table 2.2 Mean democratization scores (Polity), by peacekeeping One year out All cases No peacekeeping Consent-based PK Enforcement PK N

0.36 1.14 4.55 92

If peace lasts 0.49 1.00 5.00 62

Two years out All cases 0.46 2.36 4.73 92

Five years out

If peace All lasts cases 0.68 1.79 4.89 54

0.65 2.62 4.00 84

If peace lasts 2.25 2.67 5.20 37

Statistical significance of difference in means (results were consistent whether Bonferroni, Scheffe, or Sidak methods were used): (1) In no case is the difference in means between no peacekeeping and consent-based peacekeeping statistically significant. (2) The difference between consent-based peacekeeping and enforcement is statistically significant only for democratization one year out if peace lasts (i.e., column 2). (3) The difference between no peacekeeping and enforcement is statistically significant one and two years out, among all cases or only if peace lasts (columns 1–4) but not five years out (columns 5–6). (4) The difference in means between no peacekeeping and any peacekeeping (combined means not shown) is significant one, two, and five years out for all cases, but is only marginally significant (0.05 < p < 0.10) for cases in which peace lasts one and two years out, and is not significant for cases in which peace lasts five years out.

statistically significant, however, so while the preponderance of evidence suggests that peacekeeping fosters democratization, we cannot reject the null hypothesis of no effect with confidence.22 The bivariate relationship between peacekeeping and democratization tells us only so much, however. Because peacekeeping is not deployed to civil wars at random, and because there may be other factors that affect both whether peacekeepers are present and the prospects for democratization, multivariate analysis is essential for evaluating this empirical relationship. In the analyses that follow I control for a number of variables identified as important in the general literature on democratization (e.g., Barro 1999; Burkhart and Lewis-Beck 1994; 22

Note that the lower levels of statistical significance for differences when we restrict analysis to those cases where peace has lasted, and when we look at longer-term effects, may be the result of smaller sample sets.

56

Virginia Page Fortna

Diamond, Linz, and Lipset 1988; Inglehart 1997; Lipset 1959; Londregan and Poole 1996; O’Donnell, Schmitter, and Whitehead 1986; Przeworski et al. 1996; Przeworski and Limongi 1997; Ross 2001). These include GDP per capita, illiteracy rates, oil exports, and the level of democracy at the beginning of the war. I also include characteristics of the country or the conflict that have been found to affect either the likelihood that peacekeepers will deploy, the prospects for peace, or both. These include the war’s outcome: whether the fighting ended in a clear victory for one side, whether it ended in a peace treaty (with the omitted comparison category here being wars that end with only a truce or a cease-fire); the cost of the war in terms of lives lost or people displaced from their home; and a number of dummy variables marking in turn: wars in which rebel forces funded their fight through contraband (drugs, diamonds, etc.); conflicts in or next to the territory of one of the five permanent members of the UN Security Council (Perm-5); former colonies of the Perm-5; and as a proxy for levels of mistrust, cease-fires that take place after a previous, failed, cease-fire agreement.23 I also checked the effects of several other variables, including whether the conflict was identity-based, whether it was a secessionist conflict, whether neighboring countries provided significant aid to the rebels, whether there were only two sides involved or many factions, the size of the government’s army, and the mountainousness of the terrain in the country.24 None of these affected levels of democratization, nor affected other results, so I do not include them in the results reported here. Nonetheless, some of these negative findings are substantively important and interesting. They suggest, for example, that contrary to what conventional wisdom might lead us to expect, countries rent by ethnic wars or other identity-based divisions fare no worse in terms of post-war democracy than do other war-torn countries. Similarly, countries afflicted by complex wars that pit many factions against each other have no worse democratization records than those emerging from simpler wars that involved only two sides.

23

24

For more information on these variables, see Fortna (2008). The data are available from the author (see www.columbia.edu/~vpf4/research.htm). Mountainous terrain has been shown to be a good predictor of civil war, presumably because it eases rebellion and hinders government efforts to fight insurgency (Fearon and Laitin 2003).

Peacekeeping and democratization

57

Because the measures of democratization are continuous, I use ordinary least squares (OLS) regression. However, because not all of the cases in the data are independent of one another, I cluster cases by conflict and calculate robust standard errors.25 Table 2.3 shows the effects of peacekeeping, as well as other potentially important variables on post-war democratization one year after the conflict. Democratization is measured using both the Polity (columns 1 and 2) and Freedom House (columns 3 and 4) scores. The table shows the effects on all cases (columns 1 and 3) and on the subset of cases in which peace lasts at least one year (columns 2 and 4). Table 2.4 shows the same thing for democratization two years out; Table 2.5 for five years out (with the ‘‘if peace lasts’’ columns referring to cases in which peace lasts two and five years, respectively). It is worth discussing some of the control variables first since many of these results are quite surprising. There is a relationship between the level of democracy at the start of the war and democratization afterwards, but this relationship is negative. Antebellum experience with democracy actually makes countries more likely to move toward autocracy after the fighting stops. This does not appear to be the result of the fact that more democratic countries simply have less room for improvement after the war – the result holds even if we drop countries with high pre-war democracy scores from the analysis.26 This surprising result is perhaps less so when we consider that civil wars are watershed events in the politics of a country. They mark a clear break from the politics of the past. Many civil wars turn autocratic states into democratic ones, while those enjoying democracy before the war have experienced a cataclysmic shock to that system. The results for GDP per capita and illiteracy are similarly unexpected. One of the strongest findings in the democratization literature is that a relatively high level of wealth is a precondition for democracy. In the subset of cases that have experienced civil war, however, the opposite seems to be the case; richer countries enjoy, if anything, lower 25

26

Thus, I assume that cases are independent across conflicts, but not necessarily within them. That is, democratization in Angola is independent of democratization in Cambodia, but democratization in Angola after 1991 and after 1994 are related processes. This negative relationship is significant for the Polity measure, but not for the Freedom House measure. Note, however, that the measure of pre-war democracy is taken from Polity data in both cases.

0.643 (0.993) 0.266 (0.098) 0.0003 (0.0002) 0.017 (0.025) 1.847 (1.227) 1.027 (0.947) 0.635 (0.218) 1.435 (1.090) 2.547 (0.871) 1.957 (0.661) 1.661 (0.814) 1.555 (1.054) 8.929 (3.052)

81 0.38

All peacekeeping Democracy at war start GDP/capita Illiteracy Victory Treaty Cost of war Contraband financing Oil Failed past agreement Perm-5 contiguous Perm-5 former colony Constant

N R2

Coef. (RSE)

All cases

Table 2.3 Democratization one year out

0.52 0.01 0.24 0.50 0.14 0.29 0.01 0.20 0.01 0.01 0.05 0.15 0.01

P > |t|

If peace lasts

56 0.43

0.802 (1.204) 0.230 (0.102) 0.0005 (0.0003) 0.002 (0.029) 0.955 (1.650) 0.114 (1.301) 0.653 (0.285) 1.391 (1.327) 1.692 (0.975) 1.567 (0.717) 0.941 (1.240) 2.752 (1.282) 10.756 (4.047)

Coef. (RSE)

Polity measure

0.51 0.03 0.11 0.93 0.57 0.93 0.03 0.30 0.09 0.04 0.45 0.04 0.01 84 0.11

0.058 (0.679) 0.041 (0.057) 0.000 (0.002) 0.007 (0.019) 0.005 (0.697) 1.362 (0.732) 0.182 (0.128) 0.063 (0.883) 0.252 (0.711) 0.172 (0.522) 0.191 (0.468) 0.120 (0.927) 2.415 (1.782)

P > |t| Coef. (RSE)

All cases

If peace lasts

0.93 0.48 0.49 0.70 1.00 0.07 0.16 0.94 0.73 0.74 0.69 0.90 0.18 58 0.12

0.700 (0.871) 0.029 (0.053) 0.000 (0.000) 0.001 (0.022) 0.948 (0.780) 0.815 (0.796) 0.123 (0.170) 0.157 (0.894) 0.247 (0.800) 0.453 (0.566) 0.041 (0.733) 0.695 (1.013) 2.839 (2.124)

P > |t| Coef. (RSE)

Freedom House measure

0.43 0.59 0.37 0.96 0.23 0.31 0.47 0.86 0.76 0.43 0.96 0.50 0.19

P > |t|

0.769 (0.980) 0.244 (0.097) 0.0003 (0.0003) 0.035 (0.025) 1.766 (1.257) 1.424 (0.890) 0.574 (0.225) 0.656 (1.160) 2.987 (0.807) 1.926 (0.721) 1.867 (0.819) 2.052 (1.049) 8.066 (3.043)

81 0.41

All peacekeeping Democracy at war start GDP/capita Illiteracy Victory Treaty Cost of war Contraband financing Oil Failed past agreement Perm-5 contiguous Perm-5 former colony Constant

N R2

Coef. (RSE)

All cases

0.44 0.02 0.24 0.17 0.17 0.12 0.02 0.58 0.00 0.01 0.03 0.06 0.01

If peace lasts

49 0.48

0.238 (1.262) 0.223 (0.105) 0.0005 (0.0003) 0.020 (0.030) 0.523 (1.716) 0.012 (1.357) 0.518 (0.298) 0.237 (1.486) 2.752 (1.119) 1.863 (0.825) 0.883 (1.351) 3.786 (1.257) 10.301 (4.345)

Coef. (RSE)

Polity measure

P > |t|

Table 2.4 Democratization two years out

0.85 0.04 0.15 0.52 0.76 0.99 0.09 0.87 0.02 0.03 0.52 0.01 0.02

P > |t|

84 0.21

0.589 (0.673) 0.069 (0.066) 0.0002 (0.0002) 0.012 (0.016) 0.692 (0.816) 1.932 (0.661) 0.152 (0.133) 0.322 (0.689) 1.078 (0.496) 0.133 (0.462) 0.407 (0.508) 0.033 (0.817) 1.927 (2.060)

Coef. (RSE)

All cases

0.39 0.30 0.48 0.46 0.40 0.01 0.26 0.64 0.04 0.78 0.43 0.97 0.36

P > |t|

51 0.25

0.342 (0.818) 0.077 (0.056) 0.0000 (0.0003) 0.002 (0.020) 0.684 (0.941) 0.634 (0.855) 0.041 (0.167) 0.002 (0.802) 1.540 (0.754) 0.159 (0.572) 0.031 (0.938) 0.977 (0.908) 2.023 (2.590)

Coef. (RSE)

0.68 0.18 0.96 0.91 0.47 0.46 0.81 1.00 0.05 0.78 0.97 0.29 0.44

P > |t|

If peace lasts

Freedom House measure

0.226 (1.156) 0.285 (0.104) 0.0001 (0.0004) 0.022 (0.036) 1.394 (1.142) 0.629 (1.050) 0.669 (0.247) 1.073 (1.529) 5.348 (1.628) 1.261 (0.840) 0.675 (1.655) 2.537 (1.006) 10.458 (3.667)

76 0.37

All peacekeeping Democracy at war start GDP/capita Illiteracy Victory Treaty Cost of war Contraband financing Oil Failed past agreement Perm-5 contiguous Perm-5 former colony Constant

N R2

Coef. (RSE)

All cases

0.85 0.01 0.74 0.54 0.23 0.55 0.01 0.49 0.00 0.14 0.69 0.02 0.01 35 0.67

0.459 (1.211) 0.471 (0.124) 0.0007 (0.0004) 0.013 (0.035) 1.991 (1.328) 1.157 (1.341) 0.446 (0.293) 2.668 (1.429) 9.190 (1.641) 0.998 (1.189) 0.071 (1.427) 2.934 (1.220) 10.282 (4.291)

0.71 0.00 0.12 0.71 0.15 0.40 0.14 0.07 0.00 0.41 0.96 0.02 0.02

P > |t|

If peace lasts Coef. (RSE)

Polity measure

P > |t|

Table 2.5 Democratization five years out

84 0.26

0.014 (0.670) 0.110 (0.066) 0.0003 (0.0002) 0.010 (0.015) 1.139 (0.788) 1.362 (0.714) 0.138 (0.147) 0.535 (0.793) 2.473 (0.732) 0.268 (0.486) 0.601 (0.519) 1.367 (0.786) 3.350 (2.103)

Coef. (RSE)

All cases

0.98 0.10 0.21 0.51 0.16 0.06 0.36 0.50 0.00 0.58 0.25 0.09 0.12

P > |t|

39 0.39

0.974 (1.026) 0.073 (0.092) 0.0001 (0.0003) 0.010 (0.025) 0.192 (1.364) 0.237 (1.018) 0.047 (0.204) 0.926 (1.264) 3.335 (1.335) 0.258 (0.872) 0.761 (1.137) 2.088 (1.083) 4.256 (3.278)

Coef. (RSE)

0.35 0.44 0.89 0.71 0.89 0.82 0.82 0.47 0.02 0.77 0.51 0.07 0.21

P > |t|

If peace lasts

Freedom House measure

Peacekeeping and democratization

61

levels of democratization (that is, smaller moves toward democracy or greater moves toward autocracy) than poorer ones in the first few years. These effects are not statistically significant (though in some models they are close) and should thus be interpreted with caution, but we certainly do not see the expected positive effect of GDP in the immediate aftermath of war. This negative effect falls away by the fifth year, however, suggesting that the short-term and long-term effects of wealth are quite different. The short-term effect of literacy is the opposite of what we would expect when the Polity measure of democratization is used. Again these coefficients are not significant, but countries with higher proportions of people who cannot read appear to undergo, if anything, more democratization than those with more literate populations. By the fifth year out, this effect holds when we examine all cases but flips to the expected direction in the subset of cases in which peace has lasted to date. The relationship between literacy and the Freedom House measure is more in line with expectations, higher levels of illiteracy depress moves toward ‘‘freedom,’’ but this relationship is never even remotely significant. In short, literacy appears to have no systematic benefits for post-war democratization. The effects of cost of war are more in line with what we would intuitively expect. The more people killed and displaced by the war, the harder it is to democratize. This effect is statistically significant and quite robust for the Polity measure but is not significant for the Freedom House measure. For the Polity measure, the size of this effect is fairly consistent whether we are examining all cases or only those in which peace has lasted.27 This suggests that the negative effect of the cost of war is not just the result of the fact that high-cost wars are more likely to resume, which in turn depresses democratization. Rather the cost of war appears to have an independent effect on efforts to move toward democracy. However, for the Freedom House measure, the effect of the cost of war is generally larger for the full set of cases than for the restricted set. This difference between the two measures is telling. It indicates that the direct effect of costly wars operates more strongly on the institutional forms of democracy rather than on political rights and civil liberties. For the latter, it is the detrimental effect

27

The measure used here is the natural log of the number killed or displaced.

62

Virginia Page Fortna

on stability that undermines democracy. This may be because very costly wars undermine a state’s political and physical infrastructure in ways that make building institutions more difficult. Alternatively, it may reflect the countervailing effects suggested by the second dilemma outlined above; that more costly wars lead citizens to demand more political rights in ways that offset otherwise negative effects on democracy. Our proxy measure for levels of mistrust between the recent belligerents (whether previous agreements between them have failed) has a similar negative effect on the establishment of the institutions and procedures of democracy (as measured by Polity). Here, however, the results appear to wear off over time. One and two years out, a failed agreement in the past depresses democratization by about two points (on the twenty-point scale), all else equal, and this effect is statistically significant. By five years out, however, the effect is about half as big, and no longer significant. This effect does not hold for the expansion of civil liberties and political rights (as measured by Freedom House), however. Wars that end with a decisive victory for one side see, if anything, larger moves toward the institutions and procedures of democracy (Polity) than those that do not, consistent with earlier research (Toft 2003). However, this effect is not significant, though it is often close, so should be treated with caution. Moreover, it does not hold for the Freedom House measure of democratization. The effect of victory on civil liberties and political rights is generally negative, though not consistently so. Remember that the comparison category here is fighting that ends with only a cease-fire or truce in place – an outcome that is neither good for stable peace, nor, clearly, for democracy. When we examine all cases, we see that wars that end with a peace treaty also experience positive levels of democratization (though again, this effect is not always significant). This is true whichever measure of democratization is used. This effect is generally smaller than that for decisive victory, however, and it does not hold among only those cases for which peace lasts. This suggests that while peace treaties help make peace more stable, which in turn fosters democratization, they do not, in and of themselves, foster democratization. This result is quite surprising, as many peace treaties call for elections in an attempt to move conflict from the battlefield to the ballot box.

Peacekeeping and democratization

63

Natural resources also have interesting effects on attempts at postwar democratization. Oil-rich countries are cursed by their valuable assets. The negative effect of oil on democratization is large, statistically significant, and grows over time. On the Polity scale, oil reduces democratization by about 2 points one year out, 3 points two years out, and a whopping 5 to 9 points by the fifth year after the war ends. The pattern is much the same in the Freedom House data.28 This result confirms studies that find ‘‘oil and democracy don’t mix.’’ As Ross (2001) argues, oil wealth (a) allows governments to use low taxes and patronage to avoid pressures for reform, (b) allows them to fund efforts to repress reformist movements, and/or (c) produces economic growth that, unlike other forms of economic development, does not lead to social and cultural changes, such as education and economic specialization, that tend to foster democracy. On the other hand, and quite surprisingly, contraband financing for rebels does not appear to hurt the prospects for democracy. In fact the positive coefficient for this variable in the Polity data suggests just the opposite. This effect is not significant (though it comes close in some models), nor does it hold in the Freedom House data, so we should not conclude that conflict diamonds and drugs are good for democracy. But the fact that we see no detrimental effect and the possibility of a positive effect is quite unexpected. The relationship between a country and the great powers – the United States, Great Britain, France, China, or Russia (also known as the permanent members of the Security Council, or Perm-5) – also has an effect on post-war democratization. Conflicts that are in (e.g., Northern Ireland, Chechnya) or next door to the Perm-5 see higher rates of democratization one and two years out. But this effect is much larger and only statistically significant when we examine all cases rather than the subset for which peace has held. And in the Polity data at least, the effect is largely gone five years down the road. Interestingly, former colonies of these countries democratize several points fewer on the Polity scale than other countries, an effect that is consistent across models and generally statistically significant. While this effect is much less pronounced, it is generally still negative in the Freedom House data. This finding is particularly surprising as all of 28

The scale of the two measures is different, but the size of the effect of oil on democratization is somewhat smaller in the Freedom House data.

64

Virginia Page Fortna

these countries are former colonies of three leading democracies (the United States, Britain, and France) that we might expect to be particularly interested in fostering democracy in their former wards.29 Why former colonies of these three countries might fare so poorly in postwar democratization is unclear – it cannot be because these cases start at higher rates of democracy and so have less room for improvement, since the analysis controls for democracy levels at the start of the war. This is an interesting topic for further investigation. Turn now to our main variable of interest, peacekeeping. Overall, the results in Tables 2.3–2.5 are rather weak: peacekeeping appears to have little effect on democratization. There is no statistically significant effect in either direction.30 In the short term, one and two years after the fighting stops, there is more evidence for a positive effect than a negative one, at least when the Polity measure of the institutions and procedures of democracy is employed (columns 1 and 2). Peacekeeping does not appear to thwart democratization efforts and may have a modest positive effect. But this effect is not statistically significant – we cannot place much confidence in this positive result.31 We can, however, be fairly confident that peacekeeping does not undermine democratization, at least in the short term. In the longer term, five years out, the positive effect becomes even smaller, and when only cases in which peace lasts are considered, we see a possibly negative effect. This suggests at least weak support for the temporal dilemma described above. Although, again, lack of statistical significance makes definitive statements impossible to make. The effects are even worse for peacekeeping when the Freedom House measure of moves toward or away from ‘‘freedom’’ is used (columns 3 and 4). Here there is no evidence of a positive effect one year out and if anything peacekeeping depresses freedom among the subset of cases in which peace lasts. Two years out the coefficients are 29

30

31

This surprising negative effect is stronger in former French and US colonies (the several cases in the Philippines are the only instances of the latter) than in former British colonies (results not shown). This means that the positive and negative coefficients could be the result of chance alone. Note, however, that because the data used here represent all ceasefires in civil wars during this period, not a smaller sample, issues of inference from the data to the larger population are much smaller than in many statistical analyses. In models in which the control for former colonies of the Perm-5 is omitted, this positive effect is often significant or close to it.

Peacekeeping and democratization

65

positive for both sets of cases, but still not significantly so. Five years out, we again see no effect (the coefficient is essentially zero) in the full set of cases, and if anything a negative effect in the restricted set. This suggests that the direct effect of peacekeeping is, if anything, to thwart political rights and civil liberties, but that this is offset by the positive effects of peacekeeping on peace and stability. In short, there is evidence for the dilemmas proposed above – peacekeeping can help establish peace, which is good for democracy, but it also undermines the establishment of that democracy. Tables 2.6–2.8 distinguish between the four types of peacekeeping mission outlined above: observer missions, interpositional or traditional peacekeeping missions, multidimensional missions, and enforcement missions. As noted above, the first three categories are consent-based missions with Chapter VI mandates, while the fourth does not rely on consent and is authorized under Chapter VII of the UN Charter. As before, results are shown for both Polity and Freedom House measures, and for both the full set of cases and the restricted set of cases in which peace holds one, two, and five years out, respectively. Examining systematic differences in the effects of different types of peacekeeping is difficult given the small numbers of cases in each category, but this breakdown yields some potentially interesting results. Let us look first at the effect of mission type on the institutions and procedures of democracy, as measured by Polity (columns 1 and 2 of the tables). One year after the fighting stops, all types of peacekeeping have a positive (if insignificant) effect on democratization except multidimensional peacekeeping. This is particularly striking because it is exactly these missions, with large civilian components for election monitoring, human rights monitoring, and other attempts at institution building that we would expect to have the largest positive impact on democracy building. In the very short term, it appears that maintaining security is more important for the establishment of democracy than explicit attempts by peacekeepers to build its institutions. By two years out the picture has changed, however. Here, both multidimensional peacekeeping and more robust peace enforcement missions have, if anything, a positive effect on democracy building, while monitoring and traditional peacekeeping missions have negative coefficients. Five years out, monitoring and traditional peacekeeping continue to have a negative effect, and multidimensional peacekeeping maintains a positive coefficient. The effect of enforcement missions is

0.717 (1.653) 0.322 (1.090) 1.396 (1.048) 1.995 (1.766) 0.240 (0.104) 0.0003 (0.0003) 0.020 (0.027) 1.633 (1.330) 1.164 (1.013) 0.571 (0.222) 1.211 (1.143) 2.596 (0.995) 2.101 (0.715) 1.561 (0.850) 1.551 (1.029) 8.217 (3.194)

81 0.41

Monitoring PK Traditional PK Multidimensional PK Enforcement PK Democracy at war start GDP/capita Illiteracy Victory Treaty Cost of war Contraband financing Oil Failed past agreement Perm-5 contiguous Perm-5 former colony Constant

N R2

Coef. (RSE)

All cases

If peace lasts

0.67 0.77 0.19 0.27 0.03 0.32 0.47 0.23 0.26 0.01 0.30 0.01 0.01 0.07 0.14 0.01 56 0.48

1.221 (2.064) 0.434 (1.298) 1.448 (1.324) 2.019 (2.003) 0.199 (0.115) 0.0005 (0.0003) 0.009 (0.032) 0.666 (1.856) 0.272 (1.375) 0.550 (0.309) 1.182 (1.376) 1.802 (1.102) 1.691 (0.687) 0.869 (1.373) 2.743 (1.263) 9.514 (4.486)

P > |t| Coef. (RSE)

Polity measure

0.56 0.74 0.28 0.32 0.09 0.17 0.79 0.72 0.84 0.08 0.40 0.11 0.02 0.53 0.04 0.04 84 0.12

0.602 (1.413) 0.536 (0.937) 0.197 (0.966) 0.494 (0.866) 0.040 (0.058) 0.000 (0.000) 0.008 (0.019) 0.011 (0.709) 1.372 (0.821) 0.177 (0.137) 0.135 (0.875) 0.391 (0.696) 0.160 (0.519) 0.230 (0.489) 0.137 (0.930) 2.425 (1.990)

0.67 0.57 0.84 0.57 0.50 0.53 0.67 0.99 0.10 0.20 0.88 0.58 0.76 0.64 0.88 0.23

58 0.14

0.667 (1.887) 1.162 (0.930) 1.211 (1.176) 0.115 (1.041) 0.021 (0.056) 0.000 (0.000) 0.001 (0.023) 1.071 (0.767) 0.763 (0.852) 0.114 (0.194) 0.138 (0.908) 0.314 (0.825) 0.392 (0.568) 0.018 (0.784) 0.675 (1.031) 2.735 (2.546)

P > |t| Coef. (RSE)

If peace lasts

Freedom House measure All cases P > |t| Coef. (RSE)

Table 2.6 Democratization one year out, by peacekeeping mission type

0.73 0.22 0.31 0.91 0.72 0.47 0.97 0.17 0.38 0.56 0.88 0.71 0.49 0.98 0.52 0.29

P > |t|

0.260 (1.648) 0.037 (1.116) 0.348 (1.697) 2.025 (1.676) 0.235 (0.103) 0.0004 (0.0003) 0.034 (0.027) 1.575 (1.333) 1.355 (0.946) 0.569 (0.226) 0.497 (1.217) 3.125 (0.864) 1.976 (0.777) 1.824 (0.868) 2.043 (1.013) 8.242 (3.185)

81 0.42

Monitoring PK Traditional PK Multidimensional PK Enforcement PK Democracy at war start GDP/capita Illiteracy Victory Treaty Cost of war Contraband financing Oil Failed past agreement Perm-5 contiguous Perm-5 former colony Constant

N R2

Coef. (RSE)

All cases

0.88 0.97 0.84 0.23 0.03 0.23 0.20 0.24 0.16 0.02 0.69 0.00 0.02 0.04 0.05 0.01

P > |t|

49 0.50

1.559 (1.905) 0.813 (1.341) 0.397 (2.405) 1.585 (2.188) 0.194 (0.133) 0.0007 (0.0004) 0.019 (0.035) 0.216 (1.776) 0.295 (1.465) 0.574 (0.342) 0.012 (1.682) 2.923 (1.292) 1.777 (0.860) 0.682 (1.515) 3.624 (1.205) 11.461 (5.087)

0.42 0.55 0.87 0.47 0.15 0.13 0.59 0.90 0.84 0.10 0.99 0.03 0.05 0.66 0.01 0.03

P > |t|

If peace lasts Coef. (RSE)

Polity measure

Table 2.7 Democratization two years out, by peacekeeping mission type

84 0.23

0.698 (0.970) 0.617 (0.738) 1.262 (0.960) 0.863 (0.957) 0.070 (0.071) 0.0002 (0.0003) 0.012 (0.017) 0.534 (0.854) 1.691 (0.767) 0.178 (0.149) 0.281 (0.718) 1.009 (0.557) 0.085 (0.493) 0.358 (0.515) 0.011 (0.798) 2.386 (2.288)

Coef. (RSE)

All cases

0.48 0.41 0.20 0.37 0.33 0.42 0.50 0.54 0.03 0.24 0.70 0.08 0.86 0.49 0.99 0.30

P > |t|

P > |t|

51 0.29

1.735 (1.889) 0.36 0.814 (0.886) 0.36 0.740 (1.343) 0.59 0.584 (1.150) 0.62 0.064 (0.069) 0.36 0.0000 (0.0004) 0.91 0.002 (0.023) 0.91 .921 (1.014) 0.37 0.193 (0.925) 0.84 0.093 (0.226) 0.68 0.114 (0.821) 0.89 1.554 (0.833) 0.07 0.238 (0.679) 0.73 0.245 (1.012) 0.81 0.731 (0.936) 0.44 2.891 (3.345) 0.39

Coef. (RSE)

If peace lasts

Freedom House measure

0.120 (2.061) 1.238 (1.040 1.600 (2.043) 1.394 (1.746) 0.300 (0.119) 0.0000 (0.0004) 0.017 (0.038) 1.383 (1.216) 0.449 (1.151) 0.733 (0.256) 0.989 (1.550) 5.586 (1.683) 1.157 (0.854) 0.734 (1.665) 2.492 (1.039) 11.570 (3.860)

76 0.39

Monitoring PK Traditional PK Multidimensional PK Enforcement PK Democracy at war start GDP/capita Illiteracy Victory Treaty Cost of war Contraband financing Oil Failed past agreement Perm-5 contiguous Perm-5 former colony Constant

N R2

Coef. (RSE)

All cases

0.95 0.24 0.44 0.43 0.02 0.99 0.66 0.26 0.70 0.01 0.53 0.00 0.18 0.66 0.02 0.01

P > |t|

35 0.69

2.502 (3.217) 0.995 (1.459) 1.044 (3.028) 0.056 (1.259) 0.474 (0.139) 0.0006 (0.0005) 0.021 (0.041) 1.727 (1.511) 1.853 (1.185) 0.567 (0.318) 2.534 (1.534) 9.427 (1.588) 0.685 (1.249) 0.376 (1.622) 2.627 (1.178) 12.353 (5.305)

0.44 0.50 0.73 0.97 0.00 0.21 0.62 0.26 0.13 0.09 0.11 0.00 0.59 0.82 0.03 0.03

P > |t|

If peace lasts Coef. (RSE)

Polity measure

Table 2.8 Democratization five years out, by peacekeeping mission type

P > |t|

84 0.28

1.184 (1.378) 0.40 0.587 (0.870) 0.50 0.311 (0.907) 0.73 0.110 (0.897) 0.90 0.119 (0.070) 0.09 0.0003 (0.0002) 0.26 0.013 (0.015) 0.40 1.307 (0.779) 0.10 1.473 (0.765) 0.06 0.144 (0.159) 0.37 0.580 (0.780) 0.46 2.671 (0.689) 0.00 0.367 (0.498) 0.47 0.684 (0.541) 0.21 1.418 (0.795) 0.08 3.468 (2.271) 0.14

Coef. (RSE)

All cases

P > |t|

39 0.43

2.306 (2.169) 0.30 1.827 (1.016) 0.08 0.348 (1.441) 0.81 0.257 (1.607) 0.87 0.061 (0.086) 0.48 0.0000 (0.0003) 0.92 0.009 (0.030) 0.76 0.506 (1.366) 0.71 0.609 (1.063) 0.57 0.132 (0.228) 0.57 0.817 (0.904) 0.37 3.476 (1.331) 0.01 0.187 (0.852) 0.83 0.764 (1.074) 0.48 1.936 (1.373) 0.17 5.549 (2.745) 0.05

Coef. (RSE)

If peace lasts

Freedom House measure

Peacekeeping and democratization

69

positive for all cases, but not for the subset in which peace has lasted, suggesting that this positive effect is all about helping to maintain peace, not in a direct affect on democratization. This suggests at least some evidence for the dilemma noted above for more robust military peacekeeping. It may be better able to keep the peace effectively, but may also impinge on democracy to a larger extent. Again, none of these effects is significant, so any conclusions must be extremely tentative. The story is somewhat different, and murkier, when we examine the Freedom House measures of change in political rights and civil liberties (columns 3 and 4 in Tables 2.6–2.8). One year after the war ends, the effect of all types of peacekeeping is negative among the restricted set of cases for which peace is holding. Among the full set of cases, monitoring and enforcement missions indicate a positive relationship with democratization, while traditional and multidimensional missions indicate a negative relationship. However, any difference among mission types is most likely simply noise in the data, as none of the relationships is statistically significant. In other words, as above, peacekeeping has, if anything a negative direct effect on ‘‘freedom,’’ but this is offset by a positive effect on stability. Two years out, the coefficients for all types of mission except monitoring missions are positive in both sets of cases, but again, we see no significant effects. Five years out, the indication of any direct effect of peacekeeping on democratization is negative (as indicated in the restricted set of cases), while the effects in the full set of cases are mixed. Overall, there is little we can conclude about differences in the various mission types in terms of effects on civil liberties and political rights. The results generally echo those for peacekeeping in general, as discussed above. That is, the direct effects are generally negative, as indicated in the restricted set of cases, while differences among mission types in the full set of cases are inconsistent across time. As noted earlier, there are too few cases of transitional administration missions to examine their effects on democracy in the quantitative analysis. But a quick look at the democratization records of three cases in this category is revealing. Because the war in Cambodia ended so soon after the Cold War wound down, it gives us a chance to examine political changes over a relatively long time period (some fourteen years). The UN mission in Cambodia ostensibly put that country on the road to democracy after decades of brutality and war. In the few years after the war ended in 1991, both Cambodia’s Polity scores and

70

Virginia Page Fortna

its Freedom House scores show a slight improvement over its pre-war record (though there is little change from its Polity scores in the year before the war ends). But Cambodia’s democratization project soon stalled. Ongoing political violence and intimidation, and then a bloody coup in 1997 reversed any gains in democracy or freedom. Since then, Cambodia has democratized only slightly.32 In neither East Timor nor Kosovo do we have as long a time span to study post-war democratization. The early democratization news in East Timor seems quite good as the UN transitional authority helped East Timorese leaders set up fledgling democratic institutions. Elections for a constituent assembly in August 2001 set the stage for independence in 2002. In terms of both political rights and civil liberties, there has been enormous improvement, especially relative to the abuses surrounding the independence vote in 1999 that precipitated the crisis. While there are no Polity scores or Freedom House rankings for East Timor, as opposed to Indonesia as a whole, before the war, both now code the country as relatively democratic. Note that the rest of Indonesia has also democratized significantly in the last several years, making it somewhat harder to claim that it has been the UN mission in East Timor that accounts for democratization.33 However, the Freedom House country reports give credit to the United Nations Transitional Administration in East Timor (UNTAET) for helping to 32

33

Cambodia’s Polity score was as low as 7 and 5 before the war (in the 1970s) but was 1 by 1990. In the early 1990s it remained at 1, dipped sharply in 1997 to 7, then rose to 2 in 1998, where it has remained since. Cambodia’s Freedom House ranking for political rights/civil liberties, respectively, was 7/7 (as far ‘‘not free’’ as the scale goes) during the war, improved slightly to 6/6 in 1991 and 1992, moving to the ‘‘partly free’’ ranking of 4/5 for 1993 and 1994, before sliding over 1995 and 1996 to 7/6 in 1997, and then rebounded only slightly, to 6/6 from 1998 to 2000, and 6/5 since then. (Remember that in its original codings, higher numbers correspond to less freedom rather than more.) The country reports that Freedom House has issued for Cambodia since 2002 describe elections marred by intimidation and violence, a judiciary that is not independent, and despite some improvements in human rights in recent years, an ongoing culture of impunity. Only in 2005 has the formation of a coalition government put a slight dent in Hun Sen’s autocratic rule (Freedom House 2002–2005a). Polity gives East Timor a relatively high democracy score of 6 in 2002 and 2003. Indonesia’s score is 7 for the same years, a vast difference from the 5 in the year before the war ended, 1998. The Freedom House rankings for East Timor begin in 2002 with political rights/civil liberties at 5/3. The situation improves to 3/3 by 2003 because of the presidential elections and stays there through 2005.

Peacekeeping and democratization

71

‘‘rebuild roads and buildings and set up legislative and other basic democratic institutions’’ (Freedom House 2002–2005b). Kosovo, which continues to be administered by the UN, has democratized much less than East Timor. Because Kosovo is not independent, there is no Polity data available, but Freedom House has issued reports on the territory since 2003. These reports note some improvement in both civil liberties and political rights in 2002 ‘‘because postwar ethnic discrimination and terror has largely subsided.’’ They note ‘‘generally free and fair [municipal] elections,’’ but limited freedom of expression because of security concerns and problems with freedom of movement for ethnic minorities. The ratings stay the same through the 2004 report, but the 2005 report demotes Kosovo slightly (with an overall status of ‘‘not free’’ rather than ‘‘partly free’’ as before) because of the Serbian boycott of parliamentary elections following ethnic violence in 2004. While it is far too early to tell what the democracy trajectories of East Timor and Kosovo will be over the long haul, together these three cases of transitional administration suggest much the same pattern as other forms of peacekeeping. There is some evidence of moves toward democracy in the short term, but this is often followed by backsliding toward autocracy several years down the road. The effects of peacekeeping on democratization discussed so far make no distinction between UN missions and peacekeeping undertaken by other regional organizations or ad hoc groups of states. It is reasonable to ask whether there is a difference. UN peacekeeping might be considered more legitimate, both by the belligerents themselves and by the international community, than peacekeeping conducted by others. There might also be differences in the efficacy of UN vs. nonUN peacekeeping. Table 2.9 shows the effects of UN and non-UN peacekeeping on democratization (in both cases, relative to no peacekeeping mission). Because the effects of the control variables are substantially the same as in earlier tables, I omit them here in the interest of space. However, the results that are shown control for the same set of variables as in Tables 2.3–2.8. As before, results are shown for both the Polity and the Freedom House measures, for both the full set of cases and those in which peace has lasted one, two, and five years out. In the short term, one and two years after the war ends, it appears that non-UN peacekeeping is better (or in some cases less bad) for democratization than UN peacekeeping. Peacekeeping performed by

81 0.38

N R2

UN peacekeeping Non-UN peacekeeping [Control results omitted]* Constant

81 0.41

49 0.49

9.686 (4.452)

0.01

7.944 (2.998)

0.720 (1.564) 0.677 (1.593)

56 0.45

10.04 (4.124)

—

0.291 (1.384) 1.378 (1.462)

Coef. (RSE)

—

0.71 0.40

0.01

0.96 0.37

P > |t|

If peace lasts

—

0.454 (1.194) 0.946 (1.123)

B. Democratization two years out

N R2

A. Democratization one year out UN peacekeeping 0.053 (1.104) Non-UN 1.035 (1.149) peacekeeping [Control results — omitted]* Constant 8.660 (3.022)

Coef. (RSE)

All cases

Polity measure

Table 2.9 UN vs. non-UN peacekeeping

0.04

0.65 0.67

0.02

0.84 0.35

P > |t|

84 0.21

1.931 (2.069)

—

0.600 (0.824) 0.583 (0.683)

84 0.12

2.200 (1.707)

—

0.494 (1.016) 0.377 (0.652)

Coef. (RSE)

All cases

0.36

0.47 0.40

0.21

0.63 0.57

P > |t|

51 0.27

1.644 (2.755)

—

0.247 (1.193) 0.603 (0.794)

58 0.16

2.215 (2.058)

—

1.625 (1.116) 0.224 (0.768)

Coef. (RSE)

If peace lasts

Freedom House measure

0.55

0.84 0.45

0.29

0.15 0.77

P > |t|

76 0.38

N R2

0.02

0.90 0.63

84 0.26

3.336 (2.134)

—

0.023 (0.957) 0.035 (0.688)

0.13

0.98 0.96

39 0.41

3.851 (2.423)

—

1.689 (1.266) 0.626 (1.200)

0.12

0.19 0.61

*These regressions control for: Democracy at war start, GDP/capita, Illiteracy, Victory, Treaty, Cost of war, Contraband financing, Oil, Failed past agreement, Perm-5 contiguity and Perm-5 colony. Findings for these variables are substantively the same as in previous tables. Results are omitted for the sake of space.

35 0.67

10.39 (4.291)

0.01

10.81 (3.703)

0.293 (2.229) 0.551 (1.113) —

0.50 0.72

—

1.135 (1.647) 0.355 (0.976)

UN peacekeeping Non-UN peacekeeping [Control results omitted]* Constant

C. Democratization five years out

74

Virginia Page Fortna

organizations or groups of states other than the UN has either a positive effect on peacekeeping while UN missions have a negative effect, or non-UN peacekeeping has less of a negative effect or a greater positive effect than does UN peacekeeping. In the medium term, five years on, the same can be said for the Freedom House measure of democratization, but the reverse is true for the Polity measure. For the latter, UN peacekeeping has a positive (or in the restricted set of cases, a less negative) effect on democratization, relative to non-UN missions. In all of these regressions, neither the UN nor the non-UN peacekeeping coefficient is significant, so we should be very careful about drawing strong conclusions from these comparisons, but they provide at least weak evidence that UN peacekeeping is no more effective, and may actually be less effective, than peacekeeping by other groups at instituting democracy in post-war societies.

Conclusion Many countries do make substantial strides toward democracy in the aftermath of civil war. But despite the fact that building democracy has been a major goal of the international community when it deploys peacekeepers to war-torn societies, peacekeeping has no clear or strong positive effect on democratization, relative to cases where belligerents are left to their own devices. This stands in stark contrast to its effects on the stability of peace. Peacekeepers help keep the peace very effectively, but they do not necessarily foster democratization. The weak effects of peacekeeping in general on democratization provide evidence for the dilemmas outlined earlier in this chapter. The negative and positive effects of peacekeeping on democracy appear to cancel each other out, as the discussion above of the democratizing effects of both war and peace would suggest. There is also at least some evidence for the systemic and temporal dilemmas discussed above. The presence of outside peacekeepers can help foster democracy, for example, by providing stability, decreasing levels of mistrust and fostering democratic institutions, but this external presence is unaccountable politically and can crowd out the indigenous growth of democracy. Thus we see that peacekeeping may help foster democratization in the short term but tends to undermine it over the longer term. The small number of cases makes it hard to draw distinctions among the effects of different types of peacekeeping missions – almost none of

Peacekeeping and democratization

75

the peacekeeping effects in Tables 2.6–2.8 are statistically significant. But it is striking that we do not see positive effects on democracy for multidimensional missions, or in the examination of the few cases in this category, of transitional administrations, which represent the most focused attempts by peacekeepers to democratize post-war states. Overall, we see again that the direct effects of peacekeeping on democracy may undermine the positive indirect effects of increased stability, while short-term gains are often undone in the longer term. It is also notable that UN peacekeeping efforts, which might be thought more legitimate than non-UN missions, are no better and may in fact be worse for democracy than peacekeeping by other organizations or groups of states. Examination of the effects of peacekeeping on the establishment of civil liberties and political rights (the key components of the Freedom House measure of democracy) suggests that peacekeeping can foster democracy by fostering stable peace, but that its direct effects on political rights tend to be detrimental. It should be stressed that the lack of statistical significance for almost all of these results means we should be very cautious about interpreting these findings – the differences among types of missions or between UN and non-UN peacekeeping could easily be the result of statistical chance alone.34 Nonetheless, these findings are consistent with the complex relationship between peacekeeping and democratization hypothesized above. It is, of course, difficult to reach strong conclusions about the longterm effects of peacekeeping when in some cases we have data available to examine only five years of post-war attempts at democratization – only time will tell what happens over the next decades. There are also limits to what can be gleaned from large-N statistical surveys. This analysis has two key benefits over other types of research methods. It allows us to cover all cases in a systematic way, and it allows for multivariate analysis, which is crucial for evaluating a treatment, such as peacekeeping, that is not applied at random. But careful analysis of individual cases is required to flesh out the complex relationships and

34

As noted above, however, because these analyses represent the full universe of cases we can have more confidence in the results than if we were attempting to infer patterns from a smaller sample of cases.

76

Virginia Page Fortna

the dilemmas inherent in post-war democratization. The other chapters in this volume take on that important task. The empirical results presented here, while somewhat tentative, suggest that there are indeed trade-offs and dilemmas inherent in trying both to maintain peace and to build democracy in post-war societies. There are trade-offs between the short term and the long term, and between the need for outsiders to help put a country on the road to stable peace and democracy on the one hand, and the need on the other for outsiders to leave political space for indigenous actors to build homegrown institutions. The international community has professed two central goals when it intervenes in war-torn states – to help maintain peace and to foster the growth of democracy. These have been presented as complementary tasks: stability is meant to encourage democracy, and democracy is thought to help maintain stability. The institution of peacekeeping has been designed to help with both endeavors. But there are contradictions between these strategies, and peacekeeping is not a cure-all. The evidence is very strong that peacekeeping is effective at achieving the first goal. Peacekeeping makes peace much more likely to last. But peacekeeping has not contributed strongly to the second goal. To the contrary, the direct effects of peacekeeping (that is, separate from its effect on stability) have, if anything been negative. This is especially true for the less institutional or procedural aspects of democracy such as the expansion of political rights and civil liberties. This creates fundamental dilemmas for the international community. Because there is some truth to the conventional wisdom that stable peace fosters democracy, especially in the short term, the contribution of peacekeeping to stability indirectly fosters democratization. But by intruding in the domestic politics of countries as they try to recover from civil war, peacekeepers also crowd out local efforts to build democratic traditions over the long term. That is, the direct and indirect effects of peacekeeping on the establishment of democracy are fundamentally contradictory. The weak and often contradictory statistical results presented above indicate these competing processes at work. Peacekeeping is good for democracy, and bad for it at the same time. Because peacekeeping is clearly and unambiguously good for peace, countries emerging from civil war are better off when peacekeepers deploy than when they are left to their own devices. But the international community should be under no illusions that peacekeeping will

Peacekeeping and democratization

77

be able to transform war-torn states into ideal democratic societies. There are, rather, fundamental dilemmas and trade-offs between the goals of peace and democracy, and between the short- and long-term effects of peacekeeping on democracy.

Appendix: List of cases Mission type key 0 ¼ none 2 ¼ observation 3 ¼ interpositional 4 ¼ multidimensional 5 ¼ enforcement Note some cases experienced more than one mission type. The highest mission code is reported here. Name

Cease-fire date

Peacekeeping mission type

Afghanistan–Mujahideen Afghanistan–Taliban Algeria–FIS/AIS Angola Angola Azerbaijan–Nagorno K. Azerbaijan–Nagorno K. Bangladesh–CHT Bosnia Bosnia Cambodia Central African Rep. Chad Congo–Brazzaville Congo–Brazzaville Congo–Brazzaville Congo Dem. Rep./Zaire Croatia Croatia Djibouti Djibouti

25 Apr 1992 07 Mar 1993 15 Oct 1997 31 May 1991 20 Nov 1994 31 Aug 1993 16 May 1994 01 Aug 1992 01 Jan 1995 14 Dec 1995 23 Oct 1991 25 Jan 1997 11 Aug 1994 30 Jan 1994 15 Oct 1997 29 Dec 1999 17 May 1997 30 Mar 1994 12 Nov 1995 28 Feb 1992 26 Dec 1994

0 0 0 2 4 0 0 0 3 5 4 3 0 0 0 0 0 4 5 3 0

78

Virginia Page Fortna

Name

Cease-fire date

Peacekeeping mission type

Egypt El Salvador Ethiopia–Eritrea Ethiopia–ideology Georgia–Abkhazia Georgia–Abkhazia Georgia–Ossetia Guatemala Guatemala Guinea–Bissau Guinea–Bissau Guinea–Bissau Guinea–Bissau Haiti India–Assam India–Assam India–Sikh Indonesia–Aceh Indonesia–E. Timor Iraq–Kurds Iraq–Kurds Iraq–Shia Israel–Palestinians Lebanon Liberia Liberia Mali Mali Moldova Morocco/W. Sahara Mozambique Myanmar–Kachin Myanmar–Karen Myanmar–Karen Namibia Nicaragua Pakistan–Mohajirs Papua New Guinea Papua New Guinea Papua New Guinea

15 Sep 1997 16 Dec 1992 21 May 1991 21 May 1991 27 Jul 1993 14 May 1994 14 Jul 1994 26 Apr 1991 20 Mar 1996 26 Aug 1998 02 Nov 1998 03 Feb 1999 07 May 1999 18 Sep 1994 20 Apr 1991 17 Dec 1991 31 Dec 1993 31 Dec 1991 25 Oct 1999 01 Mar 1993 15 Oct 1996 15 Dec 1993 13 Sep 1993 13 Oct 1990 28 Nov 1990 17 Aug 1996 06 Jan 1991 31 Mar 1995 21 Jul 1992 06 Sep 1991 04 Oct 1992 15 Oct 1993 28 Apr 1992 15 Jun 1995 01 Apr 1989 19 Apr 1989 15 Oct 1999 15 Mar 1990 21 Jan 1991 10 Oct 1997

0 4 0 0 0 3 3 0 3 0 0 2 2 5 0 0 0 0 5 5 5 0 0 5 5 5 0 0 5 2 4 0 0 0 4 3 0 0 0 2

Peacekeeping and democratization

79

Name

Cease-fire date

Peacekeeping mission type

Peru Philippines–Communists Philippines–Mindanao Philippines–Mindanao Philippines–Mindanao Romania Russia–Chechnya Russia–Chechnya Rwanda Rwanda Rwanda Senegal Sierra Leone Sierra Leone Somalia South Africa Sri Lanka (Tamil) Sri Lanka (JVP II) Sudan Sudan Sudan Sudan Tajikistan Turkey–Kurds Turkey–Kurds Uganda–Kony Uganda–LRA UK–N. Ireland UK–N. Ireland Yemen Yugoslavia–Croatia Yugoslavia–Croatia Yugoslavia–Kosovo Yugoslavia–Kosovo

31 Dec 1996 31 Dec 1995 07 Nov 1993 31 Dec 1990 02 Sept 1996 23 Dec 1989 01 Jun 1996 23 Aug 1996 31 Jul 1992 04 Aug 1993 18 Jul 1994 08 Jul 1993 30 Nov 1996 07 Jul 1999 27 Jan 1991 26 Apr 1994 07 Jan 1995 29 Dec 1989 01 May 1989 28 Mar 1995 15 Jul 1998 15 Apr 1999 27 Jun 1997 20 Mar 1993 01 Sep 1999 15 Feb 1989 15 Jul 1992 31 Aug 1994 10 Apr 1998 10 Jul 1994 23 Nov 1991 03 Jan 1992 12 Oct 1998 09 Jun 1999

0 0 0 0 0 0 0 0 2 3 3 0 0 5 0 0 0 0 0 0 0 0 5 0 0 0 0 0 0 0 0 3 2 5

3

Violence in war-to-democracy transitions ¨ GLUND KRISTINE HO

In the Democratic Republic of Congo, violent clashes in the eastern Kivu region have continued unabated in spite of the transition process that began in 2003. In Afghanistan, insecurity has been a prominent feature of the post-Bonn period, and escalations in violence clustered around the presidential and parliamentary polls in 2004 and 2005. In Liberia, violent riots spread through the capital Monrovia in October 2004, in spite of the heavy United Nations peacekeeping presence on the ground. These examples illustrate that the signing of a peace deal and the introduction of democracy to a war-torn society do not prevent killings and atrocities from taking place – at least not in the short term. This chapter asks two basic questions: Does democratization unintentionally create conditions conducive to violence? And, does the management of violence sometimes undermine the process of democratization? By combining insights from the literature on peacebuilding and democratization, the chapter brings to light central dilemmas related to violence and security in countries emerging from war that recently have entered a path toward democracy. Two arguments about the problematic relationship between violence, democratization, and peace are put forward and discussed. Firstly, I argue that democratization creates new opportunities and motives for violence. The opening up of political space in the area of political competition (elections), and freedom of expression (media), as well as reforms of the security sector are crucial to democratization. However, in the transition period, these institutions are flawed or ineffective, providing opportunities for substantial maneuvering for elites who are willing to take advantage of the situation to fan violence. Secondly, I argue that measures to combat violence – for instance through the use of coercive measures and the inclusion of spoilers in politics – can undermine the democratization process. While it is necessary for domestic and international actors to deal with violence, the manner in which violence prevention and control is carried out can 80

Violence in war-to-democracy transitions

81

vary substantially, ranging from the use of force to negotiated solutions. A particular challenge stems from the fact that new forms of violence emerge as the organized political conflict subsides, rendering old ways of dealing with violence ineffective. This chapter analyzes violence and security in relation to five issue areas – elections, media reform, security sector reform, the use of force to manage violence, and inclusion as a means to regulate violence. The purpose is to identify the four types of dilemmas outlined in Chapter 2 of this book as they pertain to these issues. Horizontal dilemmas are activated when considering whether to include or exclude violence makers and spoilers in the democratic process as a way of dealing with violence. By including armed actors in, for instance, transitional governments, legitimacy is granted to violence makers, with the risk of continued reliance on violence as a political tool. Vertical dilemmas relate to the most effective ways of dealing with violence. Coercive measures may be effective in the sense that they stop violence through threat and deterrence. On the other hand, combating violence with violence might undermine the legitimacy of state institutions or peacekeepers. Systemic dilemmas concern the role of international actors. A dilemma for state institutions might arise from the need for international assistance both in terms of verifying human rights and rebuilding the state institutions for the protection of human rights. While international actors are needed to oversee peace agreements and to create a secure political environment and basic civilian security, the dependence on international assistance might undermine trust in local institutions. Finally, temporal dilemmas are particularly crucial when considering democratization efforts. For instance, reform of the police and military is needed to democratize the security institutions. However, security sector reform takes time and might in the meantime leave society vulnerable to violence. In terms of establishing a secure environment, lack of action to deal with violence might aggravate instability with the effect of creating new grievances, and thereby increase the risk of escalation of violence.

Violence, democratization and democracy Since the end of the Cold War, democracy promotion has been part of the peacebuilding strategy of the international community. Also in cases with comparatively little international involvement – as in

82

Kristine Ho¨glund

South Africa’s transition – the process of democratization has often overlapped with a transition from war to peace. The rationale for this approach – to build peace through democracy – is straightforward: a prevailing assumption both within theory and practice has been that peace and democracy are mutually reinforcing. While this is true in the long run an expanding body of scholarly work has contended that democratization processes – in particular in post-war societies – are highly conflictual. Indeed, under certain circumstances, democratization will make a return to war more likely (Mansfield and Snyder 2001; 2005; Paris 2004; Snyder 2000). From this perspective, a more reasonable starting point is to assume that democratization processes are prone to recurring outbursts of violence constituting severe challenges to war-shattered societies. This premise is also in line with evidence from peace processes worldwide indicating that transitions from war to peace are frequently accompanied by violence – both in new forms and along the old conflict lines. Political assassinations, violent riots, and extreme levels of crime are threats to the new political order and to basic civilian security. In extreme cases, failure to deal with violence can lead to a resumption of war – as was the case of Angola in 1992 – or to genocide – as in Rwanda in 1994. A growing literature on conflict resolution and war termination addresses the problem of violence and security during peace negotiations and in the post-settlement phase (e.g., Darby 2001; 2006; Ho¨glund 2005; 2008; Kydd and Walter 2002; Stedman 1997). In particular, the issues of security sector reform (SSR) (e.g., Call and Stanley 2003; Cawthra and Luckham 2003), and disarmament, demobilization, and reintegration (DDR) (e.g., Gamba 2003; Salomons 2005; Spear 2002) have been accorded a great deal of attention. There is also an awareness that violence is commonly part of democratic transitions (Dahl 1971; Huntington 1991). However, while definitions of democracy typically include some notion of non-violence, the literature on democracy oftentimes fails to include an analysis of the relationship between violence and democratization.1 As expressed by Laurence Whitehead, ‘‘[m]uch theorizing – not only on democracy but also on democratization – takes the basic security of the median citizen as a datum, a presupposition upon which liberal and 1

Important exceptions include the edited volumes by Rapoport and Weinberg (2001a), and Cawthra and Luckham (2003).

Violence in war-to-democracy transitions

83

constitutional systems can be founded, rather than as problematic social constructs’’ (2002: 163). The following section brings together insights from these two strands of research to pinpoint two key problems related to violence and security in war-to-democracy transitions. Violence commonly refers to the intentional killing, injuring, or coercion of people, or the destruction of property, and for analytical purposes a distinction is made between political and criminal violence. However, in the context of a war-affected society the boundaries of political and criminal activity become obscured, making it difficult to differentiate the two forms of violence clearly or easily. For instance, armed groups may use criminal violence, such as kidnappings or extortion, to finance the armed struggle. In such cases, criminal violence clearly has political consequences, although these might not have been intended. It also means that violence which in non-warring societies would be considered a non-political criminal act sometimes becomes politicized during the armed conflict. Bearing these caveats in mind, I will seek to distinguish between criminal and political violence in this chapter.

Democratization: political space for violence When belligerents close peace deals to end an armed conflict, violence rarely ends. There are several reasons for the continued existence of violence in post-war societies, such as the enduring legacies of violence and the uncertainties inherent in transition processes. In what ways does the democratization effort itself increase the inclination of actors to utilize violence? Scholars have found both new opportunities and new motives for violence makers in democratization processes (Cawthra and Luckham 2003; Gurr 2000a: 85; Snyder 2000).2 Ideally, increases in democracy in

2

The argument has been thoroughly developed by Jack Snyder (2000) in relation to the perils of democratization and nationalism. For instance, Snyder has emphasized the tendency of elites to mobilize electorates along ethnic divisions in order to gain support. The argument is also in line with some findings within the research on terrorism, indicating that democracies are more likely than non-democracies to experience terrorism. William Lee Eubank and Leonard Weinberg, for instance, find that ‘‘the likelihood of terrorist groups occurring in democracies is three and one half times greater than their occurrence in nondemocracies’’ (1994: 426).

84

Kristine Ho¨glund

terms of contestation and participation provide new channels for groups within society to express discontent. However, under certain circumstances, the opening up of political space may result in additional channels through which violence can be instigated and organized. The war-to-democracy transition may also create new motives for violence. Ideally, the peace accord and the democratization process address some of the grievances and conflict issues that the war was fought over. However, democratization processes may threaten the interests of elites: ‘‘democratisation raises political expectations, while at the same time tends to be actively resisted, often forcibly, by those whose power and position it threatens’’ (Cawthra and Luckham 2003: 6). The literature on conflict resolution makes similar claims. The elite may feel threatened by the peace accord and the peace itself. So-called ‘‘spoilers’’ pose a particular problem: those actors who are not satisfied with the emerging peace and therefore are intent on disrupting the implementation of a peace agreement (Stedman 1997). Dissatisfaction with a peace agreement may stem both from fear about what the peace will bring and from the extreme positions held by some sectors within society. Opportunities and motives may also interact, increasing the propensity to use violence among actors in the war-to-democracy transition. As Ted Robert Gurr (2000a: 86–87) writes: ‘‘In democratizing autocracies . . . national and minority peoples ordinarily feel a loss of security simultaneously with a transient increase in opportunities for mobilization and action. New democratic regimes usually lack the resources or institutional means to make and guarantee the kind of accommodation that typify the established democracies.’’ This section discusses measures activated as part of a democratization process that inadvertently may increase opportunities or motives for violence. It highlights that in a situation where people are used to dealing with conflict violently and where there are many grievances left unsolved – the new-won freedoms can be exploited for violent purposes. Violence in relation to elections, media reform, and security sector reform – three key features of a democratization process – will be discussed in turn. Elections Competitive elections are integral features of democracy.3 Most internationally sponsored peace agreements stipulate the holding of free and 3

See more on elections in Reilly’s chapter in this volume.

Violence in war-to-democracy transitions

85

fair elections.4 These elections serve a twofold purpose: (1) to establish a legitimate and representative democratic government, and (2) to seal the agreement and to put a decisive end to the war (Kumar 1998c: 5; Reilly 2003b). Furthermore, to the international community, elections for new political institutions provide a suitable point in time for reducing its presence in the country (Lyons 2002; 2004; Reilly 2003b).5 Many scholars would argue that elections in post-war societies are ‘‘fundamentally different from those organized under normal circumstances’’ (Kumar 1998c: 1). For instance, local conflicts may surface as refugees return to their communities to vote. The opening up of political competition, which is a prerequisite for democratic elections, creates new opportunities to organize violence. Competitive elections – to be deemed free and fair6 – require fundamental political rights such as freedom of movement and freedom of speech. These rights can be misused as a vehicle for militant political organization. In particular, the competitive nature of elections may aggravate existing conflicts and societal cleavages. In countries which have experienced protracted conflict, political mobilization is likely to be along the conflict lines (Paris 2004). Differences rather than similarities are brought to the surface to win votes. Moreover, in electoral processes the use of military metaphors is common: ‘‘the parties wage ‘campaigns’, employing ‘strategies and tactics’. Party faithful are called ‘cadre’, and areas with many supporters are known as ‘strongholds’ or ‘citadels’’’ (Rapoport and Weinberg 2001b: 31). For this reason, the rhetoric used during the war is easily available to be made use of in an election campaign to mobilize voters for support. In competitive elections, groups that have had a dominant position in society run the risk of losing some of their power if the voters do not 4

5 6

Rafael Lopez-Pintor (1997) uses the term ‘‘reconciliation elections’’ to denote such elections. There is a large literature on elections and election monitoring in general (e.g., Abbink and Hesseling 1999; Cowen and Laakso 1997; Elklit 1999; McCoy, Gerber, and Pastor 1991), and some studies – mainly case specific – on electoral violence (e.g., Damm 2003; Fischer 2002; Pausewang, Trondvall, and Aalen 2002; Rapoport and Weinberg 2001a). However, there are few studies that specifically apply the concepts to the context of war-shattered societies. For important contributions on post-war elections more generally, see in particular the edited volume by Krishna Kumar (1998c) and Terrence Lyons (2005). This constitutes the so-called ‘‘exit strategy.’’ For a discussion on the concept of ‘‘free and fair,’’ see Jørgen Elklit and Palle Svensson (1997).

86

Kristine Ho¨glund

lend them support. Thus, elections may create incentives for political actors to influence the electoral process through threat, intimidation, and violence. The situation immediately after the elections is particularly sensitive: will the contenders accept the outcome, or will they resort to violence unless they emerge as winners? Thus, elections themselves can produce outbreaks of violence (Rapoport and Weinberg 2001b: 20).7 Electoral violence has been understood as ‘‘any random or organized acts that seek to determine, delay, or otherwise influence an electoral process through threat, verbal intimidation, hate speech, disinformation, physical assault, forced ‘protection,’ blackmail, destruction of property, or assassination’’ (Fischer 2002: 8). There are surprisingly few studies that have focused on the issue of electoral violence in postwar societies. However, elections in Afghanistan and Iraq in recent years have brought attention to the difficult issue of establishing a secure environment that can facilitate free and fair democratic elections in countries emerging from violent conflict. In many cases the first elections after a civil war take place in a situation where the parties are not fully disarmed and demobilized. There is an abundance of cases where elections have prompted violence. For instance, in August 1999 the people in East Timor went to the polls to have their say regarding independence in a referendum organized by the UN. An overwhelming majority voted in favor of independence from Indonesia. However, the vote unleashed a brutal campaign of violence and looting carried out by militia groups with the support of the government of Indonesia. Prompted by the disorder, the UN Security Council authorized the deployment of a multinational force under Australian command in September 1999, to restore peace and security. In contrast, the elections to the constituent assembly in 2001 were conducted in a secure environment, although there had been fears that the elections would cause a new escalation in violence. In a democratic process, it is important that elections take place under conditions that are free from fear and intimidation. Violence and insecurity may affect the election result, if large sections of the 7

There are different suggestions about what these conditions may be. According to Jeong (2005: 105), ‘‘the military balance on the ground can determine whether elections without demobilization can pose a serious threat to stability.’’

Violence in war-to-democracy transitions

87

population refrain from voting. Threat and intimidation during the election campaign may force political contenders to leave the electoral process. For instance, in Ethiopia, the Oromo Liberation Front (OLF) withdrew its participation in the elections in 1992 due to threats, intimidation, and attacks by the Ethiopian People’s Revolutionary Democratic Front (EPRDF), which subsequently took power. Elections in 2000 and 2001 were also accompanied by harassment and violence, causing the opposition to leave the campaign. To summarize: elections in a post-war society may inadvertently give rise to new opportunities and new incentives for violence makers to take up arms or to use threat and intimidation to influence the political process. For instance, those whose power is threatened by democratic elections – often termed spoilers – may use violence to disrupt the transition process or to overthrow the election result. Other actors may use violence as a means to influence the polls and the election outcome, by preventing some actors from participating in the election campaign or by intimidating people from going to the polls. Media reform Freedom of expression and press freedom are generally considered cornerstones of a democratic society. Functioning and diverse media provide information to the citizens and critical scrutiny of political issues, crucial to democratic accountability.8 Freedom of expression is also formulated as a fundamental human right in Article 19 of the Universal Declaration of Human Rights and in other international law documents. In the course of armed conflict, media commonly become polarized and serve as a tool for propaganda to the conflict parties. Thus, in countries emerging from war, the media scene is in need of reform. As part of the transition, efforts are often made to create new laws guaranteeing freedom of expression. Reform of the media infrastructure includes the formation of independent media and widening the range of media outlets and ownership. These reforms are needed to promote a diversity of perspectives and to make these accessible to the public. However, in the words of Ross Howard (2005: 118): ‘‘[i]t is a contradiction but true that in the singular most accessible democracies, where

8

On media and economic development, see World Bank (2002a).

88

Kristine Ho¨glund

the media has complete freedom, the strongest impulses of the news media may still lead it to play a destructive role.’’9 Moreover, in a democratizing context, there is a risk that the reform of media – in particular their liberalization – can lead to the creation of irresponsible media outlets of negative and nationalistic propaganda (Paris 2004; Snyder 2000; Snyder and Ballentine 1996). In the most extreme case – Rwanda – sustained propaganda campaigns instigated genocidal violence. In particular, the hate propaganda by the Hutudominated radio station Radio des Milles Collines has been deemed particularly important in the genocide. Moreover, the newly liberalized press was extremely polarized, with Hutu-extremists fanning antiTutsi violence and anti-regime press further adding to the fears of the Hutu of being excluded from power (Snyder and Ballentine 1996). In most cases the link between media and violence is indirect in the sense that media create a political environment conducive to polarization and violence.10 However, there are examples of media in warshattered societies spurring violence more directly through biased reporting. For instance, in 2004 some twenty people died in Kosovo as violent anti-Serb riots spread across the province. These riots had been stimulated by the inflammatory media reports about the presumed drowning of three boys chased down into a river by Serbs near the ethnic flashpoint Mitrovica. These reports were filled with flaws and the claim that Serbs had been involved in the drowning proved inaccurate (Haraszti 2004; Temporary Media Commissioner 2004). The international community has instructed many initiatives to support media reform in democratization processes. Training in what has been termed ‘‘peace media’’ has been gaining ground in peacebuilding efforts. However, the international community has been criticized for doing too little to suppress hate media. One way to promote effective media is to endorse ‘‘codes of conduct’’ and licensing systems (Paris 2004: 198), which are used to regulate media in established democracies. Another tool for the international community is to interrupt the transmission or dissemination of hate media. For instance, in the summer of 1997, Serb Radio Television in Bosnia launched an 9

10

Gadi Wolfsfeld (2001; 2003; 2004) has examined the positive versus the negative effect that media can have on peace processes and has specifically analyzed the conflicts in the Middle East and Northern Ireland. For a useful overview of the incitements of ethnic violence, see Donald L. Horowitz (2001).

Violence in war-to-democracy transitions

89

anti-Nato campaign, which was interrupted through international intervention (Metzl 1997). To recapitulate, in the hands of some people, media can be a powerful tool to foster hatred and to mobilize people quickly for violence. In particular, media can be influential in the development of violent riots and communal violence – oftentimes with ethnic overtures. In the most extreme cases media have played a crucial part in genocidal violence. Security sector reform In the course of an armed conflict, the country’s security forces tend to gain extreme powers. It is often difficult to maintain the distinction between upholding law and order within the country, and providing protection against external threats. The security forces – including the police, paramilitary units, and other special units – are frequently part of the counter-insurgency and repression of the opposition. Establishing civilian control over the military thus becomes important for moving toward democratic governance.11 The reform of the police – in order for it to be the sole provider of internal security – is crucial in this process (Holm and Eide 2000). However, a dilemma arises when the size of the security sector is reduced at the same time as new threats surface. A substantial threat in countries emerging from war is the often-seen increase in violent crime, including armed robbery and looting, car hijackings, and domestic violence.12 Several factors are important in explaining the rise in violent crime in countries such as South Africa, Guatemala, El Salvador and Iraq. Among those is the lack of effective institutions to deal with crime, widespread poverty, and easy access to arms. Profitable economic activities that funded the violent campaigns 11

12

A dilemma related to security sector reform is that the military might feel threatened by the civilian initiative for reform and may take action against it (Kohn 1997). There are three areas in which civilian control over the military must prevail in order to be democratic: (1) ‘‘clear chain of command under all circumstances, with the head of the government atop of that chain,’’ (2) ‘‘decision to begin or end warfare lies in civilian hands,’’ (3) military policy more generally (Kohn 1997: 6–7). See also Huntington (1957) on ‘‘objective civilian control.’’ Few studies, have examined criminal violence in connection to war and transition periods. However, more scholars are paying attention to the topic; see, for instance, Roger Mac Ginty (2006). There are also several studies on crime in specific post-war societies, such as South Africa. See for instance, Antoniette Louw (1997) and Mark Shaw (2002).

90

Kristine Ho¨glund

of the guerillas or rebels during the conflict are commonly turned into networks of organized crime. There is also the hypothesis suggesting that as a result of the war experience, there is a higher tolerance for the use of violence more generally.13 Reform of the police and the military, in particular in the short term and in connection to rising levels of crime, makes it difficult for warshattered societies to deal effectively with violent crime (Cawthra and Luckham 2003: 314–315). In El Salvador, the ‘‘[d]isruption of the internal security system took its toll. The eventual turnover of almost the entire investigative units meant that networks of informants had to be reconstructed’’ (Call 2003: 843). There was a similar lack of resources in, for instance, the attorney’s office due to reform of the judicial system. The problem was further aggravated by the fact the demobilization of the armed forces left ‘‘a public security gap, the number of people circulating under arms decreased from 60,000 (including combatants on both sides) to only about 6,000 National Police officers’’ (Call 2003: 843). In other cases, as in Guatemala, the absence of an effective police force has led to situations where the military has been called in to cooperate with the police to deal with crime. This meant that instead of much needed resources to reform the police, funds were channeled to boost up the military budget (challenging the limits to military spending stipulated in the peace agreement). Dilemmas related to democratization efforts Democratization can provide new opportunities for those disgruntled with the emerging post-war political order. Moreover, the intrinsic uncertainties of a war-to-democracy process can create new motives for violence. In fact, in the post-war period, new forms of violence may emerge, such as electoral violence. The security vacuum resulting from security sector reform may lead to dramatic increases in criminal violence. There are several dilemmas pertaining to reforms for democratization, which may have the perverse effect of opening political space for violence. Media reforms, the holding of democratic and competitive 13

Dane Archer and Rosemary Gartner (1976) made a study of homicide rates in countries that had taken part in combat during international conflict and found that violent killings increased substantially in such countries. An explanation for this result is the so-called legitimization of violence hypothesis, which suggests that war legitimizes the use of violence in society beyond the realm of combat.

Violence in war-to-democracy transitions

91

elections, and security sector reform, all give rise to temporal dilemmas. Media reform instigated to support democratic development in a war-torn society – if changes are introduced too fast or without adequate controls – may in the short term indirectly be used to instigate violence. The vulnerability of war-torn societies going through changes in the security sector which are required for long-term democratization may undermine stability in the short term. The introduction of electoral processes in a war-to-democracy transition also encompasses issues of timing: should elections be held early with the risk that extremists come to power, or should they be postponed until the society is less polarized? Postponements of reforms or the holding of elections until security is established can activate the vertical dilemma. How can a legitimate post-war government be created without elections? Different arguments about the timing of post-war elections have been put forward. Roland Paris, for instance, contends that elections should not be held until the conditions for holding free and fair elections are institutionalized. This means the development of moderate political parties and of a ‘‘judicial mechanism to rule on election-related disputes’’ (Paris 2004: 190). Terrence Lyons, on the other hand, argues that if elections are postponed, ‘‘opportunities to assist war-termination’’ may be lost (2005: 96). Thus, from a peacebuilding perspective it may be counterproductive to postpone elections. For instance, the first post-war elections in Bosnia – carried out in September 1996 – were not only a matter of electing government officials but also served to legitimate the Dayton Peace Agreement. The vertical dilemma is a potential quandary also in relation to security sector reform. The change of uniforms – from a repressive regime or from a rebel group – does not automatically infer an increase in legitimacy of the police and military. If the same people that have committed atrocities become part of the new police force, legitimacy of the new institutions may be undermined. However, at the same time, the recruitment of former rebels and regime is needed to fill a security vacuum which might otherwise emerge. Recruitment may also be important to occupy former soldiers who are potential threats to peace if left unemployed. The question of eligibility for recruitment to the police and armed forces becomes crucial. Liberia after the peace agreement in 2003 is a case in point. The new army does not include any elements from the old military apparatus, and it is hoped that it

92

Kristine Ho¨glund

therefore will be granted a high degree of legitimacy. However, the establishment of a new force has been slow and it is therefore believed that the international community will need to maintain its support to Liberia’s security sector for a long period. On the other hand, the recruitment to the police has been considered successful, but there are fears that the newly trained policemen will immediately fall into the patterns of the old police.14 The vertical and temporal dilemmas are closely related in the sense that security sector reform means more than downsizing, a new name, and a change of uniform. For security sector reform to be successful, a change in norms and a development toward more respect for human rights are needed. This takes time. Moreover, underlying the temporal dilemma is a more fundamental dilemma related to the legitimacy of the new post-war order. If public insecurity remains one of the key problems after a peace agreement has been reached, the ability of local actors to address the issue of violence properly may be questioned. This quandary will be discussed next.

Countering violence but undermining democracy? Democracy is built on principles of non-violence. However, violence is never entirely eradicated from society, prompting reflection on ways in which violence can be prevented and brought under control. In established democracies, violence is controlled and managed by elected governments, backed by the police and military as the executive institutions. The legitimate use of force is restricted and granted to specific agencies. Those responsible for violence are held accountable through the judicial system based on the rule of law. This section highlights the difficult task of dealing with violence during a war-to-democracy transition: a situation that is often characterized by a need for fundamental transformation of the institutions responsible for security, as well as a weak adherence to the rule of law. It points specifically to the problem that efforts to control and manage violence – both by domestic and by international actors – can severely undermine the democratization process. The protection of human rights and the provision of basic civilian security is a centrepiece of liberal democracy. During an armed conflict, human rights are abused as part of the violence campaign both by 14

See more on the issue in the ICG report (2005d: 17).

Violence in war-to-democracy transitions

93

the government and rebels. The human rights abuses can take many forms, for instance forced recruitment to rebel groups, or torture and ‘‘disappearance’’ of political opponents by the state. Human rights violations often transcend the purposes of the political conflict and may, for instance, be used more generally to suppress opposition and regime critics. Thus, peace deals are struck between parties that lack an encouraging record in respect for human rights. In order to uphold basic human rights and security, the belligerents (the government side and the opposition alike) have to undergo a transformation both in their purpose and in their means to deal with conflict without violating basic human rights. Such transformation does not happen overnight. For this reason, security is a key issue during both the implementation of peace agreements and the democratization processes. International actors are often brought in to oversee the peace implementation and to provide security guarantees, because trust between the belligerents is low. Moreover, local actors frequently lack the necessary resources to sustain a secure environment. If violence and insecurity gain the upper hand in a war-to-democracy transition, a return to armed conflict is one possible scenario. Excessive levels of political and criminal violence will severely challenge the legitimacy of the governing institutions. Surveys indicate, for instance, that disproportionate levels of crime undermine people’s trust in the state institutions and democracy. Moreover, situations of extreme crime and delinquency ‘‘can stimulate authoritarian and anti-democratic reflexes among significant sectors of the electorate’’ (Whitehead 2002: 181). Promises about a restoration of law and order thus appeal to large sections of a society fraught with violent crime. In El Salvador, polls indicated increased political support for the hardliners as crime and violence escalated (Call 2003: 859). In Israel, Benjamin Netanyahu came to power in 1996 on a political platform calling for toughness against terrorists. A consequence of the failure of public institutions to provide basic civilian security is that in response to criminal violence, alternative providers of security emerge.15 For instance, vigilante groups have 15

An important issue to consider is who it is that provides security in a society. According to John Darby, the state and its agencies have the primary responsibility for maintaining law and order even in cases where they are fraudulent (2001: 44). However, during protracted conflict, the armed opposition can take control over large sections of the territory. For instance, in

94

Kristine Ho¨glund

surfaced in South Africa as a result of the extreme levels of crime. Moreover, in societies with high crime rates, private security guards have found a profitable market. Thus, they have taken over security functions, which in a working democracy should be provided by state institutions. This privatization of security is problematic for democratization since it erodes the legitimacy of the state institutions. In the following section I will discuss the precariousness of using force to control and combat violence in a democratizing context, and attempts to deal with violence makers through marginalization or inclusion in the democratic process. The use of force In established democracies, the state has the legitimate right to use coercive measures to enforce the rule of law and to protect the country from external threats. However, in countries emerging from war, the state institutions frequently lack legitimacy and the necessary resources to deal effectively with violence. Moreover, in war-shattered societies, where violence is often at high – although subsiding – levels, it is important to find means to control violence, that in themselves do not undermine the democratization process. It underlines the fact that ‘‘[g]etting violent with violence is . . . risky’’ (Keane 2004: 174). In several cases, coercive measures to deal with violence have severely called into question the government’s willingness to uphold basic democratic principles. In Northern Ireland, a set of anti-terror legislation was adopted after a bomb in Omagh in 1998 – planted by a republican splinter group – killed twenty-eight people. The legislation gave broad powers to the security forces in the pursuit of suspected terrorists and evoked memories of past abuses committed during the course of the conflict. In war-to-democracy transitions, the local institutions to deal with violence are often weak and in many cases local officials might be perpetrators of violence and human rights abuses themselves (Plunkett 2005). For this reason, abuses commonly persist after the armed conflict has ended. Human rights violations have serious consequences for democratization. For instance, if the violations are carried out by the government agencies, confidence in the post-war order the mid-1990s the rebel group LTTE in Sri Lanka had de facto control over the northern part of the country.

Violence in war-to-democracy transitions

95

is likely to erode (Kumar 1997: 9). Persistent human rights abuses – whether by the government or by paramilitaries – will run counter to attempts to reconcile social cleavages in society. For instance, human rights abuses in post-war Guatemala have clearly inhibited meaningful dialogue between the human rights community and the military. The result is a deeply polarized society. In many cases, the international community is called in or has intervened to restore and maintain security in the post-agreement phase.16 In some cases, the peacekeepers stand mainly as a buffer between the belligerents. However, if state institutions are weak, peacekeeping operations take on policing activities to uphold basic civilian security and to create conditions for peace implementation and democratization (Viggo Jakobsen 2003). ‘‘Where there is no effective local independent judiciary or police – the peacekeepers will be required to take on the task of arrest, prosecution, and trial of serious offenders’’ (Plunkett 2005: 85). In such instances, the international forces have to be strong enough to withstand challenges from spoilers and to deal with violence. James Fearon and David Laitin cite two examples of violence management by the international community in post-war societies. In Tajikistan, Russian peacekeepers were successful in dealing with violence makers because they were willing and able to utilize violence and coercive measures themselves. Police units with members recruited from the Royal Ulster Constabulary (RUC) (with experience from Northern Ireland) ‘‘were probably the most effective police unit in Kosovo, although their methods may not always have been fully in accord with human rights conventions’’ (Fearon and Laitin 2004: 23). Kosovo also provides plenty of examples of where the international actors have held individuals viewed as security risks in custody on very loose grounds and without being allowed a trial. The Kosovo case also underscores the point that international actors are oftentimes inadequately equipped to deal with the new forms of violence that emerge in post-war societies. Analyses of the riots in Kosovo during the spring of 2004 (in which nineteen people lost their lives) report that the international forces under the NATO-led Kosovo Force (KFOR) and the UN Mission in Kosovo (UNMIK) had neither sufficient capacity in terms of 16

See more on the role of international actors in the chapter on peacekeeping by Fortna.

96

Kristine Ho¨glund

number of troops deployed, nor had adequate training and equipment to respond effectively to the ensuing riots.17 Thus, the use of force is highly problematic in countries emerging from war. There is a danger that the excessive use of force – by local security institutions or by international police or peacekeeping missions – in particular if these are not considered legitimate, may undermine the democratization process. Inclusion as a means to regulate violence Key players have carried out political assassinations, bomb explosions, and other deeds to challenge peace accords and a nascent democratization process. Armed conflict commonly experiences the rise of organized groups, warlords, and politicians that gain a position of power and profit from war-related activities. These actors will only accept peace and democratization on terms under which their investments and privileges are not severely threatened. They have often been referred to as ‘‘spoilers,’’ defined by Stephen Stedman as ‘‘leaders and parties who believe that peace emerging from negotiations threaten their power, worldview, and interests, and use violence to undermine attempts to achieve it’’ (Stedman 1997: 5). While the literature on spoilers generally has focused on rebels and guerillas, actors connected to the state (the military, militias and paramilitaries, etc.) can also use violence to disrupt the new order (Ho¨glund and Zartman 2006). Spoiler violence can be directed toward the opponent, in order to instill fear and undermine political support for the peace agreement. However, violence to destabilize a peace agreement may also target moderates within one’s own group, also referred to as ‘internal feuding’ (Darby 2001). In many peace processes, deliberate efforts have been made to regulate violence through the inclusion of violence makers in the political process, in exchange for promises about the abandonment of violence.18 In numerous cases, the calling of a cease-fire has been a prerequisite for the inception of peace negotiations. In Northern Ireland, the British Government demanded a cease-fire from the Irish Republican Army to allow their political branch, Sinn Fein, to participate in the peace talks. 17

18

See, e.g., the reports by the Human Rights Watch (2004) and the International Crisis Group (ICG 2004a). A related issue is whether actors who have committed war crimes should be allowed to participate in the political process. See more in So¨derberg Kovacs’ chapter on this issue.

Violence in war-to-democracy transitions

97

In addition, the Mitchell principles were established to regulate the inclusion of parties in the negotiation forum. These constituted a set of principles of democracy and non-violence that the parties had to commit to, to be allowed to participate in the peace negotiation process. Sinn Fein and parties linked to loyalist paramilitary groups were on a few occasions expelled from the peace talks due to cease-fire violations. After the peace agreement in 1998, which set up the self-governing institutions in Northern Ireland, pledges to principles of non-violence have continued to be a means to regulate violence. Ministers to the Executive had to reaffirm their commitment to non-violence and a set of sanctions – ranging from the exclusion of a minister from his post, to non-payment of fiscal support to the party – were available if it was judged that a political party was implicated in violence. For instance, after it was established that the IRA was involved in the serious armed robbery of a bank in Belfast in December 2004, the Northern Ireland Secretary of State acted to remove financial assistance to Sinn Fein. Formal criteria for inclusion in peace processes – often through cease-fires – may result in new forms of violence, since it is in the interest of the parties not to be associated with violations of the principles agreed to. In Northern Ireland, for instance, concerns about adherence to the cease-fire and the Mitchell principles made the paramilitaries change violence tactics from the use of punishment shootings to punishment beatings (Mac Ginty and Darby 2002: 89). In the wake of the 2002 cease-fire in Sri Lanka, threats, intimidation, and killings between Tamil paramilitaries and within the LTTE outnumbered the violence between the LTTE and the government forces (Ho¨glund 2005). This has called into question the value of the cease-fire. Thus, inclusion in the political process based on commitments to peaceful means can be an important tool to prevent and manage violence. However, trust in the inclusion mechanism can be undermined if new forms of violence arise which are not as clearly linked to the parties. Dilemmas emerging from violence management The prevention, control, and management of violence in war-todemocracy transitions through the use of force or through inclusions give rise to several serious dilemmas for democratization. The question of the utility and consequences of the use of force relates both to the vertical dilemma and to the systemic dilemma. The

98

Kristine Ho¨glund

vertical dilemma is highlighted by persistent human rights abuses and the excessive use of force, which might undermine the legitimacy of those in power and the state institutions in the post-war order. The role of international actors in the control of violence may result in a systemic dilemma, by questioning the ability of local actors to perform security functions – such as upholding order and the rule of law – functions they would carry out under normal circumstances. Moreover, international peacekeepers have to be strong enough and willing to deal effectively with violence. However, in their pursuit of violence makers, measures are sometimes taken that undermine the democratization effort. Moreover, ‘‘[o]nce PKOs leave the world of neutrality . . . the risk of crime and human rights abuses committed by internationally sanctioned agents increases’’ (Fearon and Laitin 2004: 34). These activities not only undermine the local actors’ belief in the democratization process but may also undercut faith in the international community. A question of whether spoilers using violence should be included in, or excluded from, the democratic process pertains in essence to the horizontal dilemma. There is a risk involved in incorporating actors that do not adhere to democratic principles of non-violence and respect for human rights, because their inclusion might undermine the legitimacy of the democratic process. For instance, in the Democratic Republic of Congo, the transitional government put in place in 2003 is made up of ‘‘former’’ armed groups. These groupings demonstrated a continued reliance on violence as a means to pursue their economic and political interests, with a dire security situation as a consequence, in particularly in the Kivu region. However, there is also the argument suggesting that the more parties that are excluded from a peace agreement, the more likely a return to war (Darby and Mac Ginty 2000; Hampson 1996; Zahar 2003).19 Moreover, violence makers might have been guaranteed a place in power through the peace agreement. Excluding them from that position might ensure a return to war.

19

Recent quantitative research supports the notion that if rebel groups are excluded from a peace settlement, a return to armed conflict is more likely. Moreover, research also suggests that it is generally the excluded groups that engage in post-agreement fighting (Nilsson 2006).

Violence in war-to-democracy transitions

99

Conclusions This chapter points to two central problems in countries emerging from war that are simultaneously undergoing democratization and peacebuilding. On the one hand, measures taken and reforms initiated to promote democracy – such as the holding of elections, reform of media and the security sector – might have the unintended effect of producing conditions that facilitate or encourage polarization and violence. On the other hand, efforts to deal with violence – in particular by using coercive measures and including spoilers in the democratic process – may undermine the transition to democracy. The relationship between violence, democratization, and peacebuilding covers the four dilemmas highlighted in Chapter 2 of this book. While more systematic research is needed, this chapter reflects a first attempt to reach tentative conclusions about the importance of these dilemmas. In particular, the temporal dilemma – a trade-off between short-term and long-term achievements – is strongly related to issues of when reforms for democratization should be instigated. It was argued that elections, media reform, and security sector reform – if carried through prematurely – might inadvertently create conditions conducive to violence. The vertical dilemma – entailing issues over legitimacy versus efficacy – relates to the fundamental necessity to manage violence. Unless violence is controlled, the state institutions responsible for domestic security will lose credibility. If human rights are abused in the pursuit of violence makers, the democratization process may be undermined. The horizontal dilemma comes strongly out in relation to the inclusion versus exclusion of violence makers in the political process. The systemic dilemma concerns the functions performed by international actors, which may have the unintended effect of undercutting the ability of, and trust in, domestic actors and institutions. By merging the literature on peacebuilding and on democratization, this chapter prompts reflection on ways in which democratization processes in war-ravaged states can be designed in order to alleviate or prevent the potentially damaging effects of democratization efforts. Democratization processes have to be accompanied by ambitious efforts to disarm and demobilize armed actors, to prevent opportunists from using violence for economic or political purposes. Convincing arguments have been put forward by scholars for the case that democratic

100

Kristine Ho¨glund

reform has to be delayed until strong institutions have been established that can address and control the hazards of democratization in warshattered societies (Paris 2004). At the same time, a war ending opens up opportunities for democratic reform which may also help legitimize the post-war political order. Ideally, the peacebuilding process and democratization can be carried out in tandem and the question is rather a matter of design than about timing. For instance, peaceful demonstrations are a democratic outlet for frustration and demands for political change. However, to prevent peaceful demonstrations from turning violent, they have to be backed up by adequate security arrangements. In addition, this chapter calls for fresh thinking with regard to ways of dealing with violence. In particular, successful violence management requires strategies that differ from those that are used in stable, democratic societies. The reason is twofold. Firstly, in post-war societies, uncertainty is exacerbated by the fact that the rules of the game in the political sphere are not set but are still under development. Secondly, violence-induced crises become focal points where the spirit and substance of peace and democracy are put to test. Thus, strategies adopted by the international community and the local actors to prevent and manage violence will have to take into account the specific context in which it emerges. Or in Jeong’s words: ‘‘[a]n assessment of the extent and nature of local violence is necessary when intervention strategies are being designed’’ (Jeong 2005: 58). When the armed struggle comes to an end, new forms of violence emerge or are reintroduced, as was seen in Kosovo, Sri Lanka, and Northern Ireland. As the regular fighting between rebels and government forces subsided, violence on the streets and communal violence has remained a feature of these countries in the wake of cease-fires. Often a shift from military thinking to policing is needed to tackle violence in post-war societies. This means that local and international actors involved in violence management require training and equipment to use anti-riot and policing techniques. The deployment of international police missions is one response to this need. However, their usefulness shows mixed results and more research is needed on the design and operation of international police missions.20 20

Important exceptions include the edited volumes by Oakley, Dziedzic, and Goldberg (2002) and Holm and Eide (2000). On related issues from a US policy perspective, see Robert M. Perito (2004).

Violence in war-to-democracy transitions

101

One of the most important distinctions to consider is whether the violence is committed for political or criminal purposes. Criminal violence has to be countered with traditional police work firmly anchored in the rule of law. Thus, if the police are to be successful in their effort to combat crime, this has to be accompanied by a functioning judicial system. With regard to political violence, inclusion is one way of dealing with it. Through participation in the political arena, even the more extremist parties have a vested interest in upholding the political system. However, as discussed in the chapter, inclusion may also undermine faith in the new political system. Negotiations at the local level can be important to deal with violence efficiently and can be used as an alternative or complement to the use of force. Negotiations can be institutionalized through the formation of peace committees or monitoring missions as in the cases of South Africa and Sri Lanka. The Sri Lanka Monitoring Mission (SLMM) – a civil mission established to oversee the 2002 cease-fire between the Liberation Tigers of Tamil Eelam (LTTE) and the Sri Lankan Government – was initially instrumental in solving disputes regarding violence at the local level. However, in 2006 Sri Lanka de facto returned to a state of war. A final comment relates to a more fundamental dilemma pertaining to war-to-democracy transitions. Political violence is frequently a response to too little democracy. A transition period to establish a new political order is therefore needed to overcome violence. While violence during the transition process in some cases may threaten to derail the transition, in other cases the transition survives all attempts aimed at preventing it from proceeding. Thus, sometimes the two processes seem to be closely linked and can instigate degenerative spirals, with more violence sending countries down a trajectory of instability and crumbling democracy. For this reason, a key question for future research is to understand when a transition can cope with violence, and when violence will continue well beyond the transition and threaten long-term stability.

PART III

The political process

4

Power sharing: former enemies in joint government ANNA K. JARSTAD

More than a decade after the peace agreement for Bosnia and Herzegovina was signed in Dayton, Ohio, the country remains divided. The peace agreement served its main aim – to end the 1992–1995 conflict – but its power-sharing provisions are now seen as an obstacle to peacebuilding and democratization. The political representatives of the three main ethnic groups have not managed to govern jointly. Instead, the international community’s High Representative has imposed laws and removed several politicians from office. For the sake of efficacy, these measures have substituted normal procedures for accountability and democratic legitimacy. The extensive external control has undermined local ownership. Furthermore, the political system provides no incentives to form cross-ethnic and moderate political parties. The international community has therefore suggested a need for constitutional changes. The pertinent issues include the tripartite presidency and decentralization to the two entities, the Bosniak-Croat Federation and Republika Srpska. Revisions toward an integrated, centralized non-ethnic parliamentary democracy with a single president have been proposed. However, these changes upset the power balance that has regulated the conflict. Removing guaranteed governmental position threatens the power of the present elite. Hence, Serbian leaders want to maintain Republika Srpska, and many Croats believe that they should also get their own entity. In April 2006 the parliament voted against constitutional changes. The case raises pertinent questions about governance in post-war societies. In what ways does power sharing facilitate, or obstruct, the transition of divided societies toward a secure democracy? In recent years peace agreements have stipulated the inclusion of warring parties in government in countries such as Burundi, the Democratic Republic of Congo, Liberia, Sudan, and Nepal. Previous research provides explanations for why parties find power sharing acceptable. Within the field of conflict management, it has been demonstrated 105

106

Anna K. Jarstad

that combatants are much more likely to sign an agreement if it reduces uncertainty by the inclusion of guaranteed positions in the future government. Scholars on democratic theory have advocated power sharing as a school in democracy, socializing opponents into compromises and moderation, and as the only option for democratic governance for many divided societies. This leads us to expect that power sharing contributes to peace as well as democracy. At the same time, case studies illustrate that power sharing may be a source of instability, ineffective governance, and violent conflict (e.g., Rothchild 2005b). How can this puzzle be understood? My interpretation is that previous research on power sharing has underestimated the long-term negative consequences of power sharing on both democracy and peace. To date, scholars of conflict management have perceived of power sharing as a tool for short-term peace. It has been suggested that because contending parties cannot trust that the other side will uphold an agreement on democratic governance after a winner-take-all election, parties to a peace deal are likely to demand some form of power sharing (Walter 1999). However, after a peace deal is signed, power sharing may have negative effects on democratization as well as peacebuilding. Scholars of democratic theory, on the other hand, suggest that power sharing promotes democracy, while assuming that it also produces peace. Advocates of power sharing as a means for improving democracy suggest that government by many is more legitimate than mere majority rule (Lijphart 1999). Iterative cooperation in a grand coalition is expected to promote tolerance, increased trust, the development of a democratic culture and societal peace (Lijphart 1994). However, this field of research has failed to take into account how the special conditions of war-torn societies affect the functioning of power sharing. The lack of integration of the two discourses on power sharing has led to a neglect of a deeper understanding of potential dilemmas for simultaneous and long-term democratization and peacebuilding.1 1

One of the few studies that indeed discusses the dilemmas of power sharing in relation to both democratization and peace is Roeder and Rothchild 2005. However, in that volume ‘‘power sharing’’ refers to a variety of arrangements for sharing and dividing power between groups, including both formal provisions for political, military, economic, and territorial arrangements as well as informal practices of cross-communal cooperation. In the present chapter, I focus on the dilemmas related to peace accords stipulating guaranteed positions in the future

Power sharing

107

I argue that power sharing in war-torn societies entails choices between promotions of peace or democracy. Four mechanisms of power sharing condition the prospects for peace: inclusion of warring parties, intragroup contestation, international dependence, and the leveling of power relations. At the same time, power sharing can affect democratization negatively in at least four ways: by exclusion of moderate elites, by lack of popular support, by external intervention preventing local ownership of the political process, and by freezing ethnic division by group representation. Each of these eight attributes of power-sharing accords corresponds to one of the four dilemmas, outlined in Chapter 1 of the book. Firstly, expanding the political center by ensuring inclusion in parliament for key representatives of society may facilitate both peacebuilding and long-term democratization. However, many power-sharing arrangements after civil war are based primarily on inclusion of representatives of the former warring parties, which do not always give up their violent tactics despite a peace agreement. While key figures in the conflict are often included in such power-sharing arrangements, moderate actors are often excluded from a share of power. These aspects speak to the horizontal dilemma of which parties to include – and exclude – to make power sharing conducive to both peace and democracy. Secondly, the focus on a small group of elites activates the vertical dilemma of efficacy at the cost of legitimacy. This may give rise to violent splinter groups, as well as weak support for the power-sharing system among the civilian population, which may affect the prospects for peace and democracy negatively. Thirdly, power-sharing accords may entail a systemic dilemma of local versus international ownership and responsibility to uphold the peace agreement. In some cases, a third-party guarantor is deemed necessary for the functioning of a power-sharing government and to prevent a return to war. At the same time, such intervention may undermine local ownership of the political institutions. Fourthly, power sharing often implies a leveling of power between contending groups. Over time, there is a risk that the agreement does not reflect the perceived power relations and some actors may even challenge it by force. Furthermore, group representation can be an legislative and executive branch of government, i.e., a type of formal political power sharing.

108

Anna K. Jarstad

obstacle for individual choice by restricting the possibility to change political orientation over time, or voting for candidates of different ethnic affiliations. These aspects hamper the processes of democratization and peacebuilding and give rise to the temporal dilemma of power sharing. Below, I begin by discussing the concept of power sharing. Then, the effects of power sharing on the conflict dynamic as well as on democratization are discussed separately. The chapter concludes by discussing the dilemmas of power sharing and trade-offs between democratization and peacebuilding.

The meaning of power sharing Conceptual confusion has hampered research on power sharing. Two, actually separate, strands of research use the term ‘‘power sharing,’’ often without recognizing the differences in terms of democracy and conflict management. However, power sharing stipulated in part of the conflict-management literature differs from power sharing in accordance with democratic theory. Because of different definitions, there is little overlap between the characteristics, the cases and the mechanisms of these two concepts of power sharing. One strand of research stems from the field of conflict management. In this discourse the main function of power sharing is to end violence. To lay down arms, parties often demand guaranteed positions in the future government.2 Barbara Walter explains the function of power sharing as a mechanism for solving the commitment problem in a context of severe distrust and vulnerability. She outlines several reasons why combatants have difficulty in credibly committing to a peace agreement. The concessions involved in a peace deal increase the parties’ vulnerability and limit their ability to enforce the treaty’s other terms. Once the parties have laid down their arms and surrendered territory, they become vulnerable to surprise attacks. Also, the parties cannot trust each other to implement the agreement. Therefore, provisions for democratic governance do not by themselves settle the issue of uncertainty. With less control over resources, parties may fear that the other 2

Alternative demands often concern some form of transfer of power to smaller units, such as decentralization, federalism, autonomy, separate legal systems, etc. Division of power based on territorial units is sometimes called ‘‘territorial power sharing’’ (see, e.g., Hoddie and Hartzell 2005). In this volume, however, power sharing refers to a specific type of political pact, namely power-sharing accords.

Power sharing

109

party will form a one-party state if it wins the election, and that those in power would exploit minority and opposition groups. In the absence of a strong civic culture, the population may not reject parties advocating authoritarian control to foster order and economic advancement. For this reason, the losers of the first post-war elections cannot count on another opportunity to gain power. Under such circumstances, Walter (1999) argues, each side is likely to demand some form of power sharing as the price for peace. She has demonstrated that warring parties are 38 percent more likely to sign an agreement if it includes guaranteed positions in the future government (Walter 2002: 80). In this discourse, the main function of power sharing is to end violence, not to build democracy. Here, power sharing denotes all types of sharing and dividing power between former foes, with less emphasis on democratic representation and elections (Hartzell and Hoddie 2003; Rothchild 2005a; Spears 2000; Walter 1999). Ian S. Spears writes that power sharing ‘‘can be compatible with democracy’’ (2000: 105, my italics). However, power sharing is sometimes constructed as an alternative to competitive elections (Spears 2000: 108). Among quantitative research on war endings, Matthew Hoddie and Caroline A. Hartzell’s works are well cited. They suggest that of the total of thirty-eight civil wars ended by negotiated settlement between 1945 and 1998, only one did not include provisions for power sharing. Furthermore, they conclude that the more power sharing, the higher the likelihood that peace will endure (Hartzell and Hoddie 2003: 319). However, in contrast to the most common notion of power sharing, guaranteed positions in the government are not a necessary component of this definition. Rather, power sharing denotes any type of institution dividing or sharing political, economic, territorial, and military power. Military power sharing refers, for example, to provisions allowing antagonists to remain armed or retain their own armed forces. In order to qualify as political power sharing, it is sufficient that the peace deal provides for electoral proportional representation or administrative proportional representation (e.g., appointment of representatives of warring groups to courts).3 This means that cases without political pacts are included as cases of power sharing (Hartzell and Hoddie 2003; Hoddie and Hartzell 2005). Consequently, many pivotal 3

Bumba Mukherjee (2006) also regards proportional representation as a sufficient criterion for power sharing.

110

Anna K. Jarstad

issues, such as the impact of guaranteed positions in government on peace and democracy, or factors associated with the creation of political pacts after civil war, cannot be assessed based on their findings.4 To the other strand of research, which takes its starting point in Arend Lijphart’s theory on consociationalism, political pacts are central. Here the main issue is how to make democracy function in societies divided along ethnic lines. This means that the theory of consociationalism primarily addresses conflicts that are clearly ethnic. Lijphart (1968) coined the concept of consociational democracy in 1968 to denote an institutionalized form of democratic conflict management for divided societies. Majoritarian electoral systems are inapt, he writes, since they presuppose shifting majorities in parliament and fairly similar policies of major parties in order not to exclude the other parties’ interests. However, in divided societies, the political parties diverge to a great extent. Where people vote along ethnic lines, political parties representing ethnic minorities have no chance of ever forming a majority, and shifting majorities in parliament are therefore unlikely. Under such conditions, Lijphart holds that majoritarian rule is not only undemocratic, but also dangerous and risks resulting in civil strife (1999: 31–33). For countries such as Lebanon in 1985, Lijphart writes, ‘‘the choice is not between consociational and majoritarian democracy, but between consociational democracy and no democracy at all’’ (1985: 13). To avoid partition or majoritarian dictatorship, Lijphart advocates a group-based form of democracy, which he calls consociational democracy. The two main components are grand coalition, implying that all rival groups should be included in government and autonomy for each ethnic segment in all matters not of common concern. Two additional features are mutual veto rights and proportionality in political representation, civil service appointments, and the allocation of public funds (Lijphart 1993: 188–189). Donald L. Horowitz, Timothy D. Sisk, Benjamin Reilly, Andrew Reynolds, and others have recognized that there are also other forms of democratic power sharing (e.g., Horowitz 1985; Reilly 2001; 4

Power-sharing arrangements defined as political pacts remain largely unexplored in quantitative research. Walter’s (2002) study is an exception. However, this study is limited in many regards as it does not cover the entire post-Cold War period, excludes minor conflicts, and only includes those peace deals where all or the main warring parties are signatories.

Power sharing

111

Reynolds 2002; Sisk 1996). The integrative approach to power sharing eschews ethnic groups as the organizing principle for democracy.5 Instead, incentives for moderation and cooperation across ethnic divides are engineered. These constitutional designs can include majoritarian as well as non-majoritarian forms of electoral design. However, proponents of the consociational as well as the integrative approach agree on the baseline that some form of joint rule is the only option for democratic governance in divided societies. Thus, in the conflict-management discourse, power sharing is seen as a mechanism to manage the uncertainty involved in a peace process – if need be, as a substitute for elections – while research based on democratic theory treats power sharing as a mechanism to foster moderation and to improve the quality of democracy. This means that researchers of both schools advocate power sharing for war-shattered societies, albeit for different reasons. However, the lack of integration between the two discourses means that there is limited knowledge of the longterm consequences of power sharing in societies emerging from war. Of particular importance for the issues of democratization and peacebuilding in war-torn societies are provisions for guaranteed positions in government, stipulated in a peace accord. Table 4.1 provides a list of examples of such power-sharing accords concluded in the post-Cold War era.6 A power-sharing accord (PSA) is a political pact between contending parties which formally outlines how power is to be shared in the legislative and/or the executive branch of a future government. It can be noted that power-sharing provisions have been stipulated to manage several types of conflicts. Some power-sharing accords reflect ethnic divisions. For example, the Dayton agreement stipulated power sharing among Serbs, Croats, and Bosniaks in Bosnia and Herzegovina. The Burundi agreement provided for two types of ethnic power sharing: during a transitional period ethnopolitical parties shared power, whereas the permanent constitution fixed the ratio between Hutu and Tutsi representatives. By contrast, other peace accords do not make any 5

6

‘‘Centripetalism’’ denotes a similar approach. Sisk uses this term to refer to electoral engineering (in line with Horowitz) aiming for moderation and compromises by reinforcing ‘‘the centre of a deeply divided political spectrum’’ (Sisk 1995: 19). Reilly uses a wider definition including electoral incentives, arenas for bargaining and centrist, aggregative political parties (2001: 11). Less recent examples of power-sharing accords include the London Accords 1959 (Cyprus) and the Lancaster House Settlement 1979 (Zimbabwe).

112

Anna K. Jarstad

Table 4.1 Examples of recent power-sharing accords Conflict location

Name and year of power-sharing accord

Afghanistan Angola Bangladesh Bosnia and Herzegovina Burundi Cambodia Chad Colombia Comoros DRC Guinea-Bissau Ivory Coast Kosovo Lebanon Liberia Mali Mexico Nepal Rwanda Sierra Leone South Africa Sudan

Mahipar Agreement 1996 Lusaka Protocol 1994 Chittagong Hill Tracts Peace Accord 1997 Dayton Peace Agreement 1995 Arusha Agreement 2000 Paris Agreement 1991 Tripoli Accord 2006 Acuerdo Final Gobierno Nacional 1991 Agreement on the Transitional Arrangements 2003 Pretoria Agreement 20027 Abuja Peace Agreement 1998 Linas-Marcoussis Agreement 2003 Rambouillet Agreement 1999 Ta’if Charter 1989 Accra Comprehensive Peace Agreement 2003 The Pacte National 1992 San Andre´s Accords 1996 Comprehensive Peace Agreement 2006 Arusha Accords 1993 Lome´ Accord 1999 Interim Constitution Agreement 1993 Naivasha Comprehensive Peace Agreement 2005 Darfur Peace Agreement 2006 Moscow Declaration 1997 Yumbe Peace Agreement 2002

Tajikistan Uganda

Source: Uppsala Conflict Data Program, March 2007. Selection made with assistance of Ralph Sundberg.

reference to ethnic groups, for example the peace accords in Angola, Cambodia, the Ivory Coast, Liberia, and Sierra Leone. By integrating the two discourses on power sharing, ethnic as well as non-ethnic PSAs can be taken into account. The long-term consequences of many of these PSAs are yet to be seen. However, the theoretical arguments outlined below suggest that the main functions of power-sharing 7

The formal title is the Global and Inclusive Agreement on the Transition in the Democratic Republic of the Congo.

Power sharing

113

agreements – inclusion, elite-oriented regulations, international guarantee, and the leveling of power between groups – can have negative consequences for peacebuilding as well as on democratization.

Power sharing: effects on conflict dynamics The logic of power sharing as a way to manage violent conflicts can be understood in terms of a basic two-player game. In brief, conflict is then seen as a result of a situation where one party strives for total political control whereas the other demands partition. If one of the parties has its way, it will gain a hegemonic position. If the other party has its way, the game will result in partition. Under such circumstances, or when both parties strive for total political control, the only solution to manage conflict is joint rule (Jarstad 2001). For this reason, power sharing is attractive to peace negotiators. At the same time, there are several reasons why the warring parties are reluctant to end conflicts by power sharing and why PSAs fail.8 A first reason concerns the issues of insecurity and trust. One interpretation of Lijphart’s theory is that it assumes that the provisions of consociationalism change the value of cooperation for the parties involved. In a similar vein, Walter suggests that power sharing solves the commitment problem. However, in many war-shattered societies, power sharing might be inadequate to mitigate the parties’ vulnerability during the peace process.9 Under such circumstances, the parties cannot trust each other to uphold an agreement on power sharing. Secondly, power sharing entails great concessions for many parties. Per definition, power sharing levels the power relations between contending groups. A strong party is likely to be reluctant to accept a peace deal which renders all parties equal powers. In the choice between 8

9

A recent data collection, the IMPACT (Implementation of Pacts) data set, demonstrates that in the conflicts where an agreement of political power sharing was reached and implemented within five years, nearly half of all cases saw the recurrence of warfare within this time period (Jarstad and Sundberg 2007). In game-theoretical terms, this is the case when the rank orders of the parties’ preferences result in a Prisoner’s Dilemma. Consociationalism provides no mechanism to solve the Prisoner’s Dilemma. Therefore, as Walter demonstrates, a combination of third-party security guarantees and power-sharing pacts are critical to the peaceful resolution of civil wars (Walter 2002). For a review of the role of third parties in relation to power sharing, see the introduction by John MacGarry and Brendan O’Leary 2004).

114

Anna K. Jarstad

democratic options, the larger party is likely to consider a majoritarian electoral system more democratic and legitimate than a power-sharing system, as the latter gives small groups too a share of power, and in several cases even veto rights. But a minority group may also be reluctant to sign a power-sharing deal. Territorial control of part of a state may be preferable to the inclusion in a power-sharing government. This is particularly true for regions with valuable resources such as oil or diamonds.10 Furthermore, inclusion in central government implies accountability also for decisions that go against your own group. Thus, leaders may fear that if they accept a power-sharing deal, their followers would see them as traitors.11 Thirdly, even in cases where war ends in a power-sharing accord, the risk of a resumption of violence remains. This is not always a consequence of power-sharing provisions, but of conditions which obstruct any form of peace deal. For example, lingering violence may prevent power sharing from taking root. This was the case in Angola 1994, where violence continued even after the Lusaka agreement was signed. However, it would be incorrect to argue that power sharing led to the resumption of war in 1998, as the agreement was not fully implemented until many years later. In Sierra Leone in 1999, on the other hand, it was violence that caused power sharing to break down. In this case, power sharing was a condition for the peace accord and the subsequent international peace operation, which eventually contributed to the marginalization of the Revolutionary United Front (RUF).12 10

11

12

In recent years there has been an upsurge of research on what has become known as the greed and grievance motives for conflict (see, e.g., Ballentine and Sherman 2003; Berdal and Malone 2000; Collier et al. 2003). Based on several case studies, Karen Ballentine and Jake Sherman demonstrate that the combatants’ opportunities for mobilization and/or incentives for self-enrichment were not the primary causes of these conflicts. Nevertheless, access to natural and financial resources does in some cases create serious impediments to conflict resolution. ‘‘Lootable’’ resources such as diamonds and illegal narcotics have been found to prolong non-separatist insurgencies, while ‘‘unlootable’’ resources, such as oil, gas, and deep-shaft mineral deposits are associated with separatist conflicts (Ballentine and Sherman 2003). Hence, there is reason to believe that natural resources impact on the parties’ preferences, the type of power sharing agreed on, and the success of power sharing in terms of conflict management. This argument is, for example, illustrated by the case of South Africa, where the National Party withdrew from the power-sharing agreement in 1996 (Sisk and Stefes 2005). See the chapter by So¨derberg Kovacs on the failed transformation of RUF from warring group to political party in this volume.

Power sharing

115

Fourthly, power sharing affects the intergroup relations in war-torn societies. Below, I nuance the analysis of how power sharing affects the conflict dynamics. I discuss the impact of four features of power-sharing accords: inclusion of warring parties, intra-group contestation, international dependence, and the leveling of power relations between groups. Although these four mechanisms can have positive effects on conflict dynamics, I propose that under certain conditions they can also affect intergroup relations negatively.

Inclusion of warring parties The horizontal dilemma of power sharing is triggered by the choice of whom to include in a power-sharing arrangement. In many conflicts, power sharing is the key for reaching a peace deal at all. A peace deal is per definition an agreement between contending actors. Negotiated settlements result from compromises between warring parties, who often maintain the option of resumption to war should the costs of peace outweigh the prospective gains from future outcomes. For warring groups who see little chance of winning the war on the battlefield, a stake in the government, stipulated by a PSA, may work as an incentive to lay down arms. The African National Congress (ANC) in South Africa is only one example of a former rebel group that has agreed to give up violent tactics in exchange for a transition to democracy that featured power sharing, at least for a transitional period. The armed struggle against the apartheid regime in South Africa was ended by a series of pacts that included ‘‘sunset clauses’’ by which power sharing would eventually give way to majoritarian democracy. However the transition period was far from peaceful. In the transitional period from 1990–1994, large-scale violence raged throughout the country. Although political violence ended in most parts of the country when the ANC-dominated Government of National Unity came to power in 1994, the KwaZuluNatal province continued to be marred by political violence as local-level power struggles prevailed. The interim constitution stipulated transitional power sharing among all political parties that won substantial votes in elections, but the erstwhile governing party, the National Party, voluntarily withdrew from the Government of National Unity in May 1996. Nevertheless, the reliance on sunset clauses provided for a smooth end to formal power sharing and the permanent constitution

116

Anna K. Jarstad

was adopted in 1996 that does not feature formal guarantees of power sharing. Informal power sharing continues to be vital for moderation and compromise. Thus, power sharing is seen as the key to the comparatively successful transition from violent political conflict to peace in South Africa (Sisk and Stefes 2005). In Northern Ireland lingering violence and paramilitary organizations have been the main obstacles for power sharing stipulated by the 1998 Good Friday agreement. The political party Sinn Fein has never officially recognized its military wing, the Irish Republican Army (IRA), but has widely been held accountable for its violent tactics. After the agreement was signed, and Sinn Fein became part of the provincial government, a splinter group called Real IRA has been responsible for a number of violent incidents, most notably a bomb in Omagh, which killed twentyeight people (Belloni and Deane 2005). The IRA has now officially declared an end to its armed campaign and Sinn Fein, led by Gerry Adams, and the Democratic Unionist Party, headed by Ian Paisley, entered a power-sharing government in May 2007. However, exclusion of violence makers can also trigger violence. An agreement which stipulates guaranteed positions in government only for some of the warring groups provides a strong incentive for excluded groups to resort to violence, with the purpose of gaining leverage for future negotiations or winning militarily.13 The number of sufficient signatories is a particular problem in conflicts with many warring groups where there are reasons to fear that groups excluded from power sharing will use violence to fight their way into the powersharing deal. The 2000 Burundi PSA is a case of partial inclusion. After the ‘‘Arusha Agreement for Peace and Reconciliation in Burundi’’ was signed on August 28, 2000, four non-signatories continued to fight. For several years the rebel movement the National Council for the Defence of Democracy-Forces for the Defence of Democracy (CNDD-FDD) refused to participate in the peace talks. Under strong pressure from the mediation team, led by Nelson Mandela, CNDD-FDD eventually laid down their arms. The Pretoria Protocol of October 8, 2003 allotted them, among other provisions for power sharing, four seats in the transitional cabinet and fifteen seats in the National Assembly. The Global Peace 13

On the risk that exclusion leads to violence, see Darby and Mac Ginty 2000; Hampson 1996; Nilsson 2006; Zahar 2003.

Power sharing

117

Accords of November 16, 2003 stipulated CNDD-FDD should get three ministry positions and the Minister of State (Bentley and Southall 2005; Rothchild 2005b).14 The last remaining rebel group, Palipehutu–FNL, was included in peace negotiations in 2006. Another case of partial inclusion is the Sudan 2005 accord which was designed to end the conflict in the southern part of Sudan. The accord did not aim at regulating the conflicts in other parts of Sudan. However, it was incorrectly assumed that other groups in the country were represented either by the government or the Sudanese People’s Liberation Movement (SPLM). The 2005 accord unintentionally exacerbated the conflict in the Darfur region and increased the risk for conflict also in the eastern part of the country by excluding these regions from the political and economic power sharing. The Darfur crisis erupted when the peace negotiations concerning southern Sudan progressed. The peace talks – and the subsequent accord – included regulations concerning allocation of valuable resources. In Darfur, as well as in other parts of Sudan, there were fears of exclusion from the agreement on wealth sharing, which was an important aspect of the power-sharing deal. Also the 2006 Darfur agreement failed to include all warring parties: Only one rebel group faction signed the accord with the government, and violence in that region did not end. In the eastern part of Sudan, the Beja Congress demands a fair share of power and national wealth, similar to the agreement made for the southern part of Sudan. If the issue is not settled, there is a risk of violence also in this region (ICG 2006a; UCDP 2006).15 As these cases demonstrate, it is often difficult to settle all interlinked conflicts at once. Inclusion of warring parties in a power-sharing arrangement does not always end violence. At the same time, concessions to some warring parties can provide incentives for other parties to use violence in the pursuit of a share of power. When this is the case, inclusion of warring parties in a power-sharing arrangement may at best be a short-term solution to violence. In the worst case, inclusion of warring parties can escalate violence and give rise to new conflicts. 14

15

The election on August 19, 2005 rendered CNDD-FDD a massive victory for the former rebel movement (ICG 2005a). My thanks to Ken Menkhaus for comments on my interpretations of these events.

118

Anna K. Jarstad

Intra-group contestation Power sharing changes the dynamic of political contestation by what I call the ‘‘eclipse’’ mechanism. By fixing the ratio of government positions for each contending group, the primary conflict is regulated – or at least concealed. When the limelight is removed from the main incompatibilities, there is a possibility to focus on other tangible political issues that cut across the former conflictual lines. Ideally, this provides space for moderate factions in support of the peace process and increases vertical legitimacy. In some cases, it takes time for power sharing to have this effect. In Angola, the power-sharing provisions of the 1994 Lusaka Protocol contributed to an intraparty split of the National Union for the Total Independence of Angola (UNITA).16 When Jonas Savimbi, the leader of UNITA returned to war, many other UNITA members saw power sharing as more attractive and refrained from using violence.17 Partial implementation of the power-sharing provisions began by the formation of the Government of Unity and National Reconciliation in April 1997, when some UNITA deputies took up their seats in parliament (Hodges 2001: 58). However, Savimbi never assumed the position of minister, nor did he take up one of the two positions as vice-president as stipulated in a constitutional amendment (Ohlson 1998: 79, footnote 76). The power-sharing provisions of the 1994 Lusaka Protocol were not fully implemented until large sections of UNITA had been militarily defeated and after Savimbi was killed in February 2002. A subsequent cease-fire agreement reinforced the provisions of the Lusaka accord,

16

17

Another case where power sharing was used to pacify warring groups is New Zealand. The process of substituting warfare with the incorporation of Maori in the political process took some thirty years. In 1867 the British Crown offered Maori tribes reserved seats in the New Zealand parliament on condition that they agreed to lay down their weapons. Although several tribes continued their warfare, eventually all Maori were pacified. Thus, also in this case power sharing did eventually work as an incentive for moderation (Jarstad 2001). It was not power sharing per se that caused the peace agreement in Angola to break down, but rather the failure to implement the agreement. A more convincing explanation of why the conflict lingered on despite the peace accord, suggested by several authors, is UNITA’s control of the lucrative diamond mines (Hodges 2001; Zahar 2005a). Thus, for Savimbi the spoils of war outweighed any potential gains of political influence via power sharing.

Power sharing

119

and UNITA assumed four ministerial posts and began the transformation into a political party. It could therefore be suggested that the power-sharing provisions of the 1994 Lusaka Protocol have contributed to peace in Angola. However, vertical legitimacy is still weak as the present power-sharing arrangement between UNITA and the Popular Movement for the Liberation of Angola (MPLA) is not based on open contestation. In the case of Angola, the moderate faction within UNITA eventually gained ground. However, intra-group contestation does not always lead to moderation. Pierre M. Atlas and Roy Licklider have suggested that societies in civil war are often driven by a whole variety of conflicts. When the primary conflict is handled, other conflicts may come to the foreground (Atlas and Licklider 1999). For this reason, intragroup competition can also have the negative effect of resulting in extremist splinter groups. In severe cases, this can lead to a resumption of violence.18 This was the case in Rwanda 1993, where the inclusive Arusha Accords could not prevent the genocide. An extremist faction of the Hutu government feared exclusion from power sharing stipulated in the agreement and formed a splinter group, which aimed to eliminate all Hutu moderates and Tutsi (Rothchild 2005b).

International dependence Several PSAs are mediated and guaranteed by international actors. Such third-party dependence activates the systemic dilemma. In Cyprus, the constitution of 1960 was a result of the negotiations between Great Britain, Turkey, Greece, a Greek Cypriot delegation headed by Makarios, and a Turkish Cypriot delegation headed by Kutchuk. It allotted the Turkish Cypriots more than their proportional share of seats in parliament to be elected separately by the Turkish Cypriot electorate, veto rights in vital matters, and a Turkish vice-president. The consociational arrangement was a compromise between the Turkish Cypriot demand for partition (taksim) and the Greek Cypriot desire for unification of Cyprus with Greece (enosis), or at least Greek hegemonic control of political power. 18

One example of this took place in Northern Ireland, where a splinter group from the IRA planted a bomb in Omagh in August 1998, after the peace deal was signed.

120

Anna K. Jarstad

However, shortly after the formation of joint government, Makarios felt confident of international support for removing provisions enhancing Turkish Cypriot political power. In December 1963, President Makarios unilaterally announced revision of the constitution, which in fact eroded its consociational character. The reason for Makarios’ defection from the agreement was that he believed that his bargaining position had improved. As a consequence, ethnic conflict erupted all over the island and since then Turkish Cypriots have no representation in the government. Nevertheless, the United Nations decided to recognize the now all Greek Cypriot parliament as the only legitimate government of Cyprus. Still today, when a parallel Turkish Cypriot administration in the north has assumed the features of statehood, the government controlling the southern part of the island is the only internationally recognized Cypriot government (Jarstad 2001). In this case, the promise of international guarantors to intervene to restore constitutional order was not followed through and power sharing ended. In cases such as Bosnia and Herzegovina and Kosovo, the international community has decided to stay on to ensure multiethnic power-sharing governance. It is yet to be seen whether these amendments will provide for sustainable peace also without international security guarantees.

Leveling power relations Power sharing entails the leveling of power relations between contending groups. Such ensured inclusion in government contributes to conflict management by reducing the uncertainty of elections. However, changing power relations between the target groups in the powersharing arrangement activate the temporal dilemma. A PSA tends to reflect the power relations at the time of negotiations. If power relations shift, there is a risk of defection from the agreement. Power relations may shift as a consequence of changing military capabilities or shifting demographic proportions. Lebanon is an illustrative case in this regard. Lebanon has experienced four power-sharing systems since 1861. As the power balance has changed, each system has created different winners and losers amongst the seventeen religious groups in the country. The powersharing system introduced in 1943 lasted until 1975. As the Muslim share of the population in Lebanon increased, many Muslims saw the

Power sharing

121

fixed Muslim–Christian ratio in parliament as unjust. Dissatisfaction with the power-sharing balance was a major cause of the fifteen-yearlong civil war that broke out in 1975. The 1989 Ta’if Charter of National Reconciliation contributed to ending the conflict. By this agreement the constitution was amended to provide for an equal number of Christian and Muslim members instead of the previous six to five ratio in the National Assembly (Zahar 2005b).19 However, the distribution of seats and the relative balance between the various groups is again an issue of discussion. The Sunni–Shiite polarization constitutes the most serious threat to the old equilibrium. In May 2000, Israel ended the occupation of southern Lebanon after a decade of Hizbollah attacks. Hizbollah’s military achievements were rewarded during the democratic elections in 2005, when Hizbollah won 14 out of 128 seats in parliament. Hizbollah have demanded a larger share of influence in both the legislative and the executive branch of government to adjust the powersharing balance to reflect the increased Shiite population (ICG 2006b). To avoid a situation where power sharing over time gives rise to recurrence of conflict, several peace accords include provisions for abolishment of power sharing after a transitional period. The so-called sunset clause of the transitional power-sharing arrangement in South Africa is often cited as a model for designing power sharing after war (see, e.g., Sisk and Stefes 2005).20 A less known case is the transitional power sharing in Sierra Leone. The Lome´ Accord of July 7, 1999 provided for guaranteed inclusion of the rebel movement Revolutionary United Front (RUF). The post of vice-president was allotted to RUF’s leader Foday Sankoh, along with four cabinet positions and four deputy ministerial positions (Francis 2000: 363). UN troops were deployed to the country and in spite of repeated outbreaks of violence by the RUF during the implementation period, by January 2002 the disarmament and 19

20

The prior consociational system held for thirty years and during this period it managed to prevent full-scale civil war and, according to Arend Lijphart (1977: 150), also established a remarkable record of democratic stability. Colombia 1958–1974 is another case where power sharing was agreed to last for a specified period of time. However, during the time of power sharing there were no means for other actors to gain government positions. During this period, guerillas that still operate, such as the Revolutionary Armed Forces of Columbia (FARC), were formed. It has been suggested that the political blockage of political movements outside of the National Front regime was one of the major explanations for why such non-parliamentary groups were formed (see, e.g., Hartlyn 1998: 191 ff.).

122

Anna K. Jarstad

demobilization was completed and the war declared over. The transitional period ended with the holding of elections in May 2002, in which the political arm of RUF received only 1.7 percent of the votes. Without any guaranteed seats in government, no domestic popular support, and with a significantly diminished battlefield capacity, the RUF soon thereafter disappeared as a significant military or political force in Sierra Leone (ICG 2002; So¨derberg Kovacs 2007). In this case, transitional power-sharing provisions proved to be sufficient for the rebels to sign the accord and thereby provided for peace. However, in other cases, the abolishment of power sharing is a source of conflict. In some cases the actors demand permanent regulations regarding government positions, and it may therefore not be possible to convince the warring parties to agree to temporary power sharing. Also when the parties have agreed to end power sharing after a transitional period, it may be difficult to abolish it. After a period of transitional power sharing in Cambodia, stipulated in the Paris Agreement in October 1991, elections were held in May 1993. The Cambodian People’s Party (CPP) under Hun Sen refused to accept that they lost the elections. A coalition government was formed between CPP and the winning party FUNCINPEC (National United Front for an Independent, Neutral, Peaceful and Cooperative Cambodia) with a dual prime minister arrangement. This power-sharing arrangement had no basis in either the 1991 peace agreement or the constitution of 1993 (Brown 1998: 90–100). In addition, the ad hoc nature of the power-sharing arrangements probably contributed to the renewed outbreak of violence between the parties in 1997 and the attempt by the CPP to oust FUNCINPEC from power in a coup in the run-up to the next elections. Nevertheless, the CPP–FUNCINPEC coalition was revived after the 1998 elections, due to the two-thirds constitutional requirement for government formation. The 2003 elections again led to a stand-off between the two major parties (Reilly 2007). As these cases indicate, power sharing has effects on the prospects of peace as well as democracy. The consequences of power sharing regarding issues related to democratization are discussed more thoroughly below.

Power sharing and democratization in war-torn societies In terms of the quality of democracy, Lijphart has demonstrated that non-majoritarian democracy outperforms majoritarian democracies

Power sharing

123

on indicators of macroeconomic variables, gender issues, voter participation, welfare policies, environmental protection, criminal justice, and economic aid to developing countries (1999).21 Lijphart holds that the majority is dependent on the minority’s allegiance to the regime. He suggests that broad participation, bargaining, and compromise are essential for legitimate decision making. Therefore, according to Lijphart, the principle of maximum inclusion is more democratic than majority rule (1999: 2, 32–33). In addition, Lijphart expects consociationalism to promote compromise and conciliation (1994: 12). Lijphart is not clear on exactly how this transformation comes about, but interdependence and recursive negotiations seem to be key components in this process. One interpretation of the logic of consociationalism suggests that recognition of ethnic demands is expected to increase the perception of security, which in turn promotes mutual trust and reduces grievances. Iterative cooperation in a grand coalition is expected to promote an understanding of the other party.22 In the long run, the policies of the parties are expected to conform, to avoid exclusion or discrimination of the other group. In this way consociationalism is expected to depoliticize ethnicity and allow development of a common national identity. In other words, consociationalism is expected to produce moderation (Jarstad 2001: 46–48). For this reason, power sharing could ideally work as a catalyst for peaceful cooperation among contending parties after a peace deal is signed. However, Lijphart’s theory on consociational democracy is mainly based on the analysis of reforms in already democratic societies, such as the Netherlands. The theory fails to analyze how the particular conditions of a war-shattered society affect the democracy-promoting aspects of power sharing. Below, I discuss four types of potential democratic shortcomings that designers of a PSA confront: exclusion of moderates, weak popular support, lack of local ownership, and group representation.

21

22

In this work Lijphart analyzes consensus democracies. However, most of the consociational democracies are included in this category. For the difference between consensus and consociational democracy, see Jarstad (2001). Such repeated rounds of cooperation can be understood by the reciprocal strategy of tit for tat by matching the other player’s move in iterative games (Morrow 1994: 264).

124

Anna K. Jarstad

Exclusion of moderates When consociational democracy develops as a result of reform in countries with prior experience of democracy, it indeed broadens participation in decision making. According to the consociational logic, recurring rounds of bargaining and compromises foster moderation. However, in many war-torn societies, power sharing may have the adverse effect of preventing the development of moderate political parties. This issue of inclusion versus exclusion activates the horizontal dilemma of power sharing. In some peace accords, warring parties decide to share power, and at the same time effectively block other political movements from power. When this is the case, the former warring parties control economic resources as well as media and other channels to mobilize political support. Even if the government eventually opens up the political space, any opposition groups lag behind, due to the lack of resources. This uneven start for parties in a democratization process can have long-term effects on the prospects of multi-party democracy. One possibility to overcome this negative effect of power sharing is to include other actors in the peace negotiations and also in the future government. This was done, for example, in the Ivory Coast, where the 2003 PSA also included the main political parties, in addition to the warring parties. The Liberia 2003 PSA included not only all warring parties to the conflict – the government and the two rebel groups Liberians United for Reconciliation and Democracy (LURD) and the Movement for Democracy in Liberia (MODEL) – and the main political parties, but also representatives from the civil society. Likewise, the 2002 PSA for the Democratic Republic of Congo included government representatives, rebel groups, militias, opposition parties, and different civil society organizations (Nilsson 2006; Nilsson and So¨derberg Kovacs 2005). Although inclusion of a broad spectrum of society in the peace agreement is likely to improve the legitimacy of the peace process, inclusion of warring groups remains an obstacle to democratization. A typical peace deal stipulating power sharing tends to reflect power relations during negotiations, rather than the size of potential electorates. By granting warring parties a stake in government, violence is rewarded. This in itself violates a fundamental principle of democracy. The Sudan 2005 power-sharing accord only includes warring parties in

Power sharing

125

the executive branch of government. Under these conditions, the only access to political power is via the battlefield. Such arrangements lack any form of democratic legitimacy. Also when power-sharing governments are elected, reserved seats for warring parties may undermine democratic legitimacy. There is a risk that the actors continue to use violent tactics to affect the outcome of future elections. Thus, the inclusion of warring parties in government gives rise to several questions regarding legitimacy: Will they conform to democratic norms of non-violence, tolerance and compromise? How representative are these actors?

Popular support Increasing the number of political parties in decision making beyond a mere majority (or even plurality, as is the case in some systems) may enhance vertical legitimacy of the political system. However, this depends to a great extent on the level of support for the included elites among the population. When elites are appointed to positions in a power-sharing government, it is difficult to know how representative they are. But also elective power sharing runs into problems of vertical legitimacy. In divided societies power sharing freezes the conflict lines, and the parties do not need to compete for votes among their former foes. This means that most parties have a narrow support base. In Bosnia and Herzegovina, moderate parties receive little electoral support. Instead, nationalist parties prevail in elections. At the same time, the voter turnout is low – less than 55 percent (OSCE 2006) – and the trend demonstrates a decline during the last years. This reflects weak legitimacy of the political system as a whole. In November 2006, Nepal’s Maoist rebels reached a transitional power-sharing agreement with the civilian government to end the country’s ten-year-long civil war. The constitutional process has thus far concentrated on building elite consensus in closed-door negotiations at the expense of public debate. The challenge to ensure vertical legitimacy remains (ICG 2007a). Even when power sharing and peace negotiations are exclusively elite-driven, there are ways to gain public acceptance of the agreement. One way to ensure that the system as a whole is seen as legitimate, is to make power sharing the object of a referendum. In Sudan the 2005 peace agreement stipulates a referendum on self-determination for the

126

Anna K. Jarstad

southern part of Sudan to be held in early 2011 (ICG 2006a). Thus, during the interim period the North has a chance of making unity attractive to the South. For such a referendum to result in support for the peace agreement, the political leaders have to ‘‘sell’’ the agreement to the public, and explain why concessions are necessary and desirable for sustainable peace. If not, the public is likely to remain polarized. This was the reason why the Greek Cypriot majority voted against UN Secretary-General Kofi Annan’s plan to unite Cyprus under a powersharing accord in April 2004. An example of a referendum that rendered the peace deal legitimate, is the one concerning power sharing in Northern Ireland, where the Belfast Agreement won the support of both communities in 1998. Belloni and Deane suggest that the approval in the referendum is an important factor explaining why the peace deal has not collapsed, despite the many obstacles (2005).

Local ownership Internationally mediated power-sharing agreements activate the systemic dilemma of local versus external ownership of the democratization process. In extreme cases, where power sharing is imposed by third parties, local ownership is weak. In Kosovo power sharing was imposed by UN provisions, without a consensus among the formerly warring parties. In June 1999, UN Security Council Resolution 1244 turned Kosovo into a UN protectorate. It is headed by the Special Representative of the Secretary-General (SRSG), who has the ultimate political authority in Kosovo, including the right to issue legislative acts, and appoint and remove any person in the civil administration. Despite provisions for inclusion in the political bodies of representatives of the Kosovo Albanians, Kosovo Serbs, and the other communities in Kosovo, power sharing has to a very limited extent led to political cooperation across the former conflict lines. This is a result of the lack of ownership of the power-sharing arrangement. The provisions for power sharing were not issued by the central government, but by United Nations Mission in Kosovo (UNMIK) regulation 2001/9, May 15, 2001. These provisions were not put up for general consultation among the wider public. Not even the political parties involved in the UNMIK structures signed any decisions for the creation of the new political institutions. Instead, a small closed group designed the

Power sharing

127

institutions. The only input from minority groups included consultations with a few Serbs. Mass demonstrations suggest that also the plan presented by United Nations Special Envoy Martti Ahtisaari in the beginning of 2007 lacks local ownership (Baldwin 2006; Jarstad 2007). Power sharing in Bosnia and Herzegovina suffers from a similar lack of local ownership. During the negotiations in Dayton 1995 concerning Bosnia and Herzegovina, no representative from the civil society was included.23 Furthermore, the High Representative has used his mandate to remove elected officials from politics on several occasions, thus undermining local ownership as the democratic foundation for the power-sharing government. Both in Kosovo and in Bosnia and Herzegovina, there is a problem of democratic accountability. Real power is not vested in the democratically elected governments, but in the international administration. These structures are not formally accountable to the citizens in these states. Such international intervention risks increasing support for extremism and ethnochauvinism, thus undermining democratization. It has been suggested that power sharing is more likely to last when it is arrived at indigenously (Sisk 1996: 118). Even when a PSA is mediated and guaranteed by international actors, there are ways to enhance the local ownership of the agreement. An example of such an effort is the Pretoria Agreement 2002 (DRC),24 which was mediated by a third party (South Africa) and involved a broad spectrum of society. The agreement was reached after discussion under the so-called InterCongolese dialogue, which brought together the government, civil society groups, political opposition groups, and rebel movements (UCDP 2005).

Group representation It has been questioned whether power sharing at all can be democratic. One reason for this is that power sharing emphasizes representation of certain target groups, rather than representation of ideas and ideologies.25 All power-sharing systems have to settle the difficult issues of 23 24

25

See the chapter by Belloni on civil society in war-torn countries in this volume. The formal title of the agreement is the ‘‘Global and Inclusive Agreement on the Transition in the Democratic Republic of the Congo.’’ See, e.g., Phillips 1998 on the representation of ideologies versus groups.

128

Anna K. Jarstad

defining which groups should be represented and the share of seats for each group. Over time, both which groups to include and the relative share of power need to be reconsidered to address the temporal dilemma of power sharing. Democratic forms of power sharing have to regulate who is eligible to vote for the group representatives (separate or common electoral rolls), who can stand for elections, and the design of the electoral system. In addition, the issue of accountability is problematic for all types of broad-coalition governments. When all groups are included in government, it is difficult to hold politicians accountable and elections may even seem meaningless. However, elections can make a difference also in power-sharing systems, by allowing voters to choose between different candidates. Cyprus 1960 and Burundi 2000 are examples of different types of power-sharing constitutions. In Cyprus, the constitution stipulated that voters could choose between candidates on separate ethnic rolls. This meant that Greek Cypriots could only vote for Greek Cypriot candidates and Turkish Cypriots could only vote for Turkish Cypriot candidates. Political parties with candidates of different ethnic affiliation could not stand for elections. In practice, there was a lack of political alternatives, especially among the Turkish Cypriots. However, there are ways to make power sharing more democratic. In New Zealand as well as in post-transitional Burundi, the powersharing arrangements allow the voters to vote for candidates of any ethnic origin. The 2000 Burundi PSA stipulates two types of power sharing; during the transitional period, power was shared among ethnopolitical parties, while the permanent constitution stipulated quotas for ethnic representation in the democratically elected National Assembly. Both types are based on a 40/60 formula, to over-represent the Tutsi minority. In the Transitional Government, 40 percent of the seats were allocated to political parties then largely representing Tutsi, and 60 percent were reserved for the Hutu parties (ICG 2004c: 3). The permanent constitution stipulates a quota for Tutsi candidates, not Tutsi parties. This means that people are allowed to vote for any party and that the Hutu parties have an incentive to include Tutsi in order to gain more seats. In the government elected on August 19, 2005 40 percent are Tutsi, but most of them belong to traditionally Hutu parties and only two Tutsi parties remain in the official institutions. This system has thus had tangible effects on ethnic relations. The traditional Hutu–Tutsi interethnic conflict has been substituted by

Power sharing

129

violent dispute over power among Hutu parties. Furthermore, fewer exclusive Tutsi parties remain in power (ICG 2005a). Under the permanent constitution, no group has ensured ethnoexclusive representation. Typically, the minority has more to fear from such an arrangement as it risks absorption by the majority. But it also means that the extremists have no ensured inclusion in government. It is yet to be seen whether Tutsi interests are sufficiently taken care of by this government. This power-sharing arrangement results in the dilemma of providing incentives for moderate parties to attract voters from both ethnic groups, while increasing the risk that extremist Tutsi parties challenge the power-sharing setup by military means. It is often suggested that power-sharing arrangements are inefficient and that the decision-making process is slowed down when all groups have a say in the process. That is true: power sharing does entail the risk of stalemates and problems of accountability. However, this is a consequence of any coalition government and does not imply that such a system is not democratic. On the contrary, after a civil war, it is of particular importance that all interests are represented, in order to render decisions a high degree of legitimacy and to facilitate implementation. It is no easy task to define the groups and who should represent them. Even in societies where power sharing develops in the course of peaceful and democratic procedures, there is no obvious principle for defining target groups for reserved seats in parliament. In New Zealand, each individual, no matter their ethnic affiliation, can now enlist either on the Maori roll or the General roll prior to each election. This allows for some flexibility regarding definition of politicized ethnicity. In the case of New Zealand, it is clear that Maori are the indigenous people, and no other ethnic group, such as the large Chinese community, is allotted a quota. However, it is contested whether or not aboriginal status is a just basis for special representation. The number of seats depends on the share of registered voters on the Maori roll. In addition to quotas for Maori, all main political parties include Maori among their top candidates. As a result, in the last elections the number of Maori in parliament roughly corresponds to their demographic share. There is no mechanism ensuring proportional ethnic representation. At present, however, it is in the interest of all political parties to appeal also to Maori voters and it could therefore be argued that the quotas have become a self-enforcing institution of conflict management (Jarstad 2001).

130

Anna K. Jarstad

Conclusions Power sharing has become a tool for ending conflicts by the inclusion of warring parties in government. In this chapter, I have outlined how previous research expects power sharing to contribute to peace and democratization. Although it is yet too early to evaluate the long-term effects of many of the power-sharing accords discussed here, I have argued that power sharing in war-torn societies can have negative effects on war-to-democracy transitions. Power sharing may not only fail to end the violent conflict. In addition, it may enhance the risk of violence and undermine democratization. Possible negative consequences for sustainable peace were discussed in relation to four mechanisms of power sharing: inclusion of warring parties, intragroup contestation, international dependence, and the leveling of power relations. Firstly, while power sharing works as an incentive to lay down arms for included groups, it entails at the same time an incentive for excluded groups to continue fighting or to take up arms. Secondly, the ‘‘eclipse’’ mechanism may give rise to intra-group conflict. In the short run, power sharing may take the heat off the major conflict and allow other political issues to emerge. But simultaneously, there is a risk that extremist splinter groups are formed. In war-torn societies this risk is particularly acute, and its consequences may be devastating. Thirdly, international guarantees to uphold the powersharing arrangement may turn into dependence. Fourthly, in the long run, shifting power relations mean that it may be difficult to avoid a breakdown of the agreed power-sharing balance. Furthermore, four problems pertaining to democratization via power sharing were discussed: exclusion of moderates, lack of popular support, lack of local ownership, and group representation. Firstly, power sharing only among warring parties risks undermining the development of moderate political parties. PSA allotting government positions to warring parties sharply contradicts the principle of democratic legitimacy and undermines long-term democratization by limiting the space for moderate parties. Secondly, the elites appointed to share power may lack popular support. This is an impediment to vertical legitimacy. Thirdly, the lack of local ownership may hinder the development toward a democracy. Fourthly, power sharing often presupposes the definition of groups and may prove inapt to accommodate new groups that emerge.

Power sharing

131

These eight negative consequences of power sharing cannot all be addressed at the same time. Rather, the design of a PSA involves tradeoffs between the desirable goals of promoting conditions conducive to peacebuilding, on the one hand, and democratization, on the other hand. Below, these trade-offs are discussed in relation to the four dilemmas outlined in the conceptual chapter of this book. The horizontal dilemma of power sharing speaks to the issue of inclusion versus exclusion. The issue of which parties are included – and which actors are excluded – in a power-sharing arrangement has pivotal implications for long-term peace as well as democratization. A viable peace deal cannot be reached without the leaders of the warring parties. However, for the sake of democracy, exclusion of violence makers is warranted. While this is a genuine dilemma which cannot be solved, the negative consequences can be mitigated by inclusion of a broad range of sections of society in the peace process as well as in the power-sharing government. Furthermore, deliberate measures need to be taken to ensure that all parties abstain from violence, for example by disarmament and reforms pertaining to the security sector. The vertical dilemma entails the issue of legitimacy versus efficacy. In the case of power sharing, the consequences of intra-group contestation and popular support interact. For power sharing to contribute to both peacebuilding and democratization, the intra-group contestation should result in elected moderate politicians. This is highly unlikely, but not impossible, in societies shattered by war. The ways to increase the democratic legitimacy of a power-sharing system include election of the members of the power-sharing government, mechanisms allowing new political parties to enter, and making the power-sharing system the object of a referendum. Thus, under certain conditions the vertical dilemma can be avoided. The systemic dilemma entails the issue of internal versus external ownership of the power-sharing agreement.26 Ideally, the local actors arrive at a well-balanced power-sharing arrangement. However, in the midst of a civil war, third-party mediation is often needed to reach a negotiated settlement including a power-sharing deal. To ensure local ownership of the peace process and of the power-sharing agreement 26

For an analysis of dilemmas related to third-party guarantee, see the chapter by Fortna in this volume (see also Walter 2002 on this dilemma in relation to power sharing).

132

Anna K. Jarstad

itself, broad inclusion is desirable already during the negotiation phase. This can be combined with third-party guarantees to ensure a peaceful transition toward democracy. The temporal dilemma of power sharing speaks to the issue that the power balance agreed on in a peace deal may be challenged over time. A PSA tends to reflect power relations at the time of the negotiations, rather than the size of potential electorates. To end a war, it may be necessary to include warring groups in government, despite the risk of undermining the prospects of democratic legitimacy and restricting the possibility for moderate parties to develop. In the long run it is likely that new groups emerge, excluded groups demand a share of government posts, or that the demographic character of the state changes. If there are no mechanisms for adjustments to such changes, the system may be challenged. In such cases, transitional power sharing is to be preferred. However, to end violence, it might be necessary to guarantee the parties involved a permanent share of power. Warring parties may not be prepared to lay down their arms and sign an agreement if they know that transitional power sharing will end, and that they risk losing all political and military power once the transitional period is over. For this reason, there is no simple solution to the temporal dilemma of power sharing. By bringing together research on power sharing from two previously disparate strands of research – conflict management and democratization – the potential negative consequences of power sharing become evident. Under power sharing there is no easy way to avoid dilemmas: the management of one dilemma often results in new trade-offs between peacebuilding and democratization. At the same time, in many cases the alternatives to power sharing are worse. Although each power-sharing deal needs to be designed based on an analysis of the specific context of the war-torn society in focus, a few guidelines could be suggested. Firstly, strive for inclusion rather than exclusion. Design a system that allows for existing moderate parties and for new groups to join. One possibility would be to stipulate a minimum quota, not a fixed maximum quota, and to leave some positions in parliament open for election outside of the quotas. Secondly, immediately after a war legitimacy is often considered more important than efficacy. Ensure that voters have a choice between several candidates in each election. Strive to reserve seats for political parties, not for specific individuals. Thirdly, strive for long-term rather

Power sharing

133

than short-term assistance. To last, some power-sharing deals need credible international security guarantees. In other cases, difficult negotiations continue and need international facilitation and mediation, also after a power-sharing deal is reached. Fourthly, the balance between the need for a flexible system and provisions to guarantee some actors’ influence over a long time period depend on the overall peace process. External security guarantees and international monitoring are often pivotal for a smooth transition from power sharing to more open forms of elected governments. To conclude, this chapter suggests that for power sharing to promote both peace and democracy, it needs a carefully crafted design, as well as continuous monitoring. By taking into account the adverse effects of power sharing on peacebuilding and democratization, some of the worst pitfalls can thus be avoided.

5

When rebels change their stripes: armed insurgents in post-war politics ¨ DERBERG KOVACS MIMMI SO

Following the announcement of the final results of the legislative elections in Afghanistan in 2005, there was widespread concern that powerful warlords, former Mujahideen commanders, and Taliban strongmen had been elected to power in this war-ravaged country. The same year a peace agreement was signed between the Indonesian government and the guerillas in the province of Aceh, aimed at ending the prolonged civil war. The provisions of the agreement laid out the political and legal conditions for the establishment of local political parties and thus gave the amnestied rebel movement an opportunity, for the first time, to pursue its aims through the ballot box. In Nepal, the Maoist rebels agreed to join the interim government following the peace agreement in 2006. But what are the prospects for democratization and sustainable peace in war-scattered societies in which formerly armed insurgents emerge as politicians? The purpose of this chapter is to address this pivotal issue that is likely to be of great relevance to researchers and policymakers alike concerned with better understanding the conditions that facilitate and obstruct a transition to both peaceful and democratic politics in intrastate armed conflicts. Because issues related to failed governance and the unequal distribution of political power and public goods often are at the core of the causes of civil wars, conflict resolution in these contexts frequently includes efforts to introduce, reintroduce, or reinforce political reforms aiming at a transition to democratic politics. One such political reform is the effort to transform warring armies to political parties, who are to pursue their political goals peacefully and openly within the political system. This trend ought to be encouraging, as the process of rebel conversion has been identified as one of the key factors for the successful implementation of peace agreements. In war-to-democracy transitions, the inclusion of formerly armed groups in politics also fulfills an additional purpose, namely to provide the agents for the emergence of a 134

When rebels change their stripes

135

multi-party democracy in previously autocratic or semi-democratic states. Thus, in recent decades we have witnessed numerous former rebels emerge as political parties following peace agreements in intrastate armed conflicts. However, this study argues that although the inclusion of formerly armed and militarized groups in post-war politics has proven beneficial, or even necessary, for the purpose of ending the war through a negotiated settlement, it sometimes has negative consequences for the emerging democratization process. The militant, hierarchical, sectarian, and internally undemocratic nature of many of these groups work counter to the development of peaceful, democratic, transparent, and inclusive policies. In addition, the process of rebel inclusion itself may hamstring the democratization process due to the use of amnesty provisions in the peace process or by limiting the possibility for other actors to emerge and influence the post-war political agenda. In the long run, the lack of democratic progress might also impair prospects for sustainable peace. Paradoxically, therefore, efforts to promote both peace and democratization in war-shattered societies through the transformation of armed groups to peaceful parties might undermine precisely those values that it sought to encourage. In spite of its relevance, this question remains largely unexplored in the scholarly literature, as it tends to fall between two different strands of research that are rarely integrated, namely that on war termination and conflict resolution on the one hand, and that on democracy and democratization on the other. This chapter will set out by discussing the argument in favor of transforming rebel groups to political parties in civil war peace processes for the sake of peace. Thereafter, the critical role played by political parties in democracies is briefly addressed. This is followed by a closer look at some contradictory issues that may arise when the two processes of democratization and peacebuilding meet, and armed groups emerge as political parties in the aftermath of a negotiated settlement. These issues will be analyzed in the light of the four overarching dilemmas addressed in the conceptual chapter of this book: the vertical, the horizontal, the systemic, and the temporal dilemma. Empirical examples will be given from a range of different cases to illustrate the issues under discussion. Finally, in the concluding remarks, a few implications for policy on this area will be discussed.

136

Mimmi So¨derberg Kovacs

Rebel-to-party transformations: a remedy for ending the war Since the end of the Cold War, peace agreements between the warring parties have become a more common option for ending civil wars (Wallensteen 2002: 3). Following in the footsteps of this trend, researchers have attempted to identify the conditions under which parties enter into negotiations, when negotiations are likely to lead to settlements, and when peace agreements between former belligerents are likely to hold and prove durable. Within this strand of literature, two different, though sometimes interrelated, arguments have been put forward in favor of efforts to transform former rebels to political parties in civil war peace processes. The first argument is based on the notion that if the previous war was the result of legitimate grievances on the part of the armed opposition, the conflict resolution processes should strive to address these injustices, in particular the systematic exclusion of certain groups, and as far as possible try to amend them. The second argument builds on the view that in order to ensure the commitment of the rebels to the peace process, one should strive to increase their incentives for peace through granting them a legitimate political role in the new post-war order. In line with the first argument, it has been suggested that because internal conflicts begin with the breakdown of normal politics, the conflict resolution process should entail the establishment of mechanisms that allow the conflict to shift from violence back to politics. The return to normal politics should be on the basis of a new inclusive polity that brings together those who felt deprived and discriminated against, and those who were part of the old political system, to share power and benefits in a new political system. Specifically, third-party actors should strive to put to work settlements that attempt to engage the rebel group in a new role involving participation, legitimization, and allocation through, for example, a conscious effort to encourage the former rebels to pursue their goals openly in the political system (Zartman 1995: 337–338). Because civil wars ultimately concern the distribution of power in society, one of the key problems in civil war conflict resolution concerns precisely the construction of a social and political system, which ‘‘gives reasonable social and political space to all groups in a society’’ (Wallensteen 2002: 133). Democratization, in this view, becomes a way to handle the participation of parties in a post-war society and to give space to actors who have previously been

When rebels change their stripes

137

suppressed or excluded from influence. A conflict resolution process based on the establishment of a democratic political system thus encompasses the necessary procedures for transferring the struggle into constitutional and non-violent forms (Wallensteen 2002: 139). The inclusion and active participation of the former rebels may therefore symbolize the end to one-party politics and exclusive political systems, and becomes the ultimate sign of democratic politics. Cerdas Cruz argues in reference to the Central American peace processes that ‘‘the incorporation of insurrectional groups into the system, transformed into duly legalized political parties, is critical for the future of the transition and the formation of a truly representative party system’’ (1998: 15). Hence, peacebuilding in post-war societies that merely attempts to reestablish the pre-war political order has serious limitations as it ignores the imbalances between groups in existing political and economic structures. Peacebuilding strategies therefore, ‘‘must be geared towards modifying social structures and processes associated with such power imbalances’’ (Jeong 2005: 3). In addition, the formation of a legitimate government is a necessary condition for order and stability in the aftermath of war, and trust and confidence in the new government can best be gained through the establishment of democratic principles reflecting an inclusive representation that may overcome the divisions created by wartime alignments (Jeong 2005: 83–84). Thus, in accordance with this line of reasoning, peace processes should aim to address and modify the underlying grievances, in particular the politics of social, economic, and political exclusion of certain groups or regions that produced the war in the first place, in order to create stable and peaceful solutions to intrastate armed conflicts. In accordance with the second line of argument, there is another more immediate and pragmatic reason for creating and supporting solutions that attempt to engage the former antagonists in the postwar political system. It has been suggested that third-party actors concerned with the successful implementation of peace agreements in civil wars should prioritize the demobilization of soldiers and the demilitarization of politics, through, for example, the transformation of former warring armies to peaceful political parties. In fact, without doing so, ‘‘civil wars cannot be brought to an end, and important normative goals such as the creation and consolidation of democracy and the protection of human rights have little chance of success’’

138

Mimmi So¨derberg Kovacs

(Stedman, Rothchild, and Cousens 2002: 668). Lyons argues that the risk of renewed conflict increases significantly if warring parties have the option to return to armed struggle. The disbanding and disarmament of former combatants thus leave few choices for the parties but to accept the peace process and the outcome of the post-war elections. Any failure of demobilization, on the other hand, leaves the possibility of resuming fighting an option. In addition, if former warring parties have a stake in the post-war political order and believe they have a credible chance of organizational survival and political influence through peaceful means, they are more likely to remain committed to implementing the peace deal and have fewer incentives to return to arms. The successful conversion of former insurgents to political parties thus affects both the capacity and the willingness of these groups to remain committed to the peace process (Lyons 2002; 2004). Whether the rationale has been stated in light of the first or the second argument, or perhaps based on a combination of the two, international donors and concerned third party actors as well as local peace custodians have actively supported the transformation of former warring groups to political parties in several peace processes since the end of the Cold War (Kumar 1998c).1 In some cases it has even been explicitly stated in the peace agreement that the former rebels should be allowed to form a political party and participate in post-war elections as a legitimate political alternative. The outcome of these processes has, however, varied considerably. In some cases the former rebels emerged as the new government party after decades of armed struggle, as happened in South Africa, in East Timor, and in the Palestinian self-governing territories. In other cases, for example in El Salvador, Guatemala, Cambodia, and Mozambique, former rebel groups emerged as opposition parties in the post-settlement period. In some instances, these attempts to transform rebels to statesmen failed when the rebels returned to arms during the implementation of the 1

The phenomenon itself is not new. Prior to the end of the Cold War, there are many examples of armed groups that later assumed governmental power and became political parties. Most of these, however, did so through the use of force, either through victory in wars against other groups in the country or following prolonged armed conflicts with their former colonial powers. In the post-Cold War period, many former rebels have instead converted to political parties as part of a negotiated settlement to end the war and a decision to introduce democratic politics.

When rebels change their stripes

139

peace accord, as was the case with the Khmer Rouge in Cambodia.2 Finally, some rebels emerge as political parties through the establishment of power-sharing mechanisms. In Angola, following the death of their long-time rebel leader, Jonas Savimbi, and the subsequent 2002 cease-fire agreement with the government, the insurgency group National Union for the Total Independence of Angola (UNITA) completed its transformation into a political party and decided to take up its seats in both the parliament and in the cabinet that it had been granted in the 1994 power-sharing agreement. Although the specific problems and prospects these groups face when entering legal and democratic politics are quite diverse depending on the particular circumstances of their new role and position, they also share many similar challenges in terms of the expectations on them as the central agents of democratization.

Political parties as agents of democratization Because most contemporary peace processes include a simultaneous process toward democratization as part of the conflict resolution efforts, the transformation of former rebels to peaceful political actors fulfills an additional purpose. They provide the post-war political order with the political parties deemed necessary for the emergence of a multi-party democracy in previously autocratic or semi-democratic states. In the democracy literature, political parties have traditionally been regarded as the main intermediary organization of liberal democracy, linking citizens with the state. Building on this notion of political parties as the intermediary links between the people and the state, the various functions that political parties are expected to perform in democracies can be grouped into two main categories. First, political parties are expected to perform a variety of representative functions aiming at channeling the interests and preferences of the people into policy proposals. Second, they also perform a variety of procedural or institutional functions that serve to organize the political system of the 2

For an in-depth comparative study on why some rebel groups and not others successfully transform from violent protagonists to viable political parties following peace agreements in intrastate armed conflicts, see So¨derberg Kovacs 2007.

140

Mimmi So¨derberg Kovacs

state, for example in parliament or in government, or when it comes to the recruitment of leaders (Bartolini and Mair 2001: 332; Gunther and Diamond 2001: 7–8). Political parties are thus important for a number of tasks deemed critical for the democratic political system. Hence, in most contemporary democracies, political parties are considered ‘‘a sine qua non for the organization of the modern democratic polity as well as for the expression and manifestation of political pluralism’’ (van Biezen 2003: 1). In new democracies emerging from authoritarian regimes in particular, the very establishment of democratic procedures has been identified with the establishment of free competition between parties, and parties have hence been attributed a pivotal role for the expression of pluralism and political participation (van Biezen 2003). However, it is commonly recognized that the empirical reality of political parties found in many weak and developing states sometimes makes it difficult for these to function as the expected agents of democratization. According to Grugel one of the greatest problems facing party systems in new democracies is their lack of institutionalization; they enjoy low levels of legitimacy, have weak roots in society, are poorly organized, and there are few opportunities for structured interaction between parties. Thus, ‘‘they operate quite differently than the academic literature on democracy and democratization would like to suggest’’ (2002: 74–75). Randall and Sva˚sand (2002), in their discussion of the performance of political parties in Africa’s new democracies, point to these parties’ limited geographical spread, their lack of both financial and human resources, and their poor organizational underpinnings. Owing to these shortcomings, the representative role of these parties is generally limited, with small bases of support and weak links to the population. Moreover, they rarely have an aggregative function or present clearly distinguishable policy platforms. However, while most observers acknowledge the limitations of political parties in new democracies, political parties ‘‘remain an indispensable institutional framework for representation and governance in a democracy,’’ and ‘‘only political parties can fashion diverse identities, interests, preferences, and passions into laws, appropriations, polices, and coalitions’’ (Diamond 1999b: 96–98). Hence, Diamond argues that given the critical role played by parties in a democracy, ‘‘some degree of party system institutionalization – of parties with effective, autonomous organizations, and developed, relatively stable linkages to

When rebels change their stripes

141

voting blocs and social organizations – seems an important condition for democratic consolidation’’ (1999: 96–98). Thus, although the limitations on political parties in new democracies are generally acknowledged both in theory and in practice, political parties are still considered vital to the process of democratization. This is equally true for democratic transitions that emerge from authoritarian rule and those that take place in the wake of the ending of a civil war.

Dilemmas of armed insurgents in post-war politics From the perspective of both war termination and democratization, the transformation of formerly armed groups to political parties in war-todemocracy transitions is thus generally considered a positive development that ought to be encouraged and supported by local as well as international peace custodians. The parallel processes of peace and democracy are expected to reinforce each other in cases where former warring armies become political parties. However, the emergence of rebels as newborn democrats in post-war politics gives rise to a number of challenges for both democratic progress and sustainable peace. This chapter will explore a few issues that are illustrative of this phenomenon. These issues are analyzed in light of the dilemmas outlined in the conceptual framework of this book: the horizontal, the vertical, the systemic, and the temporal dilemma.

The horizontal dilemma: inclusion versus exclusion of rebel groups In terms of the horizontal dilemma, this chapter will discuss two issues that illustrate the tensions that may arise between the values of peace and democracy when former rebels are included into the post-war political system. First, the risk of a sedimentation of parties and issues based on the structures of war, and second, the risk of undermining the democratic principle of rule of law when the peace agreement includes provisions of amnesty for crimes committed during the armed struggle. First, the inclusion of warring parties into post-war politics often takes place at the expense of the exclusion of other political parties and new political issues emerging from the post-war democratic context. This is problematic from a democratic perspective because democratic consolidation in post-war societies is most likely to take place when

142

Mimmi So¨derberg Kovacs

‘‘a new leadership emerges, seeking to organize politics in a different way from those adopted by discredited parties and leaders in the past’’ (Clapham and Wiseman 1995: 226). As noted by Ottaway, democratization requires a radical shift in the nature of political power. Elections in weak and previously non-democratic states are therefore unlikely to function as a vehicle for genuine political change. In fact they may even contribute to the sedimentation of existing power structures through a ‘‘premature closure’’ of the transition process due to the establishment of formal procedures and institutions before a real change in the nature of power has taken place (Ottaway 1997: 3). In post-war societies, this means sedimentation of the structures of war. Through the inclusion of formerly warring groups, these parties come to define which questions will dictate the post-war political agenda. In many cases, these are the same, or similar, issues over which the war was fought. Hence, there is a risk that post-war politics will serve to reinforce those incompatibilities that the preceding peace process sought to reduce or eliminate, and consolidate wartime allegiances and societal divisions. This is, for example, illustrative for the situation in Bosnia, where nationalist parties like the Croatian Democratic Union (HDZ), the Serb Democratic Party (SDS), and the Bosniac Party of Democratic Action (SDA) play an important role in hampering democratic development (Reilly 2002). Second, the inclusion of formerly warring parties in post-war politics often takes place at the price of granting these individuals and groups amnesty for war crimes and human rights abuses committed during the war. The rebels often declare that their participation in the negotiations and their commitment to a peaceful solution to the armed struggle is contingent on assurances of amnesty. In a large number of peace processes, from Angola in 1991 to Aceh in 2005, armed groups have thus been granted amnesty for war crimes in the peace accords alongside recognition and legitimacy as political actors.3 As noted by Baker, 3

The general amnesty provisions are sometimes limited by certain exceptions. For example, in the 1997 peace accords in Tajikistan, the amnesty law calls for criminal charges against former civil war combatants to be dropped with the exception of certain crimes, such as the killing of non-combatants, rape, terrorism, and drug smuggling (ICG 2004b). Likewise, in the Aceh peace accords of 2005, only those charges with offences relating to the political conflict, that is, for rebellion, will be eligible for amnesty, and not those charged with criminal charges, such as rape, murder, or arson (ICG 2005c).

When rebels change their stripes

143

this practice places peace and conflict resolution before justice and human rights and may be considered a trade-off with the principle of rule of law that symbolizes a democratic political system (Baker 2001). From a democratic point of view, this practice may therefore pose an obstacle to the prospects of democratization in post-war societies. An approach that instead places justice before peace and emphasizes the need for prosecution for abuses committed during the conflict can serve several functions that further a democratic culture (Kritz 2001: 808–809). It can provide victims with a sense of justice, put an end to a culture of impunity, and provide an important focus for rebuilding the judiciary and the criminal justice system in accordance with rule-oflaw principles. In addition, settlements that build on democratic foundations ‘‘have a far better chance of achieving sustainable security,’’ while settlements that put conflict management before democracy and human rights may have ended brutal civil wars but remain inherently fragile (Baker 2001: 760–761). However, it should be noted that although peace and justice may be seen as theoretically separate goals, they are not necessarily mutually exclusive in practice. There have been many attempts to reconcile them, for example through the establishment of truth and reconciliation commissions. The most appropriate approach for the purpose of furthering both peace and democracy will differ substantially from case to case and is dependent on answers to questions such as who should be held responsible? for what? and by whom? It is not necessarily the case that bringing the offenders to justice is the strategy that is the most conducive for either peace or democracy. The outcome hinges to a large extent on the perceived legitimacy of the selected approach in the eyes of the domestic population. As noted by Arzt, ‘‘if international criminal courts are to achieve their aims – one of which is to contribute to the consolidation of democracy and the triumph of the rule of law [. . .] – perception of their legitimacy by the local population is a crucial factor’’ (2006: 227). For example, the decision of the International Criminal Tribunal for the former Yugoslavia (ICTY) to issue an indictment for war crimes against Kosovo’s democratically elected Prime Minister Ramush Haradinaj in March 2005 was met with shock and anger among most of Kosovo’s majority Albanians. The decision resulted in a number of protests and campaigns that, according to observers, could have escalated to violence had it not been for Haradinaj’s own public plea for

144

Mimmi So¨derberg Kovacs

calm (ICG 2005b). Hence, the trade-off between amnesty and justice in war-to-democracy transitions brings up issues that also strongly relate to the other dilemmas discussed in this book, most notably the vertical and the systemic dilemma. In some cases, however, the question of amnesty for the sake of peace versus justice seems to be tied directly to the issue of inclusion versus exclusion. More precisely, its seems like participation equals protection and non-participation equals prosecution. As long as the rebels comply with the terms of the agreement that provide for their inclusion into peaceful and democratic politics and succeed in their transformation to political parties, they are unlikely to face trial no matter the extent or severity of prior human rights abuses. When, however, the rebels fail to live up to their prior commitment to peace, they are more likely to be held responsible for crimes committed during the armed conflict in spite of previous guarantees of amnesty. The case of the Revolutionary United Front (RUF) in Sierra Leone is illustrative. The 1999 Lome´ peace agreement granted the rebels ‘‘absolute and free pardon and reprieve’’ for crimes and human rights abuses conducted during the armed struggle. The agreement further stated that the Government of Sierra Leone would ensure that no official or judicial action would be taken against any member of the RUF in respect to their actions during the war.4 However, the United Nations, which played a vital role as mediator during the negotiation process, issued a disclaimer at the time of the signing of the accord, which stated that the amnesty provisions were not applicable to ‘‘international crimes of genocide, crimes against humanity, war crimes and other serious violations of international humanitarian law’’ (Francis 2000: 366). Subsequently, in May 2000 the RUF ambushed and abducted hundreds of the newly arrived UN peacekeepers and the peace process broke down. In a direct response to this event, the Government of Sierra Leone turned to the UN to ask for assistance in setting up a court ‘‘to try and bring to credible justice’’ members of the RUF for crimes committed during the armed conflict (ICG 2003: 2). The Special Court became operational in 2002 and the first indictments were issued in March 2003 against the

4

Article IX, Peace Agreement between the Government of Sierra Leone and the Revolutionary United Front of Sierra Leone, available at www.sierra-leone.org/ lomeaccord.html. Last visited February 26, 2007.

When rebels change their stripes

145

leadership of several of the parties that had participated in the civil war, including most of the top RUF leaders and commanders.5 A similar development took place in Cambodia following the Khmer Rouge’s defection from the peace process and its failure to comply with the terms of the Paris Agreement of 1991. Following the dictates of pragmatism and realpolitik, the accord provided for the participation of the former rebels as a legitimate political party in a post-war democratic system built on power sharing. However, in July 1994, after Khmer Rouge’s decision to withdraw from the peace process, the Cambodian National Assembly responded by outlawing the group. Shortly thereafter, the United States Congress passed the Cambodian Genocide Justice Act. These two events signaled the beginning of a changed national and international approach toward the rebels. In 1997, at a time when the movement was at the verge of its final military and political collapse, a process toward international justice was set in motion by the Cambodian Government and the United Nations. In 2003, a tribunal was established to try the surviving leadership of the Khmer Rouge for genocide and crimes against humanity during the Pol Pot regime from 1975 to 1979 (Hammer and Urs 2005).

The vertical dilemma: popular legitimacy versus efficacy of governance In cases where former rebels have entered post-war politics, a number of issues can be identified that relate to the second overarching dilemma that speaks to the trade-off between legitimacy and efficacy. These issues center on the democratic deficits that often continue to characterize the practices of these parties after entering democratic politics due to the legacy of the armed struggle and their background as political-military organizations. In addition, they relate to the changing relationship between the former rebels and the population at large in war-to-democracy transitions. The particular issues raised in each case depend, however, to a large extent on whether the insurgent group capture government power in the post-war elections or assume the role of opposition party. These two different scenarios will be discussed in turn. The problem of transforming armed movements into democratic governments is not unique to the post-Cold War era. A large number 5

See www.sierra-leone.org/specialcourt/html. Last visited February 26, 2007.

146

Mimmi So¨derberg Kovacs

of African liberation movements were declared political parties on the eve of independence from colonial rule. As noted by Salih, many of these parties and their leaders found it difficult to adjust to the accountability and transparency democratic politics is supposed to entail. They hardly ever transformed on the basis of politics and remained essentially reliant on personalistic and clientelistic mechanisms of internal control within their parties and in their relationship with their electorate. Interestingly, the experiences of the early post-colonial liberation movements are in many ways echoed in more contemporary examples of armed movements in Africa that have come to power through negotiated settlements. These parties often blur the distinction between the party and the state and they continue to be an embodiment of nationalist/populist politics of the liberation struggle (Salih 2003: 12–14). What characterized these movements during the armed struggle was the stress on unity and the need to speak with one voice for a single purpose. Paradoxically, therefore, the more successful these groups were in terms of the liberation struggle, the more difficult they experienced the transition to become one political competitor among many in a multi-party democracy (Ottaway 1991). For example, while it has been acknowledged that the African National Congress (ANC) has come a long way in its internal transformation since the first democratic elections in South Africa in 1994, critical voices have been raised over some internal practices of the party, and warnings have been issued concerning its electoral dominance, which might indicate that South Africa is becoming a de facto one-party state (Lanegran 2001: 99). Although the ANC cannot be blamed for its electoral success, the combination of the party’s dominance and its lack of internal party democracy is a cause for concern, as this means that the fate of South Africa’s new democracy to a considerable extent is dependent on the party’s own democratic behavior (Randall and Sva˚sand 2002: 46). Likewise, it has been argued that the South West Africa People’s Organization (SWAPO) in Namibia still plays the tunes of the nationalist politics that brought it to power, which is proving a serious obstacle for further democratic developments (Melber 2002). In East Timor, the former resistance party the Revolutionary Front of Independent East Timor (FRETILIN) won the Constituent Assembly elections of 2001, following the vote for independence in 1999. According to Hohe, the party drew heavily on indigenous values,

When rebels change their stripes

147

traditions, and symbolism in its electoral campaign. In doing so, the party was able to appeal to traditional values of unity and hierarchical political authority, and the election campaign served to rationalize the emergence of a one-party state, which after the elections has been rapidly consolidated by the party. ‘‘Fretilin’s linkage of these concepts to the international multiparty exercise means, logically, that political opposition can only be expressed in a hostile manner, and there is no space for alternative and peaceful competition’’ (Hohe 2002: 85).6 After its entry into politics, FRETILIN has attempted to use the party’s own symbols as symbols of the state, which may be interpreted as ‘‘a dangerous precedent bespeaking an authoritarian tendency to conflate the ruling party with the state itself’’ (Smith 2004: 153). Former rebels and guerilla movements that have emerged as opposition parties following the outcome of the electoral process have also displayed difficulties in adjusting to democratic politics. They need to take into consideration the opinions and wishes of their former members and loyal supporters and simultaneously attempt to attract votes among a broader section of the population in order to survive in politics. In these situations, many parties are more likely to remain true to their wartime ideology and political program than to adjust their political message to the new political and economic realities. In many instances, their parliamentary behavior as opposition parties often shows tendencies of following the principles of a ‘‘reversed Clausewitz,’’ where politics becomes the continuation of war by other means. Manning has argued that the character and degree of commitment that former armed groups make to the post-war political system depends on the kinds of challenges that adaptation to the new environment presents for them as organizations. Specifically, she argues that parties that are forced to make significant adjustments to their 6

According to Smith, the only real opposition exists in the form of President Xanana Gusmao, the former leader of the FRETILIN’s armed wing and later the leader of the united resistance front formed in 1998, and the loyalty he commands from the armed forces. Since the landslide victory of Gusmao in the presidential elections of 2002, the conflict between FRETILIN and Gusmao has seen no improvement and open disagreement has surfaced on a number of political issues (2004: 154–155). In addition, unofficial security elements, possibly linked to FRETILIN, have emerged. Thus, ‘‘the conditions have been laid, therefore, for the stark alternatives of either one-party rule or violent political competition’’ (Hohe 2002: 85).

148

Mimmi So¨derberg Kovacs

collective incentive strategies in order to compete politically and survive in post-war politics also invest in necessary changes in their internal authority structures. In addition, they also make more investments in the procedural rules of the democratic game and thus make durable democracy more likely. In contrast, in situations characterized by the lack of vigorous political competition, these changes rarely take place and the prospect of democratic progress is much less likely (Manning 2004). For example, Manning argues, Mozambiquan National Resistance (RENAMO) in Mozambique has been subject to very little pressure to change its collective incentives strategies after its entry into parliamentary politics over a decade ago. The social and political cleavages that characterized the county during the war carried over into the post-war period, and the apparent failure of the government to improve living conditions for the large part of the population that lives in the areas from which RENAMO draws its support reinforces this picture. In addition, Mozambique’s essentially two-party system poses an obstacle to the emergence of viable political competitors. Hence, RENAMO continues to fight the armed struggle from within parliamentary politics and remains more anti-government than in favor of any well-defined and clearly spelled-out political or economic agenda. In addition, it attracts votes mostly among its former wartime constituencies in the central and northern parts of the country. This lack of incentives to change its appeal to the population has led to little pressure to address questions of intraparty nature, and the party has eluded any significant changes concerning its own internal organization. The old leadership continues to control the party in a hierarchical fashion that has seriously hampered the party’s democratic performance (Manning 2004). In contrast, the Kosovo Liberation Army (UCK), which became the Democratic Party of Kosovo (PDK), was obliged to make significant adjustments to its collective incentives strategies and broaden its message and identity in order to compete in post-war politics, as the political cleavages that defined the armed conflict no longer defined the political arena. This also allowed new party competitors to emerge and the PDK was forced to invest in changes to its internal authority structures, for the purpose of strengthening its position toward other parties in the political system. Consequently, the party has adapted more quickly to electoral politics and adjusted itself to the rules of the game (Manning 2004).

When rebels change their stripes

149

In the cases of the former communist guerillas in Central America, one of their central ideological characteristics during the armed conflict was precisely their anti-systemic nature. For this reason the process of integration into the political system has often been both slow and complex (Cerdas Cruz 1998: 46). For example, although the former communist guerilla in El Salvador, the Farabundo Martı´ National Liberation Front (FMLN), has been comparatively successful in its electoral performances following its entry into parliamentary politics, it has struggled with internal disagreements over issues of ideology and party governance. The internal divisions of the FMLN emerged shortly after the first post-war elections in 1994, when two of the five organizations that made up the FMLN left the party over such a dispute. According to Wade (2007), this split signaled the beginnings of the party’s post-war difficulties in keeping a united front. Since then, she argues, the FMLN has been characterized by factionalism between those who want to remain true to the character of the revolutionary movement and favor a vertical decision-making structure demanding strict adherence to the party line, and those pressing for more pragmatic and pluralistic policies and increased intraparty democracy. In Guatemala, the Guatemalan National Revolutionary Unity (URNG) has suffered from the lack of an ideological and programmatic identity following its transition to a legal political actor. The party has remained cautious on ideological issues and unsure of how to identify itself in the political system in general and to the Guatemalan left in particular. Vinegard suggests that this absence of ideological and political clarity has contributed to the URNG’s difficulties in establishing itself as an effective voice of opposition to the government. In addition, the hold of URNG’s traditional leaders on the party structure, and its seeming lack of tolerance for internal dissent, has given reason to question the party’s capacity for internal democratization (Vinegard 1998: 223–225).

The systemic dilemma: domestic versus international legitimization of parties The transformation of warring actors to legal political parties in warto-democracy transitions also raises the question of who has the right to determine which formerly warring groups are to be considered

150

Mimmi So¨derberg Kovacs

legitimate political actors and therefore should be granted a role in post-war politics. This question relates to the systemic dilemma, which concerns the trade-off between international and domestic ownership over processes of peacebuilding and democratization. Because the transformation of former rebels to political parties has emerged as a commonly recognized tool for conflict resolution and democratization in many internationally supervised peace processes, international actors have spent a considerable amount of resources on these processes of rebel conversion in cases where this has been deemed necessary. For example, in the Mozambique peace process, a UN trust fund was established and eventually US$17 million was raised for it to help RENAMO to transform itself into a political party, as the international community realized that the rebels’ participation in post-war politics and their continued commitment to the peace process depended on the provision of external resources (Vines 1996). In addition to providing financial assistance, donors have also frequently contributed with technical assistance and training to armed groups to build their capacity to select candidates, organize election campaigns, and monitor election outcomes. Assistance to these new parties has in some instances also continued in the post-election period (Kumar 1998c: 218). In some cases, international processes of legitimization of armed groups have been aborted or reversed during the implementation process due to the rebels’ inability or unwillingness to comply with the terms of the agreement. This was, for example, the case with the RUF in Sierra Leone and the Khmer Rouge in Cambodia. International actors thus play a critical role in determining the fate of these groups. The international community’s considerable influence over the dynamics and outcomes of these processes of rebel conversion raises issues of concern over ownership. Does the process of international legitimization (or marginalization) of certain warring groups always comply with the perceptions and opinions of the domestic audience whose political future is at stake? The case of the Lome´ Accord in 1999 between the Government of Sierra Leone and the RUF is illustrative of a case where the strategy of the international community stood in stark contrast to the wishes of the majority of the local population. International mediators and donors put strong pressure on the Government of Sierra Leone to extend political recognition and legitimacy to the RUF in order to facilitate a

When rebels change their stripes

151

negotiated settlement to end the armed conflict. This was in spite of the fact that most government representatives, including the president, initially were very reluctant to do so, and regardless of the fact that most evidence from the prolonged civil war seemed to suggest that the rebels had very little support among the population for their cause. According to the terms of the agreement, the rebels were granted a considerable amount of political influence over state affairs and a general amnesty for war crimes (Bangura 2000: 564–565). It was not until after the UN hostage-taking event in May 2000 that international opinion shifted toward a strategy of military and political marginalization of the rebels, a strategy that better reflected the preferences of the Sierra Leonean population. This conclusion may be drawn on the basis of the outcome of the post-war elections in 2002, where the fraction of the former rebel group that had turned into a political party only received 1.7 percent of the popular vote (Kandeh 2003). The situation might also be the reverse. When the Office of the High Representative (OHR) in Bosnia Herzegovina has stepped in and dismissed democratically elected political actors and parties with a nationalist agenda because they have been considered to be an obstacle or a threat to peace and democracy in the country, it clearly violates the democratically expressed opinions of the domestic population. When Hamas, who publicly opposes a peace deal with Israel, was asked to form a new government in the Palestinian territories following its electoral success in the parliamentary elections in early 2006, the international community faced a similar problem. In spite of having come to power through democratic elections, both Israel and the United States, who has branded the group as a terrorist organization, decided to withhold funds previously channeled to the Palestinian administration. Some critics argue that such a strategy might be counterproductive, as it risks alienating the organization and may push it even further in a non-democratic direction (ICG 2007b). Such a development would seriously hamper any future prospects for peace in the region. The dilemma concerning the trade-off between external versus internal control over critical choices in war-to-democracy transition gives no easy clues as to which approach truly is the most beneficial for either peace or democracy. However, without considering the legitimacy of the choices made, the prospects for durable peace and self-sustaining democratic progress are likely to be in jeopardy.

152

Mimmi So¨derberg Kovacs

The temporal dilemma: short-term versus long-term effects of rebels in politics Beyond posing an obstacle to democratic development, a continued lack of good governance and democratic progress in new democracies poses a long-term risk for sustainable peace and may contribute to renewed armed conflict. This problem speaks to the temporal dilemma between efforts to advance both peace and democracy following an armed intrastate conflict. In post-war societies, the lack of democratic progress is especially problematic, as the legitimacy of the peace process is contingent on the display of significant improvement in people’s daily lives. Unless peace brings significant peace dividends, both the people and the warring parties will start to question the value of the current peace accord and the political order that emerged from it. The shortcomings of the democratic system can thus provide fuel for new grievances. Besides, in addition to the often conflict-ridden transitional phase, semidemocracies, or states that fail to move beyond the initial buildup of democratic institutions and where the underlying structures of politics remain autocratic or neo-patrimonial in nature, are more prone to armed conflict than are both democratic and autocratic states (Hegre et al. 2001). This further underscores the gravity of the lack of democratic progress in post-war societies, and its implications for war recurrence. In Liberia, the post-war elections of 1997 and the developments that followed in the wake of former warlord Charles Taylor’s electoral road to the presidency are ample evidence of the possibly of such a scenario. In the elections, Taylor and his former army-turned-party, the National Patriotic Front of Liberia (NPFL), converted into the National Patriotic Party (NPP), won the national vote with a landslide. Although the elections were supposed to mark the end of seven years of brutal civil war in Liberia, they also marked the beginning of a deteriorating political and security situation that would ultimately culminate in the outbreak of a new civil war in early 2000. After assuming power, Taylor openly declared that he no longer considered himself committed to the conditions of the Abuja peace accords. He refused to carry out the reconstruction of the national army, and he took advantage of his new power platform to oppress the political opposition in the country. Soon thereafter, the new regime began to show the same

When rebels change their stripes

153

authoritarian attributes and the same social and political demise that had characterized the pre-war authoritarian regime. Subsequently, in the early months of 2000, reports of the founding of a new rebel movement began to circulate in Liberia. The group announced itself as Liberians United for Reconciliation and Democracy (LURD) and consisted mostly of political opponents to Taylor; militia groups, former politicians, and Liberian refugees that Taylor’s politics had forced out of the country. This marked the beginning of a new civil war that did not come to a conclusion until August 2003, when Taylor was forced to leave the country and step down from power. There is little doubt that the outbreak of the civil war in early 2000 was directly linked to Taylor’s misuse of his role in government power and the lack of democratic reform that had characterized the post-Abuja political order (Nilsson and So¨derberg Kovacs 2005). The example of Liberia under Taylor strengthens the notion that efforts to advance sustainable peace must be accompanied by efforts to advance democratic values in war-shattered societies. However, it also points to the difficulties that may arise when the two processes of conflict resolution and democratization are introduced simultaneously in a war-scattered society. Under such circumstances, the value of peace (in terms of absence of war) may clash with that of democracy and sustainable peace in a longer time perspective.

Conclusions The purpose of this chapter has been to address and discuss the prospects for democratization and sustainable peace in war-shattered societies in which armed insurgents emerge as political parties in the post-war political order. In the last decades, the transformation of armed groups into political parties has become an integral part of peace efforts aiming at ending civil wars. This development has reflected findings emanating from the war termination and conflict resolution literature, in which the transformation of warring groups to political parties has been identified as one of the contributing factors to the successful implementation of peace agreements in civil wars. The theoretical rationale for such an approach has been based on the notion that if civil wars are seen as generally emerging from the mobilization of legitimate grievances on part of the population, the peace processes and the post-war system should aim to address and mitigate these

154

Mimmi So¨derberg Kovacs

injustices in order to create a just, and therefore stable, peace. In addition, the successful implementation of civil war peace agreements is contingent on the continued commitment of the warring parties to the peace process and their incentive for doing so increases considerably when granted a legitimate political role. Because contemporary peace processes in civil wars frequently include a parallel transition to democratic politics as part of the conflict resolution efforts, this trend has also fulfilled an additional purpose, namely to provide the post-war state with the political parties necessary to introduce a political system based on the principles of multi-party democracy. Although the limitations of political parties in new democracies have been widely acknowledged, political parties are generally still seen as the key agents of democratization and are expected to perform a wide variety of functions deemed essential to the democratic political system. This chapter has pointed to some tentative yet intriguing findings in regard to this trend of rebel reappearance in post-war politics. Although the transformation of formerly armed groups to political parties has proven critical to the success of many peace processes in the last decades, the inclusion and participation of these groups in postwar politics has in several instances proven problematic for the democratization process. In terms of the horizontal dilemma, two issues in particular were identified. The inclusion of former warring parties in post-war politics often takes place at the expense of including new political parties into the political process. Immediately following the end of a civil war, this might be inevitable for the purpose of ending the armed conflict and due to the lack of viable alternatives. However, there is a great risk that this leads to sedimentation of the political cleavages that defined the society during the armed conflict and the issues that created the wartime alignments. A possible strategy for overcoming this problem could be found in the use of transitional arrangements, for example in the establishment of a transitional government that includes the warring parties for a predetermined and limited time period. Only toward the end of the interim period, postwar elections are held to determine the future distribution of power in the country. This was, for example, the approach chosen in Liberia following the 2003 peace agreement. Another issue of concern is the use of general amnesty provisions for former combatants in exchange for laying down their guns and committing to the peace process. In many cases this has been considered a

When rebels change their stripes

155

necessary condition for ending the war. From a democratic point of view, however, such exemptions from the rule of law and the principles of human rights risk encouraging a culture of impunity. However, it is equally true that an emphasis on justice may impinge on the emerging democratization process if the procedure is not considered legitimate by the population at large, which was illustrated by the case of the ICTY indictment of Prime Minister Haradinaj in Kosovo in 2005. In some cases, amnesty has in reality been contingent on the parties’ postsettlement behavior and their commitment to a peaceful process. In the cases of both Sierra Leone and Cambodia, the peace agreements provided the RUF and the Khmer Rouge respectively with controversial amnesty provisions and a legal status as political parties. These provisions were later revoked following these groups’ repeated failures to live up to the terms of the agreements. What may be considered the most appropriate approach will thus most likely have to be settled on a case-to-case basis. However, a general strategy that equals participation with protection and non-participation with prosecution risks undermining the very values that these processes of international criminal law are intended to uphold. In terms of the vertical dilemma, the inclusion of formerly armed groups into positions of political power has in some cases resulted in democratic deficits due to these groups’ organizational legacies as political-military organizations originally created for a different purpose. Many groups with their origins in the armed struggle have been forced to make critical trade-offs between keeping their legitimacy in the eyes of their wartime constituencies and the need for making broader appeals to a larger part of the population, and between keeping the structures of the original group intact and adjusting to the democratic demands of accountability, transparency, and efficacy. These issues may be difficult to overcome. However, an increasing awareness of this dilemma and its implications for the post-war order might help to adjust the democratic expectation on these groups as they enter peaceful politics. The transformation of warring parties also raises the critical question of who has the right to determine which groups are to be considered legitimate political actors. This issue relates to the systemic dilemma and the trade-off between international and domestic ownership of the war-to-democracy transition. This chapter suggests that the process of international legitimization or marginalization of certain

156

Mimmi So¨derberg Kovacs

warring groups needs to reflect the perceptions and opinions of the concerned domestic population in order to work to the benefit of both peace and democracy in that country. Finally, in regards to the temporal dilemma, the lack of democratic progress in a post-war society is problematic not only for the sake of democracy, but also for peace, as failed governance also serves to undermine the long-term prospects for durable peace. The case of Liberia following the elections in 1997 is illustrative. Shortcomings in the democratic process and lack of visible improvements following the transition to democracy are especially problematic in post-war societies, where there already is a high risk of war recurrence due to the extreme polarization of society and the heightened sense of insecurity. At the same time, the conditions for a successful democratization process and democratic consolidation are less favorable in post-war societies than elsewhere, due to the legacies of the war. This dilemma speaks to the critical need for a long-term policy of engagement from the side of concerned local and international peace custodians in war-shattered societies. The construction of comprehensive peace agreements which attempt to reconcile a multitude of different and sometimes competing goals, including that of ending violence and initiating a process toward greater political liberalization and democratization, might not, however, be the most conducive strategy. The construction of such all-encompassing peace and democracy packages does not ensure the viability of these processes, nor their outcomes. Perhaps a better strategy would be to introduce these different goals (and the appropriate methods and tools for reaching them) in a step-by-step manner in a phased process. There is no readily available formula for such a process, but the purpose would be to avoid or mitigate some of the dilemmas that may arise when the values of peace and democracy are aimed at simultaneously.

6

Post-war elections: uncertain turning points of transition BENJAMIN REILLY

Elections held as part of a peace deal following a violent conflict highlight several crucial dilemmas of democratization in post-war societies. Such ‘‘post-war elections’’ are now a feature of almost all efforts to democratize war-torn regions, with peace agreements routinely including provisions for elections to be held as part of the process of conflict termination, often with the assistance, supervision, or sometimes direct control of the international community. But while post-war elections have become an integral element of contemporary peace agreements, they can also themselves become the focus of increasing tension and renewed violence. Taking a comparative perspective, this chapter focuses on several inherent dilemmas of post-war elections, including issues of timing, sequencing, mechanics, political parties, and the role of the international community. In each of these areas, post-war elections force difficult choices to be made between short-term versus long-term priorities, representation versus stability, domestic versus international legitimacy, and a range of other sometimes incompatible objectives. These dilemmas are reinforced by the competing discourses that dominate both academic and policy discussions of post-war elections. On the one hand, elections and democracy are often seen as a primary means of conflict management, with theorists arguing for the benefits of democratic competition in managing the tensions inherent in all societies, including war-torn ones (see Przeworski 1991). On the other hand, an increasing body of work points to the dangers of holding elections in conflict-prone societies, and the empirical reality that societies in the early stages of democratization are often more, not less, conflict prone (Snyder 2000). The role of the international community is another important factor. Competitive elections in post-war societies are often promoted by international actors for a range of varying and even mutually contradictory reasons. Typically, such elections are expected to play a role in simultaneously terminating civil 157

158

Benjamin Reilly

wars; encouraging the transformation of warring armies into peaceful political parties; stimulating the development of ‘‘normal’’ politics; choosing members of a legislature or other kind of representative assembly; forming a government; and conferring legitimacy upon the new political order. For all of these reasons, post-war elections are today widely seen as an integral part of the process of war termination, international disengagement, and nation building. However, the success of post-war elections in achieving these goals has varied considerably. In some cases, such as Namibia in 1989, El Salvador in 1994, or Mozambique in 1994, elections clearly played a vital role in making a decisive break with the past. In others, such as Angola’s abortive 1992 elections held under the Bicesse peace accord, or Liberia’s 1997 elections, flawed elections created more problems than they solved. Haiti’s parliamentary and presidential elections in 1995 led to a first-ever transition of power but also highlighted administrative deficiencies which undermined the credibility of the broader electoral process. By contrast, in Cambodia’s United Nations administered polls of 1993, the technically successful elections were soon overwhelmed by the realities of power politics as the ‘‘losing’’ party at the elections returned to power through hard-line tactics. In post-war Bosnia, successive elections held under the Dayton Peace Accords helped nationalist parties cement an early grip on political power, while in Kosovo and East Timor a more measured electoral timetable played a constructive role in terms of political development. More recently, in Papua New Guinea’s rebellious island province of Bougainville, the combination of an extended electoral timetable, international observation, and systemic innovations have helped secure one of the world’s most successful, if little-known, cases of post-war peacebuilding. Similarly, Liberia’s 2005 elections marked the end of the transition following the country’s second civil war and resulted in Africa’s first democratically elected female head of state, former World Bank employee Ellen Johnson-Sirleaf. Finally, while it is still too early to evaluate recent high-profile elections in Afghanistan and Iraq, in both cases it is clear that elections themselves have not led to an end to hostilities, and in Iraq may have contributed to the ongoing sectarian conflict. As this brief survey indicates, there has been a considerable variation in the relative success of elections in meeting the twin goals of war termination and consolidation of democracy from country to country

Post-war elections

159

and from case to case. The tension between these two overarching goals lies at the heart of this particular dilemma of democratization. As Terrence Lyons (2004: 272) notes, because post-war elections typically carry such tremendous burdens of expectation, they are often saddled with multiple and often mutually contradictory objectives: They are designated in peace agreements as a primary instrument of implementation and hence play critical goals with regards to war termination. At the same time, they are designed to promote a process of democratization and to serve as ‘‘breakthrough’’ elections that initiate a new set of rules and institutions for competitive, multiparty politics . . . Success with relation to one goal, say war termination, does not necessarily mark ‘‘success’’ relative to another, such as democratization.

In this chapter, I address some of the core dilemmas confronting post-war elections that flow from these multiple and sometimes contradictory goals.1 I focus in particular on those specific dilemmas which tend to recur across both space and time: the inherent tension between competitive elections and conflict management; the incompatibility of short- versus long-term electoral objectives; the trade-off between efficiency and inclusion in terms of government structure; the merits of sequenced versus simultaneous local, regional, and national-level elections; the choice between party-based and independent forms of electoral administration; and the need to build local accountability while encouraging the development of national party politics.

Democratic dilemmas and post-war elections The overarching question facing all post-war elections is under what circumstances they help in building a new, peaceful, democratic order, and under what circumstances they undermine prospects for stable democracy and pave the way for a return to conflict. As one survey notes, the high expectations often placed upon post-war elections tend to be accompanied by a weakness in the preconditions for their success: ‘‘most war-torn societies lack the political climate, social and economic stability, institutional infrastructure, and even political will to mount successful elections’’ (Kumar 1998a: 7). The international community 1

This chapter also builds upon some of my earlier work focusing on other aspects of post-war elections: see, in particular, Reilly 2002 and 2004.

160

Benjamin Reilly

has often not been sufficiently cognizant of the dangers in pushing for early post-war elections, particularly in countries which have recently emerged from civil war, or given sufficient attention to the capacity of the host country to carry them out. On the other hand, the promise of early elections is often essential in getting commitments from major powers to deploy peacekeepers and fund post-war reconstruction. There are fundamental difficulties with holding competitive elections following a period of violent conflict. In such situations, a combination of acute coordination problems, information asymmetries, hardening of societal divisions, and fears for the future typically confront voters. One consequence of this can be a profound ‘‘security dilemma’’ which afflicts both voters and candidates, whereby competing ethnic, religious, and political actors will often mobilize against the possibility of future threats, triggering a cascading tit-for-tat escalation and polarization from other segments of society. In many cases, rising levels of internal conflict have accompanied or been precipitated by transitions from authoritarian rule toward democracy. Despite their essential role, post-war elections have often fomented these tensions, becoming a lightning-rod for popular discontent and extremist sentiments (Dahl 1971; Horowitz 1985; Rabushka and Shepsle 1972). Bosnia’s repeated post-Dayton elections held in 1996, 1998, 2000, and 2002 were an illustration of this process in action, as voters from different ethnic communities persistently re-elected hard-line nationalist leaders despite overt attempts by the international community to encourage moderate, pro-Western victors instead. The victory of Hamas, which the United States considers a terrorist organization, at the January 2006 elections to the Palestinian Authority is another case in point. Electoral competition in such circumstances often turns on the politics of ‘‘outbidding’’ – that is, competition for votes on the basis of reciprocally heightened extremist rhetoric – increasing ethnic tensions, and the polarization of the political spectrum (Rabushka and Shepsle 1972: 187). Taken to an extreme, outbidding cycles can lead to the outright failure of democracy, as minorities are excluded from power, restrictions are placed on opposition movements, and the ‘‘rules of the game’’ manipulated to benefit incumbents. For instance, political outbidding over access to higher education was an early motivator for the Sinhalese–Tamil conflict in Sri Lanka, fomenting a cycle of escalating ethnic hostilities which led directly to the tragic civil war and erosion of

Post-war elections

161

democracy there (DeVotta 2004). In other cases, by electing hard-line leaders committed to exclusionary visions of the country, post-war elections can become little more than ethnic censuses which increase the risk of conflict, threatening, ultimately, the failure of democracy itself – a recurrent problem that has been well documented in ethnically divided societies.2 In this view, the processes of electoral democracy in post-war societies can represent a danger to peaceful state building. Increasing recognition of these dangers has led some to contend that democracy itself is part of the problem in such highly fraught situations, and that post-war societies are too fragile to be exposed to the competitive pressures of the electoral process.3 But this oft-heard critique ignores several factors. First, elections can be purposively designed to encourage not zerosum, winner-take-all outcomes, but rather the sharing of power between groups. Indeed, many would argue that some form of power sharing is a primary requirement for successful democratization in post-war situations.4 Second, post-war polities face a real need to construct a legitimate governing authority. Not least because so many of today’s conflicts take place within states, the overarching challenge is thus to build (or rebuild) a state that can function without direct international involvement. Elections can be a crucial element in achieving this, providing that sufficient forethought is given to their purpose, timing, and likely effects.

Post-war elections and international policy Post-war democracy building is a difficult and disruptive process. Democratization by its very nature undermines established political orders, provides a pathway for new entrants to access the political system, highlights social cleavages, subverts existing power relations, and threatens incumbent authority. For all of these reasons, transitions to democracy in general and competitive elections in particular have the potential to be deeply destabilizing events. As Roland Paris (2004: 1) observes, ‘‘the process of political and economic liberalization is inherently tumultuous: It can exacerbate social tensions and undermine the 2

3

Horowitz 1985. Recent research on India has improved our understanding of this process: see Wilkinson 2004 and Chandra 2004. For an example of this sentiment, see Chesterman 2004. 4 See Sisk 1996.

162

Benjamin Reilly

prospects for stable peace in the fragile conditions that typically exist in countries just emerging from civil war.’’ Put simply, post-war democratization is a difficult, uncertain, and often dangerous business. Despite this, over the past decade many Western policymakers adopted a facile and naı¨ve interpretation of democratic elections as being a natural and unproblematic form of conflict resolution, a process which began in the 1990s and may have reached its nadir with the truly extravagant claims regarding the beneficent impact of democratization and free elections made by the Bush administration in the postSeptember 11 era. Post-war elections in Iraq, for instance, were not just a means of choosing representatives but also highly symbolic events signaling the establishment of a new political order. In other cases such as East Timor, post-war elections have even marked the assumption of a new nation into the family of international statehood. However, post-war elections can also be highly fraught exercises which highlight many of the dilemmas of democratization examined in this volume. While elections certainly may play a role in settling violent conflicts and creating new political orders, they will not inevitably do so, and the multiple goals that post-war elections are supposed to achieve means that they are often overloaded with inconsistent and sometimes mutually conflicting objectives, such as ending armed conflict and simultaneously promoting vigorous political competition. The uncritical alignment between democratic elections and national peace in the minds of many policymakers has its recent origin in the so-called ‘‘democratic peace’’ thesis much cited by former US President Bill Clinton during his term in office, which maintains that consolidated democracies both have not historically and will not in the future go to war with each other (Russett 1993). This thesis, which has strong empirical support but shaky theoretical foundations, was conjoined with the separate but related argument that democracies were more peaceful internally as well, and that intrastate conflicts were thus less prevalent and severe in democratic than autocratic environments (see, for example, Gurr 2000b: 52–64). Both arguments have since been overtaken in US policy by a broader conflation between democracy, liberalism, and security, which was elevated into an article of faith in the Bush administration – a conflation of hopes and beliefs which rests on many unspoken assumptions. One is the expectation that a move from authoritarian to democratic governance will inevitably lead to more peaceful inter-communal relations and lower levels of conflict

Post-war elections

163

in societies divided along ethnic, religious, or other lines. Another is that democratic elections are the most reliable means of generating moderate governments which do not pose a threat to others, or at least to the West. The evidence for both claims is (unsurprisingly) more complex. While there is little question that consolidated democracies are, on average, both less prone to large-scale internal conflict and much less likely to go to war with each other than their authoritarian counterparts, historical and comparative research has repeatedly found that countries undergoing the wrenching process of democratization are neither. As one such study concluded, ‘‘while mature, stable democracies are safer, states usually go through a dangerous transition to democracy. Historical evidence from the last 200 years shows that in this phase, countries become more war-prone, not less, and they do fight wars with democratic states’’ (Mansfield and Snyder 1995b: 79). Many indicators of intrastate conflict also tend to rise in the initial period of democratization (de Nevers 1993). A major insight of political science scholarship in recent years thus concerns the dangers of early democratization in fragile states.5 In such circumstances, electoral competition can quickly come to be characterized by centrifugal pressures, in which the moderate political center is overwhelmed by extremist forces, leading to zero-sum politics in which some groups are permanently included and some permanently excluded. One consequence can be the failure of democracy itself, as majority group hegemonic ‘‘control,’’ often aided by the assumption of martial law or outright military rule, is justified in part by the need to restore order and stability. This is often accompanied by the exclusion of minorities, changes to the rules of the game to benefit incumbents, restrictions on opposition movements, and the undermining of democratic institutions. The failure of Indonesia’s first, abortive experience of democracy in the 1950s period is a good example of this dilemma of early democratization before civic institutions developed. Indonesia’s fall into authoritarian rule was largely a response to the chaotic experience of democracy between 1950 and 1957, when a combination of 5

In a book-length analysis of this phenomenon, Snyder (2000) argues that democratization is most likely to stimulate internal conflict when elites are threatened by rapid political change and when the expansion of popular participation precedes the formation of strong political institutions.

164

Benjamin Reilly

religious, cultural, and regional conflicts, combined with a fragmented party system and weak institutions, led to the declaration of martial law by President Sukarno. Shifting coalitions of secular, Islamic, nationalist, communal, and regional parties had precipitated six changes of government in seven years, providing a ready pretext for the overthrow of democracy and forty years of authoritarian rule under the Sukarno and then Suharto regimes (Liddle 1997: 311). Rapid political change and insecurity are almost endemic to the circumstances in which post-war elections are held. Domestic political institutions are weak or non-existent, voters are suspicious, and elites’ hold on power is tenuous. This is a dangerous combination of factors. At different times during the 1990s, for instance, post-war elections in war-torn African states such as Angola and Sierra Leone led to a resumption of warfare as a result, in part, of the threats these elections represented to incumbent elites. Likewise, the prospect of forthcoming elections and ethnic power sharing in Rwanda has been identified as a factor in the 1993 genocide there.6 Elsewhere, in Cambodia, Bosnia, and Liberia, post-war elections cemented in power essentially nondemocratic elites – often the very same individuals who had been instrumental in the prior conflict. Finally, in post-war Rwanda, Ethiopia and Uganda, the winning parties to the conflict held elections to legitimize their victory – thus subverting the conflict-mediating functions which such elections were supposed to perform. Cases such as these lend some weight to the World Bank’s suggestion that elections in war-torn societies should be deferred by up to a decade to allow state building to occur (World Bank 2003a). None of this should be taken to mean that democracy is a negative factor for the management of internal conflicts. By providing an institutional framework for diverse social groups to gain access to government, participate in decision making, and influence policy outcomes, democracies are capable of responding to societal conflicts by accommodation rather than repression, in sharp contrast to authoritarian 6

Paris (2004: 75–76), for example, writes that ‘‘the effort to move Rwanda in the direction of democracy did not have the pacifying influence that international peacebuilders had hoped for and apparently expected. The international community had presented the plan for power sharing followed by democratic elections as a means of resolving Rwanda’s civil war, but attempts to foster peace and stability in Rwanda by promoting political liberalization ultimately backfired, in the worst possible way.’’

Post-war elections

165

regimes. This is one reason that theorists like Adam Przeworski characterize democracy as a political arrangement which processes, but never definitely resolves, social conflicts (Przeworski 1991: 10–14). Under this interpretation, a functioning democracy serves as a system of conflict management, with potential conflicts channeled into constitutional arenas, such as non-violent competition between political parties, rather than armed conflict on the streets. These arguments have been buttressed by empirical studies which emphasize the success of consolidated democracies in accommodating social cleavages and tensions through peaceful means (Hegre et al. 2001: 33–48). However, Przeworski’s arguments also highlight one – indeed, perhaps the most fundamental – of the dilemmas facing post-war elections: the problem of uncertainty. Consolidated democracy works to manage conflicts precisely because electoral outcomes are, by definition, uncertain: the players of the game do not know the outcome, but commit to the game with the understanding that they may lose in the short term but still be winners later on. Democracy requires this certainty of uncertainty to engender loyalty from all players and thus to survive over the long term. But in post-war societies, the uncertainty of election outcomes is itself a source of tension, and a major threat to incumbent elites which can make them wary of committing to the game at all. For this reason, some kind of pre-election bargain on postelection outcomes is often required, as was the case in South Africa’s transition from apartheid, when strong constitutional guarantees helped to reassure the key players (Sisk 1995). The aftermath of the 1993 elections in Cambodia is a good example of this dilemma in action. The culmination of the largest and most expensive UN peacekeeping mission to date, the massive international reconstruction of Cambodia reached its apex in 1993 when the United Nations both ran and oversaw an electoral process which was technically almost faultless, but which resulted in the election of two main parties, the Cambodian People’s Party (CPP) and the National United Front for an Independent, Neutral, Peaceful, and Cooperative Cambodia (FUNCINPEC), each of whom had expected to control power alone. Amid threats of renewed civil war from the incumbent CPP if it was excluded from government, a clumsy post-election powersharing deal brokered by the United Nations saw a coalition government featuring ‘‘co-prime ministers’’ from the two parties installed. As it reflected neither the election results nor common policy ground

166

Benjamin Reilly

between the two parties, this arrangement proved highly unstable in practice: the CPP remained in effective control of most of the armed forces, the bureaucracy, and the judiciary, while FUNCINPEC’s attempt to gain a greater share of real power paralyzed the executive branch and the National Assembly. After a prolonged succession of political crises, the coalition fell apart completely in 1997 when the CPP forces of the ‘‘second Prime Minister,’’ Hun Sen, attacked those of FUNCINPEC and the ‘‘first Prime Minister,’’ Prince Ranariddh, and claimed power alone. As Cambodia returned to its familiar politics of intimidation and authoritarian rule, Hun Sen proceeded to change the electoral system to benefit his government and restrict opposition movements. The electoral formula was changed so that seats were allocated according to the ‘‘highest average’’ method at the provincial level, rather than the nationwide ‘‘largest remainder’’ system introduced by the United Nations in 1993 – a change which wiped out smaller parties. In response to calls for greater local accountability, district boundaries were adjusted and a number of new districts created with the result that over one-third of all seats were chosen from single-member districts. These technical changes were accompanied by widespread intimidation of opposition politicians and their supporters. The net effect was the elimination of most opposition parties, to the advantage of the larger incumbents. At the 1998 elections, which the CCP won outright, calculations suggest that up to ten additional parties would have gained representation had the election been held under the 1993 electoral laws.7 Despite this, a range of international observer groups gave their stamp of approval to the 1998 elections, with the European Union particularly keen to see Cambodia’s re-elected government receive international blessing, regardless of the sharp decline in the quality of its democracy.

Short-term versus long-term objectives Despite such setbacks, there is also evidence of learning by the international community regarding the dilemma posed by the disjuncture between the short-term and longer-term objectives of post-war democratization. We now have a much deeper understanding of the 7

My thanks to Michael Maley for the data on this point.

Post-war elections

167

relationship between civil society, political institutions, and international actors as agents of post-war peacebuilding than before. Scholars and policymakers alike increasingly appreciate the importance of select incentives in moving societies from war to peace, and the self-reinforcing role of democratic procedures in helping them stay there. Yet one crucial aspect of post-war democracy building continues to be overlooked: the temporal dimension of democracy building. Given that democracy is a long-term process of political development, a key dilemma common to nearly all post-war elections is the trade-off between the short-term goals of war termination and the longer-term goals of democracy building. Over the course of the 1990s, the United Nations and other international bodies developed a kind of standard operating procedure for post-war peacebuilding as part of the new global consensus on the virtues of democracy (Newman and Rich 2004). Once a minimum level of peace had been obtained (which did not necessarily mean a full cease-fire agreement), and a basic level of infrastructure was in place, the next step was usually to hold some kind of elections – often within a year or two of the start of the mission – followed by a rapid hand-over to the newly elected local authorities, and an even more rapid departure of international troops and personnel. Under this approach, elections came to be seen as the crowning event of the post-war peacebuilding phase, enabling the reestablishment of legitimate domestic authority, and allowing international forces to disengage and, in most cases, depart. Several dilemmas are inherent in this new model of state reconstruction. One is the question of election timing. How early a fragile state should hold elections or referenda represents a fundamental choice facing almost all post-war peacebuilding and democratization efforts. Hasty or rushed ‘‘instant elections’’ have become common for several reasons: the need to ‘‘do something’’ quickly, to start the process of political development and, of course, to have an identifiable ‘‘exit strategy’’ for international involvement. A classic case is the November 1996 elections in Bosnia, a date dictated in large part not by events on the ground but rather the desire of the Clinton administration to show progress in the Balkans in time for mid-term elections in the United States. More diffuse temporal concerns also encourage premature elections. Democracy, as the mantra goes, is a long-term process, but the domestic

168

Benjamin Reilly

political pressures that weigh on the Western states that usually fund and implement peacekeeping missions are almost all short term. Quick results are required. An early establishment of home-grown institutions, such as representative legislatures and multiethnic peace and security forces, is needed to create both the shell of a state and to create a legitimate body politic for the international community to deal with. Financial pressures mean that many missions have an incentive to scale back their immediate presence and reduce the levels of their assistance as soon as it is minimally feasible to do so. All of this places considerable pressures on the post-war election timetable and frequently leads to elections being held as early as possible in the life of a peacekeeping mission in order to create some kind of legitimate government – a pressure which in Iraq led to elections being held in 2005 in the absence of popular security and in the face of a boycott from one of the country’s main ethnic groups, the Sunni. Such ‘‘premature elections’’ can also create multiple, ongoing problems for the development of peacetime politics in deeply divided societies even years after the war has ended – as demonstrated by the regular re-election of ethnic hard-liners in post-war Bosnia, where nationalist parties and elites not only have continued to be elected by the voters, but have attempted to use the democratic political process to press their sectarian aims. In general, the early application of elections immediately following a conflict increases the likelihood that the contest will become a de facto contest between the former warring armies masquerading as political parties. By contrast, an extended process of consultations and local-level peacebuilding, in which some of the real interests and concerns that provoked the conflict are addressed in a step-by-step fashion before national elections are held, may offer better prospects for a peaceful transition in post-war societies.8 A related dilemma is the virtue of referenda on deeply divisive issues, such as independence or self-determination, compared to the virtues of constructing a phased series of consultations rather than one take-it-orleave-it choice – the latter exemplified by the rolling series of independence referendums in the early 1990s which precipitated the breakup of Yugoslavia, or the 1999 autonomy plebiscite in East Timor. Because they channel complex questions of political identity into an all-ornothing, zero-sum choice, the comparative evidence suggests that 8

For a survey of these, see Harris and Reilly 1998.

Post-war elections

169

referenda are unsuited to solving deeply divisive issues and are particularly inappropriate for the combustible conditions which apply in post-war societies (Reilly 2003a). These failures suggest the need to give more attention to alternative models of post-war transition and methods of self-determination, such as those employed in the powersharing and autonomy arrangements in the Bougainville conflict, in which time periods were deliberately lengthened or left unspecified. The success of the peacemaking process in Bougainville, an island at the eastern extremity of Papua New Guinea that was the site of the largest and most violent conflict in the South Pacific since World War II, deserves more international attention than it has received to date. The conflict centered around demands for Bougainville’s independence made by rebel groups, a demand opposed not only by Papua New Guinea but also many Bougainvilleans themselves. Before the war began, Bougainville’s substantial contributions to the national economy was disproportionate to its small size and population, mainly due to an enormous open-cut copper, gold, and silver mine that operated on the island from 1972 until the conflict caused its closure in 1989. The cascading violence came to a head in 1997, when the Papua New Guinea Government commissioned an international mercenary service, Executive Outcomes, to attack the rebels. But in a surprise move, the Papua New Guinean army’s chief commander announced the refusal of his forces to work with the mercenaries, who were ejected from the country, and the incumbent Prime Minister and several key ministers involved in engaging the mercenaries stood down. Taking advantage of these changes, rebel forces began to make direct contact with the central government. Further developments resulted in the New Zealand Government facilitating talks between the Bougainvillean leaders. These talks resulted in a cease-fire agreement, followed by the deployment of an unarmed ‘‘Peace Monitoring Group,’’ led by Australia, on the island, accompanied by a UN observer team. Successive agreements – notably the so-called ‘‘Loloata Understanding’’ of March 2000 and the Bougainville Peace Agreement signed at Arawa in August 2001 – paved the way for the election in 2005 of an autonomous government and agreement on a future referendum on Bougainville’s political status to be held after an extended period of autonomy. Importantly, both the election of the autonomous government and the longer-term referendum plans involve an extended time period, in sharp contrast to the rush to elections that has taken place in Angola

170

Benjamin Reilly

and the Balkans. Bougainville’s first autonomous government elections were not held until June 2005, four years after the signing of the Bougainville Peace Agreement, while the timing of the independence referendum is set at some unspecified time in the future, after some ten to fifteen years of autonomous government. This drawn-out timetable appears to have played a role in allowing many local-level peace initiatives, which would otherwise have been subsumed to electoral considerations, to flourish. It also allowed a series of innovative reforms to the electoral process to be introduced: Bougainville’s new autonomous parliament reserves seats for specific regions, former combatants, and women (the last of whom have played a particularly important peacemaking role). These various innovations have helped to deliver one of the more successful post-war electoral processes of recent years. In a further institutional innovation, future Bougainville elections will be held under the alternative vote, a ‘‘vote-pooling’’ electoral system which some scholars advocate as a means of promoting moderation and accommodation in divided societies by coercing cooperation across ethnic lines (Horowitz 1985; Reilly 2001). A final key temporal issue, which confronts many post-war electoral processes, is the sequencing of local, regional, and national elections. The coordination of election timing at the national and sub-national level directly affects the development of local and national-level parties and the extent to which national or regional consciousness takes hold politically. Some scholars such as Juan Linz and Alfred Stepan argue that new democracies should hold national elections first, before regional or local ones, in order to generate incentives for the formation of national, rather than regional, political parties (Linz and Stepan 1996: 98–107). Others such as Larry Diamond believe that simultaneous national and local elections ‘‘can facilitate the mutual dependence of regional and national leaders. The more posts that are filled at the regional and local level . . . the greater the incentive for regional politicians to coordinate their election activities by developing an integrated party system’’ (Diamond 1999b: 158). This was the approach taken in post-Suharto Indonesia, with identical party-based ballots being presented to voters at simultaneous elections for national, provincial, and local assemblies – a strategy which strengthened the position of nationally focused parties. In recent transitional elections in East Timor, as well as other postwar cases such as Kosovo and Afghanistan, the decision was made to

Post-war elections

171

start at the local level first: rather than leading with national elections, the preferred sequence was to hold local or municipal elections as a precursor to national ones, allowing steps toward democratization to be taken gradually. The relative success of these cases suggests that scholars such as Linz and Stepan are likely mistaken in advising that national elections should be held before local ones, at least if the aim is to inculcate voters in the routines of electoral politics and party politics in new democracies.9 For transitional elections, a first-run test of municipal or local elections before national ones appears to have much to recommend it. However, if party building is the aim, then simultaneous national, regional, and local elections are likely to be more favorable to the development of strong nationwide parties. There is evidence of genuine learning over time by the UN and other international actors on most of these issues, with more recognition of the need for sustained international involvement for several years after a conflict rather than the rushed ‘‘in-and-out’’ approach of former years. In major international assistance operations such as Kosovo, East Timor, and Afghanistan, pressure to hold ‘‘instant’’ national elections has been resisted in favor of a two-year period of political development as part of a much longer process of democratization. In both Kosovo and East Timor, relatively peaceful national elections, which had been preceded by successful municipal polls, were held in 2001. In Afghanistan, presidential elections were held in 2004, almost three years after the fall of the Taliban, while the parliamentary poll was further postponed until September 2005. Even the 2002 Loya Jirga process in Afghanistan – which brought Afghan tribal representatives and elected delegates together to choose an interim government in a process that was only partially democratic – can be seen as a kind of local election. By contrast in Iraq, national elections to a constituent assembly were pushed through within a year of the conflict being declared over by the US president. In general, the comparative evidence suggests that a bottom-up approach to electoral timing is probably the best way to encourage the development of party politics and to inculcate voters in the routines of electoral politics. 9

Indeed, of the three cases cited by Linz and Stepan in support of their argument – Spain, Yugoslavia, and the USSR – only Spain held its first truly competitive elections at the national rather than the regional level. My thanks to Bethany Lacina for bringing this to my attention.

172

Benjamin Reilly

Electoral mechanics: efficiency versus inclusion A recurring dilemma animating the choice of political institutions in all democracies, including post-war states, is the trade-off between stability and governability versus inclusion and representation. Classically, ‘‘representational’’ institutions are thought to best ensure the direct translation of popular preferences and cleavages into the political sphere via political parties representing distinct social groups, proportional elections to promote the representation of minorities, and low thresholds or other barriers on the formation of new parties. Together, these institutions should ideally lead to the development of a diverse multi-party system in which all significant social groups and interests are separately represented. By contrast, ‘‘efficient’’ institutions that can deliver clear parliamentary majorities to disciplined political parties offering distinct policy alternatives are more likely to be associated with majoritarian electoral laws and the presence of ‘‘catch-all’’ parties, which can command electoral support across social cleavages. These scholarly debates have direct implications for institutional choices – particularly the choice of electoral system, which has long been recognized as one of the most important institutional choices affecting the nature of democracy. They can have profound implications for the extent to which the voices of the poor and other marginal groups can be heard and their power enhanced. For example, systems in which the parliament is elected from many small geographically defined electoral districts tend not to be as good at representing minority opinion than proportional ones, but may be better at building links of local accountability. These choices can also influence other aspects of the political system, such as the development of the party system, linkages between citizens and their leaders, political accountability, representation, and responsiveness. Because of such impacts, constitutional and electoral system choices have many long-term consequences for the process of democratic governance, and the choice of electoral system is one of the most important political decisions for any country. Electoral systems are often categorized according to how proportionately they operate in terms of translating votes cast by electors into seats won by parties. A typical three-way structure divides such systems into plurality-majority, semi-proportional, and proportional representation (PR) systems. Plurality-majority systems typically give more

Post-war elections

173

emphasis to local representation via the use of small, single-member electoral districts than to proportionality. Amongst such systems are plurality (first-past-the-post), runoff, block, and alternative vote systems. By contrast, proportional representation systems – which typically use larger, multi-member districts and deliver more proportional outcomes – include ‘‘open’’ and ‘‘closed’’ versions of party list PR, as well as ‘‘mixed-member’’ and ‘‘single transferable vote’’ systems. Semiproportional systems such as the single non-transferable vote offer yet other approaches, as do various mixtures of plurality and proportional models – such as the ‘‘mixed’’ models by which part of the parliament is elected via PR and part from local districts, a common choice in many new democracies over the past decade.10 Most of the major transitional elections conducted in recent years, including almost all of those held under UN auspices, have utilized some form of proportional representation. Prominent transitional electoral operations in Namibia (1989), Nicaragua (1990), Cambodia (1993), Mozambique (1994), Liberia (1997), Bosnia (1996, 1998, 2000, 2002), Kosovo (2001), Sierra Leone (2002), Rwanda (2003), and Iraq (2005) were all conducted under PR rules. In particular, the simplest form of proportional representation – party-list PR – appears to have become the de facto norm for UN-administered elections. But the adoption of such systems for post-war elections has usually been dictated more by administrative concerns, such as the need to avoid demarcating individual electoral districts and to produce separate ballot papers for each district, than these wider political issues. Indeed, in many post-war elections, national PR systems are the only feasible way to hold an election, as a uniform national ballot can be used, no electoral districts need be demarcated, and the process of voter registration, vote counting, and the calculation of results is consequently simplified. In Liberia in 1997, for example, population displacement and the lack of accurate census data led to the abandonment of the old system of single-member majoritarian constituencies in favor of a proportional system with a single national constituency. Indeed, the inclusion of internally displaced peoples, refugees, and ex-combatants in the electoral process is a recurring dilemma of post-war elections generally (see Lacy 2004).

10

For a survey of these options, see Reynolds, Reilly and Ellis 2005.

174

Benjamin Reilly

For Iraq’s crucial 2005 elections, experts favored a system based around provincial boundaries, to ensure greater accountability and representation of local constituencies. However, this would have entailed a lengthy national census. In the interests of time, it was therefore decided to fall back on a single, nationwide district elected by proportional representation in which 1/275th of the vote was sufficient to gain a seat in the constituent assembly. While this doubtlessly facilitated the administration of the election itself, it also had the effect of fragmenting the legislature and marginalizing numerically smaller groups like the Sunni, while encouraging ethnic polarization amongst the electorate. Many new democracies have therefore preferred ‘‘mixed’’ electoral systems, in which part of the legislature is elected on a national level by proportional representation, and part at the local level from single-member districts, so that both proportionality and accountability are maximized. For example, at the August 2001 elections for East Timor’s 88-member constituent assembly, most seats were elected on a nationwide basis by list PR, but there were also separate single-member electorates corresponding to each of the country’s thirteen districts. A similar system in Iraq may have guaranteed the Sunni minority a baseline of political representation at the provincial level, thus helping to assuage the political alienation which is at the root of Iraq’s insurgency (Diamond 2005: 269). Perhaps the most unusual electoral system choice for a post-war election in recent years has been the decision to use the single nontransferable vote (SNTV) for the September 2005 parliamentary elections in Afghanistan. Under SNTV each elector has one vote, there are several seats to be elected in the district, and the candidates with the highest number of votes fill these positions. As a result, the number of candidates a party nominates in each district becomes a critical choice: too few, and parties miss out on valuable chances to win additional seats; too many, and they risk splitting their vote too thinly and losing winnable seats. Despite being structurally majoritarian, SNTV can thus advantage smaller parties and deliver relatively proportional election outcomes. However, by forcing candidates from the same party to compete against each other for the same pool of voters, personalistic attributes are emphasized over and above those of the party. The resulting candidate-centered, intraparty competition has been widely identified as a cause of factionalism, corruption, and clientelistic politics in states like Japan, where the abandonment of SNTV in 1994 was

Post-war elections

175

fuelled by a series of corruption scandals linked to factional competition which damaged confidence in the political system. While having the virtue of simplicity, these pathologies mean that it is also an extremely risky choice for a new democracy, particularly if encouraging cohesive national political parties is an objective. In an ethnically complex, clan-based society such as Afghanistan’s, SNTV makes it unlikely that a consolidated party system can develop in the short term. Illustrating this, the 2005 Afghan parliamentary elections featured over 5,800 candidates – in Kabul alone the ballot paper displayed over 400 names – resulting in a fractionalized and incoherent parliament which is likely to remain highly divided and unable to coordinate around pressing policy challenges.11

Electoral administration: independent or party-based? Another dilemma of post-war elections is the role and responsibilities afforded to the bodies charged with running the elections themselves. While constitutional and electoral reforms have attracted a voluminous academic literature, issues of electoral administration remain understudied by scholars and under-rated in general in terms of their effect on post-war polities. There are several models of election administration used around the world. Some countries locate responsibility for the administration of elections within a government portfolio like the interior or home affairs ministry. Others situate the responsibility for administration of elections within government agencies such as the public records office, the tax department, or even the postal service. In some countries, the body responsible for running elections is created anew before each electoral event. And in some cases, as in Cambodia in 1993 or East Timor in 2001, the United Nations itself takes responsibility for running the elections. Probably the most important administrative decision concerns the composition of the body managing the elections, and specifically whether the elections are run by the government of the day or by some form of independent electoral commission. Their perceived neutrality and independence from political interference lends credibility to the electoral process, which is a crucial determinant of the success of any election. A truly independent commission is one that is able to 11

See ‘‘Democracy, sort of,’’ The Economist, September 24, 2005, 34.

176

Benjamin Reilly

operate effectively without direct ministerial control, including in terms of its financial and administrative functions, and is (ideally) composed of non-partisan appointees. In practice, many independent commissions around the world do not have complete financial independence and may comprise party representatives rather than nonpartisan appointments. However, they are still able to operate free from government interference or control. By contrast, in some countries electoral management bodies are composed not of independent civil servants, judges, or other officials, but rather of the political parties contesting the elections themselves. This practice can provide a form of non-partisan independence if the composition of party representation is balanced in such a way as to ensure genuinely neutral functioning. The influence of the United States is particularly important here, as the American form of electoral administration is based on political appointees and party representatives, and many post-war democracies, particularly in Latin America, have followed this model for their own elections. Some authorities argue that, when there is no better tradition or an existing body of widely respected independent civil servants, a party-based electoral authority may be the only realistic choice (Lo´pez-Pintor 1998: 53). But recent problems with this model in important transitional elections such as Indonesia and Haiti, as well as in established democracies (most notably, the 2000 presidential poll in the United States), emphasize its propensity for politicization and deadlock. Most established and emerging democracies have chosen non-partisan models of electoral administration. Indeed, since the world’s largest democracy, India, adopted this model at independence there is a clear trend toward the adoption of independent electoral commissions staffed by non-partisan civil servants. The comparative evidence strongly favors independent commissions run by apolitical civil servants, and the United Nations now explicitly advocates this model wherever possible. Party-based commissions have an almost inevitable tendency to split along party lines. In Haiti, for example, the Provisional Electoral Council was made up of representatives of the political parties but was also deeply divided along party lines, and internal mistrust and divisions prevented it from working efficiently (Nelson 1998: 76). In Cambodia, by contrast, a non-partisan electoral commission was widely seen as one of the outstanding elements of the entire United Nations transitional administration and elections of

Post-war elections

177

1993. Non-partisan commissions were also a prominent and successful part of United Nations missions in Namibia and in East Timor. The dangers of using party-based electoral administrations in transitional situations were graphically demonstrated by Indonesia’s transitional elections in 1999. Amid the flowering of new political movements that accompanied the democratic opening, a requirement that all political parties be represented on the National Elections Commission (KPU) resulted in a deadlocked and unwieldy body of no fewer than fifty-three officials, most of them party representatives (including some individuals who were also candidates for the election). As a result, during the preparation for one of the most important transitional elections of the 1990s, the body charged with running the elections was almost completely dysfunctional, being deeply divided along party lines and unable to take even basic decisions (at one stage, fist-fights broke out between different members of the commission). Following the elections, the Indonesians moved quickly to discard the party-based KPU and replace it with a much smaller, non-partisan body of eleven non-party and non-government representatives. Such cases underline that independent and permanent electoral management bodies are a clear best practice of electoral administration, a conclusion which has also been reinforced by a global study of electoral management bodies (Lo´pez-Pintor 2000).

Encouraging national party politics A final dilemma confronting post-war elections is the nature of the emerging party system and the extent to which party politics becomes institutionalized. Scholars of democracy have long considered political parties to play a crucial role not just in representing interests, aggregating preferences, and forming governments, but also in managing conflict and promoting stable politics. As the key agents of political articulation, aggregation, and representation, political parties are the institution which impact most directly on the extent to which social cleavages are translated into national politics. Parties perform a number of essential functions in a democracy: ideally, they represent political constituencies and interests, recruit and socialize new candidates for office, craft policy alternatives, set policymaking agendas, form governments, and integrate disparate groups and individuals into the democratic process (Diamond 1997b: xxiii). These linking, mediating,

178

Benjamin Reilly

and representational functions mean that political parties are one of the primary channels for building accountable and responsive government in new democracies. This stylized depiction of the roles parties play in terms of democratic consolidation, however, can be undermined by the reality of communalism, clientelism, and other forms of particularistic politics. In post-war situations, party politics tends to reflect the social cleavages which created the conflict in the first place. If the conflict had a strong ethnic dimension, for instance, then these differences will tend to be reflected in the new democratic system, particularly if ‘‘ethnic parties’’ are allowed to form freely. The presence of such parties, in turn, can quickly incite intergroup competition while inhibiting cooperation. As Gunther and Diamond write, ‘‘The electoral logic of the ethnic party is to harden and mobilize its ethnic base with exclusive, often polarizing appeals to ethnic group opportunity and threat . . . the ethnic party’s particularistic, exclusivist, and often polarizing political appeals make its overall contribution to society divisive and even disintegrative’’ (2001: 23–24). At the margins, the presence of such parties can lead to what Sartori dubbed ‘‘polarized pluralism,’’ where the ideological distance between the parties expands, to the detriment of the political center. Indeed, in Western democracies, the presence of parties with extremely divergent policies and preferences has historically been an important predictor of political instability. For this reason, many scholars of ethnic conflict advocate the need for broad multiethnic parties or coalitions of parties as a key mechanism for ameliorating conflict in ethnically divided societies. In such party systems, elections tend to be fought out between a small number of relatively large and cohesive parties, and politicians ‘‘crowd the center’’ in their quest for the median voter, avoiding sharp differentiation with their competitors. As a result, ‘‘they tend to have a moderating influence on the way interests are aggregated’’ (Haggard 1997: 140). But forging centrist, programmatic, political parties in a postwar society is easier said than done, as parties often spring from the same cleavages and tensions which spurred the original fighting, leading to a continuation of the former conflict through the new democratic process.12 Increasing awareness of the problems caused by such polarized or otherwise dysfunctional party systems has lately spurred 12

See So¨derberg Kovacs’ chapter in this volume.

Post-war elections

179

multilateral bodies such as the United Nations – which have traditionally been wary of direct involvement in politics, preferring more traditional kinds of development assistance – to take a more active role in assisting political party development in some countries.13 The most ambitious actors in this field have been the international democracy promotion organizations which have proliferated over the past decade (Carothers 1999). Because they are not bound by the same strictures as multilateral agencies, some of these have attempted to intervene directly in the development of the party system in recipient countries. In Bosnia, for example, the US National Democratic Institute openly and actively promoted putatively multiethnic parties such as the Unified List coalition in preference to nationalist parties such as the Serb Democratic Party (SDS) or the Croatian Democratic Union (HDZ) at the 1996 elections (see National Democratic Institute 1996). Also in Bosnia, a range of related reforms to the electoral system and other areas introduced in recent years by the OSCE have attempted to undercut nationalist parties by changing voting procedures and, in some cases, barring candidates from election (Belloni 2004). Kosovo also saw overt attempts by the international community to mandate multiethnicity in the political system (Simonsen 2004). However, despite some inflated claims to the contrary, the success of such interventions has been modest, and ethnic parties continue to dominate the political landscape. The vexed problem of transforming armies into parties after a protracted period of conflict continues to trouble international interventions in this field. As one survey of post-war elections concluded, ‘‘Democratic party building is proving to be a slow process. In all the [post-war] countries, political parties are organized around personalities, narrow political interests, and tribal and ethnic loyalties’’ (Kumar 1998b: 218). Historically, the most successful example of such a transition is probably the armies-to-parties transformation wrought by the United Nations in Mozambique, where a special-purpose trust fund and some creative international leadership succeeded in bringing the previous fighting forces of the Liberation Front of Mozambique (FRELIMO) and particularly RENAMO into the political fold. Financing political party development has been an important element of a number of other post-war elections. One approach involves 13

For a survey of these approaches, see Reilly 2006.

180

Benjamin Reilly

channeling technical or financial assistance from international donor agencies, non-governmental organizations, or multilateral agencies to party organizations in those states in which the international community has taken a prominent role, such as countries emerging from a period of violent conflict. Recent proposals for political party assistance in Afghanistan have also focused on this kind of approach. Despite the widespread agreement on the importance of strong parties for a functioning democracy, there are dilemmas inherent in the encouragement of broad-based political parties too. Post-Suharto Indonesia, for example, has seen an ambitious exercise in ‘‘political engineering’’ to promote broad-based parties with a national focus and hamper separatist groups through a complex collection of incentives and restraints on party system development, which requires parties to establish local branches in half of all provinces and municipalities. The bias in favor of national parties was so strong that regional parties were even banned from competing in elections to the regional assemblies, where again only national-level parties were permitted. While encouraging broad-based parties, these provisions also represent a restriction upon new entrants into the political system and may end up encouraging extra-constitutional action by aggrieved minorities who are unable to contest elections. Exceptions have already been made. As part of the 2005 peace deal, which appears to have ended the long-running civil war in Aceh, the rebel Free Aceh Movement (GAM) was explicitly granted the right to compete in elections. As a result, there is direct – and, at the time of writing, unresolved – contradiction between the provisions of the peace deal and Indonesia’s national party law (Reilly 2006).

Conclusion The core problem facing post-war elections in the contemporary era is the ideologically driven belief that all good things go together – a belief which ignores a great deal of contrary evidence regarding the interaction between post-war politics and democracy. While well-crafted elections may indeed be important instruments of peacebuilding, polls held in highly conflictual environments often have pernicious consequences. They can act as a catalyst for the development of parties based around cultural, linguistic, religious, or other kinds of ‘‘ethnic’’ cleavage. They can promote a focus on regional, rather than national, issues. They are inevitably an unattractive option for those groups who

Post-war elections

181

see themselves being consigned to a permanent minority status, and are therefore likely to be violently opposed by the potential losers. They can serve to place in positions of elected authority leaders committed to exclusionary visions of the country – in many cases, the same characters who started or fought the conflict in the first place. Despite these well-known dangers, the outcomes that post-war elections are expected to foster have become increasingly overloaded by policymakers and politicians. Following the fall of the Berlin Wall and the collapse of legitimate alternatives, democracy came to be seen by the United Nations and the international community more generally as an essential element of post-war reconstruction, with erstwhile conflicts to be transformed into peaceful electoral competition via ballots rather than bullets. As a result, elections have become a standard part of the prescription of contemporary peacebuilding. In addition, the post-September 11 era saw the emergence in American foreign policy of a grander rhetoric of democracy (and, by extension, elections) as essential elements in building peaceful states and combating religious fanaticism in the Middle East. Therefore, many transitional elections are now saddled with unrealistic expectations to achieve goals that are inconsistent and sometimes incompatible. A more realistic and less ideological appraisal of elections is required – one which recognizes that elections can be potentially advantageous or injurious to post-war democratization – and that success is dependent on a careful consideration of timing, sequencing, mechanics, and administration issues. On the basis of experience to date, such an appraisal would likely include a recognition that while elections cannot be postponed for more than a few years as part of a post-war peace deal, rushed elections held in situations of insecurity will almost inevitably aid extremist parties and candidates; that highly proportional PR systems may be administratively convenient but have hidden and sometimes debilitating political costs; that independent electoral commissions are demonstrably preferable to party-based models for established and emerging democracies alike; that a sequenced step from local to national elections is optimal for most post-war societies, particularly those with little prior experience of democracy; and that building coherent political parties focused on the delivery of public goods is a pre-eminent challenge. Policymakers need to pay more attention to these issues, and to the link between institutional choices and the broader goals of building stable and democratic post-war polities.

7

Civil society in war-to-democracy transitions ROBERTO BELLONI

In the 1980s Serbia, like all other Eastern European countries, witnessed a strong revival of civil society. Several non-governmental organizations (NGOs) and committees for the defense of different rights from freedom of expression to ecological protection were created – primarily among intellectual circles in Belgrade. In the second half of the decade many members of these organizations and groups began to move from demands for human rights and democratization to extreme nationalism which denied basic rights to members of other non-Serb national groups. Virtually all intellectuals and grass-root organizations supported Slobodan Milosevic’s rise to power – identifying him as the champion of the Serb national cause. The outbreak of war in the 1990s, combined with Milosevic’s controversial role in the protection of Serbs throughout the territory of the former Yugoslavia split the emerging civil society. Non-nationalist opposition to the regime’s wars in Croatia, Bosnia, and Kosovo was dubbed the ‘‘other Serbia.’’ At the same time, however, extreme nationalist groups also criticized Milosevic, blaming him for having abandoned Serbs in Croatia and for having failed to support Serbs adequately in Kosovo – an area inhabited by an overwhelming majority of Albanians. The peaceful revolution of October 2000, ending thirteen years of Milosevic’s rule, occurred because of a broad alliance among actors within civil society, including both nationalist and non-nationalist groups. The ouster of Milosevic, however, did not terminate the influence of uncivil, violent, and illegal groups – one of the legacies of the Milosevic era (Bieber 2003). In March 2003 a member of the criminal Zemun Clan assassinated pro-Western reformist Prime Minister Zoran Djindjic, complicating the reform process which had begun in October 2000. This chapter examines the role of domestic civil society, that is, the set of voluntary organizations and groups not created by the state, in the transition from conflict to peace and democratic consolidation. It is structured in two main sections. First, it provides a brief analysis of the 182

Civil society in war-to-democracy transitions

183

state of civil society in war-torn countries. As in Serbia, different groups and organizations coexist within the civil society realm. First, even the most deeply divided societies maintain multiethnic and civic constituencies. There exist associations and groups which recognize the importance of respecting human rights and promoting compromise, dialogue, and economic and social integration. These groups strive to promote civic politics instead of ethnic politics, and social and political spaces of dialogue instead of ethnic or national segregation. Not only do they provide a space to articulate citizens’ participation in public life, but also they can constrain the arbitrary exercise of state power. Supporters of the importance of civil society in war-to-democracy transitions implicitly or explicitly endorse these kinds of associations and groups. Second, people organize not only around democratic and liberal values, but also around values that can be seen as ‘‘uncivil.’’ Many civil society groups and organizations justify and engage in violent and/or illegal actions. Mafia-like groups and paramilitaries often thrive in the context of a national, ethnic, or religious divide and in the absence of a functioning state guaranteeing the framework for peaceful, non-violent coexistence. Third, between civil and uncivil groups there exists a wide spectrum of organizations divided along ethnic, religious, and national lines. Civil society in conflict areas is as polarized as political society. Many groups and associations organize around sectarian and particularistic identities. Although most of these groups and associations may openly reject violence, they often endorse and promote a worldview that considers groups’ relations in zero-sum terms and resist compromise and cooperation. Their very existence perpetuates the divisions within society and can contribute to political polarization and continuing confrontation between groups. The second part of this chapter explains how the confidence and expectations placed upon civic, cross-cutting, and pro-democracy civil society are often frustrated in practice. Of the four dilemmas outlined in the introduction to this volume, three in particular affect civil society – complicating its contribution to both democratization and peacebuilding. A vertical dilemma between legitimacy and efficacy arises in the process of the negotiation and implementation of a peace agreement. The inclusion of civil society groups in this process improves the legitimacy of an agreement but may come at the cost of efficiency by increasing the number of actors involved and thus making the decision-making process more burdensome. While this dilemma is

184

Roberto Belloni

particularly acute in the peacemaking phase, the next two dilemmas affect the post-settlement peacebuilding/democratization transition. A systemic dilemma may take place when international intervention attempts to support local, bottom-up peace constituencies. International assistance may be indispensable to guarantee the survival and further development of local civil society groups; but such assistance can result in skewed local priorities by inducing local groups to prioritize those activities likely to attract international funding. Finally, a temporal dilemma appears when short-term and long-term intervention strategies conflict with each other. International organizations have a tremendous institutional pressure to demonstrate short-term results. But ‘‘short-termism’’ limits the effectiveness of international intervention, in particular by preventing the adoption of long-term structural projects. In conclusion, this chapter asks under what conditions civil society can have a positive influence on both democratization and peacebuilding, and how to reframe and restructure international engagement to better sustain post-settlement transitions. The three dilemmas identified are genuine and difficult to resolve. They require a far-reaching adjustment in bilateral and multilateral donors’ practices. Central to this adjustment is the need to mobilize effectively existing domestic resources as equal partners to international engagement, instead of continuously identifying domestic gaps to be filled with the proper dose of international assistance. So long as international engagement is framed around the notion that individuals, groups, and local associations are objects of international engagement, rather than being active agents with resources and assets, civil society’s contribution to both democratization and peace will be limited.

Civil society between ideals and reality The current popularity of civil society in democratization and peacebuilding projects is due in large measure to the legacy of the democratic struggle in Eastern Europe and Latin America in the 1970s and 1980s. Individuals and groups fighting dictatorships in these regions viewed the struggle as one between civil society against the state, with the two confronting each other in a zero-sum relationship (see, for example, Arato 1981). When the Berlin Wall fell, civil society became endowed almost with a heroic quality. Since then, much of the literature on the

Civil society in war-to-democracy transitions

185

topic either has stressed the qualities of the ‘‘good’’ society to juxtapose against the ‘‘bad’’ state or the need of a strong civil society to make democratic institutions work more efficiently.

The civil society ideal and its limits Perhaps more authoritatively than any other scholar, Robert Putnam (1993; 2000) has argued that a healthy civil society is key to making democracy work. In his view, civil society is the microcosm for the development of democratic norms and practices. Social interaction in voluntary NGOs is a school of democracy. When members agree to play by the rules and commit to the goals of a civil society organization or community group, protracted, face-to-face interaction will develop ‘‘bonding social capital,’’ reinforcing the members’ mutual trust and cooperative behavior. Members are socialized into democratic norms through a process of learning by doing. Not only can civil society nurture trust and reciprocity, but also it fosters tolerance for diversity – a crucial aspect for societies torn by civil conflict. According to Putnam, the interaction with people from different ethnic, racial, religious, and political affiliations and social status promotes ‘‘bridging social capital.’’ By participating in civil society organizations, individuals learn how to confront divergent opinions and tend to develop greater tolerance for different interests and views. Crucially, it does not matter what type of group or organization individuals belong to. Social capital and toleration can develop regardless of the goals and aims of these groups and organizations. Accordingly, Putnam argues that choirs, hunting organizations, bowling leagues, and more broadly sport clubs are all useful tools for developing trust among its members and tolerance for diversity. Others have confirmed the existence of a strong positive relationship between social capital and democratic performance – expanding the analysis beyond civil society as a ‘‘school of democracy.’’ Larry Diamond (1999a: 230–234), for example, argues that civil society performs many important functions, including monitoring and restraining the exercise of power by the state, stimulating political participation, developing a democratic culture of tolerance and compromise, creating additional channels for the articulation and representation of interests, generating cross-cutting cleavages, recruiting and training new political leaders, improving the functioning of

186

Roberto Belloni

democratic institutions, disseminating information to citizens, and producing favorable conditions for economic reforms. Francis Fukuyama (1995) has taken Diamond’s last point further, pointing out how the existence of social capital (‘‘trust,’’ in Fukuyama’s jargon) underpins and facilitates economic activities in the market, making trustful societies not only more democratic but also more prosperous. In sum, civil society participation has a beneficial influence on individuals, who become better citizens of the democratic polity even when they participate in recreational, non-political groups. As a whole, a vibrant civil society is instrumental for a more democratic and prosperous society. Building on research conducted primarily in Western consolidated democracies, practitioners and students of democratization and peacebuilding have argued in favor of the positive role civil society can play in conflict areas (Fitzduff 2004; Gidron, Katz, and Hasenfel 2002; Paffenholz and Spurk 2006; van Tongeren, Brenk, Hellema, and Verhoeven 2005; World Bank 2003b). According to its proponents, civil society organizations are the answer to the problems confronting societies transitioning from war to peace and democracy. Not only can civic organizations increase trust within and between different communities, but also they are key to providing public goods when the state is too weak, divided, or indifferent to do so – which is often the defining condition of regions torn by civil strife. According to Daniel Posner (2004), the collapse of the state can be ‘‘liberating’’ for civil society – a view squarely within the ‘‘good’’ society / ‘‘bad’’ state tradition. Groups and organizations can patrol neighborhoods in the absence of a functioning police; they can organize a rudimentary judicial system when state courts are unable to administer justice; and they provide education to young people when schools are not working. Moreover, even where the state still maintains some capacity to deliver services to its citizens, civil society can still complement the work of domestic institutions by helping to improve economic and political performance, control crime and corruption, provide opportunities to former combatants to demobilize after war, and support the process of post-war return home of refugees and displaced persons. This ideal vision must be weighed against the reality of a fragmented, factionalized, and occasionally xenophobic version of civil society. Even in consolidated democracies, civil society is a vague and general concept, which can be filled with different contents – ranging from

Civil society in war-to-democracy transitions

187

democratic to undemocratic actors and from peaceful to violent agents. Civic action is often burdened by incivility and violence – leading at least some observers to argue that civil and uncivil aspects coexist within the civil society realm. John Keane, for example, has underlined how the persistence or the possibility of violence within and between countries survives alongside the spreading of civilizing politics aimed at reducing the incidence of genocide, ethnic cleansing, murder, and rape. As Keane (1998: 135) argues, ‘‘all known forms of civil society are plagued by endogenous sources of incivility, so much so that one can propose the empirical-analytic thesis that incivility is a chronic feature of civil societies . . . a perennial barrier to the actualization of a fully ‘civilized’ civil society’’ (emphasis in the original). Not only can domestic organizations promote cooperation and trust among its members and society at large, but also they can foment discord and violence. To refer to Putnam’s work, it does matter a great deal what kind of tunes choirs sing.1

Deeply divided societies: community polarization and state weakness The ease with which civilized coexistence breaks down in the escalation to war corroborates Keane’s thesis about the Janus-like nature of civil society. Anti-civic and violent forms of associationism prevail over those pro-democracy organizations celebrated by advocates of civil society. In Lebanon during the 1975–1989 civil war and in Rwanda prior to the 1994 genocide, local voluntary organizations fomented intergroup violence. In these and similar cases of breakdown of civil coexistence, while political authorities, intellectuals, and religious leaders provided the rationale for violence, civil society fragmented into opposite camps, while the media quickly turned into an instrument for nationalist propaganda. In Croatia, Serbia, and BosniaHerzegovina local media actively endorsed the war agendas of their main nationalist parties (Thompson 1994). In Sri Lanka mainstream media frequently promote Sinhalese nationalist perspectives while 1

Putnam has been much criticized for not considering enough the ‘‘dark side’’ of civil society in his 1993 book Making Democracy Work. Margaret Levi (1996: 52), for example, points out that the Oklahoma City bombers were members of a bowling league. In his later book, Putnam (2000: 350–363) addresses the issue directly.

188

Roberto Belloni

ignoring Tamil views – thus hardly contributing to a dialogue between the warring parties (Orjuela 2005). The war itself further polarizes civil society and more broadly the views of all those involved, leaving a bitter legacy of resentment and mistrust that complicates the process of post-war democratization and peacebuilding. The presence of divided communities suspicious of each other’s intentions hinders human rights protection and more broadly the development of the rule of law. For example, in the case of interethnic crimes, a member of one group is unlikely to report another member to the police. Not only does this attitude hamper police work, which relies heavily on the community, but also it complicates crossethnic human rights advocacy. The predominance of national and group identities can lead some to condone human rights violations, and even to accept corrupt and patrimonial relationships as the alternative to citizenship rights. Moreover, even after the signing of a peace settlement the warring parties typically continue to contest the state, thus weakening its legitimacy and capacity to provide the framework within which meaningful dialogue among human rights and democracy groups can be carried out. Weak political institutions without the support of their citizens lack the ability to make and enforce binding decisions. When the state is weak and/or contested, the influence of uncivil and even xenophobic forms of civil society becomes particularly strong. In a paradoxical reversal of the ‘‘anti-political’’ mood popular in the 1970s and 1980s in Eastern Europe and Latin America, and celebrating civil society against the state, the lack of functioning and effective state institutions in conflict areas provides an opportunity for uncivil groups to thrive. A public sphere with weak or no protection opens the way for the rule of the stronger to replace the rule of law and for patrimonial and patriarchical relations to replace the benefits of citizenship. In a few, limited cases, civil society building can be possible even in the absence of a state (see Paffenholz 2001: 3, on Somalia). In general, however, under such conditions civil society is more likely to succumb to sectarian and often violent and illegal interests. In a post-war context, at least three types of civil society groups coexist in the public sphere. While the impact of the first and second types on democratization and peacebuilding is quite uncontroversial, the role played by the third one is more complex and will be the focus of this section.

Civil society in war-to-democracy transitions

189

First, there are groups defending and promoting a politics of inclusion and civic principles, and stressing the importance of developing multiple civic identities, pluralism, equal opportunity, tolerance, and the government’s accountability. They attempt to defend and enlarge spaces for negotiation and compromise – islands of civility and dialogue in a context polarized by mutual fear and mistrust. They strive for inclusion, participation and equal access and place emphasis on the need to negotiate and cooperate in view of building consensus rather than repeating adversarial, zero-sum positions. It is not difficult to recognize in this version the positive kind of civil society commended by many scholars and practitioners and authoritatively put forward by Putnam and others in their study of the relationship between civil society, democracy, and economic prosperity. There is increasing evidence that multiethnic and diverse civil society organizations bridging national, ethnic, and social divisions have a key role in preserving and consolidating peace. In his research on Hindu–Muslim relations in India, Ashutosh Varshney (2002) found that ethnically integrated organizations, including business organizations, trade unions, professional groups, political parties, and sport clubs, stand out as the most effective way of controlling violence between ethnic groups. Intercommunal engagement in formal organizations leads, in Varshney’s words (2002: 46), to an ‘‘institutionalized peace system’’ moderating tensions and pre-empting violence. Earlier research in Northern Ireland by John Darby (1986) anticipated Varshney’s observation. In the mid-1980s Darby examined workingclass areas of Belfast finding that of four comparable neighborhoods, the one with the most developed network of community associations had the lowest level of violence. Similarly, Timothy Sisk and Christoph Stefes (2005) found that cross-cutting civil society groups were essential to sustain cooperation and moderation among members of different ethnic groups in the transition from apartheid to majoritarian democracy in South Africa. Everywhere in deeply divided societies, groups struggling to promote pluralism, tolerance, and inclusion are widely recognized as instrumental to further democratization and peace. Women groups in particular are extremely engaged in building bridges across communities and demanding increased participation and government’s accountability (Shoemaker 2005). However, the legal and political vacuum left by a weak or failing state often makes society at large subject to the second type of civil

190

Roberto Belloni

society – uncivil organizations such as mafia-like groups and paramilitary gangs engaged in illegal and violent activities. In most conflict regions the main threat to stability and obstacle to peacebuilding and democratization is not the conflict per se, but criminality. Mafia-like groups and paramilitaries represent instances of uncivil society sustaining inter-ethnic divisions or enforcing intra-ethnic private and illegal rule. Mafia-like organizations can be very successful in establishing networks of economic and political control. They often maintain a close relationship with the political establishment, which consolidated itself out of the disintegration of the previous political order and the outbreak of war. These organizations’ continuing existence and successful economic performance is predicated on the presence of ethnic separation and division (Jung 2003). Similarly, paramilitaries have a vested interest in fomenting and perpetuating ethnic and national segregation and exclusion. In Northern Ireland, for example, paramilitary groups enforce a tight control over their respective communities, control illegal trade, and resist the establishment of any effective state institution that would curtail their activities, particularly a multiethnic police force. Groups such as the Ulster Defence Association and the Provisional Irish Republican Army, to name just two, are uncivil organizations engaged in widespread illegal activities, such as the control of arms and drugs trafficking, whose negative consequences are borne by the members of the respective communities they claim to represent. The presence of the paramilitaries is a continuing source of discord, insecurity, and fear, and an obstacle to the consolidation of democratic politics. More complicated is the case with the third type of civil society, those organizations operating within the boundaries of the law, usually not engaged in violent or illegal behavior, but divisive for society as a whole. Civil society institutions in conflict areas, including religious and community groups, veteran associations and NGOs are often as divided as the society in which they are embedded. Membership in these organizations is usually determined by ascriptive criteria, such as race or religion, or by having fulfilled patriotic duties, such as military service. Similarity and cultural homogeneity provide an initial basis for trust, which in turn is the foundation of successful cooperation. Ethnic and religious groups can perform a positive task in furthering democracy and democratization. Putnam (2000: 65–79) considers these groups as a major source of social capital in the United States, and an

Civil society in war-to-democracy transitions

191

important reason explaining this country’s successful democratic tradition. In Eastern Europe and Latin America, where the civil society revival has begun, many churches and religious organizations fought for democracy and freedom. Even in divided societies and war-torn regions, ethnic, national, and community groups fulfill important tasks. In Central Asia, traditional networks comprising of village elders and councils and organized around clan membership have been indispensable in preventing or mitigating conflict – with considerable stabilizing effects in a potentially volatile region (Collins 2006). In Tajikistan, for example, local councils were effectively mobilized by international agencies to smooth the relationship between refugees returning to their villages and other local groups (Mullojanov 2001). More examples can be cited from other war-torn areas. In Cambodia, Buddhist village organizations helped reintegrate demobilized fighters and build social capital. In Guatemala, Mozambique, and Uganda, veterans’ associations played a similar function (Colletta, Kostner, and Wiederhofer 2004). Likewise, in Rwanda the gacaca, an adapted citizen tribunal system at the community level, has contributed to post-genocide reconciliation by relieving the local justice system of the impossible task of trying tens of thousands of perpetrators (Uvin and Mironko 2003). Furthermore, not only do community organizations provide an avenue for social reintegration after the war, but also they may provide their members with a sense of ontological security, thus creating the conditions for increasing bridging social capital across the ethnic/national/religious divide. In some cases, civil society organizations can even evolve and move from uncivil to civil politics. In Serbia, the Serbian Resistance Movement started out as an ultranationalist group pressuring the government into more anti-Albanian policies in Kosovo, but it slowly evolved into an organization accepting key values of civil society such as the formal acceptance of ethnic diversity (Bieber 2003). At the same time, however, ethnic, national, and religious organizations and groups sometimes uphold exclusivist values and norms that are incompatible with democratic practices. Civil society built on ascriptive criteria is frequently based on exclusion of and conflict with others, particularly in regions with a history of war. A worldview that regards identity as closed, fixed, and inherently conflictual underpins a politics of belonging. Separation and exclusion are put forward in the name of national purity and as a precondition for neighborly

192

Roberto Belloni

coexistence. When civil society organizations are not civic, multiethnic and multi-religious, their contribution to democracy and peace might be spurious. Intra-group cooperation based on a sense of belonging and kinship may be directed toward anti-social ends. Bonding social capital can become the vehicle of ethnic and parochial interests undermining social cohesion, fragmenting society, and pitting one group against another (Colletta and Cullen 2000). Many countries in Africa, for example, have long been successful in bonding associational life structured along ethnic, national, religious, and clan lines, but have often been poor in both democracy and peace (Kasfir 2004). This is true in other contexts as well. For example, war-torn Sri Lanka exhibits a high level of popular mobilization developed along racial and sectarian lines – but many civil society groups are not particularly supportive of peacemaking. Indeed, Sinhalese groups linked to the Buddhist clergy have coalesced into the most vociferous opposition to the peace process (Orjuela 2005). The potentially negative impact of ascriptive organizations is visible in other regions. The Orange Order in Northern Ireland, to mention the most gripping example, is an entirely legal association which proclaims itself ‘‘primarily a religious organization,’’ which is ‘‘Christ centered, Bible based, and Church grounded.’’ It stands officially for civil and religious liberty but remains firmly sectarian. Its members must sign a declaration before joining that both their parents are Protestants and leave if they marry a Catholic. The Order’s very existence and activities are a continuing source of tension and occasionally of violent conflict. Among these activities, the Order each year celebrates the military triumph of Protestant King William III of Orange over the Catholics. Since the signing of the Good Friday Agreement in 1998, the organization’s parades through Catholic neighborhoods have contributed to preserving the tension between parts of the Protestant and Catholic community. On the Catholic side it is difficult to find a similarly contentious organization. But the broader reality of a society divided along exclusivist and sectarian lines remains. Indeed, Catholic civil society often reflects the same political preferences of political society. The most popular Catholic organization – the Gaelic Athletic Association – is firmly nationalist (Belloni 2007). Thus, although ascriptive groups can contribute positively to peacemaking and peacebuilding, such groups may also reinforce national and social differences and perpetuate and even strengthen the divisions

Civil society in war-to-democracy transitions

193

and cleavages existing in society – instead of fostering and sustaining a public sphere hospitable to democratic and civic life. When recruitment is exclusive and based on ascriptive criteria, civil organizations tend to exacerbate existing divisions in political life and perpetuate the existence of potentially hostile communities (Hadenius and Uggla 1996: 1623). Strong in-group loyalty, while strengthening bonding social capital, may also undermine cross-cutting social capital and keep alive out-group antagonism. Organizations with little or no national, religious, or even class diversity tend to reinforce the views of its members and strengthen people’s sense of difference from other communities, dangerously perpetuating stereotypes about outsiders. This brief overview reveals how the considerable variety of ascriptive organizations, and the different contexts where they operate, complicates generalizations about their role in democratization and peacebuilding. Generally speaking, ascriptive organizations are more likely to be mobilized for democracy and peace when they arise and develop to address local issues directly pertaining to the community they are attempting to serve. Often these organizations and groups are small and locally rooted. They pragmatically attempt to mitigate the consequences of conflict and mediate between opposite groups. By contrast, bigger, nationwide groups can provide an important electoral base for sectarian political parties. For these reasons, organizations of demobilized soldiers, churches, and even sport leagues can and often are mobilized and instrumentalized on the basis of national, ethnic, or religious criteria – reinforcing societal cleavages, hindering the democratization process, and sometimes even engaging in undermining the peace process (see, for example, Bojicic-Dzelilovic 2006).2 Overall, civil society reflects and incorporates the divisions within society. Because of the legacy of the conflict, post-war societies are particularly likely to be dominated by sectarian groups. Moreover, the lack of an effective and legitimate state often leaves civil society in conflict areas prey to violent and illegal interests. In this context, associations striving to further civic politics and political, economic, and social integration may be either a minority or less influential than uncivil groups – or both. Their positive influence on war-to-democracy transition may be further hindered by the choices of international and

2

Many thanks to Roland Paris who pushed me to think harder about this issue.

194

Roberto Belloni

domestic actors. Civil society groups present peacemakers with difficult dilemmas and trade-offs, to which I now turn.

The vertical dilemma: inclusion or exclusion? The vertical dilemma entails a choice between legitimacy and efficacy. The involvement of civil society in the pre-negotiation and negotiation of a peace agreement is one of the first choices faced by negotiators. Should civil society groups participate in the peacemaking process? Which segments and interests of society should be represented? More specifically, which individuals and groups should participate and why? What criteria should be applied in the selection of civil society representatives? Should sectarian civil society be kept out? If so, on what grounds? In general, what advantages would representatives of civil society bring to the table and which problems could it create? The involvement of civil society’s representatives has both positive and negative consequences, making the decision of whether such representatives should be included a difficult one. There is no unique approach to the issue of whether and to what extent civil society should be involved in peacemaking. Rarely do civil society members directly participate in the process with a seat at the negotiating table. Often international mediators exclude non-combatants from peace negotiations. The reasons for this exclusion are multiple, including the need to focus primarily on reaching an agreement among those actors who control the means of violence (and thus can stop the war) and to conduct negotiations efficiently and successfully. The successful conclusion of peace negotiations may require a high degree of confidentiality, precluding or complicating the opening of the process to actors other than the main warring parties. This need for confidentiality can lead to the negotiation of a peace agreement behind closed doors, frequently in a foreign country, increasing the possibility that its terms will respond to the demands of the participants and their immediate constituencies, including the military and the political and economic elites. Furthermore, international mediators might fear that the involvement of a greater number of individuals could delay the process of the negotiations or even irritate domestic political elites – who often dismiss local civic groups as either instruments of improper international interference, or political dissidents on the payroll of the opposition. Finally, inclusion may also require a thorough assessment

Civil society in war-to-democracy transitions

195

of local reality and actors – a difficult task in the context of an ongoing war. Thus, civic groups are often excluded from negotiations. International mediators prefer to pay lip service to civic groups, instead of meaningfully involving them in the process. The process leading to the negotiation of the Dayton Peace Agreement (DPA) ending the war in Bosnia in late 1995 exemplifies this approach perhaps better than any other peacemaking process. Chief American peace negotiator Richard Holbrooke focused his shuttle diplomacy on reaching a compromise among those political leaders in Bosnia, Serbia, and Croatia most responsible for the war – not ever meeting with any actor who did not directly control the means of violence. During the talks that preceded the peace conference in Dayton, Holbrooke (1998: 135) endorsed the principle of secrecy to avoid ‘‘public pressure’’ from Bosnian citizens and groups. He later adopted the same approach at Dayton, where lack of communication with the press was the ‘‘most important rule’’ of the peace conference (Holbrooke 1998: 236). This ground rule was expressly aimed at insulating the main political leaders from the demands of civil society constituencies back home. The fact that no civil society member was invited to Dayton to participate in the negotiation ensured the success of the strategy.3 Although effective in reaching an agreement ending a bloodshed, which cost more than 100,000 lives, the DPA also entrenched the power of the same ethnonational elites responsible for the war, and made it very difficult for Bosnian civil society to play a positive role in the post-settlement transition (Belloni 2001). Furthermore, only one woman participated in the conference – a representative of the British delegation. Unsurprisingly, the Dayton Peace Agreement were genderblind – creating serious obstacles for the success of the post-settlement democratization and peacebuilding process. In particular, because the vast majority of those individuals displaced by the war were women, often widows who might have been physically and mentally abused during the war, failing to take into account explicitly and directly their 3

In his detailed account of the peacemaking process in southeastern Europe, Holbrooke cites domestic civil society only once. In December 1996, hundreds of thousands of Serb citizens crowded the streets of Belgrade demanding greater democracy in Serbia. As Holbrooke (1998: 345) points out, on that occasion American diplomacy ignored the demonstrators and their leaders, missing an important ‘‘chance to affect events.’’

196

Roberto Belloni

specific needs has posed a serious hurdle to the implementation of the human rights aspects of the peace agreement (Lithander 2000). The absence of civil society in the negotiation of the peace settlement resulted in a self-fulfilling prophecy, creating the conditions for future constraints in the peace implementation period. Belief in local elites’ cooperation in the implementation of the DPA proved illusory. In particular, Milosevic’s role quickly evolved from that of peacemaker and guarantor of the peace settlement to the main obstacle to its implementation.4 His political trajectory highlights the difficult choices involved in the vertical dilemma. The reliance on pacts between elites and the marginalization of civic groups might be indispensable to reach a peace settlement but at the same time, by preventing the active involvement of domestic constituencies, it can limit the legitimacy of the agreement, and complicate its later implementation. As we shall see with reference to the Democratic Republic of the Congo and the Ivory Coast, civil society’s inclusion in peacemaking might not prevent difficulties during the peace implementation, but it ensures broader domestic support. The Bosnian peace negotiations, conducted under the pressure of an ongoing bloody war, excluded civil society actors in the name of prioritizing the signing of an agreement among the main war makers. But the exclusion of civil society actors from the negotiating table also occurs in less dramatic situations. In South Africa, where the transition from apartheid occurred without the much-feared bloodshed, the principal forum for the negotiation of the ‘‘historical compromise’’ – the Convention for a Democratic South Africa (CODES) – excluded actors from civil society. Similarly, in Northern Ireland no civil society group as such was part of the peace negotiation process, which was initially held in secret and eventually led to the signing of the 1998 Belfast Agreement. In arguing against the inclusion of civil society, the British government raised the question of how representative and accountable civil society groups could be considered. As a result, in order to obtain a seat at the negotiating table, civil society members were asked to achieve a clearer legitimacy by participating in the 4

Under this respect, the evolution of Yasser Arafat’s role in the Middle East peace process resembles that of Milosevic, from guarantor of the 1993 Oslo Agreement, negotiated in secret in Norway, to the real or perceived main obstacle to its later implementation.

Civil society in war-to-democracy transitions

197

formal political process. The Women’s Coalition took up the challenge, transformed itself into a political party, contested the elections, and won one seat. One of its proposals put forward during the negotiation process led to the establishment of the Civic Forum comprising representatives of business, trade unions, and the voluntary sector (Guelke 2003: 68).5 Of course, even when civil society groups are not formally allowed to join the negotiating process, as in South Africa and Northern Ireland, they can still provide both inputs through informal channels and legitimacy and support to the peacemaking effort. In both instances, activities of civic groups have contributed to increasing the number of those in favor of the peace process and thus they have been instrumental in creating the background conditions necessary to reach a peace agreement. The proximity between the general population and civil society actors made civil society a key factor in the effort to build a peace constituency. In South Africa, community groups and progressive antiapartheid organizations were instrumental in diffusing the tension in the difficult and violent period of transition from apartheid to power sharing and, later, to majoritarian democracy (Collin Marks 2000; Sisk and Stefes 2005). In Northern Ireland prior to the Belfast Agreement two campaigns played an important role at key times. In 1992–1993 the Opsahl Commission involved the grass roots and changed the terms of the political discourse by coming forward with one of the key principles of later peace negotiations: ‘‘parity of esteem.’’ In the spring of 1998 the nonpartisan ‘‘Yes’’ campaign helped achieve a victory in the referendum that sealed the deal between Protestants and Catholics (Guelke 2003). As these examples suggest, local involvement of civic and pro-democracy groups can provide greater popular backing for a peace agreement, and facilitate its later implementation. At the same time, however, civil society’s support for peacemaking cannot be taken for granted. Both in South Africa and Northern Ireland intergroup antagonism during the negotiation process heightened tensions and divisions within society and risked derailing the negotiations. A similar pattern is currently being repeated in other conflict areas. In Sri Lanka, a cease-fire agreement between the majority Sinhalese community and the Tamils concentrated in the north of the country was 5

Tellingly, the largest organization in civil society in Northern Ireland – the Orange Order – was denied representation.

198

Roberto Belloni

reached in February 2002 – opening the way for several rounds of peace talks. Several pro-peace civil society groups mobilized support for this process and raised awareness among the masses for the need for it. At the same time, Sinhalese nationalist groups joined forces with Buddhist monks to condemn international involvement in Sri Lanka’s internal affairs and the holding of peace negotiations. They staged vociferous protests – often even more visible than demonstrations in support of peace (Orjuela 2005: 133). The formal exclusion of civil society from official ‘‘settlement negotiations,’’ which took place in Bosnia, South Africa, and Northern Ireland, is not true of all peacemaking processes. Without civic ownership, peace is unlikely to be self-reinforcing and sustainable – suggesting the need for broadening the negotiations. Because elites-based pacts can be unstable, the involvement of civil society can establish the process on more solid foundations (Wanis-St. John and Kew 2006). Accordingly, political elites can attempt to achieve broader legitimacy for peace deals by accepting and even encouraging civil society’s direct involvement. In the Democratic Republic of Congo (DRC), the peace settlement signed in December 2002 included not only government representatives and rebel groups, but also opposition parties and different civil society organizations. Similarly, in the Ivory Coast the peace deals reached in early 2003 incorporated a wide spectra of society (Nilsson 2005) As the cases of the DRC and the Ivory Coast both suggest, the inclusion of civil society is not a panacea, since there remain serious constraints on the implementation of the agreements reached. Sometimes the government encourages civil society’s consultative role, not to meaningfully involve broad societal constituencies but as a form of ‘‘window dressing’’ (Barnes 2002). Guatemala is an example of a case where civil society did not sign the peace agreement but provided non-binding inputs to it. In the early 1990s, the government solicited civil society’s contribution in an attempt to provide legitimacy to the peace deal – even though it expected that its diverse composition would prevent it from reaching a common negotiating position, and thus led to only little or no substantive input (Krznaric 1999). The Assembly for Civil Society (ACS) representing the views of the civic sector was invited to contribute to the peace accord. Although the government accepted the ACS’s participation because it assumed this newly created body would be ineffective, its presence represented a breakthrough,

Civil society in war-to-democracy transitions

199

and its influence on the outcome of the peace negotiation was significant, as many of its proposals were included in the peace accords. At the same time, however, the ACS’s proposals on socioeconomic and agrarian reform – particularly land redistribution – were ignored. Guatemala’s most important business association (the Coordinating Committee of Agricultural, Commercial, Industrial, and Financial Associations) successfully lobbied against it – leading one commentator to describe this association as the key ‘‘uncivil’’ actor bent on limiting citizenship rights in order to preserve economic privilege (Krznaric 1999). In sum, the vertical dilemma involves a choice between efficiency and legitimacy. Effective negotiations might require minimizing the number of individuals involved, and preserving a certain degree of secrecy. At the same time, however, the inclusion of civil society representatives in peacemaking negotiations can increase the legitimacy of a peace agreement and the prospects for its implementation – even when domestic elites expect civil society involvement will have only a modest impact on the drafting of the peace settlement. Whether or not civil society participated in peacemaking, and in what form, in the postagreement phase promoting local civic participation and involvement is a central strategy of the international support for post-settlement peace processes. The dilemmas involved in this strategy are the focus of the next section.

The dilemmas of peacebuilding and democratization International organizations, multilateral and bilateral donors, and international NGOs have invested considerable resources in building and strengthening domestic civil society in conflict areas. One reason is pragmatic. Years of conflict at best weaken state institutions and at worst decimate their capacity to deliver services to the citizens. Sometimes foreign occupation and rule contribute to the inexperience of domestic elites in addressing effectively societal concerns. In Northern Ireland for almost three decades direct rule from London created a dearth of debate about social policy and prevented political parties from developing public policy experience. In this vacuum, civil society stepped up its work and often replaced local institutions in service delivery. When state institutions do not function properly, and the domestic political elite does not have the ability or willingness

200

Roberto Belloni

to perform its duties, civil society presents international donors with a valuable, cheap, and overall reliable alternative. The second reason for supporting and developing domestic civil society reflects a hope, or an aspiration. International actors believe that the stronger the civil society, the easier the fulfillment of the post-settlement peacebuilding and democratization tasks, and the easier the process of international disengagement. Because civil society in war-torn countries is typically extremely weak and divided, the crucial question is whether international funding can build pro-democracy civic-networks over the short-run, or whether the existence of a thriving civil society depends on long-term historical legacies hard to change by short-term human action – particularly in war-torn regions.6 The war’s damage to social capital complicates the capacity of communal groups and/or the state to recover after the end of the hostilities. Conflict weakens the fabric of society, divides its population, and undermines the interpersonal trust necessary for effective cooperation and collective action (Colletta and Cullen 2000). In this context, while international funding may be indispensable to make it worthwhile for individuals to cooperate in local groups, it can also give rise to the systemic and the temporal dilemma. First, the absence or low levels of international aid might prevent the emergence and/or consolidation of bridging, pro-democracy civil society. At the same time, however, international support comes with a cost – in particular when local organizations become more accountable to international donors than to the communities they are supposed to serve. Second, while international donors recognize the need to support long-term structural projects, institutional constraints push them toward a shortterm and top-down approach. The issues raised by the systemic and the temporal dilemmas, respectively, are the focus of the next two sections.

6

In his research on civic tradition in modern Italy, Putnam endorses the longterm historical perspective. He argues that those Italian regions exhibiting civic strength owe their success to the legacy of republican traditions during the Renaissance. Although a civic tradition can be beneficial to democracy, its contribution can also be overstated. In Italy, civil society has strong roots in the north, but this is precisely one of the regions that supported Mussolini’s Fascist Party in the 1920s and 1930s. Similarly, civic density did not prevent democratic collapse in Weimar Germany, where pre-existing and well-organized social networks were captured and mobilized by anti-democratic movements (Berman 1997).

Civil society in war-to-democracy transitions

201

The systemic dilemma: too little or too much international support? The potential drawbacks raised by heavy international involvement for the sake of strengthening local civil society lie at the heart of the systemic dilemma. To begin with, the exact boundaries between, on the one hand, civic, inclusive, and democratic groups and, on the other hand, uncivil, exclusivist, and xenophobic organizations can be difficult to determine in theory, and even more so in practice. Intransigent nationalists may be as likely to create local organizations to take advantage of foreign funds as pro-democracy civil society activists. In the Serb Republic of Bosnia, for example, there is evidence that some local NGOs have learned how to talk the language of multi-ethnicity, dialogue, and compromise necessary to obtain foreign funding, but they remain unabashedly nationalist (Katana 1999). In some cases, opportunities in the civil society realm may even present local politicians with a profitable alternative in times of political decline. Prominent politicians can create their own NGO to benefit from political connections in obtaining funding and influence. The blurring between government and civil society is further confirmed by the influence of family ties across the two realms. The wife of indicted war criminal and former President of the Serb Republic of Bosnia Radovan Karadzic served as the Director of the International Committee of the Red Cross in the Serb Republic both during and after the 1992–1995 Bosnian war. Similarly, the wife of Croatian President Franco Tudjman was head of the Humanitarian Foundation for the Children of Croatia during the 1990s. In both cases, these civil society/humanitarian organizations enjoyed special privileges from political authorities. Although formally independent from the government, their political affiliation was never called into question. Occasionally, similar government-sponsored organizations can even be created to undermine the work of pro-democracy and advocacy groups. In Rwanda the government created false NGOs for the purpose of attributing misleading propaganda to them. For example, the organization Licrodhor spread outrageous anti-Tutsi propaganda – arguably to discredit a respected human rights organization called Liprodhor.7 7

Personal communication with Benjamin Siddle, Peacebuilding Program Manager for Trocaire in Rwanda (1999–2002), April 2005. The ruling Rwandan Patriotic Front regularly intimidates local organizations, most notably by accusing any

202

Roberto Belloni

But even when inclusive and pro-democracy organizations can be identified and supported, the financial relationship between the local non-governmental sector and international donors may set into operation the systemic dilemma. International support is often indispensable for domestic civil society groups, who face an enormous challenge in raising funds in the context of post-war societies. Private resources are rarely available, while the government is more concerned about attracting international aid for its own activities than directing such aid to civil society. However, while international support may be indispensable, such support may come at a cost. Local participation and ownership encouraged, financed, and sponsored by outside donors is limited in its capacity to create domestic social capital and ownership of the peace process. Civil society building strategies based on the attempt to support the creation and development of civil society groups from the outside face at least three major limitations. First, international engagement in divided societies reinforces the view that local communities have needs which can be met through international funding channeled via domestic NGOs. The prevailing approach to civil society development downplays the existence of local resources and knowledge. Domestic community leaders learn that the best way to attract international funding is to underline the problems and disparage community members and their individual and collective resources. Residents of conflict areas are frequently described as severely traumatized individuals in need of outside experts and funding. Furthermore, when successful in attracting international funding, local civil society groups develop an interest in perpetuating the system, while presenting themselves as necessary transmission belts between local communities, national governments, and international donors. The negative consequences of this strategy are clear. Instead of developing existing human and social resources, increasing existing social trust, and providing stronger foundations for democratization and peacebuilding, civil society groups are given incentives to downplay the existence of these resources and rely on external support (Mathie and Cunningham 2002). Moreover, support from the outside may decrease the need of leaders of an organization to rely on members’ contributions, with negative consequences on members’ voluntarism, potential critics of ‘‘ethnic divisionism.’’ On these grounds, independent human rights organizations may also be banned (Hampson 2004).

Civil society in war-to-democracy transitions

203

internal accountability, and in general on the organization’s ability to advance a democratic political culture. Second, the international donor–local group relationship has shown the dangers inherent in the financial dependency of the latter on the former. Building and strengthening civil society is often conceived as an externally driven process that is often entirely dependent upon international resources (Hadenius and Uggla 1996; Hann and Dunn 1996; Pugh 2000). Despite much talk about partnership and consultation between international agencies and domestic civil society actors and the need to develop local ownership of peace processes, donors regularly ignore local organizations’ priorities and agendas. In order to be funded, local organizations have to operate according to the frequently changing preferences of international actors, instead of the needs of the communities these organizations are supposed to serve. Rather than creating local ownership, this practice may lead to domestic dependency. As an old African proverb wisely states, ‘‘the man with his hand in someone else’s pocket will move when he moves.’’ Accountability is redirected toward the donor and away from the organization’s social base, and the idea of political and social participation and empowerment is undermined by the reality of an externally driven process, leaving domestic groups in a weak and subordinated position. Third, the technical mould of international support strategies, centered on training local trainers and creating local capacities, limits their effectiveness. Many citizens in post-war societies show antipathy toward an approach to democratization and peacebuilding centered on international experts and trainers who often do not speak the local language, and have limited direct knowledge of local customs and culture. Often the training itself is not so much in developing skills to increase the impact of local activism but more pragmatically on how to successfully write a project proposal and be funded. After domestic organizations are socialized to the language and expectations of international donors, aid delivery creates a hierarchy. There is a big discrepancy between the most developed organizations that work in the bigger cities and take advantage of their contacts and exchanges with international agencies and the larger number of small civic groups that strive for visibility and funding. Furthermore, evidence across a range of different cases, ranging from Bosnia (Belloni 2001) to El Salvador (Foley 1996: 84–86), Timor Leste (Patrick 2001), and Sri Lanka (Orjuela 2005) shows how donors tend to fund moderate, middle-class groups,

204

Roberto Belloni

whose interests might match their own, but that might not be too representative of society at large. Ordinary people often avoid participating in NGOs because they do not feel these organizations reflect their needs and priorities. As a result, as Beatrice Pouligny (2005: 501) perceptively points out, international intervention pretends to ‘‘build ‘new’ societies while excluding the large majority of their members.’’ Not only does this externally driven strengthening of civil society exclude and disempower, but it also can be interpreted by domestic groups and citizens as a continuation of a post-colonial Western ‘‘civilizing mission’’ (Kaldor 2003: 9), and because of this resented and resisted. But the main dilemma of international engagement in civil society building may come from its very success. International resources and opportunities made available for the third sector risk deterring talented and motivated citizens from joining political parties and contributing to the political process through institutional channels, driving them instead to find a role in the voluntary sector. The decision to privilege NGOs as intermediaries between citizens and the state – rather than, for example, political parties or trade unions – ends up de-politicizing society. As Paul Stubbs (1999: 31) has argued in the context of international assistance to southeastern European countries, ‘‘donor support has created a kind of ‘anti-political political opposition,’ marginalizing formal opposition parties, turning social movements into bureaucracies, and ultimately buttressing the status quo while formally seeming to support ‘democracy from below.’’’8 This suggests that, paradoxically, instead of getting around the dominance of political elites to find new opportunities for social and political change, international support may result in reinforcing the existing ethnic, national, and sectarian divisions. The promotion of ‘‘transnational democracy/human rights networks’’ is sometimes touted as a way to overcome these problems triggered by international support of domestic civil society. These networks describe horizontal, flexible, and fluid links allowing international organizations, international and domestic civil society, and bilateral donors and foundations (not all necessarily present in the network at all times) to 8

According to Julie Hearn (2000), this is not an unintended consequence of international assistance. Rather, civil society support to transitional and democratizing states seeks to neutralize potential opposition within domestic educated elites.

Civil society in war-to-democracy transitions

205

cooperate almost as if they were part of one organization to pursue common goals. Margaret Keck and Kathryn Sikkink coined the expression ‘‘boomerang effect’’ to describe how local groups, when blocked politically at the national level, can find the support of network members to clear the national situation: ‘‘Voices that are suppressed in their own societies may find that networks can project and amplify their concern into an international arena, which in turn can echo back into their own countries’’ (1998: x). The key insight relevant also for post-war societies is that when government is divided, ineffective, and not responsive to society’s needs and unable or unwilling to deliver services to their citizens, transnational networks allow for the meaningful participation of local actors in decision making. In theory, local organizations and populations can be involved in the process of institution building and reconstruction, avoiding the pitfall of the traditional hierarchical and top-down international support of domestic civil society (Holohan, 2005; Kaldor, 2003; Reinicke and Deng 2000). While transnational networks provide an interesting and potentially far-reaching way to sidestep the systemic dilemma, they are ‘‘not necessarily harmonious, democratic or effective’’ (Kaldor 2003: 96). Moreover, there is no reason to assume that the less powerful members of the network (above all local populations and civic organizations) can set the agenda against the wishes of the most powerful members – thus undermining the network participatory ethos. In Kosovo, where Anne Holohan (2005) found evidence of the efficiency of networks of democracy, international actors took the lead. When they established the framework for international intervention following the 1999 war, they ignored the ‘‘parallel society’’ that dominated Kosovo’s social life throughout the 1990s, failing to capitalize on a wealth of human resources that would have improved the legitimacy and efficiency of intervention. In sum, the unequal relationship between powerful international actors and relatively powerless domestic civil society groups is a structural condition complicating civil society’s contribution to democratization and peacebuilding. The problems raised by this unequal relationship also lie at the heart of the temporal dilemma – to which I now turn.

The temporal dilemma: short-term or long-term intervention? Civil society building strategies are also affected by a temporal dilemma. Short-and long-term approaches clash in the attempt to support

206

Roberto Belloni

exogenously the development of local civic organizations. Societies coming out of conflict require immediate support to consolidate peace, and prevent any non-state actor from challenging the post-war transition. At the same time, conventional wisdom also advises in favor of the adoption of long-term, structural, and ultimately more effective projects. International donors face a dilemma between long-term democratization and peacebuilding intervention and short-term initiatives with a strong crisis prevention focus. More often than not, international democratization and peacebuilding agencies apply a short-term time frame to their programs and projects. All international agencies, including leading ones such as the United Nations and the Organization for Security and Cooperation in Europe (OSCE), have a tremendous institutional pressure to minimize short-term risks of failure and maximize visible gains. These agencies must be able to demonstrate immediate progress. UN field operations are reviewed regularly. The Heads of Mission report to the UN Security Council every six months. Similar reporting requirements are in place for other regional organizations, such as the OSCE. Moreover, media and think tanks frequently challenge the efficacy of multilateral international organizations such as the UN and the OSCE, intensifying the pressure on these bodies to show that peacebuilding and democratization ‘‘are working.’’ This pressure is even more severe for bilateral donors whose resources are dependent on a budget allocated annually by the respective national legislatures and/or governments. The need for immediate outcomes often prevails, pushing toward a short implementation time frame and explaining three common problems of internationally driven civil society building programs. First, international involvement focuses on the creation of domestic NGOs – the easiest, short-term success to achieve for foreign donors, but also the most superficial one. As Michael Edwards (2004: 95) points out, the number of NGOs in society is the easiest thing to influence, but also the least important. A study on civil society building in Bosnia confirms that there is no direct correlation between affiliation with voluntary associations and the amount of social capital (World Bank 2002b: viii). Similarly, the presence in Rwanda of numerous local NGOs supported by external aid did not prevent the 1994 genocide. For the most part, these NGOs were apolitical, service-oriented, and did not bridge group divides (Uvin 1998). Civic values and the commitment to a common

Civil society in war-to-democracy transitions

207

life are notoriously difficult to change and require a long-term strategy that extends considerably beyond the creation of local NGOs. Second, the greater the international role and need for short-term successes, the more international agencies fear that recognizing problems may be perceived as an admission of failure. Therefore, they devote considerable attention to selling their achievements, rather than addressing deep-rooted problems and undertaking those long-term structural projects that might have a more profound impact. The more subtle side of this tendency to ‘‘show results’’ in the short term is the problematic inclination to blame local actors for delays, obstacles, and drawbacks in the process. In southeastern Europe slow progress is frequently blamed on the ‘‘Balkan mentality,’’ the supposed combined effects of socialism and war. Similarly, in the Middle East international actors frequently point to the lack of established democratic traditions to explain continuing turmoil. Third, the pressure to achieve short-term concrete results goes a long way to explain ‘‘projectism,’’ the tendency to treat democratization and peacebuilding as a set of discrete interventions incorporated into a project with a relatively clear beginning, implementation, and evaluation, usually with a six-month timespan, or at best one year. ‘‘Projectism’’ makes peacebuilding and democratization a top-down enterprise. Because of the brevity of internationally funded projects, international agencies have little scope to develop significant local partnerships and include local actors in decision making. Instead, they make important decisions about the priorities and allocation of international assistance in the initial phases of a project, when international understanding of local conditions is limited (Carothers 1999: 264). Seen in this light, the idea of developing ‘‘local ownership’’ of the peace process sounds at best naı¨ve. The pressure to show immediate and concrete results advises against the implementation of long-term structural projects. At the same time, shortterm and discrete projects militate against capacity building and domestic ownership. Thus, the temporal dilemma perversely reinforces the systemic one, making civil society building a weak element in the peacebuilding and democratization strategy.

Conclusion: beyond civil society building Civil society in deeply divided regions reflects the larger society in which it is embedded. Inclusive, civic, bridging, and pro-democracy

208

Roberto Belloni

organizations work alongside exclusivist, sectarian, and occasionally even xenophobic groups. The sectarian divide, which is often institutionalized in the peace agreement ending the war, may also be instrumental in preserving and entrenching the power of illegal and uncivil groups, such as paramilitaries and mafia syndicates with extensive connections at the political level. In this context, the attempt of international organizations and donors to promote and support civil society building is often disappointing. The hope that civil society groups may play a useful role in sustaining democratization and peacebuilding with a view to rebuilding failed states may prove misplaced. A healthy civil society is the sign of a well-functioning state, not its cause. Instead of providing the foundations for a more stable democracy and improving the functioning of state institutions, as argued by civil society theorists such as Putnam, Diamond, and others, a healthy civil society needs a functioning state, including effective policing, an impartial judiciary, the rule of law, and the capacity to guarantee personal and societal security. Only institutions and social norms making possible debate, dialogue, and the nonviolent reconciliation of differences can sustain peacebuilding and democratization. In other words, the source of social capital is not simply attributable to civil society but can also be produced by the state. This point leads to a first policy prescription: the best avenue to favor the emergence and development of domestic civil society in war-todemocracy transitions is to strengthen the state by establishing stable and efficient social and political institutions. Despite persisting clientelism and corruption in both established democracies and transitional states, public institutions still maintain some degree of transparency and political accountability. Efficient institutions have an indispensable role in securing the conditions for civic participation by keeping in check illegal, uncivil, and xenophobic groups while ensuring legal protection for domestic, inclusive, and pro-democracy organizations. At the same time, however, state building is bound to be resisted. Entrenched interests may work to prevent or slow down the process. Often domestic elites benefit from a weak state. Paramilitaries and mafia-like groups may see state building as a threat to their interests and actively oppose it. In the anti-statist mood prevailing since the 1980s, even civil society advocates may oppose this strategy. As Keane (1998) and Pouligny (2005) remind us, scholars of civil society

Civil society in war-to-democracy transitions

209

and its activists tend to be uncritical about the potentiality of civil society and privilege this realm over the idea of developing the state. The establishment of a functioning, democratic, and responsive state is only one part of the strategy to address constructively the dilemmas of democratization and peacebuilding in conflict areas. Civil society organizations need also to become progressively less dependent on foreign resources for their funding. Civil society organizations are unlikely to be an expression of grass-roots democracy and inter-ethnic reconciliation when they are more accountable to their international donors than their members and local constituencies. By providing local civil society groups with the money and resources to address needs in their own communities, international agencies might end up, albeit unwittingly, controlling local groups’ agendas. An alternative approach should seek to stimulate the agenda, not to create or control it. While there is no clear-cut solution to organizations’ dependency on their donors, it is at least possible to imagine alternatives. One simple but far-reaching initiative is to make funding available for projects with a time frame longer than the current six months to one-year period. Funding cycles should be at least three years, with periodic progress assessments. Multilateral donors like the European Union are better placed than bilateral donors to provide long-term, structural support. From 1995–2000 the EU package known as the ‘‘Peace Programme’’ for Northern Ireland was so decisive in facilitating the rapid growth and influence of the local civil society sector that it was expanded for the period 2000–2006 (Williamson, Scott, and Halfpenny 2000). Over time, however, the solution to the problem of civil society organizations’ dependency on foreign resources is to reform the very approach international agencies adopt in their civil society support programs. An effective civil society development strategy capable of overcoming the dilemmas of democratization and peacebuilding requires both the ability to identify and support inclusive, bridging, and pro-democracy organizations and the capacity to acknowledge and mobilize local knowledge and experience. The sound analysis and careful assessment of the context, taking into account the conflict history, actors, issues, and dynamics should become routine operational work for international agencies and lead to framing options for international involvement (Paffenholz 2001). Regional experts and anthropologists need to be more systematically involved in the analysis of domestic

210

Roberto Belloni

structures and in the identification of opportunities for engagement. International support must be compatible with existing local realities and careful not to legitimize uncivil and inequitable social structures and organizations. Moreover, building on reliable mapping of the local reality, international policy must shift from a needs-based strategy to an asset- or strength-based approach – in much the same way development agencies have begun conceiving community development in Western countries (Mathie and Cunningham 2002). Even in the unfavorable conditions of division and mistrust deepened by the experience of war, divided societies do possess a network of community groups and individuals committed to inclusive politics, dialogue, compromise, and peaceful change (Colletta and Cullen 2000). As Varshney (2002: 289–297) demonstrated in the case of India, small-scale, localized intervention based on local knowledge and resources can have a considerable impact. Similar examples from Mozambique, Sierra Leone, Cambodia, and Guatemala confirm that ordinary people can and often do mobilize to make sense of and overcome the challenges posed by a profoundly disrupted environment (Pouligny 2005: 502–503). Instead of continuously identifying real or perceived gaps and needs, usually to be met with the proper dose of international money and expertise, international support should be based on local strengths and resources. In sum, the role for international actors is not that of filling needs but rather that of helping local civil society to mobilize resources that work toward a self-defined community vision. International actors must recognize domestic partners as agents with their own experiences and knowledge, and work to improve the quality of domestic contribution to democratization and peacebuilding. Domestic ownership is critical to (re)building bridging social capital. Only when active citizenship in civil society is placed at the center of international engagement, rather than at its receiving end, will democratization and peacebuilding be grounded on more solid foundations.

PART IV

International engagement

8

International responses to crises of democratization in war-torn societies PETER WALLENSTEEN

Post-war situations are difficult periods of transition, involving many simultaneous dilemmas for new and inexperienced governments. Historically, the Weimar Republic in defeated Germany is a classical case, where the first democratic government led by Social Democrats had to administer the disaster created by the previous regime and also do this while abiding by the punitive stipulations of the Versailles Peace Treaty. Germany struggled through a turbulent period of the 1920s, only to be severely hit by the Great Depression with mass unemployment and the exploitation of these conditions by a ruthless new challenge, the Nazis. Central issues in Hitler’s agitation were the Versailles Treaty and the weakness of democratic governance. The international community at the time chose to take a hands-off attitude, and allowed the Treaty to be undermined parallel to the elimination of democracy. A new, even more devastating, war began, twenty years after the ending of the previous one. This example highlights the significance of peace conditions, as well as the importance for international organizations to be concerned with their broader implications. The fate of the Weimar Republic and the rise of Nazi Germany created a lasting guilt feeling among democratic states. Peacemaking today differs in many respects, but the historical lessons should always be kept in mind. The transition from war to peace is cumbersome and the hope to build a functioning democracy should be cautioned with a need for long-term thinking and an understanding of the dangers of failure. The post-World War I experience also suggests that it is necessary to survey one or two decades before trends become clearer, whether negative, as in this case, or positive, which is the experience of the Federal Republic of Germany after World War II. This also illustrates that the dilemmas for the international community are whether, when, for how long, and in what ways to engage in the face of a crisis in the democracy building one is favoring. 213

214

Peter Wallensteen

Since the late 1990s, international donor agencies have taken up democracy as a major theme, and there is an emerging literature on development assistance and promotion of democracy and good governance as part of conflict prevention (Collier et al. 2003; Sida 2005). It has been demonstrated that democratic states rarely fight wars with one another and that democracies have fewer internal wars than other forms of governance (Russett 1993; Russett and Oneal 2001). Since the Cold War, it is clear that peace agreements in internal conflicts over government often resort to democratic constitutions and elections as ways of ending conflict (Harbom et al. 2006). Supporting the development of a sustained democracy appears a reasonable approach in post-war societies. It might help to bring about stability in difficult conditions in order to ensure that a conflict does not return. But the possibilities of supporting peace through democracy building is an avenue which has become questioned (see, e.g., Paris 2004). This approach to peacebuilding may result in difficult dilemmas at critical junctures in a democratization process. The stipulations of a democracy-oriented peace agreement may be difficult to implement. Election outcomes may favor groups challenging the peace. The pace of democratization may be different in different parts of a society (socially, regionally). Parliamentary stalemates may prevent reforms of society. Some studies demonstrate that the transition from autocratic conditions to democratic ones may be the most conflict- and war-prone phase of all (Hegre et al. 2001; Human Security Report 2005). That observation might be particularly applicable to post-war conditions. Taking this together, the way in which democracy is promoted is one of the most critical issues of post-war crises of implementation. However, the literature on how international organizations (whether governmental or non-governmental) or media react to particular crises in such a transition is surprisingly limited. There is some insight suggesting that such issues rarely enter the international agenda. For instance, the armed conflicts that reach the UN Security Council are regularly less than half of all ongoing armed conflicts (Malone 2004; Wallensteen 2007). They have to pass through the screening of governments (often unwilling to allow internal conflicts in their own countries to become internationalized), regional powers (preferring to deal with matters bilaterally or in regional bodies under their control), and the veto powers of the UN Security Council Permanent Members. If we add that postwar democratization crises constitute a subset of such conflicts, it can be

International responses to crises of democratization

215

asked why they should be of concern to, for instance, the United Nations. They may not easily be understood as threats to international peace and security, which is stipulated for Security Council action. Still, there is the observation that the UN is involved in democracy promotion, and particularly in smaller countries rather than larger ones (Newman and Rich 2004). On the whole, the UN is probably more engaged in weak states than in major states, and democracy promotion is no exception to this pattern. However, in the case of UN involvement there is an interest in the long-term construction of democratic governance, for instance as part of a peacekeeping mission (Ludwig 2004; Newman 2004). The focus in this chapter is on the crisis or short-term situations: when there is an observed threat to democratization, how is that handled? If such crises are not dealt with in a constructive way, it has implications for the subsequent efforts of democratization. Clearly, there is a doctrinal development in the international understanding of when international institutions should react: democracy issues are seen to be part of the international agenda in principle. It needs to be seen in practice as well. There is a record of the UN reacting to military coups overthrowing elected governments (Haiti 1993, Sierra Leone 1997; see Farer 2004). This is not consistent, however. The military takeover in Thailand in September 2006 did not result in any official Security Council action, whereas a similar event in Fiji in November generated two press statements. This refers to a situation where the democratic system is overturned by unconstitutional means from ‘‘inside’’ the country. There is also the possibility of outside force being used in overthrowing a regime, often without prior UN approval (Panama 1989, Iraq 2003, interestingly neither case directly linked to post-war conditions), sometimes with a global understanding (Afghanistan 2001). However, the number of military coups is low, as are the military interventions. There are few instances of this kind to react to. In more common post-war situations the dilemmas may look different. Continued development toward a sustained democratic society may be disrupted, not by constitutionally external forces (military factions, outside powers) but by the participants themselves, notably a government clinging to power or a well-organized opposition using means beyond the democratic framework. These are the more limited democratization crises to which this chapter is addressed. The international community is likely to face difficult problems when confronting such a democratization crisis. The amorphous international

216

Peter Wallensteen

community consists of organizations and actors supporting the values of democracy, human rights, welfare, and peace. Its shape will vary but may include UN organs, the European Union (EU), donor states, and international civil society organizations (Wallensteen 2007: 251–258). It will, almost by definition, exhibit differences of opinion and thus be hard to move in a uniform direction. The focus in this chapter is on the actual behavior of this community and how it tends to deal with dilemmas it faces when confronting a democratization crisis. The chapter does not address what would be the ‘‘best’’ choice, but rather tries to understand which courses of action are probable under what conditions. More specifically this chapter deals with the four dilemmas of this book in the following ways: First, there is the issue whether to react, the systemic dilemma: does the international community react at all or does it prefer to let the local actors ‘‘play it out’’ among themselves? Second, it discusses when the international community is likely to react and when to stop reacting – the temporal dilemma: for how long is the international community sustaining the reactions that it chooses? Third, it raises the choice of what means: which actions are legitimate and are they also effective – the vertical dilemma – and how do the targets of action react? Fourth, by analyzing the three first dilemmas also the horizontal dilemma is brought out: which segments of the international system are included in a particular pattern of reactions and which ones are not, whether by deliberate exclusion or by their own choice? The strength of international consensus is possibly a significant determining factor for action and impact. There is no claim that this chapter will provide definitive answers to these questions. The framework of understanding post-war developments as dilemmas, however, is a fruitful way of approaching reality, and some novel issues can, hopefully, be raised. Mostly this will be done by way of illustrative examples.

A democratization crisis: Uganda 2005 Developments in Uganda constitute a plausible starting point. The victor of the Ugandan civil war, Yoweri Museveni, received solid international support for pursuing a fairly generous and inclusive policy, aimed at economic reform and social reconstruction. In the Constitution of 1995 it was stipulated that the president could stay in office for only two terms. This was applauded and donor support

International responses to crises of democratization

217

continued, in spite of signs of a less than democratic implementation of policies in the country. By late 2004, however, Museveni faced the possibility of having to step down, after being in power since his military victory in 1986. He chose instead to amend the Constitution, so as to allow himself to run for a third term. His main opponent, Dr. Kizza Besigye, previously exiled, returned in November 2005 to contest the elections, only to be arrested on charges of treason and rape (Tangri 2005). The international donor community reacted in December, when, for instance, Sweden and the United Kingdom decided to freeze further disbursement of direct budgetary support to Uganda. The respect for human rights and democracy was stipulated in the bilateral agreements, and thus Uganda was seen to have breached that stipulation. There was even a fear of a military takeover. The international community began to respond, in what it saw as significant ways to uphold the principles of democracy, and prevent a new polarization in the country, which already was involved in armed conflicts both internally and externally. These acts by donors could be seen as promoting democracy and preventing escalation of conflict at the same time. The strains on Uganda’s path to sustained democracy had by then been visible for at least a year, but also the election campaign in 2001 could have been questioned. Clearly, donors preferred a quieter dialogue with the regime, rather than making public pronouncements. As the international community had been involved in promoting democratization, it would be difficult not to react at all. Its first choice was for a less confrontational approach. When this clearly did not work, firmer reactions were slow in coming. The constitutional change as such could not be opposed, as it was regarded as a matter of Uganda’s internal affairs. However, the arrest of a leading opponent (and accompanying measures against his supporters) threatened the fairness of the electoral campaign. These seem to be the events that triggered international reaction. A free and fair electoral process is central for democratic development. Without this, the elections themselves and their outcome can be disputed. Not to react to the arrest might have reduced credibility in the international stand on democratization in Uganda. This example demonstrates how a series of events can trigger the international community’s engagement. Still, the reactions were far from concerted (individual donors taking different steps at different times) and restricted to some forms of development assistance, not all. Was this likely to be effective in stemming the erosion of democracy in Uganda?

218

Peter Wallensteen

With this example in mind, we may ask which are the issues that will be seen as a democratization crisis? The discussion on good governance suggests a number of such issues: the rule of law, free and fair elections, free media, a vibrant civil society, independence of the judiciary, independence of the representative body (parliament, national assembly), secure human rights, widespread participation. With this wide understanding of good governance and democratization, there are a number of events that could be seen as ‘‘crisis’’ or ‘‘warning signs.’’ Clearly, only some of these will trigger international reactions, although there might a logical sequence to many of the events. The removal of judges as a way of controlling the judiciary is unlikely to be seen as such a crisis but may be a first step for a government to maintain itself in power. The recruitment of staff into the central administration, and even biased or corrupt operations in the state machinery will seldom unleash concerted reactions. In the reporting – by diplomats and media – this will no doubt be observed, but not seen to be enough to single out a particular government or country in an environment where similar events have taken place before or are common in the region. For instance, the closing down of universities when students or teachers have voiced criticism seems very seldom to engage international governmental organizations. Even increased regime control over media or over civil society tends not to stimulate reactions. However, the donor community is more likely to react to this, for instance as part of the negotiations on renewal of development assistance. Here, there is another dynamics at work: civil society and media will have colleagues in neighboring or donor countries that react, thus stimulating political consideration. If there were more solidarity in the university community, actions curtailing academic freedom might also enter the arena of political reaction. A hypothesis is that the type of democratization crisis that is more likely to generate international action is the one that directly concerns the distribution of power in a country, notably local (governmental or non-governmental) actions against the parliament and against the electoral process. Thus, it is here suggested that the electoral process is the single most typical event that is seen as a carrier of democratization and a significant sign of progression toward ‘‘good governance’’ and democracy. Threats to this process may trigger international action. There are several explanations for this. In a democracy, elections have a particular standing as a way of determining the political strength

International responses to crises of democratization

219

of actors. This is so, although most analysts would agree that election outcomes may, in fact, be determined by a number of earlier actions. Furthermore, threats and/or manipulation of elections during the course of a public campaign are likely to draw considerable media attention and thus result in demands for reactions. In addition, as the international community we discuss here largely is constituted by democratic states this community is likely to understand the threats to elections and, thus, be more willing to react. These possible explanations are illustrated by the events in Uganda. This case, however, also suggests that responses to events that threaten elections might be too limited a perspective and even too late a moment to react firmly. Let us then proceed to discuss whether and when international reactions are likely in democratizing processes – by what means, with what degree of consensus, and for how long. In doing this, the different dilemmas will be illustrated, and, as will be made clear, several dilemmas may emerge at the same time. This helps to show the complexities facing political decision-makers when various triggering events make decisions necessary.

International response: react or not? The systematic dilemma refers to the issue of ownership and posits the question of whether there is (or should be) international or local control over democracy building. Ideologically, the international community would normally adhere to the position of giving primacy to local ownership. However, many democratization processes are nurtured by international actors, through, for instance, development assistance. Thus, there is international involvement, making triggering events not only a matter of local politics but also an international concern. Furthermore, from the point of view of the international system, a postwar crisis in implementation of a peace agreement or of an agreed democratizing constitution can be seen as an early warning sign of a possible return to armed conflict. To react also becomes a matter of conflict prevention. A common lesson from the 1990s is the need to react early to have a maximum impact. The primary responsibility for the UN is to prevent war. Security Council actions under Chapter VII can only be taken to deal with a threat to international peace and security. The idea of conflict prevention, however, highlights the possibility of reacting under the less demanding Chapter VI which asks for

220

Peter Wallensteen

the pacific settlement of disputes. Democracy promotion per se is not in the UN Charter, and action in this field will require broad agreement to be pursued. Since the end of the Cold War, the UN Security Council has regularly concerned itself with internal affairs and also with democratic conditions. The report by the UN Secretary-General called ‘‘An Agenda for Democratization’’ in 1996 is credited with giving the UN a firmer role in this. The UN practice since then has made democracy a significant concern (Newman 2004: 194). These are factors that would generate a willingness to respond in the face of a crisis in democratization. At the same time, democratization is a highly sensitive process. International reactions are likely to be interpreted in terms of both who pursues them (a matter of the West versus others, i.e., a horizontal dilemma) and their impact on the local power struggles in the concerned society (who gains, who loses, i.e., affecting the vertical dilemma). The international bodies can easily be seen as partial – by intention or in effect – and supporting certain social forces. This contradicts a basic tenet in international governmental donor policy to remain ‘‘above,’’ ‘‘beyond,’’ or ‘‘outside’’ such domestic power struggles. It is an issue that has faced all peacekeeping operations in internal conflicts since the first one in the Congo in 1960. Today, the inhibitions against acting and even being partial may be lower, as long as it is justified in terms of promoting democracy. The case of Uganda demonstrates this. The threat of deteriorating conditions in Uganda’s democracy has not been a matter for the UN Security Council. It has not even passed a resolution on the ongoing armed conflict in Northern Uganda, although the issue has been addressed in other ways. The UN Security Council has taken up Uganda’s relations with the neighbors to the east and south: The Democratic Republic of Congo and Rwanda. It is not likely that the electoral question in Uganda will come to the Council’s agenda unless it can be seen to lead to a serious international situation. Thus, the UN is constrained and the principle of non-involvement in internal affairs is likely to prevail. A broad international coalition behind actions is less probable. The horizontal dilemma of who should intervene was actually highlighted and the outcome in this case has been that the UN did not act at all. Consequently, the actors dealing with Uganda’s democratic crisis were instead major donors, who have their own reasons for this (not the least that it was their taxpayers’ money that was supplied

International responses to crises of democratization

221

based on agreements that were contradicted). In other words, the question of whether and when to react, for an international body, will be limited by its mandate, which is an expression of the underlying understanding of why the organization exists in the first place. In bilateral relations other paths can be followed. A hypothesis is, however, that matters will turn out differently if the international body is already involved in the situation. As noted, the civil war in Uganda had not robustly entered the UN Security Council agenda, and as a consequence, there was no record of a UN Security Council commitment to internal democratic development. This contrasts other cases. Rich (2004) provides a list of twenty such situations, all relating to wars that have had considerable and long-term UN commitment, going from Namibia (first resolution in 1978) to Afghanistan (in 2001). For different reasons – decolonization, regionally destabilizing civil wars, and global terrorism – the world body has become engaged and thus resorted to dealing with the situation by promoting democratic solutions. It can thus be suggested that the way an international organization such as the UN enters a conflict is fundamental for understanding which democratic challenges are likely to result in action by that organization. Once committed, electoral irregularities are likely to be interpreted as issues for preventive action and the world body will remain engaged. Uganda’s situation was not on the agenda of the world body. Also the country had a record that generated international support for its policies, internally as well as externally, thus, in effect, giving more autonomy to its leader. The democracy crisis of 2005 did not connect to a previous history of Uganda as an agenda item. It may now be valuable to turn to another example. An entirely different case is the international involvement in Bosnia-Herzegovina, where the Office of the High Representative (OHR) often reacts to events which are seen to threaten the 1995 Dayton Peace Agreement and prevent progress toward a multiethnic democratic society. Thus, OHR has removed candidates from electoral rolls, restricted parliamentarians, and even replaced members of the cabinet. This has made the OHR a central source of power in the country, highly contested by all the groups. In reality, OHR is a body appointed from abroad, albeit by democratic states, and has not emerged through a democratic process in the country in which it is operating. It makes Bosnia-Herzegovina a democracy with an ‘‘enlightened’’ overlord. As the heavy international

222

Peter Wallensteen

peacekeeping presence is seen as necessary more than a decade after the peace agreement, this case also illustrates the protracted nature of a ‘‘robust’’ democracy promoting effort. The issue of democracy promotion thus relates to the horizontal dilemma by asking which body is to react to a democratization crisis. An increasing number of organizations have democracy explicitly on their agenda. This includes the EU, OSCE, and NATO, all active in conflicts and democracy issues in the larger European space. The Organization of American States (OAS) takes this role in the Americas. The African Union (AU) has an expressed objective of promoting democracy in Africa. Also international bodies less directly concerned with international peace and security do have democracy on their agenda: this is true for the United Nations Development Program (UNDP), the Organization for Economic Cooperation and Development/Development Assistance Committee (OECD/DAC), as well as the different commissioners on Human Rights (UN, EU, Council of Europe). The reports of these bodies are significant in highlighting challenges to democratic development. However, the non-governmental organizations are those most unconstrained to deal with these issues, at least as a way of alerting the governmental communities on what goes on in particular countries (Amnesty International, Human Rights Watch, International Alert, International Crisis Group). The large and increasing number of actors engaged in promoting democracy and human rights means that threats to democratic developments in post-war societies are increasingly unlikely to take place without considerable attention. This, furthermore, may in itself work as a deterrent to governments contemplating, for instance, changes in the constitution favoring the incumbent. The record in Africa of peaceful transitions from one president to the next is improving (Southall, Simutanyi, and Daniel 2005). However, often issues of candidacy for re-election are seen as sensitive matters, and neighboring or even extracontinental governments are careful not to involve themselves openly in a legislative domestic process. It will require particular conditions for explicit government reaction. The fact that a country finds itself in a post-war transition with international commitment is likely to be one such event. In this situation the international community has invested considerably in future stability and will react more legitimately than otherwise would be the case. Such an investment, it is argued here, provides the general background for international concern and even

International responses to crises of democratization

223

legitimacy for international actions in a democratization crisis. The systemic dilemma does at the same time involve a horizontal dilemma. If there is a record of commitment, an international body is more likely to get involved also in a democratization crisis and thus be highly inclusive. This means more international consensus on action. However, if the democratization crisis takes place without such a prior organizational commitment, it is less likely to have universal support. Instead, action will be pursued by regional organizations (building on regional consensus), by individual donor countries (with a commitment to democracy), or by civil society organizations. Still, there is a general trend of democratization crises generating more international concern. The question whether this results in a tilting toward more international rather than local ownership of democratization will, however, require further analysis.

International response: when? The temporal dilemma involves two aspects: when will action take place and when will it end, i.e., the short-term vs. the long-term dimension? Let us begin by raising the issue of when international actors enter a democratization crisis. This requires us to ask which challenging events are expected to lead to international reaction: It is here suggested the events in electoral processes are more probable than others to trigger an international response. This is so – as we observed earlier – because the election process is regarded as central to the conduct of democracy. According to conventional democracy theory, the electoral process and elections are the occasions when the general public (the electorate) can pass judgment on the achievements and responsibilities of the incumbent government. The public’s main sanction is to vote for an alternative to the present rulers. Peaceful change is accomplished this way. Free and fair elections are important elements of democratic participation. Thus, in democracy building the first and second elections after a war are those most closely watched. For international media, elections also incorporate the drama that provides for readers’ interest. It is a peaceful battle that can be followed and understood globally. It reduces the complexities of ‘‘distant’’ societies into manageable alternatives. With this in mind, it can be concluded that there are several reasons for elections to become central indicators of the progress of democracy promotion. Challenges to the electoral

224

Peter Wallensteen

process, then, would most typically involve events that may trigger international responses. There are, however, different elements that together constitute ‘‘the electoral process’’ and they may solicit different reactions. The example of Uganda illustrates a first type of triggering event: the patent arrest of a leading opponent. This could be generalized: physical and direct threats to opponents are events that generate considerable attention internally as well as internationally. The attempt to poison a leading opposition candidate in Ukraine in 2004 generated international concern and galvanized domestic opposition, preventing the incumbent government from manipulating the elections to its advantage. This, of course, was not a case of a postwar situation, but one which was close to both the EU and Russia, which had a strategic interest. As a consequence, it became a concern to the EU, rather than to the UN (Maksymenko 2005; Solana 2005). In a post-war situation, the physical security of the leaders is likely to be a major interest, and probably an issue to which they themselves pay considerable attention. Particularly if there is a peace agreement, such threats are closely watched by international observers. This means, paradoxically, that the safety issue may become more urgent, the further the process moves along. When a democratic transition seemingly has been stabilized, the security provisions may be relaxed. After all, the war in Uganda that led to the new Constitution ended in 1986. By 2005, matters should have been more stable. In fact, they were not. This is further illustrated by the modern experience of Lebanon. The civil war ended in an agreement in 1990. A policy of reconstruction was initiated and, by many standards, a return to a new ‘‘normalcy’’ seemed to take place. For instance, in 2000, Israel decided to withdraw from the south of the country, in the expectation of stable conditions as well as in the implementation of a gentlemen’s agreement with the dominant force in the south, Hizbollah. However, in the run-up to the 2005 elections, the leader associated with the recovery – as well as with a desire for independence from Syria’s influence – Rafik Hariri, was assassinated together with twenty others in a huge bomb blast in Beirut on February 14, 2005. The electoral process had therefore lost one of its main contenders, and this led to an international as well as a national crisis. Indeed, the democratic gains in the country seemed to be threatened. In fact, they may have been strengthened as people turned out in large demonstrations to demand Syria’s withdrawal.

International responses to crises of democratization

225

Interestingly, the issue was also brought to the UN Security Council and an international investigation was ordered, as well as the withdrawal of the Syrian troops. In April 2005, the Syrian military had pulled out and the elections in May and June gave strong support to Lebanese factions that were critical of Syria, including giving a prominent role to Saad Hariri, the son of Rafik Hariri. In this case, we see that the elimination of a democratic leader, who was also regarded as successful in maintaining economic development and balancing between different factions, immediately sparked an international reaction. Certainly, Lebanon’s geopolitical location helps explain this, but it is also significant that the stability of the country has been of continuous concern to the international and regional community (including peacekeeping and peacemaking efforts). There was considerable visibility coupled with a willingness to act. The examples of threats to leaders may be extreme cases, but there is also an issue of widespread electoral violence. Often, the fate of leaders may be only the most easily observable event of systematic practices. For the international governmental community, it may be easier to act if leaders (who may be acquaintances to other leaders) are exposed to dangers. For the international non-governmental community, not least those concerned with fair electoral policies, events such as intimidation of voters, obstacles to campaign meetings, killings of local party workers, as well as interference at polling stations are equally or more worrying. Reports on such events help form the picture an international audience acquires on a particular election campaign. For a more direct international reaction, this may not be enough. Elimination of one leader may trigger more media coverage and result in more outrage internationally than the fate of dozens or hundreds of ‘‘anonymous’’ election workers. A second set of events that might result in international action relates to the election outcome. If there is international monitoring of the process, there will be reports to work from. However, if there is such a presence it is more likely that the outcomes are correct. The nonobserved cases are probably those with the most irregularities, particularly in post-war conditions. A paradox is that such elections generate less international concern, and those outcomes may be more readily accepted, although they are domestically contested. The result, as a consequence, is that internationally observed elections are also those that are the most likely to trigger international reactions. If the

226

Peter Wallensteen

electoral conduct can be criticized, and if the outcome is also affected this is likely to spur international responses. Attempts by incumbents to ‘‘steal an election’’ in a process of great international significance is likely to be resisted. A now almost classical case in point is the outcome of the presidential elections in the Federal Republic of Yugoslavia in September 2000. Yugoslavia (today Serbia) found itself in a post-war situation, following its involvement in the wars of Slovenia, Croatia, and BosniaHerzegovina (1991–1995), as well as in Kosovo in 1999. The incumbent, Slobodan Milosevic, was running for re-election. The electoral announcement of his victory was met with disbelief, a general strike was announced, and demonstrators assembled around the Parliament building in Belgrade in early October 2000. The international reactions were sharp. It was clear, for instance, that international sanctions would not be lifted unless the results were correctly reported. In the end, Milosevic had to admit that he had lost the elections and he stepped down. This development was actually not part of the peace agreements or other post-war arrangements. The events of 2000 were a matter of following the constitutional set-up arranged by the domestic political process. Still, the outcome had repercussions for the entire region, and a legitimate outcome was important for the possibilities of peacemaking in the post-war societies of the Western Balkans. The elections received intense international coverage. The manipulation of the results triggered sufficient action to make clear to the incumbent government that it would not enjoy international legitimacy if it persisted. The domestic opposition may in the end have been more important in Milosevic’s decision to step down. His power base was eroded, but also his main international allies, notably Russia, abandoned him during the first days of October 2000. The triggering event was electoral fraud: Milosevic was a key regional actor, the legitimacy issue (an element in the vertical dilemma) was no longer in his favor (but rather to the advantage of the international actors), and there was sufficient consensus among key actors to unite against him (an aspect of the horizontal dilemma). By his actions, Milosevic orchestrated his own downfall. An additional aspect is the aftermath of the elections; will the winner and loser have a chance to play the role the results demand? As the previous examples suggest, it is difficult for a national government to withstand the pressures resulting from an electoral process. It may face

International responses to crises of democratization

227

the inevitable choice of giving up power. From the point of view of the power holders this may be a difficult choice. What will happen to the outgoing leader, the losing administration, the party, his/her family? There are many elements of an internal security dilemma posing agonizing choices on the loser (Wallensteen 2007). A number of leaders have had to face this, and chosen to abide by the result. The list includes Daniel Ortega in Nicaragua, Kenneth Kaunda in Zambia, Daniel arap Moi in Kenya. In all these cases, the former leaders have been able to make a living in their own countries following the electoral defeat.1 The recent advent of war crime tribunals may suggest one more disincentive for leaders in post-war societies to step down. Milosevic found himself at the International Tribunal in The Hague less than a year after his electoral defeat. Other leaders, such as Moi, may face charges of corruption (as happened to Frederick Chiluba, who overturned Kaunda and himself lost a later election). In these instances, the international community is recorded to have acted in favor of the loser yielding power. In some instances, obeying by the outcome has been met with resistance, however, resulting in protracted stalemates. In 1988, war-torn Burma/Myanmar had a democratic election to appoint a constitutional conference. The democratic opposition clearly had the upper hand. The National League for Democracy led by Ms. Aung San Suu-kyi won overwhelming support and as a consequence should have had the dominant role in Burmese politics. The military regime, however, refused to give up its power. A stalemate resulted, still remaining. The pressure on the military regime has been strong, including industrial sanctions by the United States and targeted action by the European Union. The Nobel Peace Prize was awarded to Ms. Aung San Suu-kyi in 1991. The UN Secretary-General has appointed a special representative for this issue, as has the Association of Southeast Asian Nations (ASEAN). The pressures are obvious, and the neighboring countries have found themselves in a squeeze. Lately, ASEAN has become more vocal in demanding democratic reform in Burma. Burma’s closest ally, 1

Most remarkably, Daniel Ortega was again elected president in Nicaragua in 2006, thus, being at the same time the first president in the country’s history to step down peacefully and be re-elected. In 2007, the Sudanese businessman Mo Ibrahim announced a substantial prize for African leaders who ‘‘have taken office through proper elections and left having served the constitutional term stipulated when taking office,’’ in effect rewarding constitutional behavior.

228

Peter Wallensteen

China, however, has not imposed any action, preferring to build up its own investment in the country. The Burmese issue has been blocked from entering the formal agenda of the UN Security Council, but an informal discussion has been allowed. The military government has defended itself by pointing to its ability to conclude agreements with a number of the ethnically based insurgencies. This, it has seen as a development toward a more multinational state. However, given this, international actors have argued that now may be the time to deal with the issue of national governance and democracy, as part of a post-war process and as a way of preventing a return to serious conflict. The observations so far point to the electoral process, particularly the election campaign, as a triggering event for international action in a democratization crisis. Clearly, also the outcome will be closely monitored by international actors. Possibly, there is a sequence of attention: if the campaign has been followed, it is likely that also the outcome is of concern. In other ways, if the issue is on the agenda then it will not easily go away, be it an international or regional organization or a coalition of donors that are concerned. It still asks the question raised by the temporal dilemma: is there a preference for short-term commitment, or will it be sustained over a period of time? To this we now turn.

International response: for how long? The duration of commitment is a second aspect of the temporal dilemma. It will partly, of course, depend on what type of action is taken (to which we will turn in the next section). Normally, action would end when a clear-cut success or failure can be established. The promotion of democracy, particularly in post-war conditions, seldom offers such easy points. In reality, for instance, peacekeeping operations will be sustained as long as the troop-contributing countries are willing to put up forces, or funds are enough for continuing. As there are always some possible threats to refer to, the operations can become protracted. There is a similar tendency for economic sanctions to go on for a considerable period of time. Even when sanctions are targeted, it can be politically difficult for the initiators to terminate them, as that might be seen as yielding to the targeted country or its regime. If the demands are linked to decolonization, however, independence might be a clear event for major disengagement (e.g., East Timor in 2002). Even so, the concern for democracy may extend far beyond such points.

International responses to crises of democratization

229

Some of our examples – Lebanon, Uganda, Bosnia-Herzegovina, Burma/Myanmar, and, more recently Liberia – point to the long-term nature of democracy building in post-war conditions. Formative crises might come late in the process, even after a series of elections. As we have already mentioned repeatedly, once an issue is on the international agenda, it is likely to remain there for a considerable period. The attention may vary across time, however. One might even suggest that there are different phases of such interest. In the war period, the focus is likely to be on issues of negotiations and peace agreements. This involves international diplomacy, special missions, mediation efforts, strategic political considerations, etc. Once a war has ended, there might be a requirement for peace operations, humanitarian efforts, refugee assistance, demobilization, and technical support for the implementation of agreements. When this is seemingly managed, it will be followed by international development assistance aiming at recovering from the war effects and the creation of self-generating economic conditions. Supposedly, from this will follow the building of democratic conditions in all its aspects (for instance, media, civil society, rule of law, academic freedom, security sector reforms). After this, aid will be replaced with trade and promotion of democracy with the indigenous solidification of good governance. This is an ideal model. In fact, the international commitment may wane very quickly. For instance, once refugees have returned, armies have been reduced, and the first elections have been conducted, global concerns will shift to other crises. ‘‘Real’’ attention will not return until a new crisis unfolds. Lebanon might be a case in point. The international commitment between 1990 and 2005 was limited to particular issues. Similarly, Haiti was of great concern in 1993–1994, with sanctions and threat of military actions unless the deposed government was allowed to return. It did so in 1994, and Haiti dropped from the immediate concern of international organizations. In 2004 a new crisis took place. What had been achieved in a period of ten years did not provide for a sustained democratic form of government. In the case of Uganda, Museveni managed to win the elections, and the donors faced a new dilemma. Were they going to sustain their actions against the regime or were there other means available to deal with the government? Donors gradually reestablished relations. For instance, half a year later, Sweden restarted 90 percent of its budgetary support, arguing on the basis of an EU evaluation that there was now

230

Peter Wallensteen

multi-party representation in the parliament. Ten percent was withheld as an indication of continued concern. The conduct of the elections process was not fully acceptable, but the outcome was likely to have been the same, a victory for the incumbent. However, to work with Uganda and the new parliament would increase the chances of furthering democracy in the long run. ‘‘A democratic state is not shaped overnight,’’ said the head of Sida’s Africa Division (Sida 2006). Thus, democracy promotion in post-war conditions is a long-term commitment, but international high-level attention is impossible to maintain for such a period of time. New issues will emerge and rightfully deserve their shares of attention. The realpolitik conditions may also dictate against a sustained and coherent democracy approach: provision of aid to exposed populations requires a functioning state. Thus, Uganda could find a way back to donor funding. If Uganda, Afghanistan, or Bosnia-Herzegovina were to relapse into war, considerable efforts and resources would have been wasted. Thus, there is a need to develop a sustained international interest in post-war conditions. In 2005, the UN General Assembly and the Security Council jointly agreed to set up a Peacebuilding Commission. In the words of Assembly President Jan Eliasson (Sweden) this would mean that ‘‘post-conflict does not mean post-engagement of the international community.’’ The task of this commission was to keep countries on the ‘‘verge of lapsing or relapsing into conflict’’ on the agenda of UN action (UN DPI, December 27, 2005). This is the first organizational expression of the need for sustained efforts in peacebuilding, including the promotion of democracy. It testifies to a general recognition of the exit problem, as a problem of premature, or too early, departure by the international community. During its first year of operation, the new commission selected two situations to work with, in agreement with the governments: Burundi and Sierra Leone. Both countries are operating within a democratic framework, and refining this is one of the ambitions of this new cooperation. The temporal dilemma points to the issue of commitment. Clearly, democracy building requires long-term efforts, but management of crises requires short-term action. What might be emerging is an awareness of the need for long-term international commitment. This, however, will also require the creation of new institutions or other durable arrangements (bilateral agreements, regional organs, development cooperation, etc). It cannot solely rest on the type of early reactions

International responses to crises of democratization

231

that we see when triggering events take place. Short-term actions, often negative in nature (sanctions of different sorts), ultimately yield in favor of more lasting measures. Still, when facing a crisis, the dilemma is activated. There is a need for short-term reaction, without at the same time jeopardizing long-term investment. Skillful political leaders can exploit this dilemma to their advantage. After all, Museveni was re-elected and the outcome accepted. The donors’ choice between demonstrating dissatisfaction and jeopardizing long-term gains was balanced in the direction of a sustained commitment. The dilemma remains, however.

International reactions: by which means? There is an additional dilemma to consider: the vertical dilemma, which posits legitimacy versus efficacy: Which are the measures used by the international community to react to a democratization crisis? Are they legitimate, in the eyes of that community as well as to the inhabitants of the country exposed to the actions, and are they also effective in bringing about the changes expected? Reacting to a democratization crisis involves delicate issues of legitimacy, as it concerns the sovereignty of the targeted country. For the countries contemplating or carrying out action it has to do with their investment in the situation. There may have been a long-term commitment, for instance, through a peace process during the war period. Donors have supported the post-war developments. Thus, to them it is legitimate to take action. It might be different on the other side of the divide: government and population in the targeted country. The government is likely to be suspicious; the population may be more divided. This affects the means chosen. The issues of governance (including a democratization crisis) are within the sovereignty of a member state of the international community (be it the UN or a regional body). This may make other member states hesitant to act. The post-war conditions add to this. Humanitarian concerns will enter into the decision making as well: aid might be of critical significance for exposed groups. This differs from the typical Cold War era or to reactions to military coups. A typical reaction to military coups has been not to allow the particular regime a seat in the UN. The suspension of diplomatic recognition (or not extending relations to a new regime) was a common measure. There were cases of open or clandestine military interventions to

232

Peter Wallensteen

impact on the domestic dynamics, either favoring democratization or protecting democratic gains (Western support to armed opposition in Afghanistan against the Soviet Union may be an example of the former, the US intervention in Panama in 1989 an example of the latter). These measures are still available, as seen in the case of Iraq in 2003, but the resort to actions which aim at the removal of a regime or the exclusion of an entire country have been limited to a few situations. The postCold War era has resulted in a need for refined ways of reacting. This points to ways in which the vertical dilemma is handled: finding ways which are at the same time legitimate and effective. A solution has been to find targeted sanctions, that focus on the responsible actors rather than on the population; or affect the revenue of the government in selected areas (Wallensteen and Staibano 2005). Some examples may help us to pursue the discussion. After about fifteen years of post-war transition Lebanon could be regarded as having recovered as a functioning society, albeit under heavy Syrian tutelage. There were international contributions to reconstruct the country’s democratic power-sharing system. Financial institutions were again returning, trade relations were beginning to reestablish themselves. There are many lessons to be learned from this, in terms of long-term peacebuilding with a democratic framework in mind. With the assassinations of the former Lebanese Prime Minister Rafik Hariri and twenty-two others on February 14, 2005, however, the country was again thrown back into severe crisis. This time, the international reaction was immediate, and the murder entered the agenda of the UN Security Council. A special prosecutor was appointed and a report was delivered, pointing to the involvement of Syria in the plot. The newly elected parliament was decidedly more critical of Syria’s role in the country. The conditions did not remain calm, however. Throughout 2005 there were car bombs and political leaders critical of Syria were killed. Once the prosecutor’s report was delivered, a new crisis threatened. By the end of October the Security Council sharpened its attitude, imposing sanctions on those individuals that were mentioned in the report (e.g., travel bans and freezing of assets) in order to ensure that the legal process would be maintained. In Syria this was perceived to be part of a strategy of isolating the country. The issue of the democratic process in Lebanon began to take on new dimensions: it threatened to become a matter of dealing with state-sponsored terrorism and the regime in Syria itself, thus connecting to the entire web of

International responses to crises of democratization

233

Middle East politics. Interestingly, the Security Council maintained its uniform position.2 Targeted sanctions as well as the UN investigation that focused on individuals rather than on Syria’s involvement were instrumental in this process. The approach, in other words, was to target particular actors that could potentially undermine the democratic gains in Lebanon. The government welcomed the international engagement. Uganda provides a different story. In this case wars and repression also lasted for about fifteen years and the reconstruction took an equally long time. As we already observed, the international community embraced Museveni and for many years applauded his regime. The democratization crisis in 2005, thus, was a novel experience for a leader that had so far enjoyed good reviews internationally. Also for that community it was difficult to reverse its position. The arrest of the opposition leader was, indeed, unprovoked and became a formative event for international reactions. This was, nevertheless, less severe than the killings that were witnessed in Lebanon. However, there was no expectation of violence or human rights abuses. The fear of the elections of 2006 not meeting international standards may have triggered reaction more than anything else. The responses were different: criticism by some states at the Commonwealth summit meeting in Malta in November, the reduction of direct budgetary aid by significant donors. The resulting reaction of Museveni was not one of immediately releasing the leading opponent but instead he argued that these responses showed that ‘‘aid is arrogantly mixed up with an effort to interfere with our sovereignty’’ (Museveni 2005). Museveni, who had long been a benefactor of international support, now tried to define such support as colonial, and himself as a man of independence. The donors that reacted thus had made the arrest of the opposition leaders a major public event. The government resisted international engagement, but in the end the opponent was released while other forms of harassment continued. Through international attention and limited actions, the international community may thus have achieved some of its goals, and certainly made the elections more closely watched.

2

Obviously the war in 2006 between Israel and Hizbollah challenged many of these democracy gains, but the Lebanese government still remained in office in early 2007.

234

Peter Wallensteen

The sanctions approach in the case of Lebanon–Syria was targeted at particular individuals, thus seemingly freeing the Syrian regime itself. The regime could presumably deal with this by forcing these individuals to cooperate. By doing this, the regime might prevent itself from being the next target. The implicit threat of escalating actions by the international community may have had an impact. The actions on Uganda, however, covered budgetary support that went straight into the treasury of the government. The impact would affect Uganda’s ability to conduct national policy. Whether the sums blocked actually were large enough for such an effect can be debated. The demands built on previous bilateral agreements but may have been somewhat diffuse. It held out the possibility that the release of the arrested opponents would restore budgetary aid, at some point. At the same time there was an implicit threat of escalation of donor measures. The demands of the international community and the actions taken are, thus, two considerations which are intrinsically linked. The measures chosen are likely to be proportional, however difficult it may be to determine that balance. In addition, there will be a need for international agreement on the actions. In the Syria–Lebanon case measures has been taken by the Security Council, where such coordination is routine. The actions on Uganda lacked such a forum. The options available, furthermore, would not be global in reach. For instance, donor countries coordinate through OECD/DAC or through EU, both organizations covering about one eighth of all states. Uganda might claim an ability to replace aid lost from some donors with aid from other countries (although China showed no such record). The sums involved might not have been prohibitive either. The concerted actions of the Security Council, in other words, would have advantages of more precise demands, targeted action, and international agreement. Actions by other actors would suffer from having too wide demands, difficulties in targeting, and difficulties in achieving international (near-) consensus. One might say that the choice of action in both cases, targeted sanctions, was a solution to the vertical dilemma. They had legitimacy (being part of the UN, or being undertaken by democratic states), they avoided affecting ‘‘ordinary’’ citizens (thus not being punitive), and they were focused on particular individuals or capacities of the government (travel, financial assets, budgets, respectively). A course of action was chosen trying to combine legitimacy with efficacy, thus suggesting a way to manage the vertical dilemma. In

International responses to crises of democratization

235

the case of Syria–Lebanon some cooperation was achieved; in the case of Uganda, the actions did possibly modify the incumbent’s policy. These two cases demonstrate traditional versions of sanctions: by breaking off certain types of relations, and inflicting some pain, there is an expectation of change. There are some new options developing in international practice, however, pointing to a second category: positive sanctions (Wallensteen and Staibano 2005: ch. 15). Such sanctions have also been discussed in terms of incentives or rewards (Dorussen 2001; Rothchild 1997). A case not mentioned so far in this contribution is instructive: Liberia. The most recent civil war ended in 2003, by the removal of the incumbent, Charles Taylor (he fled to Nigeria, from where he was extradited to the court in Sierra Leone in 2006, although that court was based in the Netherlands). An interim government was installed. UN peacekeepers were sent to the country. UN sanctions had a role in achieving this change. By preventing the exports of diamonds and timber from Liberia, one of the main sources for pursuing the civil war was blocked, particularly for Taylor. In addition, Taylor and some associates were exposed to financial and travel sanctions. When the war ended there was a general expectation that the sanctions would be lifted. The Security Council did not do so, however. Instead sanctions were maintained and were regarded as a way of supporting post-war reconstruction and democracy. By maintaining sanctions, no faction in Liberia could restart war using these sources of income. The Security Council wanted to make sure that a responsible government would be in place and that these resources would be used for the benefit of the entire economy of the country. Thus, Liberia sustained itself on humanitarian support. The presidential elections of 2005 resulted in a runoff of two candidates. The Security Council held out the possibility that the winner, who in the end was Mrs. Johnson Sirleaf, would benefit from the lifting of sanctions, once the government was able to control the resources in question. A carrot was in place. The approach, nevertheless, suggested a development in the practice of international reactions. During 2006 sanctions were eased, contributing to increased governmental revenue by early 2007. Sanctions thus have legitimacy (being a measure in the UN Charter), and the efficacy of this measure has gradually improved, no longer being comprehensive and general but targeted and with positive options. There has been some learning from the negative humanitarian effects in the case of Iraq during the 1990s. Practice has improved. The

236

Peter Wallensteen

vertical dilemma expresses the problem well and the international community has seemingly understood how to deal with this for furthering democracy. Still, the record of actually achieving democratizing effects is in doubt. Lebanon remains in a fragile state, Uganda may see increasing authoritarianism, while Liberia may have a historical chance. International responses to democratization crises are often too late to be effective.

Conclusions In this chapter the experiences in international post-war democratization crisis management have been exposed. There has been a strong, global tendency for the international community to engage in the internal affairs of member states more often. This seems like a logical consequence of the fact that most of the wars since the end of the Cold War have concerned intrastate matters. In such situations, the creation or re-creation of democratic forms of governments have become central. As this chapter has illustrated, this situation generates a number of delicate dilemmas, and I have suggested ways in which they have been dealt with by the international community. First, there is a systemic dilemma: whether to react or not. A hypothesis has been suggested: local threats to the electoral process in situations where the international community has already been involved seem to trigger reaction without involving serious sovereignty problems. Also, there are particular moments in the process that draw attention: physical threats or actions against leading opponents (democratic proponents), followed by the manipulation of outcomes of such elections. In many other instances of threats to the electoral process, the international community is not likely to act strongly. Instead, threats to constitutional processes, the undermining of the rule of law and freedom of speech are often referred to non-public diplomatic dialogues. These issues may, on occasion, enter into bilateral cooperation agreements. Second, the international organizations face an obvious temporal dilemma: if one is to react, when is that to be done (early or late?) and for how long is action to be maintained? There is a tendency to crisis reaction, meaning that it might be too late to be effective (as earlier signs have been ignored), at the same time that there is an urge to maintain long-term relations (thus potentially reducing the impact of the short-term measures).

International responses to crises of democratization

237

Third, there are also key elements of the vertical dilemma: by what means, that is, which are the most effective ones that can legitimately be used? This chapter points to the resort to targeted measures (different forms of sanctions) as a way of mitigating the dilemma. These are measures supported by international agreements; at the same time they are not harming the population at large or the fragile democratization process. The targeting in itself is a problem that has to be pursued: one may suggest that if it is the opponents to democratization that are targeted, there may be more effectiveness. Fourth, the chapter has illustrated the horizontal dilemma throughout the text. Central for international bodies and for international actions is the ability to muster international support (for instance, from the UN Security Council). This might be highly generic to international action. When it is not possible to get broad-based international backing, actors have pursued other ways, for instance regional organizations or donor meetings. In such, more limited settings, the agreement on democracy promotion may be higher. At the same time actions may be undermined by other actors, who step in to prop up the target. Thus, actions will be limited and, possibly, short-lived (as illustrated in the case of Uganda). Sustained, effective, legitimate action, in other words, will have to be based on international support (e.g., through the UN, as in Lebanon and Liberia). Democratization crises are likely to activate all these dilemmas, more or less at the same time. This may often make it difficult to predict which actions will be taken when the international community faces a problem. Policymakers looking for an appropriate mixture of action will more or less instantaneously consider whether there is a sufficiently triggering event, what options are available, what international support can be marshaled, what would be effective, and for how long action can be maintained. The options available are key. Recent experience has stimulated new thinking. One is the use of targeted sanctions. There is a range of options in the mandates and composition of peacekeeping operations. There is also the creation of the UN Peacebuilding Commission, concentrating on the post-war period of situations on the UN agenda. There is the formation of more positions as Special Representatives (by the UN, EU, ASEAN, AU, etc.) with varying assignments and resources. An example is the appointment in 2004 of a special advisor of the UN Secretary-General on the prevention of genocide, which also

238

Peter Wallensteen

has to deal specifically with post-war situations as they have a higher likelihood of genocidal actions. There is now a similar position on conflict prevention. Remarkably, however, there is not yet a Special UN Envoy for democracy building. Similarly, the re-creation of the UN’s work on human rights, with the new Human Rights Council in 2006, indicates a wider support of concerted action, where also democratization crises may receive earlier attention. A Democracy Fund has been established within the UN, and international organizations demonstrate an increased willingness to entertain inputs from civil society organizations. Together, these innovations may signal new instruments for quicker action as well as a longterm readiness for the international community to remain engaged, for the benefit of democracy and peace.

9

Peacebuilding as democratization: findings and recommendations TIMOTHY D. SISK

Peacebuilding, as a leitmotif concept guiding policy strategy for international intervention in war-torn societies that have negotiated a settlement, has matured considerably since the term first entered the lexicon of the international community in the immediate years after the Cold War’s end. Peacebuilding evolved dramatically from its first articulation in United Nations Secretary-General Boutros-Boutros Ghali’s Agenda for Peace in 1992 to the landmark creation of a permanent institution at the UN, the Peacebuilding Commission in 2006, to advance the aim of preventing the recurrence of war. Likewise, policy practice has equally evolved from the initial experiences of ‘‘complex, multidimensional peace operations’’ in Namibia and Cambodia, for example, to the challenge of building states in intractable conflicts such as Afghanistan or Liberia. Common to the peacebuilding experiences of this era and today is the reliance on a principal formula for durable war termination: the introduction of democracy as a way to move the theatre of conflict off the battlefield and into the institutions and processes of politics. As the authors in this book forcefully show, this formula for durable peace is premised on a fundamental contradiction that presents policy practitioners with a serious set of dilemmas: pursuit of democracy can undermine efforts to secure peace, and efforts to secure peace can undermine the meaning and quality of democracy. Thus, in practice, the promotion of democracy and the pursuit of peace can work at cross purposes.1

1

Moreover, the conceptualization of linkages between peacebuilding and democratization have also been clouded by the intervention of the US-led coalition in Iraq in 2003, which sought to unilaterally impose a process of democratization after the fall of Saddam Hussein’s regime; this democratization process unfolded with escalating violence such that the military occupation and democratization efforts have in effect precipitated a failed state environment in

239

240

Timothy D. Sisk

In transitions to democracy following deadly wars, international mediators and protagonists alike face trade-offs that cannot be easily reconciled or avoided. Indeed, at times choices must be made between competing priorities at different moments in difficult processes; usually, and understandably, international policymakers choose peace over democratization and human rights because stopping the killing is a higher-order priority than seeking justice or promoting multi-party electoral competition (as Anna K. Jarstad points out in her conceptual chapter in this volume). It is a common mantra among policymakers that democratization is unlikely to proceed until peace is achieved (an assumption only partially supported by this research). The wide-ranging challenges of war-to-democracy transitions explored in this volume reinforce the realization that the international community must have sober expectations about the partial and restricted nature of outcomes in post-war transitions: neither the peacemaking nor the democratization aims are likely to be fully realized in transitional processes.2 The finding that policymakers often choose pursuing peace over democracy and justice should not lead to pessimism, for the research in this volume also shows that there are situations when innovation and careful planning can surmount the dilemmas such that peace and democratization can progress more or less simultaneously. The central task, then, is not whether democratization should or should not occur, whether democratization should wait for peace or vice versa, or even how long democracy processes such as elections should be delayed (a decision that will in any event vary widely according to circumstance). Confronting the deep dilemma between conflict management and democratization involves designing ways in which the conflict-inducing nature of transitional processes can be mitigated such that the initial constraints upon democratization that arise from peace imperatives can, over time, fall away as trust and legitimacy ostensibly build in the post-war period. This concluding chapter evaluates the summary findings of systematic evaluation of dilemmas in war-to-democracy transitions and explores

2

Iraq and levels of internecine conflict that have the characteristics of civil war. For an analysis, see Diamond (2005). Thomas Carothers does a good job of highlighting the partial nature of transitional outcomes in his piece, ‘‘The End of the Transition Paradigm,’’ reproduced in his book (2004), together with rejoinders from other prominent analysts.

Peacebuilding as democratization

241

the implications of these findings for international policymakers. The chapter synthesizes the ways in which each type of dilemma – horizontal, vertical, systemic, and temporal – is experienced in the issue areas explored by the authors. The chapter also seeks to make the link between research findings and policy implications. The need for more effective, theory-informed strategic directions to policy is especially acute at a time when there is significant reform of international institutions mandated to realize international peace, such as the Peacebuilding Commission and a United Nations Democracy Fund (also launched in 2006).3 For policy formulation and programming by these potentially promising new entities, and in the decision-making processes of the Security Council, understanding democratization dilemmas is the first step to improving the capacity to confront or avoid them. The policy implications of this research are organized into four categories: designing transitions from war to democracy, managing political violence, engaging civil society, and improving international stewardship. The chapter ends with a call for further evolution in the conceptualization of peacebuilding today and in particular ways to redress the myopic approach that plagues present efforts to create sustainable peace through democratization. As Peter Wallensteen observes in his chapter, crisis management by international actors such as the UN needs to have a clearer appreciation of the fundamental dilemma between external supply of, and internal demand for, democracy and greater ability to respond earlier to crises of democratization in war-torn societies when they occur . . . often years after a war-ending settlement has been reached.

Findings The contributions of this volume reveal how the four types of dilemmas articulated in Jarstad’s theory chapter are experienced by international conciliators and protagonists in contemporary efforts to settle civil wars through the democratization formula in a critical set of issue areas. What synthesis findings emerge from the analysis? The first finding is that often the conundrums of war-to-democracy transitions are addressed not once a peace agreement is signed, but 3

Analysis of the UN’s role in democracy promotion is found in Newman and Rich (2004).

242

Timothy D. Sisk

in the course of negotiating settlements themselves. As I. William Zartman has noted (1991), in the evaluation of a phase of peace process – the post-agreement phase in this instance – one must look to antecedent decisions made in prior moments of bargaining. Thus, tough choices are often made in the negotiation of peace agreements that affect the postsettlement phase of implementation. Among the critical issues that are addressed in early phases of bargaining are the following:

 The nature and relative balance of power of various protagonists in









4

the war, to include the nature of elites who organize and wield power (e.g., nationalist, ideological, or economic elites), the nature and capabilities of military factions (e.g., government forces, rebel groups, private militias); The possible types of sequences for setting up a transition – to include protagonist perspectives on interim, transitional regimes, on constitutional or final status questions, and in the imagination of how various choices among political institutions such as elections (such as electoral system choice) or territorial matters (federalism, autonomy, division, or unitary states) affect the interests of the protagonists in terms of their primary objectives; The economic base of the society and in particular the political economy of the war – over issues of absolute scarcity as in Somalia or over ‘‘lootable goods’’ in Sierra Leone or Angola – to include the international dimensions of economic support for protagonists (as in diaspora support for Sri Lanka’s rebels) or in economic dependencies in terms of trade and investment (such as Angola’s oil revenues for the government); The social divisions that exist to include the structure and nature of civil society (e.g., to change the regime, to be included in power, to advocate for minority interests, or to divide the state through secession); and The nature, extent, and resources of external involvement by international organizations such as the UN, regional organizations, pivotal global or regional states, non-governmental organizations, and eminent persons.4 The contributors to this volume are indebted to Pauline Baker, who made these observations at the 2006 Vail Symposium, ‘‘Dilemmas of Democratization in War-Torn Societies,’’ at which the book’s chapters were presented, together with Tom Farer who emphasized the economic dimension.

Peacebuilding as democratization

243

Understanding and addressing the trade-offs of the post-settlement phase thus find their origins in the difficult issues of negotiating peace settlements, where some of the early choices – such as who is included and who is excluded in peace processes or democratization pacts – is decided. Second, the trade-offs evaluated here are not dichotomous choices in most instances: there are shades of gray in ways in which the problems have been handled. For example, the evaluation of post-Dayton Bosnia is often cited as an instance in which the cause of an ideal, socially integrative democracy – not based on the ethnic divisions over which the war was fought – was sacrificed in the interests of peace. Yet postDayton Bosnia has experienced elements of democracy, to include reasonably free and fair elections, greater protection of human rights, and – as Roberto Belloni shows – resuscitation of an integrated civil society. Similarly, the presence of peace in some post-settlement situations is also not an ideal that has been fully achieved, as Kristine Ho¨glund’s analysis of post-settlement violence vividly demonstrates. Thus, most war-torn societies find themselves somewhere in between war and peace and war and democracy. The answer to this conceptual gray zone – which afflicts scholarship and policymaking alike – is to conceptualize war-to-democracy transitions not just in terms of static regime types, but as dynamic situations that can experience progress and regression in the consolidation of post-war peace through democratization processes.

Horizontal dilemmas The most pervasive of these dilemmas, readily seen in the foregoing analysis in this volume, is one of inclusion and exclusion: horizontal dilemmas. This dilemma is very much about inclusion or exclusion of ‘‘spoilers,’’ but it goes well beyond that principal concern, for example, on how electoral processes affect the inclusion or exclusion in governing coalitions, or the involvement of civil society in negotiation processes. In horizontal dilemmas, the critical question revolves around the relationships among the parties in civil war: the roles of states (political elites and military forces), rebel forces, and other factions with the capacity to engage in violence such as militias. Much has to do with these parties’ ideological or cause-of-conflict position (e.g., ethnonationalist, religious fundamentalist, or perhaps Marxist/Maoist

244

Timothy D. Sisk

orientation), their bargaining dynamics, and their worthiness as parties to agreements of the social contract type reflected in their conduct during the war and ongoing into the peace. It may also occur in considerations of their behavior during the war: should amnesty be granted to all former combatants, including those who committed gross violations of human rights? Likewise, much depends on parties’ resource endowments and the networks of political economy they use to fund their capacity for violence. In the chapter by Ho¨glund on violence, for example, we learn that efforts to create strict preconditions for inclusion (such as linking Sinn Fein participation in Northern Ireland’s political process to abandoning the use of violence by the IRA) may induce parties to evade the criteria by shifting tactics from one form of violence (punishment killings) to another form (punishment beatings). Efforts to minimize violence by bringing in those who bear responsibility for killing – such as rebel forces, splinter factions, or ‘‘rogue’’ elements of the security forces – raises concerns of appeasement, of the shelving of human rights concerns, or in some instances (such as Afghanistan or Liberia) the creation of a ‘‘warlord’’ democracy (Wantchekon 2004). Benjamin Reilly’s contribution carefully shows how electoral processes not only define the nature of governing coalitions and the validation and legitimacy of negotiated settlements, but that electoral processes fundamentally shape the nature of representation in the war-torn societies context. Whether underlying social forces are included or excluded – and how that representation actually takes place – is often a function of the critical electoral system choice. In this regard, electoral processes present clear trade-offs between desirable features of democracy that elections can provide: proportionality of representation by group or political party and accountability of individual candidates, between inclusion of representatives of a wide swath of society and the creation of cohesive, effective governing coalitions, and between the realities of identity-based voting behavior and policies designed to induce moderation in societies emerging from civil war. Innovative engineering to ameliorate these tensions, while worthy to pursue, is limited by the realities of context (Reilly 2001). It is for this reason, in part, that Jarstad’s observations about the need in some instances to include representation quotas for some groups in power-sharing pacts needs to be heeded. At the same time, there is the equally important consideration that inclusion and

Peacebuilding as democratization

245

exclusion is not just about warring parties: in most situations of wartorn societies, there are large numbers of internally displaced or refugees whose interests may not be well represented by the organized factions at the negotiating table or in an interim government; devising ways for participation of these excluded interests in electoral processes, for example, is an urgent task (Lacy 2004).

Vertical dilemmas Vertical dilemmas, these authors show, arise in relation to what can be accurately described as the elite–mass nexus. Here, in issues of voter participation or in engagement of civil society, dilemmas arise between efficacy and efficiency and the equally desirable pursuit of legitimacy. Elites make the deals, but peace and democratization – to be sustainable – need to involve the people. Whether in constitution-making processes that yield power-sharing agreements, or in an electoral process in which people formally participate in the war-to-democracy transition, a reference to the people in peace processes is costly and potentially subjects the carefully balanced elite pacts to ‘‘disruption from below.’’ Particularly, the research here shows that mid-level elites, such as rank-and-file of state military or rebel forces, or mid-level political entrepreneurs, are the under-appreciated actors in vertical dilemmas. The broad public is often too devastated from war – displaced or dejected – to have the capacity for extensive participation in the war-to-democracy transition process. Indeed, their attitudes during the war may well have hardened, leading to maximalist claims that cannot be reconciled with the imperatives of peace. Vertical dilemmas are all about the practical difficulties in pursuing a strategy of conflict transformation, in which all elements of society – from elites to grass roots – are involved in the peacemaking process; conflict transformation is so difficult to implement precisely because the causes of conflict have deep structural roots that feature ‘‘horizontal inequalities.’’ When wealth and control over resources overlaps with other identities, for conflict transformation social forces will experience a realignment that can be highly conflict-inducing.5 Thus, the project of state-building is equally about creating the means for 5

John Paul Lederach (1997), argues that three levels need to be involved to achieve ethnic reconciliation. At the top level, political and military leaders are involved in

246

Timothy D. Sisk

fairer distribution of resources and creating the capacity of the state to deliver services equitably, provide educational, health, and public sector opportunities, improve policing and public safety, and create the conditions for economic prosperity.6 Persistent human rights abuses and recourse to the use of force by governments and rebel factions undermine the impetus to peace generated by a negotiated settlement to civil wars. Electoral violence, assassinations, and public riots are all indicative that societies torn by war are prone to renewed and/or ongoing violence, the origins of which are found deep in society. Likewise, when such instances of transitional violence do occur, often it is the UN peace operation that must respond: generating the perception, and often the reality – as in the Democratic Republic of Congo – in which the UN becomes another party to a complex, ongoing violent encounter. The transformation of rebel forces is an essential part of peacebuilding but with potential negative consequences for democratization unless the parties fully give up the armed struggle and wage solely peaceful campaigns for power. As Mimmi So¨derberg Kovacs points out in her analysis, the warring groups must shed the identities of mobilization used during the war and reinvent themselves along lines that cross-cut divisive identities. This is illustrated by the more glorified cases of the transition, such as South Africa’s ANC conversion from guerilla force to political party, or the transformation of the republican cause in Northern Ireland from the Irish Republican Army to the political party Sinn Fein, and today’s cases of the makeover of Aceh’s Free Aceh Movement (GAM) to a political entity. Likewise, there is a clear political economy of security sector reform and rebel movement demobilization: unless the resource-capture dynamics that allow for the mobilization and operational effectiveness of warring militaries

6

high-level negotiations to achieve settlements between rivaling ethnic groups. At the middle level, economic, religious, and humanitarian leaders organize problem-solving workshops and peace committees, and provide training for conflict resolution. At the grass-roots level, local leaders and officials organize neighborhood committees and workshops for prejudice reduction. For discussion of the political economy of horizontal inequalities, see Frances Stewart (2000). See, the report of the Making States Work project (Chesterman, Ignatieff, and Thakur 2004), a collaboration of the International Peace Academy, the Carr Center for Human Rights Policy at Harvard University, and the United Nations University; also, see the work of the Research Partnership on Post-War Statebuilding at www.statebuilding.org.

Peacebuilding as democratization

247

and rebels are addressed, it is unlikely that peace will be sustainable. At the same time, demobilizing a security force without an appreciation of the deep, personal, economic effects such a move will have on individuals – and thus their incentives for political or criminal violence – will undermine the possibility of peace. With regard to linking politics with the people, power-sharing pacts constrain democracy horizontally, but they are also limited in the vertical vein as well. As Jarstad notes in this volume, ‘‘In divided societies power sharing freezes the conflict lines, and the parties do not need to compete for votes among their former foes . . . Instead, nationalist parties [in Bosnia and Herzegovina] prevail in elections.’’ She later writes that despite the finding that elections are ‘‘meaningless’’ in power-sharing situations, at times elections can make a difference in power-sharing outcomes, citing the cases of Burundi and Cyprus. At least one option for addressing the problems of elite domination of power-sharing agreements is through evaluation of the ways in which political leaders are linked to the elements of society they purport to represent, either through political parties or civil society. Reilly’s admonition in his chapter that large, integrative political parties are best able to contribute to peace and to the sustainability of long-term democratization is a critical finding on the ways in which the challenges of popular participation can be potentially resolved, allowing transitional power-sharing pacts to be less critical for peace and allowing for elections to be more meaningfully fought without endangering the peace. As the work of Roberto Belloni shows, after war civil society’s capacity to participate is deeply weakened at the same time that associations and organizations are needed to transcend lines of conflict. The key to sustainable peace in the long term is the integration of civil society across lines of conflict, requiring, over time, a different configuration of civil society, social structures, and political institutions: identity must cross-cut lines of conflict, leading to bargaining based on class concerns rather than those that revolve around the negotiation of competing identity.7 Belloni concludes that ‘‘An effective civil society development strategy capable of overcoming the dilemmas of 7

South Africa emerges as a case in which civil society did manage to congeal across lines of historical conflict, greatly facilitating progress in the war-to-democracy transition; see Stremlau (1998).

248

Timothy D. Sisk

democratization and peacebuilding requires . . . the ability to identify and support inclusive, bridging, and pro-democracy organizations.’’

Systemic dilemmas Systemic dilemmas relate to the very nature of contemporary internal, civil wars that cause spillovers and contagions for the entire world: invariably, these conflicts gain international attention and intervention by external forces, from neighboring states that feel the direct and tangible effects to globally powerful states that see ‘‘soft’’ threats emanating from weak state environments. The international community is compelled to intervene, wherever possible, to manage the international effects of internal civil wars (Sisk 2001). When external parties engage, as in a peace negotiation, they tacitly or explicitly are involved in recognition of parties – or exclusion of others – as recognized, legitimate parties to the conflict. When international peacebuilders include a group at the negotiating table, they implicitly provide legitimacy for these factions (Rothchild 1997). Likewise, in including rebels at the table as legitimate actors for bargaining, they also endorse the difficult transformations that such factions are intended to make as political entities. Sometimes the international community makes grave mistakes: as So¨derberg Kovacs observes in her chapter, recognition of the notoriously brutal RUF in Sierra Leone’s 1999 Lome´ peace pact had the effect of legitimating the party – and providing them with considerable power in a transitional pact – when in elections the former bandits garnered only 1.7 percent of the popular vote. Virginia Page Fortna’s contribution demonstrates that international peace operations do provide the essential credible commitment sorely needed in the volatile, post-war context. She writes that ‘‘Because peacekeeping is clearly and unambiguously good for peace, countries emerging from civil war are better off when peacekeepers deploy than when they are left to their own devices.’’ But she also notes that in some instances peacekeeping operations may actively inhibit some elements of democratization. The finding is that more than transitional, shortterm peace operations are required: to reduce violence for example, it is equally salutary to deploy large international electoral observation missions as a way to redress the uncertainty dilemmas that plague domestic protagonists as they contemplate loss, or envisage aggrandizement, in electoral processes (Bjornlund 2004).

Peacebuilding as democratization

249

Temporal dilemmas The research in this volume also points to one of the most engaging temporal dilemmas that occur in war-to-democracy transitions. Deals needed to achieve war termination and bring an end to violence may require compromises that constrain democratization later. Jarstad’s analysis of power sharing reveals that both protagonists and international mediators face powerful temporal dilemmas when negotiating peace agreements: she observes that the legitimacy of power sharing can be questioned when such arrangements are imposed by third parties. This suggests that, over time, they are prone to collapse unless a way is found to revisit the terms of settlement gradually without inducing some of the uncertainty dilemmas that have led to such pacts in the first place. Pacts that guarantee representation for warring parties run against the subsequent need in flourishing democracy to have uncertainty about who wins and loses in electoral competition. Because violence often accompanies the transition from war to peace into the post-settlement period, there are calls for the bloodletting to end before democratization can proceed (especially, elections). Security must come first. At the same time, there is often the reality that the violence is instrumental to the uncertainty of the moment, and that violence will only subside once the turbulence of the transition has passed. Thus, the sequencing puzzle – the planning and timing of peacebuilding (e.g., demobilization) and democratization (e.g., elections) is critical. As Ho¨glund concludes in her essay, ‘‘Political violence is frequently a response to too little democracy.’’ Perhaps the most appreciated, and yet continually most poignant, finding of this research is that often external agendas – often at the level of the Security Council – push electoral timetables even when the implication may be deleterious to the long-term needs of democratization. Reilly notes in his contribution to this volume, that ‘‘Hasty or rushed ‘instant elections’ have become common for several reasons.’’ ‘‘Democracy, as the mantra goes, is a long-term process, but the domestic political pressures that weigh on the Western states that usually fund and implement peacekeeping missions are almost all short-term . . . such ‘premature elections’ can also create multiple, ongoing problems for the development of peacetime politics in deeply divided societies even years after the war has ended.’’

250

Timothy D. Sisk

Recommendations for peacebuilders-as-democratizers The principal finding of this book is that, indeed, the goals of democratization and the demands of conflict management often work at cross-purposes: dilemmas of democratization and peacebuilding are ubiquitous in war-torn societies. Dilemmas of democratization arise when political violence continues into the post-war era and indeed escalates as elections approach; when rebels face the need to transform their identities from military organizations to political parties; when imperatives of justice and an end to impunity for human rights abuses collide with peacemaking and conciliation; when power sharing is accepted as an interim and, sometimes, permanent solution to contested governance but impedes accountability and popular participation; and in strategies of international democracy building, with trade-offs of short-term needs for transitional success and long-term demands of state building.

Designing war-to-democracy transitions The findings have policy implications in two important areas. First is in the analysis or context-specific assessment of war-to-democracy transitions, where there is the need for theoretically informed understanding of a given particular context. Research on dilemmas of democratization in relation to peacebuilding can help those evaluating a particular context to identify the right variables, to ask the right questions, and to take findings from comparative research into new settings. Here, there is a need to fuse two common methodologies of analysis employed by the international community to determine context appropriateness: democracy assessments and conflict vulnerability assessments. The former seeks to identify strengths and weaknesses in political processes, and the latter seeks to identify vulnerability to violence.8 8

For such a comprehensive conflict analysis model, see the World Bank’s Conflict Analysis Framework (available at lnweb18.worldbank.org/ESSD/sdvext.nsf/ 67ByDocName/ConflictAnalysis). Similarly, the World Bank’s governance indicators are an accepted set of measures for evaluating democratic practice: www.worldbank.org/wbi/governance/govdata/. International IDEA’s State of Democracy project offers a comprehensive assessment guide, and bilateral donors such as US Agency for International Development also have developed off-the-shelf democracy and governance assessment tools.

Peacebuilding as democratization

251

Assessment approaches can help address some of the most vexing questions that the war-to-democracy process raises in specific contexts. For example, the question of whether a faction or party can be excluded or included involves a decision made – often by international mediators such as a Special Representative of the Secretary-General – that a party is redeemable as a partner for peace. Sometimes, as in Uganda and the rebel Lord’s Resistance Army (LRA), the decision is made that some factions simply cannot be included either because of what they have done (in this case, gross violations of human rights, especially against children) or what they believe (espousing an extremist and indeed bizarre evangelical ideology in this instance). Even there, despite an indictment against LRA leaders by the International Criminal Court, the domestic imperatives for peace have led to a negotiation process between the government and the rebel forces. Context-based assessments are also critical for determining how a transition could be designed to more effectively minimize trade-offs between democracy and peace. The example of sequencing helps illustrate how assessments can improve policy strategies. A constant concern of researchers and policymakers alike is how to evaluate a strategy of sequenced change in outlining war-to-democracy transitions. The sequence of events – interim agreements and arrangements, local and national elections, constitution-making processes, referenda, and the like – is critical to resolving some of the challenges that peacebuildersas-democratizers face. For instance, in some situations it may make sense to have local elections first, as in Kosovo, and then follow with national-level polls to reflect a gradual, bottom-up approach to democratization that does not beg the question of national-level political authority but does provide some sense of legitimacy for governance. In other situations, however, the need is for consolidation of nationallevel authority such that relatively early national elections – as in Afghanistan – can begin to create an authoritative state to deal with local power centers that are injurious to democratization aims.9 Likewise, there is also good analysis available on how to sequence concerns regarding transitional justice on the route to peace.10 The variety of sequences addressed in this book – Northern Ireland, Bosnia, 9

10

For evaluation of local-level processes in United Nations peacebuilding, see Risley and Sisk (2005). For a thorough evaluation, see Mani (2002).

252

Timothy D. Sisk

South Africa, Sri Lanka, Kosovo, and others reinforces the principal finding of sequencing: there is no single or ideal sequence for a warto-democracy transition, and in this domain the primacy of analysis and decision making rests on the uniqueness of context.

Managing political violence The challenge of managing political violence is acute precisely because democratization processes can stimulate new bloody struggles for power: elections lead to mobilization, media openness can lead to the dissemination of hate speech, security sector reform is often inadequate, and rebels often are insufficiently transformed in their metamorphosis to becoming a political party. Democratization can, however, resolve problems of violence through the inclusion of the formerly excluded into political power, suggesting that with sufficient commitment, time, and resources the problem of violence can be managed so that peace can be self-sustaining. The challenge for external peacebuilders is to try to turn degenerative cycles of violence into generative cycles of trust, tolerance, and a willingness to play the game of democracy non-violently. Critical to this aim is the creation of credible commitment to peace agreements, primarily by international peacekeeping forces but also through the well-considered, long-term engagement of the international community. Likewise, when parties in such situations do use violence to pursue politics by non-democratic means, the international community must be willing and able to respond quickly and effectively, both in terms of military responses and in the willingness to punish (through delegitimation, sanctions, or other ‘‘negative inducements’’) violators of agreements. As Fortna demonstrates, too often peacekeeping forces are an insufficiently powerful tool to dissuade parties from turning to violence or to address spoiling when it occurs. This finding suggests that successful peace operations need two important preconditions: first is the capacity to act, and to act forcefully and convincingly, to provide a modicum of security on the ground. The second is to be deployed for much longer periods of time, and not to have six-month mandates (a common time frame for Security Council authorization of missions) for a multiyear job. Unless the international community is seriously committed to peace operations for the long term – and the creation of the Peace

Peacebuilding as democratization

253

Support Fund in 2006 for this purpose is a hopeful first step – efforts to manage transitional violence in peace operations will likely fall short, jeopardizing peace-through-democratization aims. Other proposed reforms, such as allowing the World Bank to fund peace support operations or to support programs such as police force training and reform (with a Security Council mandate and coordination with the UN force on the ground) are promising and deserve serious consideration. Another critical challenge in this domain is the question of anticipating and predicting electoral violence. The research here points to key indicators for understanding when a situation is predisposed toward electoral violence occurring, and the measures can be put in place in a preventive and post-electoral dispute resolution process to manage the likelihood of violence. An especially robust set of indicators are the residual coercive power of armed factions and the capacities of agencies of restraint – civilian authorities, police and military forces, international peacekeepers – to limit these capacities. Another key indicator to analyze is the possible outcome dimensions of electoral processes themselves: if significant factions are likely to lose the power they already have, or formerly excluded factions have no chance of meaningful representation or influence, the likelihood of ‘‘spoiler violence’’ is high.

Designing institutions: beyond ‘‘once-and-for-all’’ settlements The second implication is for those who are actively engaged in peacebuilding-through-democratization. A clearer understanding of the dilemmas that arise for the parties with which they may be working, and for themselves, can lead to more informed decision making and, over time, more successful outcomes in war-to-democracy processes. Peace agreements may contain features of a social contract, but it is not a self-enforcing one: the key to the long-term success of the war-todemocracy transition is the perpetuation of the bargaining process that led to the settlement long into the democratic future.11 Such a future, 11

This is not a new finding. For an earlier analysis on the importance of the proliferation of bargaining institutions for the management of social conflict, see Rothchild (1973).

254

Timothy D. Sisk

however, is not a quick one. Those who seek to promote war-todemocracy transitions should expect that the outcome of their efforts will be inherently partial, superficial, and in some ways deeply disappointing as the dilemmas of democratization present themselves forcefully.12 At least one solution to this problem is to build into peace settlements a longer-term transition for war-torn societies, one that envisages not just a single process of negotiation ending in a set of once-and-for-all institutional arrangements. Instead, clever inclusion of future negotiating processes, of longer-term moments of decision, can help address short-term needs for certainty on final status or on the immediate question of territorial sovereignty with longer-term needs for flexibility. At least two practical measures can accomplish this aim. One is to feature in the course of peace settlements ‘‘sunset’’ (or expiry) clauses that restrict some of the more democracy-constraining elements of peace agreements such as power sharing. Sunset clauses may sound like a good way to resolve the temporal dilemma, but often parties in conflict – particularly relative weaker parties politically (but not necessarily militarily) – demand a permanent share of power. Likewise, some parties such as governments are unwilling to agree to ‘‘down the road’’ provisions such as referendums on independence (even though the January 2005 peace agreement for Sudan included a provision for a referendum on independence for the South six years following). The second is to encourage national dialogue processes on democracy that can allow for supplementary consensus building to occur outside of formal institutions; such dialogues have the benefit of creating consensus first on possible institutional or procedural reforms, following which implementation of reforms can be less controversial. Examples of such dialogues are found both at the national level (as in South Africa’s highly participatory constitution-making process) and at the local level, for example those involving mayors in Colombia. Opening up political space for broader participation has been useful in a number of transitional arrangements, for example in the appointment of interim governments (Afghanistan, Lloya Jirga, and in Liberia, 12

Indeed, some scholars have argued that well beyond the initial peace agreement there is the need for ‘‘post-settlement settlements’’ that review the terms of peace long after the wounds of war have begun to heal and societies can withstand anew a period of political uncertainty that reopens issues over which the war was initially fought (du Toit 2003).

Peacebuilding as democratization

255

Benin, Togo, and other West African national conferences such as in Congo–Brazzaville and in Niger), in the drafting of constitutions (Eritrea, Rwanda, Afghanistan, and more recently Nepal), and in various elements of implementation of peace agreements as in Guatemala (Papagianni 2006: 23–27).

Engaging civil society One of the problems of international bilateral assistance in war-torn societies has been that of ‘‘stove-piping’’ in which various elements of donor agencies, or across different types of donors, agencies, and elements of a coalition that intervenes, narrowly focus on a single aspect of engagement of civil society without a more holistic picture of how such assistance can be better provided to support peacebuilding efforts more coherently. Assistance for health, education, or human rights entities often occurs without consideration of the implications of such assistance for conflict dynamics or for democratization objectives. Conflict mitigation needs to be more fully mainstreamed into all aspects of civil society promotion and governance processes such that a more consistent, coherent approach to addressing the causes of violence is employed. Likewise, there needs to be greater appreciation of what civil society can, and cannot, accomplish and more careful planning for how assistance to non-governmental organizations, trade unions, business associations, and religious entities can reinforce peacebuilding objectives. Belloni points out in his chapter that not all civil society groups are the same: some groups are virtuous and contribute to peace, but some civil society groups can in fact frustrate durable peace. Strategically, the promotion of civil society cannot occur in a platitudinous fashion that sees all civil society as an inherent good for peace and democratization. Quite the contrary, there needs to be a sharp strategy of differentiation in civil society promotion by which international donors are quite discriminating in identifying three types of non-state actors to support: those that cross-cut identity lines or fissures of conflict (as highlighted above), those that are moderate but reflecting primarily one perspective or protagonist social group, and those that are more extreme but which, through coaxing and inclusion, can become moderate (such as ex-prisoners, as was the case in Northern Ireland). Likewise, effective international action requires identifying and working diligently

256

Timothy D. Sisk

against those civil society groups that are deemed not constructive to peacebuilding aims, either because of their irredeemably extreme nature and positions or because they have other interests or activities (such as engagement in crime, resource exploitation, or activities such as human trafficking) that work against progress toward peace or democracy.

Rethinking international stewardship The present approach to international stewardship of war-to-democracy transitions aimed at bringing about lasting peace through democratization in societies shattered by war needs to be fundamentally reevaluated. As noted in the introduction to this chapter, the present approach has evolved in recent years but the fundamental formula remains the fusion of peacekeeping, peacebuilding, and democratization functions in a single, complex, multidimensional peace support operation. This approach does make sense in the immediate response of the international community to war-torn states that have managed to clinch a cease-fire. Indeed, where such negotiation-backstopping deployments of peacekeepers are not forthcoming for whatever reason, as in Azerbaijan (Karabagh) or Sri Lanka (2002–2006), moving beyond cease-fire to comprehensive settlement appears to be stifled, not assisted. Tough measures are needed to keep the peace and prevent violence, development and economic restructuring require early intervention, civil society (generally) needs to be included, and democratization – to include electoral processes and constitution making – cannot be indefinitely delayed. As the imbroglio in Iraq would reaffirm, success in this realm appears to require the United Nations or at least its imprimatur in the form of a Security Council mandate that enjoys a broad consensus among leading states; as Fortna shows, the actual composition of the force is less important that its response capacities, and its legitimacy. However, despite years of experience and no dearth of research, the current approach toward peacebuilding remains myopic; the causes of such myopia are found in the limits of resources (personnel, funds, and political will by leading states) for risky, longterm engagements; this is especially true when new emergencies clamor for attention and require new or diverted resources. Democratization is the long-term solution for war-torn states, but it cannot be

Peacebuilding as democratization

257

accomplished with a one-off, peacekeeping mission that ‘‘exits’’ when the political will for the mission wanes (usually, as violence begins to subside and transitional elections are over). Events in Angola, DRC, Haiti, Liberia, and Timor Leste (formerly East Timor), which have required repeated UN-mandated military interventions, reinforce this now well-appreciated observation. The fallacy of the regrettably still-current approach is the perception that a deployed, bounded mission lasting at least a year but not more than a decade can facilitate the process of war to democracy.13 The present formula: intervene, establish order through blue-helmet force, fund some civil society, facilitate a new constitution, hold an election . . . and then leave – may suffice for short-term peace but does not suffice for democratization. To redress the fallacy of myopia, and the need to sustain support for the long transition from war to democracy, international stewardship of peacebuilding will need to evolve institutionally and in practice much more significantly than it has so far. The war-to-democracy transition demands a series of successive missions to address the longer-term phases of consolidating peace, of building democratic institutions and processes, of reacting to subsequent political crises and power-grabs, of addressing the socioeconomic conditions that give rise to violence, and, ultimately, the creation of a self-sustaining social contract. Successive missions will require a seamless set of phases of international involvement that may begin with heavy-handed deployments but that changes over time into a deeper, generationslong commitment to democratization. An example of where this has been somewhat effective is the continued commitment by the international community to Mozambique, which benefited from an initial peace operation but has also been subsequently subject to extensive monitoring, electoral assistance, donor support, and other governanceimproving engagement by the international community. Such extended commitments to war-torn societies needs to be the norm, not the exception.

13

The latter is quite unusual and found only in the post-Cold War era in the Balkans where European resources and NATO troop commitments account for the longevity of commitment. Even there, initially, there was an ill-considered effort to impose a one-year constraint on NATO deployment.

258

Timothy D. Sisk

Commensurate with such a longer-term view is the need for constant monitoring and a willingness to intervene again early when progress toward peace or democratization is threatened. As Wallensteen notes, many of the crises in the cases he has evaluated – Uganda, Lebanon, and Liberia – have come a decade or more after the peace agreement was reached. Likewise, Zimbabwe is an example of a situation in which the legacies of war and social conflict threatened peace and democratization more than a decade after the war was ostensibly settled, with devastating social, economic, and regional effects of the slide into authoritarianism and state failure in that country. International reaction in this case has been too partial, too restricted, and too late. This need for a rethinking of the present approach will require the reinvigoration of instruments that exist but are insufficiently used – such as extensive United Nations civilian observer missions and of long-term UN civilian police deployments to be in place once military security seems to be achieved. Indeed, there is the need to consider the deployment of civilian missions who can readily observe when backsliding is occurring; increasingly, it is regional organizations (in Europe, Africa, and the Americas, especially) that are effectively playing this role. A new evolution of peacebuilding will also require new instruments that have been considered but not realized, especially the creation of a military rapid deployment force (such as an international gendarme or rapid reaction force) to respond quickly when new crises arise,14 new funding mechanisms for providing the resources required for long-term engagements contexts far away from prosperous Western states, and the will to react earlier when there are indicators – such as political crises or the removal of independent judicial officials – that peace or democratization is threatened anew. Measures such as these are ideal but, as the challenges of responding early to the Zimbabwe crisis or empowering Afghanistan’s nascent democratic institutions, for example, show, are hampered by the stark realities that international intervention remains limited by international consensus and domestic contexts. As a result, the outcomes of 14

This recommendation was included as a key finding of the Carnegie Commission on Preventing Deadly Conflict but has not been embraced by the permanent members of the United Nations Security Council. See the Commission’s Final Report for the rationale and specifics of this recommendation (Carnegie Commission on Preventing Deadly Conflict 1998).

Peacebuilding as democratization

259

war-to-democracy transitions are likely to remain, for the foreseeable future, inherently limited despite the best efforts of the international community to confront and address innovatively the dilemmas such transitions pose. More optimistically, however, some of the ways in which the dilemmas are addressed are amenable to innovation and implementation by the international community – as the deployment of an all-female UN peacekeeping unit in Liberia in 2006 following the election there of Africa’s first female president, Ellen Johnson Sirleaf. This deployment pairs the need for a longer-term security commitment with an implicit appreciation for human rights and conflict transformation. Likewise, the deployment of a significant security force and large-scale election-assistance mission in the Democratic Republic of Congo in July 2006 is also indicative of learning and adaptation in facilitating war-to-democracy transitions, even as post-election violence has created new insecurities in 2007. Small and incremental steps such as these do not vitiate altogether the dilemmas in war-torn societies that the international community and internal protagonists experience; however, even such small steps do indicate that innovation, a stronger will and more forceful deployments, and an abiding commitment to a longer time horizon can go some way toward minimizing the deleterious impact that dilemmas of democratization impose on the pursuit of peace.

References

Abbink, Jon, and Gerti Hesseling, eds. 1999. Election Observation and Democratization in Africa. Basingstoke: Macmillan Press. Andersson, Andreas. 2000. ‘‘Democracies and UN Peacekeeping Operations, 1990–1996.’’ International Peacekeeping 7 (2): 1–22. Arato, Andrew. 1981. ‘‘Civil Society vs. the State: Poland 1980–81.’’ Telos 47: 23–47. Archer, Dane, and Rosemary Gartner. 1976. ‘‘Violent Acts and Violent Times: A Comparative Approach to Postwar Homicide Rates.’’ American Sociological Review 41 (6): 937–963. Arzt, Donna E. 2006. ‘‘Views on the Ground: The Local Perception of International Criminal Tribunals in the Former Yugoslavia and Sierra Leone.’’ Annals of the American Academy of Political and Social Sciences 603 (1): 226–239. Atlas, Pierre M., and Roy Licklider. 1999. ‘‘Conflict Among Former Allies after Civil War Settlement: Sudan, Zimbabwe, Chad and Lebanon.’’ Journal of Peace Research 36 (1): 35–54. Baker, Pauline. 2001. ‘‘Conflict Resolution versus Democratic Governance: Divergent Paths to Peace?’’ In Turbulent Peace: The Challenges of Managing International Conflict, ed. Fen O. Hampson, Chester A. Crocker and Pamela Aall. Washington, DC: United States Institute of Peace Press (753–764). Baldwin, Clive. 2006. Minority Rights in Kosovo under International Rule. London: Minority Rights Group International. Ballentine, Karen, and Jake Sherman, eds. 2003. The Political Economy of Armed Conflict: Beyond Greed and Grievance. Boulder, Colo.: Lynne Rienner Publishers. Bangura, Yusuf. 2000. ‘‘Strategic Policy Failure and Governance in Sierra Leone.’’ Journal of Modern African Studies 38 (4): 551–577. Barnes, Catherine. 2002. Owning the Process: Mechanisms for Political Participation of the Public in Peacemaking. London: Conciliation Resources. Barnett, Michael. 2006. ‘‘Building a Republican Peace.’’ International Security 30 (4) (Spring): 87–112.

260

References

261

Barro, Robert J. 1999. ‘‘Determinants of Democracy.’’ Journal of Political Economy 107: 158–183. Bartolini, Stefano, and Peter Mair. 2001. ‘‘Challenges to Contemporary Political Parties.’’ In Political Parties and Democracy, ed. Larry Diamond and Richard Gunther. Baltimore: Johns Hopkins University Press (327–343). Belloni, Roberto. 2007. ‘‘Shades of Orange and Green: Civil Society and the Peace Process in Northern Ireland.’’ Paper presented at the 48th Annual Convention of the International Studies Association, Chicago, February 27. 2004. ‘‘Peacebuilding and Consociational Electoral Engineering in Bosnia and Herzegovina.’’ International Peacekeeping 11 (2): 334–353. 2001. ‘‘Civil Society and Peacebuilding in Bosnia and Herzegovina.’’ Journal of Peace Research 38 (2): 163–180. Belloni, Roberto, and Shelley Deane. 2005. ‘‘From Belfast to Bosnia: Piecemeal Peacemaking and the Role of Institutional Learning.’’ Civil Wars 7 (3): 219–243. Bentley, Kristina A., and Roger Southall. 2005. An African Peace Process: Mandela, South Africa and Burundi. Cape Town: Human Sciences Research Council. Berdal, Mats, and David M. Malone, eds. 2000. Greed and Grievance: Economic Agendas in Civil Wars. Boulder: Lynne Rienner Publishers. Berman, Sheri. 1997. ‘‘Civil Society and the Collapse of the Weimar Republic.’’ World Politics 49 (3): 401–429. Bertram, Eva. 1995. ‘‘Reinventing Governments: The Promise and Perils of United Nations Peace Building.’’ Journal of Conflict Resolution 39 (5): 387–418. Bieber, Florian. 2003. ‘‘The Other Civil Society in Serbia: NonGovernmental Nationalism – The Case of the Serbian Resistance Movement.’’ In Uncivil Society? Contentious Politics in PostCommunist Europe, ed. Petr Kopecky and Cass Muddes. London and New York: Routledge (19–36). Bjornlund, Eric. 2004. Beyond Free and Fair: Monitoring Elections and Building Democracy. Baltimore, Md.: Johns Hopkins University Press. Bojicic-Dzelilovic, Vesna. 2006. ‘‘War Veterans’ Associations in Bosnia and Herzegovina.’’ In Challenges to Peacebuilding: Managing Spoilers During Conflict Resolution, ed. Edward Newman and Oliver Richmond. Tokyo, New York, and Paris: United Nations University Press (200–218). Brass, Paul R. 1991. Ethnicity and Nationalism: Theory and Comparison. New Delhi, Newsbury Park and London: Sage Publications.

262

References

Brown, Frederick Z. 1998. ‘‘Cambodia’s Rocky Venture in Democracy.’’ In Postconflict Elections, Democratization & International Assistance, ed. Krishna Kumar. Boulder, Colo. and London: Lynne Rienner Publishers (87–110). Brown, Michael E., Sean M. Lynn-Jones, and Steven E. Miller, eds. 1996. Debating the Democratic Peace. Cambridge, Mass.: MIT Press. Bueno de Mesquita, Bruce and George W. Downs. 2006. ‘‘Intervention and Democracy.’’ International Organization 60 (3): 627–650. Burkhart, Ross E., and Michael S. Lewis-Beck. 1994. ‘‘Comparative Democracy: The Economic Development Thesis.’’ American Political Science Review 88: 903–910. Burnell, Peter. 2005. ‘‘Political Strategies of External Support for Democratization.’’ Foreign Policy Analysis 1: 361–384. Call, Charles T. 2003. ‘‘Democratisation, War and State-Building: Constructing the Rule of Law in El Salvador.’’ Journal of Latin American Studies 35: 827–862. Call, Charles T., and Susan E. Cook. 2003. ‘‘On Democratization and Peacebuilding.’’ Global Governance 9: 233–246. Call, Charles T., and William Stanley. 2003. ‘‘Military and Police Reform after Civil War.’’ In Contemporary Peacemaking: Conflict Violence and Peace Processes, ed. J. Darby and R. Mac Ginty. Basingstoke: Palgrave Macmillan (212–223). Carnegie Commission on Preventing Deadly Conflict. 1998. Preventing Deadly Conflict: Final Report of the Carnegie Commission on Preventing Deadly Conflict. Washington, DC: Carnegie Commission on Preventing Deadly Conflict. Available at wwics.si.edu/subsites/ ccpdc/pubs/rept97/finfr.htm. Carothers, Thomas. 2004. Critical Mission: Essays on Democracy Promotion. Washington, DC: Carnegie Endowment for International Peace. 1999. Aiding Democracy Abroad: The Learning Curve. Washington, DC: Carnegie Endowment for International Peace. Cawthra, Gavin, and Robin Luckham. 2003. Governing Insecurity: Democratic Control of Military and Security Establishments in Transitional Democracies. London: Zed Books. Cerdas Cruz, Rodolfo. 1998. ‘‘Political Parties and Party Systems.’’ In Guatemala after the Peace Accords, ed. Rachel Sieder. London: Institute of Latin American Studies (15–54). Chandler, David. 2004. ‘‘The Responsibility to Protect? Imposing the ‘Liberal Peace’.’’ International Peacekeeping 11 (1): 59–81. 1999. Bosnia. Faking Democracy after Dayton. London: Pluto Press. Chandra, Kanchan. 2004. Why Ethnic Parties Succeed: Patronage and Ethnic Headcounts in India. Cambridge: Cambridge University Press.

References

263

Chesterman, Simon. 2004. You, The People: The United Nations, Transition Administration, and State-Building. Oxford: Oxford University Press. Chesterman, Simon, Michael Ignatieff, and Ramesh Thakur. 2004. Making States Work: From State Failure to State Building. Tokyo: United Nations University Press. Chesterman, Simon, Tom Farer, and Timothy Sisk. 2003. ‘‘Competing Claims: Self-Determination, Security, and the United Nations.’’ A report of the project on ‘‘Self-Determination, Security, and the United Nations,’’ Graduate School of International Studies, University of Denver, and the International Peace Academy. Available at www.ipacademy.org/ Publications/Reports/Research/PublRepoReseCompClaims_body.htm. Clapham, Christopher, and John A. Wiseman. 1995. ‘‘Conclusions: Assessing the Prospects for the Consolidation of Democracy.’’ In Democracy and Political Change in Sub-Saharan Africa, ed. J. A. Wiseman. London: Routledge (220–232). Colletta, Nat J., Markus Kostner, and Ingo Wiederhofer. 2004. ‘‘Disarmament, Demobilization, and Reintegration: Lessons and Liabilities in Reconstruction.’’ In When States Fail: Causes and Consequences, ed. Robert Rotberg. Princeton: Princeton University Press (170–181). Colletta, Nat J. and Michelle L. Cullen. 2000. Violent Conflict and the Transformation of Social Capital: Lessons from Cambodia, Rwanda, Guatemala, and Somalia. Washington, DC: World Bank. Collier, Paul, V. L. Elliot, Ha˚vard Hegre, Anke Hoeffler, Marta ReynalQuerol, and Nicholas Sambanis, eds. 2003. Breaking the Conflict Trap: Civil War and Development Policy. Washington, DC: World Bank and Oxford University Press. Collin Marks, Susan. 2000. Watching the Wind: Conflict Resolution during South Africa’s Transition to Democracy. Washington, DC: United States Institute of Peace Press. Collins, Kathleen. 2006. The Logic of Clan Politics in Central Asia: The Impact on Regime Transformation. New York: Cambridge University Press. Commission on Global Governance. 1995. Our Global Neighbourhood. The Report of the Commission on Global Governance. Oxford: Oxford University Press. Cousens, Elizabeth M. 2001a. ‘‘Building Peace in Bosnia.’’ In Peacebuilding as Politics: Cultivating Peace in Fragile Societies, ed. Elizabeth M. Cousens and Chetan Kumar. Boulder, Colo. and London: Lynne Rienner Publishers (113–152). 2001b. ‘‘Introduction.’’ In Peacebuilding as Politics, ed. Elizabeth M. Cousens and Chetan Kumar. Boulder Colo. and London: Lynne Rienner Publishers (1–20).

264

References

Cowen, Michael, and Liisa Laakso. 1997. ‘‘An Overview of Election Studies in Africa.’’ Journal of Modern African Studies 35 (4): 717–744. Dahl, Robert A. 1971. Polyarchy: Participation and Opposition. New Haven: Yale University Press. Damm, Darlene. 2003. Managing Election Violence in Cambodia: New Strategies for the International Donor Community. Draft Report 1 December 2003. Johns Hopkins University: International Donor Community Cambodia Steering Committee. Available at www.saisjhu.edu/bwelsh/DarlenePolicyPaper.pdf. Darby, John. 2006. Violence and Reconstruction. South Bend, Ind.: University of Notre Dame Press. 2001. The Effect of Violence on Peace Processes. Washington DC: USIP Press. 1986. Intimidation and Control of Conflict in Northern Ireland. Dublin: Gill and Macmillan. Darby, John, and Roger Mac Ginty, eds. 2000. The Management of Peace Processes. Basingstoke: Macmillan Press. de Nevers, Rene´e. 1993. ‘‘Democratization and Ethnic Conflict.’’ In Ethnic Conflict and International Security, ed. Michael Brown. Princeton: Princeton University Press (61–78). DeVotta, Neil. 2004. Blowback: Linguistic Nationalism, Institutional Decay, and Ethnic Conflict in Sri Lanka. Stanford: Stanford University Press. de Zeeuw, Jeroen. 2005. ‘‘Projects Do Not Create Institutions: The Record of Democracy Assistance in Post-Conflict Societies.’’ Democratization 12 (4): 481–504. du Toit, Pierre. 2003. ‘‘Why Post-settlement Settlements?’’ Journal of Democracy 14 (3): 104–118. Diamond, Larry. 2005. Squandered Victory: The American Occupation and the Bungled Effort to Bring Democracy to Iraq. New York: Time Books. 1999a. ‘‘Toward Democratic Consolidation.’’ In The Global Resurgence of Democracy, ed. Larry Diamond and Marc F. Platter. Baltimore and London: Johns Hopkins University Press (227–240). 1999b. Developing Democracy: Towards Consolidation. Baltimore: Johns Hopkins University Press. 1997a. ‘‘Promoting Democracy in the 1990s: Actors, Instruments, and Issues.’’ In Democracy’s Victory and Crisis, ed. Axel Hadenius. Cambridge: Cambridge University Press (311–370). 1997b. ‘‘Introduction: In Search of Consolidation.’’ In Consolidating the Third Wave Democracies: Themes and Perspectives, ed. Larry Diamond, Marc F. Plattner, Yun-han Chu, and Hung-mao Tien. Baltimore and London: Johns Hopkins University Press (xiii–xlvii).

References

265

Diamond, Larry, Juan J. Linz, and Seymour Martin Lipset, eds. 1988. Democracy in Developing Countries. Boulder, Colo.: Lynne Rienner Publishers. Diamond, Larry, Marc F. Plattner, Yun-han Chu, and Hung-mao Tien, eds. 1997. Consolidating the Third Wave Democracies: Themes and Perspectives. Baltimore and London: Johns Hopkins University Press. Dorussen, Han. 2001. ‘‘Mixing Carrots with Sticks: Evaluating the Effectiveness of Positive Incentives.’’ Journal of Peace Research 38 (2): 251–262. Doyle, Michael W., and Nicholas Sambanis. 2006. Making War and Building Peace: United Nations Peace Operations. Princeton: Princeton University Press. 2000. ‘‘International Peacebuilding: A Theoretical and Quantitative Analysis.’’ American Political Science Review 94 (4): 779–802. Druckman, Daniel. 1986. ‘‘Stages, Turning Points, and Crises.’’ Journal of Conflict Resolution 30: 327–360. Edwards, Michael. 2004. Civil Society. Cambridge: Polity Press. Eide, Espen Barth, Anja Therese Kaspersen, Randolph Kent, and Karen von Hippel. 2005. Report on Integrated Missions: Practical Perspectives and Recommendations. Independent Study for the Expanded UN ECHA Core Group. Elklit, Jørgen. 1999. ‘‘Electoral Institutional Change and Democratisation: You Can Lead a Horse to Water, but You Can’t Make it Drink.’’ Democratisation 6 (4): 28–51. Elklit, Jørgen, and Palle Svensson. 1997. ‘‘What Makes Elections Free and Fair?’’ Journal of Democracy 8 (3): 32–46. Eriksson, Mikael, and Peter Wallensteen. 2004. ‘‘Armed Conflict 1989–2003.’’ Journal of Peace Research 41 (5) (September): 625–636. Eubank, William Lee, and Leonard Weinberg. 1994. ‘‘Does Democracy Encourage Terrorism?’’ Terrorism and Political Violence 6 (4): 417–443. Farer, Tom J. 2004. ‘‘The Promotion of Democracy: International Law and Norms.’’ In The UN Role in Promoting Democracy. Between Ideals and Reality, ed. Edward Newman and Roland Rich. Tokyo: UN University Press (32–61). Fearon, James D., and David D. Laitin. 2004. ‘‘Neotrusteeship and the Problem of Weak States.’’ International Security 28 (4): 5–43. 2003. ‘‘Ethnicity, Insurgency, and Civil War.’’ American Political Science Review 97 (1): 75–90. Fischer, Jeff. 2002. Electoral Conflict and Violence: A Strategy for Study and Prevention. IFES White Paper. Fitzduff, Mari. 2004. Civil Society and Peacebuilding: The New Fifth Estate? European Center for Conflict Prevention. Available at www.gppac.net.

266

References

Foley, Michael W. 1996. ‘‘Laying the Groundwork: The Struggle for Civil Society in El Salvador.’’ Journal of Interamerican Studies and World Affairs 38 (1): 67–104. Fortna, Virginia Page. 2008. Does Peacekeeping Work? Shaping Belligerents’ Choices After Civil Wars. Princeton: Princeton University Press. 2005. ‘‘Where Have all the Victories Gone? War Outcomes in Historical Perspective.’’ Paper presented at International Studies Association, March, Honolulu, HI. 2004. ‘‘Does Peacekeeping Keep Peace? International Intervention and the Duration of Peace after Civil War.’’ International Studies Quarterly 48 (2): 269–292. Francis, David J. 2000. ‘‘Tortuous Path to Peace: The Lome´ Accord and Postwar Peacebuilding in Sierra Leone.’’ Security Dialogue 31 (3): 357–373. Franck, Thomas M. 1992. ‘‘The Emerging Right to Democratic Governance.’’ American Journal of International Law 86 (1): 46–91. Freedom House. 2006. Freedom in the World Comparative Rankings: 1975–2005. Available at www.freedomhouse.org. 2002–2005a. Country Report: Cambodia: Freedom House. Available at www.freedomhouse.org. 2002–2005b. Country Report: East Timor: Freedom House. Available at www.freedomhouse.org. Fukuyama, Francis. 1995. Trust: The Social Virtues and the Creation of Prosperity. New York: Free Press. Gamba, Virginia. 2003. ‘‘Managing Violence: Disarmament and Demobilization.’’ In Contemporary Peacemaking: Conflict, Violence and Peace Processes, ed. J. Darby and R. Mac Ginty. Basingstoke: Palgrave Macmillan (125–136). Gidron, Benjamin, Stanley N. Katz, and Yeheskel Hasenfel. 2002. Mobilizing for Peace: Conflict Resolution in Northern Ireland, Israel/ Palestine, and South Africa. Oxford: Oxford University Press. Gilligan, Michael, and Stephen John Stedman. 2003. ‘‘Where Do the Peacekeepers Go?’’ International Studies Review 5 (4): 37–54. Gleditsch, Nils Petter, and Ha˚vard Hegre. 1997. ‘‘Peace and Democracy: Three Levels of Analysis.’’ Journal of Conflict Resolution 41 (2): 283–310. Greif, Avner. 1998. ‘‘Self-Enforcing Political Systems and Economic Growth: Late Medieval Genoa.’’ In Analytic Narratives, ed. R. H. Bates, A. Greif, M. Levi, J.-L. Rosenthal, and B. Weingast. Princeton: Princeton University Press (23–63). Grugel, Jean. 2002. Democratization. A Critical Introduction. Houndmills: Palgrave Macmillan.

References

267

Guelke, Adrian. 2003. ‘‘Civil Society and the Northern Ireland Peace Process.’’ Voluntas: International Journal of Voluntary and Nonprofit Organizations 14 (1): 61–78. Gunther, Richard, and Larry Diamond. 2001. ‘‘Types and Functions of Parties.’’ In Political Parties and Democracy, ed. Richard Gunther and Larry Diamond. Baltimore and London: Johns Hopkins University Press (3–39). Gurr, Ted Robert. 2000a. Peoples Versus States: Minorities at Risk in the New Century. Washington, DC: United States Institute of Peace Press. 2000b. ‘‘Ethnic warfare on the wane.’’ Foreign Affairs 79 (3): 52–64. Hadenius, Axel, and Fredrik Uggla. 1996. ‘‘Making Civil Society Work, Promoting Democratic Development: What Can States and Donors Do?’’ World Development 24 (10): 1621–1639. Haggard, Stephan. 1997. ‘‘Democratic Institutions, Economic Policy, and Development.’’ In Institutions and Economic Development, ed. Christopher Clague. Baltimore and London: Johns Hopkins University Press (120–149). Hammer, Peter J., and Tara Urs. 2005. ‘‘The Elusive Face of Cambodian Justice.’’ In Bringing the Khmer Rouge to Justice: Prosecuting Mass Violence before the Cambodian Courts, ed. Jaya Ramji and Beth Van Schaack. Ontario: Edwin Mellen Press (13–58). Hampson, Dave. 2004. ‘‘It’s Time to Open Up:’’ Ten Years After the Genocide in Rwanda. Christian Aid, March. Available at www.discover.ac.uk/ socialsciences/cgi-bin/fullrecord.pl?handle=sosig1. Hampson, Fen Osler. 1996. Nurturing Peace: Why Peace Settlements Succeed or Fail. Washington, DC: United States Institute of Peace Press. Hann, Chris, and Elizabeth Dunn, eds. 1996. Civil Society: Challenging Western Models. London: Routledge. Haraszti, Miklos. 2004. The Role of the Media in the March 2004 Events in Kosovo. Vienna: OSCE. Harbom, Lotta, Stina Ho¨gbladh, and Peter Wallensteen. 2006. ‘‘Armed Conflicts and Peace Agreements.’’ Journal of Peace Research 43 (5): 617–631. Harbom, Lotta, and Peter Wallensteen. 2005. ‘‘Armed Conflict and Its International Dimensions, 1946–2004.’’ Journal of Peace Research 42 (5): 623–635. Harris, Peter, and Benjamin Reilly, eds. 1998. Democracy and Deep-Rooted Conflict: Options for Negotiators. Stockholm: International Institute for Democracy and Electoral Assistance. Hartlyn, Jonathan. 1998. The Politics of Coalition Rule in Colombia. Cambridge: Cambridge University Press.

268

References

Hartzell, Caroline A., and Matthew Hoddie. 2003. ‘‘Institutionalizing Peace: Power Sharing and Post-Civil War Conflict Management.’’ American Journal of Political Science 47 (2): 318–332. Hearn, Julie. 2000. ‘‘Aiding Democracy? Donors and Civil Society in South Africa.’’ Third World Quarterly 21 (5): 815–830. Hegre, Ha˚vard, Tanja Ellingsen, Scott Gates, and Nils Petter Gleditsch. 2001. ‘‘Towards a Democratic Civil Peace? Democracy, Political Change and Civil War, 1816–1992.’’ American Political Science Review 95 (1): 33–48. Heldt, Birger. 2007. ‘‘Peacekeeping Operations and Post-conflict Transitions to Democracy?’’ Paper presented at the 2007 Annual Conference by the Swedish Network of Peace, Conflict and Development Research: ‘‘The Democratization Project: Challenges and Opportunities,’’ Uppsala October 23–24, 2007. Hoddie, Matthew, and Caroline A. Hartzell. 2005. ‘‘Power Sharing in Peace Settlements: Initiating the Transition from Civil War.’’ In Sustainable Peace: Power and Democracy after Civil War, ed. Philip G. Roeder and Donald Rothchild. Ithaca and London: Cornell University Press (83–106). Hodges, Tony. 2001. Angola From Afro-Stalinism to Petro-Diamond Capitalism. Bloomington and Indianapolis: Indiana University Press. Ho¨glund, Kristine. 2008. Peace Negotiations in the Shadow of Violence. Leiden: Martinus Nijhoff Publishers. 2005. ‘‘Violence and the Peace Process in Sri Lanka.’’ Civil Wars 7 (2): 156–170. Ho¨glund, Kristine, and I. William Zartman. 2006. ‘‘Violence by the State: Official Spoilers and their Allies.’’ In Violence and Reconstruction, ed. J. Darby. South Bend, Ind.: University of Notre Dame Press (11–31). Hohe, Tanja. 2002. ‘‘Totem Polls: Indigenous Concepts and ‘Free and Fair’ Elections in East Timor.’’ International Peacekeeping 9 (4): 69–88. Holbrooke, Richard. 1998. To End a War. New York: Random House. Holm, Tor Tanke, and Espen Barth Eide. 2000. Peacebuilding and Police Reform. London: Frank Cass. Holohan, Anne. 2005. Networks of Democracy: Lessons from Kosovo for Afghanistan, Iraq, and Beyond. Stanford: Stanford University Press. Holsti, Kalevi J. 1996. The State, War, and the State of War. Cambridge: Cambridge University Press. Horowitz, Donald L. 2001. The Deadly Ethnic Riot. Berkeley: University of California Press. 1985. Ethnic Groups in Conflict. Berkeley, Los Angeles, and London: University of California Press.

References

269

Howard, Ross. 2005. ‘‘The Media’s Role in War and Peacebuilding.’’ In Postconflict Development: Meeting New Challenges, ed. G. Junne and W. Verkoren. Boulder, Colo. and London: Lynne Rienner Publishers (117–128). Human Rights Watch. 2004. Failure to Protect: Anti-Minority Violence in Kosovo, March 2004. New York: Human Rights Watch. Human Security Report 2005: War and Peace in the 21st Century. 2005. New York and Oxford: Oxford University Press. Huntington, Samuel P. 1991. The Third Wave. Democratization in the Late Twentieth Century. Norman, Okla.: University of Oklahoma Press. 1957. The Soldier and the State: The Theory and Politics of Civil-Military Relations. Cambridge, Mass.: Harvard University Press. ICG. 2007a. Nepal’s Constitutional Process. Asia Report No. 128. Kathmandu and Brussels: International Crisis Group. 2007b. After Mecca: Engaging Hamas. Middle East Report No. 62. Amman, Jerusalem, and Brussels: International Crisis Group. 2006a. Sudan’s Comprehensive Peace Agreement: The Long Road Ahead. Africa Report No. 106. Nairobi and Brussels: International Crisis Group. 2006b. Lebanon at a Tripwire. Middle East Briefing No. 20. Beirut and Brussels: International Crisis Group. 2005a. Elections in Burundi: A Radical Shake-up of the Political Landscape. Africa Briefing No. 31. Nairobi and Brussels: International Crisis Group. 2005b. Kosovo after Haradinaj. Europe Report No. 163. Pristina and Brussels: International Crisis Group. 2005c. Aceh: A New Chance for Peace. Asia Briefing No. 40. Jakarta and Brussels: International Crisis Group. 2005d. Liberia’s Elections: Necessary but Not Sufficient. Africa Report No. 107. Dakar and Brussels: International Crisis Group. 2004a. Collapse in Kosovo. Europe Report No. 155. Pristina, Belgrade, and Brussels: International Crisis Group. 2004b. Tajikistan’s Politics: Confrontation or Consolidation? Asia Briefing No. 33. Dushanbe and Brussels: International Crisis Group. 2004c. Elections in Burundi: The Peace Wager. Africa Briefing No. 20. Nairobi and Brussels: International Crisis Group. 2003. The Special Court for Sierra Leone: Promises and Pitfalls of a ‘‘New Model.’’ Africa Briefing No. 16. Freetown and Brussels: International Crisis Group. 2002. Sierra Leone after Elections: Politics as Usual? Africa Report No. 49. Freetown and Brussels: International Crisis Group. Inglehart, Ronald. 1997. Modernization and Postmodernization. Princeton: Princeton University Press.

270

References

Jarstad, Anna K. 2007. ‘‘To Share or to Divide? Negotiating the Future of Kosovo.’’ Civil Wars 9 (3): 227–242. 2001. ‘‘Changing the Game: Consociational Theory and Ethnic Quotas in Cyprus and New Zealand.’’ Doctoral Dissertation, Department of Peace and Conflict Research, Uppsala University. Jarstad, Anna K., and Ralph Sundberg. 2007. ‘‘Peace by Pact: Data on the Implementation of Peace Agreements.’’ In Globalization and Challenges ¨ jendal. to Building Peace, ed. Ashok Swain, Ramses Amer, and Joakim O London: Anthem Press (73–89). Jeong, Ho-Won. 2005. Peace building in Postconflict Societies. Strategy and Process. Boulder, Colo.: Lynne Rienner Publishers. Jung, Dietrich, ed. 2003. Shadow Globalization, Ethnic Conflicts and New Wars: A Political Economy of Intra-State War. London and New York: Routledge. Kandeh, Jimmy D. 2003. ‘‘Sierra Leone’s Post-Conflict Elections of 2002.’’ Journal of Modern African Studies 41 (2): 189–216. Kaldor, Mary. 2003. Global Civil Society: An Answer to War. Cambridge: Polity Press. Kasfir, Nelson. 2004. ‘‘Civil Society, the State and Democracy in Africa.’’ In Civil Society in Democratization, ed. Peter Burnell and Peter Calvert. London and Portland, reg.: Frank Cass (117–142). Katana, Gordana. 1999. ‘‘NGOs in Republika Srpska: Bashful Support of the Regime.’’ AIM Banja Luka, September 20. Available at www. aimpress.ch/dyn/trae/archive/data/199909/90926-002-trae-sar.htm. Kaufmann, Chaim. 1997. ‘‘Possible and Impossible Solutions to Ethnic Civil Wars.’’ In Nationalism and Ethnic Conflict, ed. Michael Brown et al. (265–304). Keane, John. 2004. Violence and Democracy. Cambridge: Cambridge University Press. 1998. Civil Society: Old Images, New Perspectives. Cambridge: Polity Press. Keck, Margaret, and Kathryn Sikkink. 1998. Activists Beyond Borders: Advocacy Networks in International Politics. Ithaca: Cornell University Press. Knaus, Gerald, and Felix Martin. 2003. ‘‘Travails of the European Raj.’’ Journal of Democracy 14 (3): 60–74. Kohn, Richard H. 1997. ‘‘How Democracies Control the Military.’’ Journal of Democracy 8 (4): 140–153. Kritz, Neil J. 2001. ‘‘The Rule of Law in the Postconflict Phase: Building a Stable Peace.’’ In Turbulent Peace. The Challenges of Managing International Conflict, ed. Fen O. Hampson, Chester A. Crocker, and Pamela Aall. Washington, DC: United States Institute of Peace Press (801–820).

References

271

Krznaric, Roman. 1999. ‘‘Civil and Uncivil Actors in the Guatemalan Peace Process.’’ Bulletin of Latin American Research 18 (1): 1–16. Kumar, Krishna. 1998a. ‘‘Postconflict Elections and International Assistance.’’ In Postconflict Elections, Democratization, and International Assistance, ed. Krishna Kumar. Boulder, Colo.: Lynne Rienner Publishers (5–14). 1998b. ‘‘After the Elections: Consequences for Democratization.’’ In Postconflict Elections, Democratization, and International Assistance, ed. Krishna Kumar. Boulder, Colo.: Lynne Rienner Publishers (215–227). ed. 1998c. Postconflict Elections, Democratization, and International Assistance. Boulder, Colo.: Lynne Rienner Publishers. ed. 1997. Rebuilding Societies after Civil War: Critical Roles for International Assistance. Boulder, Colo. and London: Lynne Rienner Publishers. Kydd, Andrew, and Barbara F. Walter. 2002. ‘‘Sabotaging the Peace: The Politics of Extremist Violence.’’ International Organization 56 (2): 263–296. Lacy, Brett. 2004. ‘‘Building Accountability, Legitimacy, and Peace: Refugees, Internally Displaced Persons, and the Right to Political Participation.’’ IFES Occasional Paper, Washington, DC. Lanegran, Kimberly. 2001. ‘‘South Africa’s 1999 Election: Consolidating a Dominant Party System.’’ Africa Today 48 (2): 81–102. Lederach, John Paul. 1997. Building Peace. Sustainable Reconciliation in Divided Societies. Washington, DC: United States Institute of Peace. Levi, Margaret. 1996. Consent, Dissent, and Patriotism. Cambridge: Cambridge University Press. Licklider, Roy. 2005. ‘‘Comparative Studies of Long Wars.’’ In Grasping the Nettle: Analyzing Cases of Intractable Conflict, ed. Chester A. Crocker, Fen Osler Hampson, and Pamela Aall. Washington, DC: United States Institute of Peace (33–46). Liddle, R. William. 1997. ‘‘Coercion, Co-optation, and the Management of Ethnic Relations in Indonesia.’’ In Government Policies and Ethnic Relations in the Asia-Pacific, ed. Michael E. Brown and Sumit Ganguly. Cambridge and London: MIT Press (273–319). Lijphart, Arend. 1999. Patterns of Democracy. Government Forms and Performances in Thirty-Six Countries. New Haven and London: Yale University Press. 1994. ‘‘Democracies: Forms, Performance, and Constitutional Engineering.’’ European Journal of Political Research 25: 1–17. 1993. ‘‘Consociational Democracy.’’ In The Oxford Companion to Politics of the World, ed. Joel Krieger. New York and Oxford: Oxford University Press (188–189).

272

References

1985. Power-Sharing in South Africa. Berkeley: Institute of International Studies, University of California. 1977. Democracy in Plural Societies. New Haven and London: Yale University Press. 1968. ‘‘Typologies of Democratic Systems.’’ Comparative Political Studies 1 (April): 3–44. Linz, Juan, and Alfred Stepan. 1996. Problems of Democratic Transition and Consolidation: Southern Europe, South America, and Post-Communist Europe. Baltimore: Johns Hopkins University Press. 1992. ‘‘Political Identities and Electoral Sequences: Spain, the Soviet Union, and Yugoslavia.’’ Daedalus 121 (2): 123–139. Lipset, Seymour Martin. 1959. ‘‘Some Social Requisites of Democracy: Economic Development and Political Legitimacy.’’ American Political Science Review 53: 69–105. Lithander, Anna. 2000. Engendering the Peace Process: A Gender Approach to Dayton – and Beyond. Stockholm: Kvinna till Kvinna Foundation. Londregan, John B., and Keith T. Poole. 1996. ‘‘Does High Income Promote Democracy?’’ World Politics 49: 1–30. Lo´pez-Pintor, Rafael. 2000. Electoral Management Bodies as Institutions of Governance. New York: United Nations Development Programme. 1998. ‘‘Nicaragua’s Measured Move to Democracy.’’ In Postconflict Elections, Democratization, and International Assistance, ed. Krishna Kumar. Boulder, Colo.: Lynne Rienner Publishers (39–55). 1997. ‘‘Reconciliation Elections: A Post-Cold War Experience.’’ In Rebuilding Societies after Civil War: Critical Roles for International Assistance, ed. K. Kumar. Boulder, Colo. and London: Lynne Rienner Publishers (43–61). Louw, Antoniette. 1997. ‘‘Surviving the Transition: Trends and Perceptions in Crime.’’ Social Indicators Research 41: 137–168. Ludwig, Robin. 2004. ‘‘The UN’s Electoral Assistance: Challenges, Accomplishments, Prospects.’’ In The UN Role in Promoting Democracy: Between Ideals and Reality, ed. Edward Newman and Roland Rich. Tokyo: UN University Press (169–187). Lyons, Terrence. 2005. Demilitarizing Politics: Elections on the Uncertain Road to Peace. Boulder, Colo.: Lynne Rienner Publishers. 2004. ‘‘Transforming the Institutions of War: Post-Conflict Elections and the Reconstruction of Failed States.’’ In When States Fail, ed. Robert I. Rotberg. Princeton: Princeton University Press (269–301). 2002. ‘‘The Role of Postsettlement Elections.’’ In Ending Civil Wars: The Implementation of Peace Agreements, ed. Stephen John Stedman, Donald Rothchild, and Elisabeth M. Cousens. Boulder, Colo.: Lynne Rienner Publishers (215–235).

References

273

Mac Ginty, Roger. 2006. ‘‘Post-Accord Crime.’’ In Violence and Reconstruction, ed. J. Darby. South Bend, Ind.: University of Notre Dame Press (101–120). Mac Ginty, Roger, and John Darby. 2002. Guns and Government: The Management of the Northern Ireland Peace Process. Basingstoke: Palgrave. MacGarry, John, and Brendan O’Leary, eds. 2004. The Northern Ireland Conflict: Consociational Engagements. Oxford: Oxford University Press. MacMillan, John. 2003. ‘‘Beyond the Separate Democratic Peace.’’ Journal of Peace Research 40 (2): 233–243. Maksymenko, Serhiy. 2005. ‘‘European Neighbourhood Policy and PostOrange Revolution Ukraine.’’ In Development, Security and Conflict Prevention, ed. Anders Mellbourn. Brussels: Madariaga European Foundation Bank of Sweden Tercentenary Foundation and Gidlunds (164–174). Malone, David, ed. 2004. The Security Council: From the Cold War to the 21st Century. Boulder, Colo. and London: International Peace Academy and Lynne Rienner Publishers. Mani, Rama. 2002. Beyond Retribution: Seeking Justice in the Shadows of War. Cambridge, UK: Polity Press. Mann, Michael. 2005. The Dark Side of Democracy: Explaining Ethnic Cleansing. Cambridge: Cambridge University Press. Manning, Carrie. 2004. ‘‘Armed Opposition Groups into Political Parties: Comparing Bosnia, Kosovo, and Mozambique.’’ Studies in Comparative International Development 39 (1): 54–76. Mansfield, Edward D., and Jack Snyder. 2005. Electing to Fight: Why Emerging Democracies Go to War. Cambridge: MIT Press. 2002a. ‘‘Democratic Transitions, Institutional Strength, and War.’’ International Organization 56 (2): 297–337. 2002b. ‘‘Incomplete Democratic Transitions and the Outbreak of Military Disputes.’’ International Studies Quarterly 46 (4): 529–549. 2001. ‘‘Democratic Transitions and War: From Napoleon to the Millennium’s End.’’ In Turbulent Peace: The Challenges of Managing International Conflict, ed. C. A. Crocker, F. O. Hampson, and P. Aall. Washington, DC: United States Institute of Peace Press (113–126). 1995a. ‘‘Democratization and the Danger of War.’’ International Security 20 (1): 5–38. 1995b. ‘‘Democratization and War.’’ Foreign Affairs 74 (3): 79–97. Maoz, Zeev, and Bruce M. Russett. 1993. ‘‘Normative and Structural Causes of Democratic Peace 1946–1986.’’ American Political Science Review 87 (3): 624–638.

274

References

Marshall, Monty G., and Keith Jaggers. 2002. Polity IV Project: Political Regime Characteristics and Transitions, 1800–2002. College Park, Md.: Center for International Development and Conflict Management (CIDCM), University of Maryland. Marten, Kimberly Zisk. 2004. Enforcing the Peace: Learning from the Imperial Past. New York: Columbia University Press. Mathie, Alison, and Gord Cunningham. 2002. From Clients to Citizens: Asset-Based Community Development as a Strategy for CommunityDriven Development. The Coady International Institute, St. Francis Xavier University, Occasional Paper. McCoy, Jennifer, Larry Gerber, and Robert Pastor. 1991. ‘‘Pollwatching and Peacemaking.’’ Journal of Democracy 2 (4): 102–114. Melber, Henning. 2002. ‘‘From Liberation Movements to Governments: On Political Culture in Southern Africa.’’ African Sociological Review 6 (1): 161–172. Mendelson, Sarah E. 2005. Barracks and Brothels: Peacekeepers and Human Trafficking in the Balkans. Washington, DC: Center for Strategic and International Studies. Metzl, Jamie F. 1997. ‘‘Information Intervention: When Switching Channel Isn’t Enough.’’ Foreign Affairs 76 (6): 15–20. Morrow, James D. 1994. Game Theory for Political Scientists. Princeton: Princeton University Press. Mukherjee, Bumba. 2006. ‘‘Why Political Power-Sharing Agreements Lead to Enduring Peaceful Resolution of Some Civil Wars, But Not Others?’’ International Studies Quarterly 50 (2): 479–504. Mullojanov, Perviz. 2001. ‘‘Civil Society and Peacebuilding. Politics of Compromise: The Tajikistan Peace Process.’’ Conciliation Resources, Accord No. 10, March. Available at www.c-r.org/accord10/ peacebuilding.shtml. Museveni, Yoweri. 2005. Quoted on BBC December 21, 2005. Available at http://news.bbc.co.uk/1/hi/world/africa/4549220.stm. National Democratic Institute. 1996. Bosnia and Hercegovina (BiH): Strengthening Political Parties. Washington DC: NDI Reports. Nelson, Sue. 1998. ‘‘Haitian Elections and the Aftermath.’’ In Postconflict Elections, Democratization, and International Assis-tance, ed. Krishna Kumar. Boulder, Colo.: Lynne Rienner Publishers (71–86). Newman, Edward. 2004. ‘‘UN Democracy Promotion: Comparative Advantages and Constraints.’’ In The UN Role in Promoting Democracy: Between Ideals and Reality, ed. Edward Newman and Roland Rich. Tokyo: United Nations University Press (188–207).

References

275

Newman, Edward, and Roland Rich, eds. 2004. The UN Role in Promoting Democracy: Between Ideals and Reality. Tokyo: United Nations University Press. Nilsson, Desire´e. 2006. In the Shadow of Settlement: Multiple Rebel Groups and Precarious Peace. Uppsala: Department of Peace and Conflict Research, Uppsala University. 2005. ‘‘Inclusive Peace Agreements: A Path to Durable Peace?’’ Paper presented at a review conference October 24–26, 2005 in Uppsala on the book project From Intra-State War to Durable Peace: Conflict and Conflict Resolution in Africa 1989–2004. Nilsson, Desire´e, and Mimmi So¨derberg Kovacs. 2005. ‘‘Breaking the Cycle of Violence? Promises and Pitfalls of the Liberian Peace Process.’’ Civil Wars 7 (4): 396–414. Oakley, Robert B., Michael J. Dziedzic, and Eliot M. Goldberg, eds. 2002. Policing the New World Order: Peace Operations and Public Security. Washington DC: National Defence University Press. O’Donnell, Guillermo, Philippe C. Schmitter, and Lawrence Whitehead, eds. 1986. Transitions from Authoritarian Rule: Prospects for Democracy. Baltimore: Johns Hopkins University Press. Ohlson, Thomas. 1998. Power Politics and Peace Policies: Intra-State Conflict Resolution in Southern Africa, Uppsala: Department of Peace and Conflict Research, Uppsala University. Oneal, John R., Bruce M. Russett, and Michael L. Berbaum. 2003. ‘‘Causes of Peace: Democracy, Interdependence, and International Organizations, 1885–1992.’’ International Studies Quarterly 47 (3): 371–393. Orjuela, Camilla. 2005. ‘‘Civil Society in Civil War: The Case of Sri Lanka.’’ Civil Wars 7 (2): 120–137. OSCE. 2006. ‘‘International Election Observation Mission, Bosnia and Herzegovina – General Elections, 1 October 2006.’’ OSCE/ODIHR Election Observation Mission Final Report. Warsaw: OSCE. Ottaway, Marina. 1997. ‘‘From Political Opening to Democratization?’’ In Democracy in Africa. The Hard Road Ahead, ed. Marina Ottaway. Boulder: Lynne Rienner Publishers (1–14). 1991. ‘‘Liberation Movements and Transition to Democracy: The Case of the ANC.’’ Journal of Modern African Studies 29 (1): 61–82. Paffenholz, Thania. 2001. ‘‘Designing Transformation and Intervention Processes.’’ In Berghof Handbook for Conflict Transformation, ed. David Bloomfield, Martina Fischer, and Beatrix Schmelzle. Berlin: Berghof Center for Constructive Conflict Management. Available at www.berghof-handbook.net Paffenholz, Thania, and Christoph Spurk. 2006. Civil Society, Civic Engagement, and Peacebuilding. Washington, DC: Social

276

References

Development Papers, Paper No. 36, commissioned by the Social Development Department of the World Bank, October. Papagianni, Katia. 2006. ‘‘Transitional Politics in Post-War Countries.’’ Paper presented at the International Symposium ‘‘State, Conflict and Democracy,’’ Lund, May 12–13, 2006. Paris, Roland 2004. At War’s End: Building Peace after Civil Conflicts. Cambridge: Cambridge University Press. Patrick, Ian. 2001. ‘‘East Timor Emerging From Conflict: The Role of Local NGOs and International Assistance.’’ Disasters 25 (1): 48–66. Pausewang, Siegfried, Kjetil Trondvall, and Lovise Aalen, eds. 2002. Ethiopia since the Derg: A Decade of Democratic Pretension and Performance. London: Zed Books. Perito, Robert M. 2004. Where Is the Lone Ranger When We Need Him? America’s Search for a Postconflict Stability Force. Washington, DC: United States Institute of Peace Press. Phillips, Anne. 1998. The Politics of Presence. Oxford: Oxford University Press. Pickering, Jeffrey, and Mark Peceny. 2006. ‘‘Forging Democracy at Gunpoint.’’ International Studies Quarterly 50 (3): 539–560. Plunkett, Mark. 2005. ‘‘Reestablishing the Rule of Law.’’ In Postconflict Development: Meeting New Challenges, ed. G. Junne and W. Verkoren. Boulder and London: Lynn Rienner Publishers (73–98). Posen, Barry R. 1993. ‘‘The Security Dilemma and Ethnic Conflict.’’ Survival 35 (1): 27–47. Posner, Daniel N. 2004. ‘‘Civil Society and the Reconstruction of Failed States.’’ In When States Fail: Causes and Consequences, ed. Robert Rotberg. Princeton: Princeton University Press (237–255). Pouligny, Beatrice. 2005. ‘‘Civil Society and Post-Conflict Peacebuilding: Ambiguities of International Programmes Aimed at Building ‘New’ Societies.’’ Security Dialogue 46 (4): 495–510. Przeworski, Adam. 1991. Democracy and the Market: Political and Economic Reforms in Eastern Europe and Latin America. Cambridge: Cambridge University Press. Przeworski, Adam, Michael Alvarez, Jose Antonio Cheibub, and Fernando Limongi. 1996. ‘‘What Makes Democracy Endure.’’ Journal of Democracy 7 (1): 39–55. Przeworski, Adam, and Fernando Limongi. 1997. ‘‘Modernization: Theories and Facts.’’ World Politics 49 (2): 155–183. Pugh, Michael. 2000. ‘‘The Civil/Social Dimension.’’ In The Regeneration of War-Torn Societies, ed. Michael Pugh. London: Macmillan (112–133). Putnam, Robert. 2000. Bowling Alone: The Collapse and Revival of American Community. New York: Simon & Schuster.

References

277

1993. Making Democracy Work: Civic Traditions in Modern Italy. Princeton: Princeton University Press. Rabushka, Alvin, and Kenneth Shepsle. 1972. Politics in Plural Societies: A Theory of Democratic Instability. Colombus, OH: Merrill. Randall, Vicky, and Lars Sva˚sand. 2002. ‘‘Political Parties and Democratic Consolidation in Africa.’’ Democratization 9 (3): 30–52. Rapoport, David C., and Leonard Weinberg, eds. 2001a. The Democratic Experience and Political Violence. London: Frank Cass. 2001b. ‘‘Elections and Violence.’’ In The Democratic Experience and Political Violence, ed. D. C. Rapoport and L. Weinberg. London: Frank Cass (15–20). Rawls, John. 1971. A Theory of Justice. Cambridge, Mass.: Harvard University Press. Reilly, Benjamin. 2007. Democracy and Diversity: Political Engineering in the Asia-Pacific. Oxford: Oxford University Press. 2006. ‘‘Political Engineering and Party Politics in Conflict-Prone Societies.’’ Democratization 13 (5): 811–827. 2004. ‘‘Elections in Post-Conflict Societies.’’ In The UN Role in Promoting Democracy: Between Ideals and Reality, ed. Edward Newman and Roland Richs. Tokyo: United Nations University Press (113–134). 2003a. ‘‘Democratic Validation.’’ In Contemporary Peacemaking: Conflict, Violence and Peace Processes, ed. John Darby and Roger MacGinty. London: Palgrave (174–183). 2003b. International Electoral Assistance: A Review of Donor Activities and Lessons Learned. The Hague: Netherlands Institute of International Relations ‘‘Clingendael.’’ 2002. ‘‘Post-Conflict Elections: Constraints and Dangers.’’ International Peacekeeping 9 (2): 118–139. 2001. Democracy in Divided Societies: Electoral Engineering for Conflict Management. Cambridge: Cambridge University Press. Reinicke, Wolfgang H., and Francis Deng. 2000. Critical Choices: The United Nations, Networks, and the Future of Global Governance. Ottawa: International Development Research Centre. Reynolds, Andrew, ed. 2002. The Architecture of Democracy: Constitutional Design, Conflict Management, and Democracy. Oxford: Oxford University Press. Reynolds, Andrew, Benjamin Reilly, and Andrew Ellis. 2005, Electoral System Design: The New International IDEA Handbook. Stockholm: International Institute for Democracy and Electoral Assistance. Rich, Roland. 2004. ‘‘Crafting Security Council Mandates.’’ In The UN Role in Promoting Democracy: Between Ideals and Reality, ed. Edward Newman and Roland Rich. Tokyo: United Nations University Press (62–85).

278

References

Risley, Paul, and D. Timothy Sisk. 2005. Democracy and United Nations Peacebuilding at the Local Level: Lessons Learned. Stockholm: International IDEA. Available at www.idea.int/publications/democracy_un/ index.cfm. Roeder, Philip G., and Donald Rothchild, eds. 2005. Sustainable Peace: Power and Democracy after Civil Wars. Ithaca and London: Cornell University Press. Rosato, Sebastian. 2003. ‘‘The Flawed Logic of Democratic Peace Theory.’’ American Political Science Review 97 (4): 585–602. Ross, Michael L. 2001. ‘‘Does Oil Hinder Democracy?’’ World Politics 53 (3): 325–361. Rothchild, Donald. 2005a. ‘‘Problems in Applying Executive Power Sharing to Africa: The Impact of Sequencing on Political Consolidation.’’ Paper presented at 46th Annual International Studies Association Convention, 1–5 March, at Honolulu, HI. 2005b. ‘‘Reassuring Weaker Parties after Civil Wars: The Benefits and Costs of Executive Power-sharing Systems in Africa.’’ Ethnopolitics 4 (3): 247–267. 1997. Managing Ethnic Conflict in Africa: Pressures and Incentives for Cooperation. Washington, DC: Brookings Institute. 1973. Racial Bargaining in Independent Kenya: A Study of Minorities and Decolonization. Berkeley: University of California Press. Russett, Bruce. 1993. Grasping the Democratic Peace. Princeton: Princeton University Press. Russett, Bruce M., and John R. Oneal. 2001. Triangulating Peace: Democracy, Interdependence, and International Organizations. New York: W. W. Norton. Rustow, Dankwart A. 1970. ‘‘Transitions to Democracy: Toward a Dynamic Model.’’ Comparative Politics 2 (3): 337–363. Salih, M. A. Mohamed. 2003. ‘‘Introduction: The Evolution of African Political Parties.’’ In African Political Parties: Evolution, Institutionalisation and Governance, ed. M. A. Mohamed Salih. London: Pluto Press (1–33). Salomons, Dirk. 2005. ‘‘Security: An Absolute Prerequisite.’’ In Postconflict Development: Meeting New Challenges, ed. G. Junne and W. Verkoren. Boulder Colo. and London: Lynne Rienner Publishers (19–42). Shaw, Mark. 2002. Crime and Policing in Post-Apartheid South Africa: Transforming Under Fire. Cape Town and Johannesburg: David Philip Publisher. Shoemaker, Jolynn. 2005. Conflict Prevention and Transformation: Women’s Vital Contribution. Cambridge and Washington, DC: Women Waging Peace and the United Nations Foundation.

References

279

Sida. 2006. Press Statement, August 9, 2006. Stockholm: Swedish International Development Cooperation Agency. 2005. Promoting Peace and Security through Development Cooperation. Stockholm: Swedish International Development Cooperation Agency. Simonsen, Sven Gunnar. 2004. ‘‘Nationbuilding as Peacebuilding: Racing to Define the Kosovar.’’ International Peacekeeping 11 (2): 289–311. Sisk, Timothy D. 2001. ‘‘Intrastate Violence.’’ In Managing Global Issues: Lessons Learned, ed. P. J. Simmons and Chantal de Jonge Oudraat. Washington, DC: Carnegie Endowment for International Peace (534–563). 1996. Power Sharing and International Mediation in Ethnic Conflict. Washington, DC: Carnegie Corporation of New York. 1995. Democratization in South Africa: The Elusive Social Contract. Princeton: Princeton University Press. Sisk, Timothy D., and Christoph Stefes. 2005. ‘‘Power Sharing as an Interim Step in Peace Building: Lessons from South Africa.’’ In Sustainable Peace. Power and Democracy after Civil Wars, ed. Philip G. Roeder and Donald Rothchild. Ithaca and London: Cornell University Press (293–319). Smith, Anthony L. 2004. ‘‘East Timor: Elections in the World’s Newest Nation.’’ Journal of Democracy 15 (2): 145–159. Snyder, Jack. 2000. From Voting to Violence: Democratization and Nationalist Conflict. New York and London: W.W. Norton. Snyder, Jack, and Karen Ballentine. 1996. ‘‘Nationalism and the Marketplace of Ideas.’’ International Security 21 (2): 5–40. So¨derberg Kovacs, Mimmi. 2007. From Rebellion to Politics: The Transformation of Rebel Groups to Political Parties in Civil War Peace Processes. Uppsala: Department of Peace and Conflict Research, Uppsala University. Solana, Javier. 2005. ‘‘Prevention Must Be Concrete.’’ In Development, Security and Conflict Prevention, ed. Anders Mellbourn. Brussels: Madariaga European Foundation (9–14). Southall, Roger, Neo Simutanyi, and John Daniel. 2005. ‘‘Former Presidents in African Politics.’’ In Legacies of Power, ed. Roger Southall and Henning Melber. Uppsala: Nordic Africa Institute (1–25). Spear, Joanna. 2002. ‘‘Disarmament and Demobilization.’’ In Ending Civil Wars: The Implementation of Peace Agreements, ed. John Stephen Stedman, Donald Rothchild, and Elizabeth M. Cousens. Boulder, Colo.: Lynne Rienner Publishers (141–182). Spears, Ian S. 2000. ‘‘Understanding Inclusive Peace Agreements in Africa: The Problems of Sharing Power.’’ Third World Quarterly 21 (1): 105–118. Stedman, Stephen John. 1997. ‘‘Spoiler Problems in Peace Processes.’’ International Security 22 (2): 5–53.

280

References

Stedman, Stephen John, Donald Rothchild, and Elisabeth M. Cousens, eds. 2002. Ending Civil Wars: The Implementation of Peace Agreements. Boulder, Colo.: Lynne Rienner Publishers. Stewart, Frances. 2000. ‘‘Crisis Prevention: Tacking Horizontal Inequalities.’’ Oxford Development Studies 28 (3): 245–262. Stremlau, John. 1998. A House No Longer Divided: Civil Society in South Africa. Washington, DC: Carnegie Commission on Preventing Deadly Conflict. Stubbs, Paul. 1999. Displaced Promises: Forced Migration, Refugees and Return in Croatia and Bosnia-Herzegovina. Uppsala: Life & Peace Institute. Tangri, Roger. 2005. ‘‘Politics and Presidential Term Limits in Uganda.’’ In Legacies of Power, ed. Roger Southall and Henning Melber. Uppsala: Nordic Africa Institute (175–196). Temporary Media Commissioner. 2004. An Inquiry into the Performance of Kosovo-Wide Television (16–17 March 2004). Pristina: Temporary Media Commissioner. Thompson, Mark. 1994. Forging War: The Media in Serbia, Croatia and Bosnia-Herzegovina. London: Article 19. Thompson, William R., and Tucker Richard. 1997. ‘‘A Tale of Two Democratic Peace Critiques.’’ Journal of Conflict Resolution 41 (3): 428–454. Tilly, Charles. 1990. Coercion, Capital, and European States, AD 990–1992. Cambridge Mass.: Basil Blackwell. 1985. ‘‘War-Making and State-Making as Organized Crime.’’ In Bringing the State Back In, ed. P. Evans, D. Rueschmeyer, and T. Skocpol. Cambridge: Cambridge University Press (169–191). Toft, Monica Duffy. 2003. ‘‘Peace Through Victory?’’ Paper presented at American Political Science Association, August 27–31, Philadelphia. UCDP. 2007. Uppsala Conflict Data Program. [cited 24 March 2007]. Available at www.pcr.uu.se/database/. 2006. ‘‘Sudan Conflict Summary’’. Uppsala Conflict Data Program. Department of Peace and Conflict Research, Uppsala University [cited 24 May 2006]. Available at www.pcr.uu.se/database/. 2005. Uppsala Conflict Data Program. Department of Peace and Conflict Research, Uppsala University [cited June 2, 2005]. Available at www.pcr.uu.se/database/. United Nations. 2001. Letter dated 12 February 2001 from the SecretaryGeneral addressed to the President of the Security Council. Uvin, Peter. 1998. Aiding Violence: The Development Enterprise in Rwanda. West Hartford, Conn.: Kumarian Press. Uvin Peter, and Charles Mironko. 2003. ‘‘Western and Local Approaches to Justice in Rwanda.’’ Global Governance 9 (2): 129–231.

References

281

van Biezen, Ingrid. 2003. Political Parties in New Democracies: Party Organization in Southern and East-Central Europe. Houndmills: Palgrave Macmillan. Van Tongeren, Paul, Malin Brenk, Marte Hellema, and Juliette Verhoeven, eds. 2005. People Building Peace II: Successful Stories of Civil Society. Boulder, Colo. and London: Lynne Rienner Publishers. Vanberg, Victor, and James Buchanan. 1989. ‘‘Interests and Theories in Constitutional Choice.’’ Journal of Theoretical Politics 1: 49–62. Varshney, Ashutosh. 2002. Ethnic Conflict and Civic Life: Hindus and Muslims in India. New Haven: Yale University Press. 2001. ‘‘Ethnic Conflict and Civil Society.’’ World Politics 53 (April): 362–398. Viggo Jakobsen, Peter. 2003. ‘‘Military Forces and Public Security Challenges.’’ In The United Nations and Regional Security: Europe and Beyond, ed. M. Pugh and W. P. Singh Sidhu. Boulder, Colo. and London: Lynne Rienner Publishers (137–153). Vinegard, Anna. 1998. ‘‘From Guerrillas to Politicians: The Transition of the Guatemalan Revolutionary Movement in Historical and Comparative Perspective.’’ In Guatemala after the Peace Accords, ed. Rachel Sieder. London: Institute of Latin American Studies (207–227). Vines, Alex. 1996. Renamo: From Terrorism to Democracy in Mozambique? London: James Currey. Wade, Christine. 2007. ‘‘El Salvador: The Electoral Success of the FMLN.’’ In From Soldiers to Politicians: Transforming Rebel Movements after War, ed. Jeroen de Zeeuw. Boulder, Colo.: Lynne Rienner Publishers (33–54). Wallensteen, Peter 2007. Understanding Conflict Resolution, 2nd edn. London: Sage; 1st edn 2002. 2002. Understanding Conflict Resolution. War, Peace and the Global System. London: Sage Publications. Wallensteen, Peter, and Carina Staibano, eds. 2005. International Sanctions: Between Words and Wars in the Global System. London and New York: Frank Cass. Walter, Barbara F. 2002. Committing to Peace: The Successful Settlement of Civil Wars. Princeton: Princeton University Press. 1999. ‘‘Designing Transitions from Civil War: Demobilization, Democratization, and Commitments to Peace.’’ International Security 24 (1): 127–155. Wanis-St. John, Anthony, and Darren Kew. 2006. ‘‘The Missing Link? Civil Society, Peace Negotiations: Contributions to Sustained Peace.’’ Paper presented at the 47th Annual Convention of the International Studies Association, San Diego, March 23.

282

References

Wantchekon, Leonard. 2004. ‘‘The Paradox of ‘Warlord’ Democracy: A Theoretical Investigation.’’ American Political Science Review 98 (1) (February): 17–33. Wantchekon, Leonard, and Zvika Neeman. 2002. ‘‘A Theory of Post-Civil War Democratization.’’ Journal of Theoretical Politics (14) 439–464. Ward, Michael D., and Kristian S. Gleditsch. 1998. ‘‘Democratizing for Peace.’’ American Political Science Review 92 (1): 51–61. Waterman, Harvey. 1993. ‘‘Political Order and the ‘Settlement’ of Civil Wars.’’ In Stopping the Killing: How Civil Wars End, ed. Roy Licklider. New York: New York University Press (292–302). Weinstein, Jeremy M. 2005. ‘‘Autonomous Recovery and International Intervention in Comparative Perspective.’’ Working Paper, Washington, DC: Center for Global Development. Whitehead, Laurence. 2002. Democratization: Theory and Practice. Oxford: Oxford University Press. Wilkinson, Steven I. 2004. Votes and Violence: Electoral Competition and Ethnic Riots in India. Cambridge: Cambridge University Press. Williamson, Arthur, Duncan Scott, and Peter Halfpenny. 2000. ‘‘Rebuilding Civil Society in Northern Ireland: The Community and Voluntary Sector’s Contribution to the European Union’s Peace and Reconciliation District Partnership Programme.’’ Policy & Politics 28 (1): 49–66. Wolfsfeld, Gadi. 2004. Media and the Path to Peace. Cambridge: Cambridge University Press. 2003. ‘‘The Role of the News Media in Peace Negotiations: Variations over Time and Circumstance.’’ In Contemporary Peacemaking: Conflict, Violence and Peace Processes, ed. J. Darby and R. Mac Ginty. Basingstoke: Palgrave Macmillan (87–100). 2001. ‘‘The News Media and Peace Processes: The Middle East and Northern Ireland.’’ Peaceworks (37). Washington, DC: United States Institute of Peace (1–55). Woodward, Susan L. 2002. ‘‘Economic Priorities for Successful Peace Implementation.’’ In Ending Civil Wars. The Implementation of Peace Agreement, ed. John Stephen Stedman, Donald Rothchild and Elisabeth M. Cousens. Boulder, Colo.: Lynne Rienner Publishers (183–214). World Bank 2003a. Breaking the Conflict Trap: Civil War and Development Policy. Washington DC: World Bank and Oxford University Press. 2003b. Working Together: World Bank-Civil Society Relations. Washington, DC: IBRD/World Bank. Available at: http://siteresources. worldbank.org/CSO/Resources?WorkingTogetherBrochure.pdf. 2002a. The Right to Tell: The Role of Mass Media. Washington DC: World Bank.

References

283

2002b. Local Level Institutions and Social Capital Study, vol. I, June, Washington, DC: IBRD/World Bank. Zahar, Marie-Joe¨lle. 2005a. ‘‘Violence and War-to-Peace Transitions: A Second Look at the ‘Spoiler’ Debate in Light of African Case Studies.’’ Paper presented at 101st Annual Meeting of the American Political Science Association, September 1–4, 2005, Washington, DC. 2005b. ‘‘Power Sharing in Lebanon: Foreign Protectors, Domestic Peace, and Democratic Failure.’’ In Sustainable Peace. Power and Democracy After Civil Wars, ed. Philip G. Roeder and Donald Rothchild. Ithaca and London: Cornell University Press (219–240). 2003. ‘‘Reframing the Spoiler Debate in Peace Processes.’’ In Contemporary Peacemaking: Conflict, Violence and Peace Processes, ed. John Darby and Roger Mac Ginty. Basingstoke: Palgrave Macmillan (114–124). Zartman, I. William, ed. 1995. Elusive Peace: Negotiating an End to Civil Wars. Washington, DC: Brookings Institute. 1991. ‘‘Common Elements in the Analysis of a Negotiation Process.’’ In Negotiation Theory and Practice, ed. J. William Breslin and Jeffrey Z. Rubin. Cambridge, Mass.: Program on Negotiation, Harvard Law School (147–159).

Index

Page numbers followed by n indicate footnotes. Abuja accords, 152–153 accountability effect of foreign peacekeeping on, 45–46 electoral systems and, 172 Aceh, 180 Afghanistan, 31–32, 171, 174 African National Congress (ANC), 115–116 amnesty for war crimes, 142–145, 154–155 Angola, 25–26, 42, 114, 118–119 Annan, Kofi, 5, 27 armed conflict defined, 4[n2] armed forces, demobilization, 27 armed groups. See rebels and insurgents Aung San Suu-kyi, 227–228 authority issues. See legitimacy; systemic dilemma Bosnia and Herzegovina, 24, 31–32, 42, 91, 105, 111–112, 125, 127, 151, 160, 167, 201, 221–222 Bougainville, 158, 169–170 Boutros-Ghali, Boutros, 27, 239 Burma, 227–228 Burundi, 22, 111–112, 116–117, 128–129 Bush administration, 162 Cambodia, 22, 25–26, 69–70, 122, 145, 165–166 cease-fires, 96–97 Central America, 149

284

civil liberties, 64–65, 75 civil society blurring between government and, 201 dilemmas of democratization, 23, 183–184, 194–207 exclusion from power sharing, 23 international encouragement of, 201–207, 209–210 in peace process, 189, 191–199 as pro-democracy ideal, 184–186, 189 strategy for promoting, 208–210 uncivil and violent forms, 186–188, 189–190, 191–194 in war-to-democracy transitions, 186, 187–190, 199–200, 255–256 civil war defined, 4[n2] effect of victory on democratization, 62 legacy of violence, 31–32 partition of country following, 2–3 Clinton, Bill, 162 Clinton administration, 167 Colombia, 121[n20] colonies, 63–64 Congo. See Democratic Republic of the Congo (DRC) consociational democracy, 110–111, 122–123 constitutions, 10, 105 coordination problems, 6–7 credible commitments to democratic rules, 10

Index peace agreements, 108, 252–253, 256–259 crime, 83, 89–90, 93–94, 101 See also amnesty for war crimes; human rights abuses Croatia, 25–26, 201 Cyprus, 23–24, 119–120, 126, 128 Darfur agreement, 22, 117 Dayton Peace Agreement, 91, 105, 111–112, 195–196, 221–222 demobilization, 27 ‘‘democratic peace,’’ 5, 29, 162–163, 214 Democratic Republic of the Congo (DRC), 22, 23, 98, 124–125, 127, 198, 259 democratic transition theory, 1 democratization assessment approaches, 250–252 civil war protagonists’ views, 7–10 as conducive to conflict and violence, 12, 28–31, 80–84, 101 consociational democracy, 110–111, 122–123 country rankings, 25–26 defined, 17, 32 impact of peacekeeping case studies, 69–71 overview, 40, 44–48, 74–77 short- versus long-term effects, 51–52, 64–74, 75–76, 256–259 statistical analysis: approaches and methodology, 48–54, 55–64, 75–76 statistical analysis: findings, 54–55, 64–74 international actors’ support for, 4–7, 256–259 international versus local ownership, 24–25, 220 peace as precondition, 3–4, 5, 26, 46, 256–259 peace through, 3–5, 7–10, 181, 239–241 political parties as agents, 139–141 power sharing and, 122–129, 161 See also democratization crises; democratization versus peacebuilding dilemmas

285 democratization crises Bosnia and Herzegovina, 221–222 electoral violence and irregularities, 223–228 forms of international intervention, 229, 256–259 international versus domestic control over, 6, 219–223, 231–236 Lebanon, 224–225, 232–233 timing and duration of intervention, 223–231, 236, 256–259 triggers for international engagement, 217–219, 222–223, 236 types, 217–219 Uganda, 216–217, 220–221, 224, 229–230, 233–235 United Nations role in, 214–215, 219–221, 237–238 democratization versus peacebuilding dilemmas international actors’ perspectives, 4–7 need for trade-offs, 17–18, 35–36, 243 overview, 1–2, 4–7, 10–13, 17–20, 239–241, 250 peacebuilding given priority, 18, 26, 35–36 policy recommendations, 250–259 protagonists’ perspectives, 7–10 See also horizontal dilemma; systemic dilemma; temporal dilemma; vertical dilemma disarmament, 27 East Timor, 3, 70–71, 86, 171, 174 efficacy versus legitimacy. See vertical dilemma El Salvador, 25–26, 42, 90, 93, 149 elections contradictory goals and outcomes, 157–159, 161–163, 180–181 democratic dilemmas, 159–161, 165, 172 and democratization crises, 223–228 electoral administration, 166, 175–177

286 elections (cont.) electoral systems, 172–175 electoral violence, 84–87, 225 as indicators of democratization, 223 international support for, 161–163, 165–166, 181, 223–228 minimal security as precondition, 35–36 outbidding in, 31–32, 160–161 sequencing of local and national, 170 timing, 25, 91, 154, 167–168 in war-to-democracy transitions, 84–87 See also referendums elite-mass dilemmas. See vertical dilemma (legitimacy versus efficacy dilemma) Eritrea, 3 Ethiopia, 87 ethnic conflict Burundi, 128–129 and later democratization, 56 managed through power sharing, 110, 111–112 ethnic representation, 127–129 See also consociational democracy exclusion. See horizontal dilemma; power sharing Fiji, 215 force controlling violence with, 94–96 legitimacy, 97 peacekeepers’ use, 6, 95–96 GDP per capita, and democratization, 57–61 genocide, 88, 145 Germany, 213 great powers, 63 Guatemala, 42, 90, 95, 149, 198–199 guerillas. See rebels and insurgents Hamas, 22, 151, 160 Haradinaj, Ramush, 143–144 Hariri, Rafik, 224 hate propaganda, 88 Herzegovina. See Bosnia and Herzegovina Hizbollah, 120–121

Index horizontal dilemma (inclusion versus exclusion dilemma) civil society, 23, 183–184 definition and overview, 11, 21–23, 243–245 power sharing, 107, 115–117, 124–125, 131 rebel groups in post-war politics, 22, 81, 91, 98, 99, 141–145 human rights abuses, 92–93, 94, 142–145, 188 Human Rights Council, 238 human rights networks, 204–205 illiteracy, 61 inclusion. See horizontal dilemma; power sharing; rebels and insurgents India, 189 Indonesia, 177, 180 insurgents. See rebels and insurgents international actors coordination problems among, 6–7 democratization agendas, 3–5, 222, 256–259 and the democratization vs. peacebuilding dilemma, 3–5, 24–25, 201–205 versus local control, 7, 24–25, 44–45 myopic time frame, 205–207, 256–259 in the peacebuilding process, 4–7, 24–25, 76, 95–96 promotion of civil society, 201–205, 209–210, 255–256 promotion of parties and elections, 149–151, 161–163, 165–166, 177–180 responses to democratization crises duration of intervention, 228–231 legitimacy of intervention, 6, 219–236 range of bodies involved, 222 timing of intervention, 223–228, 230–231, 236 triggering events, 217–219, 222–223, 236 types of intervention measures, 229, 231–236

Index role in power-sharing agreements, 107, 119–120, 126–127, 131–132 role in violence management, 81, 98, 99 international criminal courts, 143 Iraq, 23–24, 171, 174, 239[n1] Irish Republican Army (IRA), 96–97, 116 Israel, 93 Ivory Coast, 124–125, 198 justice, 142–145 Khmer Rouge, 145 Kosovo, 25–26, 71, 88, 95–96, 126–127, 143–144, 148, 155, 171 Lebanon, 110, 120–121, 224–225, 232–233 legitimacy former armed groups as parties, 149–151 international criminal tribunals, 143 international intervention, 6, 219–223, 231–236 peace process, 22, 23–24, 152–153 power-sharing arrangements, 118–119, 125–126 via elections and referendums, 22, 23–24, 25, 164 violence and crime as challenge, 93–94 Liberia, 23, 91, 124–125, 152–153, 154, 158, 173, 235, 259 literacy, and democratization, 61 Lome´ accord, 121–122, 150–151 Lusaka accord, 118–119 Maori, 118[n16], 129 media, and violence, 87–88, 89, 90–91 military coups, UN reaction, 215 Milosevic, Slobodan, 182, 196, 226 Mozambique, 42, 148, 179 Museveni, Yoweri, 216–217, 229–230, 231, 233 Myanmar, 227–228 Namibia, 25–26, 40, 42 Nepal, 125

287 New Zealand, 118[n16], 129 NGOs, 206–207 See also civil society Northern Ireland, 94, 95, 96–97, 116, 126, 189, 190, 192, 196 oil resources, 63 Palestinian territories, 151, 160 Papua New Guinea, 42, 158, 169–170 Paris agreement, 145 partition, 2–3, 110 peace as precondition for democracy, 46 through democratization, 1–2, 3–5, 7–10, 27, 161–163, 181, 214, 239–241 peace agreements credible commitments, 108, 252–253, 256–259 domestic legitimacy, 5 effect on democratization, 62 failure to end conflict, 8–10 list of agreements/accords, 32–35, 111–113 negotiations conditions for entering, 136–138 critical issues, 241–242 inclusion of warring parties, 21–23, 96–97, 105–106 participation of civil society, 194–199 protagonists’ perspectives, 7–10 provision for democratization, 3–4 uncertainty for participants, 8–10 peacebuilding concept, 1, 17, 17[n1], 32, 239–241 versus democratization (See democratization versus peacebuilding dilemmas; peacekeeping: impact on democratization) given precedence over justice, 142–145 negative effects on democratization, 26–28 policy recommendations, 250–259 recent cases, 32–35 risks for protagonists, 8 role of civil society, 189

288 Peacebuilding Commission, 27, 237–238 peacekeeping defined, 41 effectiveness, 39, 52–54, 74, 76–77, 256–259 impact on democratization case studies, 69–71 overview, 40, 44–48, 74–77 short- versus long-term effects, 51–52, 64–74, 75–76, 256–259 statistical analysis: approach and methodology, 48–54, 55–64, 75–76 statistical analysis: findings, 54–55, 64–74 legitimacy and accountability issues, 44–46, 74 types of missions, 41–43, 65–69, 74–75 UN vs. non-UN, 43[n7], 71–74, 75 use of force in, 6, 94–96 police reform, 89–90 political parties as agents for democratization, 139–141, 177–178, 179 domestic versus international legitimization, 149–151 ethnic versus multiethnic, 177–178, 179 former armed rebels as, 145–149, 179–180 international promotion of, 177–180 one-party tendencies, 145–147 role in electoral administration, 175–177 sedimentation, 141–142, 178–179 weaknesses in new democracies, 140–141 political rights, 64–65, 75 power sharing arguments for, 105–106, 161 changing power relations under, 118–119, 120–122 defining the included/excluded groups, 115–117, 124–125, 127–129 democratization versus peacebuilding dilemmas, 21–23, 106–107, 115–117, 122–129, 131–132

Index guidelines, 132–133 international mediation, 119–120, 126–127 list of power-sharing accords, 111–113 meaning, 108–113 partial inclusion, 116–117 popular support for, 125–126 protagonists’ views, 8 recent agreements, 111–113 as source of conflict and violence, 106–107, 114–117, 119, 120–122, 130–132 sunset clauses, 121–122, 254 ‘‘projectism,’’ 207 proportional representation, 172–174, 175 rebels and insurgents amnesty for, 154–155 commitment to democratic politics, 145–149 contraband financing, 63 inclusion in new regime arguments for, 96, 136–138, 153 authoritarian tendencies, 145–147 continuation of wartime allegiances, 141–142, 145–149, 155 democratization dilemmas, 21–23, 81, 91, 98, 99, 141–153 domestic versus international authorization, 149–151, 155–156 horizontal dilemma, 21–23, 81, 91, 98, 99, 115–117, 141–145 impact on rule of law, 142–145 in police and security forces, 91–92 as political parties, 145–149 short- versus long-term effects, 152–153, 156 reconciliation commissions, 143 referendums, 23–24, 125–126, 168–169 rule of law, 142–145, 188 Rwanda, 25–26, 88, 201 sanctions, international, 232, 234–235

Index security issues: in war-to-democracy transition, 10, 11–12, 35 security sector reform (SSR), 89–90, 91–92 sedimentation of wartime divisions, 141–142, 154, 204 Serb Republic of Bosnia, 201 Serbia, 182, 226 short- versus long-term dilemma. See temporal dilemma Sierra Leone, 42, 114, 121–122, 144–145, 150–151 single non-transferable vote, 172–174, 175 Sinn Fein, 96–97, 116 social capital, 185, 190–191 South Africa, 115–116, 121, 127, 189, 196–197 Soviet Union, 3 spoilers, 23, 84, 96, 98 Sri Lanka, 97, 101, 197–198 Sudan, 22, 117, 124, 125–126 sunset clauses, 121–122, 254 systemic dilemma (local versus international control) civil society building, 183–184, 201–205 definition and overview, 11, 24–25, 248 democratization crises, 219–223, 236 former armed rebels in politics, 149–151 peacekeeping, 44–46, 74 power sharing, 107, 126–127, 131–132 violence management, 81, 98, 99 Tajikistan, 95 Taylor, Charles, 152–153 temporal dilemma (short- versus longterm dilemma) civil society building, 183–184, 205–207 definition and overview, 11, 25, 249 democratization crises, 223–226, 228–231, 236 election timing, 90–91, 166–171 former armed rebels in politics, 152–153 media reform, 90–91, 99

289 peacekeeping, 48, 64–74 power sharing, 120, 132 security sector reform (SSR), 91–92, 99 violence management, 81 terrorist organizations, 22 Thailand, 215 Timor Leste. See East Timor tolerance, 185 transitional arrangements, 121–122, 154, 177 transnational democracy networks, 204–205 trust, 186, 190 truth and reconciliation commissions, 143 Uganda, 216–217, 220–221, 224, 229–230, 233–235 Ukraine, 224 uncertainty dilemmas, 8–10, 165 United Nations Cambodia, 69–70, 165–166 East Timor, 70–71 election supervision, 173 Kosovo, 71, 126–127 peacebuilding agendas, 27, 230, 237–238, 239 peacekeeping missions, 43[n7] reaction to armed conflicts, 214–215 reaction to democratization crises, 214–215, 219–221, 237–238 short-term project orientation, 207 Uganda, 221 UN vs. non-UN peacekeeping, 43[n7], 71–74, 75 use of force by peacekeepers, 6 vertical dilemma (legitimacy versus efficacy dilemma) civil society building, 194–199 crises of democratization, 237 definition and overview, 11, 23–24, 245–248 election timing, 91–92 inclusion of former armed groups, 145–149 power sharing, 107, 125–126, 131 security sector reform, 91–92 violence management, 81, 97, 99

290 victory, effect on democratization, 61–62 violence democracy as conducive to, 29–31, 80–81, 101 and dilemmas of democratization, 81, 90–92, 97–101 electoral, 84–87, 225 legacy of, 31–32 management of conditions for success, 100, 252–253 dilemmas, 97–101 negative effects on democratization, 80, 92–96 through inclusion of spoilers, 96 use of force to curb, 94–96 media as instigator, 87–88, 89, 90–91 political versus criminal, 83, 89–90 under power-sharing arrangements, 106–107, 114–117, 119, 120–122, 130–132 security sector reform (SSR) and, 89–90, 91–92 in war-to-democracy transitions, 12, 31–32, 80–84, 92–101

Index war as spur to democratization, 46, 61–62 war-to-democracy transitions civil society in, 186, 187–190, 199–200, 255–256 concept and processes, 20–21 international engagement, 2–3, 4–7, 95–96, 256–259 protagonists’ perspectives, 7–10 security issues, 10, 11–12, 35 uncertainty dilemmas, 8–10 violence in, 12, 31–32, 80–84, 92–101 See also democratization; democratization versus peacebuilding dilemmas; peace agreements; peacebuilding warring parties. See rebels and insurgents Western Sahara, 42 women’s groups, 189 Yugoslavia, 3, 143–144, 155, 182, 226 Zimbabwe, 258