Ending Civil Wars: The Implementation of Peace Agreements 9781685850395

Citation preview

S-FMrev

11/25/09

11:13 AM

Page i

ENDING CIVIL WARS

S-FMrev

11/25/09

11:13 AM

Page ii

A project of the International Peace Academy and

the Center for International Security and Cooperation

S-FMrev

11/25/09

11:13 AM

Page iii

ENDING CIVIL WARS The Implementation of Peace Agreements

EDITED BY

Stephen John Stedman Donald Rothchild Elizabeth M. Cousens

b o u l d e r l o n d o n

S-FMrev

11/25/09

11:13 AM

Page iv

Published in the United States of America by Lynne Rienner Publishers, Inc. 1800 30th Street, Boulder, Colorado 80301 www.rienner.com and in the United Kingdom by Lynne Rienner Publishers, Inc. 3 Henrietta Street, Covent Garden, London WC2E 8LU © 2002 by the International Peace Academy, Inc. All rights reserved by the publisher Library of Congress Cataloging-in-Publication Data Ending civil wars : the implementation of peace agreements / edited by Stephen John Stedman, Donald Rothchild, and Elizabeth M. Cousens. p. cm. “A project of the International Peace Academy.” Includes bibliographical references and index. ISBN 978-1-58826-058-1 (hc : alk. paper) ISBN 978-1-58826-083-3 (pb : alk. paper) 1. Civil war—Political aspects. 2. Peace. 3. Intervention (International law)—Political aspects. I. Stedman, Stephen John. II. Rothchild, Donald S. III. Cousens, Elizabeth M. JZ6368.E53 2002 303.6'4—dc21 2002017817 British Cataloguing in Publication Data A Cataloguing in Publication record for this book is available from the British Library. Printed and bound in the United States of America The paper used in this publication meets the requirements of the American National Standard for Permanence of Paper for Printed Library Materials Z39.48-1992.

S-FMrev

11/25/09

11:13 AM

Page v

CONTENTS

Foreword, David M. Malone Preface

1

Introduction Stephen John Stedman

Part 1 2

3

4

5

1

Evaluating Implementation Strategies

Evaluation Issues in Peace Implementation George Downs and Stephen John Stedman

43

Strategy and Transitional Authority Michael W. Doyle

71

The Challenges of Strategic Coordination Bruce D. Jones

89

Settlement Terms and Postagreement Stability Donald Rothchild

Part 2 6

ix xi

117

Evaluating Implementation Tasks

Disarmament and Demobilization Joanna Spear v

141

S-FMrev

11/25/09

11:13 AM

Page vi

vi 7

8

9

10

11

12

CONTENTS

Economic Priorities for Successful Peace Implementation Susan L. Woodward

183

The Role of Postsettlement Elections Terrence Lyons

215

Human Rights and Sustainable Peace Tonya L. Putnam

237

Refugee Repatriation Howard Adelman

273

Civilian Security Charles T. Call and William Stanley

303

Building Local Capacity: From Implementation to Peacebuilding John Prendergast and Emily Plumb

327

Part 3 13

14

15

16

17

18

Case Studies

Peace in Stages: The Role of an Implementation Regime in Nicaragua Caroline A. Hartzell

353

Assessing El Salvador’s Transition from Civil War to Peace Charles T. Call

383

Broad Participation, Diffuse Responsibility: Peace Implementation in Guatemala William Stanley and David Holiday

421

Implementing the Arusha Peace Agreement on Rwanda Gilbert M. Khadiagala

463

Implementing Cambodia’s Peace Agreement Sorpong Peou

499

From Missed Opportunities to Overcompensation: Implementing the Dayton Agreement on Bosnia Elizabeth M. Cousens

531

S-FMrev

11/25/09

11:13 AM

Page vii

CONTENTS

19

20

21

Peace by Unconventional Means: Lebanon’s Ta’if Agreement Marie-Joëlle Zahar

567

Liberia: A Warlord’s Peace Adekeye Adebajo

599

Flawed Mediation, Chaotic Implementation: The 1987 Indo–Sri Lanka Peace Agreement Sumantra Bose

631

Part 4 22

vii

Conclusion

Policy Implications Stephen John Stedman

List of Acronyms Selected Bibliography The Contributors Index About the Book

663

673 683 699 705 729

S-FMrev

11/25/09

11:13 AM

Page viii

S-FMrev

11/25/09

11:13 AM

Page ix

FOREWORD DAVID M. MALONE PRESIDENT, INTERNATIONAL PEACE ACADEMY

This volume represents the culmination of an intense interactive process over a period of four years, bringing its authors together repeatedly with practitioners, including at the most senior reaches of the United Nations, and other scholars. We at the International Peace Academy are deeply grateful to the many participants in the book’s development, as well as to the editors and authors. We are especially grateful to the Center for International Security and Cooperation for their institutional partnership and fruitful intellectual collaboration. It is clear that international actors were on a steep learning curve on peace implementation throughout the 1990s. Thus, the purpose of the book’s editors was to ensure that the findings of the project, captured and synthesized here, would be as policy-relevant as possible. International actors have welcomed the insights presented to them in a series of briefings throughout 2000 and 2001, and they will delight in the in-depth encapsulation of these afforded in the pages that follow. The core findings of the research program, presented in the opening and concluding chapters of this volume, have proved tremendously influential at the UN. To note just one prominent example, in July 2001 key findings were cited during an informal meeting of the Security Council on peacebuilding in East Timor and found expression in several formal interventions in the council some days later. In many other ways, the lessons have become part of the background knowledge guiding policymakers in their day-to-day work on peacebuilding. It is a privilege for the International Peace Academy to have been associated with this exciting project. With the publication of this volume, we hope that many others will draw on its insights in years to come.

ix

S-FMrev

11/25/09

11:13 AM

Page x

S-FMrev

11/25/09

11:13 AM

Page xi

PREFACE

Between 1997 and 2001, Stanford University’s Center for International Security and Cooperation (CISAC) and the International Peace Academy (IPA) conducted research to better understand the determinants of successful peace implementation in civil wars. The CISAC-IPA project on peace implementation focused on three primary issues: 1. An evaluation of international actors and their strategies of peace implementation. 2. An evaluation of various tasks of peace implementation (for example, demobilization, disarmament, refugee repatriation, human rights, and reconciliation) and their relationship to overall implementation success. 3. A search for low-cost, possible high-payoff opportunities for linking short-term implementation success to longer-term peacebuilding. To strengthen the policy relevance of the research, practitioners and policymakers participated in every stage of the project. The project began with two conferences, one held at Stanford in November 1997 and the other held at the Rockefeller Brothers Fund conference center at Pocantico, New York, in March 1998. We asked experts in various aspects of civil war termination (for example, mediation, strategic coordination, transitional authority, human rights, elections, and refugee repatriation) to write short memos framing what they considered to be the key issues in ending civil wars. We then asked practitioners— policymakers with experience in the mediation and implementation of peace agreements—to comment on the memos in order to maximize their policy relevance. After the memos were revised to address practitioner concerns, the project then commissioned case studies of every peace agreement in civil xi

S-FMrev

11/25/09

xii

11:13 AM

Page xii

PREFACE

wars between 1980 and 1997 where international actors were given prominent roles in implementation. Case authors were asked to respond to questions and issues raised by the experts’ memos. Drafts of the cases were then critiqued at a third workshop at Stanford in late 1998. Upon receipt of the case studies, we asked our experts to develop their memos into papers taking into account, but not limiting themselves to, the evidence supplied by the case authors. After papers were drafted, we once again submitted them to the scrutiny of policymakers and case authors at a fourth conference in New York in July 1999. A final meeting of the experts was held at Stanford in November 1999, where a set of common themes and arguments was further hashed out and the authors were asked to revise their papers. Stephen Stedman then drafted a set of tentative findings and policy recommendations, which he presented to groups of policymakers in New York, Washington, and Geneva from May through November 2000. Based on the feedback that he received, authors were asked to submit a final round of revisions. Implementation of such an ambitious research program depended on robust support from several foundations. Stephen Stedman developed the research program and design at CISAC through a grant from the U.S. Institute of Peace (USIP). What was originally supposed to be just one chapter in a book on managing internal conflicts became the book you have here, and we thank the USIP and David Smock for their patience and support. Major funding for the research was provided by the Ford Foundation, and we are grateful to Ford and its program officer, Christine Wing, for their willingness to support such a huge undertaking. The first conference held as part of the program was made possible by a generous gift from Reuben and Ingrid Hills through the Edward E. Hills Fund. We would like to thank the Rockefeller Brothers Fund for the use of its conference facility at Pocantico for the second conference. The third conference was supported by the Carnegie Corporation of New York in honor of its former president, David Hamburg. Partial funding for the project came from a John D. and Catherine T. MacArthur Foundation grant to the IPA for “Peacebuilding: Issues and Responses” and a William and Flora Hewlett Foundation grant to CISAC for the study of conflict management. The project directors would like to express their appreciation to the many policymakers, diplomats, and scholars who read draft manuscripts and research memos and offered substantive comments and advice: Aldo Ajello, Dame Margaret Anstee, Jean Arnault, Reginald Austin, Andrea Bartoli, Derek Boothby, Michael Brown, Rex Brynen, Abram Chayes, Alvaro de Soto, Janelle Diller, William Durch, Lynn Eden, James Fearon, Shepard Forman, Alexander George, Sir Marrack Goulding, David Hamburg, Jeffrey Herbst, Terry Karl, Odin Knudsen, Chetan Kumar, Andrew Mack, David Malone, Stephen Marks, Eric Morris, Stephen Morrison, Michael O’Hanlon, Stewart Patrick, William Perry, Frank Ruddy, Yogesh Saksena, James Schear, Chandra Sriram, Mark Taylor, Ben Valentino, Christine Wallich, Robert Wasserman,

S-FMrev

11/25/09

11:13 AM

Page xiii

PREFACE

xiii

and Teresa Whitfield. The directors would also like to thank members of the project who drafted papers that we were unable to publish because of space limitations: Abiodun Alao, Virginia Page Fortna, Karen Guttieri, João Honwana, Lise Howard, and Ameen Jan. Special thanks also go to former IPA president Olara Ottunu and former codirector of CISAC David Holloway, who gave the go-ahead for the project, and to David Malone, current IPA president, and Scott Sagan, current codirector of CISAC, who have actively supported the project through its long life. We would like to thank those at CISAC and the IPA who helped organize the project and its workshops and offered assistance in bringing the book to publication, especially Nichole Argo, Karen Ballentine, Analia Bond, Charles Cater, Lisa Saad, Jens Sedemund, Helen Sutton, Diana van Walsum, and Karin Wermester. We are grateful to our editor, Lesli Athanasoulis, for shepherding the book to conclusion. A final round of thanks go to our publisher, Lynne Rienner. Year after year Lynne publishes excellent books, and, as two of us can attest, her interest in and commitment to African studies have saved publishing in that field. For this project, we are indebted to her for putting out a quality product at an inexpensive price. Much is made within foundations about the need for research to have an impact on public debate. We can think of few better ways than to have a publisher willing to produce accessible books that are affordable for use in the classroom. — the Editors

S-FMrev

11/25/09

11:13 AM

Page xiv

S-1

11/24/09

4:15 PM

Page 1

1 Introduction STEPHEN JOHN STEDMAN

When antagonists in civil war sign a peace agreement, what can international actors do to prevent a recurrence of that war? This is a life-or-death question for millions of people. The two worst outbreaks of massive violence in the 1990s—Angola in 1993 and Rwanda in 1994—followed the failure of peace agreements to end those wars. In both cases, the death and destruction were staggering: an estimated 350,000 dead in Angola and 800,000 dead in Rwanda. War went on for eight years in Liberia and took 150,000 lives because multiple peace agreements failed to end the civil war there. In 2001, Angola found itself at war again after the failure of another peace accord. Wanton destruction and killing continued in the Congo despite a peace agreement reached in 1999. In all of these cases, international actors mediated the agreements and were given prominent roles in implementation. Why did they fail? What could they have done differently? Was implementation in these cases doomed by unworkable peace agreements? Was failure a question of unfulfilled mandates or mandates inappropriate to the task at hand? Or was it caused by the lack of an appropriate strategy and the unwillingness to anticipate violent challenges and craft an effective response? How did the failures differ from successes such as Namibia, El Salvador, or Mozambique? Was success the result of less challenging environments, or did international actors do things differently? Analysts of conflict resolution and civil war have paid scant attention to the short-term implementation of peace agreements. Because the negotiated settlement of civil wars was a relatively rare phenomenon in the Cold War era, studies in the 1980s focused primarily on the conditions and tools for getting parties in civil wars to sign agreements.1 Rather legalistically, scholars assumed that a contract between state and insurgent leaders would 1

S-1

11/24/09

2

4:15 PM

Page 2

ENDING CIVIL WARS

remain binding in the postagreement phase. There was also a tendency to conceive of conflict resolution in a linear fashion, where successful negotiation signaled an irreversible reduction in conflict. Successful cases in the 1980s—Zimbabwe, Namibia, and Nicaragua—reinforced these assumptions. Before long, however, several civil wars—Angola, Rwanda, and Liberia—defied the linear view of conflict resolution and brought attention to the difficulties of getting parties to live up to their commitments to peace.2 Far from being a time of conflict reduction, the period immediately after the signing of a peace agreement seemed fraught with risk, uncertainty, and vulnerability for the warring parties and civilians caught in between. Peace implementation is the process of carrying out a specific peace agreement. It focuses on the narrow, relatively short-term efforts (three months in the case of Zimbabwe, over five years in the case of Bosnia) to get warring parties to comply with their written commitments to peace. Success is measured in relation to the ending of violence and the conclusion of the war on a self-enforcing basis: when the outsiders leave, the former warring parties refrain from returning to war. The relevant evaluation criteria are much narrower than the basket of goods associated with longerterm peacebuilding (e.g., the amelioration of root causes of conflict, and the promotion of justice, positive peace, harmony, and reconciliation of enemies), but are broader than the accomplishment of specific, mandated tasks. This chapter reviews research on peace implementation, describes the research design of this project, and summarizes our findings. It argues that current research on peace implementation suffers from several weaknesses, including a tendency to treat all civil wars as equally difficult to end, the prescription of open-ended, underspecified strategies for implementing peace agreements, a lack of attention to which cases get the appropriate resources to succeed, and a failure to prioritize among various tasks of peace implementation. To overcome these weaknesses, this project used an implementation perspective to study successes and failures of agreements to end civil wars. Such a perspective focuses on the attempts of international actors to make warring parties comply with their written and verbal commitments to peace, and emphasizes the interaction between implementation environment, strategies, resources, and incentives. To carry out the research, the project sought to combine the benefits of universal coverage of cases of international implementation of civil wars from 1980 to 1997 with the indepth understanding of process and causal relationships that one can glean from case studies. The project found that cases of peace implementation differ in two important respects—difficulty of the implementation environment and the willingness of states to provide resources and risk troops—and that these differences are predictable before implementation begins. Three factors are

S-1

11/24/09

4:15 PM

Page 3

INTRODUCTION

3

most commonly associated with a difficult environment: spoilers—leaders or factions hostile to a peace agreement and willing to use violence to undermine it; neighboring states that are hostile to the agreement; and spoils—valuable, easily tradeable commodities. For peace to prevail in difficult implementation environments, international implementers must provide greater amounts of financial resources and more peacekeepers, and pursue more coercive strategies than those associated with traditional peacekeeping—confidence-building, monitoring, and verification. Trouble arises, however, when we turn from the world of best practice to the world of political feasibility. The willingness of a state to invest blood and treasure to implement a peace agreement in any civil war depends on a prior judgment that making peace in that war is important to its vital security interests. Since the ability to coerce spoilers requires superior lethal capabilities, it is not just any state that must judge a civil war in its vital security interest, but a major or regional power. All too often in the 1990s international and regional organizations were sent to implement peace agreements in extremely challenging environments where no major state possessed a security interest. When implementers were challenged, the missions failed, usually with catastrophic consequences. The project also evaluated the contribution of various subgoals or tasks to overall implementation success. Peace agreements and the mandates of implementers usually contain myriad commitments ranging from protection of human rights to repatriation of refugees, from holding elections and creating democratic accountability to demobilizing and disarming armies. Implementers, however, face important resource constraints and must judge what tasks are essential to overall implementation success and what tasks are peripheral or even counterproductive to success. The project found that priority should be given to the demobilization of soldiers and the demilitarization of politics; that is, the transformation of soldiers into civilians and warring armies into political parties. The achievement of important normative goals such as protection of human rights and creation of accountability and democracy depend on these transformations. And more to the point, in the absence of such transformations, civil wars cannot be brought to an end. The project also identified two relatively low-cost tasks that, while not crucial to short-term implementation success, had important potential longer-term payoffs: (1) the provision of civilian security through police and judicial reform, and (2) local capacity-building for human rights and reconciliation.

Current Explanations of Success and Failure The first studies of peace implementation in the 1990s tended to an undifferentiated view of civil wars: El Salvador was Angola was Northern

S-1

11/24/09

4

4:15 PM

Page 4

ENDING CIVIL WARS

Ireland was Rwanda. The challenges of making peace were generic—overcoming the security dilemma and related commitment problems, building trust and confidence among adversaries, and problem solving to address the security needs of the warring parties. The most potent weapons for overcoming those challenges were provided by outsiders in the form of attention, resources, and guarantees. The problem for this research is that given an undifferentiated treatment of the problem, prescribed solutions tended to be open-ended. That is, more resources, more attention, and stronger guarantees are always suggested, with a resulting danger of tautology: if international actors are willing to do all it takes to make peace, then peace will be made. Hampson and Nurturing Peace Fen Osler Hampson’s Nurturing Peace was the first book-length treatment of why some peace agreements fail and others succeed at ending civil wars.3 He puts forward four possible answers to explain success: the extent of international nurturance of the peace process; the ripeness of the conflict as represented by the intrinsic desire of the parties to make peace; systemic/regional power balances that favor peace; and the quality of the peace agreement itself, in particular whether the agreement includes powersharing arrangements. To assess these broad answers, Hampson carried out a controlled comparison of five cases: two failures, Cyprus, and Angola 1992–1994; two successes, Namibia and El Salvador; and one partial success, Cambodia. Based on his research, Hampson concludes that the outcome of a peace agreement is “linked to the quality and level of support given by third parties to the peace process, especially during implementation of the agreement.”4 Hampson also finds evidence for the importance of what he calls a supportive regional and international environment. As he summarizes, “If key regional actors or outside great powers are hostile to the peace process, third parties will find themselves in an uphill battle in their efforts to make a settlement succeed. At a minimum, peaceful intervention by third parties in a civil conflict requires the support of a country’s neighbors and outside great powers that are involved directly or indirectly in the conflict.”5 Indeed, the role of outside actors is so dominant in Hampson’s account that what he begins by treating as a domestic variable—the ripeness of the conflict—is by the end a product of international nurturing and resources.6 And although Hampson insists that the poor quality of a peace agreement can be a source of implementation failure, he concludes that “all peace agreements are prone to failure regardless of how well structured and designed they appear at first blush.”7 In his analysis, well-designed agreements are as much in need of international nurturing as badly designed agreements.

S-1

11/24/09

4:15 PM

Page 5

INTRODUCTION

5

In Hampson’s account the civil wars themselves are mostly undifferentiated. With the exception of regional environment, Hampson fails to consider that civil wars may vary in ways that affect implementation success; for example, in the number of warring parties, the war aims of the parties, balance of military power, size of the country, numbers of combatants, level of death and destruction, and residual state and economic capability. Indeed, there is good reason to suspect that two of the variables that he does explore—a well-designed agreement and intrinsic desire to make peace—may be the product of some of the basic variables listed above. Without more differentiation on the conflict side, Hampson cannot provide direction on what amount of resources should be devoted to a given case to produce success. With regard to the actual provision of international attention, Hampson fails to ask why some cases get more attention than others, and why international actors provide resources to one case as opposed to another. Finally, he does not offer specific advice about how international actors should spend their resources to produce success. Hampson discusses activities that international actors may pursue, such as demobilization of soldiers, holding elections, mediating disagreements, and protecting human rights, but provides no sense of which ones may be more important than others in determining success and therefore warranting more resources.8 Hampson’s book was a pioneering effort to call attention to a major underresearched problem. Before his book, much greater attention had been paid to the mediation of civil wars, where judgments of success ended with the signing of a peace agreement. Hampson went further by showing that left untended, peace agreements would not implement themselves, and that there were important successful examples from which policymakers could learn. And given his methodology of structured, focused comparison, Hampson did not strive to test theory, but to develop theory by suggesting important relationships that needed further investigation. Walter and Security Guarantees A clear answer to the question of what international actors do to make peace implementation succeed came from Barbara Walter. According to her, the major problem for parties in civil wars is the inability to credibly commit to disarmament and demobilization of soldiers. Outside actors, by their willingness to enforce an agreement, allow local parties to overcome their commitment problems; in Walter’s words, outsiders “guarantee that groups will be protected, terms will be fulfilled, and promises will be kept (or at least they can ensure that groups will survive until a new government and a new national military is formed).”9 For Walter the key variable to explain successful implementation of a peace agreement is a third-party security guarantee defined as “any implicit or explicit promise given by an

S-1

11/24/09

6

4:15 PM

Page 6

ENDING CIVIL WARS

outside power to protect adversaries during the treaty implementation period.”10 How do we know a third-party security guarantee when we see one? Here Walter is much less helpful. She divides guarantee into three levels: weak, moderate, and strong.11 A weak guarantee “involves a formal promise to intervene should the treaty break down.” Such a promise has to be “widely known and [cannot] be recanted without negative reputational effects.” A moderate guarantee requires “the deployment of at least five hundred soldiers prior to the implementation period.” Walter argues that such a deployment is evidence of “a credible deterrent given the visibility of the endorsement.” A strong guarantee consists of the deployment of “massive ground forces (at least ten thousand) to the beleaguered country.” Strong guarantees provide “unambiguous and indisputable demonstration of intent.” Walter argues that successful negotiations to civil wars should “vary directly with the strength of outside security guarantees.” There are three reasons why this is less helpful. First, numbers of troops do not imply strategic intent, which is critical to the concept of guarantee. For a security guarantee to do the heavy lifting in explaining implementation success, implementers must possess an explicit strategy to use force to deter violations, compel compliance, or protect would-be peacemakers. In the cases in this project, for example, there are missions where third parties do just that: the Implementation Force (IFOR) and Stabilization Force (SFOR) in Bosnia, for instance, explicitly pursued a strategy of deterrence. Indian forces in Sri Lanka and Syrian troops in Lebanon used a strategy of compellence to make one side in the war comply with an agreement. On the other hand, there are plenty of military deployments by third parties that eschew protection: the United Nations in Angola, Mozambique, Rwanda, Namibia, Guatemala, and El Salvador pursued a strategy of confidence-building based on monitoring and verification of compliance. Second, numbers of troops do not imply commitment, a point made inadvertently by a quote that Walter provides to argue her case. Walter contends that deploying at least 500 troops provides a credible, though moderate, security guarantee, because early withdrawal “is viewed as potentially costly to the guarantor since it could severely damage ‘credibility with friends and allies,’ as President Clinton himself admitted when pressured to withdraw from Somalia.”12 Somalia, of course, is a telling case: the United States and the United Nations had nearly 25,000 troops on the ground when the United States unilaterally announced a six-month disengagement after losing eighteen soldiers in Mogadishu. But there are others just as telling: the United States withdrawing marines from Beirut in 1984; India withdrawing troops from Sri Lanka in 1990. The point is simple: numbers and the simple deployment of troops do not imply credibility of commitment.

S-1

11/24/09

4:15 PM

Page 7

INTRODUCTION

7

Walter does not examine cases where large numbers of troops fail to provide a credible commitment, but she does discuss how small numbers of troops can provide an adequate, credible security guarantee. By asserting that small numbers of troops can be used in a “tripwire” strategy, she implies that small numbers are a credible commitment, if it is understood that any major violation of an agreement will prompt the arrival of adequate numbers of troops to be effective. This, of course, leads us away from numbers as proxies for commitment and it requires that we ascertain the strategic intent of the implementer. But this Walter does not do. She asserts, for example, in reference to the case of successful negotiation and implementation of the Lancaster House agreement that ended the Zimbabwean civil war in 1980, that the war ended because Britain’s deployment of troops—1,319 Commonwealth soldiers to a country fifty times the size of Kosovo, with guerrilla forces and Rhodesian troops and auxiliaries equaling close to 170,000 soldiers—was a tripwire that symbolized Britain’s willingness to use force if necessary to punish violations of the agreement.13 Walter provides no evidence that this was the strategy of the British in Zimbabwe. Since strategy is the lynchpin holding together the otherwise contradictory claims of small numbers and credible enforcement, the veracity of the claim in this case resides in evidence of intent and plan. But the most comprehensive study of the Commonwealth Monitoring team (the name itself says much about its strategy) shows that the force would only help to “facilitate the agreement that the parties had voluntarily agreed to, not to seek to coerce them.” 14 In the cease-fire agreement itself, the British declared that it is “impossible for any external authority or force to guarantee that a cease-fire will be effective. Only the parties themselves can ensure this.”15 This was in opposition to the demands of the guerrillas that the force be strong enough to have a credible deterrent effect.16 Third, numbers as an indicator of credible security guarantees are misleading for the basic reason that civil wars differ in terms of the challenges that they pose for third-party implementers. Numbers of warring parties and soldiers, size of country, a hostile regional environment—all of these variables should influence how many troops it takes to credibly provide security guarantees in a peace agreement. For example, it is at least plausible that a security guarantee in Guatemala, where the rebels numbered under 2,000 and were essentially defeated, would take fewer troops than such a guarantee in Bosnia, where there were three warring parties, plenty of paramilitaries, and more than 150,000 men-at-arms. This helps to explain what would be an otherwise anomalous outcome—that a couple hundred lightly armed UN peacekeepers in Guatemala succeeded in bringing a war to a self-sustaining close, while in Bosnia 60,000 heavily armed soldiers from the North Atlantic Treaty Organization (NATO) could not. To be fair, Walter does hint that an important difference in civil wars

S-1

11/24/09

8

4:15 PM

Page 8

ENDING CIVIL WARS

may have implications for provision of security. But here she doesn’t focus on variables associated with the war, such as intensity, number of soldiers, number of warring parties, or size of country, but suggests that terms of the peace agreement matter. The more power sharing built into the agreement, Walter argues, the less the international commitment need be.17 But this seems doubtful for at least two reasons. First, empirical evidence presents contrary cases: two of the agreements in this study that are the most explicit in their commitment to power sharing, the Dayton Agreement on Bosnia and the Arusha Agreement on Rwanda, were among the more challenging implementation episodes that the international community faced in the 1990s. Second, there may be a selection problem that works against Walter’s claim: it may be that mediators and warring parties choose explicit power-sharing arrangements in precisely those civil wars that are the hardest to settle. But again note how we are back to the need to differentiate civil wars far more than Walter has done. Walter furthers the debate about success and failure of implementation by adding some precision to the word guarantee, a word that has been utterly devalued, yet frequently asserted to have great explanatory power. As Steven Ratner, an international legal scholar who helped mediate Cambodia’s peace accords, asserts, “today the term stands for a promise by a group of states that, if observed uniformly, would entail a certain outcome, coupled with pledges to respond, if only through consultations, to a violation of the agreement. It literally guarantees nothing, but has political significance and some legal content.”18 Ratner proceeds to discuss what actions can fall under the rubric of guarantee: observing and verifying compliance with agreements, building confidence among the warring parties, and the use of force to compel compliance. Needless to say, these are very different tasks, with different resource requirements and different underlying theories of conflict resolution behind them. Walter provides an invaluable service by insisting that we use the term to mean the pledge to use force to protect, deter, or compel. But as Ratner observes with regard to using force to compel compliance, “this type of strict guarantee will remain rare.”19 Even if Walter is correct, we still need to know which cases are likely to get the kind of guarantees that she recommends. Stedman and Rothchild: Barriers to Implementation Writing at the time of Walter and Hampson, Donald Rothchild and I argued that six recurrent problems plague the implementation of peace agreements: vague and expedient peace agreements, the lack of coordination between mediators and implementers of agreements, the lack of coordination among implementing agencies, the incomplete fulfillment of mandated tasks, the short time horizon and limited commitment of implementers, and the pres-

S-1

11/24/09

4:15 PM

Page 9

INTRODUCTION

9

ence of spoilers—leaders or factions who use violence to undermine implementation.20 We felt the need to call attention to organizational impediments to successful implementation of peace agreements: that the performance of mediators and implementers affected the quality and outcome of civil war termination. In hindsight, our argument seems naïve for several reasons. We never specified whether these are problems that face every case of implementation or whether they are problems that vary among cases and therefore help to explain successful or failed implementation. Nor did we probe into what accounts for these problems; that is, whether the barriers we discussed were the result of other more important factors. Finally, sharing a symptom common to Hampson and Walter, we assumed that identification of a problem leads to a remedy. These criticisms apply to all of the organizational barriers we describe, but a discussion of one will suffice to illustrate my point: the relationship between vague, expedient peace agreements and implementation failure. Moreover, it is a factor cited by Hampson and Walter in slightly different ways. Our initial formulation presented the character of a peace agreement as a variable, suggesting that some peace agreements are clear, comprehensive, and largely judicious. Our implicit argument was that when peace agreements leave important matters undecided, then the warring parties are much more likely to hedge their bets, take advantage when provisions of an agreement are not specified, and interpret ambiguous terms in ways that benefit them during implementation. The operational consequence for implementers is to require that they revert to being mediators, blurring the line between putting policy into practice and making policy through ongoing negotiation and interpretation. Two questions arise: Do peace agreements vary in their clarity, comprehensiveness, and judiciousness? Does it matter in predictable ways if they do? Almost every peace agreement has important silences, some points of agreement that are described in banal language open to multiple interpretations and, in worst cases, contradictory dictates. Correspondingly, parties’ commitment to certain provisions may be tactical, temporary, or conditional. In this sense, there are few if any cases where an implementer does not slip into the role of mediator or, if possessed with requisite capacity and legitimacy, adjudicator. This does not imply that all agreements are alike, however. What varies is not the vagueness and expediency contained in agreements, but rather the baseline amount of problem solving and detail on which implementers can build. An implementer who must work with an agreement in which every provision is contested will have a much more difficult challenge than one who must implement an agreement that is the culmination of years of problem solving, relationship-building, and inclusion.

S-1

11/24/09

10

4:15 PM

Page 10

ENDING CIVIL WARS

But why do some peace agreements contain vagaries, silences, and contradictions? First, given the complexity of making peace in civil war, no document is likely to address all of the concerns of the parties; issues that become important during implementation may simply not be anticipated during mediation. Second, civil wars involve multiple contentious issues and mediators may make progress on some only to find that disagreements still exist. Mediators may then bet that an incomplete agreement will initiate a process of trust-building that will enable the parties to tackle the toughest issues at a later date. Third, peace agreements are quintessential coalition-based documents that stem in part from the demands of constituencies beyond the warring parties themselves. To garner the support of international patrons of the warring parties or politicians in Western capitals, or nongovernmental organizations, peace agreements may need to include provisions that hinder implementation. Fourth, mediators may have stronger incentives to preside over the signing of a peace agreement than to preside over the signing of a comprehensive agreement. When war is raging, and there is pressure to stop it, the urgency to reach a peace agreement may take precedence over the quality of its provisions. But how much weight should a bad peace agreement carry in explanations of implementation failure? Certainly when implementers fail, a common target is the peace agreement and the mediator who produced it. The tendency to scapegoat the peace agreement as the cause of implementation failure is understandable, and there are cases where fault seems one-sided. A well-known distinction in public policy differentiates policy failure from implementation failure and cautions against blaming implementers for the failures of a policy (or in our case a peace agreement) that was fundamentally flawed at its conception.21 Of course, turnabout is fair play: when a policy fails, there is a maxim that deflects scrutiny from its architects: “good policy badly executed becomes bad policy.”22 That the maxim in this case was self-serving—its author is a prominent mediator—does not detract from its fundamental insight. When peace fails, blaming the mediation and the peace agreement it produced, while understandable, is often misguided for at least two reasons. First, even if it is possible to create “best practice” advice for mediators to seek clear, judicious peace agreements, such advice is often untenable for political reasons. As several scholars of implementation insist, “the frequent advice that policies should be clear seems to assume that policy makers can arbitrarily choose the level of clarity of a policy, that policies are ambiguous because of some form of inadequacy in policy-making. . . . In fact policies are negotiated in a way that makes the level of clarity no more accessible to arbitrary choice than other vital parts of policy.”23 A second reason to distrust blaming either side of the mediatorimplementer divide for failure focuses on the interaction between the two. Often mediators preside over agreements that look flimsy or incomplete,

S-1

11/24/09

4:15 PM

Page 11

INTRODUCTION

11

not because of a desire to evade—although ambiguity is sometimes a deliberate tactic with constructive effect—but because often the alternative to a weak agreement is no agreement. On its face there is nothing wrong with an incomplete, expedient agreement, as long as it is mediated with an understanding that a flimsy agreement signed under duress will likely demand much more forceful implementation than if the agreement is the product of lengthy, gradual relationship-building among adversaries. Expedient peace agreements only become problematic when coercive strategies are unavailable to the implementer. Rothchild and I may have missed the interaction between the quality of a peace agreement and the effort that is needed to implement it because we assumed that most implementers have truncated strategy sets and coercion is not available to them. In a world where implementers are limited to confidence-building, the quality of any resulting peace agreement looms large. Stedman and Spoilers My experience with UN peacekeeping as an observer in Angola in 1992, my involvement in several study groups of UN peacekeeping, and my area expertise and research in Southern Africa made me uncomfortable with both the performance of the United Nations in civil wars and the early scholarship on peace implementation.24 Much of my discomfort stemmed from three observations. First, where international actors had succeeded in implementing peace in Southern Africa, none of them had provided security guarantees of the kind that Barbara Walter described. Outsiders had made peace in Namibia, Zimbabwe, and Mozambique, and indeed insiders had made peace in Zimbabwe in 1987 and South Africa in 1994, without the provision of security guarantees. The United Nations in Namibia and Mozambique had followed a strategy of confidence-building derived from traditional peacekeeping. The British had used a strategy akin to the game of chicken in Zimbabwe, and far from reducing security fears, they played upon them to emphasize the need for compliance with the peace agreement.25 It struck me then, as it does now, that the British were far more lucky than skillful, and I vowed that I would never use their success as a lesson for others. The second observation was that confidence-building was absolutely ineffective if one or more sides decided to return to war and block the implementation of a peace agreement. I saw firsthand the absolute lack of coherent international response to repeated violations of the Angolan peace agreement in 1992, culminating in disaster when Jonas Savimbi returned to war. My observation was reinforced by following what was happening in Cambodia at about the same time, as the UN Transitional Authority in Cambodia (UNTAC) scrambled to manage defections by the Khmer Rouge from the Paris Peace Accords. And it came home horrifically as the United

S-1

11/24/09

12

4:15 PM

Page 12

ENDING CIVIL WARS

Nations stood paralyzed when a determined faction in Rwanda, including some within the government who had signed the Arusha Peace Agreement, committed genocide to destroy the implementation of peace. My third observation stemmed from an unhappiness with the depiction of warring parties contained in explanations of implementation failure that emphasized problems of credible commitment and insecurity. It seemed to me that Walter’s analysis asked implementers to take on faith that warring parties were solely motivated by insecurity and fears of future vulnerability. Such a depiction seriously underrepresented a basic recurrent problem in civil war negotiations: strategic deception by parties who sought to use an agreement and its implementation as a source of advantage to winning the war. These observations led me to want more differentiation on the conflict side: that civil wars differed by the presence or absence of spoilers who sought to use violence to undermine the agreement. The presence of spoilers then required more differentiation on the strategy side: confidencebuilding would not be enough to stop wreckers of peace agreements. In an article for International Security, I sought to make three contributions to our understanding of peace implementation.26 First, an emphasis on spoilers insists that civil war negotiations and implementation are rife with strategic deception—parties who signed peace agreements, not for imagined cooperative gains, but for opportunities to prevail. Second, even in those cases where the parties to an agreement sincerely want to implement a mutually beneficial peace, they may have to contend with spoilers who are not a part of the agreement and who attempt to destroy the incipient coalition for peace. Third, implementers face important uncertainties— what I referred to as the “fog of peace”—on such dimensions as the sincerity of the warring parties and their command and control. My research tried to add to the differentiation among cases of civil war termination by putting forth a typology of spoilers based on their position in the peace process, number of spoilers, their intent, and whether the locus of spoiling behavior lies with the leader or followers of the party. Of crucial importance is the motivation and intent of the spoiler. Does the spoiler have limited demands that can be met through inducements? Is the spoiler a total spoiler, who sees power as indivisible, holds immutable preferences, and will take strategic advantage of any inducement? Or is the spoiler greedy, that is, possessed of goals that expand based on the prospect of inducement? The key for international implementers is to diagnose spoiler type and then choose an effective strategy for managing the spoiler. Whereas limited or greedy spoilers can be addressed through inducements or socialization, total spoilers can only be managed with a coercive strategy. Several aspects of the spoiler concept have been criticized. First, it is said that since it is impossible to identify ex ante what type of spoiler one

S-1

11/24/09

4:15 PM

Page 13

INTRODUCTION

13

faces, then the concept has little predictive or prescriptive value.27 This is true, but it is not clear what harm this does to the theory. Indeed, if I understand the criticism correctly, it is that the spoiler concept doesn’t eliminate the possibility of strategic deception in peace negotiations. After all, if there were a magic indicator of type, then no party could act to deceive; they would all wear Ls, Ts, or Gs on their chests. It is impossible to know any type that is based on intention: we are not mind readers. Our attempt to assess the intent and motivation of parties during peace implementation must rest on intelligence, not indicators. The spoiler concept argues that implementers of peace agreements should pursue their tasks with an appreciation of uncertainty and incomplete information. This seems to me superior to implementers assuming that all parties who sign peace agreements do so in good faith or are equally trustworthy. This is not to say that the concept cannot be improved to warn policymakers that certain leaders or factions deserve special scrutiny. When parties to a peace agreement have a particularly egregious past, for example, such as the Revolutionary United Front (RUF) in Sierra Leone or the Khmer Rouge in Cambodia, then implementers should be particularly vigilant to greater likelihood of spoiler behavior. Implementers should also be aware that rebel leaders the likes of Foday Sankoh, Pol Pot, or Jonas Savimbi, who spend decades in the bush surrounded by sycophants who assure these individuals that the world revolves around them, may not be able to conceive of power as divisible. But since the goal at the beginning of implementation is to keep an open mind to the possibility of sincere signatories to an agreement, even where there is good reason to doubt that sincerity, implementers need to try to move forward by including the party. They should then judge compliance, and assess motives for noncompliance, best accomplished through intelligence—informants, surveillance, and reading of documents. Second, it is argued that the definition of total spoiler type as immutable is problematic, given the propensity of so-called total spoilers to change. Marie-Joëlle Zahar and John Darby argue that there are examples of factions who are labeled total spoilers at one time, such as the Palestine Liberation Organization (PLO) in the 1970s and the Irish Republican Army (IRA), who then years later are willing to negotiate a settlement. 28 Accordingly, they call into question whether there is such a thing as a total spoiler who holds immutable preferences for total power. The mere fact that some analysts describe a faction as a total spoiler at a given time seems irrelevant for two reasons. First, spoilers can only be defined in relationship to a given peace agreement. In the absence of an agreement, the concept of spoiler should not apply. Second, a history of abhorrent behavior and totalistic rhetoric does not in and of itself create a total spoiler. Most parties in civil wars engage in atrocities and many use the rhetoric of total war; the early dismissal of a faction as totalistic is usually done for political reasons to delegitimize it rather than to provide any

S-1

11/24/09

14

4:15 PM

Page 14

ENDING CIVIL WARS

objective sense of whether there is a compromise to which it might agree. Finally, there is a route whereby a party that is a total spoiler can change. When the locus of the spoiler problem lies with a single leader who, for whatever reason, sees a war in an all-or-nothing fashion, there is the potential for the party to change type if there is a change of leader. In a third criticism of the spoiler concept, it can be argued that too much attention to spoiler motivation and intent detracts from the much more important considerations of capability and opportunity to spoil. This could be an important refinement of the concept, as it calls attention to aspects of the conflict environment that make spoilers a greater threat in some cases than in others. Where spoilers have access to easily tradable valuable commodities, and where they can rely on the support of neighboring countries that oppose peace, they are a more likely and greater threat to peace. Peceny and Stanley: Learning from Central America Other researchers working from an area studies perspective have expressed dissatisfaction with undifferentiated analyses of civil wars. Mark Peceny and William Stanley argue that successful termination of civil wars in El Salvador and Guatemala occurred without the forceful security guarantees of the type that Barbara Walter describes.29 They also point out that at least several failed implementation efforts—Angola, Somalia, and Liberia— involved far greater commitment of troops than the successes of El Salvador and Guatemala. Peceny and Stanley put forward a norms-based explanation for the Central American successes. They assert that several variables distinguish the successful Central American cases from failures elsewhere: the region’s history of liberal norms, its integration into the global economy, and the presence of surrounding states that are democratic. These variables served to make these civil wars amenable to what Peceny and Stanley call an implementation strategy of liberal, social reconstruction. Although Peceny and Stanley do not treat civil wars in an undifferentiated manner, they jump to a regional-specific explanation for success and ignore the possibility that there may be more basic, general variables— number of parties, size of armies, war aims of the parties, interest of great powers—that explain the relative ease of civil war termination in Central America. They acknowledge that this poses a problem for them, for instance, in explaining why Mozambique and Namibia, in a region of the world sorely lacking in democratic, liberal norms, marginal to the world economy, and in a tough neighborhood of civil wars, authoritarian regimes, and fledgling democratic polities, successfully ended their civil wars.30 Or why Bosnia, in a region of the world with strong commitments to liberal democracy and economics, should be such a tough case for peace implementation.

S-1

11/24/09

4:15 PM

Page 15

INTRODUCTION

15

Peceny and Stanley’s discussion of strategy also suffers from important weaknesses. The first concerns whether the liberal transformative aspect of UN strategy really produced success in the cases they examined. The strategy pursued by the United Nations in El Salvador and Guatemala was a hybrid. It combined precepts derived from traditional peacekeeping such as the need for confidence-building and verification to address security problems with a transformative strategy aimed at reforming state institutions, changing political rules of the game, and redistributing power. Although the intensity and stubbornness of the security dilemma were less evident in Central America than elsewhere, they were not absent. The UN’s confidence-building strategy was successful in the Central American cases: rebels demobilized and state armies downsized. It is at least an open question, however, how much political and social transformation has taken place in these countries, and an even greater question how much the transformation can be attributed to the strategy and actions of the international implementers. This is a common problem with evaluating multifaceted strategies, where one seeks to determine what aspect of a strategy produced what results. It is possible, at least, that what was crucial for bringing the wars to a close in Central America was confidence-building, monitoring, and verification, and not the transformative aspect of the strategy. Second, if Peceny and Stanley are correct that there is something special about the Central American cases that leads the liberal transformation strategy to be successful, then we should be skeptical about their robust endorsement of the potential of the strategy for ending civil wars elsewhere. A more appropriate inference would be to assert that absent facilitating conditions like those present in the successful cases, the strategy will fail. Still, Peceny and Stanley provide insight into why Central America’s regional environment was so conducive to successful peacemaking, and how important factors within a society may diminish the urgency or direness of the security and commitment problems that Walter puts forward. In that regard the work is similar to an exciting branch of research undertaken by Elizabeth Wood, who argues that economic interdependence between or among warring parties provides a powerful incentive for them to cooperate in the face of spoilers.31 Wood contrasts South Africa, where the welldeveloped industrial economy created incentives for the parties to work together to implement their peace agreement, with Angola, where the lack of economic development and the economy’s reliance on oil and diamond exports forged a winner-take-all, zero-sum approach to the conflict. Doyle and Enhanced Consent and Multidimensionality Peceny and Stanley’s recommendation of social transformation as an implementation strategy has much in common with the work of Michael

S-1

11/24/09

16

4:15 PM

Page 16

ENDING CIVIL WARS

Doyle in the 1990s on peace implementation. Based on case research on Cambodia and comparative study of the UN there and in El Salvador, Doyle argues that outsiders in peace agreements can choose from a range of strategies to elicit compliance.32 On the one hand, traditional peacekeeping approaches stress monitoring and verification of compliance. At the other extreme is peace enforcement, where the outsider uses force to compel compliance. Doyle criticizes both approaches: traditional verification is often not enough when consent of the warring parties is weak, tactical, or eroding. Peace enforcement, however, has proven extremely risky and costly to troop contributors, who seem unwilling to go to war to make peace. As an alternative to both approaches, Doyle emphasizes “strategies of enhanced consent” in the implementation of peace agreements. Such strategies include creating arrangements such as “Friends of the SecretaryGeneral,” establishing “ad hoc semi-sovereign bodies composed of representatives of the warring parties,” and pursuing a transformative strategy akin to that described by Peceny and Stanley, appealing beyond elites to the masses who want to make peace and need assistance in changing political and social institutions. These strategies, Doyle suggests, can make a crucial difference in winning and keeping the consent of the warring parties. Doyle’s analysis raises several issues. First, are these really strategies, or are they tools for enhancing either noncoercive or coercive approaches? For example, Friends arrangements are formal coalitions. Such coalitions can be supportive of a noncoercive or coercive strategy. The arrangements are useful for implementers because they can provide consensus on how to respond to noncompliance and leverage. But there is nothing inherent in the arrangement that ensures it will do so. And there is nothing inherent in the arrangement that broadens the possible actions of the implementer. Consensus rests on the calculations of the member states that belong: for a member of a Friends group to use its leverage with a former client or to countenance the use of coercion, it must perceive that it is in its interests to do so. For a consensus to emerge, all must perceive a course of action as in their interest. To advocate the creation of a Friends arrangement is simply to say it is better to implement with a knowledge of what interested states are willing to support: good advice, but advice that does not imply strategic direction. Moreover, it is not at all clear that a strategy of a Friends group is a choice available in all instances; a Friends group implies that some powerful states have interests in a specific peace process and are willing to invest in it, and therefore we should not imagine that all civil wars are equally likely to have Friends. Nor is it is clear from Doyle’s analysis what makes the mechanisms themselves effective in enhancing compliance. For instance, from Doyle’s study of Cambodia and El Salvador, he argues that a Friends arrangement and an ad hoc semisovereign body can make a difference between failure and mixed performance. But one can argue that in the case of Angola in

S-1

11/24/09

4:15 PM

Page 17

INTRODUCTION

17

1992, the troika of Portugal, Russia, and the United States was a Friends arrangement, and that the joint monitoring group set up between the warring parties and the Friends was an ad hoc semisovereign representative body. One can go further and argue that the election of September 1992 in Angola was extremely popular among the masses of people who wanted peace. Why did the mechanisms work in Cambodia and El Salvador and fail in Angola? And to what extent was success a function of the mechanisms or the vastly greater sums of resources that were provided to UNTAC and the UN Observer Mission in El Salvador (ONUSAL)? In his later work on peace implementation, Doyle acknowledged that resources and conflict differentiation played a role in success, and sought to include them in his analysis. Transformation, on the other hand, does imply content and strategic direction. It says that when compliance of the parties is in jeopardy and when their consent to the mission is weak, implementers should try to improve the lives and institutions that govern the mass of people in a society. By doing so, they can compensate for the weak consent of the warring parties. But the same questions that we asked of Walter, and of Peceny and Stanley, apply here. How do we measure whether implementers are really pursuing such a strategy? Are all civil wars amenable to the strategy? And since transformation can take place on many dimensions, which institutions and practices should receive priority? In some cases of civil war agreements (for example, El Salvador, Guatemala, and Cambodia), the parties agree that transformation of political institutions will be part of the settlement. If the outside implementer simply monitors and verifies the transformation efforts, then transformation is squarely within the realm of confidence-building. The importance of the outsider is to verify that one or more of the parties have held up their end of a bargain, end of story. But to the extent that the outside implementer insists on reform or transformation yielding substantive results, then one can speak of a transformation strategy above and beyond narrow compliance with meeting the commitments to reform institutions. But how to judge whether an implementer treats transformation as a problem subsumed within a strategy of confidence-building or as a more ambitious strategy that complements or competes with confidence-building? Where transformation is done alongside confidence-building, one returns to the difficult evaluation problem of inferring results from simultaneous strategies. Where it is more ambitious, one should look at the priorities placed on goals such as human rights, the freeness and fairness of elections, and rule of law. When choices need to be made between whether or not to push for substantial results, especially when pushing for results that may provoke one or more of the parties to withhold compliance on security issues, one can gauge the robustness of the commitment to transformation. Here the evidence is problematic for Doyle, for in the Cambodian case, Yasushi Akashi,

S-1

11/24/09

18

4:15 PM

Page 18

ENDING CIVIL WARS

the special representative of the Secretary-General, placed limits on human rights as a subgoal for fear of provoking the warring parties. Moreover, one of the major criticisms of Akashi was that he put higher priority on simply holding an election than on holding a quality election. In its aftermath, UNTAC overrode the results of the election to give the State of Cambodia far greater power and representation than was deserved by its share of the vote. In later work Doyle revisited the nexus between strategy, context, and resources.33 In work with Nicholas Sambanis, Doyle used a data set of all civil wars since 1945 to examine what factors increase the likelihood of war termination through negotiation. In their work, variables such as presence of a treaty, scale of death and displacement, war motivations other than identity, the duration of war, a small number of factions, and lack of natural resource dependence all create easier environments for implementation. Where the environment increases in difficulty, international actors must pursue a more robust strategy, what Doyle and Sambanis call multidimensional peacekeeping, to succeed. Doyle and Sambanis have produced to date the most sophisticated, quantitative treatment of the nexus between strategies, resources, and context. However, at least two important lacunae remain. The first is captured in the very name multidimensional strategy. Other than telling implementers to do a lot of things, multidimensionality does not provide guidance about which dimensions are most important and deserve more resources, and should take precedence over basic security issues in implementation. The second flaw is that it is not enough to diagnose that more difficult cases need greater resources. Here we return to the criticism of Hampson and Walter: we need to know which cases are likely to get the resources they need. Paris and the Perils of Liberal Peacebuilding While Doyle, and Peceny and Stanley extol the strategy of liberal transformation and argue that it has successfully brought several wars to an end, Roland Paris criticizes the same strategy for contributing to instability in war-torn societies.34 What Doyle, and Peceny and Stanley describe as a means of inculcating liberal, democratic norms and empowering masses as a means to earning elite compliance, Paris sees as exacerbating already intense divisions in such societies and failing to address root causes of conflict. Although Paris is not against democracy or liberal economies, he argues that implementers, as outsiders trying to impose these goods immediately at the end of a civil war and for only two to three years’ time, actually do more harm than good. Part of the discrepant evaluation of the liberal transformation strategy can be attributed to Paris’s criteria for judging the success of implementa-

S-1

11/24/09

4:15 PM

Page 19

INTRODUCTION

19

tion missions. He argues that only one UN operation in the 1990s— Namibia—has been successful in creating the conditions for a longenduring peace. By setting a high bar for success, Paris fails to discriminate between catastrophic failures—for example, Rwanda and Angola, where hundreds of thousands of people die when war returns—and flawed successes, such as El Salvador, Cambodia, and Mozambique, where war has ended but citizens face the mundane reality of life in crime- and povertyridden insecurity. By casting such a large critical net, Paris dismisses even the triumphs of international assistance in ending civil wars. Moreover, Paris bases his evaluation of failure on long-term criteria that cannot be judged in the short term. Implicit in his evaluation is certain knowledge that even where civil wars have ended, they will return. Why? Because they have not produced viable market democracies. But since the consolidation of market democracies is a long-term process, by definition any war-torn society that embarks on a transition to a market democracy will fall short of meeting standards appropriate for the long term. The second problematic aspect of Paris’s critique is his belief that the liberal international strategy of peacebuilding forms a coherent set of directives for international implementers of peace agreements. Paris attributes much operational content to mandates that proclaim the necessity of human rights, truth and accountability, democratic elections, and market economies. But an alternative interpretation of the repeated inclusion of these virtues into mandates can be that this is less of a coherent model that the UN seeks to impose on war-torn countries and more of an expression of the basket of virtues that the UN claims to stand for—a basket of virtues that has become more important as NGOs have strengthened their ability to lobby for those goods that they hold dear. Moreover, we need to ask of Paris the same question that we asked of Doyle, and Peceny and Stanley: When implementers actually do their work and implement peace agreements, how much stock do they put in achieving their transformative goals, and do they at any time prioritize their achievement over addressing the security needs of the warring parties? Both the proponents and the critics of the liberal transformative model need to be reminded that policies and mandates need not be guidelines for action, so much as expressions of faith, acknowledgment of virtue, and instruments of education. Moreover, they need to bear in mind that the United Nations is an organization of states, and not just a carrier of norms. Since its founding more than fifty-five years ago, the balance among the UN’s core constituency has changed: the number of international NGOs has grown tremendously and universal ideas, attitudes, and norms associated with global society have spread, with corresponding pressure for the UN to legitimate itself with reference to those ideas and norms. This frequently leads to the organization mouthing the principles of global society, while continuing to follow the dictates of states. In peace operations, the result is

S-1

11/24/09

20

4:15 PM

Page 20

ENDING CIVIL WARS

often a stated rhetorical commitment to democracy, human rights, justice, and development, with a bottom-line commitment to what the most powerful, interested states desire. But there is an aspect of Paris’s critique that can sting Doyle’s, and Peceny and Stanley’s laudatory appraisal of the transformative strategy: whatever the strategy may achieve in terms of human rights, reform of police and judiciary, and transition to electoral democracy, the liberal economic model that accompanies it may take away. If we recall that the power of the transformative strategy and multidimensionality is that it is said to empower the grassroots support for peace, then we need to know the effects of the liberal package of economics that often accompanies the strategy. Even Doyle, and Peceny and Stanley acknowledge that the economic policies that guide international implementation of peace agreements are often the weakest link, running at cross-purposes with the needs of war termination. Again, we are back to the problem of evaluating complex strategies, and attempting to sort out the different effects of their strands.

The Potential Contribution of an Implementation Perspective There is a general consensus within the literature on civil war termination that international attention and resources are necessary for successful implementation of peace agreements. Beyond that, however, there is disagreement about what outside actors actually do to assist implementation. Strategies are underspecified; myriad tasks are bunched together under general strategies without any hint about which are more important for successful war termination and hence deserving of more resources. There has been progress in indicating which implementation environments are likely to be more difficult than others and an understanding that the larger the degree of difficulty, the more international attention, resources, and coercion will be necessary for implementation success. But there is no theory about which cases will receive adequate attention or resources, and certainly no theory about which cases should receive adequate attention or resources. In short, our state of knowledge is ripe for a new approach that can shed light on the successes and failures of peace agreements and bridge some of the gaps in the current understanding of peace implementation. The broader literature on policy implementation is a natural place to look for such an approach. As a branch of policy studies devoted to understanding the ability of practitioners to take written commitments and put them into practice, and to attempt to realize the potential of political and government policies, implementation emphasizes the interaction between agency and structure. Vision provides a direction for change, coalitions put a dollar

S-1

11/24/09

4:15 PM

Page 21

INTRODUCTION

21

sign on the proposed change and put forward resources according to their interests, and then implementers face a reality on the ground that may hinder or promote achieving their vision. The literature on policy implementation writ large provides several insights. First, it insists that some implementation environments are more difficult than others; accordingly, successful implementation of policy should be, in part, a function of easier environments. Second, it posits that, all things being equal, the greater the amount of resources devoted to implementation of any given case, the greater the likelihood that implementation will succeed. Third, it notes that resources must have a source; there must be a coalition that is willing to provide the resources, and there must be fundamental incentive compatibility between the interests of coalition members and their commitment to implementation. Fourth, it observes that there are important interactions between environments and action—what is appropriate in one context may be inappropriate in another; and that strategies matter. A premium is therefore placed on specifying what in a particular context may make the difference between choosing one strategy or another, and on specifying how one’s proposed strategy will accomplish its goals. Finally, an implementation perspective calls attention to the fact that in any complex endeavor, to realize results from policy, implementers will be asked to accomplish myriad subgoals and tasks, and that overall success is usually not so much a function of accomplishing all subgoals, but rather prioritizing among them. How to translate the larger implementation perspective to the world of civil wars and their ending? First, it calls for a differentiated understanding of civil wars and the need to discover what makes for an easier or more difficult implementation environment. We therefore asked the authors of this book to pay close attention to the relationship between the implementation environment and success and failure, not only in individual cases, but also in larger judgments about strategies and tasks of implementation. Second, following Hampson and Walter, an implementation perspective posits that the greater the amount of international commitment involved, be it attention or security guarantees, the greater the likelihood that peace implementation will succeed. But this perspective also demands that we observe that not all civil wars receive equal attention or resources, and that we ask which conflicts attract more international involvement in peace implementation. Third, the implementation perspective calls attention to the interaction between environment, resources, and strategies. More difficult implementation environments will need greater international commitment and, in those cases, international actors will have to do things differently than in easier cases. Fourth, given that international actors are called upon to do so much in peace implementation, an implementation perspective demands that we evaluate the relative contributions of subgoals contained in implementation mandates to the successful ending of civil wars.

S-1

11/24/09

4:15 PM

Page 22

22

ENDING CIVIL WARS

Cases Studied This project sought to identify all instances between 1980 and 1997 where the warring parties in a civil war reached a peace agreement, and international actors were expected to play a large role in the implementation of the agreement. Table 1.1 lists all of the cases that the project commissioned.35 We defined civil wars as wars fought within internationally recognized boundaries to establish who will rule. The definition encompasses wars fought over control of a central government, wars fought to secede and create a new government, and wars fought over the balance of power between center and regions within countries. We used a standard measure of war involving battlefield deaths and the ability of each warring side to inflict a percentage of the casualties.36 Thus, we did not formally include Haiti from 1991 to 1994 in our sample, believing that it was not a civil war, but largescale, state violence against citizens.37 Since our focus was on the role of outside implementers, we did not include peace agreements that were largely self-implementing, such as South Africa in 1994, or Zimbabwe in 1987.38 And since it was difficult to chase current events, we did not formally include the cases of East Timor, Northern Ireland, Sierra Leone after 1998, Kosovo, or the Democratic Republic of Congo, all of which produced peace agreements where outsiders took the lead in implementation.39 Finally, we included two examples where large-scale intervention took place before international actors brokered an agreement: Liberia and Somalia. We felt that those cases could inform us about the difficulties and potentials of making peace in an ongoing war, where the would-be implementers are looked upon by some of the warring factions with hostility and mistrust. To define a peace agreement, we wanted a standard that would serve as a formal marker of commitment to war termination and that would allow us to focus on the ability of outsiders to get parties to comply with formal commitments. We therefore wanted something more specific than ceasefires, which are common in civil war and generally do not imply a commitment to the permanent cessation of the war, but we also did not want to set

Table 1.1

Cases Commissioned

Angola, 1992–1993 Angola, 1994–1998 Bosnia-Herzegovina, 1995–2000 Cambodia, 1991–1993 El Salvador, 1993–1995 Guatemala, 1992–1998 Lebanon, 1991–2000 Liberia, 1990–1999

Mozambique, 1992–1994 Namibia, 1989 Nicaragua, 1989–1991 Rwanda, 1993–1994 Sierra Leone, 1996–1998 Somalia, 1992–1993 Sri Lanka, 1987–1988 Zimbabwe, 1980

S-1

11/24/09

4:15 PM

Page 23

INTRODUCTION

23

too strict a standard by examining only those agreements that were the culmination of intense, protracted negotiation, and that considered every possible detail of war termination. We also did not limit ourselves to agreements where all of the warring parties signed the agreement. In the end we used the definition proposed by Peter Wallensteen and Margareta Sollenberg: “Peace agreements are arrangements entered into by warring parties to explicitly regulate or resolve their basic incompatibility.”40 We felt that it would be advantageous to have variance on the quality of the agreement and the quality of the process that produced the agreement. Since, as discussed above, much of the literature on peace implementation looks to the agreement and how it was produced as the source of success and failure, we wanted a range of agreements from the flimsy, expedient, and coerced to the detailed, well-designed product of inclusive problem solving. Thus at one end of the spectrum, we included Somalia, where under pressure from the United States and the United Nations, clan elders met in Addis Ababa, and in short order agreed to a set of commitments to end their war; Liberia, where the Nigerian-led Economic Community of West African States Cease-Fire Monitoring Group (ECOMOG) militarily intervened, then presided over a series of quickly negotiated and short-lived agreements; and Sri Lanka, where the Indian government intervened to implement an agreement that omitted one of the two major warring parties. At the other end of the spectrum is Guatemala, where a series of detailed agreements and commitments emerged from an inclusive process that represented most stakeholders in society, not just the warring parties, and included gradual commitments and phased trust- and confidence-building. We also wanted variation on the interveners. We therefore did not limit ourselves solely to the United Nations, but included regional organizations, such as NATO and ECOMOG, and also individual states, such as India in Sri Lanka, and Syria in Lebanon.

Evaluating Implementation Strategies Part 1 of this book focuses on evaluating strategies of peace implementation. Strategies of peace implementation are guiding approaches to achieving the goal of a self-enforcing peace. When embarking upon peace implementation, international actors must assess the motives of the parties who signed the peace agreement and their incentives and disincentives for complying. Implementers must anticipate challenges to implementation—either by signatories to, or factions excluded from, the agreement—and plan to overcome such challenges. Such a plan must realistically calculate the resources available to implementation. Part 1 has two goals. The first is normative: to recommend, based on

S-1

11/24/09

24

4:15 PM

Page 24

ENDING CIVIL WARS

the conditions of a particular case, what strategies outsiders should pursue. The second is explanatory: to understand where strategies come from. The former purpose stems from concerns of evaluation: When do strategies succeed and when do they fail? The latter purpose stems from two mysteries: Why are poor strategies chosen? And why, when the failure of a strategy becomes apparent, do so few implementers change course? In Chapter 2, George Downs and Stephen Stedman examine various issues that arise in the evaluation of international actors who implement peace agreements in civil wars and the strategies they pursue in trying to bring wars to an end. Given the many variables that can affect the success or failure of missions and the paucity of cases, systematic evaluation of peace implementation is extremely difficult. Several difficulties stand in the way of measuring success and failure, including endogeneity, noncomparability and ambiguity of mandates, and selection effects. Downs and Stedman argue against overambitious standards of success and insist instead that success is best measured by short-term indicators directly related to the ambitions of peace implementation: (1) the stopping of violence in the short term, and (2) ending the war on a self-sustaining basis. As a supplement to these measures, the authors suggest that counterfactual analysis can play an important role in assessing the impact of interventions. Downs and Stedman then conduct a determinants study to ascertain what variables are associated with implementation success. They score the cases of peace implementation on two sets of variables—one related to the conflict environment and one related to international commitment. As the authors show, the greater the difficulty of the environment, the greater the likelihood that peace implementation will fail. The three most important environmental sources of failure are the presence of spoilers, the presence of neighboring states that oppose the peace agreement, and the presence of valuable commodities or spoils. Cases of peace implementation also differ in terms of the willingness of international actors to provide resources and risk troops, a variable, Downs and Stedman posit, that is a function of whether a regional or major power sees the war-afflicted country as affecting its vital security interest. While the United Nations can succeed in peace implementation in the easiest conflict environments, regional or major power interest is crucial to succeed in the most difficult environments. The findings of Downs and Stedman raise the issue of incentive compatibility. Difficult cases require greater international attention, resources, and coercive capability. Unless the security interests of major powers are engaged, the resources and commitment necessary for coercive strategies to succeed will not be forthcoming. Downs and Stedman contend that the gap between what is needed in some missions and what major or regional powers are willing to provide leads to organizational pathologies within the United Nations. Eager to ensure that the UN will have a role in conflicts

S-1

11/24/09

4:15 PM

Page 25

INTRODUCTION

25

around the world, the Secretariat has incentives to downplay the difficulties of some missions, and to withhold worrisome threat assessments. The need to present optimistic scenarios to the Security Council precludes contingency planning and occasionally leads the Secretariat to ignore or downplay violations of peace agreements. In Chapter 3, Michael Doyle looks at the role of international actors in supplying transitional authority during peace implementation. Similar to Downs and Stedman’s approach, Doyle differentiates cases into ecologies of conflict, and argues that the amount of transitional authority that must be supplied by implementers varies according to the degree that the parties are few or many, reconciled or hostile, and coherent or fragmented. Where the parties are few, reconciled, and coherent, transitional authority can be limited to confidence-building and verification. Where the parties are many, hostile, and fragmented, the international role must substitute for local authority, and must be willing to use coercion to assert its own authority. In Chapter 4, Bruce Jones discusses a different problem of incentive compatibility, involving mechanisms for providing strategic coherence and coordination. The more difficult the implementation environment, the greater the need for strategic coordination. When international actors lack unity, spoilers can take advantage of international splits to attack the peace process and threaten the peacemakers. But strategies available for international coordination, such as Friends groups, are also a function of great and regional power interest. The willingness of states to join Friends groups is indicative of a prior judgment that the specific case is in the state’s interest. Such mechanisms will not be available in all cases. In Chapter 5, Donald Rothchild looks at another aspect of strategy. He argues that mediators of civil wars can pursue one of two strategies of political settlement: one that emphasizes the group basis of rights and representation and one that rests on individual rights and majoritarianism. He points to a different problem of incentive compatibility: whereas warring parties have a large short-term incentive to choose group-based accommodation in order to address their security fears, the group-based strategy exacerbates problems of governance in an already weak state. Since consolidation of peace must rest on the reconstruction of state capacity, Rothchild argues that for long-term, postimplementation stability, the individual rights strategy provides greater flexibility for elites, and greater potential for strengthening the state. But mediators, whose interest lies in the reaching of agreements, have an incentive to support group-based approaches. Rothchild argues strongly for mediators and negotiators to consider the problem of state capacity-building when negotiating a peace agreement. As a means of combining the flexibility of the individual rights strategy with the greater sense of security that accrues from the group-based strategy, Rothchild puts forward several mechanisms to ensure group rights and representation without complete reliance on the group-based strategy: propor-

S-1

11/24/09

26

4:15 PM

Page 26

ENDING CIVIL WARS

tional positions and resources in government, proportional representation, and protection for cultural sensitivities.

Evaluating Implementation Tasks Part 2 scrutinizes claims that are made about the importance of various implementation tasks and their role in overall success and failure. Such claims grew in prominence in the 1990s as various international NGOs lobbied publics and governments and insisted that their single issue of concern—whether it be disarmament, elections, human rights, or refugee repatriation—was crucial to implementation success. In Chapter 6, Joanna Spear examines the tasks of demobilization and disarmament in peace implementation. Although these topics are usually merged into one, Spear points out that demobilization of former combatants is the more important goal of the two, and that success in disarmament usually follows success in demobilization. Spear argues that implementers must take care when designing demobilization and disarmament programs to take into account the local context. Societies where gun ownership is promoted by cultural norms, or where the state has collapsed and groups possess greater confidence in seeking security through self-help than the state, present much greater challenges to disarmament. According to Spear, demobilization of combatants is a lynchpin for successful peace implementation for two reasons. First, it is at the heart of the civil war security dilemma. If done poorly, demobilization can increase the vulnerability and fears of the warring parties and, in the worst case, rekindle the war. She argues that seldom do implementers act as security guarantors, in the sense of protecting the warring parties during demobilization. Rather, implementers usually employ a strategy of confidence-building in demobilization and disarmament. For such a strategy to assuage reluctant warring parties, and encourage them to overcome their commitment problems, implementers must provide robust monitoring and verification. Spear criticizes the performance of many implementers, especially their reluctance to acknowledge violations and cheating during implementation. Second, successful demobilization provides former combatants with an alternative to life with the gun. Spear argues that for demobilization to bolster the long-term prospects for peace, it must include efforts to reintegrate combatants into society. She estimates that for reintegration to succeed, there must be a three- to four-year commitment from international implementers. Such a commitment, however, often conflicts with the urge to define demobilization in narrow terms and finish it in the first year of implementation. The danger, of course, is that such a sudden discharge may strain the resources of the new post–civil war state and may pose a threat to civilian security if former combatants turn to crime.

S-1

11/24/09

4:15 PM

Page 27

INTRODUCTION

27

In Chapter 7, Susan L. Woodward describes the economic priorities of peace implementation and argues that it is an area where international implementers must change their strategies and improve their performance. As she observes, the tasks and priorities of economic reconstruction are usually afterthoughts in peace agreements; when the international financial institutions (IFIs) take a lead economic role in implementation, they tend to bring with them strategies and approaches that are inappropriate to post–civil war conditions. Indeed, in several cases, the demands of the IFIs have been contradictory to the successful implementation of a peace agreement. Woodward notes that we now have enough evidence from what works and what doesn’t on the economic side to prioritize economic goals: the rapid revival of the economy in order to create a peace dividend that warring parties and their constituents can enjoy; the funding of specific commitments of the peace agreement, especially in the area of demobilization and reintegration of combatants and in providing the new postwar state with the capacity to meet the demands placed upon it; and the creation of the economic foundations necessary to sustain peace. Woodward criticizes the IFIs and international donors for failing to create strategies appropriate to implementing and sustaining peace. Even when IFIs and donors recognize that their actions are not producing their intended effects, the standard response, says Woodward, is to treat it as a technical problem of improving performance. Ideological precommitment to neoliberal economic models, combined with bureaucratic rigidities, prevents the IFIs from proper evaluation of their policies, and forecloses the creation and adoption of strategies more appropriate to the postwar environment. In Chapter 8, Terrence Lyons observes that implementers often hold elections for multiple, sometimes contradictory goals: for war termination, for democratization, and for their own organizational purposes. When such goals are in conflict, as they often are, Lyons advocates the priority of war termination. Critics of these so-called settlement elections often compare the results of such elections against impossibly high standards or against alternatives that are not politically feasible. The problem for implementers is to gauge what is possible to achieve from holding elections and to consider what alternatives may be available. Consistent with the conclusions of Joanna Spear, Lyons emphasizes the need for substantial investment in demilitarizing politics. A concerted effort to transform armies into viable political parties can play a crucial role in persuading rebels to compete for power peaceably within the political arena, thereby strengthening the prospects for war termination. Even in elections whose short-term results are unlikely to further democratization, the process can matter. The construction of electoral authorities and institutions with legitimacy and effectiveness can serve as a bridge to long-term democratization.

S-1

11/24/09

28

4:15 PM

Page 28

ENDING CIVIL WARS

In Chapter 9, Tonya Putnam puts forward an instrumentalist approach to human rights advocacy in peace implementation that appeals to pragmatism and context-sensitivity in the inhospitable environment of post–civil war societies. Putnam argues that international human rights organizations have paradoxically asked for both too much and too little with regard to human rights and peace implementation. They have asked for too much by insisting on a doctrinaire enforcement approach to human rights protection in environments that are ill suited for such an approach. They have asked for too little by dogmatically refusing to assist new governments in working toward improved human rights performance and to see that all subgoals of peace implementation have a human rights component. Although the promotion of human rights and the establishment of institutions capable of advocating and protecting human rights are desirable for societies emerging from civil war, the demands and challenges of implementing peace in war-torn countries require more nuanced strategies than those typically chosen by international human rights organizations (IHROs). Putnam argues that an “exclusive reliance on the enforcement approach to human rights protection is entirely unsuited to the early stages of peace implementation.” Typically, IHROs have failed to appreciate the specific challenges posed by war-torn societies and have used strategies designed for environments where functioning governments and rule of law already exist. This has led IHROs to three common errors: (1) to place undue importance on achieving a formal expression of human rights provisions in the main texts of peace settlements; (2) to demand individual responsibility for human rights violations in societies that lack even rudimentary investigative, adjudicative, and compensatory institutions; and (3) to isolate human rights concerns by eschewing direct involvement with postsettlement governments and failing to integrate human rights concerns with other key tasks of peace implementation. Putnam’s critique of the IHROs does not imply that human rights are irrelevant to the achievement of peace. Rather, she points to the need for a different approach by those who care about the promotion and protection of human rights. First, human rights objectives must be linked to other instrumental tasks of peace implementation. Second, human rights groups must work to build the capacity of locally based human rights advocates, who will have to do the heavy lifting over the long haul. In Chapter 10, Howard Adelman evaluates the contribution of refugee repatriation to successful peace implementation. He concludes that the evidence from this project’s case studies defies any simplistic formulation of the role that repatriation plays in making or sustaining peace. As a refugee scholar, Adelman expresses surprise at this finding, as it is simply assumed in refugee scholarship and advocacy that there can be no peace without refugee repatriation. Indeed, Adelman cites several examples where the absence of repatriation, especially when many refugees are resettled in

S-1

11/24/09

4:15 PM

Page 29

INTRODUCTION

29

third countries and remit funds back to their home country, can assist the sustaining of peace. But Adelman cautions that international security scholars have also been amiss by dismissing the importance of refugees for making and implementing peace. Here, he warns, context makes a great difference: in wars that are fought over people, such as Bosnia, Rwanda, and Lebanon, as opposed to wars fought over the government, such as El Salvador and Mozambique, the treatment of the refugees in the peace process is crucial. International actors can endanger a nascent peace by including a right to return in the peace agreement, but then not enforcing that right. By treating return as a rhetorical commitment, rather than a real commitment, implementers create a large population who can neither return nor be resettled— a recipe for the creation of what are known in refugee scholarship as refugee warriors. This, of course, creates a daunting ethical dilemma: where implementation success is measured simply by the absence of war, the surest way to avoid new hostilities is to avoid a commitment to repatriation in the peace agreement. Yet in practice, at least in Bosnia, the unease of policymakers with the ethical implications of that choice and their reluctance to repatriate refugees to their original homes has led to a worse outcome: a proclaimed right to return without enforcement, which, in Adelman’s words, “is no real right to return [and] means the refugees are in limbo . . . the worst of the alternatives available.” In Chapter 11, Charles Call and William Stanley point to a gap in peace implementation efforts: the provision of security to the mass of civilians who are vulnerable during the implementation of peace agreements. The security gap, although not threatening to the short-term implementation of agreements, has the potential to undermine the long-term sustaining of the peace. Where civilians do not feel secure in a newly created state, they will be vulnerable to the appeals of ethnic extremists who provide communal protection in competition with the state. To build civilian security in the short and long run, Call and Stanley urge peace implementers to begin a process of judicial and police reform. Although such a process is fraught with difficulties, it is nonetheless a robust contribution that implementers can make to ensuring that peace will survive their departure. In Chapter 12, John Prendergast and Emily Plumb argue in a similar vein that implementers should strive to support and build local capacity for peace through local civil society organizations. Again, like Call and Stanley, the authors acknowledge that failure to build local capacity will probably not endanger the achievement of short-term implementation success. But like many of this book’s authors, Prendergast and Plumb stress that implementers should strive to take advantage of the implementation period to sow the seeds for sustained peacebuiliding. They discuss several local innovations in peacebuilding, including problem-solving workshops,

S-1

11/24/09

30

4:15 PM

Page 30

ENDING CIVIL WARS

trauma treatment programs, collaborative community development programs, and indigenous conflict resolution mechanisms. Although Prendergast and Plumb concede that local civil society organizations are not a panacea for reconstructing social relations and peacebuilding, they insist that, because of their commitment and local knowledge, they are better suited to the tasks of peace than their international counterparts.

The Contribution of Case Studies Part 3 presents case studies of peace implementation missions. In choosing which cases to publish in this book, we used four standards. First, we wanted global coverage, and therefore included cases from every part of the world. Second, we gave priority to less-studied cases, for example, Sri Lanka, Lebanon, Nicaragua, Guatemala, and Liberia, over more-studied cases such as Mozambique and Angola. Third, we included case studies where the narrative of the author makes a compelling contribution that goes beyond a summary of detail and events. Fourth, we wanted to ensure that the cases published here provided important variation in implementation environment and strategy. Noncoercive Cases: Nicaragua, El Salvador, Guatemala, and Rwanda In Chapter 13, Caroline Hartzell examines the Central American peace process of the 1980s and its relationship to the successful negotiated settlement of the Nicaraguan civil war. She shows that a regional peace implementation regime developed that involved heads of state, the Organization of American States (OAS), and the United Nations to address the linkages between civil wars in the region and interstate conflict. In the 1980s the Central American regional security complex consisted of several states that suffered from civil war—Nicaragua, Guatemala, and El Salvador—and international intervention in terms of supplying arms and training to guerrillas and state security institutions, and providing rebels with sanctuaries across borders. The peace process recognized that to end the region’s civil wars, action was necessary both within countries—a commitment by governments to negotiate with rebels, to initiate democratic reforms, and to implement human rights—and between countries: a commitment by governments to refuse rebel groups sanctuary and bases on their territory from which they could attack across borders, as well as to stop the flow of weapons and military supplies across borders that maintained the rebels. If done sequentially, moves by neighboring states to limit their support of rebels elsewhere would embolden those governments in the midst of war to initiate negotiations and reforms. To build confidence, the

S-1

11/24/09

4:15 PM

Page 31

INTRODUCTION

31

process created an implementation regime: norms and expectations about appropriate actions to be taken by governments at home and in the region, with international actors willing to monitor and verify compliance. Hartzell does not argue that the implementation regime was sufficient to end the war in Nicaragua. As she insists, the keys to the end of the Nicaraguan civil war were the changed geopolitical environment as the Cold War closed and a change of administration in Washington, D.C., to one willing to accept and then support regional initiatives for peace. But the creation of a regional peace implementation put in place a formula and the capacity for peaceful resolution of the region’s interconnected wars, and built confidence in the monitors and verifiers of the agreement, and between the governments of the region. Hartzell’s chapter should be read by those who seek to make peace in regions where civil wars spill over borders and intersect with and inflame interstate rivalries. Even where the specific civil wars may not be amenable to negotiation, a regional implementation regime can begin a process that untangles interstate hostility and involvement from ostensibly internal conflicts, and provide a push for the necessary reforms and dialogue at the national level that are invaluable for civil war termination. In Chapter 14, Charles Call evaluates the implementation of El Salvador’s peace agreements in the early 1990s, a case often cited as a model for the successful mediation and implementation of a peace agreement in civil war. Call acknowledges the accomplishments of the peace process: the demobilization of the guerrillas and their transformation into a competitive political party; free and fair elections in the 1990s; and thoroughgoing institutional reform of the police and judiciary and the advancement of human rights in the country. Now almost a decade after the peace agreement, Call argues that the prospect of a reignition of organized violent conflict is unlikely. Still, he admits, the country has been beset by high levels of crime, and shortfalls in public security. Call notes that the successes of implementation in El Salvador are related to the care and specificity that went into the forging of the peace agreement. The amount of detail included in the accords was a boon, not a barrier for the implementers. But Call cautions that the quality of the agreement was a function of other, more decisive variables: the balance of political and military power between the government and the guerrillas, the degree of cohesion within the parties, and the unity of purpose among key international actors, including the United Nations, the OAS, the regional heads of state, and the United States. Call maintains that the success of the United Nations, as first mediator, and then verifier and institution-builder, is best explained by the propitious environment in which it worked. As Call insists, this does not denigrate the skill and choices of the international implementers, but it does limit the applicability of the Salvadoran model to other war-torn societies.

S-1

11/24/09

32

4:15 PM

Page 32

ENDING CIVIL WARS

William Stanley and David Holiday also emphasize contextual variables in explaining the successful termination of the Guatemalan civil war. In Chapter 15, they argue that from the standpoint of war termination, the implementation of the peace agreement was an unambiguous success; the cease-fire was never breached by the combatants, demobilization of guerrillas was completed rapidly and they now compete in elections as a political party, and government troops complied with personnel and resource reductions. But Stanley and Holiday insist that looking solely at success on security dimensions is misleading, as fighting had stopped before the final peace agreement was signed, and the guerrillas were a spent force of less than 2,000 active combatants and so weak that a return to war was highly unlikely. Precisely because the conflict environment was so amenable to terminating the war, Stanley and Holiday argue that the implementation of the peace agreement should be judged by a higher standard: the degree to which the peace process contributed to the democratization of the country and addressed fundamental issues of resource equity and cultural inclusion. They contend that implementation of the tasks in the peace accords meant to accomplish these ends has lagged, and the success of the overall process is in doubt. The Guatemalan peace negotiations are often described as the most inclusive and participatory of all of the major civil war peace processes. International actors succeeded in bringing a wide range of voices from civil society into the peace talks, and these voices were awarded with what appeared at the time to be major concessions by the Guatemalan government to open politics, reform government institutions, and pursue more equitable development policies. But Stanley and Holiday contend that the very inclusiveness that seemed a strength during the negotiations has been a weakness during implementation. The commitments by the government were in hindsight broad and vague, hostage to popular referenda that were easily manipulable by conservative elites, and without specific redress if abandoned or diluted. The power of broad civil society participation in the peace talks has been diminished by their lack of unity and organization during implementation. As captured in the title of their chapter, “everyone participates, no one is responsible.” Stanley and Holiday fault the international implementers of the agreement for failing to stay aware of the imbalances of power between the parties to the peace process and how they can shift and imperil implementation. The very weakness of the guerrillas facilitated ending the war and accomplishing much on the security side. But once they had been demobilized, the government gained a dominant position. Although the government faced myriad voices and organizations in the negotiations, during implementation it faced a civil society unable to mobilize to pressure for compliance. In the cases of Nicaragua, El Salvador, and Guatemala, international

S-1

11/24/09

4:15 PM

Page 33

INTRODUCTION

33

actors followed a strategy of confidence-building, with its emphasis on monitoring and verification of compliance. In all three cases, such a noncoercive strategy was sufficient to end the wars and address the security concerns of the warring parties. In the cases of El Salvador and Guatemala, international actors went beyond confidence-building to attempt to transform political and economic institutions, with more success in El Salvador than in Guatemala. But as the authors insist, the only reason that the United Nations could attempt to be ambitious in these cases was that the conflict environment was amenable to such a strategy. The success of confidence-building as an implementation strategy in Central America and elsewhere (Namibia and Mozambique) contrasts dramatically with its failure in Rwanda. In Chapter 16, Gilbert Khadiagala examines the failure of international actors to implement the Arusha Accords, a failure that resulted in the genocidal slaughter of more than 800,000 Tutsi and moderate Hutu. Khadiagala takes issue with critics who blame the mediation process at Arusha for overseeing a process that produced an agreement that excluded an extremist faction, the Coalition pour la Défense de la République (CDR), a Hutu supremacist group instrumental to the genocide. Khadiagala insists that given the extent of polarization within Rwanda in 1993 and given that economic circumstances were intensifying the polarization, the agreement reached at Arusha was all that was possible. According to Khadiagala, responsibility for the failed implementation lies in part with representatives of the United Nations and important member states, who participated at Arusha and led the signatories and the mediators to expect that the UN would provide a robust implementation force to manage the threat posed by the CDR. When the UN limited its involvement to a small force, based on traditional concepts of confidence-building, and refused to confront the violence and obstructionism of the extremists, Rwanda’s fate was sealed. The failure to implement Arusha stemmed from the unwillingness of the United Nations, and its member states, to pursue a coercive strategy of implementation that might have protected the signatories of the agreement and deterred, or defeated, the extremists. Particularly disturbing is that it should have been predictable that the UN would not pursue the appropriate strategy. Khadiagala’s analysis raises the question of whether the mediators suffered from role-induced myopia where their focus on getting an agreement prevented them from seeing ahead to whether it had a chance of being implemented or what the consequences of failed implementation would be. Mixed Strategies: Cambodia and Bosnia Implementers need to craft their strategy with due consideration given to facts on the ground, especially the likelihood of facing spoilers, and facts

S-1

11/24/09

34

4:15 PM

Page 34

ENDING CIVIL WARS

about the coalition that supports the peace process, especially their willingness to provide troops and resources necessary to carry out the chosen strategy. The Rwanda case suggests that there are times when there is a gap between what is needed to implement an agreement and what the coalition behind implementation is willing to undertake. In situations where implementers know that they will not receive the appropriate resources to do the job, it would be better not to implement at all. The Cambodia case suggests that when confronted by spoilers, implementers may have strategic options short of defeating them militarily. As described by Sorpong Peou in Chapter 17, the UN operation in Cambodia (UNTAC) discussed the use of force to compel the Khmer Rouge to comply with its commitments to the Paris Agreement or to defeat them militarily. Yasushi Akashi, the special representative of the Secretary-General, and General John Sanderson, force commander, believed that the coalition of member states that supported the peace process would not countenance using force against the Khmer Rouge. Although they possessed nearly 16,000 troops, eight times the amount that the UN Assistance Mission in Rwanda (UNAMIR) had in Rwanda, they realized that they lacked the military capacity to confront the Khmer Rouge. Moreover, they believed that many of the countries that contributed troops for the mission did so under the condition that the mission would be a consent-based, noncoercive mission. If UNTAC were to use force against the Khmer Rouge, many member states would withdraw their soldiers. Peou contends that when the Khmer Rouge withdrew from the peace process, UNTAC switched from a strategy of pursuing a democratic peace to a strategy of pursuing a hegemonic peace, defined as political stability obtained through hegemonic control. By isolating the Khmer Rouge and going ahead with elections, UNTAC tended to ignore violence carried out by the State of Cambodia (SOC), whose compliance with the agreement was limited. This led, in Peou’s judgment, to a new imbalance of power that favored the SOC. The SOC realized that the UN needed it for elections to go forward, and it knew that if it lost the elections, it could nonetheless demand a large role in a power-sharing government. Peou concedes that the hegemonic strategy produced a partial success in Cambodia, but asserts that UNTAC could have accomplished more. Peou argues that within the international coalition of states that supported the peace process, most wanted an outcome that would reduce Khmer Rouge power and influence. Moreover, from the start, the implementers feared that the Khmer Rouge’s participation in the peace process was tactical. This, according to Peou, led UNTAC to ignore legitimate security fears of the Khmer Rouge, which further convinced its leaders that its survival would be in jeopardy if it were to disarm and comply with the peace agreement. The hegemonic strategy that Peou describes falls somewhere between

S-1

11/24/09

4:15 PM

Page 35

INTRODUCTION

35

confidence-building and compellence on the dimension of coercion. On the one hand, UNTAC clearly abandoned reliance on mere confidence-building. UNTAC reconfigured its military positions to deter and be able to defend the mission against possible Khmer Rouge attacks. It also worked with the Khmer Rouge’s main patron, China, to dissuade the Khmer Rouge from militarily challenging the implementation mission. Finally, UNTAC cooperated with and deployed alongside the armies of the remaining factions in the peace process to protect voting stations in the country. On the other hand, as Peou argues, UNTAC’s limited use of coercion did not threaten the Khmer Rouge’s survival. UNTAC left room for the Khmer Rouge to rejoin the peace process before the election—if it would comply with its overall commitments. It did not attempt to use military force to compel compliance or defeat the Khmer Rouge. Nor did it threaten to charge the Khmer Rouge with war crimes. Had UNTAC done so, Peou contends, it is unlikely that the Khmer Rouge would have collapsed after the elections, when most of their soldiers claimed amnesty and abandoned Pol Pot. In Bosnia, another case of only partial success, the implementers pursued a military strategy of deterrence that they never satisfactorily tied to a civilian strategy to win political compliance with the Dayton Agreement. As Elizabeth Cousens describes in Chapter 18, the implementers of Dayton faced a difficult conflict environment and an agreement, composed of contradictions and competing theories of war termination, that was difficult to implement. As she points out, the contradictions of the agreement reflected contradictions in the coalition of states that supported the peace process: desiring of a just peace that rolled back the gains of ethnic cleansing and held individuals accountable for war crimes, but cautious about the risks that they were willing to run to accomplish their objectives. And given that the agreement was the product of coercion, Cousens adds, it should have been clear that a good deal of coercion would be necessary to implement it. While IFOR anticipated the possibility of large-scale military spoiling of the agreement, and therefore deployed with a deterrent capacity, it also set the bar low on what the armies would be asked to do. Dayton avoided the security dilemma–inducing prospect of deep demobilization and creation of a single unified army and instead allowed the new quasi-statelets of Republika Srpska and the Bosniac-Croat Federation to keep their own army in the hopes that a balance of power could be engineered that would provide stable interethnic deterrence. The implementers did not anticipate the possibility of political spoiling of the agreement, and rested their civilian strategy on rapid elections and multiethnic institutions that would tame the ethnic statelets. When the rapid elections, a prerequisite for U.S. participation in implementation, consolidated the power of the ethnic extremists who had led the war, the door was open to widespread political spoiling of the agreement: the continuing appeal to ethnic extremism and exclusion as a basis for political support, the use of police as a means of rendering eth-

S-1

11/24/09

36

4:15 PM

Page 36

ENDING CIVIL WARS

nic and political adversaries insecure, refusal of local authorities to allow freedom of movement or repatriation of other ethnic groups to their areas, the collusion of political authorities in informal, corrupt economic activities, and general footdragging by elected authorities in creating and then working with multiethnic institutions. After three years of noncompliance on the political front, the implementers decided to overrule the democratic processes and institutions that were to be the basis of civilian peacebuilding and chose to pursue what Cousens calls a trusteeship strategy. The civilian heads of the international implementation mission would dictate political solutions to thorny problems and ensure that controversial policies were carried out over the footdragging of local authorities. Cousens argues that such a strategy would have made sense at the beginning of the mission. Coming when it did, however, it poses several large problems, chief among them that the very democratic process the implementers attempted to instill in Bosnia was discredited, extremist politicians can free-ride on the benefits of imposed cooperation without having to make the tough choices to eschew ethnic polarization, and the implementers now find no realistic prospect for a transition to self-sustaining peace that would allow them to leave. Coercive Strategies: Lebanon, Liberia, and Sri Lanka The final three cases in the book examine the application of coercive strategies to compel compliance with an agreement. All three are cases of regional powers taking a lead to impose peace; we judged two of them partial successes and one of them a failure. In Chapter 19, Marie-Joëlle Zahar examines the Syrian role in the implementation of the Ta’if Agreement on Lebanon. As Zahar points out, the Ta’if Agreement differs from most civil war peace agreements in that the negotiators were not representatives of the warring parties, but politicians who had been elected to Parliament almost eighteen years prior to the negotiations. In Zahar’s words, they were spectators rather then actors in the conflict, and this weakened their legitimacy in the eyes of several would-be spoilers. The most significant party to the conflict who objected to the agreement was General Michael ‘Awn, who argued that political reforms envisioned in the agreement were insufficient to solve Lebanon’s political problems and who demanded that the agreement call for the rapid and total withdrawal of Syrian troops from Lebanon. When ‘Awn refused to comply with provisions of the agreement, the Syrian army and Lebanese army units that were proagreement swiftly defeated ‘Awn’s militia, paving the way for implementation of Ta’if by Syria. Zahar acknowledges that Syria’s hegemonic implementation of Ta’if brought large-scale violence to a close, the gradual reassertion of Lebanese state sovereignty over the country, and successful elections and the rebuild-

S-1

11/24/09

4:15 PM

Page 37

INTRODUCTION

37

ing of the state and the economy. Because Syria has a direct security interest in a stable Lebanon, it has been willing to take risks and accept high costs in maintaining the peace. In doing so, however, Syria’s hegemonic peace has curtailed human rights, democracy, and national reconciliation. Moreover, it has only partially fulfilled the provisions of Ta’if, most importantly in the area of demobilization, where it has allowed Hizballah to continue its army, an important asset for Syria in its conflict with Israel. Zahar provides a balanced assessment of Syria’s muscled approach to implementation: the ending of war and the provision of stability, but at the costs of several political goods, such as democracy, human rights, and reconciliation, and at the cost of continuing involvement by Syria in Lebanon’s affairs. The final two case studies, Liberia and Sri Lanka, serve as cautionary examples for would-be regional implementers. In Chapter 20, Adekeye Adebajo evaluates the Nigerian-led ECOMOG intervention into Liberia in 1990. ECOMOG intervened without a peace agreement among the warring factions and at a time when one rebel, Charles Taylor, controlled 85 percent of the territory. In the absence of consent by the warring parties and even a nascent framework for peace, ECOMOG quickly found itself at war with Taylor, and in need of a diplomatic solution that would end the war. In the course of the war, fourteen separate peace agreements would be forged, but all failed until the Abuja II Agreement of 1997. Adebajo argues that peacemaking in Liberia proved much more difficult than Nigeria had imagined in 1990. The conflict environment was especially challenging with the presence of hostile factions, a collapsed state, and splintering groups. ECOMOG’s attempts to compel peace were foiled by political divisions within its coalition members and the lack of assistance and support by the United States and the major powers. Although ECOMOG found itself a combatant in Liberia’s civil war and at times felt that it was in a military quagmire that precluded any withdrawal, ECOMOG gradually learned to marry its military capability to a viable political strategy to end the war. As ECOMOG shifted its strategy to show willingness to accommodate Taylor’s demands, regional unity and international support made it a more capable implementer. But as Adebajo demonstrates, the peace that emerged from ECOMOG’s accommodating strategy rewarded the very faction that would have come to power earlier in the war if ECOMOG had not intervened. Moreover, Adebajo raises serious doubts about the likely longevity of the peace. Adebajo concludes the chapter by discussing several trends in Charles Taylor’s rule that show corruption, human rights violations, and ethnic favoritism on the rise. And his consolidation of power has coincided with an even greater role in destabilizing peace in the region, including aiding and abetting the RUF in Sierra Leone as it undermined peace in that country. In Chapter 21, Sumantra Bose provides a stinging indictment of India’s attempt to implement peace in Sri Lanka from 1987 to 1990. Bose

S-1

11/24/09

4:15 PM

38

Page 38

ENDING CIVIL WARS

describes how the government of Rajiv Gandhi, in an attempt to strengthen his political fortune at home, crafted a rushed agreement with the government of Sri Lanka to end the war in that country. That the agreement did not include any representatives of the Tamil groups that were fighting the Sri Lankan state and address even the minimal security fears or political demands of the Tamils meant that the agreement would have to be renegotiated during its implementation if it were to have a chance for success. But because of faulty intelligence and hubris by an Indian government that believed it could impose the terms of the agreement on the most radical of the Tamil parties, the Liberation Tigers of Tamil Eelam (LTTE, or Tamil Tigers), the Indian peacekeeping force quickly found itself in major open combat in Sri Lanka. At the height of the Indian intervention, it had 100,000 soldiers in Sri Lanka, with heavy equipment including tanks, artillery, and fighter planes. Nonetheless, its soldiers were fought to a standstill by the much smaller, but much more committed guerrilla force of the Tigers. As Bose points out, it is not self-evident that an agreement that attempted to include the Tigers and address their demands would have led to settlement. But given their willingness to fight the Indian army over the terms of the agreement, the only possibility for a peace agreement would have been for the Indian force to change strategies and attempt to revisit the terms of the agreement and appease the Tigers.

Notes 1. See, for example, I. William Zartman, Ripe for Resolution (New York: Oxford University Press, 1985); Richard Haass, Conflicts Unending (New Haven: Yale University Press, 1990); and Stephen John Stedman, Peacemaking in Civil War: International Mediation in Zimbabwe, 1974–1980 (Boulder, Colo.: Lynne Rienner, 1991). 2. Stephen John Stedman, “UN Intervention in Civil Wars: Imperatives of Choice and Strategy,” in Donald C. F. Daniel and Bradd C. Hayes, eds., Beyond Traditional Peacekeeping (New York: St. Martin’s Press, 1995), pp. 40–63. 3. Fen Osler Hampson, Nurturing Peace: Why Peace Settlements Succeed or Fail (Washington, D.C.: U.S. Institute of Peace, 1996). 4. Ibid., p. 210. 5. Ibid. 6. Ibid. 7. Ibid., p. 221. 8. Ibid., pp. 221–232. 9. Barbara F. Walter, “The Critical Barrier to Civil War Settlement,” International Organization 51, no. 3 (summer 1997): 340. 10. Ibid., p. 345. 11. Ibid. Quotations on guarantees from pp. 345–347. 12. Ibid., p. 347. 13. Ibid., p. 341.

S-1

11/24/09

4:15 PM

Page 39

INTRODUCTION

39

14. Jeremy Ginifer, Managing Arms in Peace Processes: Rhodesia/Zimbabwe (Geneva: UNIDIR, 1995), p. 22. 15. Ibid. 16. Ibid., p. 23. 17. Walter, “The Critical Barrier,” p. 362. 18. Steven R. Ratner, The New UN Peacekeeping (New York: St. Martin’s Press and UN Council on Foreign Relations, 1995), p. 49 (emphasis added). 19. Ibid. 20. Stephen John Stedman and Donald Rothchild, “Peace Operations: From Short-Term to Long-Term Commitment,” International Peacekeeping 3, no. 2 (summer 1996): 17–35. 21. Jeffrey L. Pressman and Aaron Wildavsky, Implementation, 3rd ed., expanded (Berkeley: University of California Press, 1984), pp. xxv, 147–162. 22. Richard C. Holbrooke, To End a War, rev. ed. (New York: Random House, 1999), p. 337. 23. Vicki Eaton Baier, James G. March, and Harald Saetren, “Implementation and Ambiguity,” in James G. March, ed., Decisions and Organizations (Oxford: Basil Blackwell, 1988), p. 157. 24. Stedman, “UN Intervention in Civil Wars”; and Stephen John Stedman, “Negotiation and Mediation in Internal Conflict,” in Michael E. Brown, ed., The International Dimensions of Internal Conflict (Cambridge: MIT Press, 1996), pp. 341–376. 25. Stedman, “Negotiation and Mediation,” pp. 367–368. 26. Stephen John Stedman, “Spoiler Problems in Peace Processes,” International Security 22, no. 2 (fall 1997): 5–53. 27. Marie-Joëlle Zahar, “Reframing the Spoiler Debate in Peace Processes,” in John Darby and Roger MacGinty, eds., Progressing Towards Settlement (London: Palgrave, forthcoming). 28. Ibid., and John Darby, The Effects of Violence on Peace Processes (Washington, D.C.: U.S. Institute of Peace, 2001). 29. Mark Peceny and William Stanley, “The Resolution of Civil Wars in Central America,” International Organization 55, no. 1 (winter 2001): 149–182. 30. Ibid., pp. 177–178. 31. Elizabeth Jean Wood, “Civil War Settlement: Modeling the Bases of Compromise,” paper presented at the 1999 annual meeting of the American Political Science Association, Atlanta, Ga., September 2–5, 1999. 32. Michael Doyle, “Strategies of Enhanced Consent,” in Abram Chayes and Antonia Handler Chayes, eds., Preventing Conflict in the Post-Communist World (Washington, D.C.: Brookings Institution, 1996), pp. 483–506. 33. Michael W. Doyle, Ian Johnstone, and Robert C. Orr, eds., Keeping the Peace: Multidimensional UN Operations in Cambodia and El Salvador (Cambridge: Cambridge University Press, 1997); and Michael W. Doyle and Nicholas Sambanis, “International Peacebuilding: A Theoretical and Quantitative Analysis,” American Political Science Review 94, no. 4 (December 2000): 779–801. 34. Roland Paris, “The Perils of Liberal International Peacebuilding,” International Security 22, no. 2 (fall 1997), pp. 54–89. 35. Originally the project included the case of Western Sahara, but upon review of the case study the project directors believed that the case did not fit the parameters of our definition. 36. The common statistical measure is 1,000 battle deaths cumulative, with a smaller yearly benchmark that indicates the continuance of the war.

S-1

11/24/09

40

4:15 PM

Page 40

ENDING CIVIL WARS

37. Since there was a major international intervention to enforce and build peace in Haiti, we did not discourage our thematic authors from studying the case for clues about the capacity and ability of international actors to make peace. 38. Again, since these cases do have potential lessons for those who seek to make peace in civil wars, we did not discourage our thematic authors from referring to them. 39. Fortunately, our project included several authors who were experts on those cases, or who had participated in their implementation as they unfolded, and therefore several of the thematic essays make reference to them when appropriate. 40. Peter Wallensteen and Margareta Sollenberg, “Armed Conflicts, Conflict Termination, and Peace Agreements, 1989–96,” Journal of Peace Research 34, no. 3 (August 1997): 342.

S-2

11/24/09

4:16 PM

Page 41

PART 1 Evaluating Implementation Strategies

S-2

11/24/09

4:16 PM

Page 42

S-2

11/24/09

4:16 PM

Page 43

2 Evaluation Issues in Peace Implementation GEORGE DOWNS AND STEPHEN JOHN STEDMAN

Compared to the formidable scholarly literature dealing with the problem of negotiating peace agreements in civil wars, the literature devoted to the problem of implementing peace agreements is modest. This is somewhat surprising, as it is painfully clear that the distance between a peace agreement and anything that warrants the term real peace is as large as the distance between the ideals expressed in the most ambitious domestic program and their concrete realization. Many more peace agreements are signed than are actually put into practice, and once signed, the stakes are often far higher. A large part of the explanation for why the literature on the implementation of peace agreements in civil wars is not more well developed is that it is difficult to think of an environment that is less conducive to the conduct of evaluation research. The number of cases is small and the measurements of potentially important variables, such as the military capability and political goals of different parties, are often unreliable. Implementation strategies tend to be poorly specified and no two are ever identical. Worse, there are literally dozens of contextual variables—ranging from the amount of international resources available to the United Nations, to the salience of a given conflict for international public opinion—that vary widely across cases and inevitably influence both the choice and the success of an implementation strategy. All of this leaves both the policymaker and the analyst continually uncertain about what a given strategy actually comprises, what would have occurred had a strategy not been employed (the counterfactual), and whether the superiority of one strategy over another reflects the greater merits of the first or the fact that the second strategy had to cope with a more demanding context. In the absence of systematic evaluation, however, our understanding of 43

S-2

11/24/09

44

4:16 PM

Page 44

EVALUATING IMPLEMENTATION STRATEGIES

what does and does not work in peace implementation is prey to any number of fallacies, many of which are brought about by a tendency to separate strategy and context. What works in Guatemala is expected to work in Bosnia, even though the cases differ dramatically in most key respects. Practitioners frequently attribute successful outcomes to their actions and strategies, and attribute failure to the implementation context. Critics, on the other hand, tend to focus on incorrect strategies as causes of failure and attribute success to context or outside factors. This chapter examines three issues associated with the implementation of peace agreements in civil wars. First, we describe some of the difficulties that plague estimating the success of individual operations and how the international system as a whole has performed to date. Second, we conduct a very modest determinants study in order to see which contextual variables are associated with implementation success as well as to describe the difficulties associated with such studies. Finally, we discuss the problems associated with determining and improving the relative effectiveness of different implementation strategies. As we shall see, this raises a host of sensitive issues, many of which are associated with the uncomfortable fact that the strategy and resources needed to succeed in some cases are unavailable because no major or regional power sees it in their interest to provide them. We argue that cases of peace implementation differ dramatically in terms of the difficulty of the implementation environment and in the willingness of international actors to provide necessary resources and troops. We find that three of the most important environmental variables in determining failure are (1) the presence of spoilers (factions or leaders who oppose the peace agreement and use violence to undermine it); (2) neighboring states that oppose the peace agreement and funnel resources to potential spoilers; and (3) the domestic availability of valuable, easily marketable commodities such as gems or timber. With regard to the importance of such commodities, no civil war peace agreement has been fully, successfully implemented where these are present. On the willingness side, we find that the most important variable in accounting for successful implementation is the extent to which a major or regional power defines a particular case as in its national security interest. Only when such interest is present has peace implementation succeeded against the opposing efforts of spoilers and hostile neighbors. Evidence suggests that in the absence of such major or regional power interest, the United Nations can only succeed in the least difficult environments.

Estimating Effectiveness: Measuring Success The first step in evaluating institutional effectiveness is to develop some reliable way to measure implementation success. Traditionally, this prob-

S-2

11/24/09

4:16 PM

Page 45

EVALUATION ISSUES IN PEACE IMPLEMENTATION

45

lem has been approached by developing one or more outcome indicators, each of which measures the extent to which a goal that is considered central to the operation’s success has been achieved. In this section we discuss various measures that have been used to evaluate successful peace implementation and describe their potential pitfalls. We begin by focusing on mandate achievement as a measure of success, and then turn to various objective indicators, such as resolution of the underlying conflict. The evaluation of subgoals such as human rights, elections, disarmament, and refugee repatriation as measures of implementation success is also discussed. Mandate Achievement Until relatively recently, the most common indicator of operation success was the degree to which an operation achieved the goals that were laid out in the UN’s mandate authorizing the mission. The fact that mandate accomplishment is defined by the United Nations itself makes it “endogenous.” This endogeneity may seem attractive, as it protects the measure from criticism that it represents a goal that the agent being evaluated never intended to accomplish in the first place, but this is actually less of a virtue than it might first seem because it makes evaluation the prisoner of ambition. This creates two quite different problems. One is that it allows an institution to be successful by doing nothing more than lowering its goals or by setting very modest goals in the first place rather than by accomplishing something substantive. Agencies and institutions that have considerable autonomy with regard to goal setting but are nonetheless operating in a politically charged environment often do this, particularly when they believe that failure may impact their funding level or future mission.1 A common criticism of the Implementation Force (IFOR) in effecting the Dayton Peace Agreements stems from how the U.S. military limited its mission to separation of the warring armies in Bosnia, and refused to take on the more ambitious tasks necessary for creating peace: the capture of war criminals, the protection of refugees who sought to return home to disenfranchised areas, and the disarmament of the contending armies.2 It makes little sense to declare IFOR a success when its aspirations are so low. While it is sometimes possible for organizations to inflate their success rate by purposely minimizing performance goals, agencies with little autonomy with respect to goal setting suffer from precisely the opposite problem. The political masters of these agencies use them as scapegoats by offloading their responsibilities onto the institution in the form of ambitious goals that the political masters want to get credit for having, but avoid the blame for not achieving. This has been a recurrent problem for the United

S-2

11/24/09

46

4:16 PM

Page 46

EVALUATING IMPLEMENTATION STRATEGIES

Nations in the 1990s as the Security Council and member states have repeatedly demanded that the institution pursue ambitious policies, such as safe havens in Bosnia, or state- and nation-building in Somalia, without providing the requisite resources to succeed. The member states and the Security Council garner symbolic value for proclaiming morally worthy goals, and are able to blame the organization when it fails to achieve them. In either of these instances, the evaluator reaches conclusions about the agency’s success that are uninformative. They provide next to no information about how well the institution performs the tasks that compose its mission relative to other institutions or its own past performance. Perhaps just as bad, using mandate performance as a measure of success keeps evaluators from asking questions about the appropriateness of the mandate, such as the extent to which the mandate is sensible given the institution’s resources—questions that are regularly asked in connection with the performance of economic and military strategy. Mandate accomplishment as a measure of success also creates comparability problems. We cannot overlook the possibility that a mission that is only partly successful in fulfilling an ambitious mandate nonetheless may generate benefits far higher than those brought about by a mission that is nominally more successful by fulfilling a weak mandate. For example, the UN Transitional Authority in Cambodia (UNTAC) has come under extensive criticism for failing to achieve all of the political, military, and economic goals of its ambitious mandate.3 But there can be little doubt that UNTAC accomplished substantially more than did the U.S. mission Operation Support Hope in Rwanda, which, while described by a former high-ranking general as a success based on its limited mandate,4 did nothing to prevent or reduce armed hostilities in Eastern Zaire—hostilities that eventually led to hundreds of thousands of deaths, refugees, and displaced persons, and an interstate war involving the armies of eight different African states. This problem of noncomparability in mandate accomplishment is only exacerbated by the fact that today’s peace operations vary substantially. Peace implementation can include a wide array of tasks. UN peacekeepers have been asked to disarm and demobilize warring troops and repatriate refugees in Cambodia, Mozambique, Guatemala, and Angola; reform legal institutions in El Salvador and Bosnia; investigate human rights abuses in Cambodia, Angola, and El Salvador; hold elections in El Salvador, Mozambique, Namibia, Cambodia, and elsewhere; and oversee truth commissions in El Salvador and Guatemala. In ongoing civil wars where the consent of the warring parties for intervention has been absent—Somalia, Liberia, and Sierra Leone—soldiers have been asked to enforce peace and create the conditions under which peace implementation can go forward. Among the other important limitations of mandate fulfillment as a measure of success is the fact that mandates are often strategically ambigu-

S-2

11/24/09

4:16 PM

Page 47

EVALUATION ISSUES IN PEACE IMPLEMENTATION

47

ous and that this calculated ambiguity increases with the complexity of the operation. Michael Doyle, Ian Johnstone, and Robert Orr argue that experience has taught the military planners connected with large, multidimensional peacekeeping operations to anticipate that the mission will unfold in ways that were not anticipated when the decision to undertake the mission was made: parties will renege on their agreements, the tasks of relocation or disarmament will prove more difficult to achieve, and so forth.5 In order to reserve the capacity to flexibly adapt to these changing conditions, planners want mandates to be more ambiguous than they might be in a simpler, more traditional mission. This is entirely sensible from the planner’s standpoint, but it makes it much more difficult for the researcher to confidently gauge the degree to which a given mission achieves its mandate. Finally, mandates for peace implementation missions in the 1990s have suffered from inflation of goals—goals that have been included in the mandate less for realistic calculation of possible accomplishment, than as a way of appeasing nongovernmental organizations and domestic public opinion. As the 1990s progressed, the lobbying of NGOs, especially those involved in human rights, pressured member states to include more and more subgoals in UN peace operation mandates. The inclusion of such goals as accountability for war crimes, de-mining, stopping the use of child soldiers, improving the plight of women in society, and protecting civilians has become a kind of humanitarian “pork,” with each NGO striving to ensure that its single issue is included in operation mandates. But given the large number of such goals in most peace implementation mandates, and given the limited resources to achieve such goals, implementers inevitably must make decisions about which subgoals are most important to their overriding task of implementing a peace agreement. Nor are all good things achievable in the short term of peace implementation, a problem we will return to shortly in discussing objective indicators of success. Other Measures of Success The limitations of mandate accomplishment as the sole measure of peace implementation success have led researchers to supplement it with other indicators. Marjorie Brown suggests two others: whether or not the operation led to a resolution of the underlying disputes of the conflict, and whether or not the operation reduced conflict in the area.6 Paul Diehl looks at whether the operation limited armed conflict and the extent to which the operation facilitated conflict resolution.7 Duane Bratt uses four indicators of success: mandate performance; the degree of conflict resolution; whether or not the operation kept conflict from spreading in the region; and the extent to which the operation reduced both military and civilian casualties below the rates that existed before the operation began.8 All of these measures of success possess the virtue of being less susceptible than mission

S-2

11/24/09

48

4:16 PM

Page 48

EVALUATING IMPLEMENTATION STRATEGIES

mandate to the problem of endogeneity. They prevent a mission from being declared a success simply because its mandate required very little, or from being declared a failure because it failed to achieve overly ambitious goals, even if it produced significant benefits. The new measures rely less on ambition and more on the kind of tangible accomplishments that dominate estimates of success in most other fields. If a conflict widens and increasing numbers of lives are lost or, conversely, if three democratic elections take place in succession over five years, it will be reflected in the level of success that these measures assign to an operation. Of course, no outcome measure is without its limitations, and these newer and more objective indicators of success have their own problems. One of these has to do with the decision about how the indicators are combined into a scale. This can be done by assuming that the indicators are equally important or by weighing them in importance. The decision is up to the analyst, and the fact that different analysts will decide the issue differently means that estimates about the overall success or failure of a given program are likely to vary more than they would in the case of an individual measure of success such as mandate accomplishment. In addition, a given multiple indicator has its own internal comparability problems in that although two separate missions are assigned the same score, it is based on the achievement of different things. Thus, Bratt, for example, classifies an operation as being partially successful if it achieves any one of three different objectives.9 This raises the potential problem that two partial successes may be as unalike as they are alike, especially with respect to the resources and strategies that were necessary to achieve partial success in the two instances. Another difficulty involves the number and breadth of indicators that are included in the multiple measure. Currently, there is an understandable tendency to include indicators that the analyst believes best characterize the successful elimination of factors that resulted in conflict in the first place. For example, a study carried out by DFI for the Pentagon requires that a successful peace operation lead to economic stabilization as well as to the installation of a second democratically elected government.10 There is no doubt that a peace operation that produced economic stabilization and two democratically elected governments would be more successful than one that simply produced a ceasefire. It is also true that one that produced economic stabilization, a growth rate of 5 percent or better, and three actively contested democratic elections would be better still. By this logic, it follows that the best measure of success and the one that will lead to the most nuanced understanding of what different peacekeeping missions have accomplished is the one that contains the most complete list of conditions believed to prevent the outbreak of conflict and improve the quality of life in a state torn apart by civil war. Yet this is almost certainly not the case. One reason lies in the fact that the passage of time is the enemy of inference. The passage of time intro-

S-2

11/24/09

4:16 PM

Page 49

EVALUATION ISSUES IN PEACE IMPLEMENTATION

49

duces the equivalent of noise in the form of a growing number of influencing factors that are not directly associated with a peace mission. The more time that passes from the conclusion of a peace mission, the more likely it is that any number of other extraneous factors (e.g., business cycles, famines, unusually good or bad weather, the policies of a neighboring state, the behavior of the first elected leaders) are actually responsible for what has taken place rather than the technology of the peace mission itself. As the potential impact of such exogenous factors increases, the quality of our inferences about the contribution of the peace operation itself tends to fall until the inferences reach the point where they break down completely. This is one of the reasons why elementary school educators tend to measure teacher and school performance by indicators such as student reading and math scores rather than by student success in getting into the college of their choice, getting a good job, or having a happy marriage. It is not that the educators value a fourth-grade reading score more than they do the impact of a high-quality university education or professional and personal success; they just believe that any correlation between the quality of fourthgrade instruction and these outcomes is likely to be more spurious than real. Some performance indicators, especially those suggested by scholars of peacebuilding, run the further danger of setting such high standards of success that we are left with a world of undifferentiated failure. Exemplary of this pitfall is the work of Roland Paris, who argues that one should judge the success of UN peace operations in the 1990s by the standard of “creating conditions that will allow peace to endure long after the peacekeepers have left” and asserts that by this standard only one operation—Namibia— can be judged a success. 11 Correspondingly, the list of failures then includes not only unmitigated disasters such as Rwanda and Angola, but also operations where war has ended, such as Mozambique, El Salvador, Bosnia, and Nicaragua. By failing to discriminate between catastrophic failures such as Rwanda and Angola, where hundreds of thousands die, and flawed successes such as El Salvador, Cambodia, and Mozambique, where war has ended but citizens face the mundane reality of life in povertyridden insecurity, Paris holds evaluation hostage to an unreasonable standard of success and ensures that it will yield very little information that can be used to improve future missions. It is not that we should avoid striving to attain good things like economic growth, equitable development, and good governance; it is that these goals form a useless standard for evaluating implementation actions that take place in a short period of time. Subgoal Achievement Given the problematic character of summary evaluations, analysts engaged in the task of assessing peace implementation often change their focus from overall mission objectives to mission subgoals such as demobilization,

S-2

11/24/09

50

4:16 PM

Page 50

EVALUATING IMPLEMENTATION STRATEGIES

human rights, and refugee repatriation because they are more manageable. They require less information because they represent only one small subroutine in a much larger program. They also promise more robustness because there are fewer dimensions that have to be matched to another case. That is, it is frequently more plausible to argue that the task of repatriation in two given states was more similar than was the overall task of peace implementation. There is a limit to the extent precision is possible, of course. The processes connected with subgoals like human rights, demobilization, or repatriation are never exactly the same in any two cases, and an insistence that every detail of two such processes be the same in every case would doom any attempt to formulate generalizations before it got off the ground. Perhaps a bigger danger lies in the fact that different subgoals interact and success in one area can affect success in another area. The importance of this interaction will be missed when analyzing subgoals in isolation. This can lead to any number of mistaken inferences. For example, Human Rights Watch not too surprisingly asserts that the failure to protect and further human rights in Angola best explains the UN’s failure to implement peace there.12 But as Tonya Putnam demonstrates in Chapter 9, the protection and furtherance of human rights during peace implementation rest completely on the ability of the implementers to persuade the warring parties to demobilize their soldiers and turn their armies into political parties. In the absence of progress in those subgoals, implementers have little chance to achieve higher standards of human rights. Moreover, much of what we have said above about mandate achievement applies to evaluation of subgoals. Again, to use the human rights example, the ability of peacekeepers to successfully pursue human rights during implementation differs greatly according to the conflict environment. Pursuing human rights in Somalia, for example, where there was no existing rule of law, no civil society, and no functioning state, and where spoilers sought to destroy the peace agreement, was much more difficult than pursuing human rights in El Salvador, where the two warring parties agreed to make reform of judiciary and police institutions a key aspect of the peace agreement. Operationalizing Success and Failure In this project we chose to operationalize mission success by scoring two variables: (1) whether large-scale violence is brought to an end while the implementers are present; and (2) whether the war is terminated on a selfenforcing basis so that the implementers can go home without fear of the war rekindling. We assessed each case of peace implementation by these measures and then further subjected the cases to an informal, counterfactual analysis about the specific contribution of the implementers to the result. If

S-2

11/24/09

4:16 PM

Page 51

EVALUATION ISSUES IN PEACE IMPLEMENTATION

51

the peacekeepers could not bring large-scale violence to a close, as in Rwanda, Angola, Somalia, or Sri Lanka, the mission was coded as a failure. If the peacekeepers brought the war to a close, but could not leave for fear of the war restarting, as in Bosnia, we coded the case as a partial success. If the peacekeepers brought the war to a close and departed without the war restarting in a two-year period, as in El Salvador, Mozambique, Guatemala, and Nicaragua, we coded the case as a success. We then went further to ask what would have occurred in the program’s absence. It is the difference between this score and what occurred in the presence of the program that represents a mission’s impact. It makes no sense to credit outside implementers with achieving something that would have happened anyway or to blame them for being ineffective when without their efforts the situation would have become far worse. In the absence of an estimate of the counterfactual, the researcher is implicitly assuming that it was the status quo at the time the mission was initiated: everything that happened after that is assumed to have been caused by the implementers. Because the task of estimating the counterfactual in an area where both theory and data are scarce is so difficult, analysts in new fields such as peace implementation tend to omit any reference to it, partly on the implicit grounds that any such estimate would be arbitrary and thus have little credibility anyway. This makes sense given the circumstances, but it risks introducing a significant upward bias in our estimate of how successful a given institution has been in addressing the problem assigned to it. To understand why, one must keep in mind that most large-scale institutional interventions take place in contexts where there is some sort of crisis, such as a recession or epidemic. In the absence of intervention some of these crises will grow increasingly worse (for example, the recession will turn into a depression), but more often than not the situation will eventually get better by itself or “regress to the mean” and the atypical state of affairs is replaced by the typical. Does the lack of counterfactual analysis lead us to credit peace operations too much or to denigrate them unnecessarily? An examination of the sixteen peace operations in this study led us to alter our initial evaluations of two cases. In the case of Liberia, which would have been coded as a success based on the criterion of ending the war on a self-sustaining basis, we downgraded it to a partial success on the basis of evidence that the actions of the implementers actually delayed resolution of the war. In the case of Cambodia, which would have been coded as a failure based on the criterion of having ended the violence, we upgraded it to a partial success because the country was much more secure when the peace operation was terminated, even though violence continued at low levels. Finally, a case could also be made that in three of the cases, Nicaragua, El Salvador, and Guatemala, the wars would have ended at some point in the 1990s without UN involve-

S-2

11/24/09

52

4:16 PM

Page 52

EVALUATING IMPLEMENTATION STRATEGIES

ment in the implementation of the peace agreements; at best, the role of outside implementers was to hasten the end of wars whose termination was ordained by a changed international and regional context. In the end, we rejected reclassifying the cases as less than successes because we felt that by quickening what may have come about in any case, outsiders contributed to saving thousands of lives and helped stabilize the agreements. Evaluating Overall Institutional Effectiveness Evaluation of how the United Nations and/or the international system as a whole have performed to date in peace implementation also suffers from vexing problems. For people who are not very interested in program evaluation, this may not seem too troubling. Can’t one simply pick the set of outcome indicators that one believes is the most informative, measure the extent to which success has been achieved in each implementation mission that has been undertaken, add them up and calculate the average performance measure on each indicator, and then either take the average score over all the indicators or weigh them as one thinks best? The answer is that this is a good place to start, but a potentially dangerous and poor place to stop because it tells us less about the overall success that the UN and the international system have had than one might first think and may be of little predictive value. Evaluation of institutional effectiveness must consider that some missions are more difficult and costly to conduct than others, and the UN is likely to have a lower success rate in these. For example, implementing peace accords in Cambodia, where the fighters grudgingly signed a peace agreement and had over 120,000 men-at-arms, was much more difficult and expensive than implementing peace accords in Guatemala, where the guerrillas were small in number (estimated between 1,500 and 3,000) and were substantially defeated. Ending the civil war in Somalia, where the state had collapsed, militias had proliferated, and none had consented to intervention, was a far more difficult mission than bringing peace to El Salvador, where the warring parties supported UN participation in implementing a settlement that emerged after several years of negotiations and no factions used violence to undermine the peace. The fact that the success rate will vary with the mix of variously difficult missions that are undertaken and that this mix has been changing over time means that any simple estimate of trends in the institution’s average effectiveness will be fundamentally ambiguous. If that estimate shows that the UN has become less effective over time, it may mean that the organization is not functioning as well as it once was, or it could just as easily mean that the organization is functioning better but the level of mission difficulty has increased at an even faster rate. Variance in the difficulty of missions causes the same problems when it comes to forecasting the likely success

S-2

11/24/09

4:16 PM

Page 53

EVALUATION ISSUES IN PEACE IMPLEMENTATION

53

of a mission that the UN is undertaking. Unless the future mission is neither more nor less difficult than the average mission, the forecast will be too high or too low. The connection between mission type and mission difficulty also means that the UN’s overall success rate may depend more on which operations it chooses to undertake than on the strategic decisions it makes about how to conduct its operations. In itself this fact is unremarkable; the history of warfare makes it clear that whom one chooses to fight is at least as important as how one conducts the subsequent war. One should be cautious about interpreting data about trends in UN peacekeeping effectiveness as measured by mission accomplishment unless the nature of the mission is controlled for—something that is difficult to do and rarely attempted. The impact of variation in the difficulty of missions that the UN or the international system chooses to undertake on its overall success rate raises the important related question of how we should deal with peace implementation opportunities that have been selected out by the UN; that is, cases that it has chosen to pass up for whatever reason (e.g., Sudan or Chechnya). This is an important issue that can make a dramatic difference in our estimate of overall success, but the boundaries of what is often referred to as the potential treatment population (i.e., the number used in the denominator of the fraction that represents the program’s success rate) are not always obvious and depend on the use to which the statistic will be put. For example, if we are trying to estimate the effectiveness or success of an employment program, does it make more sense to measure it by the percentage of people graduating from the program who have found jobs, the percentage of those entering the program who have found jobs, or the percentage of those eligible for the program who have participated, graduated, and found jobs? The people who operate the program would always prefer that the first measure be used because the smaller base rate will yield a higher percentage or rate of success. If the program is a pilot project being used to determine how well a certain type of program is likely to work if it is widely employed, the sponsors of those funding the project would find the second statistic the most appropriate because program dropouts represent program failures. If the program is a national one charged with the task of providing the unemployed with job training, voters would probably feel that the third statistic is the one that is important because the program is supposed to serve all of the unemployed, not just some of them. In the case of peace implementation, the standard approach is to look at the success rate of the operations that are undertaken, but the justification for adopting this approach seems largely one of convenience. It implicitly assumes that it makes sense to equate the mission of the United Nations with whatever it chooses to do, and that intervention opportunities that are declined have been declined for good reason. But it is likely that there are

S-2

11/24/09

54

4:16 PM

Page 54

EVALUATING IMPLEMENTATION STRATEGIES

systematic selection effects at work. Research by Michael Gilligan and Stephen Stedman indicates that the United Nations intervenes much less in civil wars in Asia than it does in other regions, and that it avoids civil wars where the government possesses a large army.13 The UN Security Council will never authorize an operation in one of the permanent five member states of the Security Council; civil wars such as Chechnya will never be addressed. The Security Council is loath to authorize peace missions when there is an absence of consent by the warring parties. It is unlikely that the Security Council would intervene in a civil war in a regional power such as India without the consent of that power.

In Search of the Determinants of Success Having devised outcome measures that they understand even if they do not trust them completely, evaluators usually turn to the task of trying to identify contextual factors that explain the variation in institutional performance that they reveal. This is considerably more difficult than devising outcome measures, for any number of reasons: the number of cases is small and difficult to handle statistically; the measurement of potentially important variables, such as extent of state authority and the extent to which parties are genuinely committed to finding a peaceful solution to their conflict, is often difficult and unreliable; and interaction and selection effects create any number of interpretation problems. With respect to interaction, the impact of a given factor is likely to vary across types of missions. To use an obvious example, the commitment of resources that are likely to be present when one or more major powers are interested in having a conflict resolved is likely to be much more important in cases where the implementers attempt to enforce rather than monitor compliance. If we decide not to worry about this and plunge ahead and obtain a regression estimate of the impact of major power interest, that estimate will be determined by the mix of consensual and enforcement missions that has accumulated to that point. Any estimate of the average effect of major power interest will be systematically too high or too low depending on the nature of the mission. To come close to an accurate prediction we would need to either estimate the impact of major power interest separately on consensual and enforcement missions or have a term in our predictive equation that represented the interaction between major power interest and mission complexity. The second technique has the advantage of conserving more degrees of freedom, but in a field where the number of cases is so small, any attempt to statistically estimate interaction is problematic. The related but more elusive problem of selection arises from the fact that peacekeeping choices are the residue of strategies. For example, suppose the United States and its major European allies have a policy of play-

S-2

11/24/09

4:16 PM

Page 55

EVALUATION ISSUES IN PEACE IMPLEMENTATION

55

ing a major military role in the implementation of peacekeeping only in cases where they believe their participation is absolutely essential in order to achieve success. If one is unaware of this fact or ignores it and looks at cases of peace implementation in the 1990s, one would see that those missions where the United States has played a preponderant military role— Bosnia and Somalia—have registered a failure and a partial success. Conversely, one would also see that U.S. military participation has been absent in the clear successes—Mozambique, El Salvador, Nicaragua, and Guatemala. The temptation is to conclude that the participation of the United States in peace operations is either irrelevant or detrimental to success when in fact its participation may have been essential to whatever was achieved in the most difficult cases and might have dramatically improved the results in the less difficult cases that were only marginal successes. In a study such as this one, where the sample is small and composed only of peace implementation missions that were undertaken, theory is too undeveloped for us to analyze the data against the backdrop of a reliable selection equation. The only way to deal with interaction and selection effects is to be extremely cautious in interpreting the findings and try to think about the effects that we might be missing. With that cautionary note in mind, we conducted a rough analysis of the determinants of the successful implementation of peace agreements in the sixteen cases included in the CISAC-IPA study. The general literature on policy implementation provides crucial insight into the conditions under which a peace agreement might succeed or fail. A policy implementation perspective requires that we pay attention to the environment surrounding implementation and recognize that some environments are more conducive to implementation than others. Such a perspective also looks to the coalitions that support implementation and their willingness to invest resources. As a first cut, therefore, our project ascertained what makes some peace implementation environments more difficult and challenging than others. Our second cut was to determine which cases get the most resources. Based on findings from the scholarly literature on peacemaking in civil war, we created a difficulty score for peace implementation based on the following eight variables: the number of warring parties; the lack of either a peace agreement before intervention or a coerced peace agreement; the likelihood of spoilers; a collapsed state; the number of soldiers; the presence of disposable natural resources; the presence of hostile neighboring states or networks; and demands for secession. • The number of warring parties. The difficulty of implementation increases when there are more than two warring parties. 14 Strategies become less predictable, balances of power become more tenuous, and alliances become more fluid. In Cambodia, for example, any action that the

S-2

11/24/09

56

4:16 PM

Page 56

EVALUATING IMPLEMENTATION STRATEGIES

United Nations might have taken against the Khmer Rouge had to be weighed against the effects such action would have had on the National United Front for a Cooperative, Independent, Neutral, and Peaceful Cambodia (FUNCINPEC), which relied on the Khmer Rouge to balance against the State of Cambodia.15 In cases where a proliferation of warring parties occurred, as in Somalia and Liberia, implementers constantly found it difficult to craft solutions that would address the concerns of all the warring factions. Where any factions found themselves excluded, the peace agreement faced their violent opposition. • The absence of a peace agreement signed by all major warring parties before intervention and with a minimum of coercion. The United Nations has usually required a detailed peace agreement among the warring parties—as a sign of their consent to a peace mission—as a precondition for UN involvement. In the 1990s, however, the United Nations intervened in many ongoing wars and, in several instances, either it or a regional organization or a state intervened in the hope of using force to compel a peace agreement: the UN in Somalia, the Economic Community of West African States Cease-Fire Monitoring Group (ECOMOG) in Liberia and Sierra Leone, India in Sri Lanka, the North Atlantic Treaty Organization (NATO) in Bosnia, and Syria in Lebanon. Intervention in the absence of a peace agreement likely will trigger violent opposition by parties who value the preintervention status quo. The absence of a peace agreement implies a lack of problem solving and trust- and confidence-building among the warring factions, thus producing a more difficult implementation environment. • The likelihood of spoilers. The presence of spoilers in peace agreements poses daunting challenges to implementation.16 One critique of the spoiler concept, however, is that spoilers are only recognized after the fact. This criticism can be addressed by attempting to gauge whether prospective implementers judged that they were likely to face violent challenges during implementation. A more sophisticated criticism of the spoiler concept is that potential spoilers are always present and whether an actor actually engages in spoiling behavior depends on the existence of a special opportunity structure.17 There is, as we shall see, some evidence that this is at least partially the case. • Collapsed state. The lack of state institutions and governing capacity places great demands on peace implementers. In addition to bringing fighting to a close, the implementers must create and build up a modicum of state capacity in order for the peace to have a chance to sustain itself. • Number of soldiers. At some level, numbers matter. High numbers of soldiers pose greater demands for verification and monitoring and, hence, a greater potential for successful cheating. Moreover, greater numbers of soldiers require more personnel for monitoring and more resources for demobilization. We scored cases where there were more than 50,000 soldiers as being difficult to implement.

S-2

11/24/09

4:16 PM

Page 57

EVALUATION ISSUES IN PEACE IMPLEMENTATION

57

• Disposable natural resources. If warring parties have access to disposable resources such as gems, minerals, or timber, implementation becomes more difficult. Such resources not only provide armies with a means for continued fighting, they also become the reward against which they weigh the benefits of peace.18 A key difference between Mozambique and Angola is that, in the latter country, the Union for the Total Independence of Angola (UNITA) had access to diamonds, which emboldened its spoiler behavior, whereas the Mozambique National Resistance (Renamo) lacked access to such resources, which effectively limited the benefits of returning to war.19 • Hostile neighboring states or networks. Civil wars rarely take place in otherwise stable regions. As Peter Wallensteen and Margareta Sollenberg observe, many civil wars today intersect with regional conflicts and interstate competition.20 From this it would follow that the attitude of the surrounding states toward a peace agreement in a neighbor’s civil war plays a key role in supporting or undermining the prospects of peace. Spoilers to a peace agreement, for example, are likely to be much stronger and more vocal if they are confident that they can count on neighboring states for sanctuary, guns, fuel, and capital.21 Likewise, in regions where weak states have little control over borders, well-organized private or semiofficial networks can allow neighboring states to take advantage of such state decrepitude to support spoilers in the war-torn country. • Wars of secession. There is a plausible argument that negotiated settlements are more difficult to attain and implement where civil wars are fought over national sovereignty.22 Such conflicts often revert to all-ornothing struggles that make the job of would-be implementers more difficult than in cases where warring parties share a common identity and at least agree on a unitary future for their country. The more these indicators are present, the greater the difficulty of bringing the conflict to an end. The conflict environment, however, is only one aspect of implementation success or failure. International willingness is also crucial; low degrees of interest and commitment lead either to no intervention or, alternatively, to an intervention with an extremely limited strategy set in the sense that implementers will be constrained by the resources they can deploy and the range of subgoals they can pursue. Constraints on the strategy set need not be a problem when the implementation environment is easy. But difficult environments and constrained strategy sets can be a recipe for disaster, for example, when the Security Council authorizes a mission to implement a peace agreement in Somalia or Angola, but does not provide adequate resources. We scored our cases on three indicators of interest and commitment, what we refer to as a “willingness score”: major or regional power interest, commitment of resources, and acceptance of risk of casualties to soldiers.

S-2

11/24/09

58

4:16 PM

Page 58

EVALUATING IMPLEMENTATION STRATEGIES

• Major or regional power interest. A key sign of a high commitment to a mission is whether large, powerful states support intervention and publicly define the conflict as important to their own vital security interests. The more remote a mission is from a powerful state’s vital security interests, the more likely it is being undertaken for symbolic reasons that are unlikely to inspire the outlay of more than a very modest amount of resources. We created a simple estimate of security interest based on whether or not a civil war was geographically proximate to a major or regional power and whether there were statements by major or regional powers at the time of intervention that enunciated security, as opposed to humanitarian, motivations for intervention. • Resource commitment. Crucial for successful implementation is the willingness of states to provide adequate financial resources for a mission. But such resources vary by case, and this is often known before a mission even begins. In Cambodia, for example, the Security Council provided extreme leeway to planners to judge what resources were needed for a successful mission. In Rwanda, on the other hand, the Security Council rejected the figures provided by its own field mission, and instead imposed extreme limits on resources available. • Acceptance of risk to soldiers. We scored this separately from financial resource commitment because policymakers evaluate the risk to their own troops differently than they evaluate financial resources. Table 2.1 lists the cases in our study, the principal implementer, the conflict difficulty score, the international willingness score, and the outcome. The higher the difficulty score, the more difficult the implementation environment; the higher the willingness score, the greater the willingness of international actors to commit to the effort. If we just look at the bivariate regressions of implementation outcome on each of these variables, we find that only four variables are significant at the 0.01 level. These are the existence of a spoiler, the presence of disposable resources, the presence of a neighboring state that is hostile to the peace agreement, and the presence of major power interest. Consistent with expectations, the presence of the first three factors reduces the chances of successful implementation, and the last improves it. The summary variables, conflict and willingness, together explain about 65 percent of the variance in mission outcome. All of the other variables have an impact that is in the predicted direction but not significant with the small number of cases. Since we were curious about the extent to which the role of spoiler was itself an endogenous strategy, the selection of which partly depends on the availability of financial resources, we also looked at the extent to which the presence of spoilers can be predicted by the presence of disposable resources and/or the presence of a hostile neighboring state. The prevalence

S-2

11/24/09

4:16 PM

Page 59

59

EVALUATION ISSUES IN PEACE IMPLEMENTATION Table 2.1

Peace Implementation Cases

Zimbabwe Sri Lanka Namibia Nicaragua Lebanon Liberia Angola I Cambodia Mozambique El Salvador Somalia Rwanda Angola II Bosnia Guatemala Sierra Leone

Time Period

Principal Implementer

Difficulty Score

Willingness Score

Outcome

1980 1987–1989 1989–1990 1989–1990 1990– 1990–1998 1991–1993 1991–1994 1992–1995 1992–1994 1993 1993–1994 1994–1999 1995– 1996–1997 1996–1998

Great Britain India UN UN Syria ECOWAS UN UN UN UN UN UN UN NATO UN ECOWAS

4 6 0 1 5 6 4 5 3 1 5 3 4 6 0 6

1.0 1.7 1.7 1.5 2.7 2.1 0.4 2.2 1.2 1.5 1.4 0.4 0.9 2.2 1.5 0.7

Success Failure Success Success Partial success Partial success Failure Partial success Success Success Failure Failure Failure Partial success Success Failure

Notes: Difficulty score: 0 = lowest, 8 = highest. Willingness score: 0 = lowest, 3 = highest.

of spoilers makes this a little like reading tea leaves, but the presence of disposable resources appears slightly related, and the presence of a hostile neighbor more strongly so. In every case where there is a hostile neighbor, there is a spoiler, and in the six cases where there is no hostile neighbor, spoilers appear only twice. Of course, one cannot overlook the possibility that instead of a hostile neighboring state providing the resources that allow a spoiler to survive, the existence of a spoiler gives a potentially hostile neighboring state reason to believe that it can effectively subvert the peace agreement. Figure 2.1 illustrates the intersection of conflict difficulty, the degree of major or regional power interest, and implementation outcome. The cases of Guatemala, El Salvador, Namibia, and Nicaragua all scored as low on difficulty of implementation contexts and high on great or regional power interest. Not surprisingly, all were successful. Mozambique and Zimbabwe fell in the low-to-medium range of implementation difficulty and, although their interest scores were not as high as the cases mentioned above, their peace agreements were successfully implemented. To the right of the graph are the more difficult cases. Where major or regional power interest was low, as in Angola, Somalia, and Rwanda, or high, as in Sri Lanka, implementation failed. In the most difficult cases, substantial major or regional power interest was enough to compensate for the difficulty of environment to produce partial success in Cambodia, Lebanon, Bosnia, and Liberia, but was not enough to avoid failure in Sierra Leone.

11/24/09

4:16 PM

Page 60

60

EVALUATING IMPLEMENTATION STRATEGIES

Figure 2.1

Interest and Difficulty: Case Outcomes Nicaragua El Salvador 1 ● ● Guatemala

Cambodia Lebanon Bosnia ■ ■

0.8 Namibia ●

Interest score

S-2

Mozambique ● ● Zimbabwe

0.6

Liberia ■ ▲ Sierra Leone ▲ Sri Lanka

0.4 Rwanda ▲

0.2

▲ Angola I Angola II

Somalia ▲

0 0

2

4

6

8

Difficulty score ● success

■ partial success

▲ failure

In terms of interaction, it is interesting to note that both cases where the level of conflict was high (i.e., where the aggregate conflict scores were 5 or 6), and where little major power interest was manifest, were total failures, while only one of the five high-conflict cases where major power interest was present was a total failure. Major or regional power interest may not be able to guarantee complete success in difficult environments, but it is usually able to guarantee at least partial success. More unsettling from the standpoint of the UN as an international body and the evolution away from realpolitik is the fact that in the absence of such interest, there is no evidence of even partial success. One might expect that the role of major powers on the Security Council acts to regularly select out difficult missions in preference to easier and less costly ones. Not much can be said about this in the absence of knowing what proportion of the total number of potential peacekeeping opportunities fall into the challenging category (and knowing whether there are other attributes that seem likely to contribute to the difficulty of a case but are entirely “selected out”), but it is nonetheless the case that almost half of the cases in the sample are classified as difficult. If major powers don’t seem to go out of their way to avoid highconflict cases, they do appear to be slightly more discriminating with respect to which cases are undertaken. The average major power interest in high-conflict cases is 0.72 as opposed to 0.58 for lower-conflict cases, and

S-2

11/24/09

4:16 PM

Page 61

EVALUATION ISSUES IN PEACE IMPLEMENTATION

61

these figures underestimate the extent of discrimination since four of the low-conflict cases are in Central America, where the United States could be expected to be interested in any case. If these are eliminated, the major power interest score of the low-conflict cases drops to 0.38.

Strategies and Implementation Success The majority of peacekeeping researchers and practitioners are less interested in the broad issue of the determinants of successful implementation than they are in the comparative effectiveness of different peacekeeping strategies. Unfortunately, the comparative evaluation of strategies is even more problematic than are determinants studies. One reason for this is that the term strategy can refer to virtually anything, from the UN’s decisions about which missions to undertake, to its decisions about what goals to pursue, to its decisions about how to pursue them. Moreover, while strategies often share much in common, they invariably differ in any number of details. This leaves the evaluator uncertain about what it is that should be evaluated as well as what has been evaluated, and creates a lot of confusion about what the conclusions of a given study actually mean. Another reason why strategic evaluation tends to be more problematic is that selection and interaction effects are often ignored or dealt with poorly. In many ways this makes little sense. The operation of selection and interaction is usually far more obvious in the case of strategy choice than it is in the operation of a given determinant. At an abstract level researchers and policymakers know that strategies are usually tailored to fit a specific context rather than the result of historical serendipity, and that when comparing strategies this makes it difficult to sort out the impact of context and strategy. They are well aware of the fact that education researchers worry that the success rates of top-rated schools or special experimental programs may have more to do with the background of the children who gravitate to good school districts or enroll in voluntary experimental programs than with the quality of the schools or programs themselves. Peacekeeping researchers know they must also worry about whether the greater success rate of one particular strategy stems from its greater superiority or the fact that the missions in which it was employed were easier than those where other strategies were used. Too often, however, individual psychology, organizational interest, and strategic misrepresentation lead policymakers and researchers to forget this. In the case of peace implementation the tendency is to look at the correlation between a grand strategy of confidence-building (reminiscent of traditional peacekeeping) and success in places like El Salvador, Guatemala, Mozambique, and Namibia, and the correlation between a

S-2

11/24/09

62

4:16 PM

Page 62

EVALUATING IMPLEMENTATION STRATEGIES

grand strategy of coercion (compellence or deterrence) and failure in places like Sri Lanka, Somalia, and Sierra Leone and only partial successes in Lebanon and Bosnia, and therefore to assume that confidence-building is a superior strategy to coercion. This, of course, ignores that the cases represent radically different implementation contexts, ignores the deadly failures of confidence-building in Rwanda and Angola, and fails to ask the counterfactual—whether confidence-building would have led to stable peace in Somalia or Sri Lanka. To put it differently, evaluations of peacekeeping in the 1990s often suffered from a belief that the UN should always employ small numbers of lightly armed troops on the basis that the rate of mission failure is correlated with troop strength and the heaviness of armor. There are exceptions, of course, but most of them are relatively recent. For example, both Doyle and Stedman make the normative case that choice of strategy should be determined by the difficulty of the case and the threat posed by spoilers.23 In Doyle’s work, as the environment of peace implementation becomes more difficult, the amount of transitional authority exerted by implementers and their willingness to act coercively must increase. In Stedman’s work, when implementers are faced with violent challenges by powerful leaders or factions who are unalterably opposed to a specific peace agreement or whose demands on the peace process are too high, implementers will have to choose coercive strategies to prevail. Confidence-building is an inappropriate strategy against ethnic extremists who are willing to commit genocide, as in Rwanda, in order to undermine a peace agreement. Nonetheless, a skeleton account of international effectiveness in peace implementation in the 1990s demonstrates a reluctance to connect strategy and context when thinking about the effectiveness of strategies. What jumps out of the data is the fact that the UN’s earliest successes in peace implementation occurred in the least difficult cases: Namibia (1989), Nicaragua (1989), and El Salvador (1990–1994). Each of these cases scored low in conflict difficulty and moderate to high in international willingness. Possibly because there was so little variance in peacekeeping context, lessons that were learned from these cases tended to emphasize the effectiveness of one universally applicable strategy. As a consequence, peacekeepers emerged with the belief that traditional peacekeeping— impartiality, neutrality, and consent—and its theoretical underpinning of confidence-building could be successfully applied to any case or mandate. This overconfidence in the ability of the United Nations to implement peace agreements led to overcommitment. Between 1991 and 1994, the United Nations authorized implementation missions to Angola, Cambodia, Guatemala, Mozambique, Rwanda, and Somalia, devoting little apparent thought to the extent to which these cases resembled those where the earlier successes took place. In short order, and with the exceptions of Guatemala

S-2

11/24/09

4:16 PM

Page 63

EVALUATION ISSUES IN PEACE IMPLEMENTATION

63

and Mozambique, the United Nations found itself in much more dangerous situations, using inappropriate strategies without adequate resources. In Somalia, the one case where the United Nations followed a coercive strategy, the institution interpreted its experience as a failure of strategy— that coercion is incompatible with peace implementation. Just as it ignored context in the case of success, it ignored context in the case of failure. By the end of 1993 the UN updated its selection of strategies based on their relative success—that is, it employed more frequently those strategies that experience had shown to be the most successful. Because the strategy associated with the most difficult case—coercion—had the lowest success rate, the UN found itself applying strategies suited to relatively easy missions in every case, with the effect of reducing the organization’s overall success rate. It was as if a hospital suddenly decided to suspend bone marrow treatments on the grounds that the patients who received them had a higher mortality rate than those who did not without bothering to figure out that part of the justification for giving a patient a very expensive and debilitating cure might be the fact that the patient is much sicker than the average person with the disease. While it is possible to attribute part of the UN’s overconfidence in the traditional peacekeeping model to its string of early successes, the decision to ignore the great contextual differences between Somalia and the other peacekeeping failures on the one hand and the early successes on the other likely stemmed from causes other than a lack of experience or analytic bewilderment. Two other closely related factors are likely to have played a role: the norms of UN decisionmakers and the nature of the UN organizational routines that they create. With respect to their personal value system, individuals are often attracted to a career at the UN in the first place because they esteem neutrality, nonpartisanship, and roles of mediator and peacemaker. They do not want to be viewed as, or view themselves as, part of a partisan militia allied against one side. As a consequence, they have a motivated bias in favor of traditional peacekeeping methods that makes them reluctant to acknowledge the limitations of their preferred intervention technology or the virtues of methods that they abhor and that invariably limit their operational autonomy. Strategic misrepresentation also seems to be operating in some of these cases. As much as UN decisionmakers are drawn to the traditional peacekeeping process, they are even more drawn to ending conflict and playing a central role in conflict intervention wherever it takes place. Both of these goals were placed in jeopardy by the failures of the 1990s, and the UN responded to this crisis in ways that were not always productive. As a consequence of the mounting number of peacekeeping failures and the UN’s failure to protect safe havens in Eastern Bosnia in 1995, leading to the slaughter of 7,000 Muslim men and boys, the United States made it a point

S-2

11/24/09

64

4:16 PM

Page 64

EVALUATING IMPLEMENTATION STRATEGIES

to exclude the UN from any substantive role in implementing the Dayton Accords in Bosnia. Regional intervention—along the lines of Nigeria and ECOMOG in Liberia and Sierra Leone—was advocated as a replacement for UN missions. As the 1990s were coming to a close, a new SecretaryGeneral, Kofi Annan, fought against what he perceived to be the marginalization of the United Nations in peace operations, and the Secretariat fought for inclusion in some of the most controversial operations of the decade: Kosovo, East Timor, Sierra Leone, and the Democratic Republic of Congo (DRC)—all extremely difficult operations that posed major challenges. One can see this as an attempt to ensure that the United Nations would regain a central role in peace implementation, but the fact that at least three of these operations are unlikely to have ever received the attention of any major or regional power without the efforts of the Secretary-General suggests that another motivation is at work. In cases such as East Timor, Sierra Leone, and the DRC, many strategic options that are theoretically available to the UN are not “incentive compatible,” in the sense that they are policies that are not in the self-interest of critical actors to implement. In more conventional language, they are not politically feasible. The fact that great and/or regional powers must play a key role in implementation creates potential incompatibilities between the strategies that the UN might like to see employed and the incentives of the major powers to support those strategies. As the difficulty of the implementation environment increases, there is a need for greater scope and assertiveness of the transitional authority that is supplied by international actors. Similarly, the greater the difficulty of the implementation environment, the greater the need for coercive strategies of implementation. This means that the strategies that are available to international implementers are a function of great and regional power interest: selection acts to truncate the strategy set. When great or regional powers do not see a particular case as affecting their security interests, they do not provide the resources and commitment necessary for coercive strategies. Incentive compatibility not only affects choice of grand strategy, but also the availability of mechanisms that are needed to provide strategic coherence and coordination. As Bruce Jones argues in Chapter 4, the more difficult the implementation environment, the greater the need for strategic coordination. When international actors suffer from a lack of unity or do not fully support an operation, would-be spoilers can take advantage of international splits to attack the peace process and threaten would-be peacemakers. Similarly, the more coercive the strategy, the greater the civil-military tensions in implementation.24 When international troops are at risk, troop-donating states tend to insist on greater day-to-day interfer-

S-2

11/24/09

4:16 PM

Page 65

EVALUATION ISSUES IN PEACE IMPLEMENTATION

65

ence in the political conduct of the mission. Such interference can defeat the ability of an implementation force to diagnose and respond effectively to spoilers. As Jones points out, strategies such as “Friends of the SecretaryGeneral” are fundamentally a function of great and regional power interest. The willingness of states to join a “Friends” groups is indicative of a prior judgment that the specific case is in the state’s interest. The result is that such mechanisms for coordination will be unavailable in some cases: there were no Friends of Somalia, as there were no great or regional powers who had security interests in a peaceful Somalia. Incentive compatibility and the gap between what is needed in some missions and what the great or regional powers are willing to provide has led to organizational pathologies within the United Nations that, unless addressed, may lead to the end of UN involvement in making peace. The Secretariat seems to have learned that there are some cases where the Security Council will only authorize a mission if it perceives that the case will be safe and easy. Where threats or dangers appear, the reaction of the Security Council is to avoid deployment altogether or to cut and run. Faced with such knowledge as early as Rwanda, UN bureaucrats became reluctant to share worrisome conflict assessments and tended only to “ask for what the traffic will bear.”25 This led to UN intervention in cases where, if anything went wrong, the UN would fail. The need to present optimistic scenarios to the Security Council precluded contingency planning, since the basic premise of such planning is that less-than-optimistic scenarios are possible. In some cases the need for the UN to present a successful face to its political masters led missions such as Angola in 1997 and Sierra Leone in 1999 to ignore blatant noncompliance by the warring parties. At its worst, as in East Timor in 1999, the UN Secretariat engaged in reckless gambling by selling the Security Council on a mission by downplaying its difficulty, in the hopes that if it went bad, the Security Council and the member states would be embarrassed into response.

Conclusion On the basis of the findings of this chapter, we recommend the following: • When selecting what peace agreements the UN should implement, objective assessments must be made of the degree of difficulty of the case, the appropriate strategy that is necessary for success, and the likelihood of major or regional powers supplying the appropriate resources. • When selecting what peace agreements the UN should implement, major power or regional power interest should be treated as a hard con-

S-2

11/24/09

66

4:16 PM

Page 66

EVALUATING IMPLEMENTATION STRATEGIES

straint. This means that without major or regional power interest, the United Nations should be extremely cautious in trying to implement cases where spoilers are present, neighboring states are hostile, and disposable resources are available. Experience suggests that UN intervention can make a bad situation on the ground even worse, and failure may both jeopardize the UN’s role in peacekeeping and increase the severity of incentive compatibility problems. • Given the importance of judgment about difficulty of implementation environment, there is a need at the United Nations for better strategic assessment concerning possible peace implementation missions. The Secretariat must act more like Consumer Reports and less like used-car salesmen, attempting to sell the Security Council every mission that arises. • Given the importance of the role that spoilers play in implementation failure, there is a need for intelligence gathering and assessment concerning the motives, intentions, and capabilities of parties who sign peace agreements and parties who are omitted from peace agreements. • Given that such intelligence is fallible and that there will be missions where unanticipated violent challenges may erupt during implementation, there is a need for the United Nations to improve its contingency planning for peace operations. • Before attempting to implement a peace agreement in a country where there are easily marketed valuable commodities (spoils) or that is adjacent to a state hostile to the peace process, the implementer should have the strategy, resources, and commitment to regulate these commodities and the inflow of assistance to spoilers. There is no easy answer to any of the connections between the UN’s success rate and the difficulty of the potential missions that arise. Regardless of how we categorize missions, some are more difficult and costly to conduct than others and the UN is likely to have a lower success rate in connection with these even with selection operating. It follows that if the UN wants to increase its success rate, the easiest way to do so will be to focus on cases where there are only two parties, a functioning state exists, the numbers of combatants are low, and there are no expected spoilers. Is this a sensible strategy? On the one hand, the answer seems obviously no. It seems like the equivalent of a hospital choosing to turn away cancer patients and people in need of transplants in favor of orthopedic procedures and gall bladder operations. On the other hand, if the UN is under constant financial pressure and there are situations where the probability of success is virtually nonexistent, there is no more to be gained by ignoring these facts in the context of peacekeeping than there is in the context of healthcare.

S-2 11/24/09

Coding of Cases by Difficulty and Willingness Variables Difficulty

Bosnia Liberia Sierra Leone Sri Lanka Lebanon Cambodia Somalia Angola I Angola II Zimbabwe Rwanda Mozambique El Salvador Nicaragua Guatemala Namibia

50,000 soldiers yes yes yes yes yes yes yes yes yes yes

Likely Hostile spoilers neighbors yes yes yes yes yes yes yes yes yes yes yes yes

yes yes yes yes yes yes yes yes yes yes

More than two Disposable Weak Collapsed Difficulty Power Resource Risk Interest commitment lives parties resources agreement state Secession scoreb yes yes yes yes yes yes

yes yes yes

yes yes yes yes

yes yes yes yes yes

yes yes

yes

yes yes yes yes yes

yes

6 6 6 6 5 5 5 4 4 4 3 3 1 1 0 0

Notes: a. Outcomes: 0 = failure, 1 = partial success, 2 = success. b. The higher the conflict score, the higher the level of difficulty. c. The higher the willingness score, the higher the willingness of international actors to end the conflict.

1.0 0.7 0.7 0.5 1.0 1.0 0.2 0.2 0.2 0.5 0.2 0.5 1.0 1.0 1.0 0.7

1.0 0.7 0.7 0.5 0.7 1.0 1.0 0.2 0.7 0.5 0.2 0.7 0.5 0.5 0.5 1.0

0.2 0.7 0.7 0.7 1.0 0.2 0.2 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0

Page 67

Civil War

Outcomea

Willingness

Willingness scorec 2.2 2.1 2.1 1.7 2.7 2.2 1.4 0.4 0.9 1.0 0.4 1.2 1.5 1.5 1.5 1.7

4:16 PM

Appendix 2.1

1 1 0 0 1 1 0 0 0 2 0 2 2 2 2 2

S-2

11/24/09

4:16 PM

68

Page 68

EVALUATING IMPLEMENTATION STRATEGIES

Notes 1. George Downs and Patrick Larkey, The Search for Government Efficiency: From Hubris to Helplessness (Philadelphia: Temple University Press, 1986). 2. See Chapter 18 on Bosnia by Elizabeth Cousens in this book. 3. See Chapter 17 on Cambodia by Sorpong Peou in this book. 4. George A. Joulwan and Christopher C. Shoemaker, Civilian-Military Cooperation in the Prevention of Deadly Conflict: Implementing Agreements in Bosnia and Beyond (New York: Carnegie Corporation of New York, 1998), pp. 19, 23–24. 5. Michael W. Doyle, Ian Johnstone, and Robert C. Orr, “Strategies for Peace: Conclusions and Lessons,” in Doyle, Johnstone, and Orr, eds., Keeping the Peace: Multidimensional UN Operations in Cambodia and El Salvador (Cambridge: Cambridge University Press, 1997), pp. 388–389. 6. Marjorie Ann Brown, “United Nations Peacekeeping: Historical Overview and Current Issues,” report for Congress (Washington, D.C.: Congressional Research Service, 1993). 7. Paul Diehl, International Peacekeeping (Baltimore: Johns Hopkins University Press, 1993). 8. Duane Bratt, “Assessing Peacekeeping Success,” in Michael Pugh, ed., The UN, Peace, and Force (New York: Frank Cass, 1997), pp. 64–81. 9. Ibid. 10. Barry M. Blechman, William J. Durch, Wendy Eaton, and Julie Werbel, Effective Transitions from Peace Operations to Sustainable Peace: Final Report (Washington, D.C.: DFI International, September 1997). 11. Roland Paris, “The Perils of Liberal International Peacebuilding,” International Security 22, no. 2 (fall 1997), pp. 54–89. 12. Human Rights Watch, Angola Unravels: The Rise and Fall of the Lusaka Peace Process (New York: Human Rights Watch, 1999). 13. Michael Gilligan and Stephen John Stedman, “Where Do the Peacekeepers Go?” paper prepared for the World Bank, October 29, 2001. 14. Gerardo L. Munck and Chetan Kumar, “Civil Conflicts and the Conditions for Successful International Intervention: A Comparative Study of Cambodia and El Salvador,” Review of International Studies 21, no. 2 (1995): 159–181; and Michael W. Doyle and Nicholas Sambanis, “International Peacebuilding: A Theoretical and Quantitative Analysis,” American Political Science Review 94, no. 4 (December 2000): 779–801. 15. See Chapter 17 on Cambodia by Sorpong Peou in this book; and Stephen John Stedman, “Spoiler Problems in Peace Processes,” International Security 22, no. 2 (fall 1997): 32–36. 16. Stedman, “Spoiler Problems.” 17. Paul Collier and Anke Hoeffler have made a similar argument in connection with the factors that inspire rebellion in the first place. See Greed and Grievance in Civil War, Policy Research Working Paper no. 2355 (Washington, D.C.: World Bank, Development Research Group, May 2000). 18. There is a burgeoning literature on the political economy of civil wars and its effect on prospects for war termination. See, for example, Mats Berdal and David Malone, eds., Greed and Grievance: Economic Agendas in Civil Wars (Boulder, Colo.: Lynne Rienner, 2000). 19. Stedman, “Spoiler Problems,” pp. 47–48. 20. Peter Wallensteen and Margareta Sollenberg, “Armed Conflicts, Conflict Termination, and Peace Agreements, 1989–96,” Journal of Peace Research 34, no. 3 (fall 1997): 339–358.

S-2

11/24/09

4:16 PM

Page 69

EVALUATION ISSUES IN PEACE IMPLEMENTATION

69

21. Stedman, “Spoiler Problems,” p. 51. 22. Chaim Kaufmann, “Possible and Impossible Solutions to Ethnic Civil Wars,” International Security 20, no. 4 (spring 1996): 136–175. 23. See Chapter 3 by Michael Doyle in this book; and Stedman, “Spoiler Problems.” 24. Karen Guttieri, “Civil-Military Relations in Peace Implementation,” draft prepared for this project. 25. See Chapter 16 by Gilbert Khadiagala in this book.

S-2

11/24/09

4:16 PM

Page 70

S-3

11/24/09

4:17 PM

Page 71

3 Strategy and Transitional Authority MICHAEL W. DOYLE

Just as civil war is about failures of legitimate state authority, civil peace is about its successful reconstruction. Transitional authority is about one of the institutions (whether formal or informal) that need to be established in between. Civil wars arise when individuals, groups, and factions discover that a police officer, judge, soldier, or politician no longer speaks and acts for them. Rather than “the local cop on the beat,” the cop becomes “the Croatian, Serb, or Muslim cop.” When the disaffected mobilize, acquire the resources needed to risk an armed contest, meet resistance, and judge that they can win, civil war follows.1 Although we can imagine purely cooperative, interest-based solutions to domestic peace,2 the confusion, “noise,” violence, and changing identification that characterize the onslaught and conduct of civil war do not seem to be promising circumstances for rational cooperation among factions. Instead the establishment of civil peace seems to require addressing directly both the defensive and the aggressive incentives that motivate faction leaders (and sometimes their followers). Defensive incentives arise in the domestic “security dilemma.” Under emerging conditions of anarchy (the collapse of central authority) each group/faction seeks to arm itself in order to be protected; but as in interstate anarchy, each defensive armament constitutes a threat to other factions.3 Offensive incentives arise because factions and their leaders will want to impose their ideology or culture, to reap the spoils of state power, to seize the property of rivals, or to exploit public resources for private gain, or all of the above. Establishing peace will thus also require the elimination, management, or control of spoilers.4 Conquest by one faction can solve the problem (but even in this case reconstruction can be vital for longer-term stability). Peace through agreement can employ territorial separation and partition to address war-prone 71

S-3

11/24/09

72

4:17 PM

Page 72

EVALUATING IMPLEMENTATION STRATEGIES

incentives.5 Civil wars can be turned into international wars or stable and relatively secure international or intercommunal balances of power (as in Cyprus, for example). To each spoiler, his or her separable pile of spoils. But in many civil wars the contest is over who or what “ideology” controls a single polity; and in some ethnic wars the costs of ethnic “cleansing” will seem too high, or a common basis for overarching civic citizenship exists or can be created. In these latter circumstances state authority needs to be cobbled together in order to resolve disputes over material interests, or to decide who or what rules, or to ensure safety. Hobbes’s Leviathan—state sovereignty, or authority—fills that role, restoring “legitimate power.”6 I. William Zartman has argued that this can be achieved by some combination of (1) reconcentrating central power (the powerful must be recognized as legitimate, or the legitimate must be made powerful); (2) increasing state legitimacy through participation (elections, power sharing); and (3) raising and allocating economic resources in support of peace. Given the devastations of civil war; all three generally require (4) external assistance in a transitional period.7 This chapter raises three issues concerning international aspects of transitional authority: the sources of authority, the degree or extent of authority, and the “ecologies,” or varieties, of effective authority. International authority is a key dimension of peace implementation. Adequate authority is not enough to ensure success. Resources, leadership, a dedicated staff, and local cooperation are just as important, if not more so. But international authority offers a license to assist and, if needed, direct, as well as an implicit strategy for aiding the difficult transition from civil war to self-sustaining peace. Without appropriately designed authority, peace implementation is headless.

Sources of Authority A transitional peace operation usually needs two authorizations; one is international, the other domestic. The two need not be always connected. An internationally authorized humanitarian intervention could proceed without host-state authorization (but it will not succeed unless it wins the support of a significant majority of the local population). And a sovereign government can invite foreign forces to assist it without recourse to the UN or a regional organization for authorization. But the two usually are connected. International authority is needed in order to permit the entry of foreign military forces and civilian officials into the domestic jurisdiction of the civil war-torn state. Domestic authority is needed to specify the bases on which the once-warring points come to cooperate and accept common rules for deciding conflicts of interest. It is worth noting that the forcible interventions in Somalia, Haiti, and Kosovo each had prior domestic

S-3

11/24/09

4:17 PM

Page 73

STRATEGY AND TRANSITIONAL AUTHORITY

73

authorizations (albeit each under duress and, in the Somali case, from factions rather than from a functioning national government). From the international point of view, peace operations—which intrude upon the domestic sovereignty of states—come to be established in two ways. First, under Chapter VI of the UN Charter, they emerge from the negotiated consent of the parties and then through a series of Status of Forces agreements that specify the legal terms for the presence of foreign forces. Or, second, they are established under Chapter VII, which permits the overriding of domestic jurisdiction (Article 2-7) without consent of the local parties. These enforcement operations draw upon the authority of Article 42, which permits the Security Council to “take such action by air, sea, or land forces as may be necessary to maintain or restore international peace and security”; Article 25, under which member states “agree to accept and carry out the decisions of the Security Council”; and Article 43, in which they agree to “make available to the Security Council, on its call, . . . armed forces, assistance and facilities.” Troop-contributing countries in these cases negotiate the terms of the participation of their forces either under UN command and thus with the Secretary-General (as in El Salvador or Cambodia), with a regional organization authorized as delegated in Chapter VIII, or with the leader of a multinational “coalition of the willing” authorized under Chapter VII (as was the case of U.S. leadership of the Unified Task Force [UNITAF] in Somalia). Many operations draw on combinations of authorizations: peace treaties among factions, backed up or supplemented by other measures (such as arms embargoes, no-fly zones) authorized under Chapter VII, as did the various UN Protection Force (UNPROFOR) and Implementation Force (IFOR) operations. 8 “Chinese Chapter Seven”—as employed to authorize the use of force for the UN Transitional Administration for Eastern Slavonia, Baranja, and Western Sirmium (UNTAES)—has emerged as a new signal of firm intent to enforce a Chapter VI operation, though in essence it reaffirms the “Katanga Rule” of the UN Operation in the Congo (ONUC): the traditional principle that force can be used in self-defense of both the peacekeeping troops and the mission (mobility of the force). From the domestic point of view, a local authority (or authorities) shares temporarily, and usually conditionally, some of its (or their) own legitimacy with the international peace operation. Domestic authority can be examined in light of the classic types of authorization and “imperative coordination.” Max Weber outlined three ideal types of imperative coordination: traditional, charismatic, and rational.9 The first two types of authority may be rare in civil war transitions. Traditional authority—an established belief in the sanctity of immemorial traditions and of the status of those exercising authority under them—has often broken down. Under the pressures of economic growth and social mobilization, tradition tends to erode and traditional states collapse. Charismatic authority—resting on

S-3

11/24/09

74

4:17 PM

Page 74

EVALUATING IMPLEMENTATION STRATEGIES

devotion to the sanctity, heroism, or exemplary character of the individual leader and the order ordained by him or her—is often in excess supply, claimed by each of the faction leaders. Usually, therefore, rational authority—the legality of patterns of normative rules and the right of those elevated under such rules to exercise command—has to do the work of reconstruction, and often in competition with preexisting but weakened traditional and charismatic sources of authority. Transitional authority must be constructed through painstaking negotiation, implementing widely recognized international human rights norms, and endorsed through negotiated schemes of power sharing or popular elections. It is difficult to imagine the success, limited as it is, that Cambodia has achieved without the leadership of Prince and later King Norodom Sihanouk. He repeatedly served as a catalyst for difficult decisions and a bridge between competing factions that would only contact each other under his auspices. The charismatic authority enjoyed by Nelson Mandela was an equally vital part of the difficult transition in South Africa. Lacking these forms of unifying authority in Somalia, El Salvador, Guatemala, and Bosnia, peace operations had to rely on enforceable or continually renegotiated agreements, which made the quality of international transitional authority a key component of success or failure.

Extent of Transitional Authority Effective transitional authority must be designed to fit the case if it is to succeed in implementing peace. The necessary extent of authority is a function: first, of the local root causes of conflict; second, of the local capacities for change; and third, of the specific degree of international commitment available to assist change. Effective transitional authority must take into account levels of hostility and factional capacities. Whether it in fact does so depends on strategic design and international commitment. Designs for transitional authority incorporate a mix of legal and bureaucratic capacities that integrate domestic and international commitments in a variety of ways. Important lessons can already be drawn from efforts to establish effective transitional authority.10 First, a holistic approach is necessary to end civil wars. Successful exercises of authority require a coordinated approach that draws in elements of peacemaking (negotiations), peacekeeping (monitoring), peacebuilding, and humanitarian action to create a coherent strategy.11 Strategies must be designed to fit the case. Their concepts of operations need to match the particular conflict and be “strategic” in the ordinary sense of that term, matching means to ends. Transitional strategies thus should first address the local causes of continuing conflict and, second, the

S-3

11/24/09

4:17 PM

Page 75

75

STRATEGY AND TRANSITIONAL AUTHORITY

local capacities for change. Effective transitional authority is the residual dimension that compensates for local deficiencies and the continuing hostility of the factions—the (net) specific degree of international commitment available to assist change. We can think of effective transitional authority as authority multiplied by resources multiplied by international institutional capacities. Local root causes, domestic capacity, and effective transitional authority are three dimensions of a triangle, whose area is the “political space”— or effective capacity—for building peace (see Figure 3.1). This metaphor suggests that some quantum of positive support is needed along each dimension, but that the dimensions also substitute for each other—more of one substitutes for less of another; less deeply rooted causes of war substitute for weak local capacity or minor international commitment. In a world where each dimension is finite, we can expect, first, that compromises will be necessary in order to achieve peacebuilding; second, that the international role must be designed to fit each case; and third, that self-determination is not only the right aim, it is the practically necessary aim of building peace. In a small community enjoying a deep and broad sense of affinity, considerable social and full political equality, substantial sources of social capital and wealth, and access to even greater resources from its national capital, peacebuilding is easy. The space for effective action is nearly boundless. Imagine a small European, Japanese, or U.S. town, struck by a tornado, typhoon, or flood. Habits of cooperation, emergency public assistance, and national relief pour in. The disaster is addressed. The community might even be strengthened as it successfully meets a natural challenge.

Figure 3.1

Peacebuilding Triangle

Effective transitional authority (the larger)

Local causes (the shallower)

Domestic capacity (the greater)

Note: Nicholas Sambanis and I have analyzed 124 cases of civil wars since 1944, and we find that this triangle is a useful model for determinants of successful and unsuccessful peacebuilding. See Michael W. Doyle and Nicholas Sambanis, “International Peacebuilding: A Theoretical and Quantitative Analysis,” American Political Science Review 94, no. 4 (December 2000): 779–801.

S-3

11/24/09

76

4:17 PM

Page 76

EVALUATING IMPLEMENTATION STRATEGIES

Imagine now a Cambodian town escaping from the devastation inflicted by the Khmer Rouge, up until recently governed by a force composed largely of Cambodia’s historical enemy, Vietnam, and lacking technical skills, medicine, education, infrastructure. Its national capital, rather than being a source of assistance, is also devastated. National GDP per capita is $200–$300 per year. Here the space for peacebuilding is thin and tenuous. Root causes and present local capacities interact in a competitive way. Many scholars, for example, have agreed that the root causes of the Salvadoran civil war were the militarization of the state and the persistent inequality of the distribution of landed wealth. Historically the two supported each other. When the landed oligarchy (the “fourteen families”) needed to suppress a peasant uprising, the military was available. The military correspondingly enjoyed a first claim on public revenue.12 When the guerrillas of the Farabundo Martí National Liberation Front (FMLN), representing the rural poor, and the government of the Alianza Republicana Nacionalista (ARENA), speaking as the traditional state, came to a military stalemate in the course of the Salvadoran civil war, they began to explore some of the parameters of a future of peace. Each had to compromise in order to arrive at a viable, mutually acceptable, long-term peace. The FMLN was losing the support of the Soviet bloc (with the collapse of the Soviet Union) and it grew to realize that it could not achieve through the peace process the social and economic revolution for which it had fought. The ARENA government refused to negotiate a more egalitarian distribution of wealth. But the government also realized that the traditional autocratic status quo was not something that it could maintain, following the end of the Cold War and the consequent reduction in U.S. support. Both, therefore, compromised on reforming the militarization of the state, one of the two root causes of civil war. The FMLN took the larger gamble, reflecting its weaker position. It gambled that if military and police impunity could be ended, the judiciary made fair and law-abiding, and a free and fair election organized, then it could win its long-term goals through electoral, democratic means. The ARENA government, on the other hand, realized that it no longer needed the military. This was partly because the old system of military protection for landed wealth simply was no longer politically viable given the FMLN’s resistance, but it was also because ARENA had come to realize that it was no longer necessary. The Salvadoran economy had shifted to an economy based on commerce and small industry, and commerce and small industry could survive very well through legal means in a democratic El Salvador. In short, the more powerful ARENA precluded an effort to address the root causes that were more important to the weaker FMLN. But addressing militarization through democratization and reform of the justice system was a compromise both could accept. In Cambodia, the root causes of civil strife were so deep and the local actors so weak that by 1990 each entered the peace as a near equal, each

S-3

11/24/09

4:17 PM

Page 77

STRATEGY AND TRANSITIONAL AUTHORITY

77

having its own form of monopoly power; and so each had to accommodate the others. For much of its recent postwar history Cambodia found itself in a dangerous neighborhood. Bombed by the United States during the Vietnam War, which radicalized the intellectuals and peasantry, it fell prey to the Khmer Rouge in 1975, among the worst fanatics in the second half of the twentieth century. Cambodia was rescued in 1978, but only by its historical enemy, Vietnam; and then it was occupied by Vietnam for a decade. As a result, Cambodia lacked the space in which to address the key challenges of modern development. It has faced crisis after crisis, each before it had time to adjust to or resolve the previous one. Cambodia was simultaneously trying to recover from a combination of trials.13 Cambodia is still seeking to overcome the legacies of colonialism. Indeed, the first generation of postcolonial leadership is still in place. King Sihanouk was first enthroned by the French in 1941. Huge inequalities between city and countryside persist, inequalities typical of export-oriented, metropolitan-based, colonial economic development. Before these inequalities and dependencies had been overcome, the 1978 Vietnamese invasion imposed a new kind of colonialism, as the State of Cambodia (SOC) regime ruled from out of the “knapsack” of Vietnam in 1979 and Vietnam continued to govern from behind the scenes until 1989.14 Second, Cambodia is still recovering from the destruction inflicted by wars, beginning with the U.S. bombing and Khmer Rouge devastations and continuing into the civil wars of the 1990s. All left deep rehabilitation needs, not unlike the needs of countries such as Vietnam and Eritrea. Third, Cambodia, too, suffers from a postholocaust syndrome. The Khmer Rouge massacres left a desperate need for social reconstruction. Only a handful of monks, intellectuals, medical doctors, and trained lawyers survived the Khmer Rouge massacres. A massive social capital deficit resulted and many survivors face deep psychological burdens that discourage reconstruction. Fourth, Cambodia is also a post–civil war survivor from the pitched battles of 1979–1991 between SOC and the unified resistance on the Thai border. Like Mozambique and Angola, the reconciliation and reintegration of 370,000 refugees challenge all the country’s efforts to rebuild. And, fifth, like the economies of Eastern Europe, Cambodia is undergoing a postcommunist transition to a market economy, begun by the SOC in 1991. Any one of these challenges could overwhelm one of the poorest countries of the world. Cambodia is unique in facing them all at once. As a result, each of the Cambodian factions—the SOC imposed by Vietnam in 1978, the Khmer Rouge, and the royalist faction following Prince Sihanouk—lacked either power or wide legitimacy. In 1990, the royalist National United Front for a Cooperative, Independent, Neutral, and Peaceful Cambodia (FUNCINPEC) had traditional legitimacy and the support of the West (but no effective army). The Khmer Rouge had discipline

S-3

11/24/09

78

4:17 PM

Page 78

EVALUATING IMPLEMENTATION STRATEGIES

and guns and the support of China (but, as a perpetrator of genocide, no legitimacy beyond the cadres it controlled). The SOC had an effective bureaucracy and a solid army and the support of Vietnam and the moribund USSR (but it was tainted with its knapsack origins). Each faction had to accommodate the others. The peace recognized each other’s core vulnerabilities. The Khmer Rouge genocide was termed “unfortunate practices of the past.” The factions could not agree on the terms of conjoint rule at the First Paris Peace Conference in 1989. They therefore called in the UN to serve as a transitional authority. In 1991 the great powers pressured the factions into signing a contradictory peace at the Second Paris Peace Conference. The peace consequently continued to reflect each faction’s strategies of victory. The Khmer Rouge hoped the UN control over the SOC bureaucracy would destroy it. The SOC hoped UN demobilization of 70 percent of the armed forces would destroy the Khmer Rouge. Rather than accommodation, the UN transitional authority was designed to hold the factions together long enough to allow the people of Cambodia to construct a new state through a UN-run national election. But the peace failed to attend to the deeper social and economic contexts of inequality and devastation that fed the factional rivalry. Indeed, the measures of rehabilitation that were included in the peace were stymied by continuing factional strife, as the Khmer Rouge soon defected when it realized that the SOC would not fall and the SOC refused to cooperate in key aspects of the peace when it realized that the Khmer Rouge would not be tamed and that FUNCINPEC would seriously challenge its authority in the election. The “war” continued in political form, with violence increasing as the election approached. The electoral victory of FUNCINPEC did not change the balance of power; FUNCINPEC had to accommodate the SOC’s bureaucratic and military capacities and form a coalition government, which unsteadily continues to govern Cambodia today. Transitional authority (or lack thereof) interacts with factional capacities and root causes of factional hostility to shape the legs of the triangular peacebuilding “space” as is illustrated in the Salvadoran and Cambodian cases. Few peacebuilding plans work unless regional neighbors and other significant international actors desist from supporting war and begin supporting peace. The end of the Cold War thus was an important precondition for the bloom of peace operations of the early 1990s. To succeed, transitional authority needs to accommodate the particular circumstances of the conflict, its causes, local capacities, and the quality of the peace agreement. These accommodations can be divided into a variety of ecologies.

Ecologies of Authority Effective transitional authority that governs how peacekeeping, peacebuilding, and peace enforcement are exercised must be designed to fit the case if

S-3

11/24/09

4:17 PM

Page 79

79

STRATEGY AND TRANSITIONAL AUTHORITY

it is to succeed in establishing a self-sustaining peace. The necessary extent of authority is a function, first, of the local root causes of conflict; second, of the local (primarily economic) capacities for change; and third, of the specific degree of international commitment available to assist change. Effective transitional authority must take into account levels of hostility and factional capacities. Whether it in fact does so depends on strategic design and international commitment. Designs for transitional authority incorporate a mix of legal and bureaucratic capacities that integrate domestic and international commitments. Authority operates not upon stable states, but instead upon unstable factions. These factions (to simplify) come in three dimensions. Examining a conceptual map of the post–Cold War world, we can categorize factions as either coherent or incoherent: that is, they do or do not follow the orders of their leaders. They reflect varying degrees of reconciliation or hostility. Having reached a “hurting stalemate,” they accept the process of peace; or (having been dragged to the conference table) they do not. And these factions are few or many.15 They also are in conflict in societies that either possess little economic and social capacity (LDCs—least developed countries in GDP, education, etc.) or have more (MDCs—more developed countries in those capacities). When one examines the mix of these factors, one can think about differing “ecologies” of transitional authority during peace implementation that represent differing combinations of those three sets of conditions and differing levels of international response to them (see Figure 3.2).16

Figure 3.2

Five Ecologies of Transitional Politics More hostile factions Few

Coherent factions

Incoherent factions

Many

More reconciled factions Few

THIRD

FIRST

Cambodia, Angola, Bosnia I, Georgia, Cyprus, Western Sahara

El Salvador, Namibia, Tajikistan

FOURTH

FIFTH

SECOND

Rwanda, Eastern Slavonia, Brcko

Somalia, Democratic Republic of Congo, Bosnia II, Congo, Liberia, Sierra Leone

Mozambique, East Timor, Haiti, Guatemala

Many

Note: Cases cover civil wars, not interstate wars or invasions by foreign forces (as in Lebanon, for example).

S-3

11/24/09

80

4:17 PM

Page 80

EVALUATING IMPLEMENTATION STRATEGIES

In the first ecology of peace implementation, factions are few, semireconciled, and coherent and capacities are substantial or minimal. And in the usual more or less ways, the case of El Salvador (more capacity) and Namibia (less capacity) fall into that pattern.17 The authority often relies on specific commitments made in the peace treaty by established governments, supplemented by new transitional institutions and an international peace operation. In El Salvador, for example, the government undertook a variety of commitments to engage in judicial, police, military, electoral, and other reforms. The peace agreements also created new transitional institutions, including the National Commission for the Consolidation of Peace (COPAZ), which was designed to promote dialogue between the business, labor, and other elements of civil society. The UN Observer Mission in El Salvador (ONUSAL) was charged with monitoring and assisting the peace process. Transitional authority, light as it is, still has a vital role to play in implementation in those circumstances. First, as in coordination games, it can create and needs to create transparency. In addition, although the factions may be reconciled, they don’t fully trust each other. The international role consists of monitoring and investigating in order to increase trust so that the parties can believe that the piece of paper they signed has operational significance. In El Salvador, ONUSAL helped to increase trust and transparency through the Ad Hoc Commission, which supervised demobilization, and through the Truth Commission, which investigated human rights violations and recommended reforms. Second, in these circumstances the international implementers can also offer capacity-building. They can bring in the technical assistance that the parties either lack or don’t quite trust one another to provide, such as electoral assistance or police training. Third, and perhaps most important, the implementers provide insurance of continuing coordination. No matter how well designed the peace treaty happens to have been, and despite whatever reconciliation of the parties may have occurred, the parties know that circumstances will arise that were not anticipated in the treaty. Those circumstances will need to be dealt with if the peace process is to be kept on track.18 In the second ecology, factions are few and reconciled, but they’re incoherent. In Guatemala, exhaustion and international pressure brought the government and the indigenous communities to the peace table, but soon thereafter the indigenous communities lapsed into their many local components, the thousands of communities from which war had mobilized them.19 In Haiti and East Timor, war and intervention radically reduced the influence of the opposition, as General Raoul Cédras’s forces fled Haiti and the Indonesian militias left East Timor. This left one massive “faction,” destitute and lacking in coherent organization. Here factions may be incapable of fulfilling their commitments, even if willing. In Mozambique, the UN mission there (ONUMOZ) appears to have stepped in to play an active,

S-3

11/24/09

4:17 PM

Page 81

STRATEGY AND TRANSITIONAL AUTHORITY

81

quasi-sovereign, implementing role through a variety of commissions for disarmament, elections, and humanitarian activities.20 ONUMOZ actually helped to organize a political party as well as to employ demobilized soldiers in building roads, a true effort at building capacity and infrastructure. Proactive peace management oriented toward capacity-building was important in both respects, employing the former soldiers and building a transportation grid. In the third ecology of peacebuilding, factions are few, hostile, and coherent. Both Cambodia and Angola fit this ecology, where the factions were and still are hostile and the country is poor, as does Bosnia, where the factions remain very hostile in a country that has greater social and economic capacity. (In Bosnia, Muslim-Croat relations resemble Cambodia’s SOC-FUNCINPEC relations; Federation-Serb relations resemble SOCFUNCINPEC relations with the Khmer Rouge or Angola I or II—which collapsed.) In this third ecology the peace implementation role includes all the functions that were exercised in that first, happier world of UN peacebuilding—the Salvadoran case—where transparency, coordinating insurance, and capacity-building are the keys. But over and above that, because (as in cooperation games) the factions are less than reconciled (the mildest description of the Cambodian experience), the peace process needs to embody more substantial transitional authority if it is to have a chance of success. No faction trusts the established government; alternatively put, the established government is nothing more than another faction, as was the “State of Cambodia.” Peacebuilding design may thus call for transitional sovereignty institutions, as in the Supreme National Council (SNC) of Cambodia, to which the sovereignty of Cambodia was temporarily entrusted. The SNC represented each of the factions, with Prince Sihanouk in a trustee, titular head-of-state role. International commitment is needed to glue the transitional institutions together and provide economic assistance. The UN Transitional Authority in Cambodia (UNTAC), under Special Representative Yasushi Akashi, was given the authority to decide—when the factions were deadlocked and Prince Sihanouk did not act. UNTAC also received the authority to “control” the administrative activities of the factions (most relevantly the SOC, which alone had substantial administrative capacity) in five areas of sovereign activity (finance, foreign affairs, national defense, information, and public security). Carrots and sticks are needed to supplement legal capacities. In the Paris Peace Agreement economic rehabilitation assistance was designed for, and only for, those factions that would cooperate within the peace process. Given that their former patrons had blocked the factions from obtaining financing, it was thought that this would be a very powerful constraint on defection and an incentive to cooperate in the peace process.

S-3

11/24/09

82

4:17 PM

Page 82

EVALUATING IMPLEMENTATION STRATEGIES

Unfortunately, one and then another of the factions discovered alternate sources of financing through illegal sales of gems and logs and other means that removed this particularly important carrot and stick from the peace process. In addition, in this less happy third ecology the international community has a very important role in direct implementation. The residual hostility of the factions means that they will not trust each other to implement any crucial element of the peace process. In Cambodia, it was absolutely vital that the UN itself had the authority to organize the electoral process from the ground up.21 An election run by one of the factions and only monitored by the UN (as in El Salvador) would have been prone to severe exploitation or manipulation. Instead, the UN organized and ran that election, giving more parties authentic access and guaranteeing a much fairer count of the vote. Despite this substantial authority, there is a growing impression that not enough peacebuilding occurred. 22 Significantly, the UN Angola Verification Mission (UNAVEM) lacked this kind of authority in Angola, and failed. In the early stages of the implementation of the Dayton Agreement on Bosnia, as Elizabeth Cousens illustrates in Chapter 18, the transitional authority available to make the parties comply with their treaty commitments was insufficient. Levels of hostility remained high after the signing of the accords: “brought to the table by varying forms and degrees of coercion, the parties had little more than a tactical commitment to settle, making any resulting accord dependent on more than the will of the parties for its implementation.”23 In this strategic situation, both formal and effective authority, including the will to use them, are needed. In the military sphere, both formal authority and an effective presence on the ground were much in evidence, and the successful separation of forces resulted. However, the narrow interpretation of what constituted security, the weak coordinating authority and “interpretive” authority given the high representative, and the lack of coordination between the civilian and military pillars of the Dayton process had debilitating effects on “civilian” implementation. Refugee return to areas in which they would be a minority was stymied and the forces that had led the campaigns of exclusion and violence during the civil war stayed in power. International civilian authority, moreover, was both unclear and divided. While the Office of the High Representative (OHR) coordinated many organizations, its authority vis-à-vis the parties and those other implementing organizations was underspecified. The OHR, the Organization for Security and Cooperation in Europe (OSCE), and others were empowered to make determinations of compliance with the Dayton Agreement but lacked quasi-sovereign authority to make determinations of policy when the parties disagreed.24 The mandate in the first phase was interpreted more along lines suited to the degree of genuine reconciliation evidenced in the Salvadoran or Namibian peace operations, rather than to

S-3

11/24/09

4:17 PM

Page 83

STRATEGY AND TRANSITIONAL AUTHORITY

83

the hostility that characterized the Bosnian factions. Only with the Bonn Summit of the Peace Implementation Council of December 1997 did the OHR begin to acquire the internationally recognized authority to take decisions against the will of the parties. As the factions themselves began to splinter, raising the need for authoritative coordination, the OHR attempted to manage the most blatant spoilers. It instituted neutral license plates, closed hostile media transmitters, targeted the more violent cantonal police forces, and more actively supported minority refugee returns. In the fourth ecology, where the factions are few, incoherent, and hostile, the prospects of implementing peace are extraordinarily small. Only exceptional multilateral and international commitment might succeed in overcoming incentives for resumed armed conflict. In Eastern Slavonia, the UN acquired “executive authority” through the Erdut Agreement that gave the transitional administrator, Jacques Klein, the authority to implement the agreement without consent of the Croatian government or the Krajina Serb entities. Disputes arose over whether or not and by what that authority was constrained. On the one hand, authorities in Zagreb argued that Croatian law constrained UNTAES. On the other hand, certain UNTAES lawyers argued that executive authority was constrained only by international human rights and humanitarian law.25 Equally noteworthy in the UNTAES operation, however, is the substantial military force at its disposal both locally (in UNTAES) and on call (from IFOR/SFOR). Ongoing operational consent, too, may have played a more significant role than legal mandates might suggest; UNTAES officials stressed the occasional cooperation they received from Presidents Franjo Tudjman (Croatia) and Slobodan Milosevic (Serbia) as well as from various local actors, both official and unofficial. At the extreme, as a potential solution to hostile factions with either coherent or incoherent leadership, the international community has established “supervisory authority”—fully sovereign rule. Limited so far to the municipality of Brcko in northern Bosnia, U.S. ambassador William Farrand ruled according to the Arbitral Order authorized by the Dayton Agreement, exercising with the assistance of SFOR troops fully sovereign authority. So far, Brcko is the only municipality with significant minority returns and the beginnings of a functioning multiethnic police, judiciary, and town council.26 Following the final arbitral award, in 1999, Brcko was established as an autonomous district, separate from both the Republika Srpska and the Bosniac-Croat Federation, in effect a third entity. To its international administrators, Brcko is an experiment in whether concentrated international authority and substantial international capacity can begin to build self-sustaining, multiethnic peace. How peace came to the small village of Klanac, just outside Brcko, can illustrate the role effective international authority can play. In May 1992, Serb paramilitaries “ethnically cleansed” the village of Klanac, a Bosniac

S-3

11/24/09

84

4:17 PM

Page 84

EVALUATING IMPLEMENTATION STRATEGIES

suburb. Following the Dayton Agreement, the Bosniac “displaced persons” (DPs, or internal refugees) repeatedly attempted to return in order either to reclaim their intact homes or rebuild the destroyed ones. As has happened many times elsewhere in Bosnia, the hopeful returnees were met each time with a hail of stones thrown by present residents. Few were surprised by the clashes. The Serbs in Klanac were thought to be the most hard-line opponents of reintegration. But the Serbs were also victimized refugees, another displaced community who feared being displaced once again. But in mid-May 2000, less than a month after the latest clash, sixty Bosniac families began work on their houses in the neighborhood. Serbs and Bosniacs formed a neighborhood committee. Serbs have expressed a willingness to vacate the houses they are occupying and Serbs are helping over 700 Bosniacs return. What happened? Part of the credit belongs to the international community. The new Brcko District of northern Bosnia is special. The sticking point at the Dayton negotiations, Brcko is both the strategic corridor linking the two halves of the Serb Republic “entity” and the northern access route from the Bosniac-Croat Federation “entity” to Croatia and Europe. Irresolvable at Dayton, the problem was handed over to international arbitration. In a Solomonic decision in 1999, the international arbitrator, U.S. attorney Roberts Owen, made it an autonomous district of Bosnia, owned by both entities but by neither exclusively. Earlier, he had given the nearly limitless “supervisory” authority described above to implement the arbitration to Ambassador William Farrand, the international Supervisor of Brcko. Relying on this authority, the backing of nearby SFOR troops, and the assistance of the UN police monitors, Farrand established the only functioning multiethnic administration and police in Bosnia. The district gave the displaced Serbs a sense that they could find a new home and be safe and not be forced back into the Federation. It was the multiethnic local police that quelled the last Klanac riot. The other part of the credit, of course, belongs to the courage and common sense of the Klanac residents, both Bosniac original and Serb current. Manipulated for years by their hard-line DP organizations and the ethnic political parties that relied on them for cheap votes, both groups of DPs stood up for themselves and stretched a hand across the ethnic divide when they saw a way to live together safely. Taking advantage of an offer from Supervisor Farrand, the Serbs agreed to vacate the Bosniac houses they occupied in return for free and secure land plots elsewhere in the district. When the DP leadership organizations balked at this sensible compromise and the new local District Assembly hesitated to pass enabling legislation, the current Serb and prospective Bosniac residents threatened to organize a multiethnic demonstration. (This surely would have been Bosnia’s first.) The assembly voted wisely and Klanac is now at peace.

S-3

11/24/09

4:17 PM

Page 85

STRATEGY AND TRANSITIONAL AUTHORITY

85

The struggle for a sustainable peace is far from over. Even in Brcko, unemployment stands at 60 percent, organized smuggling is rampant, ethnic tensions still simmer and sometimes boil over, and thousands more seek a return to their homes. But Klanac was an important step. In the fifth ecology, where there are many incoherent and hostile factions in a desperately poor economy, the prospects appear to be even more grim for effective peacebuilding. For example, in Somalia, what may have been needed was a partition and, for the remainder of the country, a substantial civilian and developmental effort with a long time horizon and trusteeship-like authority. Instead the international community offered a military mission with a constant eye for a fast exit.27 There thus appears to be a relation between the depth of hostility and the number and character of the factions, on the one hand, and the extent of effective authority needed to build peace, on the other. There is a functional progression from ONUSAL’s monitoring/assisting, to UNTAC’s “administrative control,” to UNTAES’s “executive authority,” to a Brcko-style sovereign “supervision.” Authority greater than monitoring/facilitating would have been redundant in El Salvador; authority less than supervisory and sovereign in Brcko would be insufficient. We can think of this as a simple relation in which, when we place transitional authority conceived of as legal authority and effective international capacity (troops and budget) on a the vertical axis and the various “ecologies” described above on the horizontal axis, a progressive relation holds, as in Figure 3.3. Successful operations (those that lead to an indigenously sustainable peace) demonstrate a one-to-one relation: the more challenging the factional conflict, the more transitional authority seems to be required. Optimistically, we can note that the international community seems to be learning past lessons: in the recent cases and trying circumstances of Kosovo and East Timor, extensive authority has been provided. In the equally if not more trying Congo, modest promises of monitoring as a step toward a negotiated peace are being offered. No one promises nationbuilding or peace enforcement or humanitarian protection, as was done in Somalia and Bosnia, without either the authority or the resources to make those commitments effective. This is a wise development. And while muddling and innovation toward enhanced authority on the Bosnian model is better than stagnation, ideally such authority is written into the peace agreement or imposed by Security Council fiat at the outset of a mission, rather than retrofitted after earlier efforts fail. There are few takers for the colonialist role in the late twentieth century. Since all transitional authorities must end, authority and the capacity to make it effective should fit the case in order to increase the prospects for a successful transition toward self-sustaining peace.

S-3

11/24/09

4:17 PM

Page 86

86

EVALUATING IMPLEMENTATION STRATEGIES

Figure 3.3

Transitional Authority

I Supervisory authority

II

III

UNTAES

Adminstrative authority

ONUSAL UNTAG UNMOT

V

UNMIK Brcko Arb.

UNTAETa

Executive authority

Monitoring/ facilitation

IV

ONUMOZ

UNTAC Bosnia I MINURSO

UNMIH MINUGUA

UNAVEM UNOMIG UNFICYP

Bosnia II UNOMILb

UNOSOM II

UNAMIR

UNOSOM I MUNOC UNPROFOR UNOMSIL

Notes: Italics indicates a recent mission; plain text indicates an ongoing mission; bold indicates the mission was a success; bold italics indicates the mission failed. a. UNTAET is a curious and exceptional case of excessive international authority. The Timorese factions are more than ready to cooperate according to a number of accounts, but the UN is insisting on full sovereign control. See Jarat Chopra, “The UN’s Kingdom of East Timor,” Survival 42, no. 3 (autumn 2000): 27–39. b. UNOMIL, the UN monitoring operation, was paired with and monitored an ECOMOG operation with authority to maintain order, which it achieved in a sporadic fashion.

Notes 1. Ted Gurr, Minorities at Risk (Washington, D.C.: U.S. Institute of Peace, 1993). For a thorough review of the field, see Steven David, “Internal War: Causes and Cures,” World Politics 49, no. 4 (July 1997): 552–576. 2. See, for example, James Fearon and David Laitin, “Explaining Interethnic Cooperation,” American Political Science Review 90, no. 4 (December 1996): 715–735. But note that Fearon and Laitin are not, nor do they claim to be, explaining the empirical record of domestic peace. They acknowledge that state power and domestic authority are alternative explanations (see p. 731). 3. Barry Posen, “The Security Dilemma and Ethnic Conflict,” in Michael E. Brown, ed., Ethnic Conflict and International Security (Princeton: Princeton University Press, 1993), pp. 103–124; and Chaim Kaufmann, “Possible and Impossible Solutions to Ethnic Civil Wars,” International Security 20, no. 4 (spring 1996): 136–175. 4. Stephen John Stedman, “Spoiler Problems in Peace Processes,” International Security 22, no. 2 (fall 1997): 5–53. 5. Kaufmann, “Possible and Impossible Solutions.”

S-3

11/24/09

4:17 PM

Page 87

STRATEGY AND TRANSITIONAL AUTHORITY

87

6. The Oxford English Dictionary defines authority as “right to command,” “power to influence action,” “power over the opinions of others.” An enlightening essay is Hannah Arendt, “What Is Authority,” in Arendt, ed., Between Past and Future (New York: Viking, 1961), pp. 91–141. 7. I. William Zartman, “Putting Things Back Together,” in Zartman, ed., Collapsed States: The Disintegration and Restoration of Legitimate Authority (Boulder, Colo.: Lynne Rienner, 1995), pp. 267–273. 8. For a valuable discussion of the international law on the use of force and its bearing on authority for peace operations, see Karen Guttieri, “Symptom of the Moment: A Juridical Gap for U.S. Occupation Forces,” International Insights 13, special issue (fall 1997): 131–155. 9. Max Weber, The Theory of Social and Economic Organization, trans. by M. Henderson and Talcott Parsons, ed. by Talcott Parsons (New York: Macmillan, 1947), pp. 324–333. 10. See Thomas Franck, “A Holistic Approach to Peacebuilding,” in Olara A. Otunnu and Michael W. Doyle, eds., Peacemaking and Peacekeeping for the New Century (Lanham, Md.: Rowman and Littlefield, 1998), pp. 275–295. 11. See Alvaro de Soto and Graciana del Castillo, “Obstacles to Peacebuilding in El Salvador,” Foreign Policy 94 (spring 1994): 69–83. This is the coordinating role that Japan, for example, played in Cambodia in organizing the Tokyo conference and the International Committee on the Reconstruction of Cambodia (ICORC). 12. Edelberto Torres Rivas, “Civil War and Insurrection in El Salvador,” in Michael W. Doyle, Ian Johnstone, and Robert C. Orr, eds., Keeping the Peace: Multidimensional UN Operations in Cambodia and El Salvador (Cambridge: Cambridge University Press, 1997), pp. 209–226; and Tommie Sue Montgomery, Revolution in El Salvador: From Civil Strife to Civil Peace, 2nd ed. (Boulder, Colo.: Westview Press, 1995). 13. In Chapter 17 of this book, Sorpong Peou offers a perspective sympathetic to each of the four major factions. 14. For a thoughtful discussion of Cambodia’s political legacy, see Aun Porn Moniroth, Democracy in Cambodia: Theories and Realities, trans. by Khieu Mealy (Phnom Penh: Cambodian Institute for Cooperation and Peace, 1995). 15. In Michael W. Doyle and Nicholas Sambanis, “International Peacebuilding: A Theoretical and Quantitative Analysis,” American Political Science Review 94, no. 4 (December 2000): 779–801, we found the relationship between factions and the difficulty of peacebuilding to be nonmonotonic, in fact Ushaped. “Few” make coordination problems less difficult; “many” (up to five) make them more difficult; “very many” make coordination easier, as perhaps crosscutting coalitions emerge and factional salience declines. 16. I use the term ecology as a variety of the “worlds” analogy employed by Robert Jervis in “Cooperation Under the Security Dilemma,” World Politics 30, no. 2 (January 1978): 167–214. 17. Lise M. Howard, “Namibia,” a draft prepared for this project, discusses the complexities of the Namibian case and makes a good case for its successful transition, noting two free and fair elections, little crime, and a white minority that accepts the current regime. The slow pace of economic development, however, is putting considerable strains on the political regime. 18. When it was discovered that one of the factions of the FMLN had a weapons cache, ONUSAL impartially investigated and then dismantled the cache. When it was discovered, in November 1993, that the death squads seemed to be reemerging, many asked whether the government was behind them. The UN was able to investigate, enjoying the trust of the FMLN that it would do as thorough a job as could be done. See Ian Johnstone, Rights and Reconciliation in El Salvador

S-3

11/24/09

88

4:17 PM

Page 88

EVALUATING IMPLEMENTATION STRATEGIES

(Boulder, Colo.: Lynne Rienner, 1995), and Chapter 14 on El Salvador by Charles Call in this book. 19. In Chapter 15 of this book, William Stanley and David Holiday note both the remarkable and contradictory emergence of both Mayan participation in the political process and the lack of substantive improvement in civic governance. 20. Richard Synge, Mozambique: UN Peacekeeping in Action, 1992–94 (Washington, D.C.: U.S. Institute of Peace, 1997). 21. Michael Doyle, UN Peacekeeping in Cambodia: UNTAC’s Civil Mandate (Boulder, Colo.: Lynne Rienner, 1995). For a discussion of UNTAC by an experienced international peacekeeper, see Reg Austin, “New Forms of International Intervention: The United Nations Military-Civilian Intervention in Cambodia,” in Mary Kaldor and Basker Vashee, eds., Restructuring the Global Military Sector, vol. 1, New Wars (New York: Pinter, 1997–1998), pp. 231–257. 22. For an assessment, see International Crisis Group (ICG), “Cambodia: The Elusive Peace Dividend,” ICG Asia Report no. 8 (Phnom Penh/Brussels: ICG, August 11, 2000). 23. See Chapter 18 by Elizabeth Cousens in this book. 24. Nonetheless, as in Cambodia, there were narrow areas of policy in which the international community possessed effective transitional authority. The International Monetary Fund (IMF), for example, had tie-breaking authority on the governing board of the central bank and the Council of Europe appointed a majority of the members of the Human Rights Chamber. 25. Author interviews in Vukovar, July 1997. 26. Author interviews in Brcko, June 1999 and June 2000. 27. For an outline of “conservatorship,” see Gerald Helman and Steven Ratner, “Saving Failed States,” Foreign Policy 89 (winter 1992–1993): 3–20; and Ken Menkhaus, “International Peacebuilding and the Dynamics of Local and National Reconciliation in Somalia,” and Thomas Weiss, “Rekindling Hope in Humanitarian Interventions,” in Walter Clarke and Jeffrey Herbst, eds., Learning from Somalia: The Lessons of Armed Humanitarian Intervention (Boulder, Colo.: Westview Press, 1997), pp. 43–63 and 207–228.

S-4

11/24/09

4:18 PM

Page 89

4 The Challenges of Strategic Coordination BRUCE D. JONES

International actors face recurrent challenges of coordinating their approaches and their efforts to implement peace agreements—in short, challenges of strategic coordination. Their efforts to end civil wars suffer from an inconsistency in conflict management strategies across different phases of the peace process; those who mediate agreements sometimes fail to coordinate with those who must implement them. All too often, different actors pursue divergent strategies within a given phase of the peace process. Or, when they do agree on a strategy, their efforts to operationalize it are at times diffuse and contradictory. Strategic coordination is particularly vexing for peace implementation and for postconflict peacebuilding because many more actors are engaged in implementation than in negotiation and the international policy process takes on much greater complexity. In difficult conflict environments, such problems make peace efforts vulnerable. They create opportunities for opponents of peace to maneuver between the cracks of a diffuse implementation strategy, to manipulate implementers against one another, and, sometimes, to derail a peace process altogether. In less difficult conflict environments, such problems may not be as fatal, but they can add costs, reduce effectiveness, and slow success. In short, incoherence and inconsistency can undermine the viability or the effectiveness of implementation efforts. Several past cases—including Guatemala, El Salvador, and Mozambique—demonstrate that the challenge of strategic coordination can be overcome by the use of coordination mechanisms such as Friends groups or by major powers taking a lead role in conflict management. Two aspects of these cases, however, powerfully circumscribe their applicability elsewhere. First, in each case of effective coordination, no more than a handful of actors were directly involved in political negotiations, and even 89

S-4

11/24/09

90

4:18 PM

Page 90

EVALUATING IMPLEMENTATION STRATEGIES

in the assistance aspects of peace implementation, the number of international and regional organizations was relatively small. Second, two necessary conditions for effective coordination were in place: a high degree of international commitment and a rough correspondence of interests of the major powers. The experiences of other, more recent, peace missions—Sierra Leone and Kosovo prominent among them—indicate that strategic coordination is a growing policy challenge for peace implementation. This is due to institutional proliferation at both the official and the unofficial levels, and also to the weakened authority of the UN relative to other multilateral bodies. The political negotiation field is increasingly characterized by the same phenomenon that complicates the assistance dimension of postconflict peacebuilding: namely, a multiplicity of actors with overlapping mandates, competitive relations, and minimal accountability for performance. Recent efforts to enhance structures for strategic coordination on the ground, both within the UN and beyond, have been frustrated by the sheer numbers of actors involved, the limited extent to which these actors accept the coordinating authority of the UN, or analogous body, and the absence of policy coordination structures at the headquarters level.

A Recurrent Policy Problem: Evidence from Past Cases That a lack of strategic coordination is a substantial obstacle to effective conflict management can be demonstrated by reference to three of the most deadly civil wars of the 1990s: Bosnia, Rwanda, and Burundi. Each of these cases illustrates a different type of coordination problem: diffuse intervention efforts in Bosnia; conflicting strategies in Rwanda; and fragmented international responses in Burundi. Depending on the difficulty of the implementation environment, however, any one of these factors can be sufficient to fatally undermine the prospects for peace. Bosnia: Divergent and Diffuse Efforts Among the myriad difficulties encountered in the Bosnia mediation and implementation process, one of the more trenchant was the question of coordination among the multitude of international organizations, donors, nongovernmental organizations, and others involved in various tasks leading up to or stemming from the Dayton Accords. Elizabeth Cousens highlights the problem of diffuse mediation strategies, characterized by “political mudslinging and competition,” and argues that only when the United States began asserting itself in 1995 did the lack of cohesion among mediating parties and interested states abate to the point where it was possible to get coherence around the Dayton process.1 In this case, the lead of a power-

S-4

11/24/09

4:18 PM

Page 91

THE CHALLENGES OF STRATEGIC COORDINATION

91

ful, interested state—and the informal coherence this provided—was a precondition for reaching agreement, let alone for implementation. The relative cohesion that characterized the run-up to Dayton was short-lived. After Dayton, there were still innumerable difficulties during the implementation phase relating to both strategic and operational coordination. As Cousens describes, implementation was “weakened by competing strategies among key implementing actors and contributing governments, these driven as much by bureaucratic and domestic considerations as by debate over the best way to consolidate peace.” This was at least as true on the operational side, where there were a large variety of different coordinating or lead entities among which there were very unclear reporting lines or lines of authority. Peace implementation in Bosnia suffered from a lack of coordination despite the fact that the Dayton Accords went relatively far both in specifying implementation tasks and allocating those tasks to particular agencies. Moreover, the accords created an Office of the High Representative (OHR) whose function was precisely to ensure both strategic and operational coordination. According to Cousens, neither of these mechanisms has been sufficient to the task. On the operational question of allocation of tasks, one set of tasks merged into another; therefore, a clear delineation of responsibilities did not prevent overlaps, gaps, the emergence of different strategies to implement the same program, or the emergence of divergent programs to address the same strategy. This has been particularly true in the movement and resettlement of refugees and internally displaced persons, responsibility for which within the UN system was given by the Secretary-General to the UN High Commissioner for Refugees (UNHCR), whose lead was also recognized by the Dayton Agreement. Connected to the refugee question are issues of civilian security (under the UN International Police Task Force), postwar reconstruction (the World Bank), human rights issues (multiple actors), and elections (the Organization for Security and Cooperation in Europe [OSCE]). The task of sorting out the connections, division of labor, and subcoordination arrangements on refugee issues in Bosnia has been notoriously difficult. As Cousens observes, the “organizational complexity of the implementation plan” was itself a barrier to implementation.2 Most observers have argued that the OHR has not succeeded in providing strategic coordination. According to some senior UN officials involved in the Bosnia operation, the OHR never articulated a coherent strategy that the international community could get behind.3 Although at one stage the OHR established subcoordination arrangements to help the high representative manage the coordination tasks (in the areas of civilian security, policing, refugee resettlement, human rights, and economic reconstruction and development), in the absence of a common strategic vision for implementation, these mechanisms for coordination worked only at the margins.

S-4

11/24/09

92

4:18 PM

Page 92

EVALUATING IMPLEMENTATION STRATEGIES

The lack of success in coordination in Bosnia is directly attributable to the fact that the major powers involved in Bosnia engaged from the perspective of different, even contradictory, policy goals. These different goals in turn affected the strategies of the range of financial and other multilateral institutions operating in Bosnia. This lack of cohesion among major powers emerges as the principal source of overall incoherence in the international response to Bosnia. The important point here is that the multilateral institutions, including the United Nations, were not able to function as a source of coordination in the absence of prior agreement among the major powers. Given the large international military presence that remains in Bosnia, it is difficult to say with any degree of finality what the impact of diffuse efforts in implementation has been. However, there is a widespread perception that the implementation effort has done little to create the circumstances for a self-sustaining peace. And, as Cousens makes clear, the lack of a strategy to overcome the weaknesses of Dayton means that whatever opportunities there are to build a lasting peace in Bosnia have had few chances to succeed. Rwanda: Conflicting Strategies Perhaps even more so than in Bosnia, the case of Rwanda is one where the absence of a coherent, coordinated strategy for dealing with opponents to the peace process emerges as a central flaw in the intervention process, with disastrous consequences.4 In the prenegotiation phase, international efforts focused on pressuring the one-party regime of Juvenal Habyarimana into a multiparty system. Leverage was applied primarily through aid conditionalities, and resulted in April 1992 in an invitation by the ruling Mouvement Révolutionnaire National pour le Développement (MRND) to selected opposition parties to join a coalition government. The push for multiparty democracy was based in part on an analysis conducted by the French military, and shared with the U.S. embassy, arguing that democratization was a necessary precursor to peace negotiations.5 This analysis proved correct, as the very first act of the coalition government was to launch political negotiations with the Rwandan Patriotic Front (RPF). As a prenegotiation strategy, democratization was effective. However, it bore costs that would complicate the mediation phase. The mediation phase in Rwanda centered on a sustained mediation effort facilitated by the Organization of African Unity (OAU) and the Tanzanian government, culminating (after five rounds of meetings over thirteen months) in a comprehensive peace settlement, the Arusha Accords. The accords covered issues relating to the rule of law, the return and resettlement of refugees, the creation of a national army, and the establishment

S-4

11/24/09

4:18 PM

Page 93

THE CHALLENGES OF STRATEGIC COORDINATION

93

of a Broad-Based Transitional Government (BBTG), pending national elections. The agreement called for the BBTG to be secured by a Neutral International Force (NIF). The most difficult elements of the negotiations related to participation in a joint national army and to the number of seats given to different actors in the BBTG. The latter issue proved the most difficult to negotiate, and was the last on which agreement was reached. It was on this question that the prenegotiation strategy complicated the mediation process. The challenge of striking a balance between the RPF and the regime was complicated by the presence of the opposition parties in the government’s negotiating party. What emerged from the deal was essentially a three-way split of seats: one-third for the ruling MRND, one-third for the RPF, and one-third for the combined opposition parties. While this formula was accepted by the Rwandan government’s negotiating team in the very final round of negotiations, it was poorly received by some members of the ruling party, and was rejected outright by the party of the spoilers, the Coalition pour la Défense de la République (CDR).6 There immediately began a process of jockeying for advantage under the three-way split. What emerged was a two-way alliance between the RPF and the opposition parties, an alliance of mutual interest in wrenching power away from the established oligarchy. The consequence of this, however, was that the ruling MRND began to see that it risked being a minority party in a BBTG dominated by the RPF-opposition alliance. Faced with this prospect, a number of members of the MRND moved toward a closer alliance with the CDR. The CDR played up the issue, and launched a fullscale campaign to recruit “losers” from the peace process into their burgeoning genocidal movement. At the same time, the locus of mediation efforts shifted from the OAU to the UN, after the OAU failed to get support for a proposal to establish a Neutral International Force, as provided for in the Arusha Accords. In its stead, the UN agreed to send a peacekeeping force to Rwanda, the UN Assistance Mission in Rwanda (UNAMIR), to secure the establishment of the transitional government. The mission was a small, Chapter VI peacekeeping operation with a limited mandate and few financial or physical resources, conditions that reflected the minimal political backing of the major powers. Thus, when it came to implementing the Arusha Accords, the UN and the Friends of the Arusha process were faced with an enormous implementation challenge: the opposition of a wide band of powerful political actors who had lost out in the negotiation process. It was a challenge that they were ill equipped to meet. Their efforts to establish the BBTG were rivaled by spoilers’ efforts to create a machinery for counteraction, including the creation and training of youth militias. Within months of the signing of the Arusha Accords, it was evident that the momentum and energy for imple-

S-4

11/24/09

94

4:18 PM

Page 94

EVALUATING IMPLEMENTATION STRATEGIES

menting a genocide greatly exceeded the capacity available for implementing the accords. That minimal resources would be available for implementation was arguably a foreseeable reality, given the climate surrounding peacekeeping operations at the time (especially in the aftermath of the Somalia operation). Possibly because the lead implementer (the UN) had been a minimal player in the Arusha mediation, correct judgments about the resources available for implementation were not factored into the mediation process. Equally significant, however, were the unforeseen consequences for mediation of the prenegotiation efforts at democratization. There was thus a critical gap between the prenegotiation and mediation strategies, and then between the mediation and implementation strategies. In short, the absence of a consistent strategy for dealing with hardline opposition fatally damaged the effectiveness of implementation efforts.7 The lack of strategic coordination undermined the prospects for the peace process, which ultimately fell victim to the far more effectively strategized and tightly coordinated efforts of the genocidaires. There can be no more compelling evidence of the critical importance of effective coordination. Burundi: Fragmented International Presence Since the onset of sustained civil war in 1993, Burundi has been the subject of a blizzard of international and regional peace initiatives. By late 1995, one informed practitioner estimated that there had been over 200 separate peace missions to Burundi, by official and unofficial actors.8 Since 1996, when Pierre Buyoya overthrew what was left of Burundi’s power-sharing government, the number of actors involved in conflict resolution in Burundi has increased.9 The result has been what Fabienne Hara has called a “fragmentation of the international response.”10 All the efforts described above, and many more, occurred simultaneously in the context of this small country. Various UN special representatives of the Secretary-General (SRSGs) and special envoys have sought to ensure that the efforts of the NGOs and others were complementary to the official negotiation process, with limited success. The only tool available to the SRSGs is to cajole donors into using their relationships with the NGOs to push their efforts in one direction or another—a difficult challenge in the face of the characteristic independence of the NGO sector. Ahmadou Ould-Abdallah, UN special representative from 1993 to 1995, recalls in his memoir of this period that the process of dealing with multiple NGOs and other unofficial mediators enormously complicated his own efforts.11 Hara describes the result in these terms:

S-4

11/24/09

4:18 PM

Page 95

THE CHALLENGES OF STRATEGIC COORDINATION

95

The Burundians were clearly able to profit from this confusion, with agents manipulating the different negotiators in order to gain maximum legitimacy. . . . In the end, it appears that every political tendency in Burundi has found a temporary ally among the international negotiators, who, in turn, have become part of the problem. . . . Burundians have intensified division among the various international agents by underlining and exploiting their different agendas.12

While these factors alone are not a sufficient explanation for the lack of progress in the Burundi peace process between 1993 and 1999, they have surely contributed to its overall weakness. Burundi, Rwanda, and Bosnia provide evidence that the coordination of strategy is a substantial factor in determining the cogency and effectiveness of international roles in support of peace implementation. Its importance becomes most clear when viewed from the perspective of responding to and containing opponents of a peace process, both spoilers and losers. In cases where there are significant opponents of peace, spoilers will seek to gain the advantage over mediators and implementers; they will seek to carve out and expand a political space in which they can demonize opponents, polarize political debate, and mobilize political actors and the population around an exclusionary or hostile agenda. To do so, they will exploit divisions among mediators and implementers, and take maximum advantage of any confusion or disagreement between various implementing agencies. Strategic coordination becomes a critical element of the capacity of implementers to stave off opposition. In more benign conflict environments, the relative efficiency of international efforts is less consequential for success. There are cases where the nature of the parties and the conditions, and the nature of the mediation process, are such that the cohesion of international actors is only minimally important. Equally, there will be cases where the will of parties to reach and sustain a peace accord will be minimal, and where even the best-designed international efforts will yield little success. However, there is a range of intermediate cases wherein effective international efforts should be able to help expand the space available to internal supporters of peace implementation, provide neutral space for dialogue on implementation, and contain or at least obstruct and frustrate opponents of a negotiated settlement. Strategic coherence is far from being the only requirement of effective implementation. It cannot, for example, overcome inadequate financial or military resources, the absence of commitment by major states, or poor strategy. But even with adequate resources, international commitment, and a good strategy, the success of implementation is likely to depend greatly on the ability of the lead implementers to devise a common strategy among the main international parties and on the consistent and complementary efforts to implement that strategy.

S-4

11/24/09

96

4:18 PM

Page 96

EVALUATING IMPLEMENTATION STRATEGIES

Lessons from Successful Cases While there is strong evidence that the absence of strategic coordination can undermine the effectiveness of international efforts, there is also evidence that strategic coordination does occur in civil war negotiation and implementation efforts, either by virtue of circumstance or through the deliberate application of policy. In the cases of Mozambique, El Salvador, and Guatemala, there were significant instances of strategic coordination that aided implementation. This section identifies key themes, tools, and mechanisms that emerge from a range of cases: the role of UN SRSGs, the issue of continuity of key actors, the role of “Friends” mechanisms, and the development of internal operational coordination mechanisms. The Role of SRSGs There can be little doubt that there is a high correlation between effective strategic coordination and the presence and good management of an SRSG or equivalent. 13 Indeed, effective coordination may be an essential— though, of course, insufficient—element of what it takes to succeed as an SRSG.14 It is surely no accident that the cases that serve as good examples of strategic coordination—Mozambique, El Salvador, Guatemala, and Israel/Palestine—are also those cases in which the respective SRSGs— Aldo Ajello, Alvaro de Soto, Jean Arnault, and Terje Roed-Larsen 15— emerge with strong reputations. However, it is also evident from a review of a wider range of cases that there are conditions on the deployment and management of SRSGs that frequently limit their capacity to coordinate. It is also true that, within the UN system, there have been a number of less capable SRSGs and that the procedures for recruitment and management of SRSGs have been woefully weak.16 On the wider diplomatic front, the forces that lead to the deployment of a special envoy or equivalent are various and often have less to do with the needs of conflict resolution than with domestic political considerations.17 What can be done to enhance the coordinating capacity of SRSGs? Of course, the personality and disposition of SRSGs are important factors in their performance. For an SRSG to be able to provide effective strategic coordination, however, there are several other factors that are more amenable to policy control. First, obviously, the peace process must be one in which the UN has a central role. SRSGs have been appointed to a number of conflict settings where the UN was a fairly marginal player. Not surprisingly, the impact of SRSGs in these cases has been minimal. An exceptionally effective SRSG can have an impact despite the limited scope of UN involvement, but this is uncommon. Second, SRSGs must be well supported by UN headquarters, and by

S-4

11/24/09

4:18 PM

Page 97

THE CHALLENGES OF STRATEGIC COORDINATION

97

the UN Secretary-General in particular.18 An often overlooked element of Aldo Ajello’s success in Mozambique was that, at key moments, he received strong backing from the UN Secretary-General.19 The support of the Secretary-General is critical both to an SRSG’s standing within the wider international community and to his/her ability to coordinate the multiple UN departments and agencies that have a stake in implementation.20 Within the UN system, an SRSG is typically the only figure whose role as overall coordinator of in-country efforts is generally accepted by the UN departments and agencies.21 Beyond the UN, an effective SRSG can usually generate a degree of involvement and coordination with other key players such as the World Bank and bilateral donors. Third, the timing of the deployment of the SRSG is also significant. There have been a number of cases—notably Rwanda—when an SRSG was appointed after the completion of the first phase of a negotiation process. Obviously, this late deployment limits the ability of the SRSG to build strategic linkages across phases of the peace process. This relates to the wider issue of the continuity of key actors, discussed below. Finally, the ability of the SRSG to ensure effective coordination is also a function of the degree to which he/she takes on strategic coordination as a central part of the mandate and job description. In the early stages of the UN’s operation in Kosovo, for example, the UN presence was headed by both an SRSG and a deputy SRSG (Sergio Vieira de Mello and Martin Griffiths, respectively), each of whom had both experience and an organizational interest in coordination. This is one reason that coordination of the work of both UN and other actors in Kosovo was a central part of the design of the UN mission there. Continuity of Key Actors The determination as to when the UN assigns an SRSG to a given conflict is a function of the extent of the UN’s role in that conflict, and depends greatly on the mandates given by the Security Council.22 In Rwanda, the minimal role of the UN during the mediation phase meant that the SRSG was deployed only after the completion of the Arusha Accords. That case has been described as one in which there was a significant disjuncture between the mediation and implementation phases, particularly in terms of the lead actor (first the OAU, then the UN), resulting in a lack of consistent strategy between the two phases. This highlights a broader issue relating to continuity of personnel, institutions, and international authority. The importance of continuity as a source of strategic coordination (if continuity is deliberately maintained, it could be referred to as a tool) can best be made by reference to positive examples of where it has worked. One of the most striking instances of effective continuity is the UN’s operation in Guatemala, where the UN had a substantive, even leading, role

S-4

11/24/09

98

4:18 PM

Page 98

EVALUATING IMPLEMENTATION STRATEGIES

among international actors during both the mediation phase and the implementation phase. This institutional continuity was supplemented by continuity of personnel; Jean Arnault, who had observed the Guatemalan negotiations, was left in place to verify their implementation. In turn, Arnault maintained a consistent approach toward parties to the agreement and ensured that the implementation strategy was based on the nuances of the agreement and the relationships built up during the implementation process.23 An important additional element in creating good international coordination was the involvement of key international financial institutions in the negotiating process. Not only was there a high degree of commonality between the peacebuilding concerns of the UN and the financial stability concerns of the international financial institutions (IFIs), the involvement of the IFIs in the negotiating process gave an opportunity for the two sets of international actors to remain well informed of each other’s agendas and to ensure the consistency of their actions. 24 One UN Development Programme (UNDP) official involved in these negotiations later commented that the presence and collaboration of the international financial institutions was central to the success of the Guatemala negotiations.25 The El Salvador case also is notable for a relatively high degree of collaboration between the UN and the dominant regional organization, the Organization of American States (OAS). The UN was involved from the outset, although its role was modest in the prenegotiation stages. When the locus of activity shifted toward the UN, there was already a record of involvement. Alvaro de Soto’s successful management of the mediation process is well documented. What is less well documented, but important to keeping the implementation process on track, is the role that de Soto continued to play from his position in the Department of Political Affairs in New York, first as director of the Americas Division and then as assistant secretary-general covering Asia and the Americas.26 Despite his moving from the field to headquarters, the continuous engagement by a key personality appears to have made an important contribution to the successful implementation of the El Salvador accords.27 Another form of continuity is illustrated by the case of Mozambique.28 Although there was a significant shift among lead actors—with the Italians playing a major role, along with the Rome-based Community of Sant’Egidio, in the mediation phase, and the UN coming in only at the point of implementation—continuity was deliberately maintained. First, the Community of Sant’Egidio, a Rome-based NGO, deliberately kept the UN fully informed of their efforts.29 Second, the appointment of an Italian to the job of SRSG was designed to keep the Italian government engaged.30 The support of the Italian government to Aldo Ajello, including the provision of extensive financial resources to the SRSG’s trust fund, was a vital element of Ajello’s capacity as SRSG, giving him both a degree of financial

S-4

11/24/09

4:18 PM

Page 99

THE CHALLENGES OF STRATEGIC COORDINATION

99

flexibility and an entry point to develop his relationship with the wider donor community, both of which enhanced his authority before the parties.31 When there is no continuity between mediators and implementers, the chances of maintaining a consistent strategy between phases diminishes because there exist no established mechanisms for country-specific policy dialogue between international, regional, and subregional organizations. Also, bureaucratic politics and institutional rivalry can disrupt a smooth transition between lead institutions. The Role of Friends Groups One of the striking commonalties among cases of successful implementation is the use of a Friends group, or the creation of a deliberate process of bringing together key governments, to ensure a degree of focus and commonality of approaches to the peace process. This mechanism emerges as an important source of strategic coordination among bilateral actors and a means by which SRSGs or lead mediators and implementers can bolster their authority, leverage, and coordinating powers vis-à-vis the parties and the implementing agencies.32 The Friends group has been widely used in the Latin American context. In Guatemala, a Friends group served to channel bilateral inputs into the mediation and implementation process, assisting effective strategic coordination overall. The “Friends of the Secretary-General” was an important avenue for keeping focus and pressure on the parties during negotiations in Haiti.33 In El Salvador, the members of the “Four plus One” mechanism reinforced de Soto’s role as SRSG. And in Nicaragua, the “Rio Group” served a similar function during the early stages of negotiation, as did a “Friendly Countries” group during the mediation phase. 34 In the Israel/Palestinian Authority context, a similar entry was constructed, partially through the efforts of Roed-Larsen (prior to taking on the job as SRSG): the Ad Hoc Liaison Committee (AHLC) for International Assistance to the Palestinian People was formed out of the Oslo peace process as a forum bringing together key donors, the parties, and the key multilateral agencies. The Friends mechanism has been less frequently used in the African context, perhaps reflecting the lower levels of Western commitment to peace processes in that continent. In Mozambique, however, Ajello was supported by a “core group” of ambassadors on the ground. A similar mechanism was used in Rwanda, in the form of the “Five Musketeers,” a group of the five most influential ambassadors in the country. Rwanda’s fate, however, cautions that the use of some sort of bilateral coordination mechanism is no guarantee of success. However, one of the distinctions

S-4

11/24/09

4:18 PM

100

Page 100

EVALUATING IMPLEMENTATION STRATEGIES

between a Friends mechanism and a “core group” is that a formal Friends group typically exists and meets not only in-country but also at the level of capitals and at the UN in New York. This signals a higher level of commitment to the peace process on the part of the respective members than is typically evidenced by collaboration among in-country ambassadors. This higher level of commitment typically results in higher levels of financial resources provided for the peacebuilding process. The relationship between a Friends mechanism and an effective strategy of implementation depends on preexisting levels of commitment. Certainly, it is the case that strategic coordination cannot generate political commitments or resources. While donors often have at times argued that a lack of effective coordination is the source of their reluctance to commit funds to peace implementation processes, the causal relationship is more likely the reverse. This said, it is not clear that there is a linear relationship between preexisting commitment and the level of strategic coordination. In a context where a range of states are providing resources, coordination of their strategies for peace implementation may be a critical asset. There have clearly been cases where the coinciding interests of a group of key states (for instance, in Haiti)35 or the dominant interest of a single state (as in Bosnia) will be the deciding factor. But there have also been cases where an SRSG or equivalent can play an important role in harmonizing somewhat divergent donor strategies (as in the Middle East).36 Friends mechanisms serve other strategic coordination functions. Most particularly, good collaboration between an SRSG and a Friends group lends considerable political support to the SRSG in terms of his authority as an overall leader of international efforts. For example, de Soto has observed that the use of a Friends mechanism in El Salvador helped him fend off potential competitors for the lead mediation role.37 Establishing a lead role in turn enhances the ability of an SRSG to perform coordination tasks vis-à-vis UN agencies and other actors. Indeed, recent experiences in Sierra Leone and elsewhere suggest that collaboration and coordination between key donors and the SRSG or equivalent is in fact the central ingredient in overall strategic coordination. The involvement and interest of a large number of different state actors in a negotiation or peace implementation process can, of course, present a major obstacle to coherent conflict management, particularly in the absence of a lead state. An obvious example is Germany’s unilateral move to recognize Croatia as a sovereign state, a move that complicated outside efforts to prevent the outbreak of war. Later in Bosnia, the United States took on a lead state function when it became more fully engaged in the management of the international effort to end the war there. Once the United States put the full weight of its political, military, and financial capacity behind a particular strategy, that strategy—well-conceived or otherwise—won the day.

S-4

11/24/09

4:18 PM

Page 101

THE CHALLENGES OF STRATEGIC COORDINATION

101

Coordination Mechanisms In a number of cases, SRSGs or lead political actors have created mechanisms to channel common inputs into a mediation process, such as the Civil Society Assembly in Guatemala. More commonly, lead political actors have established mechanisms to tackle operational coordination during the implementation phase. However, ensuring a common agreement to and a synchronized enactment of strategy within this phase has proven enormously challenging. Poor operational coordination has proved to be a serious constraint on peace implementation in the areas of demobilization and refugee return, although instances of effective coordination on demobilization and reintegration can be found in Mozambique, Central America, and to a lesser extent Cambodia. In a number of cases, the establishment of coordination mechanisms by lead actors—SRSGs, high representatives, lead agencies, or their equivalents—has served an important strategic coordination function. Although the mechanisms in question were designed to provide coordination in an operational sense, their existence in and of itself serves a strategic function. Requiring actors to work within a common operational framework can generate the development of common strategy. Coordination mechanisms provide a forum within which an SRSG or a functional equivalent can communicate a strategic vision; and the mechanisms offer limited oversight by political actors over the operational implementation of assistance and other peacebuilding tasks. An example of how the establishment of a coordination mechanism can serve to build support for a given strategy is Terje Roed-Larsen’s role as special coordinator in the Occupied Territories. Early on, Roed-Larsen helped create an effective architecture for coordination on the ground by reaching agreement with the World Bank and the Norwegian government to cochair a local donor coordination mechanism, the Joint Liaison Committee. With support from the World Bank and other donors, RoedLarsen also ensured that he represented the entire UN family of agencies in the local coordination structure. Bringing key players together around donor issues allowed Roed-Larsen to help shape the agenda in a way that strengthened his efforts to assist the Palestinian Authority.38 Here, RoedLarsen’s prior role in the Oslo peace process strengthened his ability to forge these relationships. The central lesson here is that there are tools available to perform the essential functions of strategic coordination, when basic conditions—a degree of international commitment and rough correspondence of the interests of the major powers—allow for it. Major weaknesses occur when there is a significant disjuncture in institutions or personnel, especially at different phases of the process. Problems also arise when there is a significant divergence of interests or strategies between key states. Within a moderate

S-4

11/24/09

4:18 PM

102

Page 102

EVALUATING IMPLEMENTATION STRATEGIES

range, however, divergent strategies can be harmonized by coordination models, either multilateral or state-based. When underlying conditions permit, the application of operational coordination tools can enhance the effectiveness and impact of international efforts to support the negotiation and implementation of a peace accord.

A Growing Challenge? The key question, then, becomes whether the basic conditions that create the possibility of strategic coordination will likely obtain in the future. Two trends suggest that they will not, which means that strategic coordination will likely become a greater problem in future international conflict management efforts. First, in Guatemala, Mozambique, and El Salvador, no more than a handful of actors were directly involved in political negotiations, while even in the assistance aspects of peace implementation, the number of international, regional organizations, and even nongovernmental organizations was relatively small. By contrast, by the mid-1990s there had been an explosion of the number of actors involved in various aspects of conflict management, including in the sphere of political negotiations, thus placing a much higher premium on creating and straightening coordination mechanisms. Second, implementers in these cases pursued a traditional peacekeeping strategy, based on consent, neutrality, and impartiality. If the international community faces more difficult conflict environments than El Salvador or Guatemala, there is greater likelihood that they will need more coercive strategies. Such strategies engender much more controversy within the United Nations and are often divisive to long-standing military alliances such as the North Atlantic Treaty Organization (NATO). Moreover, as Karen Guttieri observes, the more coercive the strategy of peace implementation, the greater the interference of states in the day-today operation of the mission.39 Tough Cases and Strategic Disagreement Recent experiences in Kosovo illustrate how difficult conflict environments and disagreement over coercive strategies pose greater challenges for peace implementation operations. The struggles between the OSCE and the UN prior to and during the NATO bombing campaign, and the process by which the UN was assigned the lead role in postconflict Kosovo, underscore the absence of any mechanism—other than simple political bargaining—through which the international community ensures that the basic conditions for strategic coordination are in place. From the outset of the Kosovo crisis, it was evident that the OSCE would play a leading role in handling this European affair. The crisis began

S-4

11/24/09

4:18 PM

Page 103

THE CHALLENGES OF STRATEGIC COORDINATION

103

to mount at a moment when UN relations with the United States had once again deteriorated over the Iraqi question.40 The U.S. and major European actors kept the UN out of the mediation process. The key political roles were played by NATO, the OSCE, the U.S. government, and the G-7 countries. At this time the UN feared that it would be sidelined both in Kosovo and in the military brinkmanship and eventual air strikes in Iraq, thereby further undermining the already corroded authority and primacy of the Security Council in the maintenance of international peace and security. In the end, of course, the settlement process returned to the UN, in part because of the need to bring Russia on board as a means of concluding an agreement with Serbia, and also because some European powers were concerned with U.S. support of NATO’s intervention outside the legalizing and legitimating framework of the UN Charter.41 The central point for strategic coordination, however, is that the decision as to which organization would take the lead in the implementation of the settlement had little to do with considerations of strategy or with the comparative advantages of the respective organizations. This created a major (if inevitable) disjuncture between those who had managed the mediation and settlement process and those who would oversee the settlement. In the end, the return of the UN was decided less because of any issues directly related to Kosovo and more because of wider political debates about the future architecture for the management of European collective security.42 Due to the scope of the mandated agreement, Kosovo is not a case of peace implementation like the others—although it may represent a widening trend toward the creation of transitional authority missions.43 It does illustrate, however, that the processes for determining which entity will have a leading role in managing the mediation or the implementation process, and thus whether there will be continuity and strategic coordination, function not on the basis of any kind of strategic decisionmaking but are negotiated on the basis of interstate and interorganizational politics and rivalry. This problem is especially pronounced when coercive strategies are pursued in the absence of a broad international consensus. Kosovo also demonstrates that even when one organization is finally assigned a lead or coordinating role, other roles within a mission area will tend to be determined according to organizational prerogative, rather than comparative advantage. International activities in Kosovo fall broadly among four international and regional organizations: civil administration, including international police, is led by the UN; humanitarian assistance is orchestrated by the UNHCR; economic reconstruction is led by the European Union (EU), in collaboration with the World Bank; and institution-building is led by the OSCE. The NATO-led Kosovo Force (KFOR) ensures an international security presence. That the division of labor between these organizations is spelled out in Security Council Resolution 1244, which authorized the UN Administration in Kosovo, masks the fact

S-4

11/24/09

4:18 PM

104

Page 104

EVALUATING IMPLEMENTATION STRATEGIES

that the division was actually worked out through an ad hoc process of high-level political negotiations in European capitals and working-level negotiations in Pristina.44 Proliferation of Actors: A Growing Coordination Challenge The challenge posed by institutional bargaining and politics has been further complicated by the growing proliferation of actors involved in peacebuilding. This proliferation occurred first in areas related to the aid aspects of peace implementation and had a number of dimensions: the creation of postconflict units in a number of multilateral development agencies, notably the World Bank, and in bilateral aid agencies; the creation of emergency response units in such development or technical assistance organizations as the UNDP, the World Health Organization (WHO), the Food and Agriculture Organization (FAO), and even the International Labour Organization (ILO); and the increased number of nongovernmental agencies involved in the business of emergency relief and reconstruction. By the late 1990s, an increasing number of actors were also getting involved in conflict resolution and mediation. This involved, first, the expansion of a number of regional or functional economic organizations into the political sphere—notably including the EU and the G-8 countries. Second, a number of regional organizations (particularly the EU and NATO) began involving themselves in political negotiations beyond the regional limits of their composition, for example, the EU special envoys to the Great Lakes and to Ethiopia/Eritrea and NATO’s role in political negotiations concerning Kosovo. Third, regional organizations that had been previously passive took on significant operational roles in the 1990s— notably the OSCE, in the form of both the High Commissioner for Minorities and direct operational missions such as the Kosovo Verification Mission (KVM). Fourth, there was a growth of a new sector of NGOs specializing in conflict resolution activities—Conciliation Resources, International Crisis Group, Responding to Conflict, and International Alert, to name just a few. Finally, a number of the development and relief NGOs began moving into the conflict resolution field, especially in the areas of peacebuilding and reconciliation. The involvement of an expanding number of actors has been a significant factor in virtually all peace missions since Rwanda. The number of actors involved directly in political negotiations varies considerably from case to case, but in almost all of them it considerably exceeds the numbers involved in the successful cases cited earlier. The proliferation of regional and extraregional organizations has been most notable in Europe, of course, with Kosovo standing as the current high-water mark of this phenomenon. However, the phenomenon is not limited to Europe, first because the

S-4

11/24/09

4:18 PM

Page 105

THE CHALLENGES OF STRATEGIC COORDINATION

105

European institutions have begun playing active roles outside Europe, and second because of the increasingly global scope of NGO activities. The most notable case of extensive NGO involvement in conflict resolution is Burundi. As early as 1995, there were already a substantial number of NGO-sponsored conflict resolution programs operating there, under the auspices of International Alert, Responding to Conflict, and various other agencies. This activity subsequently expanded. Development and relief NGOs, such as ActionAid, have taken on significant conflict resolution activities within their normal programs for assistance to displaced persons. The more established mediation NGOs, such as International Alert and Sant’Edigio, played direct roles in various back-channel negotiations. Meanwhile, a growing number of conflict resolution NGOs, including International Crisis Group and Search for Common Ground, established offices in Bujumbura. As observed above, the proliferation of unofficial actors has contributed to the fragmentation of the international presence. Proliferation of Coordination Mechanisms: A Partial Response As the number of actors has increased and the international response has become more fragmented, the UN and others have experimented with coordination models designed to promote coherence. Different mechanisms have been tried in various settings. Within the UN, where most coordination efforts have been concentrated, two distinct models have been used to attempt to enhance coordination: the Strategic Framework, and the Integrated Mission. A brief review of these efforts highlights both the prospects for and the constraints upon effective strategic coordination.

Strategic Frameworks. There are two central notions behind the Strategic Framework initiative. The first is that all major actors involved in a given country—UN political and assistance actors, NGOs, bilateral agencies, and local authorities—should come together to find common strategies and programs in support of “peacebuilding.” Second is the idea that these strategies should draw on all aspects of the international presence, incorporating humanitarian relief, development assistance, human rights, and political efforts into one common framework. To this end, the Strategic Framework initiative has sought to articulate some broad parameters to guide policy and to take account of the potential trade-offs between different policy elements. The overall political task of consolidating peace and stability is clearly the primary objective of Strategic Frameworks—as is indicated by the fact that in the two applications of the Strategic Framework to date, the UN Department of Political Affairs took the lead from the outset. However, the Strategic Framework concept is predicated on the idea that political and assistance efforts should

S-4

11/24/09

4:18 PM

106

Page 106

EVALUATING IMPLEMENTATION STRATEGIES

“inform and be informed by” one another—as articulated in the UN’s Generic Guidelines for Strategic Frameworks, a document negotiated at some length in the UN in 1999. 45 The Generic Guidelines suggest that political efforts should not be allowed to preempt essential, lifesaving humanitarian assistance, but also that rehabilitation and development efforts should be geared toward political efforts—a statement of implied conditionality. Further, noting the potential for a clash between human rights–based approaches and humanitarian approaches, the Guidelines state that the two should be reconciled where possible, but that human rights activities should not interfere with lifesaving humanitarian assistance. However imprecise, the negotiation of this document represented a significant step forward within the UN to identify priorities between competing elements of the UN’s overall response. The mechanisms for implementing the Strategic Framework have been focused around the SRSG and the UN resident coordinator, the normal incountry coordinator of operational activities. The basic idea is that these two actors undertake to chair a series of coordinating bodies that bring together UN agencies, local and international NGOs, donors, and national authorities. Decisions as to the application of a Strategic Framework are to be made by the deputy secretary-general of the UN on the joint recommendation of the Conveners of the Executive Committees responsible for Peace and Security, Humanitarian Affairs, and Development (the Department of Political Affairs [DPA], the Office for the Coordination of Humanitarian Affairs [OCHA], and the UNDP). The Strategic Framework was first implemented in Afghanistan, initially under the lead authority of the DPA, which headed the UN Special Mission in Afghanistan (UNSMA), a small political negotiation mission. However, primary responsibility quickly fell to OCHA, which had a much larger field presence than the DPA. This change of responsibility substantially altered the character of the Strategic Framework in Afghanistan. As many commentators have since noted, the Strategic Framework lost its political character. Rather than serving to shape political strategies, it evolved instead into a framework devoted exclusively to the coordination of aid (although, of course, in the Afghanistan context, aid was and is a heavily politicized issue). This transformation was reflected in the name of the coordinating body that was eventually established—the Afghanistan Programming Body. This body met under the lead of a rotating donor chair, rather than the head of UNSMA as was originally envisaged. This change also reflected the fact that, whereas the Strategic Framework was envisaged primarily as a body for orchestrating postconflict peacebuilding efforts, Afghanistan had not reached that stage. Indeed, even as the Strategic Framework was first being designed, the bulk of UN personnel were withdrawn from Afghanistan following presumed Taliban attacks on UN staff members in the wake of the U.S. cruise missile strike in Afghanistan.

S-4

11/24/09

4:18 PM

Page 107

THE CHALLENGES OF STRATEGIC COORDINATION

107

Many have attributed the problems encountered by the Strategic Framework initiative in Afghanistan to contingent circumstances rather than any inherent flaw. A second Strategic Framework experiment was launched. The country chosen for the second application was Sierra Leone, despite the fact that many of those working on the Strategic Framework felt that it should not be attempted in peacekeeping contexts, since peacekeeping missions had their own, preexisting tools for overall coordination. The Strategic Framework for Sierra Leone proved no more successful than its predecessor in Afghanistan.46 After an entire year, collective efforts had produced little more than a general statement of existing coordination problems and of the need for greater links between the political, development, and humanitarian coordination mechanisms. The establishment of a Strategic Framework did not prevent considerable coordination gaps from occurring, notably with respect to the critical task of demobilizing former combatants. Furthermore, the Strategic Framework did little to help rationalize the preexisting coordination mechanisms. At one point, there were four distinct UN coordination structures in Sierra Leone, none of which bore any formal or substantive relationship to the others. These included a multidimensional peacekeeping operation, under the responsibility of an SRSG and reporting to the Department of Peacekeeping Operations (DPKO); the Strategic Framework initiative, under the lead of the DPA; a humanitarian coordinator, who was also the UNHCR resident representative, and the UN resident coordinator (who thus reported to OCHA, the UNHCR, and the UNDP, respectively); and a Brookings Process initiative—an effort by the World Bank, the UNHCR, and the UNDP to coordinate efforts for linking relief and development initiatives to meet postconflict challenges. Given that the poor experience of the Strategic Framework initiative in Sierra Leone in many ways mirrored the experience of Afghanistan, many in New York concluded that the weakness of the Strategic Framework in Sierra Leone was due to flaws in the design of the Strategic Framework itself. Both experiences suggest that the Strategic Framework is too cumbersome a mechanism to generate serious collaboration or flexible coordination and does little to actually empower lead coordinators such as the SRSG. Partially as a result, there was within the UN a move toward another model: the Integrated Mission. Before turning to the Integrated Mission model, however, it is important to note one other feature of the Sierra Leone case. Notwithstanding the multiple coordination efforts of the UN, a significant—if unofficial—coordinating agent in Sierra Leone was the British government. Sierra Leone became, in effect, a case of “lead state coordination,” though this was never officially recognized. Britain took on a number of important coordinating and lead functions: It chaired a donor forum to galvanize international financial support; it took the lead in financing the mission of the Economic

S-4

11/24/09

4:18 PM

108

Page 108

EVALUATING IMPLEMENTATION STRATEGIES

Community of West African States Cease-Fire Monitoring Group (ECOMOG) when ECOMOG reintervened in Sierra Leone following the takeover attempt by the joint forces of the Armed Forces Revolutionary Council (AFRC) and the Revolutionary United Front (RUF); it provided critical funds for demobilization when there was a shortfall; and, in a number of other ways, it ensured a degree of consistency, forward planning, and sustained international commitment to Sierra Leone. Most critically, in May 2000 the UK government fielded a rapid-reaction force to Sierra Leone to bolster the UN and the government of Sierra Leone when efforts to demobilize Foday Sankoh’s troops, as required by the Lomé Accords, resulted in the RUF taking 500 UN troops hostage and launching a renewed attack on the government. Throughout, the UK government worked closely with the SRSG at the time, Francis Okelo, not through the formal Strategic Framework mechanism. The contrast between the critical role played by the UK government and the multiple, overlapping, and frankly ineffective coordination mechanisms promoted by UN headquarters was sharp. Integrated Missions. In a very different part of the world, another UN experiment in coordination was evolving. This was the Integrated Mission in Kosovo, deployed immediately after the cessation of NATO’s bombing campaign in May 1999. A number of aspects of the Kosovo mission have already been discussed. At the political level, Kosovo reveals that the means by which different regional or international organizations are allocated tasks and responsibilities within an overall division of labor has little to do with their comparative advantage or respective expertise, and much to do with major power competition over Atlantic security policy. The result was a hybrid mission consisting of a NATO security presence and a UN-led civilian operation with substantial components tasked to the UNHCR, the OSCE, and the EU. The EU’s presence was bolstered by a joint EU/World Bank Task Force based in Brussels, which would bring World Bank expertise in reconstruction to bear on the EU’s efforts. Having learned in Bosnia about the difficulty of ensuring policy coherence among these different regional and international organizations, the UN took the lead in establishing a structure designed to ensure greater coherence. Specifically, the UN established a so-called Integrated Mission, in which the UN special representative would lead a joint structure composed of the various key organizations. In practical terms, this meant that the EU, the OSCE, and UNHCR were each asked to provide a deputy SRSG to the mission, who would be responsible for the operation of each institution’s programs while also helping to manage the overall mission structure. In this way, the four main civilian institutions would be linked to one another, and, at least in theory, the SRSG would be able to ensure policy coherence

S-4

11/24/09

4:18 PM

Page 109

THE CHALLENGES OF STRATEGIC COORDINATION

109

and an effective division of labor among them. The OSCE, the EU, and the UNHCR all agreed to join this Integrated Mission structure. The task of delineating an effective division of labor was complicated by the breadth of tasks for which the international community was responsible: assuming basic civilian administration of health, education, energy, public utilities, and post and telecommunications; establishing a legal framework; restoring functioning public services; overseeing humanitarian activities, including the protection of minority enclaves, the evacuation of minorities under threat, the provision of emergency food assistance, and emergency shelter reconstruction; building institutions (including media, police, political parties, and judicial bodies); rebuilding the economy, including the reconstruction of customs services, legal currency, tax collection, and the provision of microcredit loans and the provision of fuel supplies; demilitarizing paramilitary forces and creating the Kosovo Protection Corps; monitoring human rights and ensuring institution-building; and safeguarding the environment. These tasks were fundamentally interrelated.47 For example, it is next to impossible to define a meaningful division between public administration (in a context where the UN Mission in Kosovo [UNMIK] is essentially a government) and capacity-building, between providing emergency shelter and longer-term economic reconstruction, between the legal protection of minority rights and the humanitarian protection of minorities, or between the UN’s law-and-order responsibilities and the public security tasks of NATO. The precise responsibilities of each organization within this potentially unwieldy superstructure were negotiated over the first weeks of its operation. What was eminently clear, however, was that any such division would be a flexible arrangement. Recognizing that policy coherence would not arise automatically from the integration of these organizations into one structure, a careful division of labor was articulated through in-country negotiations and through a series of coordination mechanisms established within the mission structure. In particular, the mission established an Executive Committee wherein the four deputy SRSGs met with the SRSG to set overall policy; a Joint Planning Team, again composed of the four organizations and serviced by a small secretariat, whose terms of reference were to ensure consistency of planning and implementation; and the post of a principal deputy SRSG (also pushed for by the U.S. government), whose primary function was to ensure effective integration of the mission elements.48 Still, effective implementation of the agreement among the UNMIK component organizations proved highly dependent on the performance of senior officials from each of those organizations. The primary power base of both the SRSG and the deputy SRSGs is their own organizations’ programs, money, staff, headquarters position, and reputation. The varying

S-4

11/24/09

4:18 PM

110

Page 110

EVALUATING IMPLEMENTATION STRATEGIES

clout of the deputy SRSGs within the European political context also weighed heavily on their ability to implement programs and their degree of responsiveness to efforts by the SRSG or principal deputy SRSG to coordinate strategy and the implementation of programs within the mission. Furthermore, some of the deputy SRSGs had preexisting relations with various Kosovo parties, and held different political views than the SRSG. The ability of the SRSG to contain separate political initiatives or relationships on the part of the deputy SRSGs was therefore heavily constrained. Interestingly, this structure brought into the open another coordination issue that proved very important, namely, competition within individual states between the bureaucrats and politicians involved, respectively, in EU, UN, and OSCE affairs. Poor coordination is usually experienced by government officials or local authorities in receiving very different messages from a series of UN organizations. In Kosovo, the tables were turned. A very strong coordination mechanism at UN headquarters—involving, inter alia, daily meetings chaired by the deputy secretary-general with all relevant UN entities—meant that there was little divergence within the UN, as disagreements were resolved internally. On the other hand, UN staff members in Kosovo frequently had the experience of being given profoundly contradictory messages by individual member-state representatives to the OSCE, the EU, and NATO. In short, Kosovo illustrates that in cases where a large number of institutions and major powers are involved, effective operations on the ground depend on strategic coordination at the level of both institutional headquarters and the member-state capitals. Short of this, actors in the field will use their own political connections and resources at headquarters to pursue different agendas and implement divergent strategies. Within one organization, this already poses a significant challenge. When dealing with multiple headquarters of multiple organizations, in a context where there is no established mechanism for policy dialogue or harmonization among them, the prospects for ensuring the basic conditions of strategic coordination are low, and thus the likelihood of forging or implementing a common strategy is minimal at best. Of course, policy disagreements in Kosovo have been amplified by the vagueness of the Security Council resolution concerning the nature of the final settlement. This uncertainty has enormously complicated the challenge of setting clear policy directions. Given both a vague end-state and diffuse international efforts, the bargaining position of the international community vis-à-vis aspirants to legitimate power, such as the Kosovo Liberation Army (KLA), is very weak. The net result has been a fairly good ability to coordinate technical policy differences between the implementing organizations, but a far more constrained capacity to coordinate overall strategic questions. In this respect, the Integrated Mission structure has been more effective than the

S-4

11/24/09

4:18 PM

Page 111

THE CHALLENGES OF STRATEGIC COORDINATION

111

Strategic Framework, but has still reflected the limits of strategic coordination in the context of political and institutional diffusion. The greater perceived effectiveness of the Integrated Mission structure in Kosovo has lent momentum to applying this model elsewhere. Notably, the Integrated Mission was used for East Timor, and elements of it have been incorporated into the UN’s mission to the Democratic Republic of Congo. More generally, efforts are under way within the UN to ensure that humanitarian and development elements are routinely incorporated into UN political missions led by SRSGs. Within the UN, some debates about the precise limits of the SRSG’s authority remain, but by this stage these are debates of fine-tuning, not of basic policy. The Integrated Mission concept is clearly becoming the basic building block around which UN coordination will occur in peace implementation settings. Of course, enhanced coordination within the UN does not solve the far greater problem of coordination between the UN and other actors. Experiences in Kosovo and Sierra Leone, as in other cases, reveal very serious limits to the extent to which the various regional organizations or NGOs recognize the coordinating authority of the UN or of any other actor.

Conclusion Ideally, strategic coordination should establish clear lead actors in the mediation and implementation of peace agreements. It should allow those lead actors to set priorities, to ensure that those priorities are pursued by all the third-party actors involved, and to provide consistency across phases of a political process, such that implementation efforts are grounded in the realities of the negotiating process. Lead actors should also be given the authority to resolve disputes between third parties about those priorities or about the strategies used to achieve them. In some cases, a coincidence of interests among major states, a degree of institutional continuity, and the deliberate application of policy tools have allowed for effective strategic coordination in the terms just described. Neither the UN nor any other actor is currently equipped to fulfill this idealized version of strategic coordination. The proliferation of third-party actors, the prevalence of political competition between major powers, and the weakened authority of the UN have all constrained the capacity of the UN to perform essential strategic coordination functions. A clear, consistent alternative has not yet emerged. The positive experiences of “lead states” in Bosnia, Sierra Leone, and elsewhere suggest an important alternative to UN coordination. But this model, too, has serious limits, both because this approach tends to be ad hoc and because coordination by a major power will tend to be coordination in support of one party, as distinct from impartial support to the peace process itself.

S-4

11/24/09

4:18 PM

112

Page 112

EVALUATING IMPLEMENTATION STRATEGIES

The limitations on the UN’s capacity to fulfill critical strategic coordination functions are unlikely to abate substantially in the near future. The diffusion of organizations involved serves a very important political purpose, because it means that any state with a particular interest can find an organization to represent that perspective in the conflict resolution process. Powerful states and institutions are not likely to give up their independent authority or their room for maneuver. On the other hand, that organizations such as the OSCE and the EU have agreed to serve under the UN’s lead (at least to a certain extent) in Kosovo suggests that there is some degree of support for coordination. If experience in the humanitarian world is anything to go by, repeated encounters with the negative consequences of incoherence will lead to a rising degree of support over time for the establishment and strengthening of more robust coordination structures. Of course, the most powerful states and organizations involved will seek to ensure that they lead those coordination mechanisms. What is clear, in any case, is that absent effective coordination by the UN, a comparable regional or international organization, or a lead state, the effectiveness of implementation efforts will be heavily constrained. This will, of course, be more consequential in cases where implementation is more difficult—but it is precisely in the most difficult cases that an effective international effort is most needed to ensure peace.

Notes 1. See Chapter 18 on Bosnia by Elizabeth Cousens in this book. 2. Elizabeth Cousens and Charles Cater, Towards Peace in Bosnia: Implementing the Dayton Accords, IPA Occasional Paper (Boulder, Colo.: Lynne Rienner, 2001). 3. Author interview, UN headquarters, November 1, 1999. 4. Bruce D. Jones, Peacemaking in Rwanda: The Dynamics of Failure (Boulder, Colo.: Lynne Rienner, 2001). 5. This analysis was read to me by Carol Fuller, former Rwanda desk officer, U.S. Department of State. It was written by Colonel Galinie, a military attaché at the French embassy in Rwanda, in 1991. 6. Jones, Peacemaking in Rwanda. 7. This rather complete strategic failure was of course compounded by the humanitarian response to the forced refugee outflow in Goma. In providing humanitarian assistance to the approximately 1.7 million refugees who fled Rwanda to Zaire in July and August 1994, humanitarian agencies inadvertently created a situation in which the genocidal regime had a stable population base from which to rebuild its political and military movement. This contributed to the renewal of warfare in Zaire in 1996. 8. Kumar Rupesinghe, address to the State Department Forum on Conflict Resolution, Washington, D.C., December 1995. 9. At the official level, there has been a long-running political negotiation—

S-4

11/24/09

4:18 PM

Page 113

THE CHALLENGES OF STRATEGIC COORDINATION

113

once again in Arusha and, until recently, also led by Tanzania—managed primarily by regional governments, though also involving some international governments. Sant’Egidio, a private foundation connected to the Vatican, famous for its work in the peace process in Mozambique, has held secret talks with the parties. The UN has had a special representative in Burundi and, at other times, a political mission to support the Arusha negotiations. The UN Educational, Scientific, and Cultural Organization (UNESCO) launched a “culture of peace” program in Burundi to counter polarizing media through broadcasts and other media work transmitting messages of reconciliation. International Alert, a conflict resolution NGO, has orchestrated a series of conflict resolution processes with parliamentarians and other government and civil society figures. Other conflict resolution NGOs active in the country include the International Crisis Group, Accord, the Carter Center, and Search for Common Ground. Additionally, humanitarian NGOs such as ActionAid, CARE, and Oxfam have incorporated conflict resolution activities into their own programs for refugees and victims of the war. 10. Fabienne Hara, “Burundi: A Case of Parallel Diplomacy,” in Chester A. Crocker, Fen Osler Hampson, and Pamela Aall, eds., Herding Cats: Multiparty Mediation in a Complex World (Washington, D.C.: U.S. Institute of Peace, 1999), p. 139. 11. Ahmedou Ould-Abdallah, Burundi on the Brink: A UN Special Envoy Reflects on Preventive Diplomacy (Washington, D.C.: U.S. Institute of Peace, 2000). 12. Hara, “Burundi,” pp. 149–150. 13. Although this chapter focuses on UN special representatives, this is simply for ease of reference. The arguments given about the role of SRSGs would for the most part refer equally to an OSCE head of mission or an OAU special envoy, to give but two alternative examples. 14. The importance of a single mediator as a lead coordinating agent is one of the principal conclusions of Chester A. Crocker, Fen Osler Hampson, and Pamela Aall’s introduction to Herding Cats, pp. 3–18. 15. At the time of publication, I am serving as special assistant to Terje RoedLarsen in the Middle East. References and judgments here about Roed-Larsen’s efforts pertain solely to his first tour as SRSG, between 1994 and 1996, and were written before I took up my present assignment. 16. For a review of issues relating to the role, functions, and management of SRSGs, see in particular Terje Roed-Larsen and Rick Hooper, Command from the Saddle (Oslo: Peace Implementation Network, 1999). 17. This section focuses on UN special representatives, but many of the thematic issues described here apply in equal measure to diplomatic representatives of governments. 18. Roed-Larsen and Hooper, Command from the Saddle. 19. Joao Honwana, “Mozambique,” draft prepared for this project. 20. Roed-Larsen and Hooper, Command from the Saddle. 21. This observation is based on experience within the UN, both in dealing with the role of SRSGs as a generic policy issue, and in implementing policy in cases such as Sierra Leone, Kosovo, the Democratic Republic of Congo, and East Timor. While the role of the SRSG still generates some dispute, the SRSG has reached broader and wider acceptance than that of resident, humanitarian coordinator, or lead agencies. During 2000, this was formalized in the form of a “Note of Guidance on the Relationship Between SRSGs, Humanitarian Coordinators, and Resident Coordinators.” 22. Technically, an SRSG can only be appointed on the basis of a Security

S-4

11/24/09

4:18 PM

114

Page 114

EVALUATING IMPLEMENTATION STRATEGIES

Council mandate. However, the Secretary-General can appoint personal representatives, representatives, special coordinators, and—in some cases—special envoys, without a direct mandate. 23. Jean Arnault, “Implementation of Peace Agreements in Civil Wars: Views of Practitioners,” conference presentation cosponsored by the Center for International Security and Cooperation (CISAC) and the International Peace Academy (IPA) at Pocantico, New York, February 28–March 1, 1998. 24. See Chapter 15 by William Stanley and David Holiday in this book. 25. A UNDP official involved in backstopping the negotiations cites the decision to involve the IFIs as one of the key factors contributing toward success. Author interview, New York, April 2000. 26. See Chapter 14 by Charles Call in this book. 27. For Alvaro de Soto’s account of this period, see Alvaro de Soto and Graciana del Castillo, “Implementation of Comprehensive Peace Agreements: Staying the Course in El Salvador,” Global Governance 1, no. 2 (May–August 1995): 189–204. On the other hand, the El Salvador case is also one where the lack of coordination between the UN and the World Bank caused problems in the later stages of peace implementation. 28. For Aldo Ajello’s own account of his role as SRSG, see Aldo Ajello, “Mozambique: Implementation of the 1992 Peace Agreement,” in Crocker, Hampson, and Aall, Herding Cats. 29. Andrea Bartoli, “Mediating Peace in Mozambique: The Role of the Community of Sant’Egidio,” in Crocker, Hampson, and Aall, Herding Cats. 30. Comment by Andrea Bartoli, “Implementation of Peace Agreements in Civil Wars: Views of Practitioners,” conference presentation cosponsored by CISAC and IPA at Pocantico, New York, February 28–March 1, 1998. 31. See Cameron Hume, Ending Mozambique’s War: The Role of Mediation and Good Offices (Washington, D.C.: U.S. Institute of Peace, 1994). 32. See Roed-Larsen and Hooper, Command from the Saddle. 33. See David Malone, Decision Making in the UN Security Council: The Case of Haiti, 1990–1997 (Oxford: Oxford University Press, 1998). 34. See Chapter 13 by Caroline Hartzell in this book. 35. Malone, Decision Making in the UN Security Council. 36. On the role of Terje Roed-Larsen, as UN special coordinator for the Occupied Territories, in shaping donor policy toward the reconstruction of the Palestinian Territories, see in particular Rex Brynen, A Very Political Economy: Peacebuilding and Foreign Aid in the West Bank and Gaza (Washington, D.C.: U.S. Institute of Peace, 2000). 37. Alvaro de Soto, “Ending Violent Conflict in El Salvador,” in Crocker, Hampson, and Aall, Herding Cats. 38. Rex Brynen and others count this as one of the most important variables in the consolidation of the peace process at a critical juncture. See Brynen, A Very Political Economy. 39. Karen Guttieri, “Civil-Military Relations in Peace Implementation,” draft prepared for this project. 40. See David Malone, “Goodbye UNSCOM: A Sorry Tale in US-UN Relations,” Security Dialogue 30, no. 4 (December 1999): 393–411. 41. Author interviews, NATO member-state diplomats, New York, December 1999. 42. Author interviews with diplomats posted to the UN revealed a range of perspectives. Many within NATO—among them Canadian, French, and even some British diplomats—were concerned about the extent to which the United States was

S-4

11/24/09

4:18 PM

Page 115

THE CHALLENGES OF STRATEGIC COORDINATION

115

willing to push the organization beyond the UN framework. Nevertheless, many others questioned the relevance of the UN to European security; younger diplomats in particular preferred to seek solutions through the OSCE and other European instruments. 43. See Chapter 3 by Michael Doyle in this book. See also Michèle Griffin and Bruce Jones, “Building Peace Through Transitional Authority: New Directions, Major Challenges,” International Peacekeeping 7, no. 4 (winter 2000): 75–90. 44. Among my responsibilities in the UN Advance Mission in Kosovo were to draft the report of the Secretary-General on the establishment of the UN Mission in Kosovo (UNMIK) and to facilitate the planning process through which the EU, the OSCE, and the UN agreed on a division of labor within the mission structure. 45. I represented the Office for the Coordination of Humanitarian Affairs (OCHA) during these negotiations and was on the core team of drafters who compiled the final draft of the Generic Guidelines for consideration by the UN deputy secretary-general and the Administrative Committee on Coordination (ACC). 46. This observation was made on the basis of my responsibility, within OCHA, of monitoring the implementation of the Strategic Framework and other similar coordination mechanisms in UN missions. 47. See UN Document S/1999/779, “Secretary-General’s Report to the Security Council on the UN Administration in Kosovo,” July 12, 1999. 48. I was responsible for facilitating the drafting of the report S/1999/779, wherein these coordination mechanisms were established.

S-4

11/24/09

4:18 PM

Page 116

S-5

11/24/09

4:18 PM

Page 117

5 Settlement Terms and Postagreement Stability DONALD ROTHCHILD

The transition from peace agreement to a self-enforcing conflict management system after a civil war is an uncertain and complex undertaking. Generalizations about the process are problematic because each case reveals different levels of conflict intensity, duration, and internal party unity. Varying mixes of insecurity and predatory behavior are also at work. If those engaged in the peace process are unable to overcome collective fears of the future, the society will be vulnerable to increased manipulation by predators; if greedy predators are not dealt with effectively, fears of exploitation and vulnerability will escalate.1 These two elements in the peace implementation process are intertwined. As chapters by Michael Doyle, Joanna Spear, and Charles Call and William Stanley in this book make clear, a primary focus on security issues during the transition to a self-enforcing peace is essential. After that is achieved and regularized patterns of relations are established, international implementers and local peacemakers, if they are determined and responsible, will be in a stronger position to deal with predatory activities. During implementation, the acute fears that arise from short-term military security-building (such as concerns over “cheating” or surprise attack) are partially superseded by the diffuse uncertainties of institution-building, where former enemies must learn to occupy the same political space and develop stable and effective constitutional structures for a durable peace. Although in some cases mediators can attempt to bridge the challenges of disarmament and institution-building simultaneously (Northern Ireland, 1998), effective short-term implementation is often required to lay the foundation for a long-term development of state norms and institutionbuilding. Patterns of cooperative behavior cannot be fully accomplished until military security-building (the verification of the cease-fire, the can117

S-5

11/24/09

4:18 PM

118

Page 118

EVALUATING IMPLEMENTATION STRATEGIES

tonment of troops, demobilization, disarmament, and initial efforts to reintegrate the armed forces) has been largely achieved, and leaders and the public concentrate on coping with the arduous challenges of institutionbuilding and economic development. For a peace agreement to be self-sustaining, adversaries must build on their own intra- and intergroup reciprocities and political exchange relations to ensure stability. Conflicts between state and society and within the parties themselves can be anticipated, but to the extent that these encounters become bounded by an informal or formal conflict management system, cooperation can be anticipated through ongoing bargaining encounters at multiple levels.2 Where such a framework takes hold and the public comes to embrace new political institutions and rules of competition, the regime can be expected to provide all sides with a stake in maintaining the agreement.3 Then, as trust in institutions and rules of relations grow, the likelihood increases that agreements will be successfully consolidated. Encouraging examples of reciprocity and bargaining success during implementation include Nicaragua, El Salvador, Guatemala, Namibia, and Mozambique. During peace negotiations, bargaining parties generally seek to maximize their political, economic, and security interests well into the future. This leads to problems regarding the designing of the rules of political relations, because the short-term security concerns of the bargaining parties may be at odds with the long-term institution-building needs of the society. The difference between short- and long-term incentive structures arises most clearly over constitutional design issues relating to the nature of individual and group political representation and participation. During negotiations, weaker parties, fearing their vulnerability in a reintegrated political order, often attempt to exchange cooperation for terms of agreement that provide an element of political certainty about the future.4 Thus, weaker parties often seek long-term constitutional mechanisms that provide them with an assured share of political power and an access to a fair allotment of public resources, regardless of which set of political elites assumes office. But this effort to achieve political certainty can create problems of statecraft that may defy solution. Not only is it extremely difficult for one cluster of leaders to bind their successors, but the protections for group security that seemed so critical to leaders at the time of the peace negotiations may be overtaken by events later on and come to appear irrelevant, even counterproductive, in the altered circumstances of the postsettlement political landscape. Constitutional provisions to ensure group participation may thus have a perverse influence over time, reifying old group identities, displaying a lack of responsiveness to changing social concerns, and failing to adjust to the shifting demographic patterns in their society. In light of this possible tension between immediate security needs and longer-term requirements of governance, this chapter focuses primarily on

S-5

11/24/09

4:18 PM

Page 119

SETTLEMENT TERMS AND POSTAGREEMENT STABILITY

119

reconciling these different thrusts in an effort to achieve durable settlements. It examines the arenas of intragroup, state-societal, and interstate relations, and asks two related questions. First, what political and institutional strategies can be expected to reassure political minorities in the short term while creating the incentives to maintain a stable and self-enforcing peace after the initial military security-building challenges have been dealt with? And second, because state institutions are often weak in post–civil war circumstances, how can state capacity be strengthened while at the same time restraining the majority-backed government’s potential exploitation of minority interests? This chapter begins with an analysis of the costs and benefits of the two main strategies for representing groups in institutions of state after civil war—security arrangements based upon group rights and representation, and those based on individual rights and participation in a unitary state—in order to learn whether they promote stable agreements. The chapter then examines three group-based mechanisms that can cut across these grand strategies—the principle of proportional distribution, electoral systems that use some form of proportional representation, and cultural and social protections—to gain an understanding of their potential to expand choice.

Grand Strategies of Intergroup Relations Of all the choices encountered by those engaged in negotiations to end a civil war, none is more crucial than designing the representational basis of political institutions.5 In attempting to craft peace agreements in civil wars, mediators can pursue one of two grand strategies for representing groups in state institutions. The first involves political and security arrangements based on formal group rights arrangements; the second consists of political and security systems resting upon individual rights and participation in a unitary, centralized state. The choice between these two strategies is particularly relevant to institution-building after civil wars, when the bargaining parties must deal with especially strong feelings of ethnic, regional, religious, linguistic, class, and ideological insecurity. In order to alleviate the short-term security fears of the combatants, mediators tend to favor group-based institutions. But do the institutions that favor reaching a negotiated settlement also favor the likelihood that a settlement will stick? To answer the question, we need to ask several more: How do these competing grand strategies structure incentive patterns to build a stable, ongoing relationship among the main parties in the post–civil war period? What are the benefits and costs of making use of these different strategies for managing conflict? And what success have third parties had in encouraging the bargaining parties to agree to institutions that will allow for a reconciliation of the needs for group security and

S-5

11/24/09

4:18 PM

120

Page 120

EVALUATING IMPLEMENTATION STRATEGIES

constitutional change as new circumstances arise? If the issue of short-term security were the only pressing concern for mediators attempting to end civil wars, there would be little controversy over appropriate strategy. But a daunting challenge for consolidating peace on a self-sustaining basis is the rebuilding of the state’s capacity to govern. Civil wars bequeath weak states with fragile administrative structures, frail norms, underdeveloped civil societies, and collapsed economies.6 Unlike a stable conflict management situation, where institutional rules are widely understood and acquiesced to (even accepted) and large political choices on state-society relations become minimal in nature, the post–civil war state frequently lacks the capacity to translate its decisions into authoritative rules applicable to the society as a whole.7 The weak state lacks legitimacy and is unable to penetrate and integrate the society nominally under its control. It therefore often has little option but to negotiate with powerful class and identity group leaders to achieve system goals. Despite their general ineffectiveness, institutions of governance can still make a difference in the way they are designed to manage societal conflict. In the context of a weak state environment at the time the peace negotiations take place as well as during the implementation process, the choice of grand design influences and reflects the balance of group power. The structuring of relations that occurs may involve a direct process of centrally imposed order or an indirect process of accommodating to the reality of decentralized power among autonomous actors, with different effects on the ruling coalition’s ability to craft its desired policy objectives. Given the limited political and economic resources at the disposal of local state-builders and third parties for reorganizing and rebuilding the country after civil war, realistic assessments of the situation on the ground therefore become critically important. Practitioners must determine how scarce resources can best be used to achieve immediate military-security objectives, but not at a price that will create constitutional rigidities that will stymie subsequent reformulation. Hence, negotiators must make critical judgments about the linkage of short-term political, economic, and social goals and long-term institutional objectives. This is a complex task, because the uncertainties of the immediate postsettlement period give rise to a political climate of inflexibility that complicates later efforts for change. Formal Group-Based Security-Building Systems Formal group-based measures of political representation are largely a response to a political environment marked by deep-seated collective security fears as well as continued state weakness and unreliable information about the intentions of rival elites. Ethnicity and related identities are viewed here as a reflection of social interactions, not a predetermined and

S-5

11/24/09

4:18 PM

Page 121

SETTLEMENT TERMS AND POSTAGREEMENT STABILITY

121

fixed reality. As the state fails to protect its citizens and to provide for their well-being, the importance of the identity group is enhanced, largely because of its ability to offer security to its members. What exacerbates tensions is not ethnicity and religion per se, but the breakdown of regularized interactions, a process that can occur before or during the implementation of agreements. Such breakdowns prolong intergroup anxieties and suspicions, gravely complicating the tasks of creating or reconstructing overarching rules and institutional arrangements. Even during the later phases of the implementation process, the state may still be unable to perform its primary task of protecting all its citizens, and group members can remain, for the purposes of their security, ensconced in the apparent safe sanctuary of their ethnic or other identity-group confines (or “containers”).8 In worst cases, where political or militia leaders magnify these residual fears to preserve the unity of their group (against internal and external rivals), the effect may be to perpetuate intense conflict. In these circumstances, one approach to the problem of gaining agreement to new institutions is to build on existing configurations of power, seeking, as in Lebanon, Bosnia, Sierra Leone, and Northern Ireland, to accommodate state and societal elites through a continuum of powersharing designs. As used here, power-sharing institutions consist of rules that guarantee both inclusive and partitioned decisionmaking, or a combination of these.9 Thus power sharing refers both to constitutional norms calling for the inclusion of minority parties and major ethnic and religious group representatives in the legislature, government, and administrative positions of the country, as well as to the laws or constitutional arrangements providing for territorial or functional decentralization or electoral systems. Unless parties feel that they have a degree of certainty about the future, they may hesitate to accept a binding agreement ending a civil war. As Ben Reilly and Andrew Reynolds argue, “Transitional democracies, particularly those moving from a deep-rooted conflict situation, typically have a greater need for inclusiveness and a lower threshold for the robust rhetoric of adversarial politics than their established counterparts.”10 Where the majority is prepared to accept some form of group representation during the negotiations on an agreement, it signals its willingness to recognize the legitimacy of its opponents and indicates its preparedness to work with these rivals to solve joint problems. The effect of such arrangements is more mixed. On the one hand, concurrence on group representation signals to minorities that they will share in the control of the state and therefore be in a position to protect their interests after the agreement is implemented. This reassures weaker parties that they will not be marginalized, giving them an incentive to sign on to the contract. Thus, provisions on power sharing were included in Sierra Leone’s 1999 Lomé Peace Accord, and in line with these guidelines, insurgent leader Foday Sankoh was selected to head the powerful commission

S-5

11/24/09

4:18 PM

122

Page 122

EVALUATING IMPLEMENTATION STRATEGIES

on minerals and national reconstruction, and four Revolutionary United Front (RUF) members were appointed to the cabinet.11 On the other hand, as such agreements are implemented and the incentives of the stronger parties change from reassuring the weaker parties to those of governance, new uncertainties may arise.12 Particularly where military commands or singleparty political machines emerge and rule in an overbearing manner or where politicians engage in ethnic outbidding practices, power sharing can prove a frail safeguard for minority interests. In addition, external pressures by extremist elements in the diaspora community, who contribute financially to communal activities in the home country, can also favor one element within the government as against another. The instability of postsettlement constitution-building in Cambodia and Angola serves as a warning about the potential brittleness of formal power-sharing institutions.13 This brittleness makes it essential for mediators and negotiators to be wary of overly optimistic assumptions regarding the durability of such power-sharing mechanisms and to concentrate as far as possible on building robust institutions that will be able to endure the uncertainties of the future.14 Although group-based security-building measures represent a relatively easy means of signaling the dominant coalition’s willingness to respect the weaker party in the period following a negotiated settlement, the structure that these arrangements impose upon future state-society relations may prove to be an impediment to stable and effective governance. To the extent that such schemes perpetuate a struggle among identity and ideological interests, they can complicate the state’s tasks of governance and statebuilding and thereby can become the source of political instability. Under group-based arrangements, collective loyalties tend to look inward to the communal group rather than outward to the state. Extreme, outbidding politicians may heighten fear-inflating themes, resulting in incentive patterns that can deflect concern away from the development of intergroup trust and the emergence of an overarching vision of a community-wide future. Not surprisingly, where political actors are identified along groupbased lines, as under Lebanon’s Ta’if Agreement, some leaders have expressed a commitment in principle to overcoming a group-based system among confessional groups, viewing civic nationalism as the preferred basis for a long-term relationship.15 Because constitutional arrangements reflect different social contexts and power balances, they inevitably vary enormously from one another. In practice many postsettlement designs have not been exclusively group- or individual-based but have been applicable to both security formulas simultaneously. The use of mixed principles reflects a concern to reconcile state authority with reassurance for minority parties, signaling to weaker parties a limited preparedness to adjust to their political, security, and economic needs in order to gain their cooperation during the post–civil war period. In Zimbabwe, President Robert Mugabe carefully avoided a formal power-

S-5

11/24/09

4:18 PM

Page 123

SETTLEMENT TERMS AND POSTAGREEMENT STABILITY

123

sharing cabinet arrangement, while accepting a proportional representation (PR) system and the inclusion of twenty white members out of one hundred in the lower house elected on a white communal roll for at least the first seven years.16 Mixed principles were also evident in El Salvador; although there was no formal provision in the peace agreement for power sharing, the peace agreement did stipulate that 20 percent of the members of the new National Civilian Police force would consist of former members of the police and insurgent troops. 17 And although the Rome Peace Accord on Mozambique did not provide for formal provisos on governmental power sharing, the agreement did stipulate that the government and the Mozambique National Resistance (Renamo) were each to contribute 50 percent of the total strength of the newly formed Mozambican armed forces.18 Such pragmatism on the part of dominant coalitions accepts the authenticity of group uncertainties and results in institutional arrangements that seek to accommodate the demands of identity, ideological, and other interests for political representation. At this point, it is appropriate to analyze two major group-based, institutional mechanisms used after civil wars to design state-society relations: elite pacts and schemes of regional autonomy and federalism. Three other group-based instruments that can be fashioned to protect minority interests—the application of the principle of proportionality in recruiting group representatives to high office, electoral formulas to ensure group political representation, and group cultural and social protections—are left for subsequent discussion because of their relevance to both group- and individual-based security-building systems. Decisions on the use of these formal and informal devices to overcome uncertainty largely reflect the interests of the ruling coalition at a particular point in time and can be expected to shift as coalitions change and new incentives emerge. The possibility of a move from a group-based to an individual-based security arrangement as confidence increases must not be ruled out, as the experiences of Colombia and South Africa indicate. I will examine the political logic of each mechanism as an institution-building arrangement, the likelihood that such a tool can survive during the late stages of the implementation process, and the impact that such mechanisms can be expected to have on a country’s political stability and on its ability to achieve the economic and social goals it sets for itself. Elite pacts. When the state is weak and unable to enforce the peace in the post–civil war period, and when influential political and ethnoregional elites feel threatened by the prospect of rule by majority-backed governments in a centralized state, the main group leaders may craft powersharing arrangements that institutionalize their participation in political decisionmaking.19 The nature of elite participation in these institutional arrangements reflects the configurations of political and military power in

S-5

11/24/09

4:18 PM

124

Page 124

EVALUATING IMPLEMENTATION STRATEGIES

their society. The leaders included in the pact may be drawn from diverse constituencies—such as political parties and factions, business interests, the military, warlords, and ethnoregional notables—while others, from less powerful groups, may be shut out of positions of power. What pact members share in common is a limited set of political rules that will promote political stability and minimize threats from rival interests. Frequently, the critical element involved in these postsettlement agreements is protection of identity-group interests from threat. Timothy Sisk, emphasizing the defensive dimension, describes pacts as “mutual security agreements in which parties forswear the use of violence to achieve their aims in exchange for protection under agreed-upon rules of the political game.”20 Preliminary research indicates that identity-based pacts (Somalia, Liberia) tend to involve greater problems of commitment than their socioeconomic counterparts (Colombia, Venezuela), possibly indicating the identity groups’ lower level of information about the intentions of their adversaries.21 In an effort to reduce conflict among themselves, the members of identity-based pacts negotiate procedural rules to protect their interests during a transitional period. These rules, often informal in nature, may provide for shared decisionmaking on specified issues, the exclusion of other issues from government jurisdiction, the assignment of administrative positions, and the allocation of state revenues. Pacts are typically very temporary institutions, dependent upon the maintenance of balanced elite power and a preparedness to resolve conflicts among pact members through ongoing bargaining encounters. Because pacts are negotiated in dynamic political contexts, when group demographics, internal and external military alliances, and economic opportunities shift, the bargaining relationship is not likely to remain a static one. Elites may have incentives to mobilize their communities to demonstrate their power at the political center, with the effect of heightening intergroup tensions. The lack of a third-party enforcer on the scene requires that elites engage in continuing bargaining interactions. Such elites fear that any significant concessions of their autonomy will be interpreted by their rivals as a sign of weakness. Consequently, elite pacts can be expected to display considerable fragility over time, as a changing balance of group power and the possibility of outbidding lead to new conflicts and institutional relationships. The case of Liberia is instructive. After years of warfare, the five main militia chiefs agreed in 1995 at Abuja to a collective presidency and cabinet. Unable to win an outright victory, these militia leaders accepted a loose power-sharing arrangement to minimize the threat that each leader posed to the others.22 As Adekeye Adebajo describes in Chapter 20, all the warlords were involved in the Council of State “and were therefore thought to have a direct stake in keeping the peace.” However, the flimsiness of this arrangement soon became apparent. Former adversaries Charles Taylor and

S-5

11/24/09

4:18 PM

Page 125

SETTLEMENT TERMS AND POSTAGREEMENT STABILITY

125

General Alhaji Kromah joined forces in 1996 to attack another pact member, General Roosevelt Johnson. A firefight erupted between their opposing militias in Monrovia and could only be halted when the foreign ministers of the Economic Community of West African States (ECOWAS) interceded and renegotiated the fragile pact. Elsewhere, pact solutions have been rejected as sources of instability, paralysis, and ineffectiveness. In Somalia, the UN, with active support from Ethiopia and the United States, sought to negotiate a pacted arrangement among the fifteen main warlords and some civil society representatives at Addis Ababa in March 1993. Although many of the assembled delegates did agree to a pact of militia leaders, the arrangement remained stillborn because the two main warlords, General Mohamed Farah Aidid and Ali Mahdi, displayed “a vested interest in continued conflict and instability.”23 In Guatemala, as William Stanley and David Holiday emphasize in Chapter 15, the pact between the two main groupings in that country did little to build a consensus to implement vitally needed reforms, while in Lebanon, as Marie-Joëlle Zahar discusses in Chapter 19, the pact of three presidents has strengthened cleavages among confessional groups, leading to a paralysis of decisionmaking on critical issues. In brief, elite pacts hold out the prospect of considerable autonomous power for the political elites included within the decisionmaking sphere, but the extension of such autonomy results in temporary and unstable solutions in a dynamic post–civil war context. Regional autonomy and federalism. In Bosnia, Kosovo, Ethiopia, Sudan (1972), Nicaragua, and the Philippines, peacemakers have used various forms of territorial autonomy to respond to minority demands to empower them with important political, social, and cultural functions at the regional level and thereby to allay their fears of majority domination and exploitation. Under a formal federal system, the constitution divides state power between central and regional governments, “each of which, in its own sphere, is co-ordinate with the others and independent of them.”24 Whereas centralized federal systems are marked by very asymmetrical state-regional relations, decentralized federal arrangements display broad powers of autonomous decisionmaking over regional affairs. Regional autonomy involves a grant by a law, treaty, or constitutional arrangement of specific powers and responsibilities to a regional authority or authorities. Success in designing autonomy arrangements after civil war entails setting out the details of this division of power precisely and requires a willingness on the part of political leaders to act with restraint.25 This grant of authority to a region can sometimes have unintended consequences. For example, because the regional government has jurisdiction over all the people interspersed within its territory, this may mean that an ethnic heartland area’s limited self-determination will result in the creation of new ethnic,

S-5

11/24/09

4:18 PM

126

Page 126

EVALUATING IMPLEMENTATION STRATEGIES

religious, or other minorities and therefore prove potentially conflictproducing.26 When a broadly autonomous Kosovo emerged after the implementation of the peace agreement, the remaining Serbs found themselves governed by a predominantly Kosovar leadership, increasing their fears of victimization and encouraging their efforts to barricade themselves in safe enclaves. In the short term at least, territorial decentralization can help to overcome group insecurity about the future by sending reassuring signals to minority interests about their autonomous powers for dealing with certain local matters. Because territorial decentralization can be formulated to accommodate well-entrenched and spatially separated identity groups, many people involved in crafting postsettlement designs viewed it as a useful institution-building mechanism. For example, political autonomy was the guiding principle underlying the 1996 agreement between the government of the Philippines and the Muslim secessionist Moro National Liberation Front; even so, the autonomy arrangement set up under this agreement is still not acceptable to one of the opposition groups, the Moro Islamic Liberation Front. In Ethiopia, following the defeat of the authoritarian and centralist-inclined regime of President Mengistu Haile Mariam in 1991, Meles Zenawi, the president of the new ruling Ethiopian People’s Revolutionary Democratic Front (EPRDF), issued a series of proclamations empowering Ethiopia’s “nations” with broad jurisdiction over affairs taking place in their regions.27 Marina Ottaway concludes that the Ethiopian government, with extensive backing from external donors, has in fact transferred bloc grants to the regions, enabling them to make and implement autonomous policy decisions.28 Territorial autonomy has also been used at times to reassure dispersed groups who did not occupy a separate political space prior to the peace settlement. In Nicaragua, for example, the Chamorro regime sent a signal of support to the Contras by establishing twenty-three development zones on land not controlled by the Contras, where they could settle and exercise a degree of autonomous control.29 Although territorial decentralization seems a logical response to minority demands, it can prove a fragile arrangement that is difficult to agree upon and maintain. In Sri Lanka, the proposed merger of the north and east under the 1987 agreement was suspect in Tamil eyes, in part because of a lack of adequate powers granted to regional authorities.30 Territorial autonomy, proposed as a possible solution to Somalia’s overcentralization, would have increased tensions, according to Ameen Jan, because it would have imposed artificial territorial divisions on the clans.31 And in Bosnia and Kosovo, the continued oversight of a third party creates doubts about the stability of the autonomy plan after the external enforcer disengages. Without the guarantee of supranational institutions, as in Europe, autonomy arrangements seem an unstable solution in post–civil war circumstances.

S-5

11/24/09

4:18 PM

Page 127

SETTLEMENT TERMS AND POSTAGREEMENT STABILITY

127

What makes regional autonomy and federalism fragile institutional arrangements is the threat of instability that may occur when the structure of incentives shifts from reassurance to governance and leads to the dominance of one of the parties during the later phases of implementation. Two possible trajectories may appear—one toward political centralization and the consolidation of majoritarian power, and the other toward further political decentralization and a possible breakup of the federation.32 In the first trajectory, political majorities, as exemplified by South Africa, hold out the prospect of federalism to reassure ethnic and racial minorities about their ability to exercise a limited authority over local matters after the transfer of power. Then, after the founding election and after the new, majoritysupported government has consolidated its position, the structure of incentives can shift toward a further hardening of the majority party’s already firm grip on power. By contrast, a trend toward further territorial decentralization was revealed in federal states, where pressure from ethnonationalist interests in the regions led to a fragmentation of the country into its major constituent republics (exemplified by the former Yugoslavia and the former Soviet Union).33 This breakup can become a source of instability, as new political and identity-based minorities emerge and the successor states establish new international alignments. Thus, an unstable union resulted from the 1995 Dayton Accords, with Bosnia-Herzegovina now loosely governed by a multiethnic, federal government based in Sarajevo. Its two provinces of Republika Srpska and the Muslim-Croat Federation exercise considerable powers at the substate level (including separate armed forces and a limited right to negotiate external agreements). This agreement created an unsteady arrangement that may further fragment once the external enforcer withdraws. Clearly, where negotiators promised to include federal-type provisions in a postsettlement constitution, this was reassuring to weaker parties at the bargaining table, as indicated by the experiences of Sudan (1972) and Ethiopia. Preliminary statistical data indicate that the inclusion of provisions on territorial autonomy and federalism in peace agreements had stabilizing effects, at least in the short term.34 This effect was carried over into the early stages of the implementation process, and for a time it helped steady an agreement. After about ten years, however, negotiated agreements to civil wars with a territorial autonomy provision did not ensure the success of an agreement any more than those negotiated agreements that lacked a territorial autonomy provision.35 Such instability ten years after the signing of an agreement is explained in part by the minority’s fear for its future and by the majority’s difficulty in committing credibly to the provisions on decentralizing political power. As memories of the peace process dim and politicians focus increasingly on consolidating effective political

S-5

11/24/09

4:18 PM

128

Page 128

EVALUATING IMPLEMENTATION STRATEGIES

control and achieving their policy goals, the long-term stability of federaltype institutions may come into doubt (Sudan, South Africa, Pakistan). This shift in priorities and power balances resulted at times in a trajectory toward either political centralization or decentralization.36 Individual-Based Security-Building Designs The possibility of establishing an individual-based, majoritarian approach, whether by conscious decision at the time of negotiations or through redesign afterward, brings us to the second type of grand strategy for post–civil war security-building. Many state leaders, indicating their misgivings about the potential immobility and instability arising from formal group-based security-building arrangements, champion instead an individualist and integral state approach. Rather than make binding concessions to the main group interests to assure their inclusion at the political center, they opt instead for a formula of individual rights within a unitary state. In doing this, they may be prepared to concentrate sufficient power in executive hands to ensure governmental primacy. Unitary government is valued for its ability to reduce transaction costs between the state and society, thereby avoiding delays and “log-rolling” practices that result from extended bargaining encounters. Although informal practices of ethnic arithmetic often become a matter of routine in soft authoritarian as well as democratic regimes, formal power-sharing arrangements are not always regarded by leaders of majority interests as an efficient or legitimate structure for future intergroup relations. 37 Lebanese constitution-makers, concerned over the divisive potential of confessionalism as experienced in the past, made specific reference to the representation of confessional groups in the Chamber of Deputies and the cabinet in the 1990 constitution; at the same time, they expressed a preference for abolishing political confessionalism over the long term.38 In Ethiopia, moreover, critics of ethnic federalism contend that its constitutionally entrenched system of decentralized, ethnic-based political authority will likely cause new conflicts and secessionist demands, because it structures decisionmaking along separate lines.39 These critics look to unitary governance to create an integrated “nation,” thereby preventing separatistinclined interests from gaining an institutionally sanctioned base for maneuver. Some scholarly research has been supportive of these contentions. Thus, Jack Snyder, raising serious doubts about power-sharing mechanisms in the context of democratizing societies, argues for a “commitment to the creation of a dense web of such ethnic-blind institutions [as the highly professionalized civil service, police, armed forces, and courts] can serve as the basis for a civic nationalism that is based not on the coexistence of ethnic groups but on the civil rights and duties of individuals.”40

S-5

11/24/09

4:18 PM

Page 129

SETTLEMENT TERMS AND POSTAGREEMENT STABILITY

129

And Philip Roeder, after examining Eurasia’s postcommunist regimes, concludes that “the most successful containment of ethnic conflict below the level of ethnoconstitutional crises has come from unitary states with extensive and inclusive civil liberties for individuals.”41 The combination of individual rights and unitary governance frequently finds formal expression either in some type of majoritarian (one-person, one-vote) democracy or in a soft authoritarian regime. Both majoritarian democracies and soft authoritarianism can offer relative stability, provided that elites and their supporters have the physical and cultural security and economic opportunity to make accommodative behavior appear beneficial.42 Provided that the democratic or soft authoritarian state can protect associational life in various domains and respond constructively to reasonable citizen demands, it will be in a position to develop public confidence in its actions.43 The state, then, has an enormous capacity to be a conflict manager on its own, so long as it determines policies in line with its norms on equity and fairness. For the most part, advocates of unitary government reject phase-byphase movement toward individual rights and democratization, preferring instead to hold founding elections based upon the majoritarian principle toward the end of the military/security-building phase of implementation. Joaquim Chissano, president of the Front for the Liberation of Mozambique (Frelimo), exemplified consistent dedication to an individual-based, unitary state approach to institution-building. Chissano rejected Western appeals for a group-based, power-sharing arrangement in the cabinet prior to the 1994 general election. His opponent, Renamo leader Afonso Dhlakama, did agree, reluctantly, to Chissano’s terms, even though a winner-take-all electoral process was viewed as posing decided political risks for him. Then, following his party’s defeat in the founding elections (based on a list system of PR), Dhlakama did accede to the norms of electoral democracy and assumed his role as the leader of the opposition.44 But not all insurgent leaders, fearing for their future under a genuinely democratic system, can be expected to commit themselves so readily to the risks of electoral defeat. The distrust that marks a postsettlement relationship magnifies the uncertainties of elections for societies emerging from civil war, frequently requiring those crafting electoral institutions to choose between competing preferences on individual- and group-based security arrangements. For well-established parties, such as Frelimo, the risks of the first course seem minimal, and the results in terms of stable societal relations may prove more than worth the perils of conceding too much space for maneuver to former insurgents. For smaller, less institutionalized competitors such as Renamo, however, some form of power sharing logically has appeal as a defense mechanism against future party domination. Depending upon the balance of power between contending forces and the

S-5

11/24/09

4:18 PM

130

Page 130

EVALUATING IMPLEMENTATION STRATEGIES

extent of social and ideological distance between the parties, then, both political arrangements can seem reasonable, and can be justified or rejected on the grounds of different benefit-cost calculations. Clearly, there is no one formula that fits all contexts when it comes to designing peace accords. Despite the fact that unitary government with protections for individual rights appears to offer some advantages in terms of effective governance, reduced transactions costs, and long-term stability in postconflict environments, a group-based system (as in Lebanon) may be the only available compromise on which a credible commitment can be reached. Political elites must have an eye to the particular political environments in which they are crafting institutions for the future, displaying a combination of realism, fairness, and goodwill as they negotiate on the specifics of policy. The goodwill of the ruling elite is critical under any type of regime adopted, because it may be able to engender public confidence, minority and majority alike, in the commitment of state leaders to public security and well-being. A fair-minded government whose processes of reaching decisions are transparent can bolster societal trust in the state and its institutions over time. This might seem easier to achieve under a unitary system that respects individual rights, but a group-based system whose government promotes interpersonal and interassociational trust also might lead over time to political legitimacy and a self-enforcing peace.45

Expanding Choice: Nonthreatening Group-Based Formulas Certainly not all political choices are dichotomous. Some that are appropriate to both formal group-based and individual-based security-building approaches are not threatening to state power and therefore have the potential to expand choice. This flexibility makes proportionality formulas or group cultural or social protections particularly relevant for increasing societal commitment to peace agreements. Three such institutional mechanisms have a potential utility for the framework of rules and expectations for long-term commitment to the peace process. First, the proportionality principle has broad relevance to soft authoritarian or democratic regimes or to formal and informal power-sharing arrangements. In legislative representation, appointments to the cabinet, legislative committees, and high party posts, and in the allocation of resources to the regional and local governments, the use of the proportionality principle in the postsettlement period can signal to minorities that the ruling coalition is prepared to share power and resources with them. Even more directly related to security concerns, its use in recruiting members of the police and military services can help to build confidence. The general acceptability of the proportionality principle lies in its routinization of deci-

S-5

11/24/09

4:18 PM

Page 131

SETTLEMENT TERMS AND POSTAGREEMENT STABILITY

131

sions on some highly sensitive issues, for allocations and appointments are distributed according to a neutral formula to all regions and major groups in terms of their proportionate number of people in the population, and irrespective of whether these units or groups are included in the ruling state coalition. Proportionality diverges from the majority-rule principle, as Arend Lijphart observes, in that it “makes a minimum winning coalition less profitable and therefore less probable.”46 Its key elements involve a ruling political elite prepared to use a guideline on balanced recruitment and allocations and to ensure that its criteria are continually updated to take account of demographic and other changes. In line with these guidelines, it seems possible that had the Lebanese Christians been more responsive to population shifts occurring in their midst, their country’s terrible civil war might never have occurred. The record of post–civil war governments in applying the formal proportionality norm has been a decidedly mixed one. Southern Sudanese have grounds for complaining that they did not get a fair share of current and long-term development allocations following the Addis Ababa settlement in the late 1970s.47 Nigerians, however, displayed a greater degree of sensitivity regarding regional allocations after the Biafran war. Statistical data for an extended time period following Nigeria’s civil war indicate that budgetary allocations tended toward proportional values in the late 1970s and early 1980s.48 As a consequence, the relatively advantaged oil-producing states showed declines in per capita expenditures on social services, while those for such relatively disadvantaged states as North-Eastern and NorthCentral rose significantly. While this move toward a needs-based formula left many peoples in the oil-producing areas dissatisfied, the general equalization of opportunity that resulted may have had an overall stabilizing effect. The willingness of governments to concede proportionality in regard to appointments to high government and administrative positions in postsettlement times also sends important signals of the majority’s preparedness to share significant decisionmaking powers with minority interests. The inclusion of weaker parties in government provides them with incentives to commit to the regime during the difficult transition to a self-sustaining peace.49 A study by Daniel C. Esty and others of significant predictors of state failure from 1955 to 1994, in fact, indicates that the ethnic character of the ruling elite is one of the best predictors of state failure (accurate 78 percent of the time). Thus, if the ruling coalition represented only one group in an ethnically divided society, it significantly increased the risk of state failure.50 Implementation experience does indicate that the preparedness of governments to include important political, ideological, or identity interests in postsettlement regimes has had a calming effect on likely minority parties. Thus Nicaragua’s Violeta Chamorro, upon coming to power, astutely

S-5

11/24/09

4:18 PM

132

Page 132

EVALUATING IMPLEMENTATION STRATEGIES

retained Sandinista-appointed military officers in high security positions. Similarly, in South Africa the 1994 interim constitution provided that parties winning over 5 percent of the seats in the National Assembly would be included in the cabinet on a proportional basis for a five-year period. As a consequence, the 1994 election resulted in a coalition cabinet consisting of eighteen members drawn from the majority African National Congress and nine members from the two minority parties—six from the National Party (NP) and three from the Inkatha Freedom Party (IFP). Although the NP leadership soon withdrew from this ruling coalition, the implementation of a proportional recruitment formula in the initial military/security-building phase was reassuring to moderate African nationalists and local whites as well as to the international investment community; also it eased the transition from formal to informal mechanisms of power sharing.51 Provided that governments continue to utilize the proportionality norm as the incentive structure shifts in the late stages of the implementation process, and provided they are prepared to appoint authentic group spokespersons to important ministries over time, the proportionality norm can be expected to enhance political stability and the ability of the state to achieve its developmental objectives. By contrast, despite momentary hopes for genuine power sharing in Sri Lanka, Cambodia, and Angola, the failure of the ruling coalition to bring a distrustful and sometimes resistant opposition into the inner circles of government has had destabilizing consequences. A second institutional means of promoting proportionality that can be relevant for individual-based security arrangements is through the electoral system. In adopting some variant of proportional representation, the proponents of such a system may consciously seek to reassure minority ideological and identity groups that they will be represented in the legislature, roughly in accordance with their population size in the country. This can give them a greater opportunity to exert influence from within the country’s ruling coalition. Those championing PR seek thereby to allay minority fears that a one-person, one-vote majoritarian system will have undesirable outcomes, leading, for example, to ethnoregional dominance in a postsettlement context or to dangerous ethnic outbidding practices by ambitious politicians. Under certain circumstances, where the party lists are ethnically balanced or ethnic parties come to play a critical role in legislature, the effect may be to increase the likelihood of political stability through inclusive politics. This may encourage moderation, giving minority leaders an incentive to work within the political system to promote change, but at a possible cost in terms of indecisive political and economic leadership. Because of their ability to assure minorities about their future access to the inner circles of power, it is not surprising that PR systems have been widely utilized in negotiated settlements. In Nicaragua, Cambodia, Bosnia, Zimbabwe (1980), Angola, Mozambique, Namibia, and South Africa, PR arrangements have largely reflected the prevailing configurations of power

S-5

11/24/09

4:18 PM

Page 133

SETTLEMENT TERMS AND POSTAGREEMENT STABILITY

133

during the transition. Where strong ethnoregional parties were in evidence, as in Angola, parties agreed on the use of the least proportional of the formulas (the d’Hondt system) in legislative elections, advantaging the major parties and virtually excluding the smaller parties from effective political representation. Where negotiators sought to further the inclusion of powerful racial and ethnic minority interests in the country’s decision process, however, the bargaining parties made use of the LR-Hare formula (in Namibia and Zimbabwe, for the common roll seats in its 1980 election) or the LR-Droop formula (in South Africa), two of the most proportional of the PR systems.52 On the whole the effects of adopting PR in these postsettlement elections have been mixed. PR reassured whites and moderate blacks in southern Africa that they would play a more meaningful political role after the transition to nonracial politics; also, because of the wide-ranging coalitions that might emerge within and between parties, it was assumed (erroneously in the case of Bosnia) that extreme nationalist parties would be inhibited.53 Nevertheless, it seems imprudent to attribute too much to PR as a means of safeguarding minority interests and promoting conflict management when clear majorities are present. Where identity groups are spatially separate and vote overwhelmingly for their party candidates, the electoral outcomes may be little different from those under a plurality system. Thus, the South West African People’s Organization (SWAPO) secured 73.9 percent of the vote in Namibia’s second parliamentary elections in 1994, well over the two-thirds majority required to amend the constitution. Inclusiveness in Namibia resulted more from the majority-backed government’s pragmatic actions when agreeing to a constitution by consensus than from electoral engineering.54 And even if it is reasonable to conclude that PR provides reassurance for minority interests about their participation in legislative activities, the survival of the PR system seems likely to depend on the continued willingness of the majority party to utilize some variant of it over time. Concerns over the lack of contact between constituents and their representatives in large, multimember constituencies have contributed in some cases to electoral redesigning that links the single-member district system with PR (for example, in South Africa). Third, the state can offer group cultural and social protections irrespective of individual- or group-based security approaches. Such “nonterritorial federalism,” which allows communities a limited autonomy on the basis of what is referred to as the “personality principle,” is not normally threatening to state power. It concedes specified responsibilities to these communities on matters of language, religion, education, and the roles of traditional authorities. 55 The granting of such autonomous rights has the effect of engendering minority-group confidence in the majority and its leaders, and at a relatively low cost to the central government. In South Africa, where President Nelson Mandela urged public sensitivity on the question of

S-5

11/24/09

4:18 PM

134

Page 134

EVALUATING IMPLEMENTATION STRATEGIES

removing Afrikaner monuments, and where the 1993 constitution made eleven languages (including isiZulu, Afrikaans, and English) official languages at the national level, these actions signaled government goodwill at the very time that the transition to majority rule was occurring.56 But majority-backed concessions on social and cultural autonomy are not tantamount to unqualified communal control; not surprisingly, therefore, both Palestinians and Israelis have refused during their negotiations to let sovereign control of the Temple Mount or Noble Sanctuary or the lands beneath it slip from their political grasp. In sum, provisions for group protections and autonomy generally represent concessions by the ruling majority coalition to the minority. Paradoxically, because communal protections and autonomy demand less of central governments than formal agreements to share political power, these limited concessions have enlarged choice, outlasting some more farreaching constitutional accommodations as elite pacts, federalism, and other formal power-sharing arrangements. The experiences of Lebanon, India, South Africa, Latvia, Israel, and elsewhere indicate that governments may be prepared to allow communal elites to exert a limited control over personal status laws and institutions, provided that this does not entail an abuse of their members’ civil liberties or a fundamental challenge to the state’s overall leadership.

Conclusion The successful implementation of peace agreements after civil wars requires overcoming predatory behavior, addressing deeply held grievances, and assuaging security fears. Civil wars are high-stakes conflicts over the organizing principles and fundamental rules of the polity. Because the issues are so basic, the means adopted are often totalistic and the outcomes fearsomely destructive. However, when the conflict is perceived as unacceptably costly and unwinnable (in the short term at least), negotiators may produce something tantamount to a “cold peace.” Even so, the diffuse fears and uncertainties that survive the war create problems of reliable information about adversary intentions and credible commitment to peace accords that are difficult to surmount. External actors may sometimes be prepared to intervene and protect the weaker parties during the initial phases of the peace implementation process, but this involvement is likely to diminish over time. This leaves the parties largely on their own to build legitimate and effective institutions for managing conflict. Distrust of adversary intentions after civil war may be logical given the widespread uncertainty that an opponent may “cheat” on its commitments and take advantage of the momentary opportunity to launch a preemptive

S-5

11/24/09

4:18 PM

Page 135

SETTLEMENT TERMS AND POSTAGREEMENT STABILITY

135

military strike. Because of intense collective fears surrounding the initial transition process, the adversaries will likely insist on knowing in detail what provisions will be made in the settlement for their security and for political participation after the agreement has come into effect. This short-term focus on security can have unfortunate consequences for later peacebuilding, however. In some cases, the very incentives required to reassure weaker parties and gain commitment to a bargain in the initial period can cause political instability later on, if rigid power-sharing institutions come into conflict with the need to concentrate power at the political center in a unitary state structure. Because the grand strategies of formal group-based security-building and individual-based security-building create their own diverse fears of domination and immobilism, it makes sense to use institutions such as proportional appointments and allocations, proportional representation, and group cultural and social autonomy arrangements that cut across both strategies. Such institutions can promote respect for group rights without threatening the majority-backed state elite’s political leadership. Clearly, enormous skill is required during the peace negotiations to reconcile the needs of short-term security with those of effective governance over time. By attempting to find a bridge between the political logics of the two grand strategies of intergroup relations, it may be possible to help stabilize peace agreements for the long duration.

Notes 1. Jack Snyder and Robert Jervis, “Civil War and the Security Dilemma,” in Barbara F. Walter and Jack Snyder, eds., Civil War, Insecurity, and Intervention (New York: Columbia University Press, 1999), p. 23; and Paul Collier and Anke Hoeffler, Greed and Grievance in Civil War, Policy Research Working Paper no. 2355 (Washington, D.C.: World Bank, Development Research Group, May 2000). 2. Robert D. Putnam, “Diplomacy and Domestic Politics: The Logic of TwoLevel Games,” International Organization 42, no. 3 (summer 1988): 433–435. 3. See Chapter 13 by Caroline Hartzell in this book. 4. Barbara F. Walter, “Designing Transitions from Civil War,” International Security 24, no. 1 (summer 1999): 127–155. 5. To be sure, classifications based on such related issues as electoral and party systems, presidential and parliamentary regime patterns, and the presence or absence of security safeguards are also important; nevertheless, in this chapter, I concentrate on the decision in the agreement to share or not to share group power as the core principle of postsettlement cooperation. 6. Donald Rothchild, “The Effects of State Crisis on African Interstate Relations (and Comparisons with Post-Soviet Eurasia),” in Mark R. Beissinger and Crawford Young, eds., Beyond State Crisis? Postcolonial Africa and Post-Soviet Eurasia in Comparative Perspective (Washington, D.C.: Woodrow Wilson Center Press, 2002), pp. 189–214. 7. Eric A. Nordlinger, On the Autonomy of the Democratic State (Cambridge: Harvard University Press, 1981), p. 8.

S-5

11/24/09

4:18 PM

136

Page 136

EVALUATING IMPLEMENTATION STRATEGIES

8. Manfred Halpern, “Choosing Between Ways of Life and Death and Between Forms of Democracy: An Archetypal Analysis,” Alternatives 12, no. 1 (January 1987): 13. 9. Donald Rothchild and Philip G. Roeder, “The Role of Power Sharing in Designing Stable Democracies,” paper presented at the Institute on Global Conflict and Cooperation (IGCC) Conference on Powersharing and Peacemaking, La Jolla, Calif., December 8–9, 2000. 10. Ben Reilly and Andrew Reynolds, Electoral Systems and Conflict in Divided Societies, Papers on International Conflict Resolution no. 2 (Washington, D.C.: National Academy Press, 1999), p. 16. 11. Abiodun Alao, “Sierra Leone,” draft prepared for this project. 12. Philip G. Roeder, “Preventing the Next Ethnic Crisis: Robust Institutions for the Management of Conflict,” paper presented at the IGCC Conference on Durable Peace Settlements, La Jolla, Calif., May 7–8, 1999. 13. Donald Rothchild, Managing Ethnic Conflict in Africa: Pressures and Incentives for Cooperation (Washington, D.C.: Brookings Institution, 1997), pp. 136–139. 14. See, for example, Carnegie Commission on Preventing Deadly Conflict, Preventing Deadly Conflict: Final Report (Washington, D.C.: Carnegie Commission on Preventing Deadly Conflict, 1997), p. 82. 15. See Chapter 19 by Marie-Joëlle Zahar in this book. 16. Rothchild, Managing Ethnic Conflict in Africa, p. 175. 17. Caroline A. Hartzell, “Explaining the Stability of Negotiated Settlements to Intrastate Wars,” Journal of Conflict Resolution 43, no. 1 (February 1999): 8. 18. Joao Honwana, “Mozambique,” draft prepared for this project. 19. This analysis of political pacts is drawn from Donald Rothchild, “Ethnic Insecurity, Peace Agreements, and State Building,” in Richard Joseph, ed., State, Conflict, and Democracy in Africa (Boulder, Colo.: Lynne Rienner, 1999), pp. 319–337; Caroline Hartzell and Donald Rothchild, “Political Pacts as Negotiated Agreements: Comparing Ethnic and Non-Ethnic Cases,” International Negotiation 2, no. 1 (1997): 147–171; Terry Lynn Karl, “Dilemmas of Democratization in Latin America,” Comparative Politics 23, no. 1 (1990): 1–21; and Timothy D. Sisk, Democratization in South Africa: The Elusive Social Contract (Princeton: Princeton University Press, 1995). 20. Timothy D. Sisk, Power Sharing and International Mediation in Ethnic Conflicts (Washington, D.C.: U.S. Institute of Peace, 1996), p. 81. 21. Hartzell and Rothchild, “Political Pacts as Negotiated Agreements,” p. 163. See also Terry Lynn Karl, “Petroleum and Political Pacts: The Transition to Democracy in Venezuela,” in Guillermo O’Donnell, Philippe C. Schmitter, and Laurence Whitehead, eds., Transitions from Authoritarian Rule: Latin America (Baltimore: Johns Hopkins University Press, 1986), pp. 197–198. 22. On the composition of the pact, see Rothchild, “Ethnic Insecurity, Peace Agreements, and State Building,” p. 332. 23. Ameen Jan, “Somalia,” draft prepared for this project. 24. K. C. Wheare, Federal Government, 4th ed. (London: Oxford University Press, 1946), p. 33. 25. Ted Gurr, Minorities at Risk: A Global View of Ethnopolitical Conflicts (Washington, D.C.: U.S. Institute of Peace, 1993), pp. 319–320. 26. Ruth Lapidoth, Autonomy: Flexible Solutions to Ethnic Conflicts (Washington, D.C.: U.S. Institute of Peace, 1996), p. 39. 27. Rothchild, Managing Ethnic Conflict in Africa, p. 56. 28. Marina Ottaway, Africa’s New Leaders: Democracy or State Reconstruc-

S-5

11/24/09

4:18 PM

Page 137

SETTLEMENT TERMS AND POSTAGREEMENT STABILITY

137

tion? (Washington, D.C.: Carnegie Endowment for International Peace, 1999), pp. 71–72. 29. Donald Rothchild and Caroline Hartzell, “Security in Deeply Divided Societies: The Role of Territorial Autonomy,” Nationalism and Ethnic Politics 5, nos. 3–4 (autumn/winter 1999): 254–271. 30. See Chapter 21 by Sumantra Bose in this book. 31. Jan, “Somalia,” pp. 20–21. 32. David A. Lake and Donald Rothchild, “Political Decentralization and Civil War Settlements,” paper presented at the IGCC Conference on Powersharing and Peacemaking. 33. Ivan Vejvoda, “Yugoslavia 1945–91: From Decentralization Without Democracy to Dissolution,” in D. A. Dyker and Ivan Vejvoda, eds., Yugoslavia and After: A Study in Fragmentation, Despair, and Rebirth (London: Longman, 1996), p. 15; and Gail W. Lapidus and Edward W. Walker, “Nationalism, Regionalism, and Federalism: Center-Periphery Relations in Post-Communist Russia,” in Gail W. Lapidus, ed., The New Russia: Troubled Transformation (Boulder, Colo.: Westview Press, 1995), pp. 80–84. 34. Caroline Hartzell, Matthew Hoddie, and Donald Rothchild, “Stabilizing the Peace After Civil War: An Investigation of Some Key Variables,” International Organization 55, no. 1 (winter 2001): 199. 35. Data analyzed for Hartzell, Hoddie, and Rothchild, “Stabilizing the Peace.” 36. Lake and Rothchild, “Political Decentralization and Civil War Settlements.” 37. Rothchild, Managing Ethnic Conflict in Africa, p. 63. 38. Art. 95 of the constitution of Lebanon; see The Beirut Review 1, no. 1 (spring 1991): 158. 39. Francis M. Deng, Sadikiel Kimaro, Terrence Lyons, Donald Rothchild, and I. William Zartman, Sovereignty as Responsibility: Conflict Management in Africa (Washington, D.C.: Brookings Institution, 1996), p. 82; and Marina Ottaway, Democratization and Ethnic Nationalism: African and Eastern European Experiences (Washington, D.C.: Overseas Development Council, 1994), p. 48. 40. Jack Snyder, From Voting to Violence: Democratization and Nationalist Conflict (New York: W. W. Norton, 2000), pp. 330, 332. 41. Philip G. Roeder, “Peoples and States After 1989: The Political Costs of Incomplete National Revolutions,” Slavic Review 58, no. 4 (winter 1999): 854–882. 42. Roy Licklider, “False Hopes? Democracy and the Resumption of Civil War,” and Elisabeth J. Wood, “Durable Settlement of Civil Conflicts in Oligarchic Societies: El Salvador, South Africa, and Guatemala,” papers presented at the IGCC Conference on Durable Peace Settlements. 43. Margaret Levi, “Trust, Trade, and the Role of Government,” APSA-CP 10, no. 1 (winter 1999): 19. 44. Joao Honwana, “Mozambique,” draft prepared for this project, pp. 31–32; Cameron Hume, Ending Mozambique’s War: The Role of Mediation and Good Offices (Washington, D.C.: U.S. Institute of Peace, 1994), p. 29; and Rothchild, “Ethnic Insecurity, Peace Agreements, and State Building,” pp. 329–330. 45. Levi, “Trust, Trade, and the Role of Government,” p. 19. 46. Arend Lijphart, Democracy in Plural Societies: A Comparative Exploration (New Haven: Yale University Press, 1977), p. 39. 47. Rothchild, Managing Ethnic Conflict in Africa, pp. 76–77. 48. The index of variation between 1975–1976 and 1979–1980 in the Nigerian road program decreased from 1.18 to 0.48, while the decline in the health and edu-

S-5

11/24/09

4:18 PM

138

Page 138

EVALUATING IMPLEMENTATION STRATEGIES

cation indexes during that period was less precipitous: from 0.96 to 0.88 in health expenditures and 0.70 to 0.51 in education expenditures. Rothchild, Managing Ethnic Conflict in Africa, p. 79. 49. Walter, “Designing Transitions from Civil War,” pp. 46–47. 50. Daniel C. Esty, Jack A. Goldstone, Ted Robert Gurr, Pamela T. Surko, and Alan N. Unger, Working Papers: State Failure Task Force Report (Washington, D.C.: CIA’s Directorate of Intelligence, 1995), p. 14. 51. On the South African experience with informal institutions of power sharing, see Timothy D. Sisk and Christoph Stefes, “When the Grass Turns Green: Negotiating Democracy in South Africa,” paper presented at the IGCC Conference on Powersharing and Peacemaking. 52. Shaheen Mozaffar, “Electoral Systems and Conflict Management in Africa: A Twenty-eight State Comparison,” in Timothy D. Sisk and Andrew Reynolds, eds., Elections and Conflict Management in Africa (Washington, D.C.: U. S. Institute of Peace, 1998), p. 90. The abbreviation LR signifies “largest remainder.” 53. Andrew Reynolds, “Elections in Southern Africa,” in Sisk and Reynolds, Elections and Conflict Management in Africa, p. 72. 54. Joel D. Barkan, “Elections in Agrarian Societies,” Journal of Democracy 6, no. 4 (October 1995): 109. 55. On Karl Renner and Otto Bauer’s writings regarding the possible uses of the personality principle in the Austro-Hungarian Empire, see Lapidoth, Autonomy, pp. 38–39. See also Arend Lijphart, “The Puzzle of Indian Democracy: A Consociational Interpretation,” American Political Science Review 90, no. 2 (June 1996): 261. 56. Donald Rothchild, “On Implementing Africa’s Peace Accords: From Defection to Cooperation,” Africa Today 42, nos. 1–2 (1995): 16.

S-6

11/24/09

4:19 PM

Page 139

PART 2 Evaluating Implementation Tasks

S-6

11/24/09

4:19 PM

Page 140

S-6

11/24/09

4:19 PM

Page 141

6 Disarmament and Demobilization JOANNA SPEAR

Effective disarmament and demobilization of warring factions contribute to the security necessary for the successful implementation of a civil war peace agreement. Lacking security on the ground, there cannot be deep implementation of such an agreement; for example, the holding of elections, reempowerment of civil society, establishment of political accountability, and redevelopment of the economy. However, not all peace settlements involve disarmament (the Yemen Arab Republic, the operation in Haiti), and some are merely armistices (Angola), while others may actually involve weapons buildups (the Dayton Accords in Bosnia). Although the tasks of disarmament and demobilization are often bundled, demobilization plays a far greater role in ensuring successful peace implementation. This chapter puts forward two arguments. First, the form and extent of disarmament in a peace agreement needs to be decided with reference to the norms of the society under consideration, rather than applying a “standard model.” There are post–civil war societies where guns play legitimate roles and therefore where complete disarmament is unfeasible and undesirable. Nevertheless, certain types of weapons—with no sociocultural justifications—must be removed from the combatants. A combination of selective disarmament and the establishment of controls over the remaining arms should be implemented. Second, implementers should place greater emphasis on effective demobilization than on disarmament. Peace requires breaking the command and control structures operating over rebel fighters—thus making it more difficult for them to return to organized armed rebellion— and reforming or integrating new state armies to act in the interest of the entire citizenry. This chapter highlights some of the key issues encountered in attempting to institute successful disarmament and demobilization programs in the 141

S-6

11/24/09

4:19 PM

Page 142

142

EVALUATING IMPLEMENTATION TASKS

aftermath of civil wars. Five factors are particularly important in determining the likelihood of successful disarmament and demobilization: the feasibility of the peace agreement and its aims; the implementation environment; the capability and resources of the international implementers; the attitudes of the warring parties; and effective verification. The chapter begins with definitions and a discussion of disarmament and demobilization.

Disarmament Two main types of disarmament can be discerned, cooperative and coercive disarmament.1 Cooperative disarmament, which is the focus of this chapter, is more commonly associated with peace settlements where there has been no clear victor in the conflict. Among the successes are El Salvador (eventually) and Mozambique. Among the failed efforts at cooperative disarmament are Angola (twice) and Cambodia. Coercive disarmament is most often practiced either by clear victors in civil wars or by intervention forces, for example, India in Sri Lanka and U.S. forces in Somalia. Nevertheless, the line between these two types of disarmament can be blurred both by peace agreements and through practice; witness the aggressive disarmament efforts of the Economic Community of West African States Cease-Fire Monitoring Group (ECOMOG) in Liberia in spring 1997. Disarmament fulfills two interactive roles in civil war termination: (1) the removal of the means by which civil wars have been prosecuted, thus preventing the reignition of conflict; and (2) the creation of a stable environment, thus strengthening confidence- and security-building among combatants. These goals are particularly difficult to achieve in the atmosphere of deep mistrust that often persists even after the signing of a peace agreement. Unless properly and equally implemented, attempts at disarmament can refuel the security dilemmas that intensified the conflict in the first place. Therefore, it is vital that there is adequate verification of the disarmament process. As Trevor Findlay notes, verification of disarmament plays three important roles in postconflict settlements. First, it is a determination of compliance or noncompliance. Second, it is a deterrent to cheating, as the parties will be aware that they will be caught. Third, it is an effort at confidence-building, enabling the parties to the agreement to demonstrate their compliance to each other.2 Often there is no explicit mention of “verification” within a disarmament mandate, but the more vague term of “monitoring” is used. It is important to discriminate here between disarmament and destruction. Disarmament may mean merely the transfer of ownership of a weapon, for example, from a militia to a newly created army. In several peace settlements, such as in Mozambique, disputes erupted over which

S-6

11/24/09

4:19 PM

Page 143

DISARMAMENT AND DEMOBILIZATION

143

activity had been mandated, disarmament or destruction. A further definitional complication is introduced by the term “disengagement,” which was used in the third UN Angola Verification Mission (UNAVEM III) in Angola.3 This involved neither disarmament nor demobilization, but the positioning of the opposing military forces beyond artillery range. The existing literature on disarmament has ignored the fact that often the process of disarmament is rendered more complex by the integral role that weaponry has come to play in the society. In many post–civil war societies there exist well-developed sociocultural roles for guns. These cultural norms often significantly predate the onset of internal conflict. In addition to the role they play in conflict, guns can be symbols of superiority and prestige; symbols of the successful transition to manhood; 4 signifiers of masculinity and machismo;5 a reliable form of currency used to obtain cows,6 land, or a wife; a means of securing a livelihood (e.g., through hunting or cattle raiding);7 and a desirable way to defend personal and family security. These sociocultural roles that guns play in some societies call for sensitive management of the light-weapons problem. Significantly, the failure to understand the sociocultural context has been identified as a key reason for the failure of the disarmament effort in Somalia.8 Before most countries descend into civil war, societal and/or governmental rules regulate gun possession and use. During civil wars, traditional values, social rules, and governmental laws (not just over gun use) are challenged and undermined. Indeed, according to Mohammed Ayoob, lightweapons proliferation among warring factions undermines not only central authority, but also the traditional foundation for order, the clan, and tribal elders.9 One of the tasks for postconflict peacebuilding is to restore the rules and values that governed gun use. The preference would be to strengthen existing tribal/clan structures, norms, and governmental laws as sources of restraint, rather than attempting to impose norms from without. Clearly, while there are some types of weapons that can be considered appropriate and legitimate, there will be others that are not, and disarmament efforts should concentrate on these. Thus, it is important to ensure that heavy weapons (e.g., tanks, armored personnel carriers, aircraft, and large-caliber guns) are taken from the former combatants. Certain forms of light weapons should also be removed (for example, grenade launchers, light rocket launchers, portable antiaircraft missiles, and land mines). Depending upon the particular society, it might also be appropriate to remove automatic and semiautomatic weapons. Knowledge of the “normal” level of armament in a society—and of the rules that governed arms use— provides a basis for making important decisions, for example, over the contentious question of whether to disarm “traditional hunters” such as the Kamajors in Sierra Leone. Disarmament strategies that account for the “normal” role of guns in a society will have a greater chance of success. Indeed, what may be required

S-6

11/24/09

4:19 PM

144

Page 144

EVALUATING IMPLEMENTATION TASKS

is a mix of selective disarmament and rigorous arms control. In some situations, all that may be possible is to institute rules about limiting the number of light weapons to one per person, creating weapons-free zones, and having rules about not carrying weapons in public and rules governing concealment of weapons. Given that light weapons can exacerbate problems of crime and insecurity in postconflict societies, these types of rules—in addition to disarmament—are important.10 For example, in the Somali region of Hoddeur, where there is a pervasive gun culture and Somali law permits ownership of weapons,11 UN officials in the early 1990s worked with clan elders and used fines and jail terms against young men wielding guns. This appeared to have at least temporarily “effectively deterred gunplay.”12 Any peace settlement that allows for the retention of arms by groups is open to charges of leaving in place the means for future conflict. However, given the global availability of light weapons and the porousness of most state borders, guns are so easy to obtain that the same charge also could be leveled at settlements that involve disarmament of the parties to the conflict without gaining explicit regional agreement for an embargo on arms transfers into the new postconflict state. This highlights the importance of implementing effective regional and international arms embargoes. In many societies that have seen extensive conflict, a key problem is land mines. The longevity of land mines makes them a menace to people and livestock and affects the ability of the society to utilize roads, arable land, and sources of water. As such, they are a tool of economic warfare that, unless eliminated, will hinder redevelopment efforts for decades. 13 Fortunately, there has been a learning curve concerning de-mining. In the de-mining operations in Afghanistan and Cambodia the mines were removed from the ground and retained, which made them vulnerable to loss or theft. By contrast, in de-mining operations in Mozambique the mines were detonated to avoid these problems.14 In some postconflict societies the threats posed by unexploded ordnance (UXO) are equally significant. For example, in Cambodia the problem of UXO is considerable and means that rather than merely focusing on de-mining (which suddenly looks relatively straightforward), programs for a full range of explosive ordnance disposal are required.15 A peace settlement that does not include effective disarmament or arms control can result in the creation of a new political economy of light weapons. This can affect both the state and the region through the smuggling of guns and the “militarization of crime.”16 Evidence of this occurring comes from Mozambique, where the disarmament effort, defined by some as “successful,”17 netted 190,000 light weapons of an estimated 6 million in the state.18 In the aftermath of the peace agreements in Mozambique and Namibia, surplus weapons outflowed to other states in the region.19 For South Africa this flow of arms has fueled crime, economic conflicts, and tribal disputes in the townships.20

S-6

11/24/09

4:19 PM

Page 145

DISARMAMENT AND DEMOBILIZATION

145

Demobilization Mats Berdal defines demobilization as “the formal disbanding of military formations and, at the individual level, as the process of releasing combatants from a mobilized state.”21 This process may involve the complete disbanding of military units, or their reduction in size.22 For success in the medium to long term, demobilization should include efforts to assist former combatants to reintegrate into society. Reintegration is the most effective way to break former combatants’ ties to their former fighting units and allows a means for them to provide for their dependents. The more modern the economy of the state, the more complex will be the reintegration process,23 and often the greater will be the gap between the skills that former combatants have and the skills in demand in the labor market. In many of the wars examined in this book, the combatants have known nothing other than fighting in their adult lives. Often they are at best semiliterate and have no marketable skills or experience of seeking work. Theoretically it should be easier to reintegrate former combatants into agricultural societies. However, there are problems here too, as the longevity of many of these wars weakens the ties of former combatants to their home areas, thus reducing both their access to land and their interest in farming. Just as societal norms affect the likely efficacy of disarmament measures, they also pose challenges for demobilization. Membership in a fighting force confers a certain status in society; losing that status in a peace agreement may be profoundly threatening to combatants. If one’s social standing depends on the role as a defender of the society, symbolized by the possession of a gun, then giving up that gun implies marginalization. In situations such as Northern Ireland, this has led terrorists on both sides of the sectarian divide to seek to perpetuate their roles in society through “punishment beatings.”24 Although ostensibly a reflection of the community policing itself and dealing with joyriders and drug dealers, this can be seen as fighters seeking to establish for themselves a position of status in the postconflict society. This suggests that any reintegration of these former combatants into society will have to provide them with a role of equivalent status to the one they had during the conflict, or at least positively recognize the role they played in the conflict. That such recognition may be provocative or insulting to the survivors of violence and those close to the victims of violence poses a tough ethical dilemma for peace implementers. Some groups of combatants in civil wars will have specific needs. For example, child soldiers face particular problems.25 Women combatants also face specific problems in reintegrating into society. For example, in Zimbabwe, “because of society’s perception of the place of women, former female combatants were effectively stigmatized more than their male counterparts after the liberation struggle.”26 There also tend to be high numbers of disabled former combatants who often require medical and other forms

S-6

11/24/09

4:19 PM

146

Page 146

EVALUATING IMPLEMENTATION TASKS

of special assistance. To date this has rarely been adequate. For example, in Uganda no special provision was made for the disabled, which caused intense resentment not just among the disabled themselves, but also among other former soldiers who felt that their colleagues’ sacrifices went unrecognized.27 A major form of disability found among former combatants is psychological, including trauma and mental illness. Problems that have been identified among former combatants include post-traumatic stress disorder, stresses created by “returning hero syndrome”’ (where the family of a returnee expects that person to now improve the family fortunes), and depression and alienation after years away from their communities.28 As John Prendergast and Emily Plumb argue in Chapter 12, few postsettlement peace processes have the resources to meet the psychological needs of either former combatants or other members of the society who are suffering from mental illness. This can create future problems if the material expectations of the demobilized are not met. Of particular concern are the security repercussions of a failed demobilization program. Paul Collier suggests that the process of demobilization can present security challenges at two levels. First, common citizens may suffer from “microinsecurity,” where individuals fear that they will be the victim of crimes perpetrated by former soldiers.29 Such fears are increased if demobilized soldiers retain their arms. Interestingly, Collier’s work on Uganda and Ethiopia led him to conclude that demobilization can actually lead to a decrease in crime if the newly demobilized soldiers have been given access to land. However, if former soldiers lack access to land, there can be a significant increase in the amount of crime in an area.30 The second level is “macroinsecurity,” which is the fear that the state might be overthrown by insurrection. Collier’s work suggests that there can be a link between demobilization and insurrection.31 This implies that it is vital for the sustainability of the new political order in the postconflict society that former combatants are effectively reintegrated into the society. The technical aspects of the demobilization must address the fears of combatants that there may be a reignition of conflict. As Berdal notes: “A key concern for guerrilla forces during an uncertain and fragile peace process will often be to ensure that their ability to resume, or credibly threaten to resume, fighting is not lost too early in the process. Consequently, locating assembly points in heavily mine-infested areas (often close, as it turns out later, to the guerrillas’ own undeclared arms caches) is sometimes a pre-condition for cooperation.”32 During demobilization there is the potential for a security dilemma to operate, but the strategic positioning of assembly points for demobilization, for example in remote, isolated areas, can reduce the fears of betrayal by an opponent, albeit by creating logistical difficulties for implementation agents. More cynically, however, positioning quartering sites to allow armies a fail-safe, self-help security option might provide them with a strategic position from which to relaunch

S-6

11/24/09

4:19 PM

Page 147

DISARMAMENT AND DEMOBILIZATION

147

the war after a period when the donor community has bankrolled their fighters. In Angola the positioning of quartering sites was a matter of strategic import. There was recognition by both sides that because the sites had clean water supplies, they would become villages, rather than being temporary entities, and would be in the sway of one faction or the other. Fierce disputes in the peace negotiations therefore took place over the location of the quartering sites. The provision of security to fighters and civilians alike during the process of disarmament and demobilization is absolutely crucial. Without a sense of security, fighters and noncombatants alike will not want to part with their weapons. Only when security has been achieved can effective disarmament and demobilization be achieved. This is where outside monitoring of the process, transparency, and independent verification can play crucial roles in preventing a backslide into a destructive security dilemma. But in addition to peacekeepers, it is vital to have in place structures (police, gendarmerie, national guard, and customs), to ensure civilian security. When the peace agreement mandates the creation of a new, single armed force, implementers should aim to achieve genuine integration within this force, rather than just the merger of units.33 In order to avoid the types of tensions that precipitated conflict, there should be political, ethnic, and religious integration of military forces to avoid the creation of definable units. Implementers should strive to create a professional military force inculcated with an inclusive sense of civic nationalism. Introducing a professional ethos is particularly important where the military has previously been a source of corruption, human rights abuses, and authoritarianism. The inclusion of the African National Congress’s Umhkonto we Sizwe (Spear of the Nation, known as MK) in the South African National Defense Force (SANDF) illustrates the challenges to successful integration. The South African government was assisted in this process by a British Military Advisory Training Team (BMATT) who, in practice, “served to certify to all the parties to the integration process that the procedures and standards applied in any individual case and to the process as a whole, were fair. To achieve this BMATT’s role had to be extended much further, since they also had to evaluate SANDF training curricula and selection standards in relation to international standards.” 34 Despite such assistance, the South Africans encountered unexpected pitfalls. In November 1994, MK fighters left their bases and traveled to Pretoria to complain to President Nelson Mandela about their comparatively poor treatment. Compared to regular government forces they were paid less and denied the ranks that they had held while fighting for the African National Congress. Only President Mandela’s promise to look into the issues persuaded these fighters to return to their bases. Moreover, the process did not cater well to the needs of those

S-6

11/24/09

4:19 PM

148

Page 148

EVALUATING IMPLEMENTATION TASKS

who were refused positions in the new force, or to those who would not join the SANDF but refused to demobilize. This latter group were left unoccupied in assembly areas for eighteen months, when the time could have been used for training and education.35 In South Africa, where there has been demobilization of both state armed forces and several armed opposition groups, part of the demobilization has involved hardened fighters retaining their “profession” by joining mercenary companies such as Executive Outcomes, and Sandline. Initially the majority of recruits to Executive Outcomes came from the 32 Battalion, a South African counterinsurgency unit notorious for human rights abuses and torture.36 Although this may have solved an internal problem—what to do with battle-hardened veterans—it merely shunted the problem elsewhere. After the activities of Executive Outcomes embarrassed the South African government, its Parliament passed antimercenary legislation.37 The role of implementing agents in ensuring fair play in the creation of new armed forces can help to build confidence in the peace process. The exclusion of implementing agents can have the opposite effect. For example, the Abuja II Accord mandated ECOMOG to restructure the armed forces of Liberia. However, Charles Taylor prevented this and asserted his right as the duly elected head of state to create his own military. By filling it with his own fighters and systematically purging other groups from the army—particularly the Krahn—Taylor hijacked the demobilization effort.38

Key Elements of the Implementation Process In this section I discuss various implementation challenges encountered in attempting to disarm and demobilize warring factions in the aftermath of civil war. Five key elements of the implementation process provide the analytical framework: the peace accord itself, the implementation environment, the implementers, the warring parties, and verification. The Peace Agreement A crucial aspect of any peace settlement concerns the degree of detail provided to guide implementation. Many peace agreements contain “silences” on key issues. Although such silences may be a means to avoid derailment of negotiations, they also may result from negotiators not appreciating what is involved in disarmament and demobilization. In either case, key terms such as “disarmament” and “demobilization” may not be defined. Thus in the case of Mozambique: “within the ONUMOZ [UN Operation in Mozambique] mandate, guidelines as to what constituted disarmament were absent and how to achieve this end was not explained. From the word-

S-6

11/24/09

4:19 PM

Page 149

DISARMAMENT AND DEMOBILIZATION

149

ing of the mandate, it would seem that the disarmament process was almost entirely subsumed under the definition of demobilization.”39 One of the consequences of these silences—regardless of why they were created—is that the onus is on the implementers to decide how to proceed. The details of the peace agreement can be regarded as a ceiling for action or a floor from which agents can move forward to implement their interpretation of the intent of the accords. In the case of Mozambique, many ONUMOZ personnel regarded disarmament as a central feature of the agreement, and acted accordingly.40 The chapter appendix (page 159) lays out the relationship between disarmament and demobilization, peace accords, and the holding of elections in this volume’s case studies. A significant issue for effective implementation of disarmament and demobilization programs concerns the data on which reductions in weapons and personnel are to be based. Given the irregular nature of many of the forces that have participated in the civil war, a number of problems arise. The first concerns numbers of fighters. Leaders may genuinely not know the strength of their forces, either because of the phenomenon of “ghost soldiers” (where a commander will have claimed a larger force than he actually heads in order to gain the wages of the ghost soldiers) or because of the use of large numbers of “casual fighters,” as for example in Liberia.41 In the latter situation, people may just melt away rather than be demobilized, thus creating a problem of verification. The second problem concerns data on weaponry. In conflicts where the factions have weak chains of command, there may be no inventory of weapons available, or it may be hopelessly out of date. The best data on armaments and troop numbers are developed by the parties to the negotiation in conjunction with the external parties who are to be charged with verification.42 However, often such figures become part of the presettlement “gamesmanship” between the negotiating parties, with each side inflating its troop and weapon numbers while disputing those of opponents in an effort to secure concessions. For example, in Angola in the negotiations over demobilization after the signing of the Lusaka Protocol in 1994, the government claimed it had 200,000 troops and the Union for the Total Independence of Angola (UNITA) claimed 50,000 fighters. By contrast, the UN later estimated the real figures to be 113,700 and 37,300 respectively.43 Nevertheless, in a sense this does not matter if all the parties to the peace agreement are prepared to accept the figures. In such a situation there is an element of collusion between erstwhile enemies. During the negotiation of a peace settlement, there is a potential clash of organizational interests over timetables for disarmament and demobilization. Political expediency and the desire to cut costs can lead to unrealistic timetables being set. As Nicole Ball notes, “for political-military reasons . . . postconflict demobilizations, particularly those governed by negotiated settlements, are often required to be completed within the space of one

S-6

11/24/09

4:19 PM

150

Page 150

EVALUATING IMPLEMENTATION TASKS

year, and the discharge of troops may be closely linked to compliance with other provisions of the peace accords.”44 Such a sudden discharge, although politically expedient, may strain the resources of the state, police, peacekeeping forces, and NGOs. Ball considers a period of three to four years more feasible for demobilizing a significant number of combatants and reintegrating them into society. If resources are only provided by the international community for the timetabled period, the new national authority may find itself incurring significant financial expenditure to finish a task that sponsors of the peace process consider complete. The failure to meet an unrealistic timetable may have a negative effect on the peace settlement as a whole. Moreover, if demobilization is not achieved in a timely fashion, combatants may choose to self-demobilize.45 Although this may solve a problem in the short term, it may complicate the important process of effective reintegration into society. Issues of landownership and land distribution often precipitate civil war. If the peace settlement has failed to address this issue (as many do), the problem can recur in negotiations with former combatants. As Berdal has noted, “In parts of Central America, problems of demobilization and reintegration have been directly linked to the issue of land redistribution and the terms of land ownership intended to benefit ex-combatants.”46 In the case of El Salvador, the promised redistribution of land to former combatants was an important element of the peace settlement.47 Even in African and Central American countries whose domobilization efforts have been successful, there are many former combatants who have neither found employment nor been given land to farm, as they had been promised. These groups can become politically problematic. The issue of distributing land and economic resources to former combatants is still haunting Zimbabwe in 2002, twenty-two years after the achievement of independence. In El Salvador a veterans group called the Association of Demobilized Army Troops has occasionally organized protests about the conditions in which former soldiers live. In January 1995 the group seized the Salvadoran legislature, demanding the demobilization payments, land, education, and other programs aimed at helping them reintegrate into society that were promised in the 1992 peace accords.48 Up until this point the demobilization of forces in El Salvador, which cut the army in half and disbanded the paramilitary police and civil defense patrols (composed of 6,000 men), was considered a success. The Implementation Environment Analysis of the implementation environment must include several levels: individual (the psychological impact that the civil war has had upon both combatants and noncombatants), local, national, regional, and international. In many ways these levels are interconnected, as a failure at one level

S-6

11/24/09

4:19 PM

Page 151

DISARMAMENT AND DEMOBILIZATION

151

can impact the pursuit of implementation at another level. For example, a reignition of conflict at the local level might result in withdrawal of support for a settlement by regional sponsors and international donors. Consideration of the local environment includes the role that guns have traditionally played in the society. This can vary from locality to locality within a state, for example, between agricultural societies, pastoral societies, and societies based upon hunting. In the latter two societies, gun ownership and use will likely be the norm. The local environment needs to inform the type and mix of arms control and disarmament initiatives that are attempted. For example, aiming for total disarmament of a society that subsists by hunting is doomed to failure. A major factor undermining attempts to disarm and demobilize in the aftermath of civil war is the instability of the national implementation environment. Two characteristics of the implementation environment within the state are particularly important. First, in a post–civil war situation where a settlement has been negotiated between two parties where there was no clear winner in the war, power is likely to be very dispersed and the society is likely to be unstable. This has implications at lower levels. For example, this may mean in practice that the sides to the peace agreement may not have a clear hold over the forces they ostensibly lead; that chains of command are weak; and that there may be renegade groups unwilling to obey even the leaders of their own faction. In such situations there are also problems of differentiating between combatants and noncombatants (exacerbated by the existence of “casual fighters,” who may appear and disappear unpredictably), as military discipline is often very poor and fighters may not wear uniforms. These practical problems can complicate the implementation process. The second important characteristic of the national implementation environment is that the state is inevitably weak, possibly nonexistent in some areas. In that situation, alternative sources of authority that have come to play statelike roles (e.g., protection of citizens and provision of welfare), such as warlords, local militias, tribes, clans, and so forth, may be unwilling to renounce these roles. Moreover, as Ball observes, the state institutions that do exist are “weak and ill-suited to the needs of participatory political systems.”49 In post–civil war situations, with state capacity very low, often the clearest power structure in the country is the one that has been imported by the United Nations or the regional powers overseeing the peace process. This environment presents a dilemma for the external implementers. On the one hand, as Michael Doyle argues in Chapter 3, in the name of efficiency it may be necessary for the implementers to take over many of the functions of the state and play a pivotal role in the country. On the other hand, such actions may further undermine the (already weak) capacity of the state, when one of the eventual aims of the peace process is capacity-building.

S-6

11/24/09

4:19 PM

152

Page 152

EVALUATING IMPLEMENTATION TASKS

As Berdal points out, the nature of the conflict (and the environment it created) will have a direct bearing on disarmament and demobilization efforts in two ways. First, where violence has been extreme and the conflict protracted, confidence and mutual trust will be more difficult to build between the erstwhile enemies. Second, the task of creating new armed forces and police forces will be rendered more difficult if the armed forces to be reformed were in the past instruments of state violence and terror. Moreover, the longer the conflict has lasted, the more profound the effects on economy and infrastructure, and thus the greater the logistical and technical challenges attached to disarmament and demobilization.50 The issue of economic resources is crucial, as effective demobilization and reintegration is an expensive business. Demobilization needs to be sustained over a number of years, and lack of economic resources in the latter stages can undermine the good work done at the outset of the peace agreement; witness the continuing problems in Zimbabwe and El Salvador. Berdal stresses the need to be sensitive to the ways in which the civil war has altered the political economy of the society, and the need for disarmament and demobilization efforts to be attuned to the needs and interests of different groups in the society.51 However, this can lead to better treatment being given to potentially troublesome groups. One of the more contested aspects of the demobilization program in Angola (UNAVEM II) was the fact that the former UNITA combatants and their families were popularly seen as being treated better than government supporters and their families.52 In Nicaragua following the peace settlement, the Chamorro government ensured that the Contra guerrillas were reintegrated into civilian life. Well-funded Contra-controlled “development zones” were created. By contrast, the soldiers dismissed from the army of the former Sandinista government were provided with little. This caused great resentment and some former soldiers turned to banditry. Despite this, the re-creation of the Nicaraguan armed forces is an overall success.53 The tasks of disarmament and demobilization are complicated when there has been a privatization of the means of violence, for example, because of the virtual independence of military forces from political authority (El Salvador, Guatemala), because of the (overt or covert) presence of foreign troops in the country (Cambodia, Liberia), the existence of militia groups (Rwanda, Haiti, Somalia, Congo-Brazzaville), armed militias effectively holding refugees as hostage populations (Cambodia, Rwanda), private security firms, or the importation of mercenary forces into the state (Sierra Leone, Angola). 54 The existence of these “nonstate military formations” will affect both the planning and implementation of disarmament and demobilization.55 The regional implementation environment is important for the implementation of a disarmament program. If an internal party to the peace

S-6

11/24/09

4:19 PM

Page 153

DISARMAMENT AND DEMOBILIZATION

153

process knows that it will never be left without access to arms, this may make it less likely to compromise to make a peace settlement work. Consequently, ensuring that parties to the peace settlement cannot get easy access to new arms can be crucial to making a settlement work. As noted above, the international market for light weapons is very permissive, and without the support of regional and international players for an embargo on light-weapons supplies into the state undertaking the peace settlement, disarmament efforts may be undermined if dissatisfied groups decide to return to the gun. Given the fact that the light-weapons market is beyond national control, support for the disarmament process might involve neighboring states having to improve customs and even undertake border patrols to prevent light weapons leeching across borders. Regional cooperation in dealing with arms cached in neighboring states may also be required. The international implementation environment is also important to the implementation of a disarmament program, as many of the initial sources of light weapons lie outside the region where the civil war took place. The failure of the international community to adopt and abide by an arms embargo may undermine a peace process. A salutary example of this is the current situation in Angola, where despite UNITA having been under an arms embargo since September 1993, the rebel forces have experienced little difficulty in rearming for renewed war. Indeed, the weapons that began to appear in the UNITA arsenal from mid-1998 onward indicate that they have been able to spend more money and obtain weapons of much greater sophistication than they previously had. Weapons are thought to have been purchased using revenue from diamond smuggling (also subject to international sanctions) and to have come from Bulgaria, other East European countries, Ukraine, and possibly China.56 Thus the unfavorable international environment undermined the already fraught implementation process. In terms of which levels of the implementation environment matter most, it is obvious that if there is no will for peace at the local and national levels, disarmament and demobilization are unlikely to be implemented. Nevertheless, the attitudes of actors at the regional and international levels can have important effects on the will for peace. To illustrate, the civil wars in El Salvador, Nicaragua, Namibia, and Mozambique were in large part sustained by the superpowers as part of the Cold War. With the end of the Cold War, the superpowers were no longer willing to support factions in the civil wars, and even began to press the parties to settle their differences. This had an important influence on the attitudes to peace of the parties on the ground, and peace settlements were eventually forthcoming. Actors at the regional and international levels of the implementation environment can have other influences, for example, empowering or disempowering a new government through the granting or withdrawal of aid and/or recognition, supporting rebel factions, or allowing arms transfers across their territories.

S-6

11/24/09

4:19 PM

Page 154

154

EVALUATING IMPLEMENTATION TASKS

The Implementers Behind the implementers acting in the field is a wide array of groups whose cooperation is required in order to facilitate effective implementation. Among the wider implementation coalition are groups whose cooperation will matter to varying degrees: various government departments including but not limited to the military, the private sector, nongovernmental organizations, international organizations, and state and nonstate donor groups. Each of them has a different stake in the implementation process. Together the agents and the coalition form an implementation chain that needs to endure if the process of disarming and demobilizing is to succeed. As Bruce Jones argues in Chapter 4, ideally there should be continuity of personnel between those involved in brokering the peace agreement and those involved in implementing it. This would provide for maximum congruity between the two processes and give the implementing agents a greater stake in a successful outcome. Among the relevant issues here are the resources available to implementing agents, and the support (political, physical, and logistical) they receive from the wider implementation coalition. Resource issues matter at a very basic level. For example, in Angola (1992–1993), attempts at disarmament were hindered by the small size of the UN peacekeeping force. With only 298 peacekeepers in the field, it was not possible to institute cooperative disarmament programs, let alone more aggressive disarmament efforts.57 Often the early phases of demobilization are handled by military forces, before civil authorities take over the task of reintegration into society. Disarmament is also often a military-led process, be it by national militaries or international peacekeeping forces. Over time, however, newly constituted police forces may come to play a role in ongoing disarmament efforts. Many of the implementation agents have little experience in disarming and demobilizing warring factions. It is clear that much of the training takes place on the ground, with the consequence that there is little learning across peace operations. Another aspect of training concerns verification. Many of the implementing agents have received no formal training in verifying disarmament, yet this is one of the most important roles they will have to play. As disarmament is the potential source of a security dilemma, implementing agents must be seen to be impartial, and disarmament and demobilization must be both transparent and verified. Where there is local or national concern over the trustworthiness of the military forces charged with disarmament, NGOs may be able to make a positive contribution to the process. There is evidence from Mozambique that NGOs played an important role in the disarmament process as people came to recognize that weapons handed to them would be destroyed (whereas it was feared that weapons given to the police would be merely

S-6

11/24/09

4:19 PM

Page 155

DISARMAMENT AND DEMOBILIZATION

155

resold for profit). Consequently, local communities placed land mines, light weapons, and ammunition by the roadside if they knew humanitarian organizations or mine clearance groups were in the vicinity. 58 Moreover, “in some cases, NGOs have been governments’ and donors’ most valuable partners in implementing demobilization/reintegration programs. For example, NGOs have organized transport to camps (Namibia), distributed food, medical care, and materials (Namibia), and provided training and job counseling (Zimbabwe). NGOs can compensate in part for weak public sector institutions.” 59 In these situations, the expertise, resources, and organization that the NGOs can draw on are an important means to “fill the gaps” in a weak state. An interesting issue here concerns the potential for international involvement in peace implementation to have a negative effect on the newly created government or reconstituted state. The fact that various international organizations and NGOs play increasingly active roles in implementing peace agreements has been passionately criticized by Joseph Hanlon as undermining government competence, autonomy, and institutionbuilding. 60 It seems that there has been some acknowledgment of this potential problem: “Donor recognition—in word at least—that the process of demobilization and reintegration must be owned by the implementing government (despite calls for more donor influence over, and involvement in, planning and design) is a positive sign.”61 The Warring Parties The degree of involvement of the parties to the conflict in implementing disarmament and demobilization varies across peace agreements. In the Angola Accords of 1991, the primary responsibility for disarmament and demobilization was to rest with the parties to the conflict, with UNAVEM II relegated to observing the parties as they monitored implementation.62 In many peace agreements, however, the implementers play a more prominent role in the process. Nevertheless, it is very useful to have representatives of the various factions also involved in implementing disarmament and demobilization and involved in the process of verifying that their side has fulfilled its obligations. It is important that all the major actors—warring parties and implementers—feel that they have a stake in the success of the implementation process. The attitude of the implementation targets toward the peace process has a significant effect on the chances of disarmament and demobilization being implemented effectively. If the warring parties are reasonably satisfied with the political terms of the peace process and think that they can live with the military balance of power being created, the chances of successful implementation are far greater. If there is a will for peace, problems

S-6

11/24/09

4:19 PM

156

Page 156

EVALUATING IMPLEMENTATION TASKS

at other points in the implementation process, for example, resource constraints and minor instances of cheating, can be overcome. Verification A crucial aspect of any disarmament and demobilization is verification of the process so that implementation agents can assure the targets that what was supposed to happen, has indeed happened. The implementing agents— in their role as impartial actors—must evaluate the extent to which the parties to the peace agreement have complied with both the letter and the spirit of the peace settlement in terms of disarmament and demobilization. This raises the question of cheating. It may be that a degree of cheating (through the storing of arms in secret caches, etc.) by all sides is actually quite stabilizing—at least in the short term—as they know that they are not completely vulnerable if the settlement falls apart. In this sense, cheating may act as a unilateral confidence-building measure. As Terrence Lyons notes: “Nearly all parties hedge and keep some portion of their fighters and weapons outside the demobilization process. The fact that disarmament is inherently imperfect need not prevent implementation if the political dimensions of the peace process are adequate.”63 Cheating may arise, however, from less benign motives. It may indicate a problem of command and control, that there are renegade factions who reject the compromises of the leadership and refuse to disarm. Parties to peace may cheat because it provides them with tactical advantages in the peace process; they want the process to go forward, but they know that by cheating, especially by holding back soldiers for intimidation purposes in elections, they may gain an advantage over their rivals. Or as Stephen Stedman contends, there are parties, total spoilers, who cheat precisely to gain a military advantage over their opponent in the hopes of defeating them through surprise.64 Where the political dimensions of a peace process are weak, cheating is both emblematic of the weakness of and a liability to the peace process, as the discovery of such activity could result in the redevelopment of a security dilemma and a reignition of the civil war. This helps to explain the different consequences of cheating in three of the different peace processes under discussion. In Mozambique the consequences of cheating were contained by a fairly robust political process; in El Salvador the political process tottered but survived; and in Angola the situation was lost due to the weak commitment to the political process. In Mozambique, it was strongly suspected at the time—and confirmed retrospectively—that there was widespread cheating, but it did not derail the peace process.65 UN special representative Aldo Ajello said of the problem: “I know very well that they will give us old and obsolete material, and they will have here and there some hidden. I don’t care. What I do is create

S-6

11/24/09

4:19 PM

Page 157

DISARMAMENT AND DEMOBILIZATION

157

the political situation in which the use of guns is not the question. So they stay where they are.”66 In El Salvador, on the other hand, cheating threatened to derail the peace process. In November 1992 the rebel Farabundo Martí National Liberation Front (FMLN) officially disarmed, handing over significant quantities of weaponry. However, in May 1993 an accidental explosion in an FMLN arms cache in Nicaragua revealed that the FMLN had hidden an enormous amount of arms and ammunition from the UN Observer Mission in El Salvador (ONUSAL), which had responsibility for verifying the postsettlement disarmament and demobilization. In five “safe-houses” in Managua, ONUSAL and Nicaraguan officials found some 1,240 rifles, 2,000 kilograms of explosives, 1.4 million bullets, 1,300 mortars, 3,900 grenades, 350 rockets, and 19 surface-to-air missiles. The embarrassed FMLN then admitted the existence of a further 120 arms caches containing tons of matériel.67 In El Salvador the law banned any armed group from registering as a legal political party. The discovery of the caches led to talk of denying the FMLN the right to participate in the crucial March 1994 elections. ONUSAL, citing the FMLN’s insistence that it had no plans to return to war and the fact that no FMLN units were actually armed (and given the distance of the caches from FMLN forces, suggesting no plans for surprise attack), argued that denying the FMLN participation in the elections would jeopardize the entire peace process. In the end, the FMLN’s legal right to participate was not challenged and the peace process continued. However, the FMLN’s leverage was weakened, undermining its ability to challenge government violations of the accords regarding the military and the police (most important, the appointment of a former military officer to run the police, in direct contradiction of the peace accords). In addition to the FMLN having lost the “moral high ground” and nearly the whole peace process, its image was tarnished in the important preelectoral period, and potential FMLN electoral candidates who had been associated with the concealment of the caches were considered too damaged to run in the elections. A clear example of cheating being an important signal of lack of commitment to peace and therefore the death of a peace process is the case of Angola and the attempt to implement the Bicesse Accords. The United Nations possessed inadequate resources to monitor the demobilization, and therefore initial instances of cheating were obscured by logistical and technical problems. But it soon became clear that there was widespread cheating, particularly by UNITA.68 This meant that when UNITA returned to war it was in a comparatively stronger position due to the uneven demobilization.69 The consequences of incidents of cheating will in part depend on the terms of the peace agreement and the scope that is granted to implementers

S-6

11/24/09

4:19 PM

158

Page 158

EVALUATING IMPLEMENTATION TASKS

to act against cheaters. If it is possible to crack down hard on early instances of cheating, this may deter more serious violations of the peace accord later. To improve detection of cheating, however, peace implementation missions need better intelligence and analysis, especially into the motivations behind instances of cheating. Given the allergy of the United Nations to intelligence gathering and analysis, UN-led missions are crippled from the start in attempts to robustly confront cheating. Lacking intelligence and enforcement capabilities, most of the implementers studied in this book have been hesitant to acknowledge cheating. In addition to their role as verifiers, implementers often find themselves continuing to mediate the terms of the agreement and therefore rigidly protect their reputations for neutrality and impartiality. The fear of putting their impartiality at risk can lead implementers to be less than vigilant in their approach to monitoring and verifying. The problem is compounded if the agreement is structured in such a way that there are no methods to deal with cheating short of abandoning the whole agreement; then the zero-sum nature of the situation will create big incentives for the implementers to ignore instances of cheating. Agents may thus ignore or challenge the veracity of information indicating that disarmament and demobilization are failing. For example, the biggest failure of UNAVEM III in Angola was its official validation that UNITA had fully complied with demobilization, even though it maintained over 25,000 of its best troops.70 The sunk costs of the implementing agents, who have invested much (physically and psychologically), may make them unwilling to recognize that the peace process—rather than proceeding from strength to strength—is being fundamentally undermined by cheating. Verification inevitably involves making value judgments, and there is the danger that in objectively assessing the compliance of the parties, the implementer may no longer be seen as impartial. However, such judgments have to be made. Trevor Findlay suggests that in order to minimize potential controversy over verification, it is important to have in place an accepted procedure for solving disputes prior to any arising.71 If such a procedure has been endorsed by the parties prior to the verification, the impartiality of the implementer is more likely to emerge intact, and assessments will be more difficult to dispute. It is also imperative for the implementers to be transparent about their assessments, and to provide evidence to support them. Arbitration procedures, perhaps involving part of the wider implementation coalition (e.g., traditional allies, sources of aid, etc.), may provide means by which pressure can be brought to bear on recalcitrant parties.

Conclusion This chapter has laid out five key aspects of disarmament and demobilization in the aftermath of civil wars: the viability and feasibility of the peace

S-6

11/24/09

4:19 PM

Page 159

DISARMAMENT AND DEMOBILIZATION

159

process and its aims, the implementation environment, the implementers, the warring parties, and verification. Together these provide an analytical framework through which a comparative analysis of attempts at disarmament and demobilization can be undertaken. This chapter has not attempted to provide any kind of “model” implementation process. Indeed, one of its key findings is that disarmament and demobilization must be attuned to the local and national situation, considering such issues as the traditional role of guns and the sociocultural role of warriors in the societies in question. This suggests that implementers will need to acquire an in-depth knowledge of the specific environment in which they operate, in order to ensure that appropriate programs for disarmament and demobilization are developed and implemented.

Appendix 6.1: The Relationship Between Disarmament, Demobilization, and Postconflict Elections ANGOLA UN Angola Verification Mission II (UNAVEM II), 1991–1995 UN Angola Verification Mission III (UNAVEM III), 1995–1997 Forces and Weaponry UNAVEM II: Total number of combatants, 151,000; Movimento Popular de Libertação de Angola (MPLA): 113,700; Frente Nacional de Libertação de Angola (FNLA) and the União Nacional para a Independência Total de Angola (UNITA): 37,300. Lack of reliable information on the quantity and quality of weaponry held by the parties to the conflict. D&D in the Peace Accord? UNAVEM II: Yes. Bicesse Accords of 1991 included a provision imposing an arms embargo on both parties. UNAVEM II was mandated to verify the cease-fire between the government of Angola and UNITA and observe and verify elections. UNAVEM III: Mandated to monitor and verify the assembly and demobilization of UNITA and formation of new army. What Degree of D&D Agreed? UNAVEM II: Embargo on lethal weapons supplies from all governments (the “triple zero” clause). The separation, demobilization, and disarmament of all forces and the standing armies of the government and UNITA. Collection and disposal of weapons. The formation of a new Angolan Armed Force (FAA) of 50,000 before the elections. UNAVEM III: Full disarmament and demobilization of UNITA. Modalities of integration of national army left to be decided at a later date.

S-6

11/24/09

4:19 PM

160

Page 160

EVALUATING IMPLEMENTATION TASKS

Who to Implement D&D? UNAVEM II: The Joint Political-Military Commission (CCPM) was responsible for the implementation of all aspects of the Bicesse Accords. UNAVEM II to monitor the collection and storage of weapons by the parties. Parties themselves reached agreement on the regulations and procedures to be followed for the collection and storage of weapons. UN not involved, but offered technical advice on the classification and storage of weapons. NGOs planned reintegration programs, as did the EU. UNAVEM III: Same as UNAVEM II except UN SRSG appointed as chair of the Joint Commission and the UN Resident Coordinator for Humanitarian Activities responsible for support of demobilization. D&D Planned to Be Completed Prior to Election? UNAVEM II: Yes. UNAVEM III: Yes. Was D&D Implemented Prior to Election? UNAVEM II: No. Elections held on September 29–30, 1992. Only 45 percent of government troops and 24 percent of UNITA troops demobilized. Formally, by September 27, 1992, the two armies were announced disbanded. By October 7, 1992, a total of 96,620 government troops had been demobilized, representing 80 percent of the projected figure. However, a much lower proportion of ex-UNITA troops were demobilized. 1 Disarmament hardly begun. UNAVEM II could not effectively monitor compliance or protect storage sites. UNAVEM could not search for caches, nor could it control supplies of weapons flowing into the country. CCPM was unable to enforce the arms embargo and both sides continued to obtain arms during the UNAVEM II mission, though through the black market rather than traditional sources of supply. Estimates that under UNAVEM II, UNITA kept about 30,000 men and MPLA 10,000 out of the disarmament process.2 UNAVEM III: No. UNITA found to withhold 15,000–25,000 troops from demobilization. Parties return to war. Was Election Judged to Be Fair? UNAVEM II: Deemed free and fair by the United Nations. UNITA disputed the results. The remaining arms caches and large numbers of illicit weapons both caused problems for UNAVEM and meant that both sides were able to easily return to war after the elections. UNAVEM III: No election held. Was Peace Maintained After Election? UNAVEM II: No. After the elections UNITA restarted the war. Consequently in April 1993 the Angolan government announced that it

S-6

11/24/09

4:19 PM

Page 161

DISARMAMENT AND DEMOBILIZATION

161

would no longer abide by the arms embargo in the Bicesse Accords. Subsequently the observing troika, the United States, Russia, and Portugal, announced that they were lifting their bans on arms sales to the Angolan government.3 Subsequent D&D Activities. In November 1999 the Angolan government announced an amnesty for ex-rebels and in January 2001 set up a special Peace and National Reconciliation Fund to oversee the reintegration of exrebels. UNITA rejected the amnesty immediately but according to governmental sources more than 800 rebels had turned themselves in by January 2001.4 In February 2001, 600 UNITA members turned themselves in, in the province of Huambo. In early 2001 the Namibian Defense Forces seized military weapons from UNITA (mortars, AK-47 rifles, G-3 rifles, R-1 rifles, RPG mortars, antitank and antipersonnel land mines) from southern Angola, following UNITA’s extended operations in Namibia as a response to the latter’s alleged logistical support to the Angolan government. 5 150,000 soldiers demobilized, but only 50,000 were socially reintegrated via agricultural, school, and housing projects sponsored by the German Society for Technical Cooperation and the EU.6

BOSNIA-HERZEGOVINA AND CROATIA UN Transitional Authority in Eastern Slavonia, Baranja, and Western Sirmium (UNTAES), 1996–1998 Multinational forces: Implementation Force (IFOR), December 1995–December 1996; NATO-led Stabilization Force (SFOR) in BosniaHerzegovina, 1995–present. Forces and Weaponry UNTAES: Population of 150,000 of which 85 percent Serbs (including 60,000 refugees) and 8 percent Croats. The Army of the Republic of Serb Krajina (ARSK), amounting to 8,000–12,000 armed soldiers but also a number of Serb paramilitary units (including Arkan and “Tiger” militia).7 IFOR: Total number of combatants, 322,000; army of Bosnia-Herzegovina: 92,000 active duty, 100,000 reserve; Croatian defense force 55,000; army of Republika Srpska, 75,000. A variety of heavy weaponry on all sides, including battle tanks, armored combat vehicles, and artillery.8 D&D in the Peace Accord? UNTAES: 1995 General Framework Agreement for Peace in Bosnia and Herzegovina (the Dayton Agreement) and Separate Serb/Croat Agreement of November 1995 (the Erdut Agreement). IFOR/SFOR: Dayton Agreement mandated to reform and bring under sin-

S-6

11/24/09

4:19 PM

162

Page 162

EVALUATING IMPLEMENTATION TASKS

gle command the Bosnian military, currently divided into three ethnically based, rival armies (the Bosnian Muslim, Croat, and Serb). Efforts to coordinate the activities of the local armed forces and the ministries of defense of Bosnia’s two entities, the Federation of Bosnia and Herzegovina, and Republika Srpska. What Degree of D&D Agreed? UNTAES: 100 percent. UN Security Council Resolution 1037 of January 1996 mandated UNTAES to supervise the demilitarization of Eastern Slavonia and its transfer to Croatia. IFOR/SFOR: D&D planned to be completed in sixteen months (in stages).9 The Bosnian Federation forces were entitled to build up their holdings (via the U.S. “Train and Equip” program) under Art. IV to correct in their favor the asymmetrical war-ending.10 Who to Implement D&D? UNTAES: It maintained a “visible link” with SFOR/NATO, including air support. IFOR: Entitled to take all necessary measures to ensure compliance with the weapons control measures that have been agreed upon by the parties. D&D Planned to Be Completed Prior to Election? UNTAES: Yes. IFOR/SFOR: No. Was D&D Implemented Prior to Election? UNTAES: Yes. D&D was completed in accordance with the schedule. By June 21, 1996, all heavy weapons of the local Serb forces had either been removed or handed over to UNTAES for disposal; ARSK was disbanded and the paramilitary units were escorted out of the region. IFOR/SFOR: Cease-fire consolidated; limited demobilization. No militaryon-military violence, but no integration of forces. “Rebadging” undercut IFOR as armed groups, reluctant to disarm, joined local police units.11 Was Election Judged to Be Fair? UNTAES: Yes. IFOR/SFOR: Yes. Was Peace Maintained After Election? UNTAES: D&D considered successful. Coercive disarmament was used, as well as buy-back scheme to collect weapons from civilians (although this was not provided for in the UNTAES mandate). Thus 5,000 rifles, nearly 1 million rounds of ammunition, 12,000 grenades, 5,000 antitank rockets, and 6 surface-to-air missiles were netted.

S-6

11/24/09

4:19 PM

Page 163

DISARMAMENT AND DEMOBILIZATION

163

IFOR/SFOR: Cease-fire consolidated and Federation and Serb military forces progressively demobilized. Foreign forces withdrawn. No militaryon-military violence. Regional arms control agreements forged. Integration of forces weak. Subsequent D&D Activities. Croatia, February 1997: UN peacekeepers instituted a buy-back campaign that swapped cash for weapons. Netted: 100,400 rifles, 253,000 antitank launchers, 6,271 grenades. Croatia and Minus, Eastern Slovenia, August 1997: UN Transitional Administration in Eastern Slovenia instituted a buyback campaign that swapped cash for weapons. Netted: 7,963 rifles and machine guns, 1,922 antitank rocket launchers, cannons, and surface-to-air missile launchers.12

CAMBODIA, 1992–199313 UN Transitional Authority in Cambodia (UNTAC) Forces and Weaponry Cambodian People’s Armed Forces: armed forces, 131,000; militia, 220,290; weapons, 273,343; heavy weapons, 877; ammunition, 79,205,175. National Army of Democratic Kampuchea: armed forces, 27,000; weapons, 20,000; heavy weapons, 176; ammunition, 516,000. Armée Nationale pour un Kampuchea Indépendent: armed forces, 17,500; weapons, 13,500; heavy weapons, n.a.; ammunition, 742,000. Khmer People’s National Liberation Armed Forces: armed forces, 27,800; weapons, 13,600; heavy weapons, n.a.; ammunition, 266,000. D&D in the Peace Accord? Yes. However, no measures specified to deal with noncompliance. What Degree of D&D Agreed? 70 percent demobilization of the four factions’ armed forces. Mandate included search and seizure of arms caches. Who to Implement D&D? UNTAC. D&D Planned to Be Completed Prior to Election? Yes. 70 percent of cantoned forces to be demobilized by September 1992, one month prior to commencement of voter registration. Was D&D Implemented Prior to Election? No. Khmer Rouge refused to participate in cantonment and demobilization. The other factions initially complied, but operation had to be suspended. UNTAC is reported to have

S-6

11/24/09

4:19 PM

164

Page 164

EVALUATING IMPLEMENTATION TASKS

taken possession of 300,000 weapons and 80 million rounds of ammunition.14 Was Election Judged to Be Fair? Partially. Campaign marred by violence. 90 percent of registered voters participated. Election declared free and fair by international monitors. However, results of election altered by State of Cambodia brinksmanship. Was Peace Maintained After Election? Yes. Operation considered a partial success. But UNTAC was unable to implement the provisions of its mandate relating to the D&D of some 450,000 combatants.15 Subsequent D&D Activities. Weapons from the civil war are believed to be owned by 500,000, and the government has engaged in a buy-back scheme (U.S.$13 for shotguns, U.S.$10 for AK-47s, and U.S.$2 for grenades). The last soldiers of the Khmer Rouge surrendered their weapons in late 1998.

EL SALVADOR, 1991–1995 UN Observer Mission in El Salvador (ONUSAL) Forces and Weaponry El Salvador Armed Forces (ESAF): 93,000; Farabundo Martí National Liberation Front (FMLN), guerrilla umbrella organization: 6,000–7,000. D&D in the Peace Accord? Yes. What Degree of D&D Agreed? The FLMN to disarm 100 percent but at a pace of 20 percent per month.16 El Salvador’s government committed to restructure, reduce, and demobilize its armed forces. A National Civil Police was to be formed and ESAF was to be downsized to 32,000. Who to Implement D&D? Actual D&D carried out by local parties. ONUSAL was mandated to monitor the 1992 cease-fire, verify the compliance of both parties with the signed agreements, and oversee the elections. In 1992 a military division of ONUSAL was set up, mandated to monitor the troops of both parties in “peacetime” locations, verify the inventories of weapons and personnel, authorize and accompany the movements of both forces, and receive and investigate complaints of violations.17 D&D Planned to Be Completed Prior to Election? Yes. FMNL was to disarm by October 31, 1992. However, the D&D process was delayed and suspended on a number of occasions.

S-6

11/24/09

4:19 PM

Page 165

DISARMAMENT AND DEMOBILIZATION

165

Was D&D Implemented Prior to Election? Yes. By November 1992 FMLN was officially disarmed, but in May 1993, after an explosion in an FMLN hidden cache in Nicaragua, ONUSAL and the Nicaraguan authorities found some 1,240 rifles, 2,000 kg of explosives, 1.4 million bullets, 1,300 mortars, 3 900 grenades, 350 rockets, and 19 surface-to-air missiles.18 This was a serious breach of the peace accord. By August 1993 the FMLN arms caches were destroyed and ONUSAL declared FMLN disarmed and demobilized. Was Election Judged to Be Fair? Elections held in March and April 1994. In 1992 FMLN was legitimized as a political party and took part in the elections. Elections declared generally fair, despite some irregularities. Was Peace Maintained After Election? Yes, but subsequent unrest organized by the Association of Demobilized Army Troops. They seized Parliament in 1995, claiming that they had not received the promised demobilization payments, land, and education support.19 Subsequent D&D Activities. Several weapons buy-back campaigns. 1996: Guns-for-vouchers program to buy food and clothes (sponsored by the New York–based Guns for Goods organization) netted nearly 3,000 guns, including 1,043 assault rifles, 1,036 hand grenades, 16 grenade launchers, 119 antitank weapons, 43 pounds of C-4 plastic explosives, 87 pounds of TNT, scores of land mines, and 50,000 rounds of ammunition. But none of the surface-to-air missiles or the M-60 machine guns, used during the war, surfaced.20 1997: Effort sponsored by “Patriotic Movement Against Crime.” Weapons swapped for vouchers redeemable for goods and food. Over 2,000 weapons retrieved. 1998: Effort sponsored by “Patriotic Movement Against Delinquency” with assistance from the OAS. Weapons exchanged for food coupons. Total of weapons netted over three campaigns: 6,634 arms including pistols, rifles, and explosives.21

GUATEMALA UN Verification Mission in Guatemala (MINUGUA), 1994–1997 Forces and Weaponry Unidad Revolucionaria Nacional Guatemalteca (URNG): total number of combatants, 1,500-3000;22 Guatemalan Army: 45,000.23

S-6

11/24/09

4:19 PM

Page 166

166

EVALUATING IMPLEMENTATION TASKS

D&D in the Peace Accord? Yes. What Degree of D&D Agreed? Verification of designated armed forces units, limitations to military flights: UNRG to hand over all types of weapons, munitions, explosives, mines, and supplementary military equipment; UNRG supposed to end military structures. Who to Implement D&D? D&D to be verified by UN troops. D&D Planned to Be Completed Prior to Election? Prior to constitutional reform and installation of electoral reform commission. Was D&D Implemented Prior to Election? Yes. The cease-fire and demobilization of guerrilla forces went virtually without incident. Guerrilla cantonment and disarmament was completed ahead of schedule, and government forces complied with troop and budget reductions. Was Election Judged to Be Fair? Yes. Was Peace Maintained After Election? Yes.

LEBANON Syria, 1989–2002 Forces and Weaponry Lebanese Army (‘Awn): total number of combatants, 8,000 (tanks, 100 armored vehicles, 200 artillery pieces, and 200 heavy mortars); Lebanese Forces (Ja’Ja): 12,000–18,500 part-time, 12,000 militarily affiliated (1/2 ‘Awn’s weaponry);24 Palestinian Liberation Organization (PLO): 14,00018,000 (light weaponry, tanks, mortars, artillery, antiaircraft missiles, rockets); 25 Druze Popular Liberation Army: 5,500–8,500 standing, 7,000 reserves;26 South Lebanon Army (SLA): 5,000;27 Hizbollah: 600–800 full time cadres, 3,000–5,000 part time cadres (mainly light weapons);28 Syrian army: 40,000;29 Iranian army: 1,500–3,000 aiding Shiite militias in south;30 UNIFIL: 4,000. D&D in the Peace Accord? Yes. What Degree of D&D Agreed? The Ta’if Agreement stipulated the disbanding of all Lebanese and non-Lebanese militias within six months of the beginning of implementation. The agreement did not specify the particulars

S-6

11/24/09

4:19 PM

Page 167

DISARMAMENT AND DEMOBILIZATION

167

of demobilization, but vaguely provided for reintegration of the militiamen under the umbrella of the Internal Security Forces. Who to Implement D&D? The Syrian army is to assist in helping the legitimate Lebanese forces extend the state’s authority, but is supposed to withdraw thereafter. D&D Planned to Be Completed Prior to Election? Yes. Disbanding and demobilization to be completed within six months after implementation begins. Was D&D Implemented Prior to Election? No. Not all militias disbanded. Hizbollah retained all its weapons, and several militias—as well as Palestinian refugee camp enclaves—retained stockpiles of matériel. Was Election Judged to Be Fair? No. In the 1992 elections, 87 percent of the electorate (mostly Christians) boycotted the vote to protest overt Syrian manipulation of the process. Was Peace Maintained After Election? Yes. Subsequent D&D Activities. Hizbollah has still not demobilized fully, although it has transitioned to a purely political actor. Israel and the SLA continued to secure the south until 2000 when Israel withdrew and the SLA was demobilized; as of 2002, 25,000 Syrian troops have not withdrawn from Lebanon.

LIBERIA Cease-Fire Monitoring Group (ECOMOG) of the Economic Community of West African States (ECOWAS), 1990–1997 UN Observer Mission in Liberia (UNOMIL), 1993–1997 Forces and Weaponry Estimated 25,500 combatants at time of intervention. National Patriotic Front of Liberia (NPFL): 12,500; Independent National Patriotic Front of Liberia (INPFL): 6,000; Armed Forces of Liberia: 7,000.31 By 1997, estimates of combatants increased to 60,000.32 D&D in the Peace Accord? Number of agreements. Most important the ECOWAS-brokered 1996 Abuja Peace Accord. ECOMOG was to monitor the borders of Liberia and monitor the main entry points by land, sea, or air

S-6

11/24/09

4:19 PM

168

Page 168

EVALUATING IMPLEMENTATION TASKS

in order to ensure that no arms and ammunition were brought into the country; to assemble and disarm combatants of all factions; to establish checkpoints to verify the movement of arms; and to assist in the return of refugees and internally displaced persons. What Degree of D&D Agreed? 100 percent. All army factions to be disbanded. The army to be reformed and trained by ECOMOG. Who to Implement D&D? UNOMIL and ECOMOG. Following analyses that meaningful D&D was not proceeding in the southeast regions, ECOMOG was reinforced (overall strength to 11,000) and it recovered weapons forcibly while UNOMIL kept open its sites for voluntary disarmament. This policy resulted in the additional capture of 458 weapons and 27,000 pieces of ammunition, and ECOMOG seized approximately 6,000 weapons and 150 rounds of ammunition.33 D&D Planned to Be Completed Prior to Election? Yes. D&D ended in February 1997. Approximately 10,000 weapons and 1.24 million pieces of ammunition surrendered by factional fighters.34 Was D&D Implemented Prior to Election? Yes. 61 percent (20,000) of the estimated total combatants were disarmed, surrendering approximately 9,500 weapons and 1.2 million pieces of ammunition. Weapons were kept in joint custody by UNOMIL and ECOMOG. Although with delays, D&D conducted from November 22, 1996, to February 1997.35 Was Election Judged to Be Fair? Elections held on July 19, 1997, and acclaimed by international observers as peaceful, free, and fair. 30,000 weapons and more than 2 million rounds of ammunition collected and destroyed. Suspicions of hidden caches.36 Was Peace Maintained After Election? Tenuous. 1999 military reforms to build a better-trained and professional military jettisoned by elected government.

MOZAMBIQUE, 1993–1995 UN Operation in Mozambique (ONUMOZ) Forces and Weaponry 82,000 combatants prior to demobilization from both warring parties: the governing party, Frelimo (Front for the Liberation of Mozambique), and Renamo forces (Mozambique National Resistance).37 Lack of reliable

S-6

11/24/09

4:19 PM

Page 169

DISARMAMENT AND DEMOBILIZATION

169

information on the quantity and quality of weaponry held by the parties to the conflict. Information that the weapons included AK-47 rifles, pistols, grenade launchers, antiaircraft guns, mortars, and mines (UN estimates that there were about 2 million land mines, most often the Chinese M-72). Paramilitary forces projected to have 49,806 weapons. Estimates that the Frelimo government had distributed 1 million AK-47s to civilian defense units in the 1980s.38 D&D in the Peace Accord? Yes. 1992 Rome Accords, modeled after the Angola peace plan: cease-fire, D&D, and elections.39 What Degree of D&D Agreed? All combatants to be disarmed during the deployment of the peace support mission. They were then to be demobilized and either reintegrated into society or recruited into the new armed forces. Integrated army of 30,000 (volunteer). Weapons and ammunition in poor condition to be disposed of safely. Weapons collected during disarmament were to be stored and later transferred to the new armed forces. Who to Implement D&D? ONUMOZ military personnel and the parties. ONUMOZ to verify withdrawal of foreign forces, separate and canton forces, demobilize and disarm combatants, collect, store, and destroy weapons, and disband private and irregular forces. The Cease-Fire Commission, chaired by UNMOZ, developed regulations and procedures to deal with the registration, classification, and storage of weapons; the transfer of weapons to the Forças Armadas de Defensa de Moçambique (FADM; Mozambican Defense Force); and the functioning of regional arms depots. Contrary to the Cease-Fire Commission’s rules, ONUMOZ was denied permission to collect and disable weapons beyond the assembly areas. D&D Planned to Be Completed Prior to Election? Yes. Was D&D Implemented Prior to Election? Mixed. Demobilization fell behind schedule. Both sides withheld forces as “insurance,” 5,000 government troops and 2,000 Renamo troops. Disarmament completed at assembly areas. ONUMOZ collected 189,827 weapons, 43,491 from paramilitary forces. However, ONUMOZ was unable to complete the verification of weapons before the mandate expired, due to delays in the demobilization process. No active disarmament efforts beyond assembly areas. Was Election Judged to Be Fair? Yes. Was Peace Maintained After Election? Yes. Despite the fact that the population had access to large amounts of weaponry, there has been no large-

S-6

11/24/09

4:19 PM

170

Page 170

EVALUATING IMPLEMENTATION TASKS

scale rearmament. However, weapons stolen from FADM storage facilities found their way to South Africa, causing instability there. Subsequent D&D Activities. During a postdemobilization verification mission, large numbers of weapons were discovered at declared and undeclared sites.40 March 1997 weapons buy-back campaign: Sponsored by the Christian Council of Mozambique. Exchanged goods and tools for weapons. Netted: 800 guns and 24,000 armaments.41 According to a 1997 UNDP study, 29 percent of the demobilized soldiers stated they were not reintegrated, with 5 percent responding that they would never be integrated (because of war traumas).42

NAMIBIA, 1989–1990 UN Transitional Assistance Group (UNTAG) Forces and Weaponry Approximate number of combatants prior to demobilization: 43,261; South West African People’s Organization (SWAPO): 20,000; South West African Territorial Force (SWATF): 21,661; Koevet (South African): 1,600+. Problems due to lack of reliable information on the quantity and quality of weaponry held by the parties to the conflict. D&D in the Peace Accord? Yes. New York Accords of December 1988. UNTAG to supervise the cease-fire, monitor the behavior of the South West African Police (SWAPOL), and observe the election campaign.43 As a South African “colony,” Namibia was subject to the 1977 UN arms embargo against South Africa. The embargo was only lifted after independence. What Degree of D&D Agreed? 100 percent. All combatants to be demobilized—SWATF and SWAPO to be disbanded, and the South African Defense Forces (SADF) to be withdrawn by November 8, 1989.44 The local military and paramilitary forces established by South Africa. Who to Implement D&D? UNTAG forces negotiated a number of “field agreements” with the South African Forces that contained detailed schedules and procedures for the collection and storage of weapons during the disarmament process. UN and parties to conduct disarmament. D&D Planned to Be Completed Prior to Election? Yes. The weapons that were collected and stored during the UNTAG mission were transferred to the new Namibian national forces (military and police) after independence.

S-6

11/24/09

4:19 PM

Page 171

DISARMAMENT AND DEMOBILIZATION

171

Was D&D Implemented Prior to Election? Despite some early delays, the demobilization of combatants from both sides was effectively carried out before the elections. By November 22, 1989, the remaining South African troops had left Namibia. Was Election Judged to Be Fair? Yes. Was Peace Maintained After Election? Yes. Despite the fact that the population had access to large amounts of weaponry, there has been no largescale rearmament. Subsequent D&D Activities. A Tripartite Military Integration Committee, with UNTAG as its chair, was established to develop a concept for the Namibian Army. The Namibian Defense Force (NDF) recruited volunteers including 7,000 ex-SWAPO fighters in a force of 10,000. Sixteen months after independence the government initiated a rather haphazard reintegration process in the face of rising banditry and high unemployment (80 percent) among ex-combatants.45

NICARAGUA, 1989–1990 UN Observer Group in Central America (ONUCA) Forces and Weaponry Sandinista People’s Army (EPS): Following the Sandinista Revolution and the 1979 ousting of president Somoza, EPS, centered on the Sandanista National Liberation Front (FSLN) combatants, increased to 80,000.46 It was supplemented by reservist and militia, thus acquiring an overall fighting strength of 125,000. Their weaponry included Soviet/Eastern European equipment, such as 122 mm and 155 mm howitzers, 122 mm multiple rocket launchers, trucks, tank carriers, and AK-47 assault rifles.47 Contras believed to number 10,000 by the time of Esquipulas II. D&D in the Peace Accord? No. The 1987 Esquipulas II merely prevented aid and support to irregular forces (U.S. support to the Contras and Soviet and Cuban support to guerrillas in Guatemala and El Salvador). It also called for free elections in Nicaragua. However, UN Security Council Resolutions 650 and 653 of March and April 1990 extended ONUCA’s mandate to include disarmament and demobilization of Contra forces. On April 18, 1990, the Sandinista government and the Nicaraguan resistance signed a peace accord under which all Nicaraguan resistance forces would be moved to the ONUCA-protected enclaves while the government was to

S-6

11/24/09

4:19 PM

172

Page 172

EVALUATING IMPLEMENTATION TASKS

withdraw all military forces to a point at least twenty kilometers from the enclave borders.48 What Degree of D&D Agreed? 100 percent D&D of the Contras by April 1990. The objective was to create a single national army under a newly elected government by one-sided disarmament and restructuring of national forces.49 Military reforms included cuts in personnel (from 97,000 to an estimated 15,200 in 1993), the abolition of conscription, and disbanding of the militia.50 Who to Implement D&D? ONUCA: Its mandate was subsequently changed to include weapons destruction. A total of 16,554 weapons were collected and destroyed, including 14,408 small arms, 106 mortars, 82 missiles, 740 grenades.51 D&D Planned to Be Completed Prior to Election? No. The elections were held in February 1990, while the first agreements on the Contras’ D&D were signed in March 1990. Was D&D Implemented Prior to Election? No. Was Election Judged to Be Fair? Yes. Electoral result facilitated successful demobilization of the Contras. Was Peace Maintained After Election? Yes. Some minor microsecurity problems. Subsequent D&D Activities. April 1998 government weapons buy-back campaign supported by France, Canada, and Sweden. Exchanged land for weapons. No information available on weaponry recovered.52

RWANDA UN Assistance Mission for Rwanda (UNAMIR), 1993–1994 Forces and Weaponry Total number of combatants, 50,000. Forces Armees Rwandaises (FAR, Rwandese Armed Forces): 30,000; Rwandese Patriotic Front (RPF): 20,000.53 D&D in the Peace Accord? Yes. What Degree of D&D Agreed? According to the peace agreement, the two

S-6

11/24/09

4:19 PM

Page 173

DISARMAMENT AND DEMOBILIZATION

173

armies were to be integrated in a 19,000-soldier new national army and 6,000-person gendarmerie.54 Who to Implement D&D? UNIMIR was to assist with disengagement, disarmament, and demobilization efforts leading to the integration of the armed forces. D&D Planned to Be Completed Prior to Election? Yes. Demobilization/ integration planned to culminate in elections. Was D&D Implemented Prior to Election? No. Implementation derailed by genocide. Was Election Judged to Be Fair? Not applicable. Was Peace Maintained After Election? Not applicable.

SIERRA LEONE Nigeria-led Cease-Fire Monitoring Group (ECOMOG) force from the Economic Community of West African State (ECOWAS), deployed in 1997 UN Observer Mission in Sierra Leone (UNOMSIL), July 1998–October 1999 UN Mission in Sierra Leone (UNAMSIL), October 1999–present As of November 2000, the UK had deployed 400 troops to train the army of Sierra Leone. Forces and Weaponry Estimated total number of combatants: 45,000, consisting of Sierra Leone Army (SLA) and Civil Defense Force (CDF), Revolutionary United Front (RUF), and various paramilitaries. AK-47 main weapon. D&D in the Peace Accord? 1997 Conakry Agreement, under which the elected civilian government had to be restored within six months and the armed forces and rebel groups disarmed. Following RUF offensive in October 1998, new talks began and the Lomé Agreement of July 1999 was signed. Under this agreement a government of national unity was established. Disarmament and demobilization were planned. What Degree of D&D Agreed? Simultaneous disarmament of CDF and RUF. UN Security Council of October 1997 imposed a UN oil and arms embargo.

S-6

11/24/09

4:19 PM

174

Page 174

EVALUATING IMPLEMENTATION TASKS

Who to Implement D&D? UNOMSIL: Mandated to monitor the security situation in Sierra Leone and execute D&D. ECOMOG: Nigerian-led ECOMOG was authorized to enforce the sanction regime, using military force if necessary. Due to domestic reasons, it withdrew and UNOMSIL descended into chaos, as the rebels managed to steal UN weapons, including tanks, and kidnap UN personnel. D&D Planned to Be Completed Prior to Election? UNOMSIL: No. Was D&D Implemented Prior to Election? UNOMSIL: No. Was Election Judged to Be Fair? UNOMSIL: RUF did not participate in the elections and did not recognize the results. The conflict continued. Was Peace Maintained After Election? No.

SOMALIA UN Observer Mission in Somalia I (UNOSOM I), April 1992–March 1993 U.S.-led Unified Task Force (UNITAF), December 1992–March 1993 UN Observer Mission in Somalia II (UNOSOM II), March 1993–March 1995 Forces and Weaponry Lack of reliable information on numbers of Somali combatants at the time of UN intervention in 1992. The Somali armed forces, which at one time numbered 60,000 soldiers, had by that time evaporated, with some forces killed, some forces self-demobilized, and others forming the backbone of the militias. Rough estimates of militia strength run to no more than 12,000, but this figure seems somewhat low. D&D in the Peace Accord? UNOSOM I: No peace accord. UNITAF: No peace accord. UNOSOM II: Strong coercive disarmament. Mandate to disarm, in line with the 1993 peace agreement. What Degree of D&D Agreed? UN arms embargo imposed in 1992. January 1993 UN-sponsored peace agreement included handing over of heavy weapons for safekeeping until legitimate government in place. To be

S-6

11/24/09

4:19 PM

Page 175

DISARMAMENT AND DEMOBILIZATION

175

completed by March 1993. Cantonment, disarmament, demobilization, and reintegration of militia forces. All other armed elements to be disarmed and rehabilitated. Who to Implement D&D? UNOSOM I: Monitor UN arms embargo; monitor cease-fire agreement between Somali factions; protect UN personnel and equipment in Mogadishu; and escort humanitarian relief supplies to distribution centers. UNITAF: Enforcement powers granted in December 1992 were designed to enable the multinational force to establish security in Somalia—using coercive disarmament if necessary. UNOSOM II: Strong coercive disarmament. D&D Planned to Be Completed Prior to Election? No elections planned. Was D&D Implemented Prior to Election? No elections planned. Was Election Judged to Be Fair? No elections held. Was Peace Maintained After Election? No peace after coercive disarmament efforts. UN withdrew without restoration of central authority. Following a national reconciliation conference in August 2000, a transitional government was formed. The government aims to demobilize up to 75,000 militia by 2003 and collect weapons through a food-for-weapons program (1 million small arms believed to be in the hands of militias).55

SRI LANKA Indian Peacekeeping Force (IPKF), 1987–1989 Forces and Weaponry Sri Lanka armed forces and paramilitaries: 70,000; Tamil Tigers (LTTE): 3,000–4,000 full time cadres, plus several thousand auxiliaries.56 D&D in the Peace Accord? Yes. What Degree of D&D Agreed? All Tamil guerrillas to surrender their weapons to specially designated Sri Lankan authorities within 72 hours of the signing of the peace agreement. Who to Implement D&D? IPKF. D&D Planned to Be Completed Prior to Election? Yes.

S-6

11/24/09

4:19 PM

176

Page 176

EVALUATING IMPLEMENTATION TASKS

Was D&D Implemented Prior to Election? No. Was Election Judged to Be Fair? Election boycotted by Tamil groups. Was Peace Maintained After Election? No.

ZIMBABWE, 1979–198057 UK-led Commonwealth Monitoring Force (CMF) Forces and Weaponry Approximately 141,000 combatants: ZANLA (military wing of the Zimbabwe African Nationalist Union) and ZIPRA (military wing of the Zimbabwe African People’s Union), 60,000; Rhodesian Security Force composed of army and territorial force: 81,000. Wide array of weaponry. D&D in the Peace Accord? No. Who to Implement D&D? Given a lack of consent by the guerrillas or government to include disarmament or demobilization in the Lancaster House Agreement, CMF officers initiated sporadic, low-level joint patrols between guerrillas and government soldiers as a first step toward reconciliation and integration. These operations were vetoed by the Rhodesian chief of staff, General Peter Walls. Was D&D Implemented Prior to Elections? No. Was Election Judged to Be Fair? Yes. Was Peace Maintained After Election? Peace maintained between former guerrillas and white military, but sporadic military clashes between troops of the former erstwhile allies, ZANLA and ZIPRA, 1980–1982. ZIPRA dissidents turn to guerrilla attacks in Matabeleland, 1985–1987, until amnesty decreed as part of peace agreement in 1987. Subsequent D&D Activities. Immediately after the 1980 election, a British Military Advisory and Training Team (BMATT) assisted in demobilization of some guerrillas and incorporation into a new Zimbabwean army. Many soldiers self-demobilized, with few reintegration programs. Ex-combatants become powerful lobby in the 1980s, demanding pensions, health care, jobs, and land.

S-6

11/24/09

4:19 PM

Page 177

DISARMAMENT AND DEMOBILIZATION

177

Appendix Notes 1. UN Department of Peacekeeping Operations, UNAVEM II, Background, www.un.org/depts/dpko/missions/unavem2/unavemiib.htm. 2. Fred Tanner, “Weapons Control in Semi-Permissive Environments: A Case for Compellance,” International Peacekeeping 3, no. 4 (winter 1996): 131. 3. Peter Batchelor, “Disarmament, Small Arms, and Intra-State Conflict: The Case of Southern Africa,” in Christopher Smith, Peter Batchelor, and Jakkie Potgieter, eds., Small Arms Management and Peacekeeping in Southern Africa, UNIDIR Disarmament and Conflict Resolution Project (Geneva: UNIDIR, 1996), pp. 69–70. 4. “Angolan Government Forms Amnesty Fund for Ex-Rebels,” AFP (Luanda), January 11, 2001. 5. “Namibia to Display Captured UNITA Weapons,” Panafrican News Agency, February 19, 2001. 6. “Demobilized Soldiers Jobless,” UN Integrated Regional Information Network, March 3, 2001. 7. “Consolidating Peace Through Practical Disarmament Measures and Control of Small Arms: From Civil War to Civil Society,” report from the Third International Workshop of the Stiftung Wissenschaft und Politik (SWP), Berlin, July 2–4, 1998, pp. 91–102. 8. International Institute for Strategic Studies (IISS), The Military Balance 1995/96 (London: Oxford University Press, 1995). 9. Fred Tanner, “Post-Conflict Weapons Control: In Search of Normative Interactions,” paper prepared for International Studies Association (ISA) meeting, Washington, D.C., February 16–20, 1999, p. 10. 10. Ibid., p. 11. 11. KFOR Online, www.kforonline.com/kfor/chronology.htm. 12. General Accounting Office (GAO), Conventional Arms Transfers: U.S. Efforts to Control the Availability of Small Arms and Light Weapons, NSIAD-00141 (Washington, D.C: Government Printing Office, 2000), p. 13. 13. All data taken from Jianwei Wang, Managing Arms in Peace Processes: Cambodia (Geneva: UN Institute for Disarmament Research, 1996). 14. Mats R. Berdal, Disarmament and Demobilization After Civil Wars, Adelphi Paper no. 303 (London: IISS and Oxford University Press, 1996), p. 18. 15. Ibid., p. 61. 16. Tanner, “Post-Conflict Weapons Control,” p. 11. 17. UN Department of Public Information, www.un.org/depts/dpko/missions/ onusal_b.htm. 18. Joanna Spear, “The Disarmament and Demobilisation of Warring Factions in the Aftermath of Civil Wars: Key Implementation Issues,” Civil Wars 2, no. 2 (summer 1999): 28. 19. “Salvador Veterans Say They May Step Up Strife,” Globe, January 26, 1995. 20. Douglas Farah, “Cash for Cashed Weapons; When Asked No Questions, Salvadorans Respond,” The Washington Post, November 1, 1996, p. A27. 21. GAO, Conventional Arms Transfers, tab. 1. 22. See Chapter 15 by William Stanley and David Holiday in this book. 23. See Central American UPDATE: http://www.eecs.umich.edu/~pavr/ harbury/archive/1996/071096.html.

S-6

11/24/09

4:19 PM

178

Page 178

EVALUATING IMPLEMENTATION TASKS

24. See Charles Winslow, Lebanon: War and Politics in a Fragmented Society (New York, Routledge, 1996), p. 272. Also, Anthony Cordesman, After the Storm: The Changing Military Balance in the Middle East (London, Westview Press, 1993), p. 314. 25. Ibid. See also Susan Smith, “Special Report: The Ta’if Agreement: Still on Schedule After 1 1/2 Years,” Washington Report on Middle East Affairs (May–June, 1991): 48. Available online at http://www.washington-report.org/backissues/0591/ 9105048.htm. 26. Cordesman, After the Storm, p. 314. 27. Ibid, p. 312. 28. Shlomo Brom and Yiftah Shapir, eds., The Middle East Military Balance, 1999–2000 (Cambridge: Jaffe Center for Strategic Studies and Belfer Center for Science and International Affairs, 2000), p. 265. 29. Ibid. 40,000 reflects Syrian numbers at start of Ta’if; today, there are 25,000. 30. Winslow, Lebanon, p. 318. 31. See Chapter 20 in this book. 32. UNOMIL, UN Department of Peacekeeping Operations, www.un.org/ Depts/dpko/dpko/co_mission/unomilFT.htm. 33. “Consolidating Peace Through Practical Disarmament Measures,” pp. 79–84. 34. UN Document S/1997/712, “Final Report of the Secretary-General Report on UNOMIL,” September 12, 1997, www.un.org/docs/sc/reports/1997/ s1997712.htm. 35. “Consolidating Peace Through Practical Disarmament Measures,” pp. 79–84. 36. Associated Press, “Liberia Destroys Guns from Civil War in Gesture for Peace,” July 27, 1999. 37. Jim Wurst, “Mozambique: Peace and More,” World Policy Journal 11, no. 3 (fall 1994): 78–83. 38. Berdal, Disarmament and Demobilization After Civil Wars, p. 18. 39. Wurst, “Mozambique: Peace and More,” pp. 78–83. 40. Batchelor, “Disarmament, Small Arms, and Intra-State Conflict,” pp. 69–75. 41. GAO, Conventional Arms Transfers, tab. 1. 42. “The Socio-Economic Reintegration of Demobilised Soldiers in Mozambique: The Soldier’s View,” UNDP, October 1997, p. 20. 43. Chris Alden, “The Issue of the Military: UN Demobilization, Disarmament, and Reintegration of Southern Africa,” International Peacekeeping 3, no. 2 (1996): 51–69. 44. Ibid., p. 53. 45. Batchelor, “Disarmament, Small Arms, and Intra-State Conflict,” pp. 79–80. 46. US Library of Congress, Federal Research Division, Country Studies— Nicaragua, http://lcweb2.loc.gov/cgi-bin/query/r?frd/cstdy:@field(docid+ni0111). 47. Ibid., http://lcweb2.loc.gov/cgi-bin/query/r?frd/cstdy:@field(docid+ ni0112). 48. Ibid., http://lcweb2.loc.gov/cgi-bin/query/r?frd/cstdy:@field(docid+ ni0098). 49. Tanner, “Weapons Control in Semi-Permissive Environments,” pp. 127–145. 50. US Library of Congress, http://lcweb2.loc.gov/cgi-bin/query/r?frd/ cstdy:@field(docid+ni0116).

S-6

11/24/09

4:19 PM

Page 179

DISARMAMENT AND DEMOBILIZATION

179

51. “Breaking the Cycle of Violence: Light Weapons Destruction in Central America,” British American Security Information Council (BASIC) Papers, Occasional Papers on International Security Issues no. 24 (London: BASIC, December 1997), p. 3. 52. GAO, Conventional Arms Transfers, tab. 1. 53. See Alison Des Forges, excerpts from Leave None to Tell The Story: Genocide in Rwanda as a report of Human Rights Watch, available online at: http:/www.hrw.org/reports/1999/rwanda/Geno1-3-11.htm#P824_314441, “The Arusha Accords.” See also, United Nations, The United Nations and Rwanda, 1993–1996 (New York: United Nations Department of Public Information, 1996), p. 224. 54. Des Forges, Leave None to Tell the Story. 55. “New Somali Government Demobilises Militia,” Pan African News Agency, November 28, 2000. 56. Library of Congress, Federal Research Division, Country Studies—Sri Lanka: National Security, http://memory.loc.gov/frd/cs/lktoc.html. 57. Zimbabwe date from Jeremy Ginifer, Managing Arms in Peace Processes: Rhodesia/Zimbabwe, UNIDIR Disarmament and Conflict Resolution Project (Geneva: UNIDIR, 1996).

Notes 1. Joanna Spear, “Arms Limitations, Confidence-Building Measures, and Internal Conflict,” in Michael E. Brown, ed., The International Dimensions of Internal Conflict (Cambridge: MIT Press, 1996), pp. 405–408; and Fred Tanner, “Consensual Versus Coercive Disarmament,” in UN Institute for Disarmament Research, Managing Arms in Peace Processes: The Issues (Geneva: United Nations, 1996), pp. 169–204. 2. Trevor Findlay, “Post Conflict Demilitarization: The Role of Verification,” paper presented to the MacArthur Seminar on Regional Security in a Global Context, Department of War Studies, King’s College, London, February 2, 1999. 3. UN Document S/1995/588, “Report to the Secretary-General on the United Nations Angola Verification Mission (UNAVEM III),” July 17, 1995. 4. “Owning a weapon has been a sign of manhood in Albanian culture for centuries.” Associated Press, “UN Program to Encourage Albanians to Give Up Weapons,” January 12, 1999. 5. “If I do not carry a gun the men despise me as a defenseless woman,” said James Wole of the local NGO Community Development Association in southern Sudan. Cited in Simon Shikangalah, “The Development Brigade: The Namibian Experience,” in Jakkie Cilliers, ed., Dismissed: Demobilisation and Reintegration of Former Combatants in Africa (Pretoria: Institute for Defence Policy, 1995), p. 77. 6. Ibid. Again according to Wole, “you can get six cows for one gun—when it used to cost ten.” 7. In Somalia, nomadic people have traditionally had the right to bear arms to protect their cattle. Hussein M. Adam, “Somalia: A Terrible Beauty Being Born,” in William I. Zartman, ed., Collapsed States: The Disintegration and Restoration of Legitimate Authority (Boulder, Colo.: Lynne Rienner, 1995), p. 79; and “Uganda’s Nomadic Region Given Ultimatum on Guns,” Xinhua, March 23, 1999. 8. Ioan M. Lewis, “Misunderstanding Somalia,” Anthropology Today 9, no. 4 (August 1993): 1–3.

S-6

11/24/09

4:19 PM

180

Page 180

EVALUATING IMPLEMENTATION TASKS

9. Mohammed Ayoob, The Third World Security Predicament: State Making, Regional Conflict, and the International System (London: Lynne Rienner, 1995), p. 172. 10. See Chapter 11 by Charles Call and William Stanley in this book. 11. Paul Claesson and Trevor Findlay, “Case Studies on Peacekeeping: UNOSOM II, UNTAC, and UNPROFOR,” in SIPRI [Stockholm International Peace Research Institute] Yearbook 1994 (Oxford: Oxford University Press, 1994) p. 62. 12. Jonathan Stevenson, “Hope Restored in Somalia,” Foreign Policy 91 (summer 1993): 151. 13. Ian Doucet, “The Cowards War: Landmines and Civilians,” Medicine and War 9, no. 4 (October 1993): 304–316; Boutros Boutros-Ghali, “The Land Mine Crisis,” Foreign Affairs 73, no. 5 (September/October 1994): 8–13; and Arms Project of Human Rights Watch and Physicians for Human Rights, Land Mines: A Deadly Legacy (Washington, D.C.: Human Rights Watch, 1993). 14. Jim Wurst, “Mozambique Disarms,” Bulletin of the Atomic Scientists 50, no. 4 (September/October 1994): 39. 15. Paul Davies, “Mines and Unexploded Ordnance in Cambodia and Laos: Understanding the Costs,” in Krishna Kumar, ed., Rebuilding Societies After Civil War: Critical Roles for International Assistance (Boulder, Colo.: Lynne Rienner, 1997), p. 258. 16. Christopher Louise, The Social Impacts of Light Weapons Availability and Proliferation, International Alert and UN Research Institute for Social Development Research Paper no. 59, March 1995. 17. Wurst, “Mozambique Disarms,” pp. 36–39. 18. Christopher Smith, “The International Trade in Small Arms,” Jane’s Intelligence Review 7, no. 9 (1995): 427–430. 19. Facts on File 54, no. 2786 (April 21, 1994): 272–273. 20. Christopher Smith, “Light Weapons and the International Arms Trade,” in Christopher Smith, Peter Batchelor, and Jakkie Potgieter, eds., Small Arms Management and Peacekeeping in Southern Africa (Geneva: UNIDIR, 1996), pp. 30–57. 21. Mats R. Berdal, Disarmament and Demobilization After Civil Wars, Adelphi Paper no. 303 (London: IISS and Oxford University Press, 1996), p. 39. 22. World Bank, “Demobilization and Reintegration of Military Personnel in Africa: the Evidence from Seven Country Case Studies,” in Cilliers, Dismissed, p. 13. 23. Jakkie Cilliers, preface to Cilliers, Dismissed, p. 10. 24. Henry McDonald, “INLA Pledges to Keep Ceasefire While Avenging Murder,” The Observer, October 17, 1999. 25. Guy Goodwin-Gill and Ilene Cohn, Child Soldiers: The Role of Children in Armed Conflicts (Oxford: Clarendon Press, 1994); and Margaret McCalin, The Reintegration of Young Ex-Combatants into Civilian Life: A Report to the International Labour Office (Geneva: International Labour Office, 1995). 26. Muchaparara Musemwa, “The Ambiguities of Democracy: The Demobilisation of the Zimbabwean Ex-Combatants and the Ordeal of Rehabilitation, 1980–1993,” in Cilliers, Dismissed, p. 50. 27. Paul Collier, “Demobilization and Insecurity: A Study in the Economics of the Transition from War to Peace,” Journal of International Development 6, no. 3 (1994): 349. 28. Jacklyn Cock, “The Social Integration of Demobilised Soldiers in Contemporary South Africa,” South African Defence Review no. 12 (1993): 5–6.

S-6

11/24/09

4:19 PM

Page 181

DISARMAMENT AND DEMOBILIZATION

181

29. Collier, “Demobilization and Insecurity,” p. 344. 30. Ibid., pp. 343–351. 31. Ibid., p. 348. 32. Berdal, Disarmament and Demobilization After Civil Wars, p. 43. 33. Ibid., p. 52. 34. Tsepe Motumi and Andrew Hudson, “Rightsizing: The Challenges of Demobilisation and Social Reintegration in South Africa,” in Cilliers, Dismissed, p. 118. 35. Ibid., pp. 119–120. 36. William Reno, “Privatizing War in Sierra Leone,” Current History 97, no. 610 (May 1997): 228; Herbert Howe, “To Stabilize Tottering African Governments,” Armed Forces Journal International 134 (November 1996): 38–39; and Jim Hooper, “Executive Outcomes,” World Air Power Journal 28 (spring 1997): 38–49. 37. Donald G. McNeil Jr., “Pocketing the Wages of War,” New York Times, February 16, 1997; and Jimmy Seepe, “SADF Men ‘Recruited’ in Angolan War,” The Sowetan (Johannesburg), March 30, 1999. 38. Terrence Lyons, Voting for Peace: Post Conflict Elections in Liberia (Washington, D.C.: Brookings Institution, 1998), pp. 43, 66; and Victor Tanner, “Liberia: Railroading Peace,” Review of African Political Economy 75, no. 25 (March 1998): 137. 39. Chris Smith and Alex Vines, Light Weapons Proliferation in Southern Africa, London Defense Studies no. 42 (London: Brassey’s for Center for Defense Studies, 1997), p. 15. 40. Smith, “Light Weapons and the International Arms Trade,” p. 33. 41. Lyons, Voting for Peace, p. 42. 42. Verification in All Its Aspects, Including the Role of the United Nations in the Field of Verification, UN report, 1995, p. 53. 43. Virginia Page Fortna, “Angola,” draft prepared for this project, p. 22. 44. Nicole Ball, “Demobilizing and Reintegrating Soldiers: Lessons from Africa,” in Kumar, Rebuilding Societies After Civil War, pp. 87–88. 45. For example, in Angola by early 1997, 26,400 of the 69,800 UNITA fighters registered at encampments had self-demobilized. Many of them had returned to their units. Kees Kingma, “Demobilization of Combatants After Civil Wars in Africa and Their Reintegration into Civilian Life,” Policy Sciences 30, no. 3 (August 1997): 152. 46. Berdal, Disarmament and Demobilization After Civil Wars, p. 16. 47. Graciana del Castillo, “The Arms-for-Land Deal in El Salvador,” in Michael W. Doyle, Ian Johnstone, and Robert C. Orr, eds., Keeping the Peace: Multidimensional UN Operations in Cambodia and El Salvador (Cambridge: International Peace Academy and Cambridge University Press, 1997), pp. 342–365. 48. Reuters, “Salvadoran Veterans Say They Might Step Up Strife,” Boston Globe, February 1, 1995. 49. Nicole Ball, “The Challenge of Rebuilding War-Torn Societies,” in Chester A. Crocker and Fen Osler Hampson with Pamela Aall, eds., Managing Global Chaos: Sources and Responses to International Conflict (Washington, D.C.: U.S. Institute of Peace, 1996), p. 608. 50. Berdal, Disarmament and Demobilization After Civil Wars, pp. 13–14. 51. Ibid., p. 15. 52. Unattributable briefing by British officer who spent two years in Angola with UNAVEM II. London, November 5, 1997. 53. Reuters, “For Nicaragua’s Army, a Peaceful Power Transfer,” Boston Globe, February 22, 1995.

S-6

11/24/09

4:19 PM

182

Page 182

EVALUATING IMPLEMENTATION TASKS

54. David Shearer, “Outsourcing War,” Foreign Policy 112 (fall 1998): 68–81. 55. Robin Luckham, “The Military, Militarization, and Democratization in Africa: A Survey of Literature and Issues,” African Studies Review 37, no. 2 (September 1994): 19. 56. Reuters, “UNITA Sells Diamonds, Breaks U.N. Embargo,” December 14, 1998; and Chris Gordon, “Eastern European Aid Boosts UNITA’S Force,” Weekly Mail and Guardian, January 15, 1999. 57. Richard Bernstein, “Sniping Is Growing at U.N.’s Weakness as a Peacekeeper,” New York Times, June 21, 1993. 58. Alex Vines, Human Rights Watch/Africa, address to the Conference on Comparative Regional Security, Institute for Security Studies, Midrand, South Africa, July 1–3, 1997. Cited in Abdel-Fatau Musah et al., Africa: The Challenge of Light Weapons Destruction During Peacekeeping Operations, BASIC Papers no. 23 (London: BASIC, December 1997), p. 8. 59. World Bank, “Demobilization and Reintegration of Military Personnel in Africa,” p. 22. 60. Joseph Hanlon, Mozambique: Who Calls the Shots? (Indianapolis: Indiana University Press, 1991), cited in Michael F. Stephen, “Demobilisation in Mozambique,” in Cilliers, Dismissed, p. 62. 61. Shikangalah, “The Development Brigade,” p. 74. 62. Fortna, “Angola,” p. 9. 63. Lyons, Voting for Peace, p. 11. 64. Stephen John Stedman, “Negotiation and Mediation in Internal Conflict,” in Michael E. Brown, ed., The International Dimensions of Internal Conflict (Cambridge: MIT Press, 1996),” pp. 364–365; and Stedman, “Spoiler Problems in Peace Processes,” International Security 22, no. 2 (fall 1997): 5–53. 65. Stephen, “Demobilisation in Mozambique,” p. 65. 66. Brian Hall, “Blue Helmets, Empty Guns,” New York Times Sunday Magazine, January 2, 1994, p. 24. Cited in Lyons, Voting for Peace, p. 12. 67. See Chapter 14 by Charles Call in this book. 68. Margaret J. Anstee, Orphan of the Cold War: The Inside Story of the Collapse of the Angolan Peace Process, 1992–1993 (New York: St. Martin’s Press, 1996), pp. 50–52. 69. Fortna, “Angola,” p. 19. 70. Stephen John Stedman, “Implementation of the Lusaka Accord,” draft prepared for this project. 71. Findlay, “Post Conflict Demilitarization.”

S-7

11/24/09

4:20 PM

Page 183

7 Economic Priorities for Successful Peace Implementation SUSAN L. WOODWARD

No international or local action in support of peace can occur without a budget or donor to tap. The organization of peace implementation requires financial resources. What donors are willing to fund will heavily influence the actual strategy of implementation, often more than the peace agreement itself. In addition, war is physically as well as psychologically and socially destructive. The first steps taken by people themselves toward peace are physical repair and reconstruction as they struggle to put their lives back together and reorient to peaceful pursuits. The best indicator of success in the first stages of implementation will in fact be the level of such activity, which is a calculated risk that will not be taken if the environment has not begun to change in the direction of greater physical and psychological security. These economic aspects of peace agreements and their implementation are obvious. But to move from the obvious to hard conclusions about the role of economic factors in the success of implementation and to policy recommendations supported by evidence from actual cases is very difficult. This is surprising, given the substantial and growing attention during the 1990s by scholars, specialized research institutes, and international organizations such as the World Bank and UN agencies to the causes of state failure and civil war, to the characteristics of war-torn societies, and to policies for postconflict reconstruction. By 1999, such attention had even provoked substantial policy initiatives and new mechanisms at the international financial institutions (IFIs), the UN, and many bilateral development agencies.1 Nonetheless, there has been no systematic analysis of the contribution of economic factors to the success or failure in the implementation of peace agreements. One explanation may be that the economic aspects of peace agree183

S-7

11/24/09

4:20 PM

184

Page 184

EVALUATING IMPLEMENTATION TASKS

ments and the staging of their implementation tend to take a backseat to security concerns. Two case studies for this project conclude that peace agreements are only achieved at the cost of ignoring the economic disputes (often identified as “root causes”) that led to war.2 Mounting evidence of the longer-term economic trajectories of postconflict cases shows that countries that have experienced a civil war never fully recover to the economic level they had before war began. Their growth paths also remain lower economically than comparable countries that did not have a civil war. But both of these findings suffer the same temporal problem in judging the role of economic factors in successful implementation; the outcomes of economic assistance are often not known for quite some time. Moreover, economic conditions usually worsen after the civil war ends. And while armed hostilities may cease, growing economic inequalities and hardship can fuel increased violence from crime or social crises after the peace agreement, thus substantially complicating the tasks of building peace and stable government. These outcomes have tended to be judged as regrettable but inevitable rather than as a matter of policy choice. Peace agreements, as a rule, do not address economic reconstruction and development. Economic objectives are often included in a list of concerns, but with little and vague discussion, particularly when compared with the attention given to security and justice. There are a few partial exceptions—Bosnia-Herzegovina, El Salvador, Sierra Leone, and Mozambique, as well as the Oslo Accords for Palestine—that addressed economic concerns directly. In Guatemala and Bosnia-Herzegovina, representatives of the IFIs were present at the peace negotiations and had some influence over the outcome. But far more common is the example of Cambodia, where the mandate of the UN Transitional Authority there (UNTAC) was “limited to rehabilitation” and where assessments rated economic reconstruction as “one of the least effective aspects,” involving “exceptionally poor planning and administrative confusion.”3 An explicit mandate to a reconstruction agency is rarely included in peace agreements. There is somewhat greater knowledge about the economic role of outsiders in peace implementation. Widespread consensus exists that economic resources do matter to success. The surest way to failure, as illustrated by the failures in Angola and Rwanda, the mixed results in Cambodia, and even by certain important aspects of the otherwise successful case of El Salvador, is not to provide sufficient external resources in support of a peace agreement. But even more influential than the level of resources available is the kind of resources provided. External assistance can be “redundant, harmful, or squandered.”4 A repeated theme of postconflict reconstruction assessments is the stark contrast between what is known to be needed in the first years after war and what is currently funded and done. A summary of the role of Bretton Woods institutions in peacebuilding

S-7

11/24/09

4:20 PM

Page 185

ECONOMIC PRIORITIES FOR SUCCESSFUL PEACE IMPLEMENTATION

185

written in 2000 concludes: “IFI efficacy cannot make a peace process, but IFI inefficacy can break one.”5 Another reason for the lack of analysis may lie in the nature of the policy response to such findings and criticism, which has tended to focus on operational aspects, particularly efforts to improve the delivery of aid—for example, faster, more efficiently, with greater flexibility and transparency, better coordination among donors, or more targeted conditionality.6 Some of these refinements in the technology of aid appear to improve chances of success. But one would not know from the actual measures of that success because these measures are also technical—such as performance criteria for the delivery of money or the completion of economic projects and programs. These technical measures do not assess the contribution of aid to the implementation of the peace agreement or its political goals. Finally, one might be able to deduce conclusions from recent efforts by the IFIs to learn lessons from experience or to concede to criticism coming from other donors in regard to the substance of economic policy and strategy. But the result has been shifts in policy emphasis from one fad to another, for example, from macroeconomic stabilization to good governance to social capital. These shifts confound the task of analysis and assessment further. They also reveal the absence of agreement on appropriate economic strategy specifically addressed to the tasks of implementing peace agreements and the first years after war.

The Economic Needs and Priorities of Peace Implementation There are three sets of economic tasks necessary to implement a peace agreement: sufficiently rapid economic revival to buy confidence in the peace process; funding to implement specific commitments in the peace agreement; and the economic foundations necessary to sustain peace over the long term. Reviving the Economy The first, immediate economic tasks are both straightforward and inexpensive, although their funding is often quite difficult to secure. The provision of services such as garbage collection, sewage, electricity, power plants, and sufficient, clean water are not only basic to health but also the best way to build confidence in government, that elusive ingredient called “social capital” that is considered essential to a successful transition and collective action of any kind.7 Such services may only require repair of existing facilities, but they may require entirely new construction.

S-7

11/24/09

4:20 PM

186

Page 186

EVALUATING IMPLEMENTATION TASKS

Because these simple economic necessities are obvious, their political role in the success of a peace agreement—and thus the significance of funds for their provision—can easily be overlooked. Peace agreements are commitments signed by leaders, but the population must also be persuaded to make a commitment to peace. Some signal that there will be a “peace dividend”—that the benefits will outweigh the costs—must be palpable for people to make even the minimal investment in rebuilding their lives and supporting politically those promoting peace. In the words of Christine Wallich, 1995–1997 director of the World Bank’s five-year program for reconstruction, rehabilitation, and recovery in Bosnia, “an accepted prior of all peace agreements” is an early peace dividend and quick transition to earned incomes that provide “relief from the cruelty of the environment and a clear signal from the international community that they stand to support them and the agreement.”8 Such apparent consensus, however, has not translated into early assistance and aid policy as often as its acceptance would lead one to expect.9 In addition to speed, this dividend must be equitable. Reviving the inequalities that caused or fueled civil war, or creating new ones, will create divisions that can be easily mobilized by populist rhetoric from politicians who are dissatisfied with the terms of the peace or its actual results. Distributional differences are inevitable, such as between areas with greater or less war damage or the needs of urban and rural populations. But if these differences are perceived as unjust, there will be at the least fuel for spoilers and for delays in demobilization. Moreover, if equal access to basic services including education and healthcare and explicit attention to land redistribution are not addressed in the first stage of peace, macroeconomic policies aimed at currency stabilization and at domestic and foreign investment in growth will institutionalize the inequalities produced by wartime and likely worsen their effects.10 The political task of achieving greater equity later will be very difficult. Success in implementation will depend on a transformation in methods of political conflict and competition, from the threat and use of violence to those of nonviolence. Generalized poverty is less threatening to such practices than are sharp inequalities.11 A contemporary vogue to view civil wars in the Cold War period as based on class conflict and those since the Cold War as wars of identity and ethnicity may have wrongly diverted attention away from economic inequality. Even if there were any truth in this tendency to essentialize identity and ethnicity, the end of a civil war requires warring factions of different identities to create a single state—not only to accept a new balance of political power but also for some to abandon hopes of independence. The ability in such cases to create one constituted state, whether through majoritarian or proportional rules, depends even more on avoiding

S-7

11/24/09

4:20 PM

Page 187

ECONOMIC PRIORITIES FOR SUCCESSFUL PEACE IMPLEMENTATION

187

the emergence of economic inequalities that can be interpreted as discriminatory on cultural grounds (religion, language, lineage). The concept of security must also change from that required in wartime to the expectations of law and order and broader human security. While most peacebuilding missions give priority early on to demobilization and reducing the capacity to make war and may even include monitors on human rights, they tend to neglect the institutions that must fill the security vacuum that arises—law enforcement and assistance to organizations that can bargain with authorities over human rights protection or wages and prices.12 Yet, overall crime rates tend to rise in the first years after war and threats to individual lives may actually be greater than during the war. Peacekeeping missions now recognize increasingly that law enforcement cannot be delayed until the separation of forces and their cantonment or demobilization are complete, but funding sources have not caught up. Moreover, as one practitioner with leading positions in many UN peaceimplementing missions has emphasized, the primary focus of the average citizen in the first year is on everyday concerns for material survival, such as jobs, pensions, schooling, and healthcare, which are perceived fundamentally as a matter of security.13 Financing the Peace The second set of tasks are specific commitments made in the peace agreement that simply will not be met if they are not funded and organized. Most prominent in studies of peace implementation are the demobilization of wartime armies and the right of refugees and internally displaced persons to return home.14 Return requires substantial resources for housing, transport, and some minimal welfare. Although spontaneous return is the most common experience, the more important the role of return is to achieving the political goals of the peace, such as the attempt to reverse ethnic cleansing in Bosnia-Herzegovina, the more that external resources will be necessary to success. And where spontaneous return depends on remittances sent by relatives living abroad, such in El Salvador, there will be a cost in the return of those who sent remittances. Even more necessary to success in implementation is funding for the demobilization of armed groups. As Joanna Spear points out in Chapter 6, without monies and programs for the reintegration of demobilized soldiers into civilian society, demobilization will fail. Although a stable government is rarely a specific element of a peace agreement, there will be no peace without it. The construction of a new government, or an entire governmental system, requires salaries for civil servants and perhaps also training, particularly, as Terrence Lyons suggests in Chapter 8, if the goal is to transform former warring factions into politi-

S-7

11/24/09

4:20 PM

188

Page 188

EVALUATING IMPLEMENTATION TASKS

cal parties capable of governing. Physical reconstruction of infrastructure—power plants, schools, health clinics—makes no sense without monies to pay the salaries of the people who run them, but many donors have rules against paying local wages and salaries. Elections figure prominently in most third-party engagements in peace implementation. The creation of a voter register, preparation of electoral rules and regulations and their enforcement, and the mounting of information campaigns cost substantial monies. Third-party mediators increasingly urge the adoption of a power-sharing agreement among the parties as an inclusionary political formula to end the war, but such arrangements also presume public expenditures. Sovereignty depends not only on international recognition and a functioning government but also on relations with the rest of the world. Embassies and ambassadorships need to be funded. Even more problematic and urgent are relations with IFIs; without full membership in the International Monetary Fund (IMF), countries cannot receive World Bank, International Development Association (IDA), or International Fund for Agricultural Development (IFAD) loans, borrow in international capital markets, or join trade associations.15 For membership, a country must first settle its debt arrears. Sustaining the Peace The third set of tasks arise because initial success in implementing a peace agreement cannot be sustained unless the bases of a functioning government and economic growth are also put in place. Both political and economic institutions must be built. Years of war will have created a genuine need for economic development, if this had not already been the case. Transforming a war economy will require legalization of property rights and the elimination of trafficking routes that provided arms, ammunition, and financing for the war, and the creation of judicial systems to protect contracts and property rights. It is likely also to require a total reorientation of a production structure and economic policy that was based on defense to civilian industries and export production. While the literature on civil war emphasizes human reintegration, the reintegration of the country in physical, economic, and administrative terms will also be necessary. Sustainability means both a continuation of peace and an increasing capacity to survive without external assistance, or at least to reach a level of economic independence in which foreign loans and credits can be serviced effectively and domestic revenues are sufficient to keep the budget deficit at a manageable level. In the short run, sustainability requires generating popular confidence that the hardships will be only temporary.

S-7

11/24/09

4:20 PM

Page 189

ECONOMIC PRIORITIES FOR SUCCESSFUL PEACE IMPLEMENTATION

189

Strategy Until the second half of the 1990s, donors responded to these economic tasks and their financing with long-established programmatic approaches developed for other circumstances. The tasks were divided into two categories: relief and development. Relief was provided on humanitarian principles by humanitarian organizations or emergency response divisions of development agencies. Development was provided by development agencies, both bilateral and multilateral. The increasing number of cases of third-party-assisted implementation of peace agreements during the 1990s, and the growth of humanitarian aid and peacekeeping missions, however, led to a recognition that there was a gap between these two types of assistance. Very specific lessons drawn from the mounting number of cases seemed to demonstrate that quite different policies were needed for the immediate transition that were neither relief nor development. The result was to reconceptualize the division between relief and development as a continuum and the stages in that continuum in terms of transitions and postconflict reconstruction. Nevertheless, by the year 2000, only one government, Germany, had taken this lesson on board in terms of budget lines; all others perpetuated the inherited approaches and distinctions between relief and development through separate budget lines. Some governments did create new sections, such as the Office of Transition Initiatives (OTI) within the U.S. Agency for International Development (USAID), or the addition of the word “conflict” to the humanitarian affairs department (CHAD) of the UK’s Department for International Development (DFiD). Canada established a Peace-Building Fund, and similar special budgetary lines for policy coordination under these circumstances were introduced by Denmark, the Netherlands, and Norway. But the distinction between relief and development and the resulting bureaucratic stovepipes remains the dominant influence over funding and field operations. These two aid strategies, relief and development, are based on different economic models: the natural disaster model for crisis relief, and the postwar stabilization and reconstruction model for development. The Natural Disaster Model The natural disaster model provides the conceptual framework for most postagreement assistance in the first year or two, as it does for most humanitarian assistance and emergency relief during wartime. Organized by crisis-response divisions of bilateral development agencies and implemented by international NGOs, its principles are rapid response and the use of local labor. Its aims are to provide an immediate, visible peace-dividend

S-7

11/24/09

4:20 PM

190

Page 190

EVALUATING IMPLEMENTATION TASKS

in food, shelter, and basic utilities until longer-term developmental projects can be implemented, and to move the population as rapidly as possible toward self-sufficiency and away from aid dependence. The goal of buying commitment to peace overrides economic objectives, for small repair tasks do not by themselves generate jobs or local economic activity, except in the rare cases where the international mission decides to favor local produce and purchase locally produced materials. Its political aspects are similarly narrow. The connection between this approach and wartime relief is largely organizational. Agencies that have the budgetary authority to act quickly, taking aid decisions within hours or days, dispersing monies and matériel quickly, without cumbersome bureaucratic procedures, are the emergency response and disaster relief agencies that deploy under both circumstances. Adaptation to the special conditions of a peace implementation mission did introduce another organization—civil affairs or civil-military cooperation (CIMIC) officers within military units of the peacekeeping troops— into the early delivery of aid. An innovation of the UK’s Overseas Development Agency (ODA, predecessor of the DFiD), first in BosniaHerzegovina in early 1996, and now widely copied by other countries and the European Union (EU), CIMIC aimed to improve cooperation in emergency circumstances between development agencies and military forces— two organizations that represent two cultures and conscious distancing. Impressed by the importance of showing immediate, visible progress and by the fact that the first representatives of the international community to deploy in large numbers are military forces, officials at the ODA decided to develop a program of microreconstruction projects that soldiers would identify with the help of local communities and then supply. In turn, confidence would be built between the peacekeeping troops and the local population.16 Organizationally equipped to respond rapidly and often directly knowledgeable about the local conditions and the immediate needs of the population, humanitarian workers and the military nonetheless have a short-term approach, which they tend to continue beyond the initial stages of peace implementation because they gain a vested interest in the program’s continuation. World Bank staff argue that this is a general problem, that “in the transition from relief to development, too much time has been spent in the relief phase.”17 The debate over how fast one should move from emergency relief to development, however, often obscures the fact that development strategy for postconflict reconstruction has also been based on the same assumptions. The World Bank, the dominant force globally in defining postconflict reconstruction strategy, acknowledges in its 1998 review of Bank experience in postconflict reconstruction that the model underlying that strategy from its origins in the 1970s and 1980s to 1998 was also that of recovery from natural disasters.18

S-7

11/24/09

4:20 PM

Page 191

ECONOMIC PRIORITIES FOR SUCCESSFUL PEACE IMPLEMENTATION

191

Postwar Stabilization and Reconstruction Model The postwar stabilization and reconstruction model is a neoliberal economic strategy emanating from the IFIs, particularly the IMF. The IMF determines the conditions for foreign confidence in a country’s economic prospects and places priority on macroeconomic stability, both as the first task of economic policy and as the context within which all other aid and policy takes place. Its stabilization model has been criticized frequently for being in conflict with the goals of social peace and reconciliation. The tension has organizational and strategic aspects as well, between representatives of the IFIs and their role in financing reconstruction, on the one hand, and the representatives of the peace mission (most often the UN) and its mandate to implement the peace agreement, on the other.19 As a postwar model, this strategy emerged in response to the particular needs of Western Europe after World War II, when short-term balance-ofpayments constraints on foreign trade threatened the import needs of postwar reconstruction and the revival that depended on it. The IMF response in 1947 to the British liquidity crisis evolved over the next decades into a development strategy based on neoliberal principles of economic growth and free trade. The Articles of Agreement of the World Bank include the express purposes of “the restoration of economies destroyed or disrupted by war” and “the reconversion of productive facilities to peacetime needs.”20 The two-pronged strategy of macroeconomic stabilization and structural adjustment begins with an IMF conditionality program and credits drawn on a country’s special drawing rights (SDRs) at the IMF in support of stabilization based on orthodox deflationary principles of monetary and fiscal restraint. It is followed by sovereign loans mobilized from donors by the World Bank in support of Bank-designed projects for large-scale physical infrastructure and for reform of economic institutions and policies to effect liberalization and privatization that are said to promote economic growth and attract foreign investment.

Barriers to Effective Economic Support of Peace Implementation Sequencing Although there is internal coherence to these economic models and the strategies based on them, their application to contemporary postconflict settings creates a series of practical problems. Current practice is to treat this as a matter of tactics and, in particular, a question of sequencing. The parallel is strong with the debates in the 1990s over the sequencing of market

S-7

11/24/09

4:20 PM

192

Page 192

EVALUATING IMPLEMENTATION TASKS

transitions in postcommunist countries between advocates of shock therapy, who insisted on rapid stabilization, price liberalization, and privatization, and the gradualists, whose local expertise led them to emphasize the institutional and human infrastructure necessary for the expected supply response. Evidence from the successful postsocialist transitions and the developmental failures in Africa reinforce this challenge to the neoliberal model.21 Its relevance to postconflict countries, where the focus was on achieving economic sustainability without losing the peace, is even more questionable. First, the alleged success of the orthodox model in post–World War II Europe occurred in dramatically different conditions than current cases of peace implementation. Post–World War II Western Europe faced the tasks of physical reconstruction and demobilization after an interstate war, in countries that were victors, and whose governmental and economic institutions were intact. Their tasks were not those of state-building, nation-building, and reconciliation after a civil war, integration of former enemies into one army and one society, overcoming underdevelopment, and the implementation of a peace agreement. Moreover, the characteristics of a war economy then were entirely different than the war economies created by civil war in poorer countries now.22 Priority on macroeconomic stabilization assumes, for example, that the economy is fully monetized when barter, informal economies, and illegal trafficking actually predominate. In contrast to mass mobilization for wartime production, the war economies that must be transformed to peacetime economies in contemporary cases of civil war are not emergency adjustments to an otherwise normal economy but an entire transformation of social and political institutions. The role of women is critical in return, reconciliation, and the reestablishment of households, but “the Bank [not to speak of the IMF] has done little to incorporate gender issues in its post-conflict portfolio.”23 The obstacles to foreign investment in postwar Western Europe did not include the high level of uncertainty surrounding the prospects for peace and the absence of a capacity to enforce contracts, property rights, and repatriation of profits. While some countries then might have been at risk from social revolution from below, they did not risk U.S. withdrawal or a collapse of the peace. Moreover, the neoliberal orthodoxy of IFI policy and practice in the 1990s even ignores crucial lessons of the post–World War II experience, particularly “the role that, historically, social policy has played in the development of western European economies.” Although “strong institutions that deliver public policy, the rule of law, and sustain open societies are fundamental to the functioning of a modern economy . . . they have been consistently underestimated by market fundamentalists.”24 This tendency has been powerfully reinforced by the “exclusionary clause” and the “doctrine of economic neutrality” of the charters of both the IMF and the World Bank, which, according to interpretations dominant in the legal

S-7

11/24/09

4:20 PM

Page 193

ECONOMIC PRIORITIES FOR SUCCESSFUL PEACE IMPLEMENTATION

193

offices of the organizations, explicitly prevents them from being involved in anything political, including funding for explicit peacemaking functions.25 In postconflict conditions, there are several deleterious consequences. Privatization before business infrastructure exists makes no sense, while privatization of economic assets before the transfer of power from armies to accountable civilian governments and successful legal control over criminal networks will seriously delay—if not prevent—the political goals of the peace agreement. The method of privatization also has political consequences and should be subject to political as well as economic criteria if it is to contribute to successful implementation. Only belatedly have the IFIs begun to recognize that rapid privatization may be counterproductive to peace in many instances.26 Likewise, IFI policies of macroeconomic restraint prevent the public expenditures that are essential to peace, such as building a new, competent civilian administration, financing demobilization, and providing social infrastructure such as healthcare and schools. Although in this aspect the IMF and the World Bank are frequently in conflict in their loan criteria and proposals, as manifested in Mozambique, Nicaragua, and Macedonia, the World Bank tends to pull back, deferring to IMF targets on budget deficits despite their negative political effects. In its 1998 evaluation, for example, the Bank admitted that “the Cambodia case study finds that the Bank has continued to push for downsizing the civil service when the political coalition arrangement under the peace accords was based in part on raising the size of the civil service to absorb large numbers of the incoming parties’ functionaries. The Bank’s position was not politically realistic from the outset.”27 Orthodox stabilization policies also tend to increase and exacerbate economic inequalities, certainly in the short run, and tend to create more unemployment, when achieving the reverse is essential. Tight monetary policy inhibits cheap credit for the promotion of local small and medium enterprises (SMEs), even though their critical importance early on in the peace process is now widely acknowledged even by World Bank staff. Privatization and the promotion of agricultural exports work directly counter to the crucial task of land reform and redistribution and of poverty reduction. IFI Rigidities As mentioned above, the ability of IFIs to design and implement effective economic assistance to peace implementation has been hamstrung by strict interpretations of the charters of the IMF and the World Bank, which prevent funding for explicitly political tasks like peacemaking. By charter, too, the IMF and the World Bank deal only with sovereign lending, that is, they

S-7

11/24/09

4:20 PM

Page 194

194

EVALUATING IMPLEMENTATION TASKS

work only with recognized governments and are bound to respect their sovereignty. Although this is also true of the UN agencies, they are less constrained than the IFIs. For example, where there is no government (as in Somalia), where a government exists but is not a member of the IMF (as was true for a time of Bosnia-Herzegovina), where relations with the IMF have been suspended because of default (as in Liberia), or where circumstances such as these prevail, as is likely in many immediate postwar settings, the IFIs cannot act at all or cannot remain engaged once they have spent their small grant monies. If a peace agreement applies only to a region of the country (as in Eastern Slavonia in Croatia, and Kosovo in Serbia), then the fact that the IFI can only work through the recognized government may actually provide the means for that government to undermine the peace agreement. Indeed, this is precisely what the Croatian government attempted to do with the Erdut Agreement, an obstructionism that became an “open sore” for the mission of the UN Transitional Administration for Eastern Slavonia, Baranja, and Western Sirmium (UNTAES). Coordination Decisions on sequencing and priorities are also the source of conflicts in the field operation among organizations and approaches to peacebuilding. As Bruce Jones observes in Chapter 4, such obstacles to strategic coordination are more often than not resolved by pluralism—donors are left to pursue different approaches in different parts of the same country. Although this practical solution may reduce obstacles to the flow of aid, it introduces inconsistencies and potential political problems into the implementation strategy. In Mozambique, for example, some regions continued to receive food aid free while people in another region were required to pay, a disparity that can actually slow the peace process. Similarly, bilateral donors and UN agencies tend to leave leadership on macroeconomic stabilization and policy reform to the IFIs, while designing and financing development projects, either as donors to a World Bank program or as independent actors, along sectoral lines within that framework. They do not use their disagreements to propose an alternative strategy. Meanwhile, sectoral coverage can be hit-or-miss and lead to wasteful duplication.28 The “let a hundred flowers bloom” approach of these other agencies and donors does not mean that they support the IFI approach. In fact, the IFI approach generates much criticism. The problem, as Alvaro de Soto and Graciana del Castillo made abundantly clear in their seminal study, is that the economic approach and decisions of the IFIs, particularly the IMF, and the political tasks of implementing a peace mission, are often directly in conflict.29 Although the World Bank acknowledges that it has paid insufficient attention to sequencing and coordination in postconflict countries, this has

S-7

11/24/09

4:20 PM

Page 195

ECONOMIC PRIORITIES FOR SUCCESSFUL PEACE IMPLEMENTATION

195

not influenced the package of economic reforms that make up its advice and loans, or its conviction that these reforms should take place as early as possible so as to attract foreign investment. Perhaps the best evidence that these approaches to peacebuilding are inadequate was the acknowledgment by donors, particularly the World Bank, that their involvement in postconflict missions is ad hoc and in need of a framework. In 1996 the Operations Evaluation Department (OED) of the Bank was tasked to review its involvement in postconflict cases (18 countries with 157 Bank-supported operations representing $6.2 billion in lending at the time)30 and to draw lessons for future policy. Its findings, based on three field studies (BosniaHerzegovina, El Salvador, and Uganda) and six desk studies (Cambodia, Eritrea, Haiti, Lebanon, Rwanda, and Sri Lanka), emphasized the key importance of speed and flexibility. Yet its recommendations suggest these two criteria are in conflict. While emphasizing speed, including more flexible and simplified procurement procedures to reduce delivery delays, the study urges greater flexibility on pace to be more sensitive to the social and political context. Sequencing of policy reforms had to be sensitive to the fact that in the early transition from war to peace, governments are fragile and unstable, and can rarely push rapidly on economic reforms. “Bank and Fund insistence on rapid increase in tax effort (ratio of tax revenues to GDP) may well have been counterproductive, constraining growth of the economy and the size of the tax base.”31 Similarly with regard to tax policy in the cases of Guatemala and Uganda, “the Uganda case study finds that owing to a history of predatory government tax policy during the conflict periods, pressure by the Fund and Bank on tax effort (often included in a standard stabilization package) has had a chilling effect on private investment, driving economic activity into subsistence, or along with investable funds, abroad.”32 Rapid privatization may make the “prospects for sustained, equitable development” worse. And many elements of conventional developmental wisdom, such as a focus on primary education and healthcare, may be incorrect for these circumstances when the effects of war may make secondary and tertiary education and hospitals far more urgent.33 Nonetheless, these assessments evaluate the success in implementing specific donor projects, not their contribution to the overall implementation of the peace agreement or to sustaining the peace. Indeed, the three main recommendations in this World Bank study focus not on altering its approach but on increasing the role of the Bank’s expertise in the peace process: first, that the Bank should encourage invitations to participate in the peace negotiations itself, to advise on the economic implications of particular choices; second, that it should act to coordinate donors and assistance programs in each case because external aid coordination is more important where a government has been weakened by war; and third, that unusually large, resident World Bank missions should be established to gain the flexibility and sensitivity possible from decentralized operations.

S-7

11/24/09

4:20 PM

196

Page 196

EVALUATING IMPLEMENTATION TASKS

Similarly, the World Bank and the IMF have taken on board the recommendations made by de Soto and del Castillo in 1994: to be more transparent in their interactions with noneconomic peacebuilding actors and agencies through the systematic and regular exchange of information; to accept the need for coordination in the field; and to be flexible in their conditionality when it appears directly contrary to immediate peacebuilding goals.34 But this has meant that the World Bank seeks, with IMF agreement, to become the lead agency on reconstruction and development in the mission. Special funds for demobilization, de-mining, budgetary and recurrent expenditures (the Holst Fund for the West Bank and Gaza Strip), a political risk guarantee facility (developed for Bosnia-Herzegovina), “peace technology,” and postconflict projects have all been created in the late 1990s, but within the context of Bank programs and relatively “rich” cases (Bosnia-Herzegovina, Palestine).35 Despite the negative lessons from El Salvador and Mozambique, for example, on the restraints on demobilization and reintegration imposed by IMF loan conditions on inflation, budget deficit, and credit targets, the IMF has not shown willingness to adjust loan criteria in the first crucial years so that peace goals can be financed. Moreover, none of these changes amount to a strategy designed for the purpose of peacebuilding. As Jonathan Stevenson concludes after an analysis of the changes made between 1995 and 2000, “[The Bretton Woods institutions] have not critically explored the connection between structural adjustment and peace, have only circumspectly approached the sensitive issue of security-sector conditionality, and have not optimally synchronized structural adjustment and development strategies with political efforts aimed at consolidating fragile peace agreements following protracted periods of armed conflict.”36

Who Pays for Peace? There is an underlying disequilibrium in the political economy of the first postwar years: a huge gap between popular expectations and peace requirements, on the one hand, and government revenues and fiscal capacity, on the other. Initial repair and political tasks are not particularly expensive relative to later capital investments, but the demand on public expenditures is huge in relation to the low capacity of new governments to extract resources from the population and of civil servants to use available resources efficiently. This problem of capacity is politically serious because of the central focus of external third parties on the signatories to the peace agreement and their commitment to implement it. An increasing emphasis on early elections as a major component of the peace process, aimed at increasing leaders’ accountability for implementation, makes the political implications even more immediate and consequential. Instead, the insecuri-

S-7

11/24/09

4:20 PM

Page 197

ECONOMIC PRIORITIES FOR SUCCESSFUL PEACE IMPLEMENTATION

197

ties of the first stage of peace implementation encourage politicians to seek votes on populist grounds, promising improved welfare or less onerous policies. Therefore the critical question of economic strategy and policies, as de Soto and del Castillo posed it in 1994, is: Who pays for peace?37 The choices of policy and sequencing of tasks will be hostage to the source and availability of funds. If external resources are to fill part of the gap between inevitably insufficient local resources and the immediate needs of peace, early negotiations with the IMF and its stamp of approval are necessary. As the World Bank assessment in 1998 emphasized: the comparative advantage of the Bank and the Fund is in establishing a “sound macroeconomic framework,” and they should take the lead on macroeconomic and external debt issues. Their comparative advantage on these issues does not, however, guarantee that the IFIs will accept that responsibility. The World Bank nearly always takes the lead in designing a program for economic recovery. Bank officials are inclined to insist that it is better suited to do a needs assessment than any other agency. Bilateral donors also tend to defer to the Bank’s sheer technical capacity for all aspects of a financial project. In addition, the World Bank Group has a special lending instrument for poorer countries—its IDA window—that is of particular use for most countries emerging from war. The very size of World Bank funds and reconstruction programs tends, in turn, to give it dominant influence in the field mission over the economic program chosen for peace implementation. There are also reasons of international power why the IMF and the World Bank are likely to be the dominant shapers of economic strategy and sequencing. The U.S. Treasury Department consistently argues successfully for IMF leadership because of its policy of conditionality and its capacity to use conditionality in obtaining reform. Because the World Bank, in turn, does not operate on budgetary assessments of its member states, major powers such as the United Kingdom and the United States lobby as well, and usually successfully, for World Bank leadership so that they do not have to pay additional budgetary assessments. Like most countries, they prefer to use their monies for visible bilateral aid projects for which they can claim direct credit and show taxpayers at home what they are doing. As banks, the IFIs take account of exposure to risk and the political will to reform in their decisions to become involved and lend. Unless there is strong political pressure from a major member of the executive board of the IMF or the World Bank,38 the IFIs will be reluctant to support what appears to be a risky peace. This, in turn, can lead to self-fulfilling prophecies about the success of aid. The minimal amounts of aid in support of the peace agreement itself that flowed into Cambodia, Sierra Leone, Liberia, and Angola can be explained by initial assessments that the prospects for sustainability and reform were low. In the case of Bosnia-Herzegovina, the

S-7

11/24/09

4:20 PM

198

Page 198

EVALUATING IMPLEMENTATION TASKS

World Bank sought to reduce this problem by developing a new financial instrument, the political risk guarantee facility, but this is aimed at attracting private investors, not at easing IFIs’ reluctance to engage. Even where the IFIs choose to act, this caution often leads to very significant and harmful delays in financing. When they do act, it is often because the country has already been engaged in negotiations or has actual loan programs with the IMF and the Bank that preceded or accompanied the war. Although these programs are unrelated to the peace negotiations and the tasks of peacebuilding, their legacy imposes a burden on postwar policy formation on both sides. For the IFIs, there appears to be an inclination to restore their reputation against those who would identify connections between their policies and the outbreak of conflict. Unfortunately, it is also not so easy bureaucratically to refashion the content of ongoing or stalled programs to the new demands of peace. For the government, the burden can be a psychological wariness based on hard experience with IFI programs in the past. This can dilute the prime political advantage claimed by the IFIs in imposing harsh conditions, that governments can more easily withstand public criticism and focus on reconciliation and social peace if they can blame inescapable budgetary stringency and austerity on the IMF. Furthermore, an IMF program depends on a willing bilateral donor or group of donors. The problem of debt must be addressed. If new borrowing is not to be used simply to settle past debts, but is to go toward rebuilding, then a patron willing to risk an early commitment to the success of the peace must be found (such as the Netherlands did for Bosnia-Herzegovina). But this is not easy. In addition, if monetary stability is to be achieved without at the same time restricting public expenditures essential to peace implementation, then monies must also be found from a donor willing to finance a higher budget deficit level than the IMF would otherwise allow. These additional funds are also critical because donors tend to follow closely the terms of the IMF agreement in their own lending policies, despite their freedom to do otherwise and their frequent criticism of IMF conditionality. The result is a culture of conditionality toward a particular country, in some instances inflexible, while in others less so, that becomes fixed very early, influencing the long-term path of the peace process and its outcomes regardless of its detrimental effects.39 Despite the lesson that a too rapid pace of economic reform will be detrimental to a peace process, the sharp decline in official foreign aid during the 1990s and its replacement by foreign direct investment and commercial lending still appear to drive donor policy; major donors, led by the World Bank and the United States, place high priority on economic reforms aimed at attracting private foreign investors. In conclusion, the relief-development “gap” actually represents the vacuum in financial mechanisms designed for peace implementation and the transition between emergency relief and long-term development. Relief

S-7

11/24/09

4:20 PM

Page 199

ECONOMIC PRIORITIES FOR SUCCESSFUL PEACE IMPLEMENTATION

199

occurs because humanitarian organizations and agencies of governments and international organizations exist to provide it. Long-term development also has its internationally institutionalized mechanisms for mobilizing resources and expertise. As Marika Fahlen wrote when she headed the Development Assistance Committee (DAC) of the Organization for Economic Cooperation and Development (OECD): “Unlike fund-raising for emergencies and annual pledges for long-term development, there is no specific resource mobilization mechanism for post-conflict recovery of a combined developmental and humanitarian nature. . . . The UNDP-led Round Tables are sometimes late in coming about and the World Bankchaired Consultative Group meetings do not make much sense without a legitimate government in place.”40 As Shepard Forman and Stewart Patrick demonstrate, there is no “international regime” for postconflict or peace implementation tasks: to the extent that resources flow, it is voluntary and ad-hoc.41 But the question remains: Will the power of the purse continue to drive strategy, or will there be a recognition first that “the tool kit international donors bring to post-conflict transitions was not designed for peace building”?42

Five Emerging Lessons The experience of peace implementation says far more about economic reasons for failure or severely handicapped success than about economic aspects of success. The overall record points clearly to five emerging lessons: the need for broad-based impact assessments; the critical importance of an early emphasis on employment; the need for institution-building and public-sector strengthening; the political influence of sovereign lending on the peace process; and the economic distortions introduced by the international presence. Measuring Impact A search for evidence about the relative contribution of economic factors and assistance to the implementation of peace agreements reveals the limited character of most assessments. The overwhelming lesson is that these assessments rarely examine overall impact. The idea that success should be judged in terms of the political mission to establish sustainable peace tends to be lost in a world of symptomatic monitoring and a priori assumptions about the effects of a particular strategy, policy, or program. Both in-house and independent evaluations perform audits of recordkeeping and financial transactions within the particular donor organization and its subcontractors, such as NGOs, and ascertain whether a particular aid project or program was implemented as planned. But “aid flows them-

S-7

11/24/09

4:20 PM

200

Page 200

EVALUATING IMPLEMENTATION TASKS

selves tell us little about the impact of external assistance . . . performance is often evaluated according to success in transferring funds, rather than the appropriateness of the design of the aid—and its likely impact on recipient populations.”43 Similarly, “project-by-project assessments often do not provide the broader picture [and] delays as long as a year or more have reduced the usefulness of post-conflict completion reports.”44 For bilateral donors whose aid portfolio may be composed of a large number of small projects, often administered by autonomous NGOs, the problem of impact assessment is magnified many times.45 The World Bank evaluation of 1997–1998 recommends that postconflict projects need a “process” rather than “blueprint” design,46 but process is far less amenable to standard evaluation techniques. A more serious problem is that donors have no reason to evaluate their impact on goals they never had in the first place, as the forthright Norwegian evaluation for the Mozambican case suggests: “Norwegian authorities gave little systematic attention to the links between supporting the peace process in the short run and sustaining the peace in the long run.”47 Because the effect of developmental assistance is intended to be longterm, there is also often little way to assess its positive contribution to the first years of a peace agreement. Short-term negative effects are excused as necessary for long-term sustainability on the basis of economic philosophy, not on the determinants of peacebuilding. Likewise, the longer-term effects of projects designed for the first years after war are also rarely assessed, with the consequence that the opportunity cost of wasted investments is not calculated. It is as if “quick impact projects” are never intended to have longer-term benefits. Vague benchmarks of progress substitute for analysis. The result in both cases is to provide little basis on which to propose midcourse adjustments in long-term programs or corrections to the method of planning and choosing immediate postwar projects and priorities. It is even difficult to gain a clear sense of the total amount of aid provided, so as to evaluate aid’s contribution to the mission. In their pledges of assistance, many donors repackage funds previously committed or list amounts they are already contributing to their standing obligations so as to appear more generous than they are. Most peace missions have either no database of donors and projects or multiple databases, each with different assumptions and criteria for recording those data.48 The structure of much assistance, by which each donor insists on controlling its own accounts and projects and on channeling funds through a foreign intermediary, creates multiple budgetary systems that are difficult to capture in one snapshot. Even breakdowns of global flows that distinguish aid money that remains within the country from money to repay former debt (which does not remain within the country), or sums spent to facilitate refugee return that go to relieve the costs of their upkeep in host countries rather than cash and materials allocated to the communities of return, are hard to come by.

S-7

11/24/09

4:20 PM

Page 201

ECONOMIC PRIORITIES FOR SUCCESSFUL PEACE IMPLEMENTATION

201

In 1999, the World Bank assigned to its Post-Conflict Unit the task of preparing quarterly monitoring reports on countries and regions affected by conflict, to a Post-Conflict Management Steering Group the task of reviewing these reports, to country desks the task of conducting Country Portfolio Performance Reviews every twelve months or less for postconflict countries (instead of the normal twelve to eighteen months), and to the OED the task of developing guidelines “on how to apply evaluation criteria with greater sensitivity to the post-conflict political and economic environment when conducting completion reports.” Watching briefs are now required for every country emerging from conflict.49 It is too early to say whether these innovations will address the relationship between a peace process and economic assistance strategies of the IFIs. Social surveys now ordered by the Bank in postconflict countries on a regular basis do not measure that relationship, while the emphasis on country-specific assessments, in place of a standardized system for monitoring and reporting impact, may work against drawing generalizable inferences on which new strategy and policies could be developed. One thing is clear: systematic monitoring and assessments of actual impact are sorely neglected, and without research designed to assess how aid contributes to peace, refinements of existing practice make little sense. Employment: The Most Obvious but Most Neglected Lesson A common problem for all cases of peace implementation is high unemployment that affects a large majority of the population in the first years after war. Lessons drawn about success or failure all point to the lack of employment opportunities. For example, the poor records on demobilization and reintegration into civilian life of soldiers and on the return of refugees and the internally displaced to their homes, both key aspects of all peace agreements, turn on the lack of jobs. Conversely, where partial success is claimed on demobilization, as in Mozambique, analysis points to specific funds designed for employment. High unemployment is a clear threat to peace, whether through disillusionment, lack of alternative activity and status to war, or the continued availability of the unemployed for mobilization by spoilers. The critical role of active employment in redirecting behavior and commitments toward peace is so obvious that no one disputes its importance. Yet economic strategies are not aimed to overcome the problem. Neither the IMF approach to macroeconomic stabilization nor the World Bank emphasis on large-scale infrastructure promotes employment. Development assistance and advice are still focused on laying the basis for economic growth in the long run, and assume that employment will naturally follow. This is surprising in the case of the World Bank, which has long

S-7

11/24/09

4:20 PM

202

Page 202

EVALUATING IMPLEMENTATION TASKS

come to view SMEs as a core component of development, particularly to promote employment, and it would seem to be even more central in promoting peace. Only by 1999 did Bank officials draw this lesson from Bosnia-Herzegovina, proposing earlier and greater attention in a peace process to SMEs (for example, in revising their strategy for Kosovo, which was then in its first stages). At the same time, only twelve donors for Bosnia had committed funds for emergency projects designed to create jobs. But even here they were slow to deliver: only $146 million was committed (out of more than $5 billion) and only $116 million was actually disbursed, leading the Bank to conclude that the “primary obstacle” to reducing Bosnia’s “painfully high” unemployment was “lack of credit.”50 Alternatively, the problem of employment may be the consequence not of policy but of sequencing. For example, in the case of Sierra Leone, the reduction of resources to subsidize the army led to rapid cuts in pay and personnel, with the result that 8,000 newly unemployed soldiers defected to the guerrillas and the peace was lost. Arguably, had there been more coordination between the economic and political halves of the mission, such “mis-timing” might not have occurred. But if insistence on early cuts in public expenditures to establish macroeconomic stability always wins the day, and is not accompanied by explicit attention to alternative employment, it is difficult to see what benefit coordination would bring. Equally costly is the reluctance—and downright refusal, for some donors—to finance core and recurrent budgets. This is costly because a peace agreement is, more than anything, about the transformation of wartime political structures or re-creation of government, the provision of public services, and a restoration of trust in public institutions. Government and other public institutions, however, need staff. Without monies to finance the salaries of public officials, police, judges, teachers, doctors, and others who will actually restore basic public order, the most fundamental tasks will not occur. Initial reconstruction of schools and clinics is a waste if they remain empty for lack of recurrent costs for staffing and operation, a common experience. This donor reluctance goes beyond the generally recognized problem of the IFI’s priority in economic strategy on macroeconomic balance through cutting public expenditures and, therefore, publicsector employment. Furthermore, donors to peace missions overwhelmingly favor technical assistance so as to avoid making political judgments in the sensitive early stages of peace implementation. Such assistance pays salaries to their own nationals to train and advise locals, which further reduces the possibilities for local employment.51 Donor preference for visible showcase projects also tends to neglect the employment aspect—that people are needed to make these projects operate, whether schools, hospitals, or factories. This requires monies for local salaries until there is local purchas-

S-7

11/24/09

4:20 PM

Page 203

ECONOMIC PRIORITIES FOR SUCCESSFUL PEACE IMPLEMENTATION

203

ing power sufficient to make World Bank cost-recovery policies actually work. Finally, because people do have to find ways to survive, and relief is never sufficient for those who obtain it or fully comprehensive, the slow growth of paid employment also leaves many with little option but to turn to illegal or informal sources of earnings. The inevitable result is to create new problems—crime, patronage, and corruption—that undermine the rule of law and are particularly difficult to root out later. Informalization also slows monetization and other aspects of economic normalization, delaying the effectiveness of IFI policies, and reduces the state’s tax base further at exactly the moment when social trust and public confidence in government and the future are at a premium. Institution-Building and Budgetary Support Beyond demilitarization, a peace agreement must construct a new political order—government, a public sector and institutions, and public services. As many peace implementers attest, the primary lesson is “politics first.” By the late 1990s, based largely on the experience of market transitions in formerly communist or developmentalist states, major donors including the IFIs did begin to stress the importance of institutions and “good governance” to successful implementation of macroeconomic stabilization and structural reforms. Recently, the World Bank has taken this aspect even further by emphasizing the importance of social trust to the creation and success of institutions. The problem in the case of peace implementation is that macroeconomic stabilization is a result of certain monetary and fiscal policies and their effective implementation. That is, it cannot even begin, let alone become sustainable, if there is no government to adopt those policies or proper financial and legal institutions to make markets work and enforce contracts, property rights, and monetary and fiscal targets. Moreover, social capital is acquired. It is a response to functioning, trustworthy institutions, and a decline in social trust results from nonfunctioning or poorly functioning public institutions. Even the ability of a postconflict country to use the aid monies and technical assistance provided—called its absorption capacity—depends on functioning institutions. Similarly, a major step by the World Bank in 1998 in the direction of new strategy, its proposal to precede the preparation of a Country Assistance Strategy with a Transitional Support Strategy that would be prepared “as soon as resolution is in sight,” would require Bank staff to “collaborate with the government and other partners to prepare a national recovery program as an initial step toward a more comprehensive, fullscale reconstruction program.”52 The Paris Peace Agreement on Cambodia

S-7

11/24/09

4:20 PM

204

Page 204

EVALUATING IMPLEMENTATION TASKS

in 1991 mandated that the Cambodians do just that, take the lead in determining their own recovery needs and formulating a national recovery plan. But its formulation had to await the formation of a postwar government, and delays in the latter meant delays in everything else.53 Despite these lessons learned during the 1990s and what appears to be a new consensus, actual assistance strategies for economic development and peacebuilding continue to be based on the assumption that governments and economic institutions already exist—at most, they may need reform, but not wholesale construction. While the IMF and the World Bank appear to be open to those who say that macroeconomic stability is important but not sufficient, they do not appear yet to have accepted that institutions must be built first, or at least simultaneously. The result is usually to place this power in the hands of interim bodies or simply individual officials. But the type of macroeconomic stabilization policy chosen is one of the most consequential decisions that a government can take, with huge political consequences for the quality and durability of the peace. Because of the political consequences, in fact, it is becoming increasingly common to rely on an IMF-managed currency board in the first years after the peace agreement, as was done in Bosnia-Herzegovina, so as to deprive local authorities of any policy discretion. This goes counter to the IMF’s own better judgment, the very narrow conditions in which currency boards are known to be effective, and the advice of most economists with an understanding of the political and social environment in which economies function. The World Bank did move in 1996 to accept within its development mandate in postconflict countries some tasks that it had previously considered “political” (and, by some interpretations, prohibited by its charter)— specifically, de-mining, demobilization and reintegration, public-expenditure realignment, and human and social capital. In its 1996–2000 Priority Reconstruction and Recovery Program (PRRP) to support the Dayton Peace Accords on Bosnia-Herzegovina, which has been the most innovative and well-endowed example of postconflict engagement in the Bank’s history, it even classifies some projects as “peace implementation activities,” specifically those relating to the mass media, elections, human rights, and local security, including local police reform and a national border service. Nonetheless, by the end of 1998, after three years’ operation of the $5.1 billion PRRP, only $320 million had been pledged for these activities and only $217 million had been disbursed, amounting to about 8 percent of total disbursements of the PRRP.54 In 1999 the Bank’s assessment was that “institutional and policy reform has generally lagged behind physical reconstruction.”55 It is also in the Bosnia program where the importance of budgetary support for recurrent costs was first accepted.56 Nonetheless, even there, the Bank acknowledged that “emergency programs often failed to take into account the need for sufficient funding in recurrent costs.”57

S-7

11/24/09

4:20 PM

Page 205

ECONOMIC PRIORITIES FOR SUCCESSFUL PEACE IMPLEMENTATION

205

The importance to peace implementation of the development and maintenance of the public sector, both for critical institutional development and to give all parties a stake in reconciliation, is so great that some have proposed that all external aid be viewed as public-expenditure support. Rather than multiple, off-budget sources of reconstruction and capacity-building, the aim should be to strengthen the public sector.58 This would imply a major change in the way aid is delivered to peace processes. At the very least, others say, the focus should not be on how to control budgetary expenditures but on how to generate new budget revenues.59 The initial lack of institutions should be taken seriously in policy design as well, for example, to the kind of taxation that does not require sophisticated systems of collection and revenue management as are required for income taxes. Countries should be urged, argues Vito Tanzi of the IMF, to begin simply, say, with excise taxes, which are the easiest to raise, and not to add complexity until state agencies are able to manage that complexity.60 Finally, the general absence of funding for key political tasks of peace implementation, such as police, judicial reform, civil administration, and democratization beyond elections, has been a particular grievance among the special representatives of the Secretary-General (SRSGs). To a person, they argue that the most important innovation in peacebuilding practice would be to ensure a discretionary budget available to the SRSG to finance political goals as they arise. SRSG Aldo Ajello attributes his greatest successes in implementing the Mozambican peace agreement to the $17.5 million trust fund he managed to obtain and use to transform the Mozambique National Resistance (Renamo) from an army into a political party even though it allowed little discretion.61 According to Derek Boothby, deputy SRSG in UNTAES, implementing the Erdut Agreement on Eastern Slavonia, Croatia, a true discretionary budget amounting to perhaps 2 percent of the total budget, would have made all the difference in what the mission could accomplish.62 The lesson may be getting through. Bilateral donors (particularly the UK’s DFiD) did provide a trust fund for the UN Mission in Kosovo (UNMIK) in 1999, and the World Bank set one up for the UN Transitional Administration in East Timor (UNTAET) the same year. Assistance to Whom? The distinction between political aspects of a peace mission and economic aspects is hardwired into thinking about peacebuilding and into analyses of the role that economic actors and assistance do and should play. In fact, the putatively economic actions of donors are highly political, with independent effects on implementation. The most obvious effect is from decisions about whom to assist. While the idea that neutrality is a myth is gaining ground in the humanitarian

S-7

11/24/09

4:20 PM

206

Page 206

EVALUATING IMPLEMENTATION TASKS

community, economic assistance is still constrained either by the norm of sovereignty governing intervention or by the rules of sovereign lending that now dominate the international monetary system. Whether giving relief or loans, donors need sovereign partners. This has several consequences. Humanitarian organizations and bilateral donors empower local actors by their decision of whom to aid, whether the criterion is who appears most like a government or which political parties and leaders they want to support and strengthen politically. The IFIs must lend to governments, which will be held responsible for meeting loan conditions and eventually for repayment. But the search for sovereign partners, or the delay while governments get established, introduces substantial delays into the delivery of assistance and continues humanitarian relief long beyond its constructive role. To avoid these two problems, there is a growing resort to limit aid to technical assistance, which is seen as apolitical (in the sense that it appears not to be openly supporting one faction over another), but which takes the form of salaries to expatriate advisers, not to local actors and projects. Similarly, foreign donors increasingly seek out international NGOs to implement projects where they do not trust either the politics or the capacity of local governments. The result, again, is that a large proportion of the actual assistance goes to salaries and overhead of the international NGOs, not to locals in need of assistance. But in all cases, the decisions about whom to finance, whom to establish an aid relationship with, and whose preferences to consult have direct influence on the peace process and the success of implementing the provisions of a peace agreement. A second set of effects results from the fact that money is power and, aside perhaps from infrastructural finance, is intended to influence behavior. Third-party assistance in implementation is a set of principal-agent relationships in which external actors seek not only to assist parties to a peace agreement but also to gain their compliance. A primary instrument of leverage over parties’ behavior is through offers of assistance (or threats to withhold assistance), and no aid is given without some conditions attached. The ongoing debate about conditionality—to what extent and in what way aid should be used as leverage over parties’ behavior—is based more on assumptions about influence than on studies of actual effects. No one doubts that there are effects, however. The primary criticism of the policies of the IFIs grows out of their use of economic conditionality—conditions such as agreed levels of wages, prices, budget deficits, and sectoral expenditures (such as on the military or on pensions) that must be met to receive each distribution of a loan— because of the limits they impose on political flexibility, choice, and expenditures that may be necessary to building peace. A second criticism comes from those who believe that economic assistance should be used as political conditionality, to punish or reward political leaders for noncompliance or compliance with provisions of their signed agreement, which is in conflict

S-7

11/24/09

4:20 PM

Page 207

ECONOMIC PRIORITIES FOR SUCCESSFUL PEACE IMPLEMENTATION

207

with economic or loan conditionality. The IFIs generally oppose political conditionality for that reason, namely, that it is in violation of their legal charters, their economic philosophy, the conditionality they impose, and the efficiency and effectiveness of development plans. Some advocates go further in calling for peace conditionality, through which the enormous leverage of the IFIs and the sums that they or other major donors control are used primarily to nurture constituencies for peace by rewarding supporters, punishing opponents, encouraging vacillators, and holding parties specifically accountable to the terms of the agreement they signed.63 In addition, there is major disagreement about the best strategy for dealing with potential spoilers.64 Should they be co-opted into the peace process, as UNTAC attempted to do with the Khmer Rouge and the Lomé Accords attempted to do with Foday Sankoh in Sierra Leone, or should they be isolated, as Western policy over the Dayton implementation chose to do with two Bosnian Serb political parties and their leaders (the Serb Democratic Party and the Serb Radical Party)? Even the identification of potential spoilers is not immune from the political preferences of major external actors and donors. But even economic conditionality requires political choices. Leverage over the compliance of governments by the IFIs, particularly of the World Bank, is limited by their market relationship with borrowing governments. They cannot lend if governments are not willing to sign off on project loans that are proposed, while to succeed in a loan program, they need to maintain good relations with the government. The danger of cozy relations—including a government’s refusal to borrow for projects that might aid its rivals or former enemies, to act in ways that would harm its constituency, and to escape too much scrutiny over corrupt practices—is one of the motivations behind those who call for explicit peace conditionality with IFI financing. A third political effect of economic assistance is the choice of political tactics by third-party implementers that result from what donors are willing or not willing to fund in the first years. For example, the tendency to focus their tactics on spoilers may result from the lack of donor attention to institutional capacity, although it may be a far better explanation of a leader’s apparent refusal to comply than any strategic choice to obstruct an agreement, as the concept of spoilers assumes. If local institutional capacity is not an explicit target of assistance in the first year or two of a peace agreement, a leader’s refusal to comply with specific aspects of the peace agreement may not be a choice. Whereas a peace mission may be present for several years—say, between two and five—a leader calculates his political survival over a longer period of time. Any admission that he does not have sufficient power and authority to implement a task and obtain compliance from his followers or citizens is a far riskier political step than appearing strong by refusing compliance with outsiders. Similarly, doubts about the commitment of parties to a peace agreement have led donors and peace

S-7

11/24/09

4:20 PM

208

Page 208

EVALUATING IMPLEMENTATION TASKS

missions increasingly to emphasize bottom-up peacebuilding instead of top-down, that is, projects that are part of a loan package to governments but that emphasize local communities, NGOs, and popular participation as ways of circumventing national leaders. One effect, however, is to compound the problem of weak capacity at the center and to delay the “ownership” that outsiders demand. Another is that punishment—isolation, withholding aid, or circumvention—actually reduces one’s ability to influence behavior. Distortions Induced by the International Presence The very presence of an international peace mission, military forces, and aid agencies has economic consequences that are directly contrary to the political goals of self-governance and economic and political sustainability. This problem is rarely discussed, but it emerges vividly in every case study for this project. The first set of distortions are those introduced into local labor, housing, and retail markets through the high salaries paid to expatriates and international civil servants, the rapid jump in demand for local translators and drivers, and the demand for local housing and services attuned to foreign tastes and salaries. Because the new peacetime governments cannot compete with the international organizations in the salaries offered, they find it difficult to attract the best professionals to government employment and public services, which delays local capacity- and institution-building further. In time, many locals employed as translators and drivers progressively lose their skills because they are not practicing their profession. The high prices paid for housing and food crowd out locals, while internationals do not consume enough locally to stimulate and sustain businesses. A second set of distortions comes from the interaction between international and local actors in the peace process. The visibility of expatriates in their white foreign vehicles, eating at expensive restaurants reoriented to foreign clients, and buying in privileged, subsidized stores generates greater local resentment over economic hardships. The relative cost of an international presence—for example, $1 million a day for the UN Operation in Mozambique (ONUMOZ), $4–5 billion a year for the Dayton implementation in Bosnia-Herzegovina for the military deployment alone—provokes doubts about priorities when evidence regarding success tends to emphasize the importance of local initiative. For example, in Mozambique, the international mission concluded that where locals took on the tasks of return, reconciliation, and reconstruction themselves, the result was more durable. Robert Chavez, resident representative of the World Bank in Mozambique during ONUMOZ, warned that “relief agencies need to be more sensitive to the contribution of society to reconstruction and not try to impose too much order on the process.”65 According to a World Bank

S-7

11/24/09

4:20 PM

Page 209

ECONOMIC PRIORITIES FOR SUCCESSFUL PEACE IMPLEMENTATION

209

assessment of Bosnia-Herzegovina in early 1999, “implementation of reconstruction programs has been most effective in those sectors (e.g. in transport and energy) where priorities of donor assistance have been established jointly with the authorities.”66 And while much attention in peace missions is now drawn to the problem of avoiding aid dependency, or a “culture of dependency,” there is also a risk of distortion from the timing and manner of the mission’s departure. The precipitous departure of ONUMOZ in Mozambique—where a “resource-intensive, high-profile operation engulfed Mozambique for two years and then left as swiftly as it had come”—is much criticized for its lack of regard for the effects of the rapid hand-over to locals.67 Although scholars of peacekeeping pay much attention to “exit strategies,” their criteria for exit (aside from political pressures for disengagement in troopcontributing countries) tend to be measures of progress along a list of benchmarks in implementing the peace agreement. The effect on that path of the exit itself, in its timing and its manner, does not enter calculations.

Conclusion: The Need for a New Economic Strategy for Peace Implementation The economic strategy behind the implementation of peace agreements has been adapted from other purposes, and not designed explicitly for ending civil wars. Increasing criticism of the consequences, particularly the conflict between the priorities and tactics of the political mission and those of the economic actors, during the second half of the 1990s led to substantial efforts by donors to seek lessons and make improvements. The primary result has been to add new financing instruments for peace-related activities, such as for elections, police reform, and de-mining, and supplementary projects such as arms-for-land or trust funds for core and recurring budgetary expenditures.68 New financial facilities also aimed at gaining flexibility on IMF restrictions, while increased coordination among donors in the field and between the main officials in the economic and political tracks became the common denominator of reform. None of these responses addresses the differences in approach, priorities, and tasks that do or do not get funded. The fact that so little evaluation takes place of the impact of assistance on peace tasks makes it very difficult to demonstrate why and how these reforms misunderstand the problem. Moreover, the selective interest of outsiders in postconflict cases means that even these remedies will not be available in many countries. There is, however, substantial fragmentary evidence from actual cases of what does not work and what needs to be changed. This evidence does not point toward larger resources in most cases, but toward the need for a strategy appropriate to the task that would, inter alia, promote far more efficient use

S-7

11/24/09

4:20 PM

210

Page 210

EVALUATING IMPLEMENTATION TASKS

of resources and greater likelihood of success. Indeed, the examples of Bosnia-Herzegovina and Palestine suggest that too much money and too assertive an international presence may be counterproductive, just as the lessons drawn from the 1996 World Bank study of Bank-financed demobilization and reintegration programs in Africa were that low-cost solutions may be more effective than costly interventions, if one is sufficiently attuned to local context and culture. 69 But as James Boyce and Manuel Pastor write, “unless the peace process is allowed to reshape economic policy, both will fail.”70 The vested interests in technical knowledge and in organizational autonomies and jurisdictions appear to be standing in the way of the substantial rethinking and collaborative design that would integrate both political and economic aspects into a peace strategy, building on the actual experience of peace missions and their lessons such as the five discussed above. The high proportion of failures (or political stalemates and worsening economic conditions) in current experience would seem to demand as much.71

Notes 1. For example, the World Bank’s Post-Conflict Unit, the Emergency Response Division (ERD) of the UN Development Programme (UNDP), and the Development Assistance Committee (DAC) of the Organization for Economic Cooperation and Development (OECD), as well as the U.S. Agency for International Development (USAID), the UK’s Department for International Development (DFiD), and the international development agencies of Canada (CIDA) and Sweden (SIDA), among others. 2. See Chapter 14 on El Salvador by Charles Call in this book, and Lise M. Howard, “Namibia,” draft prepared for this project. 3. Michael W. Doyle, UN Peacekeeping in Cambodia: UNTAC’s Civil Mandate (Boulder, Colo.: Lynne Rienner, 1995), pp. 49–51. 4. Shepard Forman and Stewart Patrick, introduction to Forman and Patrick, eds., Good Intentions: Pledges of Aid for Postconflict Recovery (Boulder, Colo.: Lynne Rienner, 2000), p. 30. 5. Jonathan Stevenson, Preventing Conflict: The Role of the Bretton Woods Institutions, Adelphi Paper no. 336 (London: IISS, 2000), p. 68. 6. The best-known such study, Alvaro de Soto and Graciana del Castillo, “Obstacles to Peacebuilding in El Salvador,” Foreign Policy 94 (spring 1994): 69–83, identifies the difference in mandate and approach between those normally assuming the responsibility for peacebuilding, such as the United Nations or regional security organizations, and those normally given the task of economic reconstruction, above all the international financial institutions and regional development banks. 7. Martin Raiser, “Informal Institutions, Social Capital, and Economic Transition: Reflections on a Neglected Dimension,” Working Paper no. 25 (Washington, D.C.: World Bank, 1997); and Nat J. Coletta and Michelle L. Cullen, Violent Conflict and the Transformation of Social Capital: Lessons from Cambodia, Rwanda, Guatemala, and Somalia (Washington, D.C.: World Bank, 2000). 8. In her comments on an earlier version of this chapter made at the second

S-7

11/24/09

4:20 PM

Page 211

ECONOMIC PRIORITIES FOR SUCCESSFUL PEACE IMPLEMENTATION

211

conference of this project, at Pocantico, New York, February 28, 1998 (emphasis in original). 9. For an analysis of the slow delivery of pledged aid, see Forman and Patrick, Good Intentions. 10. On the importance of the distribution issue, see Elizabeth J. Wood, Forging Democracy from Below: Insurgent Transition in South Africa and El Salvador (Cambridge: Cambridge University Press, 2001). 11. Frances Stewart, “The Root Causes of Humanitarian Emergencies,” in E. Wayne Nafziger, Frances Stewart, and Raimo Varynen, eds., War, Hunger, and Displacement: The Origins of Humanitarian Emergencies, vol. 1 (Oxford: Oxford University Press, 2000). 12. On the security gap, see Chapter 11 by Charles Call and William Stanley in this book. On human rights, see Chapter 9 by Tonya Putnam in this book. 13. Derek Boothby, commenting on an early version of Michael Doyle’s chapter on transitional authority from his experience as deputy SRSG in UNTAES, at the second conference of this project at Pocantico, New York, February 28, 1998. 14. On demobilization, see Chapter 6 by Joanna Spear in this book. On refugee return, see Chapter 10 by Howard Adelman in this book. 15. Notably, “the majority of countries in arrears to the Bank are countries in conflict.” The World Bank’s Experience with Post-Conflict Reconstruction (Washington, D.C.: World Bank Operations Evaluation Department, 1998), p. 8. 16. An analysis of this innovation can be found in Centre for Defence Studies, Evaluation of the Western Bosnia Rehabilitation Programme 1996–1998 (King’s College, London: Centre for Defence Studies, November 25, 1999), a report prepared for the Department for International Development, the primary authors of which were Susan L. Woodward and Michael Clarke. 17. Sven Sandstrom, managing director of the World Bank, address to the Peace Implementation Network (PIN) Forum on “Public-Sector Finance in PostConflict Situations,” Washington, D.C., August 1999. 18. The World Bank’s Experience, p. 5, esp. para. 2.1. 19. The locus classicus of this literature is de Soto and del Castillo, “Obstacles to Peacebuilding in El Salvador.” 20. Stevenson, Preventing Conflict, p. 51. 21. On the postcommunist transitions, Martin Wolf, citing a study by Joel Hellman, Geraint Jones, and Daniel Kaufmann, “Seize the State, Seize the Day,” World Bank Policy Research Working Paper no. 2444, argues that privatization in the absence of appropriate institutional conditions leads to “insider privatisation” that “can be worthwhile for those who [are thereby enriched] but do great damage to the economy as a whole,” as it creates vested interests with a strong incentive to prevent further reform. “To have the right sort of market-oriented democratic society, one needs the right sort of state.” See “Avoiding the Trap of Transition,” Financial Times, October 11, 2000, p. 25. On Africa, see Michael Holman, “Rethink Urged in Africa’s Strategy for Development,” Financial Times, August 1, 2000, p. 5, reporting on a study by the UN Conference on Trade and Development (UNCTAD), Capital Flows and Growth in Africa, www.unctad.org. 22. See David Malone and Mats Berdal, eds., Greed and Grievance: Economic Agendas in Civil Wars (Boulder, Colo.: Lynne Rienner, 2000), and work by Mark Duffield and by David Keen—for example, their contributions to Joanna Macrae and Anthony Zwi, eds., War and Hunger: Rethinking International Responses to Complex Emergencies (London: Zed Books, 1994), and David Keen, The Economic Functions of Violence in Civil Wars, Adelphi Paper no. 320 (Oxford: Oxford University Press, 1998). 23. The World Bank’s Experience, p. xv (emphasis in original).

S-7

11/24/09

4:20 PM

212

Page 212

EVALUATING IMPLEMENTATION TASKS

24. John Eatwell, Michael Ellman, Mats Karlsson, D. Mario Nuti, and Judith Shapiro, Hard Budgets and Soft States: Social Policy in Central and Eastern Europe (London: Institute for Public Policy Research, 1999), pp. 20–21. 25. A 1998 internal World Bank review of experience with postconflict operations maintains: “The Board has clearly stated that peacemaking and peacekeeping fall under the mandate of the UN and are not responsibilities of the Bank.” The World Bank Experience, p. 22. 26. Reflecting evidence from postconflict countries, Nat Coletta, former head of the World Bank’s Post-Conflict Unit, stated that “in general, privatization is proposed too early in post-conflict situations, before business infrastructure exists.” Comments made at the PIN Forum on “Public-Sector Finance in Post-Conflict Situations.” 27. The World Bank’s Experience, p. xvi. 28. For example, an external evaluation of Norwegian Assistance to Mozambique, where Norway was a major donor, noted: “The macro-economic context for peace and reconstruction was shaped by the Bretton Woods institutions and seems to have been largely accepted as a given in the development of Norwegian aid strategies toward Mozambique.” Its “aid portfolio was extremely fragmented . . . with some 900 aid disbursements, many of them very small.” Moreover, “four and a half years after the peace agreement was signed, the Ministry and NORAD [Norwegian Agency for Development Cooperation] still had not articulated an aid strategy that explicitly confronted issues of sustaining the peace.” The evaluation summarizes a more general problem elegantly: “In practical terms, Norwegian policy tended to fold the peace process into the aid programme rather than vice versa.” See Alistar Hallam, Kate Halvorsen, Janne Lexow, Armindo Miranda, Pamela Rebelo, and Astri Suhrke (project leader), Chr. Michelsen Institute in association with Nordic Consulting Group, Evaluation of Norwegian Assistance to Peace, Reconciliation and Rehabilitation in Mozambique (Oslo: Royal Ministry of Foreign Affairs, May 1997), pp. x–xii. 29. De Soto and del Castillo, “Obstacles to Peacebuilding in El Salvador.” 30. The World Bank’s Experience, p. ix. 31. Ibid., p. 33. 32. Ibid., p. xv. 33. Ibid., pp. 33–35. 34. Stevenson, Preventing Conflict, pp. 58–60. 35. The OED evaluation of Bank experience suggests that “BiH [BosniaHerzegovina] and the West Bank and Gaza may have set precedents for the establishment of trust funds for post-conflict reconstruction activities,” but also admits that “the final report of the Framework [for World Bank Involvement in PostConflict Reconstruction of April 1997] does not provide much detail on budgetary arrangements and financing instruments for post-conflict reconstruction, indicating that this issue warrants further discussion.” The World Bank’s Experience, p. 8. 36. Stevenson, Preventing Conflict, pp. 12–13. 37. De Soto and del Castillo, “Obstacles to Peacebuilding in El Salvador.” 38. In some cases, World Bank staff can be influential, as Kemal Dervi, director in 1995 of the Central Europe Department within the Europe and Central Asia Region of the World Bank, was in helping to mobilize the Bank to plan a program for Bosnia-Herzegovina even before the end of war and the peace negotiations. 39. The political consequences are particularly striking in the example of El Salvador, which decided nine years after the peace agreement of 1992, on January 1, 2001, to adopt dollarization—taking advantage of its heavy dependence on remittances but also accepting the resulting constraints on domestic economic and social

S-7

11/24/09

4:20 PM

Page 213

ECONOMIC PRIORITIES FOR SUCCESSFUL PEACE IMPLEMENTATION

213

policy. See Andrew Bounds, “Salvadoreans Embrace the Dollar,” Financial Times, December 29, 2000, p. 2. 40. Marika Fahlen, “Mobilisation and Allocation of Resources for PostConflict Assistance and Identification of Appropriate Modalities for Implementation,” note by the Delegation of Sweden submitted for discussion at the seventh meeting of the DAC Task Force on Conflict, Peace, and Development Cooperation, January 9–10, 1997 (Paris: Development Cooperation Directorate of the Organization for Economic Cooperation and Development, DCD (97)1/ANN1, OLIS, December 20, 1996), p. 5. 41. Forman and Patrick, introduction to Good Intentions, p. 13; see also their summary of the findings of Good Intentions and their proposal for the creation of such a mechanism, which they call a Strategic Recovery Facility, in Shepard Forman, Stewart Patrick, and Dirk Salomons, Recovering from Conflict: Strategy for an International Response (New York: Center on International Cooperation, New York University, 2000). 42. James K. Boyce, “Beyond Good Intentions: External Assistance and Peace Building,” in Forman and Patrick, Good Intentions, p. 369. 43. Forman and Patrick, introduction to Good Intentions, p. 30 (emphasis in original). 44. This particular criticism is made by the World Bank Operations Evaluation Department of its assessments, in The World Bank’s Experience, p. xvii. 45. Here too, “the autonomy and diversity of NGO operations . . . made it difficult for the Embassy to assess the cumulative impact of the Norwegian contribution and extract political mileage accordingly.” Evaluation of Norwegian Assistance, p. xiii. 46. The World Bank’s Experience, p. 40. 47. Evaluation of Norwegian Assistance, p. xi. 48. On this problem, see in particular the discussion in Zlatko Hurtic, ´ Amela ˇ cˇ anin, and Susan L. Woodward, “Bosnia and Herzegovina,” in Forman and Sap Patrick, Good Intentions, pp. 328–337. 49. The World Bank Experience, p. 47; Stevenson, Preventing Conflict, p. 61. 50. Bosnia and Herzegovina: 1996–1998; Lessons and Accomplishments; Review of the Priority Reconstruction Program and Looking Ahead: Towards Sustainable Economic Development, report prepared for the May 1999 Donors Conference Co-Hosted by the European Commission and the World Bank, esp. pp. 14–15. 51. For example, one-third of all aid to Mozambique was technical assistance to expatriate professionals (Roberto Chavez at the PIN Forum on “Public-Sector Finance in Post-Conflict Situations). 52. The World Bank’s Experience, p. 6. 53. Forman and Patrick, introduction to Good Intentions, p. 17. 54. Bosnia and Herzegovina: 1996–1998, p. 10. 55. Ibid., p. 6. 56. Similar measures were also applied in the West Bank and Gaza (the Johan Jørgen Holst Fund for Start-up and Recurrent Costs) and other budgetary assistance. See Barbara Balaj, Ishac Diwan, and Bernard Philippe, “External Assistance to the Palestinians,” Politique Étrangère (autumn 1995), for the various attempts to fund their unexpectedly large budget deficit and the problems this caused. 57. Ibid. 58. Shanta Devarajan from the World Bank, at the PIN Forum on “PublicSector Finance in Post-Conflict Situations.” 59. Odin Knudsen, on the basis of his experience as resident director for the

S-7

11/24/09

4:20 PM

214

Page 214

EVALUATING IMPLEMENTATION TASKS

World Bank in West Bank and Gaza, in comments made to the discussion at the second conference of this project at Pocantico, New York, February 28 and March 1, 1998. 60. In remarks made to the PIN Forum on “Public Sector Finance in PostConflict Situations.” 61. Joao Honwana, “Mozambique,” draft prepared for this project. 62. Derek Boothby in discussion at the second conference of this project at Pocantico, New York, February 28–March 1, 1998. 63. Boyce, “Beyond Good Intentions.” See also James K. Boyce and Manuel Pastor Jr., “Aid for Peace: Can International Financial Institutions Help Prevent Conflict?” World Policy Journal 15, no. 29 (summer 1998), pp. 42–49; and James K. Boyce, Peace Conditionality: External Assistance and Post-Conflict Transitions (forthcoming). 64. Stephen John Stedman, “Spoiler Problems in Peace Processes,” International Security 22, no. 2 (fall 1997): 5–53. 65. At the PIN Forum on “Public-Sector Finance in Post-Conflict Situations,” and at a seminar on Donor Coordination in Post-Conflict Countries at the Overseas Development Council, Washington, D.C., October 22, 1997. 66. Bosnia and Herzegovina: 1996–1998, p. 6. 67. Evaluation of Norwegian Assistance, p. xiv. 68. However, the World Bank, at least, has tended to underestimate the actual costs of the budgetary support that will be needed, such as occurred in El Salvador and in the West Bank and Gaza programs. 69. Stevenson, Preventing Conflict, p. 60. 70. In Krishna Kumar, ed., Rebuilding Societies After Civil War: Critical Roles for International Assistance (Boulder, Colo.: Lynne Rienner, 1997), p. 287 (emphasis in original). 71. As Nicole Ball warns, “Because peacebuilding activities are inherently political and are implemented in highly politicized environments, the selection, design, and implementation of programs cannot be approached from a purely technical perspective.” Nicole Ball with Tammy Halevy, Making Peace Work: The Role of the International Development Community, Policy Essay no. 18 (Washington, D.C.: Overseas Development Council, 1996), p. 48.

S-8

11/24/09

4:20 PM

Page 215

8 The Role of Postsettlement Elections TERRENCE LYONS

In nearly all cases of negotiated settlements of civil wars, elections have been designated in the peace accord as the mechanism for ending the transition from war to peace.1 As a crucial part of a peace process, such elections aim to terminate the civil war and contribute to sustainable peacebuilding. At the same time, postconflict elections attempt to instill democracy by enshrining new political institutions and rules of competition. For the international community, postsettlement elections serve as symbolic endpoints for the high-profile peacekeeping or peace implementation phase of their involvement. Christopher Clapham suggests that in the post–Cold War era the terms of a peace agreement have “to meet the moral expectations of those who mediated it, and who would in turn be required to guarantee its implementation.”2 In cases where civil war coincided with state collapse, the return of a “normal” recognized government allows international financial institutions and bilateral donors to begin making loans for reconstruction. Postsettlement elections therefore try to advance multiple and often contradictory goals both within the state in question and for the international community. Such elections carry a tremendous burden. They are called upon to settle the contentious issues of internal and external legitimacy and must be organized under the difficult circumstances of societal disorder, general insecurity, and institutional breakdown—conditions, according to comparative studies of democracy, that seldom favor democratic transition. 3 Yet there seem to be few options other than elections to organize the transition from war to peace. Moreover, as demonstrated in the case studies of this project, the ability of postsettlement elections to serve these multiple roles is mixed, with more successes than the literature on democracy would predict. 215

S-8

11/24/09

4:20 PM

216

Page 216

EVALUATING IMPLEMENTATION TASKS

I begin this chapter by acknowledging that postsettlement elections relate to multiple goals and that these goals may not all be equally possible in the short run. In particular, some elections may be able to further war termination but do little to advance democratization. I then address powersharing pacts and electoral systems designed to contribute to the goals of both war termination and democratization and argue that in the most difficult cases these efforts are unlikely to make a significant difference. I contend that instead of placing confidence in pacts and electoral systems, implementers can increase the chances that elections will both end a war and begin a transition toward democracy by “demilitarizing politics.” In particular, strengthening interim institutions that provide the context for postsettlement elections, building electoral administration that creates confidence that the vote will be managed effectively, and engaging in programs to transform militias into political parties all strengthen the ability of postsettlement elections to serve their multiple goals. Finally, I urge policymakers to recognize that in the most difficult cases, they must judge whether to support elections that may further war termination but bear little potential to promote democratization. In the most difficult cases the “least bad” policy outcome may be to accept postsettlement elections that end a war and create some degree of security and stability, and recognize that democratization must be pursued over a longer time frame.

The Multiple Goals of Postconflict Elections Postsettlement elections are designed to advance sometimes contradictory goals relating to war termination, democratization, and international intervention and legitimacy. These multiple goals operate on different time lines and require different preconditions. In some cases priorities among these important objectives must be established, at least in the short term. A postsettlement election succeeds or fails relative to war termination to the extent that it contributes to ending the fighting. Some elections have succeeded relative to this goal, most notably Zimbabwe, Mozambique, El Salvador, Namibia, and Nicaragua. Others perhaps have contributed to war termination in the short term but with less clear results with regard to longterm security, such as Liberia, Cambodia, Lebanon, and BosniaHerzegovina. In Angola, the 1992 elections sparked a renewal of violence and failed even with respect to the limited objective of war termination. War resumed following elections in Sierra Leone as well. Another set of goals relates to the process of democratization. As has been analyzed in cases of “breakthrough elections” following authoritarian rule, multiparty elections may mark the beginning of a new pattern of contesting politics. Postsettlement elections, like any transitional elections, have the purpose of serving as the first step in and foundation for a longer-

S-8

11/24/09

4:20 PM

Page 217

THE ROLE OF POSTSETTLEMENT ELECTIONS

217

term process of democratic consolidation through which new rules of the political game are institutionalized. While consolidation is a complex and long-term process, at least some postsettlement elections seem to have made considerable progress in laying a basis for democratic practice, most notably in El Salvador, Mozambique, and Nicaragua, where competitive elections have continued to take place after the first postsettlement election. In other cases subsequent elections have had an ambiguous relationship to democratization, as in Zimbabwe, Bosnia-Herzegovina, and Cambodia. In all cases analysts must wait until several rounds of elections and leadership transfers have taken place before reaching reliable conclusions regarding the vitality of the democratization process. The danger with regard to democratization is that postsettlement elections may entrench and provide electoral legitimacy to authoritarian parties who continue to rely upon coercion, fear, and chauvinism to remain in power. Leaders and political organizations who derive their power from the structures of war often remain powerful at the time of the elections and therefore win the vote. Elections under these circumstances therefore risk increasing the power of nondemocratic actors. This occurred in Liberia, in Bosnia-Herzegovina, and to an extent in Cambodia. In Angola, the postsettlement election did even more damage, creating a sense of cynicism about the democratic process and making it much more difficult to implement democratic institutions in the future. It is important to recognize, however, that in these cases the election did not cause these results but reflected either a conflict that had not been resolved in the negotiations or a poorly designed implementation process that failed to demilitarize politics prior to the vote. Even when long-term consolidation of democracy remains questionable, a postsettlement election may at least represent a return to constitutional rule, an important step in the most difficult cases where the state has collapsed. Constitutional restraints on power by no means are sufficient to prevent authoritarian rule, but they at least represent a set of norms against which rulers may be judged. In some cases an election that merely puts in place a constitutional government, despite questionable democratic credentials, may begin to create stability and serve as a first step in a longer-term agenda of democratization. An additional set of goals relates to the broader international community. In some cases, the inclusion of an election in a peace agreement was critical to convincing states to contribute troops for an implementation mission. The United States would have been extremely reluctant to commit to involvement in Bosnia-Herzegovina without elections serving as a convenient point to structure an (eventually unused) exit strategy. Similar imperatives were important to the UN force that entered Sierra Leone in 2000. In Liberia, the Economic Community of West African States Cease-Fire Monitoring Group (ECOMOG) insisted that elections take place in 1997 so

S-8

11/24/09

4:20 PM

218

Page 218

EVALUATING IMPLEMENTATION TASKS

that it could withdraw. The costs of a peace implementation force are high and the dangers are considerable. Without some event to mark the end of the transition and allay concerns regarding open-ended commitments, states are less likely to contribute peacekeeping forces. In the most difficult cases, prospects for either peace or democracy are minimal without such forces. Postsettlement elections also provide the context for major and regional powers to disengage from support to insurgent groups that no longer serve their interests. Disentangling external powers from local conflicts may be a critical step in resolving an internal war. The peace processes and elections in Cambodia and Nicaragua served this purpose. The elections in Liberia provided Nigeria with a mechanism to withdraw from its conflict with Charles Taylor and to end its support for other insurgent groups. As Susan L. Woodward points out in Chapter 7, the international financial institutions and important bilateral donors have great difficulty in designing projects with an interim regime that lacks a clear mandate. Lending institutions require a recognized government to serve as their partner in making large sums of capital available. The daunting tasks of postconflict reconstruction and the imperatives to provide jobs to consolidate the peace often remain on hold until the transition is “over.” Postconflict elections provide that endpoint and serve as a signal to lenders that they may begin to make binding loan commitments. These goals need not be complementary. For a postsettlement election to serve the purposes of democratization, for example, a long transition that provides sufficient time and appropriate institutions to encourage security and the creation of peacetime political and civil structures prior to voting may be necessary.4 In some cases, however, the nature of the conflict and the peace agreement requires quick elections in an atmosphere in which leaders continue to derive power from their control over fighters and the structures of war remain effectively in place. In other cases the international community is unwilling to provide the resources and troops necessary to support a long transition. Waiting until conditions are more propitious relative to the democratization goals may put the peace process at risk, encourage international peacekeepers to withdraw, and delay needed funding from international donors. Under such difficult circumstances, the “least bad” policy outcome may be to have an election that serves important goals with relation to war termination but that makes less progress in promoting democratization. Such war-termination elections may open limited but nonetheless important opportunities for a new political order that may be a prerequisite to advance the longer-term goals of democratization, rule of law, and strengthening civil society. Policymakers should recognize that while elections inherently and appropriately should be associated with democratization, they also serve other important roles. Although the international community should not cynically accept meaningless “demonstration elections” that legitimize

S-8

11/24/09

4:20 PM

Page 219

THE ROLE OF POSTSETTLEMENT ELECTIONS

219

authoritarian regimes and provide little scope for voter choice, 5 neither should it insist that democratization is the only criterion by which to judge the value of postsettlement elections. Other goals, most notably war termination and in many cases a desire by the international community to have a clear exit strategy, are also important and legitimate.6 The potential of a given election to serve as a tool of war termination may be greater than its potential with regard to democratization. It may also be that in the most difficult cases, ending the war is the only goal that can be accomplished in the short run. In addition, the value of a postsettlement election should be judged relative to the alternatives available. If the transitional period can be prolonged to provide the time, space, and security for civil society and political parties to organize, such a transition is more likely to result in meeting the goals of war termination and democratization. If, however, postponing an election until democratization goals are more likely to be met risks derailing the peace process and returning the country to war, it may be necessary to hold the election, implement the peace agreement, and promote democratization by focusing on the postsettlement government. The willingness of the international community to support the transition process is a critical factor in determining whether a long interim period is possible. Finally, if holding quick but flawed elections is more likely to ignite a new round of conflict resulting in neither war-termination nor democratization goals being advanced, then supporting the election is clearly dangerous and counterproductive. Judging whether an election is more likely to promote one, both, or neither of the twin goals of war termination and democratization requires careful analysis and is inevitably easier to judge in hindsight. That, however, is the policy challenge facing implementers of peace agreements in civil wars.

Power-Sharing Pacts and Electoral Systems There is a significant literature on the relationship between power-sharing pacts, electoral systems, and the potential for conflict mitigation in divided societies. Several scholars argue persuasively that to the extent that a regime can be inclusive and provide all significant constituencies with a voice in policymaking, chances of peace are increased. The experience of postsettlement elections, however, suggests that these policy tools are not likely to play a decisive role in determining success or failure in the most difficult cases. The characteristics that make some conflicts difficult to end will prevent agreement on power sharing. Electoral systems designed to disperse power and provide parties who win a minority of votes a significant role in governance are unlikely to provide sufficient assurances to those that fail to win the election.

S-8

11/24/09

4:20 PM

220

Page 220

EVALUATING IMPLEMENTATION TASKS

Power-Sharing Pacts The literature on political transitions suggests that the best way to manage conflicts in the context of breakthrough elections is to negotiate powersharing arrangements or political “pacts” to construct the broadest possible coalition in government. Political pacts are a set of negotiated compromises among competing elites prior to elections that institutionalize the distribution of power and thereby reduce uncertainty.7 Several transitions in Latin America, the Round Table talks in Poland, and the Convention for a Democratic South Africa were organized around implicit or explicit pacts designed to provide powerful actors with sufficient guarantees so that they would accept major political change. Pacts, however, ultimately rely upon other mechanisms for enforcement and do not by themselves end uncertainty or resolve the difficulty in making credible commitments. The strategic problems in reaching the initial cease-fire (information failures, inability to make credible commitments, spoiler problems, security dilemmas) will recur in the pact-making negotiations.8 Pacts are more likely among elites with relatively clear and loyal constituencies, such as traditional political parties, labor unions, or other institutions in a corporatist setting. In the aftermath of a civil war, such political and social organizations generally are absent and the ability of militia leaders to deliver the compliance of their own fighters is often questionable. Finally, the ability to assess the political power of a military faction is difficult and the identity of the critical constituencies to include in a pact is unclear immediately following a conflict. Even if the military balance among factions is relatively clear (as indicated by a stalemate on the battlefield), the relative political power of these factions and the extent to which they are capable of representing significant civilian constituencies in peacetime may be unknown.9 In Liberia, for example, the principle behind the negotiating process was that each faction had an equal number of representatives in the transitional government. Without any mechanism to assess relative strength, it was difficult to construct any alternative power-sharing arrangement. Following the 1997 elections, however, it was clear that only Charles Taylor’s National Patriotic Party represented any significant political power and that the other factions were politically marginal.10 The ability to assess the relative strengths of parties and thereby create a formula for a powersharing pact prior to elections is inherently difficult. Many observers expected the Union for the Total Independence of Angola (UNITA) to win the Angola elections, and many others were surprised by the ability of the Mozambique National Resistance (Renamo) in Mozambique, the National United Front for a Cooperative, Independent, Neutral, and Peaceful Cambodia (FUNCINPEC) in Cambodia, and the Farabundo Martí National Liberation Front (FMLN) in El Salvador to win significant votes. Assessing

S-8

11/24/09

4:20 PM

Page 221

THE ROLE OF POSTSETTLEMENT ELECTIONS

221

how fear and perceptions of the peace process will shape voter behavior is extremely problematic and therefore makes assessments of relative strength difficult. It is noteworthy that the best examples of power-sharing pacts in the cases of this study took place after the postsettlement election, when the relative political power of the parties was clearer. In Nicaragua the Sandinistas negotiated control over the military and in Cambodia the Cambodian People’s Party (CPP) forced a power-sharing arrangement after the elections. Negotiating a power-sharing pact in the most difficult cases may put the peace agreement and cease-fire at risk. Parties often accept a cease-fire and agreement to hold elections in the expectation that they will win the vote and gain power. Such parties will resist additional constraints on their power, preferring to compete in a winner-take-all election. The Carter Center tried to engage Charles Taylor in discussions of postelection power sharing in Liberia, but Taylor (correctly) believed that such an agreement would limit his power after he won the elections. To insist that Taylor accept a power-sharing pact would have placed the cease-fire and the rest of the peace process at risk. Furthermore, most Liberian citizens also opposed power-sharing pacts. The violent record of interim power-sharing governments established under the previous twelve peace agreements left many Liberians convinced that a return to “normal” government with a single president was better than another set of divided interim institutions. In Angola and Mozambique the international community pressed the parties to agree to a government of national unity, but the parties to the conflict refused to accept such limits, anticipating that they would achieve more power through winner-take-all elections.11 Even when a powerful international implementation force compelled power sharing in BosniaHerzegovina, parties to that conflict simply refused to cooperate with the new multiethnic and interentity institutions.12 In these most difficult of cases, negotiating a power-sharing pact, while desirable, may not be possible without placing the entire peace process at risk. Electoral Systems A substantial body of literature analyzes the relationship of institutional design in general and electoral-system choice in particular on the potential for conflict.13 Timothy Sisk and Andrew Reynolds argue that “electoral systems—the rules and procedures under which votes are translated into seats in parliament or the selection of executives—are a critical variable in determining whether elections can simultaneously serve the purposes of democratization and conflict management.”14 Proportional representation systems are generally regarded as preferable for divided societies because they more often create broad, inclusive governments. Other schemes involve various types of vote-polling and federalism.15

S-8

11/24/09

4:20 PM

222

Page 222

EVALUATING IMPLEMENTATION TASKS

As with power-sharing pacts, however, electoral systems and constitutional checks on power are particularly difficult following civil war and generally fail to overcome the security dilemmas and fears that shape such transitions. The same inability to make credible commitments that trouble the security dimensions of the transition will make agreements to limit power unpersuasive to parties who fear elimination by their rivals after the election. The influence of electoral systems is highly contextual and unlikely to alter the outcome of postsettlement elections. In the toughest cases of civil war termination, it is difficult to imagine what type of electoral system or separation of powers might increase the attractiveness of an election to competing parties. In Angola, for example, Stephen John Stedman argues that “Savimbi and UNITA received more from the Angolan settlement than any of the losing parties in El Salvador, Mozambique, Nicaragua, and Zimbabwe received in theirs.”16 Yet Jonas Savimbi defected from the agreement after losing the election. Even if provincial governors had been allocated according to provincial voting patterns, UNITA would have won only four out of eighteen provinces, too meager a prize to persuade Savimbi from trying his luck at winning power through war.17 The political and economic structures of many postconflict states are such that even an offer of local or regional leadership positions is unlikely to be sufficient to a party that lost the election for national leadership. Most resources and international aid in particular flow through the capital and are under the control of the executive, making the position of regional governor or mayor a weak power base. Renamo decided not to compete in 1998 local elections in Mozambique in part because it perceived that such posts were unimportant and subject to budget allocations under the control of their rivals in Maputo.18 Seats in a parliament or even cabinet positions meant little to the opposition in Liberia, given the strong presidential system enshrined in the Liberian constitution. In Sierra Leone the insurgents demanded control over the state’s diamond-marketing organization, recognizing that real resources flowed from gems, not regional political office.

Demilitarization of Politics Power-sharing pacts and electoral systems are unlikely to improve the chances that elections will advance both war termination and democratization, particularly in the most difficult cases where a return to war is a real risk. There are often other, more promising ways, however, to shape the transitional period to encourage more effective postsettlement elections. To overcome the legacies of civil war, implementers of peace agreements must initiate a process of “demilitarizing politics.” While consolidating the

S-8

11/24/09

4:20 PM

Page 223

THE ROLE OF POSTSETTLEMENT ELECTIONS

223

demilitarization of politics takes time, initial steps and precedents can be built into the transition in the short term and thereby channel the transition along a path that supports sustainable peace and democratization. To the extent that politics can be demilitarized during the interregnum between the cease-fire and the postsettlement election, such elections are more likely to advance the goals of conflict resolution and democratization. Postsettlement elections are shaped by the legacy of fear and by institutions developed and sustained during the conflict.19 To demilitarize politics and overcome this difficult heritage prior to elections entails building norms and institutions that bridge wartime structures based on insecurity and fear (such as militias and chauvinistic identity groups) to structures based on security and trust that can sustain peace and democracy (like political parties and civil society). The powerful actors that developed and were sustained during a protracted civil war cannot be wished away. Neither can the enabling environment for peaceful political competition be proclaimed into existence. A strategy to demilitarize politics includes strengthening interim institutions, reinforcing processes to demobilize combatants in a way that builds new institutions and norms, and supporting processes to build new political institutions such as political parties and effective electoral commissions. If the legacies of the conflict can be overcome through such developments, then the transitional period can better support elections that promote peacebuilding and democratization. This was the pattern in Mozambique, in El Salvador, and to an extent in Cambodia. If demilitarization of politics fails and the legacies of wartime continue to shape expectations at the time of the elections, then the elections will result in either a return to war, as in Angola, or the electoral ratification of the structures of the conflict and the strengthening of the most militant leaders, as in BosniaHerzegovina and Liberia. The process of demilitarizing politics therefore plays a critical role as the intervening variable between the legacies of war and the ability of elections to serve the role of ushering in a new institutional structure that supports peace. Interim Governments Interim governments are by their nature not legitimated by democratic processes but derive their authority from the extent to which they prepare the country for meaningful elections and turn power over to the winners.20 In the meantime, however, the country needs to be governed. Critical and contentious policy decisions relating to the electoral framework, demilitarization, reintegration of the displaced, and reestablishment of functioning economic and legal institutions must be made and implemented. The process through which such policies are made will shape the expectations

S-8

11/24/09

4:20 PM

224

Page 224

EVALUATING IMPLEMENTATION TASKS

of the major actors and may inspire confidence or ignite fears. Welldesigned and well-supported interim institutions can help encourage effective postsettlement elections. Disputes are inevitable during the transition, as the broad (and often vague if not contradictory) principles listed in the peace agreement must be made operational in a difficult and tense atmosphere. Negotiated settlements, signed under pressures to reach a cease-fire and end humanitarian suffering, generally include what they can, leave out what cannot be settled, and gloss over some important differences. As suggested by analysts of security dilemmas in civil war termination, a key to successful interim administration is to build institutional frameworks that bind the parties in self-restraint and mutual cooperation without increasing the risk of exploitation from a spoiler who does not comply.21 In a number of cases, provisions for ongoing negotiations, bargaining, and collaborative problem solving during the implementation period have helped build norms of nonviolent governance that foster confidence, legitimize decisions, and thereby reinforce the momentum for successful peacebuilding and democratization.22 Interim regimes therefore create new institutional contexts that can channel political action toward a path that supports both peacebuilding and democratization. The period between the signing of an agreement and an election provides the context for testing and assessing the risks and benefits of cooperation and the intentions and commitments of each party. In many cases, parties to a conflict adopt the extreme rhetoric of total war during the conflict but shift their language and tactics (if not their goals) during the peace implementation process. During the period between the cease-fire and elections, each party will look for evidence to confirm its fears that its rival is cheating. Noncompliance, however, may be the result of fear or a poorly designed or implemented demobilization or electoral process rather than a lack of commitment to the new rules of the game.23 Many analysts doubted that Renamo in Mozambique could make the transition from a fighting organization (some would even characterize it as a bandit or terrorist organization) into a viable political party able to play a constructive role in a multiparty democracy. At the same time, many anticipated that UNITA in Angola had a solid base of support and could function as a political party. It was only through the process of implementation that the capacity of each to contribute to peace and democratization became clear. The transitions in El Salvador and Mozambique illustrate how interim institutions based on joint decisionmaking encourage transitions that advance both conflict resolution and democratization. In El Salvador, the peace accords led to the creation of a National Commission for the Consolidation of Peace (COPAZ), a body with representation evenly split between the government and its allies, and the opposition including the FMLN insurgents, with observer status for the UN and Catholic Church.

S-8

11/24/09

4:20 PM

Page 225

THE ROLE OF POSTSETTLEMENT ELECTIONS

225

COPAZ debated and passed implementing legislation under the peace agreement, including constitutional amendments that redefined the role of the armed forces and the electoral law. Because the commission was evenly split between the government and the opposition (including the insurgents), “hammering out compromises became a political necessity—and a newly acquired skill for many politicians.”24 When problems arose over the electoral system or when an FMLN arms cache was uncovered after the deadline for disarmament, COPAZ was able to keep the parties talking and to keep the process moving toward elections. In Mozambique “the construction of a lasting peace . . . required still more negotiation and planning” following the Rome Accord.25 Joint decisionmaking bodies such as the Supervisory and Monitoring Commission (CSC) and the Cease-Fire Commission (CCF) brought together the major political actors with the major donors in a consultative process chaired by the special representative of the UN Secretary-General. Other specialized commissions dealt with reintegration of former combatants, reform of the Mozambican defense forces, and preparations for the election.26 These interim institutions created the context for overcoming some of the legacies of the civil war and thereby promoted the demilitarizing politics. In both El Salvador and Mozambique the international community supported these transitional arrangements through effective and well-funded UN operations. Angola, Liberia, and Bosnia-Herzegovina, in contrast, demonstrate how poorly constructed peace agreements and ineffective implementation can lead to transitional arrangements that fail to build the confidence, norms, or institutions that are associated with power sharing and a wellmanaged transition. Under the 1991 Bicesse Accords in Angola, pro forma meetings among the parties within the Joint Political-Military Commission did not change behavior and may have created a false sense of confidence in the parties’ commitment to the agreement.27 Demobilization failed, the military structures remained in place at the time of the elections, and war resumed after the vote. Under the Yamoussoukro, Cotonou, and Abuja Agreements in Liberia, interim administration was parceled out to each of the factions with only a weak and generally stalemated Council of State assigned the impossible job of coordination. The dysfunctional interim government, the failed demobilization process, and wartime perceptions that political rivals were enemies created a context in which most voters feared a return to war unless the most powerful military faction won the election. Under the Dayton Accords in Bosnia-Herzegovina, a Joint Interim Commission that included the prime ministers of Bosnia-Herzegovina, the Bosniac-Croat Federation, and the Republika Srpska met regularly, with the high representative of the Organization for Security and Cooperation in Europe (OSCE) as the chair.28 The commission discussed a broad range of issues, but working with such deeply divided institutions proved difficult.

S-8

11/24/09

4:20 PM

226

Page 226

EVALUATING IMPLEMENTATION TASKS

Each party continued to pursue unilateral advantage, thereby preventing effective forms of joint decisionmaking and forcing the OSCE to act unilaterally on a number of contentious topics. Interim institutions responsible for administration during the period between a cease-fire agreement and elections have the opportunity to begin the process of demilitarizing politics. To the extent that such transitional regimes operate on the basis of joint decisionmaking and collaborative problem solving, and create norms that protect the interests of key constituencies, they can build confidence in the peace process. Effective interim regimes will be better positioned to promote demilitarization of politics, particularly with regard to demobilization and the development of democratic institutions. Security, Disarmament, and Demobilization The first priority of a civil war peace settlement is to stop the fighting, manage security during the transitional period, and thereby provide the preconditions for democratization. Unless the difficult security dilemmas inherent in ending civil wars and the threats to the peace process posed by potential spoilers opposed to the agreement can be managed in the short run, the implementation process will fail. The manner by which security is managed in the interim period, however, can have important implications for the prospects of successful elections and longer-term peacebuilding. The institutions, norms, and precedents established to implement the security provisions of a peace accord can shape how governance will operate following implementation and an election. A process of demobilization that builds institutions and norms that can help sustain democratic governance after the transition can both contribute to reducing the threat of insecurity and increase the prospect for democratization. For the political process, the extent to which new rules are developed on cooperative and transparent processes to manage security questions is more important than the absolute quantities of arms decommissioned or reduction of force levels.29 The rationale for demobilization and disarmament is the exchange of military capacity for political benefits. 30 The demobilization and disarmament process may serve as the context for important confidence-building among the parties and the development of new institutions and procedures of decisionmaking that can encourage democratization and reinforce the peace process in important ways. There is a subtle interplay between the dynamics of a postconflict political transition and the manner in which the disarmament, demobilization, and reintegration provisions associated with that process are implemented.31 On the one hand, demobilization and related activities reduce the means by which civil wars are prosecuted and thereby reduce the chances that war will be reignited. At the same time and perhaps more importantly, as Joanna Spear

S-8

11/24/09

4:20 PM

Page 227

THE ROLE OF POSTSETTLEMENT ELECTIONS

227

points out in Chapter 6, demobilization helps create an environment for confidence- and security-building and for the development of institutions and norms for nonviolent political competition. Political Parties It is extremely difficult for militias and institutions that derived their power from the conflict to play the role of competing political parties in a democratic system if they remain organized and are led as they were during the period of armed conflict. In the more successful cases of transition, processes to demilitarize politics encouraged military organizations to transform themselves into political parties able to operate effectively in a multiparty context. In the less successful or failed cases, militias retained the ability to operate as military forces, weakening the ability of postconflict elections to mark a transition to civilian rule. In El Salvador the insurgent FMLN successfully converted itself to a legal political party capable of winning significant votes. 32 As the election neared, the FMLN focused its resources on building a national network of cadres and building alliances with other parties on the left. The United Nations called this transformation “the very core of the Peace Accords.”33 At the same time that the FMLN was moving away from its origins as an insurgent movement, the incumbent party, Alianza Republicana Nacionalista (ARENA), shifted its base of support. Initially founded by individuals closely associated with death squads and the repression of the left, it evolved into an effective political party that represented a broad range of landowners, bankers, merchants, and industrialists. Leadership of ARENA shifted from Roberto D’Aubuisson, who had ties to right-wing death squads, to the more moderate, probusiness Alfredo Cristiani.34 The postsettlement election in El Salvador therefore provided the opportunity for two key organizations involved in the civil war to make the transformation from military to political organizations able to compete effectively in elections. In Mozambique, the international community recognized that assisting Renamo to be a viable partner in the peace process was a condition for successful implementation.35 UN special representative Aldo Ajello stated that “it was necessary to help Renamo to achieve a minimum level that could allow the functioning of the whole mechanism” of the peace agreement.36 Renamo insisted that there could be “no democracy without money,”37 and Ajello agreed: “Democracy has a cost and we must pay that cost.”38 After initial concerns from donors reluctant to fund a party with a particularly brutal reputation, a $17.5 million fund was established to help Renamo transform itself into a political party. In addition to financial inducements, the international community worked to “socialize RENAMO into the rules of democratic competition, and to make its legitimacy contingent on fulfilling its commitment to peace.”39

S-8

11/24/09

4:20 PM

228

Page 228

EVALUATING IMPLEMENTATION TASKS

In Cambodia, the royalist FUNCINPEC developed during the transition into a relatively effective political party with a grassroots base, while the incumbent CPP continued to behave as it had when it was the sole legal party and the Khmer Rouge continued to operate as a guerrilla movement.40 In Angola, neither the ruling Popular Movement for the Liberation of Angola (MPLA) nor the insurgent UNITA ever made the transformation from a military organization to a competitive political party. In Liberia, the skill and ease with which the National Patriotic Front for Liberia transformed itself from an insurgent military organization into the populist National Patriotic Party suggests that some skills and structures developed during the armed struggle can be used as the base for an effective political organization.41 The creation of political parties able to represent key constituencies and compete effectively in an electoral process is an important component of any democratic transition. During peace implementation the viability of political parties plays an additional role with relation to war termination. To the extent that powerful leaders and their followers perceive that they have the option to operate as a political party rather than as a militia or military government, the chances of successful implementation are increased. Demilitarizing politics through the transformation of militias into political parties therefore promotes both war termination and democratization and is one of the keys to successful postsettlement elections. Electoral Commissions Institutions to organize and manage postsettlement elections combine functions relating to both conflict resolution and democratization. The primary purpose of any electoral administrative body is to deliver credible election services to the candidates and voters.42 Issues relating to impartiality and independence, efficiency, professionalism, and transparency are particularly important in the context of suspicion and mistrust that characterizes postsettlement elections. Furthermore, the precedents established in such an election—as in any “breakthrough” or “founding” election in a process of democratization—are likely to shape popular perceptions for years to come. Yet another challenge to postsettlement electoral authorities are the memories of earlier electoral fraud, as in El Salvador and Liberia. In cases where a stolen election was a cause of the conflict, the character of the postconflict election will be closely watched. There is no single best model for electoral commissions, and international practice is diverse.43 In a number of postconflict cases, an international organization has managed electoral administration as a means to increase public confidence in the process. The United Nations played a major role in Cambodia and Mozambique, a smaller but critical role in El Salvador, and a lesser role in Angola. The OSCE supervised elections in

S-8

11/24/09

4:20 PM

Page 229

THE ROLE OF POSTSETTLEMENT ELECTIONS

229

Bosnia-Herzegovina, and the Economic Community of West African States (ECOWAS) had oversight along with the UN in Liberia. All of these postconflict elections had international observers, from both international organizations and nongovernmental organizations, along with local domestic observers. Mozambique demonstrates that the manner by which disputes relating to electoral administration are managed can provide the context for building new institutions and norms that promote the demilitarization of politics and increase the chances of an effective postconflict election. The Rome Peace Accords left many issues relating to the elections vague and subject to decisions made by the parties through the Commission for Supervision and Control. A series of controversies between Renamo and the Front for the Liberation of Mozambique (Frelimo), over the composition of the National Elections Commission (CNE) in particular, contributed to the postponement of the election. After a series of discussions and with the active involvement of the UN special representative, the parties reached an agreement on the composition of the CNE. Frelimo received ten seats, Renamo seven, and other political parties three. The partisan balance made efficient decisionmaking difficult but increased the confidence of each party in the process. Over time and under the leadership of the Brazão Mazula, its independent chair, the CNE developed a reputation for nonpartisan decisionmaking.44 Unlike the Cease-Fire Commission and other commissions that included international actors along with representatives of the parties, the CNE was an independent and wholly Mozambican organization. The presence of UN technical advisers working within the Technical Secretariat, however, provided the parties with additional confidence in the process.45 El Salvador demonstrates that a consultative interim administration and other components that demilitarize politics can counterbalance the suspicion created by a weak and distrusted electoral administration. To administer the elections, the peace agreement relied upon the Supreme Electoral Tribunal (TSE), composed of representatives from five political parties, including the leftist Democratic Convergence (CD). Partisanship, “institutionalized distrust,” and incompetence, however, created gridlock, a crisis over voter registration, and chaos on election day.46 A multiparty Board of Vigilance that had oversight over the tribunal fortunately alleviated some of the suspicion toward the TSE.47 Despite these concerns, the presence of COPAZ and success in demobilizing combatants and creating political parties resulted in a successful process of demilitarization of politics. The period between the signing of a peace agreement and the implementation of the settlement through elections provides a number of opportunities to demilitarize politics and increase the chances that the voting process will serve to advance the goals of peacebuilding and democratization. Strong interim institutions that establish patterns and norms of bar-

S-8

11/24/09

4:20 PM

230

Page 230

EVALUATING IMPLEMENTATION TASKS

gaining and collaboration by developing confidence through the process of disarmament and demobilization, the precedents and habits put in place in the establishment of electoral administrations, and by the opportunity to encourage the transformation of militias and militarized organizations into political parties, can begin to demilitarize politics and increase the opportunity for postsettlement elections to advance both peacebuilding and democratization. Recognizing the imperatives of demilitarizing politics is critical to designing and supporting successful implementation of settlements through postconflict elections.

Conclusion: Policy Choices Postconflict elections play multiple roles in the context of peace implementation following civil war. While elections are inherently and appropriately associated with ideas of peaceful competition, participation, accountability, and other attributes of democracy, they often also perform other important roles. In some cases, most notably Mozambique and El Salvador, postconflict elections can simultaneously advance these multiple goals. In the most difficult cases, such as Liberia, Cambodia, and Bosnia-Herzegovina, the ability of elections to serve as a mechanism to initiate a process of democratization was limited, but their ability to terminate a particular period of conflict and to serve as a diplomatic stratagem to permit international peacekeeping forces to organize their intervention was important. These partial successes are often criticized and characterized as failures because the process of democratization did not advance significantly or indeed may have suffered a setback. This criticism is largely accurate but misses an important dimension of postconflict transitions. As recent studies have suggested, a significant number of factors are necessary for a postconflict election to advance the processes of war termination and democratization. Demobilization, rule of law, refugee repatriation, and a commitment by the warring parties to adhere to their agreements are often mentioned.48 In the most difficult cases, however, these elements are not present. This is unfortunate—even tragic—and the international community should examine what it can do to create a more propitious context for a given peace process, and why sufficient resources and political will are so rarely available. Regardless of the cause, however, when policymakers are faced with making decisions in a case that lacks the components most likely to promote both sustainable war termination and democratization, they must accept that the best they can do is to select the “least bad” option then available. Policymakers charged with addressing the massive challenges that face postconflict societies cannot afford to make successful democratization the criterion for all policies and must accept that in many of these hard cases war termination may be the only available short-term option that

S-8

11/24/09

4:20 PM

Page 231

THE ROLE OF POSTSETTLEMENT ELECTIONS

231

at least provides the potential for long-term stability and eventual transition to more liberal and democratic governance. To set expectations too high may lead policymakers to miss opportunities to assist in managing conflict. The dilemmas relating to the ability of an election to serve all the multiple goals assigned to them in a peace process may be illustrated with regard to the question of timing. The timing of postsettlement elections is critical to the capacity of such events to promote the dual goals of war termination and democratization. In some cases, the pace of implementation may be different for the two distinct challenges. Democratization requires time and will be encouraged by a slower pace that places elections at the end of a sequence of events such as demilitarization, repatriation of refugees and displaced persons, and rebuilding the basics of a functioning state. War termination, however, may require a more rapid tempo to obtain the consent of the warring parties, to build momentum from a cease-fire, and to ensure sufficient participation by international peacekeepers. Waiting too long may lead to the collapse of the peace process and a return to war. Whether the international community should press for a longer transition and a realistic opportunity to accomplish the dual goals of war termination and democratization simultaneously or accept that the former requires moving quickly will be case-specific and require difficult judgments about the nature of the peace agreement and underlying conflict. If war termination cannot wait, the international community should accept this limited but vitally important goal, since ending the fighting is a prerequisite for democratization and the other attributes of postconflict reconstruction and reconciliation. The potential to achieve these limited but important potential benefits by supporting flawed elections should be balanced against the potential for such elections to strengthen authoritarian leaders. If nondemocratic leaders win postconflict elections, as in Bosnia-Herzegovina, Liberia, and elsewhere, they will have the opportunity to entrench their power and use state institutions to weaken their opponents and strengthen their hold on power. The ability of future elections to be competitive may be reduced if a de facto single-party government is created and legitimized by the first round of elections. In addition, if the public concludes that flawed elections indicate that voting is a tool by which powerful forces in society legitimize their control without offering the voter a meaningful choice and that the international community is prepared to sanctify such exercises as “free and fair,” then distrust of democratic institutions will become pervasive, as in Angola. Flawed elections may be the “least bad” alternative in a postsettlement environment, but the price for long-term democratization may be high. For the international community to recognize the opportunity to support war-termination elections in contexts where democratization will receive at best a modest boost, policymakers and analysts will have to

S-8

11/24/09

4:20 PM

232

Page 232

EVALUATING IMPLEMENTATION TASKS

accept that elections can serve multiple purposes. The traditional judgment on elections—were they “free and fair”?—should give way to a more nuanced and contextual interpretation. Free and fair elections are important, but under the difficult circumstances of civil wars a more immediate judgment may be whether the elections were meaningful or effective with relation to war-termination goals. In cases like Angola, elections characterized as “free and fair” by international observers failed miserably with respect to war termination. In other cases, notably El Salvador, Nicaragua, and Mozambique, elections succeeded to advance both war termination and democratization. In Liberia and Bosnia-Herzegovina, the elections should be understood as marking an important stage in transforming the conflict (with the long-term sustainability of conflict management still very much in doubt) while failing to advance significantly the democratization process.

Notes 1. See, for example, Krishna Kumar, ed., Postconflict Elections, Democratization, and International Assistance (Boulder, Colo.: Lynne Rienner, 1998). 2. Christopher Clapham, “Rwanda: The Perils of Peacemaking,” Journal of Peace Research 35, no. 2 (1998): 195. 3. For a classic statement, see Dankart A. Rustow, “Transitions to Democracy: Towards a Dynamic Model,” Comparative Politics 2, no. 2 (1970): 337–363. 4. Kumar, Postconflict Elections, Democratization, and International Assistance. 5. Edward S. Herman and Frank Brodhead, Demonstration Elections: U.S.Staged Elections in the Dominican Republic, Vietnam, and El Salvador (Boston: South End Press, 1984). See also Terry Lynn Karl, “Imposing Consent? Electoralism vs. Democratization in El Salvador,” in Paul Drake and Eduardo Silva, eds., Elections and Democratization in Latin America, 1980–85 (San Diego: CLAS/Center for U.S.-Mexican Studies, 1986), pp. 9–36. 6. Timothy D. Sisk, “Conclusions and Recommendations,” in Timothy D. Sisk and Andrew Reynolds, eds., Elections and Conflict Management in Africa (Washington, D.C.: U.S. Institute of Peace, 1998), p. 147. 7. On pacts, see Terry Lynn Karl, “Dilemmas of Democratization in Latin America,” Comparative Politics 23, no. 1 (1990): 1–21; Guillermo O’Donnell, Philippe C. Schmitter, and Laurence Whitehead, eds., Transitions from Authoritarian Rule: Latin America (Baltimore: Johns Hopkins University Press, 1986), pp. 37–47; Frances Hagopian, “Democracy by Undemocratic Means? Elites, Political Pacts, and Regime Transition in Brazil,” Comparative Political Studies 23, no. 2 (1990): 147–170; and Giuseppe Di Palma, To Craft Democracies: An Essay on Democratic Transitions (Berkeley: University of California Press, 1990), pp. 86–90. 8. For a discussion of the strategic issues that make pact-making difficult, see Adam Przeworski et al., Sustainable Democracy (Cambridge: Cambridge University Press, 1995), pp. 24–30.

S-8

11/24/09

4:20 PM

Page 233

THE ROLE OF POSTSETTLEMENT ELECTIONS

233

9. J. ‘Bayo Adekanye, “Power-Sharing in Multi-Ethnic Political Systems,” Security Dialogue 29, no. 1 (March 1998): 33. 10. Taylor received 75.3 percent of the total vote and a majority in every county, while his rival factional leaders, Alhaji Kromah and George Boley, received only 4.0 percent and 1.3 percent respectively, both concentrated in their ethnic heartlands. Militia leaders treated as equals in the negotiations leading to the peace agreement clearly were not equal when measured by their ability to win votes in an election. See Terrence Lyons, Voting for Peace: Post Conflict Elections in Liberia (Washington, D.C.: Brookings Institution, 1998). 11. Joao Honwana, “Mozambique,” and Virginia Page Fortna, “Angola,” drafts prepared for this project. 12. See Chapter 18 by Elizabeth Cousens in this book. 13. Kenneth D. McRae, “Theories of Power-Sharing and Conflict Management,” in Joseph Montville, ed., Conflict and Peacemaking in Multiethnic Societies (Lexington, Mass.: Lexington Books, 1990). Classic studies include Arend Lijphart, “Consociational Democracy,” World Politics 4, no. 1 (January 1969): 207–225; and Arend Lijphart, Democracy in Plural Societies (New Haven, Conn.: Yale University Press, 1977). 14. Timothy D. Sisk and Andrew Reynolds, “Democratization, Elections, and Conflict Management in Africa: Exploring the Nexus,” in Sisk and Reynolds, Elections and Conflict Management in Africa, pp. 3–4. Ben Reilly and Andrew Reynolds, Electoral Systems and Conflict in Divided Societies, Papers on International Conflict Resolution no. 2 (Washington, D.C.: National Academy Press, 1999), similarly argue that “electoral systems can be powerful levers for shaping the content and practice of politics in divided societies” if the design is sensitive to context (p. 1). 15. Donald Horowitz, Ethnic Groups in Conflict (Berkeley: University of California Press, 1985). 16. Stephen John Stedman, “Negotiation and Mediation in Internal Conflict,” in Michael E. Brown, ed., The International Dimensions of Internal Conflict (Cambridge: MIT Press, 1996), p. 370. 17. Thomas Ohlson and Stephen John Stedman, The New Is Not Yet Born: Conflict Resolution in Southern Africa (Washington, D.C.: Brookings Institution, 1994), p. 193. 18. John Blacken and Terrence Lyons, Mozambique: From Post-Conflict to Municipal Elections (Washington, D.C.: Management Systems International for USAID, 1999). 19. For more on the legacy of fear in the context of postconflict elections, see Terrence Lyons, “Postconflict Elections: War Termination, Democratization, and Demilitarizing Politics,” paper presented at the annual meeting of the American Political Science Association, Atlanta, Ga., September 1999. 20. Yossi Shain and Juan J. Linz, Between States: Interim Governments and Democratic Transitions (Cambridge: Cambridge University Press, 1995), pp. 3–21. 21. Jack Snyder and Robert Jervis, “Civil War and the Security Dilemma,” in Barbara F. Walter and Jack Snyder, eds., Civil War, Insecurity, and Intervention. (New York: Columbia University Press, 1999), pp. 15–37; and Stephen John Stedman and Donald Rothchild, “Peace Operations: From Short-Term to LongTerm Commitment,” International Peacekeeping 3, no. 2 (summer 1996): 17–35. 22. Donald Rothchild, “Bargaining and State Breakdown in Africa,” Nationalism and Ethnic Politics 1, no. 1 (1995): 54–72. See also Timothy D. Sisk, Power Sharing and International Mediation in Ethnic Conflicts (Washington, D.C.: U.S. Institute of Peace, 1996), pp. 85, 115.

S-8

11/24/09

4:20 PM

234

Page 234

EVALUATING IMPLEMENTATION TASKS

23. Stephen John Stedman, “UN Intervention in Civil Wars: Imperatives of Choice and Strategy,” in Donald C. F. Daniel and Bradd C. Hayes, eds., Beyond Traditional Peacekeeping (New York: St. Martin’s Press, 1995), p. 57; and Stedman, “Negotiation and Mediation in Internal Conflict,” pp. 345–351. See also Rui J. P. de Figueiredo Jr. and Barry R. Weingast, “The Rationality of Fear: Political Opportunism and Ethnic Conflict,” in Walter and Snyder, Civil War, Insecurity, and Intervention, pp. 261–301. 24. Tommie Sue Montgomery, Revolution in El Salvador: From Civil Strife to Civil Peace, 2nd ed. (Boulder, Colo.: Westview Press, 1995), pp. 233–234. David Holiday and William Stanley criticize COPAZ for its slow and cumbersome decisionmaking, compelling endless rounds of negotiations among parties. From the perspective of building new norms to demilitarize politics, such continuous discussion is a strong asset. See David Holiday and William Stanley, “Building the Peace: Preliminary Lessons from El Salvador,” Journal of International Affairs 46, no. 2 (winter 1993): 427–429. 25. Richard Synge, Mozambique: UN Peacekeeping in Action, 1992–94 (Washington, D.C.: U.S. Institute of Peace, 1997), p. 52. 26. J. Michael Turner, Sue Nelson, and Kimberly Mahling-Clark, “Mozambique’s Vote for Democratic Governance,” in Kumar, Postconflict Elections, Democratization, and International Assistance. 27. Margaret J. Anstee, Orphan of the Cold War: The Inside Story of the Collapse of the Angolan Peace Process, 1992–1993 (New York: St. Martin’s Press, 1996). 28. See Chapter 18 by Elizabeth Cousens in this book. 29. Fred Tanner, “Post-Conflict Weapons Control: In Search of Normative Interactions,” paper presented at the International Studies Association Meeting, Washington, D.C., February 1999, p. 2. 30. Annika S. Hansen and Lia Brynjar, The Role of International Security Assistance in Support of Peace Agreements in War-Torn Societies, FFI Report no. 98/05291 (Kjeller, Norway: Forsvarets Forskingsinstitut [Norwegian Defense Research Establishment], December 1998). 31. Mats R. Berdal, Disarmament and Demobilization After Civil Wars, Adelphi Paper no. 303 (London: IISS and Oxford University Press, 1996), p. 73. 32. Gerardo L. Munck, “Beyond Electoralism in El Salvador: Conflict Resolution Through Negotiated Compromise,” Third World Quarterly 14, no. 1 (1993): 87; and Montgomery, Revolution in El Salvador, p. 253. 33. UN Document S/26005, “Report of the Secretary-General Concerning Illegal Arms Deposits Belonging to the FMLN,” June 29, 1993, para. 11. 34. William Stanley, The Protection Racket State: Elite Politics, Military Extortion, and Civil War in El Salvador (Philadelphia: Temple University Press, 1996), pp. 220, 254–255. 35. Carrie Manning, “Constructing Opposition in Mozambique: Renamo as Political Party,” Journal of Southern African Studies 24, no. 1 (March 1998): 161–190. See also Donald Rothchild, Managing Ethnic Conflict in Africa: Pressures and Incentives for Cooperation (Washington, D.C.: Brookings Institution, 1997). 36. Aldo Ajello, “O papel da ONUMOZ no processo de democratização,” in B. Mazula, ed., Moçambique: Eleições, democracia e desenvolvimento (Maputo: Inter-Africa Group, 1995), p. 127, cited in Martinho Chachiua and Mark Malan, “Anomalies and Acquiescence: The Mozambican Peace Process Revisited,” African Security Review 7, no. 4 (1998): 22. See also UN Document S/25518, “Report of the Secretary-General on ONUMOZ,” April 2, 1993, para. 58.

S-8

11/24/09

4:20 PM

Page 235

THE ROLE OF POSTSETTLEMENT ELECTIONS

235

37. Alex Vines, Renamo: From Terrorism to Democracy in Mozambique? (London: James Currey, 1996), p. 146. 38. “Mozambique: Funding for Peace,” Africa Confidential 34, no. 10 (May 14, 1993): 4. 39. Stephen John Stedman, “Spoiler Problems in Peace Processes,” International Security 22, no. 2 (fall 1997): 41. 40. Kate Frieson, “The Politics of Getting the Vote in Cambodia,” in Steve Heder and Judy Ledgerwood, eds., Propaganda, Politics, and Violence in Cambodia: Democratic Transition Under United Nations Peace-Keeping (Armonk, N.Y.: M. E. Sharpe, 1996), pp. 183–207. 41. For more details, see Lyons, Voting for Peace. The flip side of this pattern was seen in Lebanon, where a number of political parties, such as Jumblat’s Druze-based Progressive Socialist Party, proved adept at making the transformation from partisan politics to militia politics, suggesting that change in the other direction is also possible. See Michael C. Hudson, “Power-Sharing in Post-Civil War Lebanon,” International Negotiation 2, no. 1 (1997): 111. In the former Yugoslavia, Georgia, and South Africa, most political parties had their own armed wings. 42. Peter Harris, “Building an Electoral Administration,” in Peter Harris and Ben Reilly, eds., Democracy and Deep-Rooted Conflict: Options for Negotiators, Handbook Series no. 3 (Stockholm: International Institute for Democracy and Electoral Assistance, 1998), p. 310. 43. Guy S. Goodwin-Gill, Free and Fair Elections in International Law (Geneva: Inter-Parliamentary Union, 1994). 44. When Renamo leader Afonso Dhlakama charged fraud and announced a boycott of the election, the Renamo representatives on the CNE joined their colleagues in rejecting it unanimously. 45. Turner, Nelson, and Mahling-Clark, “Mozambique’s Vote for Democratic Governance.” 46. Montgomery, Revolution in El Salvador, pp. 248–259; UN Document S/26606, “Report of the Secretary-General on the Activities of the ONUSAL Electoral Division,” October 20, 1993; Madalene O’Donnell, Jack Spence, and George Vickers, El Salvador Elections 1994: The Voter Registration Tangle (Cambridge, Mass.: Hemisphere Initiatives, July 1993); and Madalene O’Donnell, El Salvador Elections 1994: Voter Registration Progress and the Tasks Ahead (Cambridge, Mass.: Hemisphere Initiatives, November 1993). 47. Enrique A. Baloyra, “El Salvador: From Reactionary Despotism to Partidocracia,” in Kumar, Postconflict Elections, Democratization, and International Assistance, p. 21. 48. Krishna Kumar and Marina Ottaway, “General Conclusions and Priorities for Policy Research,” in Kumar, Postconflict Elections, Democratization, and International Assistance.

S-8

11/24/09

4:20 PM

Page 236

S-9

11/24/09

4:21 PM

Page 237

9 Human Rights and Sustainable Peace TONYA L. PUTNAM

The internationalization of settlement mechanisms for civil wars has ushered in a heightened awareness of and concern with the human rights aspects of peace implementation. However, as the poor human rights records of many postsettlement governments readily attest, awareness alone is hardly sufficient to begin to correct the serious human rights problems that obtain in most postconflict settings. Part of the difficulty associated with establishing a robust regime of human rights protection undoubtedly lies with the magnitude and complexity of the problems prevailing on the ground during and immediately following peace implementation. However, some measure of responsibility for the failure to more effectively promote and protect human rights in the initial stages of peace must be placed on human rights organizations themselves, particularly those operating at the international level. The heart of the problem is the apparent failure of many international human rights organizations (IHROs) to recognize that effective promotion and protection of human rights early in postconflict settings requires different tactics than those typically applied in response to human rights abuses occurring in stable societies with established governments.1 In the latter situation, IHROs have achieved positive results by calling international attention to gross shortfalls between human rights ideals embodied in international legal instruments, and the actual practice of governments in order to prompt corrective action to enforce those international standards. The specific IHRO tactics referred to include gathering and publicizing information about human rights violations; expressing formal protest to violator governments by mechanisms within the United Nations and regional structures; and lobbying outside governments to pressure the violator to comply with international standards. Where possible, many IHROs have also advo237

S-9

11/24/09

4:21 PM

238

Page 238

EVALUATING IMPLEMENTATION TASKS

cated using the justice system to enforce individual accountability and redress for violations. For shorthand purposes, I refer to this set of legal and political tools and tactics as the “enforcement approach” to human rights protection. I argue in this chapter that exclusive reliance on the enforcement approach to human rights protection is entirely unsuited to the early stages of peace implementation. The methods encompassed by the enforcement approach are misapplied in these circumstances because they assume the functionality of the very institutions that peace implementation operations are tasked to help bring into existence. Among the required features for effective application of the enforcement model that are likely to be either nascent or altogether absent in the early stages of peace implementation are a nonpartisan police force, an impartial judiciary, and various other public institutions for the administration of health, safety, and basic services. Attempts to make standard international human rights “solutions” fit postconflict human rights “problems” introduce a degree of dysfunction into the task of peace implementation and the role of IHROs within it. This dysfunction is reflected in three increasingly common modes of IHRO action that, when applied in the contexts studied here, are more likely to detract from the goals of substantive international human rights protection than to bring them about. First, the central importance IHROs have attached to expanding and perpetuating the formal reach of international human rights instruments often leads these organizations to place undue importance on achieving formal expression of human rights provisions in the body of peace settlements. Yet evidence suggests strongly that the formal enumeration of human rights provisions in settlements correlates only weakly, if at all, with the quantity and quality of human rights protection during and after peace implementation. The drive to achieve formal elaboration of human rights commitments, at best, expends political and organizational resources that could be used more effectively elsewhere. At worst, such pressure may actually interfere with the achievement and successful implementation of a peace settlement, particularly when questions of accountability for past human rights violations arise. Second, a pure enforcement approach to human rights protection makes the assignment of individual responsibility for human rights violations imperative. Viewed through the lens of enforcement, to ignore that imperative is to sanction acts of impunity, which in turn undermines the future deterrence function of legal enforcement. However, prosecution of human rights violations requires investigative and adjudicative bodies and compensatory structures that do not exist in most postconflict settings, and that can take months, years, or even decades to develop. As a result, IHROs have in recent years begun to advocate what amounts to the importation of fully functional judicial proceedings in the form of specially constituted

S-9

11/24/09

4:21 PM

Page 239

HUMAN RIGHTS AND SUSTAINABLE PEACE

239

international criminal tribunals (ICTs) for the prosecution of war crimes and other serious violations of human rights. However, the ICTs for both the former Yugoslavia (ICTY) and Rwanda (ICTR) have failed the expectations of the populations they serve, and have become entrenched as a separate and, in the case of the ICTR, even competing model of justice. Meanwhile, national justice institutions limp along, incapable of handling and sometimes even perpetuating the multitude of violations that continue to occur on a daily basis inside these countries, as documented by many of the very same IHROs that have supported the ICT model.2 Finally, the enforcement approach to human rights protection brings with it procedural biases that pressure strongly for the isolation of IHROs and human rights concerns from direct involvement with postsettlement governments, and even from other tasks within the implementation mission. From the IHRO side, the adversarial ethos embodied in the enforcement approach reinforces the idea that IHROs cannot and should not consult directly with governments regarding human rights violations. 3 However, IHROs and, for that matter, entire human rights divisions that are sequestered away from other components of peace implementation missions occupy a comparatively weak position vis-à-vis both mission headquarters and the parties on the ground. Neither peace implementers in the field nor postsettlement governments are likely to attempt to alter the situation, since incorporating human rights considerations would only complicate their already demanding daily tasks. In almost all cases, IHRO agreement during the planning stage to a coordinated role within specific implementation tasks will provide much greater opportunity to bring about lasting structural and institutional change in post–civil war societies than will reliance on monitoring and reporting alone. Each of these three propositions will be treated in further detail following a brief examination of the operational nexus between peace implementation, IHROs, and effective mechanisms for the protection of fundamental human rights in the period immediately following the signing of a peace agreement. Thereafter, I turn to a discussion of positive courses of action that IHROs can adopt to ensure more effective advancement of the international human rights agenda in postconflict settings.

IHROs and Peace: A Contextualization Whereas peace is necessary for the robust protection of fundamental human rights and freedoms, an actual or nascent protective regime for human rights may not be necessary to achieve or maintain the cessation of armed hostilities.4 Examples abound, both historically and in the present day, of stable governmental regimes with very poor human rights records. By contrast, no state wracked by armed hostilities and internal conflict is able to

S-9

11/24/09

4:21 PM

240

Page 240

EVALUATING IMPLEMENTATION TASKS

ensure that respect is accorded to the full slate of human rights and freedoms of the populations concerned. This observation is intended only to lay bare the hierarchy of priorities during peace settlement and implementation. It is not to suggest that “peace” should be purchased at any price— clearly it should not. Nor is it to suggest that human rights conditions play no role in determining social and political stability—clearly they do. However, the functional dependence of human rights protection on the absence of war means it is unreasonable to demand, as many from the international human rights movement voice, that when human rights and political expediency require different courses of action, human rights considerations should be automatically accorded precedence. Peace implementation is a complex undertaking, and post–civil war settings contain a number of unfortunate political realities (many of which constitute the focus of other chapters in this book) that cannot be ignored or glossed over. Human rights constitute but one component in successful peace implementation, and the organizations that make it their business to oversee human rights must be able to work together with other elements in order to be ensured a meaningful place in the implementation formula.5 The proclivity of IHROs to overplay their hand should be a matter of concern to planners and policymakers. In the fragile conditions that prevail during peace negotiations and in the first stages of implementation, if influential actors misjudge a situation the consequences can be far from benign. When IHROs attempt to superimpose their own priorities on implementation settings without sufficient sensitivity to the complexity of the endeavor as a whole, they may place the entire enterprise of peace at risk. For instance, some IHROs, particularly Amnesty International, have been highly vocal in opposing amnesties against prosecution for high officials suspected of war crimes and crimes against humanity,6 even when amnesty is explicitly held out as a condition of settlement.7 In an ideal world, such trade-offs would not have to be made,8 yet in settings as diverse as Angola, Guatemala, Namibia, Sierra Leone, and Cambodia, explicit or de facto amnesties have proven key to attaining settlement.9 In another example, at least one insider to the Bosnian peace process attributes part of the blame for the Bosnian Muslims’ last-minute rejection of the proposed September 21, 1993, settlement to the outcry of international human rights NGOs that the accord constituted an implicit endorsement of ethnic cleansing.10 Considering the patent improbability that the ethnic partition that formed the basis of the outcry could be reversed in the near term either by diplomacy or by continued fighting, it is difficult to see what this outcry was meant to achieve. The implication of the IHRO arguments in both types of examples is that safeguarding a principle of international human rights is more important than attaining a peace settlement. It should be recalled, however, that in these situations, the price of maintaining the moral and rhetorical high ground will be paid in additional lives lost from continuation of the

S-9

11/24/09

4:21 PM

Page 241

HUMAN RIGHTS AND SUSTAINABLE PEACE

241

conflict, often with no possibility of further progress being made with regard to the issue at stake.11 Clearly, IHROs have important roles to play in effective peace implementation. The promotion and protection of human rights both during and after implementation should be an important measure of the success of an implementation mission as a whole, and to the quality of the subsequent peace. However, too much of a focus on the long view at the beginning of an operation can obscure the difficulty of the steps needed to get there.12 “Big picture” myopia may cause features of the political and strategic landscape of critical importance to other implementation actors to be filtered out of the IHRO picture. This in turn can lead IHROs to ride roughshod over delicate facets of peace negotiation and implementation in pursuit of a single, albeit highly important issue. An even stronger likelihood is that IHRO failure to take into account other tasks and priorities will reinforce the pronounced tendency to isolate human rights within implementation missions to the detriment of all involved. IHROs, no less than other actors engaged in peace implementation, need to be able to respond in each new situation in the manner most likely to achieve the overall goal of sustaining peace, even if doing so entails adopting unfamiliar practices and entering into unfamiliar partnerships. Admittedly, the challenge is not an easy one.

Hard Bargains: Human Rights Provisions and Settlement Agreements The urgency with which many in the international human rights community insist upon the need for parties emerging from civil war to commit formally to uphold human rights is an artifact of the era before the internationalization of peace implementation, when a line in a settlement agreement constituted the best, and sometimes the only, toehold into a peace process. However, since the early 1990s a de facto consensus on the constitutive elements of comprehensive peacebuilding has emerged under UN auspices from which the commitment to the protection and promotion of human rights in the early stages of implementation can be easily read. The consensus is reflected not only from within the UN, but also in the practices of international financial institutions and several key donors of aid and reconstruction funds, including the United States and many European countries. Still, human rights field manuals, 13 and recommendations drawn from analyses of specific UN human rights operations,14 emphasize the importance of including explicit provisions for the protection of human rights in the body of agreements between warring parties. In actuality, the enumeration of human rights provisions offers little in the way of leverage in the absence of institutions or actors willing and able to implement them. At the same time, premature or misplaced attempts to force inclusion of enumerat-

S-9

11/24/09

4:21 PM

242

Page 242

EVALUATING IMPLEMENTATION TASKS

ed human rights provisions may create additional stumbling blocks to settlement—by opening up new areas of disagreement, or by providing uncooperative or obstructionist parties additional pretexts for failure to sign a proposed agreement. Nature of the Bargain The bargain struck among conflicting parties that permits them to lay down their arms is obviously a key feature in the resolution of any civil war. However, the cases summarized in the appendix of this chapter (see page 261) indicate that a carefully articulated regime of human rights guarantees is usually not a substantive part of those bargains. Only in rare circumstances where the parties themselves choose to make human rights a centerpiece of the peace process (e.g., El Salvador) does this conclusion cease to apply. Short of situations in which a settlement is imposed from outside, mediators are limited in their ability to force issues and provisions onto the negotiating table. Starting conditions for negotiation impose serious constraints on the type and range of measures—human rights provisions among them—that parties are willing to agree to, and also indicate a great deal about the probability of their fulfillment once accepted. For example, the parties to peace talks may be cooperative or noncooperative; 15 the issues over which the war was fought may be resolved or still outstanding; economic conditions may be conducive or not conducive to a rapid postsettlement improvement in the lives of former combatants, citizens, and refugees. A high-profile approach to human rights may operate as a lever in negotiations, providing points of agreement to lift parties out of deadlock on difficult issues. However, human rights issues may also function as a stumbling block, serving only to multiply points of disagreement and thereby to reduce overall the likelihood of reaching agreement. In the El Salvador peace negotiation, the shared drive among the parties to eliminate “systematic practices of human rights violations” was instrumental in breaking their deadlock. The introduction of human rights into the negotiation at the instigation of UN special representative Alvaro de Soto spurred what Ian Johnstone describes as a “symbiotic relationship between human rights and the peace process”: human rights eased the way to a sustainable political settlement, and the achievement of political settlement in turn paved the way for progress in human rights.16 Without taking anything away from the parties or the negotiators of the peace agreement, the unity of purpose behind the parties received considerable help from the acknowledged fact of military stalemate, and from the ongoing effort of other governments in the immediate region to encourage a settlement. In Guatemala and Nicaragua, two further success stories, regional players were key to galvanizing the initial effort toward

S-9

11/24/09

4:21 PM

Page 243

HUMAN RIGHTS AND SUSTAINABLE PEACE

243

peace, and only at a later stage did human rights formally become part of the peace agenda.17 By contrast, the starting conditions in conflicts such as those in Liberia, Sierra Leone, and Somalia suggest an extremely low probability that human rights considerations could have functioned as a lever in reaching a settlement. These conflicts share the characteristic of multiple competing factions with no armed party that necessarily aspires to a protective political role over the general population. Instead, what began in each case as a reaction against ethnic-based economic inequalities deteriorated into a competition for control over economic resources.18 Introducing human rights considerations into the mix under such conditions could not have transformed these conflicts into situations ripe for settlement. Moreover, in Liberia and Sierra Leone the practices that the international human rights agenda have sought most to curtail—extrajudicial killing and maiming of noncombatants, the use of child soldiers, systematic rape—functioned (and continue to function) as operative means by which parties control populations and extract resources. In short, unless imposed by outside actors possessing both the will and the means to force an end to the conflict, human rights qua human rights are simply unlikely to figure into first-order security calculations among parties that would prefer, when all is said and done, to continue the fight. Thus far the discussion in this section has focused on settlement processes that, with the notable exception of Somalia’s Addis Ababa Agreement, have been achieved largely on the initiative of the parties themselves, or the parties plus an interested regional power.19 In contrast, the negotiation of the 1991 Paris Agreement on Cambodia and the 1995 Dayton Agreement on Bosnia-Herzegovina occurred under radically different conditions. Both settlement agreements are noteworthy not only with respect to the comprehensiveness of the human rights provisions they contain, but also because of the degree of coercion with which they were imposed upon the signatory parties. In both cases, the agreements were negotiated with a high degree of international participation, and the overall structure of the settlement was decided for the parties, not by the parties. Moreover, in both cases at least one powerful party to the conflict was an unwilling participant in the process.20 Both the Paris Agreement and the Dayton Agreement were followed by the deployment of international implementation forces of unprecedented size and scope. In short, external conditions distinguish these two situations from those discussed above. Here again, human rights considerations had no opportunity to operate either as a lever or a stumbling block to reaching a settlement. In both Bosnia and Cambodia it was the identity of the implementing parties—not the presence of formal provisions in the settlement per se— that ensured that the protection of human rights would constitute a substantive focus in the postsettlement implementation process.

S-9

11/24/09

4:21 PM

244

Page 244

EVALUATING IMPLEMENTATION TASKS

UN Formula An important consequence of the internationalization of conflict settlement mechanisms is that comprehensive settlements to civil wars almost always entail UN involvement, either as the primary implementer or as a formal observer to the process. In cases where the United Nations or the United States takes primary responsibility for implementation, human rights goals can be read directly from the list of substantive implementation goals in the implementation mandate. Thus, the division of labor under the implementation mandate constitutes a far more appropriate focus for IHRO lobbying efforts than do direct negotiations between conflicting parties. If instead the UN will have only a partial mandate during peace implementation, as was the case, for example, in Liberia and Sierra Leone,21 the IHRO argument for insisting that the parties make reference to human rights in the text of the settlement is somewhat stronger. In these cases, however, attention to starting conditions still suggests that direct appeals to the parties to include a substantive human rights component will not be effective, and may in some cases damage the peace process. Post–Cold War peacebuilding actively encourages postconflict governments and societies to rebuild along liberal democratic lines—to adopt democratic institutions and democratic standards of openness and transparency; to establish civilian control over the military; to encourage market economies; to actively reform police organizations and judiciaries; to educate the public about their rights before such institutions; and to ensure the protection of fundamental human rights and freedoms. Functionally, issues of concern to human rights permeate every aspect of comprehensive peacebuilding. Most operational components of peace implementation can be easily reframed in human rights terms, offering opportunities for human rights to piggyback on other goals. Thus, weapons decommissioning and curtailing trade in small arms can be framed alternately as removing the means for groups to use violence arbitrarily or for their own narrow ends. Similarly, the task of disbanding a police force complicit in the abuses of the old regime and designing a new legal framework for an impartial, disciplined police organization is replete with human rights concerns regarding procedural fair treatment and substantive limits on the legitimate use of state police powers. The ability to conduct what the UN considers to be “free and fair” elections in a postsettlement setting effectively requires active observance on the ground of freedom of speech, movement, and assembly, and free participation in the political process.22 The list could easily go on. The important point is that these requirements hold whether or not they are spelled out explicitly in the settlement or in the implementation mandate: they are integral to the substantive fulfillment of the tasks themselves. As a case in point, the settlement in Namibia did not feature an explicit

S-9

11/24/09

4:21 PM

Page 245

HUMAN RIGHTS AND SUSTAINABLE PEACE

245

human rights component in either the peace agreement or the mandate of the UN Transitional Assistance Group (UNTAG). Nonetheless, Namibia is widely considered to have been an exceptional UN peace implementation success on several dimensions, including human rights promotion and education. Although the implementation operation in Namibia was, arguably, unique in a number of respects,23 outcomes in Mozambique under the 1992 Rome Agreement and in Lebanon under the 1989 Ta’if Agreement, where human rights also did not feature in the text of the agreements, have generally developed in encouraging directions. In Mozambique, positive results were achieved by overtly tying the implementation tasks of the UN mission there (ONUMOZ) to the simultaneous development of civilian institutions, including UN training of Mozambican nationals to provide security and to administer justice.24 In Lebanon, the impetus for aggressive protection of human rights in the postsettlement context came largely from below with the creation of a large number of local NGOs immediately following the 1989 settlement. Interestingly, these local human rights organizations relied largely on rights and guarantees embodied in the Lebanese constitution, and not international instruments, to press their human rights agenda.25 Of course, there are also failed settlements in which human rights considerations were excluded from the negotiations, including those in Sri Lanka, Liberia, Sierra Leone, and Angola. Among this latter set of cases, only in Angola did the United Nations have more than observer status in the brokering and implementation of the settlement. However, the reasons for the collapse of the 1992 Bicesse Accord are highly unlikely to have been prevented through greater involvement of human rights actors and organizations.26 Implementation Mandates When it comes to the provisions contained in implementation mandates, the need for formalism becomes more pronounced. The need arises because, in the move from agreement to mandate, the primary audience shifts from the parties in the conflict to the gaggle of organizations and agencies jockeying for resources and position in the implementation mission. Under Chapter VII operations, implementation mandates are drafted entirely by outside parties. The situation is somewhat more complicated under Chapter VI Observer and Verification missions, since care must be taken not to overstep the formal boundaries of what has been agreed by the parties. In these cases too, a lead state or organization, and not the parties to the conflict, typically drafts the mandate. The main function of the mandate is to set the scope and objectives of the implementation mission for contributing parties.27 Mandates are typically short (a few paragraphs in length) and therefore are not intended to elaborate fully the proposed division of responsibilities among implementers. The concept of operations, the formal designa-

S-9

11/24/09

4:21 PM

246

Page 246

EVALUATING IMPLEMENTATION TASKS

tion of lead agencies for various operations, together with the allocation of implementation tasks, are left to be worked out less formally (although often almost equally brutally) among the organizations and agencies involved. Where, in the cases under consideration, implementation mandates have included a human rights component, the relevant clauses have been, with the exception of those in the Dayton Agreement and the San José Agreement on El Salvador,28 drafted in quite general terms. Even in the mandate attached to the Paris Agreement on Cambodia, human rights provisions consume a mere three lines in what is otherwise a relatively elaborate mandate. However, the consideration, or lack thereof, afforded to human rights in an implementation mission does not depend upon the number of column inches devoted to the topic in the mandate. Rather, it depends upon the support of other components of the implementation mission and, as is the case so often with respect to legal questions, upon how those with agenda-setting power choose to interpret the relevant provisions. For example, in the Implementation Force (IFOR), the implementation mission for Bosnia-Herzegovina, enormous domestic political pressures existed in the United States for the U.S.-led NATO force to interpret its mandate narrowly—unduly so from a human rights and civilian implementation perspective. As Elizabeth Cousens makes clear in Chapter 18, from the outset of the mission, IFOR’s objective was to move toward military completion (i.e., elections and a hand-over of power) as quickly as possible and to leave. In practical terms, this meant that implementers on the “military side” were directed to minimize responsibilities and commitments on the “civilian side,” irrespective of damage this might entail to civil reconstruction efforts, and to the quality of the elections themselves. Constitutional Order A peace agreement is only an intermediary step between armed conflict and the design and consolidation of new, postconflict structures of government. Michael Doyle observes that “signing [a peace treaty] does not end political bargaining. After the parties agree to the creation of a peacekeeping operation, they continue to compete for advantage. The agreement becomes, as do so many other constitutional texts, an agreement to struggle.”29 Over the longer term, domestic institutions and domestic arrangements will determine the degree of protection afforded to the fundamental human rights and freedoms of a country’s citizens. These arrangements find their formal expression first and foremost in constitutions. A survey of postconflict constitutions in force in states previously engaged in civil war should therefore indicate whether and to what degree the inclusion or exclusion of formal human rights provisions from the settlement agreement impacts upon the later formalization of human rights in the domestic order. In every instance

S-9

11/24/09

4:21 PM

Page 247

HUMAN RIGHTS AND SUSTAINABLE PEACE

247

where peace implementation proceeded far enough to permit the drafting of a postconflict constitution or the reinstatement of a previously existing constitution, those documents have included substantive human rights elements (see chapter appendix). This outcome holds regardless of whether human rights provisions were included explicitly in the peace accords. That being said, the actual pattern of human rights protection in postconflict societies is, not surprisingly, far more checkered than a survey of the formal provisions embedded in constitutions would predict. To identify the reasons behind this variation, one must move beyond formalism at all levels to an examination of subsequent institutional expressions and embodiments. Winning Some, and Losing Some Within the universe of peace agreements summarized in the chapter appendix, in only three of the six successful cases did either the peace agreement or the implementation mandate contain an explicit human rights component, and in only one of those cases (El Salvador) were the aforementioned human rights provisions broadly enumerated. Of the ten failed or partially successful cases, two settlement agreements (Cambodia and BosniaHerzegovina) included broadly enumerated human rights components. Additionally, Sierra Leone included a human rights component in the mandate of its UN Observer Mission there (UNOMSIL), as did Liberia in the modified mandate for UNOMIL, for the implementation of its abortive settlement. The raw numbers reveal little about whether and to what degree human rights considerations were instrumental to the ultimate success or failure of each respective agreement, or whether they were necessary to the concrete improvements that were made to human rights conditions on the ground after the settlement. In many cases these outcomes are overdetermined, and therefore one should be cautious about assigning causality to human rights provisions without considering the full range of contributing factors. As George Downs and Stephen Stedman observe in Chapter 2, peace implementation operations do not come into being with equally likely chances of success. The ability of IHROs to take effective action likewise differs with the implementation context, as does the recipe for what constitutes effective action. Peace agreements sometimes fail for reasons that no set of human rights initiatives, particularly those in their beginning stages, could reasonably be expected to prevent: military opportunism in the face of changing conditions, political spoiling,30 the failure of military confidencebuilding measures, or the evaporation of external pressures for parties to adhere to the agreement, to name but a few.31 At the same time, in cases of implementation success where formal human rights provisions were included in the settlement agreement, it may be unclear whether those provisions,

S-9

11/24/09

4:21 PM

248

Page 248

EVALUATING IMPLEMENTATION TASKS

as distinct from deeper forms of commitment, in fact contributed to the outcome. Even in El Salvador, the formal commitment to the human rights ideals contained in the San José Accord did not represent a “new” development in a legal sense, but instead a political commitment to return to the rule of law as it existed before the conflict.32 According to the evaluation of the Lawyers Committee, the “real innovation” in the agreement was not the formal commitment to human rights per se, but the fact that the parties pledged to allow international verification of the implementation effort.33

Trying Times: Human Rights and Turnkey Justice Although international instruments provide standards for the protection of human rights, they alone cannot provide for the promotion and protection of those standards. The default approach for operationalizing international human rights standards, at least among the larger IHROs with the greatest transnational reach, is modeled heavily on Western liberal ideas of judicial enforcement. However, effective use of the enforcement approach presupposes the existence of an entire set of functioning institutions to investigate, prosecute, and punish individuals who commit human rights violations. In societies emerging from civil wars, such institutions are normally weak or nonexistent, if indeed they existed beforehand.34 Institutions must be built before they can function; police, advocates, and judges must be trained in due process before they can administer due process; members of the public must understand their rights as individuals, as well as the system designed to protect those rights, before they can begin to exercise them. Moreover, success in the protection of human rights is highly dependent upon progress in a variety of peace implementation tasks. For example, in Guatemala, human rights institutions such as the Ombudsman for Human Rights and the Human Rights Council, although created early in the UN mandate process, began to operate effectively only after the security context stabilized, civil society had again begun to function, and institutions such as the postreform judiciary and the police had begun to function.35 International organizations and other third parties can take up some of the slack in human rights enforcement initially, but the presence of third parties is a temporary factor and cannot substitute for the development of local capacities to promote and protect human rights.36 Among the important determinations every implementation force must make is which tasks to attempt independently, and which to attempt in coordination with the parties to the settlement. Michael Doyle, Ian Johnstone, and Robert Orr argue that one of the central lessons from the UN Transitional Authority in Cambodia (UNTAC) is that UN operations should “seek to design into the mandate as much independent implementa-

S-9

11/24/09

4:21 PM

Page 249

HUMAN RIGHTS AND SUSTAINABLE PEACE

249

tion as the parties will agree to in the peace treaty.”37 However, this option is, in general, not realistic when it comes to human rights: the open-ended demands of protecting human rights contrast sharply with the temporally limited character of third-party involvement. Achieving anything approaching immediate compliance with international human rights standards in the sphere of policing, judicial operations, or the distribution of economic and social resources is feasible only if outside implementers assume responsibility for day-to-day operations in these areas. But this option creates its own set of problems down the road, since the greater the reliance upon third-party implementers for the day-to-day administration of civil affairs, the less likely the chance for domestic institutions to develop. If, as is highly likely, the “takeover” method hinders the development of a domestic capacity to protect human rights, the threat to human rights over the long term may be greater.38 To be effective, third-party human rights inputs must be geared toward the establishment of local institutions capable of assuming those functions, and not simply toward investigating and punishing violations as they occur until the mandate expires and the third parties withdraw. The dilemma described above is particularly acute with respect to the question of how to treat severe human rights violations committed before a settlement was reached. At the outset it should be noted that under classic formulations of international law and the laws of war, no legal obligation exists to prosecute individual human rights violators following the conclusion of a peace treaty.39 Increasingly, however, IHROs are reading such an obligation out of various international human rights treaty instruments.40 Almost invariably, this disposition has been tied, implicitly or explicitly, to a bias favoring Western-liberal legal models of human rights enforcement with its attendant requirement of formal legal proceedings against indicted suspects. In recent years the pull of this approach has proven so strong that even organizations such as Human Rights Watch and Africa Watch, which have traditionally focused largely on monitoring and reporting, have entered the judicial fray in Rwanda under the auspices of facilitating prosecutions at the ICTR.41 International assistance in the prosecution of war crimes and other abuses of human rights, some would argue, constitutes a practical means by which outside actors can help to jump-start domesticlevel judicial institutions and move formerly divided societies toward rapprochement.42 Prosecutions are purported to signal the return to the rule of law in postconflict societies, to facilitate reconciliation by creating a public record of violations,43 to help institutionalize mechanisms for punishing past violations (thereby minimizing vigilante justice), and finally, to deter future violators by assigning individual responsibility for criminal acts. However, there are strong reasons to doubt whether any of these goals are actually best served through the prosecutorial mechanism.44 Insistence upon prosecution of suspected war criminals immediately

S-9

11/24/09

4:21 PM

250

Page 250

EVALUATING IMPLEMENTATION TASKS

begs the question of who or what institution will take on the task. The national legal and judicial systems in post–civil war societies will not be immediately up to the job of handling wide-scale war crimes prosecutions with anything approximating full adherence to Western-liberal standards of due process. This presents advocates of prosecution with a tough choice: postpone the call for prosecutions or risk subjecting the accused to a judicial gauntlet that is highly likely to transgress another set of standards of core importance to the rule of law. International criminal tribunals have been touted as a way out of this dilemma. Set up on what is effectively a turnkey basis, an ICT can be up and running in a matter of months, as opposed to the years it may take to develop a domestic capacity for the fair and orderly prosecution of war crimes and other grave human rights abuses.45 However, there are problems with the prosecutorial mechanism that extend beyond the character of the forum.46 To begin, the extraordinary nature of the crimes in question raises doubts about the ability of courts, particularly national courts in the country where the crimes allegedly occurred, to prevent those proceedings from becoming politicized, thereby undermining the move to the rule of law. Second, the rule of law informed by due process requires that violations be punished swiftly and equitably. However, in settings where mass atrocities have occurred, the number of full-blown prosecutions that a single legal system can conceivably manage will be minuscule in relation to the actual scope of prosecutable acts.47 For example, Rwandan governmental officials estimate that, at the current rate of proceedings, it would take over 130 years to prosecute the full docket of criminal indictments stemming from the 1994 genocide.48 This high degree of selectivity obviously undermines the purported deterrence function of individual war crimes prosecutions, while also leaving the impression that many individuals have escaped punishment. Finally, it is highly dubious whether individual prosecutions create a public record of atrocities sufficient for purposes of national reconciliation, due both to the small number of prosecutions, and to the mode of presentation of evidence in adversarial forums. Under the adversarial model, parties are not encouraged to present a balanced version of the events in question, but rather to characterize all relevant facts in the manner most favorable to one’s own position and to withhold information that does not support one’s own case. In addition, establishing an ICT to prosecute suspected war criminals may create fresh problems. In Rwanda at least, the international tribunal has emerged as a competing—not a complementary—model of justice. As a result of the relatively small number of cases being handled by the ICTR relative to the number of those incarcerated pending trial, war crimes prosecutions have gone forward simultaneously at the national level.49 The ICTR has indicted or prosecuted only those individuals accused of having played leading roles in planning or executing the genocidal massacres of

S-9

11/24/09

4:21 PM

Page 251

HUMAN RIGHTS AND SUSTAINABLE PEACE

251

April–May 1994, leaving the vast majority of suspects to be processed in the national courts. As a result, individuals accused of lesser crimes than those whose cases end up in the ICTR must defend themselves with lesser counsel, under less stringent due process guarantees, and face the potential of much harsher penalties if convicted.50 Tensions between the two systems came to a head in early November 1999 when the ICTR released JeanBosco Barayagwiza, on grounds that he had been detained too long without trial.51 As a result of Barayagwiza’s release, the Rwandan government officially suspended its cooperation with the ICTR for several months.52 Prosecutions at the international level may therefore facilitate “justice” in an abstract sense, but still fail miserably to resonate in the societies in which human rights violations occurred. Less legally formalistic institutions such as truth commissions have proven to be far superior to trials in avoiding many of the problems identified above.53 Moreover, evidence gathered by truth commissions can also provide leverage necessary for purging high officials from positions in the new governments, even if those officials are barred from formal prosecution by political amnesty, 54 and they can deliver specific recommendations for preventive reform. 55 Still, the most compelling argument against the advisability of IHRO insistence upon prosecutions for war crimes during peace implementation is that resources devoted to dealing with past wrongs are not being applied to present problems and to prevent future wrongs. Although some would argue that postconflict societies cannot move forward until they have come to terms with the past,56 before the past can be dealt with it must truly be considered the “past.” Without institutions and institutional barriers (judicial, social, security) to ensure against a repetition of past abuses, postconflict societies cannot move toward rapprochement in a meaningful way. Emphasis on the prosecution of past violations is justifiable, therefore, only if accompanied by parallel efforts at building institutions capable of punishing human rights abuses perpetrated in the postconflict environment and of helping to prevent future abuses. To the extent that there are important questions of timing, or trade-offs to be made between the two, the choices are better left to the societies in question, not imposed from the outside.57

No Human Rights Organization Is an Island Another serious drawback associated with the enforcement approach to human rights protection is its tendency to encourage the isolation of IHROs from other strategic and operational components inside implementation missions. Under the enforcement approach, IHROs document human rights abuses, and demand, either indirectly through lobbying or directly through the courts, acknowledgment and redress for victims.58 However, the greatest human rights challenge in postconflict situations rarely consists in

S-9

11/24/09

4:21 PM

252

Page 252

EVALUATING IMPLEMENTATION TASKS

demonstrating that human rights violations have occurred. Rather it lies in discovering root causes behind systematic violations, and in devising viable options for prophylactic response. Without a workable institutional framework capable of responding to demonstrated human rights violations, even the most stringent international advocacy in support of enforcement will have few direct repercussions for either perpetrators or victims. Even more important, the adversarial ethos that underpins the enforcement approach sets IHRO advocates in formal opposition to suspected violators, which in turn limits severely the level of constructive engagement IHROs can undertake with violator institutions and governments.59 Because most other components of peace implementation missions require active engagement with governing bodies and with the populace to carry out their missions, an enforcement approach to human rights protection also limits the ability of IHROs to integrate with other implementation agencies in pursuit of common goals. The large degree of congruence between the goals of the human rights agenda and the goals of successful implementation of comprehensive peace settlements offers fertile ground for IHROs to cultivate early institutionalization of human rights ideas and standards in the design of key civil and administrative structures and procedures.60 But it is an opportunity that is all too rarely acted upon. Even within UN-led implementation missions in which human rights feature in both the settlement agreement and the mandate, the open-ended character of human rights goals, their breadth, and their process-orientation all dictate that, if isolated from other implementation goals, human rights objectives will receive low priority during implementation.61 Inside the UN, the situation is not helped by the competition between New York–based human rights operations and Geneva-based operations, with the latter category under the authority of the UN High Commissioner for Human Rights (UNHCHR) tending to be far more assertive than the former. In addition, UN human rights missions, particularly those administered from New York, have been faulted for being excessively remote from the situation on the ground, especially when it comes to incorporating local NGOs in the implementation process.62 In most implementation settings, the price of maintaining the structural autonomy of a human rights mission, although attractive from an agenda control perspective, is to “doom the division to irrelevance.”63 In the words of Marrack Goulding, UN Undersecretary-General for Political Affairs, “Standing alone, a human rights division would lack influence, room to maneuver and status. . . . Being within the mission gives it very direct status as an interlocutor.”64 The comparative weakness of IHROs within the UN structure, at least with regard to their role in peace operations, has been recognized as a problem by Secretary-General Kofi Annan.65 This is fortunate, insofar as securing a more integrated role for human rights in UN peace implementation

S-9

11/24/09

4:21 PM

Page 253

HUMAN RIGHTS AND SUSTAINABLE PEACE

253

missions over the short term will have to be instigated from the top down, beginning at the planning stage.66 Despite ubiquitous rhetoric in support of human rights, the agencies and units charged with the operational aspects of peace implementation can hardly be expected to beat down IHRO doors to demand an integrated IHRO presence in their respective spheres of responsibility. The reasons are several. And they are attributable at least as much to the nature of organizations in general, as to the United Nations in particular, although the importance of each factor may be heightened or lessened in specific cases depending upon particular organizational biases and the personalities involved. First, at the level of strategic planning, introducing a new and different set of priorities and concerns will undoubtedly complicate decisionmaking, at least until functional roles and formulas for cooperation become better understood and more established. Agencies and organizations that are regular contributors to peace implementation missions, such as the Department of Peacekeeping Operations (DPKO), regional organizations such as the Organization for Security and Cooperation in Europe (OSCE) and the Economic Community of West African States (ECOWAS), and even the UN High Commission for Refugees (UNHCR) and the Office for the Coordination of Humanitarian Affairs (OCHA), are likely to resist widening the decision circle to include IHROs. At least, that is, until elements within these organizations are able to perceive IHROs as potential allies in helping to further their own agendas. Second, like all “rational” organizations, peace implementation missions prefer to structure operations according to finite tasks that lend themselves to clearly demonstrable successes, both for the benefit of international audiences, and to win support within the postsettlement societies they serve. The unbounded character of human rights protection, coupled with its greater emphasis on process (i.e., how specific goals are accomplished, as opposed to merely whether they are accomplished), lends itself neither to finitude nor to easily definable indices of success. Therefore, the net effect of incorporating IHROs into implementation planning will be to elevate the criteria on what missions must achieve in order to be considered successful, and to complicate the design of many mandate tasks. Third, turning now to the operational side of the equation, incorporation of human rights tasks and concerns into the implementation mix will make already difficult operational tasks even more complex. It will, at a minimum, add to or change established implementation repertoires, which in turn is likely to demand additional resources and additional training of implementers. Active collaboration with IHROs may also complicate relationships between peace implementers and authorities on the ground. IHRO scrutiny is often unpopular with postsettlement political authorities who, in spite of their often precarious positions, find themselves the target of criticism for failing to quickly eradicate pervasive and long-standing human

S-9

11/24/09

4:21 PM

254

Page 254

EVALUATING IMPLEMENTATION TASKS

rights problems. To the degree that the active cooperation of postconflict authorities is normally required to ensure the safety and security of the implementation mission, together with necessary guarantees such as freedom of movement, and cooperation with regard to technical verification, adding a human rights dimension may not be simple or costless. Further underlining the danger that human rights will fall by the wayside in peace implementation if not integrated with other tasks is the often noted discrepancy between the ambitiousness of mandate goals and the resources allocated to achieve them. For example, the 1993 UNTAC mandate in Cambodia was unprecedented with respect to the scope of its human rights provisions. Nevertheless, a mere ten officers were appointed initially to oversee implementation of the mandate for the entire country, a number that later increased to thirty-one.67 To put this in perspective, at its height, the human rights unit of UNOMIL in Liberia consisted of only three officers.68 Therefore, given the problem of chronic underfunding and low prioritization, how can IHROs operating at all levels have the greatest positive impact in the sphere of human rights protection? The answer, already suggested, is to ensure that human rights organizations and interests are integrated into other implementation tasks of higher, immediate organizational priority. As difficult as the IHRO struggle for influence may appear from inside UN politics, the view from the outside is more so.69 Regardless of their respective resource bases and preferred approach to human rights protection, all international human rights NGOs share a general lack of direct leverage vis-à-vis the parties to the conflict. Even for human rights NGOs that also undertake humanitarian functions, the provision of resources rarely translates into an ability to induce parties to significantly improve the human rights situation. Humanitarian aid is by nature short-term, and the organizations that provide it often have a visible interest in seeing it delivered to the intended recipients—a factor that can itself be manipulated by parties on the ground to undercut efforts directed at curtailing practices that violate human rights.70 Initiatives have been launched to begin to integrate NGOs more fully into UN activities generally. However, it is unclear how successful such efforts will be across the board given the demonstrated unwillingness of many international human rights NGOs to tie themselves to the UN mast, even if only on a short-term basis. Of course, from an IHRO perspective there are potential drawbacks to becoming integrated into an implementation mission, beyond the reduction in agenda control already mentioned. First, agreeing to an integrative strategy may require IHROs to appear to tolerate a higher incidence of human rights violations at the beginning of the mission than would be the case if those organizations maintained their status as stand-alone participants. Coordination with other agencies may occasionally require “tactical” handling of information (e.g., delaying action against certain limited violations

S-9

11/24/09

4:21 PM

Page 255

HUMAN RIGHTS AND SUSTAINABLE PEACE

255

in order to broker deals on other issues with greater preventive potential). Trade-offs of this type are anathema to actors and organizers dedicated to pure forms of enforcement, and will undoubtedly create a great deal of antagonism within specific IHROs faced with the choice of whether to cooperate. Cooperation of this sort may also pose difficulties vis-à-vis some financial supporters, even if, over the medium to long term, an integrative strategy is likely to save more lives and to deliver a higher quality of life through the creation of more robust and sustainable civil and human rights structures. Integration into peace implementation missions demands that IHROs habituate themselves to the pragmatics of policymaking, a requirement that may not fit comfortably with their normal modes of operation. IHROs should not expect to lead peace operations merely on the strength of the ideals they proclaim; rather, they need to pitch in with generating and helping to implement potential solutions to the problems they so thoroughly document. For example, Kenneth Cain, a former UN human rights officer in Liberia, is highly critical of IHRO policy performance in the Liberian conflict. Cain argues that the policy inputs of these organizations amounted to nothing more than “human rights cheerleading.” Cain notes pointedly that IHROs “failed entirely . . . to offer any realistic, pragmatic, meaningful program, plan, or strategy to actually address human rights abuses on the ground.”71 My own research suggests that this pattern is the rule, rather than the exception. In summary, the enforcement approach to human rights protection exacerbates an already pronounced tendency for human rights components to become isolated within implementation missions, and for IHRO inputs to be relegated to the sidelines. However, IHROs willing to expand or adapt their repertoire of activities beyond those encompassed by the enforcement approach can open possibilities for more directly influencing key implementation processes so as to minimize or eliminate probable sources of future human rights problems. If integrated wisely, the benefits of IHRO inputs in the formative stages of key institutions will be of long duration— far longer than the existence of the implementation mandate.

Human Rights by Other Means Implicit in the argument thus far has been the notion that human rights protection can be furthered through modes of operation outside the enforcement approach, and that these alternative modes are more suited to effective human rights work in postsettlement situations. Human rights education, standard-setting, and institution-building represent three approaches to building human rights capacities in postconflict societies—the same capacities required for effective human rights enforcement. In fact, nonenforce-

S-9

11/24/09

4:21 PM

Page 256

256

EVALUATING IMPLEMENTATION TASKS

ment modes of human rights advocacy constitute a significant proportion of the human rights work undertaken by NGOs from the Global South, in contrast to NGOs and IHROs from North America and Europe, which tend to rely more on the enforcement approach.72 To be clear, the recommendation here is not that IHROs seeking involvement in peace implementation completely set aside traditional monitoring and reporting roles. 73 To do so would run counter to the core mission of many of these organizations, and would, moreover, deprive missions of invaluable informational resources. Rather, a balanced approach to human rights protection asks IHROs to play both sides of the human rights field simultaneously: to contribute to the institution-building and education effort, while continuing to monitor the progress of the peacebuilding enterprise as a whole. As expressed in one particularly apt formulation, the challenge for IHROs is to “buy in” to peace implementation without “selling out.”74 A degree of tension exists between the role of institutional oversight traditionally performed by IHROs, and the involvement of these organizations in the task of institutional design.75 However, Diego Garcia-Sayan, director of the Human Rights Division in the UN Observer Mission in El Salvador (ONUSAL), argues that while these tensions are “logical,” they are also “manageable” and need not detract from the quality of human rights verification.76 The fact that some IHROs, or parts of them, opt to work within the framework of peace implementation in no way constitutes a brake on other organizations continuing to work independently. In Cambodia, even local NGOs have succeeded in combining both functions. For example, one local organization provides human rights training to officers in Cambodia’s national police force, and still manages to be one of the most outspoken critics of human rights violations by the police.77 Indeed, IHROs can maximize their presence and influence in peace implementation missions by taking on both roles. What, then, are some of the specific contributions IHROs can make in the early stages of peace implementation within an integrative approach to human rights promotion? Of course, a detailed answer to this question for any specific context requires an evaluation of the particular postsettlement situation in question, including the starting conditions, the nature of the settlement, and the scope of the implementation mandate, together with the resources and configuration of the implementation mission. Even so, it is possible to offer some general recommendations that can be tailored to specific contexts. Grassroots Level Human rights promotion and protection at a basic, grassroots level are about educating individuals and groups regarding their rights and obliga-

S-9

11/24/09

4:21 PM

Page 257

HUMAN RIGHTS AND SUSTAINABLE PEACE

257

tions under international treaties and within their own national constitutional structures. It is also about developing mechanisms for the realization and protection of those rights, including, most especially, civil institutions that function equitably and according to democratic precepts. Clearly, these objectives cannot be accomplished in the course of a peace implementation mission; they constitute ongoing projects even in the world’s most established liberal democracies. Still, IHROs, working in conjunction with other components of peace operations, can orient their training and lobbying efforts toward facilitating the performance of key institutions such as the police and the judiciary, without which the prospects for sustainable peace often appear dim, and the prospects for sustainable rights, even dimmer. A sizable number of activities undertaken in the early stages of an implementation mission offer opportunities for IHROs to establish contacts and build working relationships from the government level down. In many cases, taking advantage of these opportunities can actually help to facilitate the tasks of other implementation components. From the opening phases of an implementation mission, IHROs should be actively engaged in helping to train and monitor locally recruited police, security forces, judges, and public prosecutors.78 IHROs can help to jump-start judicial administration for handling property settlements, family issues, and postsettlement criminal cases by coordinating with civil reconstruction agencies and local authorities to ensure working space and materials for courts to begin functioning on at least a local level. At the same time, IHROs should actively solicit local NGO participation in each of these areas, both to encourage greater coverage and to ensure that this work continues beyond the termination of the mission. In addition to the Cambodian example already mentioned, local NGOs in El Salvador and Guatemala have achieved considerable success in this area, despite difficult beginning relationships (particularly in El Salvador) with their IHRO counterparts. IHROs can also team up with agencies charged with economic reconstruction and local NGOs to combine human rights education with the provision of basic resources, such as clean water and electricity and other public utilities; reopening hospitals and schools; refugee resettlement; distribution of construction materials and farming implements—or any other programs that will enable contact with large numbers of people. IHROs as Interlocutors In return for the integration of IHROs into implementation missions, IHROs can offer access to a valuable stream of reliable information regarding developments in the field. Insofar as successful peace implementation is dependent upon parallel developments in several areas, reliable monitoring and exchange of information regarding activity at all levels is essential.

S-9

11/24/09

4:21 PM

258

Page 258

EVALUATING IMPLEMENTATION TASKS

In many contexts, IHROs may also be able to facilitate the flow of information in the opposite direction, downward into the population. Working in tandem with local NGOs, contacts with the “grass roots” can be utilized to educate host populations about new institutions and their rights within them, to register voters and disseminate election information, and to explain and offer assurances about the peace implementation mission itself. As large organizations with sizable budgets, a wide geographical embrace, and a strong commitment to human rights ideals, IHROs are able to undertake initiatives, and to tap resources that smaller, more locally based human rights organizations cannot. Local human rights NGOs, particularly those that are new or that operate from small resource bases, can often benefit from a reliable interlocutor at the international level, especially when it comes to establishing transnational contacts or seeking out funding. Contact with IHROs can also bring broader comparative perspectives to locally based human rights activists and enrich the available store of problem-solving tools. Networking with local NGOs can open channels of communication that will enable outside implementers to avoid needless and costly mistakes based on misinformation or cultural ignorance. Although IHROs often have a strong grasp of human rights conditions on the ground in postsettlement settings, as outsiders they may not be able to contextualize this knowledge sufficiently to be able to put it to the best use for institution-building. For this, coordination and cooperation with local human rights actors and organizations is crucial.79 Local NGOs are significantly more “context-driven” than IHROs, meaning that their raison d’être is the promotion and protection of human rights in a particular setting in contrast to the more universalist drive behind many of the larger IHROs. Perhaps the greatest service IHROs can render to local and national NGOs both during and after the mission is to act as watchdogs regarding the treatment these organizations and their members receive from government authorities. If local human rights NGOs are to flourish in postconflict societies, they need room to maneuver as well as to speak openly regarding their governments. Postconflict governments therefore need to know that if fault-finding organizations or their outspoken members are “eliminated,” they will be held accountable at the international level. However, for such arrangements to be credible, there must be at least some degree of familiarity with the organizations in question, and some measure of communication. Here too there is room for creativity within implementation missions. For example, as part of the implementation process in Cambodia, a country with no previous NGO experience, UNTAC sponsored an international conference to support emerging local NGOs and to bring them into contact with IHROs at the regional and international levels.80 Tellingly, Cambodia has sustained an active community of human rights NGOs, even through the period of renewed armed hostilities in 1997.

S-9

11/24/09

4:21 PM

Page 259

HUMAN RIGHTS AND SUSTAINABLE PEACE

259

Mission Level At the mission level, IHROs are in a position to act credibly as guardians of the long term regarding institutional and structural choices early in the implementation process. In contrast to many of the organizations contributing personnel and expertise to implementation missions, IHROs are likely to remain engaged with the governments and societies in question over the long term and therefore can claim a more direct interest in the state of affairs on the ground after the mission. Anticipation of eventual requirements for the protection and promotion of human rights is needed from the beginning of the institution-building process, even if those institutional goals are unlikely to be realized, either fully or partially, for some time. Finally, IHROs working within the implementation process are better positioned to monitor progress in implementation than those operating outside the fold, and better positioned to communicate their recommendations to others in the mission with greater effect. IHROs can forward the cause of human rights by pushing lead actors to define specific human rights goals and to specify steps for their realization in conjunction with progress in other areas of implementation. Considerable scope for the inclusion or exclusion of human rights considerations exists in the operationalization of objectives agreed to by the parties in the peace agreement. Gaps in the substance of the peace agreement can be filled by reading into goals already agreed to by the parties (e.g., civilian reconstruction) consent as to the concrete steps necessary to reach the goals. These may include everything from planning programs for refugee resettlement, to humanitarian aid, opening up space for economic activities, de-mining, and so on. Not all measures may be equally popular with the parties to the agreement, but when presented as a package, measures that are less popular with authorities on the ground, such as human rights oversight, have a greater chance of being accepted. As pointed out by the Lawyers Committee, “It’s important that the good cop/bad cop is within the same institution.”81

Conclusion Achieving both peace and human rights in the immediate aftermath of a civil war entails a series of balancing acts between complex demands and crosscutting tensions. First, mediators must balance between what are objectively desirable elements in a peace settlement, and what parties are willing to agree to. Next, implementers must balance between the breadth of tasks specified in the implementation mandate, and the limited resources allocated for their fulfillment. Third, human rights actors must balance the functional tensions that exist between their traditional oversight role and

S-9

11/24/09

4:21 PM

260

Page 260

EVALUATING IMPLEMENTATION TASKS

the more integrated demands of effective peacebuilding. Finally, and most importantly, human rights actors must balance a sensitivity for the fragility of postconflict political settings against the impetus to react forcefully against all forms of human rights violations using familiar enforcement tactics—that is, those tactics developed to extract compliance from stable, established governments. Examination of the balancing acts undertaken in the sixteen cases included in this project, together with their relative successes and failures, yields the following policy conclusions: • The absence of explicit human rights provisions in a peace settlement in no way inhibits their incorporation into subsequent laws or constitutions. Therefore, the positive utility of pressing to include such provisions in a settlement instrument should be approached in each new context as an open question—not as a starting assumption. • Approaches to human rights protection that ignore the paucity of institutional preconditions for effective human rights enforcement offer little to improve the conditions on the ground during the first stage of peace implementation and should be substituted or heavily supplemented by other approaches. • Given the time and resource constraints inherent in third-party peace operations, IHRO efforts to improve human rights conditions in postconflict societies are likely to fail unless those efforts are linked to other tasks within the peace implementation mission. Peace implementation missions, in the words of the Lawyers Committee, “must be tailored to the politics of what is possible.” 82 Clearly, post–civil war situations present some of the most inhospitable conditions for attempting to construct a robust regime of human rights protection. Responsible participation in situations in which human rights are being violated broadly and with impunity demands that IHROs choose their tactics according to the conditions actually prevailing in those situations, and not according to an abstract ideal. In many cases, following this precept will require IHROs to move beyond traditional international human rights enforcement roles and methods toward more flexible, consultative approaches to human rights protection that focus on building and enabling institutions on the ground in postconflict settings. In the vast majority of peace implementation settings, the long-term goal of securing respect for and observation of international standards of human rights will be best served if IHROs adopt an integrative strategy in the short term.

S-9 11/24/09

Years of Negotiationc

HR in Settlement?

HR in Mandate?

HR in Constitution?d

Angola I (1991)

2

3 [1991–1993]

Angola II (1994)

2

ICCPR/ICESCR Strategy During Postimplementation Signatoriese Implementationf Recordg

no [Bicesse]

no

yes [1992: Pt. II]

yes/yes

1

3

3 [1994–1998]

yes [Lusaka]

yes

yes [1992: Pt. II]

yes/yes [1983]

1

3

Bosnia-Herzegovina 3 (1995)

1 [1995]

yes [Dayton]

yes (split)

yes [1993: Ann. 4]

yes/yes [1993/1992]

3

2

Cambodia (1991)

3

5 [1987–1993]

yes [Paris]

yes

yes [1993: Chap. IV]

yes/yes [1992]

2

3

El Salvador (1992)

1

3 [1989–1992]

yes [San Jose]

yes

yes [1994: Tit. II]

yes/yes [1979]

1

2

Guatemala (1990)

1

6 [1990–1996]

yes [Guatemala City]

yes

yes [1985: Tit. II]

yes/yes [1992/1988]

1

2

Lebanon (1989)

2

1 [1989]

no [Ta’if]

no

yes [1990: Pt. I]

yes/yes [1972]

3

2

Liberia (1995)

3

5 [1991–1996]

no [Abuja]

no

NA [susp. 1980]

yes/yes [1967]

3

3 (continues)

Page 261

Coercion in Settlementb

Countrya

4:21 PM

Appendix 9.1: Human Rights Implementation Scorecard

S-9 11/24/09

Years of Negotiationc

HR in Settlement?

HR in Mandate?

HR in Constitution?d

Mozambique (1992)

1

3 [1989–1992]

Namibia (1990)

2

Nicaragua (1987)

ICCPR/ICESCR Strategy During Postimplementation Signatoriese Implementationf Recordg

no [Rome]

no

yes [1990: Pt. II]

yes/no [1993]

1

2

11 [1979–1990]

no [UN]

no

yes [1990: Chap. III]

yes/yes [1994]

1

1

1

4 [1983–1987]

yes [Esquipulas II]

no

yes [1995: Tit. IV]

yes/yes [1980]

1

2

Rwanda (1993)

2

3 [1990–1993]

yes [Arusha]

no

yes yes/yes [1993: Arts. 12–34] [1975]

1

3

Sierra Leone (1996)

3

1 [1996]

no [Abidjan]

yes (split)

yes [1991: Chap. III]

yes/yes [1996]

3

3

Somalia (1993)

3

less than 1 [1993]

no [Addis Ababa]

no

yes [1979: Chap. II]

yes/yes [1990]

3

3

Sri Lanka (1987)

3

less than 1 [1987]

no [ISLA of 1987]

no

yes [1996: Chap. III]

yes/yes [1980]

3

3

Zimbabwe (1979)

2

5 [1975–1979]

no [Lancaster House]

no

yes [1980: Chap. III]

yes/yes [1991]

1

3

Page 262

Coercion in Settlementb

Countrya

4:21 PM

Appendix 9.1 continued

S-9 11/24/09 4:21 PM Page 263

Notes: a. Under each country name is the date of the peace settlement that constitutes the focus of the case study in this project. For each country, the year represents the year of settlement agreement for the particular settlement process analyzed as part of this project, although in some cases it may not represent the most recent agreement. Cases classified as peace implementation successes are indicated by underlining, and cases classified as undetermined at the time of writing are indicated by italics. b. Settlement coercion ratings: 1 = cooperative settlement (i.e., the parties themselves formed the main drive behind settlement); 2 = partially imposed settlement (i.e., at least one of the parties desired settlement and the rest of the impetus came from outside actors); 3 = imposed settlement (i.e., no agreement would have occurred without the pressure of the external actors). c. The tally of the negotiation process includes all negotiations that can be considered part of a single, sometimes extended, effort to end an armed conflict. Thus, in both Liberia and Guatemala, a series of agreements were concluded, building one upon the other, until the effort ultimately failed (Liberia) or can be considered to have succeeded in bringing an end to the conflict (Guatemala). d. In each case the date listed indicates either the date the constitution entered into force, or the date of the most recent constitutional amendment with a substantial impact on the rights of individuals. “Substantial impact” is presumed whenever a new right is added or entirely eliminated from a text, but not necessarily when an amendment further clarifies a right, as was the case, for example, with Article 11 of El Salvador’s constitution, which was amended in 1996 to expanded application of the right to habeas corpus to situations “when any authority attacks the dignity or physical, mental or moral integrity of detained persons” (Decree no. 743). e. The ICCPR is the International Covenant on Civil and Political Rights, and the ICESCR is the International Covenant on Economic, Social, and Cultural Rights. Data for this column are taken from “Status of Ratifications of the Principal International Human Rights Treaties,” May 15, 2000, Office of the UN High Commissioner for Human Rights. f. Implementation strategy ratings: 1 = little or no coercion (military or political) applied in attempt to fulfill the implementation mandate; 2 = moderate coercion applied; 3 = a high degree of coercion applied during implementation. Examples of developments that would move a strategy up the scale range from political appeals among elites (including UN resolutions), to threats to withdraw aid or to withhold future aid, to the actual use of force to enforce the terms of the agreement. At one extreme of this variable is Guatemala, where peace implementation went forward almost entirely on the basis of the agreements and confidence-building measures laid out in various settlement instruments negotiated over a period of years. At the other extreme is Bosnia, where the constant threat of a resumption of the NATO bombing campaign overshadowed the IFOR operation together with the UN-led civilian operation. g. Postimplementation ratings: 1 = good, 2 = moderate, 3 = poor. The emphasis here is on civil and political rights. Examples of practices that would produce a “poor” rating include a pattern of toleration or perpetuation of extrajudicial killings by security forces; arbitrary arrest and detention practices; torture and physical abuse of persons in custody; interference with free speech; governmental obstruction of human rights monitoring and investigations. The data for each evaluation are taken from 1999 Amnesty International country reports, cross-checked where available with 1999 Human Rights Watch country reports.

S-9

11/24/09

4:21 PM

264

Page 264

EVALUATING IMPLEMENTATION TASKS

Notes 1. At the international governmental level, the central actor in the sphere of international human rights is the United Nations. Although the Commission for Human Rights is, formally speaking, the lead agency in the human rights sphere at the UN, other, more specialized agencies have been far more active in the field during peace implementation. These include the High Commission for Human Rights (UNHCHR), the High Commission for Refugees (UNHCR), the International Children’s Emergency Fund (UNICEF), and the Development Programme (UNDP). The UN system is supplemented at the regional level by the human rights activities of organizations such as the Council of Europe, the Organization of African Unity (OAU), and the Organization of American States (OAS). On the international NGO side, the primary human rights players include Human Rights Watch (HRW)/Africa Watch, Amnesty International (AI), the Lawyers Committee for Human Rights (LCHR), Oxfam, and International Alert. 2. See, for example, Amnesty International’s country report series on BosniaHerzegovina, FRY, 1993–2000, and for Rwanda 1994–2000; and Human Rights Watch’s World Report for these same years. 3. Under the adversarial system as it functions in the Anglo-American legal tradition, a “cause of action” is initiated by one party, the plaintiff, and the other party, the defendant, has the right to contest the claim, with each side attempting to present the most convincing case in its favor. A decision in favor of one party based on the “facts” presented is then handed down by a neutral judge or third party. The adversarial system is often criticized for its “winner-take-all” character, which polarizes the parties involved, and drives litigants to present every aspect of a case in the light most favorable to themselves, regardless of the distortions of “truth” this produces for the legal record. 4. Cedric Thornberry, former assistant UN secretary-general and deputy chief of the UN Protection Force (UNPROFOR), argues that “the establishment and maintenance of peace and order is, in the great majority of situations, a pre-requisite for conditions in which human rights can develop in a strong and healthy manner.” Cedric Thornberry, “Peacekeeping, Peacemaking, and Human Rights,” public lecture at Magee College, University of Ulster, February 20, 1995. 5. Of course the business of building peace and the business of building a robust regime for the protection of human rights are hardly mutually exclusive endeavors. The high degree of overlap between human rights and the goals of sustainable peace should be viewed as a source of mutual reinforcement, not as a source of confusion between human rights and other goals, or as a recipe for the probable diminution of human rights in the implementation process. 6. See, for example, Amnesty International, “Kosovo Crisis Recommendations,” EUR/91/99, June 1999. Also, both AI and HRW campaigned against granting a blanket amnesty to the faction leaders in the conflict in Sierra Leone as part of the Lomé peace settlement signed in June 1999. 7. “Sierra Leone Rebels Demand Amnesty,” World African Network, July 7, 1999, www.wanonline.com/news/news7326.html; “Peace Process on Trial: Testing Guatemala’s Amnesty in the Courts,” Lawyers Committee for Human Rights Advisor, spring 1997; and “Chissano Says Stop,” The Economist, June 17, 1999. 8. It should, however, be noted that under international law, grants of amnesty are expressly allowed in the context of settlement of internal conflicts and, some would argue, even encouraged. See, for example, Yoram Dinstein and Mala Tabory, eds., War Crimes in International Law (The Hague: Martinus Nijhoff, 1996). Dinstein finds that a recommendation for amnesty “follows from Article 6(5)

S-9

11/24/09

4:21 PM

Page 265

HUMAN RIGHTS AND SUSTAINABLE PEACE

265

of the Additional Protocol to the Geneva Convention of 12 August 1949, and Relating to the Protection of Victims of Non-International Armed Conflict (Protocol II)” (p. 319). 9. This is hardly surprising, given that in most cases the very individuals likely to top the list for immediate indictment are those seated at the negotiating table and their closest political allies. To expect individuals who would perpetrate such crimes in the first place to then sign away the opportunity to escape punishment for those crimes defies logic. 10. Anonymous, “Human Rights in Peace Negotiations,” Human Rights Quarterly 18, no. 2 (1996): 249–258. See also responses by Francis A. Boyle, “Negating Human Rights in Peace Negotiations,” Human Rights Quarterly 18, no. 3 (1996): 515–516; and Felice D. Gaer, “UN-Anonymous: Reflections on Human Rights in Peace Negotiations,” Human Rights Quarterly 19, no. 1 (1997): 1–8. 11. Seventeen months and thousands of deaths later, the same parties met in the United States to broker a new agreement. The resulting Dayton Agreement that ended the war differed little from the September 1993 agreement on the issue of the de facto ethnic partition. 12. Organizations are generally able to handle complex tasks precisely because they develop standard operating procedures and established repertoires for identifying, classifying, and responding to issues and problems. But these same mechanisms also lead organizations to paint the world using a narrow spectrum of colors—all the more so when the organizational mission is focused on a single issue, such as human rights. Organizational missions, organization-specific forms of knowledge, and organizational structure can all act as blinders to opportunities and hazards present in each new implementation context, constraining artificially the perceived menu of choices for action. 13. Janelle M. Diller, Handbook on Human Rights in Situations of Conflict (Minneapolis: Minnesota Advocates for Human Rights, 1997). 14. See, for example, Binaifer Norowjee, “Joining Forces: United Nations and Regional Peacekeeping Lessons from Liberia,” Harvard Human Rights Journal 8 (spring 1998): 149; and Andrew Clapham and Meg Henry, “Peacekeeping and Human Rights in Africa and Europe,” in Alice H. Henkin, ed., Honoring Human Rights and Keeping the Peace: Lessons from El Salvador, Cambodia, and Haiti (Washington, D.C.: Aspen Institute, 1995), p. 136. 15. A “cooperative” party is one that genuinely desires to reach an agreement to end armed hostilities, and that is willing to enter into at least some level of compromise with other parties to bring about that result. However, parties may be forced to the negotiating table due to pressures exerted by battlefield stalemate, external constraints such as resource exhaustion, or the preferences of outside patron states on which parties are highly dependent. These conditions are generally unfavorable to achieving sustainable peace: as soon as the opportunity for advantage presents itself, the resource shortfall is resolved, or the attention of the patron is diverted, the assumption is that a noncooperative party will prefer to resume fighting. 16. See Ian Johnstone, “Rights and Reconciliation,” in Michael W. Doyle, Ian Johnstone, and Robert C. Orr, eds., Keeping the Peace: Multidimensional UN Operations in Cambodia and El Salvador (Cambridge: Cambridge University Press, 1997), p. 314. 17. See Chapter 13 by Caroline Hartzell, and Chapter 15 by William Stanley and David Holiday, in this book. 18. William Reno, “Privatizing War in Sierra Leone,” Current History 96, no. 610 (May 1997): 227–230; William Reno “The Business of War in Liberia,”

S-9

11/24/09

4:21 PM

266

Page 266

EVALUATING IMPLEMENTATION TASKS

Current History 95, no. 601 (May 1996): 211–215; and John L. Hirsch, Robert Oakley, and Chester B. Crocker, Somalia and Operation Restore Hope: Reflections on Peacemaking and Peacekeeping (Washington, D.C.: U.S. Institute of Peace, 1995). 19. These conflicts are coded in the appendix as settlements of “little” or “moderate” coercion. 20. In Cambodia the uncooperative parties were the Khmer Rouge and the State of Cambodia, and in Bosnia-Herzegovina the uncooperative parties were the leaders of the Bosnian Serbs and the president of the Yugoslav Federation, Slobodan Milosevic. 21. Unlike implementation missions deployed to Namibia and Cambodia, the missions in Liberia and Sierra Leone were UN Observer Missions, UNOMIL and UNOMSIL respectively. 22. Yasushi Akashi, special representative of the UN Transitional Authority in Cambodia (UNTAC), defined “free and fair” elections as having three dimensions. First, to be free and fair, there can be no technical irregularities in the actual conduct of the poll. Second, an election is not considered free and fair if the antecedent campaign is marred by excessive violence and intimidation. A final requirement is that of a substantially level playing field, where all parties, and not merely the party in control of the state apparatus, have access to media and a chance to make their views heard. See Michael W. Doyle, UN Peacekeeping in Cambodia: UNTAC’s Civil Mandate (Boulder, Colo.: Lynne Rienner, 1995), pp. 55–56. However, even in cases where the United Nations has formal control, widespread incidences of human rights violations have been documented in connection with the conduct of elections. For example, in Cambodia members of some political parties were beaten and even killed in the run-up to the election, and on election day Khmer Rouge troops intimidated villagers on their way to vote by shelling them. Still, over 90 percent of Cambodia’s eligible voters turned out over the course of the five-day election. See Doyle, UN Peacekeeping in Cambodia, p. 57. 23. The process by which the guerrilla war between South Africa and the South West African People’s Organization (SWAPO) was brought to a close, allowing Namibia (formerly South West Africa) to gain its independence in 1990, is unique in international law. Beginning in 1960, South West Africa had been designated an “international protectorate” under UN auspices when the Security Council failed to renew South Africa’s mandate to administer the territory in light of the system of apartheid that began to be formalized legally following World War II. Despite its official status, Namibia continued to be administered de facto by South Africa until 1989, when South African authorities agreed to withdraw and to allow a plan for transition that had been carefully crafted over a period of almost a decade to go forward. 24. Clapham and Henry, “Peacekeeping and Human Rights in Africa and Europe,” p. 143. 25. Marie-Joëlle Zahar explains that the effectiveness of local Lebanese NGOs was predicated, at least in the earliest stages of implementation of the Ta’if Agreement, upon their not having any formal international ties to governments or IHROs. She explains that international ties would have immediately opened these organizations to accusations that they were acting as “agents” of an external interest, which in turn would have destroyed their ability to function. Author interview at Stanford University, May 15, 2000. 26. Virginia Page Fortna, “Angola,” draft prepared for this project. 27. An important point to recall is that only the agencies formally part of the implementation mission are formally bound by the mandate. Thus, in all cases gov-

S-9

11/24/09

4:21 PM

Page 267

HUMAN RIGHTS AND SUSTAINABLE PEACE

267

ernmental IHROs will be bound, whereas this is not necessarily the case with respect to nongovernmental IHROs. 28. Again, developments in El Salvador were largely sui generis insofar as the human rights aspects of the peace settlement preceded even agreement to cease-fire. Because the mandate for international verification contained in the San José Agreement formed the original basis for the UN Observer Mission in El Salvador (ONUSAL), human rights provisions, understandably, received considerable elaboration. However, as ONUSAL expanded into more traditional peacekeeping roles, even the elaborate nature of the human rights mandate in the San José Agreement could not prevent the political agenda of the UN mission from overshadowing what were perceived by those on the ground as the more “pure” human rights imperatives. See Improvising History: A Critical Evaluation of the United Nations Observer Mission in El Salvador, report of the Lawyers Committee for Human Rights, December 1995, pp. 15–19. 29. Doyle, UN Peacekeeping in Cambodia, p. 66. 30. See Stephen John Stedman, “Spoiler Problems in Peace Processes,” International Security 22, no. 2 (fall 1997): 5–53. 31. Note that of the sixteen cases covered in this survey, twelve of the fifteen countries had signed both the International Covenant on Civil and Political Rights (ICCPR) and the International Covenant on Economic, Social, and Cultural Rights (ICESCR) before or contemporaneously with the conclusion of the peace settlement in question. Interestingly, the three exceptions fall into the implementation “success” category—Namibia, which acceded to both agreements in 1994, and Mozambique, which acceded to the ICCPR in 1993 and has yet to sign the ICESCR, and Zimbabwe, which acceded in 1991. See chapter appendix, p. 261. 32. Improvising History describes the situation as follows: “The signatories pledged their commitment to human rights standards contained in Salvadoran laws; in the international treaties to which El Salvador is a state party; and with ‘the basic declarations and principles on human rights and international humanitarian law adopted by the United Nations and the Organization of American States.’ In short, nothing substantively new was required of the Salvador government, which in signing the document simply renewed it commitment to its own people and the international community to comply with these national and international standards” (pp. 6–7, emphasis added). 33. Improvising History, p. 7. 34. Surviving institutions, which in many cases include military and other security units, are almost guaranteed to require substantial reform and restructuring before they can be constructively reintegrated into a new government. 35. See Johnstone, “Rights and Reconciliation,” and Leonardo Franco and Jared Kotler, “Combining Institution Building and Human Rights in Guatemala: The Challenge of Buying In Without Selling Out,” in Alice H. Henkin, ed., Honoring Human Rights: From Peace to Justice (Washington, D.C.: Aspen Institute, 1998), pp. 39–70. 36. The tactic of temporary institutional substitution can backfire if social expectations become too high too quickly, and if popular opinion writes off key institutions as useless or irrelevant. A serious hazard associated with overreaching in the human rights sphere, therefore, is the potential that human rights considerations will begin to contribute to social and political instability, rather than to mitigate it in contexts where the human rights record is actually improving. 37. Michael W. Doyle, Ian Johnstone, and Robert C. Orr, “Strategies for Peace: Conclusions and Lessons,” in Doyle, Johnstone, and Orr, Keeping the Peace, p. 383.

S-9

11/24/09

4:21 PM

268

Page 268

EVALUATING IMPLEMENTATION TASKS

38. As long as an implementation force is present, clear limits can be set for the allowable scale of violations, but not once it is gone. In my opinion, it is preferable to allow those institutions that will ultimately bear responsibility for protecting human rights in postconflict societies to test themselves under UN (or other thirdparty) tutelage, rather than bending to pressures for immediate compliance. 39. For example, G. Fitzmaurice writes that “a treaty of peace . . . has no inherently necessary character as a solvent of illegalities which have been committed during the course of war.” The Juridical Clauses of the Peace Treaties (1948) 73 R.A.D.I., pp. 259, 342. 40. Naomi Roht-Arriaza and Lauren Gibson, “The Developing Jurisprudence on Amnesty,” Human Rights Quarterly 20, no. 4 (1998): 844. 41. Alison Des Forges of Human Rights Watch defends this move. She argues the organization’s involvement in gathering information for prosecution dossiers represents not so much a change in direction, but instead merely an increase in the rigor with which information is gathered. Presentation at the Carr Center for Human Rights at the Kennedy School of Government, Harvard University, October 20, 1999. 42. See, for example, Stephen Landsman, “Alternative Responses to Serious Human Rights Abuses of Prosecutions and Truth Commissions,” Law and Contemporary Problems 59, no. 4 (1992): 81–92; and Payam Akhavan, “Justice and Reconciliation in the Great Lakes Region of Africa: The Contribution of the International Criminal Tribunal for Rwanda,” Duke Journal of Comparative and International Law 7, no. 2 (spring 1997): 325. 43. “GA Told Rwanda Tribunal Important for Reconciliation,” Africa News Online, December 8, 1997. See also Juan E. Mendez, “Accountability for Past Abuses,” Human Rights Quarterly 19, no. 2 (1997): 278. 44. See Jose Alvarez, “Crimes of States/Crimes of Hate: Lessons of Rwanda,” Yale Journal of International Law 24, no. 2 (1999): 365; and Martha Minow, Between Vengeance and Forgiveness: Facing History After Genocide and Mass Violence (Boston: Beacon Press, 1998), pp. 25–51. 45. It should be noted that the ICTY and the ICTR have not even taken place in the countries in which the violations occurred. Rather the ICTY has carried out its work while being based in the Hague, Netherlands, and the ICTR while being based in Arusha, Tanzania. 46. To be clear, the concern here is not with the substance of the international human rights standards to be applied, but rather with the rigidity of form those standards assume under the enforcement model. 47. Diane Orentlicher, “Settling Accounts: The Duty to Prosecute Human Rights Violators of a Prior Regime,” Yale Law Journal 100, no. 8 (June 1991): 2537–2615. 48. Author interview with Theogene Rudasingwa, adviser to the vice president and minister of defense on foreign affairs and economic policy, at the Kennedy School of Government, Harvard University, October 29, 1999. 49. The high number of detainees is in part due to the adoption on August 30, 1996, after government consultation with international legal and human rights experts, of a four-tiered national law to cover crimes during and leading up to the 1994 genocide. This framework categorizes suspects as (1) those responsible for planning and leading the massacre, (2) “ordinary” killers, (3) those who wounded without killing, and (4) looters. Some within the Rwandan government criticized this framework as unworkable, precisely because of the time and resources it would demand. They argued instead for a law that would have covered crimes of the first and second categories (roughly 1,000 persons), but that would have relied on alter-

S-9

11/24/09

4:21 PM

Page 269

HUMAN RIGHTS AND SUSTAINABLE PEACE

269

native mechanisms such as truth commissions, or so-called participatory justice (gacaca), for the remainder. It would have also permitted a more rapid repatriation of Rwandan refugees from neighboring states, a factor recognized as having contributed to wider regional instability, particularly in eastern Zaire. See International Crisis Group, “Five Years After the Genocide in Rwanda: Justice in Question,” www.crisisweb.org/projects/cafrica/reports/rw01emaina.html; and December 18, 1996, remarks by Undersecretary for Global Affairs Timothy E. Wirth at the Center for National Policy in Washington, D.C., www.state.gov/www/global/global_issues/ 961218tw.html. 50. Rwandan national tribunals can and do impose the death penalty, whereas death sentences are prohibited according to the international human rights standards that inform ICTR sentencing. 51. Barayagwiza was a former minister of state and cofounder of the radio station that played a key role in inciting the slaughter of Tutsi during April and May 1994. 52. “Rwandan Fury As Genocide Suspect Freed,” BBC Online Network, November 6, 1999. 53. Minow, Between Vengeance and Forgiveness, pp. 58–59. 54. Johnstone, “Rights and Reconciliation,” pp. 318–319, 325. 55. Priscilla B. Hayner, “Fifteen Truth Commissions, 1974–1994: A Comparative Study,” Human Rights Quarterly 16, no. 4 (1994): 609. 56. See Richard Goldstone, “Exposing Human Rights Abuses: A Help or Hindrance to Reconciliation?” Hastings Constitutional Law Quarterly 22, no. 3 (1995): 607–621; and Orentlicher, “Settling Accounts.” 57. In some postconflict societies, such as Guatemala, there was a push from inside to establish a mechanism for truth-telling. See “Accord on the Establishment of the Commission for the Historical Clarification of Human Rights Violations That Have Caused Suffering to the Guatemalan People,” Oslo, June 1994. By contrast, in Cambodia attempts on the part of IHROs and others to initiate a truth commission received little encouragement from the population at large and were subsequently dropped. Stephen P. Marks, “Forgetting ‘The Policies and Practices of the Past’: Impunity in Cambodia,” The Fletcher Forum of World Affairs 18, no. 2 (summer–fall 1994): 21. 58. In functioning Western-liberal political and social systems, demonstration that a legal wrong has been committed constitutes a valuable service to the aggrieved party insofar as it triggers injunctive or remedial responses enforced through the appropriate institutions. 59. The LCHR provides a notable exception to the standard legal enforcement approach, with its practice of working together with violator governments to address and correct the structural and institutional roots of systematic violations of human rights. 60. However, the degree of overlap, while substantial, is not perfect. With integration, the need for special rapporteurs or other specially appointed independent human rights officers and some measure of budgetary independence will not disappear. Even in highly integrated implementation settings, such mechanisms should still exist to ensure that human rights issues do not systematically fall through the cracks. 61. See, for example, Johnstone, “Rights and Reconciliation,” p. 331, for commentary regarding low prioritization of human rights offices, such as the Office of the Ombudsman, during implementation in Bosnia-Herzegovina. 62. See, for example, Kenneth L. Cain, “The Rape of Dinah: Civil War in Liberia and Evil Triumphant,” Human Rights Quarterly 21, no. 2 (May 1999):

S-9

11/24/09

4:21 PM

270

Page 270

EVALUATING IMPLEMENTATION TASKS

294–295; and Andrew Clapham and Florence Martin, “Smaller Missions, Bigger Problems,” in Henkin, Honoring Human Rights, pp. 133–162. 63. Interview with anonymous senior official in ONUSAL. Improvising History, p. 20. 64. Improvising History, pp. 20–21. 65. In a reorganization effort in 1997 the United Nations sought to correct the situation by recommending that human rights functions be distributed across the UN’s four core missions—peace and security, economic and social affairs, development, and humanitarian affairs—and by promising to “enhance [the UN’s] human rights programme and fully integrate it into the broad range of the Organization’s activities.” See UN Document A/51/950, Secretary-General’s Report “Renewing the United Nations: A Programme for Reform,” July 16, 1997. 66. Toward this end, on November 5, 1999, Mary Robinson, UN High Commissioner for Human Rights, and Bernard Miyet, UN Undersecretary-General for Peacekeeping Operations, signed a Memorandum of Understanding between the Office of the High Commissioner for Human Rights and the Department of Peacekeeping Operations for cooperation in peace implementation. Specific points of cooperation included: “Planning, design and establishment of human rights components of peace-keeping operations”; “Institutional arrangements for human rights components of peace-keeping operations”; “Reporting and public statements”; and “Human rights training for staff of peace-keeping operations.” 67. Doyle, UN Peacekeeping in Cambodia, pp. 45–46. 68. See Clapham and Martin, “Smaller Missions, Bigger Problems.” 69. The situation is not, however, entirely one-sided: IHROs have resources and skills within their normal repertoires that are in high demand by implementation missions at both the strategic and the operational level. 70. There is often a strong tension running between the providers of humanitarian aid and those whose primary concern is with developing the means to protect human rights over the longer term. In the drive to “stop the dying,” humanitarian organizations may unwittingly supply the perpetrators of violence with the means to continue their activities. Conversely, the finger-pointing of IHROs can make it difficult for humanitarian relief organizations to maintain the impartiality necessary to ensure access to noncombatants on all sides of the conflict. 71. Cain, “The Rape of Dinah,” p. 297. Cain goes on to argue: “Instead, these organizations tended to advocate that abstract, ideal standards be met and to articulate aspirational human rights goals that had no hope of actually being implemented in the real world. The moral authority of the message was thereby neutered, and relevant parties could easily and blithely ignore the toothless critique in toto, without repercussion.” 72. NGOs from the South are significantly more likely than Northern NGOs to be involved in promotion of human rights through national institutions and through institutional reform. Southern NGOs are also more likely to be involved in national judicial processes, drafting national legislation, election monitoring, and implementation of international human rights standards. They also devote more attention to economic and social rights, such as the right to development, the right to a clean environment, and promotion of cultural rights. See Jackie Smith, Ron Pagnucco, and George Lopez, “Globalizing Human Rights: The Work of Transnational Human Rights NGOs in the 1990’s,” Human Rights Quarterly 20, no. 2 (1998): 388–390, 395. 73. Monitoring is undoubtedly a core human rights activity, and for this reason even IHROs that agree to cooperate with implementation missions need to retain independent reporting functions within the mandate. But monitoring alone is insuf-

S-9

11/24/09

4:21 PM

Page 271

HUMAN RIGHTS AND SUSTAINABLE PEACE

271

ficient to bring about change quickly in postconflict settings. Without mechanisms capable of using the information gathered by monitors to correct for shortfalls, unanticipated developments, and outright mistakes in postconflict governance and institution-building, the most reliable information in the world is ineffective. 74. Franco and Kotler, “Combining Institution Building and Human Rights in Guatemala.” 75. These pressures and tensions may be heightened when the completion of a specific mandate task (e.g., holding an election) is linked to the exit strategy of the implementation force, as was the case in Bosnia, Cambodia, Mozambique, and Angola. 76. Improvising History, pp. 88–89. 77. Private communication, Stephan P. Marks, head of the Human Rights Education, Training, and Information Section of the Human Rights Component of UNTAC, and human rights adviser for the Asia Foundation/Cambodia. 78. This constitutes an excellent example of where individual IHROs may need to separate their opinions regarding an arrangement between parties to the peace settlement regarding “housecleaning” from the presettlement regime, and the practical necessity of establishing working police and judicial systems. Although a clean slate would clearly be preferable, in reality, as Charles Call and William Stanley describe in Chapter 11, postsettlement judicial and security institutions are likely to contain members who played less-than-untarnished roles in the conflict. However, by working within the implementation fold, IHROs can attempt to educate such individuals, while at the same time monitoring closely their behavior and, if necessary, reporting continued human rights breaches to the government and the international community. 79. For example, widespread violence, even brutal violence, is rarely “senseless”—that is, it is rarely committed without some instrumental objective. What that objective is, and how it is connected to particular atrocities, may be more apparent to local actors than to outside observers and interveners. 80. Sidney Jones, “Human Rights: The Organic Growth,” Far Eastern Economic Review 156, no. 24 (June 17, 1993): 23. 81. Improvising History, p. 20. 82. Ibid., p. 88.

S-9

11/24/09

4:21 PM

Page 272

S-10

11/24/09

4:21 PM

Page 273

10 Refugee Repatriation HOWARD ADELMAN

The conventional wisdom of most refugee experts holds that there is a necessary connection between forging and implementing a peace agreement and ensuring the successful return of refugees.1 Peace depends on refugee repatriation, and every peace agreement must provide for it, or so it is widely believed.2 For example, in Central America in August 1987, the Esquipulas II Accord claimed that there would be “no lasting peace without initiatives to resolve the problems of refugees, returnees and other displaced persons.”3 Just over ten years later, in the African context, the intimate connection between peace accords and refugee repatriation was reiterated, but with a slight twist. Refugee repatriation was said to be a sign that peace was in process. In this chapter I scrutinize such claims and conclude that refugee repatriation is neither a necessary nor a causal condition of peace. A peace agreement need not even address the refugee issue. Further, however horrifying this may appear to refugee scholars like myself, there can be peace without the refugee issue even being addressed. It may be that the refugee issue cannot be left in limbo; otherwise refugee warriors will develop. But as long as the refugee issue is solved one way or another, sometimes quite independently of the peace agreement, the refugee issue need not even be addressed in a peace agreement. Nor is peace a necessary or a causal condition of successful refugee repatriation, however counterintuitive that may first appear. Peace may be in process without refugee repatriation, and refugee repatriation need not even be a manifestation of peace. Based on the case studies examined by this project, as well as other cases of cessation of hostilities that produced refugees, in this chapter I question any causal, necessary, signifying, or even inherent type of relationship of refugee repatriation to peace agreements. In keeping with the 273

S-10

11/24/09

274

4:21 PM

Page 274

EVALUATING IMPLEMENTATION TASKS

contention of the book that the implementation environment affects the prospect of realizing peace, I argue that the type of war that is fought and the type of peace that is sought matter greatly for whether repatriation affects the prospects for peace. I begin the chapter by sketching various claims that are made about the relationship between repatriation and peace implementation. I then examine these general claims in light of the case studies of this project, and find that the issue of refugees is mostly an afterthought or, when addressed, is tertiary to overall implementation success. I then attempt to identify relationships between refugee repatriation and peace implementation that the case authors ignored, and show that there are certain cases where refugees are crucial to war termination.

Peace Agreements and Repatriation: Soft and Hard Positions Distinctions between four positions (what I will call Soft I, Soft II, Hard I, and Hard II) on the relationship of implementing peace agreements to the repatriation of refugees may be helpful. There is a difference between the view that refugee repatriation is a sign of the movement toward peace and the conviction that there is a necessary connection between refugee repatriation and peace. The first espouses a soft connection between peace and repatriation, the second a hard connection. The soft position is exemplified by the following: “The return and reintegration of people who have been uprooted or affected by violence is an important manifestation of the process whereby national protection is restored and human security reinforced.”4 But this position (which I have labeled Soft II) must be distinguished from an even softer position (Soft I). In Soft II, refugee repatriation is a manifestation of peace. In Soft 1, resolving the refugee issue, however it is solved, is a manifestation of peace; refugees need not be repatriated. In Soft II, the issue of repatriation (and usually reintegration as well) must be addressed in a peace agreement. In Soft I, only the resolution of the refugee issue, for which repatriation is only one option, needs to be addressed in a peace agreement. However, the hard view remains the overwhelming conviction and presupposition of those who connect refugee repatriation to the implementation of peace agreements. Resolving the issue of refugees and displaced persons appears to be a core element in the terms and successful implementation of a peace agreement. As one scholar summarized the belief: One of the goals of modern peace accords is to create a stable environment for the return of refugees. The United Nations High Commission for Refugees (UNHCR) has identified the following dangers, problems and uncertainties which must be addressed in order to facilitate the repatria-

S-10

11/24/09

4:21 PM

Page 275

REFUGEE REPATRIATION

275

tions process: Physical security; social and psychological security; legal security; and material security. Ideally a peace accord should address each of these problem areas.5

The hard relationship between peace agreements and refugee repatriation can be disaggregated into two different positions. In Hard I, peace and refugee repatriation are necessary conditions of one another. In Hard II, they are viewed not only as necessary conditions of one another, but also as causes of one another. That is, refugee repatriation is not only a necessary condition of peace, but it helps bring about that peace. On the other hand, peace is not only a necessary condition of refugee repatriation, but is a cause of refugee repatriation. To summarize, Soft I: (1) Resolving the refugee issue is a sign that peace is really in place. (2) In a proper peace accord, the refugee issue must be addressed. Soft II: (3) Successful refugee repatriation is a sign that peace is really in place. (4) Hence, the issue of refugee repatriation must be addressed in a proper peace accord. Hard I: (5) Unless there is refugee repatriation, there will be no lasting peace. (6) Peace is a necessary precondition for successful refugee repatriation. Hard II: (7) Refugee repatriation alone is an essential element in the social construction of peace.6 (8) Peace is a cause of successful refugee repatriation. I have studied refugee issues for many years and written extensively on the subject, and always assumed that there could be no real peace without dealing with the refugee issue. My studies of Palestinian refugees, however, did not lead me to believe that refugee repatriation necessarily had to be a part of a peace agreement. Still, the refugee issue could not be ignored. There would be no permanent peace unless the refugee issue was addressed as part of the peace agreement. My perspective on the relationship between peace and refugee repatriation is exemplified by Soft I. The most important reason could be drawn from the Palestinian and other cases. If a solution were not found for the refugee issue, there would be a good chance that refugees would evolve into refugee warriors, perpetuating a cycle of violence.7 If the refugee issue is not resolved successfully, the refugees may resort to violence directed at their former enemies and/or the government. At the very least, they often resort to violent criminal activity. For example, in the Lebanese civil war, refugees, and later the displaced, were the source of recruits for the militias that destabilized the state. “The strongest militias recruited their members from among the refugee peoples who were themselves the product of the Lebanese Civil War and other Middle Eastern wars.”8 This was also true of the disaffected and unemployed youth in Somalia. Similarly, in Bosnia, refugees, supported by the respective receiving states, were recruited and mobilized into military units.9 But the connection between peace and repatriation meant that the refugee problem had to be addressed, not that the refugees had to be repatriated.

S-10

11/24/09

276

4:21 PM

Page 276

EVALUATING IMPLEMENTATION TASKS

This chapter provides evidence that the hard positions (either I or II) are not supported by empirical facts. Refugee repatriation is neither a necessary nor a causal condition of peace. Peace is neither a necessary nor a causal condition of successful refugee repatriation. The evidence does not even support the Soft II position. This does not mean that refugee repatriation cannot be an important aspect of some peace processes and in fact serve as a sign that peace is in process. But peace may be in process without refugee repatriation, and refugee repatriation need not even be a manifestation of peace. However, based on my analysis of the cases in this study, I find that even my former Soft I position is incorrect. A peace agreement need not address the refugee issue, and violence can be stopped without the refugee issue being addressed.

The Marginality of Refugee Repatriation to Peace Accords Of the case studies of civil wars that were part of the project, eight indicated that the refugee issue was totally marginal to the outcome of the peace agreement. In fact, little attention was paid to the issue of refugees in the scholarly case studies of peace implementation with respect to Angola, El Salvador, Lebanon, Liberia, Sierra Leone, Somalia, Sri Lanka, or Zimbabwe, though the agreements sometimes made provision for refugee repatriation. There seems to be no necessary connection between repatriation, peace, and peace agreements. In the Mozambican civil war, a proposal was made to include repatriation as a major part of the peace agreement; it was rejected in the peace negotiations. 10 Yet Mozambique is considered to have been one of the few successful peace agreements to end a civil war. Further, in this and two other cases—Guatemala and Nicaragua—scholars make clear why refugee repatriation was not essential to the success or failure of the peace agreement. Many refugees repatriated prior to the signing of the agreement.11 Even after the peace agreement was signed, return depended on implementation of certain key factors of the agreement, particularly the provision of security.12 In the long term, the key factor for analysts of the peace process was not repatriation, but land and political reform.13 Sometimes refugee repatriation and peace are conjoined. In the case of Namibia, under the provisions of the peace agreement worked out between 1978 and 1988, all refugees were to be welcomed back into the country at specific entry points. In this and two other of the remaining five cases, where the repatriation was essential rather than apparently marginal to the implementation of a peace agreement, the repatriation was intimately and directly tied to the success of the election or referendum provided for in the

S-10

11/24/09

4:21 PM

Page 277

277

REFUGEE REPATRIATION

agreement. Repatriation was not a goal in itself but a means of establishing a legitimate successor government. This was the case in Cambodia, where subsequent events subverted the election results. It was certainly the case in Namibia. And in Western Sahara, the determination of the membership of the polity was the central issue in order to determine who was eligible to vote in the referendum on independence. Sorting out that issue was, for the longest time, considered a precondition for refugee return. In only two peace agreements could the return of refugees in itself be considered a central issue. In Bosnia-Herzegovina, refugee repatriation was at the heart of the agreement. Yet a majority of the refugees were not able to return to their homes if those homes were located in areas where the returnees would be a minority. Still the peace held—in the sense of an absence of resumption of the civil war. In Rwanda, the war continued after the breakdown in the peace agreement, a central issue of which was the return of the refugees. Those refugees did repatriate, because of who won and not because of the peace agreement. Further, another, even larger group of refugees (almost 1.5 million) fled in the renewed conflict. Separating those refugees from the former government forces and militias that controlled them became the new problem. The relationship in the studies between repatriation and peace agreements can be summarized in Table 10.1.

Table 10.1

Refugee Repatriation and Peace Agreements, 1980–1997

Relationship

Number of Cases

Cases

Explanation

Repatriation marginal

8

Angola El Salvador Lebanon Liberia Sierra Leone Somalia Sri Lanka Zimbabwe

None. Resettlement. None. None. None. None. None. None.

Disjunction

3

Guatemala Mozambique Nicaragua

Occurs before and long after. Reintegration rather than repatriation critical.

Conjoined

3

Cambodia Namibia Western Sahara

Related to legitimization of government.

Central not carried out carried out

2 Bosnia Rwanda

Reestablish multiethnicity.

S-10

11/24/09

278

4:21 PM

Page 278

EVALUATING IMPLEMENTATION TASKS

In almost all the cases where the authors of the studies, by implication and inattention, considered the refugee issue marginal to the success of the peace agreement, no explanation was offered for that marginality. Charles Call’s study of the peace agreement ending the El Salvador civil war was the exception (see Chapter 14). According to Call, the refugees did not need to be repatriated, either because they were permanently settled elsewhere, or because many had repatriated prior to the signing of the agreement as had been the case somewhat in three of the other civil wars— Guatemala, Mozambique, and Nicaragua. Therefore, rather than the peace agreement being a condition of refugee repatriation, the nonrepatriation of some refugees could possibly have made the implementation of peace easier. In sum, the refugee issue was not central to either the success or the failure to implement the peace agreement even in the two cases where the refugee issue was central to the war itself. In some cases, most peace agreements were silent on the issue of refugees (e.g., Liberia). In other cases, the studies themselves were silent on whether the agreement even made any provision for refugees (e.g., Sri Lanka). Even in cases where the agreements and studies both mention refugee repatriation (e.g., UN Resolution 814 regarding Somalia), the studies provided no analysis of its import or consequences. Where discussion is provided, it is generally terse and mentioned as a factual aside rather than perceived as a key to understanding the core of the peace agreement and its effects on implementation strategies. If refugee repatriation is marginal to the implementation of peace agreements, why is this so? One possible answer is that refugees, as long as they remain legitimate refugees, are perceived as victims rather than perpetrators of the violence. The key issue in maintaining the peace, then, is stopping the perpetrators of violence, not assisting the victims, as laudable and humanitarian as the latter task is. Refugees would certainly be the beneficiaries of the cessation of violence just as they were its victims, but qua refugees they have no role to play in whether the violence stops. However, before we discard either of the hard or soft theses on the connection between repatriation and the implementation of peace agreements in civil wars and try to explain the disconnect between scholars of refugees and scholars of civil war termination, we must ask: What evidence might falsify or at least raise questions about the conclusion we have just drawn?

Probing Deeper into the Relationship Between Refugees and Peace Implementation Numbers alone indicate that refugee repatriation should be central to implementing a peace agreement. According to Richard Black and Khalid

S-10

11/24/09

4:21 PM

Page 279

REFUGEE REPATRIATION

279

Koser, at least 12 million refugees were repatriated in the 1990s.14 In 1992 alone, in twenty operations, 1.7 million self-repatriated refugees were assisted and 720,000 were directly repatriated under UNHCR auspices, mostly in South Africa, Cambodia, the Horn of Africa, and Afghanistan.15 The June 1990/April 1991 settlement plan for Western Sahara provided for “the return, within six weeks, of about 120,000 Saharan refugees (at least half the population) from Algeria and Mauritius starting eleven weeks after D-Day under UNHCR supervision.”16 The 1992–1994 peacebuilding operation in Mozambique involved the repatriation of 1.6 million refugees.17 Despite the scale of these numbers, Rosemary Preston suggests that the percentage of returnees to refugees has actually declined in the late nineties.18 How can this be? Large numbers of refugees and displaced persons are a by-product, if not an actual intended product, of violent conflict.19 In Bosnia, for example, by the time the Dayton Agreement was initialed, half of Bosnia’s prewar population of 4.4 million had fled their original homes, either to live as refugees outside the country or as internally displaced persons elsewhere in Bosnia. 20 The Sierra Leone war of the 1990s produced 400,000 refugees (9 percent of the population); one-third of the 4.6 million population was internally displaced. In turn, Sierra Leone hosted 69,000 Liberian refugees while Guinea held 225,000. By the second phase of refugee production, 1992–1995, 500,000 had fled Liberia to Côte d’Ivoire as well as Guinea and Sierra Leone; by 1995 half the population of Liberia had been displaced.21 In Angola the numbers of refugees and internally displaced escalated from the original 425,000 refugees and 800,000 internally displaced recorded in 1991 to over 1 million in each category in 1999. The Cambodian conflict produced 500,000 refugees. El Salvador’s civil war produced 1 million refugees and an additional 750,000 displaced. The Nicaraguan civil war resulted in 500,000 refugees and another 360,000 internally displaced; the half million refugees represented 12 percent of the population. In Guatemala, the 1 million internally displaced overwhelmed the 150,000 who managed to escape the country to seek asylum elsewhere. In Namibia, there may only have been 44,000 to 90,000 refugees, but in Western Sahara, at least half the population became refugees. In Sri Lanka, there were 200,000 refugees, and in Sierra Leone, 400,000. Rwanda had 350,000 and almost 1 million internally displaced before the signing of the Arusha Agreement. A new batch of at least 1.5 million refugees fled the country after the genocide. After the fall of Mogadishu in 1991, in the northern part of the Ogaden, there were 594,000 Somali refugees and 217,000 displaced out of a total population in the region of 1 million. In light of these proportions, how could there be peace unless such large numbers of individuals are repatriated and successfully reintegrated into their homes? Perhaps we need to move beyond simple formulations of the roles that refugees do or do not play in peace processes.

S-10

11/24/09

280

4:21 PM

Page 280

EVALUATING IMPLEMENTATION TASKS

A survey of a week (from July 31 to August 6, 1999) indicates how refugees can return and be produced at the same time as peace agreements are being forged and implemented, or are being sabotaged and undermined. While a peace agreement in the Sierra Leone civil war held, refugees were not repatriating, perhaps wary of the depth of commitment of the various parties to a peaceful resolution of the conflict. Similarly, while refugees in Sierra Leone were not repatriating, in Guatemala, many years after the civil war had ended, a final group of Guatemalan refugees who had lived for sixteen years in refugee camps in Mexico’s Yucatan peninsula and Chiapas returned home under UNHCR auspices. Similarly, in the same period, UNHCR, Thailand, and Laos agreed to the final return of 1,300 Lao refugees still in Thailand. Thus, refugee repatriation often does not occur following the cessation of hostilities, but may occur many years later. Refugees need not return when or in order for hostilities to cease, but may return years later. Refugees may also repatriate, or be forced to repatriate, before the cessation of hostilities. As an example of the latter, in the beginning of August 1999 the Thai government determined that, despite an ongoing civil war, 250,000 Burmese refugees would be repatriated over the next three years.22 If some refugees repatriate many years after a war is over and others return or are forced to return even before a war ends, some wars never seem to end and refugees continue to be produced. 23 At the end of July 1999, Malanje in Angola was packed with 130,000 people living in appalling conditions; they had been displaced by the resumption of civil war. At the same time, Namibia was itself suffering from renewed violence as a result of an allegedly Angolan rebel–sponsored secessionist movement led by Mishake Muyongo, head of the Caprivi Liberation Army in Northern Namibia’s Caprivi Strip.24 As refugees repatriated to Kosovo as a result of the victory of NATO forces, 160,000 Serbs and Roma fled the same area while the Albanian Kosovars returned. In the DRC (Democratic Republic of Congo), in the same sample period, civil war continued with fresh clashes between Balendu and Wagerepue, with thousands of homes torched and thousands of new refugees produced just as the Lusaka Peace Agreement was being negotiated; in the interim, 95 percent of the displaced people from South Kivu returned home before the agreement was concluded. However, at the very same time and in the very same region where these returnees were coming home, violent incidents instigated by the Interahamwe militia, a major perpetrator of the 1994 genocide in Rwanda, produced more refugees. In another region of the DRC, again in the same period, fighting along the Angola-DRC border produced a new, large outflow of refugees from the DRC. In another region of the DRC, 30,000 refugees had recently fled into Gabon.25 How does one make sense out of this chaos and confusion? Four major

S-10

11/24/09

4:21 PM

Page 281

REFUGEE REPATRIATION

281

themes emerge from these observations and are discussed widely in relationship to refugee repatriation. First, do refugees enjoy a right of return?26 Second, is the return voluntary or not?27 Third, is the return organized by the international community or is it “spontaneous”; more precisely, do the refugees instigate and organize their own repatriation (self-repatriation), or are they repatriated under the auspices of some international organization?28 Though none of these issues appear to have a direct concern on how repatriation and peace agreements to end civil wars affect one another, the discussion of the latter issue will have some bearing on these three major themes. Further, as we shall see, some misconceptions about these three themes certainly affect the formulation and implementation of some peace agreements. A fourth theme, whether refugees return during or after a conflict, does have a direct and obvious bearing on the relationship of repatriation and peace agreements.29 This project is focused on refugee repatriation during the implementation of a peace accord. So we must differentiate between those cases where the repatriation takes place prior to the signing of the accord and those that follow such a signing. If repatriation takes place after a peace accord (sometimes many years after), is this repatriation a result of the peace accord? Many refugees wait in camps until they feel that the peace is truly secure; the peace accord in itself provides no such assurance.30 Right of Return The right of refugee return is specified in some peace agreements. Part 5 of the Cambodian peace agreement recognized the right of return of the refugees to Cambodia, where they could live in safety, security, and dignity, free from intimidation and coercion of any kind. The Lebanon Ta’if Agreement sets out the right of each person displaced to return to “the place from whence he was displaced,” but, as Judith Harik noted, the provision occupied a second-tier status in the agreement.31 Yet the right of return has only an indirect connection to the successful implementation of a peace agreement. In fact, insisting on a right of return to homes and not just to homelands complicates and makes more problematic both the making of peace and the ability to enforce the right and is generally only specified when return is problematic, normally in situations where the removal of part of the population was either a specific war aim or a welcomed even if unintended consequence of a conflict between two ethnic groups. Voluntary Return The issue of voluntary return also has an indirect bearing on the relationship of refugee repatriation to the implementation of peace agreements. When they focus on repatriation, peace agreements usually affix the modifier

S-10

11/24/09

282

4:21 PM

Page 282

EVALUATING IMPLEMENTATION TASKS

“voluntary” to that repatriation exercise even when the repatriation is promoted, induced, and in some cases even forced. For example, in the movement of Mozambicans back from South Africa after the peace agreement was signed, the UNHCR resorted to phasing out food rations distributed in South Africa to encourage return. In other cases, food-for-work programs and efforts to make camps less hospitable are techniques used to induce repatriation.32 A great deal of evidence suggests that the Cambodian repatriation was induced and not strictly voluntary.33 Thus, the norm established in the agreement and the actual practice may be at odds. More important, the enunciation of the norm may complicate the return process, as was evident in the case of the return of Rwandan refugees after the genocide. However, induced return sometimes exacerbates a conflict. In the repatriation of Sri Lankan refugees between 1987 and 1989, it is unlikely that many refugees wanted to return (except those who volunteered to fight).34 Did pressure to return make the possibility of peace less likely? Tamils with no options on where to go became more desperate, more determined, and less inclined to compromise on the principle of self-governance. In some cases, individuals have no choice whatsoever. They may be forced to return to their home countries or other countries following a peace. Palestinians from Kuwait went to Jordan following the Gulf War, with no provisions or international assistance.35 In such cases, refugees are produced as a result of the “peace” even when those individuals may not have been de facto refugees any longer, since they had resettled in that host country and had been living there in relative security for years. Pakistan wanted to induce repatriation of Afghan refugees, but lacked the de facto power;36 the 1989 UN Plan of Action to foster repatriation was ineffective.37 Insecurity in the homeland inhibited return. In 1990, 473 repatriation projects and subprojects were launched with little result.38 Yet in 1992, 1.4 million Afghan refugees suddenly did return. However, the continuing interethnic fighting among the rebel groups that overthrew the Najibullah government inhibited the return of several million more remaining in Pakistan.39 Repatriation: Organized or Spontaneous? Most peace agreements that refer to refugees assume that the international community will organize the repatriation. In the case of refugees from Western Sahara, the UNHCR opened a field office in Tindouf in July 1996 with a budget of $50 million to provide documentation and reception centers for returning refugees and to monitor events following repatriation. Not one refugee returned, given the political difficulties that arose between the parties to the conflict in agreeing to an identification process. Most returns are in fact spontaneous.40 They precede or immediately

S-10

11/24/09

4:21 PM

Page 283

REFUGEE REPATRIATION

283

follow the signing of an accord.41 In Mozambique, for example, some repatriation preceded the conclusion of a peace agreement. As Koser puts it, “Even before the Peace Accord, quite large scale self-repatriations were estimated to have occurred from Malawi. The Government of Mozambique estimated that by the end of 1989 up to 200,000 refugees had repatriated.”42 The role of the UNHCR in Mozambique, according to most scholars, was a failure with respect to the official repatriation program.43 As Chris Dolan notes, “the majority of refugees who returned from other countries to Mozambique did so without UNHCR assistance.”44 Overall, of 1.7 million refugees, only 378,000 (22 percent) were repatriated under UNHCR auspices.45 Refugees may return without assistance or guarantees of protection, more or less as per their original flight. Or refugees may organize their own return on a systematic basis, as in El Salvador. There, the refugees who were repatriated were self-repatriated, but used the organized international community as logistic aids for travel and food en route.46 The fact that the process was entirely self-governed was an unusual feature of the El Salvador repatriation. Possibly, the feature of self-organization helped provide one of the bases for self-governance and democracy. Then again, the international community may be used to provide witnesses and enhance security in the repatriation process. In the Guatemalan refugee repatriation, NGOs played a major role as observers of safe return through an international accompaniment program and emergency response network to stop acts of violence, intimidation, and injustice. The main decisionmakers in repatriation are the refugees themselves and not always or even mainly the international peace brokers or the parties to the conflict. Peace agreements rarely make provision for this fact or for the suddenness of refugee movements and the calculus refugees make on relative risks.

Different Senses of Peace, Different Types of Civil Wars Perhaps the issue of refugee repatriation is central to a peace agreement only in those cases where the issue of refugees played a central role in the continuation of the conflict, such as the Israel/Palestinian fifty-year war, the Rwandan civil war, and the secession wars following the breakup of Yugoslavia. In other civil wars, where refugees are by-products of the violence, but not central to the original or continuing conflict, the success or failure of their repatriation might not be central to the success in implementing a peace agreement. This might be the case even though a peace agreement might, understandably, provide for the repatriation of those refugees.

S-10

11/24/09

4:21 PM

284

Page 284

EVALUATING IMPLEMENTATION TASKS

In keeping with this project’s emphasis on understanding how different environmental factors influence the implementation of peace agreements, evidence discussed above, as well as derived from the case studies, suggests that two variables play a critical role in determining the importance of refugee repatriation for peace implementation: the type of war that produced the refugees, and the type of peace intended by the agreement. Wars over People vs. Wars over Governments The issue of repatriation in connection with the implementation of a peace agreement is closely tied to the type of civil war fought. The most basic difference is between wars fought between peoples vying for sovereignty in a state, versus civil wars simply fought over control of the government of a sovereign state (see Table 10.2). The former are wars fought over people. One subtype of such wars is wars based on inclusion and exclusion, for example, the wars in Bosnia-Herzegovina, Rwanda, and Kosovo. A second subtype of such wars involves conflicts over the control of the governing structure where two or more ethnic groups are fighting for either control of the state (as in Lebanon) or secession from the state (as in Sri Lanka). The parties, however, have not taken that extra step of defining the body politic to exclude its opponents totally by means of mass expulsions or, in the extreme, genocide. Repatriation of refugees is a central issue when the conflict is over the

Table 10.2

Peace Implementation Cases and Type of War, 1980–1997

Type of War Ethnic wars Anti-people wars of exclusion Pro-people wars Power wars (wars fought for political control only)

Cases

Bosnia-Herzegovina (ethnic cleansing) Rwanda (genocide) Lebanon (segregationist) Sri Lanka (secessionist) Angola Cambodia El Salvador Guatemala Mozambique Namibia Nicaragua Sierra Leone Somalia Western Sahara Zimbabwe

S-10

11/24/09

4:21 PM

Page 285

REFUGEE REPATRIATION

285

definition of who constitutes the sovereign people and/or the territory that people may govern. Civil wars over membership and sovereignty issues are wars targeted against members of the body politic. In wars of exclusion, a specific war aim is the elimination of some people as part of the polity by creating separate, uni-ethnic states or ethnic enclaves, or undertaking ethnic cleansing and genocide. In anti-people and pro-people civil wars, refugees are a direct consequence of the violent conflict rather than merely a byproduct. A central war aim is to produce the refugees. Refugee repatriation is presumed to be only an issue of management of the process when the war is over. This presumption is false. In anti-people and pro-people wars, repatriation is a central issue in the peace itself, for it is the definition of the sovereign people constituting the body politic that is at stake. There is thus a large qualitative difference on the issue of repatriation between anti-people wars of exclusion or inclusion, pro-people wars over control, and the variety of civil wars fought over governance. In the latter type, repatriation itself is not central to implementing any peace agreement as in anti-people wars, though sentiment is important in fostering or undermining political support for one party or another. The type of war has a direct impact on the four themes of repatriation—the right to return, voluntary repatriation, organized or spontaneous return, and whether refugees return before a conflict is over. In civil wars against people fought over the constituency of the body politic, if the victorious parties are exclusionists in defining the body politic, the powers in control will not grant any right to return whatever the peace agreement says. Perhaps they could be coerced to do so by more powerful forces willing and able to commit military power to undertake such a task, as in Kosovo, but this is not something observed in any of these case studies. There are simply very few opportunities for any significant voluntary repatriation to the original homes of the refugees no matter how hard the international community tries to facilitate the return of refugees. In most cases, the best that can be achieved is repatriation to the homeland in areas where the power in control belongs to the same ethnic group. People Wars and Refugees: Cases The dynamic interplay between the aims of the warring parties in wars fought over people, the negotiation and implementation of peace agreements, and the role of refugee repatriation can be seen in the cases of Rwanda, Bosnia-Herzegovina, Lebanon, and Sri Lanka. Rwanda. In the case of Rwanda, the issue of refugee repatriation was central. In the Mwanza regional meeting in Tanzania on October 17, 1990, following the invasion of Rwanda by the Rwandan Patriotic Front (RPF),

S-10

11/24/09

286

4:21 PM

Page 286

EVALUATING IMPLEMENTATION TASKS

the resolution of the refugee issue was put forth as a key issue. That problem had been left unresolved since hundreds of thousands of Tutsi fled between 1959 to 1966.47 Before 1991, President Juvenal Habyarimana of Rwanda claimed that the population density of Rwanda and the shortage of land meant that the refugee issue needed to be resolved in an international context. Only in early 1991 did Habyarimana begin to shift in his rhetorical response to the refugee issue. Although he had agreed in the Dar es Salaam Conference of February 1991 to remove obstacles to voluntary repatriation and to give an amnesty to refugees who had allegedly committed crimes against the state, he continued to see the resolution of the refugee issue in a regional context. This was unlikely to be realized since Habyarimana made easing of restrictions conditional upon Uganda, Burundi, Tanzania, and Zaire all agreeing to naturalize and integrate those refugees who chose not to return. The refugee issue continued to be a central theme in the peace negotiations. Between March and June of 1992, the repatriation of the refugees was one of the major items, but it remained linked within a larger context of settlement in countries of first asylum. By October 1992, the protocols on power sharing in the October 30 agreement set repatriation of refugees as one of the primary goals for the transitional government. Finally, the Arusha Accords included a protocol on the repatriation of refugees, signed on June 9, 1993, that referred to the inalienable rights of refugees to return home to the region of their choice. However, the peace agreement was not implemented. The war resumed, and the RPF won. The government fled into exile with about 1.5 million new refugees. The victors in the civil war believed in an integrated vision of membership in the state. That is, membership of the state was defined as Rwandan rather than Hutu or Tutsi. The enemies of that idea held to a belief that the essential membership of the state was reserved for Hutu; Tutsi, if they were even to be tolerated, were to be restricted to second-class status in the state. The war fought until the Arusha Peace Agreement was signed in August of 1993 had as a central component these two visions. When the accords were signed between the invading forces, led by refugees wanting to return and moderates who accepted the integrated vision, extremists who did not agree, and who held to a vision of the state that was essentially Hutu, set out to undermine the peace. When the accords were on the verge of putting in place a new government based on that agreement and integrated vision, the extremists staged a coup and initiated the genocide of the Tutsi. When the extremist forces were defeated, they not only went into exile, but also led a large group of the Hutu population into exile with them. The repatriation of those refugees became part of the fundamental conflict. Would they remain in exile as part of the extremist camp unequivocally opposed to the integrated vision of the Rwandan

S-10

11/24/09

4:21 PM

Page 287

REFUGEE REPATRIATION

287

state, or would they return to live with an integrated definition of the sovereign people of Rwanda? In 1994 the bulk of the Tutsi refugees returned with the victory of the RPF. The leaders of the RPF espoused an inclusionist ideology. Nevertheless, 1.5 million Hutu refugees fled, induced by extremist propaganda and forced to remain in exile. By 1996 the issue was largely resolved when the refugee warriors intractably opposed to an integrated idea of the state were militarily defeated in Zaire and separated from the bulk of the remaining refugees, who returned to Rwanda. Bosnia-Herzegovina. Precisely the same issue was at stake in BosniaHerzegovina. However, the results were the opposite. According to the UN High Commissioner for Refugees: “The conflict in the former Yugoslavia and the difficulty in ensuring refugee return in its aftermath demonstrate that the international community has yet to come to grips with contemporary wars fuelled by territorial and ethnic rivalries. In such conflicts the expulsion of the ‘other’ ethnic group is a goal in itself and the perpetuation of the results of ‘ethnic cleansing.’”48 The results in Bosnia contrasted with those in Rwanda because the victors in the war and the implementers of the peace agreement in Bosnia were all ethnic nationalists. Yet the peace accords called for an integrated state, even if that state was constituted of very powerful parts, each dominated by a different ethnic group. To advance an integrated vision, provision was made in the Dayton Peace Accords for the repatriation of all refugees (and displaced persons) to their homes and not just their homelands.49 No repatriation took place to any significant measure.50 The parties in power on the ground in Bosnia were not signatories to the refugee convention that obliged states to protect refugees. Only one regional state was a signatory—Croatia. As Elizabeth Cousens notes in Chapter 18, full implementation of Annex 7 would amount to a flat-out reversal of the demographic course of the war. The agreement itself provided no guidance on how to manage its chief contradiction: namely, that it both stabilized the lines of confrontation between Serb and Bosniac-Croat Federation areas and within the Federation, and derived political rights from this division, at the same time as it obligated parties to respect a right of return that could amount to reversing those lines entirely. The irony occurred when the international community brokering the peace and the cessation of hostilities helped construct an agreement that prevented the repatriation of refugees to their homes from taking place. The parties to the agreement, who were given power through the agreement, had a vested interest in ensuring the nonreturn of refugees to their homes. The international community, understandably wedded to an ideology and vision of a multiethnic society, pursued the dream of minority returns in

S-10

11/24/09

288

4:21 PM

Page 288

EVALUATING IMPLEMENTATION TASKS

spite of all the evidence that the international community was not willing to do what was necessary to make minority returns possible. 51 Those who wanted the Implementation Force (IFOR) to enforce the rights of returnees in minority areas of Bosnia failed to recognize the military unwillingness to do so, the political unwillingness to do so, and the unprecedented nature of holding international peacekeepers responsible for providing security to repatriated refugees.52 IFOR was successful in suppressing any resumption of violent conflict except when that violent conflict was a response to efforts to return refugees to minority areas. Even the moderates, such as President Biljana Plavsic in Republika Srpska, opposed minority returns. Because of the huge amount of resources invested in a fruitless (and expensive) pursuit of fostering returns to minority areas, most refugees (and internally displaced) were left in limbo. Lebanon and Sri Lanka. The same effects can be followed by state government policies in pro-people sovereignty wars. In Lebanon, the villages that were first chosen for resettlement by the government in Beirut were those that were uniconfessional, based on the belief that these would be the easiest to resettle, while mixed areas would by dealt with later.53 However, Lebanon produced the same result as in Bosnia-Herzegovina, but on a municipal rather than a provincial level. Lebanon became a series of ethnically homogeneous enclaves within a single state. The reasons are clear if we simplify the depiction of the war. The Lebanese civil war was one fought between ethnic/religious groups for control of the state. The Muslims won, the Christians lost. The demographic balance in Lebanon shifted irreversibly as many Christians immigrated. Those who repatriated returned to religious enclaves. In Sri Lanka, the story is somewhat different again, but it remains a war of control between ethnic groups. The government insists on maintaining, as an inclusive part of the state, the territory in the north where the Tamils predominate. The separatist movement, led by the Tamil Tigers in the 1990s, consists of ideological extremists. Their battle for an exclusivist Tamil territory and state in the north of Sri Lanka is fought even more viciously than the war waged by the Sri Lankan state. The result is a war to maintain territorial inclusion versus very militant secessionists. Wars over Governments and Refugee Repatriation In other kinds of civil wars, mostly fought over who is to govern the state, the issue of repatriation does not go to the heart of the makeup of the body politic. In some cases, though, it may go to the heart of the issue of electing the government of that body politic. Repatriation is often closely linked to the holding of elections, as in Cambodia and Guatemala. However, in

S-10

11/24/09

4:21 PM

Page 289

REFUGEE REPATRIATION

289

Liberia only 75,000 refugees returned home to participate in the 1997 elections. One-fifth of the population, 550,000, were left disenfranchised. Repatriation was a consideration in the elections in Mozambique; Protocol 3 of the Mozambique General Peace Agreement of October 4, 1992, between the government and the Mozambique National Resistance (Renamo) provided for the return of refugees and displaced persons. In addition to the linkage with elections, there is a connection of the timing of return in relation to elections. In Mozambique, and in contrast to Cambodia, repatriation was not required before the elections could be held.54 In Nicaragua, although the Sandinista government encouraged repatriation during the war, significant repatriation did not take place until after the national elections of 1990. The war over Western Sahara may have appeared to be a war of inclusion or exclusion, with Morocco on the side of the inclusionists and the rebels characterized as exclusionists. After all, the central issue in the peace negotiations was over inclusion: Who was to be defined as members of the body politic? But the war only appeared to be about inclusion and exclusion. The war over Western Sahara was primarily one between the indigenous population and those who were proxies of a regional hegemon, many with dubious claims to Western Saharan residence. The issue of the constituent membership of the body politic—who had a right to membership in Western Sahara—became an important aspect of implementing the peace agreement because that agreement made provision for a referendum. Only the residents of the region were eligible to vote to determine whether the people of the region wanted self-rule or wanted to be an integral part of an adjoining state, Morocco in this case. The implementation of the peace agreement has been very protracted because of the questionable status of many of the names that Morocco put forth as constituent members of the Western Saharan region. The fights over status and membership only arose because the Moroccan government seized control of the Western Saharan territory. Negative and Positive Peace and Relationship to Implementation In reading the case studies of the implementation of peace agreements, it becomes readily evident that most commentators are concerned with negative peace, that is, the cessation of violent organized hostilities between or among the parties involved in the conflict. Indeed, this project defined successful implementation to consist of ending the violence in a manner that international actors judge will be self-enforcing. Although the project is interested in the relationship between short-term success in implementation and long-term peacebuilding, the thrust of the analysis is not concerned with positive peace, the creation of conditions that turn the hostility

S-10

11/24/09

290

4:21 PM

Page 290

EVALUATING IMPLEMENTATION TASKS

between and among the conflicting parties into a cooperative working arrangement. Which end-goal one chooses to pursue is obviously related to judgments about success and appropriate ways to address refugee repatriation. In some cases, refugee repatriation may have in fact jeopardized the possible “success” of a peace agreement, where success is simply defined as the absence of violence. In Bosnia, where refugees attempted to return to their homes, and not just parts of the homeland, and where the ethnic cleansers were de facto victorious, the return of the refugees to their homes was a stimulant to the resumption of violence. This is quite aside from the issue of the justice of that return and the injustice perpetuated by the nonreturn. Intimidation by the majority against minority returnees may be ethically unacceptable, but with the exception of Kosovo, the international community showed no inclination to stop the ethnic cleansing with the force necessary to accomplish the task. In Kosovo, when they did stop the ethnic cleansing, the effect was to ensure the repatriation of the expelled refugees, but only at the cost of the expulsion of the other ethnic groups. In other words, in many civil wars there may be an inverse correlation between refugee repatriation and the successful implementation of a peace agreement defined as the cessation of violence. Put another way, in some cases, the terms of the agreement may be implemented but there is more violent conflict. On the other hand, the terms may be effectively ignored, and the result is short-term peace and longterm injustice. The relationship between exclusionary wars and internationally preferred solutions should be seen in its historical context. In the aftermath of what was then called the Great War (World War I), the official solution to a refugee crisis concentrated on internationally endorsed population exchanges. After World War II, following the establishment of the UNHCR and the signing of the Refugee Convention, the solution to refugee issues legally shifted to one based on refugee-status determination for persecuted individuals and humanitarian action focused on resettling refugees often rooted in Cold War politics. The 1990s gave birth to a third predominant solution—unsanctioned population exchanges and temporary protection for masses of refugees— while the rhetoric endorsed multiethnic societies. As a consequence, many refugees were left in limbo. The provisions in the peace agreements did not recognize this reality. They focused on one solution (putting enormous resources behind minority returns in Bosnia) while not facilitating the real one available—relocation—a conclusion apparent in the studies of the efforts to return refugees and displaced peoples to minority areas governed by those who led the ethnic cleansing in the aftermath of the exclusionary wars won by the ethnic cleansers.

S-10

11/24/09

4:21 PM

Page 291

REFUGEE REPATRIATION

291

Other Findings Beyond the importance of the type of war that is fought and the type of peace that is sought, the case evidence suggests several important findings regarding the relationship between refugees and peace implementation. These include the consequences of peace agreements, the regional context surrounding a war, and the roles of the media and other international actors. Refugees as Products of Peace Agreements As well as refugees being produced by civil wars fought over the definition of the sovereign people, they are also both the intentional and the unintentional products of the end of violent conflict in such civil wars. Even in anti-people sovereign wars, when the victors are those with an integrationist rather than ethnic definition of the sovereign people, refugees are produced. When violence ends with victory by one side—either one ethnic group or the other, or even those who hold a nonethnic definition of the body politic—many supporters of the defeated side flee. They may be Serbs and Roma from Kosovo or Hutu from Rwanda or Christians from Lebanon. This phenomenon occurs even when violent conflict ceases with the signing of a peace agreement rather than being based on victory of one side. In Bosnia-Herzegovina, tens of thousands of Serbs fled or were induced to flee Sarajevo after the Dayton Accords were signed. Thus, in these types of wars refugees may be the product of the peace agreement as well as its beneficiaries (different groups in each case, of course). Regional Context Instability in one state reverberates on neighboring states. In the extreme case, the Congo, six wars were being fought at the same time, each in relation to a different neighbor or set of adjacent states. Refugees have a dramatic impact on the stability and economic health of a region. Thus, even when refugees are not an objective of the conflict itself, they may engender conflict in neighboring states or use their camps in neighboring states as bases from which to launch attacks on their former homeland. Thai stability, domestic politics, and security were crucial motives in Thai support for Cambodia’s peace agreement and the repatriation of the refugees.55 The risk of refugee camps being used by refugee warriors is determined, in part, by the location of the refugee camps. That location is often predetermined by environmental, social, political, logistical, and financial constraints rather than by the security of the region and of the refugees themselves. Further, when the rule of law is weak in the neighboring state where the refugees have obtained temporary asylum, and the refugees are

S-10

11/24/09

292

4:21 PM

Page 292

EVALUATING IMPLEMENTATION TASKS

prevented from repatriation, the propensity to metamorphose into refugee warriors became much greater. Even if refugees do not become refugee warriors, but certainly if they do, there will be a strong likelihood that many refugees will become involved in illegal entry, drug trafficking, crime, prostitution, and the arms trade. Any involvement of this type will add to regional destabilizing propensities. A solution to the refugee problem is especially crucial before refugees are associated (even marginally) with criminal activities. When there is a determined interest of one party to prevent repatriation—because the conflict is over the definition of the body politic or possibly even where the refugees will likely support one side in an electoral battle provided for in the agreement—it is important to make provision for a permanent resolution to the refugee problem for the security of the region. However, in exclusionary civil wars over the definition of the sovereign body politic in which ethnic cleansing is a central element, and where processes of ethnic homogenization in specific territories have been a result of the civil war, refugee repatriation to homes is unlikely to be one of the permanent solutions. Quite aside from the humanitarian concern for the refugees, long-term regional stability will require another solution to the problem of refugees than repatriation to their homes. Beyond the immediate spatial context, there are larger regional factors at work that make the resolution of the refugee issue necessary when drawing up and implementing a peace agreement. There are regional economic factors. But necessity and easy feasibility are not necessarily in sync. If refugees are not repatriated, if ethnic separation becomes the de facto solution when the ethnic cleansers are the de facto winners, then the regional distrust between and among the ethnic enclaves will engender heightened distrust and hamper, if not sabotage, efforts to get regional trade moving again. Since regional interstate trade usually constitutes the major part of economic exchange between and among states, the failure to repatriate refugees may deform local trading and seriously hamper the prospects of economic recovery from the conflict. The International Role in Refugee Repatriation In forging a peace agreement, repatriation may be more of an issue for regional interveners than for the combatants. Neighboring states may push for a faster peace agreement without key details of the peace being worked out because these states are anxious to see the refugees off their soil and back across the border. In the war in Sierra Leone, the 1996 Abidjan Peace Agreement made no provision for repatriation. In contrast, when interveners played a much more significant role in the peace negotiations, the 1997 Conakry Agreement included such a provision. The Economic Community

S-10

11/24/09

4:21 PM

Page 293

REFUGEE REPATRIATION

293

of West African States (ECOWAS), with power to enforce the agreement, was intent on covering all the bases rather than simply avoiding explosive issues. In Central America, the Conferencia Internacional sobre Refugiados Centroamericanos (CIREFCA) was committed to repatriation throughout the region. It was also an effort to strengthen the presence of the UN as well as the working relationship between the UN and NGOs. Globalization of the market system has also been a force for putting local conflicts within a larger context, as indicated in some of the provisions of recent peace agreements. Major Western powers have both a low tolerance for receiving masses of refugee flows at the same time as they practice a low-risk strategy in dealing with the civil wars that produce such flows.56 Therefore, they seek strategies that try to limit the irritability produced by the threat of a large influx of refugees. At the same time, major powers adopt strategies that minimize the risk to their own military, since they are unwilling to risk the lives of their own soldiers even to save large numbers of lives at risk in intrastate wars. There is another aspect of the globalization issue that receives little attention. Refugees resettled in the diaspora, particularly in developed countries, serve to exacerbate a conflict by funding different parties to the conflict, and act as propaganda agents, while they may also assist in reintegrating refugees through the transfer of remittances to relatives. This funding may be the greatest source of foreign earnings for a state recovering from civil war. The Media The legal context may stress the rights that refugees possess, but the ability to exercise those rights may depend more on perception than on any abstract rights. Media provide the vehicle for transmitting most of those perceptions. In Kosovo, if Slobodan Milosevic had not greatly accelerated the rate of ethnic cleansing already under way before NATO started its bombing campaign, there would have been a good chance that a large and effective protest movement against the bombing would have developed. The pictures of wave after wave of fleeing and distraught masses crossing the borders, destabilizing the local states, and carrying with them tales of atrocities and horrors, provided the passion and support for the bombing campaign against a sovereign state and the infrastructure that made it operate. In the Rwanda genocide of the Tutsi, it was not the genocide that instigated an international response, but the plight of the Hutu refugees at Goma. Refugees provide a magnet for the media. The plight of the refugees has a powerful effect on whether Western powers become involved. The media coverage also affects how they become involved and the reaction to that involvement by their publics.

S-10

11/24/09

4:21 PM

294

Page 294

EVALUATING IMPLEMENTATION TASKS

Conclusion There has been a failure to critically examine widely held assumptions that refugee repatriation is essential to achieving and maintaining peace. This is certainly the case when refugees are not repatriated, for there is a real danger of creating refugee warriors. Further, if the refugees are repatriated but not reintegrated, they may become refugee warriors anyway. There are a number of reasons why this may be the case. Sometimes when the refugees are repatriated but the reintegration of those refugees is not successful, the result is the creation of a breeding ground for future violence and even recruits for a new violent uprising. But there are many other cases where it appears that successful refugee repatriation is not essential to the successful implementation of a peace agreement. On the other hand, the implicit and unexpressed contrary position that repatriation is a marginal issue in the implementation of most peace agreements has also not been proved. The conclusion I draw is more surprising than either of the above positions. Successful refugee repatriation, in many cases, may have been neither central to the success of a peace agreement nor marginal to that success. Contrary to both positions, in many cases the nonrepatriation of refugees may have been central to the success of a peace agreement, such as in El Salvador. It facilitated the successful repatriation and reintegration of the small proportion of refugees who did return. The refugees who remained in the diaspora provided a source of remittances to help rebuild the economy. Could it be the case that the failure of repatriation was an essential part in securing the cessation of violence? Indeed in wars that are fought over people, and where the desired peace is defined solely as the absence of violence, an insistence on repatriation can produce violence and destabilize an agreement. When ethnic cleansers are de facto victorious, but as part of a peace agreement find that the displaced attempt to return home, violence will likely resume. There are two ways of managing such potential violence, and their desirability must be based on an objective assessment of what international actors are realistically prepared to do in the situation. On the one hand, a negative peace that implicitly acknowledges the fact of ethnic cleansing offends our sense of justice. On the other hand, to change facts on the ground, international actors must be willing to use coercion, protect those who seek to return home, and displace and relocate those who benefited from the ethnic cleansing. If one knows a priori that international actors lack the incentive and capacity to reverse demographic facts on the ground, then the secondbest alternative is to seek alternatives to repatriation to homes. If the right of return is insisted upon, at least in the terms of the agreement, but there is no intention of enforcing that right, then there is no real right of return. There are merely rhetorical and hypocritical gestures that make a mockery of the whole principle of rights. More significantly, insist-

S-10

11/24/09

4:21 PM

Page 295

REFUGEE REPATRIATION

295

ing on certain rights in certain situations, as in the case of the Palestinian refugees or the refugees from the wars following the breakup of the former Yugoslavia, means the refugees are left in limbo. They are neither resettled nor repatriated to their homes. The result is the worst of the alternatives available. This is not just the case in exclusionary wars. In the peace agreements to end both the Cambodian and Lebanese civil wars, the right of return was included. But in each case, it effectively meant the opposite. The refugees did not have the right to return to places of their choice. In Cambodia, they were located where it suited the political parties, for very different reasons. In Lebanon, the refugees were relocated to areas where they would be homogeneous with the dominant ethnic/religious group. One begins to suspect that when the “right of return” is articulated in an agreement, it is a cover for taking away rights to freely locate anywhere in the country. The phrase “voluntary repatriation” also seems to have developed into a rhetorical device to disguise many situations where refugee return is induced or even coerced. In other situations, as in the case of the Rwandan refugees in Zaire, it is used to impede a return that might have otherwise taken place. And when returns are truly voluntary, they often occur outside the terms of the agreement and often before the conflict has really ended. Other repatriation issues emerged that have been seldom discussed. A great deal of repatriation is clearly political, aimed at enhancing the support for one side or another in the electoral battles that are scheduled to ensue following the cessation of armed battles. In one sense, such a transformation from violent to peaceful ways of settling conflicts is certainly to be welcomed. On the other hand, it clearly affects the pace of repatriation, often to the detriment of the refugees. If the goal is to get the refugees “home” so they can vote, as in Cambodia, the pace may be too fast. If the goal is to delay the vote, the repatriation may be held up for years, as in Western Sahara. In either case, the process of repatriation becomes very politicized and, as in Cambodia, undermines the reintegration process more than usual. The most important observation, however, is that it is crucial to link repatriation to the type of war fought and who the real winners are. If the ethnic cleansers remain in de facto power after a war of exclusion, attempting to move refugees back to their homes in minority areas is a chimera driven by supporters of a multiethnic philosophy who, understandably, do not want to give the ethnic cleansers a postgenocidal victory. With the exception of the war in Kosovo intended to finally diminish the military capacity of Milosevic, those intent on putting Humpty-Dumpty back together again have not succeeded in getting the developed countries to put up the military forces and use them to accomplish such a goal. Further, even in Kosovo, where supporters of multiethnicity ensure that the Muslims from Kosovo return, the cost is the ethnic cleansing of Serbs and

S-10

11/24/09

4:21 PM

296

Page 296

EVALUATING IMPLEMENTATION TASKS

Roma from Kosovo. In either case, the idea of a multiethnic polity has been destroyed and refugees who cannot be repatriated result. There are other factors that have an effect on the support, push for, and pace of refugee repatriation. The most important of these are the policies and actions of neighboring states hosting the refugees. In some cases, for example in Sierra Leone, the very issue of repatriation may be inserted into the peace agreement because of the interests and pressures of the host state or states. In addition, the role of the international media in covering such events and the development of the globalized economy have an effect on whether and how repatriation is linked to a peace agreement and its implementation. The most surprising lesson of the study, however, is that repatriation may sometimes operate to the detriment of positive peace. The absence of repatriation of the majority of the refugees who become resettled in countries of asylum may enhance the prospects for peace in both the short and the long term. The prospects for successful reintegration of the refugees who do repatriate may also be enhanced, given the reduction in competition for scarce land and the increased flow of remittances to assist in the rebuilding effort. Thus, evidence and arguments challenge the presumption of humanitarian scholars that successful repatriation is essential to the successful implementation of a peace agreement. At the same time, they also challenge the possibly implicit assumption of most scholars of security studies that the issue of repatriation is a marginal factor in the successful implementation of a peace agreement. Instead, repatriation may be a complicating factor, with many different dimensions, in the successful implementation of a peace agreement.

Notes 1. “Refugees,” as used here, are defined as persons who cross the border of the country in which they normally reside either because of persecution (as defined in the refugee convention) or, more broadly, to escape the perceived threat of violence from a conflict under way. Repatriation of refugees entails the return from a host country, either one neighboring the state from which refugees fled or a more remote country in which the refugees received some form of temporary or permanent asylum. Though refugees may return to their homes, repatriation does not necessarily entail a return home, but only a return to one’s homeland. 2. UNHCR, Repatriation, Reintegration, and Rehabilitation (Geneva: UNHCR, 1989); Astri Suhrke and Aristide R. Zolberg, “Beyond the Refugee Crisis: Disengagement and Durable Solutions for the Developing World,” Migration 5, no. 89 (1989): 69–119; Aristide R. Zolberg, Astri Suhrke, and Sergio Aguayo, Escape from Violence: Conflict and the Refugee Crisis in the Developing World (New York: Oxford University Press, 1989); UNHCR and UNDP, Progress Report on the Implementation of the Joint Plan of Action for Refugees, Returnees, and Displaced

S-10

11/24/09

4:21 PM

Page 297

REFUGEE REPATRIATION

297

Persons in Central America (San José, Costa Rica: UNHCR, 1990); and UNHCR, International Conference on Central American Refugees: Repatriation, Reintegration, and Rehabilitation (Guatemala, UNHCR, 1992). 3. Elizabeth Ferris, “Central American Refugees: Hopes for Reconciliation, Peace, and Repatriation,” Refugees 113 (May 1991): 6. 4. J. F. Crisp, “The Challenge of Return and Reintegration in the Great Lakes Region,” paper presented at the Regional Meeting on Refugee Issues in the Great Lakes Region, Cambridge, Mass., 1999, p. 3. 5. Ken Watkin, “Building a Legal Framework for the Repatriation of Refugees: ‘Where the Rubber Meets the Road,’” in Ann L. Griffiths, ed., Building Peace and Democracy in Post-Conflict Societies (Halifax: Centre for Foreign Policy Studies, Dalhousie University, 1998), p. 120 (emphasis added), citing UNHCR, “Rebuilding a War-Torn Society: A Review of the UNHCR Reintegration Programme for Mozambican Returnees,” Refugee Survey Quarterly 16, no. 2 (1997): 47–50. 6. Finn Stepputat, “Repatriation and Everyday Forms of State Formation in Guatemala,” in Richard Black and Khalid Koser, eds., The End of the Refugee Cycle? Refugee Repatriation and Reconstruction (New York: Berghahn Books, 1999), pp. 223–224, argues that the Guatemalan refugee repatriation “has been demonstrated to have contributed to the social construction of peace and the development of a ‘common discursive framework’ which includes the extension of a network of modern rationalities through which techniques of government work.” He furthermore argues that repatriation is crucial to the reconstruction of civil society as well. On the other hand, Gonzalo De Villa and W. George Lovell, “Land and Peace: Two Points of View,” in Liisa L. North and Alan B. Simmons, eds., Journeys of Fear: Refugee Return and National Transformation in Guatemala (Montreal and Kingston: McGill-Queen’s University Press, 1999), p. 54, argue that that long-term permanent peace is jeopardized by the failure to provide for extensive land reform in the peace agreement. “The fundamental issue of unequal ownership of land, however, can be resolved only if it is actually addressed. If it is not, then the peace that has supposedly been signed into being in Guatemala may prove neither firm nor lasting.” It is not the refugee repatriation, but the conditions for repatriation, that ensure long-term peace, according to Lovell. This is an even harder position than Hard II depicted above. 7. Zolberg, Suhrke, and Aguayo, Escape from Violence; Edward KhidduMakubuya, “Voluntary Repatriation by Force: The Case of Rwandan Refugees in Uganda,” in Howard Adelman and John Sorenson, eds., African Refugees: Development Aid and Repatriation (Boulder, Colo.: Westview Press, 1994), pp. 143–158; UNHCR, The State of the World’s Refugees: In Search of Solutions (Oxford: Oxford University Press, 1995); Mark Gibney, Clair Apodaca, and James McCann, “Refugee Flows: The Internally Displaced and Political Violence (1980–1993): An Exploratory Analysis,” in Alex P. Schmid, ed., Whither Refugee? The Refugee Crisis: Problems and Solutions (Leiden: PIOOM, 1996), pp. 45–61; and Howard Adelman, “Why Refugee Warriors Are Threats,” Journal of Conflict Studies 18, no. 1 (spring 1998): 49–69. 8. Nabil Beyhun, “Population Displacement in the Metropolitan District of Beirut: Are the Displacements Over?” in Seteney Shami, ed., Population Displacement and Resettlement: Development and Conflict in the Middle East (Staten Island, N.Y.: Center for Migration Studies, 1994), p. 16. 9. Sarah Kenyon Lischer, “Militarized Refugee Populations: Humanitarian Challenges in the Former Yugoslavia,” 1999, http://web.mit.edu/cic/www/migration/milit.htm.

S-10

11/24/09

298

4:21 PM

Page 298

EVALUATING IMPLEMENTATION TASKS

10. Marjoleine Zieck, UNHCR and Voluntary Repatriation of Refugees: A Legal Analysis (The Hague: Martinus Nijhoff, 1997), p. 362; and UNHCR, “Report on Mozambique, South Africa, and Malawi,” February 4–12, 1992, n. 139. 11. With respect to Guatemala, “Refugees gradually began to return as the peace process evolved. At first the numbers were small, since those returning were still largely viewed as enemies of the army. Following the signature of accords between the Permanent Commissions of Guatemalan Refugees in Mexico [known by the acronym CCPP] and the government of Guatemala in October 1992, the numbers returning increased, yet the return remained troubled by continuing human rights abuses and scarce evidence of deep transformations that would assure political freedoms, personal security, economic well-being, and respect for indigenous cultures.” Liisa L. North and Alan B. Simmons, “Fear and Hope: Return and Transformation in Historical Perspective,” in North and Simmons, Journeys of Fear, p. 4. 12. In Guatemala, “The slow pace of return was due in large part to concerns about the security of the refugees. . . . The problems faced by refugees wishing to return to their former positions and lands within cooperative communities applied both to those returning to Petén and those returning to Ixcán. In both areas the returnees were often met with open hostility and distrust on the part of local military authorities and other residents.” Brian Egan, Somos de la Tierra: Land and the Guatemala Refugee Return (Toronto: Centre for Research on Latin America and the Caribbean, May 1996), p. 1. 13. Liisa L. North and Alan B. Simmons, “Concluding Reflections: Refugee Return, National Transformation, and Neo-Liberal Restructuring,” in North and Simmons, Journeys of Fear, p. 288: “There are many good reasons for believing that the economic problems of Guatemala and other Central American countries cannot be solved without first, or at least simultaneously, addressing problems of equity, social development, and democratization in order to build the peace required for sustainable economic growth.” 14. Khalid Koser, “The End of the Refugee Cycle?” in Richard Black and Khalid Koser, eds., The End of the Refugee Cycle? Refugee Repatriation and Reconstruction (New York: Berghahn Books, 1999), p. 3. 15. C. Berthiaume and F. Mundo, “Going Home,” Refugees 88 (1992): 6–10. 16. Adekeye Adebajo, “Western Sahara,” draft prepared for this project, p. 15. 17. Joao Honwana, “Mozambique,” draft prepared for this project. 18. Rosemary Preston, “Researching Repatriation and Reconstruction: Who Is Researching What and Why?” in Black and Koser, End of the Refugee Cycle? p. 19. 19. The issue of numbers is complex, political, and fraught with difficulty. In refugee-like situations, the numbers flowing in and out are in constant flux and monitoring is often daunting and sometimes impossible. The numbers change in relationship to the degree of violence and the perceived sense of security. There is also the problem of inaccuracy in the sources for the figures. In the case of the Palestinian refugees, there were over 900,000 registered Palestinian refugees originally, and 2.8 million currently, although scholarly figures establish the original range between 660,000 and 720,000. Howard Adelman, “Palestinian Refugees,” in L. Y. Luciuk and M. S. Kenzer, eds., Under Threat: Essays on the Refugee Experience (Lanham, Md.: Rowan and Littlefield, 1998), pp. 119–142. In the case of the Rwandese refugees in Zaire in 1996, there were supposed to be roughly 1.2 million according to the UNHCR. An estimated 650,000 to 750,000 returned. What happened to the other over 400,000 refugees? Were the refugee warriors included in the counts? Were the original figures overestimated either way? According to Human Rights Watch–Zaire 1997, there were evidently a number slaughtered, and

S-10

11/24/09

4:21 PM

Page 299

REFUGEE REPATRIATION

299

even the most pessimistic view suggests a figure as high as 100,000. What happened to the other 300,000 “phantom refugees”? 20. The Bosnian Institute of Public Health estimated that of the 2,214,000 people in the territory controlled by the Bosnian Federation, 1,025,000 were displaced. Mustafa Imamovic suggested an even higher figure of 60 percent; 1,245,000 were abroad. See Steven L. Burg and Paul S. Shoup, The War in BosniaHerzegovina: Ethnic Conflict and International Intervention (Armonk, N.Y.: M. E. Sharpe, 1999), pp. 171–172, nn. 179–180. 21. Wim van Damme, “How Liberean and Sierra Leonean Refugees Settled in the Forest Region of Guinea,” Journal of Refugee Studies 9, no. 3 (September 1999): 252–267. 22. In this case, the UNHCR was faced with what may have appeared to the organization as the lesser of two evils—cooperating in the forced repatriation of a refugee population lest the refugees be bereft of any humanitarian assistance or legal protection whatsoever. 23. The UN’s Integrated Regional Information Network (IRIN) information unit in the Department of Humanitarian Affairs (DHA) reported that as of July 27, 1999, there were over 1.7 million displaced in Angola, of whom 930,000 were newly displaced as a result of the renewed violent conflict. Of those displaced, 4,000 were in Namibia, mainly in Osire. 24. During the previous year, 2,000 refugees had fled the area to Botswana. The UNHCR had just assisted the return of 1,300 refugees when the violence flared up once again. 25. During this same short time frame, Sadako Ogata, the UN High Commissioner for Refugees, claimed that “the most pressing refugee problem for the Great Lakes region was the need to tackle the problem of the Rwandese who had not yet returned after fleeing their country in the aftermath of genocide (1994).” Ogata went on to say that the UNHCR would resume support for the repatriation of Rwandese still in eastern DRC. Rwandans in the Republic of Congo would be offered either repatriation or the opportunity to settle in the north of their host country. See UN, DHA IRIN information unit, July 30, 1999. 26. For discussions on the right of return, see Howard Adelman, “Refugees, the Right of Return, and the Peace Process,” in Bashir Al Khadra, ed., Economics of Peace in the Middle East (Irbid: Yarmouk University, 1996); Howard Adelman, “The Right of Repatriation—Canadian Refugee Policy: The Case of Rwanda,” International Migration Review 30, special issue (spring 1996): 289–309; and Howard Adelman, “Refugees: The Right of Return,” in Judith Baker, ed., Group Rights (Toronto: University of Toronto Press, 1994), pp. 164–185. 27. Voluntary repatriation is the most discussed theme in the refugee repatriation literature. The analysis by André Fabienne, “Le HCR et le repatriement volontaire” (Ph.D. thesis, Augborg University, 1988), provides the best dissection that I have studied of the premises of voluntary repatriation. 28. Tom Kuhlman, “Organized Versus Spontaneous Settlement of Refugees in Africa,” in Adelman and Sorenson, African Refugees, pp. 117–142; B. N. Stein, F. C. Cuny, and R. Reed (eds.), Refugee Repatriation During Conflict: A New Conventional Wisdom (Dallas, Tex.: Center for the Study of Societies in Crisis, 1995); and T. Skari, “Spontaneous Return to Ethiopia,” Refugees 19 (1985): 9–12. 29. Frederick Cuny and B. N. Stein have written the most on this issue. See Frederick Cuny, Barry Stein, and Pat Reed, Repatriation During Conflict in Africa and Asia (Dallas, Tex.: Center for the Study of Societies in Crisis, 1992); B. N. Stein and F. C. Cuny, “Repatriation Under Conflict,” in V. Hamilton, ed., 1991 World Refugee Survey (Washington, D.C.: U.S. Committee for Refugees, 1991), pp.

S-10

11/24/09

300

4:21 PM

Page 300

EVALUATING IMPLEMENTATION TASKS

15–21; B. N. Stein and F. C. Cuny, NGOs and Repatriation During Conflict (Dallas, Tex.: Center for the Study of Societies in Crisis, 1992); and Stein, Cuny, and Reed, Refugee Repatriation During Conflict. There is also a very good discussion in Ove Narvesen and Oona Solberg, Repatriation During Conflict (Oslo: Norwegian Refugee Council, 1992). 30. Taking Mozambique as an example again, although 20 percent of the refugees repatriated three years before the peace agreement was signed, only 30 percent had returned a full year after the agreement was signed. Repatriation was not completed until 1994. See V. Bonga and K. B. Wilson, “Repatriation of Mozambicans to Malawi: The Current Situation,” report for the Refugee Studies Programme, 1993. 31. Judith Harik, “The Effects of Protracted Social Conflict on the Resettlement of the Displaced in Postwar Lebanon,” in Paul Salem, ed., Conflict Resolution in the Arab World: Selected Essays (Beirut: American University of Beirut, 1997), pp. 384–410. 32. Hiram Ruiz, When Refugees Won’t Go Home: The Dilemma of Chadians in Sudan (Washington, D.C.: U.S. Committee for Refugees, 1987). 33. B. Bernander, J. Chany, M. Eastmond, C. Lindahl, and J. Ojendal, “Facing a Complex Emergency: An Evaluation of Swedish Support to Emergency Aid in Cambodia,” Evaluation Report no. 4 (Stockholm: Swedish International Development Agency, 1992); and S. Chantavanich and M. Benyasut, “Thailand and Indo-Chinese Refugees’ Problems,” Journal of Asian Review 10, no. 2 (1989): 98–112. 34. F. C. Cuny and C. R. Cuny, “The Return of Tamil Refugees to Sri Lanka, 1983–1989,” in Stein, Cuny, and Reed, Refugee Repatriation During Conflict, pp. 23–91, esp. pp. 60, 85; Court Robinson, Something Like Home Again: The Repatriation of Cambodian Refugees (Washington, D.C.: U.S. Committee for Refugees, 1994); and Zieck, UNHCR and Voluntary Repatriation of Refugees, p. 10, fn. 29. 35. Anthony H. Richmond, Global Apartheid: Refugees, Racism, and the New World Order (Toronto: Oxford University Press, 1994), p. 215. 36. S. Irfani and F. Rahman, Afghanistan: Looking to the Future (Islamabad: Institute of Strategic Studies, 1991). 37. M. Knowles, Afghanistan: Trends and Prospects for Refugee Repatriation (Washington, D.C.: Refugee Policy Group, 1992). 38. R. English, Interim Report on Repatriation Planning for Afghan Refugees (Geneva: UNHCR, Operational Unit for Repatriation to Afghanistan, 1989); and UNOCA, “Operation Salam: Third Consolidated Report,” UN Coordinator for Humanitarian and Economic Assistance Programs Relating to Afghanistan (Geneva: UNOCA, 1990). 39. Hiram Ruiz, Left Out in the Cold: The Perilous Homecoming of Afghan Refugees (Washington, D.C.: U.S. Committee for Refugees, 1992). 40. Joshua O. Akol, “Southern Sudanese Refugees: Their Repatriation and Resettlement After the Addis Ababa Agreement,” in John Rogge, ed., Refugees: A Third World Dilemma (Totawa, N.J.: Rowman and Littlefield, 1987), pp. 143–157; and Roger Winter, “Ending Exile: Promoting Successful Reintegration of African Refugees and Displaced People,” in Adelman and Sorenson, African Refugees, p. 168. 41. Narvesen and Solberg, Repatriation During Conflict; and Stein, Cuny, and Reed, Refugee Repatriation During Conflict. 42. Koser, “End of the Refugee Cycle?” p. 5. 43. K. B. Wilson and J. Nunes, “Repatriation to Mozambique: Refugee

S-10

11/24/09

4:21 PM

Page 301

REFUGEE REPATRIATION

301

Initiative and Agency Planning in Milange District, 1988–1991,” in Tim Allen and Hubert Morsink, eds., When Refugees Go Home: African Experiences (Geneva: UNRISD, 1994), p. 167; Khalid Koser, “Information and Repatriation: The Case of Mozambican Refugees in Malawi,” Journal of Refugee Studies 10, no. 1 (1997): 1–18; and Chris Dolan, “Repatriation from South Africa to Mozambique— Undermining Durable Solutions?” in Black and Koser, End of the Refugee Cycle? pp. 85–108. 44. Dolan, “Repatriation from South Africa to Mozambique,” in Black and Koser, End of the Refugee Cycle? p. 95. 45. K. Dixon-Fyle, “Prevention as the Best Solution,” Refugees 96 (1994): 22–25. 46. B. Edwards and G. Tovar Siebentritt, Places of Origin: The Repopulation of El Salvador (Boulder, Colo.: Lynne Rienner, 1991), p. 97; Zieck, UNHCR and Voluntary Repatriation of Refugees, p. 121; and Fagin P. Weiss and J. T. Eldridge, “Salvadoran Repatriation from Honduras: El Salvador Mobilized and Poor,” in Mary Ann Larkin, Fred Cuny, and Barry Stein (eds.), Repatriation Under Conflict in Central America (Washington, D.C.: Center for Immigration Policy and Refugee Assistance, Georgetown University, 1991), p. 177. 47. Though the causes for the initiation of the October 1, 1990, Rwanda war were more complex, Deborah Kobak, “Rwanda: Never Again?” in Roderick von Lipsey, ed., Breaking the Cycle (New York: St. Martin’s Press, 1997), p. 156, stressed the importance of refugee repatriation: “The Habyarimana regime’s reluctance to permit the return of exiled Rwandan Tutsi combined with the Ugandan government’s refusal to grant the Tutsi refugees citizenship precipitated the 1990 RPF military invasion of northern Rwanda.” 48. Statement of Sadako Ogata, General Meeting, New York, December 1, 1998, “On the Humanitarian Front Lines: Refugees Crises and the Unwinnable Wars,” a thesis she first presented before the working group for humanitarian questions (HIWG) regarding peace for Bosnia-Herzegovina on December 17, 1997, in Geneva. 49. This is not quite true, although it is true of the primary interpretation given to the agreement by a high proportion of the international NGO community concerned with Bosnia. Annex 7 of the Dayton Accords obliges the parties to guarantee either the voluntary return of the displaced or compensation for property that cannot be restored to them. 50. This was in spite of the fact that the following steps were to be taken to ensure this repatriation and integration: refraining from harassment, intimidation, persecution, and discrimination. These provisions not only applied to state authorities; others were to be prevented from doing the same. Further, legislation was to be repealed and administrative practices stopped that discriminated against minorities either in intent or effect; hate speech was to be prevented or suppressed; those who violated rights of minorities were to be prosecuted. Instead of the authorities fostering repatriation, the reverse was true. Intimidation against return was pervasive by state and quasi-state authorities. Empty houses were burned most dramatically in Sarajevo upon the exodus of 60,000 Serbs following the signing of the Dayton Accords. Those who controlled the different parts of the so-called integrated polity undermined and prevented the realization of that goal. Thus there was virtually no repatriation to areas in which the returnees would constitute the minority. 51. In the late 1980s and early 1990s, Western countries were accused of pursuing a repatriation agenda because of domestic concerns and the growing restrictive attitude with respect to refugees arriving on their soil. See Barbara HarrellBond, “Repatriation: Under What Conditions Is It the Most Desirable Solution for

S-10

11/24/09

302

4:21 PM

Page 302

EVALUATING IMPLEMENTATION TASKS

Refugees?” African Studies Review 32, no. 1 (1989): 41–69; and Arne Piel Christensen, Repatriation: A New Approach to European Refugee Policy in the Nineties (Copenhagen: Danish Refugee Council, 1993). In the case of BosniaHerzegovina, however, the NGOs, international organizations, and Europe were pursuing a specific form of repatriation for ethical ideological reasons rather than self-interested domestic politics. But the results may have been even worse— refugees who could not be repatriated or, if repatriated, were left with little means to successfully reintegrate. 52. Watkin, “Building a Legal Framework for the Repatriation of Refugees,” in Griffiths, Building Peace and Democracy in Post-Conflict Societies, p. 125. 53. See Chapter 19 by Marie-Joëlle Zahar in this book. 54. D. Cammack, Mozambiquan Refugees: Flight and Refuge in the 1980s, mimeograph (Oxford: Refugee Studies Program, 1992), chap. 7, p. 21. 55. Chantavanich and Benyasut, “Thailand and Indo-Chinese Refugees’ Problems.” 56. Kathleen Newland and Demetrios Papademetrious, “Managing International Migration: Tracing the Emergence of a New International Regime,” UCLA Journal of International and Foreign Affairs 3, no. 3 (fall–winter 1998–1999): 644, argue that the “clear and widespread tendency to interpret obligations more narrowly than heretofore and restrictive practices both here [in the United States] and Europe are being emulated by developing countries like Thailand, Iran and Tanzania, which feel justified in their actions because they are less able than the developed nations to handle massive refugee flows.” For evidence of the shift from humanitarian acceptance to rejection, see L. Hitchox, “Repatriation: Solution or Expedient? The Vietnamese Asylum Seekers in Hong Kong,” Southeast Asian Journal of Social Science 18, no. 1 (1990): 11–131; and Leonard Davis, “Mandatory Repatriation Is Not the Answer to Hong Kong’s Problem,” Refuge: Canada’s Periodical on Refugees 9, no. 3 (1990): 10–11.

S-11

11/24/09

4:22 PM

Page 303

11 Civilian Security CHARLES T. CALL AND WILLIAM STANLEY

Analysts of civil war endings have devoted much attention to reducing the vulnerabilities of warring parties in order to enhance their security and foster peace. Less work has been done, however, to address the security of the general population. Once safe from the cross fire of warring armies, civilians often face new threats from violent criminals, former combatants, vigilantes, or members of other ethnic groups with whom they must cohabit under a peace agreement. In the aftermath of many civil wars of the 1980s and 1990s, civilians perceived greater insecurity, often as a result of increases in violent crime. In El Salvador and South Africa, for example, civilians faced greater risk of violent death or serious injury after the end of the conflict than during it.1 Even where the end of civil wars has reduced the dangers to civilians, postwar crime waves and civil disturbances have been common. Security reforms remain one of the most important mechanisms for preventing political violence and common crime in many postwar settings. As police scholar David Bayley puts it, “The police are to government as the edge is to the knife.”2 Where police are repressive or politically biased, formal democratic rules and equality before the law can mean little in practice. Where the police are dominated by one ethnic group, members of excluded groups may turn to ethnically based alternatives to provide security. And where no public security provisions are made, organized crime can assume such proportions as to produce corrupt mafia states, possibly with the complicity of international actors. Conversely, police who are effective in protecting individuals from criminal threats, who respect individuals’ rights, and who protect vulnerable groups regardless of ethnicity or political orientation, can create a positive climate even when broader political arrangements are uncertain or less than democratic.3 303

S-11

11/24/09

304

4:22 PM

Page 304

EVALUATING IMPLEMENTATION TASKS

Despite the myriad dangers for public security after conflict, relatively few civil war settlements include restructuring security forces in ways that reorient public safety toward protecting citizens rather than the regime. Of eighteen recent cases, minimal police reforms were mentioned in twelve agreements. In most of these cases, few specifics were stipulated beyond the inclusion of former enemies into police forces and brief commitments to human rights standards and professionalization. The general neglect of public security provisions in peace accords reflects the logic of peacemaking: the parties, and outside mediators, tend to focus on the postsettlement security of the warring parties, since this is what will make or break a peace process in the short run. Indeed, inattention to public security issues has seldom, if ever, caused renewed civil war. It has, however, contributed to extreme hardships and undermined longer-term prospects for both peace and democracy. Most civil war peace accords depend on liberal political institutions, including elections with at least two competing parties, constitutional systems with checks and balances, and baseline human rights. Often, vulnerable groups are asked to settle for procedural guarantees under governments likely to be dominated by their former adversaries. In such settings, biased, undemocratic, neglectful, or violent conduct by the police can undermine a fledgling democracy. Making individual-level protections effective requires that police and judiciaries protect individual citizens rather than regimes or dominant groups. Accomplishing this may require new personnel drawn from social groups with security fears; new internal regulations, procedures, and norms; new training procedures and retraining of existing personnel; and new mechanisms of internal discipline, external review, and civilian control. The ending of wars offers special opportunities to reform internal security institutions. For example, of twenty-three transitions toward democracy in Latin America, only ten experienced major police reforms.4 Of these ten democratic transitions, eight coincided with the ending of civil war or major international interventions. Of the thirteen cases of democratization without major reform, only one was a “war transition.” This pattern suggests that civil wars and foreign interventions disrupted state institutions and ideas and facilitated the implementation of new security models and doctrines. Other studies have also found police institutions to be resistant or “tenacious” in the face of major regime changes in Africa.5 Civil war settlements offer a chance to break this tenacity, by creating greater political space for both domestic and international actors to place major reforms on the agenda. If police reforms are omitted from civil war settlements, these issues must be addressed later, when it will usually be more difficult to find the support for reorienting systems of security and justice. International actors need to engage in ambitious agenda-setting during peace negotiations to help warring parties envision new ways of policing

S-11

11/24/09

4:22 PM

Page 305

CIVILIAN SECURITY

305

rooted in human rights, ethnic tolerance, and citizen service, and to help the parties incorporate such a vision into peace accords. To provide such an agenda-setting role, international mediators need to look beyond the immediate requirements to achieve peace. The challenges of postconflict democratization exceed the requirements for peace, even where peace processes may be fragile and dangerous. This agenda-setting needs to be followed by significant capacity to invest in institution-building over the long run, a capacity that the international community currently lacks. We begin this chapter by addressing the ways in which civil wars and postwar conditions contribute to public insecurity. We then examine eighteen cases of peace accords to assess whether public security reforms were included, and whether such reforms were implemented or not. Next we turn to choices facing international implementers of peace settlements. We distinguish between the requirements of transitional public security during initial implementation of peace plans, and the subsequent long-term development of new or reformed institutions. Each phase contains its own dilemmas, and there are potential contradictions between measures to improve transitional security and measures to facilitate long-term institutional development. We conclude with a few broad recommendations for international actors, as well as a cautionary note.

The Effects of War on Public Security In civil wars, police and judicial organizations often become instruments to protect the state and the regime rather than the general population. Their functions are subsumed into counterinsurgency efforts, and their activities are circumscribed or simply eliminated in areas under enemy control. During war, the militarization and arbitrariness of policing can so intimidate civilians as to suppress criminal activity. Even as civilians may fear to engage in criminal or other activities, however, their acceptance and expectation of violence as a means of resolving conflicts may increase. Moreover, military and police forces tend to engage in and cover up human rights abuses related to the war, and impunity and secrecy become the currency of everyday policing and justice. Consequently, even as civil wars can reduce crimes and insecurity not related directly to the causes behind war, they leave behind shattered social and community ties and weakened institutional and cultural constraints on crime and violent conduct. The termination of war, of course, removes many of the threats to civilians of displacement, death, and human rights violations. However, the very process of ending war creates additional factors that contribute to the potential for intensified violence and common crime. War termination involves the demobilization of combatants and the release of arms into society. Former combatants encounter few job opportunities in war-

S-11

11/24/09

306

4:22 PM

Page 306

EVALUATING IMPLEMENTATION TASKS

shattered economies, and even where work is available, many lack marketable skills other than soldiering. They therefore have strong incentives to resort to criminal activity. Many former combatants, especially those who served for many years or began fighting when they were children, have become psychologically accustomed to danger and violence. Criminality is made easier, and more dangerous, by the availability of military weaponry after wars. Disarmament is incomplete, particularly with respect to small arms, which can be readily cached, retrieved, and sold. Clandestine paramilitary and intelligence structures once used for repression can easily be converted into criminal organizations. Their networks of contacts, information, safe houses, vehicles, aircraft, weapons dealers, and financiers provide the means to kidnap for ransom, murder for hire, or traffic in stolen vehicles or contraband. Aside from overt criminality, various organized groups—ranging from ethnic groups to demobilized former combatants—may resort to violence to express grievances, to exert control over territory or resources, or to punish their opponents. Peace agreements that end civil wars can actually contribute to postwar insecurity. Peace accords often involve reductions in the capacity of states to deter or constrain criminality and unrest. Civil war settlements often dramatically reduce the numbers of security personnel, as armies—which often play de facto security functions—are demobilized or cantoned, or as existing police forces are demobilized or purged. In El Salvador, the overall forces of vigilance declined from 75,000 to around 6,000.6 In some cases, police forces were not particularly well-developed before the civil war, and have deteriorated, or been distorted, as a result of the conflict. The collapse of other elements of the justice sector, including the judiciary or the prisons, can render police activity useless, with suspects released because of judicial incapacity or corruption. Any combination of these weaknesses can create a permissive environment for criminal activity to escalate. In those cases where police forces remain intact and effective, other problems arise. Preexisting forces encounter difficulties in shifting their orientation and resources from political control to citizen protection, and from serving privileged groups to the entire population, including previously demonized groups.7 Criminal investigative units, in particular, have often been dedicated to political spying or covering up abuses by state agencies. The existing forces’ prior involvement in the armed conflict can make them a threat to former rebels, opposition political groups, or ethnic minorities. This is particularly true where police have been dominated by, and subordinated to, the military. Police are likely to violate human rights, and are also likely to conduct themselves in a partisan manner, thus undermining transitions to democratic rule. In Bosnia, for example, police forces have displaced civilians, prevented refugee returns, and engaged in abusive conduct since the Dayton Accords. Public insecurity presents a political opportunity for any group that has

S-11

11/24/09

4:22 PM

Page 307

CIVILIAN SECURITY

307

sufficient organization and weaponry to present itself as a protector of a given community. The extreme form of this is the protection racketeering of competing clans and associated “technicals” in Somalia. Other examples include the efforts of the Kosovo Liberation Army (KLA) to present itself as a domestic security force in the wake of Yugoslav Serb withdrawal from Kosovo. Sometimes these protection efforts are overtly political—with promises of protection serving to reinforce claims to political legitimacy (as in the case of the KLA and other de facto ethnic police forces in the Balkans). In other cases, it is simply an opportunity for economic entrepreneurship, as private organizations offer, for a price, the protection the state is failing to provide. Such privatization of security tends to accentuate inequities between social classes in the degree of security enjoyed.

Peace Agreements and Public Security To address these causes of insecurity requires a comprehensive approach. In the short term, states, transitional authorities, or international implementers must minimize the interim security gap and maximize the security of disarmed former combatants and their supporters, returning refugees, and other vulnerable groups. In the long run, they need to pursue economic and social development, create services for former combatants and other potentially volatile groups, and reform or build anew security institutions capable of protecting the public. Here we focus especially on reforms to organizations that de facto or de jure provide security and justice for the general population, be they formal police forces, judicial systems, paramilitary forces, party-based police groups, informal community-based elders or other bodies, or private security agencies. Most peace accords do not grant public security questions a level of attention and detail commensurate with their importance. This is especially true of plans for long-term public security reforms. Of the twelve recent cases in which police reform was included in the peace agreements (see Figure 11.1), in only eight cases was any reform actually implemented. Even those accords that do address public security offer little more than basic outlines for security provisions during the initial transition, plus some measure related to composition of police forces after the cease-fire. Several call for international monitoring of local police. Only in El Salvador and Northern Ireland were extensive details for institutional changes included in peace agreements. In Cambodia and Nicaragua, substantive police reforms were omitted from the accords, but were implemented later because governments and/or foreign donors determined that institutional changes were needed. None of these latter cases qualify as extensive reforms. The range of what “reforms” were set forth reveals the ambiguity of

S-11

11/24/09

4:22 PM

Page 308

308

EVALUATING IMPLEMENTATION TASKS

Figure 11.1 Post–Civil War Police Reforms Police Reform Implemented? Yes Yes

El Salvador Guatemala South Africa Croatia/Eastern Slavonia [Haitia] Mozambique

No

Cambodia Nicaragua

Police Reform Included in Peace Accord?

No Namibia Northern Ireland Bosnia [Kosovoa]

Angola (Lusaka) Rwanda Somalia Sierra Leone I (Abidjan)

Liberia Western Sahara Sri Lanka Sierra Leone II (Conakry)

Note: a. Though not cases of civil war settlement, Haiti and Kosovo are included because of their importance in shaping international approaches to police reform and because prior peace agreements (for Haiti the 1993 Governor’s Island Accord, and for Kosovo the 1999 Rambouillet Accord) included significant police reform provisions. Although the Haitian government and the Yugoslav government, respectively, failed to observe (or sign) these agreements, in each case the accords served as roadmaps for subsequent international implementation of police reforms.

the concept of “police reform” in postwar settings. National and international actors tend to define police reforms according to their own institutional interests and habits. Intervening militaries, for instance, are most interested in establishing some force to maintain order as quickly as possible, whatever the composition or (often) illegitimacy that such a force may enjoy. Armed opposition movements are frequently most interested in incorporating some of their members into any new governmental security forces so that their militants and supporters can enjoy some group protections following disarmament. Such approaches to police reform reflect the imperatives of reaching peace, with less concern for the longer-term foundations for democratic regime-building. We define “police reforms” to include not only the incorporation of former combatants or enemies, but also the reorientation of doctrine to emphasize citizen protection rather than protection of the state or specific political parties; an explicit commitment to internationally accepted human rights standards; and some efforts at professionalization or modernization. Police reforms generally include the adoption of mechanisms of oversight and accountability for police misconduct as well. In Figure 11.1, we categorize peace implementation cases according to whether police reforms were included in peace accords, and whether reforms have been implemented to date (the yes/no categories obviously oversimplify characteristics that are really shades of gray, and the figure is intended as a heuristic device). In many cases, agreed-upon police reforms

S-11

11/24/09

4:22 PM

Page 309

CIVILIAN SECURITY

309

were limited to brief references to incorporating former enemies into the police, to enhanced training or professionalization, and to conformity with international human rights standards. Croatia’s Erdut Peace Agreement (1995), for example, provides only that a transitional administration in Eastern Slavonia will “help establish and train temporary police forces, to build professionalism among the police and confidence among all ethnic communities.” In other cases, such as Northern Ireland’s “Good Friday” agreement (1998) and Angola’s Lusaka Protocol (1994), lengthy provisions called not only for incorporating former enemies, but also for fundamental reorientation of policing along new models emphasizing citizen service. Police Reforms Absent in Word and Deed In six cases, public security reforms were not mentioned in peace agreements. In some cases, the omission of provisions for interim and long-term security for the general population stems from the lack of serious intent on the part of civil war adversaries to reach an agreement. In cases where the two sides are far apart and can only agree upon vague or minimal commitments, police reforms often do not make the agenda. In such cases, it is also not surprising that a peace process breaks down due to the distrust that underlay the peace accord. Where commitment of the parties to a settlement was questionable from the outset, one would not expect the parties to subsequently undertake security reforms or other sorts of institutional reforms, given the breakdown of the process. Hence, no implementation would ensue. In four cases—Sri Lanka, Western Sahara, Liberia, and Sierra Leone’s Conakry Agreement—no police reforms were included in peace agreements or implemented afterward. In each case, the overall failure of the peace process accounts for the absence of any security reforms. Police Reforms Not in Accord, but Implemented In some cases where police reforms were absent from an agreement, the parties may subsequently implement police reforms. Of the eighteen cases categorized in Figure 11.1, such a situation occurred in two cases: Nicaragua and Cambodia. These peace processes included some elements of reform, such as the incorporation of some enemy forces into the Cambodian police force. However, peace agreements failed to stipulate any change of national police doctrine. Instead, for the initial years after war, police culture remained largely intact, despite the presence of international observers and the incorporation of some former enemies into the police force. Eventually, however, some governments may undertake broader efforts at police reform than originally prescribed in peace settlements. The governments of Cambodia and Nicaragua subsequently embraced efforts to

S-11

11/24/09

310

4:22 PM

Page 310

EVALUATING IMPLEMENTATION TASKS

reconfigure policing along the lines of nonpartisanship and respect for human rights. Police Reforms in Accord, but Not Implemented A well-designed, apolitical police force can represent a source of security to both sides after a civil war settlement only if there is little risk of renewed military conflict. Lightly armed police, no matter how politically neutral and professional, are of little consequence if there is a significant risk that either side will attack again with machine guns, rocket-propelled grenades, mortars, and artillery. In four cases—Rwanda, Angola, Sierra Leone (Abidjan), and Somalia—agreements contained provisions for the reform of police institutions. Yet the collapse of these peace agreements left police reform efforts a distant memory, sacrificed amid the renewal of armed conflict. These cases illustrate a central argument of the policing literature: Policing is inherently political and, while police organizations can have some autonomous influence upon broader political developments, they tend to reflect rather than influence the broader political environment. Thus, international actors cannot expect tinkering with police forces to prevent the breakdown of peace processes in the face of a determined spoiler. Police Reforms Present in Word and Deed In several cases, including El Salvador, Guatemala, Mozambique, Bosnia, Eastern Slavonia, Northern Ireland, and South Africa, police reforms were included in written agreements and implemented. Northern Ireland’s “Good Friday” agreement (1998) stipulated that postwar policing would be impartial, nonpartisan, committed to the rule of law and human rights, representative of diverse communities, and professional. Agreements in El Salvador and Guatemala, as well as international interventions in the nonnegotiated cases of Haiti and Kosovo, resulted in the replacement of old security forces with new police forces trained in new academies under new doctrines. In Bosnia, agreed-upon “advisory” roles for the International Police Task Force opened the door to retraining, purging, inclusion of ethnic minorities, and acceptance of “democratic principles” among BosniacCroat Federation and Republika Srpska police forces. Security reforms in these cases were not entirely successful, although they generally represent more extensive implementation of security reforms than cases in the other three categories. In every case significant shortcomings of internationally backed police reform efforts emerged and gained prominence in the press and in public discourse. In these cases, as well as in other cases of implemented reforms such as Nicaragua, failure to effectively address public security problems undercut the benefits of peace for much of the population. These cases point to the difficulties and opportuni-

S-11

11/24/09

4:22 PM

Page 311

CIVILIAN SECURITY

311

ties for police and security reforms that confront local and international decisionmakers during peace processes. Several tensions confront implementers who want to address public security in both the short and the long run. Coping with these tensions requires planning, realistic assessment of public security threats and available resources, and institutional engineering to minimize the costs of implementing peace accords. Given the difficulties of getting warring parties to focus on such mundane, public-interest issues during negotiations, international actors need to be prepared to provide agenda-setting leadership on this issue, and must prepare to act quickly to assist in institutional development efforts once political support emerges. In El Salvador, for instance, the UN provided significant guidance to the parties in designing a new police institution that would benefit the public as a whole as well as in helping to protect the disarming guerrillas. The following sections address the challenges of short-term and longterm public security for post–civil war societies. We identify the tensions faced at each stage, assess the capacity of implementers to manage these tensions, and identify lessons learned and some recommendations for reinforcing international capacities. In doing so, we draw heavily upon the cases where signatories both agreed to and sought to implement public security reforms.

Protecting the Public During Initial Transitions to Peace Two sorts of interim public security challenges directly affect the success of peace implementation. First, if former combatants deem integration into new police or military forces to be vital for their physical security after demobilization, then problems within these security forces might spell failure for the overall mission. In El Salvador and Northern Ireland, some of the toughest aspects of peace implementation concerned the restructuring of old and new police forces. In contrast to deeper reforms crucial for longterm peacebuilding, failure to incorporate former enemies into reconstituted security forces can jeopardize combatants’ faith in the process and lead them to return to war. Second, if public security deteriorates so precipitously that ethnic hatreds can resurface or organized crime can dominate political life and economic development, then both peace and democracy are at risk. In cases where entire social groups became involved in warfare, ethnic cleansing, or genocide, an absence of interim security may permit vengeance attacks to surface. Furthermore, war economies often concentrate economic power in the hands of very few people who seek to preserve their gains and translate them into political protection after war. The entry of powerful armed factions into organized or transnational crime fuels divisions over political and

S-11

11/24/09

312

4:22 PM

Page 312

EVALUATING IMPLEMENTATION TASKS

economic spoils in the short run, jeopardizing commitment to the peace process for those excluded from licit or illicit wealth opportunities after conflicts. In the long run, of course, international implementers risk leaving the country in the hands of powerful blocs of thugs or combinations of political parties, armed factions, and organized mafias. If exploitative, powerful political mafias dominate the postconflict landscape, then peace itself (and expensive international efforts to preserve it) may be worth little to those who survived the war. The challenge of organized crime requires an effective interim security and justice mechanism to deter, investigate, adjudicate, and punish criminal behavior. But postwar settings present a series of dilemmas and poor organizational options for addressing the “transitional public security gap.”8 Someone needs to provide security, and international implementers may cobble together an emergency civilian force, rely upon preexisting indigenous security forces, constitute new security forces, or provide interim public security with international military or police forces. Every option has its merits and its drawbacks, and political circumstances shape which is chosen for a particular situation. The absence of effective interim security capability at the international level has become a more serious and recognized problem for complex peace operations. Interim Security Option 1: A Quick-Start Local Civilian Force Because preexisting forces may be in disarray or lack popular support, implementers might try to assemble a hastily formed transition force from civilian personnel. Yet such a force is likely to be hopelessly ineffective if built from scratch. For example, some 900 Haitian refugees recruited in Cuba proved to have too little education and experience to perform any policing functions beyond static duties after the U.S.-led occupation of Haiti.9 A more successful model was the Transitory Auxiliary Police (PAT) in El Salvador, in which academy cadets were deployed temporarily under UN leadership to provide basic security in former conflict areas where the National Police were not welcome. This regime worked well, but largely because it applied to relatively isolated rural areas with virtually no organized crime and where the public had strong incentives to support the interim police. In the end, any emergency local civilian force will only be able to confront serious security threats with the backing of some more experienced public security force, either civilian or military. Interim Security Option 2: Preexisting Local Forces Wary of assuming interim policing duties themselves, international implementers have generally turned to existing (or reconstituted) indigenous

S-11

11/24/09

4:22 PM

Page 313

CIVILIAN SECURITY

313

forces, with international monitoring. Drawing on existing security forces or victorious rebel groups (such as the KLA in Kosovo) holds several advantages. Such forces are experienced, already deployed, organized, and paid for. They have generally received some technical training before or during the conflict, though often such training was more oriented toward political policing or military operations than conventional law enforcement.10 Even in cases where preexisting forces are no longer intact, their remnants may offer potentially valuable resources. In Somalia, for instance, former personnel of the police had skills, and a positive reputation for professionalism, that international implementers sought to exploit by reconstituting the police, even in the absence of a working government or legal framework.11 This option has several drawbacks. Existing local police forces often include many individuals with histories of political violence, provocation, and human rights violations. Sometimes vetting is done to remove the most egregious offenders. Given the urgency of having some kind of interim force, screening processes are usually superficial and based on inadequate information. Thus existing forces can pose a threat to the security of disarmed rebels, their supporters, returning refugees, or minority ethnic groups. They are also likely to violate human rights more generally, and may pose a threat to nascent democratic institutions. Often, governments attempt to transfer military personnel into existing police forces, circumventing military demobilization plans. UN Civilian Police (CIVPOL) monitors therefore must vigilantly regulate the actions of indigenous police, who may otherwise undermine the peace process. Interim Security Option 3: International Military Forces Despite the drawbacks, international actors often depend on existing forces, because they lack the resources and interest to provide civilian security during peace operations. Most military forces are not appropriate to public security tasks, since their training, equipment, and doctrine emphasize use of overwhelming force, rather than the controlled application of force necessary for police work. Military planners tend to view policing as a morass, a role that is likely to involve peacekeepers in conflictive situations with the local population, and from which it will be difficult for military forces to extract themselves.12 If military peacekeepers do take on interim security, as in Somalia, Bosnia, and other places, they attempt to hand off that role to some other actor as quickly as possible. This aversion to the public security role is driven largely by the political importance in donor countries, especially the United States, of avoiding casualties and minimizing the duration of deployments.13 Military reluctance to undertake public security tasks has sharply increased the prominence of policing issues in international peacekeeping.

S-11

11/24/09

314

4:22 PM

Page 314

EVALUATING IMPLEMENTATION TASKS

Driven largely by military protection concerns, policymakers (especially in the United States) have started to try to augment the roles and availability of UN CIVPOL and other international police personnel in interim public security. One novel approach is the use of constabulary or gendarme-type police units who combine police training and military capabilities. Such forces are trained in law enforcement and military tactics, deployed in units rather than individually, and are more heavily armed in preparation for well-organized attacks and disturbances. In Bosnia and Kosovo, “Mobile Support Units,” similar to the Italian Carabinieri and the French Gendarme, have been used effectively to defuse tense public-order situations through dialogue rather than the use of force. 14 Unfortunately, the need for gendarme-type forces for deployment abroad in peace operations outstrips available resources. Short of serving as temporary police, military forces can contribute significantly to reassuring civilians and reestablishing a measure of public security. That reassuring effect can be short-lived if military forces are unwilling or unable to act in the face of mob violence or ethnically motivated attacks. The degree of public security provided by military forces during initial phases of a peace operation depends heavily on the rules of engagement (ROE) for the military forces. If they are empowered to act to protect the public, they can be quite effective. If their ROE only allows for selfdefense, they may not act to protect civilians. Interim Security Option 4: International Civilian Police Personnel One of the most widely discussed options for public security in transitional settings is international civilian police personnel. The most prominent pool of such professionals are the UN CIVPOL, who were seen in 2000 as the most promising alternative to local forces for providing interim security after conflicts. Between 1998 and mid-2000, the number of CIVPOL deployed around the world rose from 2,800 to 5,500. Yet CIVPOL are often unprepared for serious security threats and unable to confront violent organized criminal networks. Only in 1999, in both Kosovo and East Timor, did CIVPOL first act as the sole authorized executive law enforcement authority in a territory. Previously, CIVPOL had in cases such as Cambodia and El Salvador conducted investigations and arrested suspects, at times on an ad hoc basis without clear legal status, but always in nominal support of local police forces. Yet these international police officers are generally recruited with the expectation that they will serve as monitors, not as law enforcement officers in foreign societies. Indeed it seems naïve to expect effective campaigns against well-organized criminal networks, from cops who are thrown together from disparate countries, who do not speak the local language, who have little contact or

S-11

11/24/09

4:22 PM

Page 315

CIVILIAN SECURITY

315

experience with the local population, who are deployed for only six-month rotations, and who operate within murky or nonexistent legal systems. In the Bosniac-Croat Federation, even well-planned efforts by heavily armed military units to take down the organized criminal network encountered resistance and obstacles. Organizational limits on CIVPOL are compounded by problems with their availability and qualifications. Recruitment of civilian police for international missions is difficult: police have constant peacetime duties and thus cannot be spared quickly and for periods of several months. Quality is also a problem. Even after the UN raised recruitment standards for CIVPOL contingents in the late 1990s, some international police officers reported for duty in Kosovo without requisite sidearms, language ability, or any experience walking a beat. Countries such as the United States and Great Britain that lack national police forces must call upon officers from small, resource-scarce local police forces, over whom national governments have no authority. U.S. efforts to hire active or retired police officers through private recruiting firms have resulted in very uneven quality of personnel.15 As a result of these constraints, the UN coordinating office for CIVPOL in New York has often been unable to fill all the positions authorized for peacekeeping missions, settling for slow deployments and “second- and third-tier choices.”16 Supplements to CIVPOL merit exploration as well. The deployment of Mobile Support Units in Bosnia and Kosovo has helped bolster weak CIVPOL capabilities and might eventually be deployed under civilian, rather than military, command. “Police Support Units” drawn from specialized riot-control units in donor countries were deployed for similar duties in Kosovo. However, if these units are called out of their barracks only to repress unruly crowds, rather than to constantly patrol and proactively interact with the population to deter public-order crises, they may prove counterproductive. In conclusion, transitional security arrangements may in some cases be critical for former combatants’ perceptions of their own security and for stemming violent crime waves that may solidify war economies. More generally, however, they are important because they impact how the public views the benefits and costs of peace. Unbridled crime waves can undermine popular faith in democracy and empower authoritarian responses and actors. Although all available options involve some trade-offs and disadvantages, some progress could be made in management of transitional security problems if the international community were to reinforce its institutional capacity in some areas. Through a combination of increased public safety training for military peacekeeping forces, increased use of military police capabilities already available, and wider use of gendarme-type support units, the international community could enhance its own ability to provide

S-11

11/24/09

316

4:22 PM

Page 316

EVALUATING IMPLEMENTATION TASKS

transitional security. This is particularly important in contexts where the available local forces are too biased, too undisciplined, or otherwise threaten public safety and political stability. In addition, the UN should continue its efforts to improve the availability, quality, consistency, and management of CIVPOL. International implementers can link delivery of assistance to local cooperation with CIVPOL, thereby enhancing the ability of CIVPOL to obtain information, effectively monitor local forces, and exert leverage regarding local police conduct.

Protecting the Public in the Long Term: Institution-Building Preexisting police forces, rather than international actors or new local personnel, are usually called upon to provide public security on an interim basis after peace settlements. Yet these forces are often held in disrepute, implicated in horrific abuses, and oriented toward protecting a state and regime that are no longer legitimate. Given the serious flaws of existing public security institutions in most post–civil war contexts, major efforts are generally needed to reform these institutions, or replace them, if international actors seek to leave behind a society that enjoys the minimal rule of law, observance of human rights, and democratic processes. Unfortunately, long-term reform and development of police are even more likely to be neglected in peace accords than interim security, and the capacity of the international community is even weaker in this area. Reform of public security institutions presents implementers (whether international or domestic) with a series of difficult dilemmas. This section reviews the choices encountered, illustrates them with examples, and suggests explanations for patterns observed. Interim Security and Long-Term Development Decisions regarding interim security can impose significant costs on prospects for long-term public security. Where existing forces are retained for interim security, they are more likely to be effective if their members anticipate the chance to be part of whatever new force is to be created. Where the vast majority of the police on duty knew that they would be excluded from a new force, as in El Salvador and Haiti, interim police morale and motivation were poor, and many police engaged in crime. While purging most of the existing force makes it easier to establish a new institutional culture in the police, it may also contribute to such an increase in criminal activity that the new force faces an impossible challenge just as it begins to deploy. Conversely, in cases where the logic of interim security dominated, as

S-11

11/24/09

4:22 PM

Page 317

CIVILIAN SECURITY

317

in Guatemala, former members of the old forces completely dominate the new institution, making it difficult for the new force to break from corrupt and abusive practices of the past.17 The Guatemalan attempt to develop a National Civilian Police has been shaped by the government’s overriding concern with the interim security challenge. Influenced by a powerful military, and pressured by a rightist opposition party demanding law and order, the Guatemalan government recycled existing personnel, recruited through the personal networks of provincial governors, allowed low standards for selection, abbreviated training, and neglected internal control mechanisms.18 These measures, though helpful in preventing a security vacuum, have likely compromised the long-term effectiveness and professionalism of the force. There are no cost-free solutions to this “demobilization dilemma,” but having adequate interim security provisions can help provide a climate in which longer-term institutional development can be done with less expediency. International forces could be particularly useful in buying new governments time to develop better national forces. Since this is seldom possible, an alternative is to have national military or guerrilla forces provide interim public security functions, preferably under close international supervision. This is only possible, of course, where sufficient confidence has developed between the former rebels and the government that military deployments for interim security are not seen as an immediate threat to the former rebels or their supporters. Involving the military in public security may set an unfortunate precedent by blurring the line between military and police roles, but it may be less harmful than expedient approaches that result in unreformed, abusive police institutions in the long run. Models of Policing Those peace agreements that do address public security reforms have often granted wide latitude to implementers, especially international actors, in designing, training, and equipping new police forces. What model should be adopted? Adoption of foreign models means that policing may not respond to the realities of the society, including failing to address the particular security problems associated with the causes of conflict in the first place. Yet reliance upon local structural and doctrinal models may simply re-create exclusionary structures and recruitment patterns, and reinforce doctrines that might sow the seeds of future conflict and leave sectors of the civilian population unprotected. Balancing between the imperative for change and the imperative for cultural appropriateness is not easy. International technical assistance donors must make the effort to familiarize themselves with local conditions, and seek input, rather than attempt to transplant complete policing systems.19 In situations where several bilateral donors are involved in developing

S-11

11/24/09

318

4:22 PM

Page 318

EVALUATING IMPLEMENTATION TASKS

new or reformed police institutions, there is potential for confusion as different national contingents provide distinct and sometimes contradictory advice. This can be dealt with in part by having different nationalities focus on different specializations—one on general policing, another on investigations, and so forth. Where multiple nationalities participate in training, as has occurred where CIVPOL missions have provided training, confusion and contradictions can be minimized if the mission prepares a common field-training syllabus. The development over the past decade of international norms regarding basic standards for police conduct, and especially for police use of force, helps provide guidance to different training contingents.20 Police Force Composition Some warring parties insist that their combatants have the option of joining state military or police forces. By reconstituting such forces, guerrillas hope that their members will not suffer persecution from state forces once they lay down arms. Former opposition members generally make up a minority of new or reconstituted forces, but even relatively small representation seems to be sufficient to generate greater transparency and greater confidence on the part of former rebels or their supporters. Thus, the incorporation of previously disenfranchised political and social groups into the police can be both a means of political reconciliation as well as a source of legitimacy for the new police among some popular sectors. This popular or regional legitimacy comes at a price, however, as elite groups affiliated with the old regime may distrust a more inclusive police force, and deny it needed political support. Although some international peacekeeping personnel recognize the importance of including former enemies in new or reformed police forces, other international personnel (especially military and police) may equate attention to composition with politicization. Since many professional police are disinclined to emphasize composition, CIVPOL missions and police advisers from bilateral donors often press these issues tentatively, even where failure to incorporate opposition or distinct ethnic groups into the police represents a serious violation of a peace accord. In Bosnia, most international police officers were more comfortable enhancing police skills and capabilities than enforcing agreements to incorporate minority ethnic groups into cantonal police forces.21 Similarly, in Guatemala a European Union project implemented by the Spanish Civil Guard (GCE) deferred to the ladino-dominated government’s low priority on recruitment of Mayans into the police.22 High merit-based standards of entry into reformed police forces are important for the effectiveness and reputation of a new police force, or for the future evolution of forces that remain largely intact. Yet relaxing such

S-11

11/24/09

4:22 PM

Page 319

CIVILIAN SECURITY

319

standards to ensure representation of important political, religious, gender, or ethnic groups into the police force may be necessary to ensure that such groups neither become systematic victims of violence nor resort to violence. In several peacekeeping operations, international actors have been crucial to the unprecedented incorporation of women into new permanent police forces. Anecdotal evidence in some cases suggests that demobilized soldiers accustomed to violence and left jobless tend to engage in domestic violence at a higher rate, aggravating a serious problem in most societies.23 Other standards, such as stature requirements, could often be lowered to accommodate groups such as Mayan Guatemalans, without in any way compromising the future professionalism of the police. Unfortunately, governments sometimes resist these measures, even those that have no plausible costs in terms of professionalism, suggesting the use of standards as a means of exclusion. Competing International Interests International actors’ interests in keeping the accords on track in the short run sometimes conflict with longer-term interests in peace, security, and democracy. Keeping the parties satisfied sometimes requires international verifiers to ratify arrangements that may jeopardize future stability. Often, violent and unscrupulous characters are parties to peace agreements. Allowing them to participate in permanent institutional arrangements (such as political parties, police forces, or military command structures) may be necessary for keeping the peace process from faltering, but may undermine the future legitimacy and effectiveness of the institutions. International third parties also face resource limits and a reluctance to risk the safety of their personnel. Postconflict states are often hesitant or unable to pay for agreed-upon institutional reforms. Short-term international interests in keeping the peace at the lowest cost may mean giving short shrift to expensive, long-term projects such as building new police academies and equipping police forces, and for the extensive field training that is crucial for effective performance by newly deployed cops. Whereas institutional development requires long-term commitment of advisers and trainers, other interests may dictate withdrawing personnel and declaring success. In several cases, new or reconstituted police forces have deployed with less classroom and field training than would normally be expected, because of resource and time constraints imposed by international actors. Finally, international actors also have separate security interests that may not converge with the overall development of citizen-oriented, rightsrespecting systems of public security and justice. The very agencies that provide much of the personnel for international police advising, such as the Spanish Civil Guard or the U.S. Federal Bureau of Investigations, may care more about getting help in pursuing criminals they seek abroad than in

S-11

11/24/09

320

4:22 PM

Page 320

EVALUATING IMPLEMENTATION TASKS

helping long-term development. Any involvement of foreign agencies in assisting police brings the risk that the local police may be transformed into a “transmission belt” for the national interests of the donor state. 24 Participation of multiple donors can help protect the host country against excessive subordination of its police institutions by any particular outside power. Finding the Right International Police for the Job The international community is not well organized for deployment of specialists in training and developing new or reformed police forces. Because CIVPOL are recruited mainly for the monitoring tasks of reporting on the performance of transitional public security forces, they are not well prepared or organized to advise and support institutional development.25 The tasks of creating new police academies, drafting doctrine, restructuring police forces, and establishing specialized police units require specialized senior police managers and supervisors whose concepts of policing are as important as their skills. Bilateral programs such as that of the U.S. Justice Department may be properly structured to do this kind of work, but are few in number and limited in capacity. Moreover, no bilateral effort possesses the capacity to deploy field trainers to supervise and build upon classroom training. The Spanish Civil Guard has played a crucial role in countries such as Guatemala and Mozambique, but has made poor choices of personnel and has tended to transplant Spanish models without adaptation to local conditions.26 Commonwealth countries have provided assistance in Sierra Leone and South Africa, but these programs are relatively small. The European Union, though increasingly supportive of police development projects, depends on national agencies such as the GCE to implement projects. The emergence of new multilateral programs has done little to expand the pool of qualified technical assistance personnel, or improve the coherence and rationality of their deployment. It has also failed to contribute to improved continuity across missions. The Organization for Security and Cooperation in Europe (OSCE), for example, took over police development efforts from the United Nations in the Eastern Slavonian region of Croatia, and is responsible for training in Kosovo. Yet very few “lessons learned” are retained from mission to mission, and fresh senior police officers, new to international police development, continue to dominate police missions, reinventing the wheel with each new operation. In many cases, police development was an afterthought, not integrated into peace implementation. This reflects the absence of an institutional home for police development within the UN bureaucracy. CIVPOL, as already noted, is not capable of police development. The UN Development Programme (UNDP), though it has growing experience in this area, lies outside the Department of

S-11

11/24/09

4:22 PM

Page 321

CIVILIAN SECURITY

321

Peacekeeping Operations, and coordination has sometimes been poor between the UNDP and peace missions. Reform of the Judicial System Attention here has focused on the police, because they are the front line of defense for public safety and order. Yet police must be supported by prosecutors who can effectively pursue legal cases against suspects, courts that can process cases in a timely and just fashion, and correctional facilities with adequate capacity and conditions. The legitimacy of the police depends upon their representing a judicial system recognized by the public as being fair. If judicial reforms lag, and reformed police find that suspects are regularly released due to judicial process errors or corruption, there is a risk that police will turn to vigilantism out of frustration. Unfortunately, police reforms can be implemented more quickly than judicial reforms, largely because of the much lengthier training times needed to prepare prosecutors and judges. For all their complexities, police forces are simpler to change than the complex of laws, organizations, procedures, personnel, and traditions of judicial institutions. International actors have even less capacity to promote judicial reform, and domestic constituencies for judicial reform are often slow to form, especially in societies that have never experienced a well-functioning judiciary. Peace settlements are generally negotiated by the executive branch (in cases where a government exists), whose commitments often do not bind autonomous judicial branches of government. For all of these reasons, and because civil war adversaries do not typically view the establishment of dispassionate judicial institutions as a priority, judicial system reforms are neglected in most civil war settlements. Here again, international capacities are diffuse, consisting of various bilateral and multilateral programs that have little connection to the institutions generally responsible for peacekeeping or peace enforcement operations. External and Internal Controls Some police reform efforts have addressed external and internal controls on the police, which are crucial for helping to minimize corruption and abuse by police. However, such cases have been the exception, reflecting again the lack of detail in most agreements with respect to public security. In El Salvador, three different mechanisms were established under the draft police law. Delays in actually putting these into place allowed serious disciplinary problems to appear. In South Africa, the government has established the Independent Complaints Directorate, with its own funding, investigative powers, and offices. In other cases, such as Mozambique and Guatemala, provisions for internal controls were inadequate.27 Organized

S-11

11/24/09

4:22 PM

322

Page 322

EVALUATING IMPLEMENTATION TASKS

crime quickly gained significant influence within the Mozambican police, and groups such as Amnesty International have attributed dozens of cases of torture to the police. In Guatemala, the accords failed to mention internal controls for the civilian police. The government transferred a weak agency known as the Office for Professional Responsibility from the old police into the new, but its investigative authority, training, and independence were inadequate. Because internal controls are often weak, external controls can prove even more important in postwar settings. El Salvador’s experience demonstrates the importance of human rights ombudsmen and a relatively free press in instigating investigations of police corruption and abuse. International actors need to play a stronger agenda-setting role in this area, and develop greater capacity to assist in the development of effective oversight and control mechanisms. Failure to do so may result in tremendous waste, as corruption and criminality within police forces can quickly transform internationally provided resources into a menace to public safety rather than a source of protection.

Conclusion Most civil war adversaries, as well as international intermediaries and implementers, focus on ensuring the security of former combatants and preventing renewal of civil war. This emphasis is appropriate—after all, if civil war resumes, public safety is, in general, much reduced, especially in situations where the civilian population will be targeted by one or both sides. But where prospects for settlement are reasonably good, attention to who will provide public security after the fighting stops, and under what rules, can contribute to a more livable postwar environment for most of the population. Properly executed, police reforms can also enhance the security of the politically active population, facilitating the consolidation of democratic practices and helping to prevent renewed conflict. Unfortunately, civil war antagonists have few incentives to formulate detailed plans for apolitical police and justice institutions, and have considerable incentives to keep their options open in hopes of gaining control of these functions and using them for political advantage. In a few cases, with El Salvador the clearest example, the international community has convinced the parties to accept such arrangements as part of the peace settlement. In a few others, such as Northern Ireland and South Africa, domestic parties have made pacts regarding police reform on their own initiative. Too often, however, international actors have joined the domestic parties in ignoring public security. In cases such as Cambodia and Nicaragua, the initial conception of policing and police reform proved inadequate. International actors took their cues excessively from local actors in these initial efforts, which focused on incorporating former enemies. In all three cases,

S-11

11/24/09

4:22 PM

Page 323

CIVILIAN SECURITY

323

it subsequently became clear that more profound institutional changes were needed. We have argued that international actors need to pay more attention to public security, both in guiding the negotiations of peace agreements and in developing international institutional capacities to provide interim security and support long-term development of new or reformed police forces in postconflict countries. International actors can do more to shape the agenda and definition of police, military, and judicial reforms in postconflict settings beyond the immediate interests of the parties. In some situations, it will not be possible to incorporate these concerns into formal agreements, and some processes will go forward in the absence of an agreement (e.g., Haiti, Kosovo). In those cases, the international community should use its informal powers to encourage local actors to stretch the boundaries of their thinking beyond composition of police and purging bad elements, extending it to include a more democratic notion of state-society relations, based on accountability, transparency, and participation. Restructuring police forces is no silver bullet. No amount of training or institutional development work will produce positive results where domestic actors are not really interested in changing the status quo. Nor can security reform survive in the face of a determined spoiler. Policing and administration of justice are more reflections of the environment within which they occur than shapers of that environment. Yet in those situations where public security reforms are incorporated into peace accords, or where peace implementation has ended and where a critical mass of local actors perceive a need to revisit public security issues, the international community has the opportunity to assist long-term conflict prevention by generating public security structures that can protect the people from abusive, unaccountable, and criminal behavior at the hands of governmental, criminal, or antagonistic social actors.

Notes 1. Charles T. Call, “Why the World’s Most Successful Peace Processes Produce the World’s Most Violent Countries,” paper delivered at the International Studies Association conference, Chicago, March 1999. 2. David Bayley, Patterns of Policing: A Comparative International Analysis (New Brunswick, N.J.: Rutgers University Press, 1985), p. 189. 3. Examples include Somali police during the Siad Barre regime, who enjoyed a positive public reputation despite the authoritarian nature of the regime; and Nicaragua, where since 1979 the police have achieved a positive reputation for effectiveness and good community relations, despite governments of questionable legitimacy and stability. 4. Charles T. Call, “From Soldiers to Cops: ‘War Transitions’ and the Demilitarization of Policing in Latin America and the Caribbean” (Ph.D. diss., Stanford University, August 1999), pp. 62–65.

S-11

11/24/09

324

4:22 PM

Page 324

EVALUATING IMPLEMENTATION TASKS

5. Alice Hills, Policing Africa: Internal Security and the Limits of Liberalization (Boulder, Colo.: Lynne Rienner, 2000); and Otwin Marenin, Policing Change, Changing Police: International Perspectives (New York: Garland Press, 1996). 6. William Stanley and Charles T. Call, “Building a New Civilian Police Force in El Salvador,” in Krishna Kumar, ed., Rebuilding Societies After Civil War: Critical Roles for International Assistance (Boulder, Colo.: Lynne Rienner, 1997), pp. 107–134. 7. Mark Malan, “Peace-Building in Southern Africa: Police Reform in Mozambique and South Africa,” International Peacekeeping 6, no. 4 (winter 1999): 171–190. 8. On different “public security gaps,” see Robert B. Oakley, Michael J. Dziedzic, and Eliot M. Goldberg, eds., Policing the New World Disorder: Peace Operations and Public Security (Washington, D.C.: National Defense University Press, 1998). 9. Call, “From Soldiers to Cops.” 10. Martha Huggins, Political Policing: The United States and Latin America (Durham, N.C.: Duke University Press, 1998); and Michael McClintock, The American Connection, vol. 1, State Terror and Popular Resistance in El Salvador (London: Zed Books, 1985). 11. The Unified Task Force (UNITAF) and the second UN Operation in Somalia (UNOSOM II) attempted to reconstruct a Somali police force using personnel who had served in the police under the Siad Barre government. Despite the authoritarian character of that regime, the police were widely respected. Unfortunately, the new police could not overcome the absence of a government to lend them authority, and they became irrelevant in much of the country as heavily armed factional armies resumed open warfare. Lynn Thomas and Steve Spataro, “Peacekeeping and Policing in Somalia,” in Oakley, Dziedzic, and Goldberg, Policing the New World Disorder, pp. 175–214. 12. Michael Dziedzic, introduction to Oakley, Dziedzic, and Goldberg, Policing the New World Disorder, pp. 3–18; and Erwin A. Schmidl, “Police Functions in Peace Operations: An Historical Overview,” in Oakley, Dziedzic, and Goldberg, Policing the New World Disorder, pp. 19–40. 13. Conversations with military participants in the National Defense University project on public security during peace operations, August 1996, September 1997. 14. Personal interview by Charles Call with Enzo Coppola, deputy director of the Multinational Specialized Unit (MSU) under the Stabilization Force (SFOR) in Bosnia; Washington, D.C., 1998. 15. U.S. officials referred to the U.S. police contingent in Haiti as “near the bottom of quality” and as “unemployed mall guards.” Author interviews with U.S. officials who requested anonymity, Port-au-Prince, August 1996; Washington, D.C., October 1996. A UN consultant to the Civilian Police Unit in New York reported in a telephone author interview in June 1997 that some 40 of the 200 U.S. police serving in the UN mission in Bosnia had deserted. 16. Author interview with UN CIVPOL official in New York who requested anonymity, October 1996. 17. Roughly 90 percent of the old National Police (PN) was “recycled” in Guatemala, and the entire office corps of the new National Civilian Police (PNC) consists of former PN members. Interviews by William Stanley with UN and PNC officials, Guatemala, May 1999. 18. Author interviews with UN officials and international police advisers, Guatemala, May 1999.

S-11

11/24/09

4:22 PM

Page 325

CIVILIAN SECURITY

325

19. One illustration of this problem is the work of the Spanish Civil Guard (GCE) in Guatemala, who wrote a disciplinary code for the new Guatemalan civilian police that is a near replica of the GCE code and, in practice, is far too complex and cumbersome to work effectively given the low level of education of most Guatemalan police officers and commanders. 20. UN Document ST/CSDHA/16, “Compendium of United Nations Standards and Norms in Crime Prevention and Criminal Justice.” 21. Author interviews with several UN CIVPOL officers, Sarajevo, March 1999. 22. The term ladino is used in Guatemala to refer to people who identify themselves as belonging to the dominant Hispanic culture, as distinct from the various (mainly Mayan) indigenous cultures. 23. NGOs in El Salvador reported this finding in interviews in 1993–1995. Tracy Fitzsimmons, “The Post-Conflict Postscript: Women in Central America, Haiti, and Bosnia,” paper presented at Latin American Studies Association conference, Miami, March 2000. 24. Huggins, Political Policing, p. 4, quoting Hannah Arendt, The Origins of Totalitarianism (New York: Harcourt Brace, 1951). 25. Charles Call and Michael Barnett, “Looking for a Few Good Cops: Peacekeeping, Peacebuilding, and UN Civilian Police,” International Peacekeeping 6, no. 4 (winter 1999): 43–68. 26. William Stanley, “Building New Police Forces in El Salvador and Guatemala: Learning and Counterlearning,” International Peacekeeping 6, no. 4 (winter 1999): 113–134; and Malan, “Peace-Building in Southern Africa.” 27. Ibid.

S-11

11/24/09

4:22 PM

Page 326

S-12

11/24/09

4:22 PM

Page 327

12 Building Local Capacity: From Implementation to Peacebuilding JOHN PRENDERGAST AND EMILY PLUMB

The ending of civil war does not end societal suffering, division, and conflict. Most postwar countries continue to be beset by ethnic, political, economic, social, and religious rifts. Large populations of these countries have been traumatized by the horrors of war. The consequent psychosocial impacts that invariably result from protracted civil conflict can be as harmful as the physical damage wrought by the war itself. Economic and political root causes of prior divisions remain intact or are even exacerbated long after the fighting has ceased. All of these factors point to the need for aid and assistance to go beyond the signatories to peace agreements and to build a strong capacity for peace within civil society. Peace implementation has mostly concerned itself with national-level peace agreements, international intervention, and peacekeeping forces. Less attention has been paid to local reconciliation processes and initiatives, which makes the task of evaluating these latter interventions all the more challenging.1 Yet it is our contention that local civil society organizations (CSOs), due to their proximity to local-level actors at the end of the implementation chain, can play a key role in ensuring that peace is sustainable. This chapter aims to address the value that local capacity in the form of CSOs can contribute to postsettlement peacebuilding. We argue that in order for peace agreements between warring parties to lead to durable peace, there needs to be, alongside the top-down implementation of the peace agreement, concurrent bottom-up processes aimed at constructing a new social contract and healing societal divisions. CSOs can have an impact through creating or supporting these bottom-up processes and through engendering societal ownership of the peace agreement. Critics of our argument will point out that many of the examples of innovative civil society peacebuilding activities cited in this chapter oper327

S-12

11/24/09

328

4:22 PM

Page 328

EVALUATING IMPLEMENTATION TASKS

ate in environments of failure of the national-level peace agreement. They are therefore quick to dismiss the value of civil society activities in postconflict reconciliation. We disagree for three reasons. First, admittedly CSOs have little immediate impact on the success of peace agreements. Widespread civil society activity in support of peace cannot compensate for a process in which one or more parties to an agreement make a tactical decision to return to war. Nevertheless, as this chapter will show, where a good-faith effort toward peace is made, civil society efforts can make a difference. By building intercommunal links, initiating dialogue, and engaging people traditionally left out of the peace process, CSOs bring war-torn societies closer to reconciliation. Second, as several examples mentioned in this chapter attest, grassroots peacebuilding activities can flourish even in the midst of ongoing conflict at the national or regional level. Since CSOs help mobilize resources in ways that the war-ravaged state is often unable to do, their efforts should not be disregarded. Moreover, the reconciliation skills and programs introduced by CSOs may help defuse the roots of potential future conflicts (a contribution that is difficult to measure, but nonetheless important). Finally, as noted above, CSOs can promote societal buy-in for a peace settlement. Thus, even if past attempts at lasting peace have failed, grassroots pressure to reach an agreement may compel leaders to return to the negotiating table. This phenomenon is clearly demonstrated in the cases of South Africa and Northern Ireland. As Marina Ottaway claims, it was not only sanctions, international diplomatic pressure, and strong leadership that dismantled apartheid in South Africa. The activity of hundreds of CSOs was also a critical ingredient in the largely peaceful transition to a postapartheid political system.2 Similarly, the ongoing efforts to achieve peace in Northern Ireland are anchored in firm grassroots pressure on political leaders to end the conflict. Thus, the work of CSOs should not be considered irrelevant if the peace process fails at the national level. A national failure does not render local successes moot. Before we move to the substance of our argument, it is important to state a caveat. CSOs are not a panacea for rebuilding war-torn societies and building peace. CSOs can reinforce negative, conflict-producing elements of the economic and social structure of a given state, particularly one consumed by a war economy. CSOs are not by definition virtuous; they can be political, corrupt, and as prone to exacerbating conflict as they are to resolving it. The mere fact that an actor is local does not excuse international implementers from making sound judgments about an organization’s credentials as a credible and positive force for peace. It is obviously incumbent upon international actors to do their homework before creating partnerships on the ground. But having said all of this, we want to make a strong case for why it is in the interests of international actors to look for

S-12

11/24/09

4:22 PM

Page 329

BUILDING LOCAL CAPACITY

329

strong local partners for peace. Such partners will possess resources, perspectives, knowledge, and commitment that are unavailable to outsiders. Moreover, in the end, international implementers of peace agreements leave, and to the extent that there is local capacity to build peace, the greater the likelihood that short-term implementation success will lead to self-enforcing, self-sustaining peace.

CSOs: Advantages, Roles, and Facilitating Conditions The composition of civil society varies from country to country. In general, it encompasses both formal and informal interest groups and associations larger than the family, which act independently of the state to promote diverse interests in society. Civil society can consist of religious, traditional, and business leaders, as well as women’s organizations, student associations, unions, and NGOs. Civil society also includes commercial actors and professional associations. CSOs can help build intercommunal links that are critical to reconciliation. Although any of the CSOs described above can be involved in peacebuilding, the majority of reconciliation activities of which we are aware are undertaken by local NGOs. Therefore, this chapter focuses mostly on the challenges and accomplishments of grassroots NGOs. Moreover, the chapter emphasizes the activities of domestic groups rather than international NGOs. When we look at the former, it is with an eye toward how international agencies can support the involvement and participation of CSOs. The greatest comparative advantage that CSOs possess in peacebuilding is local knowledge and deep contextual understanding of barriers and opportunities to making peace at the local level. This local knowledge translates into a set of roles that CSOs can play to foster the implementation of peace agreements and longer-term peacebuilding: as transmission belts between communities and national governments, international NGOs, and international implementers of peace agreements, and as advocates for local communities in national policy debates. The ability of any given CSO to play this role effectively is enhanced by the presence of several facilitating conditions, including experience, coordination, and good-faith efforts by the new postsettlement government to initiate policy reforms. One distinct advantage CSOs can have over their international counterparts is their inherent understanding of the postconflict situation. CSOs not only are likely to be more familiar than a foreign NGO with the history of the conflict and the issues that surround it, but they are also likely to be better equipped to understand the needs of the community and more sensitive to cultural norms. As other chapters in this book argue, too often relief, rehabilitation, and reconciliation programs undertaken by international organizations lack clear insight into the sociocultural context in which they

S-12

11/24/09

330

4:22 PM

Page 330

EVALUATING IMPLEMENTATION TASKS

are operating and end up producing ineffective or harmful results. Thus it is critical to involve representative elements of civil society in peacebuilding activities, from the planning stages through implementation. The local knowledge and understanding that CSOs bring to peacebuilding allow them to fulfill the role of transmission belt between communities, national elites, and international implementers, gathering information about what is really happening and what is needed at the local level and passing it on to policymakers at higher levels. A key question is the extent to which CSOs can build links between the state and local communities. CSOs have a tendency to portray themselves as being at odds with the state, sometimes for good reason. Sometimes, however, creating a buffer between the state and the local community is counterproductive and leads to unnecessary competition and suspicion. CSOs face a major challenge in establishing their legitimacy as partners with the government in the postagreement consolidation process. Offering realistic ideas and alternatives is critical to establishing credibility. CSOs must develop networks to maximize their influence and should not shy away from advocacy. CSOs must equip themselves to identify bottlenecks to the consolidation of peace and keep the implementing authorities accountable. At the national level, it is imperative for a CSO and its international partner to realistically assess the commitment the formerly warring parties have to the peace accord, the sustainability of any peace agreement, the legitimacy of institutions and leaders in the governmental and nongovernmental sectors, and the degree to which aid and investment will either help consolidate peace or become a target of formerly warring parties.3 This is all further predicated on a CSO’s understanding of the nature, source, and effects of conflict. Having this kind of in-depth understanding of a society also allows CSOs and international partners to support peacebuilding efforts at the subconstitutional level, the level below the main signatories to the peace accord and the interests they represent. Given the erosion of sovereignty and the breakdown of institutions that have accompanied complex emergencies, it is critical that this level be accessed.4 From professional associations in Bosnia to elders councils in Somalia, there are opportunities to consolidate peace and promote reconciliation through cross-line activities. Although it would be ideal to have a set of preconditions met for CSO involvement in peace agreement implementation, postconflict situations often require risk-taking. A self-evaluation by the U.S. Agency for International Development (USAID) of its programs in postwar Mozambique found that projects had to proceed before ascertaining their technical feasibility, “a situation that would be unlikely in a more traditional development program.”5 It is better, therefore, to think of facilitating conditions or their

S-12

11/24/09

4:22 PM

Page 331

BUILDING LOCAL CAPACITY

331

absence for CSO engagement, rather than dealing in the more rigid category of preconditions. One of the most important facilitating conditions for CSO involvement in reconciliation activities is experience working on local problems well before peace accords are signed. Having in-depth familiarity with local conditions in zones of operation is critical for intelligently supporting reconciliation processes, an arena in which external interveners can easily do more harm than good. With a history of presence and interaction in waraffected regions, reconciliation activities undertaken by CSOs can expand on past local initiatives, rather than start from scratch. A base of experience also allows a civil society organization to be in a position to recognize positive openings in a postwar society that might provide room for reconciliation. A second facilitating condition for CSO success in reconciliation activities is coordination among the various implementing agencies to reinforce peace. Humanitarian, developmental, security, and political initiatives should operate in the context of a coordinated policy framework. As shown in other chapters in this book, such coordination is frequently absent and poses a challenge for policymakers. This argues for ensuring maximal CSO participation in implementation strategies and coordination mechanisms, particularly when there already is a UN mechanism for coordination. The most important facilitating condition for any reconciliation initiative, though, is the willingness of the postsettlement government and the former warring parties to address some of the root causes of the war. In the absence of at least a political opening to addressing past wrongs, reconciliation efforts can waste resources and energy. For example, if a civil war is caused by poverty, economic and political inequities, food insecurity, environmental degradation, overpopulation, and competition for scarce resources, then CSO reconciliation interventions must by definition address these economic and political roots of war, as well as other social and psychological factors. If structural factors are not directly addressed by reconciliation initiatives, such initiatives will have little impact, and may perhaps be counterproductive in environments where extreme injustice survives.

Integrating Local Capacity-Building into Peace Implementation International NGOs and domestic CSOs play their largest roles during the postwar implementation period in the arenas of humanitarian relief and economic rehabilitation. It is in these areas that the creative use of external resources in support of civil society’s role in postwar reconstruction is perhaps most promising. Given the decline in foreign aid, the likelihood that

S-12

11/24/09

332

4:22 PM

Page 332

EVALUATING IMPLEMENTATION TASKS

there will be plentiful new resources for reconciliation initiatives is remote. Consequently, a hopeful innovation will be the integration of reconciliation objectives into existing postwar priorities and programs. Mary Anderson points out that international NGOs should undertake their activities “in a way that provides space and voice and motivation for local people to disengage from the conflict that surrounds them.”6 Working closely with domestic civil society constituencies is required if international agencies are interested in supporting postconflict peacebuilding. To enhance this, international NGOs are utilizing or considering special training initiatives in conflict prevention and resolution for their program staff in countries that have experienced or are experiencing violent conflict. Besides skill-building, this kind of capacitating of staff reinforces the importance of postconflict aid in consolidating peace processes and promoting reconciliation.7 Anderson calls for aid strategies that consciously support existing local “capacities for peace.” Programmatically, she recommends that agencies build into their programming the objectives of providing safe space for people across lines of division to meet (by hiring locals from or targeting benefits to all sides of the conflict), providing voice to those who want to discuss reconciliation (by supporting peace conferences or publications), and providing incentives to promote disengagement from conflict (by focusing postconflict resources on collaborative initiatives).8 If international agencies create parallel labor markets and microeconomies that collapse when they depart, they have clearly contributed nothing to building the institutional foundations for peace consolidation. But if they utilize structures and processes that emanate from existing civil society, strengthening those “capacities for peace,” then the cause of implementing peace agreements can only be enhanced. At the operational level, donors like USAID can move beyond treating the symptoms of crises and use aid to mitigate conflict and address the economic disruptions at the root of conflict. As Susan L. Woodward argues in Chapter 7, agencies must continue to be more flexible in removing often artificial barriers between relief and development activities. Working with CSOs can play a key role in bridging the relief-development gap. For example, postwar reconstruction entails rebuilding infrastructure and economies, absorbing traumatic changes in social relations (as well as demobilized fighters), and rehabilitating every level of society—household, community, civil group, professional association, and government—with a grasp of social relations to avoid reinforcing inequality or deepening marginalization.9 The local knowledge that CSOs provide is crucial for reconstruction to contribute to peace. There are many options for agencies to integrate CSO peacebuilding efforts into their relief and development programming. In areas where communities or contesting military forces have frequently clashed, agencies

S-12

11/24/09

4:22 PM

Page 333

BUILDING LOCAL CAPACITY

333

might support existing intercommunal mechanisms within civil society to discuss needs and strengthen cross-line communication. These mechanisms often already exist, but are under extreme pressure. They take such diverse forms as kinship ties between neighboring communities that are activated during periods of extreme stress, local religious committees or communities who incorporate multiple groups, or border chiefs or elders whose responsibility it is to negotiate between communities in the aftermath of intercommunal conflict. Strengthening cross-line communication may have no impact on the leaders of formerly warring parties, but perhaps will lead the peaceseeking elements of neighboring communities to see mutual interests in cooperation. With communication and cooperation partially restored, neighbors can continue to trade, graze animals, and maintain other ties even while war leaders continue to fight. As Ken Menkhaus notes, “Often this simple provision of political space and a line of communication is vital to the process of conflict resolution and can be the greatest contribution international organizations can make. . . . It focuses community attention on the very tangible benefits of cooperation and conflict resolution, rather than on peace as an abstraction.” 10 These cross-line mechanisms can be critical support structures for the implementation of peace agreements. Interventions can also support local peacemakers and constituencies within civil society by virtue of whom the agencies utilize as local partners. Local institution-building that supports participatory democratic processes and pluralism can be an important contributor to peace agreement implementation. Some forms of social organization are geared toward service provision; others, such as elders councils, are appropriate for conflict resolution; still others make sense as profit-seeking ventures. Relief and development aid not preceded by or including the strengthening of domestic conflict management systems can create and exacerbate conflicts, such as the cattle rustling endemic to pastoral groups, whose development as much concerns conflict management as economic improvement. Similarly, aid to farmers can raise tensions where it seems to legitimate disputed occupancy. Supporting community institutions that can resolve competing claims before distributing inputs is an important prerequisite for success in some areas.11 Despite their immediate exacerbation of poverty and risk, wars in impoverished countries can be positive agents for change. This can be both direct—when people fight for justice and equality—and indirect, when war prompts women, community, and other elements of civil society to emerge.12 War can particularly be a positive catalyst vis-à-vis women and gender relations, as women’s groups are often among the new civil organizations that arise to challenge conflict and promote peace and rights. Aid organizations can support this process and help women to consolidate their gains against pressures to retrench when war ends.13

S-12

11/24/09

334

4:22 PM

Page 334

EVALUATING IMPLEMENTATION TASKS

Nevertheless, it is important not to be uncritically sanguine about the role of women in peacemaking. Just as it is easy to see all CSOs as virtuous partners for peace, humanitarian organizations and other interveners may tend to mythologize the intentions and role of women, who often crave justice, even if that means the continuation of war. Hugo Slim of Oxford Brookes University elaborates: “I think we need to be aware that a new fashion of women peacemaking projects is emerging in NGOs. These may be to the 1990s what women’s crafts and basket weaving projects were to the 1980s, and be based on similarly simplistic assumptions and generalizations about women’s roles in society. I am wary of this new wave of ‘designer peace projects’ for women.”14 A second topic relevant to war and change concerns traditional authorities and their role in peacebuilding. The marginalization of traditional authority in some locations has had profound consequences, including the severing of channels for transmitting basic values and a consequent increase in the raping and killing of women and children.15 In the efforts of international agencies to build local capacity and enhance participation, questions need to be constantly asked about whether traditional authority structures are being undermined and—given their repressive nature in some places and their role in preserving the social fabric in other places— whether they should be. External interlocutors cannot choose not to address these issues in their quest for neutrality, because most interventions have the potential to empower or disempower certain segments of the community based on the composition and structure of planning and delivery mechanisms. Constantly monitoring these effects should be a responsibility assumed by outside agencies.

CSO Initiatives in Peacebuilding Beyond the value that they add to economic and social reconstruction, CSOs can contribute to long-term processes of reconciliation, conflict resolution, and recovery. Since the early 1990s, CSOs have become involved in important, innovative projects to address issues of long-term peacebuilding, including addressing trauma, organizing problem-solving workshops, training for conflict management, creating peace media, assembling peace committees, resurrecting indigenous mechanisms for conflict management, encouraging collaborative community activities, and supporting democracy and human rights. Psychosocial Trauma Initiatives CSOs can play important roles in restoring individual and community mental health. At least 10 million children have suffered from war trauma dur-

S-12

11/24/09

4:22 PM

Page 335

BUILDING LOCAL CAPACITY

335

ing the past decade, according to the UN International Children’s Emergency Fund (UNICEF).16 War-related trauma experienced by children is typically divided into two categories: acute and chronic. Acute trauma is defined as a child’s exposure to a single abnormal incident in which the parent/child relationship remains intact and the child, with the parent’s support, is able to readjust his or her perceptions and return to a normal, safe environment, dismissing the unusual experience as an “incident.” Chronic trauma (post-traumatic stress disorder—PTSD) occurs over a longer period and includes a multitude of symptoms that impede a child’s ability to develop emotionally. A combination of circumstances such as displacement from parents, prolonged exposure to violence, and combat participation in combination with unmet food and material needs result in personality and emotional alterations in many children.17 Conventional responses to diagnoses of trauma (individual or communal) include a variety of psychosocial projects. In Bosnia and Croatia, one evaluation found 117 local and international organizations running 185 psychosocial projects. Two-thirds offered direct psychological services. Based on a loose definition, the reviewers found nearly 1.5 million Croatians and Bosnians to be suffering from trauma.18 Political and social reality dictates that program initiatives should be placed in a broader social context if the programs are to be beneficial to the community.19 After the Rwandan genocide, dozens of NGOs undertook psychosocial programs driven by an assumption of mass traumatization, despite little familiarity with local culture and conditions. Western diagnostic systems and treatments were applied, with little relevance to Rwandan realities. Much of the distress that was diagnosed as trauma was in fact normal coping mechanisms in operation.20 (Furthermore, interventions aimed at traumatized children in centers for unaccompanied minors did not utilize standardized criteria and practices, according to the multidonor evaluation, and suffered from little monitoring and follow-up.)21 Similarly, PTSD diagnoses in Nicaragua were not reliable indicators of the need for psychological treatment.22 In fact, revisiting the interviews and data from PTSD studies in Rwanda finds that a very high percentages of people were not sad, were interested in work and play, and felt able to protect their family and self. Asking questions solely about PTSD limits responders to addressing the victimization and wounds of affected individuals, rather than their resilience. It reinforces the passivity of “victim” and the knowledge of the “expert.”23 Partnerships between domestic CSOs and international NGOs with resources often produce the most productive results. Soon after the Rwandan genocide, children were being warehoused, and the trauma of what had preceded their arrival was being compounded by the horrible conditions of their new setting. In Kigali, international NGOs and local CSOs teamed up to provide counseling services and support local forms of social

S-12

11/24/09

336

4:22 PM

Page 336

EVALUATING IMPLEMENTATION TASKS

organization for the children, such as scouts, traditional dance groups, football teams, and church groups. The partner organizations held community discussions about the future of the children. When partnerships prosper between international NGOs and CSOs, aid agencies can act as a guide or facilitator in helping the community expand and develop indigenous strategies. Culturally sensitive programs will enable children to more freely and constructively process their trauma and incorporate their emotions into their worldview. Assisting communities in devising strategies to help protect children from the effects of war is an important role for aid agencies, who can help members of the community (teachers, parents, social workers, police) better understand aggressive behavior displayed by children suffering from chronic trauma and how a parent’s reaction can influence a child’s behavior. The establishment of daily routines and stable structures by teachers and parents can also help former child combatants work through their emotions in a more adjustable manner. Psychosocial initiatives are relevant to the process of postconflict reintegration. In Chapter 6, Joanna Spear raises the challenge of reintegrating demobilized soldiers into their home communities. Spear cautions against standardizing these programs, noting that they should be customized to meet local needs. While most demilitarization, disarmament, and reintegration programs rely on the support of international organizations, CSOs can be ideal in implementing them and seeing them through to the more advanced stages of reintegration. For instance, CSOs can help former combatants find new employment, teach them marketable skills, treat PTSD cases, and meet the special needs of former child and female soldiers. Such programs are the focus of current international and regional planning in peace agreement implementation in Sierra Leone, the Democratic Republic of Congo, Angola, Rwanda, Congo-Brazzaville, and Burundi. Problem-Solving Workshops and Processes Problem-solving workshops (PSWs), also known as “third-party consultations” or “interactive problem solving,” create channels for unofficial representatives of conflicting parties who seek to collaborate in analyzing and finding solutions to particular issues of division. PSWs are not intended as a substitute for ongoing negotiations, but rather represent an informal widening of participation in peace processes and a deepening of the analysis of problems and potential solutions. The building of relationships in a politically safe space is prioritized. Some recent peace processes have had PSWs operating in the background, including the Palestinian-Israeli conflict as well as peace efforts in Northern Ireland and Spain.24 Postconflict PSWs are held with a variety of segments of civil society.

S-12

11/24/09

4:22 PM

Page 337

BUILDING LOCAL CAPACITY

337

For example, the International Center for Peace Initiatives in Bombay held a process workshop for members of Parliament from a number of South Asian countries to deal with postconflict situations in the region. Parliamentarians explored how legislation and resolutions might help reduce tensions and consolidate peace, how their public speaking might advocate peacebuilding, and how to conduct election campaigns that do not fan the flames of violence. Catholic Relief Services (CRS) and Kanyarwanda—a Rwandan human rights organization—sponsored conferences on reconciliation in Rwanda following the genocide, convening morally influential people to discuss strategies for reestablishing trust and confidence between Rwandan communities. Some of their conclusions include the importance of developing a culture that speaks the truth; the need for rehabilitation of Rwandan society; recognition of justice as critical for Rwanda’s recovery; the need to remove impediments to the return of refugees; and the need for a culture of sacrifice and goodwill. Following this meeting, CRS later sponsored a workshop organized by the Rwandan Association for Christian Workers that involved church, media, human rights, and development groups in examining the role of the church and its justice and peace mission. The InterAfrica Group (IAG) held workshops on “Women and the Making of Regional Constitutions” in postwar Ethiopia. The IAG brought together fifty women from five different regions within Ethiopia who work with women’s bureaus to examine the creation of regional constitutions. Issues examined included the role of regional women’s bureaus, NGO policies related to women at the regional level, and women’s involvement in the making of the national and regional constitutions. PSWs usually are held in academic contexts, providing what many consider to be safe space for open debate and discussion. The aim is to create a venue for interaction that would not otherwise exist under normal conditions. Some scholars have observed that contacts increase between participants after the conclusion of the workshop, and new networks are either born or strengthened. An obvious limitation of PSWs is what takes place when the workshops are over and participants return to their communities. Benefits and insights are often limited strictly to participants, leaving the responsibility on their shoulders for ripple effects. The results from these initiatives are clearly very long-term and modest. 25 Given the extremely limited impacts achievable from one isolated workshop, it is much more desirable to undertake a series of workshops, thus committing participants to a process and creating the option of including more participants.26 Distinct from PSWs are problem-solving processes (PSPs), collaborative initiatives focused on addressing a particular problem issue. An example of a PSP took place in Central America, where the Development Program for Displaced Persons, Refugees, and Returnees (PRODERE)

S-12

11/24/09

338

4:22 PM

Page 338

EVALUATING IMPLEMENTATION TASKS

attempted to promote and facilitate the postwar social and economic reintegration of uprooted people, while consciously supporting peacebuilding processes. This program had a number of positive impacts that prioritized reconciliation: “The program was a civilizing force in areas of weak civil institutions and a strong, pervasive military. It strengthened civil society by establishing humanitarian spaces that allowed for consensus-building and reconciliation.”27 Groups who were historically left out of political processes were consulted, and joint problem-solving processes for addressing health, education, income generation, and planning activities are being undertaken. Under PRODERE, numerous problem-solving and cooperative processes were initiated, an approach that is superior to the usual one of PSWs, which are often guilty of little implementation and follow-up. The World Bank has shown that networks of war veterans who form discussion groups and collaborative ventures can assist the reintegration of former combatants. Such initiatives “can be extremely helpful when social capital has been depleted,” concludes a recent Bank study.28 But although PSWs and PSPs may potentially contribute to peacebuilding, they have escaped hard evaluation. Ascertaining their specific impact in larger peace processes is next to impossible. No evaluative criteria have been developed to fully account for the disparate effects of these initiatives. Current evaluation is impressionistic and anecdotal, with little consideration of potential ripple effects.29 Conflict Management Training Some CSOs specialize in conflict management training, in order to educate communities and individuals in how best to address conflict. The Center for Conflict Resolution in South Africa, for example, trains political and community leaders in conflict resolution skills, and has brought together previously conflicting segments of society such as police and antiapartheid leaders for training sessions. John Paul Lederach observes that this and other initiatives enhance peacebuilding capacities of middle-range leaders and provide a forum to assemble people from across lines of division.30 Participants of numerous conflict management and resolution training forums have reported that one of the principal benefits of such training has been the learning of listening techniques, and more fundamentally, the importance of listening in order to understand other points of view and to resolve differences. When properly planned, training exercises can strengthen capacities of the state and civil society to search for creative solutions collaboratively. ABuGiDa, an indigenous Ethiopian NGO, has held training in conflict management in postwar Ethiopia, in which community mobilizers and party members participated. Also in Ethiopia, the Ad Hoc Committee for Peace and Development has held workshops in Harar, South Arssi,

S-12

11/24/09

4:22 PM

Page 339

BUILDING LOCAL CAPACITY

339

Wollaita, and Wollega at which a mixed group of elders and party officials met and discussed their differences. Africans for Humanitarian Action (AHA), an Ethiopian-run NGO, is conducting capacity-building in conflict management in Rwanda. AHA has conducted workshops for government, civic groups, and agencies, again giving people a chance to talk openly about the genocide. The participants formed a contact group out of one of the sessions to follow up the discussions and pursue the topic of conflict resolution. Managing ethnic, cultural, and social differences was a major theme. In Burundi, the Catholic Bishops Conference initiated a series of workshops focusing on mutual respect, tolerance, self-critical reflection, and forgiveness. The workshops have been offered in a variety of locations to different groups. Ken Hackett of Catholic Relief Services notes, “Major religious institutions in different societies are well placed with the capacity through natural structures to minimize ethnic, economic, racial, or religious divisions by reaching all levels of society.”31 Promoting Peace: Radio, Television, and Theater The media provide a venue for accessing millions of people in postconflict societies. It is estimated that there are over 50 million radios spread throughout the developing world, including millions in areas currently or formerly controlled by rebel or government forces where propaganda has been the main source of information. Until recently, the aid community paid little attention to the potential role of media, but that is changing. For example, there is increasing recognition of the impact that radio programming can have on education, as well as on reconciliation processes. Debates are intensifying about the roles of international NGOs and domestic civil society in promoting reconciliation through media outlets. Increasing legitimacy is being lent to ideas such as training journalists in sensitivity to conflict issues, providing airtime to local peacemakers rather than just the combatants, and highlighting the economic benefits of peace through the media. This all parallels a debate within the journalism community about whether reporting is meant to reflect or change society.32 Various examples provide a flavor of the diversity of programming options. In early 1994, local churches in Burundi began broadcasting radio programs on topics related to alternatives to violence for dispute resolution, respecting differences, cycles of violence, and other related issues. In Cambodia after the peace agreement was signed, the UN worked with CSOs to initiate radio broadcasts designed to counter rumors and inaccurate information. By many accounts this lowered tensions and suspicions.33 Similarly, Radio Agatashya in Rwanda, established in August 1994, aims to combat the partisan—and often destructive—nature of local media. Peace radio has been initiated in Somalia and Rwanda. From Ethiopia,

S-12

11/24/09

340

4:22 PM

Page 340

EVALUATING IMPLEMENTATION TASKS

the InterAfrica Group—on behalf of UNICEF—broadcasts a peace program into Somalia on a daily basis. Radio is by far the most influential means of communication in Somalia, a highly oral society. Links on the peace radio programs are made specifically between peace, protection of women and children, and religious traditions. A symposium held by the UN Educational, Scientific, and Cultural Organization (UNESCO) in San‘a on Somali peace and reconciliation urged the presence of more media programs presenting positive images of such components of peacebuilding as traditional conciliation mechanisms and human rights rather than violence. Somali participants in the UNESCO symposium also urged an increase in dissemination of traditional values, which promote a more peaceful environment.34 Local theater is yet another medium. UNICEF and local CSOs initiated a project called the “Circus of Peace” in Mozambique, an effort to utilize theater to educate communities on the possibilities of reconciliation and on skills involved in conflict resolution. UNICEF also sponsored a puppet show for Hutu and Tutsi children in a Burundian displaced persons camp, which is aimed at promoting the peaceful resolution of disputes. Kenneth Bush, who has studied this and other similar processes in Northern Ireland and Sri Lanka, concludes, “While such initiatives at the street- or field-level may not have immediate, large-scale results, they do contribute to a gradual, bottom up process of reconciliation.”35 Besides radio and theater, television has also been employed as a mode of transmission for ideas of tolerance, such as Search for Common Ground’s programming in South Africa. Providing an independent source of information through a medium with as widespread dissemination as radio helps address the misinformation that often refuels a resumption of conflict in postwar societies. Radio programming geared toward peace is a low-cost intervention that can—if necessary—be undertaken from a great distance. Countering hate radio, for example, is critical for providing alternative points of view in highly charged societies. Although anecdotal evidence suggests the value of such programs, there is no empirical, quantifiable method to test the success of such interventions. Evaluation methods should be developed for initiatives of this kind, measuring attitudinal change and violence prevented. Peace Committees, Commissions, and Conferences Assembling peace committees or commissions is another strategy aimed at bringing together middle-range actors to bolster peace. The Central American peace agreement signed by all five countries in the region called for the creation of National Peace Commissions in each country. In Nicaragua, regional and local commissions were established as well, including one composed of midlevel religious leaders for the conflict between the Sandinista government and the Yamata umbrella group for the

S-12

11/24/09

4:22 PM

Page 341

BUILDING LOCAL CAPACITY

341

East Coast opposition. Similarly, regional and local peace committees were formed in South Africa after the National Peace Accord. The successes of the South African commissions are not well advertised, but in many cases they contained potential violence. For example, after the assassination of Chris Hani in April 1993, peace commission personnel worked successfully to restrain revenge-seeking individuals and groups. The number of casualties would have been far higher than 200 in the absence of the commissions.36 Peace committees allow participants to come together to help consolidate a peace process. The South African and Nicaraguan experiences have shown that communication between rival groups can be facilitated by such committees. Further benefits are derived from committees that help promote plurality and other central elements of a peace agreement. For conflict prevention and resolution processes to take hold, they require the participation of all segments of society. Traditional authorities (elders, chiefs, etc.), women’s organizations, local institutions, and professional associations have critical roles to play in the development of grassroots peacebuilding.37 External agencies can assist commissions to “approach issues as problems to solve rather than as political opportunities.” For example, peace commissions were established to help implement the Mozambican peace agreement. Donors supervised some of these commissions, such as the Commission on Reintegration of former combatants. Nicole Ball suggests that “external personnel who participate in the work of formal commissions should receive training in conflict resolution and consensus-building techniques.” USAID was actively involved in technical working groups that dealt with credit, land tenancy, and housing.38 In Ahmedabad, India, an aid agency assisted in the establishment of peace committees to address tensions and counter rumors that might provoke violence. The committees initiated a series of street plays to educate communities about the drawbacks to local violence. 39 Partners for Democratic Change has developed another model, which links conflict resolution centers in Bulgaria, Hungary, Poland, and Russia. The network has developed local conciliation commissions to promote cross-cultural communication and stress conflict prevention. The organization trains local bureaucrats in joint problem solving and negotiation and creates courses for teaching conflict resolution in the schools. The government and the main separatist movement in Sri Lanka established four peace committees in different regions of the country. The committees included representation from three donor countries. The committees provided a forum to bring people together to address ways to overcome significant obstacles to peace, and allowed for exchange across ethnic and geographic lines. Until the collapse of the cease-fire, the committees kept channels of communication open between the two sides.40 There are also organic, local variations on the peace committee con-

S-12

11/24/09

342

4:22 PM

Page 342

EVALUATING IMPLEMENTATION TASKS

cept that are involved in peacemaking. For example, the Committee for Restarting Pastoral Initiatives in Butare, Rwanda, led by a Tutsi priest and a Hutu intellectual, promotes the concept that reconciliation must start at the grassroots level. The committee has a newsletter for public education purposes, and is facilitating the return of refugees. Holding seminars and conferences is another way to help consolidate peace agreements. But unless attached to some broader postconflict reconciliation process, peace conferences are limited by the small numbers of people attending the conferences themselves and the usual lack of followup. Unless part of a broader strategic initiative, these conferences are of little relevance. A good example of linking peace conferences with the larger peace implementation mission occurred in Mozambique, where the Christian Council of Mozambique and UNICEF created a program called “Preparing People for Peace,” involving church representatives from a number of provinces at a national seminar on peace issues. These representatives then went back to their home areas and held further seminars at the local level, and dealt with themes such as religious perspectives on peace and church involvement in conflict resolution.41 Like problem-solving activities, peace conferences and peace committees are ultimately about rebuilding trust in order to reintegrate societies. By opening dialogues between formerly warring factions, facilitating information sharing, and allowing adversaries to express their intentions, CSOs can play an integral part in the trust-building process. Future challenges for peace committees include the need to improve coordination between local branches of such committees. Furthermore, peace committees are unable to address fundamental governance issues that fuel conflict. They can mediate disputes, but are ultimately nonviable if serious political and economic reforms are not simultaneously being undertaken. There is also the added danger of politicizing local disputes that are largely solvable outside the government’s reach, and the potential of becoming just another state bureaucracy. To counter this, local peace committees should develop strong relations with local organizations and deep familiarity with local conflict management processes.42 Indigenous Mechanisms for Conflict Management Indigenous mechanisms for conflict management can involve a variety of groups and processes at the local level. 43 These mechanisms build on indigenous concepts of social justice and approaches for managing conflict. Implementation is usually controlled by the communities themselves. In most cases, both parties agree to the verdict. Traditional social organizations are often the means by which local communities resolve conflicts. In Ethiopia, for example, Eddir, Equb, and Mehaber are the primary indigenous formal associations. Eddir is a welfare

S-12

11/24/09

4:22 PM

Page 343

BUILDING LOCAL CAPACITY

343

institution; Equb is a savings association; and Mehaber is a mechanism for communal responsibility.44 All of these provide avenues for inter- or intragroup reconciliation. For decades, “border chiefs” between Dinka and Nuer areas in Sudan have been the mechanism for ameliorating intercommunal conflict and bringing people back together. In Nuer areas, the concept of cieng naath (covenanting) is a key mechanism for solving conflict in the family as well as between larger social groups. Due to their protracted civil war, Somalis have been thrown back on clan and subclan structures to meet basic needs, including security, and can adapt a wealth of traditional mechanisms to resolve interclan killings and scarce-resource conflicts.45 Describing a breakdown into “warlordism” as a turning back to feudal or clan relations, David Keen notes that while the reascendance of clan politics in Somalia has encouraged certain patterns of conflict, it also promises the revival of certain patterns of conflict resolution.46 Aid can enable democratic, peaceful elements in society to compete for the loyalty of young men at risk of being lured into violence absent other opportunities—for example, channeling aid through Somali elders.47 In recent years, a distinct decline in traditional authority in many wartorn societies has undermined the ability of local mechanisms to manage conflict. Successful local peace conferences in Somalia and Sudan, however, show that such initiatives can renew “indigenous forms of peacemaking and conflict resolution to restore the balance in society” destroyed by modern internal war.48 Such initiatives rebuild indigenous peacemaking capacity from the bottom up, and from the periphery in. For example, in 1995, community elders in southern Sudan held the Akobo Peace Conference in order to address intercommunal conflict between two groups of the Nuer. In the successful aftermath of the conference, mobile peace committees were formed that included community and church leaders, who traveled to fishing holes and cattle camps to explain, promote, and monitor the peace agreement. In certain locations the committees reenacted the sealing of the agreement by sacrificing animals. Since the agreement, small peace meetings have been held throughout the Upper Nile region, during which traditional methods of conflict resolution are discussed. Local Sudanese women’s organizations have been instrumental in moving this process forward, and women’s leadership potential has been consciously developed, building the confidence and elevating the status of women’s organizations.49 A study of the Akobo Peace Conference produced the following lessons for indigenous peace processes: • Key authority structures of conflicting parties must be included in the process, including traditional, military, administrative, and religious leaders.

S-12

11/24/09

344

4:22 PM

Page 344

EVALUATING IMPLEMENTATION TASKS

• Women’s groups and leaders must be a part of the process from the beginning. • Traditional conflict resolution mechanisms should be utilized. • Those with moral authority in the community must be central to the process. Educated exiles have a role to play as well. • External support should be minimal and fill gaps, taking care not to replace indigenous leadership. • Indigenous processes represent long-term interactions between traditional and modern societies, and cannot act as quick fixes. They must be placed firmly in historical context for the participants and outside observers.50 In many of the examples described above, civil wars have continued to rage. However, negotiations that have utilized indigenous conflict resolution mechanisms have produced agreements that keep broader intercommunal relations positive, creating environments where nomads can graze together, townspeople can live together, and merchants can trade together, even though military personnel remain unreconciled. Collaborative Community Activities A critical issue for strategies encouraging intercommunal collaboration is whether the collaboration will last beyond the activity itself. One approach to improving the chances of sustainability of cooperation is to encourage economic interdependence by linking local production and consumption initiatives between formerly warring groups. For example, in response to a simmering resource conflict between Luo, Kuria, and Maasai ethnic groups in the border area of South Nyanza and Narok Districts in Kenya, Maasai and Luo elders donated land for a demonstration center for agriculture and livestock development. The pilot projects brought together farmers and pastoralists from the different groups, which were followed by a series of meetings initiating a reconciliation process. Ethnic clashes reportedly declined, as did cattle raiding. Project activities within the demonstration center were slowly decentralized throughout participating communities. “The place that used to be a battle ground has been turned into a booming market,” notes Moses Sika of Lutheran World Relief.51 Using economic collaboration as a means of fostering reconciliation, however, has its limitations. Save the Children–U.S. devised a program in Tajikistan to provide food for work for local groups reconstructing wardamaged homes in the hopes of maximizing intercommunal cooperation. Significant problems arose that highlight potential barriers to such programs. First, questions were raised about the potentially counterproductive nature of attempting to have communities so recently at war work together, especially when such work brought to the surface memories of recent atroc-

S-12

11/24/09

4:22 PM

Page 345

BUILDING LOCAL CAPACITY

345

ities. Second, the monoethnic nature of most villages in the project area made common work projects logistically challenging. Third, reconciliation goals were largely formulated by the expatriate country director, while local staff persons avoided direct, conscious involvement in support of these objectives. Fourth, the food payments tended to individualize program benefits, whereas a cash payment might have encouraged broader enterprise and consequently impacted a wider set of families. Fifth, the structure of the program inadvertently produced a situation in which the welfare of each community became a focus of competition with other communities. On this last point, Mary Anderson has suggested that a way to overcome this is to develop enterprises in which each group’s success depends on the other’s success, such as interlinked systems of production.52 There are other, less production- and trade-oriented initiatives that can encourage intercommunity collaboration. In Jerusalem, an agency helped establish two interlinked but autonomous women’s centers—one for Palestinians and one for Israelis. Although they primarily address issues in their own communities, the two centers also cooperate on common issues. Also in Jerusalem, the Israeli-Palestinian Center for Research and Information is coordinated jointly by Israelis and Palestinians. Its objective is to explore mutual interests of the two peoples by sponsoring three roundtables composed of economists and industrialists, water scientists, and a cross section of the two communities, who discuss the future of Jerusalem. Collaborative community activities, if designed and implemented with great sensitivity to local conditions and concerns, can make several contributions to peacebuilding. Not only do they bring former warring parties together and create projects for the community’s development needs, but these projects may also serve a symbolic function. Rebuilt homes and roads send a strong message that the war is over and that there is a tangible benefit to peace. Democracy and Human Rights Chapters 8 and 9 by Terrence Lyons and Tonya Putnam in this book address the subjects of democracy and human rights in great detail, but it is important to note that domestic CSOs, in partnership with international NGOs, are well positioned to undertake activities in these areas that support peace agreement implementation. When carried out by local organizations, civic education, election monitoring, human rights reporting, and advocacy for democratic reforms not only bolster the chances of short-term implementation success, but also begin to create capacity to sustain the pursuit of these goods when international implementers return home. As Tonya Putnam describes in Chapter 9, the reliance of international human rights organizations (IHROs) on an enforcement strategy of human rights standards fails to develop local organizations that will be critical in

S-12

11/24/09

346

4:22 PM

Page 346

EVALUATING IMPLEMENTATION TASKS

sustaining human rights in the long term. Involving CSOs in institutionbuilding ensures that a more flexible and consultative approach to human rights protection will be taken. CSOs can also advise IHROs on the complexity of a postconflict situation and sensitize them to local needs. Furthermore, CSOs can be utilized in educating people on their human rights, an idea that was incorporated into the “Preparing People for Peace” program in Mozambique.

Conclusion: Local Capacity as a Bridge Between Implementation and Peacebuilding External interventions on behalf of consolidating peace have dominated analyses of conflict resolution initiatives. International and regional institutions have received most of the attention, bypassing important nationaland local-level response capacities, particularly those of civil society. Nevertheless, internal initiatives continue, including those of traditional authorities, local NGOs and associations, religious bodies, commercial leaders, and the state itself. External interventions that build local peace capacities and/or processes likely stand the best chance of creating a selfenforcing peace. There is no disagreement in the aid community about the importance of supporting local capacity and structures, although in practice—particularly in emergencies—the support is often more rhetorical than practical. The Rebuilding War-Torn Societies Project found that “external assistance . . . tends to become a substitute, and worse, destroys local institutions and solutions.” “A policy of betting on the local,” the project concludes, “may in the short term be more laborious, less spectacular, and take more time, but in the long-term may be the only realistic option.”53 As Susan L. Woodward argues in Chapter 7, reconciliation requires the rebuilding of social capital, involving organizations, networks, and social codes. The World Bank in a report on postconflict transitions concludes: “Rebuilding social capital means a revitalization of civil society, and revitalizing civil society entails the promotion of local associations, community participation, and peer accountability. It reduces the level of individual fear, enables the collective censure of violence, and promotes local security.”54 Despite the microexamples of positive civil society contributions, CSOs face several obstacles to greater involvement in peace agreement implementation. First, CSOs represent a threat to the bases of power and fundraising of many authorities, which severely complicates any strategy of local capacity-building for peace. Second, there is a consistent absence of sustainable funding for CSO activities in peacebuilding, as funds are much more readily available for humanitarian crisis response. Third, the potential contributions of CSOs are often ignored or unrecognized by academics and

S-12

11/24/09

4:22 PM

Page 347

BUILDING LOCAL CAPACITY

347

policymakers, who look principally at larger strategic issues and do not view those actors working on “softer” issues such as people-to-people reconciliation as relevant to the larger picture of peace implementation. There are many opportunities to support indigenous nongovernmental forms of social organization. In every society, there are traditional mechanisms of kinship and self-help, which are often the primary contributors to a community’s survival in the context of a conflict and its aftermath. Even in the structure of a food distribution, promoting alternative representation (women’s groups, technical committees, traditional social networks) can build local capacity and create an alternative to the military structures. International implementers of peace agreements have a major role to play in empowering dispossessed elements. Aid should continuously be reevaluated to see how external involvement can support empowerment. Time horizons must be adjusted as well. One- or two-year program goals and expectations are misguided and sometimes dangerous. CSO reconciliation initiatives should be viewed as long-term processes that must be put into place—building on locally determined priorities and resources—and supported over time, with emphasis on promoting the self-sustainability of programs, initiatives, and institutions.

Notes 1. David Smock and John Prendergast, “NGOs and the Peace Process in Angola,” U.S. Institute of Peace Special Report, April 1996, p. 2. 2. Marina Ottaway, “African Democratization and the Leninist Option,” Journal of Modern African Studies 35, no. 1 (1997): 1–15. 3. John Prendergast, Front Line Diplomacy: Humanitarian Aid and Conflict in Africa (Boulder, Colo.: Lynne Rienner, 1996). 4. Mark Duffield, “Complex Political Emergencies: An Exploratory Report for UNICEF,” New York, March 1994, p. 117. 5. Kimberly Mahling-Clark, “Mozambique’s Transition from War to Peace: USAID’s Lessons Learned,” Africa Bureau Information Center, USAID, April 1996, p. ii. 6. Mary Anderson, “The Experience of NGOs in Conflict Intervention: Problems and Prospects,” Collaborative for Development Action mimeograph, April 1995, p. 20. 7. USAID, “Project Evaluation: El Salvador: Peace and National Recovery,” March 16, 1994, p. ix-1. 8. Mary Anderson, Do No Harm: How Aid Can Support Peace or War? (Boulder, Colo.: Lynne Rienner, 1999). 9. M. Adams and M. Bradbury, “Conflict and Development,” Background Paper no. 2 for UNICEF/NGO workshop, New York, April 27, 1995, p. 10. 10. Ken Menkhaus, “Conflict, Peace-Building, and International Aid,” Life and Peace Review, February 1995, p. 12. 11. Adams and Bradbury, “Conflict and Development,” p. 37. 12. Ibid., p. 10. 13. Ibid., p. 35.

S-12

11/24/09

348

4:22 PM

Page 348

EVALUATING IMPLEMENTATION TASKS

14. Comments of Hugo Slim, International NGOs and Complex Political Emergencies: Perspectives from Anthropology (London: Royal Anthropological Institute, 1995), p. 21. 15. Author interview with Iain Levine, Nairobi, Kenya, August 20, 1995. 16. UNICEF, State of the World’s Children 1996 (New York: UNICEF, 1996). 17. Neil Boothby, “Displaced Children: Psychological Theory and Practice from the Field,” Journal of Refugee Studies 5, no. 2 (1992): 112. 18. I. Agger, S. Vuk, and J. Mimica, Theory and Practice of Psycho-Social Projects Under War Conditions in Bosnia-Herzegovina and Croatia (Brussels: European Commission Humanitarian Office, 1995). 19. Boothby, “Displaced Children,” p. 116. 20. Derek Summerfield, “The Impact of War and Atrocity on Civilian Populations: Basic Principles for NGO Interventions and a Critique of Psychological Trauma Projects,” Network Paper no. 14, Relief and Rehabilitation Network (London: Overseas Development Institute, April 1996), pp. 15–16. 21. Krishna Kumar, Rebuilding Societies After Civil War: Critical Roles for International Assistance (Boulder, Colo.: Lynne Rienner, 1997), p. 65. 22. F. Hume and D. Summerfield, “After the War in Nicaragua: A Psychosocial Study of War Wounded Ex-Combatants,” Medicine and War 10 (1991): 4–25. 23. Ibid., pp. 16–17. 24. John Paul Lederach, Building Peace: Sustainable Reconciliation in Divided Societies (Washington, D.C.: U.S. Institute of Peace, 1997), p. 47. 25. For a recent attempt to evaluate the effectiveness of PSWs and other types of interactive conflict resolution, see Nadim N. Rouhana, “Interactive Conflict Resolution: Issues in Theory, Methodology, and Evaluation,” in Paul C. Stern and Daniel Druckman, eds., International Conflict Resolution After the Cold War (Washington, D.C.: National Academy Press, 2001), pp. 294–337. 26. Michael Lund and John Prendergast, Preventing and Mitigating Violent Conflict: A Guide for Practitioners (Washington, D.C.: Creative Associates International, 1997), pp. B22–B24. 27. Peter Sollis and Christina Schultz, “Lessons of the PRODERE Experience in Central America,” in RPG Focus (Washington, D.C.: Refugee Policy Group, November 1995), pp. 4–6. 28. Nat Colletta, Markus Kostner, and Ingo Wiederhofer, The Transition from War to Peace in Sub-Saharan Africa (Washington, D.C.: World Bank, 1996), p. 3. 29. Lund and Prendergast, Preventing and Mitigating Violent Conflict, p. B24. 30. Lederach, Building Peace, pp. 48–51. 31. Ken Hackett, “The Role of International NGOs and Civil Society in Preventing Conflict,” speech to UN symposium on preventing conflict, New York, April 1996. 32. Clare Pointon, “Small Wars and Smart Relief,” Crosslines Global Report, July 1996, pp. 33–34. 33. Michael W. Doyle, UN Peacekeeping in Cambodia: UNTAC’s Civil Mandate (Boulder, Colo.: Lynne Rienner, 1995), pp. 54–55. 34. UNESCO, Symposium on the Culture of Peace in Somalia (San‘a, Yemen: UNESCO, 1995), paras. 14, 16. 35. Kenneth Bush, “From Bullet Holes to Bird Nests: The Role of NGOs in the Peacebuilding Process,” annual meeting of the Canadian Political Science Association, Calgary, June 1994. 36. Lund and Prendergast, Preventing and Mitigating Violent Conflict, p. B24. 37. Adams and Bradbury, “Conflict and Development,” p. 35.

S-12

11/24/09

4:22 PM

Page 349

BUILDING LOCAL CAPACITY

349

38. Nicole Ball and Tammy Halevy, Making Peace Work: The Role of the International Development Community, Policy Essay no. 18 (Washington, D.C.: Overseas Development Council, 1996), pp. 62–63. 39. Anderson, Do No Harm, pp. 126–127. 40. Canadian International Development Agency (CIDA), “CIDA’s Experience in Arbitration, Mediation, Negotiation, and Reconciliation,” notes for the Paris Meeting of OECD DAC Task Force Working Group on Conflict Prevention, February 1996, p. 5. 41. Lederach, Building Peace, pp. 53–54. 42. Lund and Prendergast, Preventing and Mitigating Violent Conflict, p. B35. 43. For a critical yet sympathetic review of African traditional conflict resolution mechanisms, see I. William Zartman, ed., Traditional Cures for Modern Conflict: African Conflict Medicine (Boulder, Colo.: Lynne Rienner, 2000). 44. UNICEF, Children and Women in Ethiopia (New York: UNICEF, 1993), p. 24. 45. Lederach, Building Peace, pp. 52–53. 46. David Keen, The Benefits of Famine: A Political Economy of Famine and Relief in Southwestern Sudan (Princeton: Princeton University Press, 1994), p. 27. 47. David Keen and Ken Wilson, “Engaging with Violence: A Reassessment of Relief in Wartime,” in Joanna Macrae and Anthony Zwi, eds., War and Hunger: Rethinking International Responses to Complex Emergencies (London: Zed Books, 1994), p. 216. 48. Mark Bradbury, The Somali Conflict: Prospects for Peace, Oxfam Research Paper no. 9 (London: Oxfam, 1994), p. 6. 49. Wal Duany, William O. Lowrey, and Julia Duany, “The Indigenous Peace Process: A Case Study of the Jikany-Lou Nuer Conflict and Its Resolution,” workshop on Political Theory and Policy Analysis, Indiana University, Bloomington, June 1997. 50. Ibid. 51. Moses Sika, “Case Study—Ogwedhi Sigawa Project,” mimeograph, July 1994. 52. Anderson, Do No Harm, pp. 85–90. 53. UN Research Institute for Social Development (UNRISD), Rebuilding War-Torn Societies (Geneva: UNRISD, March 1995), mimeo, p. 23. 54. Colletta, Kostner, and Weiderhofer, The Transition from War to Peace in Sub-Saharan Africa, p. 73.

S-12

11/24/09

4:22 PM

Page 350

S-13

11/24/09

4:23 PM

Page 351

PART 3 Case Studies

S-13

11/24/09

4:23 PM

Page 352

S-13

11/24/09

4:23 PM

Page 353

13 Peace in Stages: The Role of an Implementation Regime in Nicaragua CAROLINE A. HARTZELL

The peace Nicaragua enjoys today is not the outcome of a single agreement or central accord signed by the parties to the conflict, but rather the product of a peace process that evolved through three stages. The first of these stages, the Esquipulas peace process, saw five Central American presidents reach a series of agreements on the steps necessary to bring peace to Nicaragua and other involved Central American countries. The second stage of the peace process centered on Nicaragua’s 1990 elections and the transition accord that formed the basis of the political settlement between the Sandinistas and the incoming government of Violeta Chamorro. The final stage consisted of a series of disarmament and demobilization accords between the Nicaraguan Resistance, or Contras, and the government of Nicaragua. Early in the Nicaraguan peace process, implementation became a central issue for two reasons: the conflict environment and the exclusion of key actors from parts of the negotiation process. The civil war in Nicaragua was but one part of a regional conflict in which fighting in some countries spilled over into others. As a result, the willingness and ability of states to stop aiding insurrectionist movements in other countries or prevent their territory from being used for attacks were fundamental to the peace process. As it became increasingly clear that the presidents of Costa Rica, El Salvador, Guatemala, Honduras, and Nicaragua would be unable to implement the entire peace process, these architects of the Central American peace plan turned to third-party actors—the United Nations and the Organization of American States (OAS). These actors, they hoped, would have the necessary resources to help implement the accords and be considered legitimate by the parties to the conflict. This last point would prove particularly important given the exclusion from many of the negotia353

S-13

11/24/09

4:23 PM

354

Page 354

CASE STUDIES

tions of two parties to the conflict—the Nicaraguan Resistance and the U.S. government. Despite a difficult implementation environment characterized by Cold War politics, challenging physical terrain, an often hostile U.S. government, and regional tensions, peace was successfully implemented in Nicaragua. Not only has the country experienced years of peace, but it has also sustained this peace without outside assistance. Central to the successful implementation of the Nicaraguan peace accords was the creation by the Central American presidents of an implementation regime focused on monitoring and verification of progress toward peace. That the Central American parties acknowledged their limitations, and were willing to divide implementation responsibilities accordingly, strengthened this regime and helped surmount difficulties faced by the peace process, as did the expectation that implementation would be a long-term process. In addition, although the regime provided implementers with neither the mandate nor the means for forcing recalcitrant actors to comply with the terms of the agreement, the high profile of the regime and its successful track record in implementing the first two phases of the peace process appear to have persuaded the Contras that it was better to disarm and demobilize while the regime was still functioning than to be completely marginalized from the peace process. This chapter briefly describes Nicaragua’s civil war before moving on to the roles implementers played during the three stages of the Nicaraguan peace process. The first stage focuses on the creation of an implementation regime by regional and extraregional actors, the second stage on the actions of Nicaraguan actors within the framework of that regime, and the third stage on the challenges the Contras posed to the implementation regime and the manner in which domestic and international implementers responded to them. The chapter then concludes by evaluating the peace in Nicaragua.

The Conflict Scarcely two years prior to the outbreak of civil war in Nicaragua in 1981, the country experienced a revolution in 1978–1979 that toppled the dictatorship of Anastasio Somoza Debayle and brought to power the Sandinista National Liberation Front (FSLN—also known as the Sandinistas), a Marxist revolutionary organization. After decades of corrupt rule by the Somoza family and political repression at the hands of the Somozacontrolled National Guard, most Nicaraguans supported the new FSLN government and its stated goal of addressing economic inequality and curbing the economic and social power of Nicaragua’s private-sector elite. 1 Domestic dissatisfaction with the Sandinista government soon began to

S-13

11/24/09

4:23 PM

Page 355

THE ROLE OF AN IMPLEMENTATION REGIME IN NICARAGUA

355

grow, however. Members of the economic elite fled the country, nonSandinista members resigned from the ruling junta in 1980, and opposition and demands for autonomy grew among indigenous peoples of the Atlantic Coast who resisted Sandinista efforts to incorporate them into the new national development process. 2 These factors, when joined with U.S. antipathy to the Sandinista government, sowed the seeds for the outbreak of civil war in Nicaragua. Seeking to prevent Nicaragua from becoming “another Cuba” that might destabilize countries in the region, the Reagan administration suspended all aid to Nicaragua on January 23, 1981.3 Later that year the administration authorized the disbursement of $19.95 million to support former members of the National Guard, who set up camps in southern Honduras as their training area.4 The Nicaraguan Resistance grew in numbers as peasants from the northern part of the country and indigenous groups from the Atlantic Coast joined the counterrevolution. By the mid1980s, the various military factions of the Nicaraguan Resistance comprised about 15,000 troops.5 The bulk of these, about 80 percent, belonged to the Honduras-based Nicaraguan Democratic Forces (FDN), led by Adolfo Calero and Enrique Bermudez. This faction of the Nicaraguan Resistance focused its efforts on social and economic disruption. Another Contra group, the politically moderate Southern Opposition Bloc (BOS), was based in Costa Rica, reflecting the porousness of borders in Central America, and by the late 1980s had a few hundred armed members. The third major Contra contingent, comprising a number of indigenous groups, many of whom sought some form of regional autonomy, consisted of three factions—MISURA (Miskitu, Sumu, Rama), MISURASATA (Miskitu, Sumu, Rama, and Sandinistas Together), and Kus Indian Sut Aslika Nicaragua (KISAN; Union of Coastal Indians of Nicaragua).6 With financial and logistical support of about $400 million, much of it funneled through the U.S. Central Intelligence Agency, and initial military training supplied by Argentine military officers, Contra forces became increasingly active in Nicaragua through the mid-1980s.7 Armed with weapons ranging from small arms to mortars, grenade launchers, shoulderfired missiles, and mines, the Contras pinpointed their attacks on economic targets and symbols of the revolution (health centers, agrarian cooperatives) while avoiding direct contact with Nicaragua’s army, the Sandinista People’s Army (EPS).8 These actions, in addition to the presence of former National Guard members among the Contras and, in particular, the FDN’s often brutal treatment of civilians, made the Nicaraguan Resistance unpopular among many Nicaraguans.9 Although it never coalesced into an effective fighting force, the Nicaraguan Resistance did come to pose a threat to the FSLN government through its damage to the country’s economy. In response to increased U.S. military funding of the Contras as well as growing fears of a possible U.S. invasion of the country, the Sandinista

S-13

11/24/09

356

4:23 PM

Page 356

CASE STUDIES

government created an active-duty militia and instituted a military draft. As a result of the draft, both the militia and the EPS grew in size, with the latter numbering 86,810 in 1990. 10 Armed with Soviet-built MI-17 (Hip) transport helicopters and armored MI-25 (Hind) helicopters that were part of the more than $3 billion in military and economic assistance provided by the Soviet Union between 1979 and 1989, the EPS functioned as a large mobile infantry used to pursue Contra forces.11 By 1987 the Nicaraguan civil war had reached a military stalemate. Lacking support in urban areas, dependent on airdrops for supplies, and constantly forced to drop back across the borders to Honduras and Costa Rica in response to Sandinista offensives, the Contras clearly were unable to win the war. In addition, many of the indigenous Contra troops lost interest in armed struggle as the Sandinistas accommodated their demands for autonomy.12 The Contras were also affected by the aftermath of the IranContra scandal and the U.S. congressional decision to cut off all military support with the exception of “nonlethal” aid.13 The Sandinista government, for its part, was finding it increasingly difficult to fight a war that was destroying the country’s economy.14 Nevertheless, 1987 and 1988 saw heavy fighting as both sides in the war responded to the signing of the Esquipulas II Accord in August 1987 by intensifying combat in an effort to improve their relative positions before further negotiations took place. Although the Contra demobilization and disarmament process was not completed until the middle of 1990, the bulk of the fighting between the EPS and the Contras ended in 1989. By the war’s end, an estimated 30,000 people, or slightly less than 1 percent of the population, had died, with approximately 5 percent of the dead and wounded consisting of civilians.15 In addition, during the course of the war in Nicaragua 354,000 people were internally displaced and an estimated 500,000 refugees, approximately 12 percent of the population, fled the country. The bulk of the refugees were located in Costa Rica, the United States, and Honduras, with the latter country, in particular, home to the dependents of many Contras. Although the Sandinista government actively encouraged repatriation of Nicaraguans during the war, refugee concerns about returning to a country in the midst of a civil war as well as Contra anxieties about safety meant that significant repatriation did not take place until after the national elections in 1990, when various accords had been signed by the Chamorro government and the Contras that provided security zones and safety assurances for former members of the Nicaraguan resistance.16

The Esquipulas Peace Process: Creating an Implementation Regime Several efforts were made in the early 1980s to negotiate an end to conflict in Central America, with the most significant of these being the Contadora

S-13

11/24/09

4:23 PM

Page 357

THE ROLE OF AN IMPLEMENTATION REGIME IN NICARAGUA

357

peace process.17 Initiated by Colombia, Mexico, Panama, and Venezuela in January 1983, and later supported by the “Lima Group” countries— Argentina, Brazil, Peru, and Uruguay—Contadora emphasized regional support for and involvement in the peace process. It also focused on economic and social inequality in each country, rather than East-West tensions, as the primary cause of conflict in the region. Although many of the central points that later were included in the Esquipulas II Accord first appeared in Contadora documents, the Contadora peace process came to an end three and a half years after it had begun, the victim of Cold War pressures. Complaining that initial drafts of the peace plans neglected implementation, verification, and enforcement, the Reagan administration and its Central American allies, Costa Rica, El Salvador, and Honduras, argued that the agreements would be vulnerable to violation by the Sandinista government. Reacting to these complaints, the Contadora group asked Canada to help create verification and confidence-building measures. Subsequent drafts of Contadora plans provided much more detail on these issues. By the final draft Contadora Act of June 1986, so many efforts had been made to ease concerns about implementation that the document suffered from excessive specificity. Such flaws, continuing concerns among the United States and its allies that the peace plans favored the Sandinistas, and the opposition of the Reagan administration to any agreement that would allow the Sandinistas to remain in power, eventually scuttled Contadora. Its demise led to a heightened desire among Central American nations to take the lead in drafting a regional peace plan, and set the stage for a new peace process. A transition mechanism from Contadora to the Esquipulas peace process arose unexpectedly in late 1986 when the Secretaries-General of the UN and the OAS offered their good offices to the Contadora Group. The Rio Group (the original four Contadora countries plus the Lima Group four) proposed that the foreign ministers of the eight Latin American countries visit the capitals of the Central American countries embroiled in conflict, accompanied by the two Secretaries-General. The visit, which took place on January 19–20, 1987, laid the groundwork for future cooperation among the international organizations and the Central American countries. It also aroused the ire of the Reagan administration, now mired in the IranContra scandal, by relegating the United States to a secondary role in the Central American peace process. In an atmosphere of growing doubts about U.S. policy toward the region, Central American leaders held a meeting in Esquipulas, Guatemala, in August 1987 to discuss a peace proposal drafted by Costa Rican president Oscar Arias Sanchez earlier that year. Motivated at least in part by an unexpected bilateral peace initiative issued by President Ronald Reagan and House Speaker Jim Wright (D–Tex.) on August 4, just two days before the Esquipulas meeting was to begin, the Central American presidents (who

S-13

11/24/09

358

4:23 PM

Page 358

CASE STUDIES

apparently interpreted the U.S. plan as an attempt to kill or co-opt the Central American peace process) signed the “Procedure for the Establishment of a Firm and Lasting Peace in Central America” on August 7, 1987. Esquipulas II, as the peace accord became known, consisted of a series of confidence-building measures agreed to by the signatories. These measures, which were general in nature, included establishing cease-fires in countries experiencing armed hostilities; called for each country to enter into a process of national reconciliation, engage in democratization, hold free elections, terminate aid to irregular forces and insurrection movements, and refuse the use of its territory for attacks on other states; and stipulated that the countries enter into joint negotiations on security, verification, and arms control, commit themselves to assist refugees and displaced persons, and establish an environment of cooperation, democracy, and freedom for peace and development. The two most specific measures in the agreement were an implementation timetable that specified steps to be taken 15, 90, 120, and 150 days after the signing of the accord and the creation of an International Commission for Verification and Follow-Up (CIVS), whose duties included monitoring and verifying compliance with the commitments made in the document. 18 Conspicuously absent from Esquipulas II were guidelines stipulating how the signatories would comply with the various components of the accord. If they were truly interested in achieving a durable peace in the region, why did the Central American presidents commit themselves to such lofty goals without specifying how they meant to go about complying with them? The architects of Esquipulas II sought to formulate a plan that would make it possible for the previously divided countries (Costa Rica, El Salvador, Guatemala, Honduras, and Nicaragua) to build confidence among themselves. This could only be done if, at the outset, blame for conflict in the region was not apportioned among the countries, all five countries agreed to abide by their commitments, and none of the signatories faced the immediate prospect of an intrusive implementation and verification regime. The Central American leaders also observed that there were both regional and national components to the crisis in Central America and that different means were needed to address them. Distinct types or sets of actors would likely be required to implement and verify peace at domestic and regional levels. Acknowledging such complex and multilayered implementation requirements, however, would have resulted in a document as detailed as the final Contadora drafts, which the signatories sought to avoid. In light of these circumstances, the architects of the Esquipulas process avoided specific mechanisms too politically challenging to enact. Rather, they created an implementation regime centered on verification and monitoring of the peace. This regime initially consisted of three sets of implementers. Two of these implementers were the National Reconciliation Commissions and the CIVS, the first responsible for implementing and ver-

S-13

11/24/09

4:23 PM

Page 359

THE ROLE OF AN IMPLEMENTATION REGIME IN NICARAGUA

359

ifying the national elements of the peace process and the latter for doing the same for the regional and interstate layers. The third set of implementers was the Central American presidents, who used summitry to spur implementation. The ill-conceived CIVS eventually was dismantled and, once the Central American peace process was reconfigured to focus on Nicaragua, the National Reconciliation Commissions, too, faded in significance. The central actors during the life of the implementation regime thus proved to be its designers, the Central American presidents, and, once they realized their inability to implement the peace accords themselves, the presidents called upon third-party implementers to help them, the UN and the OAS. Esquipulas II’s emphasis on national solutions to conflict in the region is clear in the document, the bulk of which focuses on actions to be taken within each of the Central American countries. As initially designed, the principal implementers of many of the provisions of Esquipulas II were to be the governments of the Central American countries. Because the accord required only that the governments “initiate a dialogue with all unarmed political opposition groups and with those who have availed themselves of the amnesty,” rebel groups in the countries at war did not participate in negotiating the accords. Although this exclusion of rebel groups from the peace process would pose problems for establishing peace, it highlights the Central American presidents’ strategy of improving the prospects for peace by fostering better relations with one another and respecting one another’s sovereignty. Because each country was sensitive to interference by others in its affairs, Esquipulas II did not force the issue of negotiations with armed opponents or spell out how each country was to implement the various steps it was supposed to take in order to create peace. Rather, the agreement left it up to each government to establish a National Reconciliation Commission to verify its compliance with commitments to cease-fire, dialogue, amnesty, democratization, and free elections. Recognizing regional and interstate dimensions of the Central American conflict that required the cooperation of other actors to solve, Esquipulas II also provided for the CIVS to be composed of the Secretaries-General of the UN and the OAS, the foreign ministers of the signatory countries, and the foreign ministers of the Contadora and Lima Group states. Rather than directly implement the peace accord, the CIVS was to provide international monitoring and verification of the five governments’ compliance with the domestic and regional components of the agreement. The CIVS was to report its findings to the Central American presidential summits that were to be held as part of the peace process. In keeping with the Central American presidents’ sensitivity to their countries’ sovereignty, each of the five governments had the power to veto objectionable elements of the CIVS reports, thus rendering the commission relatively impotent.19

S-13

11/24/09

4:23 PM

Page 360

360

CASE STUDIES

The Central American presidents and their use of presidential summits were key to the Esquipulas process. Although each president sought to advance his government’s position and interests with respect to the conflict in the region, the presidential summits developed into an institution in which efforts were made to solve regional problems by searching for common interests.20 Aware that the Esquipulas peace process had become the focus of international attention, the Central American presidents attempted to resolve differences behind the scenes whenever possible, going so far as to cancel planned summits when it appeared that these would not advance the peace process. In the face of some of their weaknesses as implementers, the presidents’ strategy of fostering regional consensus on issues related to the peace process proved important to keeping the peace process alive. This was especially true during the early stages of the implementation regime, when the lack of specificity regarding goals and means for achieving them made it difficult to ascertain what implementation really involved. Having visited Central America following his country’s offer to help design and operate confidence-building systems in the region, Canadian external affairs minister Joe Clark observed: “That kind of vagueness was useful in getting agreement on the peace plan in the first place, but it is not good enough now. The regional presidents should be encouraged as loudly as possible to decide what they mean by peace, and just exactly how they propose to go about maintaining it.”21

Esquipulas in Action After signing Esquipulas II, the Sandinistas moved quickly to comply with the accord. They formed a National Reconciliation Commission and asked outspoken government critic Cardinal Miguel Obando y Bravo to head it, repealed the “Ley del Ausente” (a law providing for the confiscation of property of absentee property owners), and unilaterally announced a partial cease-fire with the Contras on September 22, 1987. They also authorized the reopening of the opposition newspaper La Prensa, allowed Radio Católica to resume broadcasting, and announced the end of press censorship.22 International reaction to Esquipulas II, and to the Nicaraguan government’s actions, varied. European powers, as had been true since the late 1970s, expressed their support for the multilateral peace initiative and continued to criticize U.S. policy toward Nicaragua. Although there were conflicting reports on the issue, Soviet leader Mikhail Gorbachev apparently made an offer, during a meeting with President Reagan on December 10, 1987, to suspend military aid to the Sandinista government as long as the Central American peace process continued to move forward and the United

S-13

11/24/09

4:23 PM

Page 361

THE ROLE OF AN IMPLEMENTATION REGIME IN NICARAGUA

361

States did not resume arms shipments to the Contras.23 The Reagan administration, continuing to argue that the Sandinistas could not be trusted, announced its intention soon after Esquipulas II was signed to seek an additional $270 million in funds for the Contras, an action directly in violation of the accord’s request that extraregional governments cease “military, logistical, financial, [and] propagandistic aid in manpower, armaments, munitions and equipment to irregular forces or insurrectionist movements.” In addition, the U.S. government attempted to pressure the presidents of the other Central American nations to abandon the peace process if the Sandinistas did not fully implement the democratization provisions of the peace agreement. U.S. pressure took the form of a high-level delegation from Washington, including the head of the National Security Council, General Colin Powell, which visited Costa Rica, El Salvador, and Honduras prior to the January 15–16, 1988, Central American presidential summit in Costa Rica planned as a follow-up to Esquipulas II. During these meetings, as well as during bilateral visits the four Central American presidents held with the Reagan administration prior to the January summit, the United States threatened to cut off its military and economic aid to those countries that remained committed to the peace process.24 That some of these countries took this pressure seriously and wavered in their commitment to the peace process (Honduras, for example, continued to allow Contra bases to operate on Honduran territory) highlights one of the difficulties of having states that were themselves embroiled in regional Cold War politics act as implementers of the accord. Within months of signing Esquipulas II, a series of events produced a standstill in the implementation of the accord. A joint UN/OAS Technical Assistance Commission visited the five Central American countries from October 21 to October 27, 1987, in order to evaluate how the two organizations could monitor compliance with the security commitments of the agreement. This initial visit was followed by a second round of consultations in November with representatives of the Central American governments. As a result of these visits, the commission concluded that conditions were not yet ripe for on-site verification. It made clear that before the commission could begin verification, the roles of the “peace observers” must be defined precisely, that it would be necessary to secure a cease-fire that would be respected by all parties, including rebels who had been excluded from the negotiations, and that agreement among the parties must be reached regarding disarmament.25 Implementation of Esquipulas II suffered another blow when the CIVS issued its first report on January 14, 1988, prior to the presidential summit. Although the report focused on all five Central American countries, it criticized El Salvador, Guatemala, and Honduras for their lack of progress in complying more fully with Esquipulas than had the other countries. The four Central American countries (Costa Rica, El Salvador, Guatemala, and

S-13

11/24/09

362

4:23 PM

Page 362

CASE STUDIES

Honduras) and the United States criticized the report as biased in favor of Nicaragua.26 It was thus decided at the summit to dissolve the CIVS and assign its verification duties to an Executive Commission composed of the five countries’ foreign ministers. The commission was to be assisted by a Technical Advisory Group, supported by the governments of Canada, Germany, Spain, and Venezuela, and to act under UN auspices.27 Assigning verification of Esquipulas II’s security commitments to the Central American foreign ministers weakened the monitoring of implementation. This was particularly true of the transborder security commitments that required monitoring the movement of armaments and troops. Involved as they were in the region’s problems, subject to external pressure, and lacking the resources—particularly vehicles, helicopters, and personnel skilled in the use of surveillance equipment—necessary to verify incursions across borders, the Central American countries could not be expected to perform this function reliably for themselves. Indeed, until they agreed to rely on outside actors to perform these tasks, the peace process stalled, remaining paralyzed for most of 1988. Other decisions reached at the presidential summit in Costa Rica reinforced this paralysis in the Esquipulas process. These included decisions not to set a timetable for further compliance with or evaluation of Esquipulas II and to allow each country unilaterally to follow through on its commitments, particularly those regarding cease-fires, dialogue, and amnesty and democratization, without insisting on reciprocal regional compliance. In addition, a new emphasis on democratization emerged from the Costa Rica summit. This refocused the peace process on Nicaragua, an outcome the Sandinistas were not happy with but tolerated in the hope of securing an end to aid to the Contras.28 Despite the apparent stalling of Esquipulas, Nicaragua took an important step toward peace with the signing of a cease-fire accord between the Sandinista government and the Contras in Sapoa, Nicaragua, on March 23, 1988. The Sapoa Agreement guaranteed the Contras’ political rights, including their representation in a prospective national dialogue; identified five neutral zones in Nicaragua where the rebel troops would regroup; and provided for amnesty, repatriation, and access to humanitarian aid.29 The agreement took the Reagan administration by surprise, with one official noting that the “Contras jumped the gun, got out of control, by signing.”30 That the Contras entered into such an agreement apparently reflected a division within their leadership between Contra leader Alfredo Cesar and hard-line military commander Enrique Bermudez. The Sapoa Agreement fell apart in less than three months, the result of new Contra demands including the forced resignation of the Supreme Court and allowing the Contra army to remain armed in its zones through January 31, 1989, concessions the Sandinista government refused.31 A few months later, the

S-13

11/25/09

11:34 AM

Page 363

THE ROLE OF AN IMPLEMENTATION REGIME IN NICARAGUA

363

Sandinistas clamped down on the legal opposition, and cease-fire violations increased after only months of relative peace. The Esquipulas peace process moved forward in 1989, prompted by a change of administration in the United States and growing war-weariness in Central America. Incoming president George Bush was perceived by the Central American presidents to have a less ideological approach to the region than his predecessor. Apparently concurring with a U.S. official’s observation that, unlike Ronald Reagan, George Bush “does not have the teflon” and would thus not interfere in the peace process to the extent the Reagan administration had, representatives of the Central American countries held meetings at the end of 1988 to plan for a presidential summit in early 1989. Among the topics to be discussed at the meeting was the creation of a multinational force to “verify and control” borders in the region.32 Having seen the regional peace process stall for nearly a year, the Central American leaders were now willing to work more closely with implementers who could provide the technical help necessary to verify the accords. This became evident on February 8, 1989, at a meeting between the foreign ministers of the five Central American countries and the UN Secretary-General when the ministers asked the Secretary-General to appoint a technical group to help them devise a means to verify the transborder security elements of Esquipulas II.33 Also helping to jump-start the peace process was the fact that the Sandinistas were now willing to enter into political compromises with the country’s unarmed internal opposition. Alarmed by the economic costs of the war and sensing the growing weakness of the Contras, who had been without U.S. military aid for one year, the Sandinistas believed the moment was ripe to end the war on terms that would provide the FSLN with the opportunity to remain in power. The confluence of these factors made it possible to reach a number of important agreements in 1989. The first of these, the Costa del Sol Declaration or Tesoro Beach Agreement, emerged from the presidential summit held in Costa del Sol, El Salvador, on February 12–14. The accord called for the presidents to draw up a plan within ninety days for the voluntary demobilization, repatriation, or relocation of the thousands of Contras and their families who were in third countries. Largely in exchange for the closing of Contra bases in Honduras, Nicaraguan president Daniel Ortega agreed to advance the constitutionally mandated date for elections from November to February 25, 1990, and to allow international observers to monitor all stages of the elections. These were important concessions in a country sensitive to breaches of its sovereignty. The agreement also called for small, mobile, unarmed UN observer teams composed of Spanish, German, and Canadian troops to prevent cross-border raids by guerrilla groups in the region. 34 With the Tesoro Beach Accord, the Central American presidents provided the UN with the working document it needed

S-13

11/24/09

364

4:23 PM

Page 364

CASE STUDIES

to begin planning for operations.35 The UN’s plan for a 160-person Central American peace-observing unit to monitor the cessation of aid to rebels in the region was tentatively approved by the Central American countries on March 15, 1989. This was formalized at the Esquipulas Executive Commission’s May 1989 meeting when the commission requested that the UN and the OAS take total responsibility for dismantling the Contra army. Responding to pressures from Latin American leaders, President Ortega agreed to hold talks with the Sandinistas’ political adversaries regarding their demands for legal and political guarantees of a fair campaign and elections. On August 4, the Nicaraguan government signed a forty-point accord with opposition parties ranging from agreements on the repeal of national security legislation to electoral procedures.36 In return, the opposition called on the Central American presidents to demobilize the Contras.37 These agreements apparently gave Ortega leverage at the Central American presidential summit on August 5–7 in Tela, Honduras. Having shown that Nicaragua was complying with the terms of Esquipulas II, Ortega could ask the other Central American countries to move forward on the issue of greatest importance to the Sandinistas—demobilization of the Contras. As part of the Tela Agreement, signed on August 7, the Central American leaders therefore accepted a Contra demobilization plan to be implemented under UN supervision. Although Contra demobilization was supposed to be voluntary, the Contras resented the decisions others had made during the Tela process regarding their future. Not only was it the Contras’ sense that important concessions had been made to the Sandinistas during the Tela process without any reciprocal concessions from the Sandinistas, but both the Contras and the legal opposition in Nicaragua feared that the Sandinistas would not follow through on the concessions they had made if the Nicaraguan Resistance was demobilized too rapidly.38 The Tela Agreement also created confusion by leaving unspecified how Contra demobilization was supposed to take place. Despite these problems, however, there was little feeling of urgency among the Tela Agreement’s signatories regarding the need to secure the Contras’ compliance. Rather, there was some sense that the tide was turning against the Contras. Not only had military aid to the Contras been cut off by the United States, but Honduras was also tiring of their presence in that country. Moreover, key members of the Contras’ Political Directorate, prominent among them Alfredo Cesar, Pedro Joaquín Chamorro Barrios, Azucena Ferrey, Adolfo Calero, and Alfonso Robelo, had abandoned the armed resistance in order to take part in the upcoming electoral contest between the FSLN and the legal opposition, leaving control of the Contras to peasant leaders with military experience but little interest in politics or diplomacy.39 After sending a reconnaissance mission to Central America from September 3 to September 23, 1989, the Secretary-General submitted a

S-13

11/24/09

4:23 PM

Page 365

THE ROLE OF AN IMPLEMENTATION REGIME IN NICARAGUA

365

report to the UN Security Council on October 11 recommending the deployment of the UN Observer Group in Central America (ONUCA). In Resolution 644 (1989) of November 7, 1989, the Security Council approved the Secretary-General’s report and decided to deploy ONUCA at once for a period of six months. Command in the field was to be the responsibility of a chief military observer (CMO), Spain’s Major-General Agustin Quesada Gomez, who in turn would report to the SecretaryGeneral and, through him, to the Security Council.40 ONUCA’s mandate would be to conduct on-site verification of the cessation of aid to rebel forces and the nonuse of one state’s territory for attacks on others, as agreed to in the Esquipulas II Agreement. Accordingly, ONUCA military observers would be located in all five Central American countries. Because the observers were unarmed, the plan was for them to initiate informal contact with irregular forces in their patrol areas in order to seek the latter’s cooperation, including their agreement to abstain from using force against ONUCA members. The plan called for ONUCA to be deployed in four phases, with numbers rising from an advance party of 30 officers and civilian support staff led by the chief military observer in Phase I, to 99 observers in Phase II, to 162 members in Phase III, and culminating with the addition of 98 more military observers during Phase IV.41 The Tela Accord also called for the establishment of an International Commission of Support and Verification (CIAV) by the UN and the OAS that would work with governments and international organizations to facilitate Contra demobilization and resettlement. This would include setting up temporary camps and receiving and storing arms and military equipment from the Nicaraguan Resistance until the five Central American presidents decided how to dispose of them. Since the UN was assigned repatriation duties in Costa Rica, El Salvador, and Honduras, countries from which the Contras were to depart in relatively short order, its involvement in CIAV would prove relatively short, lasting only through mid-1990. The OAS contribution to CIAV, on the other hand, would involve longer-term responsibility for repatriating all Contras and their family members once they had crossed back into Nicaragua. In addition, the CIAV-OAS mandate included verifying and investigating human rights violations and the exercise of justice, a set of tasks that kept the mission in Nicaragua several years longer than envisioned in the original mandate.42 The Esquipulas stage of the peace process ended successfully with the Sandinistas securing the other Central American countries’ agreement to dismantle the Contras. The potential for peace in Nicaragua looked promising at this stage. Key actors had agreed upon the steps necessary to build peace and had constructed an effective implementation regime to follow through on the peace plan. Recognizing the practical and political difficulties they faced in verifying the international security components of the Esquipulas II Accord, the Central American presidents had turned to the

S-13

11/24/09

366

4:23 PM

Page 366

CASE STUDIES

UN and OAS as third-party implementers for help. In accordance with the structure of Esquipulas II, which left progress on internal democratization up to each government in the region, the mandate of the third-party implementers was relatively narrow. The UN and OAS were not called upon to broker or enforce constitution-making, for example, or agreements on autonomy, power sharing, or restructuring national institutions. On the other hand, the exclusion of rebel forces from the peace agreement would prove to be a challenge for external implementers. ONUCA and CIAV could not be sure that the Contras would not act as spoilers and attack the peace process. This concern was reinforced by the fact that the third-party implementers lacked the authority and capacity to prevent the movement of armed personnel across borders.43 Despite the promise of the Esquipulas peace process, it did not succeed in bringing an end to the conflict in Nicaragua. In effect, the architects of the Esquipulas process failed to address two questions central to establishing long-lasting peace: Would the Sandinistas, who were responsible for implementing the internal “democratization” elements of the Esquipulas II Accord, fulfill their commitments to the unarmed opposition, including respecting the results of free elections? Could one ensure the “voluntary” demobilization of the Contras called for in the Tela Accord when the Contras professed themselves unwilling to disband and when neither ONUCA nor CIAV had made plans to carry out Contra demobilization as implementers of the international security elements of the agreement? Because Esquipulas failed to answer these questions, two additional stages in the peace process were needed to secure a stable peace in Nicaragua.

Elections and Transition Accords Responding to domestic demands as well as regional and extraregional pressures, the FSLN agreed to advance national elections from the November 1990 date called for by the country’s constitution to February 25, 1990. Because Contra demobilization had not taken place prior to the election and the Contra leadership had further stated their refusal to disarm unless the Sandinistas were voted out of power, there was no guarantee that the elections would produce an end to the conflict. Nevertheless, the majority of Nicaraguan voters went to the polls believing that the main issue at stake in the election was war and peace. The results of a national postelection survey indicate that 75.6 percent of the voters believed that “the war was never going to end if the FSLN won,” given U.S. antipathy to the Sandinistas, while an electoral victory by the broad-based coalition of fourteen opposition parties known as the National Opposition Union (UNO) would pave the way for peace as both the coalition and the Contras enjoyed U.S. support.44

S-13

11/24/09

4:23 PM

Page 367

THE ROLE OF AN IMPLEMENTATION REGIME IN NICARAGUA

367

In March 1989 the Sandinista government took it upon itself to ask the UN Secretary-General for an observer mission to monitor the electoral process in Nicaragua. Several missions were sent to Nicaragua to observe the revision of electoral laws and laws regulating the mass media. These missions were followed by the establishment of the UN Observer Mission for the Verification of Elections in Nicaragua (ONUVEN), which began to operate on August 25, 1989.45 Electoral assistance was also provided to Nicaragua by the U.S. government, the OAS, the “Friendly Countries Club” (Costa Rica, Spain, and Venezuela), and former president Jimmy Carter’s Council of Freely Elected Heads of State. U.S. electoral assistance, which consisted of a $12 million appropriation by Congress, funded not only observer missions such as the OAS’s and Carter’s, but also Nicaragua’s electoral authority, opposition parties, trade unions, and independent media. According to Rafael LopezPintor, a scholar with experience in UN electoral observer missions, the international missions contributed significantly to the implementation of the electoral process. Responsibilities were divided among the groups, with ONUVEN working with the Supreme Electoral Council, the OAS mission operating at the regional level, and Carter’s mission focusing on conflict resolution, particularly between the government and opposition leaders.46 Despite some violence and intimidation during the campaign, the elections themselves were certified as “free and fair” by international observers. The results of the elections, in which 85 percent of registered voters participated, were as follows: presidential—elected by simple majority in a single round of voting, UNO candidate Violeta Chamorro garnered 54.7 percent of the votes while the FSLN’s Daniel Ortega received 40.8 percent; National Assembly—elected on the basis of proportional representation from party lists and with some seats allocated according to population, the UNO won 56 percent of the seats while the FSLN gathered 42 percent; municipal—elected from party lists and on the basis of a mixed majority–proportional representation formula, the UNO took 75 percent of the mayoral seats and the FSLN 24 percent.47 These results proved to be something of a shock not only to pollsters, who had shown Ortega leading Chamorro in the polls, but also to the FSLN: during a nationally televised appearance following announcement of the electoral results, Ortega admitted that he had not thought he would lose the election. Although the FSLN promised to accept the voters’ verdict and step into the role of opposition party, many questioned its sincerity. The Sandinista regime had expressed concern over UNO threats to oust Sandinistas from properties confiscated after the collapse of the Somoza dictatorship, and the reaction of the Sandinista army to election results was a cipher, prompting anxiety both in Nicaragua and abroad.48 Sandinista behavior in the period following the elections and before the April 25 transfer of power to the UNO also spurred nervousness about the transition. For example, in what

S-13

11/24/09

368

4:23 PM

Page 368

CASE STUDIES

has been called the “piñata,” Sandinista officials seized and destroyed equipment and supplies, and the Sandinista-controlled legislature legalized the distribution of property to members of government and occupants of seized or abandoned properties.49 Concern about the transfer of power also stemmed from recognition that the FSLN continued to be Nicaragua’s single largest political party and controlled the military, the police, and the country’s largest trade unions. The UNO, on the other hand, was a fractious coalition of parties united by little other than their desire to remove the Sandinistas from power. Committed to reconciling a divided society, and aware that implementation of this stage of the peace process rested in the hands of Nicaraguans, despite interest among some international missions to help oversee the political transition, Violeta Chamorro opted to practice inclusive politics and negotiate a political pact with the Sandinista opposition.50 During the course of a one-month period following the elections, transition teams for the incoming and outgoing governments negotiated the Protocol for the Procedure of the Transfer of Executive Power, also known as the Transition Accords.51 As part of the pact, the government-elect promised to retain the country’s 1987 constitution, particularly the political institutions and rights secured by the Sandinista revolution; to respect the institutional and command structure of the EPS and the Ministry of the Interior; and to accept the distribution of property that had taken place during the Sandinista era. For its part, the FSLN accepted that the EPS be converted into a nonpartisan and nonideological institution; that its size and budget be reduced; and that properties confiscated during the previous decade be reviewed to identify cases of unjust expropriation. The pact also made provisions for a nonpartisan amnesty in order to foster national reconciliation.52 One of the most controversial elements of President Chamorro’s decision to practice inclusive politics was her announcement on inauguration day, April 25, that she would retain General Humberto Ortega, the Sandinista defense minister, as army chief. Although Chamorro became commander in chief of the EPS and controlled the portfolio of the Ministry of Defense, her authority over the EPS was limited. Ortega’s powers, set out in the Law of Military Organization of the EPS, enacted shortly before the February 25 elections, were extensive, ranging from authority over military promotions to force deployment, contracting for weapons procurement, and drafting the military budget for presentation to the government. Sandinista dominance of the military also took the form of command by Sandinista officers of all military regions and general staff directorates.53 In addition, the National Police force continued to be controlled by the Sandinistas after Chamorro assumed office. Finally, the judiciary was also dominated by the Sandinistas, from the Supreme Court to trial and appellate judgeships. Chamorro argued that her inclusive strategy would be able to rein in

S-13

11/24/09

4:23 PM

Page 369

THE ROLE OF AN IMPLEMENTATION REGIME IN NICARAGUA

369

the Sandinistas. Despite criticism from Contra leaders, the Bush administration, and her own UNO coalition regarding her decision to retain Humberto Ortega as EPS commander, she cited her success in dismantling the influential Sandinista Ministry of the Interior. 54 Although the Sandinistas challenged some reforms of the security forces and judiciary, the Chamorro government ultimately was able to implement a number of changes in these institutions. The Sandinista army was downsized under the stewardship of Ortega from 86,810 troops in January 1990 to 14,553 in 1994, in the process becoming the smallest army in Central America.55 Furthermore, what began as a crisis, when Chamorro sought to replace Ortega but realized that neither Nicaragua’s constitution nor the Transition Accords gave her a mechanism for doing so, turned into an opportunity for reform. On August 24, 1994, the National Assembly approved a new military code that enabled the president either to remove or appoint a new army chief.56 The Chamorro government also responded robustly to accusations of human rights violations by the police. Her administration created a Civil Inspection Unit, appointed a civilian without Sandinista ties as vice minister of government responsible for police supervision, removed some Sandinista police commanders from power, and passed a new police law focused on professionalism and subordination to civilian authority. In addition, she succeeded in somewhat depoliticizing the judiciary when appointments to the Supreme Court secured a non-Sandinista majority by 1993.57 President Chamorro’s policy of inclusion to secure Sandinista cooperation with electoral outcomes and the rule of law proved a viable strategy for implementing the peace agreements at this stage. Although working with the Sandinistas posed political problems for her government, Chamorro was not in a position to construct a new political system and institutions, since the Sandinistas had not been militarily defeated, as Chamorro herself noted. Rather, she had to work within the framework of the institutions she had inherited from the Sandinistas.58 Through the Transition Accords, the Chamorro government and the Sandinistas took what one Nicaraguan newspaper called “the first firm step toward the institutionalization of democratic order, peace, and the orderly exercise of power in Nicaragua” and thereby moved one step closer to consolidating a peace that reflected the reality of divided power in Nicaragua.59

The Disarmament Accords Neither the elections nor the Transition Accords were able to resolve the final issue confronting the Nicaraguan peace process, the disarmament and demobilization of the Contras. At their meeting in Tela, Honduras, in August 1989, the five Central American presidents had agreed that the

S-13

11/24/09

370

4:23 PM

Page 370

CASE STUDIES

insurgent group should disband by November 4. When the Contras, who had not been present at the Tela meeting, did not comply, the five heads of state, meeting in December 1989 in San Isidro de Coronado, Costa Rica, rescheduled the final date for Contra demobilization to February 5, 1990. This date, too, passed without Contra compliance, and when elections were held later that month, the Contras were still in arms. Even after the UNO’s electoral victory, Contra demobilization continued to look doubtful. Citing security concerns, Contra leaders said they would demobilize only after Violeta Chamorro replaced high-level Sandinista security-force commanders. “We don’t consider the fight to be over,” stated one Contra field commander, “until the Sandinistas’ means of repression have disappeared.”60 At this time, the first expansion of ONUCA’s mandate took place. Responding to a request made by the Central American presidents following their meeting in San Isidro de Coronado, ONUCA consented to verify any cease-fires or demobilizations agreed to in the region. Some three weeks after the elections in Nicaragua, UN Secretary-General Perez de Cuellar informed the Security Council that consultations between the Nicaraguan government, Chamorro’s government-elect, and the United Nations had produced an agreement regarding the manner in which Contra demobilization would proceed. Contras in Honduran camps were to be demobilized there and repatriated, at which time they would become the responsibility of CIAV. Contra demobilization in Nicaragua would take place at temporary assembly points established by ONUCA. Because Contra disarmament was to take place at these sites, ONUCA was also charged with ensuring the security of the Contras. In the face of continued Contra resistance to demobilization, the Secretary-General asked the Security Council to allow ONUCA to add armed personnel to its numbers in order to perform these tasks. The Security Council formally authorized this expansion of ONUCA’s mandate on March 27, 1990, and members of a Venezuelan armed infantry battalion arrived in the region two weeks later.61 In the interim, efforts were made to secure an agreement on demobilization on the part of the Contras. On March 27, representatives of Chamorro’s government-elect and leaders of the Nicaraguan Resistance met in Honduras and signed the Toncontin Accord, in which the Contras agreed to disarm and hand in their weapons before Chamorro’s inauguration on April 25. Expressing ongoing concerns about their security, however, the Contras continued to resist disarmament and demobilization. This prompted further negotiations, this time between the Nicaraguan government, representatives of President-elect Chamorro, and the Contras, with the participation of Cardinal Miguel Obando y Bravo, CMO Quesada Gomez of ONUCA, and Iqbal Riza, the Secretary-General’s alternate personal representative for the Central American peace process.62 The product of these negotiations was the Managua Agreement, signed on April 18 by

S-13

11/24/09

4:23 PM

Page 371

THE ROLE OF AN IMPLEMENTATION REGIME IN NICARAGUA

371

representatives of the government, the Nicaraguan Resistance, Presidentelect Chamorro, and Cardinal Obando y Bravo. It was on the basis of the Managua Agreement that ONUCA’s mandate was expanded a second time.63 This agreement called for a cease-fire by April 19, set June 10 as the “latest” date for the completion of Contra disarmament, and addressed Contra concerns regarding their security during demobilization by specifying five (later seven) security zones in which Contras would assemble and surrender their weapons to ONUCA. Stipulating that the EPS had to withdraw twenty kilometers from the perimeter of each of these zones, the only military forces or weapons that were to be allowed in the security zones were those of the Venezuelan battalion that had joined ONUCA. Once disarmed, CIAV would become responsible for providing humanitarian aid to the Contras and assisting them in their reintegration into society.64 On May 4, the Security Council expanded ONUCA’s mandate for an additional six months, stating that ONUCA’s new tasks of monitoring the cease-fire and separating forces and demobilizing the Contras would cease by June 10, 1990.65 Although the security zones called for in the Managua Agreement were established, the Contras continued to drag their feet on disarmament. Emphasizing that Contra disarmament was voluntary and that ONUCA’s role was limited, General Quesada greeted the Venezuelan troops on their arrival in Central America by stating: “Your mission is exclusively military. You will be in charge of verifying, providing protection for, and supervising the demobilization. . . . Nothing will happen with the combatant who does not want to demobilize. That will be his problem and that of his country, not of ONUCA or the Venezuelan battalion.”66 With ONUCA unable to jump-start demobilization, it was only after President Chamorro met with Contra leaders, gave her personal assurances of their security, and promised reforms to the Nicaraguan military, that the Nicaraguan Resistance began to disarm. The Contras suspended demobilization just two weeks later, however, claiming that Chamorro’s government had not “kept its part of the bargain” and expressing concern about the “climate of uncertainty and social instability” Nicaragua was experiencing as FSLN-affiliated unions went on strike and demanded wage increases.67 New negotiations between the Chamorro government and Contra leaders produced the Managua Protocol on Disarmament, signed on May 30. The Contras renewed their commitment to complete demobilization by June 10 while the government renewed its promises to provide for their security, agreeing to withdraw its armed forces even further from the security zones. The government also offered the Contras additional incentives and confidence-building measures, including requesting an extension of the mandates of both ONUCA and CIAV; committing to the creation of special development zones (called “development poles” in the protocol) for the

S-13

11/24/09

372

4:23 PM

Page 372

CASE STUDIES

demobilized Contras in northern and central Nicaragua and the Atlantic region, totaling some 8,500 square kilometers; establishing a rural police force in the development zones to be staffed by former combatants of the Nicaraguan Resistance; including Contra representatives in government ministries such as Health and Education perceived to have a direct impact on the former combatants; and extending economic aid to all demobilized Contra forces.68 Contra demobilization proceeded much more smoothly after the signing of the Managua Protocol. Reporting to the Security Council in early June on the progress of demobilization, the Secretary-General acknowledged that the June 10 deadline could not be met and requested an extension of ONUCA’s mandate until June 29. On July 6, the day following the return home of the last members of the Venezuelan battalion and the official end of the demobilization process, the number of Contras disarmed and demobilized was reported at 2,759 in Honduras and 19,614 in Nicaragua.69 Weapons collected from the demobilized soldiers included 15,144 smallcaliber arms, 137 mortars, 1,282 grenade launchers, and 9 Redeye missiles, among others.70 At this point ONUCA’s mandate reverted to its original and more limited task of monitoring the borders for possible violations of the transborder security component of Esquipulas II.71 In contrast to ONUCA’s shrinking mandate, CIAV’s responsibilities during this period grew, moving from its first phase of assistance to the demobilized, including for relocation, to the second phase, in which the CIAV was to facilitate the material and social reinsertion of Contras and their families into society. During a twelve-month period from 1990 to 1991, CIAV dispensed food, clothing, and medical assistance, farming and cooking implements, and construction materials to former combatants and their families, as well as the equivalent of U.S.$50 to each demobilized soldier. CIAV estimates that it extended humanitarian assistance to 111,481 former combatants, their families, and other refugees before moving on to the third phase of its efforts, establishing the development poles called for in the Managua Protocol and initiating a reconstruction program.72 The first two phases of CIAV’s resettlement efforts were deemed generally successful by analysts.73 Nevertheless, the CIAV did face obstacles in performing its work. One of these stemmed from the composition of the CIAV-OAS mission in Nicaragua. Headed by an Argentine OAS official and with key positions staffed by Argentines, CIAV provoked some resentment among Sandinistas who were aware of Argentine military assistance to the Contras in the early 1980s. Although the Argentines who were involved in helping to dismantle the Contra forces were primarily civilian bureaucrats, rather than the military officials who had initially been involved in training the Contras, this did not serve to diminish Sandinista suspicion regarding the political leanings of the CIAV.74 Sandinista criticism of the CIAV would prove inevitable, however; charged with helping

S-13

11/24/09

4:23 PM

Page 373

THE ROLE OF AN IMPLEMENTATION REGIME IN NICARAGUA

373

the former Contras and their families, the organization could not help but become involved in postelection politics in Nicaragua.75 The CIAV also had to contend with a scarcity of resources to finance the costly programs for which it was responsible, due in large measure to its reliance on the United States for funding. Of the $49.75 million in humanitarian aid for the Contras that the U.S. Congress had voted in April 1989, the United States initially allocated only $3 million to the CIAV for Contra relocation.76 The resource gap proved particularly detrimental to CIAV’s implementation of the development poles. In this instance, the Chamorro government’s financial inability to follow through on agreements relevant to the development zones left thousands of demobilized Contras without land and hindered the CIAV’s capacity to fulfill this phase of the mandate. When coupled with deteriorating economic conditions in the countryside, the land issue contributed to a rise in rural violence and the emergence of what were called Recontras (former Contra troops) and Recompas (former EPS troops) in 1991.77 The CIAV was also charged after July 2, 1990, with verifying security conditions and guaranteeing that demobilized former combatants, their families, and refugees could exercise fundamental rights and liberties. Between July 1, 1990, and April 30, 1991, CIAV reported 52 deaths of demobilized rebels, half of which were attributed to the army or police; 218 detentions of former rebels; 159 threats; and 72 attacks that produced injuries.78 Because of the level of instability in Nicaragua at the end of the civil war, CIAV’s security mandate was widened considerably over time. The CIAV thus found itself dealing with hostage rescue situations, lending support to judicial and other government institutions, and acting as a mediator between the rearmed Recontras, the Recompas, and the Nicaraguan government.79 Although the Nicaraguan government ultimately achieved the negotiated surrender of the last of the Recontras in 1994, a number of armed bandits continued to operate in the countryside, demanding government time and attention.80

Evaluating the Peace Despite some ongoing clashes between government security forces and rural criminal bands, Nicaragua has now experienced several years of peace. In the years since the successful completion of the Esquipulas II peace process, the country has held a second round of general elections in 1996, reformed the 1987 Sandinista-constructed constitution in 1995, and undertaken structural adjustment reforms of the economy. Although these and related events have produced some conflict in Nicaraguan society, the country’s ability to manage that conflict is a testament to the consolidation of the peace in Nicaragua.

S-13

11/24/09

374

4:23 PM

Page 374

CASE STUDIES

In addition to ending the suffering associated with war, peace has brought other improvements to the lives of Nicaraguans. The country’s human rights record has continued to improve, despite, for example, persistent problems with soldiers’ performance of law enforcement duties in rural areas, incidents of police brutality, and violence against women. Nicaragua’s judiciary can now be considered independent, although still weak, susceptible to political and material influence, and therefore limited in its prosecution of human rights abuses.81 By some measures, Nicaragua has also turned the corner economically, with inflation under control and GDP growing since 1994. On the other hand, a study published in March 1995 found that 36.6 percent of the urban population was “severely impoverished” and the rate of illiteracy was higher in 1996 than it had been in 1981 after the Sandinistas launched an ambitious literacy campaign. 82 Finally, civil society organizations, many of which were established by the Sandinistas (for example, the Sandinista Defense Committees [CDS], the Association of Nicaraguan Women Luisa Amanda Espinosa [AMNLAE], and labor organizations such as the urban Sandinista Workers Federation [CST] and the Rural Workers Association [ATC]), have declared their autonomy or otherwise distanced themselves from the FSLN, thus broadening the basis for citizen participation in Nicaragua. Regarding Nicaragua’s progress in consolidating democracy and the rule of law, the 1996 elections are instructive. The day after the elections, some political parties alleged discrepancies between official returns and the results tabulated by party poll-watchers. Following a three-week review process in which international monitors participated, the results were finalized a month after the elections. Arnoldo Aleman of the Liberal Alliance was named winner of the presidency with 51 percent of the vote, with the FSLN’s Daniel Ortega finishing second at 38 percent. The Liberal Alliance secured forty-two seats and the FSLN thirty-six seats in the National Assembly; the fifteen remaining seats were split by nine other parties. Ninety-one mayoral posts were won by the Liberal Alliance and fifty-two by the FSLN. 83 Two days later, Ortega publicly declared the elections fraudulent and the Sandinistas agreed to acknowledge the Aleman government only as “legal,” but not legitimate. This impasse was only resolved when the two men personally proceeded to negotiate a common agenda on two key issues facing the country—the ongoing conflict over property seized and distributed during the Sandinista era and reform of the electoral law.84 Although polls after the 1996 elections indicated declining public confidence in electoral institutions, an electoral turnout of 80 percent of registered voters and the patient manner in which the population, political parties, and their candidates awaited the results of the election demonstrated Nicaraguans’ commitment to the democratic process and the rule of law.85 What still remains to be seen is whether Nicaraguans can move to institu-

S-13

11/24/09

4:23 PM

Page 375

THE ROLE OF AN IMPLEMENTATION REGIME IN NICARAGUA

375

tionalize conflict management rather than relying on behind-the-scenes bargaining by political opponents to resolve crises.

Conclusion The years of peace Nicaragua has experienced since the end of the Esquipulas peace process attest to the successful implementation of that country’s peace agreements. In Nicaragua’s civil war settlement, the implementation process was influenced both by the strategic environment in which it took place and by organizational factors. The end of the Cold War and the organizational flexibility of ONUCA and the CIAV were variables central to the successful implementation of the Nicaraguan peace accords. These factors are unlikely to have exerted a positive effect, however, were it not for the creation by regional actors of an implementation regime whose framework provided a means of directing and managing the implementation process. Although united in their desire to bring peace to Central America, the five presidents who signed Esquipulas II soon realized they were unable to implement the transborder security commitments called for in the agreement. Signaling their ongoing commitment to peace via a series of presidential summits in which they negotiated accords that were increasingly pragmatic in nature, the regional leaders constructed a regime focused on implementation. This high-profile implementation regime, which increasingly centered on Nicaragua, brought in the United Nations and the OAS, international actors with the resources and skills to verify and monitor progress toward peace. It also attracted the support of a variety of countries and organizations including Canada, several European and Latin American countries, and Jimmy Carter’s Council of Freely Elected Heads of State. In this case, a strategy of focusing worldwide attention on the peace process in Nicaragua appears to have convinced key actors with interests in the outcome—among them the Sandinistas, the superpowers, and some elements of the Contras—to engage in moderate behavior. The successful implementation of the peace in Nicaragua resulted in no small measure from changes in the global and regional strategic environments. During the first phase of the peace process, implementation proved difficult largely because of continued U.S. involvement in what it defined as a Cold War conflict. The end of the Cold War created more favorable conditions for implementing regional peace as Soviet support for the FSLN declined, security concerns among Central American countries decreased, and the Bush administration in the United States embarked on a less ideologically inclined foreign policy. Honduras, for example, grew impatient with the continued presence of Contra bases on its territory. The Sandinista regime, meanwhile, was pressed by the dismal state of the country’s econo-

S-13

11/24/09

376

4:23 PM

Page 376

CASE STUDIES

my to enter into political dialogue with opposition parties in an effort to cope with the country’s economic problems. The United States, in turn, lent its economic support to Nicaragua’s elections and, though somewhat grudgingly, the demobilization of the Contras. The organizational flexibility demonstrated by the United Nations and the Organization of American States also contributed to the successful implementation of the peace in Nicaragua. Although ONUCA was initially conceived as a verification and peace-observation mission, the UN acceded to the Central American presidents’ requests for expansion of ONUCA’s mandate as conditions changed in the region. The CIAV-OAS also demonstrated considerable organizational adaptability, taking on tasks in the process of Contra relocation that it had never tackled before and working to strengthen institutions in postconflict Nicaragua, foremost among them the judicial institutions, in order to improve their ability to guarantee justice and security for the demobilized forces. Both the UN and the OAS also recognized their place within a broader implementation regime. These organizations understood that—indeed the UN initially had insisted upon— responsibility for the ongoing political management of the Esquipulas peace process rested with the Central American presidents, while implementation of the democratization elements of the peace accords was in the hands of respective regional governments, and centralized management of the administrative and logistical support necessary for implementation was the principal task of the UN. Such factors further help to explain why and how the Nicaraguan peace agreements could be implemented successfully despite the exclusion of one of the major antagonists to the conflict, the Contras, from the negotiation process. Indeed, implementation of the peace accords began before the Contras had even agreed to disarm. By structuring an implementation regime around the role that governments should play in bringing peace to the region, the Central American presidents ensured that implementation would at least partly succeed without the Contras’ assent to the peace agreements. In addition, this framework demonstrated the credible commitment of the Central American governments to the peace process. Once implementation began in earnest, the Contras increasingly feared marginalization from the peace process, prompting leading figures among the Nicaraguan Resistance to join the National Opposition Union. With the ability of the Contras to “spoil” the peace process additionally undermined by the end of the Cold War and the changing political climate in the United States, the perception grew that the Contras were finished as an effective fighting force. What lessons can be learned from the successful implementation of the Nicaraguan peace agreements? One is that a well-constructed framework which clarifies the responsibilities of domestic, regional, and international implementers and which structures the division of labor among them is use-

S-13

11/24/09

4:23 PM

Page 377

THE ROLE OF AN IMPLEMENTATION REGIME IN NICARAGUA

377

ful as a device not only to signal commitment but also to bring coherence to a complex organizational undertaking. Ideally, this implementation regime should be a nested one, created by local parties to the conflict with a vested interest in and commitment to the peace process and expanded to include regional and extraregional actors, as well as outside implementers, once the regime is firmly established and as conditions warrant. The case of Nicaragua also demonstrates the value of flexibility in an implementation regime, and among its members, to be able to adapt to changing circumstances. Had the United Nations not been willing to shift from its original mandate of verification and peace-observation to peacekeeping, the Contras are unlikely to have been willing to gather in security zones and disarm. Finally, the Nicaraguan case shows that even in the absence of ideal conditions for peace implementation, committed domestic, regional, and international implementers can move the implementation process forward, building on previous successes until the conditions for implementation improve. Although adopting such a strategy can entail a lengthy implementation process, it can also build a solid base for implementation and, more, lay the groundwork for the longer-term endeavor of building peace.

Notes 1. For a description of the FSLN’s statist or mixed-economy model, see Rose J. Spaulding, Capitalists and Revolution in Nicaragua: Opposition and Accommodation, 1979–1993 (Chapel Hill: University of North Carolina Press, 1994). 2. On opposition groups, see Thomas W. Walker, introduction to Walker, ed., Reagan Versus the Sandinistas: The Undeclared War on Nicaragua (Boulder, Colo.: Westview Press, 1987), pp. 1–17; and Martin Diskin, Thomas Bossert, Salomón Nahmad, and Stéfano Varese, Peace and Autonomy on the Atlantic Coast of Nicaragua: A Report of the LASA Task Force on Human Rights and Academic Freedom (Pittsburgh: Latin American Studies Association, September 1986). 3. E. Bradford Burns, At War in Nicaragua: The Reagan Doctrine and the Politics of Nostalgia (New York: Harper & Row, 1987), p. 30. 4. Peter Kornbluh, “The U.S. Role in the Counterrevolution,” in Thomas W. Walker, ed., Revolution and Counterrevolution in Nicaragua (Boulder, Colo.: Westview Press, 1991), p. 330. 5. John A. Booth, “Nicaragua,” in Abraham F. Lowenthal, ed., Latin America and Caribbean Contemporary Record, vol. 6, 1986–1987 (New York: Holmes & Meier, 1989), p. B359. 6. Ibid. 7. The figure on U.S. military assistance to the Contras includes $300 million of military and nonmilitary assistance that the State Department reports was provided between 1982 and 1990, and an additional $100 million solicited from private organizations and foreign governments and gathered by the U.S. National Security Council. Jean R. Tartter, “National Security,” in Tim L. Merrill, ed., Nicaragua: A Country Study (Washington, D.C.: Library of Congress, Federal Research Division, 1994), pp. 205–206. 8. Details on the Contra forces’ weapons can be found in Paulo S. Wrobel,

S-13

11/24/09

378

4:23 PM

Page 378

CASE STUDIES

“Case Study: Nicaragua,” in UNIDIR, Disarmament, and Conflict Resolution Project, Managing Arms in Peace Processes: Nicaragua and El Salvador (New York: United Nations, 1997), pp. 29–32. 9. Although not all members of the Contras were former national guardsmen, the FDN’s tactics, including the kidnapping and murder of unarmed civilians, were reminiscent of the repressive strategies employed by Somoza’s National Guard. For further information on FDN actions, see Americas Watch, Human Rights in Nicaragua (New York: Americas Watch, 1982–1991). 10. On the expansion of the Sandinista armed forces, see Roberto J. Cajina, Transición política y reconversión militar en Nicaragua, 1990–1995 (Managua: CRIES, 1996). 11. Booth, “Nicaragua,” p. B361; and Wrobel, “Case Study: Nicaragua,” p. 11. 12. Because the populations of the Atlantic Coast and the Sandinistas were in general agreement regarding such “cultural” issues as religious practice and the use of indigenous languages, autonomy demands centered on economic rights, including claims regarding land and natural resources, and political representation in institutions with defined powers and jurisdiction. Diskin et al., Peace and Autonomy on the Atlantic Coast of Nicaragua, p. 2. 13. The Reagan administration’s policies of selling arms to Iran in the hope of securing the freedom of U.S. hostages in Lebanon and providing secret military support to the Contras were linked in late 1985 when National Security Council officials diverted profits from the arms sales to the Contras. These funds were provided to the Contras after Congress terminated secret CIA military assistance to the group in 1984. The scandal, which broke in the fall of 1986, weakened the foreign policy credibility of the Reagan administration. See Lee H. Hamilton, “Iran-Contra Affair,” in Bruce W. Jentleson and Thomas G. Paterson, eds., Encyclopedia of U.S. Foreign Relations, vol. 2 (New York: Oxford University Press, 1997), pp. 421–422; and The Tower Commission Report (New York: Bantam Books, 1987). 14. Conservative estimates of the direct and indirect costs of the war for the 1980–1986 period range from $1 billion to $1.5 billion, a sum equal to between 6 and 9 percent of GDP. Booth, “Nicaragua,” p. B362. 15. Cynthia Arnson and David Holiday, “Fitful Peace: Human Rights and Reconciliation in Nicaragua Under the Chamorro Government,” Americas Watch Report (New York: Americas Watch, July 1991). 16. For data regarding the Sandinista government’s repatriation program, see Ministerio del Interior, “Acogidos al decreto de amnistía después de Esquipulas II,” Managua, November 6, 1987. For aggregate data on Nicaraguan refugees, see U.S. Committee for Refugees, World Refugee Survey 1991 (Washington, D.C.: American Council for Nationalities Service, 1991), pp. 87–88. 17. An overview of other Central American peace processes is provided in Francisco Rojas Aravena and Luis Guillermo Solis, Súbditos o aliados? La política exterior de Estados Unidos en Centroamerica (San José, Costa Rica: Editorial Porvenir—Facultad Latinoamericana de Ciencias Sociales (FLACSO), 1988). 18. Esquipulas II Agreement, August 7, 1987. 19. Francisco Rojas Aravena, El proceso de Esquipulas: El desarrollo conceptual y los mecanismos operativos (Costa Rica: Universidad Nacional, Escuela de Relaciones Internacionales, 1989), p. 13. 20. Fernando Flores Pinel, “De esquipulas a tela: Por los inciertos caminos de la guerra y de la paz,” ECA—Estudios Centroamericanos 492 (October 1989): 816; and Rojas Aravena, El proceso de Esquipulas, p. 13. 21. Quoted in Jack Child, The Central American Peace Process, 1983–1991: Sheathing Swords, Building Confidence (Boulder, Colo.: Lynne Rienner, 1992), p. 51.

S-13

11/24/09

4:23 PM

Page 379

THE ROLE OF AN IMPLEMENTATION REGIME IN NICARAGUA

379

22. Dario Moreno, “Nicaragua,” in James M. Malloy and Eduardo A. Gamarra, eds., Latin America and Caribbean Contemporary Record, vol. 8, 1988–1989 (New York: Holmes and Meier, 1996), pp. B309–B310. 23. David B. Ottaway and Helen Dewar, “Differing Reports of Soviet Offer on Nicaragua,” Washington Post, December 12, 1987, p. A12. 24. Facultad Latinoamericana de Ciencias Sociales, Consejo Superior Universitario Centroamericano (CSUCA) y Universidad para la Paz, “Informe blanco sobre los avances logrados en el proceso de cumplimiento del acuerdo de paz para Centroamerica, ‘Esquipulas II’ a los noventa dias de haberse firmado,” FLACSO, 1987; and Hans Petter Buvollen, “Low-Intensity Warfare and the Peace Plan in Central America,” Bulletin of Peace Proposals 20, no. 3 (1989): 331. 25. United Nations, The Blue Helmets: A Review of United Nations Peacekeeping, 2nd edition (New York: UN Department of Public Information, 1990), p. 390; Child, The Central American Peace Process, pp. 49–50. 26. William Goodfellow and James Morrell, “From Contadora to Esquipulas to Sapoa and Beyond,” in Walker, Revolution and Counterrevolution in Nicaragua, pp. 380–381. 27. Cristina Eguizábal, “Regional Leadership and Universal Implementation in El Salvador’s Quest for Peace,” in Thomas G. Weiss, ed., The United Nations and Civil Wars (Boulder, Colo.: Lynne Rienner, 1995), p. 177. 28. Rojas Aravena, El Proceso de Esquipulas, p. 14. 29. Amnesty was to be granted to all former National Guard members who were cleared of crimes by the OAS. In addition, the government agreed to release half of Contra prisoners when the Contras arrived in their security zones, with the remainder to be released when a permanent cease-fire was signed. Center for International Policy, “Cease-Fire Primer: A Report on the Progress of the CeaseFire Negotiations in Central America,” International Policy Report (Washington, D.C.: Center for International Policy, July 1988). 30. Jill Smolowe, “Taking a Baby Step Toward Peace: Talks Inch Forward As Congress Approves More Aid for Contras,” Time, April 11, 1988, p. 36. 31. Center for International Policy, “Cease-Fire Primer.” 32. Douglas Farah, “Central Americans Pressing Peace Quest at OAS Assembly,” Washington Post, November 17, 1988, p. A29. 33. United Nations, The Blue Helmets, 2nd ed., p. 391. 34. Mark A. Uhlig, “Nicaragua Plans Earlier Elections,” New York Times, February 14, 1989, p. A10; Storer H. Rowley, “Nicaraguan Proposal Snags Latin Summit,” Chicago Tribune, February 14, 1989, p. C1; and Lindsey Gruson, “Latin Presidents Announce Accord on Contra Bases,” New York Times, February 15, 1989, p. A1. 35. Child, The Central American Peace Process, p. 66. 36. Richard Boudreaux, “Nicaragua’s Ortega to Discuss Foes’ Election Demands,” Los Angeles Times, July 15, 1989, p. 6; and Mary Speck, “Sandinistas, Opposition OK Accord on Elections,” Chicago Tribune, August 5, 1989, p. 3C. 37. Goodfellow and Morrell, “From Contadora to Esquipulas to Sapoa and Beyond,” p. 388. 38. Ernesto Palazio, a Contra spokesman, declared: “They [the Contras] will stay intact until they see the results of the elections. They will infiltrate Nicaragua, going in clandestinely. They will relocate, but not demobilize.” Robert Pear, “Pact in Nicaragua: U.S. Resists Too,” New York Times, November 3, 1989, p. A8. 39. Abelardo Morales, Oficios de paz y posguerra en Centroamerica (San José, Costa Rica: FLACSO, 1995), pp. 51–52. 40. United Nations, The Blue Helmets, 2nd ed., p. 394.

S-13

11/24/09

380

4:23 PM

Page 380

CASE STUDIES

41. Colonel John D. Joly, “ONUCA—A Story of Success in the Quest for Peace,” Canadian Defence Quarterly (June 1991): 13. 42. In June 1993 the Secretary-General of the OAS widened the CIAV’s mandate to include providing services to demobilized Sandinistas and extended the mandate to June 1994. For further details on the CIAV’s actions in Nicaragua, see Morales, Oficios de paz y posguerra en Centroamerica, p. 123. 43. The limits on UN and OAS authority in the region are illustrated by an incident in which a CIAV official who spoke bluntly to Contra rebels, telling them to give up fighting and return home, was criticized by James Baker, U.S. secretary of state, and Javier Perez de Cuellar, UN Secretary-General, because Contra demobilization was supposed to be voluntary. Lee Hockstader, “Ortega’s Move Designed to Rivet Attention on War,” Washington Post, October 29, 1989, p. A30. 44. Paul Oquist, “The Sociopolitical Dynamics of the 1990 Elections,” in Vanessa Castro and Gary Prevost, eds., The 1990 Elections in Nicaragua and Their Aftermath (Lanham, Md.: Rowman and Littlefield, 1992), pp. 36–37. 45. United Nations, The Blue Helmets, 2nd ed., pp. 391–392. 46. Rafael López-Pintor, “Nicaragua’s Measured Move to Democracy,” in Krishna Kumar, ed., Postconflict Elections, Democratization, and International Assistance (Boulder, Colo.: Lynne Rienner, 1998), pp. 41–42. 47. Krishna Kumar and Marina Ottaway, From Bullets to Ballots: Electoral Assistance to Postconflict Societies (Washington, D.C.: U.S. Agency for International Development, Center for Development Information and Evaluation, 1997), p. 22. 48. Anne-Marie O’Connor and Nancy Nusser, “Ortega Vows to Bow Out Gracefully,” Atlanta Constitution and Journal, February 27, 1990, p. A1; and Brook Larmer, “Voters Oust Sandinistas,” Christian Science Monitor, February 27, 1990, p. 2. 49. The “piñata” allusion is a reference to the papier-mâché figure that bursts when hit with a stick, spilling prizes that people then rush to gather. 50. Both Chamorro and Ortega requested that the Carter mission, as well as the UN and OAS electoral observer missions, remain in the country to oversee the transition process. Robert Collier, “Chamorro Wins, Pledges Unity,” The San Francisco Chronicle, February 27, 1990, p. A1. For more on Chamorro’s policy of inclusive politics, see Jennifer L. McCoy, “Nicaragua’s Second Elections,” in Kumar, Postconflict Elections, Democratization, and International Assistance, pp. 59–60. 51. The transition negotiation teams were headed by, respectively, Chamorro’s campaign manager and son-in-law, Antonio Lacayo, and Sandinista defense minister Humberto Ortega, brother of Daniel Ortega. 52. Angel Saldomando, “Nicaragua: Conflictos y negociación—La difícil construcción de la paz,” in Gabriel Aguilera Peralta et al., Procesos de negociacion comparados en Africa y America Latina (Guatemala: FLACSO, 1994), pp. 76–85; and Cajina, Transición política, pp. 334–335. The issue of accountability played virtually no role in the implementation of the Nicaraguan peace process given the Sandinista-led National Assembly’s decree of unconditional amnesty in March 1990, a piece of legislation that received strong opposition support. The amnesty extended to crimes committed by public officials, members of the military, and those who “committed crimes against the public order and the interior and exterior security of the state.” See Arnson and Holiday, “Fitful Peace,” for further details on the amnesty and the reaction it has provoked on the part of human rights organizations. 53. Tartter, “National Security,” pp. 207–209.

S-13

11/24/09

4:23 PM

Page 381

THE ROLE OF AN IMPLEMENTATION REGIME IN NICARAGUA

381

54. Mark A. Uhlig, “For Unity’s Sake, Chamorro Risks Losing Friends by Embracing the Foe,” New York Times, April 29, 1990, p. 2. 55. Cajina, Transición política, pp. 347–348. 56. Chamorro named Joaquín Cuadra Lacayo the new commander of the army on December 21, 1994. Cuadra Lacayo assumed command on February 21, 1995, the date the Sandinistas commemorate the assassination of General Agusto Sandino. 57. Tartter, “National Security,” pp. 217–218, 223. 58. Nina Serafino, “Government and Politics,” in Merrill, Nicaragua, p. 147. 59. Editorial, “Un acuerdo trascendental,” La Prensa, Managua, March 28, 1990, cited in Cajina, Transición política, pp. 342–343. 60. Nancy Nusser, “Wary Contras Say They Won’t Disarm Yet,” Atlanta Constitution and Journal, March 1, 1990, p. A12. 61. United Nations, The Blue Helmets: A Review of United Nations Peacekeeping, 3rd ed. (New York: UN Department of Public Information, 1996), p. 416. 62. United Nations, The Blue Helmets, 3rd ed., p. 417. 63. The second expansion of ONUCA’s mandate was approved by Resolution 653 (1990) of April 20, 1990. United Nations, The Blue Helmets, 3rd ed., p. 417. 64. “Texto del acuerdo de alto al fuego y desmovilización de los rebeldes,” Diario las America, April 21, 1990, p. 6A. 65. The Security Council extended ONUCA’s mandate by Resolution 654 (1990) of May 4, 1990. United Nations, The Blue Helmets, 3rd ed., p. 417. 66. Quoted in Child, The Central American Peace Process, p. 95. 67. Gabriela Selser, “The Never-Ending Story,” Barricada Internacional, Managua, June 2, 1990, p. 5. 68. “Protocolo de Managua sobre el desarme,” Centroamerica Hoy no. 42 (1990): 10; and Sergio de Castro, “End of a Chapter,” Barricada Internacional, June 16, 1990, pp. 3–4. 69. Data from Morales, Oficios de paz, p. 64, tabs. 1–2. 70. For data on weapons, see ibid., p. 65, tab. 3. Although Resistance leaders assured ONUCA that all weapons had been turned in by their troops, the fact that military weapons were used in clashes by the Recontras (former Contra troops) and Recompas (former EPS troops) in 1991 and 1992 and have been associated with a deterioration in public order and the rise of organized crime since the end of the war indicates that a number of fighters retained their weapons. For an analysis of post–civil war violence in Nicaragua, see Saldomando, “Nicaragua.” For data on post–civil war incidents of violence in Nicaragua, see the tables in Xochitl Lara and Rene Herrera, La pacificación en Nicaragua (San José, Costa Rica: FLACSO, 1996), pp. 101–116. 71. ONUCA’s mandate subsequently was extended a number of times, initially at the request of the five Central American governments and later at the recommendation of the Secretary-General. Over time, there were reductions in the size of ONUCA strength and a refocusing of its tasks on liaison and exchange of information with the security authorities of the five Central American states. On January 16, 1992, by Resolution 730, the Security Council decided to terminate ONUCA’s mandate. United Nations, The Blue Helmets, 3rd ed., pp. 419–421. 72. Morales, Oficios de paz, pp. 75–76. 73. See, for example, ibid., p. 78. 74. Child, The Central American Peace Process, p. 86. 75. Ibid., pp. 119–120. 76. The offer by the United States of $3 million followed a call by the Central American presidents at their December 1989 summit in Costa Rica for all U.S. sup-

S-13

11/24/09

382

4:23 PM

Page 382

CASE STUDIES

port of the Contras to be channeled through the CIAV. John M. Goshko, “U.S. Offers $3 Million for Contras’ Relocation,” Washington Post, January 6, 1990, p. A16. 77. The Recontras and the Recompas not only clashed with one another but also took up arms against the government in order to exert pressure for further distribution of land and other material benefits. For more information on these issues, see Hablan los desmovilizados de guerra: Nicaragua, El Salvador y Mozambique (Managua: Centro de Estudios Internacionales, n.d.). See also Arnson and Holiday, “Fitful Peace,” p. 7; and Morales, Oficios de paz, p. 86. 78. Arnson and Holiday, “Fitful Peace,” p. 6. 79. In effect, a second negotiation process between the government and armed groups making material and economic demands of the government took place during 1991 and 1992, with forty-one accords signed during a three-month period alone. Government promises to provide the armed groups with houses and land, as well as money for each weapon that was turned in, succeeded in bringing a degree of peace to the Nicaraguan countryside. Saldomando, “Nicaragua,” p. 99. 80. Morales, Oficios de paz, pp. 118–119. 81. U.S. Department of State, “Nicaragua Country Report on Human Rights Practices for 1997,” released by the Bureau of Democracy, Human Rights, and Labor, January 30, 1998, www.state.gov/www/global/hu...ghts/1997_hrp_report/ nicargua.html. 82. Economist Intelligence Unit, Country Profile—Nicaragua and Honduras, 1997–98 (London: Economist Intelligence Unit, 1997), pp. 23, 29–31, 33. 83. Election results reported in McCoy, “Nicaragua’s Second Elections,” p. 64. 84. Ibid., pp. 64–66. 85. McCoy makes this point in her analysis of the 1996 elections in Nicaragua. Ibid., p. 66.

S-14

11/24/09

4:23 PM

Page 383

14 Assessing El Salvador’s Transition from Civil War to Peace CHARLES T. CALL

In January 1992, the leadership of the leftist Farabundo Martí National Liberation Front (FMLN) and the negotiating team of the right-wing Salvadoran government, joined at the last minute by President Alfredo Cristiani, signed a peace accord ending twelve years of civil war in El Salvador. Many analysts consider El Salvador’s transition from civil war to peace among the most successful implementations of a peace agreement in the post–Cold War period. Following the signing of the peace accord, the agreed-upon cease-fire between the two sides was never broken. Two years later, the guerrilla forces were demobilized and reconstituted as a political party, significant demilitarization of society and the state had taken place, and “founding” elections were held. In the mid-1990s the former guerrillas became the second most powerful party in the country. One study of post–civil war agreement elections rates El Salvador’s 1994 presidential and legislative elections as the only free and fair elections carried out without crises or contested results.1 The peace accords provided a catalyst for the incipient institutionalization of democracy in El Salvador. Despite the existence of elected civilian rule since 1984, sweeping institutional reforms negotiated as part of the accords permitted the country for the first time to be considered a procedural democracy.2 The United Nations played an important role throughout the negotiation and the implementation of El Salvador’s peace accords. It served first as the mediator between the two sides, then as the verifier of a series of peace agreements, and finally as “institution-builder” and reconstruction agent through its coordination of financial and technical assistance. Although the principal implementers of the Salvadoran peace accord were the two parties to the agreement—the Salvadoran government and the FMLN—the United Nations played a prominent and perhaps determinant 383

S-14

11/24/09

384

4:23 PM

Page 384

CASE STUDIES

role in the implementation process. The UN’s role in El Salvador was unprecedented in many respects: it was involved in the resolution of an internal conflict from start to finish; it emphasized human rights verification as an element of UN peacekeeping; it deployed UN observers before a cease-fire was in place; and, above all, it aimed at preventing future conflict through institution-building, or peacebuilding.3 To what extent is El Salvador’s successful transition to peace the result of these unprecedented efforts by external verifying actors? In this chapter, I make two arguments. First, based upon the Salvadoran case, I argue that the character of a conflict and the conditions that facilitate its termination are at least as important as the role of implementing actors. More specifically, the class-based nature of the conflict, the presence of a strategic stalemate between the parties, the number and cohesion of the parties, the changing nature of the international environment, and the shifting interests of major international actors all converged to foster a political will and capability for a negotiated settlement whose character facilitated successful implementation. In the Salvadoran case, the United Nations and other implementing parties made important contributions to implementation of the peace accords. However, analyzing the choices and strategies of the United Nations and other international implementers without tying them back to the conditions of the war and the process of negotiation would be misleading. Second, I show that El Salvador’s transition from civil war to peace was highly contested, and that the success of the Salvadoran peace process was not without important shortcomings or qualifications. These shortcomings do not mitigate the important achievements of the peace process; indeed, to a large degree the flaws of the process stem from the very same conditions that led to the agreement and paved the way to peace, political democratization, improved human rights, and strengthened state institutions and legitimacy. However, the nature of the conflict and the negotiated agreement shaped the process of implementation in ways that continue to constrain success.

Nature and Consequences of the Conflict In one of the few statements with which his leftist opponents agreed, President Cristiani stated in 1992 that the cause of El Salvador’s war lay in the absence of democratic political space and the lack of economic hopes in a country characterized by exclusion and authoritarianism.4 Civil war was rooted in class and ideological, rather than religious or ethnic, divisions, in an economic system whereby a small number of landed elites controlled the state in alliance with a powerful military. Despite the holding of periodic

S-14

11/24/09

4:23 PM

Page 385

EL SALVADOR’S TRANSITION FROM CIVIL WAR TO PEACE

385

elections under a formal constitution between 1948 and 1979, authoritarianism reigned in El Salvador. The military guaranteed stability and order for the elite, and received resources and autonomy in exchange.5 Although five main guerrilla opposition groups had been formed by 1979, their disparate efforts did not pose a threat to the state—and the situation did not constitute civil war—until 1979 and 1980. Spurred by a desire to avoid an experience like Nicaragua’s July 1979 revolution and emboldened by U.S. support for reform, a group of junior officers executed a coup in October that resulted in a “Revolutionary Governing Junta,” whose progressive members startled military hard-liners and conservative elites. The junta, promising an era of openness and social and economic reforms, dissolved the National Democratic Organization (ORDEN), the paramilitary spy network, and the Salvadoran Security Agency (ANSESAL), a state intelligence agency involved in running ORDEN’s death-squad activities. However, death-squad killings and disappearances increased as the military showed its unwillingness to heed the junta.6 As the junta’s lack of effective control over the armed forces became apparent, the more leftist members of the junta resigned in disillusionment, and a second junta gave way to a third in March 1980.7 A record number of political murders—over 1,000 per month—were recorded in 1980. Although kidnappings and killings by guerrilla groups increased, the bulk of these killings were conducted by right-wing death squads and the security forces, including the assassination of the country’s archbishop, Oscar Arnulfo Romero, in March 1980.8 The military’s wave of repression deepened polarization in the country, which strengthened and unified the armed and unarmed opposition. In 1980 the five guerrilla groups—the Popular Liberation Forces (FPL), the Popular Revolutionary Army (ERP), the Communist Party’s Armed Forces of Liberation (FAL), the National Resistance (RN), and the Workers Revolutionary Party (PRTC)—formed the FMLN, which, despite internal tensions, would remain a united armed opposition until after the war. The government’s repression focused principally upon peasant, worker, and Catholic church organizations associated with reform efforts rather than on the guerrilla combatants. Many members of these organizations joined the ranks of the FMLN, and in May 1980 almost all of the important leftist and centrist popular organizations and political parties formed a large coalition called the Revolutionary Democratic Front (FDR), which quickly formed an alliance with the FMLN that would last throughout the war.9 Between 1980 and 1983 the FMLN’s forces operated in rural areas in large units of dozens or even hundreds, driving the armed forces from what became known as FMLN “control zones.”10 From 1983 through the end of the war, the FMLN operated mainly in smaller units with a highly political strategy aimed at consolidating its support among the population through the provision of education and health services and the establishment of

S-14

11/24/09

386

4:23 PM

Page 386

CASE STUDIES

local “popular” governments. The guerrillas’ relationship with popular organizations proved indispensable to its strength and strategy. The FMLN’s military ranks numbered about 12,300 at the war’s end.11 The United States played a decisive role in the conflict. When the FMLN was on the verge of winning the war in 1983, matériel provided by the United States helped prevent the government’s collapse. Military aid per fiscal year rose from $6 million in 1980 to $35 million in 1981 to $197 million in 1984, staying between $80 million and $137 million per year for the remainder of the war.12 U.S. officials promoted the drafting of a new constitution in 1983, and pumped $1.8 million into the 1984 elections to ensure the victory of Christian Democrat José Napoleón Duarte, subsequently using the centrist image of Duarte to convince a skeptical U.S. Congress to permit the provision of $1.2 million per day into the war against the FMLN.13 All told, U.S. economic aid from 1980 to 1991 totaled $3.15 billion, and military aid totaled $1.1 billion.14 The government’s military strategy was rooted in isolating the FMLN from its logistical and arms supplies and attacking the civilian population perceived to be the support base for the guerrillas, a strategy that ultimately failed, though the armed forces gained experience and improved their fighting capability. Their ranks and air power grew dramatically. In 1979 the armed forces, including all the internal security forces, totaled 10,000 persons with 28 airplanes and 5 helicopters. By 1989 the armed forces totaled 56,000 with 63 airplanes and 72 helicopters.15 The government’s strategy accentuated the tide of refugees and internally displaced persons (IDPs) produced by the war. Most of those forced from their rural homes sought refuge either across the border in Honduras, in the United States, or in camps erected near the capital. An estimated 750,000 persons were internally displaced, and approximately 1 million persons sought refuge in the United States.16 Because the Honduran border areas served as one of the few refuges for both combatants and civilians during the war, the several camps erected there were especially sensitive. An estimated 32,000 persons eventually returned from Honduras, Guatemala, Nicaragua, and Mexico.17 Repatriation and resettlement in El Salvador were less contentious than in other civil war situations, due both to shifting U.S. immigration policies and to an effort by government, nongovernmental organizations, and international agencies to resettle before the war had ended. The war also involved numerous human rights violations that became the focus of intense international scrutiny and debate. An estimated 75,000 persons were killed directly because of the war, the majority of them civilians. Over half of this total were killed during the first four years of the war, when death squads targeted a broad range of civilians presumed to be active in “subversive” popular organizations, and when the armed forces

S-14

11/24/09

4:23 PM

Page 387

EL SALVADOR’S TRANSITION FROM CIVIL WAR TO PEACE

387

executed thousands and conducted ground sweeps targeting civilians as well as combatants. The peace accord in El Salvador emerged from a so-called strategic stalemate on the battlefield. 18 The U.S.-designed government strategy proved unsuccessful, yet it gave the government a measure of stability, as the FMLN proved unable to defeat the military. At the same time, the war accentuated the economic crisis experienced throughout Latin America during the 1980s. By 1989 the real minimum wage for the poorest class had declined to 1980 levels. Direct attacks upon strategic economic targets by the guerrillas also took their toll. The inability of the Duarte government (1984–1989) to provide prosperity was accompanied by its failure to secure peace. Despite a lack of U.S. support for a negotiated settlement during his term, Duarte initiated peace talks with the FMLN in 1984, 1986, and 1987, all of which failed. During this period, the FMLN continued to believe it could win the war, and both the economic elite and the military, backed by the United States, were unwilling to make the concessions demanded by the FMLN. Principally, the FMLN demanded participation in a temporary power-sharing arrangement that would arrange truly open elections, reorganize the military, and replace the 1983 constitution.19 By 1987, polls showed that 83 percent of the national population supported an end to the war through negotiated settlement. In that same year, the Duarte government signed the Esquipulas II Agreement, which, as Caroline Hartzell describes in Chapter 13, outlined procedures to end the region’s armed conflicts and humanitarian crises. As a result of Duarte’s inability to provide peace or prosperity, the right-wing Alianza Republicana Nacionalista (ARENA) party took control of the legislature in 1988 and the presidency in 1989. In elections in which the FMLN did not participate, landowner Alfredo Cristiani was elected president in March 1989. Although he pledged to seek peace, his government immediately intensified the war and ousted more moderate elements in the military leadership. Two changes would prove necessary for both sides to fully recognize a stalemate and begin the negotiating process. First, the end of the Cold War changed the strategic environment for both sides. The gradual collapse of the Soviet bloc reduced the likelihood of continued support for the FMLN from allies such as Cuba. More important, it removed a powerful ideological framework that had influenced the FMLN’s political thinking and vision for postwar El Salvador. The end of the Cold War also coincided with the change from the Reagan administration to the Bush administration, which soon pronounced its support for negotiated settlements in Central America. This shift in U.S. policy prevented the Salvadoran military from counting on unlimited U.S. support and reduced its ability to ignore the peace process.

S-14

11/24/09

388

4:23 PM

Page 388

CASE STUDIES

The second decisive event that opened the way to peace was the FMLN’s November 1989 offensive, which was a military failure but a political success. The guerrillas launched coordinated attacks upon the country’s key military bases and brought 2,000 combatants into the capital, convincing significant sectors within the economic elite that a negotiated solution might be necessary after all.20 At the same time, the guerrillas’ inability to win through a massive insurrection was laid bare, allowing guerrilla commanders who had previously supported a negotiated settlement to convince hard-liners a negotiated settlement was necessary. During the offensive, six Jesuit priests and two colleagues were killed in their residences at the prestigious Central American University in San Salvador. Speculation that a military unit was responsible would eventually be confirmed. The killing shocked the nation and eventually elicited a suspension of U.S. military aid conditioned upon both sides making progress in peace talks. Within weeks of the offensive, FMLN representatives asked the United Nations to widen its role, and in early 1989 the presidents of various Central American countries asked the United Nations to mediate a series of talks between the two sides.21 After a series of diplomatic moves, including a joint letter supporting the UN’s mediation in the conflict signed by James Baker, then U.S. secretary of state, and Eduard Scheverdnaze, Soviet foreign minister, UN Secretary-General Javier Perez de Cuellar announced on April 4, 1990, that the United Nations would act as mediator.22 Alvaro de Soto assumed the role of mediator on behalf of the Secretary-General. At government insistence, and despite a long-standing FMLN demand of joining a provisional government before agreeing to a permanent ceasefire, the negotiations took place in the context of the 1983 constitution, which established formal democratic rights and a series of multiparty elections that, in contrast to the authoritarian regime of 1948–1979, were not dominated by the military’s party and candidates. Thus, the negotiation and the implementation of the accords, especially the transition of the FMLN to civilian life and into a political party, took place in anticipation of the constitutionally slated elections of March 1994. Because presidential, legislative, and municipal elections coincided during this cycle, these “founding” elections of El Salvador’s postwar democracy came to be referred to as the “elections of the century.” The negotiated settlement involved a series of agreements between the government and the FMLN. The first important accord was the Human Rights Agreement, signed in July 1990, in which the parties recognized a wide range of internationally guaranteed human rights and established a UN verification mission on human rights issues. Despite concerns about opening a human rights verification mission before a cease-fire had been agreed upon, the UN opened its Observer Mission in El Salvador to verify the human rights agreements one year after the agreement was signed.23

S-14

11/24/09

4:23 PM

Page 389

EL SALVADOR’S TRANSITION FROM CIVIL WAR TO PEACE

389

Known by its Spanish acronym ONUSAL, the mission commenced in July 1991, with a budget of $23 million and a staff of 101 persons.24 The United Nations played a crucial role in reaching peace. De Soto repeatedly proposed solutions to impasses during the two-year negotiations, achieving recognition by the parties and among outside observers for his initiative. In addition, Undersecretary-General Marrack Goulding deftly mediated discussions over the complex details of the cease-fire and military aspects of the accords. After a series of other preliminary peace agreements, including one that paved the way for constitutional reforms in April 1991, the final Chapultepec Accord was signed on January 16, 1992.

The Peace Agreement The peace accord in El Salvador was remarkable for its level of detail and specific timelines. The final accord fills 100 pages in book form, including nine chapters and two series of annexes. Over a hundred deadlines were set with precise reference to the date of the cease-fire, known as “D-Day” (e.g., the deadline for the legalization of the FMLN as a political party was “D+90”). Although radical change of the economic structure of the country was a central FMLN objective throughout the war, only about 10 percent of the final accords were dedicated to social and economic issues. Instead, the majority of the text was dedicated to reforms of the military and the police forces and to the details of the military cease-fire. The FMLN viewed reduction of the military as key to its own safe reentry into civilian life, and to its ability to exercise power through electoral democracy. Cease-Fire and Demobilization A formal cease-fire would begin on February 1, 1992, two weeks after the signing of the accords. Within five days, both sides would fall back from positions, and within thirty days the FMLN and all its arms, munitions, and personnel, including clandestine units, would be concentrated in fifteen sites. The armed forces would be restricted to sixty-two specific barracks or sites except under special circumstances. After the concentration of forces, 20 percent of the FMLN would reintegrate into civilian life in specified, periodic intervals of several weeks, culminating in the end of demobilization and disarmament on October 31, 1992. Military and Intelligence Reforms Significant military reductions were stipulated. The military would abdicate all internal security functions except in exceptional circumstances. The three security forces—the Treasury Police, the rural-oriented National

S-14

11/24/09

4:23 PM

Page 390

390

CASE STUDIES

Guard, and the urban-oriented National Police—would be dissolved. Paramilitary civil defense patrols were disbanded and legally barred. The army counterinsurgency units were dissolved, forced recruitment was suspended, and civilian input and human rights issues were integrated into military doctrine. The military-controlled National Intelligence Directorate was replaced by a State Intelligence Office run by a civilian reporting directly to the president and supervised by the legislature. Police Reforms The accords created a new National Civilian Police (PNC) outside the Ministry of Defense, to be the sole national-level public security force. The PNC had responsibilities for maintaining order and protecting citizens, a doctrine that explicitly emphasized human rights and a civilian leadership. The legislature could remove the PNC director for human rights abuses committed by the force. The two sides eventually agreed that at least 60 percent of both the officer-level and the basic agent personnel of the PNC would be “civilian” applicants, that is, persons not having served as combatants during the war. No more than 20 percent of the PNC could be former members of the old National Police; and no more than 20 percent could be former guerrillas. Applicants were required to pass an entrance exam administered by the new civilian-run National Public Security Academy (ANSP), responsible for recruitment, selection, initial training, and inservice training. The accords included a draft law for the structure and doctrine of the PNC and a draft law for the new academy. Human Rights and Judicial Issues The agreement created the Ad Hoc Commission to review the files of military officers and to recommend names to be purged; the UN-supervised Truth Commission to prepare a report assigning responsibility for the most egregious human rights violations of the war; and the Human Rights Advocate’s Office, headed by a mediator chosen by the legislature and independent of the executive, to receive complaints and investigate and report on human rights violations. Judicial reforms, less extensive than the reforms to coercive forces, included increasing the autonomy of the National Judicial Council from the powerful and politicized Supreme Court. Six percent of the national budget was set aside for the judicial branch. Socioeconomic Issues The accords established mechanisms to improve the quantity and the legal status of lands in the hands of poor farmers, especially those in the FMLN-

S-14

11/24/09

4:23 PM

Page 391

EL SALVADOR’S TRANSITION FROM CIVIL WAR TO PEACE

391

controlled zones. The government agreed not to dislodge those in guerrillacontrolled zones from lands they had occupied. Procedures for landownership dispute and credit and technical assistance were established. A National Reconstruction Plan was agreed to in principle, including resources for demobilized combatants and those civilians most affected by the war. The text of the accords regarding social and economic issues was considerably more vague than that regarding security and justice issues. A number of issues were left to a tripartite Social and Economic Forum composed of government, business, and union/peasant representatives who would submit draft laws to the legislature.

The Implementers and Verifiers of the Accords The accords stipulated that a National Commission for the Consolidation of Peace (COPAZ), comprising two representatives each from the government and the FMLN, plus one representative from each political party, would be “responsible for overseeing the implementation of all the political agreements reached by the Parties.”25 COPAZ’s powers included drafting legislation for the Legislative Assembly. The parties to the accords were bound to its findings. In accordance with the agreements, COPAZ was formally constituted on February 1, 1992. However, the accords were explicit in denying COPAZ any executive powers, “since it is the [responsibility of the] Parties, through their internal machinery, to carry out the peace agreements.”26 With the signing of the final accord, ONUSAL’s mandate and observation tasks were expanded “to include the verification and monitoring of the implementation of all the agreements” that the two parties were to sign (SCR 729). ONUSAL’s existing personnel were largely transformed into a Human Rights Division, and a new Military Division and Police Division were added. This marked the first time divisions formally dedicated to human rights and to police observation constituted part of a UN peacekeeping operation.27 An Electoral Division was added in 1993 before the March 1994 elections, and dismantled by May 1994. The first ONUSAL chief of mission and the special representative of the Secretary-General (SRSG) was Iqbal Riza, who served until March 1993, and who was succeeded by Colombian diplomat Augusto Ramírez-Ocampo, who served until April 1994, when he was succeeded by ONUSAL’s last chief of mission, Venezuelan diplomat Enrique ter Horst.28 Although UN mediator Alvaro de Soto was not part of ONUSAL, he contributed through continued oversight of El Salvador from his subsequent posts within the United Nations. Facing pressures to downsize from UN member states in order to reduce peacekeeping obligations and budgets, many senior UN officials worried that closing ONUSAL would jeopardize the progress attained in

S-14

11/24/09

392

4:23 PM

Page 392

CASE STUDIES

implementing the accords. With the victory of the governing ARENA party in the presidential and legislative elections of 1994, some in President Armando Calderón Sol’s new administration and in conservative sectors were eager to end the UN’s verification functions. A compromise yielded the closure of ONUSAL as an operation of the Department of Peacekeeping Operations, but a downsized mission would continue under the Department of Political Affairs. Once again, the UN’s operations in El Salvador were pioneering, ushering in a practice of replacing a formal peacekeeping operation with a downsized mission under a new name. The UN Mission in El Salvador (MINUSAL) operated from May 1, 1995, through April 30, 1996, when a small UN Verification Office (ONUV), which operated until December 31, 1996, replaced it. ONUV consisted of only eight staff members plus three civilian police consultants, and its mandate was “to follow up implementation of pending aspects of the peace accords in El Salvador.” Relative to other peacekeeping operations, ONUSAL utilized a relatively low number of military observers. The number reached 15 in late 1991, and jumped to 295 by March 1992, just after the cease-fire occurred, staying above 200 through 1992, dropping to 101 by March 1993, and gradually dropping from 46 in June 1993 to 3 during the last months of the ONUSAL operation.29 Police observers totaled 27 in late 1991, and their number jumped to 279 within three months, staying between 270 and a maximum of 327 between 1992 and the elections of March 1994, after which their number dropped to 147 by September 1994 and to 32 from December 1994 onward.30 The United Nations adopted a strategy during the peace talks of working with a group of four Friends of the process who weighed in with the two sides on crucial matters. Rather than relying upon a relationship with the regional intergovernmental organization, the Organization of American States (OAS), the United Nations depended upon an informal relationship with states that held influence with one or both sides, including Mexico, which had long hosted the FMLN High Command, Venezuela, Colombia, and Spain.31 The United States later joined the group, which became known as the “Four plus One.” Although the Friends did not always agree during the negotiations and implementation phases, they served principally as a means of pressuring the two sides to stick with the process and make necessary concessions.32 During the last-minute negotiations, for example, the four Friends’ diplomats ran back and forth from the parties’ hotels to the UN headquarters to urge progress.33 Although perhaps not explicitly planned, the roles played by senior UN diplomats Alvaro de Soto and Marrack Goulding facilitated implementation. De Soto, having served as mediator, was perhaps the individual most identified with the achievement of a peace agreement. However, having another individual serve as the on-site chief of mission permitted that person to dedicate himself to verification tasks and did not raise the specter of

S-14

11/24/09

4:23 PM

Page 393

EL SALVADOR’S TRANSITION FROM CIVIL WAR TO PEACE

393

constant negotiations, which de Soto’s continued presence might have done were he kept on as the SRSG. But equally important, de Soto and Goulding did not disappear from the implementation process. In their capacity as senior diplomats whose official responsibilities included reporting and advising the Secretary-General on El Salvador, the two kept the Security Council’s attention focused upon the Salvadoran peace process and signaled the Salvadoran government and the FMLN that the United Nations would not take implementation for granted. Furthermore, the periodic “parachuting” into San Salvador of these senior diplomats, who knew the process and the actors intimately, proved crucial to unblocking the implementation process when it was stagnant.34

Implementation of the Accords Disarmament, Demobilization, and Reintegration The first focus of implementation concerned the concentration of the military forces. Within weeks of the February 1, 1992, cease-fire, the implementation was threatened.35 Although the two sides met the first deadline of pulling back to specified locations, the full concentration of government armed forces did not occur within thirty days (by March 2). The government, which faced dozens of deadlines within the first weeks following “DDay,” encountered resistance from the military in demobilizing the National Guard and the Treasury Police, and the FMLN cited the government’s lack of compliance and infrastructure difficulties in fulfilling its pledges.36 Further serious problems concerned land, as the government forcibly evicted FMLN supporters who had occupied land after the ceasefire.37 In addition, in April, the ARENA-dominated legislature passed a law that appeared to extend the National Guard and Treasury Police under new names. The FMLN viewed this as a violation of the agreed dissolution of the two bodies, and ONUSAL agreed. Both sides, once again, turned to the United Nations as mediator. UN envoy Marrack Goulding, who had originally helped negotiate the ceasefire provisions, convinced the two parties to cease activities that most threatened the accords, especially regarding land invasions. ONUSAL also provided a frank assessment of the state of implementation of the accords, which apparently embarrassed the government into taking its commitments more seriously, increased donors’ pressure for compliance, and demonstrated the moral and political influence of the UN’s verification judgments.38 The first of periodic “recalendarization” agreements was signed in June 1992 and within days both sides had largely complied with the provisions. The government had pledged to rescind the law regarding the old security forces and to commence training for new PNC recruits, and to submit legis-

S-14

11/24/09

394

4:23 PM

Page 394

CASE STUDIES

lation for the legalization of the FMLN as a political party.39 According to one ONUSAL official, the UN mission’s high-profile presence “made it difficult for the two sides to intensify the fighting or to walk away from the negotiating table,” and neither side wanted to be perceived as the “spoiler.” 40 The United Nations would continue to serve as occasional mediator, communications channel, and interpreter of the accords, as well as institution-builder and verifier, throughout ONUSAL’s existence. Because the FMLN had fewer commitments under the accords—essentially the concentration, demobilization, and disarmament of its forces—it delayed scheduled demobilizations when it perceived a lack of compliance by the government. Under the cumulative effect of the delays, a second recalendarization occurred in August 1992, which resulted in the demobilization of the second fifth of FMLN troops in September (almost five months after the original deadline), demobilization of the third and fourth contingents according to the newly agreed upon schedule (by October 31 and November 20, respectively), and delayed compliance with handing over arms. The government then launched the National Public Security Academy for the PNC in early September, dissolved the old National Guard and Treasury Police, and legalized the FMLN as a political party—all sticking points for the FMLN. The lack of specificity regarding land issues and the rigid and unrealistic deadlines in the accords stalled the establishment of land tenure, as well as the allocation of land beneficiaries and the purchase and distribution of legal titles by June 1992. With technical assistance from the International Monetary Fund (IMF), the World Bank, and the Food and Agriculture Organization (FAO), ONUSAL successfully prepared and proposed a plan in October 1992 to grant landownership to 15,000 former government soldiers, 7,500 former guerrilla combatants, and 25,000 civilians in conflict zones.41 On December 17, 1992, two months after the original deadline, the FMLN demobilized the last of its 12,362 combatants, officially bringing to a close the armed conflict.42 Within eight months the FMLN had turned in its arms; however, the accidental explosion of an FMLN arms cache in Nicaragua on May 23, 1993, revealed that the FMLN had concealed an enormous amount of arms and munitions from ONUSAL. This transgression became the most serious violation by either side during the entire implementation process.43 An embarrassed FMLN revealed over 120 additional caches containing tons of matériel.44 A law prohibiting the legal registration of any “armed” group as a political party prompted talk of denying the FMLN participation in the crucial March 1994 elections. ONUSAL, citing the FMLN’s insistence that it had no plans to return to war and that no FMLN units were actively armed, argued that denying the FMLN participation in the 1994 elections would place the entire peace process in grave danger. The FMLN’s legality was not challenged, and the peace process continued, but the FMLN’s leverage was weakened, undermining its ability

S-14

11/24/09

4:23 PM

Page 395

EL SALVADOR’S TRANSITION FROM CIVIL WAR TO PEACE

395

to challenge government violations of the accords regarding the military and police.45 In addition, the incident damaged the FMLN’s public image during the preelectoral period, and potential electoral candidates of the FMLN who were associated with the arms cache and its cover-up were subsequently considered too damaged to run. The armed forces generally complied with their force-reduction commitments, which were less specific than the wholesale dissolution of the FMLN military structure. In the New York Accords, the government agreed to reduce its armed forces to a size appropriate for its new doctrine and functions. The parties generally understood this to be a reduction by roughly half, but the military leadership was reluctant to agree to a specific number and instead agreed to submit a reduction plan. The plan called for a 50.2 percent reduction of overall forces, including the dissolution of five rapid-reaction infantry battalions known for their counterinsurgency work, between February 1, 1992, and January 1994.46 In fact, the armed forces adopted an accelerated schedule, demobilizing fifteen battalions in January 1993 alone, completing a 54.4 percent reduction by March 31, 1993. The civil defense units organized throughout the country were also dismantled. Citing the FMLN’s refusal to demobilize on time in late 1992, the military delayed dissolution of the rapid-reaction forces for several weeks, complying once the FMLN resumed demobilization of its troops in late 1992.47 Total ranks of the armed forces dropped from a claimed 63,170 soldiers in early 1992 to some 30,500 by 1994, although the military’s size per capita remained the largest in Central America. One additional factor in the armed forces’ agreement to downsize and its continued compliance with the accords was the role of U.S. military assistance in the early 1990s. As late as 1990, annual U.S. military assistance totaled $81 million, but by 1993 it had been reduced to $12 million, and would decline to $1.2 million for 1996.48 The peace accords were not specific about how reconstruction would proceed, and lacked specific plans for assisting the reintegration of former combatants. The government unilaterally developed a five-year National Reconstruction Plan, which included training and aid programs for reintegrating former guerrillas and former soldiers, separate from the land transfers arranged for former combatants.49 Short-term programs included the provision of farming tools and inputs, basic furniture and equipment for homes, and vocational training in agriculture, small industry, and services.50 Long-term programs ranged from credit for rural farming and small business, to university scholarships. But these programs proved inadequate, reaching only 11,000 of the estimated 42,000 former combatants from both sides. More extensive resources channeled through NGOs in zones of FMLN influence left them and civilian sympathizers relatively better off.51 Land transfers to former combatants and civilians in FMLN-controlled zones were an important part of the reintegration plans. The process was

S-14

11/24/09

4:23 PM

Page 396

396

CASE STUDIES

complicated by a procedure in which most beneficiaries had to sign up for land plots as groups of families.52 The number of estimated beneficiaries was reduced to 35,782, of whom only 8,000 were former government soldiers or police.53 Problems ranged from administrative “nightmares” at the Land Bank to inaccurate original lists to poor government outreach. ONUV reported in December 1996 that 98.93 percent (34,413) of registered beneficiaries had received title to their land, 85 percent of whom had been recorded in what proved to be an arduous process for most. The average amount of land per beneficiary was 4.01 manzanas (6.82 acres), costing about $2,948 each. Verification of land transfers proved to be among the most technically exigent and longest-lasting tasks for the United Nations, remaining a priority throughout MINUSAL and ONUV. Discontented former combatants proved to be one of the most serious problems in the implementation of the accords. The agreements provided that demobilized members of the military and security forces would receive one year’s salary as severance pay and receive preference in land-transfer programs.54 However, for reasons that appeared to be partly financial, the government failed to pay former soldiers promptly and interpreted the beneficiaries in a narrow manner, excluding civil defense patrol members, soldiers discharged for bad conduct (whose number grew as the conflict declined), and the large number of troops discharged before the signing of the peace accords.55 Over two years after the accords were signed, only 6,000 of 18,000 former soldiers had received their severance pay.56 Soon after the accords were signed, disgruntled former soldiers, sometimes allied with former guerrillas, organized themselves to press the government for compliance with its commitments and for expansion of the pool of beneficiaries. Military Reforms The downsizing of its military forces was among the very few commitments the government carried out regarding the armed forces without encountering resistance. Unlike the FMLN, the government did not agree to dismantle its military forces; therefore, the institutional reforms to the military were considered extremely important by the guerrillas and other groups as a guarantee not only for the FMLN’s return to civilian life but also for the possibility of full democratization. Most military leaders, not having been defeated on the battlefield, continued to view the FMLN as an illegitimate force and to view with scorn any intrusion by civilians into military affairs. Several also were skeptical of the FMLN’s true commitment to disarm. Consequently, the armed forces proved an important obstacle in government compliance with its commitments. Perhaps the most serious stumbling blocks in the implementation of the entire peace agreement concerned the work of the Ad Hoc and Truth

S-14

11/24/09

4:23 PM

Page 397

EL SALVADOR’S TRANSITION FROM CIVIL WAR TO PEACE

397

Commissions, established to foster accountability for past human rights violations. Because of the Ad Hoc Commission’s composition and its daunting task of reviewing 2,200 officer files in four months, few expected that its work would have much effect.57 In the end, an unprecedented purging of the senior officer corps resulted, including almost the entire high command of the armed forces, sparking a serious challenge to the accords from the military leadership. The report of the Ad Hoc Commission, which was kept confidential, recommended the removal of 102 officers, including the minister and vice minister of defense, most of the generals, and many colonels.58 The sweeping purge of the high command recommended by the Ad Hoc Commission took most observers by surprise. Senior officers alleged that the report was a leftist plot to decapitate the military.59 Although bound to comply with the commission’s findings, the government initially announced it would put off action on the Ad Hoc Commission’s findings pending FMLN demobilization. In January 1993, after UN mediation secured an agreement to proceed, President Cristiani again reneged, telling UN Secretary-General Boutros Boutros-Ghali that eight senior officers, including the minister and vice minister of defense, would be kept on through the end of Cristiani’s term in 1994, and that seven others named in the report would be transferred.60 The FMLN was slow to criticize the government’s intractability, leaving the United Nations to demand compliance. Eventually the subsequent report of the Truth Commission would secure the ouster of all the armed forces officers listed in the Ad Hoc Commission’s report. The Truth Commission’s report identified by name the military officers and other persons responsible for the worst human rights violations during the war. Most damaging was the report describing the role of the defense minister, Rene Emilio Ponce, the vice defense minister, General Juan Orlando Zepeda, and virtually the entire high command in ordering the widely publicized 1989 massacre of the Jesuit priests and their assistants. Within two weeks of the Truth Commission’s report, President Cristiani reported that Generals Ponce and Zepeda would be removed by June 1993, and on June 30, 1993, they and the rest of the military high command retired. The government’s eventual compliance was influenced by pressure from ONUSAL, as well as from the Clinton administration, which withheld $11 million in U.S. military aid contingent upon the government abiding by the Ad Hoc Commission’s report. Although the retirement of the officers named in the Ad Hoc Commission’s report represented a violation of the report’s recommendations, the consequences of the Ad Hoc and Truth Commissions’ reports signaled the most thorough housecleaning ever carried out of a Latin American military not defeated in war. Despite the progress in eventual implementation of the Ad Hoc Commission’s recommendations, other elements of military reforms were

S-14

11/24/09

4:23 PM

Page 398

398

CASE STUDIES

formally adopted, but less clearly and completely enacted. As agreed, the military’s constitutional mission was reduced to the “defense of the sovereignty of the State and the integrity of the territory,” conferring upon the president the power under exceptional circumstances to use the armed forces to keep internal peace (Article 212). The armed forces also published their new doctrine in newsprint media in accordance with the accords. The doctrine explicitly stated that the armed forces shall carry out their mission in accordance with the rule of law and with respect for human rights, and stipulated that internal security activities fall outside the normal mission of the armed forces.61 The doctrine made the delineation between internal security and external defense among the strongest in Central America. Nevertheless, the military retained many of its prior roles and its sense of institutional autonomy. In July 1993, President Cristiani called out the armed forces to patrol the highways as a means to curb rising violent crime, including highway robberies.62 Additional internal security roles followed, including patrols of the year-end coffee harvest in 1993 and subsequent years, of polling places during the 1994 elections, and of rural development projects known as “civic actions,” as well as joint patrols with the new police force in which military personnel greatly outnumber police officers.63 The implementation of the provisions regarding military reforms presents a deeply ambivalent picture overall. On the one hand, remarkable progress has been achieved. For the first time in history, a Latin American military submitted its officer corps to external review and vetting.64 As of 1998, the army was roughly the same size as the National Civilian Police, and its missions and doctrine reflected significant emphasis on classic external defense functions and respect for human rights and for civilian control. Its worst human rights violators were purged, its budget was reduced, and new levels of accountability and civilian input were reached. As one opposition figure maintained, the armed forces have generally opted not to challenge their postwar place in society.65 On the other hand, the armed forces continue to carry out internal security functions and bristle at the notion of civilian oversight or input. Interviews with officers reveal that most reject both legislative oversight of defense matters and the notion of a civilian defense minister.66 In general, the military’s political power has been greatly reduced, but its autonomy remains very high.67 It remains to be seen whether a crisis among civilian political actors might undermine the accomplishments of the peace processes. Intelligence Reforms Control over the gathering and processing of intelligence on internal political matters was a critical component of the military’s social control and repressive functions during and before the war. As part of the early wran-

S-14

11/24/09

4:23 PM

Page 399

EL SALVADOR’S TRANSITION FROM CIVIL WAR TO PEACE

399

gling over compliance with demobilization, the government failed to dismantle the military-controlled National Intelligence Directorate (DNI) by its March 1992 deadline, but did so after the first recalendarization in June, appointing a director of the new State Intelligence Office (OIE) only three days beforehand. However, the accords said nothing about the armed forces’ intelligence section (C-2), and several DNI agents were transferred there. Of more concern to some observers, the bulk of the OIE personnel were simply transferred from the DNI with no public information on selection or vetting procedures.68 In general, ONUSAL and the FMLN proved reluctant to press publicly for a more rigorous selection process or more transparency in the country’s new intelligence system. Some OIE agents were implicated in human rights violations after mid1992, and the lack of effective oversight of the OIE, under the direction of a controversial ARENA political appointee named Mauricio Sandoval, increasingly worried international and national human rights observers. In addition, the accords did not address the training of intelligence operatives, and the military continued as of 1997 to operate the National Intelligence School, effectively expanding its influence over civilian as well as military recruits to the OIE and the military’s C-2. In an apparent violation of the accords criticized by the United Nations but never remedied, the Defense Ministry’s C-2 intelligence activities continued to focus upon internal intelligence activities.69 Reforms to the Police and Public Security System The creation of a new National Civilian Police with a doctrine of protecting citizen rights rather than state interests was crucial for consolidating peace and democracy in El Salvador. Public security reforms sought to curb the armed forces’ internal security role; to create unprecedented external mechanisms of accountability for human rights abuses; and to refashion the relationship between the state’s coercive bodies and its citizens. They opened the way for FMLN participation in an armed state organization to protect its members’ physical safety and to prevent misuse of the powerful and newly centralized internal security force for political ends. Government negotiator David Escobar Galindo called the PNC the most significant (“transcendental”) institution to come out of the reforms, and UN Secretary-General Boutros Boutros-Ghali hailed the new police force in 1995 as “one of the fundamental elements of the peace accords and perhaps the single component with greatest hopes.”70 The most immediate public security priority in the implementation phase was the provision of security in the face of the reduction of forces. The combining of government troops and FMLN forces, as well as the demobilization of the Treasury Police, the National Guard, and the civil defense patrols, effectively cut the coercive forces available for deployment

S-14

11/24/09

400

4:23 PM

Page 400

CASE STUDIES

from some 60,000 to 6,000 in only a few weeks.71 The accords provided that during a roughly two-year transition period, the old National Police would provide security and be demobilized as the new PNC was gradually deployed. Delays occurred in the scheduled events because the new public safety institutions became a prominent and persistent locus of political contention among various social and institutional actors. The military not only saw the new police force as a threat to its ability to continue corrupt behavior, but also worried that former guerrillas inside the PNC might use the new police force itself to destabilize the state.72 The government and the economic elite, while not entirely trusting the integrity of the old security forces, had relied upon them for decades to protect elite interests and were reluctant to dissolve them before the PNC was fully deployed and able to control crime. In addition, death squads from the old regime undoubtedly sought to infiltrate the PNC with officers who might protect them.73 The first difficulty was the government’s failure in March 1992 to demobilize the Treasury Police and the National Guard. Even as that issue was under discussion, however, the military impeded the establishment of the renamed National Public Safety Academy (ANSP) by taking over the old Public Security Academy installations that had been contemplated for the new ANSP. When a military base belonging to the Treasury Ministry was selected for the base-level training center, the military stripped it clean as it withdrew, taking even the light bulbs and doorknobs.74 Once the June 1992 recalendarization secured the dissolution of the old security forces and the training for the first senior officer class commenced, ONUSAL learned that the government had transferred more than 1,000 former security forces personnel into the National Police as a means of gaining them “back-door” entrance into the ANSP. Furthermore, almost half of the former government component of the first class of senior officer candidates sent abroad for training came from the other security forces, in direct violation of the accords. An agreement was worked out with UN assistance by which those already present could remain, but no further such cases could occur. The process of appointing a director for the new PNC showed the first signs of problems to plague the operation of the multiparty COPAZ. COPAZ was slated to submit to President Cristiani three nominees for the post of PNC director. However, the number of parties allied with the government’s two representatives equaled the number of those allied with the FMLN, resulting in deadlock and delay in this process.75 This parity would be repeated in other arenas such as reconstruction, seriously undermining the role COPAZ was assigned in the accords. Although COPAZ served as an important official mechanism for the FMLN to obtain information about and criticize problems in the implementation of the accords, the inability of COPAZ to reach decisions allowed the government to delay in carrying out its responsibilities and augmented ONUSAL’s verification role.

S-14

11/24/09

4:23 PM

Page 401

EL SALVADOR’S TRANSITION FROM CIVIL WAR TO PEACE

401

Perhaps the most important challenges to the integrity of the new civilian police occurred in 1993. In June 1993, taking advantage of the embarrassment caused by the explosion of the FMLN’s arms cache, the government appointed Captain Oscar Peña Durán, a military officer who led the U.S.-backed antidrug unit of the old regime, to become operations director of the PNC.76 Peña Durán appointed many of his own military-trained officers from the Anti-Narcotics Executive Unit to key posts in the PNC, ignored requirements that all personnel pass through the ANSP to become trained in the new citizen-oriented doctrine, and encouraged spying by former soldiers within the PNC on their colleagues who came from the guerrillas’ ranks.77 Peña Durán also ignored and then discontinued an agreement by which ONUSAL’s Police Division provided field advice and mentoring to freshly deployed PNC agents. The FMLN accepted Peña Durán’s appointment in exchange for benefits for its former combatants in the landtransfer and reintegration programs.78 ONUSAL persisted in opposing the appointment as a violation of the peace accords. When opposition to Peña Durán emerged from Salvadoran conservatives and from the previously supportive U.S. embassy, he was forced to resign.79 In the meantime, the PNC had forgone field mentoring from experienced international police officers at a crucial time, just before the United Nations began downsizing its police presence. A second challenge to the integrity of the PNC occurred in 1993 when two units, the Special Investigative Unit (SIU) and the Anti-Narcotics Executive Unit (UEA), were transferred intact into the PNC structure. In a move it later regretted, the FMLN agreed in the December 1992 recalendarization to permit the transfer of these units after screening and retraining at the ANSP. The screening was perfunctory, and the retraining was delayed for months and then years, as the government resisted implementing these provisions. During Peña Durán’s tenure, when the units were transferred, human rights violations by the PNC increased markedly, and former UEA agents were implicated in many of the abuses. Moreover, some thirteen of the hundred SIU agents transferred to the PNC were eventually implicated in politically motivated murders or in their cover-up.80 After tedious negotiations involving the United Nations and the two parties, provisions for compliance with the accords were reached regarding the members of these two units. Offended by the requirements, however, many members of the units participated in a hunger strike that eventually resulted in the mass resignations of most of the UEA and of many SIU investigators. In the end, some members remained in the new Criminal Investigative Division and the Anti-Narcotics Division of the PNC. The PNC’s creation and deployment, one of the central institutionbuilding tasks stipulated in the accords, not only was fraught with active resistance from the military and other sectors, but also suffered from logistical setbacks and funding shortfalls. The ANSP scrambled to meet its obli-

S-14

11/24/09

402

4:23 PM

Page 402

CASE STUDIES

gation to produce 5,700 new recruits within two years, but did so nonetheless. The PNC experienced a serious lack of equipment upon its first deployments.81 The specialized divisions of the PNC, such as white-collar crimes, anti-kidnapping, and traffic control, were slow to develop. International funders, more willing to fund less controversial social and economic projects than the training and equipping of coercive forces, failed to comply with commitments to fund the development of the ANSP and the PNC. 82 This delay played into the hands of governmental and military actors who wanted to cling to the old security forces. In 1994, however, a videotaped bank heist apparently involving a top National Police officer led the newly elected government of President Calderón Sol to speed up dissolution of the old National Police, leaving internal security exclusively in the hands of the PNC and those army troops called out to assist. Due in part to the security gap created at the end of the war, levels of violent crime spiraled and became the top concern of the population by 1993,83 and remained so through 1997, when 95 percent ranked it as “very serious.”84 In 1995, deaths by homicide exceeded the average annual number of deaths during the twelve-year civil war, and by 1996 the country’s homicide rate reached 138 per 100,000, the second highest in the world.85 The crime wave created incentives to speed up PNC training and deployment, to relax selection standards, to rely upon the army for internal security, and to weaken laws aimed at protecting suspects’ rights—all of which posed threats to the achievements of the peace process. As of 2000, these tensions had not been resolved, as crime remained high and debate persisted over the role of the police. Nine years after the accords were signed, the PNC was more humane and accountable than the old security forces. In contrast to the past, PNC officers accused of torture, killings, vigilante activities, and excessive use of force were subject to internal sanctions and to judicial prosecution. The PNC was a more transparent and civilian-controlled institution than any of its predecessors, though it was rife with many of the same faults. Corruption was widespread, fueled by inconsistent and weak internal disciplinary procedures and by a government decision to sacrifice selection standards in order to attract and deploy more police agents in 1996 and 1997.86 While Salvadorans tended to fear PNC agents less than the old security forces personnel, they also did not fully trust or respect the new police officers. Furthermore, evidence persisted that organized crime had penetrated upper-level ranks of police officers, despite a leadership that was not perceived as corrupt itself. A high-profile purge by the PNC leadership in 2000–2001 resulted in the ouster of over 1,200 police personnel. Human Rights and Judicial Reforms Although human rights violations have been features of almost all internal armed conflicts, El Salvador’s civil war was unusual in the degree to which

S-14

11/24/09

4:23 PM

Page 403

EL SALVADOR’S TRANSITION FROM CIVIL WAR TO PEACE

403

human rights were a prominent element of the national discourse surrounding the war and of international debate about the war. The U.S. Congress made military assistance contingent upon a semiannual “human rights certification” process, and even low-level soldiers of both sides were aware that their conduct was under scrutiny. 87 Consequently, provisions both to address past human rights abuses and to institutionalize future protections were central to the agreements. Accountability for past abuses. The Ad Hoc Commission was the first major element of the implementation process to address past human rights violations, followed by the Truth Commission. Because the armed forces adamantly opposed any provision by which they might be tried for human rights violations committed during the war, the mandate of neither the Ad Hoc Commission nor the Truth Commission included judicial powers of indictment, prosecution, or trial, although the information gleaned from the latter’s investigations could, in principle, be used by authorities to bring charges. The Truth Commission’s mandate was to investigate “grave acts of violence” since 1980.88 To carry out its investigation within the six-month time frame allocated by the accords, the commission chose some thirty-two especially egregious or well-known human rights violations as exemplary cases.89 These cases included the better-known abuses attributed to the military or army-linked death squads, such as the 1981 massacre at El Mozote, the 1989 murder of six Jesuit priests and their two colleagues, and the 1980 assassination of the country’s archbishop, Oscar Romero, as well as some violations attributed to the FMLN. Although it did not hold hearings, the commission received testimony from over 2,000 people. The Truth Commission’s report, released March 15, 1993, had a tremendous impact on national and international opinion. Its findings affected the military institution, the reputations and electoral possibilities of leaders from both sides of the war, and the pace and depth of reforms to the institutions of human rights and justice. In addition to describing the nature of death-squad organizations, the report named sixty-two military officers, six FMLN leaders, and several civilians for committing or covering up the country’s most serious wartime human rights violations.90 The report estimated that the military and paramilitary death squads were responsible for 95 percent of all human rights abuses committed between 1980 and 1992, with the FMLN responsible for the remainder. As described earlier, the report effectively ended the military careers of those named within it. Anticipating that El Salvador’s weak judicial system would not be capable of prosecuting those named in its report, the Truth Commission recommended instead that offenders be barred from holding public office for at least ten years, with a permanent ban on holding defense or internal security posts. The commission surprised the government by including a

S-14

11/24/09

404

4:23 PM

Page 404

CASE STUDIES

series of recommendations for reforms to the judicial system that went beyond what the original accords included. These recommendations concentrated on further measures to decentralize and depoliticize the court system, as well as a call for the resignation of the entire sitting Supreme Court. Citing the old SIU’s involvement in covering up the Jesuit murders and other crimes, the report also called for dissolution of that unit. Although both sides had agreed to abide by the Truth Commission’s findings, the government quickly made clear that it would not do so. Members of the government’s negotiating team criticized the commission’s work as biased and exceeding its original mandate.91 The chief justice of the Supreme Court, ARENA-appointed Mauricio Gutiérrez Castro, defiantly announced that the peace accords were an agreement of the executive branch and did not touch the judiciary, and that “only God” could remove him from his post. In an official press conference on behalf of the armed forces high command (charged in the report with murdering priests), General Ponce blasted the report as “unjust, incomplete, illegal, unethical, partial and insolent.”92 Within days, the ARENA-dominated Legislative Assembly overrode the opposition parties to pass an amnesty covering all crimes. It was the first law related to the peace process that had not been passed by consensus.93 The only two military officers ever convicted in court of a human rights violation, those who supervised the murder of the Jesuit priests, were subsequently released from jail. The government and the FMLN eventually decided to formally ignore the suggested ban on holding public office, and some sought and won seats in the Legislative Assembly. Nevertheless the stain of the report damaged the electoral possibilities of most of those named therein, such as the FMLN’s best-known military commander, Joaquín Villalobos of the ERP. Some weaknesses of the Truth Commission complicated its reception among the Salvadoran left and international human rights groups. On the one hand, international human rights organizations felt that the report confirmed everything they had been saying for years, revealing the disingenuousness of denials by the Reagan administration and the Salvadoran government about the armed forces’ role in the war’s most notorious human rights violations. On the other hand, despite the tremendous effort by qualified personnel, the commission’s work was constrained by a six-month lifespan and limited information, and its quality suffered. Its analysis of death squads, which went beyond its investigation of exemplary cases, was challenged by some human rights experts as incomplete.94 Its criteria for selection of exemplary cases were questioned, as it omitted hundreds of extrajudicial executions and kidnappings committed by the FMLN in the early years of the war. In addition, the Truth Commission was not constituted in such a way as to encourage confessions by perpetrators. Consequently, it perversely punished those who came forward and acknowledged their role. Such was the case with the leadership of the ERP, whose acknowledgment

S-14

11/24/09

4:23 PM

Page 405

EL SALVADOR’S TRANSITION FROM CIVIL WAR TO PEACE

405

of its role in planning the kidnappings and murder of eleven mayors during the war meant that it was the only one of the FMLN’s five organizations named in the report. Although the other FMLN groups committed human rights abuses, their lack of cooperation in investigations seemed to be rewarded with the omission of their names. The report was initially hailed by the FMLN, but its findings deepened divisions within the organization. Furthermore, both human rights groups and the FMLN tended to mute their criticisms, believing that doing so might undermine the moral force of the remainder of the Truth Commission’s report, whose findings represented perhaps the most important public instrument of truth, accountability, and an end to impunity ever to emerge in El Salvador. Human rights violations and ONUSAL’s role. Beginning in mid1993, ONUSAL and other observers became alarmed at an increase in the number of apparently politically motivated killings, which especially targeted FMLN leaders.95 Between August 1 and November 15, ONUSAL recorded forty-seven cases in which political motivations may have been present, although many of these cases had the appearance of common crimes. 96 Several factors may have contributed to this trend, including right-wing disgruntlement over the military purges, the publication of the Truth Commission’s report, and the approach of the March 1994 elections. Once again, the United Nations took measures to redress the problem. On December 8, 1993, Secretary-General Boutros Boutros-Ghali announced the formation of a special “Joint Group for the Investigation of Politically Motivated Illegal Armed Groups,” constituted jointly by the United Nations and the Salvadoran government. Recognizing that the Truth Commission had been an international effort that did not contribute to institution-building within the Salvadoran state, UN officials sought to create ownership of this new effort and to lay the groundwork for judicial prosecutions by sharing direction and staffing of the Joint Group with the Salvadoran government. The creation of the Joint Group, suggested by the Truth Commission, also sought to redress the criticism that the latter’s report did not go far enough in exposing the nature of the death squads behind many of the war’s atrocities. The Joint Group comprised the director of ONUSAL’s Human Rights Division, the government Human Rights Advocate, and two lawyers named by President Cristiani. The six-month mandate of the Joint Group was to investigate the activities of all illegal armed groups since the 1992 accords. The Joint Group’s report, released in July 1994, contained details of the nature of death squads during and before the war, their financing and ties to the military, and the ways in which their operations, financing, and relationship to the state had changed with the peace process. Its “confidential annex” contained the names of those alleged to be organizers, funders, and operators of the death squads. Ultimately, however, little action was

S-14

11/24/09

406

4:23 PM

Page 406

CASE STUDIES

taken against those named in the report, and its use in judicial proceedings was practically nil. Instead, its main contribution seemed to be the stemming of the tide of high-profile political killings in the preelectoral period. ONUSAL marked an improvement in the human rights situation in the early months of 1994, unusual for an immediate preelectoral period, and a continuing decline in the four months after the March 1994 elections.97 Building institutions for human rights protections. The provisions of El Salvador’s peace agreement sought not only to account for past human rights abuses, but also to construct new institutional guarantees to prevent future human rights violations. These included doctrinal and training reforms in military and police institutions, and the creation of inspectors general to oversee the state’s coercive forces. In addition, the accords ended the fuero militar, or military jurisdiction, under which all military and police personnel were subjected to military courts for any crime or misdemeanor. The accords also provided for the creation of external oversight for the protection of human rights, and for judicial reforms to strengthen accountability, nonpartisanship, and efficiency. The most important of these new mechanisms for human rights protections was the National Human Rights Advocate’s Office. The Advocate’s Office was created as a governmental office, largely independent of the executive branch, to receive and investigate complaints, issue reports, and make nonbinding recommendations of administrative sanctions against offenders employed by the state. The Advocate’s findings could also be forwarded to the appropriate authorities for judicial proceedings. Although the Advocate was placed administratively under the executive-branch Public Ministry, the law stipulated that the Advocate would be selected by a two-thirds majority of the legislature for a renewable three-year term. The supermajority requirement was designed to induce multiparty support for the Advocate’s selection. The Advocate’s Office, created by law in 1991 but not filled until just after the accords were signed, got off to a slow start. The government had little interest in developing the office, and its work was weak due to severe underfunding, low technical capacity, and lack of a clear sense of mission.98 Much of its efforts were spent setting up regional offices. Without international funding from the Dutch, Swedish, Norwegian, and Finnish governments administered through the UN Development Programme (UNDP), the office would have been moribund. And technical assistance from ONUSAL’s Human Rights Division, especially the practice of jointly taking initial complaints, provided training and mentoring that proved indispensable. As ONUSAL began in 1994 to anticipate its own closing, its Human Rights Division became increasingly cognizant of the need to leave behind a strong national capacity to carry out the work of human rights protection.

S-14

11/24/09

4:23 PM

Page 407

EL SALVADOR’S TRANSITION FROM CIVIL WAR TO PEACE

407

ONUSAL had signed an agreement with the Advocate in July 1993 to collaborate in institution-building; however, joint verification did not commence until 1994. ONUSAL’s downsizing to MINUSAL in April 1995 coincided with the election of a new Human Rights Advocate, Victoria de Aviles, under whose guidance the office would become much more active. The Advocate’s Office initially assumed ONUSAL’s tasks of receiving and processing complaints at the departmental and regional levels, and eventually throughout the country. De Aviles increased the professionalism of the office’s work and issued detailed annual reports that became the country’s most authoritative source on human rights violations. The office also began investigating complaints against police, judicial, and military personnel, and actively visited communities to receive complaints.99 During de Aviles’s tenure, her personal profile and that of the Advocate’s Office increased precipitously. Public opinion polls in 1995 and 1998 showed that the office was regarded as the institution in the country that most contributed to protecting human rights, more than the police, courts, or even nongovernmental human rights organizations.100 Indeed, the migration of nongovernmental human rights personnel and international resources to the Advocate’s Office between 1995 and 1998 created new challenges for human rights NGOs whose work had been somewhat taken up by the office. Unfortunately, controversy over de Aviles’s successor and other problems weakened the office in 1998. Judicial reforms. The judicial reforms contained in the accords were agreed upon in a hurried fashion during the final stages of negotiations, as reflected in their lack of specificity. The highly politicized and conservative Supreme Court resisted the decentralizing and depoliticizing thrust of the accords and the Truth Commission’s recommendations. Little change occurred in the judicial system until after the new government and Legislative Assembly took office in mid-1994. After much wrangling, a new Supreme Court was elected in conformity with the process contained in the constitutional amendments passed as part of the peace process. Civil society, namely the Bar Association, participated in the nomination and ranking of possible Supreme Court magistrates, and the Legislative Assembly approved its choices by a two-thirds majority to staggered terms of nine years, rather than the previous terms that coincided with the electoral cycle. The resulting Supreme Court was decidedly more professional, more politically plural, and less partisan, although political party ties continued to enter into the selection process. Despite some important steps toward devolving the power of the Supreme Court to lower courts and to the National Council on the Judiciary, the judicial system remains highly centralized. In particular, the Supreme Court retains a determinant voice in the selection and removal of lower-court judges. In 1994 the National Council on the Judiciary was able

S-14

11/24/09

408

4:23 PM

Page 408

CASE STUDIES

to commence a process of reviewing judges. Investigations launched into judges’ records by a Department of Investigations created by the Supreme Court resulted in a high number of suspensions and other sanctions against lower-court judges. Public opinion polls in 1996 and 1997 showed the judiciary near the bottom of public regard for state institutions, and news editorials reflected a general sense that the pace of housecleaning in the judiciary remained slow in 1997. In 2002 the judicial system remained weak, inefficient, antiquated, overly partisan, and subject to corruption. The vagueness of the accords’ provision regarding the judiciary played a role, as did political wrangling and a virtual stalemate within the legislature over constitutional and legal reforms and judicial procedures. Nevertheless, as the enacting of a new code of criminal procedures in April 1998 indicated, a slow process of modernization, professionalization, and expansion of legal protections is under way. Elections and Electoral Reforms Because the negotiations and the peace process occurred under the 1983 constitution—albeit with important amendments—the electoral cycle was nonnegotiable and thus presented a constraint on the timelines and deadlines of the peace process. When serious delays occurred, the possibility of the FMLN’s competing in the March 1994 elections, and thus the entire peace process, were jeopardized. Having already demobilized its troops and turned over most of its arms by the beginning of the campaign season in late 1993, the FMLN had little leverage to ensure its ability to participate in elections except international pressure and the government’s concerns about its own legitimacy. Despite concerns about violence, the elections occurred without any serious incidents. However, the process of registering to vote, long a problem in El Salvador’s elections, became a significant weakness of the entire electoral period. ONUSAL and the UNDP conducted a study that found that some 786,000 persons, or 27 percent of the eligible voting population, were not registered in August 1993.101 An arduous application process for a voter registration card was complicated by a weak Supreme Electoral Tribunal (TSE) that rejected 74,000 applicants for reasons that generally reflect a deficient process rather than deficient or cheating voters. Separate calculations by an ONUSAL observer and by an Argentine polling firm both concluded that some 6 percent of those with voter cards who tried to vote were unable to do so.102 To their credit, ONUSAL and the international community focused a good deal of attention on the preelectoral registration and campaign period, recognizing the fallacy of a technically “free and fair” process on election

S-14

11/24/09

4:23 PM

Page 409

EL SALVADOR’S TRANSITION FROM CIVIL WAR TO PEACE

409

day if the voter lists were deficient or the campaign were conducted under unsuitable conditions. ONUSAL, whose verification mandate arose not from the accords, but from a government request in January 1993 to observe the elections, formed an Electoral Division in early 1993.103 As the elections neared, ONUSAL was viewed by many observers as carrying the TSE “across the finish line” of the registration and the elections processes.104 In the end, the election results were uncontested, with ARENA’s Armando Calderón Sol winning the presidency after a run-off against the candidate of the FMLN-Democratic Convergence coalition, Rubén Zamora, who had been a leader of the FDR during the war. The problems with the 1994 electoral process deepened pressure for serious electoral reforms. An official multiparty Vigilance Board, formed in 1993 to oversee the performance of the TSE, reached consensus on a number of reforms, and the United States and the United Nations also suggested reforms. After some delay, President Calderón Sol formed a presidential commission, which included the losing opposition candidate, Rubén Zamora. The commission made a number of recommendations, including issuing a single document that would serve as both an official identity card and a voter identity card; establishing voting by more decentralized neighborhoods to reduce the travel distance required to cast a vote; injecting nonpartisan, merit-based performance criteria into the TSE staff, rather than relying entirely upon party-based appointees to police one another; and adopting proportional representation within municipal council elections.105 Unfortunately, the ARENA-dominated Legislative Assembly did nothing with the reforms, instead approving in late 1996 an electoral law making it more difficult for small parties to remain legal. Tensions between factions of the FMLN, always present during the war, led to a split in the party within months of the 1994 elections. The Popular Revolutionary Army and the National Resistance, which adopted a more “pragmatic” or “centrist” Social Democratic position, broke off to form the Democratic Party (PD). The harder-line elements of the Popular Liberation Forces, the Communist Party, and the Workers Revolutionary Party remained under the FMLN banner. Legislative elections in 1997 showed that the FMLN had retained much greater support than the PD, as the latter performed extremely poorly in those elections. The FMLN did significantly better in 1997 than in 1994, winning control in coalition with other leftist parties of the departmental mayoralties of the capital and the second largest city, as well as forty-six municipal governments in the country. In the 2000 elections the FMLN gained control of the largest cities in the country, despite having again lost the presidency to ARENA in 1999. The elections established ARENA and the FMLN as the country’s two most powerful parties, reflecting a remarkable transformation of the armed conflict to the electoral arena.

S-14

11/24/09

4:23 PM

Page 410

410

CASE STUDIES

Socioeconomic Reforms During the last-minute negotiations regarding social and economic issues, the FMLN failed to obtain specific language regarding reconstruction and development. In contrast to the FMLN’s voice in the appointment of the Ad Hoc Commission and its role in COPAZ, the FMLN gained no formal role in either administering or overseeing reconstruction programs. The UNDP was granted a consultative role, nothing more. Consequently, the government saw the planning and execution of reconstruction as its own responsibility and prerogative. When the accords were signed, the government formed a National Reconstruction Secretariat (SRN) to carry out reconstruction programs. In March 1992 the World Bank organized a Consultative Group meeting of donors for postwar reconstruction in Geneva. At the meeting, the Cristiani government unveiled its five-year reconstruction plan, and donors pledged some $800 million for the first two years. Later the total amount pledged would increase to $943 million, with the U.S. Agency for International Development (USAID) as the single largest donor with $336 million, followed by the Inter-American Development Bank (IDB) with a pledge of $205 million, Japan with $190 million, and the European Economic Community (EEC) with $79 million.106 The National Reconstruction Plan developed by the Salvadoran government included two phases: a short-term phase of six months for urgent needs and a longer-term five-year phase. Its four objectives were the reintegration of former combatants, the improvement of the areas most affected by the war, the reconstruction to infrastructure damaged by the war, and the promotion of participation by all sectors in national reconstruction efforts.107 The areas targeted by the plan were 115 municipalities comprising 1,645,756 inhabitants. Other targeted populations included some 60,000 displaced persons, 24,000 repatriated refugees, and 56,000 demobilized combatants.108 The plan contained three sorts of projects: “investment” projects such as land-transfers for former combatants and civilians, and infrastructure projects; “technical” projects of smaller size generally administered by the UNDP; and strengthening “democratic institutions” such as the Office of the Human Rights Advocate, the ANSP, and the PNC. The entire reconstruction plan was undertaken by a government committed to a neoliberal model, including reductions of state expenditures in certain areas. Although consultations were held, the government granted neither the FMLN nor NGOs a formal voice in the content of the plan. Only pressure from international donors ensured that opposition views were eventually incorporated to a degree. The land-transfer program and reintegration programs were central parts of the reconstruction plan. In contrast to the remainder of the socioeconomic provisions, the land provisions were extremely specific on some

S-14

11/24/09

4:23 PM

Page 411

EL SALVADOR’S TRANSITION FROM CIVIL WAR TO PEACE

411

points, and specificity at times undermined implementation. For example, the accords stated that land should be “voluntarily offered for sale by its owners” and that transactions should occur at “market value.” 109 Other details that would have aided implementation, such as the size of plots, went unspecified. Despite the eventual successful transfer of land to 98.9 percent of the beneficiaries, some 600 persons still faced a need to relocate at the end of 1996, and 14,000 families of combatants who died during service were waiting for the necessary documents to obtain loans.110 It is difficult to measure the effectiveness of the reconstruction programs. Some analyses have concluded that most beneficiaries of the landtransfer program would be unable to pay off their mortgages with the expected rents from farming, and would eventually lose the land.111 In 1996 the government passed a debt relief law that forgave 70 percent of agrarian debt, plus $575 to individual parcel holders, if the debt was paid off within one year.112 The government also passed a second law aimed at breaking up collectively held lands, a characteristic of many of the lands transferred under the accords. Although the remaining 30 percent of debt held by beneficiaries was later paid off with funding from USAID, beneficiaries remained burdened with debt from loans for production. The impact of other aspects of reconstruction is even less readily assessed. Most of the reconstruction financing went to infrastructure investments, whose results are diffuse and long-term. GDP growth in the years after the war has been relatively strong: over 5 percent from 1993 through 1995. However, the government’s neoliberal policies have elicited foreign investment, and the global and regional dynamic of growth undoubtedly played a role. Moreover, the continued economic impact of remittances from Salvadorans working in the United States has made a huge difference in economic performance. In 1993 and 1994, Salvadorans in the United States sent home over $1 billion in 1993 and 1994 each, triple the amount of coffee export earnings and the equivalent of $200 per Salvadoran—this in a country where $1,300 is the per capita GDP.113 At the same time, El Salvador ranks among the lowest Latin American countries on the UNDP’s Human Development Index and maintains a high concentration of wealth, a situation that the peace process did not significantly alter. One of the factors in ARENA’s loss of support in the 1997 elections was the perception that economic growth was not reaching the majority of the populace. The hopes placed in the Social and Economic Forum were largely misplaced, as the group proved unable to negotiate further changes in the country’s economic system that would favor more equitable outcomes. As Alvaro de Soto and Graciana del Castillo have pointed out, the absence of coordination between the United Nations and the international financial institutions (IFIs) led to elements of the international system working at odds with one another.114 The World Bank, the International

S-14

11/24/09

412

4:23 PM

Page 412

CASE STUDIES

Monetary Fund, the Inter-American Bank, and even the UN Development Programme were not part of peace talks regarding social and economic reconstruction. Consequently, IFI policies of downsizing the state coincided with a concentrated need for reconstruction assistance and an expansion of health, education, and other social services to rebuild the country in the wake of war. The UNDP, for example, suffered for this oversight when it was forced to develop reintegration programs for demobilizing combatants practically overnight. Subsequent missions have sought to redress these deficiencies with improved, though still suboptimal, results.

Conclusion: Assessing the Consolidation of Peace and Democracy It might be an exaggeration to say that in 2002 the process of peace is “consolidated” in El Salvador. However, the war has definitively ended. Any new armed conflict might draw upon the remnants of the prior conflict— disgruntled former combatants, old political-military networks, or frustrated peasants who have lost their land—but it would be a different conflict, with different causes, actors, and ideologies. Such a conflict is unlikely at this writing. Some Salvadorans say that peace was never achieved because of the extraordinary levels of violent crime that plagued the country after the war. These problems with public safety and justice, which have not abated even with the full deployment of the PNC, remain perhaps the most serious after-effects of the conflict and of the implementation process. High levels of poverty and unemployment are also cited as sources of discontent in public opinion polls.115 The peace accord did not contain provisions sufficient to redress the structural inequalities within the economy that helped cause the war. However, charges that the country’s skewed postconflict economy is due to faulty implementation of the peace accord seem misplaced.116 Had the FMLN, which fought for twelve years to transform the socioeconomic structure of the country, insisted upon such changes as part of the negotiations, it is doubtful that a peace agreement would have been possible. Although the country’s poor and unequal economic conditions may contain the seeds for future armed conflict, such conditions must be viewed in a historical context and, especially today, in an extremely constraining international context for very small, dependent states. The degree of consolidation of El Salvador’s democratic institutions is more difficult to assess than its level of peace. The military retains significant autonomy, but its political influence has been curtailed. The country’s electoral regime is well established, but not fully free or fair. While no election results have been contested since the war ended, such an outcome is

S-14

11/24/09

4:23 PM

Page 413

EL SALVADOR’S TRANSITION FROM CIVIL WAR TO PEACE

413

likely if needed reforms are not adopted. Human rights problems persist, but are much more likely to be complaints about undue process and arbitrary detentions than murder, “disappearance,” or torture. Common crime has replaced political crime as the main internal security concern, and state forces have not been exempt from corruption and abuses related to the fight against crime. Despite oversight and accountability, deep flaws will likely characterize newly reformed civilian security and judicial institutions for the foreseeable future. In the face of crime and unsatisfactory economic performance, a majority of Salvadorans polled in 1996 backed the need for a “strongman” to resolve the country’s problems. The social problems generated by blatant inequities pose a continuing threat to the institutionalization of participatory mechanisms. Perhaps most pressingly, if the country’s economic and growing environmental problems are left unaddressed, the consequences for quality of life and for popular tolerance of any reigning government or regime might be jeopardized. El Salvador’s peace process has been hailed as a model policymakers have utilized in other peacekeeping settings.117 The unusually high level of detail in the agreements proved much more help than hindrance in the implementation process, as evident in the implementation of provisions regarding the cease-fire, military reform and purges, and public security. But this level of detail is not simply a technical operation or the result of the UN mediator’s efforts, although the latter clearly contributed. I have argued here that more significant were the conditions surrounding the war and the two parties: the balance of political-military forces, their degree of cohesion, and the unity of the particular international actors involved in supporting a negotiated settlement. Without a strategic stalemate and the very real power of the FMLN to disrupt the economy, to control territory, and to hold the armed forces in check, the peace accord would never have been so detailed. Nor would the FMLN have achieved so influential a role for itself and for the United Nations in the implementation process. The end of the Cold War was a fundamental facilitating condition for ending the conflict, especially by permitting the reversal of U.S. opposition to a negotiated settlement. The ability of the FMLN and, less surprisingly, the armed forces, to retain their unity in the face of the very predictable tensions between hard-liners and soft-liners also conditioned the detailed accords. Even when the FMLN split in mid-1994, after the bulk of the peace accord’s provisions had been implemented, the level of detail in the accords kept the process on track. At the same time, the United Nations was vital to both reaching and implementing the peace agreements. It did this through its various roles as mediator, verifier, and institution-builder. These roles at times called for contradictory actions on the part of ONUSAL, creating dilemmas for its staff. For example, some UN staff believed that ONUSAL was too lax in its

S-14

11/24/09

4:23 PM

414

Page 414

CASE STUDIES

verification responsibilities regarding certain appointments or actions involving the PNC.118 Others believed that the UN’s institution-building role required not joining what seemed to be a campaign of criticism from conservative sectors designed to undermine the development and image of the new police force.119 These dilemmas were difficult to manage, but overall ONUSAL balanced its multiple roles well. William Stanley and David Holiday have argued that despite the trade-offs presented by trying to wear more than one hat simultaneously, the UN’s multiple roles had a mutually reinforcing effect, giving it greater leverage to carry out each function individually.120 They also point out that the United Nations contributed to the overall implementation process by tending to support implementation of all the peace accord’s provisions, even when the two parties themselves renegotiated them in ways that undermined the spirit of the original accords. By adopting a principled position, ONUSAL eventually won greater compliance with the original accords than might have occurred had it simply agreed to the rewritten provisions. Several analysts have drawn insightful and useful lessons from the Salvadoran peace process. These have touched upon the way in which a UN mission is initiated, the interrelationship among the UN’s various roles from peacemaking to peacebuilding, the relationship of the United Nations to powerful regional actors such as the United States and Mexico, the internal structure of the UN mission, and the choices made by the UN staff at crucial junctures.121 What seems important to keep in mind is that the agency of third-party implementers and verifiers of peace agreements is constrained by the agreements reached, and that these agreements and the implementation process are highly contingent upon the balance of forces, the relative positions of global and regional powers, and the internal characteristics of the parties to the armed conflict. Extraordinary efforts and wise choices by actors inside and outside a conflict always carry the hope of peace. They should not be neglected. In El Salvador favorable circumstances combined with such efforts to produce generally positive results. But where such circumstances are absent at a particular historical moment, peace may simply be impossible. Or, sadly, peace may entail such injustice as to be unwise.

Notes 1. See Krishna Kumar and Marina Ottaway, “From Bullets to Ballots: A Summary of Findings from Six Post-Conflict Election Studies,” paper presented at the Conference on Promoting Democracy, Human Rights, and Reintegration in Post-Conflict Societies, sponsored by USAID, Washington, D.C., October 1997. 2. See Phillip J. Williams and Knut Walter, Militarization and Demilitarization in El Salvador’s Transition to Democracy (Pittsburgh: University of Pittsburgh Press, 1997); Joseph S. Tulchin with Gary Bland, Is There a

S-14

11/24/09

4:23 PM

Page 415

EL SALVADOR’S TRANSITION FROM CIVIL WAR TO PEACE

415

Transition to Democracy in El Salvador? (Washington, D.C.: Woodrow Wilson Center, 1992); and Elisabeth J. Wood, Forging Democracy from Below: Insurgent Transitions in South Africa and El Salvador (Cambridge: Cambridge University Press, 2000). 3. David Holiday and William Stanley, “Building the Peace: Preliminary Lessons from El Salvador,” Journal of International Affairs 46, no. 2 (winter 1993); and Tricia Juhn, “A Conspiracy of Peace: Civil-Military Relations and the Salvadoran Peace Negotiations” (Ph.D. diss., University of Miami, 1996). 4. Cristiani’s characterization, not a direct quote, is found in Hector Dada, “Procesos de negociación comparados: El caso de El Salvador,” in Rafael Vergara at al., Procesos de negociación comparados en Africa y América Latina (Guatemala: Facultad Latinoamericana de Ciencias Sociales, 1994), p. 30. Dada, a prominent Salvadoran social scientist and an opposition figure, affirmed his agreement with the summary. 5. William Stanley, The Protection Racket State: Elite Politics, Military Extortion, and Civil War in El Salvador (Philadelphia: Temple University Press, 1996); and Tommie Sue Montgomery, Revolution in El Salvador: From Civil Strife to Civil Peace, 2nd ed. (Boulder, Colo.: Westview Press, 1995). 6. See Stanley, The Protection Racket State. 7. Michael McClintock, The American Connection, vol. 1, State Terror and Popular Resistance in El Salvador (London: Zed Books 1985), pp. 259–260. 8. For the most authoritative account of the principal human rights violations of the war, see UN Commission on the Truth, De la locura a la esperanza: La guerra de 12 anos en El Salvador (San José, Costa Rica: Editorial DEI, 1993). 9. The Christian Democratic Party (PDC), whose alliance with the military from mid-1980 through 1987 would make them the dominant political party during that period, was the only major centrist or leftist political force to remain outside the FDR. However, the resignations of several prominent Christian Democrats from both the government and the party in March 1980 provoked a mass exodus of the most left-leaning party rank-and-file to parties associated with the FDR, leaving behind a more conservative PDC led by new junta member José Napoleón Duarte. 10. Dada, “Procesos de negociación comparados,” p.32. 11. Montgomery, Revolution in El Salvador. 12. James Dunkerley, The Pacification of Central America: Political Change in the Isthmus, 1987–1993 (New York: Verso, 1994), app. 7. 13. Terry Lynn Karl, “El Salvador’s Negotiated Revolution,” Foreign Affairs 71, no. 2 (spring 1992): 150. 14. Dunkerley, The Pacification of Central America, app. 7. 15. Mario Lungo Ucles, El Salvador in the Eighties: Counterinsurgency and Revolution (Philadelphia: Temple University Press, 1996). 16. Lungo Ucles, El Salvador in the Eighties, p. 194. 17. Personal correspondence from Adela Quesada, last director of the Salvadoran office of the UN High Commission for Refugees (1995–1998), February 2, 1999. 18. The term is from Karl, “El Salvador’s Negotiated Revolution,” p. 149, and reflects widely accepted analysis of the conditions that made peace possible in El Salvador. 19. Ibid., p. 150. 20. See Stanley, The Protection Racket State, p. 246, and author interviews with Luis Cardenal, San Salvador, October 1995, and with Héctor Vidal, president of the National Association of Private Enterprise (ANEP), San Salvador, November 1995.

S-14

11/24/09

416

4:23 PM

Page 416

CASE STUDIES

21. Karl, “El Salvador’s Negotiated Revolution,” p. 154. 22. This letter came in response to a series of suggestions made by Perez de Cuellar to the two superpowers about how to revive languishing talks. See Teresa Whitfield, “The Role of the United Nations in El Salvador and Guatemala,” in Cynthia Arnson and Carlos Basombrío, eds., Comparative Peace Processes in Latin America (Stanford: Stanford University Press, 1999), pp. 257–290. 23. A small, preliminary presence was established in January 1991. According to the agreement, “The purpose of the Mission shall be to investigate the human rights situation in El Salvador as regards acts committed or situations existing from the date of its establishment and to take any steps it deems appropriate to promote and defend such rights.” Official translation of para. 13 of the San José Accord, from UN Document S/24375, “Report of the Director of the Human Rights Division,” August 12, 1992, cited in Holiday and Stanley, “Building the Peace,” p. 419. 24. Americas Watch, El Salvador’s Decade of Terror (New Haven: Yale University Press, 1991). 25. This mandate is from Section 1.1 of the New York Accords, signed on September 15, 1991, the contents of which were explicitly reaffirmed in the Chapultepec Accord. Two of the main contributions of the accords reached in New York were agreement upon the establishment of COPAZ as a mechanism to provide the FMLN participation in the implementation phase, and the adoption of what was termed “compressed negotiations,” whereby all unresolved issues would be negotiated simultaneously and prior to a cease-fire (see “Negociaciones Comprimidas” agreement, September 25, 1991). The agreement upon “compressed negotiations” represented the government’s abandonment of its long-standing demand that the rebels lay down arms as a condition of negotiating a full peace agreement, and was viewed by Alvaro de Soto as crucial in the ability of the two sides to reach tradeoffs and agreements on the remaining unresolved issues. Author interview, New York, October 1996. 26. Ibid., Section 1.4, “Powers” of COPAZ. 27. Montgomery, Revolution in El Salvador. 28. UN, The United Nations and El Salvador, Blue Book Series no. 4 (New York: United Nations, 1995). Brigadier General Victor Suanzes Pardo, chief of the Military Division at the time, served as interim chief of mission from March 6, 1993, until Ramírez-Ocampo arrived on April 14, 1993. UN Document S/25812, May 21, 1993. 29. All numbers of military and police observers are from Annexes 1 and 2 of UN Document S/1995/220, “The Secretary General’s Report to the Security Council,” March 24, 1995, issued on the eve of ONUSAL’s transformation to MINUSAL. 30. Ibid. 31. Mark Levine, “Peacemaking in El Salvador,” in Michael W. Doyle, Ian Johnstone, and Robert C. Orr, eds., Keeping the Peace: Multidimensional UN Operations in Cambodia and El Salvador (Cambridge: Cambridge University Press, 1997), pp. 227–254; and Whitfield, “The Role of the United Nations in El Salvador and Guatemala.” 32. Ibid. 33. Juhn, “A Conspiracy of Peace”; and Levine, “Peacemaking in El Salvador.” 34. Author interview with Alvaro de Soto, New York, October 1996. 35. The problems cited in this paragraph are taken from the first report of the UN Secretary-General to the Security Council after the cease-fire, UN Document S/23999, May 26, 1992.

S-14

11/24/09

4:23 PM

Page 417

EL SALVADOR’S TRANSITION FROM CIVIL WAR TO PEACE

417

36. The chief of the Military Division, who formed part of a three-person working group along with one representative from each side, had already made the decision on the geographic limits of the FMLN’s fifteen concentration sites when the two sides could not agree. See ibid., paras. 7–8. 37. Holiday and Stanley, “Building the Peace,” p. 421. 38. Ibid., p. 422. 39. See Addendum to the Secretary-General’s May 26, 1992, report on ONUSAL, dated June 19, 1992. 40. Anonymous quote cited in Holiday and Stanley, “Building the Peace,” p. 422. 41. UN Document S/24833, November 25, 1992, paras. 53–60. 42. Montgomery, Revolution in El Salvador, p. 227. 43. The discovery of the weapons caches prompted a special report from the UN Secretary-General to the Security Council terming the incident a “grave violation,” the only time this serious a phrase was used by the United Nations. See UN Document S/26005, June 29, 1993. 44. UN Document S/26790, November 23, 1993. 45. Author interview with two members of ONUSAL staff, San Salvador, July 1993 and January 1996. The most pertinent government violations were its lack of compliance with the recommendations of the Ad Hoc Commission and its appointment of a former military officer to run the operations of the PNC. 46. UN Document S/26790, para. 16. 47. Williams and Walter, Militarization and Demilitarization in El Salvador’s Transition to Democracy. 48. Jack Spence, George Vickers, Margaret Popkin, Philip Williams, and Kevin Murray, A Negotiated Revolution? A Two-Year Progress Report on the Salvadoran Peace Accords (Cambridge, Mass.: Hemisphere Initiatives, March 1994), p. 15. 49. The information in the paragraph is from Montgomery, Revolution in El Salvador, pp. 231–232. 50. Ibid., p. 232. 51. Ibid., p. 233. 52. For good explanations of these complex procedures, see Jack Spence et al., Chapultepec Five Years Later (Cambridge, Mass.: Hemisphere Initiatives, January 16, 1997). 53. See reports by the Secretary-General, UN Documents S/1995/220, March 24, 1995, and A/51/693, November 25, 1996. Other information in this paragraph is from Spence et al., Chapultepec, pp. 36–37, unless otherwise noted. 54. Williams and Walter, Militarization and Demilitarization in El Salvador’s Transition to Democracy, p. 180. 55. Ibid. 56. UN Document S/1994/561, May 11, 1994, para. 16. 57. Holiday and Stanley, “Building the Peace,” p. 425. 58. Ibid. 59. Ibid. 60. UN Document S/25078, January 1993. 61. For a convenient compilation of pertinent legal texts and doctrine of the post-accord armed forces, see ONUSAL, Relaciones civiles-militares en el nuevo marco internacional: Fuerza armada de El Salvador, ed. by Colonel Prudencio Garcia (San Salvador: ONUSAL, Human Rights Division, November 1994). 62. Despite fears that the reinitiated internal security role for the military would open the door to political involvements, to police actions such as arrests, or

S-14

11/24/09

418

4:23 PM

Page 418

CASE STUDIES

to excessive use of force, the military’s “dissuasive” patrols provoked no such problems and seemed to deter some highway crimes. See Charles T. Call, “From Soldiers to Cops: ‘War Transitions’ and the Demilitarization of Policing in Latin America and the Caribbean” (Ph.D. diss., Stanford University, August 1999). 63. William Stanley, Protectors or Perpetrators: The Institutional Crisis of the Salvadoran Civilian Police (Washington, D.C.: Washington Office on Latin America [WOLA] and Hemisphere Initiatives, January 1996); and Call, “From Soldiers to Cops.” 64. Williams and Walter, Militarization and Demilitarization in El Salvador’s Transition to Democracy. 65. Author interview with Rubén Zamora, San Salvador, January 1996. 66. Spence et al., A Negotiated Revolution? p. 16. 67. Williams and Walter, Militarization and Demilitarization in El Salvador’s Transition to Democracy, p. 186. 68. Ibid., p. 161. 69. ONUSAL, Ninth Report of the Human Rights Division (San Salvador: ONUSAL, Human Rights Division, August–October 1993), p. 25. 70. UN Secretary-General Boutros Boutros-Ghali in report to the General Assembly, UN Document A/50/517, October 6, 1995, para. 7. 71. Call, “From Soldiers to Cops.” 72. Ibid. 73. Author interview with an inspector within the PNC Control Unit, San Salvador, February 1996; anonymity requested. 74. William Stanley, Risking Failure: The Problems and Promise of the New Civilian Police in El Salvador (Washington, D.C.: WOLA and Hemisphere Initiatives, August 1993). 75. Ibid. 76. Author interview with U.S. ambassador Alan Flanigan, San Salvador, September 1995. 77. Charles Call, Recent Setbacks in the Police Transition: El Salvador Peace Plan Update No. 3 (Washington, D.C.: WOLA, 1994). 78. Author interviews with FMLN commanders, San Salvador, fall 1995, and with government negotiator Colonel Mauricio Vargas, San Salvador, December 1995. 79. Author interviews with U.S. ambassador Alan Flanigan, with three other U.S. embassy staff, and with UN officials and Rodrigo Avila, then deputy director, San Salvador, summer 1993, November 1993, and fall 1995, respectively. 80. Author interview with MINUSAL and ONUV staff, San Salvador, fall 1995 and February 1996. 81. Call, “From Soldiers to Cops”; and Stanley, Risking Failure. 82. Stanley, Risking Failure. 83. Call, “From Soldiers to Cops.” 84. See polling by the Instituto Universitario de Opinion Publica (IUDOP) from 1993 through 1997. The 95 percent figure is from a January 1997 CID-Gallup poll cited in George Vickers, “Renegotiating Internal Security: The Lessons of Central America,” paper presented at conference, Promoting Democracy, Human Rights and Reintegration in Post-Conflict Societies, USAID, Washington, D.C., October 1997. 85. See World Bank study cited in Los Angeles Times, September 6, 1997, p. 2. See also José Miguel Cruz, Alvaro Trigueros Argüello, and Francisco González, El crimen violento en El Salvador: Factores sociales y económicos asociados (San Salvador: IUDOP, 2000).

S-14

11/24/09

4:23 PM

Page 419

EL SALVADOR’S TRANSITION FROM CIVIL WAR TO PEACE

419

86. Call, “From Soldiers to Cops.” 87. In a personal conversation, one civil defense patrol leader in La Paz spoke to me with venomous anger about the “human rights shit” that he was forced to hear from the government and military leaders. Conversation with Corporal Oscar Barrera Trejos, Candelaria, February 5, 1988. 88. The mandate of the Truth Commission is contained in the Mexico Accords, signed in April 1991. 89. For some discussion of the approach to the Truth Commission’s work, see the introduction to UN Commission on the Truth, De la locura a la esperanza. 90. Ibid. 91. Spence et al., A Negotiated Revolution? p. 6. 92. Montgomery, Revolution in El Salvador, p. 243. 93. Ibid. 94. See Lawyers Committee for Human Rights, El Salvador’s Negotiated Revolution: Prospects for Legal Reform (New York: Lawyers Committee for Human Rights, 1993). 95. On October 23, 1993, Francisco Velis, a candidate for the legislature, was shot and killed as he dropped off his daughter at daycare in a manner clearly reminiscent of old death-squad-style killings. A few days later, Eleno Hernan Castro, a well-known regional commander, was shot after an apparent traffic altercation. The landowner who was implicated fled into hiding, and it remained unclear whether the killing was politically motivated. Attacks on two other FMLN leaders, one of whom died, occurred over the next two months. See Spence et al., A Negotiated Revolution? p. 4. 96. Ibid. 97. See early paragraphs of the tenth and eleventh reports of ONUSAL’s Human Rights Division, covering November 1993–February 1994 and March–June 1994 respectively. 98. This paragraph is based upon Laura O’Shaughnessy, Michael Dodson, and Donald Jackson, “Political Will and Public Trust: El Salvador’s Procurator for the Defense of Human Rights and the Dilemmas of Institution Building,” Human Rights Review 2, no. 3 (2000). 99. Ibid. 100. Instituto Universitario de Opinion Publica, “Los Salvadorenos opinan sobre los derechos humanos y la gestion de la procuraduria,” Estudios Centroamericanos (ECA) 594 (April 1998): 357–375; and “Sondeo sobre la procuraduria para la defensa de los derechos humanos,” Serie de Informes no. 50 (1995), both of San Salvador: UCA. 101. Ian Johnstone, Rights and Reconciliation in El Salvador (Boulder, Colo.: Lynne Rienner, 1995), p. 52. 102. Spence et al., A Negotiated Revolution? p. 7. 103. Johnstone, Rights and Reconciliation, p. 51. 104. The phrase is from ONUSAL staff person Madalene O’Donnell, personal conversations, San Salvador, March 1993. 105. Spence et al., Chapultepec, p. 12. 106. Other donors included Germany with $48 million, Canada with $21 million, and others with $67 million. See Kevin Murray et al., Rescuing Reconstruction: The Debate on Post-War Economic Recovery in El Salvador (Cambridge, Mass.: Hemisphere Initiatives, May 1994). 107. Peter Sollis, Reluctant Reforms: The Cristiani Government and the International Community in the Process of Salvadoran Post-War Reconstruction (Washington, D.C.: WOLA, 1993), p. 3.

S-14

11/24/09

420

4:23 PM

Page 420

CASE STUDIES

108. Murray et al., Rescuing Reconstruction, p. 10. 109. Graciana del Castillo, “The Arms-for-Land Deal in El Salvador,” in Doyle, Johnstone, and Orr, Keeping the Peace, p. 346. 110. UN Document A/51/693, November 25, 1996. 111. Jack Spence, George Vickers, and David Dye, The Salvadoran Peace Accords and Democratization (Cambridge, Mass.: Hemisphere Initiatives, March 1995), p. 21. 112. Spence et al., Chapultepec. 113. Ibid., p. 21; and del Castillo, “The Arms-for-Land Deal in El Salvador.” 114. Alvaro de Soto and Graciana del Castillo, “Obstacles to Peacebuilding in El Salvador,” Foreign Policy 94 (spring 1994): 69–83. 115. See, for example, the quarterly polls of 1996–1998 conducted by the Instituto Universitario de Opinion Publica of the Central American University in San Salvador, routinely published in ECA. 116. In their otherwise excellent analysis of the lessons of relatively successful cases of peacekeeping operations, Doyle, Johnstone, and Orr, Keeping the Peace, fault the peace process for El Salvador’s poor socioeconomic condition after the war. 117. Ibid.; and Holiday and Stanley “Building the Peace.” 118. Author interviews with ONUSAL staff, San Salvador, July 1993, November 1993, and fall 1995. 119. William D. Stanley and David Holiday, “Peace Mission Strategy and Domestic Actors: United Nations Mediation, Verification, and Institution-Building in El Salvador,” paper presented at the North-South Center Conference on the United Nations and Peace Processes in Central America, University of Miami, May 1997. 120. Ibid. 121. See, for example, Doyle, Johnstone, and Orr, Keeping the Peace; Stanley and Holiday, “Peace Mission Strategy and Domestic Actors”; Holiday and Stanley, “Building the Peace”; Johnstone, Rights and Reconciliation; de Soto and del Castillo, “Obstacles to Peacebuilding in El Salvador”; Gino Costa, “The United Nations and Reform of the Police in El Salvador,” International Peacekeeping 2, no. 3 (August 1995): 365–390; and Lawyers Committee for Human Rights, El Salvador’s Negotiated Revolution.

S-15

11/24/09

4:24 PM

Page 421

15 Broad Participation, Diffuse Responsibility: Peace Implementation in Guatemala WILLIAM STANLEY AND DAVID HOLIDAY

In December 1996 the government of Guatemala and the Guatemalan National Revolutionary Unity (URNG) signed the final accord of a series of peace agreements designed to put a “firm and lasting” end to an armed confrontation that had continued for thirty-six years. During the conflict an estimated 150,000 people were killed, some 50,000 “disappeared,” tens of thousands fled abroad, and some 1 million were internally displaced.1 Although the guerrillas had been essentially defeated militarily and the conflict had sputtered to low levels of fighting by the late 1980s, the war’s definitive end was nonetheless an important achievement. The war, even at a low level, had been a major obstacle to democratization, since it enhanced the political position of the military and served to justify retention of abusive institutions and practices associated with counterinsurgency. Implementation of the peace accords between January 1997 and this writing in 2002 presents examples of both success and failure. On the positive side, the cease-fire and demobilization of guerrilla forces went virtually without incident. Guerrilla cantonment and disarmament were completed ahead of schedule, and government forces complied with troop and budget reductions. Few cases of political violence were directed at demobilized combatants. The former guerrillas transformed themselves into a political party and competed in elections in 1999. Refugee repatriation was completed, though land and other issues remained unsolved. Yet the mechanics of ending the armed conflict were never the primary challenge in Guatemala. Fighting stopped before the final accord was signed, and the URNG was so weak that there was no serious risk of renewed war. Within Guatemalan civil society, the peace process was perceived as useful in making the country more democratic, equitable, and cul421

S-15

11/24/09

422

4:24 PM

Page 422

CASE STUDIES

turally inclusive. The terms of democratization were the primary issues debated by three consecutive presidential administrations and the URNG for six years. The government and the URNG transformed a negotiated surrender into an internationally supported peace process. The Guatemalan parties sold the international community an image of a war-weary country seeking to excise the roots of conflict through political and social transformation. The accords promised major political, institutional, social, and economic reforms aimed at democratization and greater equity. This chapter evaluates the implementation of the Guatemalan accords against this standard. As a result, we focus on shortcomings of the process that do not threaten the basic success achieved early on—the end of organized armed conflict—but that may well threaten achievement of the democratizing goals laid out in the accords. The peace agreement set forth ambitious but attainable standards of democratization; against these benchmarks, the peace process began stagnating by 1999 and in some areas showed signs of reversal. Many of the reforms in the accords have not been implemented, and overall political stability is questionable. The government committed itself to doubling notoriously low tax revenues as a percentage of GDP by 2000. This was later rescheduled until 2002, but that goal appears unlikely to be met. Without adequate revenue, commitments to increased social spending, support for small-scale agriculture, and state support for multilingualism in the schools and the courts have faltered. Excessive military influence within the state continues to be a problem, as illustrated by the 85 percent increase in the 2001 military budget, which greatly exceeded both initial budget approved by Congress and the military spending limit set by the accords.2 The government initially fulfilled its obligation to create a new civilian police force by “recycling” existing police personnel from the prior militarized, abusive system. In May 1999 voters in a popular referendum rejected complex constitutional reforms that had been attacked by right-wing opponents, halting the implementation of many key promises of the accords. Following the first national elections held since the signing of the peace accords, the conservative Guatemalan Republican Front (FRG) opposition party took control of the presidency and a majority in the legislature in January 2000. FRG leaders had not themselves signed the accords, and despite the new president’s initial pledge of support for the peace accords, there has been little success in delivering on the remaining commitments. For international peace implementers, Guatemala presents both positive and negative lessons. On the positive side, international mediation appears to have been helpful. When the Guatemalan parties agreed to UN mediation, the peace process gained momentum. UN mediator Jean Arnault gained the trust of both parties, and succeeded in translating government proposals into draft language acceptable to the URNG. At the same time,

S-15

11/24/09

4:24 PM

Page 423

PEACE IMPLEMENTATION IN GUATEMALA

423

direct contact between the parties was crucial as well: extensive meetings late in the process between military officers and rebel leaders generated a high degree of confidence on both sides regarding the other’s intentions. Both sides credit these direct contacts with facilitating the rapid and uneventful cantonment and demobilization process. Coordination within the UN system was exceptional in Guatemala, reflecting lessons learned in El Salvador. The Bretton Woods institutions participated during the negotiations phase, and have been supportive of the peace process and its financial requirements. The United States, despite being the neighborhood superpower, acted as just another interested power during much of the process, being supportive but not overbearing. The six Friends—Norway, the United States, Mexico, Venezuela, Spain, and Colombia—were actively supportive during the negotiations, though their role declined once implementation began. On the negative side, the difficulties during implementation derive in large part from a basic problem in the Guatemalan peace equation: both parties that negotiated the accords—the National Advancement Party (PAN) and the URNG guerrillas—were political actors with very limited representational claims.3 The URNG rebels were strategically defeated as early as 1983. Although the URNG managed to survive and operate sporadically, it posed no threat to the survival of the state, and could not pressure the government to make concessions. The PAN government, which had come into office by a tiny margin, continued to confront the traditional weaknesses of civilian governments: a powerful military, a conservative business elite, a fragmented and unstable political party system, and one of the weakest fiscal bases in the hemisphere. Although three consecutive civilian governments negotiated with and made concessions to the URNG, these moves were motivated by a desire to gain international approval and hoped-for investment and development assistance. For the PAN government, which negotiated the last several accords, and which was responsible for implementation from 1997 to 1999, the accords also represented an opportunity to gain international support for changes that could strengthen the Guatemalan state and produce institutions that might generate greater stability in the long run. The partisan interests and weaknesses of the URNG and civilian governments converged to produce a set of accords that were unusually sweeping in their identification of key national issues, but that provided few specific measures. The accords distribute responsibility for implementation so broadly as to dilute the accountability of any one actor. That the URNG demobilized so quickly, thus discarding its (meager) military leverage, meant that implementation depended largely on the intentions and political capacity of the government. The government’s desire for international support has in practice been outweighed by its unwillingness or inability to

S-15

11/24/09

424

4:24 PM

Page 424

CASE STUDIES

incur domestic political costs. When opposition has arisen to such measures as tax increases, the government has retreated. The UN and international donors have so far been unable to muster sufficient leverage to persuade the government to ignore such opposition and fulfill its commitments. The leverage of the UN Verification Mission in Guatemala (MINUGUA) has been further eroded by the fact that it lacks a UN Security Council mandate and instead operates under the General Assembly. While the Department of Political Affairs monitors MINUGUA in the same way as it would a Security Council mission, the lack of Security Council involvement weakens the international diplomatic pressures brought to bear on the Guatemalan government. The Guatemalan parties expected the United Nations to be handholders rather than hard-nosed verifiers, and the United Nations has generally accepted this role. Other features of the Guatemalan process are difficult to assess. Government officials and UN observers lauded the extensive participatory mechanisms called for by the accords.4 Dozens of special commissions were formed to debate and make recommendations on issues ranging from reform of the educational system, to judicial reform, to official multilingualism and multiculturalism. These commissions generated a rich debate and resulted in extensive mobilization of civil society to formulate positions. The increase in participation was particularly noteworthy within the Mayan community. But questions remain regarding the impact of participation through these commissions: some deadlocked because of deep divisions; others produced reports that did not result in legislation or changes in policy; some disbanded entirely. It remains to be seen whether participation in these extra-parliamentary mechanisms, in and of itself, results in greater citizen interest in and enthusiasm for democratic governance or further dilutes political energy. The complexity of the consultation and implementation process diffuses responsibility. The government of Alvaro Arzú asserted that final responsibility rested with Guatemalan society in general, thus deflecting attention away from the executive and legislative branches of government. The multiplicity of actors not only slowed or blocked progress toward specific measures, it also made it difficult to apply pressure for more timely progress toward implementing the accords. The defeat of the constitutional reforms illustrates the vulnerability of such a diffuse process and the inherent difficulty presented to international actors wishing to assist with peace implementation. This chapter has three sections. The first provides historical background on the conflict and the negotiations process. The second section examines peace implementation, focusing on the strengths and weaknesses of the implementation process. In the third section we draw tentative conclusions.

S-15

11/24/09

4:24 PM

Page 425

PEACE IMPLEMENTATION IN GUATEMALA

425

Civil War and Peace Negotiations Military Context for Negotiations Guatemala’s civil war pitted an insurgency that sought to achieve socialist revolution against a state devoted to suppressing the revolutionaries and excluding leftist parties from political competition. Over the thirty-six years of conflict, the makeup of the revolutionary forces changed substantially, as did the nature of the official political regime. The Revolutionary Armed Forces (FAR) were joined years later by the Guerrilla Army of the Poor (EGP) and the Organization of the People in Arms (ORPA). The Communist Party of Guatemala, known as the Guatemalan Workers Party (PGT), also had armed units during some periods of the struggle. In 1982, these diverse organizations formed the Guatemalan National Revolutionary Unity. An initial insurgency by the FAR in the eastern part of the country in the 1960s was suppressed by the government by 1967.5 After a period of relative quiescence, during which the EGP began to develop a political base in indigenous communities in the highlands, active rural insurgency resumed in the late 1970s and early 1980s, mainly in the highland areas and in the remote northeastern department of El Petén. Government forces quickly and ruthlessly suppressed this rebellion, killing tens of thousands of noncombatants and displacing hundreds of thousands of people. By the mid-1980s the war was a prolonged, sputtering conflict in which the URNG managed to survive and periodically attack government forces or economic targets.6 The guerrillas, though resilient, never achieved sufficient military strength to pose an imminent threat to the state. They possessed only small arms, never obtained antiaircraft weapons, and made minimal use of mines. Their equipment was eclectic, and of variable and generally poor condition. By the 1980s, AK-47 and M-16 rifles of various vintages and origins became the most common equipment, reflecting improved logistics, but by this point rebel forces were quite small. While obviously a threat to the individuals and businesses that it attacked, the URNG was little more than a nuisance to the state by the late 1980s. The initial defeat of the guerrillas in the 1960s was partly the result of their narrow civilian support base, and their limited area of operations in the western part of the country. When the EGP and ORPA organized in indigenous areas, this greatly expanded the rebels’ potential support base. From the mid-1970s onward, the revolutionaries formed alliances with, and helped foment, social movements that emphasized the grievances of indigenous Mayan peoples. The relationship between the revolutionaries and Mayan communities was complex. The revolutionary movement never became a Mayan movement, though many Mayan people supported the

S-15

11/24/09

426

4:24 PM

Page 426

CASE STUDIES

revolutionaries at times. Some Mayan individuals, communities, and organizations viewed the URNG suspiciously and the URNG lacked the military capacity to protect the civilian population against government reprisals.7 The government’s military forces adopted a brutal, low-technology approach to counterinsurgency, using ground forces and draconian measures against civilian populations to deprive the guerrillas of support and limit their area of operations. In the 1960s, the military received significant assistance from the United States in organizing special counterinsurgency forces, including irregulars known as “commissioners.” Government and paramilitary forces were involved in mass killing of civilians in the eastern region of Guatemala, particularly in 1966 and 1967.8 Guatemala terminated its military assistance relationship with the United States in 1977, in reaction to President Jimmy Carter’s human rights policies. During the 1980s, the United States provided sporadic assistance under various antiterrorism and anti-narcotics waivers, and Israel provided weapons and technical assistance, particularly in the area of identifying urban safe houses. In the early 1980s, during the peak of guerrilla activity, the governments of Romeo Lucas García and Efraín Ríos Montt massacred tens of thousands of civilians and eliminated some 400 villages, relocating hundreds of thousands of people into new government-controlled villages and forcing additional hundreds of thousands of men to participate in ostensibly voluntary “self-defense” patrols. This approach quickly forced the URNG into a defensive posture. Political Context for Negotiation Until 1985 Guatemala was mostly ruled by authoritarian governments influenced by the military. During an exceptional period from 1944 to 1954, Guatemala enjoyed two consecutive elected governments. The second of these, that of Jacobo Arbenz, was overthrown in a military coup following intense pressure from the United States and an invasion of the country by a small armed force trained and equipped by the United States.9 Military officers governed until 1966, when civilian president Julio César Méndez Montenegro took office under a pact that gave the armed forces carte blanche with respect to internal security matters and an effective veto over governmental policy. From 1970 onward, presidents were military officers, mostly veterans of the counterinsurgency campaigns of the 1960s. Elections were held to create a semblance of electoral competition. This period of institutional military rule became unstable in the early 1980s, and a coup by junior officers in 1982 brought to power retired General Efraín Ríos Montt, who was subsequently replaced in another coup by General Humberto Mejía Victores. As part of a strategy by the military to achieve long-term stability by creating the perception of democratic competition within narrowly circumscribed boundaries, the Mejía govern-

S-15

11/24/09

4:24 PM

Page 427

PEACE IMPLEMENTATION IN GUATEMALA

427

ment initiated a transition to elected civilian rule.10 In 1984 a constituent assembly was elected, followed by the election in 1985 of civilian Vinicio Cerezo as president and subsequent elections for legislative and local offices. This did not end the military’s predominant role in national politics, however. Repeated coup attempts constrained the Cerezo regime.11 The military controlled the information available to the president through the Presidential Staff, which managed the president’s agenda, briefed him on issues, and disbursed presidential discretionary funds.12 The military exercised a de facto veto over most policies, and for a period in the 1980s its reach extended to the local level as part of its counterinsurgency program. Ambiguity regarding the governing authority of civilians and the military continued under the next civilian president, Jorge Serrano Elías, inaugurated in 1990. In 1993, Serrano Elías attempted to seize dictatorial powers following the lead of Peruvian president Alberto Fujimori’s self-coup (autogolpe). The so-called Serranazo, which initially enjoyed the support of sectors of the military, triggered near-unanimous rejection by a remarkable cross section of Guatemalan society. The negative, organized elite response, which included individuals and groups across the political spectrum, coupled with prompt imposition of international sanctions, convinced key sectors of the military to withdraw support for Serrano. After ten days of intense uncertainty, a civilian former human rights ombudsman, Ramiro de León Carpio, was elected by Congress and sworn in as interim president, apparently with the blessing of the military. De León was succeeded by Alvaro Arzú of the center-right PAN, which narrowly defeated the candidate of the right-wing FRG in regularly scheduled elections in November 1996. As might be expected following a prolonged period of military intervention in politics, civilian institutions in Guatemala are extremely weak. No political party is well organized at the national level. The Guatemalan Christian Democracy (DCG) came closest to being a permanent, national party in the 1980s, but was badly hurt by the corruption of many of its prominent members and the disgrace in which the Cerezo government left office. The leading parties, the PAN, the FRG, and the leftist New Guatemala Democratic Front (FDNG), are relative newcomers to electoral competition, and the loyalties of the electorate are very fluid. Earlier parties, including traditional parties of the right such as the National Liberation Movement (MLN), declined in the 1980s and are no longer competitive. The Guatemalan Congress has been notoriously corrupt, divisive, and incapable of producing well-drafted legislation in a timely way. President Serrano’s autogolpe in 1993 was at least partly motivated by his frustration with Congress, where he had no governing coalition. Serrano’s own party was nothing more than a personal vehicle of recent creation and had minimal representation in Congress, so during his brief tenure in office he

S-15

11/24/09

4:24 PM

Page 428

428

CASE STUDIES

depended on vote-buying to pass legislation. His successor, provisional president Ramiro de León Carpio, faced similar difficulties in Congress, as he too had no organized party backing. He called for a congressional reform referendum that reduced the size of Congress, but this resulted in little improvement. Although Presidents Alvaro Arzú (1996–1999) and Alfonso Portillo (2000–present) each had working majorities in Congress, resulting in lessened legislative corruption and vote-buying, both nonetheless faced difficulties in gaining qualified majorities needed to approve constitutional measures. The Guatemalan state is extremely weak; taxes are an exceptionally low share of GDP (approximately 7.7 percent, versus 15 percent for most Latin American countries and 30 percent for Western industrial states), thus producing perpetual financial crises.13 Line ministries have scant capacity to implement programs and find it difficult to use what little money they do have. These weaknesses contribute to the poor performance of the state in such areas as crime, healthcare, and education. Steps Toward Negotiations In this context of tenuous civilian government, an entrenched but internationally vilified military, and fiscal weakness of the state, three consecutive presidential administrations negotiated to end the civil war. The process took six years and involved three different negotiating protocols, two different mediators, and several different negotiating teams. Three factors slowed progress: the military incapacity of the URNG to create a climate of urgency; resistance to specific agreements by elements within the armed forces, the economic elite, the URNG, and its followers; and lack of leverage on the part of international actors because of Guatemala’s limited dependence on international assistance. Talks began in 1987 with a national dialogue, as called for by the Esquipulas Agreement signed by the Central American presidents. 14 In 1990 the National Reconciliation Commission (appointed by the government but not directly representing the government) reached agreement in Oslo, Norway, with the URNG on a process of negotiation. This initiated an expanded dialogue, called the “Oslo process,” which consisted of meetings in various locales in Latin America and Europe between the URNG and various representatives of the political parties, the business community, churches, unions, and universities. During Jorge Serrano’s presidency (1991–1993), the government and the URNG met formally, and produced two agreements (“Mexico” and “Queretaro”) that established general procedures and substantive themes, respectively, toward democratization and the establishment of peace. During the caretaker government of Ramiro de León Carpio, a “Framework” accord established more specific procedures, including UN “moder-

S-15

11/24/09

4:24 PM

Page 429

PEACE IMPLEMENTATION IN GUATEMALA

429

ation,” civil society input (the Civil Society Assembly [ASC]), and international support and consultations (the Six Friends).15 Substantive agreements on human rights followed: the resettlement of uprooted populations, establishment of a truth commission (the Historical Clarification Commission), and recognition of the identity and rights of indigenous peoples. The balance of the accords were signed during the government of Alvaro Arzú, addressing socioeconomic and agrarian issues; strengthening of civilian power and changes in the role of the military; constitutional and electoral reform; the calendar for implementation; and a final accord activating all the accords previously reached (excluding the human rights accord, and those relating to human rights in the indigenous accord that went into effect immediately upon their signing).16 Negotiations under UN “moderation” proved more fruitful than earlier talks, in part because the Guatemalan parties were better prepared to make progress, in part because the UN focused more international attention on the process, and in part because UN moderator Jean Arnault gained the confidence of the two sides. The UN’s strategy during the talks, particularly the last phase, was to help the government to translate its proposals into terms acceptable to the URNG. Arnault also used shuttle diplomacy, minimizing direct contact between the parties on sensitive issues until some degree of consensus was achieved. The Parties’ Incentives to Negotiate For the URNG leadership, a primary motive to negotiate was the impossibility of military victory and the need to buy time in hopes of generating more favorable military conditions.17 Later, URNG leaders perceived that real gains could be achieved at the table that exceeded those likely to be won on the battlefield.18 By participating in an internationally sanctioned process, the URNG gained greater visibility and leverage than their military strength warranted. The Oslo process provided a public forum for social sectors in which the URNG had significant influence, which both benefited and hurt the URNG’s long-term political prospects. Civil actors that supported the URNG’s goals took uncompromising positions on such issues as accountability for past human rights crimes, socioeconomic reforms, and indigenous issues.19 When the URNG agreed to terms that fell short of its constituents’ demands, further negotiations stalled while the URNG attempted to mend relations with disgruntled supporters. Negotiations at times jeopardized the government’s relations with powerful forces in Guatemalan society, such as the military, the business community, and the traditional right. Any agreement that would satisfy the URNG was anathema to conservative power groups. Though major domestic political dividends would not necessarily accrue to the administration that ended the conflict, there were, nonetheless, strong incentives for the

S-15

11/24/09

430

4:24 PM

Page 430

CASE STUDIES

civilian governments (and especially the Arzú administration) to end the war through negotiations. The civilian governments had only tenuous authority and autonomy. Civilians had been allowed to take office at the sufferance of the armed forces, which orchestrated the partial democratization that began in 1984. Four consecutive civilian presidents struggled to attain real power for elected civilian authorities. For these weak civilian leaders, peace talks, with their emphasis on compromise, consensus-building, and international support, provided an opportunity to counteract the tutelary authority of the military and the resistance of reactionary domestic political forces. Peace talks brought international attention and political pressures, and increased the salience of international norms of democracy and good governance. If peace could be achieved, civilian governments could challenge the military’s claims on powers and prerogatives, many of which were based on the counterinsurgency mission. The peace accords also provided an opportunity to increase the financial resources available to the chronically weak Guatemalan state. Historically, the fiscal constraints on the state not only inhibited the state’s role in developing and modernizing the economy (a role advocated by some in the private sector), but have also limited the ability of elected governments to play patronage politics and expand their support. The accords presented two partial solutions to this problem: First, the UN-sponsored talks brought expectations of substantial international assistance. Indeed, some $1.9 billion was promised by international donors, and while the Arzú government was slow to take advantage of this generosity, the funds did allow the government to move forward with infrastructure reconstruction. Second, the agreement on socioeconomic reforms required the government to raise taxes from 8 percent of GDP to 12 percent by 2000. Though this move appears to be a concession to the URNG, and proved politically difficult to implement, it was actually proposed by the government’s negotiators.20 For the government of Alvaro Arzú, the peace process offered a political strategy that might overcome the tremendous resistance to social, economic, and institutional change that has characterized the Guatemalan political system. The accords promised the government a framework to rally support for modernization, while retaining the option of blaming unpopular measures on international pressure and the demands of economic globalization. At the same time, the implementation processes agreed to were so complex and involved so many different actors that the government could easily allow individual reforms to languish if opposition proved too strong. The peace negotiations could not have gone forward without the support and cooperation of the Guatemalan military. Given the weakness of the URNG, the Guatemalan military was not compelled to agree to the accords,

S-15

11/24/09

4:24 PM

Page 431

PEACE IMPLEMENTATION IN GUATEMALA

431

and military officers sometimes questioned why the government was negotiating with a defeated insurgency. Yet in the final stages of the negotiations, the military led the way in directly contacting URNG commanders, establishing procedures and communications to avoid unwanted clashes, and building sufficient trust on the part of the URNG to make its demobilization possible. In the end, the military accepted an agreement that would eliminate a crucial portion of its institutional mission—its role in internal security. This support by the military for the peace process suggests that for at least portions of the military leadership, the peace process offered important benefits. The military’s acceptance of the negotiated settlement reflects factional shifts within the military in favor of moderate officers whose primary concern was the long-term stability of the country and the integrity of the armed forces. This group believed that national stability required a settlement, and that the interests of the military as an institution were not served by continuation of the conflict. The military’s efficient but brutal counterinsurgency tactics, which flagrantly violated international human rights standards, were reviled both at home and abroad. Moreover, the secrecy and institutionalization of illegal practices were leading to widespread criminality and decay within the military itself. While most in the military leadership considered their violent actions necessary and were willing to ignore their public image during the peak of the conflict, in the long run the military could not afford to remain a pariah. The best way to avoid ongoing institutional erosion, condemnation, and loss of legitimacy was to accept and even support a negotiated settlement.21 From this perspective, acceptance of the negotiations extended the military’s orchestrated transition to civilian rule in 1984–1985, and was based on the same analysis of how the military could best restore its legitimacy. The accords enabled the military to make these gains without sacrificing its core interests. The military was not required to undergo a purge by outsiders, nor was military education or doctrine subject to civilian scrutiny. In interviews with scholars and journalists, several officers stated that the military gained by losing its public security role, since it was one that would most likely expose them to accusations of human rights violations.22 In any event, the military’s role has not changed given the defeat of the referenda on constitutional reforms. Serrano and the Start of Direct Talks During the national dialogue that began in 1989, and the subsequent Oslo process, a consensus emerged in the government that it should treat the URNG as a counterpart in peace negotiations, and that the talks should go beyond disarmament of the rebels. At the same time, the private-sector elite and the military opposed direct talks. Although President Vinicio Cerezo

S-15

11/24/09

432

4:24 PM

Page 432

CASE STUDIES

named a “conciliator” and announced that his government would begin direct, unconditional talks with the URNG in 1990, he was too weak to follow through, while Jorge Serrano Elías had greater leverage with the military. Serrano’s presidential victory in 1991 represented the first transfer of power from one elected civilian president to another since 1950, and Serrano, a fundamentalist Christian close to retired General Efraín Ríos Montt, also had significant promilitary credentials.23 Serrano demonstrated his authority within the military by firing a defense minister and bringing corruption charges against other high-ranking officers. In addition, the United States in 1990 began to pressure for human rights improvements, and multilateral institutions such as the World Bank and the European Parliament began to condition aid on progress in human rights, social investment, and peace negotiations. 24 Some business and opinion leaders began to see that ending the war might break Guatemala’s international isolation and thereby improve investor confidence and Guatemala’s prospects in regional trade talks. With these enabling conditions, Serrano proceeded to initiate direct talks with the URNG in April 1991. Serrano included military officers as part of the governmental delegation, with Msgr. Rodolfo Quezada Toruño acting as conciliator and the United Nations and Organization of American States (OAS) as observers. These direct talks lasted two years, and produced two agreements that established the agenda for negotiations: democratization and human rights, identity and rights of indigenous people, constitutional and electoral reforms, socioeconomic issues, the agrarian situation, resettlement of uprooted populations, the reincorporation of the URNG into political life, cease-fire, a schedule for implementing and verifying the peace accords, and the signing of a final accord. This agenda was affirmed by the Framework Agreement of January 1994 and was used as the structure for the subsequent UN-mediated talks. Serrano could not go beyond setting the agenda, however, because of reactions from the military and business associations such as the Coordinating Committee of Commercial, Industrial, and Financial Associations (CACIF). These groups objected to the substance of the URNG’s proposals and to negotiating with the rebels before they disarmed, and bridled at perceived international pressures. In response to the latter complaint, Serrano demanded the removal of UN observer Francesc Vendrell on the pretext that he had acted improperly in arranging secret meetings with the URNG. Vendrell was replaced by a then relatively low-level UN official, Jean Arnault, in a concession to government pressures to lower the UN’s profile in the talks. Beyond these difficulties, at this stage in the process the URNG appears to have been negotiating largely for tactical reasons, making substantive achievements unlikely.25 Serrano’s domestic political position eroded quickly, as his legislative alliance with two other parties (the Christian Democrats and National

S-15

11/24/09

4:24 PM

Page 433

PEACE IMPLEMENTATION IN GUATEMALA

433

Centrist Union) collapsed in early 1993, and he faced increasing accusations of corruption. This instability culminated in the autogolpe of May 25, 1993, which provoked a ten-day constitutional crisis and halted negotiations. De León Carpio and UN-Moderated Negotiations Congress selected former human rights ombudsman Ramiro de León Carpio to complete Serrano’s term. His position was not auspicious for the peace process. De León belonged to no political party, lacked a disciplined base of support in Congress, and lacked the legitimacy that would be accorded by a popular election. Thus it seemed unlikely that he would have the clout to negotiate an end to the war. He did, however, have resources that enabled him to achieve progress in the talks. First, the military had incentives to support de León, since backing a former human rights ombudsman would help convince international actors of the military’s new democratic vocation. Second, he had initial support from civil society and popular groups, largely because of his favorable reputation as ombudsman and because he was seen as an improvement over Jorge Serrano. Third, international actors that had firmly opposed the autogolpe gave strong support to de León in the form of increasing pressures to reinitiate the peace talks under the auspices of the United Nations.26 Countries advising the government and the United Nations argued that having the UN as mediator and verifier would induce both parties to negotiate more seriously and make it less likely that the parties would use the talks for tactical purposes. The successful performance of the United Nations in El Salvador influenced the international community’s vision of the UN’s potential in Guatemala. De León chose an effective new head for the governmental peace commission, Héctor Rosada-Granados. The government commission and the URNG worked closely with Arnault to formulate a new framework, which included an ambitious year-end 1994 deadline, provided for the UN to serve as “moderator” (widely understood to mean “mediator”), and established a formal role for civil society organizations through the ASC.27 The ASC was to be headed by former conciliator Rodolfo Quezada Toruño, and was designed to produce consensus documents that would feed into the peace process and to certify the accords reached between the two parties. These terms were combined in a Framework Agreement signed in January 1994. The new procedure produced an agreement on human rights just two months later. As in El Salvador, agreement on human rights was seen as a confidence-building measure, facilitated by the government delegation’s willingness to let a UN verification mission deploy prior to a cease-fire.28 Much of the content of the human rights accord had already been informally agreed to in earlier negotiations, since it largely reiterated

S-15

11/24/09

434

4:24 PM

Page 434

CASE STUDIES

legal commitments the government had already subscribed to in the 1985 constitution and ratified in various international human rights treaties. Two sticking points had been international verification (which elements of the state viewed as highly intrusive) and accountability for past human rights crimes. The breakthrough was reached by postponing discussion of a truth commission until a later time. Shortly thereafter, in June 1994 in Oslo, the negotiating teams also produced two further accords, one dealing with the displaced populations affected by the war, and the other regarding the creation of a Historical Clarification Commission (CEH), known colloquially as the “Truth Commission.” The latter prompted intense political reactions. Signed by the URNG under intense international pressure, particularly from the Nordic countries, the CEH agreement called for the creation of an investigative body (with both international and national staff) that would look into cases of violence during the armed conflict, going back to 1960. The commission would work for six months, with the possibility of renewal for another six months. The commission would not begin its work until after the final peace accord was signed and, most important, would not name names (as did the commission in El Salvador). Moreover, the CEH’s findings would have no juridical effect.29 Civil society groups criticized this last point, calling it a capitulation by the left regarding historic claims for justice for the victims. The CEH accord caused a nine-month delay in further progress. Political backlash against the URNG for having conceded too much pushed the rebels to adopt a harder line and to withdraw from the talks until the United Nations could establish a verification presence that would provide greater assurances of respect for human rights. The United Nations hesitated to deploy a mission, as some in New York were skeptical that the negotiations would succeed. If the talks did not progress fairly quickly, the United Nations could find itself with an open-ended human rights verification role.30 The United Nations eventually deployed a human rights mission, MINUGUA, in November 1994. During the last quarter of 1994, as talks resumed, UN mediation played an active role in brokering the Accord on the Rights and Identity of Indigenous Peoples, signed on March 31, 1995. This agreement was enriched by the ASC’s presentation of a broad consensus document on this issue, as well as by the receptivity of the government delegation to open discussion. Like many of the accords, however, this agreement contained more general principles than concrete reforms, and set into motion debate over how indigenous values should be incorporated into public policy. The country’s increasingly organized indigenous population approved the accord, calling it a “minimal but important step forward” in the recognition of their rights and identity. Further negotiations were held on socioeconomic and agrarian issues,

S-15

11/24/09

4:24 PM

Page 435

PEACE IMPLEMENTATION IN GUATEMALA

435

and a wide-ranging document on this matter was drafted and debated. This draft agreement prompted determined opposition from the business community, which objected to language concerning property rights. Even diplomats and aid officials close to the talks considered the proposal too complex to implement. These difficulties might have been overcome with time, but with elections scheduled for 1995, it became clear that the de León Carpio government would not be able to finalize the negotiations his government had set into motion. It was not in the URNG’s interest to negotiate any further agreements with a lame-duck administration, especially given the lack of assurances that the next government would respect any agreements reached. To address this concern, Arnault, with the support of international actors, negotiated an agreement among all the political parties in August 1995 that the accords signed thus far would be considered by all future governments as “agreements of the Guatemalan state.” This was insufficient to permit further progress during the remainder of the de León administration. The 1995 Elections, the Arzú Administration, and the Final Phase Although the 1995 elections temporarily stalled the negotiations, the results of these elections created conditions for the completion of the negotiations. First, the presence of MINUGUA opened political space for the left. For the first time since the 1954 overthrow of Jacobo Arbenz, an openly left party, the New Guatemala Democratic Front, participated with a hastily organized campaign and little money. The result, which might appear to be dismal for the left—six out of eighty congressional seats, and 8 percent of the popular vote—was interpreted by the URNG leadership as a hopeful sign for their electoral potential. For the first time, the URNG leadership was heartened that they might be able to achieve at the polls what they could not achieve in battle or at the negotiating table. Up to that point, the URNG had tried to negotiate terms that would justify their thirty-six years of armed struggle, yet they were not strong enough to exact such concessions. The election results eased this impediment. Second, the election of Alvaro Arzú to the presidency (and a majority representing his PAN party in control of Congress) provided the URNG with an acceptable interlocutor to conclude the negotiations. From one of the country’s ultra-elite families, Arzú appeared to have the support of business and the dominant sectors of the military, yet signaled his intentions to negotiate a final settlement. A month after Arzú assumed office in early 1996, he and the URNG jointly announced that peace talks would resume in late February. The government also revealed that since December, Arzú and key advisers had held four secret meetings with the URNG under the auspices of the Sant’Egidio

S-15

11/24/09

436

4:24 PM

Page 436

CASE STUDIES

community in Italy. When the talks began again on February 24, Arzú headed the government delegation, an unprecedented move that signaled his commitment to peace. The peace process now moved at a rapid pace: five different agreements and the final peace accord were signed within a year. The sudden celerity of the negotiations was due to the URNG’s assessment that the Arzú government was an honest broker. Over the course of the year, measures were taken to bolster the confidence of both sides. Arzú named a former guerrilla ideologue (and personal friend), Private Secretary to the President Gustavo Porras, to head the government delegation, along with two former finance ministers and a top military officer. Porras was perhaps the first negotiator to understand the URNG’s positions. Confidence between the new government and the URNG rebels following their initial talks was such that, one month later in March, the guerrillas announced a unilateral suspension of offensive military actions, followed the next day by a presidential order for the army to cease all counterinsurgency operations. Steps were taken by both parties throughout the year that contributed to the sense that a final peace accord might be reached. In May, the accord on socioeconomic and agrarian issues was signed in Mexico. This agreement, approved only after the removal of those sections deemed most unacceptable to the private sector, was sufficiently bland to receive broad support, even from the CACIF. Like the other accords, it spelled out a framework for policy in broad terms, committing the government to increase tax revenues from 8 to 12 percent of the GDP, and to increase health, education, and housing expenditures by 50 percent by the end of 1999.31 Arzú took several immediate steps to demonstrate his seriousness in seeking military reforms, which provided incentives to the URNG to continue negotiating. In April 1996 the military proposed a revision to the military code so that common crimes committed by the military would be tried in civilian courts. In August 1996 the government announced that the notorious civil defense patrols would be demobilized by November 15 of that year.32 Through these initiatives, the government increased the confidence of the URNG in the government’s good faith, but also satisfied military concerns by preempting the negotiation of issues that the military leadership felt it should not have to discuss with an insurgency that it had defeated years earlier. Some critics have charged that the negotiation process proceeded too rapidly in the final stages. The lack of specificity in the latter accords, compounded by the relative lack of specific benchmarks in the Agreement on the Implementation, Compliance, and Verification Timetable for the Peace Accords, made it less likely that the accords would produce reforms of ade-

S-15

11/24/09

4:24 PM

Page 437

PEACE IMPLEMENTATION IN GUATEMALA

437

quate depth and consistency to achieve the lofty goals laid out in the preambles. Once the URNG decided to end the war and pursue its goals through elections, it proceeded with almost indecent haste to reach agreements that were short on substance.33

Implementation The Context of Peace Implementation On May 16, 1999, a popular referendum was held on a set of fifty constitutional reforms essential to the peace accords. Implementation of these reforms had already been slowed because Congress could not pass the package by a two-thirds majority.34 To the chagrin of Guatemalans who supported the reforms, as well as international observers, all four ballot questions were defeated by margins of roughly two to one, reflecting a particularly strong rejection in the capital city.35 Turnout was only about 18 percent (21 percent in the capital), despite months of publicity and voter education, much of it funded by international donors. The defeat was attributable to at least two main factors: a powerful, expensive, antireform publicity campaign during the final few weeks before the vote, which portrayed the reforms in a negative, almost apocalyptic light; and public ignorance regarding the content of the reforms, which were so complex as to be easily misunderstood. With regular elections scheduled for November 1999, there were no further opportunities to pass a revised package of constitutional reforms in 1999. The Congress and executive that took office in January 2000 proved unable to develop a pared-down set of reforms that could be passed by referendum. As of early 2002, constitutional reforms had still not passed, and key elements of the peace accords had become moot. This is obviously a significant failure of the peace process. This setback reflects the implementation context where neither of the parties to the accords is a hegemonic actor capable of setting and implementing policy. Though the PAN and FRG governments had majorities in Congress, they showed little capacity to deliver legislation called for in the accords. The legislature failed to draft constitutional reform legislation for voter approval until strong pressures were brought to bear by international donors. Both governments had little capacity to provide public goods and had difficulty garnering the trust of citizens, so virtually any major initiative was opposed. A key weakness of the state throughout the 1980s and 1990s was the parallel power exercised by the military and private-sector elite over civilian authorities. The peace accords did little to diminish such power, and there are signs, though proof is elusive, that

S-15

11/24/09

4:24 PM

Page 438

438

CASE STUDIES

these parallel powers blocked implementation of specific elements of the accords and actually gained influence under the Portillo government (2000–present). The entire political party system in Guatemala is weak. Vast numbers of Guatemalans do not participate in the formal political system. Moreover, powerful conservative forces operating outside the party system can mobilize substantial support through the mass media and social networks. One striking feature of the defeat of the constitutional referendum was that all of the political parties supported a yes vote, yet a strong majority of those who bothered to vote chose to reject the referendum. No party was willing to invest substantial financial and political capital to support a measure that might be defeated, so there was no counterweight to the publicity campaign against the reforms. Implementation of the accords depended on this fragile political system, supplemented by multiple special commissions, and diffusion of responsibility made it difficult to assign blame or credit for failures or successes. New public policies that should result from the peace process passed through a variety of institutions before they could be approved and implemented. At times this meant that for the government to comply it merely needed to create a commission or present a draft law to Congress. The accords provide no criteria for the content of the discussions held or the documents presented, and set few explicit standards for assessing the proposals and policies eventually produced. The structure of the process accentuated tensions between inclusiveness and the need for bureaucratic efficiency. When the peace accords were in their final stages, the government tried to reduce popular expectations by commenting that they were, in fact, only a “point of departure” and an “agenda” for further discussion. Nevertheless, as the process moved forward, more groups began voicing preferences and demands and the government, in response, began emphasizing the need for results. At different points, the government (and the URNG) opted to cut some actors out of consultations, and proposed legislation and policies that at best fulfilled minimal requirements of the accord. Constitutional Reform The case of the constitutional reforms illustrates the tension between the accords’ participatory rhetoric and their elitist origins, and demonstrates the difficulty of implementing major changes in a timely way when a large number of institutions and organizations are involved. At issue was whether opposition political parties and civil society actors that had not participated in the negotiations should accept the agreements made by the government and the URNG, or whether constitutional reforms required further discussion to build a broader societal consensus. The government had committed

S-15

11/24/09

4:24 PM

Page 439

PEACE IMPLEMENTATION IN GUATEMALA

439

itself to present a package of reforms to Congress within ninety days of the signing of the final accords. This generated an expectation, particularly on the part of international donors, that the constitutional reforms would be passed quickly so that various elements could be implemented. The PAN’s own timetable compounded the urgency, since the party hoped to capitalize on implementation of the accords—and associated international aid flows—to strengthen its electoral prospects. The easiest route to passing the reforms was to get Congress to vote on the minimal agreements negotiated between the government and the URNG, yet this attempt sparked strong objections within Congress, as well as from civil society and political party leaders who had not participated in the peace negotiations and thus had not had a say in the content of the reforms. In particular, the second largest party, the FRG, resented the PAN/URNG alliance in pushing through these agreements. The requirements of the Guatemalan constitution that reforms be approved by a two-thirds vote in Congress, followed by a popular referendum, made it impossible for the government to confine discussion to the core twelve reforms. During the first phase of discussion, the appropriate commission held hearings for political and civic organizations to present their positions on the proposed reforms. In the process, several proposals were made by the FRG and the Coordinator of Mayan Peoples’ Organizations of Guatemala (COPMAGUA).36 In an effort to limit the debate and generate sufficient consensus among the parties to pass reforms, the (PAN) president of the Congress convened an extra-parliamentary Instancia Multipartidaria (Multiparty Body) in September 1997, including representatives of the main political parties. Many who participated in these meetings were not members of Congress. This arrangement was criticized for undermining the constitutionally mandated role of Congress, but reflected the fact that most important political decisions are often made by party leaders who are not members of Congress. The Multipartidaria strategy broke down in May 1998, largely because the FRG had attempted to introduce proposals that would allow the head of the FRG, retired General Efraín Ríos Montt, to run for the presidency.37 This was a divisive issue within the PAN, but on balance the party was reluctant to make this concession in exchange for the FRG’s support. The government was under increasing pressure to pass the reforms. A Consultative Group meeting of donors was scheduled for mid-1998, then postponed until October, primarily because of the failure to pass the reforms in a timely fashion. The government shifted the debate back to Congress, and to negotiations between political parties and civil society organizations, especially indigenous groups such as COPMAGUA, in order to ensure the votes of the FDNG and the Christian Democrats. In the process, the package of proposed reforms grew from the minimum twelve required to implement the

S-15

11/24/09

440

4:24 PM

Page 440

CASE STUDIES

accords to thirty-nine substantive reforms, three reforms that proposed minor changes in wording, and eight transitory articles. In the end, the FRG withdrew from the negotiations, and on October 16 the package was approved by the PAN, the FDNG, and Christian Democrats. This package still needed to be approved by popular referendum, and the legislative decree calling for the referendum proposed to have voters respond to a simple “yes” or “no” on the entire package. The FRG accused the government of trying to ram through a package that reflected little more than a pact between the PAN and the URNG, while various civic groups felt that an up or down vote would not allow a conscientious vote on the issues. The proposed approach also triggered a complaint before the Constitutional Court by the Center for the Defense of the Constitution (CEDECON), a small group of conservative lawyers opposed to changing the constitution. The court ruled in mid-February 1999 that the single-ballot question was unconstitutional, forcing a new legislative decree that grouped the fifty reforms under four ballot questions. All together, the negotiations in Congress, in the Multipartidaria, and between parties and civil society, combined with delays attributable to Hurricane Mitch and legal challenges, resulted in the popular referendum taking place nearly two and a half years after the signing of the final accords. In the end, the package of reforms was defeated. The broadened debate on the reforms, and the addition of various other proposals to the package, shifted some normally conservative groups, such as the primary private-sector association, the CACIF, to come out in support of the basic reforms required by the accords. This support proved insufficient to revive the defeated reform package. Accompaniment Commission The Timetable Agreement signed on December 29, 1996, included the formation of an Accompaniment Commission to facilitate the implementation of the peace accords. The commission consisted of two persons from both the government and the URNG, four notable citizens, a representative of Congress, and the chief of MINUGUA (with voice, but without vote). Its role was to analyze the difficulties and obstacles encountered in the application of the timetable, review legal proposals before they went to Congress, schedule the different commitments set forth in the timetable agreement, communicate with MINUGUA, support international fundraising for the financing of the accords, and present periodic reports on compliance with the agreement. It worked closely with the government’s Secretariat for Peace (SEPAZ), responsible for implementation by the government. This commission played a low-profile, but occasionally useful, role in pushing the peace process forward. It often intervened to resolve problems

S-15

11/24/09

4:24 PM

Page 441

PEACE IMPLEMENTATION IN GUATEMALA

441

arising in the various participatory commissions, and was able to vet proposals from the executive branch. However, the commission had several significant weaknesses. It was not sufficiently accessible to those wishing to influence the outcome of policy proposals. In part because it met only once or twice a week, the commission was frequently overwhelmed with requests for meetings. The commission had no designated staff, and the government marshaled greater resources than the URNG, which had trouble interesting international donors in its work. The civil society representatives didn’t really represent organized civil society groups and brought little to these meetings other than their personal knowledge in their respective fields.38 The commission was generally overwhelmed and incapable of fulfilling its role as facilitator of accord implementation. Finally, the makeup of the commission—excluding political parties other than the PAN and the URNG (which wasn’t yet represented in Congress)—reinforced the image that the peace process belonged to the PAN, the URNG, and the international community. The International Role in Implementation By all accounts, the United Nations played a crucial role in shepherding the Guatemalan parties to an agreement, and virtually all the Guatemalan participants and observers interviewed for this study credited UN Moderator Jean Arnault with having been an effective mediator. The UN’s role as verifier has been more controversial, and it is less clear whether the UN’s role has been as effective as it might have been. The key challenge in assessing the UN’s impact during implementation is the difficulty of separating the limitations imposed by the weakness of domestic actors, the vagueness of the accords, and the nationalism of Guatemalan conservatives, and those limitations imposed by choices made by the mission and the broader UN system. This is the strategic challenge facing the mission: How much room for action does the mission actually have? How strongly can it criticize the government without creating a counterproductive backlash either against itself or against the peace process more broadly? We will not attempt here to reach judgment on this issue, but will merely present the dilemmas, since these may be instructive regarding the constraints likely to face future UN missions where weak accords combine with weak domestic implementation capacity. During the human rights verification phase, from November 1994 until January 1997, MINUGUA’s mandate was fairly straightforward and the human rights norms and instruments reiterated in the accords provided a clear framework for verification. When MINUGUA took over a broader verification mandate, however, its tasks multiplied, and it had to work with much less concrete benchmarks. While the commitments in the peace accords were numerous, many of them related to procedural issues (setting

S-15

11/24/09

442

4:24 PM

Page 442

CASE STUDIES

up commissions, presenting reports, establishing procedures, introducing legislation, designing programs, and “supporting initiatives” by unnamed parties to achieve particular goals), with few firm guidelines for assessing the substantive results. Others involved initial steps in institutional reforms, such as establishing new offices or agencies. There were a few substantive commitments, such as increased tax revenues and social spending to be achieved by the end of 1999. On these and many other issues the government simply failed to meets its commitments, despite pressure from the UN and international donors. MINUGUA’s broader verification role has been further complicated by a lack of understanding by most Guatemalans regarding the mission’s mandate and functions. The mission retained the same name after it took on verification duties beyond human rights, rather than signaling its new role with a new name. Its low profile after the signing of the final accords contributed to the average citizen having little idea what the mission is, beyond associating it with human rights. Even the business elite often failed to recognize that since January 1997 the mission had a mandate to verify all the accords. This misunderstanding has led to accusations by prominent Guatemalans that MINUGUA is an illegitimate foreign interloper that has exceeded its mandate. MINUGUA’s leverage was further compromised by the fact that the main challenge to the incumbent government came from a right-wing party, the FRG, with a reputation for opposition to the peace accords. The United Nations and international donors believed that they faced a political catch22: with elections coming at the end of 1999, they could not afford to criticize the incumbent PAN too strongly, as any criticism could strengthen the PAN’s opponents, who, if victorious in 1999, might scuttle the remaining elements of the peace accords. The existence of a strong challenge to the PAN from the right constrained the United Nations in criticizing the government and made international donors reluctant to support the present government, despite shortfalls in implementation. The mission’s image was damaged in early 1997, just as MINUGUA was taking on its more comprehensive mandate, by accusations in the Guatemalan press that in 1996 Jean Arnault—at the time moderator of the negotiations—had suppressed information regarding the “disappearance” of a URNG member arrested by the Presidential Military Staff. Irrespective of their merits, these accusations combined with the mission’s low profile to reduce the mission’s authority and clout within the peace process.39 Whatever the challenges of the Guatemalan context, many observers believe the UN mission’s response and strategy were less than ideal. Many observers have questioned the mission’s reluctance to criticize government actions and insist upon closer compliance with the accords, particularly during the first eighteen months following the signing of the final peace accords; in 1998 and 1999 the mission began to take stronger positions. Yet

S-15

11/24/09

4:24 PM

Page 443

PEACE IMPLEMENTATION IN GUATEMALA

443

the government has appeared unresponsive to these criticisms. MINUGUA belatedly improved its public information strategy, and made greater efforts beginning in 1998 to meet with key political and social elites to explain the mission’s mandate and views. Both MINUGUA staff and members of the business community credit this greater outreach with reducing misunderstandings. But in the face of continued unfavorable press coverage, MINUGUA has been unable to overcome the damage already done to its public image in the early stages of the comprehensive verification mandate. Some observers have suggested that the selection of Jean Arnault as the first head of the comprehensive verification mission was ill advised: he had too strong a personal stake in defending the peace accords, and was reluctant to acknowledge and criticize failures by the government to fulfill its commitments. It might have been more effective to have selected a different chief of mission, reserving Arnault for periodic visits to Guatemala to provide authoritative interpretations of what was agreed to during the talks, and push the parties for compliance. However, we are not convinced that the choice of leadership for the mission was decisive in the difficulties MINUGUA has encountered. Despite the difficulties that the United Nations experienced during implementation, it achieved important successes, and established approaches that may be applicable in other contexts. First, there was excellent coordination within the UN system, and good cooperation between MINUGUA and international financial institutions such as the World Bank, International Monetary Fund, and the Inter-American Development Bank. The international financial institutions were consulted and involved in the negotiating phase, contributing to a greater understanding on their part of the financial requirements of the peace process. Coincidentally, Guatemala’s state revenues and spending have been so low that effectively all of the relevant international actors and institutions agreed that the Guatemalan state should tax and spend more—an unusual situation given the usual tensions between peacebuilding and fiscal austerity.40 Within the UN system, MINUGUA, the UN Development Programme (UNDP), the UN High Commission for Refugees (UNHCR), and other agencies coordinated their work effectively, drawing on lessons from neighboring El Salvador. Also reflecting lessons from El Salvador, MINUGUA initiated efforts to strengthen state and civil institutions. Notwithstanding these important efforts, international donors have not always worked together to apply pressure on the government to comply with the accords. Implementation of Specific Tasks Human rights. The Comprehensive Agreement on Human Rights did little more than to reaffirm the government’s responsibility to protect human rights, as already stipulated in various international instruments to which

S-15

11/24/09

444

4:24 PM

Page 444

CASE STUDIES

Guatemala was signatory. Specifically, the accords required the government to strengthen judicial institutions and the Office of the Human Rights Ombudsman (the first of its kind in Central America, created by the 1985 constitution), update criminal codes, regulate the bearing of arms, fight against clandestine irregular groups (death squads), guarantee the freedom of movement and association, end forced military recruitment, provide protection to those working in human rights, and provide reparations to those harmed by the armed conflict. Both sides declared that they would respect the civilian population and those wounded or captured in combat. A key provision of the Comprehensive Agreement was the deployment, before the cease-fire, of a UN verification mission. MINUGUA, deployed in November 1994, quickly established a presence throughout much of the country. MINUGUA’s mandate reflected two lessons from the human rights verification effort in El Salvador. First, MINUGUA was to supplement its verification work with efforts to strengthen human rights monitoring capability within Guatemala. Second, the accord called on MINUGUA to give priority to certain rights over others, emphasizing rights to life, integrity, and security of the person, individual liberty, due process, freedom of expression, freedom of movement, freedom of association, and political rights, in order to avoid being swamped with cases unrelated to fundamental liberties. The human rights accord and, more specifically, MINUGUA’s presence provided a deterrent against abuses. Observers also credit the United Nations with having increased the political space available to the left, making it safer for the newly formed FDNG party to participate in the 1996 elections. MINUGUA’s human rights reports (twelve as of early 2002) reflected improvements in the human rights situation until mid-1998; since then, they note signs of deterioration in the form of increased numbers of lynchings; the continued existence of clandestine security structures; increased threats against human rights workers, judicial officials and witnesses, and the news media; and continued impunity. Three military officers were convicted—for the first time ever for a politically motivated killing—in the case of the murder of Msgr. Juan Gerardi, who had led the Catholic Church project that investigated the history of human rights abuses during the conflict. But this high-profile success did not reflect broader systemic change. The persistence of these institutional problems highlights MINUGUA’s limited institutional strengthening efforts. MINUGUA’s efforts in these areas are small in comparison to other international actors such as the U.S. Agency for International Development (USAID), the World Bank, and the Inter-American Development Bank. The courts, the public ministry, and the police have shown little willingness to implement changes. The institutional strengthening role has, however, provided MINUGUA with important

S-15

11/24/09

4:24 PM

Page 445

PEACE IMPLEMENTATION IN GUATEMALA

445

insights into the real workings of the judicial system and the security apparatus. Resettlement of refugees and displaced persons. In the early 1980s, an estimated 150,000–200,000 persons fled Guatemala and another 1–1.5 million were internally displaced due to the intense violence of the armed conflict. By 1983 some 46,000 Guatemalans were officially recognized as refugees in camps in Mexico. Refugees trickled back at a rate of 1,000–2,000 a year in response to a repatriation program begun in 1987, but the overall numbers remained approximately the same due to the high birth rate in the camps.41 International attention to the issue of returning refugees in Guatemala has been extraordinary, leading one observer to note that they “have fared relatively well over the years, compared to the internally displaced and compared, in many cases, to other campesinos.”42 A final group of Guatemalan refugees returned in June 1999, and the return process was declared complete by the Guatemalan and Mexican governments. The Accord for the Resettlement of the Populations Uprooted by the Armed Conflict was one of the first accords signed (June 1994). The refugee accord was less a set of explicitly verifiable commitments than a statement of overall development policy to be pursued in support of these populations’ integration and resettlement. It built upon a series of prior agreements between refugee organizations and the government dating back to 1991. By the time the final peace agreement was signed in December 1996, most of the refugees had already returned or were negotiating their return to Guatemala. Although the accord on refugees in and of itself contributed little to the return process, the broader peace process provided incentives for the government to be more accommodating to refugees, as a way of showing that political conditions were improving. In November 1991, President Serrano signed a “Letter of Understanding” with the UNHCR outlining basic guarantees for returning refugees, and in October 1992 the government signed an unprecedented agreement directly with refugee representatives, outlining a mechanism for land acquisition. Between 1993 and 1997, some 37,000 refugees returned to Guatemala, and more than thirty-five communities were resettled.43 The refugee return process in the 1990s was shaped by the 1992 agreement, which is even stronger than the 1994 peace accord in specifying the government’s responsibility to provide land to the displaced. As called for in the 1994 accord, a Technical Commission was established with two representatives from the government and two from the displaced populations, with the international community present as nonvoting participants. This commission began meeting in July 1994 to prepare a strategy for the implementation of the agreement, and by early 1997 the

S-15

11/24/09

446

4:24 PM

Page 446

CASE STUDIES

parties agreed to set up a trust fund under the aegis of the UNDP. By mid1998, international donors had contributed $8 million to the fund, which would be used for projects in health, agriculture, infrastructure, documentation, and land acquisition. The two-and-a-half-year delay in reaching an agreement on the trust fund stemmed from distrust between the parties, exacerbated by the government’s questioning the legitimacy of certain representatives of the displaced and twice making unilateral proposals, sidestepping the displaced populations. The 1994 accord refers to the refugees’ right to recover their land, but subsequent agreements have weakened their status. A September 1997 agreement (and a subsequent February 1998 joint declaration) between the government and the refugees determined that only refugee groups that had signed up by December 29, 1997, for new lands would qualify for the advantageous terms provided through a government-sponsored revolving fund.44 Aid for displaced populations has also met with few successes. Representatives of the government and displaced persons revised a plan of the UN Educational, Scientific, and Cultural Organization (UNESCO) for meeting the educational needs of the displaced in early 1998, but lacked funds for implementation. Also, in August 1997 a law was passed to facilitate the personal documentation of displaced persons, but civil authorities lack resources and training to implement the law. 45 In 1998 displaced groups who had previously served as the social base of the guerrillas in the Ixil triangle of northern Quiché and in department of El Petén were successfully relocated, although additional lands still needed to be purchased for a similar group of families in Ixcán. Guerrilla demobilization and disarmament. The timetable for URNG demobilization was very tight: the United Nations was requested to install a military verification component as quickly as possible following the signing of the final accords, and the full demobilization of the URNG was to follow within sixty days. By agreeing to such a rapid demobilization, the URNG gave up the opportunity to use its weapons to encourage government implementation. Security for the URNG during the transition was ensured by the 155 unarmed MINUGUA observers, who remained with URNG combatants throughout the demobilization phase, monitoring their activities as well as those of government forces, and providing both sides with information and assurances regarding the status of the other. No violations of the cease-fire or separation of forces occurred. The entire demobilization of the URNG was completed from March 3 to May 3, 1997. The few security complaints from the URNG during the demobilization phase related to high-altitude flights over rebel encampments. Several factors contributed to the remarkable lack of major problems. First, a ceasefire had already been in place for nine months before concentration and

S-15

11/24/09

4:24 PM

Page 447

PEACE IMPLEMENTATION IN GUATEMALA

447

demobilization of troops took place. Both sides had incentives for quick rebel demobilization. According to UN sources, most of the URNG combatants were part-time militia members who needed to return home in time to plant crops before rains began in May. The URNG leadership was concerned that militia members would abandon camps if the demobilization process was delayed. The government armed forces, for their part, preferred a rapid demobilization because it limited the time during which their movements were restricted. According to a UN official, the two sides had chosen URNG assembly areas that the government didn’t care much about, reducing the risk of government violation.46 The government made no public objection to the apparently inflated numbers of combatants, as limiting formal demobilization to the core group of 500 guerrillas would have embarrassed the government, which could have been criticized for having negotiated with such a small force. In this regard, the demobilization process involved an element of political theater. The willingness of former guerrillas to disarm represented a leap of faith in the judgment of their commanders, the good faith of the government and the military, and the general influence of the international community. A series of meetings between military officers and guerrilla leaders, which began as secret meetings at the highest levels and gradually evolved into numerous meetings between local field commanders, helped build confidence. In these latter meetings, combat leaders hovered over tactical maps to indicate the positions of their respective troops, exchanged radio frequencies and cell-phone and beeper numbers, and agreed to cooperate to avoid unintentional clashes.47 The spirit of collaboration that emerged in these meetings helped convince the rebels that the military sincerely wanted to end the conflict and would not take advantage of the URNG’s vulnerability after demobilizing, and this confidence appears to have been generally well founded. There have been a few killings and attacks against demobilized URNG members that may have involved political motives, but there has been no evidence of a systematic campaign against them.48 Reinsertion of the URNG was the responsibility of a Special Incorporation Commission made up of representatives of the government, the URNG, and observers from the OAS, the European Union, USAID, and the UNDP. The OAS representative was subsequently replaced on the commission by a representative of Spain. The URNG’s Guillermo Torriello Foundation (FGT) has been responsible for implementing many of the projects for former URNG combatants, but the new foundation lacked the institutional capacity to carry out the rapidly expanding projects, and international donors needed to focus attention on bringing the FGT up to speed.49 Despite the various institutional, political, and logistical problems encountered, the social reinsertion of former URNG combatants was not a serious challenge for the peace process, given the small numbers and the ample international funding available.

S-15

11/24/09

448

4:24 PM

Page 448

CASE STUDIES

Demobilization of government forces. Aside from moving its forces to facilitate the URNG’s demobilization, the government had four obligations under the accords to demobilize and disarm its own forces.50 First, the government agreed to demobilize its Voluntary Civil Defense Committees (CVDCs; known more commonly by their Spanish acronym, PACs). PAC membership is difficult to estimate, but it may have been as high as 393,132 prior to 1996; Defense Ministry sources claimed that a total of 279,421 former patrollers were demobilized. President Arzú began the process of demobilization in late 1996, before the accord came into effect. The military attempted to circumvent the accord by converting some patrols into unarmed, local-level Peace and Development Committees (CPDs). Sources disagree regarding the number so “converted,” but even the converted CPDs were ultimately demobilized.51 The second commitment by the government was to reduce the military by one-third during the course of 1997. Based on a putative benchmark force of 45,000, this implied a post-accord force of 31,000. Shortly after the signing of the accords, the army reported its actual force level at 35,000, so it needed only to reduce its forces by 4,000 troops to meet this target and did so easily and on schedule. The real reduction was thus markedly less than one-third. The officer corps was not reduced systematically, though there was some attrition. Third, the government eliminated the Mobile Military Police (Policía Militar Ambulante, PMA), which had a small component (699 agents) that actually performed police functions, and a larger component (1,722 agents) that functioned as a parastatal security company, providing guards for banks and other installations. Despite a mutiny by members of the PMA, all 2,421 were demobilized on schedule, with some incorporated into the “new” National Civilian Police as well as private security firms.52 Fourth, the government agreed to “begin a progressive process aimed at achieving” the redeployment of the army in a manner consistent with its mission, now to be confined to territorial defense rather than internal security. As of this writing, this commitment has not been implemented, as the army has maintained its deployment in fifteen military zones that correspond to the internal regions of the country rather than border and coastal defense zones.53 Some posts that had been abandoned by the military in 1997 were subsequently reoccupied by troops as part of a governmentordered transitional role for the army in public security. The government has been unresponsive to MINUGUA criticism of this violation of the accords. Public security reforms. Institutional weakness of the justice and public security systems contributed to the conditions for civil war. Police forces were weak, poorly staffed and equipped, and subordinated to the military. The police were employed in counterinsurgency roles and gained a

S-15

11/24/09

4:24 PM

Page 449

PEACE IMPLEMENTATION IN GUATEMALA

449

reputation for brutality and ineffectiveness. The military was extensively involved in internal security missions, including domestic political spying. The military enjoyed effective immunity from prosecution in civilian courts, and the military never punished personnel involved in human rights abuses, and seldom prosecuted those involved in corruption. Such impunity was perhaps a necessity in counterinsurgency war, which included violence against noncombatants, summary executions of suspected regime opponents, “disappearances,” and torture. Many of these actions were illegal and required a strategy of periodic amnesties for military personnel, as well as weakness of the courts and subservience by the police. One of the effects of institutionalized impunity was widespread criminality within the military, involving kidnapping, murder for hire, car theft, timber theft in the Petén region, illegal expropriation of lands, and contraband.54 The judiciary was notoriously corrupt, ineffective, and inaccessible to the majority of Guatemalans. The Public Ministry, which oversees prosecution, lacked the resources and capable personnel needed to prosecute crimes effectively. The courts were highly centralized, poorly administered, and hampered by cumbersome written procedures. Higher courts controlled the administration of lower courts, wresting autonomy from judges and politicizing the judicial process. Judges were often poorly qualified, unfamiliar with the law, minimally devoted to their duties, poorly paid, and easily influenced by bribes. All proceedings were in Spanish, which limited the ability of indigenous people to make use of the courts. The Agreement on the Strengthening of Civilian Power and on the Role of the Armed Forces in a Democratic Society acknowledged that the institutional weakness of the justice and public security systems contributed to the conditions for civil war. It proposed constitutional amendments that would take policing and domestic security out of the hands of the military and instead establish a National Civilian Police (PNC) that would be “under the direction of the civil authorities and shall maintain absolute respect for human rights in carrying out its functions.”55 The accord stipulated only broad outlines for the nature of the new PNC: it was to be formed under the authority of the Ministry of the Interior; include a new academy to train its personnel; take account of the multiethnic nature of Guatemalan society; form whatever specialized divisions were deemed necessary; strengthen criminal investigation; and achieve a force level of 20,000 by late 1999 (a very ambitious goal). Police officers were to receive six months of training, and serve a minimum of two years. The agreement called for a series of steps “to reform the administration of justice in order to put an end to inefficiency, eradicate corruption and guarantee free access to the justice system, impartiality in the application of the law, judicial independence, ethical authority and the integrity and modernization of the system as a whole.”56 More specifically, the accord proposed constitutional amendments that would guarantee citizens “free access

S-15

11/24/09

450

4:24 PM

Page 450

CASE STUDIES

to the system of justice in the person’s own language; respect for the multiethnic, multicultural and multilingual nature of Guatemala; legal assistance to those who cannot afford their own counsel; the impartiality and independence of judges; reasonable and prompt resolution of social conflicts and provision of alternative conflict-resolution mechanisms.” Other proposed amendments would establish a more professional corps of judges, appointed by merit and required to uphold higher standards of training and competence. The penal code would be revised, and a new, independent public defender’s office would be established with sufficient funding to provide defense attorneys for all accused persons needing representation. Implementation of the accord on strengthening civilian power depends heavily on the enactment of constitutional reforms, and there can be little progress until the relevant constitutional reforms are passed. One area of progress was in the creation of a new civilian police force. By mid-2000 the PNC largely completed its initial deployment and assumed full responsibility for public security, a mere three and a half years after the signing of the final accords. This is an important achievement and, in the words of one police official, “the one area of the accords in which the most implementation has taken place.”57 Yet the government has been unwilling to conform to the few specific commitments that it did make regarding police. The Civilian Power Accord is too vague to provide an operational framework for substantive reforms and for evaluation by the United Nations: no standards are set for educational or other qualities of recruits; no distinction is made in the amount of training for officers compared to basic recruits; no limits are set on the proportion of the new force that are members of existing police forces; no standards nor mechanisms are proposed for vetting former security force members with records of human rights abuses, corruption, or criminality; no details are included regarding organization of the new force, or the content of training; no reference was made to internal or external disciplinary mechanisms Before the Civilian Power Accord was even signed, the government had reached an agreement with the Spanish Civil Guard (GCE) to train and advise the new force. GCE advisers proceeded to write a draft of the enabling legislation for the PNC, without taking into account the accord being negotiated between the government and the URNG. The law passed in February 1997 disregarded many points in the Agreement on the Strengthening of Civilian Power: it set out no professional standards or human rights requirements for the new force, and ignored the requirement that the new police incorporate members of the diverse cultures in Guatemala. It also omitted any standards regarding the human rights content of curriculum at the academy. Following strong criticism by the United Nations, international human rights groups, and Guatemalan civic groups, some of these deficiencies were corrected by regulation. Most of the short-

S-15

11/24/09

4:24 PM

Page 451

PEACE IMPLEMENTATION IN GUATEMALA

451

comings remain, however, including grossly inadequate provisions for internal discipline and control. The government has largely ignored criticism by the United Nations, although some progress was made in the functioning of the academy. There has been little effort by the GCE to make its programs conditional on government compliance with the accords or responsiveness to UN criticism. In a few instances, GCE officials appear to have assisted the government in sidestepping requirements of the accords.58 The U.S. International Criminal Investigations Training Assistance Program (ICITAP), in turn, was accused by officials within MINUGUA of attempting to assert unilateral control over the development of the investigative department of the PNC, to the point of attempting to force the removal of two MINUGUA instructors assigned to the department. 59 With such disunity among international donors and verifiers, the Guatemalan government was able to proceed without regard for the accords or the lessons of other police development projects, including those in El Salvador. The government, for its part, seemed obsessed with managing the transition from the old forces to the new without allowing a vacuum to develop. While this is in fact an important consideration and one that reflects lessons learned in El Salvador, the transition could have been managed without resorting to measures that appear likely to produce a police force little better than the one it is replacing. Socioeconomic issues. Extreme socioeconomic inequalities were among the causes of the civil war. Despite having lower-middle-income status according to World Bank criteria, Guatemala has exceptionally adverse social indicators.60 Seventy-five percent of Guatemalans live below the poverty line; 58 percent (and 81 percent of the indigenous population) are extremely poor.61 With the exception of Brazil, Guatemala has the highest Gini coefficient of inequality (59.6) among middle-income countries. The lowest 20 percent of the population receives 2.1 percent of income, and the two lowest quintiles combined receive only 7.9 percent of income, while the top 20 percent receive 63 percent of income and the top 10 percent receive 46.4 percent of income.62 Land distribution is extremely concentrated, and production has emphasized export products, contributing to a high degree of food insecurity.63 Throughout the post–World War II period, infant mortality rates have been very high, as have rates of malnutrition and illiteracy.64 More than half of the women and 38 percent of the men are illiterate.65 According to the UNDP, the Human Development Index scores for most indigenous communities in Guatemala are equal only to those of Haiti and some places in Africa and Asia.66 Major landholders have been politically powerful in Guatemala. A land reform in 1953 and 1954 was reversed following the U.S.-orchestrated coup of 1954, and only token efforts at land reform have occurred since then. With the growth of cattle production in the 1970s, confrontations

S-15

11/24/09

452

4:24 PM

Page 452

CASE STUDIES

between peasants and ranchers (some of whom were senior military officers) became increasingly common, particularly as ranchers encroached on frontier areas colonized earlier by peasants.67 Bitter land disputes persist, as does the practice of powerful individuals seizing lands over which they have no legal claim. Historically, Guatemala has been hostile to organized labor, thanks to restrictive labor laws and the practice of murdering labor leaders. The Guatemalan economic elite has consistently opposed taxation, and enjoys one of the lowest tax rates in the Western Hemisphere; the government often is unable to collect even the lenient taxes. As a consequence, the state is chronically underfunded, and has little capacity to carry out programs that could alleviate poverty or improve social conditions for the majority of the population.68 The URNG’s ambitions for the Agreement on Social and Economic Aspects and the Agrarian Situation (hereinafter the SE Accord) far outstripped its bargaining leverage. The accord falls short of dealing with the extreme inequalities in land distribution, and rejects any sort of expropriation-based land reform. What the accord does propose are state-assisted, market-oriented mechanisms with limited mandates, limited funding, uncertain jurisdiction, and questionable future political support. In the SE Accord, the government agreed to raise the budget allocated to education and health as a percentage of GDP by 50 percent by the end of 1999; strengthen the financial solvency of the social security (public health) system; devote 1.5 percent of tax income to public housing initiatives; reallocate health spending so that 50 percent of total health spending goes to preventive care; create a lands trust fund to regain control over and subsequently redistribute state lands; increase peasants’ access to credits, technical assistance, and marketing; grant 100,000 hectares of state lands to poor and middle-income peasants; carry out $50 million per year of infrastructure improvements in rural areas; simplify, decentralize, and update the land registration and titling system, while carrying out a new survey of agricultural lands; restore lands usurped from individuals and community, or failing that, establish rates and mechanisms for compensation; pass a new municipal property tax law with a new rate scale that taxes underutilized lands more heavily and provides a market-based assessment of property values; create an overall progressive tax system; and finally, raise the overall tax burden as a percent of GDP by 50 percent by 2000.69 The accord also called for the creation of multiple mechanisms to increase popular participation in policy formation on SE issues, as well as a series of new institutions designed to increase the points of access for citizens with needs and grievances. The timetable for implementation of these measures was generally long, giving the government until the end of 1999 to proceed on most points, with a few intermediate benchmarks.70 Annual targets for tax revenues are listed in the accord for 1997

S-15

11/24/09

4:24 PM

Page 453

PEACE IMPLEMENTATION IN GUATEMALA

453

through 1999, but these were later rescheduled for 2000–2003. In 1997 the government managed to slightly exceed the benchmark by introducing two temporary tax measures, but for 1998 and 1999, tax revenues fell short. A crucial turning point on implementation of the tax commitment was the government’s aborted attempt in 1997 to establish a new land tax, which would replace existing, flawed land-tax statutes.71 The new law would have increased state revenues, increased municipalities’ share of that revenue, and taxed unused lands more heavily. The government succeeded in negotiating the tax with the private-sector elite and Congress passed the law. Shortly thereafter, legal challenges and sporadic protests in rural areas prompted the government to repeal the law. The protests by themselves were not sufficient in magnitude to justify canceling the law, so most likely some sort of elite opposition had developed. Some opposition from peasants appears to be based on fears that the tax would affect them negatively. The government clearly failed to educate the public regarding the provisions of the law. A broad-based discussion resulted in the signing of a Fiscal Pact in May 2000 that included some sixty agreements related to tax issues and budget transparency. However, a year later, MINUGUA found little progress had been made, with only two agreements implemented, and another seven well under way. One of these measures, the increase from 10 to 12 percent in the value-added tax, was widely scorned by both labor and the private sector, which organized a nationwide strike. In early 2002, MINUGUA predicted that the accords’ rescheduled target tax-to-GDP ratio of 12 percent would once again fall short, to 10.6 percent in 2002. Overall governmental credibility suffered throughout 2001 and 2002 due to a plethora of corruption scandals involving high-level officials. Without progress on the tax front, all other measures the government takes to implement the SE Accord are unsustainable. By 1998, government expenditures to implement elements of the peace accords were equivalent to 75 percent of tax revenue. Were it not for extensive international funding of these programs, the state would be left with only 25 percent of its revenues for all other state activities.72 International donors forged a united front around the need to increase tax revenue. However, following the virtual failure to move forward on the fiscal pact, donors have grown increasingly skeptical about Guatemala’s commitment to generating its own funds. Other issues in the SE Accord are potentially explosive and the government has moved with predictable caution. The accord seems to approach land distribution as if it were largely a technical problem of providing adequate documentation, establishing legal rights, and facilitating market transactions through increased financing. Yet such a sanitized view ignores the role of power, corruption, and violence in establishing both de facto and legal status of many landholdings in Guatemala. Observers estimate that some 40 percent of landholdings are not registered,73 and many of the property holdings that are registered are probably registered to individuals and

S-15

11/24/09

454

4:24 PM

Page 454

CASE STUDIES

corporations whose claims are based, somewhere in the past, on forceful usurpation, corrupt legal maneuvering, or laws designed to expropriate land from certain groups of people, especially the Mayans. In light of this, the government’s commitment to carry out a comprehensive process of land surveying, re-registration, titling, and resolution of conflicting claims is fraught with practical difficulties and political hazards. So is its potentially open-ended commitment to restore lands usurped from individuals, communities, municipalities, and the state. The government formed its Inter-Institutional Commission for the Development and Strengthening of Land Property Rights (PROTIERRA), which includes a technical-juridical unit working to lay the basis for a national cadastral survey. But problems abound, and the government has few incentives to solve these quickly, since the sooner PROTIERRA is ready for a full-scale survey, the sooner the government will face serious political backlash from landed interests. It would be relatively easy for opponents of any modernization or legalization of the land tenure system to whip up fear and hysteria among the hundreds of thousands of small holders who hold tenuous, irregular title to their land.74 One success on land issues is the Rural Development Bank (BANRURAL), designed to finance small-scale agriculture. Its board of directors includes representatives of small producers’ associations, and it has attracted numerous depositors and proven agile in delivering credit in small amounts. Indigenous rights and identity. The rights of indigenous peoples, their relationship to the state, and their identity within multicultural but Hispanicdominated Guatemalan society are one of the central challenges of the peace process. Scholars and observers have debated the role of indigenous grievances in generating the civil war: regardless of whether these issues drove the violence, they remain one of the central causes of social conflict, and one of the ways in which Guatemala does not conform to emerging international norms. The Agreement on Identity and Rights of Indigenous Peoples lays out a series of general desiderata and specific goals to address the racism, discrimination, segregation, poverty, and cultural vulnerability faced by the indigenous community. The parties agreed that the indigenous people have been particularly subject to de facto levels of discrimination, exploitation and injustice, on account of their origin, culture and language and that, like many other sectors of the national community, they have to endure unequal and unjust treatment and conditions on account of their economic and social status. . . . Until this problem affecting Guatemalan society is resolved, its economic, political, social and cultural potential will never be able to develop fully and neither will it be able to take the place in the community of nations due to it by virtue of its ancient history and the spiritual grandeur of its peoples.

S-15

11/24/09

4:24 PM

Page 455

PEACE IMPLEMENTATION IN GUATEMALA

455

As a starting principle, the government undertook to “promote a reform of the Constitution in order to define and characterize the Guatemalan nation as being of national unity, multi-ethnic, multicultural and multilingual.”75 This principle was to be realized by a series of proposed reforms that would officially recognize indigenous languages; promote the use of indigenous languages in schools, provision of social services, and the courts; outlaw discrimination against indigenous people; recognize and permit the application of indigenous “customary law”; recognize indigenous religious beliefs and practices; increase indigenous influence in local governance; and address past grievances regarding usurped lands and other resources. These initiatives were to be implemented under guidelines established by special commissions comprising representatives of the government and indigenous organizations. Both the government and the United Nations commented in very positive terms about the value of the dialogue these commissions have allowed between the government and Mayan groups. Some of the commissions were fairly successful in producing draft legislation, while others have stalled over insurmountable differences on basic issues. Many of the points included in the accord are controversial, prohibitively costly, or in other ways impractical. The indigenous groups involved in the commissions tried to make the most of the opportunity to have input into the process, pressing quite challenging demands for expanded rights and benefits, often ignoring the very limited capacity of the Guatemalan state, as well as the likely intensity of political opposition to some of the proposals. The promise to address land grievances is also unlikely to be fully implemented. As written, the accord could be taken to hold out the promise of massive restoration of usurped lands. The Mayan umbrella organization COPMAGUA, represented on the Parity Commission on Land Rights, proposed that the government begin the new cadastral survey in the areas where massive land usurpations have been most common—the southern coast, the “Northern Transverse Strip” of the Huehuetenango, Quiché, and Alta Verapaz provinces, and the remote Petén. COPMAGUA expected the survey to produce significant restoration of lands, and was prepared to pressure the government to make good on its commitments.76 Instead the government began the survey at a very small scale in areas where land disputes are smaller in scale, and emphasized sorting out overlapping claims rather than making major corrections. The existence of the indigenous accord, with its provisions for extensive consultation and participation through the various commissions, has contributed to increased indigenous political mobilization and activism. This in itself is an important achievement given the historical exclusion and political alienation of indigenous peoples in Guatemala. But the unlikelihood that the agreement will ever be implemented in a way that addresses the depth of conflict and difference between the indigenous and ladino

S-15

11/24/09

4:24 PM

456

Page 456

CASE STUDIES

communities may make this expanded participation little more than another lesson in the futility of interacting with the national state. Accountability for past human rights crimes. The Commission for the Historical Clarification of Human Rights Violations and Acts of Violence That Have Caused Suffering to the Guatemalan Population (CEH) was to clarify such acts with “objectivity, equity, and impartiality” and provide recommendations to the state. The CEH was not to “individualize responsibilities” or have any judicial effect. The head of the CEH was Christian Tomuschat, a German law professor and former UN independent expert on human rights for Guatemala. The other two members of the commission were Guatemalans, Otilia Lux Cotí, a Mayan educator, and Alfredo Balsells Tojo, a prominent lawyer. The CEH was unable to commence operations with the signing of the final peace accord because of funding and logistical problems. It finally began work in August 1997, and asked for two six-month extensions, for a total of eighteen months. The time was needed to collect testimony and other evidence relating to acts of violence during the entire armed conflict (from 1961 through 1996). The commission received 8,000 testimonies, but also received little cooperation from Guatemalan military authorities, in contravention of the accord. The conclusions to the final report presented on February 25, 1999, shocked the Guatemalan political landscape.77 The commission’s basic findings—that the military was responsible for 93 percent of the total human rights violations and other acts of violence it documented—were not unexpected. But the charges of genocide and racism committed by the armed forces in their ruthless campaign against the guerrillas in the early 1980s came as a surprise, issued as they were from a UN-sponsored effort. The commission also issued a series of recommendations relating to reparations for war victims and reforms to the judicial system and security apparatus, and proposed that Congress create a foundation, with government and civil society participation, to monitor the implementation of these recommendations. However, the government refused to implement any of the key reforms proposed by the commission, saying that important reforms were already under way, including reparations for war victims.

Conclusion In some ways, the Guatemalan peace process has already succeeded. The war is over, the URNG has become a political party, and there is essentially no chance of renewed civil warfare. Though there are occasional reports of new politico-military groups, there is no evidence of sufficient organizational base or resources for such bands to have a significant impact. But

S-15

11/24/09

4:24 PM

Page 457

PEACE IMPLEMENTATION IN GUATEMALA

457

stopping the fighting and demobilizing the few remaining URNG guerrillas was not the primary challenge: rather, the more complex objectives of political and social change laid out in the accords were the more difficult achievement. Against the standard that the accords themselves set, the peace process is stagnated, and significant elements of its provisions may never be implemented. Two central features of the Guatemalan peace process have contributed to this mixed outcome. First, the accords have more the form of comprehensive peace accords than the substance. They touch on a sweeping range of issues, propose lofty democratizing goals, but don’t reflect a broadbased consensus even among elites. Moreover, the accords lack the details needed to give both domestic and international actors a basis for verification. Second, to fill in the missing details, the accords called for a broad and many-faceted process of consultation, and depended heavily on burdensome constitutional reform procedures. Neither the PAN nor the FRG governments, not to mention the URNG, have been strong or united enough to deliver political support to elaborate and implement the accords. Consensus has been difficult to develop and sustain, and the government has been unable to overcome conservative opposition, from both inside and outside the formal political system. Throughout the negotiations and implementation, the parties to the accord have shown ambivalence regarding popular participation in the process: participation has often been very selective, yet when participatory mechanisms proved cumbersome, the parties have tried (often unsuccessfully) to avoid being constrained by them. In the end, the Guatemalan accords are little more than an undemocratic elite pact, without the efficiency and decisiveness that elite-based agreements sometimes bring. For international implementers and donors, the Guatemalan process presents a number of dilemmas. Having encouraged, mediated, and advised the parties, the key international actors such as the UN and the Friends of the peace process have felt an obligation to carry through in supporting the process, despite its limited achievements to date. Donors continued to support the PAN and FRG governments even if their support for the peace process at times rang hollow, since the alternative would have been to abandon the process entirely. Following the voters’ rejection of the constitutional reform package, it became even more difficult for international donors to withhold aid, although it did help focus their efforts around fiscal reform. The government’s failure to increase tax revenue, however, has already dampened donor enthusiasm: only a little over half of the $2.3 billion in international assistance promised Guatemala in 1997 has been disbursed, having slowed dramatically in 2000 and 2001.78 There is a certain irony to this outcome, given that obtaining foreign assistance and investment was among the government’s main incentives for signing the accords in the first place.

S-15

11/24/09

4:24 PM

458

Page 458

CASE STUDIES

Among the main lessons from the difficulties of implementing the Guatemalan process is that international peacemakers should pay close attention to the relative bargaining power of the two sides. If one side (especially the government) is very weak, it may not make much sense to endorse negotiation of multidimensional peace accords. Unless there is sufficient leverage by the opposition to extract significant, concrete concessions, the result is likely to be vague accords that produce little substantive change and that are difficult to effectively verify. International mediators should also recognize that there are definite costs, during the implementation phase, to the inclusion of elaborate schemes for further consultation and specification of measures to be taken after the signing of accords. These mechanisms—while obviously necessary to broaden ownership and develop societal consensus—will slow implementation significantly, and risk distributing responsibility so broadly that it is difficult for international verifiers to apply pressure to get things done. This issue points to an important distinction between implementation of peace agreements and what might be called normal democratic political process. Peace agreements are not democratic, but may lay foundations for democratic processes. To lay those foundations, and to prevent the stronger of the warring parties from simply dominating the postwar environment, peace implementation needs to move quickly, rearrange political institutions so that newly incorporating elements of the polity have sufficient guarantees, and exploit the very short attention span of international donors and political actors. Peace implementation is disruptive of the existing order: it creates a context in which old institutions (such as repressive police forces) can be dismantled and new ones created. Normal democratic political processes, in contrast, are generally slow-moving, evolutionary in changing institutional arrangements, and dependent on having reasonably equitable and accessible institutions already in place. If the lines are blurred between peace-plan implementation and normal democratic process, the result is a hybrid that is unlikely to produce positive results. Implementation will have too little momentum to transform entrenched institutions, and international support and pressure for change will quickly evaporate. Moreover, all the popular participation in a hybrid model will take place within a context that is still so inequitable that popular political efficacy is impossible. Guatemala appears to have become just such a hybrid, and international peacemakers should probably avoid others.

Notes 1. This from a population of approximately 7 million in 1980, and 11 million in 1997. World Bank, World Development Report 1998 (Washington, D.C.: World Bank, 1998), p. 194. Since most of the deaths occurred in the few years before and

S-15

11/24/09

4:24 PM

Page 459

PEACE IMPLEMENTATION IN GUATEMALA

459

after 1980, we can use the 1980 population estimate for calculating the percentage of the population killed during the conflict, roughly 2 percent. 2. Report of the UN Verification Mission in Guatemala (MINUGUA) for the Consultative Group meeting for Guatemala, January 18, 2002. 3. See David Holiday, “Guatemala’s Precarious Peace,” Current History (February 2000): 78–84. 4. See MINUGUA, Informe del Secretario General de las Naciones Unidas sobre la verificación de los acuerdos de paz en Guatemala (15 de Abril–31 de Diciember de 1997) (Guatemala: MINUGUA, February 1998); and MINUGUA, Informe del Secretario General de las Naciones Unidas sobre la verificación de los acuerdos de paz en Guatemala (1 de Enero–31 de Julio 1998) (Guatemala: MINUGUA, October 1998). See also UN Document A/51/936, “Misión de verificación de las Naciones Unidas en Guatemala, informe del Secretario General,” June 30, 1997. 5. See Edelberto Torres-Rivas and Gabriel Aguilera Peralta, Desde el autoritarismo a la paz (Guatemala: Facultad Latinoamericana de Ciencias Sociales [FLACSO], 1998), pp. 62–65. 6. Ibid., pp. 62–68. 7. On the relationship between Mayans and revolutionaries, see, for example, Edgar Ruano Navarro, Esbozo historico del movimiento guerrillero guatemalteco, 1962–1972, typescript, Guatemala 1998; Torres-Rivas and Aguilera Peralta, Desde el autoritarismo a la paz; Mario Payeras, Los días de la selva (Guatemala: Casa de las Americas, 1980); and Mario Payeras, Los pueblos indigenas y la revolución guatemalteca: Ensayos étnicos (Guatemala: Luna y Sol, 1997). 8. See Michael McClintock, The American Connection, vol. 2, State Terror and Popular Resistance in El Salvador (London: Zed Books, 1985). 9. See Piero Gleijeses, Shattered Hope (Princeton: Princeton University Press, 1993). 10. Héctor Alejandro Gramajo Morales, Tesis de la estabilidad nacional (Guatemala: Editorial del Ejercito, 1989). 11. For detailed discussions, see Torres-Rivas and Aguilera Peralta, Desde el autoritarismo a la paz; and Jennifer Schirmer, The Guatemalan Military Project: A Violence Called Democracy (Philadelphia: University of Pennsylvania Press, 1998). 12. See David Dye, unpublished typescript on Guatemalan military, Managua, 1998. 13. See Juan Alberto Fuentes, ed., Guatemala: Los contrastes del desarollo humano (Guatemala: Sistema de las Naciones Unidas en Guatemala, 1998), pp. 76–83. 14. See Chapter 13 by Caroline Hartzell in this book for details on Esquipulas. 15. The Guatemalan government was apparently reluctant to use the term “mediation” to refer to the UN’s role, in part in an effort to differentiate the Guatemalan process from that in El Salvador. We will use here the uncommon terms “moderation” and “moderator” following their use in Guatemalan discourse, to refer respectively to the UN’s mediation role and to the UN’s mediator, Jean Arnault. 16. The accords are available in English from the UN Public Inquiries Office, and at www.usip.org/library/pa/index. 17. This interpretation is confirmed by an internal URNG document dated July 12, 1994, “Memorandum de la CG [Comandancia General] de la URNG,” copy in authors’ possession. 18. Author interview, URNG commander Rodrigo Asturias, Guatemala, June 1998. 19. See Tania Palencia Prado and David Holiday, Towards a New Role for

S-15

11/24/09

460

4:24 PM

Page 460

CASE STUDIES

Civil Society in the Democratization of Guatemala (Montreal: International Centre for Human Rights and Democratic Development, February 1996). 20. See J. Fernando Valdéz and Mayra Palencia Prado, Los dominios del poder: La encrucijada tributaria (Guatemala: FLACSO, 1998) p. 339. 21. Author interview, former defense minister Julio Balconi, Guatemala, June 1998. 22. Author interview, Colonel Otto Noack, Guatemala, January 1998. 23. Serrano had headed the Council of State during Ríos Montt’s military government of 1982–1983. 24. See Tania Palencia Prado, Peace in the Making (London: Catholic Institute of International Relations, 1996), p. 15. 25. “Memorandum de la CG de la URNG.” 26. See Rachel McCleary, The Constitutional Crisis in Guatemala: The Responses of the International Community and Guatemalan Society (Washington, D.C.: U.S. Institute of Peace, January 1994). 27. See Héctor Rosada-Granados, El lado oculto de las negociaciones de paz (Guatemala: Friedrich Ebert Stiftung/Proyecto Nacional Guatemala, Hacia la Paz, 1998), pp. 75–89. 28. The significance of this concession by the government cannot be overstated. Just several weeks prior to the signing of this agreement, one top military officer told a Western diplomat that Guatemala would “never” accept the presence of an international verification mission. Author interview, Guatemala, June 1994. 29. According to former government negotiator Héctor Rosada-Granados, this limitation was unavoidable, since the CEH could not be part of the judicial organ of the state. Any attempt to give it prosecutorial powers would render it unconstitutional. See Rosada-Granados, El lado oculto de las negociaciones de paz, p. 93. 30. Author interview with UN officials, New York, January 1995. 31. This sounds more dramatic on paper than it is in practice, given the extremely low government expenditures at the time the accord was signed. 32. For its part, the URNG announced mid-1996 that they would restrict their activities in collecting war taxes. 33. Author interviews with Tania Palencia and Héctor Rosada-Granados, Guatemala, May 1999. 34. Constitutional reforms require the support of two-thirds of Congress, followed by a majority vote in a popular referendum. 35. If one discounts the capital-city vote, two of the four ballot questions (those related to social and cultural rights, and the reform of the judiciary) would have passed. 36. According to the Guatemalan constitution, political parties or organizations can offer constitutional reforms if they obtain 5,000 signatures from registered voters. 37. Article 20 of the 1985 constitution prohibits persons who have been de facto heads of previous governments from running for office, a provision explicitly aimed at individuals like Ríos Montt, who had governed the country through military coups. 38. These included a university vice rector (and indigenous expert on education), a banker, a cooperative leader, and an economist who had worked for international organizations. 39. According to one account, no reliable evidence has emerged linking Arnault personally to MINUGUA’s failure to investigate. At the time of the disappearance of José Cabrera Rodas (“Mincho”), Arnault as moderator of the negotiations was under orders to keep his activities separate from those of MINUGUA,

S-15

11/24/09

4:24 PM

Page 461

PEACE IMPLEMENTATION IN GUATEMALA

461

although he was a logical target of these accusations given his earlier knowledge and subsequent position as chief verifier of the accords. See Jack Spence et al., Promise and Reality: Implementation of the Guatemalan Peace Accords (Cambridge, Mass.: Hemisphere Initiatives, August 1998), pp. 59–62. 40. Alvaro de Soto and Graciana del Castillo, “Obstacles to Peacebuilding in El Salvador,” Foreign Policy 94 (spring 1994): 69–83. 41. Inforpress, “Guatemala: Refugee Returns Near End,” Central America Report 25, no. 9 (March 5, 1998): 4–5. 42. Spence, Promise and Reality, p. 21. 43. Much of the analysis presented here is attributable to Paula Worby, a longtime observer of the Guatemalan refugee return. The section authored by her in Spence, Promise and Reality, provides further detail on this issue. 44. Inforpress, “Guatemala,” p. 4. 45. Ibid. 46. Author interview, UN official, Guatemala, June 1998. Anonymity requested. 47. Author interviews with former defense minister Julio Balconi, Guatemala, May 1998; retired Major Mauricio López Bonilla, May 1999; and URNG leader Rodrigo Asturias, May 1998. 48. MINUGUA, Informe del Secretario General (1 de Enero–31 de Julio 1998). 49. See Comisión Especial de Incorporación (CEI), Balance del programa de incorporación de la URNG a la legalidad (Guatemala: CEI, 1997); MINUGUA, Informe del Secretario General (15 de Abril–31 de Diciember de 1997); MINUGUA, Informe del Secretario General (1 de Enero–31 de Julio 1998); and UN Document S/1997/432, “Report of the Secretary-General on the Group of Military Observers Attached to MINUGUA,” June 4, 1997. 50. The government’s obligations were laid out in the Agreement on the Strengthening of Civilian Power and on the Role of the Armed Forces in a Democratic Society. The government’s only obligations under the Agreement on the Definitive Ceasefire were to relocate their forces, allow UN inspection of facilities, and limit movements to those who would not impinge on the URNG’s assembly points or security zones. 51. See Spence, Promise and Reality, pp. 28–29. 52. See MINUGUA, Informe del Secretario General (15 de Abril–31 de Diciember de 1997), p. 7. 53. See MINUGUA, Informe del Secretario General (1 de Enero–31 de Julio 1998), p. 19. 54. See Torres-Rivas and Aguilera Peralta, Desde el autoritarismo a la Paz. 55. Agreement on Strengthening Civil Power 1996, para. 23. 56. Ibid., para. 10. 57. Author interview with PNC official, Guatemala, May 1999. 58. Author interviews with UN officials, Guatemala, June 1998. 59. Author interview with UN and U.S. officials, Guatemala, June 1998. 60. Guatemala shares this status with several other Latin American countries, including Bolivia, Ecuador, the Dominican Republic, El Salvador, Paraguay, Colombia, Peru, Costa Rica, Panama, and Venezuela. 61. Spence, Promise and Reality, p. 47. 62. World Bank, World Development Report 1997 (Washington, D.C.: World Bank, 1997), p. 222. 63. Charles D. Brockett, Land, Power, and Poverty: Agrarian Transformation and Political Conflict in Central America, 2nd ed. (Boulder, Colo.: Westview Press,

S-15

11/24/09

462

4:24 PM

Page 462

CASE STUDIES

1998); and Robert G. Williams, Export Agriculture and the Crisis in Central America (Chapel Hill: University of North Carolina Press, 1986). 64. Ibid. 65. World Bank, World Development Report 1997, p. 226. 66. Fuentes, Guatemala, pp. 11–23. 67. Williams, Export Agriculture and the Crisis in Central America. 68. Gustavo Porras, address at conference titled “Un año de la firma de la paz en Guatemala” (Guatemala: FLACSO, 1997); and Valdéz and Palencia Prado, Los dominios del poder. 69. Agreement on Social and Economic Aspects and Agrarian Situation. Summary drawn mainly from Spence, Promise and Reality, p. 52. 70. See Rosalinda Hernández Alarcón, The Land Issue in the Peace Accords: A Summary of the Government’s Response (Guatemala: Inforpress Centroamericana, 1998), p. 16. 71. MINUGUA, Suplemento sobre la verificación del acuerdo sobre aspectos socioeconómicos y situación agraria (Guatemala: MINUGUA, October 1998). 72. MINUGUA, Suplemento sobre la verificación, p. 7. 73. Spence, Promise and Reality, p. 49. 74. Ibid., p. 55. 75. Agreement on Identity and Rights of Indigenous People, pt. 4, sec. A. 76. Spence, Promise and Reality, p. 42. 77. See Mireya Navarro, “Guatemalan Army Waged ‘Genocide,’ New Report Finds,” New York Times, February 26, 1999. The actual report was available within weeks on the Internet (http://hrdata.aaas.org/ceh), but it was not published until some six months later. 78. “Peace and Development for All,” document prepared by the Guatemalan government for the Consultative Group meeting, Guatemala, December 19, 2001.

S-16

11/24/09

4:25 PM

Page 463

16 Implementing the Arusha Peace Agreement on Rwanda GILBERT M. KHADIAGALA

The Hutu-dominated government of Rwandan president Juvenal Habyarimana and the Tutsi-led Rwandan Patriotic Front (RPF) signed the Arusha Peace Agreement in August 1993 in an internationally sponsored effort to end Rwanda’s civil war, which began in October 1990. Nudging the combatants toward the Arusha Agreement was a host of regional and international actors, convinced about the wisdom of a peaceful settlement to avert ethnic conflagration. Efforts to implement the agreement, however, faltered from the start and finally collapsed in April 1994 when the death of Habyarimana was followed by the massacre of more than 1 million Tutsi and moderate Hutu in a three-month-long rampage. The genocide marked the end of four years of international intervention in the mediation and implementation of the Arusha Agreement. Rwanda’s genocide resulted in part from the weakness of the Arusha Agreement and the paltry efforts to implement it. As a product of a deeply polarized society, the Arusha Agreement faced severe obstacles. First, the agreement challenged the dominant ethnic basis of power without creating a solid alternative in the short term, and thus caused an extremist reaction. Second, the agreement’s internal fragility was compounded by an unstable regional environment that had a debilitating effect on the implementation process. Third, both the mediators and the parties to the Arusha Agreement staked its implementation almost exclusively on international actors who were unwilling and unprepared to expend resources required to meet its provisions. Saddled with a precarious power-sharing agreement with uncertain outcomes, the parties unrealistically sought deliverance from a lessthan-benign external environment.

463

S-16

11/24/09

4:25 PM

Page 464

464

CASE STUDIES

Background to the Crisis The colonial era set the stage for most of the contemporary conflicts in Rwanda. Although Tutsi and Hutu shared a common language and traditions in a precolonial centralized kingdom, the status differences between a pastoralist Tutsi aristocracy and a subject Hutu peasantry were solidified by the colonial powers—first Germany and then Belgium—into a rigid caste system that favored the elite minority Tutsi. As the demands of the colonial state strengthened Tutsi feudal hegemony, so did the power gap widen between Tutsi rulers and Hutu commoners. However, socioeconomic change during the colonial period produced a Hutu elite that was the center of Hutu political consciousness throughout the 1950s. Coalesced as the Movement for the Emancipation of the Hutu (Parti du Mouvement de l’Émancipation des Bahutu, PARMEHUTU), these elites led the bloody revolt that overthrew the Tutsi monarchy in 1959, culminating in the deaths of an estimated 10,000 Tutsi and causing an additional 130,000 to flee into exile in neighboring countries. In July 1962, after a series of elections, Rwanda attained independence under a government based on a power-sharing agreement between the PARMEHUTU and progressive Tutsi. 1 Decolonization inaugurated an era of endemic communal violence and massive movement of refugees. In the early 1960s, Tutsi refugees in neighboring countries attempted several armed invasions to seize power, effectively ending the postindependence power-sharing agreement. In 1963, following a rebel attack from Burundi, between 10,000 and 14,000 Tutsi lost their lives in revenge killings, and thousands more joined the ranks of refugees. By 1964 the UN High Commission for Refugees (UNHCR) estimated that 200,000 Tutsi were in the region, half of them in Uganda. As threats of a Tutsi invasion receded in the late 1960s, political conflicts emerged among Hutu and class dissension arose in the face of centralization of power in new Hutu elites.2 In addition, regional differences arose between Hutu ruling elites in the south and central heartland and those to the north, and these differences became the principal axes of conflict in the 1970s and 1980s. The first Hutu president, Grégoire Kayibanda, drew his support from the southern Hutu, who came to dominate all facets of the state and economy, causing resentment among northern elites who suffered economic marginalization.3 Widespread massacres of Hutu in Burundi in 1972 reignited ethnic tensions in Rwanda and led to reprisals against the Rwandan Tutsi. These tensions provided the pretext for a military coup staged by Army Chief of Staff Habyarimana in July 1973. Although he promised to restore national unity, Habyarimana erected a one-party state, Mouvement Révolutionnaire National pour le Développement (MRND). Habyarimana’s rule marked a shift in power from southern to northern Hutu elites and from the civilian to the military. As the Habyarimana government grew more authoritarian, mil-

S-16

11/24/09

4:25 PM

Page 465

THE ARUSHA PEACE AGREEMENT ON RWANDA

465

itary recruitment and promotion veered more narrowly toward the president’s northern kinsmen, creating what Claudine Vidal calls a “fourth ethnicity.”4 Intra-Hutu cleavages were heightened by the inequitable access to resources between the ruling groups and the predominantly peasant majority. With the highest population density in mainland Africa (256 persons per square kilometer), Rwanda typifies the dilemma of overpopulation and resource scarcity magnified by a severe dependence on coffee production. Between 1963 and 1993, the population rose from around 3 million to 7.5 million at an average growth rate of 3.7 percent per annum. Demographic pressure brought about overexploitation of land, 72 percent of which is devoted to agriculture. The additional pressures from deforestation and soil erosion gradually decreased productive land available for cultivation, subjecting the peasants to constant threats of famine. By the second half of the 1980s, with economic growth rates falling behind a burgeoning population, the government admitted that it could only feed 5 million people. The economic situation worsened the class and regional polarization and contributed to the general weakening of the Habyarimana state.5 In 1989, Rwanda faced a harsh famine stemming from a prolonged drought. At the same time, worldwide prices for coffee plunged as much as 50 percent, decimating the major source of government revenues. The impact of the price collapse was demonstrated in the sharp drop of coffee export earnings from $150 million in 1986 to $70 million in 1989. 6 Following this economic crisis, the government announced a 40 percent budget reduction, resulting in deep cuts in social services. World Bank data indicate that aggregate GDP per capita decreased from an estimated $335 in 1983 to $250 in 1990, and the incidence of poverty increased from 40 percent in 1985 to 53 percent in 1992.7 Economic decline exacerbated the class distinctions and ruptured the postindependence myth of a Hutu “social revolution” that Habyarimana had assiduously cultivated. Adding to these internal strains was external pressure for democratization, which gathered momentum in early 1990 with the French threat to link development assistance to liberalization. Growing external pressure also galvanized domestic opposition groups to demand political reforms, leading to strikes and demonstrations by university students in June 1990. In response, Habyarimana agreed to separate the party and the state in July 1990. As southern Hutu elites appropriated the label of democracy to agitate for genuine power sharing, Habyarimana appointed a commission in September 1990 to work out a National Political Charter that would allow the establishment of different political parties.8 The widespread perception of state paralysis that occurred amid Habyarimana’s internal reforms provided an opportunity for the RPF to strike from Uganda in October 1990. As Gerard Prunier notes: “By the early fall of 1990 the Rwandese political scene was one of deep and perva-

S-16

11/24/09

466

4:25 PM

Page 466

CASE STUDIES

sive crisis. . . . The crisis acted as a multidimensional spur on the RPF preparations in Uganda. . . . [Reports] told the same story: the Rwandese political system was on the verge of collapse and any strong push from the outside would complete the process.”9 Three weeks after the invasion, the government announced austerity measures mandated by the World Bank and International Monetary Fund (IMF) structural adjustment program. These policies entailed devaluation of the currency and more cutbacks in government spending on education, healthcare, and the civil service. Structural adjustment measures plunged the majority of Rwandans into the insecurity of unemployment and landlessness, alienating a population considerably weakened by the economic shocks of the late 1980s. These conditions further deteriorated in the course of the civil war, as meaningful economic activity ground to a halt and thousands of people lost their lives or were internally displaced.10 Rebel demands coalesced around a program calling for restoration of citizenship rights, national unity, and an end to a dictatorial “system that generates refugees.”11 The desire for return by the Tutsi refugees arose from the fact that, while Habyarimana’s government had rhetorically advocated their return, it did little to facilitate this process, instead invoking Rwanda’s demographic pressure to induce neighboring states to naturalize the refugees. In exile, RPF units that had been part of Uganda’s National Resistance Army (NRA) took the initiative, at the opportune instance of regime weakness, to force the issue of return. In the initial phase of the conflict, the RPF launched a conventional campaign, scoring quick victories by overrunning government defenses in northern Rwanda. Heavy casualties, in particular the loss of its three senior military commanders, overshadowed these successes.12 The Habyarimana government depicted the invasion as a bid by forces that sought “to restore a minority and feudal regime which was abolished in 1959 under the guise of liberation and democracy.”13 It also invited foreign military support from its traditional allies, Belgium, France, and Zaire (now the Democratic Republic of Congo). Meeting Rwanda’s emergency appeal for military help, Belgium sent 535 troops and France sent 300 troops ostensibly to protect their nationals in Rwanda. Zaire’s Mobutu Seso Seko dispatched about 1,000 troops, which were deployed in direct combat against the RPF.14 External support enabled the Rwandese army to inflict heavy casualties on the RPF. Soon, however, Zaire and Belgium withdrew their troops, leaving France as the principal supporter of the government. French military commitment to the government included the provision of troops and military advisers, and the supply of heavy weaponry such as armored personnel carriers, reconnaissance vehicles, communications equipment, and helicopters. More important, France provided financial guarantees for purchases of small arms, mortars, and grenade launchers

S-16

11/24/09

4:25 PM

Page 467

THE ARUSHA PEACE AGREEMENT ON RWANDA

467

from Egypt and South Africa.15 Military assistance and training enabled Habyarimana to boost the government army, Forces Armées Rwandaises (FAR), which grew from 5,200 in October 1990 to 15,000 by mid-1991, and to 30,000 by the time the Arusha negotiations began in June 1992. In addition, following the RPF invasion, the government distributed assault rifles to municipal authorities, working in collaboration with the ruling party’s armed militias, the Interahamwe (“those who fight together”) and Impuzamugambi (“those who have the same goal”), creating an armed militia of between 20,000 and 30,000.16 While France armed the Habyarimana regime, Uganda did the same for the RPF. Uganda was indispensable to the RPF in geographical, military, and political terms, roles that stemmed from the RPF’s long-term alliance with Ugandan president Yoweri Museveni’s National Resistance Army (NRA). Throughout the war, the RPF benefited from Uganda’s supply of arms, food, and gasoline, as well as use of Uganda’s southern border as a military launching pad and place of refuge. Moreover, in the face of condemnation for Uganda’s logistical and material support, Museveni remained a dependable RPF ally in regional and international diplomatic circles. From its northern operational base, the RPF under its new leader, Paul Kagame, reorganized its forces, shifting from conventional to guerrilla warfare at the end of 1990, and the RPF attempted to broaden its ethnic and political base by recruiting Hutu. Partly from these efforts, the RPF grew from 4,000 in 1990 to about 25,000 by early 1994.17

Negotiations: Cease-Fire, Refugees, or Power Sharing? Foreign military assistance in October 1990 allowed Habyarimana to contain the RPF advance, but he could not dislodge the RPF from its operational bases in northern Rwanda, the most productive agricultural region. The civil war compounded the food crisis by disrupting Rwanda’s northern transport corridor through the Kenyan port of Mombasa. Budgetary allocations underscored the mounting costs of the war on the economy: in 1990 military expenditures accounted for 15 percent of the budget, but by 1993 they had risen to an astronomical 70 percent of the state’s operating expenses.18 Habyarimana embarked on a regional offensive in the aftermath of the invasion to isolate the RPF by bringing pressure on Uganda, which the government repeatedly described as the “source of the invasion.”19 This strategy entailed mobilizing the support of Belgium and its Western allies to use their economic muscle on regional actors, and enlisting Tanzania and Zaire as mediators between Rwanda and Uganda. Three weeks after the invasion, Belgium sent its prime minister, Wilfried Martens, to East Africa to per-

S-16

11/24/09

468

4:25 PM

Page 468

CASE STUDIES

suade regional leaders to prevail on Museveni. Martens proposed the creation of a regional intervention force to supervise a cease-fire, with Europe furnishing logistical and financial support.20 Belgian diplomatic intervention led to regional consultations on a cease-fire and the establishment of a peacekeeping force under the leadership of President Hassan Mwinyi of Tanzania. Habyarimana tried to use Tanzania’s influence over Museveni to curtail Uganda’s logistical and military support to the RPF. At a summit meeting on October 17, 1990, in Mwanza, Tanzania, Habyarimana promised to “initiate a dialogue with both the internal and external opposition under the auspices of the SecretaryGeneral of the Organization of African Unity [OAU] to end the conflict.” This dialogue, however, was predicated on Tanzania and Uganda persuading “the armed opposition (RPF) to observe a cease-fire . . . that should be monitored by neutral troops mutually agreed upon by the parties concerned.”21 Couching the cease-fire in terms of the RPF’s withdrawal from Rwandese territory enabled Habyarimana to define the conflict as one of Ugandan aggression. Marginalized from the Mwanza deliberations, the RPF scoffed at the cease-fire arrangement, vowing to continue the war: “We can’t withdraw from our positions,” one of its leaders explained. “It is totally unacceptable that we can leave our own country. Most of us are deserters from the Ugandan army, and we can’t go back to Uganda.”22 And according to Museveni: “We won’t let the Rwandan rebels back into Uganda, and I hope that they don’t want that either.”23 Zaire’s Mobutu Seso Seko embarked on a series of diplomatic initiatives to break the impasse of the cease-fire. The first regional summit, on October 26, 1990, in Gbadolite, Zaire, proposed a cease-fire agreement to be supervised by a fifteen-member OAU Neutral Military Observer Group (NMOG) drawn from Zaire, Burundi, Tanzania, and Uganda. A subsequent meeting in Goma on November 20 restated the terms of the Gbadolite agreement.24 Mobutu’s mediation created a negotiating process without altering fundamentally the government’s view of a cease-fire as a prelude to the RPF’s withdrawal back to Uganda. During these talks, the government either refused to participate or sent representatives who had no negotiating mandate.25 The government’s unyielding stance was demonstrated when it scuttled the deployment of OAU military observers, insisting on stationing a military team on the Rwanda-Uganda border to prevent further incursions by the RPF.26 With no appreciable movement on a cease-fire, Habyarimana’s Western allies stepped up pressure on regional states, highlighted by regional visits of French and Belgian ministers in November and December 1990. In early 1991, Habyarimana shifted his diplomatic efforts from the quest for a cease-fire to what he now described as the “regional crisis of Rwandese refugees,” as a way to undermine the RPF’s primary objectives of refugee return and power sharing. The proposal for a “regional” solution

S-16

11/24/09

4:25 PM

Page 469

THE ARUSHA PEACE AGREEMENT ON RWANDA

469

seemed designed to reinforce the status quo and received tacit backing from the French. The conference on Rwandese refugees in Dar es Salaam, Tanzania, in February 1991, brought together Rwanda’s neighbors, the OAU, and the UNHCR. Habyarimana promised to remove all obstacles impeding the voluntary return and reintegration of refugees. In addition, he pledged to grant general amnesty to all refugees who had committed crimes against the state after concluding a cease-fire agreement with the RPF. In return, Uganda, Burundi, Tanzania, and Zaire agreed to naturalize and integrate refugees who opted to settle outside Rwanda.27 Excluded from the Dar es Salaam talks, the RPF derided the final agreement and demanded direct dialogue as a way out of the civil war.28 Following the Dar es Salaam conference, Mobutu reinvigorated his efforts to find a durable cease-fire, culminating in an agreement at N’sele, Zaire, on March 29, 1991, that proposed the immediate cessation of hostilities as a prelude to negotiations on power sharing. At N’sele, the Habyarimana government, for the first time, agreed on direct negotiations with the RPF, lending legitimacy to the latter’s quest for recognition.29 Although military pressure had yielded the RPF’s recognition as a serious negotiating partner, it was not sufficient on its own to lead to a negotiated settlement. Political changes engendered by Habyarimana’s internal reforms played an equally significant role in the maturation of the conflict. The RPF invasion forced Habyarimana to hasten political reforms that had begun in July 1990, and in June 1991 the legislature approved a multiparty constitution with executive power shared between the president and the prime minister. These constitutional changes emboldened the growing domestic opposition, which saw an opportunity to press for power sharing.30 The political changes had mixed impact on the course of the conflict and subsequent negotiations. Political reforms gradually opened up the political system, challenging the ruling northern Hutu’s grip on power, and as moderate political parties tried to straddle the Hutu-Tutsi divide, they became the RPF’s allies in the struggle for democracy. However, the ruling elite, which could barely muster support beyond its northern confines, now used the specter of the invasion to strengthen its power by appealing for Hutu ethnic solidarity against perceived Tutsi onslaught. Habyarimana began to rely increasingly on an exclusive group of military and political institutions for his authority, primarily the akazu (little house), a network of personal friends, and the presidential guard. Thus, at the start of powersharing negotiations, political liberalization had transformed the previously bilateral conflict between the government and the RPF into a genuinely multilateral competition, but it also solidified a conservative alliance that perceived a negotiated outcome to be inimical to their power.31 The onset of pluralism led to the emergence of five major political par-

S-16

11/24/09

470

4:25 PM

Page 470

CASE STUDIES

ties in addition to Habyarimana’s MRND: Mouvement Démocratique Républicain (MDR), Parti Social Démocrate (PSD), Parti Libéral (PL), Parti Démocrate Chrétien (PDC), and Coalition pour la Défense de la République (CDR). Over time, the main moderate parties, the MDR, PSD, and PL, formed a loose alliance, the Democratic Force for Change, while the progovernment radical party, the CDR, allied with the MRND in the Alliance for the Reinforcement of Democracy. In the fall of 1991, Habyarimana began negotiations with opposition parties on a transitional government of national unity pending multiparty elections.32 As the economic strains from the guerrilla war escalated, prominent members of moderate civilian opposition parties held secret discussions with the RPF in a bid to force the government to open direct power-sharing negotiations. In a major breakthrough in April 1992, the major political parties signed a protocol that established a coalition transitional government, with the MDR’s Dismas Nsengiyaremye as the prime minister. As part of this agreement, Habyarimana conceded to the start of negotiations with the RPF.33 By June 1992, the eighteen-month-long political and military stalemate had been joined. With no military solution in sight, demands for negotiations became irresistible. Furthermore, internal reforms intended by the government to nullify the RPF’s demands for democracy had altered the domestic balance of power by gradually circumscribing Habyarimana’s options in negotiations. As Filip Reyntjens writes, the internal reforms had created alliances critical to the negotiations: The Arusha negotiations were held between quite different partners. On the one hand, the RPF was united and showed a clear sense of purpose. On the other, the Rwanda government delegation included the old regime and the internal opposition, that is, two partners with different agendas. As the negotiations were essentially about power sharing, the internal opposition concluded an objective alliance with the RPF in order to weaken the former single party as much as possible.34

Mediating the Arusha Peace Agreement Tanzania, under the broad mandate of the OAU, mediated the Arusha peace talks from their inception in June 1992 to their conclusion in August 1993. To provide both a semblance of multilateralism and additional sources of leverage, representatives from France, Germany, Belgium, the United Nations, and the United States attended the talks as observers. Similarly, Burundi, Uganda, Zaire, and the OAU sent observers to Arusha to give the talks an African imprimatur and demonstrate regional interest in a settlement. The multilateral nature of the Arusha talks provided a wider international context of power and reassured the disputants who were distrustful of

S-16

11/24/09

4:25 PM

Page 471

THE ARUSHA PEACE AGREEMENT ON RWANDA

471

each other. In reality, however, the essentially token participation of major Western countries and the United Nations from the outset helped shape expectations about the extent of international commitment to Rwanda’s peace process. Multiple actors at the negotiating table exaggerated the mediators’ and the parties’ perceptions of the international community’s ability to remain engaged and mobilize resources beyond the negotiations. During the negotiations, inordinate expectations about international actors weighed heavily in decisions about the resources for implementation, postwar economic reconstruction, and refugee resettlement. In June 1992 the mediators first focused on reaching agreement on a durable cease-fire as a means to build confidence. The urgency of a ceasefire was demonstrated by an RPF military offensive in several northern communes intended to ensure its strong position on the eve of the peace talks. This attack drove the Rwandese army from strategic positions, displaced some 350,000 civilians, and led to the deployment of additional French troops.35 Previous attempts to establish a cease-fire in the N’sele Agreement of March 1991 provided a blueprint for the negotiations, facilitating a quick agreement on establishing a fifty-five-member OAU NMOG from Mali, Nigeria, Senegal, and Zimbabwe. The task of the NMOG was to create a buffer zone between RPF-held and government-held areas in the north. The cease-fire agreement also created a Joint Political Military Commission of representatives from Burundi, Tanzania, Uganda, Zaire, Belgium, France, and the United States to oversee the work of the NMOG and coordinate the implementation of a final peace agreement. Following the cease-fire agreement, Uganda and Rwanda began parallel steps to ease tensions along their common border, culminating in a security agreement on joint border monitoring in August 1992.36 With the cease-fire holding and the NMOG structures in place, the challenge to Tanzanian mediators was to hasten power-sharing talks to secure against the resurgence of animosities. At the top of the agenda of the second round of negotiations, which started on August 10, 1992, was the establishment of the rule of law based on national unity, democracy, and respect for human rights. Tanzanian mediators saw an agreement on these broad principles as a way to commit the parties to an “ideological set of values” that would form the basis for negotiating the details of powersharing procedures within the framework of a broad-based transitional government. The signing of a protocol on the rule of law and establishment of democratic rule on August 18 enabled the parties to proceed to the third round of negotiations on the issue of power sharing. Negotiations for the substantive power-sharing protocols, which began in September 1992, revealed the precariousness of Habyarimana’s coalition government. Internal conflicts in Kigali affected the negotiations because coalition partners had to reach decisions on the government’s positions before submitting them to Arusha. Apart from conflicts in the coalition

S-16

11/24/09

472

4:25 PM

Page 472

CASE STUDIES

government creating a paralysis in Arusha, fragmentation in the government delegation along the factional fissures in Kigali hampered the pace of the negotiations. For the mediators, it was much easier to control events at the negotiating table than in Kigali.37 The major issue of contention in the parties’ proposals for power sharing was the structure for managing a transition period before elections, particularly the powers of the president. While the government was determined to keep intact the coalition government along the lines of the June constitution of 1991, the RPF proposed a transitional government based on new institutions. Seeking to control the transition without devolving central power, the government proposed a twofold division of presidential powers: presidential prerogatives that would remain unshared and nonnegotiable; and executive prerogatives that would be shared with the prime minister and be subject to negotiations. The RPF, however, proposed the complete removal of executive power from the presidency, contending that “democracy cannot prevail if the president of the Republic remains with excessive powers. These powers should be shared out.”38 As the talks proceeded haltingly throughout September and October 1992, virulent conflicts between allies of Habyarimana and the prime minister, Nsengiyaremye, surfaced and became the barometers of the government’s overall commitment to a negotiated settlement. In a memorandum issued in early October 1992, moderate civilian parties accused Habyarimana and extremist political forces of obstructing the peace talks. Backing this position, the RPF complained that the delays in Arusha stemmed from an inadequate mandate by government negotiators and urged the government to confer more authority to the negotiators.39 Habyarimana’s supporters, on the other hand, accused the prime minister of entering into an alliance with the RPF to undermine the authority of the state. In a petition to Habyarimana in October 1992, members of the MRND and the CDR complained about “the unacceptable attitude of the prime minister who, instead of doing what is necessary to bring the positions of his team members closer in order to reach decisions that would reflect a consensus, has opted to giving the Rwandan delegation in Arusha unlimited powers to decide on behalf of the government on matters which are as delicate as they are important for the future of the country.”40 The mediators made efforts to reconcile the various positions over power sharing, yielding a compromise on the institutions of a future transitional government. The protocol on power sharing within the framework of a Broad-Based Transitional Government, signed on October 30, 1992, comprised the constitutional provisions that became the heart of the Arusha Agreement. Its provisions included the following: Habyarimana would remain president during the transition, but most of his powers would be shifted to a coalition cabinet led by a prime minister; as the head of government, the prime minister would exercise executive power through a consen-

S-16

11/24/09

4:25 PM

Page 473

THE ARUSHA PEACE AGREEMENT ON RWANDA

473

sus in the cabinet; a transitional national assembly would be established to assist the cabinet in governing; and a judiciary independent of the executive and the legislature would be established. The goals of the transitional government were to implement democracy, consolidate national security and reconciliation, repatriate refugees, and establish the mechanisms for postwar reconstruction.41 With the signing of the power-sharing protocol, the balance of power in the coalition government shifted toward MRND and CDR extremists, wary of the consequences of stripping Habyarimana of his power. At the end of October and early November 1992, Habyarimana’s supporters organized street demonstrations against the Arusha negotiations and its supporters, events that unleashed counterattacks that paralyzed commercial activities in Kigali. At the same time, as the RPF uncovered preparations by MRND’s armed militias to exterminate political opponents, Habyarimana stoked the flames of extremism by denouncing the Arusha protocols as “scraps of papers” and appealing to Interahamwe to give him full support.42 As the political crisis in government and the escalating violence threatened to sink the negotiations, Tanzanian mediators brought the power of regional heads of state to bear on the parties. During a summit meeting between Museveni and Mwinyi in late November 1992, both publicly supported the Arusha Protocols and renewed their call for an urgent negotiated solution to the conflict.43 During the resumption of power-sharing talks in November, the bone of contention became the government’s insistence on the inclusion of the CDR in transitional institutions. The RPF, however, insisted that since the CDR was not a formal political party, it did not deserve representation. The RPF’s position prevailed, paving the way for the signing of protocols on the division of seats in the cabinet and Parliament in December 1992 and January 1993. 44 The MRND and the CDR immediately repudiated the power-sharing protocols, charging that they had granted enormous power to the RPF and its civilian supporters.45 Habyarimana supported this position, stating that the agreement was unlikely to be implemented because the negotiations “did not take into account the proposals of the majority of Rwandans . . . the agreements did not come out of a consensus from the government side. . . . Instead of its directives being taken into account by the Rwandan delegation, the delegation gets directives from institutions which sometimes have no authority.”46 Habyarimana’s denunciation of the power-sharing protocols gave the armed factions of the MRND and the CDR the upper hand in launching war on proponents of negotiations. From mid-January 1993, Kigali descended into chaos as MRND supporters crippled commercial and government activities, demanding the prime minister’s resignation. In communities beyond Kigali, government death squads attacked Tutsi and moderate Hutu in killings foreshadowing the genocide that engulfed Rwanda a year later. In February 1993, as the government demanded negotiations to revise the

S-16

11/24/09

474

4:25 PM

Page 474

CASE STUDIES

protocols previously agreed upon, the RPF broke the cease-fire and launched a large-scale offensive against government troops, doubling the territory under its control and advancing to a few miles from Kigali; French authorities saved Habyarimana from the RPF advance by sending more troops. Although the number of French troops swelled to at least 680, comprising four companies, including paratroopers, this deployment could not undo the RPF’s military gains.47 The failure of the cease-fire was a turning point in the conflict, testing not only the military capabilities of both parties, but also, more critically, the unity of the multiparty political consensus that had sustained the Arusha negotiations. Two factors had thus far propelled the peace process: the mediators’ use of the military stalemate to arrange an uneasy but successful cease-fire, and the moderate coalition’s desire for peace. Habyarimana’s abrogation of the protocols, followed by organized ethnic massacres and the RPF’s breach of the cease-fire, highlighted the gulf between negotiations and the parties’ commitment to a peaceful settlement. The military confrontation created a propitious moment for the mediators to prod the parties back to negotiations. Yet, as Prunier points out, the RPF’s military offensive widened ethnic polarization, fraying the alliance between moderate Hutu opposition parties and the RPF. Although most of the opposition groups had ignored the MRND’s campaign of portraying the RPF solely as Tutsi seeking to seize power, the February attack stoked fears of the RPF’s motives.48 Extremism found fertile ground in the rural population suffering the economic consequences of the civil war. In March 1993 an international commission report estimated that since 1990 a total of 900,000 people, or 13 percent of the population, had been displaced by the war.49 Growing Hutu extremism in early 1993 induced internal fragmentation in otherwise united political parties, particularly the MDR, PSD, and PL. As radical factions emerged in the mainstream moderate opposition parties, the latter were unable to mount a countervailing challenge to forces of ethnicity and militarism. In later phases of the peace process, extremists used the fragmentation in moderate parties to raise doubts about the wisdom of ethnic reconciliation and the establishment of the transitional institutions. On a visit to Kigali in February 1993, the French minister for cooperation, Marcel Debarge, fueled extremism by asking the main opposition parties to “make a common front” with Habyarimana against the RPF.50 Seen in the context of French military reinforcements, this statement reignited the public animosities between France and the RPF that the Arusha process had partially tamed.51 Before the Arusha talks resumed, Tanzania organized a high-level summit in Dar es Salaam on March 7, 1993, to obtain the parties’ commitment to conclude a peace agreement. The Dar es Salaam summit reconstituted the cease-fire and agreed on the mechanisms for the gradual withdrawal of

S-16

11/24/09

4:25 PM

Page 475

THE ARUSHA PEACE AGREEMENT ON RWANDA

475

French troops and their replacement with an international monitoring force. The UN Security Council adopted Resolution 812 on March 12, 1993, authorizing the formation of an international force “under the aegis of the OAU and the United Nations entrusted with the protection of, and humanitarian assistance to, civilian population and [to] support the OAU force for the monitoring of the cease-fire.” A UN reconnaissance team recommended that this mission, the UN Observer Mission Uganda-Rwanda (UNOMUR), be deployed on the Uganda side of the border to prevent the resupply and reinforcement of the RPF.52 Creating UNOMUR boosted the role of international actors in the peace process, strengthening the already growing perception about the UN’s commitment to Rwanda. In negotiations between March and June 1993 dealing with the integration and formation of a new national army, the resettlement of refugees, and the details of the transitional arrangements, the government strongly favored a UN force in the implementation stage. Even after the SecretaryGeneral sent two military experts in May 1993 to provide technical assistance to the OAU in preparation for donor funding of an expanded OAU NMOG, Habyarimana derided the OAU’s role, telling a Belgian newspaper that he preferred UN troops.53 Despite occasional skirmishes, the NMOG under General Ekundayo Opaleye of Nigeria reconstituted a buffer zone and repositioned troops in the summer of 1993. Demonstrating the improvement of security in the conflict zone, negotiations on demilitarization of the buffer zone and resettlement of refugees were moved to Kinihira, in northern Rwanda. During the negotiations, however, the debate tilted decisively toward a neutral international force with more extensive responsibilities than the NMOG. Thus, in early June 1993, Tanzania informed both parties of the decision by the United Nations and the OAU that the UN would assume responsibility and command of the force. In response, the government and the RPF in a joint letter to the United Nations on June 14, 1993, suggested a rapid UN deployment after the signing of the agreement to “permit its speedy implementation and, in particular, the establishment of a broad-based transitional government thereby avoiding excessively long intervals, which might be detrimental to the peace process.”54 By June 24, 1993, when the negotiators had completed all outstanding issues and the parties were ready to sign the agreement, the Rwandese cabinet suddenly reversed course, pleading for a postponement to allow sufficient time to study all protocols. In a pattern reminiscent of the quarrels over the protocols on power sharing, Habyarimana objected to the provisions that provided for equitable distribution of military command posts, a move that would lead to the progressive erosion of his power.55 Habyarimana’s options for avoiding the final agreement were narrowing due to mounting international and regional pressure. Having suspended all but humanitarian assistance since early 1993, donor nations threatened to

S-16

11/24/09

476

4:25 PM

Page 476

CASE STUDIES

halt all funds if Habyarimana continued to procrastinate. The OAU warned Habyarimana of the “ominous consequences for Rwanda and the region” caused by delay in reaching a peace agreement, and reiterated the “need and maximum urgency for a peace agreement to be concluded as soon as possible.”56 This concerted pressure enabled the parties to narrow their remaining differences on military command structures and transitional arrangements in time for the signing of the Arusha Peace Agreement on August 4, 1993. Both sides pledged to “eradicate the deep-seated causes which gave rise to the war,” but Habyarimana was a reluctant signatory.57 Despite the pattern of defiance and denial, military and economic pressure had delivered Habyarimana to the table and kept him there. Political liberalization and international forces had constrained Habyarimana’s room to maneuver. From this perspective, the Arusha Agreement was a temporary triumph of political forces in the coalition government and the RPF. Adelman and Suhrke characterize it as a political, rather than a peace, agreement.58 After signing the agreement, the moderate coalition of parties faced the formidable problem of maintaining unity and coherence in the untested terrain of implementation.

The Arusha Peace Agreement The Arusha Peace Agreement incorporated six protocols signed by the parties over the twelve months of talks: the N’sele cease-fire agreement, the rule of law, power sharing, repatriation of refugees, integration of armed forces, and miscellaneous provisions. Central to the transition was a multiparty Broad-Based Transitional Government (BBTG) and Transitional National Assembly (TNA), which would be established thirty-seven days after the signing of the agreement. Before their establishment, the agreement allowed Habyarimana’s government to retain power on condition that it would neither usurp the mandate of the BBTG nor introduce new legislation. The transitional institutions would supervise local, parliamentary, and presidential elections in 1995, twenty-two months after the establishment of the BBTG.59 The agreement stated that Rwanda’s 1991 constitution and the Arusha Peace Agreement would form “indissolubly” the Fundamental Law that would govern the country during the transitional period and that the agreement was in fact superior, because, as the act indicated, “in case of conflict between the other provisions of the constitution and those of the Peace Agreement, the provisions of the Peace Agreement shall prevail.” Similarly, the provisions of the Fundamental Law would prevail if they came into conflict with other laws and regulations. The Fundamental Law embraced principles that the framers hoped

S-16

11/24/09

4:25 PM

Page 477

THE ARUSHA PEACE AGREEMENT ON RWANDA

477

would restrain the competitive interests of the disputants as they structured long-term cooperative relationships. Captured in the protocol on the rule of law were principles of national unity, democracy, pluralism, and respect for human rights. National unity was predicated on the Universal Declaration of Human Rights, which precluded economic and political discrimination. Under democracy and pluralism, the protocol recognized the sovereign rights of people to form associations, join political parties, and elect leaders of their choice: “Elections shall be organized in such a way that transparency is guaranteed and fraud eliminated through the establishment of efficient supervision mechanisms, including if the need arises, enlisting the assistance of International Observers.” In seeking to end human rights violations, the protocol proposed the establishment of an independent commission of human rights to “investigate human rights violations committed by anybody on Rwandese territory, in particular by organs of the state and individuals in their capacity as agents of the state or of various organizations.” There was an additional provision to establish an international commission of inquiry to investigate the human rights violations committed during the war. The protocol on power sharing within the framework of the BBTG laid the foundation for the agreement. This protocol pruned presidential power, consigning executive power to collegial decisions between a multiparty cabinet and a prime minister. The agreement named Faustin Twagiramungu, head of the smaller of the two MDR factions, as the prime minister, to take office after the installation of the BBTG. As the source of effective power, the prime minister was required to prepare the government’s program, lead government business, determine the functions of the head of ministries, and appoint civil servants. The compromises over the numerical division of seats among political parties created a cabinet of twenty-one ministers allocated as follows: five cabinet seats, including the minister of defense, to the MRND; five to the RPF; four to the MDR, including the prime minister and the minister of finance; three to the PSD; three to the PL; and one to the PDC. The MRND would retain the presidency, while one of the five RPF seats would bear the title of deputy prime minister. There was a similar factional balancing in the seventy-member TNA, which would consist of representatives chosen by the respective political parties: eleven members each from the MRND, MDR, PSD, PL, and RPF, four members from the PDC, and one member each from the other recognized parties. The power-sharing protocol comprised a political code of conduct for all participants in transitional institutions that enjoined them to support the agreement, promote national reconciliation, and reject exclusionary ideologies. The provision of a strong cabinet system operating through a consensus sought to insulate decisionmaking from the one-party dominance that would potentially delay or disrupt the transition. Yet multi-

S-16

11/24/09

478

4:25 PM

Page 478

CASE STUDIES

party cooperation depended on a modicum of stability within each political party. The bitter splits in moderate parties that had begun in early 1993 were later to provide entry points for peace opponents to derail the delicate multiparty arrangement.60 Conspicuously absent from the power distribution was the CDR, which, although a participant in the coalition government, had opposed the agreement from the start. Its leadership refused to sign the code of conduct for inclusion in the TNA and warned that implementation of the agreement would “bring even more bloodshed.”61 In response to the perception that the RPF had obtained a larger share of seats in the cabinet, the signatories held an olive branch to the CDR and other extremist parties by offering them participation in the TNA after they had demonstrated their commitment to abide by the principles governing the Protocol of Agreement on the Rule of Law, indicated their support for the peace process, and agreed to avoid engaging in sectarian practices and in any form of violence. The framers provided for an independent judiciary that would exercise its powers through the courts, tribunals, and other jurisdictions. One of the most important provisions of the protocol on power sharing was the designation of a commission for national unity and reconciliation. This commission was to launch a national dialogue on unity and prepare information on socializing the population about reconciliation. In addition, the protocol spelled out a program of postwar national rehabilitation that would provide humanitarian assistance, rebuild areas devastated by war, and stimulate the economy by promoting food security and diversifying exports. Upon the assumption of power, the transitional government was required “as a matter of urgency and priority . . . to rid the administrative apparatus of all incompetent elements as well as authorities who were involved in the social strife or whose activities are an obstacle to the democratic process and to national reconciliation.” The transitional institutions sought to balance national reconciliation and effective government. The transition was to be a deliberate learning phase of power sharing in which the values of interethnic accommodation would gradually take hold. In addition, an orderly transition promised to establish a foundation for a more durable constitutional structure. Beyond the transition period, the protocol on power sharing laid down the procedures for writing a new constitution. The TNA would appoint a constitutional commission of experts that would, after extensive consultation with all the strata of the population, prepare a preliminary draft constitution. After review by the government and the legislative branch, the draft constitution was to be submitted to a national referendum for adoption. The BBTG’s detailed constitutional provisions offered a map to navigate an uncertain political future, but in the short to medium term their implementation hinged on solid cease-fire arrangements. The cease-fire provisions incorporated the N’sele cease-fire agreement and the protocol on

S-16

11/24/09

4:25 PM

Page 479

THE ARUSHA PEACE AGREEMENT ON RWANDA

479

the integration and demobilization of armed forces and the gendarmerie and the establishment of a Neutral International Force that would “assist in the implementation of the Peace Agreement, more especially through the supervision of the implementation of the Protocol of Agreement on the Integration of Armed Forces of the two parties as well as the provision of all kinds of assistance to the competent authorities and organs.” The protocol envisaged that demobilization, disengagement, and integration of a new army would be completed within seven to nine months. Under this plan, there would a new integrated army of 19,000 men, divided on a 60–40 percent basis in favor of the government. Command posts were to be shared equally on an alternating basis, meaning that while each would have an equal number of brigade and battalion commanders in the five units of the army, “neither force would be entitled to occupy simultaneously the posts of commander and deputy commander of any given formation.” The demobilization and gradual integration of the rest of the soldiers in civilian life were to start with the voluntary publication of lists of soldiers targeted for demobilization, the establishment of assembly points, monetary payments to individual soldiers, and the formation of a secretariat for rehabilitation and social integration. Throughout this exercise, the International Force was to demarcate assembly points, including the establishment of an expanded Demilitarized Zone (DMZ), create the demobilization procedures to supervise the disengagement of forces, and train the new armed forces. The protocol on integration of armed forces spelled out the reforms of the gendarmerie, including the demobilization of approximately 6,000 gendarmes. Subsequently, the UN force would screen all candidates for positions in a new gendarmerie. The rest would, like the surplus army, be demobilized and reintegrated into society. Similarly, the International Force was to assist in maintaining public security and monitoring the activities of the gendarmerie and communal police. To verify that law and order were effectively and impartially maintained, the protocol proposed that a UN police force, led by a police commissioner, be deployed in Kigali and the nine prefectures. These meticulous details ignored the problem of financing the estimated 35,000 demobilized soldiers from both armies. Moreover, it glossed over the long-term question of absorbing soldiers into a constrained civilian labor market. Hartmut Diessenbacher observes: “As non-inheriting farmer’s sons, they [the soldiers] had found in the army a secure existence and a purposeful outlet for their youthful energies. Released into the hopelessness of an over saturated labor market, they could become a destabilizing factor.”62 It was for this reason that soldiers, particularly senior ones who were to be demobilized first because of their age, opposed the agreement. The protocol also provided additional security roles for the Inter-

S-16

11/24/09

480

4:25 PM

Page 480

CASE STUDIES

national Force: to guarantee the overall security of the country, especially verifying the maintenance of law and order; to ensure the security of distribution of humanitarian aid; to assist in the tracking of arms caches and neutralization of armed gangs; to aid in the recovery of all weapons in the hands of civilians; to undertake mine clearance operations; and to monitor the observance of cessation of hostilities. The protocol stipulated that the Neutral International Force shall be “informed of any incidents or violation and shall track down the perpetrators.” In its initial role, the Neutral International Force was to provide security in Kigali, a task that would be linked to the formation of the BBTG and the TNA: “The establishment of transitional institutions shall take place after the deployment of the Neutral International Force or the Expanded NMOG and after the withdrawal of foreign troops.” Subsequent to the withdrawal of French troops according to the March 7, 1993, Dar es Salaam communiqué, the RPF was to send a 600-strong infantry battalion to Kigali to protect its political leaders participating in the transitional institutions. The protocol on the repatriation of refugees and the resettlement of displaced persons recognized the inalienable right of refugees to return home and settle in the region of their choice, so long as they did not encroach on the rights of others. The protocol stated that all refugees had a right to repossess their property on return, but it recommended that for the sake of social harmony and national reconciliation, “refugees who left the country more than 10 years ago should not reclaim their properties, which might have been occupied by other people. The government shall compensate them by putting land at their disposal and shall help them to resettle.” Like everything else in the overall agreement, the protocol assumed the transitional government would create a secretariat for state rehabilitation and social integration to deal with procedures for repatriation. Under this plan, the BBTG, on the recommendation of the secretariat, was to set up a commission for repatriation composed of government, UNHCR, OAU, and refugee representatives. Some of the tasks of the commission were to include conducting a survey of refugees, organizing a prerepatriation census, identifying centers for repatriation, and supervising assistance to returnees. Given the enormous costs entailed in resettlement, the protocol proposed that six months after the establishment of the BBTG, the OAU and the UNHCR would organize a donor’s conference on the financing of the repatriation program. The other significant element of the repatriation protocol was the return of persons displaced by war and social strife, a process that was to follow the deployment of UN peacekeepers, disengagement of forces in the war zones, the clearance of mines, and the provision of humanitarian assistance. While the BBTG would reconstitute the local administrative authorities in these zones, the secretariat for state repatriation and humanitarian agencies were slated to rehabilitate the socioeconomic infrastructures and

S-16

11/24/09

4:25 PM

Page 481

THE ARUSHA PEACE AGREEMENT ON RWANDA

481

ensure close coordination between the resettlement of refugees and displaced persons. Negotiations on repatriation of refugees were built on the framework of the Dar es Salaam refugee conference of February 1991 and on the expert advice received from organizations already involved in relief work. As a consequence, most of the protocol on repatriation was based on recommendations from diverse regional and external actors. For instance, Article 32 of the protocol conceded: “For the technical implementation of the various components of the repatriation program, the government of Rwanda and the UNHCR shall preferably resort to those NGOs with an established reliability, taking also their respective specialization into account. As such, one or several NGOs shall undertake site development activities, building activities, and the distribution of food aid.” Limited local resources dictated the dispersion of responsibilities for repatriation and resettlement, but this still begged the question of how much international goodwill would be marshaled for the enormous rehabilitation tasks.

Implementation The Arusha Agreement described the United Nations, the OAU, and Tanzania as the “primary third parties for implementation.” In the world of Africa’s resource constraints, however, this tripartite mandate meant that Tanzania and the OAU would play essentially secondary roles to the United Nations, particularly since the mediators and parties had, in June 1993, agreed on the UN’s responsibility and command of an international force. The notion that the talks had been a multilateral partnership for Rwanda’s peace and democracy undergirded the UN’s leadership role. The parties assumed that, having supported the talks and contributed to forging the outcome, multilateral actors would be prepared to devote additional resources to bring the process to fruition. The perception of a beneficent “international community” responsive to Rwanda’s transitional pains was emphasized by both parties from the start of the implementation. The RPF’s Alexis Kanyarwenge said, “Each step in the implementation of the Peace Agreement is dependent upon the presence of an International Neutral Force.”63 Habyarimana’s letter to the Secretary-General carried the same message: “I would appreciate steps that you may take for the rapid deployment of the neutral international force, with a view to the strictest respect for the Peace Agreement that has just been signed.”64 In the preliminary phase of the implementation, the gap between the parties’ optimism and the reality of international bureaucracy centered on the timing of establishing the UN Neutral Force. Conscious of the fragile nature of the settlement, the RPF and government had, in their joint letter to

S-16

11/24/09

482

4:25 PM

Page 482

CASE STUDIES

the Secretary-General on June 11, 1993, requested an accelerated deployment. As a result, the agreement set a thirty-seven-day timetable (that is, September 10, 1993) for establishing transitional institutions. In setting the timetable, the parties received prior warning about possible delays, according to Boutros Boutros-Ghali: “The United Nations made clear to the parties that the decision to deploy a United Nations force rested with the Security Council and that, if approved, the deployment could take up to three months. . . . The parties had none the less decided to adopt the accelerated timetable.”65 Given the relationship between UN deployment and transitional institutions, the edifice of the transition became essentially hostage to the unrealistic timetable. On August 19, 1993, the UN Secretary-General sent a reconnaissance mission to Rwanda led by General Romeo Dallaire, the chief military observer for UNOMUR, to examine the functions of what became the UN Assistance Mission in Rwanda (UNAMIR) and evaluate the human and financial resources required to carry them out. The RPF and the government had in June 1993 envisaged an extravagant reconnaissance mission that would “disengage the two forces involved in the conflict, . . . demarcate the assembly areas, identify the sites to be used as assembly points and billeting stations and also establish the security parameters for the city of Kigali as part of its neutralization in order to protect the transitional institutions.”66 Both parties, however, received a purely fact-finding mission that hastened to disabuse them of meeting the September 10 deadline.67 As an interim measure, the OAU Secretary-General extended the mandate of the NMOG and increased the number of observers from 55 to 132. At the same time, UNOMUR, which had deployed on the Uganda side of the border in mid-August 1993, became fully operational.68 The second area in which the expectations of the Rwandese parties diverged from the realities of decisionmaking at the United Nations was in the nature of resource commitment. The Arusha Agreement envisaged UNAMIR performing broad security roles including: to guarantee “overall security” in Rwanda, and especially to verify the maintenance of law and order by the competent authorities; to monitor the cease-fire and establish demobilization procedures; to “assist in tracking of arms caches and neutralization of armed gangs throughout the country”; to “assist in the recovery of all weapons distributed to, or illegally acquired by, the civilians”; and to “assist with mine clearance, including training and mine-awareness programs.” In addition to these military and security responsibilities, the Arusha Agreement expected UNAMIR to assist with the coordination of humanitarian assistance activities in conjunction with relief operations, and with repatriation and resettlement programs. The reconnaissance mission, however, weakened key provisions of UNAMIR’s mandate. Thus, instead of “overall security” in Rwanda, the

S-16

11/24/09

4:25 PM

Page 483

THE ARUSHA PEACE AGREEMENT ON RWANDA

483

mission suggested that UNAMIR “contribute to” security only in the city of Kigali: “Owing to the presence of several battalions of government forces in Kigali and the introduction into the city of the RPF leadership with a fully equipped RPF battalion, the Mission would establish a weaponssecure area in and around Kigali.”69 Outside Kigali, while UNAMIR’s role in monitoring the cease-fire and demobilization of troops was to remain intact, the mission proposed a reduction of the assembly points, cantonment points, and integrated training centers from a total of forty-eight, as suggested by the two parties, to twenty-six. With regard to maintaining civilian security through monitoring the activities of the gendarmie and communal police, the mission changed the mandate to “investigate and report on incidents regarding the activities” of the police. On the security components that provided for disarmament of militias and civilians, the mission, conscious of the difficulties caused elsewhere, simply ignored them.70 The scaling down of major provisions of the agreement inevitably led to a relatively small number of peacekeepers, compared to what the parties, and some UN military experts, assumed would have been needed to guarantee implementation of the agreement. UN military experts proposed 8,000 troops, or 5,000 at the very least, while the United States suggested 500. In the end, there was agreement on 2,548 troops, of which 2,217 would be staff officers and troops, and 331 military observers.71 The reconnaissance mission recommended the progressive deployment in four phases. The first phase provided for the deployment of an advance party of twenty-five military personnel, eighteen civilian personnel, and three civilian police after formal authorization by the Security Council. This force would secure Kigali and establish the “essential condition needed to permit the secure installation of the transitional Government,” which, the mission acknowledged, “may not be installed until the end of 1993.” Slated to take ninety days, this phase would also lead to the integration of UNOMUR and the NMOG in UNAMIR, concluding with a total military force of 1,428.72 The second phase, taking ninety days, was to focus primarily on disengagement, demobilization, and integration of the armed forces and gendarmie. The end of this phase would coincide with full deployment of the 2,548 military personnel. During the third phase, which would last about nine months, UNAMIR was to complete the integration of the forces and reduce its staff to approximately 1,240 personnel. In the final phase, lasting about four months, a reduced mission of approximately 930 military personnel would assist in securing the atmosphere required in the final stages of the transitional period leading up to the elections. Three weeks beyond the thirty-seven days (October 5, 1993), the UN Security Council passed Resolution 872 endorsing the plan of the reconnaissance mission and established UNAMIR for an initial period of six months. The UNAMIR author-

S-16

11/24/09

484

4:25 PM

Page 484

CASE STUDIES

izing resolution warned that its deployment beyond ninety days would be contingent on a determination that “substantive progress has been made toward the implementation of the Arusha Peace Agreement.”73 UNAMIR was created in a global context characterized by apprehensions about the wisdom of international peacekeeping. Boutros-Ghali took the lead in presenting the Arusha Agreement as an “opportunity for the international community to contribute to the successful implementation of the peace process,” and constantly warned that delays would “seriously jeopardize” the agreement.74 But in the Security Council, where it mattered most, Rwanda was not a high priority. Growing U.S. apprehension about funding peacekeeping operations was heightened by the killing of eighteen U.S. soldiers in Mogadishu, Somalia, two days before Security Council debate on UNAMIR began. Following the Somalia debacle, the Clinton administration came out with a new peacekeeping policy, termed Presidential Decision Directive 25, that stressed limiting the costs and risks of peacekeeping. 75 Human Rights Watch broadly indicts the Security Council: “Partly because they counted on an easy success, partly because they were not disposed to invest much in resolving the situation in Rwanda anyway, the Security Council failed to devote the resources necessary to ensure that the hard-won Accords were actually implemented.”76 Before the deployment of UNAMIR, the caretaker government led by Agathe Uwilingiyimana faced a volatile domestic situation that the expanded NMOG had neither the mandate nor the power to manage. Extremist opponents of the Arusha Agreement organized by the CDR, the MRND, and their militias redoubled their efforts to defeat the implementation of the agreement at three levels. First, their main propaganda instrument, the Radio Télévision Libre des Mille Collines (RTLMC), began broadcasting in August 1993 by openly calling on the population to reject the agreement and to prepare to fight against the installation of an RPF-dominated government. Second, the extremists increased the recruitment and training of militias in ostensibly “self-defense” programs as well as expanding the arming of the presidential guard and gendarmes. Third, the delay in setting up the transitional institutions furnished Habyarimana’s government with the breathing space to reorganize government ministries that the RPF was scheduled to assume. For instance, the government reorganized the powerful Ministry of Internal Affairs by redistributing its responsibilities to government agencies sympathetic to Habyarimana.77 Tanzania and the OAU participated in efforts for an effective peacekeeping force, but their roles diminished considerably once the United Nations made the decision on UNAMIR. As one of the steps to lessen bilateral mistrust, Uganda and Rwanda restored the northern transport corridor. Museveni, however, was less successful in inducing a rapprochement between Habyarimana and the RPF during a meeting in Kampala at the end of August 1993. As a result, despite the presence of UNOMUR on the

S-16

11/24/09

4:25 PM

Page 485

THE ARUSHA PEACE AGREEMENT ON RWANDA

485

Uganda border, the differences between Museveni and Habyarimana persisted.78 The Rwandese peace process was dealt a severe blow by the military coup in Burundi on October 21, 1993, in which the first democratically elected Hutu president, Melchior Ndadaye, was assassinated, tens of thousands of people died, and Hutu fled to Rwanda. That Ndadaye was killed by the Tutsi military only helped to inflame ethnic passions in Rwanda, giving opponents of power sharing more ammunition to discredit the Arusha Agreement. As Ian Linden notes, “The message of these events to many around Habyarimana was doubtless that the Tutsi would never genuinely accept [Hutu] majority rule within the context of a government of national unity.” 79 Most observers regard events in Burundi as the most important trigger to the unraveling of the agreement, because it undercut the moral and organizational positions of Hutu political parties that previously had been the core of the consensus for negotiations and power sharing. Already reeling from internal fractionalization engineered by the MRND and the CDR, the MDR, PSD, and PL were hard pressed to support the agreement.80 The first contingent of UNAMIR under General Dallaire arrived in Kigali at the end of October 1993 in the face of radical internal realignment and unfavorable regional conditions. UNAMIR became operationally effective at the end of December 1993 after the integration of the NMOG into UNAMIR and the creation of a weapons-secure area in Kigali manned by Belgian and Bangladeshi troops. With the deployment of nearly 1,300 peacekeepers, 400 Belgian soldiers ushered the RPF civilian leaders and 600 soldiers into Kigali, the first move in the establishment of the BBTG. Lurking in the background, though, were reports of escalating ethnic massacres conducted by Hutu militias. In November 1993, for instance, UNAMIR noted that the killing of civilians in the north was done by a “well-armed and reportedly ruthless group” that was trying to disrupt or even derail the peace process.81 And in early December 1993, peacekeepers received an anonymous letter from Rwandan army moderates warning that Habyarimana had conceived a “Machiavellian Plan” for countrywide political assassinations and ethnic massacres. These reports were boosted later that month by a disclosure by a Rwandan magazine of details of a “final solution” with “identification committees” to compile death lists and distribute arms to militias and civilians.82 In mid-December 1993, the special representative of the SecretaryGeneral in Kigali, Jacques-Roger Booh-Booh, held talks with the RPF and the government in which both sides issued a joint declaration reaffirming their commitment to the Arusha Peace Agreement and the installation of transitional institutions by the end of the same month. The SecretaryGeneral’s December 1993 report to the Security Council noted that the cease-fire had generally been respected, despite the potentially disruptive

S-16

11/24/09

486

4:25 PM

Page 486

CASE STUDIES

moments in the wake of the Burundi crisis and signs of increasing tensions. He recommended the early deployment of troops designated for the second phase of the operation, thereby increasing UNAMIR to its peak strength of 2,548 military personnel. In light of pressure within the Security Council to reduce the strength of UNAMIR, the Secretary-General warned: “I am convinced that, under the present circumstances, a reduction in the projected resource levels would negatively affect the performance and credibility of UNAMIR in discharging its mandate. It could also jeopardize the peace process in Rwanda.”83 Attempts to constitute transitional institutions, however, stalemated in late December 1993 as a result of severe disagreements over representation in the BBTG and the TNA. In a challenge to the power-sharing provisions of the Arusha Agreement, Habyarimana sought the nomination of members of extremist splinter factions of the MDR and PL in the cabinet and the National Assembly. In addition, Habyarimana demanded the revision of the provisions of the agreement that barred the CDR’s representation in the TNA. Having continually maneuvered to undercut the moderate coalition through aggressive appeals to Hutu ethnicity, Habyarimana was now ready to reap the benefits of a deeply divided opposition to determine the shape of the transitional government. But the RPF rejected these moves, claiming that they entailed renegotiating the agreement. As various factions in the MDR, PL, and PSD lined up either on the MRND or the RPF side, the moderate center that had restrained extremism faded. Thus, even though Habyarimana was sworn in as the interim president on January 5, 1994, the implementation process was essentially stalled. Predictably, the political impasse resulting from Habyarimana’s jockeying for political advantage contributed to the deterioration of the security situation, compounding the fragility of the implementation plan. UNAMIR had the mandate to contribute to the security of Kigali in collaboration with the local police, including enforcing the ban on weapons. Yet UNAMIR’s modest mandate could not meet the flagrant distribution of arms to militias and civilians by roving Hutu death squads responding to the RTLMC’s call to exterminate Tutsi. In January and February 1994, there was an explosion of violent demonstrations, roadblocks, assassination of political leaders, and murders of civilians. On January 11, 1994, a close confidant to Habyarimana reported to General Dallaire about a plot by the regime to marshal the Interahamwe to exterminate Tutsi in Kigali and about the intensified stockpiling of arms. This report led General Dallaire to seek permission from the Department of Peacekeeping Operations (DPKO) in New York to use military force. The government’s overall scheme, General Dallaire reported, was “to start a civil war” during which UN troops “were to be provoked, and if Belgian soldiers resorted to force, then a number of them were to be killed.” The DPKO, however, instructed General Dallaire to report this information to the ambassadors of France, Belgium, and the

S-16

11/24/09

4:25 PM

Page 487

THE ARUSHA PEACE AGREEMENT ON RWANDA

487

United States, and to plead with Habyarimana to investigate the matter. Furthermore, as Boutros-Ghali reported, “DPKO informed UNAMIR headquarters that such action went beyond the UNAMIR mandate which authorized the mission to contribute to the security of Kigali within a weaponsfree zone, but which also made it clear that such a zone had to be established ‘by’ the parties.”84 As violence escalated, the DPKO relented in early February 1994, authorizing the peacekeepers to assist Rwandese police, on a case-by-case basis, to recover illegal arms. In spite of this action, UNAMIR’s capacity to seize weapons and provide security was hampered by the rules of engagement that required it to work with local police, which was allied with the Interahamwe. UNAMIR’s glaring impotence forced General Dallaire to send another cable to New York on February 1994 warning that the success of the peacekeeping operation would be in jeopardy without prompt confiscation of arms stockpiled by the militia. He predicted “more frequent and more violent demonstrations, more grenade and armed attacks on ethnic and political groups, more assassinations and quite possibly outright attacks on the U.N. peacekeepers.”85 When General Dallaire’s requests for a robust mandate fell on deaf ears, Belgian authorities made pleas to the Security Council. The United States and the United Kingdom, however, vehemently opposed enlargement of the UNAMIR operation, citing the costs and risks of exceeding the original peacekeeping mandate. In a letter to the Secretary-General on March 14, 1994, Belgian foreign minister Willy Claes echoed General Dallaire’s concerns: “The Rwandese army appears increasingly annoyed by the parties’ procrastination, while information on the stockpiling of weapons by the various militias is becoming even more compelling. Even some of the leaders admit that a prolongation of the current political deadlock could result in an irreversible explosion of violence.”86 UN debates about boosting UNAMIR’s mandate were certainly not lost on Habyarimana and his allies. With the credibility and efficacy of UNAMIR in question because of Western reluctance to commit more resources, opponents of the peace agreement could bide their time, as was reflected in their delays on setting up transitional institutions. After being sworn in as the president of the transitional government in January 1994, Habyarimana seemed to be in no hurry to break the stranglehold over party nominations. In Gerard Prunier’s apt description, he “settled down again to his usual round of postponements, manipulations, denials and jerry-built pseudo-solutions in the hope of surviving, Mobutu-like, through a state of stagnating turmoil.”87 UN special representative Jacques Booh-Booh convened a series of all-party negotiations in February 1994 to try to narrow the differences regarding representation in the government. In a blunt message to the parties, Booh-Booh warned that unless Rwanda’s feuding parties showed seriousness in implementing the Arusha Agreement, UN forces would be withdrawn.88

S-16

11/24/09

488

4:25 PM

Page 488

CASE STUDIES

The February negotiations produced a compromise whereby the BBTG and the TNA would be established on February 14, 1994, temporarily leaving out the parties with internal feuds until they could resolve these problems; the PL opted to settle its conflicts through an internal process, while the MDR sought the intervention of courts. This agreement was, however, vetoed by the MRND, which contended that all outstanding problems needed to be resolved before the installation of transitional institutions. Additional initiatives in mid-February 1994 to reach a compromise failed with the outbreak of violence in Kigali and other parts of the country, and a final attempt on February 23, 1994, to install the institutions was boycotted by the RPF and other political parties, which claimed that Habyarimana was trying to impose a unilateral solution on the PL and the MDR.89 With little progress made on the inauguration of the proposed administration, Tanzania intervened in March 1994 to resolve the crisis. Against renewed warnings from Booh-Booh that UNAMIR would cease its activities on April 4 or 5, 1994, if “there has not been substantial progress in the implementation of the Arusha Agreement,” 90 Foreign Minister Joseph Rwegasira of Tanzania mediated a new deal among the rival factions of the PL and the MDR. Soon after these consultations, Prime Minister–Designate Faustin Twagiramungu announced a compromise list of members of the cabinet and National Assembly, and asked Habyarimana to swear in the new government on March 21, 1994. Habyarimana rejected the new cabinet list, just as he had done before, renewing the demand for the inclusion of the CDR in the TNA. The RPF bitterly resisted this move, since the CDR had breached the Arusha Agreement by inciting and condoning the massacre of Tutsi and had not signed the political parties’ code of ethics, suggesting instead that it should be banned.91 The RPF and other parties insisted that the agreement provided for representation of only those parties that subscribed to the agreement, which the CDR initially and vociferously did not. When in mid-March the CDR altered its previous opposition to the agreement and finally subscribed to a code of ethics for political parties, the opposition parties were unwilling to believe its pledges.92 As the political uncertainty prevailed, UNAMIR as well as Western and African diplomats issued a joint appeal in March 1994 calling for representation in the TNA of all political parties in existence at the time of the signing of the agreement, but the appeal did not break the impasse.93 Facing the impending expiration of the UNAMIR mandate at the end of March 1994, leaders of Rwandan human rights associations and other NGOs pleaded with the Security Council “to maintain and reinforce” UNAMIR because its withdrawal “would be interpreted as abandoning the civilian population to the worst of calamities.”94 The Secretary-General’s report at the end of March 1994 recommended the extension of UNAMIR’s mandate for an additional six months with a new deadline and a dire warn-

S-16

11/24/09

4:25 PM

Page 489

THE ARUSHA PEACE AGREEMENT ON RWANDA

489

ing: “In the event that the transitional institutions are not installed within the next two months and if, by that time, sufficient progress in the implementation of the next phase of the Agreement has also not been achieved, the Council should then review the situation, including the role of the United Nations.”95 On April 5, 1994, the Security Council approved the extension of UNAMIR’s mandate until July 29, 1994. The missing link remained movement on domestic reconciliation, a task that regional states shouldered in a last-minute effort to prevent what President Mwinyi of Tanzania called a “Bosnia on our doorstep.”96 Summoned before his regional peers in Dar es Salaam to explain his unwillingness to implement the transitional institutions, Habyarimana promised to honor his word. However, Habyarimana’s plane was shot down on its return to Kigali on April 6, 1994, and his death triggered the beginning of the genocide that killed more than 800,000 Tutsi and moderate Hutu in ninety days. The civil war resumed until RPF troops routed the Rwandan army and seized Kigali in July 1994.97 International response to the unfolding genocide followed the pattern that had prevailed since the start of the implementation process. Of critical significance was whether UNAMIR, under its constrained mandate, would stay the course following the disintegration of civil authority. Two days after the crisis started, General Dallaire sent a cable to UN headquarters about a “campaign of terror, well planned, organized, deliberate, orchestrated,” and directed against opposition leaders and “particular ethnic groups.” In calling for a revision of the mandate to counter the genocide, he noted that the situation “would be a good deal worse without the presence of UNAMIR.” But, in secret and informal discussions that ensued in New York, the priority was how to pull out of Rwanda.98 In the first few days of genocide, the United States and the United Kingdom led the move in the Security Council to cast doubt on the viability of UNAMIR. On April 12, a U.S. delegate is reported to have indicated that the “U.S. would not be held responsible for the gradual depletion of an isolated force,” while Britain was wondering whether the force “could continue to perform a useful function.” Britain’s UN representative, David Hannay, opposed reinforcements for UNAMIR to try to stop the bloodshed, since this would be “a repetition of Somalia with its well-known and dire consequences.” In the same tone, a U.S. delegate to the United Nations, Karl Inderfurth, spoke of a “strong feeling in Washington” that the peacekeeping force “was not appropriate now and never will be.”99 In Rwanda, the RPF suggested conducting a joint operation with UNAMIR troops and supportive sections of the Rwandese army to provide security to civilians, but the United Nations objected to this proposal. With the resumption of civil war, UNAMIR focused more on achieving a ceasefire than protecting civilians. After the Interahamwe brutally killed ten Belgian peacekeepers, the security of foreign troops took precedence as

S-16

11/24/09

4:25 PM

490

Page 490

CASE STUDIES

Belgium withdrew its 420 soldiers and, shortly thereafter, several hundred Bangladeshi soldiers. The Belgian withdrawal deprived UNAMIR of its best troops and strengthened international pressure for total withdrawal. Two weeks into the genocide, Boutros-Ghali presented three options to the Security Council for UNAMIR. The first option proposed the immediate and massive reinforcement of UNAMIR, changing its mandate and authority so that it could coerce the opposing forces into a cease-fire. The second option called for the reduction of UNAMIR to about 270 personnel, who would remain to “act as an intermediary between the two sides in an attempt to bring about a cease-fire.” The third option was a complete withdrawal. Boutros-Ghali claims that he had a strong preference for forceful action since it would “give UNAMIR the credibility to deal effectively with the situation. . . . I reminded the Council that the consequences of a complete withdrawal of UNAMIR, in terms of human lives lost, could be very severe.”100 On April 21, the Security Council passed Resolution 912 to withdraw the peacekeepers, leaving a token force of 264. During three weeks of genocide, the Security Council had, in the words of the Czech Republic’s representative, Karel Kovanda, spent “80 percent of its efforts deciding whether or not to pull out the peacekeepers” and “20 percent trying to get a cease-fire.”101 As Human Rights Watch aptly concludes: “The Americans were interested in saving money, the Belgians were interested in saving face, and the French were interested in saving their ally, the genocidal government. All of that took priority over saving lives.”102 UNAMIR’s withdrawal gave extremist forces an opportunity to wage the genocide for two more months. Boutros-Ghali described the withdrawal as a “scandal.”103 Yet it merely ended a phase that had begun with unrealistic expectations enshrined in the Arusha Agreement. As General Dallaire has admitted: “The U.N. Mission, and those Rwandans it was intended to secure, fell victim to an inflated optimism to which I contributed, thereby creating expectations that the U.N. did not have the capacity to fulfill.”104 By the time the Security Council reversed itself in late May 1994 and authorized the establishment of UNAMIR II with a proposed force of 5,500, the RPF was poised to capture Kigali. As an RPF spokesman commented: “It is cynical of the United Nations now to say they are coming to save people when there are no people left to save.”105

Conclusion Analyses of the Arusha Agreement revolve around whether it unraveled because of the flaws in the negotiation process, the nature of the agreement, or its implementation. Critical to these questions is whether there were any leadership coalitions and strategies that would have improved the chances of success. Was the agreement doomed from the outset, or could it have

S-16

11/24/09

4:25 PM

Page 491

THE ARUSHA PEACE AGREEMENT ON RWANDA

491

been made to work? Did the parties expect the broader international community to save them from an essentially fragile agreement to which they were not committed? Would more forceful and aggressive international action have made a difference during the implementation? Given the limited information about the motives of major actors in Rwanda, analyses of failure have understandably been conjectural, depending for the most part on counterfactuals.106 Most critics of the Arusha Agreement assert that it could have been “strengthened” by bringing the extremists, particularly the CDR, on board. These critics have cast aspersions on the Tanzanian mediators, charging that they should have been concerned about the lack of representation of extremists in the power-sharing provisions and the RPF’s disproportionate control of important transitional institutions. Alan Kuperman, for example, blames the mediators for failing to appreciate “how much Rwanda’s entrenched elite had to lose under political pluralization, and the lengths to which it would go to preserve the status quo. The mediators’ application of leverage succeeded in compelling Rwanda’s President to sign and begin to implement the Arusha accords, but this very success raised the insecurity of Rwanda’s elite.”107 Addressing the weaknesses of the Arusha Agreement, Bruce Jones distinguishes between the process and the outcome of the negotiations. While conceding that the Arusha process was effective, Jones nevertheless sees major shortcomings in the outcome: “The Arusha accords violated key tenets of conflict resolution in the context of protracted social conflicts by effectively excluding the CDR from power, and thus marginalizing even the MRND, diverging from the central tenet that forces which comprise the problem must be made part of the solution.” 108 There is evidence of Tanzanian mediators and Western observers demanding the CDR’s inclusion in the agreement to make it more inclusive.109 Yet this evidence is selectively and singularly deployed with the benefit of hindsight to separate process from outcome and to conceal the limited control mediators and other external actors had over the parties. The negotiation process constrained outcomes solely because the mediators operated within the frame of available information about the motives of the principal parties. Short of dictating the terms of the agreement, there does not seem to be much that the mediators could have done to induce the parties to reach a more “inclusive” and “realistic” settlement. The RPF resisted entreaties to include the CDR in the agreement not just because of its alliance with Habyarimana, but also because, throughout the negotiations, the CDR presented the most consistent public face of Hutu ethnic chauvinism and militarism.110 In Lindsey Hilsum’s view, the MRND and the CDR were instruments in Habyarimana’s two-track strategy of moderation and extremism, a strategy that he could juggle successfully during the negotiations.111 The pattern of prevarication that Habyarimana had established toward the peace

S-16

11/24/09

492

4:25 PM

Page 492

CASE STUDIES

process since 1990, in addition to the deliberate destabilization of the PL and the MDR in the implementation phase, revealed that his agenda was not that far removed from the broader one of the CDR and militias. UN reports point to the collaborative links between the formal security organs under Habyarimana, the CDR, and the Interahamwe.112 Even the CDR’s last-ditch effort to moderate its position by signing the code of ethics for political parties was clearly linked to Habyarimana’s desperate efforts to boost his power in the TNA to derail major provisions of the Arusha Agreement. As stated by Alison Des Forges: “Habyarimana was determined to have the CDR seated because it could provide him with the final vote necessary to block any effort to impeach him.”113 This pattern reveals that Habyarimana largely fostered the growth of extremism. While he could sustain the two-track strategy in the negotiations due to internal and external pressures, this strategy was untenable during the implementation and consequently cost him his life. The Arusha Agreement ended in genocide rather than peace because it was constructed on a moderate political center that could not withstand the countervailing pressures of ethnic mobilization and militarization beyond the negotiating table. Criticisms that the mediators did not have a sound strategy of dealing with extremism are correct, but they ignore the fact that the mediation process hinged largely on cultivating a moderate center based in the constellation of existing political alliances. Military and economic pressures on Habyarimana’s core Hutu alliance helped to launch a negotiated process that, the mediators hoped, would build a common ground for reducing ethnic antagonisms. The agreement was an outcome of political moderation, but it had implacable foes. In a polarized socioeconomic environment in which the extremist coalition was fast losing power to the moderate coalition, there did not seem to be any compensations and guarantees by which the mediators could have prevented the genocide. As Christopher Clapham argues: “Given the ideology of ethnic exclusion promoted by the Hutu power factions . . . and the genocidal length to which these were prepared to go in order to uphold it, no pact involving these groups was possible, and any settlement would have had to rest either on their supremacy or on their defeat.”114 The Rwanda peace process occurred within a regional environment where conflict in Burundi severely shook an already unstable peace agreement. The Tutsi-led coup emboldened Hutu extremists in Rwanda and compounded the fragility of implementation. Could the mediators have better prepared for the unforeseen events in Burundi and made contingency plans to prevent their deleterious impact on the implementation process? There was a clear lack of regional leadership on Rwanda beyond the Arusha negotiations because of expectations about UN commitment to the implementation. Despite the limited role of Tanzania and other regional countries, one

S-16

11/24/09

4:25 PM

Page 493

THE ARUSHA PEACE AGREEMENT ON RWANDA

493

is hard pressed to contemplate what their role would have been in insulating Rwanda from civil conflagration in Burundi. There is more clarity on the motives of major players at the United Nations toward the implementation process, captured in the postgenocide consensus about the abandonment of Rwanda. Recognizing the history of conflict and the deep mistrust between them, both the RPF and the government sought a stronger international peacekeeping force than could be marshaled in the less charitable international environment that prevailed after Somalia. The mediators also staked the implementation process on substantial UN commitment, partly because of the roles of Western countries in the negotiations. But the tentative approach by the Security Council on crucial aspects of the peacekeeping force demonstrated international disinterest in Rwanda by those who mattered. The United Nations failed because of the fundamental lack of political will to muster the resources to launch a credible enforcement effort. The hostility of key Western countries in the Security Council to strengthen UNAMIR’s mandate showed a clear absence of international commitment to the implementation process. Paltry international effort in turn gave further ammunition to opponents of the peace agreement. General Dallaire later summed up the central problem of political will: “The United Nations wanted to send me more troops, but sovereign states made sovereign decisions not to do so.”115

Notes 1. For analysis of Rwanda’s fractured history, see Gerard Prunier, The Rwanda Crisis: History of a Genocide, 1959–1994 (New York: Columbia University Press, 1995). 2. Hartmut Diessenbacher, “Explaining the Genocide in Rwanda: How Population Growth and a Shortage of Land Helped to Bring About the Massacres and the Civil War,” Law and State 52 (1995): 76. 3. David Norman Smith, “The Genesis of Genocide in Rwanda: The Fatal Dialectic of Class and Ethnicity,” Humanity and Society 19, no. 4 (November 1995): 65–67; and Alison Des Forges and Alain Destexhe, “Rwanda and Genocide in the Twentieth Century,” Times Literary Supplement no. 4924 (October 1997): 26. 4. Quoted in Smith, “The Genesis of Genocide in Rwanda,” p. 67. 5. For a summary of the economic and demographic arguments, see Catherine Andre and Jean-Philippe Platteau, “Land Relations Under Unbearable Stress: Rwanda Caught in the Malthusian Trap,” Journal of Economic Behavior and Organization 34 (1998): 1–47; and Diessenbacher, “Explaining the Genocide in Rwanda,” pp. 58–88. 6. For data on coffee exports, see “Rwanda: Subsistence Farming,” Africa Research Bulletin, October 16–November 15, 1990, pp. 10151–10153. 7. World Bank, Rwanda: Poverty Reduction and Sustainable Growth

S-16

11/24/09

494

4:25 PM

Page 494

CASE STUDIES

(Washington, D.C.: World Bank, Population and Human Resources Division, Africa Region, 1994). 8. On the impact of accumulating local pressure, see Rakiya Omaar and Alex de Waal, Rwanda: Death, Despair, and Defiance (London: African Rights, 1994); and Prunier, The Rwanda Crisis, pp. 88–90. 9. Prunier, The Rwanda Crisis, p. 90. 10. Catherine Newbury, “Background to Genocide in Rwanda,” Issue 23, no. 2 (1995): 12–17. 11. For a summary of RPF demands, see David Buchan, “Rwanda Peace Bid Gets Under Way,” Financial Times, October 16, 1990, p. 6. 12. William Cyrus Reed, “Exile, Reform, and the Rise of the Rwandese Patriotic Front,” Journal of Modern African Studies 34, no. 4 (1996): 479–502. 13. “Statement of the Minister for Foreign Affairs and International Cooperation During a Meeting with Heads of Diplomatic and Consular Missions Accredited to Kigali,” Kigali, October 8, 1990. 14. The role of France and Belgium is amply discussed in Prunier, The Rwanda Crisis, pp. 93–100; “Belgian Premier Arrives: Front Spokesman Confirms,” Foreign Broadcast Information Services (FBIS)-AFR-90-194, October 5, 1990, pp. 4–5; and “Mobutu, Habyarimana Discuss Rwandan Crisis,” FBIS-AFR90-196, October 10, 1990, p. 6. 15. Omaar and de Waal, Rwanda, pp. 60–67; Prunier, The Rwanda Crisis, pp. 100–108; and Stephen D. Goose and Frank Symth, “Arming Genocide in Rwanda,” Foreign Affairs 73, no. 5 (September 1994): 86–89. 16. Alison Des Forges, Leave None to Tell the Story: Genocide in Rwanda (New York: Human Rights Watch, 1999), pp. 116–120. 17. Scott R. Feil, Preventing Genocide: How the Early Use of Force Might Have Succeeded in Rwanda: A Report to the Carnegie Commission on Preventing Deadly Conflict (New York: Carnegie Corporation, April 1998), p. 38. 18. Cited in Howard Adelman and Astri Suhrke, with Bruce Jones, International Response to Conflict and Genocide: Lessons from the Rwanda Experience, vol. 2, Early Warning and Conflict Management (Copenhagen: DANIDA, 1996). 19. Jonathan Clayton, “Rwanda to Appeal to U.N. Security Council on Rebel Invasion,” Reuters, October 15, 1990. 20. Buchan, “Rwanda Peace Bid Gets Under Way,” p. 6. 21. “Cease-Fire Agreement Reached: E.C. Help Sought,” FBIS-AFR-90-202, October 18, 1990, p. 2. 22. “Rebels Say Fighting Continues: Leader Comments,” FBIS-AFR-90-206, October 24, 1990, p. 3; and “Rebels State Conditions for Cease-Fire Talks,” FBISAFR-90-206, October 24, 1990, p. 3. 23. “Museveni Attends a Rwanda Summit in Zaire,” New Vision (Kampala), October 27, 1990, p. 1. 24. “Joint Communique Issued,” FBIS-AFR-90-209, October 29, 1990, p. 1. 25. For a summary of Mobutu’s mediation, see “Zairian Envoy Meets with RPF on Cease-Fire,” FBIS-AFR-90-214, October 5, 1990; and “President Mobutu Receives Patriotic Front,” FBIS-AFR-90-221, November 15, 1990, p. 5. 26. “Foreign Minister Comments on Goma Talks,” FBIS-AFR-90-231, November 30, 1990, p. 4; and “Observer Officers Meet in Goma on Cease-Fire,” FBIS-AFR-90-223, November 19, 1990, p. 4. 27. “The Dar es Salaam Declaration on the Rwandese Refugee Problem,” Dar es Salaam, February 19, 1991. 28. “Rebel Colonel Details Cease-Fire Provisions,” FBIS-AFR-91-203, February 19, 1991, p. 3.

S-16

11/24/09

4:25 PM

Page 495

THE ARUSHA PEACE AGREEMENT ON RWANDA

495

29. “Details on N’sele Accord,” FBIS-AFR-91-062, April 1, 1991, p. 2; and “Government, Patriotic Front Sign Cease-Fire,” FBIS-AFR-91-062, April 1, 1991, pp. 1–2. 30. For accounts of the reform process, see “Press Release: The Rwandese Head of State Decides to Accelerate the Process of Political Reforms,” Rwandese embassy, Dar es Salaam, April 30, 1991; and “Habyarimana on Multiparty System, Elections,” FBIS-AFR-91-128, July 3, 1991, pp. 2–5. 31. For critical analyses of the outcomes of political reforms, see Filip Reyntjens, “Constitution-Making in Extreme Crisis: The Case of Rwanda and Burundi,” Journal of African Law 40, no. 2 (1996): 234–236; and Rene Lemarchand, “Managing Transition Anarchies: Rwanda, Burundi, and South Africa,” Journal of Modern African Studies 32, no. 4 (1994): 581–604. 32. “Habyarimana Addresses Legislators on Reform,” FBIS-AFR-90-221, November 15, 1990, p. 4. 33. For accounts of opposition perspectives on negotiations and the formation of the coalition government, see “Opposition Parties Reject Amnesty Law,” FBISAFR-91-194, October 7, 1991, p. 1; “Bizimungu Discusses Peace Process with Opposition,” FBIS-AFR-92-022, February 3, 1992, pp. 5–6; “Negotiations Under Way for Union Government,” FBIS-AFR-92-038, February 26, 1992, p. 3; “Regime, Opposition Agree on Future Government,” FBIS-AFR-92-036, February 24, 1992, p. 2; and “Opposition Parties, Ruling Party Sign Agreement,” FBIS-AFR-92-051, March 16, 1992, p. 2. 34. Reyntjens, “Constitution-Making in Extreme Crisis,” p. 23. 35. “Defense Minister on Fighting,” FBIS-AFR-92-110, June 8, 1992, p. 3; “French Troops Brought In,” FBIS-AFR-92-110, June 8, 1992, p. 4; and “Council of Ministers Meets on War Escalation,” FBIS-AFR-92-120, June 22, 1992, pp. 2–3. 36. “Cease-Fire Pact Reached in Rwanda,” Chicago Tribune, July 13, 1992, p. 4; Tsegaye Tadesse, “OAU Names Official to Monitor Rwanda Cease-Fire,” Reuters, July 29, 1992; “Foreign Minister on Cease-Fire, Peacekeepers,” FBISAFR-92-149, August 3, 1992, pp. 7–8; and “Rwandan, Ugandan Presidents Meet in Bujumbura,” FBIS-AFR-92-165, August 25, 1992, p. 2. 37. Author interviews with participants, Kigali, Rwanda, and Dar es Salaam, Tanzania, May–June 1996. 38. For analysis of each party’s position, see “Rwandese Government, Rebels Begin Third Round of Talks,” Daily News (Dar es Salaam), September 9, 1992; “Sides Present Proposals at Rwandan Peace Talks,” FBIS-AFR-92-177, September 11, 1992, pp. 5–6; and “Habyarimana Speaks on Political Situation, RPF,” FBISAFR-92-163, August 21, 1992, p. 4. 39. See “President ‘Blocking’ Democracy: Resignation Sought,” FBIS-AFR92-198, October 13, 1992, p. 4, for analysis of internal opposition views on Habyarimana’s attitude toward Arusha. For the RPF’s position, see “RPF Chairman Commemorates 1 October Rebellion,” FBIS-AFR-92-194, October 6, 1992, p. 4. 40. “Habyarimana Meets Eleven Party Representatives,” FBIS-AFR-92-207, October 26, 1992, p. 8; and “MRND Bureau Cites ‘Lack of Consultations,’” FBISAFR-92-211, October 30, 1992, p. 5. 41. “Protocol of Agreement on Power-Sharing Within the Framework of a Broad-Based Transitional Government Between the Government of Republic of Rwanda and the Rwandese Patriotic Front,” Arusha, October 30, 1992. 42. For reports of planned massacres, see “RPF Communique Warns of Government Massacres,” FBIS-AFR-92-208, October 27, 1992, p. 5; and “Government Accused of ‘Exterminating Youths,’” FBIS-AFR-92-208, October 27, 1992, p. 5.

S-16

11/24/09

496

4:25 PM

Page 496

CASE STUDIES

43. “Mwinyi, Museveni Discuss Relations,” Daily News (Dar es Salaam), November 17, 1992; and “Museveni Returns from Tanzania, Comments,” New Vision (Kampala), November 18, 1992. 44. “Protocol of Agreement on Cabinet Portfolios and the Transitional National Assembly Between the Government of the Republic of Rwanda and Rwandese Patriotic Front,” Arusha, January 6, 1993. For an account of the process, see Bruce D. Jones, “The Arusha Peace Process,” in Howard Adelman and Astri Suhrke, eds., The Path of a Genocide: The Rwanda Crisis from Uganda to Zaire (New Brunswick, N.J.: Transaction Press, 1999), pp. 138–139. 45. “MRND ‘Refuses’ Role in Government,” FBIS-AFR-93-006, January 11, 1993, p. 5. 46. Ibid. 47. “MRND Continues Demonstrations: Kigali Under Curfew,” FBIS-AFR93-012, January 21, 1993, pp. 4–5; “Youth Block Roads: Hinder Arusha Talks,” FBIS-AFR-93-012, January 4, 1993, p. 2; and “Displaced Tutsis Fear Massacre by Hutu Neighbors,” FBIS-AFR-93-021, February 3, 1993, pp. 2–4. 48. Prunier, The Rwanda Crisis, p. 180. 49. UN, The United Nations and Rwanda, 1993–1996, Blue Book Series vol. 10 (New York: United Nations, 1996), pp. 20–22. 50. Prunier, The Rwanda Crisis, p. 178. 51. In futile negotiations between the RPF and opposition parties in late February 1993 in Bujumbura, Burundi, to explore the resumption of talks, the RPF described French troops as the “major hindrance to the peace process” and demanded their immediate withdrawal. See “‘Total Stalemate’ in Bujumbura Talks Reported,” FBIS-AFR-93-039, March 2, 1993, p. 1. 52. “Interim Report of the Secretary-General on Rwanda, Recommending the Establishment of a United Nations Observer Mission Uganda-Rwanda (UNOMUR),” in the UN, The United Nations and Rwanda, pp. 162–168. 53. “President on Talks with Rebels, Reconciliation,” FBIS-AFR-93-087, May 7, 1993, pp. 15–16. 54. “Joint Request by the Rwandese Government and the Rwandese Patriotic Front to the Secretary-General of the United Nations Concerning the Stationing of a Neutral International Force in Rwanda,” in UN, The United Nations and Rwanda, p. 166. 55. “Habyarimana Postpones Signing of Peace Agreement,” Daily News (Dar es Salaam), June 24, 1993; and “Government-RPF Peace Agreement ‘Postponed Indefinitely,’” FBIS-AFR-93-121, June 25, 1993, p. 5. 56. “Tanzania Calls for Signing of Rwandan Peace Accord,” Daily News (Dar es Salaam), June 26, 1993; and “Minister Comments on Return from Talks in Arusha,” FBIS-AFR-93-123, June 26, 1993, pp. 11–12. 57. “Rwanda: Peace Agreement Signed,” FBIS-AFR-93-149, August 5, 1993, p. 3. 58. Adelman and Suhrke, Early Warning and Conflict Management. See also Prunier, The Rwanda Crisis, pp. 159–191. 59. References to the agreement are drawn from UN, The United Nations and Rwanda, pp. 170–200. 60. Des Forges, Leave None to Tell the Story, p. 125; and Prunier, The Rwanda Crisis, pp. 192–212. 61. Quoted in Des Forges, Leave None to Tell the Story, p. 126. 62. Diessenbacher, “Explaining the Genocide in Rwanda,” p. 75. 63. “Arusha Peace Agreement,” Daily News (Dar es Salaam), August 5, 1993. 64. UN, United Nations and Rwanda, p. 217.

S-16

11/24/09

4:25 PM

Page 497

THE ARUSHA PEACE AGREEMENT ON RWANDA

497

65. Ibid., p. 24. 66. The two former adversaries, in an unprecedented gesture, sent a delegation to the UN in mid-September 1993 to stress the urgent need for the deployment of UN forces. 67. “Report of the Secretary-General on Rwanda, Requesting Establishment of a United Nations Assistance Mission for Rwanda (UNAMIR) and the Integration of UNOMUR into UNAMIR,” in UN, United Nations and Rwanda, p. 222. 68. Ibid., p. 223. 69. Ibid., p. 224. 70. Ibid., p. 226. 71. Des Forges, Leave None to Tell the Story, pp. 130–133. 72. “Report of the Secretary of the Secretary-General on Rwanda,” p. 226. 73. “Security Council Resolution Establishing UNAMIR for a Sixth-Month Period and Approving the Integration of UNOMUR into UNAMIR,” in UN, The United Nations and Rwanda, p. 232. 74. “Further Report of the Secretary-General on Rwanda, Concerning the Arusha Agreement and the Possible Role of the United Nations in Its Implementation,” in UN, The United Nations and Rwanda, p. 219. 75. Linda Melvern, “Genocide Behind the Thin Blue Line,” Security Dialogue 28, no. 3 (1997): 335; Michael N. Barnett, “The U.N. Security Council, Indifference, and Genocide in Rwanda,” Cultural Anthropology 12, no. 4 (1997): 551–578; and Bruce D. Jones, “‘Intervention Without Borders’: Humanitarian Intervention in Rwanda, 1990–1994,” Millennium 24, no. 2 (1995): 225–249. 76. Des Forges, Leave None to Tell the Story, p. 130. General Dallaire remarked: “UNAMIR suffered several important shortcomings from the very beginning. This was due, in part, to a conscious decision by the Security Council seemingly to classify the crisis in Rwanda as a low risk priority.” Quoted in Romeo Dallaire, “The Changing Role of UN Peacekeeping Forces: The Relationship Between UN Peacekeepers and NGOs in Rwanda,” in Randolph Kent and Shashi Tharoor, eds., The Role of Peacekeeping (New York: St. Martin’s Press, 1996), p. 208. 77. Omaar and de Waal, Rwanda. 78. For regional diplomatic efforts after the Arusha Agreement, see “Joint Communique Issued Between President Habyarimana and Museveni on Improving Bilateral and Regional Relations,” New Vision (Kampala), September 2, 1993. 79. Quoted in Adelman and Suhrke, Early Warning and Conflict Management, p. 32. 80. Des Forges, Leave None to Tell the Story, p. 135. 81. UN, The United Nations and Rwanda, p. 22. 82. See the catalog of events reported by several organizations between November and December 1993 in Des Forges, Leave None to Tell the Story, pp. 142–144. 83. “Report of the Secretary-General on UNAMIR,” in UN, The United Nations and Rwanda, p. 240. 84. UN, The United Nations and Rwanda, pp. 25–26. For accounts of General Dallaire’s fax, see Adelman and Suhrke, Early Warning and Conflict Management, pp. 40–46. 85. Quoted in Linda Melvern, “Genocide Behind the Thin Blue Line,” Security Dialogue 28, no. 3 (1997): 336. 86. “Letter Dated March 14, 1994 from the Minister of Foreign Affairs of Belgium to the Secretary-General Expressing Concern that Worsening Situation in Rwanda May Impede UNAMIR’s Capacity to Fulfill Its Mandate,” in UN, The United Nations and Rwanda, p. 244.

S-16

11/24/09

498

4:25 PM

Page 498

CASE STUDIES

87. Prunier, The Rwanda Crisis, p. 208. 88. John Baptiste Kayigamba, “UN and Rwanda Peace: United Nations Threatens to Quit,” Inter Press Service Feature, February 24, 1994. 89. Paul Chintowa, “Rwanda-Politics: Embattled President Fails to Form Government,” Inter Press Service Feature, March 8, 1994; and Augustin Hatar, “Rwanda-Politics: Politicians Bicker over Power-Sharing,” Inter Press Service Feature, March 8, 1994. 90. “Political Crisis Deepens, Famine Worsens,” FBIS-AFR-94-051, March 16, 1994, p. 5. 91. “Rwanda: UN Envoy, Mediators Urge Seating of Transitional Bodies,” Inter Press Service Feature, March 28, 1994. 92. Prunier, The Rwanda Crisis, p. 208. 93. “Rwanda: UN Envoy, Mediators Urge Seating of Transitional Bodies.” 94. John Baptiste Kayigamba, “Rwandan Human Rights Groups Plea to International Community,” Inter Press Service Feature, March 26, 1994, p. 3. 95. UN, The United Nations and Rwanda, p. 251; and “Rwanda: Extended Mandate for UN Troops Brings Some Relief,” Inter Press Service Feature, April 6, 1994. 96. Paul Chintowa, “Rwanda-Burundi: Regional Summit Planned to End Violence,” Inter Press Service Feature, April 4, 1994, p. 2. 97. Prunier, The Rwanda Crisis, p. 211. 98. Linda Melvern, “How the World Failed to Save Rwanda,” Mail and Guardian, May 5, 1998; Adelman and Suhrke, Early Warning and Conflict Management; and Des Forges, Leave None to Tell the Story, pp. 174–175. 99. Quoted in Melvern, “How the World Failed to Save Rwanda,” p. 6. 100. UN, The United Nations and Rwanda, p. 43. 101. Quoted in Melvern, “How the World Failed to Save Rwanda,” p. 6. See also Melvern, “Genocide Behind the Thin Blue Line,” pp. 338–339. 102. Des Forges, Leave None to Tell the Story, p. 25. 103. Farhan Haq, “Rwanda: U.N. Hopes to Expand Force to 3,000 by Mid-August,” Inter Press Service Feature, July 28, 1994. 104. Romeo Dallaire, “The Changing Role of UN Peacekeeping Forces,” p. 208. 105. Farhan Haq, “UN Now Ready to Deploy Peacekeeping Force in Rwanda,” Inter Press Service Feature, May 3, 1994, p. 2. 106. See, for instance, Jones, “The Arusha Peace Process,” pp. 146–153. 107. Alan J. Kuperman, “The Other Lesson of Rwanda: Mediators Sometimes Do More Harm Than Good,” SAIS Review 16, no. 1 (1996): 228. 108. Jones, “The Arusha Peace Process,” p. 150. 109. Ibid., pp. 140–141; and Des Forges, Leave None to Tell the Story, p. 177. 110. See Des Forges, Leave None to Tell the Story, pp. 53–54, and pp. 176–177 for a comprehensive discussion of the CDR. 111. Lindsey Hilsum, “Rwanda: Settling Scores,” Africa Report 39, no. 2 (May–June 1994): 15. 112. Des Forges, Leave None to Tell the Story, pp. 101–108. See also Turid Laegreid, “U.N. Peacekeeping in Rwanda,” in Adelman and Suhrke, The Path of a Genocide, p. 234. 113. Des Forges, Leave None to Tell the Story, p. 177. See also Prunier, The Rwanda Crisis, p. 208. 114. Christopher Clapham, “Rwanda: The Perils of Peacemaking,” Journal of Peace Research 35, no. 2 (1998): 209. 115. Quoted in “Rwanda: UN Commander Says More Troops May Have Saved Lives,” Inter Press Service Feature, September 7, 1994, p. 2.

S-17

11/24/09

4:25 PM

Page 499

17 Implementing Cambodia’s Peace Agreement SORPONG PEOU

When the United Nations accepted the invitation of Cambodia’s four warring parties to implement a peace agreement reached on October 23, 1991, it embarked upon a mission of unprecedented ambition. After fifteen years of civil war and large-scale repressive violence, Cambodia’s belligerents had finally reached a comprehensive settlement, but recognized their need for a third party to play a central role in putting it into practice. Due to the intense distrust and hostility among the parties, and given the extent of social and economic damage wrought by the civil war, the United Nations was expected to provide transitional authority during implementation and take the lead in carrying out many of the provisions of the Paris Agreement. Although much has been written on the broader subject, few scholars have examined systematically the process of implementing the Paris Agreement and the UN’s critical responsibility for accurately identifying the key obstacles to implementation and in developing appropriate strategies to overcome them. Many analysts, instead, approach Cambodia’s troubled peace process simply with varying degrees of pessimism that yield few practical prescriptions for policy.1 While useful for their cautionary notes about hasty and overambitious intervention in war-torn states, these pessimistic perspectives fail to thoroughly explore the UN intervention in order to derive lessons for future efforts to help build a more peaceful world. In contrast to the pessimists, this study argues that the UN’s performance in Cambodia between 1992 and 1993 is best seen as a “qualified success.” Why the UN’s success was only limited thus becomes the most important question. The UN Transitional Authority in Cambodia (UNTAC) accomplished more than the pessimists had anticipated. Elections took place, and the country has since experienced greater political stability than 499

S-17

11/24/09

500

4:25 PM

Page 500

CASE STUDIES

in the years preceding the operation. That said, the mission did not accomplish many of its goals. The disarmament process envisioned in the Paris Agreement failed completely, the Khmer Rouge withdrew from the electoral process, and civil war continued after the elections in May 1993. Subsequently, one of the parties, the State of Cambodia (SOC), reasserted hegemonic control in Cambodia and threatened the country’s nascent democracy. Why, then, did UNTAC only succeed to the extent that it did? Several explanations give a partial account for UNTAC’s limited effectiveness. To begin with, the mission contended with a challenging political context: lack of democratic tradition, authoritarian habits, and a prevalence of antidemocratic ideologies. At the operational level, UNTAC had to deal with problems not unfamiliar to peacekeeping operations: insufficient resources, lack of qualified personnel, and a mandate that, however robust in some respects, was not strong enough to allow the UN to play an optimal role in a weak state like Cambodia. Most problematic, the parties were fundamentally unwilling or unable to abide by their commitment to the agreement. This unwillingness, in turn, derived significantly from the parties’ concerns over security, which were aggravated rather than lessened by choices made during implementation. Understandably perhaps, external actors worried that the Khmer Rouge might return to power and sought ways to marginalize it. This strategy, however, rendered the Khmer Rouge only more insecure and therefore less willing to accept UNTAC authority. Meanwhile, because external actors were preoccupied with the Khmer Rouge, they did little to restrain the SOC. The emerging imbalance of power in favor of the SOC after the agreement was signed limited UNTAC’s potential success and put the process of democratization at risk. Study of the Cambodian conflict has by and large neglected a thorough examination of the motives and interests of the various parties, especially the Khmer Rouge. This chapter offers a course correction in this respect, and turns to the emerging literature on “spoilers” to gain a more accurate understanding of the Cambodian belligerents. The upshot of Stephen Stedman’s recent work is that analysts and policymakers need a richer typology of spoilers, their interests, and their consequent behavior in order to develop more effective strategies for “managing” spoilers in the context of peace processes. 2 According to Stedman, parties can alternatively be driven by insecurity or a lust for power, and have limited or total goals. Custodians of a peace process must adopt varying strategies and tactics to deal with different types of peace spoilers: inducement where spoilers have limited aims, socialization where spoilers are greedy, and coercion where spoilers’ aims are total. To enrich this theory, one needs to ask more about where adversaries’ behavior comes from and the relative shaping influence of intentions, beliefs, and ideologies, on the one hand, in contrast with desire for security,

S-17

11/24/09

4:25 PM

Page 501

CAMBODIA’S PEACE AGREEMENT

501

on the other. The argument here is that the international community misread the Khmer Rouge and treated them as “total spoilers,” leaving marginalization or coercion as their only policy options. Instead, evidence indicates that the Khmer Rouge were not total spoilers and that greater attention to their minimal security and power requirements (including the concomitant advantage of a measure of international legitimacy) had the capacity to induce more cooperative behavior.

Prelude to UNTAC: From Civil War to the Paris Agreement What has been termed the “Third Indochina War” broke out in 1975 following the communist takeover of three Indochinese states (Cambodia, Laos, and Vietnam) and the ensuing border conflict between two of them: Vietnam and the new Cambodian regime of the communist Khmer Rouge and its leader Pol Pot. The Khmer Rouge began a systematic and brutal “revolution” of Cambodian society, whose toll over their three years in power was over 1 million dead. The regional war that developed has been explained in several ways, including Pol Pot’s “irredentist ambitions,” irrationalism,” and ideological competition between rival communist regimes whose models of socialist society-building differed. The most compelling explanation, however, begins with Cambodia’s domestic vulnerabilities, Khmer Rouge perception of domestic and foreign enemies, and its leadership’s willingness—especially within the army—to respond to threats preemptively. There is evidence indicating that Pol Pot did not fully control his army even after his movement took power in 1975, and his military commanders often showed more enthusiasm about taking military action against Vietnam than he did.3 The border conflict also rapidly became internationalized. The Soviet Union lent support to Vietnam with technical personnel and military hardware; China similarly assisted Cambodia as an instrument against Vietnam, which it saw as a Soviet client. China’s position moved closer to that of the United States, which shared an interest in containing expansion of the Soviet Union into Southeast Asia. Foreign intervention demonstrably escalated the war, prompting the massive Vietnamese invasion of Cambodia in late December 1978. Vietnam quickly installed a friendly Cambodian government, ending the Khmer Rouge reign of terror. In early 1979, the People’s Republic of Kampuchea (PRK, later changed to SOC in 1989) was thus established, led first by Heng Samrin and later by Hun Sen. Vietnamese occupation lasted until 1989.4 The Vietnamese occupation of Cambodia did not end the war, however, as regional powers condemned Vietnam’s violation of international law and aggression against a sovereign state. China and the Association of Southeast Asian Nations (ASEAN) were especially vocal. China launched a

S-17

11/24/09

502

4:25 PM

Page 502

CASE STUDIES

military incursion into Vietnam in February 1979, with the aim of teaching the latter a “lesson.” ASEAN also mobilized international support at the UN to isolate Vietnam diplomatically. The regional grouping also mobilized support for Cambodian resistance groups, particularly the National United Front for a Cooperative, Independent, Neutral, and Peaceful Cambodia (FUNCINPEC), founded by Prince Norodom Sihanouk; the antiroyalist Khmer People’s National Liberation Front (KPNLF); and the Khmer Rouge. Although ASEAN disliked the Khmer Rouge’s past, it made good use of the faction’s military expertise to counterattack Vietnamese offensives on the Thai-Cambodian border. Under pressure from China and some ASEAN states, a Coalition Government of Democratic Kampuchea (CGDK) was formed, allying the Khmer Rouge with the other two resistance forces. In the CGDK, FUNCINPEC’s Sihanouk acted as president, KPNLF leader Son Sann as prime minister, and Khmer Rouge leader Khieu Samphan as foreign minister. Beijing believed that such a coalition would help the factions gain international recognition, as did ASEAN, which supported the CGDK at the United Nations. The coalition government was also seen as a way to improve military coordination against PRK/Vietnamese troops. The United States provided direct financial assistance to the noncommunist resistance forces, and Thailand helped China channel military aid to the resistance, particularly to the Khmer Rouge. The PRK, meanwhile, enjoyed steady backing of the Soviet bloc.5 The battlefield favored the PRK/Vietnam alliance through the mid1980s, despite external support for the CGDK. In 1984–1985, some Vietnamese and PRK troops (at 60,000 and 10,000 strong, respectively) launched an extensive, dry-season offensive against major resistance encampments and at refugee camps, which harbored 30,000 resistance troops.6 By 1986, the PRK had consolidated its strategic position.7 Politically, it had outlasted the previous two regimes of Pol Pot (1975–1978) and President Lon Nol (1970–1975) put together. The PRK regime was unable to eliminate resistance, however. The Khmer Rouge army, in particular, recuperated from its defeats. By 1981, Khmer Rouge troops numbered roughly 50,000. Although this number may have declined to 30,000 by the mid-1980s, they continued to fight alongside the KPNLF and FUNCINPEC, each of which had approximately 5,000 troops. While the larger PRK army had managed to destroy several key resistance bases and retained the backing of 100,000 Vietnamese troops until the mid-1980s, it failed to conquer the guerrilla forces. By the end of 1986, neither Vietnamese nor resistance forces were in a military position to dictate terms.8 Faced with military stalemate, a weakened political and diplomatic position, and domestic economic problems, the Vietnamese army withdrew in 1989. The PRK—now known as the State of Cambodia—could no longer hope for victory, despite its continuing numerical military dominance.9 The SOC outnumbered all its enemies

S-17

11/24/09

4:25 PM

Page 503

CAMBODIA’S PEACE AGREEMENT

503

combined, with around 150,000 troops.10 The SOC’s police force was also larger than the combined strength of the resistance, with over 47,000 police personnel (at the provincial level and below, the number is estimated at 40,000). By comparison, the Khmer Rouge’s police totaled 9,000, FUNCINPEC’s 1,700, and the KPNLF’s 1,000.11 Numbers do not tell the whole story, however. Despite numerical superiority, the SOC force could not keep the resistance at bay without the Vietnamese army. Many reports indicated numerous SOC army defections (especially on the border with Thailand), desertions, and mutinies, as well as draft-dodging, poor-quality conscripts, lack of discipline, low morale, and corruption. Pay was poor. Some units also acquired a reputation for raping and looting their way through remote villages. Meanwhile, the Khmer Rouge army was “disciplined and widely feared,” leading some to see it as the strongest armed faction.12 The result was an effective stalemate by the end of the 1980s, with the opposing armies focused “on holding on to what they had captured.”13 The human toll was also costly to both sides. Casualties were enormous among Cambodians. The estimate of lives lost between 1975, when the Khmer Rouge began their short reign of terror, and the Paris Agreement varies between 1.5 and 2 million. Beginning in 1987, several attempts were made to get the Cambodian factions to negotiate, including a promising 1988 meeting in Jakarta that brought together for the first time not only the four Cambodian factions but also their allies in the region.14 Until June 1991, the negotiation process was marred by a series of contentious issues, however. The First Paris Conference (August 1989) brought all of the Cambodian factions to the table, along with a number of foreign powers, but produced no breakthrough. The main obstacle to settlement appeared to be disagreement among the Cambodian factions about a viable power-sharing formula. The stalemate continued: at meetings in Jakarta (February 26–28, 1990) and Tokyo (June 4, 1990), the factions continued to disagree over the role of a proposed Supreme National Council (SNC). Other stumbling blocks included differences over disarmament provisions and the formation of a transitional civilian authority to be accompanied by a UN peacekeeping force during the transition to a new government.15 An initial breakthrough came in September 1990 with an agreement among the parties on the establishment of an SNC. On October 23, 1991, the Cambodian factions and nineteen other states signed “The Agreements on a Comprehensive Political Settlement of the Cambodian Conflict,” known commonly as the Paris Agreement for the city in which it was signed. The Paris Agreement was divided into nine parts.16 Part 1 defined the transitional period and the role of UNTAC and the SNC, along with military provisions related to cease-fire, withdrawal of foreign forces and their verification, and cessation of outside military assistance. Any “foreign forces, advisors, and equipment remaining in Cambodia shall be withdrawn

S-17

11/24/09

504

4:25 PM

Page 504

CASE STUDIES

from Cambodia, not to be returned [and their] withdrawal and non-return will be subject to UNTAC verification.” Part 2 recognized the Cambodian people’s political right to determine their own political future via the free and fair election of a Constituent Assembly. The assembly was to draft a new constitution, transform itself into a legislative assembly, and create a new democratic government. Part 3 covered human rights, stipulating that “[all] persons in Cambodia and all Cambodian refugees and displaced persons shall enjoy the rights and freedoms embodied in the Universal Declaration of Human Rights and other relevant international human rights instruments” (Article 15). Signatories also agreed to undertake the following tasks: (1) to ensure respect for and observance of human rights and fundamental freedoms in Cambodia; (2) to support the right of all Cambodian citizens to undertake activities that would promote and protect human rights and fundamental freedoms; (3) to take effective measures to ensure that the policies and practices of the past shall never be returned; (4) to adhere to relevant international human rights instruments. During the transitional period, UNTAC “shall be responsible . . . for fostering an environment in which respect for human rights shall be ensured” (Article 16). The UN Commission on Human Rights “shall continue to monitor closely the human rights situation in Cambodia” (Article 17). Part 4 provided for international guarantees, and Part 5 was designed to ensure that Cambodian refugees and displaced persons had “the rights to return to the country, to live in safety, security, and dignity, free from intimidation and coercion of any kind.” Release of all prisoners of war and civilian internees, under the direction of the International Committee of the Red Cross (ICRC), was covered under Part 6. Part 7 focused on the postelection constitution, which was to include basic principles, including those concerning human rights, fundamental freedoms, and Cambodia’s status of neutrality. Part 8 dealt with long-term development. According to Annex 5, the constitution was to contain fundamental rights, including the rights to life, personal liberty, security, freedom of movement, and freedom of religion, assembly, and association; due process and equality before the law; protection from arbitrary deprivation of property or deprivation of private property without just compensation; and freedom from racial, ethnic, religious, and sexual discrimination. The constitution would confirm that Cambodia would adopt a “system of liberal democracy, on the basis of pluralism.” The Paris Agreement was deliberately pragmatic. The retroactive application of criminal law was prohibited. The agreement also omitted any discussion of accountability mechanisms for atrocities, war crimes, and human-rights violations. Instead, the Khmer Rouge, despite a reign of terror that caused more than 1 million deaths, was recognized as a legitimate party; no provisions in the agreement suggested that its leaders be tried for crimes against humanity.

S-17

11/24/09

4:25 PM

Page 505

CAMBODIA’S PEACE AGREEMENT

505

UNTAC’s Mandate The chief challenge for implementation was to ensure that the factions abided by their agreement to compete peacefully through elections and accept any result without returning to the battlefield. In anticipation of the seriousness of this challenge, the Paris Agreement provided the UN with an unprecedented set of powers to rebuild the war-torn state through its intervention. True to its name, the UNTAC mission was given “transitional authority” essentially to govern Cambodia in the absence of a legitimate national regime. For the first time in UN history, a sovereign state’s administrative agencies, bodies, and offices in the ministries of foreign affairs, national defense, public security, and information were to be placed under the direct control of the United Nations.17 UNTAC was also assigned a series of functional tasks following specific provisions of the peace agreement. UNTAC had a major peacekeeping role, which included responsibility for cantoning, disarming, and demobilizing 70 percent of the factional forces. UNTAC was also to supervise, monitor, and verify the withdrawal of foreign forces, adherence to the cease-fire, and related measures. In addition, UNTAC would have a police component to help promote public security by supervising, controlling, and training factional police personnel. Moreover, UNTAC was to play an ambitious role in peacebuilding. It was asked to organize and conduct postwar national elections, and was also mandated to ensure respect for human rights and to help prepare Cambodia for economic reconstruction, with Part 8 of the agreement specifically urging the wider international community to provide support for the process of rehabilitation and reconstruction.18 UNTAC would be headed by a special representative of the Secretary-General, Japanese diplomat Yasushi Akashi; its military component would be led by General John Sanderson of Australia as force commander. Several key items were not in UNTAC’s mandate, however. UNTAC was not invited to develop any long-term plan for economic reconstruction; here, Cambodia was to take responsibility for determining and adopting its own priorities. The mission also was not authorized to take any action that would bring Khmer Rouge leaders to justice for the crimes against humanity they had committed in the past. After all, these now were recognized as legitimate players in Cambodia’s postwar politics. The source of legitimacy for this comprehensive and unprecedented mission was twofold. Consent from the factions provided the first and most critical source of UNTAC’s legitimacy, which derived from the UN’s commitment to create a neutral political environment for free and fair elections and to exercise its authority without favor toward any specific Cambodian party. The agreement made this very clear: “UNTAC will make every effort to ensure that the system and procedures adopted are absolutely impartial”

S-17

11/24/09

506

4:25 PM

Page 506

CASE STUDIES

(Annex 1, Section D, 6). Any recommendation from the SNC would thus only be accepted by UNTAC if SNC members had reached consensus and only if that consensus was consistent with the objectives of the agreement. UNTAC itself would “determine whether advice or action of the SNC is consistent with the present Agreement” (Annex 1, Section A, 2E). The second source of legitimacy derived from international commitments to the peace plan, especially from the nineteen foreign signatories to the agreement. In the text, representatives of these states “invite[d] the United Nations Security Council to establish a United Nations Transitional Authority in Cambodia . . . with civilian and military components under the direct responsibility of the UN Secretary General” (Article 2). They also pledged “to recognize and respect in every way the sovereignty, independence, territorial integrity and inviolability, neutrality and national unity of Cambodia” (Annex 5). The extent to which UNTAC could effectively exercise its authority was another question and difficult to measure, especially in relation to the Supreme National Council. On the one hand, the SNC was not the new government of Cambodia, did not enjoy executive powers, and had delegated “to the United Nations all powers necessary to ensure the implementation of this Agreement.” On the other hand, the SNC was declared “the unique legitimate body and source of authority in which, throughout the transitional period, the sovereignty, independence and unity of Cambodia are enshrined” (Article 3); and it was internationally recognized as the representative of Cambodia externally, including occupying the country’s seat at the United Nations, in UN specialized agencies, and in other international institutions and international conferences (Article 5). Overall, the Paris Agreement provided no clear guidance as to how UNTAC could effectively carry out its mandate, beyond the fact of Cambodian consent and international recognition of the mission’s legal and political legitimacy. Instead, the agreement legalistically assumed that the signatories would fulfill their obligations. The agreement assumed that the adversaries would compete at the ballot box in a free and fair manner simply because they had initially agreed to it, and failed to provide for any sanctions against the signatories should they undermine the process. Perhaps the greatest weakness of the agreement was that it contributed to a breakdown of the rough balance of power among the four Cambodian factions that had made a peace settlement possible in the first place. It recognized the SOC, but delegitimized the CGDK as a coalition government. The SOC, which had failed to restore the Cambodian seat at the United Nations throughout the 1980s, now gained international recognition as a legitimate player in Cambodian politics. Meanwhile, the agreement did not recognize the CGDK as a single entity: the three resistance coalition partners signed the agreement as separate parties who would therefore compete in elections individually vis-à-vis the SOC. The Paris Peace Agreement, in

S-17

11/24/09

4:25 PM

Page 507

CAMBODIA’S PEACE AGREEMENT

507

effect, gave the SOC a competitive advantage by dividing the SOC’s enemies.

Successes and Setbacks During Implementation UNTAC can claim a number of significant accomplishments, especially in the context of Cambodian history, the devastating effects of the war, and the framework of the Paris Agreement itself. These include the repatriation of close to 400,000 refugees, the reasonably effective management of the electoral process, and the commencement of a measure of civil political life.19 Each achievement had its limitations, however, and was counterbalanced by significant failings. In the end, UNTAC succeeded only in reducing the scale of the Cambodian conflict rather than fully resolving it, because the mission could not disarm the factions or manage to prevent the Khmer Rouge faction from withdrawing from electoral contest. Before addressing these larger issues, it is worth reviewing the implementation of other provisions. Refugees and Repatriation The war had generated a large number of Cambodian refugees. At least half a million Cambodians fled their homes, most to Thailand and the area near the Thai-Cambodian border. By the early 1990s, the number of refugees in Thailand was estimated to be 370,000 and the number of internally displaced persons in Cambodia was thought to be 180,000. The UN’s plans to reintegrate returnees into mainstream society faced a number of complications but were ultimately considered a success. Initially, each refugee family was to receive two hectares of land and was allowed to settle where they liked; single women were also promised assistance for house-building, grants to buy housing materials, food supplies, and other forms of short-term aid. The UN had to change tack, however, when faced with the lack of adequate quantities of mine-free, fertile, and available land. As a result, the vast majority of returnees ended up receiving a cash grant of $50 per adult and $25 per child under twelve, food for 400 days, and some assistance with transport, housing materials, and agricultural implements. As of April 1993, more than 339,000 refugees had been repatriated.20 Land Mines The record of mine clearance during the transitional period was mixed. The de-mining process was slowed down in part by the factions’ refusal to disarm and demobilize their troops, and by ongoing military clashes. At the

S-17

11/24/09

4:25 PM

Page 508

508

CASE STUDIES

time of UNTAC’s departure, more than 4 million square meters of land had been de-mined, but with only about 37,000 of the estimated 6–10 million mines removed. Yet 2,000 Cambodians had been trained by the UN in mine-clearance techniques, which was a considerable accomplishment.21 Economic Revival The war and occupation had left the Cambodian economy in tatters. Just one year before the Vietnamese pullout in 1989, the SOC seemed economically desperate. It quickly moved to introduce the Cambodian version of Soviet perestroika, a decision allowing the Cambodian people for the first time to run private enterprises. 22 Despite some progress, the economy remained one of the world’s poorest: stagnant, rife with corruption, and suffering from cutbacks in Soviet aid. Average life expectancy was 49.7 years. Infant mortality was high: 123 out of 1,000 children died in their first year and 70 more died before the age of five. From 1988 to 1991, the currency’s official value depreciated from 142 to 711 per U.S. dollar and dropped to 1,100 to the dollar in May 1992. According to a 1992 report by the UN Development Programme (UNDP), the country ranked 140th among 160 countries in UNDP’s human development index.23 The dilapidated economy was also hurt by the widespread presence of land mines, estimated to number between 6 and 10 million. Given this context, UNTAC’s few accomplishments are worth noting. It managed to control inflation, facilitate a ban on logging, and create new jobs for unemployed Cambodians, including returnees. Sadly, the UN did not contribute to Cambodian infrastructure or general productive capacity beyond putting in place capacity needed for its own operational requirements.24 Human Rights The Cambodian factions continued to infringe on human rights, and UNTAC was largely unable to prevent such abuse. Although all factions violated human rights, especially in the lead-up to the elections, the KPNLF and FUNCINPEC were typically victims rather than perpetrators, with the SOC and the Khmer Rouge in a race for chief violator. Throughout the transitional process, the SOC was accused of election violations, arbitrary violence, and political intimidation against its opponents and the population at large. From November 1992 to January 1993, the SOC force in Battembang Province repeatedly attacked opposition-party offices belonging to FUNCINPEC and the KPNLF. According to David Ashley: “In these typically night-time attacks, grenades or rockets were fired indiscriminately into newly established party offices.”25 The SOC launched a highly orchestrated propaganda campaign and violence against opposition parties, which

S-17

11/24/09

4:25 PM

Page 509

CAMBODIA’S PEACE AGREEMENT

509

subsequently intensified.26 Its violence peaked in December 1992 and continued unabated throughout the election campaign, causing more than 200 deaths between January and May 1993 alone.27 In its tally of casualties, the SOC was second only to the Khmer Rouge, whose abuses inside its own territory were little known but whose army was known to attack ethnic Vietnamese and UNTAC personnel. More than 100 Vietnamese were massacred, and a large number of those who survived fled for Vietnam in fear of further violence. Overall the Khmer Rouge was deemed responsible for the bulk of the violence and for the largest number of deaths.28 This violence did not mean that the UN was unable to maintain the peace. Importantly, levels of violence and political intimidation were lower than at any other time in recent Cambodian history. Disarmament and Demobilization The greatest failure was UNTAC’s inability to disarm the four factional forces as agreed at the Paris conference. The United Nations had deployed a large number of military and police personnel in anticipation of factional disarmament: 15,500 armed infantry and engineering troops in 270 locations across the country, 480 unarmed military observers, and 3,600 civilian police. The Khmer Rouge initially accepted some UNTAC military observers, though it limited their activities in its zones. When cantonment, disarmament, and demobilization began on June 13, 1992, however, the Khmer Rouge stalled, insisting that the withdrawal of Vietnamese troops be verified by UNTAC, that the SOC’s administrative structures be dismantled, and that the SNC be given the power to run the country. For their part, the SOC, FUNCINPEC, and the KPNLF were partially willing to comply with the agreement. Together, they cantoned almost 55,000 soldiers. Yet they refused to participate fully. The Khmer Rouge’s refusal to disarm and demobilize in good faith was undoubtedly the main rationale. The disarmament process eventually failed after UNTAC rejected the Khmer Rouge’s demands and agreed to return surrendered weapons to the other three factions for self-defense.29 Elections Many observers view UNTAC’s Electoral Component as the greatest success of the mission. Several new political parties sprang to life; altogether, twenty parties were registered, including those of the four former warring factions. In spite of threats from the SOC and its political party, the Cambodian People’s Party (CPP), opposition parties were able to establish their party offices across the country. Although KPNLF leader Son Sann threatened to pull out of the electoral process, citing SOC/CPP violence and political intimidation, the KPNLF stayed. In the end, the Khmer Rouge,

S-17

11/24/09

510

4:25 PM

Page 510

CASE STUDIES

which had refused to disarm, was the only Cambodian signatory of the Paris Agreement to withdraw from the electoral arena.30 Most dramatic was voter registration and turnout, which surprised most observers. More than 4.6 of 5.2 million eligible voters had registered to cast their ballots by January 1993. On the days of actual election in May 1993, more than 4 million people turned out at the polls, representing close to 90 percent of registered voters.31 After the elections, FUNCINPEC and the CPP emerged as the dominant parties, receiving 58 and 51 seats respectively in a 120-member Constituent Assembly.32 A coalition government of four elected parties was formed: the CPP, FUNCINPEC, the KPNLF (renamed the Buddhist Liberal Democratic Party, or BLDP), and Moulinaka (another resistance army loyal to Sihanouk, which had fought the SOC in the 1980s but had not been a party to the Paris Agreement). FUNCINPEC president Norodom Ranariddh became first prime minister; CPP leader Hun Sen emerged as second prime minister. The electoral process can be viewed more critically, however. It was deeply marred by the Khmer Rouge’s refusal to disarm and demobilize. Along with continuing cease-fire violations, human rights abuses by all sides (especially the Khmer Rouge and the SOC), and the Khmer Rouge’s withdrawal from the electoral process, the failure of disarmament and demobilization made it impossible for UNTAC to establish the “neutral political environment” deemed necessary for free and fair elections. Tensions between the Khmer Rouge and the SOC escalated after disarmament failed. As the election date drew near, violence further “escalated to a level unprecedented during UNTAC’s presence.”33 Of the Cambodian signatories to the agreement, the Khmer Rouge was the most belligerent and most destructive, followed closely by the SOC. Their behavior shifted from reluctant cooperation to defiance, resistance, and ultimately violence against both political opponents and UN personnel. UNTAC’s electoral mission was, at best, half-accomplished.

Explaining Factional Noncompliance There are two broad sets of explanations for the unwillingness of Cambodia’s factions to comply with their commitments: one set focuses on the nature and aims of the factions themselves; the other focuses on UNTAC’s strategy for trying to manage the factions. Elements of both are needed for a full account of Cambodia’s experience with peace implementation. Factional Nature, Aims, and Behavior To understand the range of noncompliance by Cambodian factions, one can look at three basic influences on their behavior: culture, ideology, and secu-

S-17

11/24/09

4:25 PM

Page 511

CAMBODIA’S PEACE AGREEMENT

511

rity. Each factor played a demonstrable role. Ultimately, however, an interest and security-based explanation is the most persuasive when one examines closely the progressive breakdown of the power equilibrium among the armed factions during implementation. Culture. Cambodian culture provides one explanation of the Cambodian factions’ inability to compromise. According to some scholars of Southeast Asia, Indochinese societies in general have no tradition of genuine compromise. Power sharing among adversaries has no legitimate place because power is seen as a possession that can only be enjoyed exclusively.34 Cambodian behavior, in particular, is often characterized as egoistic, arrogant, uninterested in consultation, deceptive, and vengeful. 35 Importantly, culturalist accounts of Cambodia also point to an unwillingness to accept either defeat or a partial victory, meaning that combat only ends when an opponent is totally destroyed. As former Cambodian politician Bunchan Mol writes in Charret Khmer (Khmer Practice, or Khmer Conduct), perhaps the most widely read book among Cambodians: “We Khmers . . . [i]f one knocks down another person, one will not stop there; one will rush to finish him off by beating him until he either loses consciousness or even dies. . . . [But] if the loser is still alive, it would mean that victory has not yet been won.”36 Furthermore: “A Khmer does not want any others other than himself to become more popular and cannot trust anyone else enough to let the latter soar higher than himself. He must be the supreme leader who stands over and above everyone else.”37 Such notions of defeat and victory help explain why factions have not always spared their enemies, for fear that the latter will seek retribution. They are insecure of their positions, expect revenge, and see competition as inherently zerosum. The cultural perspective alone is inadequate, however, in addition to being reductionist. It cannot explain varying degrees of violence by different factions, who presumably all share Cambodian cultural traits; nor can it account for the Khmer Rouge’s changing behavior from initial cooperation through violent defiance of UNTAC authority. Ideology. Most analysts of the UNTAC operation blame the Khmer Rouge as the most intransigent faction that was willing to manipulate the domestic political situation to gain popular support and return to extreme violence to achieve its goals.38 Such behavior is often explained as rooted in the Maoist radicalism that shaped its history. After years of insurgent warfare and of close relations with Chinese communists, the faction had developed “a strongly nationalist, xenophobic, radical agrarian ideology, and a respect for Maoist tactics of guerrilla warfare.”39 Its stalling tactics and eventual violent recalcitrance against UNTAC, then, can be explained as a continuation of its earlier beliefs and behavior. Even after the Paris Agreement, the Khmer Rouge’s goals were principally to drive ethnic

S-17

11/24/09

512

4:25 PM

Page 512

CASE STUDIES

Vietnamese out of the country, to topple the SOC, and to reestablish its own exclusive rule. When it became clear that UNTAC posed an obstacle to its tactical manipulation of domestic politics, the Khmer Rouge simply changed tactics by withdrawing from the implementation game. While this ideology-based perspective offers some insight into Khmer Rouge behavior, it too is an insufficient and even misleading explanation. Analysts who focus on Khmer Rouge ideology tend to exaggerate the faction’s insubordination. The faction did not fail to abide by its commitments under the agreement to the extent that has been generally assumed. Throughout the transitional period, the Khmer Rouge leadership demonstrated a degree of cooperation and seemed prepared to participate in disarmament, demobilization, and electoral contest. 40 Janet Heininger argues similarly that: “Contrary to most predictions, and in abrupt reversal of its actions leading up to the elections, the Khmer Rouge did not engage in a sustained campaign of terror and intimidation against Cambodia’s voters.”41 Evidence given by Khmer Rouge members who did demobilize further “belies any simple assertion that the PDK [Party of Democratic Kampuchea] never intended to go along with the terms of the Paris Agreements.”42 In fact, the Khmer Rouge leaders and their troops, who never renounced the Paris Agreement, “believed themselves ready to participate in demobilization and elections.”43 The Khmer Rouge also, according to UNTAC force commander John Sanderson, demonstrated cooperation with UNTAC’s repatriation efforts: they “were keen to assist the Repatriation component and receive large numbers of returnees into the areas under their control. . . . [T]hey also welcomed the work of other aid agencies.”44 The emphasis on Khmer Rouge noncompliance and its ideological roots also overlooks the resistance of other factions to peace proposals and implementation efforts that put it at a disadvantage. The SOC had earlier opposed the UN Security Council’s call for a nationwide system of proportional representation, under which resistance forces could have obtained a sizable representation in the National Assembly. The SOC argued instead for a “winner-take-all” electoral system based on single-member constituencies, which would help it secure control over the country. The SOC wanted an election, but only one that helped it monopolize political power. The SOC was intransigent during the transition period as well, offering stiff resistance to UNTAC’s attempts to exercise control over key ministries. Documents further confirmed the SOC’s escalating use of coercion, intimidation, and propaganda to try to defeat FUNCINPEC and other opposition parties at the polls.45 General Sanderson observed that there “is a tendency to blame the [Khmer Rouge] for initiating breaches of the peace process, and their historical baggage, including the secretive nature of the party’s leadership, made them easy targets for such claims.” In fact, according to Sanderson, the “corruption of the peace process” began in Phnom

S-17

11/24/09

4:25 PM

Page 513

CAMBODIA’S PEACE AGREEMENT

513

Penh.46 On his return to Cambodia, Prince Sihanouk quickly attempted to form an alliance between Ranariddh (his son) and Hun Sen of the CPP. A series of clearly orchestrated demonstrations were directed against Khmer Rouge representatives arriving in Phnom Penh, and immediately the angry crowd ran them out of the city on November 27, 1991. The SOC strategy against FUNCINPEC, in turn, was to treat it as though it were a front for the Khmer Rouge, and to frighten people into believing that the Khmer Rouge could thereby fight its way back to power. The SOC’s propaganda campaign against its enemies “offered no hope of reconciliation, only the prospect of a never-ending search for ‘enemies’ and ‘targets,’ with all the violence that this implied.”47 Even after the elections, according to Sanderson, it “was not the Khmer Rouge but rather these armed forces, particularly those of the SOC, which had the capacity to undermine the election or to overturn the verdict of the people.”48 While both the SOC and the Khmer Rouge arguably shared an ideological root and an unwillingness to play the democratic game—core SOC leadership were former Khmer Rouge who had fled to Vietnam in 1978 to avoid Pol Pot’s relentless purges—it is misleading to exaggerate the role their “ideology” played in thwarting UNTAC efforts to implement the peace agreement. In neither case was ideology immutable: for its part, the Khmer Rouge had already publicly renounced communism, and its leaders were willing, at least in principle, to compete in the elections. Instead, evidence indicates that ideological stance depended on factors of interest and self-preservation. It is also difficult to determine the relationship among ideology, personality, and policy. The Khmer Rouge leadership’s communist ideology did not serve the interest of either the peasantry or the working class.49 Neither did its alleged “racialist nationalism” reveal much about Khmer Rouge behavior. 50 Throughout the 1980s, Hun Sen was identified as a “moderate”; yet before he rebelled against Pol Pot in 1978, he had allegedly been one of the dictator’s most enthusiastic lieutenants in massacring Vietnamese. In a raid into a Vietnamese village in 1977, for instance, he ordered his troops to kill all the villagers, including women and children.51 Chea Sim, the top CPP leader and chairman of the National Assembly after the 1993 election, was seen as a “hard-liner.” In recent years, he has enjoyed a new political status as a “moderate.” Ideology is not irrelevant, but it is also not the primary driver of behavior. Finally, the role of ideology, if seen as the only obstacle to peace, cannot explain why all sides cheated on the peace process to a significant extent.52 Interest and security. The central weakness of the culturalist and ideological explanations is straightforward. Neither approach offers an account of variations in spoiler behavior, either between different factions or within one faction over time. Indeed, the cultural factor itself can better be under-

S-17

11/24/09

514

4:25 PM

Page 514

CASE STUDIES

stood by seeing culture in a context of military capabilities and power relations. Throughout Cambodian history, kings and political leaders have almost always rejected calls for accommodation that would diminish their power and security. Compromise was possible only when the adversaries could not destroy each other by force. In Cambodia’s highly unpredictable and factionalized political environment, interest and self-preservation mattered most. All parties tended to view peace tactically, and their ultimate objective was to maximize security and ensure survival. Even those who see Khmer Rouge radicalism and xenophobia as a primary obstacle to the peace process recognize that the faction’s survival was at stake. That the SOC came second only to the Khmer Rouge in political violence only reinforces the assertion that questions of relative security best explain factional behavior. Both cooperation and defiance can be explained by reference to security concerns. To the extent that the Khmer Rouge cooperated in implementation, for example, it did so for strategic reasons. The return of refugees to its zones of control, to take one of Sanderson’s illustrations, would strengthen its local power base. Such security-driven cooperation also highlights the likelihood of the Khmer Rouge resisting full demobilization, since it had a security interest in defending itself in the case of continued war. The Khmer Rouge faced a severe dilemma: it had agreed to a peace accord that required disarmament and demobilization, yet its nearly exclusive source of power was military might.53 The faction’s pre-Paris insistence on a role in a nonelected government was premised on precisely this concern, in order to help it avoid losing an electoral contest where its assets—being military—were significantly less useful. During implementation, then, once the Khmer Rouge realized how negligible were its prospects of securing political power, demobilization of its troops could only be seen as an act of self-destruction. Consequently, the military faction “immediately started defying UNTAC’s preparations for cantonment, disarmament, and demobilization.”54 The Khmer Rouge’s overall defiance of UNTAC was similarly motivated. It believed that it “could get away with stretching the political terms of the agreements to the breaking point,” that the SOC and UNTAC had already seriously violated them, and that the political situation was not developing to its advantage.55 Its leadership appeared to be divided on the question of whether to participate in the forthcoming elections. Its final decision to withdraw from the electoral process suggests that, at that stage of implementation, the Khmer Rouge no longer saw any strategic benefits from compliance with the agreement. That the Khmer Rouge was the most destructive force may have more to do with its increasing level of vulnerability when compared with others. There seemed to be a direct relationship, for example, between increasing violence committed by SOC members and the Khmer Rouge’s ongoing

S-17

11/24/09

4:25 PM

Page 515

CAMBODIA’S PEACE AGREEMENT

515

defiance. The SOC’s attacks on its political opponents began to intensify around November 1992, immediately after the Khmer Rouge had stalled on disarming and was believed to have infiltrated the opposition parties. The SOC’s attacks on opposition parties were concentrated mostly in Khmer Rouge zones, especially in Battembang Province. Moreover, the Khmer Rouge had some reason to fear reprisals from its enemies. After the Vietnamese invasion of Cambodia earlier in 1979, when the Khmer Rouge army was retreating, villagers slaughtered former Khmer Rouge officials with sticks and axes. Some were arrested and executed. 56 According to Kevin Rowley, “some Khmer Rouge cadres admitted privately that the legacy of 1975–78 meant that their leaders could not be safe in areas of Cambodia they did not control.”57 UNTAC’s Effective Loss of Neutrality To explain the Khmer Rouge’s changing behavior toward UNTAC, one also needs to examine the UN’s operational limitations as well as the diplomatic coalition supporting the implementation process. Factional defiance corresponded closely to opportunities created by UNTAC’s obvious inexperience in an operation of this magnitude. Such lack of experience was perhaps understandable, since the world organization had “no precise precedents from which to work.”58 Nonetheless, one can only fully explain the dynamics of noncompliance during implementation in relation to UNTAC’s various operational limitations and, then, to the mission’s eventual loss of effective neutrality as a result of the larger diplomatic context. Operational limitations. UNTAC’s operational problems, especially at the start-up phase of the mission, increased the factions’ opportunities for noncompliance as well as decreased any embryonic confidence they might have had in the mission’s capabilities. First, the United Nations did not have any real understanding of Cambodia’s needs. It was not alone: other international agencies, along with potential donor countries, lacked basic knowledge about Cambodia. A UN document itself acknowledged the need for more and better information when it noted that “no effective program of national reconstruction can be initiated without detailed assessments of Cambodia’s human, natural and other economic assets.” The UN document added that “it will be necessary for a census to be conducted, developmental priorities identified, and the availability of resources, internal and external, determined.”59 After the Paris Agreement was signed, several survey and fact-finding missions were sent to the country, the first of which visited Cambodia from October 31 to November 13, 1991. This interagency humanitarian mission, led by Deputy Special Representative Dennis McNamara, aimed to assess Cambodia’s immediate needs for the first six months of implementation.

S-17

11/24/09

516

4:25 PM

Page 516

CASE STUDIES

The mission comprised representatives from almost every component of the UN system and examined virtually all sectors of Cambodian society: health, nutrition and sanitation, drinking water, and food and agriculture; education, vocational training, and employment; shelter and physical infrastructure; and internally displaced persons and other vulnerable groups, including flood victims.60 An earlier survey conducted in 1990 also aimed to gather data on Cambodia, but it had focused more on the potential UN role in resolving the Cambodian conflict.61 Second, UNTAC faced numerous staffing problems. It was generally agreed that training and experience were severely lacking, and that there was a serious shortage of qualified personnel in important areas ranging from public security to civil administration. According to Sanderson, “many were selected for UNTAC and elevated to positions for which they were not equipped by either training or experience.”62 With such an ambitious peace mission taking shape, the United Nations could not afford to be selective, and the organization essentially recruited anyone willing to work in Cambodia. The United Nations wanted quick action, so even concerns about minimum standards of discipline, sensitivity to religion, and appreciation of local customs were not top priorities. Third, UNTAC suffered from a lack of planning and coordination, both in New York and in the field. The United Nations had been caught unprepared when the Paris Agreement was signed. As Sanderson described the situation at headquarters: “No substantial planning had taken place in anticipation and, other than on the initiative of some contributing countries, no contingents had been earmarked for the operation.” Although a UN Advance Mission in Cambodia (UNAMIC) was present, it “appeared to lull the United Nations bureaucracy into a false sense of security.” UNAMIC “was not organized even to look after itself” and “staggered from one crisis to another as it attempted to react to circumstances for which it was illprepared, leaving it unable to execute the important task of planning for UNTAC.”63 Within UNTAC itself, “there never was any strategic planning,” and from “the beginning, each component conducted a separate survey mission and there was no coordination beforehand to determine the strategic method.”64 In the early stages of the transitional period, such dilatory planning and ambiguity about timetables made “reluctant cooperation” from the parties, especially from the Khmer Rouge, a rational response. Civil administration, police, and peacekeeping. Factional defiance toward UNTAC became more obvious when UNTAC’s key components proved unable to take effective, timely action due to belated deployment and a paucity of personnel. Regarding Civil Administration, the United Nations had “no precedents, no experience, [and] no guidelines upon which to draw,” and “[in] some ways it was doomed from the start.”65 This mission component had an enormous mandate, and an extremely small staff. Roughly 170 mission staff were expected to oversee more than

S-17

11/24/09

4:25 PM

Page 517

CAMBODIA’S PEACE AGREEMENT

517

100,000 Cambodian civil servants under SOC control alone, with no guidelines as to how they should exercise oversight. Dependent on local interpreters, they ended up learning by trial and error. Understaffed, the component was also unable to operate effectively outside of Phnom Penh. This lack of geographical reach limited UNTAC control over Khmer Rouge administration, allowed the SOC to dominate provincial administrative structures and gave SOC officials reason to complain about UNTAC’s bias against them. UNTAC’s Civilian Police (CIVPOL), perhaps the least effective of all mission components, failed to fulfill its primary mandate as supervisor and controller of local police. Although UN police were given powers of investigation, arrest, and detention for the first time in UN history, they were generally ineffective. As Charles Call and William Stanley note in Chapter 11, police contributions to peace missions vary dramatically, from local cops in countries without national police forces through border guards, military police, and paramilitary forces. Rarely do they meet even such basic skill requirements as relevant language and driving ability. According to Chris Eaton, this “was particularly true of UNTAC in Cambodia and seriously debilitated the capacity of its CIVPOL component to efficiently and effectively meet its task.”66 Several of the UNTAC police posed discipline problems. Worse, no effective command system existed to support them, rendering even their better efforts less effective.67 Moreover, local police far outnumbered them, thus limiting CIVPOL’s operational effectiveness. Although they were the largest of all UNTAC components, the Military Component of peacekeepers (MILCOM) were not fully deployed until ten months after the signing of the agreement. By April 1992, they had only 3,694 troops in Cambodia, leaving them unprepared for the cantonment, disarmament, and demobilization (second-phase operations) scheduled to begin at the end of May. Full deployment was not completed until August, two months after this most important phase had already started. This slow troop deployment proved detrimental to UN peace efforts, contributing to “an environment of instability which made all parties—particularly the skittish Khmer Rouge—reluctant to demobilize and disarm.”68 Cumbersome and inflexible procurement and budgetary processes, problems of command and control, and poor communication between New York and Phnom Penh further restricted MILCOM’s ability to operate to full advantage.69 Coalition bias. Noncompliance with the agreement on the part of the Cambodian signatories requires still another factor to be fully explained. UNTAC could not have succeeded to the extent that it did without continuing support from the external signatories of the Paris Agreement. Coordinated diplomatic support for UNTAC was provided through many channels. A group called the Extended Permanent Five (EP5) consisted of ambassadors from Indonesia, Japan, Australia, Thailand, and Germany,

S-17

11/24/09

4:25 PM

Page 518

518

CASE STUDIES

alongside the five permanent members of the UN Security Council. These ten governments played a key role in instilling confidence among the Cambodian factions. The EP5 also worked closely with another group of ambassadors in New York known as the Core Group and occasionally attended meetings of the Mixed Military Working Group. Various governments also undertook specific confidence-building measures. Before the elections, for example, the United States and Thailand, both of which had been criticized for earlier inaction against the Khmer Rouge, conducted joint military exercises near the Thai-Cambodian border. Some believe this kind of military presence saved the UNTAC-organized elections.70 Still, evidence suggests that, at the very least, factions perceived a lack of impartiality on the part of external sponsors of UNTAC. Although the United States had always been biased against the SOC, for instance, it also opposed the Khmer Rouge’s return to power, a policy seen by the faction as working against its interests. Indeed, by the end of the 1980s, all major parties interested in the conflict, except China, agreed on the need to neutralize the Khmer Rouge, from Vietnam and ASEAN members to the United States, Britain, Japan, and other influential states.71 Nate Thayer also argues that the West was moving toward legitimizing the Vietnamese-installed SOC for fear of a militarily strong Khmer Rouge army, and even ignored internal conditions that continued to feed anti-Vietnamese resistance.72 Few scholars argued that the agreement would work to the Khmer Rouge’s advantage. Not surprisingly, the faction was observed by at least one analyst to feel isolated and unhappy with the new political arrangement, increasing the likelihood that its leaders would create problems for UNTAC.73 During the transitional period, the former resistance forces and some observers saw major external players (especially Western powers) as lacking impartiality toward the factions. Much evidence indicates that Australia, France, and the United States were far from neutral.74 No other great power has been as active as the United States in amassing evidence of war crimes against the Khmer Rouge leadership.75

Alternative Strategies? The last section of this study examines whether there were alternatives for implementers in order to have ensured a greater degree of success. Should UNTAC have adopted a more coercive strategy toward noncompliance, as many observers have advocated? Was UNTAC’s judgment sound when it continued with elections on schedule even after the Khmer Rouge had withdrawn? The argument here is that coercive measures, or peace enforcement, would have done more harm than good, and that leaving the Khmer Rouge behind was the second-best option. One could argue that UNTAC’s failure to enforce compliance was the

S-17

11/24/09

4:25 PM

Page 519

CAMBODIA’S PEACE AGREEMENT

519

main reason for the factions’ increasingly belligerent behavior. The UNTAC leadership abided by traditional peacekeeping doctrine that force be used minimally and then only in self-defense.76 Both realist and liberal scholars view UNTAC’s unwillingness to use enforcement measures in the face of noncooperation by the parties as a serious weakness. 77 Gerard Porcell, chief of UNTAC’s Civil Administration, also faulted the UN for its unwillingness to force the Khmer Rouge to abide by the agreement. In his words: “We [the UN] don’t have the will to apply the peace accords. This absence of firmness with the Khmer Rouge was a sort of signal for the other parties who saw there the proof of UNTAC’s weakness towards the group that from the start eschewed all cooperation.”78 It is debatable, however, whether “firmness” with the Khmer Rouge would have generated more cooperation, given its security concerns. Strict enforcement would have undermined not only what legitimate authority UNTAC enjoyed in Cambodia but also its credibility internationally. A number of countries contributing troops to the mission were unwilling to engage in enforcement action. According to Michael Doyle, some of the Southeast Asian states even expressed eagerness to accommodate the Khmer Rouge, a position not shared by the other contributors.79 UNTAC also did not have the necessary staff to conduct enforcement effectively. It did not possess an independent judiciary and had to rely on the SOC’s heavily biased courts. The Cambodian police and court system did not bother to investigate and prosecute offenders, especially when the accused belonged to its party. UNTAC did establish an Office of Special Prosecutor, which had the power to arrest and detain suspects, but its efforts faltered when the Cambodian courts refused to hear the special prosecutor, who became “the only lawyer ever to be thrown out of the Phnom Penh Municipal Tribunal of Justice.”80 Military enforcement, in particular, might have done more harm than good. Enforcement action arguably would have led UNTAC into a protracted war with the Khmer Rouge. It is worth remembering that more than 100,000 heavily armed Vietnamese troops had failed to pacify the Khmer Rouge army in the 1980s. Heininger persuasively argues that “[refraining] from using force was more effective in isolating the Khmer Rouge than any overt use of force.”81 This does not suggest that all forms of sanction were impossible or undesirable. Several leaders were penalized for violating the Electoral Law, for example, including Prince Norodom Chakropong (SOC deputy prime minister), Kim Bo (SOC governor of Sihanoukville), and Prum Neakareach (Moulinaka Party). UNTAC’s legal action against SOC members, however, did not threaten the party’s survival. Here, firmness could be effective because it was proportionate and because the adversaries did not perceive the level of their insecurity to be too high.

S-17

11/24/09

520

4:25 PM

Page 520

CASE STUDIES

Cooperation and compromise in Cambodia resulted from a rough balance of power between the parties. By contrast, hegemonic power in the country tended to discourage compromise. The UN needs to be aware of local power dynamics and, to the extent that a power balance is a necessary condition for the continued peace process, needs to try to prevent shifts in the security environment that advantage a single party. Should the United Nations fail to sustain such an equilibrium of power, its only alternative is to make hegemonic politics stable. Hegemonic peace, meaning political stability obtained through hegemonic control or domination, should be seen as second-best to democratic peace as an outcome. Stedman touches on something very important when arguing that UNTAC’s improvised strategy toward the Khmer Rouge “was imaginative and effective.”82 The UN’s change of strategy (from persuading the Khmer Rouge to stay involved, to leaving it behind) is seen as the second-best option. According to Stedman, UNTAC’s commitment to holding the elections remained unchanged, even without Khmer Rouge participation. UNTAC then worked to delegitimize the faction’s demands by seeking international consensus and by changing its military role to defend the polls against Khmer Rouge attack. In the end, the strategy was “vindicated” as the Khmer Rouge leadership split and disintegrated. This was “the prototype of the departing train strategy for managing spoilers.”83 While UNTAC’s strategy kept the SOC/CPP in the electoral process, it reinforced the opposition parties’ perception that UNTAC was either unwilling or unable to create a neutral political environment for free and fair elections. UNTAC’s failure to gain more control of the SOC’s civilian administration further resulted in its heavy dependence on the latter’s cooperation. The SOC ruled 80 percent of the country, and UNTAC did not negotiate with it.84 A new imbalance of politico-military power between the SOC and its adversaries contributed to the SOC’s willingness to allow the elections to take place, as it became convinced that it would win this time. In case it did lose, the SOC worked to ensure that it would not have to transfer power to the winner voluntarily unless the latter was willing to share power with the SOC as well as to work within the SOC-dominated political and security environment. Why the “departing train strategy” succeeded to the extent that it did has to be put in the broader security context. UNTAC’s “coercive” strategy did not increase the vulnerability of the Khmer Rouge and did not threaten its survival, thus making it acceptable for its leaders to wait for a chance to renegotiate with a new government, since its chance of winning in the elections was quite slim. As has been discussed, however, the international community rejected any inclusion of Khmer Rouge leaders in the coalition government. More notably, the post-UNTAC Khmer Rouge disintegration did not result from the application of a coercive strategy alone. It occurred

S-17

11/24/09

4:25 PM

Page 521

CAMBODIA’S PEACE AGREEMENT

521

after a series of negotiations that led to the granting of amnesty and inclusion of Khmer Rouge defectors, including top guerrilla officials. This supports one of my arguments that the strategy of inclusion may provide the faction with short-term security, but may ultimately lead to its fragmentation and elimination. The extent to which the Khmer Rouge’s disintegration and the CPP’s reassertion of hegemonic control in the country have affected Cambodian politics remains to be seen. Evidence, however, points to Hun Sen’s increasing authoritarian behavior. Again, Stedman touches on an important point: “Time has not been so kind in evaluating UNTAC’s strategy toward SOC, [whose] steadily increasing grip on power since 1993 caused several experts on Cambodia in 1996 to warn of a ‘creeping coup.’”85 Hun Sen’s preemptive coup in early 1997 was possible because his military power (rebuilt and strengthened over a period of four years after the elections in 1993) was far superior to that of Ranariddh.

Aftermath: The Price of Qualified Success Assessments of Cambodia after the UN’s withdrawal have alternated between optimism and pessimism. When Yasushi Akashi left Cambodia in September 1993, he asserted that UNTAC had succeeded in laying “a firm foundation for Cambodian democracy.” Former Australian foreign minister Gareth Evans, after visiting Cambodia in 1994, also voiced his “strong impression” that the country had returned to normal.86 To others, however, the costly UN operation had largely gone to waste, since most of the enemies of this liberal democracy—corruption, warlords, Khmer Rouge terrorism, and banditry—had returned.87 A more measured examination shows a Cambodia that has made progress, if limited, in the areas of economic development, military conflict, human rights, and democracy. Renewed armed conflict between the new coalition government and the Khmer Rouge army initially followed the elections, though it gradually came to an end. After the government granted the Khmer Rouge amnesty and some autonomy, thousands of guerrilla troops defected en masse in late 1996, led by former Khmer Rouge minister of foreign affairs Ieng Sary. Shortly thereafter, the entire movement began to disintegrate. Broader political developments remain discouraging, according to most analysts. Soon after the new government was established, the two dominant parties—the CPP and FUNCINPEC—began to fragment. By the end of 1994, Cambodia appeared to be entering a vicious cycle of events similar to that of the Lon Nol era.88 One observer saw 1995 as the year in which Cambodia moved “from hope to despair” as traditional elites moved to reinforce factional hierarchy and the monarchy as well as to marginalize democratic forces.89 Personal rivalries in the CPP and FUNCINPEC, and

S-17

11/24/09

522

4:25 PM

Page 522

CASE STUDIES

popular discontent with both parties, spawned a factional realignment in 1996. Factionalism within the government surged when Ranariddh decided (in March 1996) to form a National United Front (NUF) with the BLDP and other parties in their joint struggle against the CPP. Early July 1997 was a watershed. After two days of street fighting in Phnom Penh, Hun Sen’s troops swiftly defeated the royalists. The CPP leader wasted no time ousting Ranariddh. Immediately after Ranariddh’s overthrow, Hun Sen ordered his troops to push royalist remnants north toward the Thai-Cambodian border. By August, they were desperately defending their last border stronghold of O Smach. Resistance military leaders then admitted that this last possession might fall at any time, but claimed that their 3,000 troops were resolved to fight back. With both Ranariddh and his military leaders considerably weakened, royalist party members found their party up for grabs. Top officials not only competed to take over the job of the ousted Ranariddh, but also formed their own political parties. Meanwhile, the CPP-dominated government began to prepare for the elections of July 26, 1998. The CPP won the largest number of seats (64), followed by FUNCINPEC (43) and the Sam Rainsy Party (15). Claiming that the CPP’s victory had come through political intimidation and fraud, the opposition refused to accept the results and organized protests against the government. Social unrest intensified and resulted in clashes between protesters and government troops. By mid-September, the crisis remained unresolved, as the CPP refused to make concessions to the opposition. It was not until December that the CPP and FUNCINPEC agreed to form another coalition government, this time with Hun Sen as the sole premier. The election of 1998 can only be explained as a result of CPP tactics to weaken its political opponents before election day. To be fair, any assessment of post-UNTAC human rights conditions must be placed in the historical context. To argue that there is no improvement whatsoever on this particular front is misleading. As discussed earlier, factions violated human rights, especially during and after the coup in July 1997, but the scales are nowhere comparable with the atrocities committed under the Sihanouk, Lon Nol, Khmer Rouge, and PRK/SOC regimes.90 Some scholars may have overstated UNTAC’s impact on Cambodian society, but there is nonetheless hope that measures to promote human rights education, social justice, and civil society will have a long-term positive impact on this unstable political environment.91 Moreover, apprehension about violence during the 1998 elections proved exaggerated, as the level of violence was low. The post-UNTAC economy initially made some progress, but later became stagnant, as political instability worsened. In 1995, total exports increased ninefold from the preceding year, and the government planned to export rice for the first time in twenty-five years. Inflation rates dropped

S-17

11/24/09

4:25 PM

Page 523

CAMBODIA’S PEACE AGREEMENT

523

from 26.1 percent in 1994 to only 3.5 percent in 1995 (lower than the official forecast of 5–6 percent). It is estimated that inflation rates for 1996 and 1997 would be roughly the same as that in 1995 or a bit higher. GDP growth for 1995 was about 7.5 percent. The year 1997 witnessed an economic downturn, however, with the growth rate estimated to be falling to between 0 and 2 percent.92 The year 1998 saw a growth rate of only 1 percent. The government remained deeply cash-strapped. In short, UNTAC’s qualified success created a number of uncertainties and has left political problems unresolved. The country’s ongoing political crises have so far limited the potential for the growth of civil society, economic development, and democratic maturation. The 1993 election, while it can be characterized as “perhaps the most democratic election in Southeast Asian history,”93 may not have been sufficient to break the powerful chain of violence and authoritarian rule in a war-torn society like Cambodia.

Conclusion This study has described the uphill battle faced by UN efforts to implement the Paris Agreement. Cambodia had undergone an uninterrupted series of antidemocratic regimes, from Prince Sihanouk’s paternalistic authoritarianism through Lon Nol’s republican authoritarianism and Pol Pot’s revolutionary totalitarianism to Hun Sen’s socialist dictatorship in the 1980s. Cultural, ideological, and institutional legacies limited the UN potential to end war on a self-enforcing basis. This chapter has examined UNTAC’s partial success and has investigated why compliance was so lacking on the part of the warring factions. I have argued that security fears were central to noncompliance. Traits that can be interpreted as cultural or ideological appear to have been malleable in the Cambodian case. Former foreign minister Ieng Sary was seen as a hawkish Khmer Rouge leader (akin to Pol Pot and Ta Mok), but it was he who cooperated with the government in 1996, a move that brought down the guerrilla force. Pol Pot and Ta Mok remained hard-liners, but their position may have been shaped more by the fact that they were the main targets for prosecution. It was doubtful whether they would ever obtain the kind of amnesty Ieng Sary received. Moreover, if cultural and ideological factors are taken too seriously, international actors in war-torn countries like Cambodia risk overlooking adversaries’ security fears. In contexts of fluctuating power relations, security and insecurity are far better indicators of likely spoiling or potential cooperation. After the Paris Agreement unintentionally helped erode the rough politico-military balance of power between the CGDK and the SOC, the implementation process ran into several difficulties. The CGDK disintegrated into three different parties, thus making it much easier for the SOC

S-17

11/24/09

524

4:25 PM

Page 524

CASE STUDIES

to reassert its hegemonic control in all major cities, especially the capital. This is perhaps why UNTAC had difficulty convincing the resistance factions (particularly the Khmer Rouge) that it could create a neutral political environment. UNTAC’s slow deployment of its civilian, security, and military personnel further reduced their confidence. External powers also contributed to UNTAC’s operational problems by not disbursing their financial contributions quickly enough, and by failing to act impartially in the eyes of the resistance forces. The Western states’ primary strategy was to prevent the Khmer Rouge from returning to power; this strategy weakened the CGDK and strengthened the SOC. Even after the elections in 1993, all external actors involved in Cambodia (and donors in particular) refused to accept any inclusion of the Khmer Rouge in the new government. Meanwhile, they were either unwilling or simply unable to prevent Hun Sen from repressing his coalition partners. This is not to argue that the incumbent Hun Sen regime is less preferable than the murderous Pol Pot regime, but rather to explain UNTAC’s partial success and to shed light on what more can be done to maximize the operational effectiveness of peace missions. Liberal solutions for a country like Cambodia are elusive, especially in the short to medium term. It is up to the international community to choose between democratic peace and hegemonic stability. In either case, coercive strategies have limitations. To argue in favor of meeting all parties’ security needs collides with the ethical judgment that some parties’ needs are more defensible than others. After all, the Khmer Rouge regime killed more than 1 million people and its leaders may deserve different treatment. While recognizing the powerful merits of this view, it is important to keep in mind the following points: First, the Khmer Rouge “killing fields” were the byproduct of political violence and war that had begun in the 1960s. The Khmer Rouge’s rise to power was also a by-product of unrestrained passion against decades of foreign interference in Cambodia and injustices that had afflicted the country. Second, there is no conclusive evidence to support the argument that the Khmer Rouge leadership, because of its unchanged ideology and insatiable lust for power, was still genocide-bound. The killings during the transitional period and after resulted from the growing threat to its survival and the politics of exclusion. Last and most important, the Khmer Rouge rebellion came to an end after its leaders had received pardons and security guarantees. This indicates that the movement was far from monolithic and seemed united only by a common threat. Take this common threat away, and such a movement is more likely to collapse, which it did. The lesson of the Cambodian case is that compromise is possible when factions reach a balance of power. Arguably, the danger of a power equilibrium is that it can also entrench civil war and perpetuate a war economy. Stalemate is far more likely to facilitate peace, by contrast, if foreign pow-

S-17

11/24/09

4:25 PM

Page 525

CAMBODIA’S PEACE AGREEMENT

525

ers jointly exert pressure on adversaries to reach settlement as well as refrain from exacerbating the conflict themselves or altering the conditions that make peace possible in the first place.

Notes 1. The main schools of thought on Cambodia can be characterized as follows. “Culturalists” assume that ambitious international action is futile in a traditional society characterized by personal politics, unwillingness to compromise, and hostile factionalism. “Institutionalists” see Cambodia as a failed or failing state that cannot be transformed into a liberal democracy because warring factions relentlessly pursue power with no higher authority to regulate their behavior. “Moralists/legalists” inveigh against the agreement itself because it is seen to legitimize a barbarous party, the Khmer Rouge, whose leaders were poisonous to peace and democracy. “Liberal internationalists,” however, argue that there is some utility in making peace in Cambodia. For a review of these perspectives, see Sorpong Peou, Conflict Neutralization in the Cambodia War: From Battlefield to Ballot-Box (New York: Oxford University Press, 1997). 2. Stephen John Stedman, “Spoiler Problems in Peace Processes,” International Security 22, no. 2 (fall 1997): 5–53. 3. See David Chandler, Brother Number One: A Biography of Pol Pot (Boulder, Colo.: Westview Press, 1992), pp. 120, 134. 4. See Richard Thornton, “Distant Connections: Superpower Rivalry in the Middle East and Southeast Asia, 1977–1979,” Asian Affairs 11, no. 3 (fall 1984): 1–27. 5. On the Chinese and Soviet supply of weapons and equipment to the Cambodian factions, see Russell R. Ross, ed., Cambodia: A Country Study (Washington, D.C.: Library of Congress, Federal Research Division, 1990), pp. 271–276; and Nayan Chanda, Brother Enemy: The War After the War (San Diego: Harcourt Jovanovich, 1986), pp. 348–349, 381–382. 6. Justus M. Van Der Kroef, “A New Phase in the Cambodian Conflict,” Issues and Studies 21, no. 7 (July 1985): 109–131. 7. Tim Huxley, “Cambodia in 1986: the PRK’s Eighth Year,” Southeast Asian Affairs (1987): 161–173. 8. William Duiker, “Vietnam in 1985: Searching for Solutions,” Asian Survey 26, no. 1 (January 1986): 102–111; and Nayan Chanda, “Cambodia in 1986: Beginning to Tire,” Asian Survey 27, no. 1 (January 1987): 115–124. 9. Thaddee In, “Cambodia: L’Impasse?” Politique Internationale 43 (spring 1989): 117–128. 10. As of 1991, the combined military forces of the factions totaled around 200,000, deployed in some 650 locations; their militias totaled 250,000, operating across the country; altogether, they were armed with more than 300,000 weapons of many types and with about 80 million rounds of ammunition. UN Document S/23613, in UN, The United Nations and Cambodia, 1991–1995, Blue Book Series vol. 2 (New York: United Nations, 1995), p. 167. 11. Ibid., p. 173. 12. Jai Bhagwan, “Khmer Rouge: The Bone of Contention in Kampuchea,” Strategic Analysis 12, no. 8 (November 1988): 893–903. See also Sina Than, “Cambodia 1990: Towards a Peaceful Solution,” Southeast Asian Affairs (1991): 83–104.

S-17

11/24/09

526

4:25 PM

Page 526

CASE STUDIES

13. Peter Wallensteen, “Peace-Making in Cambodia: An Analytical Framework,” in Ramses Amer, Johan Saravanamuttu, and Peter Wallensteen, eds., The Cambodian Conflict, 1979–1991: From Intervention to Resolution (Penang, Malaysia: Sinaran Bros. Sdn. Berhad, 1996), p. 7. 14. On the negotiating history, see Dipankar Banerjee, “The Kampuchean Questions: Nearer to a Solution,” Strategic Analysis 12, no. 11 (February 1989): 1283–1298; Friedemann Bartu, “Kampuchea: The Search for a Political Solution Gathers Momentum,” Southeast Asian Affairs (1989): 171–184; Pao-min Chang, “Kampuchea Conflict: The Diplomatic Breakthrough,” Pacific Review 1, no. 4 (1988): 429–437; and Khatharya Um, “Cambodia in 1988: The Curved Road to Settlement,” Asian Survey 29, no. 1 (January 1989): 73–80. 15. Justus M. Van Der Kroef, “Cambodia in 1990: The Elusive Peace,” Asian Survey 31, no. 1 (January 1991): 94–102. 16. This description of the contents of the agreement is derived from UN, The United Nations and Cambodia. 17. Steven R. Ratner wrote, “For the first time, the international community has empowered the United Nations to undertake key aspects of the civil administration of a member state.” See Ratner, “The Cambodian Settlement Agreements,” American Journal of International Law 87, no. 1 (January 1993): 12. 18. Friedrick Z. Brown, “Cambodia in 1992: Peace at Peril,” Asian Survey 33, no. 1 (January 1993): 83–90. 19. According the Janet Heininger, “UNTAC’s military presence discouraged the [Cambodian] parties from resuming combat and gave Cambodia nearly two years of breathing room for political development. By providing an internationally sanctioned time-out from fighting, UNTAC reinforced the people’s desire for peace.” See Heininger, Peacekeeping in Transition: The United Nations in Cambodia (New York: Twentieth Century Fund Press, 1994), p. 137. 20. Eva Arnving, “Women, Children, and Returnees,” in Peter Utting, ed., Between Hope and Insecurity: The Social Consequences of the Cambodian Peace Process (Geneva: UNRISD, 1994), pp. 156–157. 21. In the early 1990s, Cambodia experienced the world’s highest percentage of mine victims, with amputees alone numbering over 30,000. Human Rights Watch, Landmines in Cambodia: The Coward’s War (New York: Human Rights Watch, September 1991). 22. Friedemann Bartu, “Kampuchea: The Search for a Political Solution Gathers Momentum,” Southeast Asian Affairs (1989): 171–184. 23. Van Der Kroef, “Cambodia in 1990.” See also Mya Than, “Rehabilitation and Economic Reconstruction in Cambodia,” Contemporary Southeast Asia 14, no. 3 (December 1992): 269–286; and UNDP, Socio-Economic Consequences of the UNTAC Operation, mimeograph, November 1992. 24. Grant Curtis, “Transition to What? Cambodia, UNTAC, and the Peace Process,” in Utting, Between Hope and Insecurity, p. 60. See also Elisabeth Uphoff Kato, “Quick Impacts, Slow Rehabilitation in Cambodia,” in Michael W. Doyle, Ian Johnstone, and Robert C. Orr, eds., Keeping the Peace: Multidimensional UN Operations in Cambodia and El Salvador (Cambridge: Cambridge University Press, 1997), p. 186. 25. David Ashley, “The Nature and Causes of Human Rights Violations in Battembang Province,” in Steven Heder and Judy Ledgerwood, eds., Propaganda, Politics, and Violence: Democratic Transition Under United Nations PeaceKeeping (Armonk, N.Y.: M. E. Sharpe, 1996), p. 170. 26. Judy Ledgerwood, “Patterns of CPP Political Repression and Violence During the UNTAC Period,” in Heder and Ledgerwood, Propaganda, Politics, and Violence in Cambodia, pp. 114–133.

S-17

11/24/09

4:25 PM

Page 527

CAMBODIA’S PEACE AGREEMENT

527

27. Kate G. Frieson, “The Cambodian Elections of 1993: A Case of Power to the People?” in Robert H. Taylor, ed., The Politics of Elections in Southeast Asia (Cambridge: Cambridge University Press, 1997), p. 235. 28. Ledgerwood, “Patterns of CPP Political Repression,” p. 132. 29. Peou, Conflict Neutralization in the Cambodia War, pp. 61–66. 30. Ibid., pp. 68–70. 31. Ibid., pp. 66–68. 32. This body became the National Assembly in September 1993. 33. C. M. Lee Kim and M. Metrikas, “Holding a Fragile Peace: The Military and Civilian Components of UNTAC,” in Doyle, Johnstone, and Orr, Keeping the Peace, p. 121. 34. See Michael Leifer, “Obstacles to a Political Settlement in Indochina,” Pacific Affairs 58, no. 4 (winter 1985): 626–636; and Michael Leifer, “PowerSharing and Peacemaking in Cambodia?” SAIS Review 12, no. 1 (winter–spring 1992): 139–153. See also Abdulgaffa Peang-Meth, “The United Nations Peace Plan: The Cambodian Conflict and the Future of Cambodia,” Contemporary Southeast Asia 14, no. 1 (June 1992): 33–46. 35. According to one journalist, the common view was that “behind the famous ‘Khmer smile’ of a people noted for their gentle manners and simple lifestyle lay a capacity for the most savage violence and tendency for self-destruction.” Agence France Presse, September 9, 1998. 36. Bunchan Mol, Charret Khmer (Phnom Penh, Cambodia: Residence No. 79, Hem Cheam Boulevard, 1973), p. 180. In this book, Mol describes the bestial nature and divisions of the Khmer Issarak liberation movement (meaning Khmers who are their “own masters”), formed in 1946, which he helped lead against the French colonists. In writing about political events through 1973, he laments the inability of Khmers to work together with mutual tolerance. 37. Ibid., p. 217. 38. David Roberts, “Toward Peace in Cambodia?” Politics 12, no. 2 (October 1992): 9–13. 39. Michael Wesley, “The Cambodian Waltz: The Khmer Rouge and United Nations Intervention,” Terrorism and Political Violence 7, no. 4 (winter 1995): 61. 40. See Peou, Conflict Neutralization in the Cambodia War, p. 270. 41. Heininger, Peacekeeping in Transition, p. 1. 42. Steven Heder, “The Resumption of Armed Struggle by the Party of Democratic Kampuchea: Evidence from National Army of Democratic Kampuchea ‘Self-Demobilizers,’” in Heder and Ledgerwood, Propaganda, Politics, and Violence in Cambodia, p. 109. 43. Ibid. 44. John M. Sanderson, “UNTAC: Successes and Failures,” in Hugh Smith, ed., International Peacekeeping: Building on the Cambodian Experience (Canberra: Australian Defense Studies Centre, University College, 1994), p. 23. 45. Ledgerwood, “Patterns of CPP Political Repression,” p. 116. 46. Sanderson, “UNTAC,” p. 18. 47. Ledgerwood, “Patterns of CPP Political Repression,” p. 130. 48. Sanderson, “UNTAC,” p. 27. 49. See Ben Kiernan, The Pol Pot Regime (New Haven, Conn.: Yale University Press, 1996). 50. Sorpong Peou, Holocaust and Genocide Studies 11, no. 3 (winter 1997): 413–425. 51. See Bangkok Post, January 25, 1999, p. 11. 52. Roberts, “Toward Peace in Cambodia?” pp. 9–13. 53. Michael Wesley, “The Cambodian Waltz,” p. 70.

S-17

11/24/09

528

4:25 PM

Page 528

CASE STUDIES

54. Ibid., p. 70. 55. Heder, “The Resumption of Armed Struggle,” p. 109. 56. This information is based on my personal experience living in Cambodia during and after the Pol Pot regime. 57. Kevin Rowley, “The Making of the Royal Government of Cambodia,” in Viberto Selochan and Carlyle A. Thayer, eds., Bringing Democracy to Cambodia (Canberra: Australian Defense Studies Centre and Australian Defense Force Academy, 1996), p. 7. With the exception of Michael Wesley, Rowley, and a few others, few scholars have paid attention to the Khmer Rouge’s fear of reprisals from its enemies as an explanatory factor in its behavior. Apparently, Pol Pot did not cooperate because he was despised and feared death. When his colleagues sought to negotiate with the government at his expense, the violent power struggles began in Anlong Veng and perhaps led to his death, despite reports that he died of a heart attack. 58. Ratner, “The Cambodian Settlement Agreements,” p. 13. 59. UN Document A/46/608-S/23177, October 30, 1991, p. 56. 60. UN, “Report of the UN Inter-Agency Humanitarian Mission to Cambodia, 31 October–13 November 1991.” Agencies involved in the mission were the UN International Children’s Emergency Fund (UNICEF), the World Food Program (WFP), the Food and Agriculture Organization (FAO), the UNDP, the UN High Commission for Refugees (UNHCR), the World Health Organization (WHO), the International Labor Organization (ILO), the UN Educational, Scientific, and Cultural Organization (UNESCO), and the Economic and Social Commission for Asia and the Pacific (ESCAP). 61. UN, “Report of the United Nations Fact-Finding Mission on Present Structures and Practices of Administration in Cambodia: 24 April–9 May 1990,” New York, June 1990, p. 1. 62. Sanderson, “UNTAC,” p. 19. 63. Ibid., p. 18. 64. Ibid., p. 22. 65. Lyndall McLean, “Civil Administration in Transition: Public Information and the Neutral Political Environment,” in Smith, International Peacekeeping, p. 48. 66. Chris Eaton, “The Role of Police in Institution Building,” in Smith, International Peacekeeping, p. 61. 67. Sanderson, “UNTAC,” p. 19. 68. Lee Kim and Metrikas, “Holding a Fragile Peace,” p. 125. 69. Ibid., pp. 128–129. 70. Jonathan Tombes, “Cambodia: Lessons for UN Peacekeeping,” American Enterprise 5, no. 3 (May–June 1994): 46–57. 71. Bhagwan, “Khmer Rouge,” pp. 893–903. 72. Nate Thayer, “Cambodia: Misperceptions and Peace,” Washington Quarterly 14, no. 2 (spring 1991): 179–191. 73. Ganganath Jha, “Security Concerns of the Emerging New Order in Cambodia,” Strategic Analysis 4, no. 11 (February 1992): 1313–1325. 74. Regarding the French position, President François Mitterrand declared in January 1990 that “no compromise is acceptable with the Khmer Rouge.” This was further echoed by UNAMIC’s military chief (General Jean-Michel Loridon of France), who made known his willingness to militarily compel the Khmer Rouge to comply with the agreement. Apparently, France encouraged the alliance between FUNCINPEC and the SOC—a political development considered by General Sanderson as an instance that corrupted the peace process. France, according to one

S-17

11/24/09

4:25 PM

Page 529

CAMBODIA’S PEACE AGREEMENT

529

closer observer, was also “eager to re-establish its influence in Cambodia and Vietnam.” Consequently, France “condemned the Khmer Rouge and was cool towards the KPNLF but fairly supportive of [FUNCINPEC].” Furthermore, France “was keen to see co-operation between Sihanouk and Hun Sen [and] became increasingly supportive of the SOC. It [was] the first country to give official aid to the SOC and to send its highest-ranking foreign ministry official to Phnom Penh before the signing of the Peace Agreement,” See M. H. Lao, “Obstacles to Peace in Cambodia,” Pacific Review 6, no. 4 (1993): 389. 75. Stephen Solarz, a prominent U.S. congressman active in the Cambodian peace process, sought to avoid a choice between the SOC and the Khmer Rouge, but it should be noted that he argued that the return of Pol Pot to power was to be prevented. See Stephen Solarz, “Cambodia and the International Community,” Foreign Affairs 69, no. 2 (spring 1990): 99–115. Some U.S. interest groups and individuals also sought to pressure Washington to accommodate Vietnam and to pressure China to end its support to the Khmer Rouge. See John McAuliff and Mary Byrne McDonnell, “Ending the Cambodian Stalemate,” World Policy Journal 7, no. 1 (winter 1989–1990): 71–105. On the day of the signing of the agreement, James Baker, U.S. secretary of state, asserted that his government would support the future Cambodian government in any efforts to bring Khmer Rouge leaders to justice. This implied strongly that the Khmer Rouge was expected to lose in the elections and would be subject to punishment for its crimes against humanity. 76. Trevor Findlay, Cambodia: The Legacy and Lessons of UNTAC, SIPRI Research Report no. 9 (New York: Oxford University Press, 1995), pp. 157–158. 77. See, for example, Mats Berdal and Michael Leifer, “Cambodia,” in James Mayall, ed., The New Interventionism, 1991–1994: United Nations Experience in Cambodia, Former Yugoslavia, and Somalia (Cambridge: Cambridge University Press, 1996), p. 57; and Michael W. Doyle, UN Peacekeeping in Cambodia: UNTAC’s Civil Mandate (Boulder, Colo.: Lynne Rienner, 1995), p. 66. 78. Doyle, UN Peacekeeping in Cambodia, p. 67. 79. Ibid., p. 68. 80. Mark Plunkett, “The Establishment of the Rule of Law in Post-Conflict Peacekeeping,” in Smith, International Peacekeeping, p. 72. 81. Heininger, Peacekeeping in Transition, p. 137. 82. Stedman, “Spoiler Problems in Peace Processes,” p. 31. 83. Ibid. 84. Doyle, UN Peacekeeping in Cambodia, p. 68. 85. Stedman, “Spoiler Problems in Peace Processes,” p. 36. 86. Gareth Evans, “The Comprehensive Political Settlement to the Cambodian Conflict: An Exercise in Cooperating for Peace,” in Smith, International Peacekeeping, p. 12. 87. See, for instance, Laurie Barber, “Restoring Cambodia: Democracy by Injection,” New Zealand International Review 20, no. 5 (September–October 1995): 2–6. 88. Frederick Brown and Laura McGrew, “Cambodia: The Royal Government on Trial,” Southeast Asian Affairs (1995): 127–146. 89. Pierre Lizée, “Cambodia in 1995: From Hope to Despair,” Asian Survey 36, no. 1 (January 1996): 83–88. 90. See Sorpong Peou, Foreign Intervention and Change in Cambodia: Towards Democracy? (New York: St. Martin’s Press, 2000). 91. Terence Duffy believes that UNTAC had a profound impact on Cambodia and that the prospects for increasing respect for human rights are good in the future. See Duffy, “Cambodia Since the Election: Peace, Democracy, and Human Rights,”

S-17

11/24/09

530

4:25 PM

Page 530

CASE STUDIES

Contemporary Southeast Asia 15, no. 4 (March 1994): 407–432. There are groups of specialized UN agencies and other local, regional, and international human rights organizations in Cambodia. See Duffy, “Toward a Culture of Human Rights in Cambodia,” Human Rights Quarterly 16, no. 1 (February 1994): 82–104. 92. Sorpong Peou, “Cambodia in 1998: Back to Square One?” Asian Survey 38, no. 1 (January 1998): 69–74. 93. Frieson, “The Cambodian Elections of 1993,” p. 225.

S-18

11/24/09

4:26 PM

Page 531

18 From Missed Opportunities to Overcompensation: Implementing the Dayton Agreement on Bosnia ELIZABETH M. COUSENS

By mid-2000, the enormous international effort to implement the agreement that ended Bosnia’s civil war in November 1995 had gone on longer than the war itself. Two basic concerns animated international activities in 1995: first, that war would not resume, and second, that Bosnia would rebuild for itself a just peace, which international observers by and large considered a multiethnic one. Five years later, neither concern had been conclusively resolved. Massive hostilities had not resumed, but conflict over more targeted objectives remained a sufficient worry that international peacekeepers show no inclination to leave. More troubling, the parties to Bosnia’s peace resisted common political institutions, leaving most of Bosnia’s population under the governance of monoethnic authorities and the country’s unity unrealized. This chapter examines the first four years of peace implementation and asks whether, and in what ways, international efforts contributed to the continuing irresolution of the Bosnian conflict. These years were not without significant accomplishments: several rounds of internationally certified elections were held at national, subnational, and local levels; the powersharing institutions designed to reunify the country were running, if with debatable effectiveness; nearly 650,000 of Bosnia’s forcibly displaced citizens had returned to the country, if not primarily to their original homes; significant portions of the country’s infrastructure were repaired; and not least, the military cease-fire that took hold at the end of 1995 was still intact. Each of these achievements had a subversive element, however. Early elections, for example, militated against broader democratization; in the absence of civilian security, the cease-fire did as much to deepen Bosnia’s internal divisions as to heal them. Moreover, much of the Dayton 531

S-18

11/24/09

532

4:26 PM

Page 532

CASE STUDIES

Agreement remained to be realized even after massive and continuing expenditure of international personnel, resources, and energy. 1 Indeed, international implementers faced iterated obstruction from Bosnian authorities, so much that even as they drew down military presence, they ramped up civilian political involvement. Several villages even acquired special “envoys” just to implement local election results. To explain persistent divisions in Bosnian politics, one need not look far for competing explanations. To list just a few: a centuries-old history of interethnic antagonism and bloodshed; the “artificiality” of the Yugoslav state and the Bosnian Republic within it; an intensive recent war aimed at segregating populations; political leaderships whose commitment to an ongoing peace process was questionable; and a peace agreement whose compromises—some would say contradictions—may have made it unimplementable from the start. Each of such factors played some role in Bosnia’s postwar developments. In this sense, Bosnia’s condition in 2000 was overdetermined: too many explanations chasing a regrettably limited range of outcomes. Within this matrix of contributing factors, an important share of the explanation for limited success in implementing the Dayton Agreement belonged to strategic tensions in international implementation, both at the level of policy priorities set in capitals and agency headquarters and in operational judgments made by implementers on the ground. These were easily exploited by local parties, and failure to resolve them helped consolidate the tripartite division of Bosnia. In this chapter I argue the following: First, the Bosnian war ended with a deeply dissatisfying compromise among the warring parties to which they had only acceded under intense international pressure, especially from the United States. They could be predicted to try to turn the agreement to their respective aims during implementation or otherwise use implementation as an opportunity to revise or obstruct the settlement to which they had pledged themselves. An obvious implication was that international implementers would need a clear strategy for dealing with obstruction and an adequate set of tools with which to pursue such a strategy. Second, the settlement reached at Dayton was sufficiently ambivalent between its separatist and integrationist components that it placed an enormous burden on implementers to decide their relative weight. This ambivalence was largely expressed in the tension among different provisions of the accord. Such tensions were not insuperable but could only be overcome by a deliberate effort among multiple implementers to render their collective efforts coherent. Third, international implementers actually enjoyed great potential influence in this regard. The agreement formally mandated to international third parties an extraordinary level of involvement, and with it, leverage and authority in postwar Bosnian politics. Implementers had many opportu-

S-18

11/24/09

4:26 PM

Page 533

THE DAYTON AGREEMENT ON BOSNIA

533

nities to pressure the parties to overcome weaknesses in the agreement, provided that they recognized the need to do so and could overcome their own institutional constraints. Fourth, however, more opportunities were missed than seized by international implementers during this period. Early policy choices—such as the decision to decentralize implementation among multiple international organizations (especially in the absence of a robust mechanism for coordination), the unwillingness to use military resources in support of civilian implementation, and the sequencing of key provisions once implementation began—worked against the Dayton Agreement’s integrationist goals, even while the latter were publicly championed. Implementation was also weakened by competing strategies among key implementing actors and contributing governments, these driven as much by bureaucratic and domestic considerations as by debate over the best way to consolidate peace. Finally, the principal result of such shortcomings was a growing gap between accomplishment and aspiration that generated great pressure on implementers to overcompensate. The midterm outcome was the adoption of a “trusteeship” strategy in which international implementers arrogated to themselves increasing authority to make binding decisions in Bosnian politics where the indigenous peace process fell short. To argue that implementation mattered is not to underestimate the weaknesses of the peace agreement or its impact on shaping outcomes. Far from it. If negotiated settlements were placed along a spectrum of “implementability,” the Dayton plan would rest at the more difficult end. But the Dayton Agreement was neither all nor most of what mattered. Rather, what better explains frail progress in implementation and the manifestly flawed peace agreement with which it has had to contend are political and organizational imperatives of key member governments and agencies. When these coalesced around a sensible strategy, international engagement sped the Bosnian war to conclusion in the form of the Dayton Accord; to keep the accord from unraveling, to compensate for its many limitations, and to settle the many issues left unresolved by it, an analogous combination of common interest and coherent strategy was required.

War and Settlement Bosnia-Herzegovina found itself fully at war in the spring of 1992 after a year in which the Yugoslav federal state to which it belonged had come sequentially and violently apart. The Bosnian war was the third and most destructive stage of Yugoslavia’s disintegration, following a short scuffle over Slovenia’s secession and a serious, though also short, war over Croatia’s secession. Both the dynamics of Yugoslavia’s breakup and the circumstances that brought the war to an end had specific implications for

S-18

11/24/09

4:26 PM

Page 534

534

CASE STUDIES

what would be required to implement the peace agreement that had just been reached. The Onset of War From at least the late 1980s, Yugoslavia’s complex balance of powers among its six republics, two autonomous provinces, and six constituent nationalities had become increasingly unable to function as an effective state as well as an increasingly fertile ground for competition among political leaders emerging in Yugoslavia’s republics.2 In what was at the time perceived as a progressive response to demands for local autonomy, Yugoslavia’s third constitution (in 1974) shifted the political center of gravity from the country’s federal institutions to its republics and provinces.3 As Yugoslavia lost its economic and strategic status with the end of the Cold War, the Yugoslav state’s capacity to deliver basic goods shriveled,4 creating fertile ground for nationalist complaints and ambitions. By the end of the 1980s, conditions in Yugoslavia put it at high risk for ethnic conflict: declining overall and relative socioeconomic standards, weakening state institutions, social uncertainty and anxiety about “plausible futures,” along with such longer-term attributes as a history of intergroup violence and affinities with groups on other sides of political or administrative boundaries.5 By 1990, when the first democratic elections were held in all six republics, nationalist politicians and parties were clear winners. Serbian president Slobodan Milosevic emerged as a champion of the Serb people, whose rights he encouraged them to believe were under assault everywhere from the province of Kosovo—which was 90 percent Albanian Muslim to 10 percent Serb—to the republic of Croatia, whose 1990 draft constitution appeared to grant rights only to Croats and not to the 12 percent of its population that was Serb. Slovenia openly declared its desire for national independence, directly challenging Milosevic, who expressed a fierce commitment to keep Yugoslavia together. Croatia, meanwhile, experienced its own nationalist revival, electing the well-credentialed nationalist Franjo Tudjman as its first president of the postcommunist era, and its new ruling party—the Croatian Democratic Union (HDZ)—replacing the emblems of Yugoslavia with Croatian symbols last seen during Croatia’s collaboration with Nazi Germany. At the same time, hard-line Croatian Serbs had begun a militant autonomy movement, whose core became the new Serbian Democratic Party (SDS). Bosnian politics also turned in nationalist directions. The Muslim-dominated Party of Democratic Action (SDA) was founded in the spring, with Alija Izetbegovic as its president; and sister parties of the Croat HDZ and the Serb SDS were founded shortly thereafter. Still more ominous, segments of the political population in Croatia and

S-18

11/24/09

4:26 PM

Page 535

THE DAYTON AGREEMENT ON BOSNIA

535

Bosnia had already begun to prepare for armed confrontation, both at the boundaries of the republics and within them. The immediate sequence of events that culminated in full-scale war in Bosnia was swift and brutal. Slovenia declared its independence on June 25, 1991, which it won after an extremely short “war” with the Yugoslav National Army (JNA)—more of a halfhearted police action conducted by military troops. Croatia simultaneously declared independence but met far more serious resistance from the JNA, Serb paramilitary units, and its own autonomy-seeking Serbs. As war in Croatia continued through late 1991, Bosnia’s Serb and, to a lesser extent, Croat communities also began to mobilize for conflict. When the European Community recognized Slovenian and Croatian independence in January 1992, largely as a gambit to end the war in Croatia, it worked. Within two months, however, war had engulfed neighboring Bosnia. Once Slovenia and Croatia had seceded, Bosnia was left with a Hobson’s choice: remain in a much smaller Yugoslavia that would be overwhelmingly dominated by Serbia and, by implication, its own large Serb minority; or leave the Yugoslav Federation, leaving Bosnian Serbs, and Croats to a much lesser extent, analogously worried about domination by the country’s Muslim plurality. Anxiety about the status of non-Muslim nationalities was not without basis, but it was also extravagantly stoked and manipulated by Serbian and Croatian leaders. Among Yugoslav republics, Bosnia uniquely resembled the Yugoslav Federation in its mix of nationalities, lack of an absolute ethnic majority, and intricate power-sharing formulas for managing ethnicity. Whereas the Slovenian population was close to 90 percent Slovene, and Croatia 78 percent Croat, the Bosnian population was 44 percent Muslim, 31 percent Serb, 17 percent Croat, and 8 percent denominated “Yugoslav and Other.”6 Within the framework of Yugoslavia, Bosnia had worked out a delicate balance of political power among these constituent nationalities. Once the Bosnian republic became an independent state, however, these constitutionally guaranteed relationships would be called into question: Croat, Serb, and Muslim status as constituent nations of the Federal Republic of Yugoslavia would not automatically guarantee them equal status as constituent nations of Bosnia. Uncertainty about national status in an independent Bosnia was only intensifed by Izetbegovic’s announcement of SDA opposition to national power sharing in favor of “one man, one vote”—a change that appeared to favor the larger Muslim population.7 In any event, Bosnia was recognized as an independent state by the European Community on April 6, 1992, one month after barricades were first raised in Sarajevo city streets, two days after President Izetbegovic ordered a general mobilization of Bosnia’s territorial defense forces, and one day after Serb paramilitary forces besieged Sarajevo’s police academy

S-18

11/24/09

536

4:26 PM

Page 536

CASE STUDIES

and the JNA seized its airport. In short, recognition occurred just as the country was being plunged into war. During the next three and a half years, Bosnian government forces fought to preserve an independent, unitary state that would enjoy the same borders as the former Bosnian republic and, ostensibly, offer the same rights to its Serb and Croat citizens that they had enjoyed when Bosnia was a constituent republic of Yugoslavia. The government found itself at war on at least two fronts. From the beginning, and principally, it battled radical Bosnian Serbs, closely tied to and actively supplied by Serbia, who fought to “cleanse” large portions of Bosnian territory of non-Serbs and declare an independent Serbian republic that might eventually join Serbia proper. Second, it faced Bosnian Croats, closely tied and actively supplied by their patron Croatia, who launched their own ethnically driven landgrab in central and southern Bosnia in April 1993. International Peacemaking Efforts From 1991, the international community became intimately involved in efforts to settle the cascading conflict in former Yugoslavia. Prior to mid1995, however, the international community seemed as internally divided as the Bosnians, and none of the peace plans developed by different international mediators was sufficiently backed by major powers or sufficiently acceptable to the parties to become the basis for a comprehensive settlement.8 Europeans saw the Yugoslav crisis as a test case of its capacity for common foreign and security policy—which proved possible, though only at a very low denominator of commonality. The United States, meanwhile, bemoaned European ineffectiveness but was unwilling to step into the breach. For their part, international organizations and agencies also split over such issues as the question of impartiality toward the parties, the analysis of the war’s origins, the respective culpability of the parties, and the proper role of UN peacekeeping. International actors were further divided between the diplomatic goal of a negotiated settlement and the human rights objective of a just one. This last was more than just a philosophical dispute; it was also a crudely operational one, with peacekeepers and diplomats on one side and human rights actors on the other.9 In the lead-up to Dayton, by contrast, international engagement in the conflict acquired an increasing coherence that, in turn, helped create and reinforce a changing balance of military power among the parties on the ground. These factors together made an agreement possible where it had earlier been elusive. A first element of this emerging coherence was the changing role of the United States, which began to recognize that it had an interest in active involvement to end the conflict. For the first time, the United States developed a serious strategy to bring about an end to the war, based on the assumption that Bosnia would contain both a Bosniac-Croat Federation and a quasi-independent Republika Srpska (RS), and that both

S-18

11/24/09

4:26 PM

Page 537

THE DAYTON AGREEMENT ON BOSNIA

537

entities would be allowed to establish “special” relationships with neighboring Croatia and Serbia, respectively. A second, arguably more critical element was a lurch into operational coherence among the various components of international Bosnia policy. International mediation, military, and humanitarian efforts began to function more productively to support a common strategy to end the war. In large part, this coherence was forced upon the international community by the Bosnian Serb Army. Previous calls for more muscular diplomacy in the Balkans had been persistently undermined by the vulnerability to reprisals of UN peacekeepers and other international personnel on the ground—primarily those deployed in six UN-designated “Safe Areas.”10 That a way out of this bind had to be found became shamefully evident when Serb forces took hundreds of UN peacekeepers hostage following air strikes by the North Atlantic Treaty Organization (NATO) in May 1995.11 In the brutal event, this dilemma between the vulnerability of UN peacekeepers and the need for forceful diplomacy was eased when the enclaves of Srebrenica and Zepa fell to the Bosnian Serbs in July, removing, along with the thousands of civilians expelled or killed, two of the least tenable deployments of UN peacekeepers in the country. Authorized by the UN Security Council, a military Rapid Reaction Force deployed around Sarajevo by the end of July. A third element smoothing the path to Dayton was a new international willingness to use force, especially air power, as a partner to diplomacy. By the end of summer 1995, obstacles to the use of NATO air power had been removed with the repositioning of UN Protection Force (UNPROFOR) troops, both forced, in the case of Srebrenica and Zepa, and voluntary, in the case of Gorazde. Bosnian Serb forces also provided NATO with a rationale for air strikes, with a mortar attack on August 28 that killed thirtyseven in Sarajevo’s marketplace. Finally, the most powerful pressure to reach settlement came from the dramatically changing military balance on the ground between the spring and fall of 1995, itself in no small way a product of U.S. efforts, which brought the territorial holdings of the warring parties into remarkably close alignment with the proposed basis for negotiation.12 Starting in May, Serb forces suffered a series of defeats at the hands of the Croatian and Bosnian armies that significantly changed their calculus at the bargaining table and that also represented a new level of U.S. commitment to the Croatian military, which at this stage was being openly trained by unofficial U.S. advisers. On May 1, the Croatian Army retook Serb-occupied territory in Western Slavonia, defying resident UN forces in this “UN Protected Area.”13 On August 4, it launched Operation Storm, an offensive that toppled the self-proclaimed “Krajina Serb Republic” in just two days, sending close to 150,000 Serb civilians fleeing into Serb-held Bosnia and Serbia. Through the early fall, the Bosnian Army advanced dramatically on Serb positions. As negotiator David Owen of the European Union (EU)

S-18

11/24/09

538

4:26 PM

Page 538

CASE STUDIES

described the situation from September: “Thereafter, day by day, the map altered.”14 Alongside NATO’s air campaign, these Bosniac and Croat military gains rapidly established the conditions necessary for a settlement, including on the status of remaining Serb holdings in Croatia. By early September, talks in Geneva led to an agreement that the basis for settlement was a political subdivision of Bosnia into two coequal “Entities,” and a territorial split between 51 percent under Bosniac-Croat control and 49 percent under Bosnian Serb control.15 By November 20, three weeks of proximity talks at Wright-Patterson Air Force Base in Dayton, Ohio, produced a comprehensive agreement among all three warring parties and their regional guarantors. The resulting General Framework Agreement for Peace (GFAP) in Bosnia-Herzegovina was formally signed in Paris on December 14. Setting the Stage for Implementation Several features of the war and its conclusion bore heavily on the prospects for peace implementation. First, the rapid escalation of conflict had been due partly to the sheer factor of uncertainty about the boundaries of the political community within which basic rights would be guaranteed and the preemptive mobilization of ethnic groups that ensued. To the extent that such questions remained unresolved by the peace agreement, one could expect similar violence to continue. Second, the war occurred at the earliest stages of Yugoslavia’s democratization, after a first round of multiparty elections, but before most other ligaments of a democratic polity could be established or strengthened. The nature of the warring parties was also semiauthoritarian, which remained the case in postwar Bosnia. Third, in prosecuting their respective war efforts, the movement of civilian populations was an important instrument, as was the use of police and extramilitary groups, and both would come to have their peacetime equivalents. Fourth, coercion played a critical role in ending the war. This coercion was both external, in the form of NATO air strikes and diplomatic pressure, and internal, in the form of the Bosniac-Croat ground offensive. Fifth, coercive instruments were only effective after greater operational coherence was achieved among international actors in the field. Finally, international intervention was decisive, both in helping to produce a new military balance of forces on the ground and in pressuring parties to sign an agreement. As a result, and arguably most important, the Bosnian war did not end with a “mutually hurting stalemate” but with what is better called a “coerced compromise.” The central implication for peace implementation was clear: brought to the table by varying forms and degrees of coercion, the parties had little more than a tactical commitment to settle, making any resulting accord dependent on more than the will of the parties for its

S-18

11/24/09

4:26 PM

Page 539

THE DAYTON AGREEMENT ON BOSNIA

539

implementation. To end the war, international third parties deployed the considerable forms of leverage they had always held in potential; to implement the peace, they needed to do the same, at least until conditions could be created that would exert an analogous effect on the parties and render the peace self-sustaining.

The Dayton Framework The Dayton Agreement essentially had two goals: to end the fighting and to rebuild a viable Bosnian state. To accomplish the former, the agreement detailed an elaborate calendar of commitments for separating and drawing down the armed forces of the Bosnian Serbs, on the one side, and the Bosniac-Croat Federation, on the other. To accomplish the latter, the agreement included a postwar constitution and a wide range of provisions to deal with such varied aspects as refugee repatriation, elections and democratization, human rights, and reunification. Role of International Implementers Toward its twin grand objectives, Dayton committed the international community to an ambitious, intricate, and highly decentralized set of implementation roles (see Table 18.1). Compliance with the accord’s military provisions (Annexes 1A, 1B, 2) were to be supervised by a multinational, NATO-led Implementation Force (IFOR). Authorized under Chapter VII of the UN Charter, IFOR was commanded by the North Atlantic Council (NAC), 60,000 strong at first deployment and supposed to complete its work by December 1996.16 IFOR’s function was to separate armed forces,

Table 18.1

The Dayton Agreement and Its Implementers

Annexes

Key International Implementers

1A, Military Aspects 1B, Regional Stabilization 2, Inter-Entity Boundary Line (IEBL) and Related Issues 3, Elections 4, Constitution

NATO-led IFOR OSCE International Arbitrator OSCE European Court for Human Rights International Monetary Fund

5, Arbitration 6, Human Rights 7, Refugees and Displaced Persons 8, Commission to Preserve National Monuments 9, Bosnia-Herzegovina Public Corporations 10, Civilian Implementation 11, International Police Task Force

OSCE, Council of Europe UNHCR UNESCO EBRD International High Representative UN

S-18

11/24/09

540

4:26 PM

Page 540

CASE STUDIES

oversee the cantonment of troops and heavy weapons to agreed-upon areas, and stabilize the cease-fire.17 IFOR differed from traditional peacekeepers in its capabilities and its doctrine, being prepared to deter military violations of the agreement. Importantly, IFOR’s secondary responsibilities ran the gamut of implementation activities. In overseeing compliance with the full range of commitments made under Annex 1A, IFOR’s responsibilities were linked to the parties’ promise to ensure the safety of all civilians under their respective jurisdictions, to provide humane and nondiscriminatory law enforcement, and to cooperate with the international criminal proceedings at the Hague.18 IFOR was also asked to support other components of implementation, with specific reference to the return-related responsibilities of the UN High Commissioner for Refugees (UNHCR). Moreover, IFOR was expressly enjoined “to observe and prevent interference with the movement of civilian populations, refugees, and displaced persons, and to respond appropriately to deliberate violence to life and person.” This particular responsibility carried enormous significance if fully embraced, especially given IFOR’s authority to augment its role “without interference or permission of any Party” on the sole judgment of the IFOR commander (ComIFOR) and the NAC. ComIFOR was also given “final authority in theatre” to interpret all military aspects of the settlement. The tasks of civilian implementation, by contrast, were parceled out annex by annex to numerous agencies, though some tasks, like human rights, had no single steward, and some agencies—like the Organization for Security and Cooperation in Europe (OSCE)—had multiple responsibilities (see Table 18.1). The OSCE would oversee preparation and conduct of elections. The UNHCR would handle return of refugees and internally displaced persons (IDPs). The UN peacekeeping operation that had been in Bosnia since 1992 would be transformed into the backbone of an International Police Task Force (IPTF) to monitor and help reform Bosnia’s police. Oversight of human rights provisions would be collectively undertaken by the OSCE, the Council of Europe, the UN High Commission on Human Rights, and the European Court of Human Rights, while provisions for missing persons would fall to the International Committee of the Red Cross (ICRC). The World Bank would lead on postwar reconstruction, a role not articulated in the text of the agreement but widely acknowledged. Within the Federation, the EU would oversee the knitting back together of the divided city of Mostar, a responsibility that it had assumed before Dayton. International actors also became interim members of a wide array of Bosnian institutions and bodies and usually with decisionmaking authority. These included the Constitutional Court, the Central Bank, human rights bodies, a Provisional Election Commission, and other commissions (Real

S-18

11/24/09

4:26 PM

Page 541

THE DAYTON AGREEMENT ON BOSNIA

541

Property Claims, Public Corporations, Independent Media, National Monuments; see Table 18.2). To coordinate this panoply of civilian roles and responsibilities, an international Office of the High Representative (OHR) was established. The high representative was to enjoy final interpretive “authority in theatre” of Dayton’s civilian provisions, a textually coequal role to that of the ComIFOR, though one that was unlikely to enjoy equivalent real authority. All civilian agencies reported separately to their respective governing bodies, not to the OHR, and none had the incentive to put their operational resources under the high representative’s direction. More strategic oversight was meant to come from a Peace Implementation Council (PIC) composed of over fifty implementation-friendly governments, to whom the high representative would report.19 The PIC’s principal body was a Steering Board that aimed to balance between including as many key governments as possible, especially major donors, and remaining small enough to be agile and decisive. 20 The five-nation Contact Group also provided an opportunity to harmonize the diplomatic efforts of its members and exert appropriate leverage on the parties, though its role declined relative to the PIC after 1996, largely because it did not include major donors.21 Constraints upon Implementation From the beginning, implementers had every reason to expect an uphill climb. Four basic challenges faced them. First, it was no secret that the parties to Dayton were ambivalent about signing and could be expected to try to obstruct implementation, at least tactically. The Bosnian Serbs got a demographically sweet deal (49 percent of Bosnian territory when they represented only 31 percent of the prewar population) on the one hand, but a territorially disappointing one on the other. Their entity was effectively divided into two halves at Brcko;22 Sarajevo was highly unlikely to be open to them in any serious way; and as a result of the Bosniac-Croat offensive in late 1995, their population had been swollen by approximately 200,000 refugees. Bosnian Croats did comparatively well, sharing power and land within the Federation in significant disproportion to their numbers (17 percent of the prewar population), but their leadership was brazen in its lack of interest in real integration with the Bosniacs. Their hard-line constituency, centered in southwestern Bosnia, was publicly committed to the Croat “parastate” established during the war—Herzeg-Bosna—which, they argued, deserved no less autonomy as an effective third “entity” than the RS. The Croat position was demographically complicated by the existence of Croat communities sprinkled throughout central Bosnia—often in towns that the war had left ethnically divided—but they also had a distinct advantage over

S-18

11/24/09

4:26 PM

Page 542

Table 18.2 International Membership in Bosnian Offices Office

International

Authority/Term

State

Federation

RS

Provisional Election 4, including the Commission (PEC) OSCE Head of (out of a total of 7, Mission as Chair later expanded to 16)

Chair’s decisions are binding

1 (then 2, then 4)

1 (then 2, then 4)

1 (then 2, then 4)

Constitutional Court (out of a total of 9)

3, Appointed by the European Court of Human Rights

5 years; thereafter, Parliamentary Assembly can amend selection process

4

2

Governor of the Central Bank

Appointed by the IMF

6 years

Human Rights Ombudsman

Appointed by the OSCE Chairmanin-Office

5 years, not renewable; thereafter, BiH President

Human Rights Chamber (out of a total of 14)

8, appointed by the 5 years, renewable Council of Europe Committee of Ministers, including President

4

2

Electoral Appeals Sub-Commission (EASC) (out of a total of 4)

1

1

1

1

Joint Interim Commission

OHR representative chairs

4

3

Commission on Real 3, appointed by the 5 years, renewable Property Claims European Court of (out of a total of 9) Human Rights, including Chair

4 (3- and 4-yr. terms)

2 (3- and 4-yr. terms)

Commission on Public Corporations (out of a total of 5)

5 years, renewable

2

1

5 years, renewable

2

1

3 on each of Council and Panel

1

2, appointed by EBRD, including Chair

Commission on 2, appointed by National Monuments UNESCO, (out of a total of 5) including Chair Independent Media 3 each, plus the Commission (IMC) IMC Director(out of a total of 7 on General its Council and Enforcement Panel)

S-18

11/24/09

4:26 PM

Page 543

THE DAYTON AGREEMENT ON BOSNIA

543

the Serbs, sharing a long, porous border with Croatia, which backed them unambiguously. The Bosniacs were, in turn, deeply dissatisfied by the peace, though desperate to see the war end. At Dayton, they were asked to cede almost half of the country and on behalf of a decentralized state that many doubted would survive. They also generally believed—as did Bosnian Croats—that their forces had been on the verge of victory over the Serbs when the United States vigorously appealed to them to stop short of retaking western RS. The Bosniac leadership accepted the Dayton deal with misgivings and, according to most accounts, only after promises were made to heavily invest in Bosnia’s postwar reconstruction and to equip and fully train Bosnia’s still young army. Although the Bosniacs maintained that their preference remained a united and multiethnic Bosnia, the war had strengthened the position of hard-liners within the SDA, which led some to doubt the authenticity of their commitment to a unitary state. Indeed on all sides, the unsurprising effect of the war was to fortify nationalist over moderate voices. Second, that the peace agreement was deeply problematic was also no secret. Its calendar of obligations was dramatically unbalanced, between a highly detailed military schedule and what can only be called a minimalist political and civilian one. Indeed, Dayton’s only firm deadline on the civilian side was that national elections be held within nine months, a timetable that many argued was unrealistic and even counterproductive. Other than deadlines, the settlement gave no direction on the relative importance of its many and varied provisions, no mechanism to set priorities among them, and no anticipation of likely consequences should those priorities remain unestablished or unwisely set. The accord also depended upon a continuing level of international commitment that was not only unlikely to be sustained but whose evanescence was actually written into the text of the accord. Perhaps most problematic, Dayton effectively offered no clear, single political outcome for the country. In this respect, it should more accurately be considered an “interim” than a final agreement, more akin to the Oslo Accords in the Middle East, where final status issues remained unresolved, than something like the Chapultepec settlement in El Salvador. Despite its 130 pages of detailed text, Dayton’s mediators had crafted an accord based on a fundamental ambivalence between its partitionist and its integrative elements. On the one hand, Dayton stabilized lines of confrontation and derived political rights from them; on the other, it aspired to override such divisions both from above—in its joint institutions—and from below, in its provisions for return. It also essentially neglected the question of intra-Federation conflict, which had presumably been resolved from 1994 but which many anticipated posing significant obstacles to peace implementation.

S-18

11/24/09

544

4:26 PM

Page 544

CASE STUDIES

A third set of challenges came from the nature of actors who engaged in third-party implementation: international organizations and agencies, donor governments and ministries, and NGOs. Each of these, if in different ways, typically experience operational kinks that will impact upon a peace implementation process. These ranged from start-up delays and poorly qualified staff, to coordination and information problems, to resource scarcity. Such traits were widely enough known that they could be explicitly factored into plans for implementation. More problematic was the abiding tendency among such actors to compete for what might crudely be termed implementation “market share.” Several factors encouraged such competition: difference among mandates, even to the point of contradiction; different methods of financing, forms of governance, and lines of accountability; distinct organizational cultures; divergent conceptions of what constituted legitimate forms of international engagement; and generic rivalry over roles and turf. In Bosnia—as elsewhere—lead roles were allocated and defined primarily on the basis of institutional and national bargaining rather than comparative advantage.23 A few examples: The EU was asked to “administer” the divided city of Mostar, more because European governments were willing to foot a large bill than because there was reason to think that they could advance the peace process between Bosniacs and Croats. The OSCE, instead of the UN, was asked to supervise Bosnia’s elections, even though the former had no experience in doing so, largely because key member governments saw the UN as discredited by its peacekeeping performance during the war. For similar reasons, Dayton’s crafters never seriously considered naming a UN special representative of the Secretary-General (SRSG) instead of a high representative, despite the logic of not starting such an office from scratch. Meanwhile, it has been understood that the high representative would always be European; at least two of his or her chief deputies would always be German and American, respectively; and the OSCE head of mission would always be American. Such patterns are not unique to peace implementation—witness the allocation of senior positions within international organizations. Moreover, their consequences are not always negative if the placement of key staff reflects the political commitments of key governments. Whether the effects are positive or negative, however, turns on whether implementers can establish policy coherence at the strategic level and operational coordination in the field, with capacity for real-time evaluation and course correction as needed. Unfortunately for Bosnia and its implementers, mediators opted for a model of implementation that was divided (between military and civilian/political) and decentralized (among civilian/political provisions), with all coordination among international organizations being essentially voluntary. Particularly given the recalcitrance of Bosnia’s parties, this choice was risky at best, negligent at worst.

S-18

11/24/09

4:26 PM

Page 545

THE DAYTON AGREEMENT ON BOSNIA

545

A final issue centered on domestic imperatives among key governments—particularly the United States—and the question of their stamina to see through a peace process that might require more than one year of engagement. Precisely because of its premature, ambiguous, and coerced quality, the Dayton settlement could be predicted to need sustained international engagement, and most of those working on Bosnia policy recognized that peace would take time as well as troops.24 Getting such a commitment from the United States was of disproportionate importance, especially since European governments, burned by the experience of UNPROFOR, had refused to join any postwar implementation force without U.S. participation. To get congressional support, however, the Clinton administration had to ensure that the Dayton Agreement contained the promise of a quick exit for U.S. troops; thus Dayton’s premise that the bulk of peace implementation would be accomplished by the end of 1996. Whether Dayton was the best agreement that could have been made to end an ugly war is difficult to say. Dayton’s mediators had cobbled together a flawed peace to stop a war whose humanitarian and political consequences had finally become unbearable to international observers. They accomplished what none had yet been able to—a cease-fire—amid daunting challenges. The critical question was whether implementation could compensate for such an unpromising beginning.

Security The first and central task of Dayton implementation lay in the field of security. The war left Bosnia with two broad security challenges: first, to ensure that the parties would not reengage one another’s armed forces; second, to extend the cease-fire to civilians and ensure that they were no longer targets of organized violence. Both security considerations had in some fashion to embrace the full spectrum of potential belligerents, or what Dayton labeled “all personnel and organizations with military capability.” This included reservists, national guards, military police, internal security forces, and armed civilian groups as well as the “foreign forces” that were to be withdrawn within thirty days after the peace accord was signed. To round out the security picture, the parties were obliged to conduct law enforcement “in accordance with internationally recognized standards and with respect for internationally recognized human rights and fundamental freedoms.”25 Peace Between the Parties At the end of the war, the immediate task was to consolidate the cease-fire line between Serb forces and the Bosniac-Croat alliance and thereby stabi-

S-18

11/24/09

546

4:26 PM

Page 546

CASE STUDIES

lize the 49–51 percent territorial allocation of the country between them (Annex 1A); the longer-term goal was to ensure that the parties would have neither the capability nor the incentive to return to war. Dayton most comprehensively addressed the former, with a detailed calendar of obligations on the parties to sequentially draw down their forces under the supervision of IFOR. Longer-term stability was to be ensured through committing the parties to a modest package of regional arms control and confidence-building measures under the supervision of the OSCE (Annex 1B). Importantly, Croatia and Yugoslavia were also bound by arms control provisions. No provisions were made, however, for the possibility of renewed hostilities between Bosniacs and Croats within the Federation. The settlement was also territorially vulnerable at two major points. First, the Bosniac-majority town of Gorazde, which had been a Safe Area during the war, was connected to the Federation by only a narrow spit of land. Gorazde was treated within Annex 1A largely as a matter of territorial demarcation that would be guaranteed by outside implementers. Second, the northern town of Brcko, to which both Serbs and Bosniacs laid primary claim, sat astride the even narrower Posavina land corridor connecting western and eastern RS.26 Because negotiations at Dayton were unable to settle the Brcko question, the dispute was remanded to an international arbitrator for resolution within one year. Outside of the Dayton framework, a curious coalition of the United States and several Islamic governments pledged themselves to a program of military assistance to the Bosniac-Croat Federation in order to render its capabilities comparable to that of the RS. This “Train and Equip” program was meant to create an internal balance of forces that would either deter renewed Serb attacks or enable an effective defense against them if deterrence failed, and in any case the program theoretically allowed for the timely exit of international peacekeepers. The strategy also aimed to supplant with Western assistance the ties that had developed between Iran and the Bosniacs over the course of the war. Finally, the program was to be a vehicle for integrating the Army of Bosnia-Herzegovina (ABiH) and the Croation Defense Force (HVO). At one level, Dayton’s strategy to keep military peace was successful. Under IFOR’s supervision, the cease-fire was consolidated almost to the letter of Dayton’s provisions by the end of 1996. Since then, it has held firm. Federation and Serb military forces were progressively demobilized.27 “Foreign forces”—a polite term for Islamic mujahidin—by and large left the country.28 The Inter-Entity Boundary Line (IEBL) was also quickly stabilized, with a two-kilometer buffer zone on either side known as the Zone of Separation (ZOS) along which international troops heavily deployed, and there have been no postwar incidents of military-on-military violence or uses of force by one entity against the other. At the regional level, the Bosnian parties and their neighbors also reached an arms control

S-18

11/24/09

4:26 PM

Page 547

THE DAYTON AGREEMENT ON BOSNIA

547

agreement within six months that set numerical ceilings on heavy weapons.29 In short, those military provisions to which NATO leadership devoted their resources were efficiently implemented according to the agreed-upon timetable. Less successful were efforts to integrate Bosnian armed forces. As feared initially by many critics, Train and Equip was more successful at equipping Federation forces than at training them, at least as a joint army: common exercises between the ABiH and HVO were essentially superficial, leaving the forces in all significant respects—training, administration, doctrine, command—separate. Each force was tied to its respective national leadership, with doctrine, capabilities, and intentions relatively opaque to each other and to the international community. There was even greater distance between Federation forces and the Republika Srpska Army (VRS), leading NATO and the OHR in mid-2000 to make a vocal appeal for a single, joint army. Even moderate Serbs, however, rejected this appeal.30 After four years of implementation, the tripartite military division of the country remained the primary obstacle to building common institutions for the postwar Bosnian state. Dayton had empowered IFOR and SFOR to decide where armies deployed and from which areas they withdrew, meaning that NATO forces had the power to sever the link between military control of territory and political jurisdiction. Instead, NATO relied early upon a force separation strategy that consolidated this nexus and thereby missed a critical opportunity to advance Dayton’s main political agenda. Peace for the Public Although more concerned with security between the entities than that within them, Dayton did acknowledge security concerns beyond the narrowly military or the strictly interentity. Under the agreement, the parties were obliged to “provide a safe and secure environment for all persons living in their respective jurisdictions,” ensure the “highest level” of effective respect for human rights, establish a “politically neutral environment,” and “recreate as quickly as possible normal conditions of life in Bosnia and Herzegovina.” IFOR, in turn, was authorized “to observe and prevent interference with the movement of civilian populations, refugees and displaced persons and to respond appropriately to deliberate violence to life and person.” The parties also promised to conduct all law enforcement according to “internationally recognized standards and with respect for internationally recognized human rights and fundamental freedoms” (Annex 11), in which they would be assisted by an IPTF.31 Yet in the critical first eighteen months after Dayton, NATO command was reluctant to deal with extramilitary forces that fell ambiguously in between “military” and “civilian.” They initially opted not to inspect any forces other than regular armies, leaving paramilitary groups and internal

S-18

11/24/09

548

4:26 PM

Page 548

CASE STUDIES

security forces with a peculiar freedom of movement. Not until mid-1997 did SFOR show interest in apprehending “persons indicted for war crimes” (PIFWCs) or accept any significant role in providing public security (in 1997 a Multinational Specialized Unit of military police began operation, under SFOR command). Meanwhile, the multilateral, unarmed IPTF was the sole actor to contend with the quasi- and paramilitary Bosnian police, an inappropriate vehicle at best to fill an enforcement gap or reform state security forces. This was all the more so, due to the ambivalence of the IPTF’s mandate, which required it to assist and advise local law enforcement forces while also keeping a watchful eye on them. While the IPTF and the parties performed reasonably in drawing down the number of police, they were far less successful in monitoring human rights abuses or preventing ethnic violence in which local police were either directly or indirectly involved. By the same token, efforts to integrate police forces ethnically or otherwise render them responsive to citizens beyond those of their own ethnic group also produced poor results. Even after the IPTF acquired new investigatory powers in December 1996, the tension in its mandate remained.32 Only two years into the peace did NATO-led troops seriously focus on protection of civilians from political violence, especially in the context of return; apprehension of PIFWCs; or creation of a broader sense of security: secondary tasks that were arguably critical to the political consolidation of peace in the country. The resultant security gap indirectly but powerfully affected other areas of implementation, from refugee and IDP return to the integrity of elections.

Refugees and Internally Displaced Persons Forced migration was the primary instrument of Bosnian Serbs and Bosnian Croats against the prospect of a Muslim Bosnia. When the war ended, in addition to the roughly 250,000 persons estimated dead or missing and the over 200,000 wounded, over half of Bosnia’s 4.3 million citizens had been forcibly displaced, either as refugees in host countries (1.2 million) or as IDPs within Bosnia (1 million).33 Dayton proposed to reverse the effects of this deliberate mass displacement (Annex 7). It stipulated an unqualified right of return for refugees and IDPs to their “home of origin,” setting a benchmark for success that extended beyond existing standards or practices for the rights of the displaced. Dayton obligated the signatories to create conditions necessary for peaceful repatriation and reintegration. This included confidence-building and minority protection measures, including repeal of discriminatory legislation and suppression of hate speech, prohibition of intimidation, and prosecution of any public authority who engaged in it. Moreover, in its human

S-18

11/24/09

4:26 PM

Page 549

THE DAYTON AGREEMENT ON BOSNIA

549

rights provisions (Annex 6), Dayton committed the parties to a comprehensive range of international human rights and humanitarian standards.34 If implemented in full, Dayton’s refugee provisions had dramatic potential to reverse the entire demographic course of the war. For the UNHCR, the principal challenge beyond extraordinary logistical hurdles was that nationalist politicians on all sides remained in power, arguably strengthened and uninterested in a remixing of populations that would dilute their political power. Further complicating the situation were tensions among international actors about how best to promote multiethnicity and the extent to which this goal should be a priority in implementation. Majority Versus Minority Return The question of “minority return” lay at the heart of Dayton’s return provisions. Yet in the immediate aftermath of the war, security concerns and restricted freedom of movement gave little choice to the UNHCR but to repatriate the displaced to areas where they belonged to the ethnic majority. The UNHCR nonetheless took some measures to offset restricted movement among minorities, introducing, for example, an interentity bus service to facilitate safe inspection visits by potential minority returnees.35 Only a year and a half into implementation did freedom of movement significantly increase through such factors as reduction of police harassment, removal of illegal checkpoints, and, more positively, introduction (in 1998) of a common license plate that allowed Bosnians to travel throughout the country without revealing ethnicity by their vehicle’s plate. After nearly two years, of course, many displaced persons had sought durable solutions elsewhere. Increasingly, it became clear that minority return also depended on more deliberate coordination among major elements of the international implementation effort. The UNHCR launched an Open Cities Initiative (OCI) in 1997, for example, that relied upon positive conditionality to reward those municipalities receptive to minority returns with increased donor funds, though the program did not result in significant minority returns given the resources expended.36 The UNHCR and OHR also established a Reconstruction and Return Task Force (RRTF) in January 1997 to develop joint approaches that would avoid the perceived failures of earlier initiatives. The RRTF also aimed to deethnicize the minority return process by prioritizing enforcement of individual rights rather than brokering returns of groups per se.37 Cumulatively, these efforts had some positive impact: for instance, voluntary compliance with property laws increased, improving the prospects for minority returns in aggregate. However, on balance, minority return was far less than anticipated. Only 24,000 Bosniacs and Croats returned to their homes in the RS during the first four years of Dayton implementation. The situation was only marginally better for Serbs; 35,000 managed to return to the Federation during

S-18

11/24/09

550

4:26 PM

Page 550

CASE STUDIES

this period, which is only slightly over half the number who fled postDayton Sarajevo during the hand-over to Federation authorities. Overall, most postwar population movements resulted in consolidating ethnic homogeneity.38 Political, Legal, and Socioeconomic Barriers to Return While security and political considerations were the determining factors for individual decisions about return, legal, economic, and social issues also played a role. Dayton anticipated that establishing legal ownership of residential property would be a central challenge to prospects for return, and called for a Commission for Real Property Claims of Displaced Persons and Refugees (CRPC) to handle competing claims to property alleged to have involuntarily changed ownership or occupancy during the war (Annex 7). The CRPC was arguably one of the most important mechanisms for successful implementation of return provisions and, given the centrality of minority return to a unified Bosnia, for successful implementation overall. Yet it was negligibly resourced for years—compared, for example, to funds for physical reconstruction of housing—and establishing postwar ownership and occupancy rights throughout the country remained a real challenge. Both RS and Federation parliaments initially imposed legislative roadblocks to deter returnees from reclaiming prewar property. In 1998, under intense pressure from the PIC and OHR, they eventually either repealed or amended these property laws to conform to the principles outlined under Dayton while also setting up new municipal structures to handle property claims. Despite such reforms and the fact that both the CRPC and municipal offices increased productivity over time, only 35 percent of 225,000 property claims had been resolved by June 2000. Only 13 percent of claimants had physically regained possession of their homes.39 Lack of employment prospects further dampened return, both majority and minority. Unemployment was high and significantly worse among returnees, especially minorities. Competition for jobs was so fierce that even majority returnees experienced resentment from co-nationals, whose job security was often equally precarious. Finally, the nearly unchecked power of municipal authorities over distribution of a wide range of social benefits and community services also was a major obstacle to return. Bureaucratic roadblocks to return were common, particularly directed against minorities, such as authorities demanding identification forms that were unattainable in order to register for international assistance.40 Access to appropriate education was also limited, with Bosniacs, Croats, and Serbs maintaining three different curricula, strongly discouraging minority enrollment, and resisting efforts such as the OHR’s to develop a more inclusive and integrated system.

S-18

11/24/09

4:26 PM

Page 551

THE DAYTON AGREEMENT ON BOSNIA

551

Overall, Bosnia’s ruling parties were adept at developing instruments to resist the repatriation and return of the displaced, including pressuring members of their own ethnic group from attempting minority return. The instruments available to international agencies to counter such tactics were blunt and therefore frail. Thus, four years after Dayton, less than one-third of Bosnia’s original pool of wartime refugees had returned.41 Close to half of the initial refugee population had, in fact, found durable solutions abroad.42 IDP return produced similar results: by 1999, just under one-third of the total wartime pool, 295,991 persons, had moved back to their homes, a number qualified by continuing internal displacements after the war ended.43 Over time, management of Bosnia’s displaced populations nonetheless improved, especially after 1999. Increasingly, Bosnians appeared to be more interested in returning to normal life, whether multiethnic or not,44 and implementers ramped up practical efforts to enable individual choice. However, much work remained to realize individual rights in practice given the unavoidable interdependence among major arenas of implementation: in particular, security as guaranteed by IFOR, SFOR, and the IPTF; reconstruction as spearheaded by the World Bank and donors; and political reform as sought by the PIC and OHR.

Reunification At the end of the war, Bosnia was de facto divided among three authorities backed by three respective armed forces: the Bosnian Serbs, led by the Serb Democratic Party of Radovan Karadzic; the Bosniacs, led by the Party of Democratic Action of Alija Izetbegovic; and the Bosnian Croats, led by the Croatian Democratic Union, with a clutch of leaders tightly tied to Croatian president Franjo Tudjman. Bosnian Serbs very clearly controlled the RS, but between Bosniacs and Bosnian Croats, there was greater ambiguity: although the two had been ostensibly at peace since 1994, much of central Bosnia was a patchwork of towns and villages held respectively by one or the other—sometimes divided between them—while portions of the country’s southwest were held singularly by Croats. Dayton’s primary mechanisms for unifying the country were institutional and demographic. First, a new constitution incorporated into the Dayton text (Annex 4) detailed how power was to be shared among Bosnia’s three parties and two entities within a single state. Second, this political craftsmanship was to be bolstered from below by Dayton’s provisions to reverse ethnic cleansing. A steady flow of returning refugees was supposed to be well under way by the time of the first elections, which ostensibly would empower nonnationalist parties and increase momentum toward reunification (Annex 3).

S-18

11/24/09

4:26 PM

Page 552

552

CASE STUDIES

Unification from Above Beyond the efforts to rejoin military and police forces, Dayton called for the Bosnian parties to share power in a postwar state through joint political institutions. The Bosnian state would have a three-person presidency composed of a Bosniac and Croat elected by Federation voters and a Serb elected by RS voters; and a bicameral parliament composed of a House of Representatives and a House of Peoples with a ratio of two Federation members to one RS. Significantly, the Bosnian state was given no independent judiciary—though Dayton did provide for a constitutional court— and judicial functions were left to the entities or, in the Federation case, to ten subentity cantons. At the entity level, the RS had no explicit obligation to share power, whereas the Federation was like a mini-Bosnia, explicitly constituted in order to balance the interests and office of the Bosnian Croats and Bosniacs whom it governed. The RS had a president and National Assembly, both directly elected by voters registered in the RS. In the Federation, on the other hand, the president was appointed by the Federation’s two legislative bodies, and the president and the vice president—a Bosniac and a Croat, respectively—would rotate roles on an annual basis. Below the entity level, the Federation consisted of ten cantons whose assemblies were directly elected by Federation voters. Both the Federation and the RS also contained municipalities, whose assemblies were directly elected by voters in each entity, which bodies in turn appointed mayors. The weakest of these levels of government, institutionally, was the single state at the top. Dependent on budgetary transfers from the entities, without an army, police, or judiciary of its own, it had relatively little leverage over lower levels of government. At the entity level, the Federation was also weak vis-à-vis its ten cantons. Cantons enjoyed the power to tax and had exclusive authority over police, internal security, and judicial matters. In some areas, cantons shared with the Federation a coequal authority that they were only enjoined to divide “as appropriate” after consultation, including over immigration and asylum, implementing legislation related to citizenship, social welfare, and communications and transport policy.45 Getting Bosnia’s joint institutions to function even minimally was an ordeal, however, requiring the use of political conditionality and establishment of additional forums for consultation and monitoring of compliance, such as a Joint Military Commission and a Joint Civilian Commission. The sheer complexity of the system and constant rotation of roles made it easy for parties to obstruct the functioning of various offices. Every issue from the timing of meetings to their location, let alone their agenda, became an occasion for resistance. Meetings were boycotted, as in 1996, when the newly elected Serb member of the Bosnian presidency, Momcilo Krajisnik, was absent at his own inauguration ceremony in Sarajevo, or when HDZ

S-18

11/24/09

4:26 PM

Page 553

THE DAYTON AGREEMENT ON BOSNIA

553

members walked out of the inaugural session of the Sarajevo canton assembly.46 Even when legislation has been developed, it was frequently ignored or remained unadopted in the absence of further implementing legislation. Unification from Below As frustration with the slow pace of reunification grew in 1997, greater attention was devoted to reforms that would effectively empower individual Bosnians to begin reintegrating the country from below. A major element of a bottom-up approach was the strategy developed to enable “minority” return. More intriguing were the series of directives issued by the PIC in May 1997 that became known as “Sintra” issues, for the town where the PIC met. These concerned symbolic items (the design of a common flag, composition of a common anthem, and appointment of international ambassadors) but also powerfully concrete measures (agreement on common currency, common citizenship laws and passport, a joint telephone system, and uniform car registration). In December 1997, implementers went even further at Bonn, strengthening the authority granted to the high representative by the PIC. Textually, the high representative had always enjoyed final authority over interpretation of Dayton’s civilian provisions. Yet PIC governments had never before indicated their willingness to see the high representative exercise it robustly. With these “Bonn powers,” the high representative was now given both creative authority to develop and enact laws otherwise blocked by the Bosnian leadership, and enforcement powers to take action against any public party, including members of the media, who failed to abide by the terms of Dayton implementation.47 Since then, the OHR, as well as other agencies such as the OSCE, have resorted to Bonn powers with regularity to respond to perceived noncompliance by the parties to Dayton, whether to enact integrationist legislation, issue binding directives, or dismiss officials. Rejoining Bosnia from either the top or the bottom has been an arduous and only partly successful effort. Formally, joint political institutions eventually began to function as originally planned, with some concrete accomplishments. More enduring progress, however, has been elusive. The top-down power-sharing institutions operate, even in buildings flying a common flag, but they rarely make decisions of significance. Progress on the Sintra issues, meanwhile, occurred comparatively late in the game to affect freedom of movement at the critical early stage. Moreover, achievements on both fronts have been possible only by being imposed through the authority of the high representative and other key implementers.48 As late as 2000, this pattern showed no sign of diminishing: for instance, Bosnia’s new draft election law was only rendered operative for the fall 2000 elections by OHR action.49 The roles played by the OHR and the OSCE, particularly when combined with the wide range of international involvement in

S-18

11/24/09

4:26 PM

Page 554

554

CASE STUDIES

other arenas of the Bosnian polity and economy, make for a heavy footprint of international engagement in the country. Indeed, the most salient feature of peace implementation in Bosnia has been the emergence of what has been variously called a “trusteeship in all but name” and a “creeping protectorate.” The principal strategy, as Susan L. Woodward put it, has been to override Bosnian sovereignty in the short term in order to establish a preferred foundation for building it in the long term, 50 with implications for the sustainability of any unity accomplished.

Democratization Postwar Bosnia faced daunting challenges for a newly democratic state. Its three leading parties were monoethnic, more habitually authoritarian than democratic, and predictably, even more so as a result of their war experience, when political and economic power was heavily concentrated among a small politico-military elite. The prospect of conducting national elections within a few months was also slim in a damaged and “ethnically cleansed” country with over half of its voting population either displaced or refugees. Furthermore, the country faced a new need to protect human rights, both to account for their past abuse and to prevent their future violation. Elections According to Dayton (Annex 3), national elections were to be held within six to nine months of the accord’s signing, by September 1996, with allowance but not insistence that municipal and cantonal elections be held as well. A national ballot was only to take place if conditions pertained in the country that ensured a comparatively free and fair election, specifically referring to freedoms of expression, press, association, and movement; right to vote without fear or intimidation; and importantly, “a politically neutral environment.” The OSCE was to supervise all aspects of elections, including verification of acceptable conditions. Its head of mission would chair a Provisional Election Commission (PEC) composed of both international and Bosnian members that would establish all electoral rules and regulations until Bosnia could establish a Permanent Election Commission of its own.51 The PEC was given the mandate to establish subsidiary bodies and has in effect created a potent network of election-related offices, such as a judicial Election Appeals Sub-Commission (EASC), a National Election Results Implementation Commission (NERIC), and a Media Expert Commission, supplanted in 1998 by an Independent Media Commission (IMC). Dayton put the first round of elections on an exceptionally tight sched-

S-18

11/24/09

4:26 PM

Page 555

THE DAYTON AGREEMENT ON BOSNIA

555

ule. Furious debate arose over the wisdom of holding the election on schedule, not only because of operational concerns (for instance, voter registration for refugees did not begin until June, and a full set of rules governing electoral conduct was published only in July), but, more importantly, due to the manifest absence of “politically neutral” conditions.52 The prevailing environment was still one of generalized insecurity and partisanship, with indicted war criminals dominating political life,53 opposition politician figures and minority communities being targets of intimidation and attack, and minimal freedom of media and movement. Some such as OSCE Chairmanin-Office Flavio Cotti predicted, accurately as it turned out, that holding national elections in such circumstances would restore to power the wartime leaderships least likely to commit themselves to building peace, only this time with democratic legitimacy. However, the OSCE was under enormous pressure to go through with elections on time. Many U.S. and European diplomats expressed the view that elections were an essential first step to get Bosnia’s new joint institutions off the ground, all the more so in the absence of a mandate for the international exercise of transitional authority and a consequent need to put Bosnian interlocutors in place to begin postwar reconstruction.54 Even more decisive was the imperative of holding elections before IFOR’s mandate came to an end.55 Importantly, 1996 was also a presidential election year in the United States, and the Clinton administration had publicly promised that U.S. troops would be out of Bosnia by December. The elections were thus held in September. Operational problems undoubtedly existed, though the extent and impact of these have been debated.56 The widely regarded principal flaw in the process, however, was the handling of voter registration. The OSCE wanted to enable maximum turnout, including for the displaced, who could not register in person in their place of residence. To accommodate this situation in which large numbers of voters were dispersed both in and out of the country, the PEC devised a “P-2” form in which voters could register in municipalities where they simply “intended” eventually to reside. Because the criteria for establishing intention were so minimal, the P-2 effectively became a mechanism for ethnic gerrymandering on a large scale. The results of the elections surprised no one: the three leading nationalist parties were victorious in all offices.57 Between the 1996 national ballot and 2000, Bosnia conducted four more statewide rounds of polling for various levels of government. At each stage, while the security situation was improving, the OSCE made technical improvements and eventually established a regulatory framework for electoral conduct that provided a solid, legal foundation for Bosnia’s continuing political evolution. Moderate opposition parties also made steady if still small inroads, with their best showing in the municipal ballot of April 2000. Implementers faced a major challenge, however, in the constitutional

S-18

11/24/09

4:26 PM

Page 556

556

CASE STUDIES

framework established at Dayton, which inadvertently inhibited the formation of nonnationalist constituencies and the electoral success of nonnationalist parties. Insofar as many offices could only be elected by residents of one entity or the other, politicians had no incentive to build a base that crossed ethnic lines. A 1999 Draft Election Law sought to remedy some of these shortcomings, but had not been ratified as of 2000 by Bosnia’s two parliamentary bodies or implemented by entity parliaments.58 Reform has been delicate, since it likely entails amending Bosnia’s constitution, which opens the door to revising the Dayton Agreement itself. A final drawback has been the complexity of the electoral system, even with simplifications that have been made over time. Over seventy parties and nearly 600 candidates were represented in the 1998 elections.59 Elections swollen by large numbers of parties representing very small constituencies tend to engender fragmented, divisive electioneering and perception among bewildered voters that the new system is unresponsive to their interests. Beyond Elections Elections are only one element of democratization, of course. Over the long term, successful democratization will involve a dense network of political, institutional, and even cultural changes, including reform of civil-military relations, establishment of a legal environment that fosters accountability, political party and civil society development, and growth of independent media. In the short term, however, the democratic process in Bosnia was particularly challenged by the perceived and actual impunity of war criminals, the practical frailty of the human rights institutions established under Dayton, and a highly politicized media environment. A central question immediately upon the war’s end was whether and how quickly persons indicted for war crimes would be apprehended for trial at the International Criminal Tribunal for Yugoslavia (ICTY) at the Hague.60 Dayton obligated Bosnian authorities to comply with all international human rights and enforcement efforts, expressly including the ICTY (Annexes 1 and 6). The agreement also authorized the NATO-led force to apprehend any indictees. Yet in the first year and a half of Dayton implementation, PIFWCs enjoyed near total impunity, and those Bosnian wartime leaders who had been indicted, from Karadzic through midlevel officials, exercised either overt or covert influence over Bosnian politics.61 Few surrendered to the Hague voluntarily and both Bosnian entities and neighboring governments gave them safe haven. IFOR essentially abdicated its authorized responsibility to apprehend indictees with whom it was in effective contact, although over time NATO forces showed greater willingness to apprehend them. At an early stage, the ICTY was also faulted by some critics—and the bulk of the Serbian population—for pursuing indictments and prosecutions disproportionately against Serbs. The ICTY eventu-

S-18

11/24/09

4:26 PM

Page 557

THE DAYTON AGREEMENT ON BOSNIA

557

ally responded to this criticism by showing greater commitment to investigations of Croatian transgressions during the offensives of 1995 as well as alleged NATO transgressions during the air war over Kosovo.62 Apart from the issue of war criminals, Dayton also sought to set up sustainable mechanisms to address violations of human and civil rights (Annexes 4 and 6). A Commission on Human Rights was established, the Federation had its own Human Rights Ombudsman, and Dayton’s Human Rights Annex included an unprecedented array of provisions to which the authorities agreed to be party. In a significant derogation of sovereignty, the Bosnian constitution also gave primacy to the European Convention for the Protection of Human Rights and Fundamental Freedoms (and its protocols) over domestic legal provisions. From the perspective of implementation, however, the human rights provisions were less robust. No one implementer had the lead. Both the OSCE and the Council of Europe had authorized roles, as did the European Court of Human Rights. The UN High Commissioner for Human Rights had a special rapporteur in the region and there was a Human Rights Office for the IPTF. Meanwhile, observance or knowledge of human rights abuses, simply because of their mandated activities, could not be avoided by the UNHCR, the UN Mission in BosniaHerzegovina (UNMIBH), and European Community Military Monitors. Perhaps unsurprisingly, Dayton’s human rights provisions generally lacked sufficient commitment by or coordination among international actors, despite the creation of a Human Rights Coordination Center (HRCC).63 Beyond the sheer complexity of the system and lack of funding, the constitution did not specify enforcement mechanisms, a responsibility left to the entities rather than the state. With police estimated to be the main violators, this effectively left enforcement to international efforts, a role, however, that NATO-led forces assiduously avoided and for which the IPTF was ill equipped. 64 More broadly, the Bosnian judicial system remained compromised by factors such as overt political interference, insufficient funding, a lack of security for judicial personnel, shortages of competent personnel, equipment, and legal resources, and serious caseload backlogs in a number of courts.65 Finally, to foster democratization, the international community actively supported media independence. Its strategy changed over time, from ambitious but disappointing initial efforts, such as the first year of the Open Broadcast Network, to lower key but more effective efforts, such as support for independent radio stations.66 More generally, however, implementers used their powers under Dayton—and Bonn—to foster an environment within which independent media could flourish. This, at times, involved shutting down media seen as nondemocratic,67 but over time a regulatory framework has emerged from the decisions and rules of an Independent Media Commission, set up in 1998 as a subsidiary body of the PEC. A wider spectrum of activities surrounding human rights, rule of law,

S-18

11/24/09

4:26 PM

558

Page 558

CASE STUDIES

media independence, and civil society development has now bolstered the prospects for deepening democracy in Bosnia. However, strategically, international implementers have handled poorly the question of managing Bosnia’s least democratic individuals and parties. Even though the ICTY progressively became more effective, the prolonged initial period of inaction toward indictees had an immeasurably destructive impact. Moreover, the approach alternately to back individual favorites or use Daytonauthorized institutions to sanction or remove officials from office when they were seen to be violating the spirit of Dayton has undermined popular confidence in the fairness of the system.68 At the very least, this encouraged the view among many Bosnians that accountability to voters was less important than accommodation to whoever wielded power, which presently remained the international community.

Conclusion To evaluate the Bosnian operation requires clarifying premises about its objectives and thus the appropriate criteria for assessing success or failure. Postconflict peacebuilding for civil wars, arguably, should give primacy to three objectives: first, securing a viable and sustainable cease-fire; second, creating or facilitating the conditions necessary for internal conflict resolution, or “self-sustaining peace”; and third, concluding the international peace presence in a country in such a way that normalization of domestic affairs can occur.69 Other priorities, including long-term democratization and postwar justice, should demonstrate their relationship to these primary goals. To advocate alternative priorities, one would thus need to show that a different ordering of objectives would be more likely to prevent resumption of large-scale armed violence. If one accepts this approach, then the criteria for success in Bosnia entail, first, a sustained cease-fire that can deepen into long-term stability; second, establishment of the conditions necessary for a self-sustaining peace; and third, the exit or significant downsizing of the international presence in Bosnia. These three criteria, or course, beg additional questions. To argue the importance of establishing conditions favorable to internal conflict resolution is only the first, and easiest, step. Identifying what those conditions are, how they relate to the provisions of a peace agreement, and how international actors can help foster them is the challenge; arguably, this should also be the central task of implementation. By mid-2000, only the first criterion—consolidating the cease-fire of October 22, 1995—had been fulfilled. Despite major achievements in multiple arenas of Dayton implementation, the second and third criteria—fostering conditions for self-sustaining peace and the prospect of timely international exit—had not been met. Moreover, even the cease-fire so ably

S-18

11/24/09

4:26 PM

Page 559

THE DAYTON AGREEMENT ON BOSNIA

559

stabilized by IFOR was not fully self-sustaining; thus the continuing military presence of SFOR. Residual military and public insecurity remained and, in combination with the continued power of nationalist parties, suggested that peace might not last long if the international presence was significantly downsized in the near term. Hence, five years after Dayton, the international community still maintained a formidable security presence as well as an expansive civilian administration. Why, then, this apparently “permanent international presence” in postwar Bosnia? First, the mediators of Dayton believed that a balance of power would best provide and maintain long-term military security between the two entities. In the effort to bolster the military capacity of the Bosniacs (primarily) and Croats (secondarily) as a deterrent against future aggression by the army of the RS, the main early accomplishment of the Train and Equip program of military assistance was to replace the risk of a Serb offensive with the risk of a Bosniac one. Moreover, although the program portrayed its efforts as an initiative designed to enhance military cooperation and interforce confidence, it was not perceived as such by the Serbs, nor did it bear significant fruit in terms of integrating Bosniac and Croat forces within the Federation. Second, the absence of public security contributed heavily to the continuing dependence of Bosnia’s peace on an international presence. The consolidation of the cease-fire among Bosnia’s three armies was not effectively extended to civilian populations, especially in the early years of implementation, despite the fact that both IFOR and SFOR had the mandate and operational capacity to do so. Despite tactical adjustments over time in civilian-military coordination, 70 early inaction among military implementers set a decisive tone for the ensuing years of the peace operation, negatively affecting the possibilities for a self-sustaining peace—particularly those aspects of international strategy dependent upon the interrelated tasks of returning Bosnia’s displaced populations and promoting political pluralism. Third, an emphasis on rapid physical reconstruction in the Federation required working through existing channels of authority and further empowered politicians and bureaucrats hostile to Dayton’s integrationist goals. More, the aid conditionality against the RS led to at least a two-year delay of its economic recovery.71 The consequent lack of jobs, housing, and social benefits, compared to the Federation, inhibited IDPs and refugees from going back to the RS and undermined the international community’s strategy to gain both return and democratization through a demographic remixing of the population. Fourth, the decision to allow national elections to go ahead in September 1996 had a decidedly negative impact on efforts to facilitate a self-sustaining peace. Bosnia’s voters, faced with a measure of physical insecurity and a particularly uncertain future, understandably voted for

S-18

11/24/09

560

4:26 PM

Page 560

CASE STUDIES

nationalist politicians who (if nothing else) had consistently promised to protect their own ethnicity’s interests. Early elections made it even harder to find mechanisms to compensate for the hardwiring of monoethnicity that had been built into the Dayton constitution. Finally, the international community’s increasing exercise of its own authority to impose decisions in Bosnian politics rendered an exit on good terms elusive. It also set a troubling precedent that risked undermining both the legitimacy of the international community’s presence in Bosnia and long-term democratization. There is a contradictory and self-perpetuating nature to the exercise of international authority to create Bosnian state unity: the more it was used to combat perceived obstructionism, the more external actors were subsequently relied upon and needed to maintain control. The emergence of such a “trusteeship” strategy of implementation raises several issues, the very first of which is at the crux of the challenge of peace implementation. Third parties presumably only lend themselves to such efforts when belligerents are seen as unable to implement a settlement on their own. The parties may need a third force to establish a buffer and build confidence during a fragile early peace, which is the classic peacekeeping model. The parties may also need institutional or technical assistance that is simply unavailable indigenously within a short time frame for everything from de-mining to running a postwar election. In Bosnia, a third factor was principally at work: the cease-fire, negotiations, and signing of the Dayton Agreement had only been accomplished through third-party coercion. The parties were far from committed to the agreement, and there was thus likely to be heavy continuing negotiation, even enforcement, needed to ensure that they abided by the terms of the accord. Hence, from the first day of Dayton implementation, there were nearly irreconcilable tensions between the substantial power given to the entities, the right of return promised to displaced persons, and the legal sovereignty and responsibilities accorded an ostensibly unitary Bosnian state. In this sense, Bosnia was one of the toughest of tough cases among post–civil war environments, replete with parties who were likely to act as spoilers and needed to be managed as such by international implementers. As George Downs and Stephen John Stedman argue in Chapter 2, the challenge for implementers, if they recognize a tough case at the outset, is to structure their involvement in such a way that they will be well equipped to overcome anticipated noncompliance and resistance. Dayton’s implementers, however, were constrained by domestic politics (especially U.S. preoccupation with its exit strategy) and their allocation of roles on the basis of institutional and national bargaining rather than comparative advantage. This made it impossible from the beginning to pursue the strategy most observers knew would be required and resulted

S-18

11/24/09

4:26 PM

Page 561

THE DAYTON AGREEMENT ON BOSNIA

561

instead in an initial strategy of decentralized civilian implementation that was manifestly the weakest foundation possible in a context requiring tough, ongoing negotiation with the parties from a position of strength. Over time, Dayton’s implementers adopted the strategy with which they arguably should have begun: tight coordination among civilian efforts with a panoply of carrots and sticks at their disposal, capacity to override noncompliance and establish certain ground rules for a continuing peace and political process, and integrated civil-military command. Adopting such a strategy incrementally, however, had a very different effect than if it had been used from the outset. First, time in which progress is not made for want of a more robust implementation effort is simply time lost. More troubling were the signals sent to Bosnian parties and the population at large about democratic process and legitimacy: if due process does not yield results, override it, at least if one has the political backing to do so. When the implementers shifted to the trusteeship strategy to accomplish one set of implementation objectives, they risked undermining another, principally those of fostering institutions and a culture of democratic accountability. The trusteeship strategy had at least two other weaknesses as practiced in Bosnia. Instead of using international authority early and in an interim fashion to establish a framework that could then be handed over to Bosnian authorities, it was used to compensate for inaction or intransigence among Bosnian leaders, with the result that international implementers appeared increasingly and intimately enmeshed in domestic Bosnian politics. The second, related weakness is that of exit, with the international community likely to be in Bosnia for a very long time.

Notes 1. In 2000, the international military presence was scaled down to approximately 20,000 troops, while international police were maintained at just over 2,000, still considerable in a country slightly smaller than West Virginia (51,233 square kilometers). 2. The Yugoslav state comprised six republics (Croatia, Serbia, BosniaHerzegovina, Slovenia, Montenegro, and Macedonia); two semiautonomous provinces linked to Serbia (Kosovo and Vojvodina); and six constituent nationalities (Croatian, Serbian, Slovenian, Montenegrin, Macedonian, and Muslim). 3. Under the 1974 constitutional amendments, each republic acquired its own central bank, communist party, educational system, judiciary, and, very importantly, police. The only institution that still operated exclusively at the federal level was Yugoslavia’s national army (the JNA), though it now acquired, as commander in chief, an eight-member, rotating federal presidency. Unique among the republics, Bosnia also had a collective presidency composed of two Muslims, two Serbs, two Croats, and one Yugoslav. 4. Susan L. Woodward discusses the country’s dissolution as an example of

S-18

11/24/09

562

4:26 PM

Page 562

CASE STUDIES

state failure, with particular emphasis on its inability to manage fiscal and economic challenges. See Woodward, Balkan Tragedy: Chaos and Dissolution After the Cold War (Washington, D.C.: Brookings Institution, 1995). 5. See Stuart Kaufman, “The Irresistible Force and the Imperceptible Object: The Yugoslav Breakup and Western Policy,” Security Studies 4, no. 2 (winter 1994–1995): 281–329; David A. Lake and Donald Rothchild, “Spreading Fear: The Genesis of Transnational Ethnic Conflict,” in Lake and Rothchild, eds., The International Spread of Ethnic Conflict: Fear, Diffusion, and Escalation (Princeton: Princeton University Press, 1998); and Barry R. Posen, “The Security Dilemma and Ethnic Conflict,” in Michael E. Brown, ed., Ethnic Conflict and International Security (Princeton: Princeton University Press, 1993), pp. 103–124. 6. Woodward, Balkan Tragedy, p. 33. 7. See Laura Silber and Allan Little, Yugoslavia: Death of a Nation (New York: TV Books, 1995), p. 209. 8. For accounts of the negotiating history, see Richard C. Holbrooke, To End a War, rev. ed. (New York: Random House, 1999); David Owen, Balkan Odyssey (New York: Harcourt Brace, 1995); and Bertrand de Rossanet, Peacemaking and Peacekeeping in Yugoslavia, Nijhoff Law Specials vol. 17 (The Hague: Kluwer Law International, 1996). 9. See Manfred Nowak, “Lessons for the International Human Rights Regime from the Yugoslav Experience,” Collected Courses of the Academy of European Law 8, no. 2 (The Hague: Kluwer Law International, 2000). 10. A UN Protection Force (UNPROFOR) was established in February 1992 to facilitate cease-fire in Croatia and to complement ongoing international mediation. UNPROFOR’s mandate was subsequently extended to Bosnia and included responsibility for six designated “Safe Areas,” the first of which was Srebrenica. Security Council Resolution 819, UN Document S/Res/819, 1993, paras. 1–4. The Safe Area concept was extended to Sarajevo, Tuzla, Zepa, Gorazde, and Bihac with Security Council Resolution 824, UN Document S/Res/824, 1993, paras. 3–4. 11. UN Document S/1995/444, “Report of the Secretary-General Pursuant to Security Council Resolutions 982 and 987,” 1995. 12. U.S. and European mediators advocated as a basis for settlement that Bosniac and Croat forces would hold 51 percent of Bosnia’s territory, and Serb forces the remaining 49 percent. 13. The UN Protected Areas (UNPAs), all in Croatia, were part of UNPROFOR’s original mandate and distinct from the Safe Areas later established in Bosnia. 14. Owen, Balkan Odyssey, p. 335. 15. Statement of Agreed Basic Principles. See also Holbrooke, To End a War, pp. 133–141. 16. When IFOR’s mandate was first renewed beyond December 1996, its name was changed to the Stabilization Force (SFOR). 17. Dayton’s military provisions also covered arms control and confidencebuilding—a job given to the Organization for Security and Cooperation in Europe (OSCE). 18. An International Criminal Tribunal for Yugoslavia (ICTY) was established by the Security Council in 1993 to prosecute and try persons indicted for violations of international humanitarian law, primarily: grave breaches of the Geneva Conventions, violations of the laws or customs of war, genocide, and crimes against humanity. 19. The high representative also simultaneously reported to the UN Security Council.

S-18

11/24/09

4:26 PM

Page 563

THE DAYTON AGREEMENT ON BOSNIA

563

20. The PIC Steering Board consisted of the G-8 (which amounts to the Contact Group plus Japan and Canada), the OSCE chairman-in-office, and the EU president. 21. Confidential author interviews, Sarajevo, August 1998. 22. The fate of Brcko—the town located along a narrow territorial corridor connecting the two halves of the RS and claimed by both the RS and the Federation—was to be decided by international arbitration, as provided for under Dayton’s military provisions. 23. Confidential author interviews, IFOR, OHR, OSCE, UN Mission in Bosnia-Herzegovina (UNMIBH), U.S. Departments of State and Defense, 1996. See also Georgios Kostakos, “Division of Labor Among International Organizations: The Bosnian Experience,” Global Governance 4, no. 4 (October–December 1998): 461–484. 24. Confidential author interviews, U.S. Departments of State and Defense, Washington, D.C., 1996 and 1997. 25. “General Framework Agreement for Peace,” Annex 1A, Articles 2-1, 2-3, and 3. 26. Brcko had been seized and held by Bosnian Serbs since the spring of 1992, when they drove out the bulk of the Bosniac and Croat population and sought— without success—to expand the five-kilometer-wide land corridor between the two halves of the Serb republic. The area was perceived as strategically critical for both sides: to take Brcko away from Serbs would effectively cut Republika Srpska in half; to give it to them would, beyond sanctioning ethnic cleansing, deprive the Federation of access to the Sava river and the trade and communications links that it represented. 27. Active-duty troops were reduced according to schedule from a wartime high of over 300,000 to 86,000 in 1997, in a ratio of two (Federation) to one (Republika Srpska). 28. However, a sizable number remained in Bosnia, especially near Zenica. 29. Agreement on Sub-Regional Arms Control (Florence Agreement), June 14, 1996, modeled on the Conventional Forces in Europe (CFE) Treaty. 30. See RFE/RL Newsline, July 18, 2000. 31. That there was no ready text or convention to outline these internationally recognized standards of law enforcement became a sticking point in the first stages of implementation, and the IPTF found itself having to draft the standards to which Bosnian police—and international police monitors themselves—would subsequently be held. Confidential author interview, Sarajevo, November 1996. 32. UN Document S/1996/1012, “Conclusions of the Peace Implementation Conference,” December 6, 1996, para. 77. Security Council Resolution 1088, December 12, 1996, para. 28. 33. Displacement figures from UNHCR Sarajevo, Statistical Summary, November 1998. 34. Fifteen conventions are included in the text. 35. By 1997 the bus service had transported 460,000 Bosnians across the IEBL—an average of nearly 9,000 people per week. USCR, “Country Report: Bosnia and Herzegovina,” World Refugee Survey 1998. A different initiative—a “Pilot Return Project” to swap Bosniac and Croat IDPs within the Federation by reciprocally exchanging the populations of designated towns—essentially failed due to obstruction from local authorities. 36. UNHCR, “Open Cities Status Report,” August 1, 1999. 37. European Stability Initiative (ESI), Interim Evaluation of Reconstruction

S-18

11/24/09

564

4:26 PM

Page 564

CASE STUDIES

and Return Task Force (RRTF), September 14, 1999; OHR, “Property Law Implementation Plan (PLIP),” unpublished paper. 38. As of early 2000, significantly higher levels of minority return appeared to have been occurring, due to such factors as refugee impatience, increased international effectiveness, and a notable shift in majority- and minority-group psychology. See ICG, Bosnia’s Refugee Logjam Breaks: Is the International Community Ready? May 30, 2000. 39. OHR, “Property Law Implementation Plan,” p. 4. 40. UNHCR, Registration of Repatriates in the Republika Srpska and Entitlement to Identity Documents, Food Assistance, and Medical Care, April 1999; and UNHCR, Extremely Vulnerable Individuals: The Need for Continuing International Support in Light of the Difficulties to Reintegration upon Return, November 1999. 41. Between 1996 and 1999, only 349,969 Bosnian refugees returned; the UNHCR had anticipated the return of as many as 400,000 for 1996 alone. UNHCR, “Returns Summary to Bosnia and Herzegovina from 01/01/96 to 31/12/99.” 42. The ICG estimated that 381,000 refugees remained without durable solution in late 1999 (the CRPC offers a slightly lower estimate of 324,000), with 223,000 Bosnian Serbs in the Federal Republic of Yugoslavia, 30,000 Bosnian Croats in Croatia, and another 128,000 Bosnians throughout Western Europe—the majority in Germany. ICG, The Balkan Refugee Crisis: Regional and Long-Term Perspectives, June 2, 1999, pp. 2–3. Commission for Real Property Claims of Displaced Persons and Refugees, Return, Local Integration, and Property Rights, November 1999. 43. UNHCR, “Returns Summary to Bosnia and Herzegovina from 01/01/96 to 31/12/99.” The most prominent instance of postwar displacement occurred when 60,000 Serbs fled from Sarajevo’s suburbs during their transfer from Serb occupation to Federation control, leaving only 8,000 Bosnian Serbs remaining in an SDAdominated Sarajevo. For more details, see Louis Sell, “The Serb Flight from Sarajevo: Dayton’s First Failure,” Eastern European Politics and Societies 14, no. 1 (winter 2000): 179–202. 44. Confidential author interview, OHR, Sarajevo, November 1996. 45. Constitution of the Federation, III: Division of Responsibilities Between the Federation Government and the Cantons, Article 2, points d–g. 46. For some time, the HDZ continued to resist cooperation with parliamentary and presidential bodies, demanding a renegotiation of power-sharing arrangements with the SDA that would give them a legal status within Bosnia closer to that of the RS. 47. Peace Implementation Conference, “Conclusions,” Bonn, December 10, 1997, Section 11, paras. 2, 2b, and 2c. 48. The PIC expressed anxiety over this pattern early, noting “with concern” in mid-1998, “almost all progress has required the continued and intensive efforts from the international community.” Declaration of the Ministerial Meeting of the Steering Board of the Peace Implementation Council, Luxembourg, June 9, 1998, para. 5. 49. Despite pleas and pressure from the international community, the Bosnian parliament had not ratified the law by May, prompting the OHR to enact it de facto by incorporating it into the existing PEC Rules and Regulations. 50. Susan L. Woodward, “Compromised Sovereignty to Create Sovereignty: Is Dayton a Futile Exercise or an Emerging Model?” in Stephen D. Krasner, ed., Problematic Sovereignty: Contested Rules and Political Possibilities (New York: Columbia University Press, 2001).

S-18

11/24/09

4:26 PM

Page 565

THE DAYTON AGREEMENT ON BOSNIA

565

51. The PEC, with seven members at the beginning, has expanded over time, first to ten and then in 1998 to a total of sixteen. The last enlargement was explicitly designed to include more moderate Bosnian voices. 52. Representative of this debate: Misha Glenny, “Decision Time in Bosnia,” New York Times, September 8, 1996; Stephen S. Rosenfeld, “Sticking to the Dayton Accords,” Washington Post, June 23, 1996; and Morton I. Abramowitz, “Bosnia: The Farce of Premature Elections,” Washington Post, May 24, 1996. 53. U.S. special envoy Richard Holbrooke brokered a deal on July 18, 1996, in which Radovan Karadzic agreed to step down as president of the Republika Srpska and refrain from public political activities. He nonetheless remained an active figure behind the scenes. 54. Arguably, implementers acquired the equivalent of such a mandate over time, between the judicial mechanisms of the OSCE and the Bonn powers of the OHR, but in late 1996 this was not politically plausible. Confidential author interviews, OHR, Sarajevo, March and August 1998. 55. Chris Hedges, “Despite Doubts, Official Gives Go-Ahead to Bosnian Vote,” New York Times, June 26, 1996. 56. ICG, Elections in Bosnia and Herzegovina, ICG Bosnia Report no. 16, September 22, 1996. 57. For the Bosnian presidency, 60 percent of the Bosniac vote went to the SDA’s Izetbegovic, 67 percent of the Bosnian Croat vote to the HDZ, and 67 percent of the Bosnian Serb vote to the SDS. For the RS presidency, 59 percent of the vote went to the SDS. Similar results were obtained for the Bosnian House of Representatives, the Federation House of Representatives, and the RS National Assembly. 58. Improvements included the use of open party lists, the assignment of some parliamentary offices to fixed constituencies, a preferential ranking system of election of the Bosnian presidency, a ban on candidates who occupy property belonging to refugees or displaced persons, and tight regulation of campaign expenditures. For a concise critique of the draft law, which argues for it to go further, see Balkan Action Council, “Improving the Electoral Law for Bosnia-Herzegovina,” December 22, 1999. 59. Confidential author interview, OSCE, Sarajevo, August 1998. “Bosnia Ballots Too Bewildering for Many: NGOs,” Agence France Press, October 1, 1998. 60. The ICTY was the first international war crimes tribunal established since Nuremberg and Tokyo and was authorized by the UN Security Council to prosecute “persons responsible for serious violations of international humanitarian law committed in the territory of the former Yugoslavia since 1991” (Security Council Resolution 808, February 22, 1993). While it could issue indictments, conduct prosecutions, render judgments, and impose sentences, the tribunal was entirely dependent on others, governments and international enforcers, to bring indictees into custody unless these voluntarily surrendered, as a few have. 61. Recall that Momcilo Krajisnik, apprehended in June 2000, was elected as the Serb representative on the Bosnian presidency in the 1996 elections. 62. See 1999 ICTY press releases. 63. The HRCC consisted of the OHR, OSCE, Office of the High Commissioner for Human Rights (OHCHR), UNMiBH, and UNHCR. Critics assert that it still does not have adequate authority for coordination. See, for example, Michael O’ Flaherty, “International Human Rights Operations in Bosnia and Herzegovina,” in Michael O’Flaherty and Gregory Gisvold, eds., Post War Protection of Human Rights in Bosnia and Herzegovina (Boston: Martinus Nijhoff, 1998). 64. Manfred Nowack, “Shortcomings of Effective Enforcement of Human

S-18

11/24/09

566

4:26 PM

Page 566

CASE STUDIES

Rights in Bosnia and Herzegovina,” in O’Flaherty and Gisvold, Post War Protection of Human Rights in Bosnia and Herzegovina. 65. While a relative lack of institutional capacity is an obstacle faced by even the best judicial systems, consistent patterns of bribery and improper exertion of political influence will ultimately undermine the rule of law and impair the democratization process. See OHR, HRCC Human Rights Semi-Annual Report, April–September 1999; and several ICG reports, including Denied Justice: Individuals Lost in a Legal Maze, February 23, 2000. 66. ICG, “Media in Bosnia and Herzegovina: How International Support Can Be More Effective,” March 18, 1997. 67. In 1997, for instance, SFOR seized Serb transmitters broadcasting inflammatory content. 68. In a dramatic episode, High Representative Carlos Westendorp called in March 1999 for the removal of the legitimately elected president of Republika Srpska, Nikola Poplasen. Poplasen’s offense was his refusal to accept as prime minister a moderate who had majority support from the RS National Assembly and was favored by the West. That the occasion of his removal was his refusal to accept as prime minister an obvious Western favorite only heightened cynicism about the legitimacy of political processes. OHR press release, “Removal from Office of Nikola Poplasen,” March 5, 1999. 69. Elizabeth Cousens, Introduction to Elizabeth Cousens and Chetan Kumar, with Karin Wermester, eds., Peacebuilding as Politics: Cultivating Peace in Fragile Societies (Boulder, Colo.: Lynne Rienner, 2000). 70. See Michael Williams, Civil-Military Relations and Peacekeeping, Adelphi Paper no. 321 (New York: Oxford University Press, 1998); and George A. Joulwan and Christopher C. Shoemaker, Civilian-Military Cooperation in the Prevention of Deadly Conflict: Implementing Agreements in Bosnia and Beyond (New York: Carnegie Corporation of New York, 1998). 71. Between 1995 and 1997, the RS received between 7 and 8 percent of total donor funds expended in Bosnia, though it also was comparatively less physically ˇ canin, and Susan L. Woodward, damaged by the war. See Zlatko Hurtic, ´ Amela Sapˇ “Bosnia and Herzegovina,” in Shepard Forman and Stewart Patrick, eds., Good Intentions: Pledges of Aid for Postconflict Recovery (Boulder, Colo.: Lynne Rienner, 2000), p. 348. Some months after the RS regime changed in November 1997, aid began to flow to western RS but continued to be withheld from eastern RS due to allegations of harboring war criminals.

S-19

11/24/09

4:26 PM

Page 567

19 Peace by Unconventional Means: Lebanon’s Ta’if Agreement MARIE-JOËLLE ZAHAR

In September 1989, sixty-two of the seventy-one surviving members of Lebanon’s last prewar parliament (voted to office in 1972) met in Ta’if, Saudi Arabia, to hammer out a peace agreement to end the Lebanese civil war.1 Although it faced severe challenges, the Ta’if Agreement brought violence to an end in Lebanon. Ta’if also paved the way for rebuilding state institutions. Parliamentary elections were held in 1992 and 1996; general municipal elections were held in 1998; the transfer of power between postTa’if presidents is an orderly and peaceful process. The Lebanese state has extended its authority over most of the national territory and economic reconstruction is well under way. On the other hand, not all militias have been disbanded. Syrian troops have not withdrawn from Lebanon. Displaced persons have for the most part not returned to their towns and villages. The Parliament has approved a number of bilateral treaties that, according to critics, put Lebanon squarely in the orbit of Syria. More importantly, internal opposition has been met with repression. Human rights organizations have widely documented violations in the authorities’ dealings with political opponents of the regime. The state has also curtailed basic freedoms of expression by applying pressure on the media, nongovernmental organizations, and civil society. This quick balance sheet begs the question: In view of Ta’if’s mixed record, what accounts for the agreement’s successful implementation? I argue that Ta’if’s success is a function of the Syrian presence in Lebanon. Syria’s stake in the stability of Lebanon provided the country with the requisites to weather a number of crises. In the short term, this stability has come at the expense of human rights, democracy, and national reconciliation. In the medium term, however, it has provided a basis for economic recovery and for the strengthening of state institutions, paving the way for 567

S-19

11/24/09

568

4:26 PM

Page 568

CASE STUDIES

self-enforcing implementation. The implementation of Ta’if was largely determined by the context surrounding the negotiation of the peace agreement and the environment in which it was implemented. The text of the agreement, the context of negotiations, and the environment of implementation acted as constraints on the achievement of real democracy in the short to medium term.

The Lebanese Civil War: Participants, Issues, and Consequences Dissatisfaction with the Christian-dominated power-sharing formula was a major cause of the Lebanese civil war. The prewar political system was based on a consociational formula, the National Pact, which enshrined communal representation—or political representation on the basis of religious denomination—as the fundamental premise of the Lebanese republic.2 Despite their agreement to share power, the two “founding” communities disagreed on the identity of the country. Maronite Christians underlined the uniqueness of the Lebanese historical experience and the distinctiveness of the Lebanese people, while the Sunnis proclaimed their affiliation to the larger Arab and Muslim world.3 The various Lebanese communities have a long history of drawing outsiders into domestic politics to redress internal inequalities or to counter perceived threats from other communities.4 The resulting linkage between domestic politics and regional conflicts is a hallmark of Lebanese politics. Prior to 1975, two crises illustrated this entanglement. 5 When President Camille Sham‘un decided to join the Baghdad Pact in 1958, this was strongly opposed by the Lebanese National Movement (LNM),6 which identified with President Gamal Abdul Nasser’s discourse of Arab unity. Disagreement over Lebanon’s foreign policy orientation turned into a crisis over the extensive prerogatives granted by the constitution to the Maronite president. The opposition demanded political reforms to prevent the Maronite vision of Lebanon from dominating. In 1967, after the Arab armies’ defeat against Israel, Palestinians took up guerrilla warfare and started operating from Lebanon. According to Christian leaderships, the guerrillas threatened Lebanon’s stability. The LNM parties wanted Lebanon to embrace the Palestinian cause. Once again, differences on regional issues sparked an internal crisis in 1969 when the Lebanese Army confronted the guerrillas. The move was applauded by Christians and heavily condemned by the LNM. Egypt mediated a settlement, the Cairo Agreement, which gave the guerrillas a wide margin of maneuver.7 On April 13, 1975, twenty-six Palestinians were gunned down in retaliation for an assassination attempt against Kata’ib party leader Pierre

S-19

11/24/09

4:26 PM

Page 569

LEBANON’S TA’IF AGREEMENT

569

Jumayyil.8 This spark ignited the Lebanese civil war. The two warring factions are often labeled Christian versus Muslim; this is only partially correct. It is more accurate to describe them as pro– and anti–status quo. Those in favor of the status quo came to be known as the Lebanese Front.9 The anti–status quo forces revolved around the Lebanese National Movement and some Palestinian guerrillas. The commonly held notion that this was a religious conflict masked the fact that religious divisions roughly overlapped with power and wealth differentials. The more powerful traditional elites (mostly the Maronites) fought to maintain their privileges, and the more socioeconomically disadvantaged groups in society (mostly the Shia community) fought for more power and hence access to state resources. But sectarian-based acts of violence against civilians drove the latter to seek refuge among co-religionists.10 While the right-wing Christians took the lead in purging the Palestinian and Muslim presence from the areas under their control, this confessional homogenization of territory was not a preserve of Christian-controlled areas.11 Both Syria and Israel had stakes in the civil war. Neither country wanted to see a weak Lebanon fall under the orbit of the other. As early as January 1976, Syria mediated cease-fires and a fourteen-point Constitutional Document, the first of many spoiled peace settlements. That same year, President Suieyman Franjiyya and the Lebanese Front requested the deployment of Syrian forces to prevent an all-out victory of the LNM and its allies. In May 1976, the first Syrian forces of interposition entered Lebanon. Their presence was further legitimized a few months later when the Riyadh Conference established the Arab Deterrent Force.12 In March 1978, the Israel Defense Forces (IDF) entered southern Lebanon in retaliation for a guerrilla attack on a bus near Tel Aviv. The UN called on Israel to withdraw and the UN Interim Forces in Lebanon (UNIFIL) were sent to replace the Israelis as they pulled back. However, Israel retained a tenkilometer “security belt” entrusted to a Lebanese client, the South Lebanon Army. Israel invaded Lebanon again in 1982 ostensibly to eliminate the Palestinian presence that threatened the security of Northern Galilee. 13 Israel and Syria sought to thwart efforts to end the Lebanese conflict when either perceived these efforts as giving the other the upper hand in the country. Syria torpedoed the May 1983 agreement between Lebanon and Israel while the Lebanese Forces (LF), Israel’s closest Lebanese clients, spoiled a number of Syrian-mediated peace settlements, notably the 1985 Tripartite Agreement.14

The Ta’if Agreement In summer 1988, the situation in Lebanon had reached an unprecedented political deadlock. The protagonists could not agree on a compromise can-

S-19

11/24/09

4:26 PM

570

Page 570

CASE STUDIES

didate for the presidency of the republic. President Jumayyil appointed Army commander General Michel ‘Awn at the twenty-fourth hour to head a cabinet of transition. Muslim leaders rejected the constitutionality of this appointment and they extended recognition to Jumayyil’s last prime minister, Salim al-Huss. For the first time since the outbreak of civil war, the remaining functioning state institutions were in jeopardy. Until that time, Lebanon’s executive had continued to function, although it did not possess effective control on the ground. The main militias—the LF, Amal, Hizballah, and the Progressive Socialist Party (PSP)— effectively “governed” their respective zones of control, sidelining the army and police forces. However, some state institutions, notably the Central Bank, the Foreign Ministry, and the Presidency of the Republic, remained active throughout. The cabinet was often paralyzed by warlord ministers who boycotted its sessions, and Parliament met episodically, although it had little control over the situation. In August 1988, however, the presidency was vacant, there were two parallel governments, the Central Bank refused to take sides with either government, and the army split along confessional lines with troops loyal to ‘Awn and others closing ranks with Huss. The crisis reached such proportions that in January 1989 the Arab League’s foreign ministers meeting in an emergency session established a six-member committee to negotiate with the parties, but to no avail. General ‘Awn demanded the withdrawal of foreign troops from the country as a precondition to negotiations, whereas Prime Minister Huss and Speaker al-Hussayni insisted on the primacy of internal reforms. Against this backdrop, ‘Awn launched a “war of liberation” from Syrian occupation. His firepower was no match for the might of the Syrian armed forces, which enforced a naval blockade of the Christian enclave. However, the military escalation drew international attention. Meeting in Casablanca in May 1989, Arab leaders established a Tripartite Committee that criticized Syria as an obstacle to the restoration of Lebanese sovereignty.15 The Arab mediation effort was successful in formulating a seven-point truce plan. It stipulated that a cease-fire would come into effect on August 29, followed by a meeting of the Lebanese parliamentarians in Ta’if. Process Negotiations in Ta’if were meant to address a critical political deadlock that threatened to unravel the remaining functioning state institutions. The talks were also meant to defuse a tense military situation between the forces of General ‘Awn and his allies on the one hand and the Syrian troops in alliance with the militia of the PSP on the other.16 The Ta’if Agreement is atypical of peace agreements in civil wars. The parties to the conflict were not the main negotiators. Lebanese members of Parliament who had been voted into power in 1972, and who for the most

S-19

11/24/09

4:26 PM

Page 571

LEBANON’S TA’IF AGREEMENT

571

part had been spectators rather than actors during the conflict, negotiated the agreement. The limited power base of these elected representatives allowed potential spoilers to cast doubt on their representativeness and challenge the outcome of negotiations. The negotiators did not represent the parties responsible for either the political deadlock or the military escalation, yet they were entrusted with finding a way out of both quagmires. Here lies one of the most peculiar aspects of the Ta’if Agreement; the accord is “a package deal in which the internal and external aspects of the Lebanese are inextricably linked.”17 Foreign mediators did not directly partake in the negotiations. The Tripartite Committee prepared the ground for the meeting and Saudi Arabia provided neutral ground. However, the parliamentarians were left to their own devices to hammer out the details of the agreement. Ta’if is also unusual in that negotiations did not follow a hurting military stalemate but rather a sharp escalation between one of the internal Lebanese parties to the conflict and an outside power that acted as the main patron of the other faction. In summer 1989, General ‘Awn and his allies were clearly hurting because the Syrian-imposed naval blockade made it difficult, at times impossible, for them to replenish their arsenals. The blockade also undermined the morale of civilians in the regions under ‘Awn’s control as it resulted in shortages of medications, fuel, hospital supplies, and basic commodities. However, the Syrian troops and their allies did not perceive that they were in a hurting stalemate. Content The Ta’if Agreement addressed the contentious internal issue of power sharing and the highly sensitive issues of Lebanon’s identity and foreign policy orientation. Ta’if redistributed political power between the three major Lebanese confessions. It curtailed the powers of the Maronite Christian president,18 entrusted most executive powers to the confessionally mixed Council of Ministers (thus yielding significant power to the prime minister, who is traditionally a Sunni Muslim),19 and increased the power of the legislature and especially that of the Shiite House Speaker. Ta’if stipulated that the traditional six-to-five distribution of seats in Parliament would be replaced by an equal distribution between Christians and Muslims. None of these proposals were particularly new; most of them dated back to the first attempt by Syria to propose constitutional reforms in 1976. Militarily, Ta’if sought to extend Lebanese state authority over the whole Lebanese territory. It stipulated the disbanding of all Lebanese and non-Lebanese militias within six months of the beginning of implementation. Although the agreement did not specify the particulars of demobilization, it vaguely provided for reintegration of the militiamen under the

S-19

11/24/09

572

4:26 PM

Page 572

CASE STUDIES

umbrella of the Internal Security Forces (the Lebanese police), which would “open conscription to all Lebanese, with no exception.” While the political aspects of the agreement are self-enforcing, Ta’if provides for Syrian Army assistance in helping the legitimate Lebanese forces to extend the state’s authority.20 This provision is unique because it legitimates the military intervention of a regional power in assisting implementation. The agreement clarified the nature of relations between Lebanon and Syria. The two countries have “distinctive relations which derive their force from the roots of propinquity, history, and common filial interests.” These, in turn, formed the foundation on which Lebanese-Syrian coordination and cooperation were to be based. Ta’if stipulates that any agreements between Syria and Lebanon shall “realize the interests of the two filial countries within the framework of the sovereignty and independence of each.”21 As already noted, Ta’if also set a deadline for the withdrawal of Syrian armed forces from Lebanon. Alongside these core issues, Ta’if sought to introduce a number of reforms—including administrative decentralization, a new electoral law, the establishment of an economic and social development council, education and teaching, regulation of the media—and to reassert the objective of liberating Lebanon from Israeli occupation. These were clearly secondary to the other points discussed above. Nonetheless, Ta’if is also atypical in that some of the most important issues in peace agreements are dealt with in general statements. For example, the issue of displaced persons is considered part of the broader task of extending the state’s authority over the Lebanese territory. It is estimated that one-quarter of the Lebanese population are internally displaced while another quarter, mostly professionals and skilled workers, have emigrated as a result of the war and may never return (permanently) to the country. The problem is more acute in South Lebanon and in al-Shuf Mountain. U.S. Department of State data put the initial numbers of southerners displaced from their homes by the establishment of the Israeli “security belt” in 1977 at 200,000 people, or one-third of the region’s population.22 According to the Internal Security Forces, there were 13,326 families displaced from that region, while the Bureau of the Displaced put the number at 30,948 families.23 Yet Ta’if has only this to say about the issue: “The problem of displaced persons shall be solved in depth, and the right of each Lebanese person displaced since 1975 to return to the place from whence he was displaced shall be recognized; laws that will ensure these rights shall be enacted and the means to help in their return shall be provided.” Implementation Environment In 1989, the situation in Lebanon was dire.24 The war had claimed over 150,000 casualties (about 5 percent of the population), and more than

S-19

11/24/09

4:26 PM

Page 573

LEBANON’S TA’IF AGREEMENT

573

300,000 people were injured. A quarter of the Lebanese population (approximately 750,000 people) were internally displaced and living in substandard housing while another quarter had left the country. Unemployment was at 40 percent. The Lebanese currency’s purchasing power eroded with the currency exchange rate falling from 0.5 Lebanese pounds per dollar to approximately 1,800 Lebanese pounds per dollar. Physical capital losses in excess of $15 billion were incurred. Lebanon’s service economy was severely disrupted by extensive damages to the country’s infrastructure—including roads, ports, communications, power, water, and waste management. However, Ta’if was backed by Arab pledges of reconstruction aid. At the Baghdad Arab Summit of 1990, Arab countries approved the creation of an International Fund for the Reconstruction of Lebanon with a $2 billion capital allotment. This was not the first time that reconstruction plans were drawn or money was earmarked for this purpose. The Council for Development and Reconstruction, the superministry in charge of coordinating the reconstruction effort, was established in January 1977, and the World Bank’s first reconstruction program for Lebanon dates back to 1983.25 Ta’if also had strong international support. The United States, France, and the Vatican issued communiqués to this effect as soon as the agreement was made public. The Tripartite Arab Committee endorsed the results of the negotiations, pledging to help Lebanon emerge from its crisis and to restore the sovereignty of the state.26 It also expressed willingness to serve as an intermediary between Syria and Lebanon in future negotiations over the size and duration of Syrian troop presence on Lebanese soil. In Lebanon, most factions welcomed the outcome of the Ta’if negotiations with the exception of General Michel ‘Awn.27 During the negotiations, ‘Awn was kept appraised of the details, and he apparently gave Christian members of Parliament negotiating on behalf of the eastern regions (the region under his control) the green light to accept the terms of the agreement.28 Yet upon their return home, ‘Awn publicly disavowed the Christian parliamentarians. He objected to the fact that Ta’if did not commit the Syrian armed forces to a rapid and complete withdrawal from Lebanon. He also argued that the political reforms were unacceptable because they did not solve any fundamental political problems.29 The LF was also equivocal about Ta’if, having been put on the spot by General ‘Awn, who attempted to use the agreement to strengthen his position in the Christian areas. ‘Awn requested the LF to make its position public, intimating that failure to do so meant approval of Ta’if, hence treason to the Lebanese cause. Initially, the LF reserved its opinion on the agreement and sought instead to revive its plan to solve the Lebanese conflict on the basis of a federal solution.

S-19

11/24/09

574

4:26 PM

Page 574

CASE STUDIES

Compellence and Compliance In summer 1990, regional and international developments radically transformed the context of implementation of Ta’if. These were instrumental in giving Syria free rein in using military force to subdue the rebellion of General ‘Awn. The manner in which this rebellion was tamed unequivocally altered the balance of forces on the domestic political scene in Lebanon, heralding the marginalization of Christian political forces in the establishment of the Second Republic of Lebanon. General ‘Awn’s attempt at spoiling the Ta’if Agreement was facilitated by his absence at the negotiating table. Throughout the civil war, “the most important deal in the Lebanese conflict revolved around who was included in the negotiations and who was excluded or chose not to join the negotiations.”30 Citing dissatisfaction with the content of the agreement, ‘Awn moved to spoil Ta’if by dissolving Parliament, thus blocking the election of a new president and the enactment of the agreement into law. The attempt failed. Members of Parliament elected Deputy René Mu‘awwad to the presidency. 31 General ‘Awn refused to relinquish power to President Mu‘awwad and later to President Ilyas Hrawi, both legitimately elected within the constitutional bounds of Ta’if. Hrawi relieved General ‘Awn of his command in November 1989, but ‘Awn refused to give up. ‘Awn also condemned the LF for not clearly supporting his position. On January 30, 1990, a major LF–Lebanese Army confrontation erupted. The long and bloody struggle left both sides weakened and vulnerable. 32 The Syrian Army took advantage of this vulnerability and on October 13, 1990, moved against ‘Awn (with the support of the Lebanese Army units loyal to the proTa’if political establishment), forcing his troops to surrender.33 The swiftness of Syria’s victory was mainly due to the asymmetry of power between the two parties. While the 35,000-strong Syrian forces were backed by air power, General ‘Awn’s troops—approximately 8,000 men— were exhausted by their long struggle with the LF. Iraq, ‘Awn’s former weapons supplier, was already entangled in the Gulf War, while France, ‘Awn’s foremost political ally, refrained from intervening. Why did the international community condone the Syrian attack in October 1990? This tacit endorsement is puzzling given the international support for ‘Awn’s “war of liberation” in 1989. Two factors account for this reversal of fortunes. The first is internal, the second external. The bloody struggle between ‘Awn’s troops and the LF “exasperated those few influential friends he still had left.”34 Iraq—Syria’s regional foe and rival—had thrown its weight behind General ‘Awn because, as the leader of a strong and united Christian faction, he could have counterbalanced Syria’s influence. France and the Vatican had also supported him as the representative and protector of Christians. But the “war between brothers” was costly, not

S-19

11/24/09

4:26 PM

Page 575

LEBANON’S TA’IF AGREEMENT

575

only economically, but also because it weakened the Christian political factions vis-à-vis Syria and its allies. At the same time, developments in the Gulf cleared the way for Syrian ascendancy in Lebanon. Iraq’s invasion of Kuwait in August 1990 diverted the international community’s attention away from Lebanon. Preoccupation with events in the Gulf region distracted the Tripartite Arab Committee from its supervisory role in the implementation of Ta’if. Syria’s participation in the Desert Storm coalition also bridged the gap between the Syrian regime and the international community. When Syrian troops joined the U.S.-led coalition against Iraq, “this precipitated a new Syrian-U.S. understanding that not only gave Syria the green light to move against General ‘Awn but also appeared to tacitly acknowledge Damascus’s special interest in Lebanon.”35

Syria’s Priorities in Implementing Ta’if Upon the forcible removal of General ‘Awn, Syria effectively became the external implementer of Ta’if, and the agreement’s implementation strategy reflects Syrian priorities. Syria has and will continue to have interests in the internal stability of Lebanon. Damascus has always considered Lebanon as its weak spot. An unstable Lebanon poses two kinds of threats to Syria. It can harbor political movements hostile to its ruling regime. It can also become a security liability in the conflict with Israel. The potential establishment in Lebanon of a regime friendly to Tel Aviv has long been a concern of the Syrian authorities. This was the reason for Syria’s adamant objection to the May 17 agreement between Lebanon and Israel. This concern also explains the Syrian insistence on linking the Syrian and Lebanese tracks in the Arab-Israeli peace talks. Lebanon is Syria’s trump card in negotiations with Israel. The presence of Syrian and Israeli troops in parts of Lebanon allows both countries to use their weaker neighbor as the theater for their confrontation. Both states can “put pressure on each other without risking a destabilization at the Golan front.”36 Although security concerns are at the core of the Syrian policy toward Lebanon, there is an important economic dimension to the Syrian calculus. Lebanon is one of Syria’s main windows to the outside world. The Lebanese market has served as a source of Western consumer goods for Syrians unable to acquire these goods in their own country. Lebanon’s freemarket economy allows Syrians to satisfy consumerist urges without endangering or putting pressure on the socialist economy of Syria. Lebanon has also provided employment for half a million to one million Syrians, about one-seventh of the Syrian labor force. The jobs reduce unemployment in Syria and provide a source of remittances estimated at $1–3 billion

S-19

11/24/09

576

4:26 PM

Page 576

CASE STUDIES

per year as well as foreign exchange earnings.37 Lebanon has also provided Syrian upper- and middle-rank officers with opportunities for personal enrichment. A Wall Street Journal article describes the situation as follows: “The army that came in to make peace [among the various Lebanese factions] has dug in to make money. . . . Lebanese merchants say that ‘rackets run by officers in Syria’s army and secret police control thriving trades in vegetables, cigarettes, clothing and others. There is large scale smuggling from Lebanon to Syria.’”38 The regime has turned a blind eye to exactions committed by the Syrian armed forces in Lebanon but it was safer in the knowledge that the contentment of soldiers and officers decreased any threats to the regime from within army ranks. Although Syria has 35,000 troops in Lebanon, this does not seem to add substantially to the already heavy military burden of Syria. Rather, it has been suggested that Syria’s armed presence in Lebanon might in fact be a net gain rather than a loss.39 Syria used its new position as implementer of the Ta’if Agreement to negotiate a number of bilateral coordination and cooperation agreements with Lebanon, spanning every conceivable domain from security to labor. The terms of these agreements effectively tie Lebanon to Damascus’s definition of its own strategic and military interests. For example, the 1991 Treaty of Brotherhood, Cooperation, and Coordination practically legalizes Syrian involvement in matters of Lebanese internal security. Syria also reinterpreted the provisions of the Ta’if Accord pertaining to the withdrawal of its armed forces from Lebanon. Whereas the accord provided for redeployment of the Syrian troops within a two-year period ending in September 1992, the Syrian interpretation is that this withdrawal is contingent upon the implementation of sweeping deconfessionalization in Lebanon.40 In light of these facts, analysts have characterized the relationship as one of irreversible domination of Lebanon by Damascus. Others, however, point out that the Syrian influence is far from irreversible.41 They remark that in spite of the extreme imbalance of power between the two countries, the agreements are clearly interstate accords and, in that capacity, they recognize the sovereignty and independence of the two parties.

Implementation: A List of Tasks or a Menu for Choice? The implementation of the Ta’if Agreement reflected the nature of the agreement as well as changes in the environment of implementation. Ta’if does not only address domestic Lebanese problems; it seeks to define Lebanon’s relations with Syria. The Syrian emergence as the external implementer of the agreement meant that the priorities of the Asad regime colored the implementation of Ta’if. The selective focus on some aspects of the agreement to the detriment of others reflected Syrian concerns and

S-19

11/24/09

4:26 PM

Page 577

LEBANON’S TA’IF AGREEMENT

577

wider regional objectives. It also reflected the strategy of the new dominant political forces in Lebanon. Indeed, Syria’s intervention in the process of implementation shifted the balance of power in Lebanon away from the Christian factions. The Lebanese allies of Syria have, however, remained true to the Lebanese propensity to invite external intervention in domestic affairs by communities seeking to strengthen their political power at home. In brief, both Syria and its Lebanese allies considered Ta’if as a menu of choice rather than a list of tasks. The items that were given priority reflect the interests of the domestic and external implementers. In the short term, the heavy-handed strategy of implementers raised concerns about the success of the Ta’if Agreement. Political Reforms The bulk of the Ta’if Agreement constitutes projected changes to the Lebanese power-sharing formula. These changes, written into amendments to Lebanon’s constitution, were approved by Parliament on August 21, 1990, and signed into law by President Ilyas Hrawi on September 21 prior to the Syrian operation against General ‘Awn. In compliance with the provisions of the agreement, the first government of national unity appointed new parliamentarians to fill the seats vacated by members who had passed away during the war and bring the representation of Christians and Muslims to an equal ratio. Of all the aspects of the Ta’if Agreement, this was by far the most successful and the least controversial. This relative ease of implementation can be attributed to the nature of the proposed reforms. Indeed, most of the amendments introduced by Ta’if had already been proposed in earlier peace negotiations. As Hani Faris aptly puts it, “although Ta’if is the first negotiated accord since 1975 that has endured, the settlement it embodies did not evolve in a vacuum; rather it rests on principles discovered during earlier, unsuccessful searches for peace.”42 Demobilization The implementation of demobilization fell to the first government of national unity. This cabinet was formed on December 24, 1990, and it presented its policy statement to Parliament on January 9, 1991. In it, the government expressed its intention to implement the “Greater Beirut” security plan, the first stage in the disarmament of militias. The plan provided for the evacuation of all militias from Beirut and the surrounding area. It also stipulated that the government would take over control of all ports, a major source of militia revenue. That part of the plan won the cooperation of most of the main militias, which secured the evacuation of their fighting units

S-19

11/24/09

578

4:26 PM

Page 578

CASE STUDIES

from Beirut by December. The demobilization effort was carried out under the supervision of the Lebanese Army, although Syria pledged to stand ready in assisting the government should resistance be encountered. In a second stage, the cabinet planned the dissolution of all Lebanese and non-Lebanese militias and the confiscation of their weapons. It also pledged to “find an alternative for the young men belonging to those militias and assist them in integrating public civil or military life.” In March 1991, the main Lebanese militias were declared officially dissolved. Much of their heavy equipment was collected, although some of it was sold outside the country.43 A substantial number of militiamen were inducted into the Lebanese Army and the Internal Security Forces. 44 Initially, the Christian Lebanese Forces, Hizballah, and the Palestine Liberation Organization (PLO) resisted disbanding. The LF linked their approval to the disarming of Palestinian fighters and the redeployment of Syrian troops from the Beirut area to eastern Lebanon. Later, the militia hinted that it would maintain its military apparatus in northern Lebanon as long as Hizballah militants were allowed to carry weapons. 45 Ultimately, the LF agreed to surrender heavy weapons before the government deadline, set for April 30. As for Hizballah, it tied its own disbanding to Israel’s compliance with UN Security Council Resolution 425. The party’s refusal to disarm was ultimately endorsed following an agreement between Syria and Iran. Under the terms of the agreement, Hizballah was allowed to function as a resistance movement in southern Lebanon and the eastern Biqa‘ Valley, but vowed to function only as a political party in Beirut itself and other areas under the control of the Lebanese authorities.46 The PLO invoked the same reasons as Hizballah when it refused to disarm. However, the Palestinian refusal to surrender weapons triggered a very different response. A number of Lebanese officials dismissed PLO assertions that their weapons were necessary to recover territory from Israel.47 They accused the organization of trying to build a state within a state, and 3,000 Lebanese Army soldiers were deployed in southern Lebanon. When the PLO failed to abide by a July 1 deadline, the army moved on Palestinian camps in the vicinity of Sidon and Tyre. The clashes claimed 80 fatalities and 150 wounded, most of the dead being Palestinians. 48 Ultimately, Fatah—the largest faction within the PLO— ordered the demobilization of its regular forces. The imbalance in implementation reflects both Syrian and Lebanese interests. Syria has sought to use Hizballah as a strategic card in its interaction with Israel, whereas the PLO was an ally of Iraq, Syria’s historical regional rival. This uneven policy is even more evident when comparing the treatment of the PLO to that of other Palestinian organizations. While PLO camps were besieged with barbwire and PLO fighters were forced to disarm, the Popular Front for the Liberation of Palestine and the

S-19

11/24/09

4:26 PM

Page 579

LEBANON’S TA’IF AGREEMENT

579

Democratic Front for the Liberation of Palestine, both close to Syria, were allowed to transfer their heavy weapons to bases in the Syrian-controlled Biqa‘ Valley.49 However, Syrian interests alone do not account for the relative ease with which the Lebanese Army was able to subdue the much larger Palestinian armed presence, estimated at 6,000 men. The emergence of a strong Lebanese consensus opposing extensive Palestinian freedom of action in Lebanon is also an important factor in this equation. For a number of reasons, including the heavy toll exacted by Israeli retaliation for Palestinian raids, Palestinian excesses committed during the war, and the Lebanese refusal to allow the creation of a state within the state, a broad interconfessional consensus on the Palestinian issue developed in post-Ta’if Lebanon.50 The Law and Practice of Parliamentary Elections Unlike constitutional reforms, the crafting of a new electoral law and the holding of parliamentary elections would be among the most politically charged issues in the post-Ta’if era. Initially, Ta’if stipulated that parliamentary elections shall be held according to a new electoral Law on the basis of the Muhafaza. The new Law shall respect the principles which ensure: a) coexistence among the Lebanese communities; b) political representation for all classes and age groups in the population; and c) the effectiveness of that representation, after redrawing the administrative map within the framework of the unity of the land, the people, and the institutions.

“The electoral law devised by the government and approved by Parliament on 16 July made a mockery of the Ta’if Agreement and the basic principles of consistency.”51 The law was sent to Parliament over the objections of the Christian ministers, thus violating the stipulation that major issues required a two-thirds majority in government to be settled. Moreover, the number of deputies was arbitrarily raised to 128. Even more important, the principle of holding elections on the basis of the muhafaza (governorate) was violated. The governorate principle would have encouraged moderate multiconfessional voting. It was only applied in regions where there was no doubt about the political loyalties of the would-be parliamentarians. In regions where the elections were expected to be contested by anti-Syrian forces, or where Syria sought to reward one client over another, the muhafaza was abandoned in favor of the smaller, more homogeneous qada’ (district). Thus in 1992 and 1996, a special status was accorded for the Druze in the predominantly Maronite governorate of Mount Lebanon to secure the election of Syrian ally, PSP leader Walid Jumblatt. Mount Lebanon, a stronghold of Christian opposition, was divid-

S-19

11/24/09

580

4:26 PM

Page 580

CASE STUDIES

ed into a number of constituencies to decrease the electoral chances of opposition candidates. This blatant gerrymandering casts serious doubts over the functioning of the institution. However, the clearest illustration of Ta’if’s uneven implementation remains the holding of parliamentary elections in August 1992 over the objections of the majority of the country’s Christian political forces. The Syrian-backed Lebanese government decided to ignore the Christian decision to boycott the process, a decision prompted by grave concerns about the holding of elections prior to the redeployment of the Syrian armed forces scheduled for September of that year. Thus the government was creating the first serious breach in the process of national reconciliation.52 It has been convincingly argued that Syria also had its eyes on the September 1992 deadline when pressing for the elections to be held. From the vantage point of Damascus, the election of a pro-Syrian parliament would have provided the Asad regime with additional leverage over Lebanon in the event of the withdrawal of the Syrian armed forces.53 Christian political forces were left with little to no access to the formal structures of power as a consequence of this boycott, which further tipped the balance of communal power in favor of the Muslim community. Furthermore, parliamentary elections have been plagued with irregularities. Procedural problems include the frequent and unexpected change of electoral laws, the rigging of ballot returns, and the use of intimidation and coercion at the ballot box. The scope of violations is substantial. The 1996 elections reflected this state of affairs. According to a Saudi daily, “the dead rose and voted freely in front of observers and heads of polling centers.”54 This trend has not abated and analysts have described the 2000 parliamentary elections as “in some respects the most dominated by Damascus.”55 The Armed Forces, the Courts, and Internal Order Under the provisions of the Ta’if Agreement, the Lebanese Army was entrusted with assisting the Internal Security Forces in extending the authority of the state over the entire territory. However, in the post-Ta’if era, the army and its intelligence service (in conjunction with the Syrian armed forces and intelligence services) have played an increasingly larger role in the repression of internal opposition. Most of the repression has been justified on national security grounds. The October 1990 operation that ousted General ‘Awn from power was a harbinger of things to come. The brief and brutal operation was accompanied by widespread violations of the laws regulating the conduct of war. Soldiers loyal to ‘Awn were summarily executed, others were tortured after having surrendered. Between 1991 and 1994, hundreds of pro-’Awn activists were detained and held for questioning, often incurring in the process severe exactions including instances of physical torture.56

S-19

11/24/09

4:26 PM

Page 581

LEBANON’S TA’IF AGREEMENT

581

Upon ‘Awn’s defeat, the Lebanese Forces were the most vocal and organized opposition to the uneven implementation of Ta’if. As early as 1991, the LF voiced its dissatisfaction with the reinterpretation of the terms of the Syrian troop redeployment. The LF boycotted the cabinet and the 1992 elections to register its displeasure with the course of events. In March 1994, the LF was accused of masterminding a church bombing to destabilize the country and spoil the Ta’if Agreement. The state waged a campaign against the party, banned it, jailed most of its officials, and took over its institutions, offices, and other assets. In spite of a 1991 amnesty law pardoning political crimes committed during the civil war, LF leader Samir Ja‘ja‘ stood trial for the murders of a number of personalities, was found guilty, and was sentenced to a life term. This exclusion was made possible under the provisions of Article 3 of the Lebanese penal code, which allows crimes “against external state security” to be excluded from the general amnesty. The trial has been criticized by a number of Lebanese personalities as well as local and international human rights organizations for its opacity and its failure to abide by due judicial process.57 Since 1994, public demonstrations have been banned in Lebanon. Many individuals suspected of connections to opposition groups have been arrested. Loyalists of General ‘Awn and of the LF were harassed by the intelligence services, detained for questioning, held in contempt of due process, tortured to disclose information, and prevented from seeking legal recourse. Arrests have also been commonplace among members of the proIraqi wing of the Arab Socialist Ba‘ath party, and the Nasserite-oriented Lebanese Popular Congress.58 Most of the charges speak of subversion, threats to national security, attempts at destroying relations between Lebanon and a friendly state (Syria), and (in the case of the LF and the Guardians of the Cedars—another Christian militia) collusion with Israel. Nor was the use of the army limited to quelling political opposition. In 1996, Prime Minister Rafiq al-Hariri entered into a serious confrontation with the main labor confederation, the General Confederation of Lebanese Workers (GCLW). The GCLW called for a general strike and demonstrations on February 29 to protest against the government’s social and economic policies. The government responded by threatening to resort to force, and the prime minister announced that the army would take over security in selected areas of Lebanon for a three-month period to implement government decisions and force the confederation to abide by the ban on demonstration. The crisis was averted when the army announced on the eve of the demonstrations that Lebanon would be under curfew the following day.59 Syria was also said to have exerted pressure to prevent political instability from erupting in Lebanon on the eve of the Israeli and U.S. elections. The preservation of internal order and national security is not only the preserve of the armed forces and the police. Increasingly, the Lebanese judicial system has become involved in this task. But there have been ongo-

S-19

11/24/09

4:26 PM

Page 582

582

CASE STUDIES

ing concerns about the judiciary’s independence and impartiality.60 Most of the cases brought before the judiciary are tried in military courts that depend directly on the Ministry of Defense. Although the majority of defendants are civilians, military courts have a negligible civilian judiciary presence. The army officers who preside over these courts usually lack adequate legal training, the proceedings are summary, and lawyers state that the modus operandi of such courts prevents them from preparing proper defense for their clients. The courts have been known to violate the Lebanese code of criminal procedures and the provisions of the International Covenant on Civil and Political Rights, to which Lebanon is a signatory. The Justice Council, the highest criminal court in Lebanon, covers crimes relating to state security, terrorism, and unlawful associations. The Justice Council is a special court to which cases are referred at the discretion of the Council of Ministers rather than as a result of normal judicial procedure. Its powers are such that there is no right of judicial review of the sentences it passes, including the death penalty. The Justice Council has tried a number of high-profile cases between 1994 and 1997, most of which involved prominent members of the LF. Its proceedings have come under severe criticism, especially as relates to torture and ill treatment during pretrial interrogations, access of defendants to lawyers, and nature of the evidence mustered in support of the prosecution’s case.61 Muzzling the Press In a largely authoritarian Middle East, media pluralism and freedom of expression set prewar Lebanon apart. With the outbreak of civil war, Lebanon witnessed an explosion of illegal radio and television stations controlled by the militias. The Ta’if Agreement purported to reorganize all media “in line with the Law and within the framework of responsible freedom to serve the goals of national reconciliation and the termination of the state of war.” Ta’if thus expressly dealt with the divisive role that the media played during the war. However, the authoritarian manner in which the reorganization of the media was carried out illustrates once again the muscled approach to compliance that was the trademark of the implementation process. Since 1993, various Lebanese governments have attempted to rein in the press in Lebanon. The onslaught began in the name of “protecting state security” when the government of Prime Minister Hariri shut down the independent left-wing As-Safir in May 1993. The paper was accused of having published an article about the Israeli-Arab peace talks carrying details of confidential proposals allegedly made to Lebanon by Israeli peace negotiators. 62 The government then imposed a decree restricting political broadcasts and programming to state television and radio. It was

S-19

11/24/09

4:26 PM

Page 583

LEBANON’S TA’IF AGREEMENT

583

overruled by Parliament in 1994. In the wake of the 1996 parliamentary elections, however, the government passed another decree implementing the recommendations of the Higher Council of the Audio-Visual Media (HCAVM) and granting licenses to only four privately owned television and eight radio stations. Other stations were instructed to cease all news broadcasts within twenty-four hours of the decision and to sell off their assets and shut down within a six-week deadline. The four licensed television stations were all linked to leading figures in the political hierarchy.63 The government’s decision to implement the media law was strongly criticized by leading public figures. The government was accused of seeking to curb free speech by restricting the transmission of political information to progovernment media. This assessment was further confirmed by the announcement that although additional networks might be licensed in the future, all satellite transmissions would be subject to censorship and news that might offend friendly Arab governments would be banned. In the wake of this decision, President Hrawi attempted to prohibit Hizballah’s television station, Al-Manar, from broadcasting news. Al-Manar and the Christian Télé-Lumière had been granted exceptions from the ban of unlicensed media. This effort by the president to win Christian sympathy created a crisis that would be resolved in Damascus. President Asad argued that, at a time when Syria was facing a regional challenge from Israel, the decision of the Lebanese president was detrimental to the resistance effort. 64 Consequently, the government allowed Al-Manar to resume broadcasting. Commonalities linking the various incidents involving the media confirm the use and abuse of the media to promote a hegemonic political discourse. Journalists accused of having made contact with Israel, attempting to interview General Michel ‘Awn, or criticizing the performance of the Hariri cabinet, have routinely faced suspension, interrogation, and sometimes trials ending in prison terms.65 Displaced Persons Although mentioned only in passing by the Ta’if Agreement, the return of displaced persons figured prominently on the agenda of the first government of national unity.66 This government sought to support the right to return through appropriate legislation, in addition to seeking local and foreign assistance for reconstruction. Post-Ta’if governments have consistently included a “Minister of State for the Displaced,” whose job was to oversee the process of return. However, the early stages of the process stumbled upon the realization that practical steps in the direction of resettlement needed to keep the social and political ramifications of the process in mind. Indeed, upon the disarming of militias in 1991, the Ministry of the Interior began to execute court

S-19

11/24/09

4:26 PM

Page 584

584

CASE STUDIES

orders to evict some 40,000 families illegally occupying properties. It soon became clear that the eviction orders could only be implemented in the framework of an integrated resettlement process. In May 1992, a National Congress of Displaced People set out to establish the institutions necessary for such an integrated resettlement scheme. The Congress established a judicial committee to oversee legal problems, a special court to settle conflicts resulting from displacement, a national council for the displaced, a state ministry, and a resettlement and reconstruction fund. The bill officially creating these state organs was passed by Parliament on July 21. Officials involved in the process also decided to start the first phase of the resettlement program with easy cases that would generate a positive momentum for the rest of the country. They chose villages in the Harf district of al-Shuf Mountain. The Harf is a rich agricultural belt where the majority of displacements were caused by random shelling. The villages elected as part of this first phase were also Christian, a choice premised on the belief that it would be easier to resettle religiously homogeneous areas first. The Lebanese Army and the Internal Security Forces were dispatched to provide returnees with a sense of security, and symbolic reconciliation meetings were held to bring together the religious leaders and lay notables of the Harf and neighboring communities. Although this first phase was “completed” by June 1993, it did not go without hiccups, including “nightly visits” to the villages by Druze militiamen following the assassination of a prominent Druze official.67 In general, the process was considerably slowed by the lack of state resources and consequent reliance on private funds. Caritas, for example, provided most of the funding for the first phase of the resettlement. Another problem that hinders the return of displaced persons is the perceived politicized nature of the process. The choice of Walid Jumblatt, a Druze warlord, as minister of the displaced in the Hariri government heightened this perception.68 Armed with exceptional powers, Jumblatt chose to exercise them with regard to the resettlement of Christians in the Shuf in a manner that suggested that Christians were returning to “his” areas and that he “controlled” the security forces placed at his disposal to allay the fears of the returnees. Lebanon’s Reconstruction Lebanon emerged from the civil war all but devastated.69 However, it has also made a remarkable initial recovery. Real GDP has grown 75 percent since 1990, growth averaged 7.2 percent over 1993–1995, and the recovery brought per capita incomes from less than $1,000 in 1990 to an estimated $2,700 in 1995.70 Syria has a clear interest in Lebanon’s reconstruction, and this is one area in which the Syrians deferred to the Lebanese. The Lebanese market is

S-19

11/24/09

4:26 PM

Page 585

LEBANON’S TA’IF AGREEMENT

585

perceived as a complement to Syria’s economy, a complement even more valued as it stands in stark contrast with what is available at home. The Syrian economy is slated to benefit from the job opportunities that this process affords to its own labor force. It is estimated that more than threequarters of the construction workforce in Lebanon are Syrian workers. However, it was not until the appointment of Prime Minister Rafiq al-Hariri in October 1992 that large-scale rehabilitation and reconstruction efforts were undertaken.71 Hariri’s ten-year reconstruction plan, “Horizon 2000,” divided the reconstruction scheme into two phases. The first focused on infrastructure, the second would divert spending to education, healthcare, agriculture, and industry. The plan revolved around a central assumption: that peace in the Middle East would lure private-sector investment and reestablish the country as a regional turntable and provider of services. But peace was nowhere in sight and the Lebanese government has landed an estimated $15 billion in the red. The budget deficit runs 40–50 percent of state spending, and debt service stood at around 40 percent of the 1997 budget. Nevertheless, the power grid and the communications and transportation networks, as well as other critical infrastructures, have been rehabilitated. The war-ravaged commercial downtown of Beirut is under reconstruction and the industry, agriculture, and service sectors are undertaking in-depth rehabilitation. Although Lebanon was successful at mobilizing external funding for its reconstruction, the enormity of the task (estimated at $25 billion by the United Nations) meant that the country incurred a large external and internal debt in the process. Pledges of aid from Arab and Western sources remained below the $2 billion promised by the Arab League in 1989 and below the $3.5 billion estimated cost of the first stage of reconstruction. In 1991, the World Bank resumed its operations in Lebanon by providing technical assistance to the National Emergency Recovery Program (NERP). Lending was only resumed in early 1993 with a $175 million Emergency Reconstruction and Rehabilitation Project to support the NERP. In 1996, a supplemental emergency loan of $50 million was made to fund additional NERP works and expenditures resulting from damages incurred in the wake of Israel’s “Operation Grapes of Wrath.” By mid-1996, total commitments from official and private sources, including Eurobonds, amounted to $4 billion, of which about $3 billion has come from official sources (including $500 million in grants). Regional donors accounted for $1 billion (of which 40 percent came from the Arab Fund for Economic and Social Development). Other major donors include the European Investment Bank, the International Finance Corporation, and more recently the Japanese government. Lebanon’s economic recovery requires the active participation of the private sector. In 1994, the Investment Development Authority of Lebanon was created to support foreign investors and promote build-operate-transfer

S-19

11/24/09

586

4:26 PM

Page 586

CASE STUDIES

projects. Although foreign private investment is returning to Lebanon, the Lebanese investors are more cautious. This is partly due to the nature of the Hariri project, which was criticized for its “social coldness.” The project favors the wealthier parts of society—as demonstrated by the reduction of income and corporate taxes—and does not seem to care much for the plight of the increasing numbers of Lebanese poor.72 The lost wager on peace made it even less likely that scarce resources would be diverted into social programs. This raises a number of issues, not the least of which is the reintegration of combatants who did not join the army but are increasingly incapable of making a decent living in Lebanon. 73 It also means that Lebanon is witnessing a growing exodus of its youth in search of better living conditions. The paradox inherent in this outcome is that the government’s policies have directly contributed to scaring away those Lebanese professionals and skilled workers who had left the country during the war. Yet they constitute the most likely source of private revenue and expertise needed to ensure the success of the reconstruction process. Not only is the policy self-defeating, it also has a negative impact on the return of displaced persons. Political Deconfessionalization Although confessional compromise and intercommunal cooperation, emphasized by Ta’if, are meant as temporary measures until such time as deconfessionalization is effected, no steps have been taken in this direction. In the meantime, Ta’if has further institutionalized confessionalism in a new guise. Sectarian balance and equal sectarian representation have replaced erstwhile Maronite hegemony. This has produced a “three-man show” consisting of the three “Presidents”: the President of the Republic, the “President” of the Council of Ministers, and the “President”’ of Parliament.74 Whereas Ta’if sought to replace the rule of individuals (as exemplified in the powers of the president of the republic) by the rule of institutions, in practice the country was ruled for much of the 1990s by a “troika.”75 Moreover, each of these three individuals still perceives himself as the representative of his community, engaging in attempts at strengthening his position and enhancing his power vis-à-vis the others. This has resulted in frequent paralysis of decisionmaking. It is unclear how individuals committed to the perpetuation of a particular system of representation would simultaneously work to eradicate this system. Far from working in this direction, Lebanese leaders have attempted to use customary practice to challenge the provisions of the new constitution. For instance, President Hrawi used to insist on attending all the meetings of the Council of Ministers to assert his control over the executive branch.76 More recently, House Speaker Nabih Birri’s proposal to convene a broad national front was interpreted in Beirut as an attempt to counter

S-19

11/24/09

4:26 PM

Page 587

LEBANON’S TA’IF AGREEMENT

587

what Birri perceived as an alliance between President Emile Lahoud and Prime Minister Rafiq al-Hariri and to secure a say on several domestic political issues.77 This state of affairs has been instrumental in drawing the Syrian authorities deeper into domestic politics in Lebanon as Damascus was called upon repeatedly to act as an arbiter between the three “Presidents.” Ironically, the Lebanese political elite was—wittingly or unwittingly—helping Syria by delaying the implementation of the provision that would have strengthened the Lebanese request for the redeployment and ultimate withdrawal of the Syrian armed forces.

An Evaluation of the Second Republic The implementation of the Ta’if Agreement prioritized order and stability over reconciliation and justice. This was the logical consequence of events in the region. Syria’s unchallenged ascendancy in Lebanon following the Gulf War and the paralysis of the Syrian-Israeli track of the Middle East peace talks heightened the Syrian concern with Lebanon’s potential instability and the need to prevent Israel from using this weak neighbor to threaten Syria. This was also a function of the nature of the Ta’if Agreement, an agreement that consecrated the linkage between domestic Lebanese political issues and larger regional developments. Furthermore, Lebanese politicians remained faithful to the tradition of relying on external actors to gain advantage vis-à-vis domestic opponents. The road to Damascus thus became the key to a number of internal political deadlocks as Syrian officials were asked to arbitrate between members of the Lebanese troika. Could things have gone differently, and where do they stand now? The implementation of Ta’if would undoubtedly have looked very different had the regional situation not witnessed radical transformations. However, given the enthusiastic support of a number of European countries for General ‘Awn and his movement, and given Arab condemnation of Syria’s role in Lebanon in 1989, it is not clear that Ta’if would have survived the challenge posed by ‘Awn. The tacit decision to allow Syria to use muscled compliance in dealing with the primary spoiler of the Ta’if Agreement was a function of the changed regional setting. Moreover, the challenge posed by a number of militias to the crucial tasks of demobilization and disbanding would have also appeared more daunting in the absence of Syria’s military might. In these concluding remarks, I attempt a general assessment of the costs and benefits of the strategy of implementation adopted by Syria and its Lebanese allies. Although there have been clear short- to mediumterm costs in terms of civilian security, national reconciliation, and democracy, the muscled approach to compliance has provided fledgling state institutions with the requisite internal stability to recover their strength.

S-19

11/24/09

588

4:26 PM

Page 588

CASE STUDIES

The Dangers of Half-Measures The partial demobilization of former militias has served Syria and Lebanon in their dealings with Israel. Syria uses instability on the southern Lebanon front as a card in its deterrence dialogue with Israel. For the Lebanese authorities, the ongoing guerrilla warfare is a double-edged sword. On the one hand, Lebanon’s infrastructure has suffered heavily from Israeli retaliation for Hizballah attacks on Northern Galilee. On the other hand, in spite of Israel’s compliance with UN Security Council Resolution 425, Lebanon does not want to disavow its national resistance movement.78 But the exception granted Hizballah creates its own set of problems. Lebanon’s once most “demonized” militia has transformed into a legal party, “the most effective and efficient political party in the country.”79 Initially, Hizballah vowed to restrict its military activities to the South and the eastern Biqa‘ Valley only; however, party members used their military arsenal for nonresistance-related purposes on several occasions. In March 1993, Hizballah and the Amal Movement clashed in the southern suburbs of Beirut over differences on the manner in which resistance operations should be carried out. The Lebanese Army had to intervene to stop the fighting, warning that no such breaches of security would be tolerated and that no group responsible could expect to enjoy immunity. Beirut newspapers commented that the incident came as no surprise, since “excessive appeasement [of Hizballah] by the government has enabled Hizballah to move its resistance boundaries from the south to the capital.”80 Likewise, in summer 1997, Shaykh Subhi al-Tufayli, a member of Hizballah’s consultative council, threatened to lead a hunger march from Ba‘albak to Beirut to draw attention to the plight of the poor. Hizballah’s efficient and responsive organization had earned it wide social support, mainly in the Biqa‘ Valley and in southern Lebanon.81 The party was a relentless critic of the government’s social policies at a time when other critics such as trade unions saw their freedom of maneuver seriously curtailed by bans on demonstrations and other restrictive measures. Hence, when Tufayli issued a call for civil disobedience and took over Hizballah offices in Ba‘albak, the revolt was not only backed by a substantial number of inhabitants of the area, it was also backed by the might of an armed militia. The conflict escalated into a confrontation with the Lebanese Army in early 1998. Casualty reports indicated that at least fifty people were killed or wounded in the clashes. The army emerged victorious from the confrontation and its victory strengthened the moderate faction of Hizballah, which prefers to position itself as the “positive domestic opposition” in the Lebanese arena. This situation continues in spite of Israel’s decision to withdraw its troops from South Lebanon in July 2000. While the Lebanese-Israeli border witnessed relative quiet following the withdrawal, the Lebanese govern-

S-19

11/24/09

4:26 PM

Page 589

LEBANON’S TA’IF AGREEMENT

589

ment did not move to disarm Hizballah, nor did it agree to deploy troops at the border with Israel. Further, Lebanon disagrees with the UN assessment that Israel fully withdrew in accordance with UN Security Council Resolution 425. At stake are ten farms situated in the hamlet of Sheb‘a, which Beirut claims as a Lebanese territory and Israel claims as a Syrian territory outside the purview of Resolution 425. This disagreement and the military escalation inside the occupied Palestinian territories have provided Hizballah with justifications to continue launching mortar attacks against Israel. In April 2001, Israel retaliated directly against Syrian troops, bombarding a Syrian radar station in Lebanon. One year later, Israeli prime minister Ariel Sharon reiterated his warnings to Beirut and Damascus that Israel may take military action if the Lebanese and Syrian governments did not act to restrain the militia. These and previous incidents speak to the fact that until the regional dimension of the Lebanese crisis is solved—and with it the presence of Syrian troops on Lebanese soil—Lebanon’s de facto sovereignty remains hostage to wider regional developments. In spite of the transformation of militias into functioning political parties, the Lebanese armed forces do not have the monopoly on the use of force in the country. It must be reiterated, however, that the problem of partial sovereignty was inherent to the Ta’if Agreement, regardless of its manner of implementation. The implementation strategy only heightened this problem. State Institutions and the Political Space for Dissent In the short to medium term, the implementation of the Ta’if Agreement exacted a heavy toll on national reconciliation, civilian security, accountability, and legitimacy. The increasing marginalization of Christian political forces, the use and abuse of the armed forces and the judiciary to quell dissent, and the prioritization of order and security over justice and democracy have all been discussed above in detail. The choices of Ta’if’s internal and external implementers have cast doubt on the legitimacy of post-Ta’if political leaderships. This was most obvious in the results of the 1998 municipal elections, which yielded victories for opponents of the regime, notably Hizballah and pro-LF candidates. These elections, carried out for the first time in thirty-five years, “sketched out a new picture of the Lebanese domestic political situation.” 82 Hizballah candidates swept the polls in Mount Lebanon, Beirut, the South, and the Biqa‘ Valley. The party put up the strongest show of all parties running in the elections. One of the important observations on the margins of the results was “the rising star of the ‘Lebanese Forces’ which completely swept the municipal councils against the retreat of the movement supporting Michel ‘Awn”83 The startling result of the 1998 municipal elections can partially be attributed to the slow but steady strengthening of civil society throughout the period 1992–1998. The increasing repression of political dissent and the

S-19

11/24/09

4:26 PM

Page 590

590

CASE STUDIES

state’s infringement on the rights of free speech and association triggered a very strong reaction among a population traditionally used to and fiercely protective of its constitutional freedoms. A number of Lebanese NGOs and some parliamentarians have relentlessly worked to improve the human rights situation in Lebanon, including the Bar Association and the Parliamentary Committee for Human Rights. The Association of Journalists and the GCLW (the main labor union) have challenged governmentimposed restrictions at high personal cost to their members, many of whom were detained and brought before court in the process. Others have put pressure on the government to respect the rule of law, especially with respect to parliamentary and presidential elections. In March 1996 the Lebanese Association for the Democracy of Elections (LADE) was established to monitor the parliamentary elections scheduled for that summer.84 LADE’s establishment was rejected by the government. 85 However, the association was instrumental in denouncing the excesses that occurred in the course of the election campaign. LADE also spearheaded the public opinion campaign to prevent the postponement of the June 1998 municipal elections. Though stymied by a number of setbacks, civil society associations also scored several successes. When in 1996 several deputies and leading opposition figures decided to challenge the constitutionality of the electoral law, they were strengthened in their resolve by the activities of LADE and other NGOs. The Constitutional Court upheld the appeal and declared the law unconstitutional. However, the government chose to ignore the ruling, specifying that the contentious exceptions were “for one time only.”86 Looking to the Future This unexpected ruling brings us to an encouraging aspect of the state–civil society dialectic. During the 1990s, the responsiveness of state institutions to popular pressure for accountability has increased proportionally to the growing strength of the state apparatus. A number of developments indicate a potential for midcourse change. In August 1997, the government granted additional licenses to television and radio stations, thus relaxing the stringent application of the 1994 law regulating the audiovisual media. In December 1998, two months after the accession of army commander Emile Lahoud to the presidency of the republic, the government of Prime Minister Salim al-Huss decided to permit authorized demonstrations banned since 1993, and to review the audiovisual media law. Accountability was the hallmark of President Lahoud’s first year in office as a number of high-ranking officials in the former administration were the focus of judicial inquiries into the squandering of public funds and improper use of state power. The inquiries, dubbed “Operation Clean Hands,” were part and parcel of the president’s solemn commitment during his inaugural address to establish a

S-19

11/24/09

4:26 PM

Page 591

LEBANON’S TA’IF AGREEMENT

591

state of laws and institutions.87 While this commitment has only been partially fulfilled, it is important to note that consecutive governments have also publicly acknowledged the need to strengthen public freedoms, deepen democracy, and limit transgressions of liberties.88 The Committee to Protect Journalists still considers that the freedom and independence of the Lebanese press have yet to return; however, it acknowledges that “overall, . . . few journalists have been arrested or prosecuted since President Emile Lahoud took office in 1998.”89 And while opposition members are still subjected to arrest, imprisonment, and other harassment, it must be recognized that public criticism of the government and of the Syrian presence in Lebanon reached new heights in 2000, including an open letter from the editor of Lebanon’s premier Arabic-language daily, An-Nahar, to the newly elected president of Syria, Bashar al-Asad, urging him to recalibrate Syrian-Lebanese relations, as well as a similar appeal issued by the Maronite Council of Bishops. The widely criticized September 2001 crackdown on anti-Syrian activists nevertheless underscores the limits of change.

Conclusion I have argued in this chapter that the implementation of the Ta’if Agreement left much to be desired in the areas of national reconciliation, justice, accountability, and democracy. These shortcomings can be blamed on the external and internal implementers. For Syria, order and stability in Lebanon were paramount, hence the muscled approach to compliance that became the hallmark of Ta’if’s implementation. In turn, Syria’s Lebanese allies often resorted to political expediency in the search for power. Instead of displaying statesmanship, many of the postwar Lebanese political leaders resorted to the time-honored tradition of dragging their external allies into internal squabbles. However, the prioritization of order and stability was also instrumental in dealing with potential spoilers and giving state institutions the space needed to regain strength and authority. There have been indications that a strong and secure state is more likely to be open to political dissent and responsive to demands for accountability. However, the fact remains that the Ta’if Agreement, by explicitly linking the external and internal facets of Lebanese politics, has limited the universe of possible implementation outcomes. Lebanon’s full recovery depends on a solution to the Arab-Israeli conflict. Without such a solution, the contentious issues of Israel and Syria’s military presence on Lebanese territory are not likely to be resolved. Israel’s withdrawal from Lebanon is a requisite for Hizballah’s disarmament and full conversion into a political party. In turn, the withdrawal of the Syrian forces and the recalibration of the Lebanese political system are necessary to bring the Christian opposi-

S-19

11/24/09

4:26 PM

592

Page 592

CASE STUDIES

tion back into the fold. There are hopeful signs that Lebanon may finally acquire the rule of law and the institutions needed to turn the page on the civil war. However, the impact of recent flare-ups in the conflict with Israel underscores the fact that internal developments remain hostage to the larger regional context.

Notes 1. The agreement’s official name is the Document of National Reconciliation. 2. The National Pact consists of three guiding principles: Muslim-Christian sociocultural coexistence, minority guarantees (segmental proportionality on the basis of a six-to-five Christian/Muslim ratio in Parliament, and segmental autonomy allowing the religious communities to conduct their own internal affairs), and accommodation of the communities’ diverging notions of identity. Lebanon’s population is composed of seventeen different sects (also called “confessions,” or religious communities). The latest official census (1932) placed the Maronite Christian community at 32 percent of the population. No census has been conducted since. 3. See Albert Hourani, “Visions of Lebanon,” in Halim Barakat, ed., Toward a Viable Lebanon (London: Croom Helm; Washington, D.C.: Center for Contemporary Arab Studies, Georgetown University, 1988), pp. 3–10. 4. See Samir Khalaf, “From a Geography of Fear to a Culture of Tolerance: Reflections on Protracted Strife and the Restoration of Civility in Lebanon,” in Paul Salem, ed., Conflict Resolution in the Arab World: Selected Essays (Beirut: American University of Beirut, 1997), pp. 354–383; and Edward Azar, “Lebanon: The Role of External Forces in Confessional Pluralism,” in Peter Chelkowski and Robert Pranger, eds., Ideology and Power in the Middle East: Studies in Honor of George Lenczowski (Durham, N.C.: Duke University Press, 1988). 5. “A close examination of the 1975–76 ‘civil war’ will reveal that then—as at every time there has been a civil or religious war in Lebanon—it ‘coincided’ with at least one regional problem for which the internal socio-economic or political crisis in Lebanon served as vehicle.” See Ghassan Tuéni, “Lebanon: A New Republic?” Foreign Affairs 61, no. 1 (fall 1982): 86. 6. The Lebanese National Movement is an umbrella organization of all the opposition parties during the 1958 crisis, headed by Kamal Jumblatt, leader of the Progressive Socialist Party and a prominent Druze feudal leader. 7. The PLO was to coordinate activities with the Lebanese Army. Some territorial restrictions were placed on the guerrillas. In exchange, the Lebanese government recognized the legitimacy of a Palestinian armed presence in Lebanon and the PLO’s right to establish autonomous institutions in the camps. 8. Al-Kata’ib al-Lubnaniyya, or the Lebanese Phalanges, was the largest main Maronite political party in Lebanon and one of the most outspoken proponents of the Maronite vision of Lebanon. 9. The Lebanese Front included political forces associated with Maronite traditional political families, the Jumayyils, the Sham‘uns, and the Franjiyyas, in addition to the militias of Maronite religious orders. Pierre Jumayyil headed the Kata’ib party. Former President Camille Sham‘un led the National Liberal Party. Thenpresident Suleyman Franjiyyah led the Marada. 10. Although carried out on a smaller scale than in the former Yugoslavia, ethnic cleansing was practiced in Lebanon. In 1975, the village of Damur, south of

S-19

11/24/09

4:26 PM

Page 593

LEBANON’S TA’IF AGREEMENT

593

Beirut, was the site of one of the earliest confessionally based massacres. The most sadly famous massacres of civilians based on their communal belonging are the massacres of Palestinians in the refugee camps of Sabra and Shatila in the wake of the 1982 Israeli invasion. 11. The 1983 “battle of the mountain” purged the Shuf from Christian presence while the 1984 battles of the Iqlim al-Kharrub and Eastern Sidon resulted in the displacement of the majority of the Christian population of these areas. 12. The Riyadh Conference was held in Saudi Arabia in October 1976. It sought to find a solution to the Lebanese civil war. In January 1977, a 30,000-strong Arab Deterrent Force was positioned in Lebanon, of which 27,000 were Syrian forces. 13. The operation, “Peace for Galilee,” actually followed a PLO self-imposed six-month moratorium on guerrilla attacks into Israel. 14. For an overview of the various attempts at conflict resolution, see MaryJane Deeb and Marius Deeb, “Internal Negotiations in a Centralist Conflict: Lebanon,” in I. William Zartman, ed., Elusive Peace: Negotiating an End to Civil Wars (Washington, D.C.: Brookings Institution, 1995), pp. 125–146. 15. The six-member committee of the Arab League established the Tripartite Committee, comprising the kings of Morocco and Saudi Arabia and the president of Algeria. Effectively, their respective foreign ministers carried out most of the negotiations. 16. Although the Druze militiamen of the Progressive Socialist Party fought pitched battles against ‘Awn’s forces, most of the other Lebanese militias allied with the Syrian forces participated only tangentially in the military operations. 17. Joseph Maila, “The Ta’if Accord: An Evaluation,” in Deirdre Collings, ed., Peace for Lebanon? From War to Reconstruction (Boulder, Colo.: Lynne Rienner, 1994), p. 37. 18. Christians argue that the extensive powers of the president are a security guarantee for the Maronite community and for Christians at large against being engulfed in the Arab World. The prerogatives are considered by many Maronites as an essential political tool to implement their “vision” of Lebanon. See Iliya Harik, “The Maronites and the Future of Lebanon: A Case of Communal Conflict,” in Steven Dorr and Neysa Slater, eds., Security Perspectives and Policies: Lebanon, Syria, Israel, and the Palestinians (Washington, D.C.: Defense Academic Research Support Program, 1991), pp. 45–56. 19. Safeguards were built in to prevent any community from establishing de facto control of the executive branch. Ta’if stipulated that a majority of two-thirds of government was needed to vote on major issues including the state of emergency, war and peace, general mobilization, international treaties, the budget, comprehensive development and long-range plans, the dissolution of Parliament, the election law, the naturalization law, and personal statutes’ law. 20. Ta’if sets the schedule of redeployment as follows: “A specified time period not to exceed two years beginning after the ratification of the Document of National Reconciliation, the election of the President of the Republic, the formation of the Government of National Reconciliation, and the incorporation of the political reforms into the Constitution.” 21. Paul Salem, “Documents: The Ta’if Agreement—Annotated Text,” The Beirut Review 1, no. 1 (spring 1991): 171. 22. Thomas Collelo, ed., Country Report: Lebanon (Washington, D.C.: U.S. Department of State, Bureau of Public Affairs; Library of Congress, Federal Research Division, December 1987). 23. The Bureau of the Displaced is an organization dealing primarily with persons displaced from Mount Lebanon. See Judith Harik, “The Effects of Protracted

S-19

11/24/09

594

4:26 PM

Page 594

CASE STUDIES

Social Conflict on Resettlement of the Displaced in Postwar Lebanon,” in Salem, Conflict Resolution in the Arab World, p. 390. 24. The data in support of this contention are taken from Nasser Saidi, “The Economic Reconstruction of Lebanon: War, Peace, and Modernization,” in Collings, Peace for Lebanon? p. 199. 25. For more details about the contents of these reconstruction programs, see Ahmed Sbaiti, “Reflections on Lebanon’s Reconstruction,” in Collings, Peace for Lebanon? pp. 166–170. 26. The committee issued a communiqué to this effect on October 24, 1989. 27. In September 1989, ‘Awn had publicly accepted the seven-point agreement that included provisions for the holding of peace negotiations. 28. George Sa’adah, Qissati ma’ al-Ta’if [My story with Ta’if] (Beirut: Matabi’ al-Karim al-Haditha, 1988), p. 121. 29. For a good discussion of the ‘Awn phenomenon, see Paul Salem, “Two Years of Living Dangerously: General Awn and the Precarious Rise of Lebanon’s Second Republic,” The Beirut Review 1, no. 1 (spring 1991): 62–87. 30. Deeb and Deeb, “Internal Negotiations in a Centralist Conflict,” p. 138. 31. Mu‘awwad was murdered within weeks of his election in a booby-trapped bomb attack on his motorcade, on Independence Day, November 22, 1989. 32. UN estimates put the losses from the fighting at 1,500 killed, 3,500 wounded, 25,000 houses damaged, and 300 industrial establishments damaged. See Annie Laurent, “A War Between Brothers: The Army-Lebanese Forces Showdown in East Beirut,” The Beirut Review 1, no. 1 (spring 1991): 88–101. 33. ‘Awn took refuge at the French embassy and then went into exile. 34. Salem, “Two Years of Living Dangerously,” p. 74. 35. Judith Palmer Harik, “Democracy (Again) Derailed: Lebanon’s Ta’if Paradox,” in Bahgat Korany, Rex Brynen, and Paul Noble, eds., Political Liberalization and Democratization in the Arab World, vol. 2, Comparative Experiences (Boulder, Colo.: Lynne Rienner, 1998), p. 140. Theories abound in Lebanon about an undisclosed agreement between the United States and Syria that secured Syrian support for Desert Storm in return for giving Damascus free rein in Lebanon. 36. Volker Perthes, “Syrian Predominance in Lebanon,” in Rosemary Hollis and Nadim Shehadi, eds., Lebanon on Hold: Implications for Middle East Peace (Oxford: Royal Institute of International Affairs and the Centre for Lebanese Studies, 1996), p. 33. 37. Eliyahu Kanovsky, “The Middle East Economies: The Impact of Domestic and International Politics,” Middle East Review of International Affairs 1, no. 2 (July 1997): 16. 38. Wall Street Journal, July 19, 1995. 39. Kanovsky, “The Middle East Economies,” p. 16. 40. In a declaration to the Lebanese Arab-language daily As-Safir on November 17, 1992, Syrian vice president ‘Abd al-Halim Khaddam made the withdrawal of the Syrian forces conditional on complete deconfessionalization of Lebanese politics, thus postponing this issue indefinitely. 41. See Perthes, “Syrian Predominance in Lebanon.” 42. Hani A. Faris, “The Failure of Peacemaking in Lebanon, 1975–1989,” in Collings, Peace for Lebanon? p. 28. 43. The Lebanese Forces, for example, sold some of their equipment to the Bosnian Serbs. 44. There are no available figures to estimate the number of inductees into the army and the police forces.

S-19

11/24/09

4:26 PM

Page 595

LEBANON’S TA’IF AGREEMENT

595

45. Juan Carlos Gumucio, “Lebanon Orders Militias to Disarm,” The Times (London), March 29, 1991. 46. Alan Cowell, “Syria and Iran Agree Militias Can Remain in Parts of Lebanon,” New York Times, April 30, 1991, p. 6. 47. Ihsan Hijazi, “Lebanese Cabinet Tells P.L.O. Guerrillas to Disarm,” New York Times, April 4, 1991, p. 13. 48. “Lebanese Troops Flush Out Rebels,” New York Times, July 10, 1991, p. 6. 49. David Rudge, “Syrian Threats Force PLO to Hand Over Heavy Arms,” Jerusalem Post, July 12, 1991. 50. See Rex Brynen, “Palestinian-Lebanese Relations: A Political Analysis,” in Collings, Peace for Lebanon? pp. 84–87. 51. See Paul Salem, “The Wounded Republic: Lebanon’s Struggle for Recovery,” Arab Studies Quarterly 16, no. 4 (fall 1994): 56. 52. For a detailed analysis of the background, context, and results of the 1992 elections, see Joseph Bahout, “Liban: Les Élections législatives de l’été 1992,” Monde Arabe-Maghreb Machrek no. 139 (January–March 1993): 53–81. 53. Augustus Richard Norton and Jillian Schwedler, “Swiss Soldiers, Ta’if Clocks, and Early Elections: Toward a Happy Ending?” in Collings, Peace for Lebanon? p. 50. 54. Al-Sharq al-Awsat, September 18, 1996, cited in Hilal Khashan, “Lebanon’s 1996 Controversial Parliamentary Elections,” Journal of South Asian and Middle Eastern Studies 20, no. 4 (summer 1997): 40. The article provides ample evidence of the procedural irregularities during parliamentary elections. 55. See Gary Gambill and Daniel Nassif, “Lebanon’s Parliamentary Elections: Manufacturing Dissent,” Middle East Intelligence Bulletin 2, no. 8 (September 2000), www.meib.org/articles/0009_11.htm. 56. See Amnesty International, Lebanon: Human Rights Developments and Violations (October 1997), p. 8, www.amnesty.org/ailib/aipub/1997/mde/51801997. htm. 57. Ibid., pp. 19–22. 58. Ibid. 59. “Reports suggested that the GCLW and the army command had coordinated their actions in order to prevent a confrontation. The brother of the secretarygeneral of the GCLW, Elias Abu Rizq, is a senior army officer.” See “Playing the Army Against the Labor,” Lebanon Report 4, no. 1 (spring 1996), www.lcpslebanon.org/pub/tlr/96/sp96/army.html. 60. Amnesty International, Lebanon, pp. 16–17. 61. Ibid., pp. 19–22. 62. Robert Fisk, “Beirut Newspaper Defies Closure,” The Independent, May 13, 1993, p. 12. 63. Future Television is owned by Rafiq al-Hariri. Parliament speaker Nabih Birri owns the National Broadcasting Network (NBN). It had not yet been launched when the decree was passed. The brother of interior minister Michel al-Murr owns Murr Television (MTV). The Lebanese Broadcasting Corporation International (LBCI) has several shareholders, including deputies ‘Issam Faris and Sulayman Franjiyya. 64. “The Media Law: Switching Off the Competition,” Lebanon Report no. 3 (fall 1996), www.lcps-lebanon.org/pub/tlr/96/all65/media_law.html. 65. The most prominent case is that of An-Nahar staff writer Pierre Atallah, sentenced in absentia to three years in jail for entry into enemy land (Israel) without permission, and for contempt of Lebanese security and judicial authorities. 66. It is interesting to note that the issue of displaced persons is one of the

S-19

11/24/09

596

4:26 PM

Page 596

CASE STUDIES

least discussed aspects of the Ta’if Agreement. This academic neglect parallels the second-tier status that the issue of resettlement was accorded in the text of the agreement. One of the best assessments and the source for much of the detail in this discussion is Judith Harik, “The Effects of Protracted Social Conflict.” 67. These incidents did not result in a significant number of casualties. Their significance lay in their intimidation value and psychological impact. 68. Most of the displacement of Christians from the Shuf (which represents the bulk of displaced persons) occurred at the hands of the Progressive Socialist Party, led by Walid Jumblatt. This was one of the most violent episodes of interLebanese ethnic cleansing during the civil war. 69. For details of the extent of damage, see Sbaiti, “Reflections on Lebanon’s Reconstruction,” p. 163. 70. World Bank Group, Country Report: Lebanon (1998), p. 1. 71. For evaluations of the reconstruction scheme, see Volker Perthes, “Myths and Money: Four Years of Hariri and Lebanon’s Preparation for a New Middle East,” Middle East Report no. 207 (spring 1998): 30–32; and Michael Young, “Two Faces of Janus: Post-War Lebanon and its Reconstruction,” Middle East Report no. 209 (winter 1998): 4–7. 72. One-quarter of the Lebanese population lives under the poverty line. See Antoine Haddad, “The Poor in Lebanon,” Lebanon Report 3 (summer 1996): 36–42. For an evaluation of the Hariri years, see Perthes, “Myths and Money.” 73. There are no figures to this effect, but unemployment is still high in Lebanon and violent petty crimes have increased in the years immediately following the Ta’if Agreement. 74. The titles of “Prime Minister” and “House Speaker” respectively translate from Arabic into English as “President of the Council of Ministers” and “President of Parliament.” 75. The word troika has actually made its way into the Lebanese political lexicon. This argument was first expounded by Hassan Krayem, “The Lebanese Civil War and the Ta’if Agreement,” in Salem, Conflict Resolution in the Arab World, pp. 426–427. 76. Albert Mansour, Al-inqilab ‘ala al-Ta’if [The Coup Against Ta’if] (Beirut: Dar al-Jadid, 1993), pp. 204–207. 77. Adib Abi-Aql, “Birri’s Proposal for Broad National Front to Direct Political Trends,” al-Anwar (Beirut), June 29, 2001, cited by the Foreign Broadcast Information Service, Near East Section, July 1, 2001. 78. Hizballah has increasingly become a genuine national resistance movement, as witnessed by the growing support of all Lebanese, regardless of their religious belonging, for resistance against Israel during the incidents of April 1996 and June 1999. 79. Augustus Richard Norton, “Hizballah: From Radicalism to Pragmatism,” Middle East Policy 5, no. 4 (January 1998): 148. 80. An-Nahar newspaper cited in David Rudge, “Lebanese Army Rounds Up Hizbullah, Amal Fighters in South Beirut,” Jerusalem Post, March 9, 1993. 81. Following the Israeli “Grapes of Wrath” onslaught in April 1996, Hizballah “claims to have repaired 5,000 Lebanese homes, rebuilt many roads, and paid compensation to 2,300 farmers.” See “Lebanon, Hizbullah [sic] in Politics,” The Economist, September 7, 1996. 82. Fadhilu Hadaya, “Municipal Elections in Lebanon: The Government and the Opposition Fall; Hizbullah to Executive Authority,” Al-Majallah, June 27, 1998, p. 24, cited in FBIS-NES-98-183. 83. Ibid.

S-19

11/24/09

4:26 PM

Page 597

LEBANON’S TA’IF AGREEMENT

597

84. Under Lebanese law, an association is merely required to inform the government of its establishment before beginning to operate. However, in recent years, the Lebanese authorities have reinterpreted the law on associations, allocating themselves the right to either accept or reject notifications of establishment. 85. For an overview of the relationship between LADE and the government, see “Observing Elections,” Lebanon Report 1 (spring 1996), www.lcpslebanon.org/pub/tlr/96/sp96/elections.html; and “Election Monitor: Getting No Recognition,” Lebanon Report 3 (summer 1996), www.lcps-lebanon.org/pub/tlr/96/ sum96/election_monitor.html. 86. “The Parliament of Yes-Men? Lebanon’s Supreme Soviet,” Lebanon Report 3 (fall 1996), www.lcps-lebanon.org/pub/tlr/96/fall96/yes_men.html. 87. Nadim el-Hachem, “Enjoying the Backing of the Administration, the ‘Judicial Machine’ Strikes Unsparingly and More Strongly . . . An Armenian MP Calls for Hariri to Be Questioned,” Monday Morning no. 1374, April 26, 1999. 88. See the government’s policy statement read by Prime Minister Rafiq alHariri in front of the Lebanese Parliament on November 2, 2000. Radio Lebanon (Beirut), as cited by the Foreign Broadcast Infformation Service, Near East Section, November 2, 2000. 89. See www.cpi.org/attacks00/mideast00/lebanon.html.

S-19

11/24/09

4:26 PM

Page 598

S-20

11/24/09

4:27 PM

Page 599

20 Liberia: A Warlord’s Peace ADEKEYE ADEBAJO

Liberians overwhelmingly elected Charles Taylor as their president in July 1997, ending a seven-year civil war that had claimed an estimated 200,000 lives.1 In the course of the war, Liberia’s warlords signed fourteen peace agreements. The Abuja II Agreement, which led to the 1997 elections, was the only accord to be implemented. The main external implementer of the various agreements, the Economic Community of West African States Cease-Fire Monitoring Group (ECOMOG), strove for seven years to win the compliance of the myriad warring factions in Liberia. As the first subregional, and indeed first multilateral, attempt to use military muscle to enforce peace in the post–Cold War era, the ECOMOG intervention in Liberia demands close scrutiny. Three questions are of particular significance to this case: What explains the failure of thirteen peace agreements to end Liberia’s war? What accounts for the ultimate success of ECOMOG in bringing the war to a close? How durable is the peace that emerged in 1997? We argue in this chapter that three major obstacles hampered ECOMOG’s efforts at achieving peace. First, the implementation environment in Liberia was exceptionally challenging. When ECOMOG intervened in 1990, the Liberian state had collapsed, warring factions were splintering and proliferating, and the most powerful faction, the National Patriotic Front of Liberia (NPFL), opposed ECOMOG’s intervention, condemning it as a Nigerian-led attempt to deny the NPFL the political fruits of its military labors. In response to NPFL leader Charles Taylor’s armed resistance, ECOMOG initially chose a coercive strategy and attempted to force him to make peace. This strategy failed, however, due to the second constraint on ECOMOG: its political masters, the Economic Community of West African States (ECOWAS), were hopelessly divided over the inter599

S-20

11/24/09

600

4:27 PM

Page 600

CASE STUDIES

vention.2 The NPFL was supported by several members of ECOWAS, who opposed any attempt by ECOMOG to enforce peace, while several ECOMOG states supported various anti-NPFL factions. Finally, the reduced strategic value of Africa after the end of the Cold War, coupled with international opposition to military rule in Nigeria (the key contributor to ECOMOG), resulted in the international community, particularly the United States, failing to provide the necessary logistical and financial support for the intervention. The war in Liberia ended because of the removal of these three obstacles to peace. Eventually, the growing strength of Taylor’s opponents eroded his territorial dominance and provided him with an incentive to regard a peace agreement as a means of winning power. ECOMOG changed its strategy and offered concessions to Taylor and other Liberian warlords as a means of inducing their compliance. ECOWAS itself grew more united over time on the necessity of bringing the war to an end. The participation of the francophone members of ECOWAS, the Organization of African Unity (OAU), and the United Nations in the peacekeeping mission led key international actors like the United States and the European Union (EU) to rethink their positions on the war and finally to commit logistical and financial resources to assist ECOMOG’s efforts to end the war. After providing a brief background to the Liberian civil war, this chapter divides the peace process in Liberia into three distinct phases. The first phase, between May 1990 and August 1994, saw ECOMOG unsure of its mandate and switching uneasily between keeping and enforcing peace while simultaneously attempting to overcome subregional disunity and negotiating peace. Enforcement actions taken against the NPFL and collaboration with two NPFL foes, the Armed Forces of Liberia (AFL) and the United Liberation Movement of Liberia for Democracy (ULIMO), compromised ECOMOG’s neutrality during this phase. This deprived the force of the cooperation of the NPFL, hindered subregional consensus, and deprived the peacekeepers of substantial external support. During the second phase of the peace process, between September 1994 and September 1996, Charles Taylor negotiated more seriously with ECOMOG as he lost control of Liberian territory to his rivals and as a regime more conciliatory to Taylor took power in Nigeria. During this period, the UN and the OAU increased their involvement in the peace process as a way of calming the fears of the NPFL about ECOMOG’s lack of neutrality. The third phase, between September 1996 and July 1997, saw the restoration of ECOMOG’s impartiality and the provision of substantial international assistance to disarm Liberia’s fighters and end the war. The chapter concludes by considering the quality of the peace that has emerged in Liberia since 1997. Alarming trends concerning Taylor’s politicization of the security forces, the use of repressive measures against political opponents and government critics, continuing appeals to ethnic divi-

S-20

11/24/09

4:27 PM

Page 601

LIBERIA

601

sions, and the emergence of a growing rebel movement that controls parts of the country, all raise serious doubts about prospects for a durable peace in Liberia.

The Causes of the Liberian Civil War Liberia, a republic of 2.5 million inhabitants on West Africa’s coast, was founded in 1847 by freed black American slaves. Although only 5 percent of the population, this coastal settler elite established an oligarchy that excluded and oppressed the indigenous inhabitants. Belated attempts at reform came under the presidencies of William Tubman and William Tolbert (1944–1980), who increased educational opportunities and government positions for indigenous “country” people. Such reforms proved too little, too late, and in April 1980 a young, semiliterate master-sergeant, Samuel Doe, led a bloody coup d’état that put indigenous people in power for the first time in Liberia’s history. Doe filled strategic positions of the army and cabinet with members of his Krahn ethnic group, themselves only 5 percent of Liberia’s population. He established a political alliance with the Mandingos, whose members he appointed to senior government positions and whose traders enjoyed government patronage. During the 1980s, Liberia witnessed widespread corruption and the assassination and jailing of political opponents, while the economy spiraled toward the brink of insolvency.3 Doe at first ruled by co-opting civilian politicians into government to compensate for the inexperience of his soldiers. He eventually removed troublesome politicians and soldiers from government, employed his army to suppress students, and cracked down on the media. Doe sought a veneer of legitimacy to his despotic rule through a blatantly rigged, U.S.-condoned election in October 1985. Even after this electoral charade, he continued to maintain law and order by banning, harassing, and punishing opposition to his rule. On November 12, 1985, General Thomas Quiwonkpa, a Gio from Nimba county and former commander of Doe’s army, led a failed military coup. Doe executed Quiwonkpa and his fellow putschists and purged the army of Gios. His Krahn-dominated soldiers then went on a rampage in Nimba county, burning villages and indiscriminately killing a reported 3,000 Gios and Manos, related ethnic groups that account for about 15 percent of Liberia’s population. 4 This episode, more than any other, set the stage for the exploitation of ethnic rivalries that would eventually culminate in Liberia’s civil war. On December 24, 1989, a band of 168 NPFL fighters crossed into Liberia’s Nimba county from Côte d’Ivoire in order to topple Doe’s regime. The rebels rallied support among disaffected Nimba citizens, building an ethnic army as they marched toward the capital of Monrovia. Ethnic scores were settled, as Gio and Mano NPFL fighters

S-20

11/24/09

602

4:27 PM

Page 602

CASE STUDIES

killed Krahns and Mandingos, and Krahn AFL soldiers killed Gios and Manos.5 Liberia’s civil war was mainly fought by eight factions. Charles Taylor, an Americo-Liberian, led the 12,500-strong NPFL. He had escaped from a Massachusetts prison in 1985, where he was being held pending possible extradition to Liberia to face charges of embezzling $900,000 while he served as chief of Liberia’s General Services Agency under Doe. The nucleus of the NPFL consisted of Gio soldiers and farmers from Nimba county, as well as Burkinabè, Gambian, and Sierra Leonean citizens. Two other splinter groups later emerged from the NPFL. The 6,000strong Independent National Patriotic Front of Liberia (INPFL) was founded by Prince Yeduo Johnson, who broke away from Taylor in July 1990. A violent and unpredictable warlord prone to public acts of cruelty, Johnson captured Doe in September 1990 and videotaped his grisly torture and murder. The INPFL ceased to exist after 1993 when ECOMOG banished Johnson to Nigeria and many of his fighters joined other factions. The NPFL Central Revolutionary Council (CRC) was led by Tom Woewiyu, Laveli Supuwood, and Sam Dokie, senior NPFL officials who broke with Taylor in 1994, luring hundreds of combatants away from the NPFL. The AFL, with about 7,000 soldiers, was the remnant of Doe’s Krahndominated army. Headed by General Hezekiah Bowen, the AFL was based in Monrovia for most of the war. ULIMO was founded in May 1991 in Sierra Leone by leading Krahn and Mandingo politicians who recruited the bulk of their fighters from the Liberian refugee population in Sierra Leone. ULIMO split into two factions in 1994, with Roosevelt Johnson, a Krahn, heading the 3,800-strong ULIMO-J, while Alhaji Kromah, a Mandingo, headed the 6,800-strong ULIMO-K. The Liberia Peace Council (LPC) emerged in 1993. It had about 2,500 fighters and was headed by George Boley, a prominent Krahn politician who had served as a minister under both Tolbert and Doe. The Lofa Defense Force (LDF) emerged in 1993. It had about 400 fighters and was headed by Francois Massaquoi, an indigene of Lofa county. Battles were often fought for control of areas rich in economic resources in a land blessed with diamonds, gold, timber, rubber, and iron ore. Taylor derived an estimated $75 million annually from these exports, including $10 million a month from a consortium of North American, Japanese, and European miners,6 as well as $300,000 a month from foreign timber firms.7 ULIMO-K was keen to restore the Mandingo’s diamondtrading links with Sierra Leone, from which it had been excluded in March 1991. ULIMO-J was involved in diamond mining in Bomi county, while the LPC exported rubber from Buchanan port.8 In 1995 alone, an estimated $300–500 million worth of diamonds and gold, $53 million worth of timber, and $27 million worth of rubber were exported to markets in Europe and Southeast Asia by Liberia’s warlords.9

S-20

11/24/09

4:27 PM

Page 603

LIBERIA

603

The legitimacy of the warlords was based on building ethnic support and fighting for control in areas where much of their fighters came from. Some populations supported the factions if only, in some cases, to secure protection from attacks launched by rival ethnic groups. Thus, Charles Taylor, although an Americo-Liberian, built his army around Gios and Manos in Nimba county; George Boley built his army among his fellow Krahns in the southeast; Alhaji Kromah controlled Mandingo-populated territory in the north; Francois Massaquoi recruited fighters among Lofa citizens in the northern county. But territory did not always conform neatly to ethnic coalitions: Roosevelt Johnson, a Krahn, fought for control of western territories, while Charles Taylor controlled over half of the country for much of the war. None of the faction leaders spelt out their ideological reasons for waging the war. None expressed any interest in fundamentally changing Liberia’s political and economic structures or in radically reordering Liberian society. All had vague, broad justifications for their struggle that were based on expediency. Taylor claimed to wage the war to remove Doe, but initially opposed holding a democratic election. Anti-NPFL factions claimed to fight for the democratic rights of all Liberians, but were essentially ad hoc ethnic armies led by individuals with dubious democratic credentials. Individual warlords concentrated power in their own hands and dominated decisionmaking in their factions, which were little more than temporary alliances to fulfill the political and economic agendas of their leaders. The war witnessed widespread human rights abuses and atrocities.10 Underfed and underpaid soldiers, many of them children, were often only nominally controlled by their leaders.

Friends and Foes In postindependence West Africa, institutions are notoriously weak and leaders notoriously strong, and a change of regime can radically and suddenly alter foreign policy alliances.11 An intricate network of personal relationships and shifting alliances often determined the policies of individual states toward the Liberian conflict. NPFL rebels entered Liberia through Côte d’Ivoire in 1989. They had received training in guerrilla warfare in Burkinabè and Libyan military camps, and both countries supplied most of Charles Taylor’s arms during the war. These actions stemmed from personal networks and relationships. Ivorian leader Félix Houphouet-Boigny had never forgiven Doe for executing A. Benedict Tolbert, the husband of his adopted daughter and son of the murdered Liberian president, during the bloody 1980 coup. During the Liberian civil war, the NPFL enjoyed access into Liberia through Côte d’Ivoire, and Ivorian-based commercial interests benefited from NPFL

S-20

11/24/09

604

4:27 PM

Page 604

CASE STUDIES

mineral and timber concessions.12 Burkina Faso lent several hundred of its soldiers to Taylor in the early stages of Liberia’s war.13 The Burkinabè leader, Blaise Compaoré, had obtained Taylor’s release from a Ghanaian jail a few years earlier and later introduced him to Libyan leader Muammar Qadaffi,14 who had his own reasons for wanting a change of regime in Monrovia. Doe had closed down the Libyan embassy in Liberia in 1981 and supported Washington’s anti-Libyan policies. Nigeria provided 80 percent of ECOMOG’s troops and 90 percent of its funding during much of the war, giving it the lead in determining ECOMOG’s policies. Abuja had several reasons for its involvement in Liberia. Some Nigerian citizens had been killed by NPFL fighters in Monrovia in early 1990 and Nigerians were held hostage by Taylor’s forces. Historically, Nigeria’s military leaders have been keen to portray their country as the indispensable power in West Africa.15 They were concerned about the impact of the civil war on the stability of the subregion and on ECOWAS, an economic integration scheme launched with strong Nigerian leadership in 1975. Nigeria remained involved in Liberia in part because its generals personally benefited from revenues written off as ECOMOG expenses,16 and because the ECOMOG mission helped Nigerian leader General Sani Abacha to ward off the threat of severe international sanctions against his regime. Several ECOWAS states also had their own interests in supporting ECOMOG. A few NPFL fighters had been involved in the unsuccessful Gambian coup of 1981. NPFL ally Burkina Faso sheltered Gambian dissidents, and the unsuccessful 1981 coup in Gambia was widely believed to have been sponsored by Libya, another strong supporter of the NPFL. Guinea and Sierra Leone, as two of the three states most affected by the spillover of refugees from the war, also had reason to support the ECOMOG intervention: by August 1990 Guinea hosted 225,000 refugees and Sierra Leone 69,000. Furthermore, the NPFL had Sierra Leonean dissidents within its ranks who vowed to destabilize the country. Guinean and Sierra Leonean soldiers fought alongside ULIMO against the NPFL during the war, while Alhaji Kromah, the ULIMO-K chief, enjoyed close political ties with Guinea. The United States, a Cold War ally and historical godfather of Liberia, was the only non-African power that could have provided an alternative to the ECOMOG intervention. Washington had provided $500 million to the brutal regime of Doe during the 1980s. But the Liberian war occurred at the end of the Cold War when Washington perceived little strategic interest in the country and was distracted by the annexation of Kuwait by Iraq. The United States sent military advisers to AFL positions in Nimba county at the start of the conflict, concerned about Libya’s links with the NPFL, but congressional complaints soon led to their speedy withdrawal. Thereafter,

S-20

11/24/09

4:27 PM

Page 605

LIBERIA

605

Washington limited itself to supporting the ECOWAS peace plan, and contributed $500 million in humanitarian assistance to Liberia during the civil war. The administration of President George Bush ordered its officials to avoid a high-profile role in Liberia.17 The United States merely evacuated its citizens and ignored widespread calls from Liberians and some of its European allies to stage a military intervention. We will next examine in detail the fourteen peace agreements of the Liberian civil war, starting with the nine accords reached between 1990 and 1994 and concluding with the five accords of 1994 to 1997. These sections will focus especially on the domestic, subregional, and external obstacles to implementing the peace agreements.

The Peace Agreements, May 1990–August 1994 Between May 1990 and August 1994, Liberia’s warring parties signed nine peace agreements and implemented none of them. Barriers to settlement included the unwillingness of Taylor to share power, the mutual suspicions of rival warlords and their fears of disarmament if others reneged on the deal, and significantly, the lucrative bounties from economic resources that were derived from areas under their control. Subregional actors remained divided on the best approach for achieving peace and were compromised by their support for rival Liberian factions. International donors failed to commit the resources needed to implement the agreements. ECOMOG, the implementing party, was robbed of some of its legitimacy at its creation by the dissent of key francophone states, particularly Côte d’Ivoire and Burkina Faso, but also Senegal, Niger, Mali, and Togo. ECOMOG’s legitimacy was also weakened by the refusal of the NPFL to recognize the peacekeepers’ authority to intervene militarily in Liberia. Even by ECOWAS’s own charter and defense protocols, ECOMOG was on shaky legal foundations, with no specific clauses allowing for military intervention in a member state’s internal conflict.18 The intervention was largely justified on humanitarian grounds as well as by citing the destabilizing effect of the civil war on the entire subregion.19 UN and OAU diplomatic support, and eventually military support, did serve to bolster ECOMOG’s international legitimacy, as did the addition of Senegalese and Malian military contingents. But the continued opposition of Burkina Faso and the NPFL to the peacekeepers, and ECOMOG’s collaboration with anti-NPFL factions, continued to deprive the intervention of universal legitimacy and support. ECOMOG’s efforts were stymied by noncompliance. The chief “spoiler” was Charles Taylor, who used violence to attack the peace process, from the entry of ECOMOG in 1990 to the invasion of Monrovia in 1992.

S-20

11/24/09

606

4:27 PM

Page 606

CASE STUDIES

ECOMOG initially attempted a coercive strategy, launching attacks against the NPFL in October 1990 and January 1993, blockading its ports, capturing its territory, and attempting to cut off its economic resources. But the peacekeepers simply lacked the means to prosecute a guerrilla war against the NPFL. Besides, successful coercion would have required not only military capability bolstered by external logistical support, but also subregional consensus and the willingness of Taylor’s allies to abandon him. Diplomatically divided, ECOMOG lacked the capacity and grew increasingly unwilling to confront the factions militarily. There were unsuccessful efforts to seal Liberia’s porous borders to prevent arms flows, but there were no attempts to stop European and Asian commercial firms from paying Liberian warlords for illicit exports of minerals and timber. Subregional efforts to secure an economic embargo at the UN Security Council were scuttled by France and several other states whose companies benefited from this trade.20 No penalties were spelled out for the failure of parties to implement peace agreements until ECOWAS leaders threatened economic sanctions against the NPFL in 1992. But ECOMOG obtained the support of neither Burkina Faso and Libya nor foreign firms to stop the flow of arms and resources to the factions. The peacekeepers reevaluated their implementation strategy by attempting to bridge subregional divisions and by including the UN and the OAU in the peace process to act as more impartial supporters of the agreements. Short-term confidence-building measures included the diversification of ECOMOG to include francophone contingents from Senegal and Mali, OAU peacekeepers from Uganda and Tanzania, and extraregional UN peacekeepers. But the role of the UN and the OAU was circumscribed by their limited involvement, and their participation did not go much beyond legitimizing ECOMOG’s actions, while their presence did not elicit the required cooperation of the factions. The establishment of a buffer zone on the border with Sierra Leone was intended to build confidence between the NPFL and ULIMO in order to encourage them to implement the Yamoussoukro Agreement. But the warlords still refused to make peace, Compaoré and Qadaffi still refused to stop arming Taylor, and the international community still refused to provide substantial support to ECOMOG. From Banjul to Lomé, May 1990–February 1991 As NPFL rebels marched toward Monrovia in May 1990, an ECOWAS summit in Banjul established a five-member Standing Mediation Committee (SMC) with a mandate to mediate the Liberian civil war. Nigeria, Ghana, Gambia, Mali, and Togo were elected as the first members, and they met in Banjul on August 6–7, 1990. At that time nearly 500,000 Liberian refugees had flooded into neighboring Guinea, Côte d’Ivoire, and

S-20

11/24/09

4:27 PM

Page 607

LIBERIA

607

Sierra Leone; at least 5,000 people had been killed in Liberia, and 3,000 Nigerian, Ghanaian, and Sierra Leonean citizens were being held hostage by the NPFL. The August meeting produced a plan to establish a peacekeeping force, ECOMOG, with a mandate to supervise a cease-fire and, following Doe’s resignation, to establish an interim government and organize elections after twelve months. None of the faction leaders would be eligible to join the interim government. A $50 million Special Emergency Fund was proposed to finance ECOMOG and to provide for the immediate humanitarian needs of Liberians. ECOWAS, the OAU, and other members of the international community were asked to contribute to the fund.21 The UN and the OAU pledged the support of their organizations to ECOMOG, but the two francophone members of the SMC, Mali and Togo, as well as Côte d’Ivoire, declined to contribute troops to ECOMOG. ECOMOG initially deployed 3,000 soldiers from Nigeria, Ghana, Guinea, Sierra Leone, and Gambia, who arrived in Monrovia on August 24, 1990. The force immediately confronted the key problem in its mandate: conceived as a peacekeeping operation, there was no peace to keep. ECOMOG came under immediate fire from the NPFL, and the peacekeepers concluded that they would need to use force to gain the NPFL’s compliance. Within a month, ECOMOG deployed 3,000 more troops, changed its mandate from peacekeeping to peace enforcement, and ordered its soldiers to establish a buffer zone to protect Monrovia from NPFL attacks. Under Nigerian general Joshua Dogonyaro, ECOMOG expelled the NPFL from the capital but compromised its stated neutrality by fighting alongside the INPFL and the AFL. The peacekeepers still lacked francophone support in November 1990 when the first extraordinary ECOWAS summit was held in the Malian capital of Bamako. Mali would soon announce its decision to contribute officers to ECOMOG in order to dilute the anglophone dominance of the force. ECOMOG’s decision to establish an Interim Government of National Unity (IGNU), headed by Amos Sawyer, the exiled head of the Association for Constitutional Democracy in Liberia (ACDL), caused further friction with the francophone states that had not given their consent, and Taylor dismissed the IGNU as a puppet of ECOMOG. Taylor attended the ECOWAS summit, where Compaoré convinced him to sign the Bamako agreement. Bamako called for a cease-fire, a cessation of arms purchases, the creation by ECOMOG of a permanent buffer zone to separate the belligerents, and the election of an interim government, followed by the disarmament of the factions by ECOMOG.22 At another SMC meeting in Togo on February 12–13, 1991, the NPFL, INPFL, and IGNU signed the Lomé Agreement to implement Bamako. A stalemate remained, however, as the NPFL insisted on a new interim government before disarmament, while the IGNU insisted on disarmament

S-20

11/24/09

608

4:27 PM

Page 608

CASE STUDIES

before the installation of another interim government. This issue would frustrate every peace conference during Liberia’s war. Yamoussoukro and After, June 1991–June 1993 Four rounds of negotiations took place in the Ivorian town of Yamoussoukro between June and October 1991 to resolve the continuing deadlock. The use of francophone capitals as mediation sites, the selection of a new Committee of Five that included Côte d’Ivoire, Senegal, and Togo, and the election of Senegal’s Abdou Diouf as ECOWAS chairman in July 1991 all confirmed the ascendancy of the Francophonie in subregional mediation. A division of labor seemed to be emerging within ECOWAS: the francophones took the lead in mediation and left military operations largely to the anglophones. But despite signs of growing subregional unity, the implementation of Yamoussoukro was further complicated by continuing subregional divisions and a lack of international support for ECOMOG. Ghana, unlike Nigeria, was prepared to tolerate a Taylor presidency and wanted its troops to engage strictly in peacekeeping functions and avoid a peace-enforcement role. Washington forcefully pushed the Yamoussoukro peace process while denying ECOMOG substantial military support, accusing the peacekeepers of supporting anti-NPFL factions. The UN and the OAU continued to lend moral rather than material support to the subregional effort. The main points of the Yamoussoukro Agreement included the deployment of ECOMOG to all parts of Liberia; the encampment and disarming of factions under ECOMOG’s supervision; the creation of a buffer zone on the Liberia–Sierra Leone border; the monitoring of all air- and seaports by ECOMOG to stem the flow of arms; and an expansion of ECOMOG, with Senegal promising to contribute troops to the mission. All this was expected to have been completed within two months. The agreement also called for the creation of an interim government, an Electoral Commission, and an ad hoc Supreme Court to oversee elections in six months.23 Having signed and broken two previous accords, Charles Taylor continued to act as a spoiler. He called for the introduction of UN troops into Liberia and a reduction of ECOMOG troops from 6,000 to 1,500. Taylor sought to increase the military costs of intervention to ECOMOG, while preventing his own total diplomatic isolation. Regarding militarily weak Sierra Leone as ECOMOG’s Achilles’ heel, he encouraged NPFL elements, led by a group of Sierra Leoneans dubbed the Revolutionary United Front (RUF), to enter Sierra Leone on March 23, 1991, to destablilize the government and expand his own economic base through revenues derived from Sierra Leonean diamonds.24 By taking the war to Sierra Leone, Taylor also hoped to splinter the ECOMOG coalition and force a withdrawal of that country from the war. He sought to render ECOMOG’s rationale for inter-

S-20

11/24/09

4:27 PM

Page 609

LIBERIA

609

vening in Liberia a sham by showing that the force could neither keep the peace nor prevent the war from spreading to the subregion. Although Taylor partly succeeded in some of his goals, he also suffered military reversals to ULIMO in western Liberia. He failed to convert his early military successes into diplomatic capital, and no government recognized his National Patriotic Reconstruction Assembly Government (NPRAG) in Gbarnga. Taylor refused to disarm until ULIMO had halted its advances, and his support for Yamoussoukro was largely based on his expectation that ECOMOG’s creation of a buffer zone between Liberia and Sierra Leone would ease ULIMO’s military pressure on the NPFL. The next ECOWAS conference on Liberia took place in Geneva on April 6–7, 1992. The meeting set new timetables for the deployment of ECOMOG and the completion of disarmament, and acceded to Taylor’s request to maintain a personal security unit of company strength. 25 Although officials had been appointed to Liberia’s Election Commission and Supreme Court, ECOMOG was unable to establish a buffer zone between Sierra Leone and Liberia due to continued fighting between the NPFL and ULIMO. As part of the implementation of Yamoussoukro, 1,500 Senegalese troops arrived in Liberia in early 1992. Washington contributed $15 million for the contingent’s logistical needs, while continuing to withhold substantial support from other ECOMOG contingents. Although Taylor had first suggested that Senegalese troops join an expanded ECOMOG in order to dilute its anglophone dominance, he complained that “the US is using the Senegalese for their surrogate activities in Liberia.”26 ECOMOG deployed its peacekeepers to the Liberian countryside but faced many difficulties: ULIMO incursions into NPFL areas forced ECOMOG to withdraw from the Liberia-Sierra Leone border zone; ECOMOG’s freedom of movement was restricted; and its soldiers’ heavy weapons were disallowed. On May 28, six Senegalese troops were killed by NPFL commandos in the Lofa village of Vahun in northern Liberia, and 500 ECOMOG peacekeepers were thereafter held hostage by the NPFL for one week. By the end of 1992 ULIMO had managed to gain control of about 20 percent of Liberian territory. Taylor responded by staging a military offensive to defeat ECOMOG or to inflict enough casualties to force its withdrawal from Liberia. On October 15, 1992, the NPFL launched “Operation Octopus,” a full-scale attack on Monrovia. The offensive surprised ECOMOG and exposed the weaknesses of its intelligence capabilities. It was not until a week later that ECOMOG counterattacked, bombing NPFL positions from the air and sea. Units of the AFL and ULIMO again fought alongside ECOMOG to defend Monrovia. By the time ECOMOG declared a unilateral cease-fire on November 10, 3,000 people had died as a result of the fighting.

S-20

11/24/09

610

4:27 PM

Page 610

CASE STUDIES

By the middle of January 1993, the Senegalese contingent withdrew from Liberia, citing pressing domestic problems.27 This resulted in the loss of a respected francophone state that had increased ECOMOG’s credibility as a subregional force. ECOMOG’s estimated strength was thereafter increased to 16,000 troops, of which 12,000 were Nigerian. In the next three months ECOMOG, under Nigerian general Adetunji Olurin, seized NPFL strongholds in the country. The loss of strategic NPFL assets was a serious blow to Charles Taylor, reducing his access to arms and cutting into his profits from iron ore, rubber, and timber exports. Elsewhere in Liberia, the proliferation of factions intensified. ULIMO seized control of most of northwest Liberia from a distracted NPFL, but lost several towns in Lofa county to the Lofa Defense Force. The Liberia Peace Council was also waging war against the NPFL for control of the southeast. Cotonou and After, July 1993–March 1994 The next efforts to end Liberia’s war occurred when the UN, in cooperation with ECOWAS and the OAU, sponsored new peace talks in Geneva from July 10 to 17, 1993, producing what became known as the Cotonou Agreement. ECOMOG was to expand its force to include UN and OAU peacekeepers; establish a cease-fire by August 1, 1993; set up a fivemember executive Council of State; put in place a Liberian National Transitional Government (LNTG) by the end of August 1993; hold presidential elections in seven months; and establish buffer zones on Liberia’s borders with Guinea, Côte d’Ivoire, and Sierra Leone. The most detailed peace agreement of the Liberian civil war, Cotonou specified ECOMOG’s and the UN’s supervisory and monitoring roles, the prohibitions on the parties, acts that constituted cease-fire violations, and the most far-reaching details on the structure and mandate of an interim government. A Joint Cease-Fire Monitoring Committee, to be chaired by the UN Observer Mission in Liberia (UNOMIL) and involving ECOMOG and representatives of the armed factions, was established to investigate and resolve cease-fire violations. The Cotonou Agreement explicitly gave ECOMOG peace enforcement powers with the approval of a UN-chaired Cease-Fire Violations Committee. With the signing of the Cotonou Agreement on July 25, 1993, ECOMOG had effectively bombed Charles Taylor to the negotiating table. The accord was signed barely two months after ECOMOG’s capture of the key NPFL port of Buchanan. NPFL control of Liberian territory had shrunk from 95 percent in 1990 to 50 percent, even as it faced continuing challenges from ULIMO, the LPC, and the LDF. Concomitant with the installation of a transitional government in Monrovia on March 7, 1994, ECOMOG started deploying its troops to the Liberian countryside, accompanied by 1,500 OAU troops from Tanzania

S-20

11/24/09

4:27 PM

Page 611

LIBERIA

611

and Uganda and 368 unarmed UN military observers. ECOMOG, however, had not received the logistical and financial support it had consistently requested from the international community and was deployed in less than 15 percent of the country.28 Continued fighting in the countryside hampered ECOMOG’s disarmament efforts and by August 1994, only 3,612 of an estimated 33,000 combatants had been disarmed. Several ambushes of OAU and ECOMOG soldiers by Liberia’s factions forced the UN to withdraw its observers to Monrovia and reduce UNOMIL’s peacekeepers from 368 to 90. Tensions between ECOMOG and UNOMIL further hampered implementation efforts. ECOMOG soldiers often complained that UNOMIL withheld its vehicles and helicopters from their use. UN personnel were paid much higher stipends than ECOMOG’s, fueling further envy.29 Four other disagreements emerged between the organizations. First, ECOMOG wanted UNOMIL to have a passive rather than a lead role in disarmament, observing rather than directing the process. Second, ECOMOG officials were angered by what they regarded as UN special representative Trevor Gordon-Somers’s unilateral disarmament negotiations with the parties. Third, in contrast to UNOMIL’s chief military observer, General Daniel Opande of Kenya, ECOMOG’s field commander, General John Inienger of Nigeria, preferred a hard-line approach toward disarming the NPFL. Finally, ECOMOG criticized UNOMIL for its failure to consult with it before deploying troops, while UNOMIL complained that ECOMOG was incapable of protecting its military and civilian personnel.30 Aside from these military difficulties, there were political obstacles to the implementation of the Cotonou Accord. Professor David Kpomakpor, the IGNU nominee, was installed as chairman of the Council of State in March 1994. The council also included representatives of the NPFL and ULIMO. But the LNTG was unable to extend its authority beyond Monrovia and remained entirely dependent on foreign donors for its financial survival. Council members sharply disagreed over appointments to public corporations and agencies, as the warlords struggled for their share of the limited spoils. As with previous peace agreements, Liberia’s warlords signed Cotonou for tactical reasons. Charles Taylor agreed not because of a sincere commitment to implementing it, but largely due to external political pressure from ECOWAS members and internal military pressure from ULIMO, the LPC, and the LDF, which had reduced his control of Liberian territory significantly. ECOMOG’s military offensive had also weakened his economic base. The IGNU signed the peace deal in the hope that it would bring an end to the war and result in a settlement that would give civilians a role in a new political order. Liberia’s politicians realized that most of their country lay under the control of the warlords and regarded the signing of peace agreements and the ECOMOG presence as necessary diplomatic and mili-

S-20

11/24/09

612

4:27 PM

Page 612

CASE STUDIES

tary tools to pressure the warlords to give up territory and agree to an interim government. As for the other Liberian factions, ULIMO, the LPC, and the LDF hoped to obtain a share of political power in a future government through the conquest of territory, which would then provide them with some leverage during negotiations. ULIMO’s presence at the Cotonou talks in 1993, after its exclusion from Yamoussoukro in 1991, was a clear sign to other factions that gaining territory was the most viable way of winning a place at the negotiating table. New factions had much to lose and nothing to gain from the successful implementation of Cotonou. It was in their interest that the agreement failed, as it had been in ULIMO’s interest that Yamoussoukro failed in 1991.

The Peace Agreements, September 1994–September 1996 From Akosombo to Accra, September 1994–January 1995 In September 1994, barely a month after becoming ECOWAS chairman, Ghanaian president Jerry Rawlings summoned Taylor, Kromah, and General Hezekiah Bowen, the leaders of Liberia’s three largest factions, to Akosombo, Ghana. The meeting produced an agreement calling for a new five-member Council of State, on which the three warlords would sit along with one representative jointly nominated by Taylor and Kromah, with the remaining seat going to a nominee of the civilian Liberian National Conference (LNC). This agreement marked the start of ECOMOG’s strategy of appeasing the warlords to win their compliance in recognition of the peacekeepers’ inability to enforce peace in Liberia. Akosombo gave joint responsibility for disarmament, monitoring of Liberia’s borders, and arms searches to ECOMOG, the UN, and the warlord-dominated LNTG. The transitional government was granted the right “to use the necessary force available to compel compliance . . . in collaboration with ECOMOG.” The agreement also gave the LNTG responsibility for restructuring a new Liberian army with the assistance of ECOWAS, the UN, and “friendly governments.” Akosombo was a warlord’s agreement. Charles Taylor had long argued that the leaders of the warring factions should be directly represented on the Council of State, and that the LNTG be allowed to supervise disarmament: Akosombo fulfilled both of these wishes. But Akosombo was an agreement that was stillborn. At the time of its signing, fighting was raging throughout Liberia, and rival factions had militarily expelled two of its three signatories from their headquarters: Taylor from Gbarnga and Kromah from Tubmanburg. The exclusion of the LPC, LDF, ULIMO-J, and

S-20

11/24/09

4:27 PM

Page 613

LIBERIA

613

CRC-NPFL from the Akosombo Agreement, despite their control of large swaths of Liberia, meant that the accord bore no resemblance to the military and political realities in Liberia and was doomed to unravel. Rawlings amended the agreement in response to mounting criticism from within Liberia and the subregion, especially from Nigeria. Whereas Ghana believed that if the warlords could run the government, disarmament would automatically follow, Nigeria believed that if the warlords could be convinced to disarm, an effective interim government would follow. ECOWAS foreign ministers met in Accra on November 22, 1994, to rectify the Akosombo Agreement. Three weeks of discussions in Accra led to agreement on a fivemember Council of State. Seats were to be divided between the NPFL, ULIMO-K, the coalition of anti-NPFL factions (AFL, CRC-NPFL, LDF, LPC, and ULIMO-J, founded in September 1994), the Liberian National Conference, and a fifth member elected by the electoral college of the LNTG. ECOMOG-protected “safe havens” were to be established in Gbarnga, Greenville, Harper, Tubmanburg, and Voinjama. The AFL and other security forces would be restructured, with ECOMOG and UN assistance, to include all factions. The warlords signed the Accra Agreement on December 21, 1994. It was never implemented. The cease-fire was consistently flouted as fighting raged in the southeast between the NPFL and the LPC, and in central Liberia between the NPFL and ULIMO-J. Nor could the amorphous anti-NPFL coalition agree on its candidate for the Council of State. The Road to Abuja, June–August 1995 On June 2, 1995, an unexpected breakthrough emerged between the Nigerian government and Charles Taylor that would prove decisive in ending Liberia’s civil war. Coaxed by Compaoré and Rawlings, the NPFL warlord spent four days with Nigerian leader General Sani Abacha in a rapprochement between ECOMOG’s dominant force and Liberia’s strongest faction. Abacha was less secure within the Nigerian army than was his predecessor, General Ibrahim Babangida, and was forced to focus more attention on domestic issues. He depersonalized the rivalry between Taylor and Babangida, and was more pragmatic in seeking a solution to the conflict. With Nigeria no longer opposing his ambitions and with the NPFL having lost territory to rival factions, Taylor was also more amenable to a solution. On August 15, 1995, Rawlings met with Abacha in Abuja. They decided to break the diplomatic impasse in Liberia by convincing the anti-NPFL coalition’s nominee, Tom Woewiyu, to relinquish his Council of State seat to LPC leader George Boley.31 All the warlords attended the meeting, as did representatives of the Liberia Women Initiative and the Inter-Faith

S-20

11/24/09

614

4:27 PM

Page 614

CASE STUDIES

Mediation Committee of Liberia, civil society groups that were fast losing influence over ECOMOG to Liberia’s warlords. In Abuja, the Nigerian foreign minister, Tom Ikimi, called for a Council of State that could provide strong and effective leadership and would be able to control the whole territory of Liberia.32 Nigeria was finally coming around to Ghana’s position that only the inclusion of the warlords in the government would bring peace to Liberia. During a meeting of ECOWAS leaders in Abuja on August 18–19, 1995, Taylor, Kromah, and Boley were nominated to be council vice chairmen and were asked to nominate individuals for council chairman.33 Along with the three powerful warlords, there were three civilians on the Council of State: its chairman, Wilton Sankawulo, a former literature professor, Chief Tamba Tailor, an aged traditional leader, and LNC representative Oscar Quiah. As only Quiah had intimate experience of Liberian politics, the civic groups feared the warlords would dominate the council. Ministerial posts were shared on an inclusive basis, with faction leaders Roosevelt Johnson, Hezekiah Bowen, Tom Woewiyu, and Francois Massaquoi obtaining cabinet posts. Positions to the Supreme Court, autonomous agencies, and public corporations were also divided among the armed factions and civilians. The Abuja Agreement was signed on August 19, 1995. It returned sole authority for military enforcement action to ECOMOG, and removed the role that Akosombo had assigned to the LNTG. A cease-fire was to come into effect on August 26, 1995. ECOMOG and UNOMIL would then deploy throughout the country. The disarmament and demobilization of the belligerents was expected to be completed by January 1996. Despite the failure of a dozen previous peace agreements, there was widespread optimism that the Abuja Agreement would finally end Liberia’s war. Nigeria and the NPFL had made peace, removing a major obstacle that had frustrated earlier implementation efforts. All the warlords were directly involved in the Council of State for the first time and were therefore thought to have a direct stake in keeping the peace. And there was warweariness among the 33,000 faction fighters, some of whom had started voluntarily disarming to ECOMOG even before the Abuja Agreement had been signed. The Making of Abuja II, September 1995–September 1996 The implementation of Abuja, however, like that of previous agreements, proved more difficult than its negotiation. On October 27, 1995, a donor conference on assistance to Liberia was held in New York. The conference sought support for humanitarian assistance, disarmament and demobilization, recovery and rehabilitation, and assistance to ECOMOG. However,

S-20

11/24/09

4:27 PM

Page 615

LIBERIA

615

donor fatigue and the low priority that Western policymakers accorded to Liberia ensured that only $145.7 million was pledged for Liberia’s reconstruction, even as the international community pledged $6 billion for the reconstruction of Bosnia.34 Rawlings had asked for $195 million for ECOMOG and UNOMIL’s disarmament and demobilization tasks alone.35 Two months after the donor conference, ECOMOG’s logistical shortcomings were again glaringly exposed when ULIMO-J combatants attacked its troops in the western Liberian town of Tubmanburg, killing sixteen Nigerian peacekeepers, wounding seventy-eight others, and capturing some of ECOMOG’s heavy weapons and other equipment. The warlords used the Council of State as a platform for the presidential campaign, and there were increasing complaints about Sankawulo’s weak and ineffectual leadership. Without an army or popular support base, he was constantly outmaneuvered by the warlords. The faction leaders also tried to increase their own authority by arguing, in contradiction of Abuja, that they were part of a collective executive presidency and that they were therefore responsible for disarming fighters and approving ECOMOG’s deployment plans.36 Such behavior eventually led Taylor and Kromah to attempt to exploit a split within the leadership of ULIMO-J by sending elements of the national police, backed by NPFL and ULIMO-K fighters, to arrest Roosevelt Johnson on April 6, 1996. Johnson refused to surrender, resulting in street battles between NPFL and ULIMO-K forces on the one hand and a Krahn alliance of ULIMO-J elements supported by the AFL and the LPC on the other.37 After two months of fighting in Monrovia, 3,000 people had died. In response to these events, OAU leaders threatened sanctions against Liberia’s recalcitrant warlords during their summit in Yaoundé, Cameroon, in July 1996. A resolution was passed warning that, “should the ECOWAS assessment of the Liberian peace process . . . turn out to be negative, the OAU will help sponsor a draft resolution in the UN Security Council for the imposition of severe sanctions . . . including the possibility of the setting up of a war crimes tribunal to try the leadership of the Liberian factions on the gross violations of the human rights of Liberians.”38 Although the resolution reflected the growing regional frustration with the Liberian conflict, as long as the warlords remained in the interim government and as long as ECOMOG’s implementation strategy required continued cooperation with them, its implied threat remained unconvincing. In recognition of the precarious situation then prevailing in Liberia, ECOWAS decided to postpone elections from August 1996 to May 1997. The United States and the European Union promised to provide logistical support to ECOMOG. Disarmament, demobilization, and the repatriation of Liberian refugees were now expected to occur by January 1997. The

S-20

11/24/09

4:27 PM

Page 616

616

CASE STUDIES

ECOWAS heads of state designated Ruth Perry, a former Liberian senator, as the new chair of the Council of State. This agreement came to be known as Abuja II. The implementation schedule of the new agreement was divided into five stages, with three assessment meetings and two ECOWAS foreign ministerial meetings scheduled for Monrovia between October 1996 and April 1997. During these meetings, sanctions, including travel and residence restrictions, freezing of business activities and assets, and exclusion from the electoral process could be recommended against Liberia’s factions by ECOWAS foreign ministers for noncompliance with the agreement. For repeated noncompliance, ECOWAS could invoke the OAU 1996 summit resolution threatening a war crimes tribunal. ECOWAS leaders also adopted a code of conduct for the Council of State that gave ECOWAS the power to replace erring members.39 These sanctions revealed a new ECOWAS approach of monitoring the implementation more closely and adding a punishment option for noncompliance. But success still depended on a more united subregion, some of whose members continued to support individual warlords. On September 3, 1996, Ruth Perry was inaugurated as chair of the Council of State. Although Perry proved to be an energetic council chair, power still lay in the hands of the warlords, who controlled most of Liberia’s territory, and the council remained divided. The three warlords blocked serious dialogue on key issues of the peace process,40 and the political gridlock was further exacerbated by their continued control of Liberia’s ministries, which were now run by their appointees. We next discuss the third and final phase of Liberia’s tortuous peace process.

Implementing Abuja II, September 1996–July 1997 Disarming the Factions ECOMOG’s commitment to disarming Liberia’s factions was aided tremendously by newfound international assistance. During the mayhem in Monrovia in April 1996, the United States helped to create the International Contact Group on Liberia to identify how international actors could assist Liberia and ECOMOG. Between August and December, Washington provided the logistical support it had long denied ECOMOG’s peacekeepers, releasing $40 million for, among other things, helicopters, communication equipment, uniforms, and medical equipment. This, along with the arrival of 119 trucks as well as helicopters and communication equipment from EU states, gave the peacekeepers the logistical support to deploy confidently to the countryside for the first time since the start of their mission. The Dutch minister for development cooperation, Jan Pronk, launched a

S-20

11/24/09

4:27 PM

Page 617

LIBERIA

617

fundraising process in October 1996 that culminated in three donor conferences that eventually provided vital funds to support Liberia’s disarmament, demobilization, and electoral process.41 By November 1996, ECOMOG’s Nigerian force commander, General Victor Malu, had deployed his troops to disarmament sites all over Liberia. The disarmament process started on schedule on November 22. Fighters were given food rations and provided transportation to chosen destinations in exchange for the surrender of a serviceable weapon or a hundred rounds of ammunition. Despite the scale of the disarmament exercise, many of the weapons surrendered in the early stages were unserviceable. The second Abuja accord did not provide even small assistance packages for former combatants, who complained about the meager food rations. Unlike the first Abuja agreement, which envisaged demobilized fighters spending one week at encampment centers, demobilization under Abuja II saw disarmed fighters leaving the camps within twenty-four hours. But despite these difficulties, the warlords largely cooperated with disarmament. There were, however, reports of arms caches being buried in the Liberian countryside by the factions.42 By February 9, 24,500 of the estimated 33,000 fighters, 74 percent, had been disarmed and demobilized. These included 4,306 child fighters and 250 adult female fighters. Over 9,570 weapons and 1.2 million pieces of ammunition were also surrendered, while ECOMOG’s cordon-and-search operations around the country yielded another 122,162 pieces of ammunition and 917 weapons.43 Most roadblocks were cleared and manned by ECOMOG soldiers. Cordon-and-search operations continued to recover weapons in various parts of the country. With the arrival, between February and April 1997, of 650 Malian, 500 Ghanaian, 320 Burkinabè, 321 Nigerien, and 250 Beninois peacekeepers, along with a 35-member medical team from Côte d’Ivoire, ECOMOG’s troop strength grew from 7,500 to 10,500 with an additional 1,000 Nigerian troops also sent to Liberia. The arrival of new national contingents diversified the force’s composition to include six francophone and four anglophone states, though Nigeria still had 7,000 of the 10,500 troops. But for the first time since its arrival in Liberia, ECOMOG’s francophone contingents outnumbered their anglophone counterparts. Social, Economic, and Humanitarian Assistance As civil administration and social services collapsed and the economy declined during the civil war, Liberia depended on the United Nations and various NGOs to provide humanitarian assistance. The European Commission in Brussels donated $2 million to the UN International Children’s Emergency Fund (UNICEF) for primary education and provided offices and logistical and technical support to Liberia’s Independent

S-20

11/24/09

4:27 PM

Page 618

618

CASE STUDIES

Elections Commission. The United States contributed $7.4 million to the elections. UN agencies and international NGOs provided humanitarian relief and social and economic support. The World Food Program (WFP) distributed food to 252,000 displaced persons and refugees and conducted feeding programs for other vulnerable groups. The UN Development Programme (UNDP) provided vehicles, computers, office supplies, and communication equipment to Liberian government agencies. The Food and Agriculture Organization (FAO) distributed agricultural products and provided jobs to war-affected farmers. UNICEF provided community centers for vocational and literacy training as well as shelters, transit homes, and trauma counseling to demobilized child fighters. The World Health Organization (WHO) undertook a national vaccination campaign and provided medical equipment and medicines. The International Labour Organization (ILO) launched a vocational training program to increase skilled labor. International NGOs in Liberia that assisted the peace process included Catholic Relief Services, Médecins du Monde, International Rescue Committee, Save the Children Fund, Action Contre la Faim, and Oxfam. The WFP, EU, and UNDP established a Civil Reconstruction Team bridging program, to support demobilization. The UN spent $3.3 million on bridging activities: the UNDP’s 110 public works projects created 10,000 jobs for civilians and former fighters, while the UN Humanitarian Coordination Office funded 44 projects involving 8,352 temporary jobs. The EU’s 128 microprojects also provided short-term employment for nearly 8,000 civilians and former combatants. By June 1997, 30,000 former fighters and civilians had benefited from the short-term employment and training opportunities, which included rehabilitating rural roads and bridges; restoring health and education infrastructure; building latrines, wells, bridges, and shelters; farming; civic education; and skills development. Although these programs sound impressive, they fell far short of the assistance Liberia needed to restore economic and social life and ensure a durable postwar political order. By June 1997, only 25 percent of Liberia’s public and private health facilities were functioning; electricity supply was nonexistent in most parts of the country; and 60 percent of children were absent from school.44 The July 19, 1997, Election On April 7, 1997, a month behind schedule, the Independent Elections Commission and the Supreme Court were installed in Monrovia. Three weeks later, an ECOWAS foreign ministers’ delegation met with the Elections Commission and Liberia’s political parties to discuss the draft electoral law, the electoral timetable, and the Elections Commission’s

S-20

11/24/09

4:27 PM

Page 619

LIBERIA

619

budget. During the meeting, the delegation reduced the commission’s proposed budget from $9.5 million to $5.4 million and decided that ECOWAS should assume greater responsibility for the physical and logistical aspects of the electoral process, including transportation and the procurement of electoral material.45 ECOWAS’s assertiveness reflected the tension that pervaded the electoral preparations. Following tense bilateral relations between Washington and the military junta in Abuja, Nigeria was keen not to allow external actors to steal the glory for ending the Liberian war. A minimum $100,000 levy was imposed on ECOWAS members to help pay for the elections in a bid to raise $1.5 million on top of the $5.4 million earlier approved for the electoral process. ECOWAS and the UN established a Joint Electoral Coordination Mechanism, which met regularly with Liberia’s Elections Commission to discuss operational and other issues related to electoral preparations. On May 21, 1997, the ECOWAS heads of state held an extraordinary summit in Abuja. The meeting postponed Liberia’s election from May to July and approved the new Liberian electoral law, which was based on Liberia’s 1985 Constitution.46 Only about 75,000 refugees returned home to participate in Liberia’s elections, leaving over 550,000 disenfranchised refugees, a fifth of the population, in neighboring countries. Many of these refugees still felt it was unsafe to return to Liberia. However, based on the scale of Taylor’s electoral victory, their participation would not have made much difference to the eventual outcome of the presidential election. On July 19, elections were held in Liberia, with ECOMOG providing security at the 1,864 voting stations and 500 international observers from the UN, EU, OAU, the Carter Center, and Friends of Liberia observing the poll.47 Final results were announced on July 24, with Charles Taylor scoring a landslide victory. He won 75.3 percent of the presidential vote, while his closest rival, Ellen Johnson-Sirleaf, won 9.5 percent. Taylor’s National Patriotic Party (NPP) also won 21 of the 26 Senate seats and 49 of the 64 seats in the House of Representatives. An impressive 85 percent of the 750,000 registered voters turned out to cast their ballots on polling day. ECOWAS and the UN declared the elections “free and fair.” There were some logistical and technical deficiencies: voter education had been inadequate; no census was conducted; and ballot secrecy was sometimes compromised by assistance given to illiterate voters. But some of these difficulties were to be expected after seven years of civil war in the first Liberian election in twelve years. Three reasons account for Taylor’s spectacular victory. First, Liberians saw Taylor as a guarantor of peace and stability and feared a return to war if he lost. During the campaign, Taylor had repeatedly told journalists that he could not imagine himself losing the election and that the issue of accepting electoral defeat was irrelevant. Second, with an estimated $450

S-20

11/24/09

4:27 PM

Page 620

620

CASE STUDIES

million in earnings from natural resources in areas he controlled during the war, Taylor had an electoral war chest that his rivals simply could not match. This enabled him to reach more voters through private radio stations and newspapers while also allowing him to distribute largesse to voters. Finally, Taylor’s political opposition was hopelessly divided and failed to present a united front against him. On August 2, Charles Taylor was inaugurated as Liberia’s twenty-first president, finally achieving his long-held ambition at the tremendous cost of a devastating seven-year civil war. Taylor had finally won the prize that had long eluded him, and his spoiler tactics had paid off handsomely. It was, however, clear that there were difficult times ahead: the new government was inheriting an empty national treasury with only $17,000, a domestic debt of $200 million, and an external debt of $3 billion.48

Evaluating Implementation The failure to implement thirteen major peace agreements in Liberia for six years can best be explained through an assessment of complex interactions at the domestic, subregional, and external levels of analysis. Domestically, the implementation of peace agreements was rendered difficult by the proliferation of armed factions and the manipulation of ethnic rivalries by power-seeking warlords, a situation exacerbated by the refusal of the strongest faction, the NPFL, to share power with other groups, and the determination of other factions to gain a stake in power through territorial conquest. Mutual suspicions and fears rendered the factions unwilling to commit to disarmament. Economic profits derived from territory under their control proved a powerful disincentive to end the war. At the subregional level, ECOWAS, the mediator and implementer of the various peace accords, disagreed on how to resolve the conflict. Burkina Faso and Côte d’Ivoire supported the NPFL militarily for much of the war, and complained about Nigeria’s ambitions to dominate West Africa; Guinea and Sierra Leone backed ULIMO; Nigeria provided military assistance to the AFL and the LPC; and Ghana and Nigeria disagreed on fundamental issues like the representation of the warlords on the Council of State and whether ECOMOG should pursue a peacekeeping or peace enforcement mandate. Due to these subregional divisions, peace agreements calling for economic sanctions and arms embargoes against Liberia’s warlords were difficult to implement. While ECOWAS lacked political unity, allowing the warlords to exploit subregional divisions, ECOMOG lacked the military tools to engage the warlords in a protracted guerrilla war. At the external level, international actors largely ignored subregional efforts to resolve the conflict. The main external actor, the United States, channeled its support largely toward humanitarian relief. Washington criti-

S-20

11/24/09

4:27 PM

Page 621

LIBERIA

621

cized ECOMOG for lack of impartiality and its deteriorating diplomatic relations with Nigeria after annulled elections in 1993 resulted in inconsistent support for ECOMOG. The United Nations had less than 100 unarmed peacekeepers in Liberia for most of the war, and the OAU’s 1,500 peacekeepers withdrew after a year, following the failure of the international community to deliver on promised logistical support. Initially, inexperienced mediators and subregional divisions led to vague peace agreements that papered over fundamental differences. But even after timetables and sanctions were spelled out in more detail by 1993, warlords who benefited from the export of Liberia’s resources and who knew that ECOMOG could not engage them in a costly guerrilla war simply refused to fulfill their commitments. The mediation process therefore became delinked from the implementation process. Peace meetings took on a surreal air with mediators and parties repeatedly signing agreements that all sides knew were unlikely to be implemented. By 1997 most of these difficulties were finally resolved, allowing the implementation of the Abuja II Peace Agreement. At the domestic level, the warlords cooperated with ECOMOG on disarmament, with the most powerful faction leader, Charles Taylor, confident of winning power through elections. At the subregional level, Nigeria made peace with Taylor and no longer supported his enemies, resulting in Burkina Faso and Côte d’Ivoire contributing troops to ECOMOG. At the external level, the United States and the European Union provided crucial logistical support to the West African peacekeepers, enabling ECOMOG to disarm the factions. The four post-Cotonou peace agreements did not specify a peacebuilding role for international actors in rebuilding Liberia’s institutions and restoring social services, beyond a supporting role in restructuring its security forces. The first Abuja agreement did, however, envisage a role for ECOWAS, the UN, and the OAU in the work of the Elections Commission. The agreements were also silent on the repatriation of refugees, again emphasizing the urgent concern of the implementers with military and political issues of disarmament and interim administration. The peace agreements of the Liberian civil war were not inherently unworkable. They failed largely due to the difficult implementation environment and the mismatch between ECOMOG’s strategy and the resources needed to bring the war to an end. Initially, ECOMOG lacked the means to force compliance on the parties and was unable to obtain substantial subregional and international support for its efforts. ECOMOG demonstrated flexibility in attempting to prioritize its goals by focusing early efforts on gaining the consent of the factions, bridging subregional divisions, and attracting international support. But these efforts failed due to the reasons outlined above. Implementation strategies varied over time, and emphasized both sticks and carrots. After ECOMOG realized that it could not defeat

S-20

11/24/09

4:27 PM

Page 622

622

CASE STUDIES

Liberia’s strongest faction, the NPFL, regional actors attempted to isolate the NPFL diplomatically and to create and support an interim civilian government in Monrovia. ECOMOG, however, always left open to the NPFL the option of participating in an interim government if it agreed to disarmament and elections. Yet ECOMOG remained too weak to force disarmament on the warlords and eventually changed its strategy from supporting an interim civilian government in Monrovia to appeasing the warlords by including them in the interim government. In order to achieve peace, ECOMOG harnessed to its own efforts the strength of warlords it had earlier attempted to marginalize. This reflected the hard-headed pragmatism of subregional mediators who were now determined to secure peace by sharing out positions in an interim government in exchange for disarmament. The implementers abandoned their four-year strategy of protecting a government of civilians in Monrovia while leaving the countryside to the warlords. This led to the demobilization and disarmament of Liberia’s fighters; the organization of the first election in twelve years; and the start of peacebuilding activities to reintegrate fighters into civilian life and to rebuild Liberia’s shattered institutions. In the end, historical contingencies combined with learning by the implementers to end the war in Liberia. ECOMOG’s strategy was largely improvised, building on acquired experience, and only the decision to appease the warlords by granting them political power eventually terminated the conflict. The proliferation and strength of military factions that successfully challenged NPFL control over its territory—not the existence of more detailed peace agreements—led the NPFL to negotiate seriously. The changed and ultimately successful Nigerian strategy of rapprochement with the NPFL was itself only possible with a military coup and change of regime in Abuja in November 1993.

The Quality of the Peace How stable has this peace of warlords been? President Taylor’s first five years in office have been difficult financially and marred by controversy over human rights violations.49 The UN established its first peacebuilding support office in Liberia (UNOL) after the 1997 elections to coordinate postconflict programs of various UN agencies, support the rehabilitation of demobilized soldiers, and promote international support for Liberia’s reconstruction efforts. But foreign assistance has only trickled in. Liberia’s National Reconstruction Program (NRP) sought $433 million over the first two postelection years and aimed to revive government institutions to provide essential social services, rehabilitate public infrastructure, protect civil liberties and national and personal security, repatriate and resettle refugees

S-20

11/24/09

4:27 PM

Page 623

LIBERIA

623

and internally displaced persons, and reintegrate demobilized fighters into society and provide them with increased employment opportunities. The program talked of “transparency,” “accountability,” and “broad-based participation.”50 But the reality of post-1997 Liberia has not matched these noble objectives. Though Taylor at first embarked on a policy of national reconciliation by inviting members of rival parties to join his cabinet, he has gradually cracked down on opposition and attempted to institutionalize his dominance of the Liberian state. Former Taylor ally and later opponent Sam Dokie and members of his family were brutally murdered in November 1997 after being arrested by Taylor’s security forces. The suspects were acquitted in April 1998 as a result of an apparent lack of evidence. Journalists have been harassed and jailed for short periods for criticizing the government. Two radio stations were closed down in March 2000, while four newspapers were closed down and several journalists were arrested in February 2001. Liberia’s security situation remains precarious five years after the end of the war. Armed robbery, looting of food aid, and banditry have thrived in rural Liberia. Crime and insecurity continue to plague Monrovia. The mobilization of armed ethnicity in support of the agendas of rival warlords during the civil war has led to continuing strains in post-1997 Liberia. A shooting incident involving Taylor’s security men and his former Krahn minister of rural development, Roosevelt Johnson, at the U.S. embassy in Monrovia in September 1998 led to fifty-two deaths and the evacuation of Johnson from Liberia. There were reports of the harassment of Krahns in Monrovia by Taylor’s security forces following this incident, forcing over 4,000 Krahns to flee to Côte d’Ivoire.51 Another disturbing trend involves continuing reports of Mandingos being violently attacked by Lorma in Lofa county and by Gios and Manos in Nimba county. These clashes involve disputes over land and resources as refugees return to areas abandoned during the civil war. By June 1998, arsonists had reportedly burnt six mosques throughout the country. Liberia’s ethnic and religious powderkegs could be ignited, with disastrous results, if these conflicts are not carefully managed. Nor has security returned to Liberia’s borders. After 200,000 Sierra Leonean refugees flooded into Liberia in September 1997 following clashes in that country between soldiers, rebels, and Kamajors (traditional hunters), Taylor sent troops to guard his border. The Liberian president is particularly sensitive about his border areas since ULIMO launched its invasion against the NPFL from Sierra Leone, and since Taylor himself invaded Liberia from Côte d’Ivoire. In April and August 1999, Liberian dissidents believed to be former ULIMO fighters, describing themselves as Liberians United for Reconciliation and Democracy (LURD), invaded the towns of Voinjama and Kolahun in Liberia’s Lofa county, from Guinea, before being repelled by Taylor’s security forces. Taylor-backed RUF

S-20

11/24/09

624

4:27 PM

Page 624

CASE STUDIES

rebels launched attacks into Guinea shortly before May 1999. In September 1999, two Guinean villages on the border with Liberia were attacked by the LURD. An ECOWAS meeting was swiftly convened in Abuja, following this last incident, at which Taylor and Guinea’s Lansana Conté pledged to “embrace good neighbourliness and non-aggression.” But despite such promises, Taylor continued to accuse Guinea of trying to destabilize his regime by providing Kromah a sanctuary in Conakry, while Conté accused Taylor of backing LURD dissidents along Liberia’s northwest frontier with Guinea.52 At the time of writing in April 2002, LURD rebels were reported to be only thirty miles from Monrovia. Instability on Liberia’s borders had earlier helped to sour relations between Taylor and ECOMOG. ECOMOG, along with the United States and Britain, criticized Taylor’s continued military support for RUF rebels who were fighting ECOMOG in Sierra Leone. Taylor in turn accused ECOMOG of supporting some of his rivals in Sierra Leone, refused ECOMOG jets permission to use Liberian airfields for missions into Sierra Leone, called for RUF leader Foday Sankoh’s release from Nigerian detention, criticized ECOWAS sanctions against the AFRC/RUF military junta, and publicly opposed the Nigerian-led intervention in Freetown to restore the elected government of Tejan Kabbah to power in February 1998. ECOMOG finally withdrew its peacekeepers from Liberia by the end of 1998. The most disturbing aspect of the post-Abuja era is the failure of Taylor to restructure the Armed Forces of Liberia and other security forces, with the assistance of ECOMOG and the UN, as envisaged in the Accra Agreement of 1994. After his election in 1997, Taylor refused to allow the peacekeepers a role in restructuring and training his security forces. He ordered the demobilization and retirement of 2,628 soldiers, including many Krahn officers, from the AFL in January 1998, though he still maintains prominent Krahns in his government. Taylor has included at least 2,000 of his former NPFL fighters in the new Liberian army. 53 The restructuring process has been criticized for being partisan and lacking transparency. Taylor’s decision to fill his new army with loyal lieutenants is hardly unique in contemporary African politics, where leaders in Burundi, Rwanda, Côte d’Ivoire, Togo, and Zaire traditionally placed kinsmen in strategic military positions. The fragile state of post-1997 Liberia and the continuing sources of insecurity make his preponderant concern with security understandable even if not justifiable. But in the long run, Taylor runs the risk of repeating the same fatal mistake that Doe made, which the current president himself was able to exploit: by filling the army with ethnic loyalists and using it as a tool against political opponents, Taylor could also create conditions for the mobilization of ethnic groups to protect their own people against a partisan army. The institutionalization of violence as a way of life that was a hallmark of the Doe era has been further entrenched dur-

S-20

11/24/09

4:27 PM

Page 625

LIBERIA

625

ing Liberia’s savage civil war.54 The Armed Forces of Liberia has historically been an instrument of partisan rule, first defending the interests of the Americo-Liberian oligarchy, then keeping the autocratic regime of Doe in power, and now attempting to ensure the survival of Taylor’s regime. Most of Taylor’s opponents view the state apparatus as an extension of his own personal power, and the dispensation of justice is not perceived as neutral. This creates potential conditions for the breakdown of law and order. If the Liberian president extends his influence to other state institutions like the Electoral Commission in a bid to maintain himself in power beyond the next election in 2003, he could create a situation similar to Doe’s fraudulent election of 1985, which left opponents with no legitimate way to challenge him other than through a resort to violence. All of Taylor’s main political rivals now live outside the country. On November 12, 1998, Taylor accused thirty-two people, including self-exiled former warlords Alhaji Kromah, Roosevelt Johnson, and George Boley, of a plot to overthrow his government, and tried the suspects. Some of these suspects were subsequently convicted and jailed. Taylor still has the psychology of a warlord having waged a guerrilla campaign for seven years. He has so far failed to make the transformation from warlord to statesman. Good governance in a democratic setting has been far more difficult than the unencumbered autocracy of warlord politics. During the civil war, Taylor briefly lost his headquarters in Gbarnga to his enemies, saw other factions gang up against him, witnessed Nigeria, Sierra Leone, and Guinea providing assistance to rival factions, and heard calls by several Nigerian generals for his assassination. He is still profoundly scarred by these experiences. Having himself led an ultimately successful rebellion into Liberia based on the exploitation of ethnic grievances, and having witnessed Doe’s grisly end, he is deeply aware of his own vulnerability in the precarious Liberian political cesspit. His paranoia and obsession with security are not altogether surprising. Based on Liberia’s history and the fragility of the post-1997 state, Taylor does have genuine reasons to be concerned about his security and does need a state that wields a monopoly over the means of violence. But this Leviathan must be seen as legitimate and its army must be perceived to be acting in the overall interests of its citizens rather than in the parochial interests of a paranoid leader. Having amassed so much illicit wealth and gathered so many enemies in his bloody ascent to power, Taylor, like Doe before him, is determined to hold on to power in order to keep any vindictive rival from probing his past practices. Events in post-1997 Liberia still seem to parallel those in pre-1990 Liberia: widespread insecurity, a weak economy, patronage-fueled corruption, harassment of the press and civic groups, interethnic clashes, and trumped-up coup plots. The lack of security, the collapse of public infrastructure, and the absence of the rule of law have made donors cautious and kept away for-

S-20

11/24/09

626

4:27 PM

Page 626

CASE STUDIES

eign investors. South African and Canadian firms are involved in prospecting for Liberian minerals, but such interest has been rare. Taylor has maintained close political relations with Burkina Faso and Libya and sought to court Taiwanese and French investors. Traditional trading partners like the United States, Japan, and Germany continue largely to stay away. Liberian diamonds continue to be smuggled out of the country,55 and there are accusations of the awarding of lucrative contracts to close political associates of Taylor. A donor conference for Liberia’s reconstruction held in Paris in April 1998 led to pledges of $200 million. But these funds were made conditional on progress on security and human rights, and have not been released.56 Little of Liberia’s infrastructure has been restored and the promised provision of social services is yet to occur as donor wariness replaces donor fatigue. Amid these domestic difficulties, Taylor faces unprecedented external pressure. Following reports of continuing collaboration between Taylor and the RUF,57 the UN Security Council, on May 7, 2001, imposed a ban on the export of diamonds from Liberia as well as travel sanctions on senior government officials and their spouses. The council also tightened an existing arms embargo by prohibiting the sale or supply of arms and related material to Liberia and banning the provision of military training to the government. These sanctions, which many of Liberia’s civil society groups and opposition politicians supported on the basis that they would hurt the leadership in Liberia more than the people, are to be reassessed every fourteen months. In order for the UN sanctions to be effective, however, it is important that the UN Security Council secure ECOWAS’s cooperation in implementing them. The diamond sanctions may not be watertight since Liberian gems can easily be smuggled through third countries and since no effective enforcement mechanism exists on the ground. The travel sanctions and arms embargo will also need the support of ECOWAS states to be implemented. But the sanctions have clearly rattled Taylor. He has accused their main architect, Britain and the United States, of attempting to overthrow his regime. More devastating economic sanctions on Liberia’s $13 million annual timber trade were blocked in the UN Security Council by France and China, which together import about 45 percent of Liberia’s timber.58 Taylor currently faces the greatest challenge to the survival of his regime. Liberia’s total government revenue for 2000 was estimated to have been only $85 million, even as the Liberian president announced that $15 million of this sum had been diverted to the war against the LURD. 59 Meanwhile, civil service salaries in Liberia went unpaid for almost one year. Based on this analysis, it is probably safe to conclude that peace in Liberia is unlikely to be self-sustaining in the long run unless there is a dramatic change of course by the current regime. The warlord’s peace of 1997 may yet become another full-blown war.

S-20

11/24/09

4:27 PM

Page 627

LIBERIA

627

Notes 1. UN Document S/1997/712, “Final Report of the Secretary-General on the United Nations Observer Mission in Liberia,” September 12, 1997, p. 5. 2. Members of ECOWAS include Benin, Burkina Faso, Cape Verde, Côte d’Ivoire, Gambia, Ghana, Guinea, Guinea-Bissau, Liberia, Mali, Niger, Nigeria, Senegal, Sierra Leone, and Togo. 3. For accounts of Doe’s Liberia, see Christopher Clapham, “Liberia,” in Donal Cruise O’Brien, John Dunn, and Richard Rathbone, eds., Contemporary West African States (Cambridge: Cambridge University Press, 1989), pp. 99–111; D. Elwood Dunn and S. Byron Tarr, Liberia: A National Polity in Transition (Metuchen, N.J.: Scarecrow Press, 1988); and Gus Liebenow, Liberia: The Quest for Democracy (Bloomington: Indiana University Press, 1987). 4. This figure is an estimate based on the 1974 Liberian census. 5. For accounts of the Liberian civil war, see Adekeye Adebajo, Liberia’s Civil War: Nigeria, ECOMOG, and Regional Security in West Africa (Boulder, Colo., and London: Lynne Rienner, 2002); Abiodun Alao, The Burden of Collective Goodwill: The International Involvement in the Liberian Civil War (Aldershot: Ashgate, 1998); Stephen Ellis, The Mask of Anarchy (London: Hurst, 1999); Karl Magyar and Earl Conteh-Morgan, eds., Peacekeeping in Africa: ECOMOG in Liberia (Hampshire, London: Macmillan, 1998); Klaas Van Walraven, The Pretence of Peace-Keeping: ECOMOG, West Africa, and Liberia, 1990–1998 (The Hague: Netherlands Institute of International Relations, 1999); and Margaret Vogt, ed., The Liberian Crisis and ECOMOG: A Bold Attempt at Regional Peacekeeping (Lagos: Gabumo Press, 1992). 6. William Reno, “The Business of War in Liberia,” Current History 95, no. 601 (May 1996): 115. 7. Africa Confidential 32, no. 10 (May 17, 1992): 3. 8. William Reno, Warlord Politics and African States (Boulder, Colo.: Lynne Rienner, 1998), pp. 104–105. 9. Philippa Atkinson, The War Economy in Liberia: A Political Analysis (London: Overseas Development Institute, 1997), p. 9. 10. See “Waging War to Keep the Peace: The ECOMOG Intervention and Human Rights,” Human Rights Watch/Africa 5, no. 6 (June 1993); “Liberia: Emerging from Destruction,” Human Rights Watch/Africa 9, no. 7 (November 1997); and the UN Secretary-General’s Reports to the Security Council on Liberia. 11. See Henry Bienen, Armed Forces, Conflict, and Change in Africa (Boulder, Colo.: Westview Press, 1989); Samuel Decalo, Coups and Army Rule in Africa, 2nd ed. (New Haven, Conn.: Yale University Press, 1990); and William Foltz and Henry Bienen, eds., Arms and the African: Military Influences on Africa’s International Relations (New Haven, Conn.: Yale University Press, 1985). 12. Paul Richards, “Rebellion in Liberia and Sierra Leone: A Crisis of Youth?” in Oliver Furley, ed., Conflict in Africa (New York: Tauris Academic Studies, 1995), p. 143. 13. Emmanuel Kwezi Aning, “Managing Regional Security in West Africa: ECOWAS, ECOMOG, and Liberia,” Working Paper no. 94.2 (Copenhagen: Centre for Development Research, February 1994), p. 12. 14. Stephen Ellis, “Liberia 1989–1994: A Study of Ethnic and Spiritual Violence,” African Affairs 94, no. 375 (April 1995): 181. 15. See, for example, Joseph Garba, Diplomatic Soldiering: Nigerian Foreign Policy, 1975–1979 (Ibadan, Nigeria: Spectrum Books, 1987); Yakubu Gowon, The Economic Community of West African States: A Study of Political and Economic

S-20

11/24/09

628

4:27 PM

Page 628

CASE STUDIES

Integration (Ph.D. thesis, Wawrick University, UK, February 1984); and Olusegun Obasanjo, Not My Will (Ibadan, Nigeria: University Press, 1990). 16. See General Chris Alli, The Federal Republic of Nigerian Army: The Siege of a Nation (Lagos: Malthouse Press, 2001), pp. 346–347; and Arthur Nwankwo, Nigeria: The Stolen Billions (Enugu, Nigeria: Fourth Dimension, 1999), pp. 184–185. 17. Author interview with Herman Cohen, former assistant secretary of state for African affairs, Washington, D.C., July 1997. 18. See the ECOWAS charter of 1975 and security protocols of 1979 and 1981. 19. See David Wippmann, “Enforcing the Peace: ECOWAS and the Liberian Civil War,” in Lori Fisler Damrosch, ed., Enforcing Restraint: Collective Intervention in Internal Conflicts (New York: Council on Foreign Relations, 1993), pp. 157–203; and Abass Bundu, “The Case Against Intervention,” West Africa no. 4156 (June 30–July 6, 1997). 20. Wippmann, “Enforcing the Peace,” p. 174. 21. See First Session of the ECOWAS Standing Mediation Committee, Final Communiqué, Banjul, Gambia, August 6–7, 1990. 22. See First Extraordinary Session of the Authority of Heads of State and Government, Final Communiqué, Bamako, Mali, November 27–28, 1990. 23. See Second Meeting of the ECOWAS Committee of Five, Final Communiqué, Yamoussoukro, September 16–17, 1991; and Third Meeting of the ECOWAS Committee of Five, Final Communiqué, Yamoussoukro, October 29–30, 1991. 24. See Colonel Festus Aboauye, ECOMOG: A Subregional Experience in Conflict Resolution Management and Peacekeeping in Liberia (Accra, Ghana: SEDCO, 1999), pp. 229–266; Adekeye Adebajo, Building Peace in West Africa: Liberia, Sierra Leone, and Guinea-Bissau (Boulder, Colo.: Lynne Rienner, 2002), pp. 79–109; and Robert Mortimer, “From ECOMOG to ECOMOG II: Intervention in Sierra Leone,” in John W. Harbeson and Donald Rothchild, eds., Africa in World Politics, 3rd ed. (Boulder, Colo.: Westview Press, 2000), pp. 188–207. 25. See Informal Consultative Group Meeting of the ECOWAS Committee of Five on Liberia, Final Communiqué, Geneva, April 6–7, 1992. 26. Quoted in West Africa no. 3892 (April 20–26, 1992): 674. 27. Lindsay Barrett, “Why Senegal Withdrew,” West Africa no. 3931 (January 25–31, 1993). 28. For further details on ECOMOG’s logistical shortcomings, see Herbert Howe, “Lessons of Liberia: ECOMOG and Regional Peacekeeping,” International Security 21, no. 3 (winter 1996–1997): 145–176; and Cyril Iweze, “Nigeria in Liberia: The Military Operations of ECOMOG,” in Margaret Vogt and A. E. Ekoko, eds., Nigeria in International Peacekeeping, 1960–1992 (Lagos: Malthouse Press, 1993), pp. 216–243. 29. Author interview with Trevor Gordon-Somers, UNDP, New York, May 1997. 30. See “Liberia: Problematic Peacekeeping,” Africa Confidential 35, no. 5 (March 4, 1994): 2–3; Clement Adibe, “The Liberian Conflict and the ECOWASUN Partnership,” Third World Quarterly 18, no. 3 (1997): 471–488; Binaifir Nowrojee, “Joining Forces: UN and Regional Peacekeeping, Lessons from Liberia,” Harvard Human Rights Journal 8 (spring 1995): 129–151; and Funmi Olonisakin, “UN Cooperation with Regional Organizations in Peacekeeping: The Experience of ECOMOG and UNOMIL in Liberia,” International Peacekeeping 3, no. 3 (autumn 1996): 33–51.

S-20

11/24/09

4:27 PM

Page 629

LIBERIA

629

31. West Africa no. 4067 (September 18–24, 1995): 1477. 32. Economic Community of West African States Consultative Meeting on the Liberia Peace Process, Final Report, ECW/MINFA/CTTE9/VI/2, August 16–19, 1995, p. 8. 33. Ibid., p. 10. 34. See Adekeye Adebajo, “Rich Man’s War, Poor Man’s War,” The World Today 52, nos. 8–9 (August–September 1996). 35. West Africa no. 4073 (November 6–12, 1995): 1717. 36. Seventh Meeting of ECOWAS Committee of Nine Foreign Ministers, Final Report (Restricted), ECW/MINFA/CTTE9/VII/2/Rev.1, Accra, May 7, 1996, p. 7. 37. For a detailed account of the events in Monrovia, see UN Document S/1996/362, “The Seventeenth Progress Report of the Secretary-General on the United Nations Observer Mission in Liberia,” May 21, 1996. 38. Quoted in West Africa no. 4109 (July 22–28, 1996): 1139. 39. See Fourth Meeting of ECOWAS Heads of State and Government of the Committee of Nine, Final Communiqué, Abuja, August 17, 1996, pp. 4–7. 40. Friends of Liberia, “Liberia: Opportunities and Obstacles for Peace,” December 1996, p. 28. 41. See Klass Van Walraven, The Netherlands and Liberia: Dutch Policies and Interventions with Respect to the Liberian Civil War (The Hague: Netherlands Institute of International Relations, 1999). 42. For further details, see Victor Tanner, “Liberia: Railroading Peace,” Review of African Political Economy 75, no. 25 (March 1998): 133–147. 43. UN Document S/1997/237, “Twenty-second Progress Report of the Secretary-General on the United Nations Observer Mission in Liberia,” March 19, 1997, pp. 3–4. 44. UN Document S/1997/478, “Twenty-third Progress Report of the Secretary-General on the United Nations Observer Mission in Liberia,” June 19, 1997, p. 9. 45. Ibid., p. 2. 46. See Independent Elections Commission, Special Elections Law for the 1997 Elections (Monrovia: Sabanoh Press, 1997). 47. See Terrence Lyons, Voting for Peace: Post Conflict Elections in Liberia (Washington, D.C.: Brookings Institution, 1998). 48. Economist Intelligence Unit, country report on Liberia, 4th quarter 1997, pp. 1–10. 49. See Jon Lee Anderson, “The Devil They Know,” New Yorker, July 27, 1998; and Baffour Ankomah, “Knives Out for Taylor,” New African, September 1998. 50. See Norwood Langley, “The National Reconstruction Program in Liberia,” in State Rebuilding After State Collapse: Security, Democracy, and Development in Post-War Liberia, Report of the Strategic Planning Workshop on Liberia, June 19, 1998 (London: Center for Democracy and Development, 1998). 51. Economist Intelligence Unit, country report on Liberia, 1st quarter 1999, p. 8. 52. Ibid., 4th quarter 1999, pp. 36–40; and 2nd quarter 1999, p. 36. 53. Africa Confidential 40, no. 4 (February 19, 1999): 7. 54. Amos Sawyer, “Foundations for Reconstruction in Liberia: Challenges and Responses,” in State Rebuilding After State Collapse p. 69. 55. Economist Intelligence Unit, country report on Liberia, 3rd quarter 1998, p. 11.

S-20

11/24/09

630

4:27 PM

Page 630

CASE STUDIES

56. Ibid., 2nd quarter 1998, p. 11. 57. First Report of the Secretary-General Pursuant to Security Council Resolution 1343 (2001) Regarding Liberia, S/2001/424, April 30, 2001, p. 3. 58. Economist Intelligence Unit, “Liberia,” March 2001, p. 49. 59. Ibid., p. 42.

S-21

11/24/09

4:27 PM

Page 631

21 Flawed Mediation, Chaotic Implementation: The 1987 Indo–Sri Lanka Peace Agreement SUMANTRA BOSE

The July 1987 Indo–Sri Lanka Peace Agreement (ISPA) completely failed to achieve its stated objective—a negotiated end to the civil war between the central government, dominated by Sri Lanka’s Sinhalese majority, and the country’s Tamil minority, concentrated in the island’s northern and eastern regions. This failure graphically illustrates two points of relevance to a comparative study of implementation of peace agreements designed to terminate civil wars. First, it underscores the important nexus between the mediation and implementation phases. The ISPA’s mediation process failed to involve a key domestic belligerent—the powerful guerrilla movement known as the Tamil Tigers—without whose cooperation, implementation on the ground in northern and eastern Sri Lanka could not succeed. Because of the failure to involve them in the mediation phase, however, the Tigers’ core security concerns and political objectives were not addressed by the agreement (which was concluded on a bilateral basis between representatives of the Indian and Sri Lankan governments) in a manner even minimally acceptable to the guerrillas. The subsequent failure of implementation suggests that when such a crucial “gap” exists in the negotiations leading to the peace agreement, and is embodied in the document itself, it becomes vital that the strategy or process of implementation attempt to bridge this gap, rather than ignore or—even if inadvertently—aggravate the deficiency. The Sri Lanka case suggests that this was essential to make implementation of the ISPA even minimally viable. However, this was not done. Second, the failure of the ISPA points to possible limitations to the efficacy of peacemaking intervention by regional powers (in this case, India) in the internal conflicts of smaller, weaker countries in the geostrategic neighborhood. When the main mediator and/or implementer is such a 631

S-21

11/24/09

4:27 PM

Page 632

632

CASE STUDIES

regional hegemon, with interests of its own in the internal affairs of neighbors as well as a possible history of prior meddling in the conflict itself, it is not likely to be perceived as a neutral, genuinely disinterested broker by any of the parties to the civil war. India’s peacemaking intervention as mediator and implementer in the Sri Lanka crisis in July 1987 was undoubtedly tainted—and troubled—by these factors. However, this is not to say that this alone condemned the ISPA to failure. India’s “hegemonic” status as a huge and powerful country vis-à-vis tiny Sri Lanka and its warring parties, as well as India’s record of prior involvement in the conflict, implied not just disadvantages but certain potential advantages as well— notably a measure of leverage over both parties to the civil war. However, this leverage had to be used appropriately in order to pay dividends. The Indians signally failed to do so. In particular, the massive Indian intervention in northern and eastern Sri Lanka after the signing of the ISPA on July 29, 1987, failed to acknowledge the critical gap in the agreement outlined above and made no efforts to seriously compensate for it. As a result, the agreement’s implementation bogged down by October 1987 in renewed fighting—this time between the Indian implementation force (Indian Peace-Keeping Force [IPKF], totaling 100,000 troops at its peak in 1989, supported by heavy armor, artillery, and air power) and the Tamil Tigers— and Sri Lanka reverted to civil war once the IPKF withdrew, having failed in its mission, in the first half of 1990. In other words, India failed to implement peace in Sri Lanka not simply or necessarily because it was a regional hegemon, but because it was a blundering regional hegemon.

The Strategic Context The ISPA was concluded in July 1987 with the purported dual intent of terminating four years of steadily escalating civil war between all-Sinhalese Sri Lankan armed forces and Tamil guerrilla groups, and of implementing a lasting political solution to the underlying causes of the conflict. The immediate catalyst to the outbreak of a full-fledged armed conflict in the Tamil-populated areas of northern and eastern Sri Lanka had been a massive pogrom against Tamil civilians living in Sinhalese-majority regions of the island in July 1983. That violence, especially severe in and around the capital, Colombo, left some 3,000 Tamils dead and about 150,000 homeless. Following these killings, a small hit-and-run Tamil separatist insurgency in the north, involving at most a couple hundred committed militants, was transformed by 1984 into a mass-based armed struggle with thousands of fighters drawn from radicalized Tamil youth. Two factors contributed to the rapid escalation of the conflict between 1983 and 1987. First, as a strengthened Tamil insurgency dramatically

S-21

11/24/09

4:27 PM

Page 633

THE INDO–SRI LANKA PEACE AGREEMENT

633

stepped up the frequency and scale of armed attacks in the north and east from 1984 onward, the Sri Lankan state and its armed forces responded with draconian repression, often directed indiscriminately at the Tamil civilian population living in those areas. As massacres and other atrocities multiplied alarmingly, tens of thousands more Tamils fled the war-torn north and east, most to the southern Indian Tamil province located just across the seas from northern Sri Lanka, others to a host of other countries in western Europe, Canada, and Australia. Thousands of young men among these refugees chose to join the exponentially multiplying ranks of various guerrilla groups fighting to create an independent Tamil state in northern and eastern Sri Lanka.1 The second factor that was critical in the rapid escalation of the civil war was India’s role in Sri Lanka’s internal conflict after 1983. The Indian government of Prime Minister Indira Gandhi, as well as its provincial counterpart in Tamil Nadu (the southern Indian province populated by over 50 million ethnic Tamils) adopted a policy of lending active support to the burgeoning Sri Lankan Tamil guerrilla forces. The Indians turned a blind eye as scores of training camps for guerrillas were established on the Tamil Nadu coastline just across from northern Sri Lanka, and as Tamil Nadu was converted into a virtual safe haven, logistical and supply area, and staging base for offensive armed operations for the exploding Tamil insurrection in Sri Lanka. But India’s role went further. India’s external intelligence agency (known as the Research and Analysis Wing [RAW]), its millionplus regular army, and other Indian state-security organizations became deeply implicated in a systematic program of directly providing weapons, training, and other facilities to Tamil guerrilla groups. This policy, begun by Indira Gandhi’s government, was continued after she was assassinated in the fall of 1984 and her son Rajiv succeeded her as prime minister. This partial and strongly interventionist policy was the result of several interlocking perceptions and motives. First, it was a response to genuine outrage among Indian Tamils at the persecution and violence suffered by their Sri Lankan cousins. Popular sympathy for the Sri Lankan Tamil separatist cause was widespread in Tamil Nadu at this time, and was a major factor determining the response of the Tamil Nadu provincial government (run by a party with strong Tamil nationalist roots and leanings) as well as that of the central government run by the Congress party, to which the Tamil Nadu government party was allied at the time. Second, there was a strategic perception involved. Sri Lanka’s internal troubles were viewed as an excellent opportunity to extend India’s strategic influence in the Indian Ocean rim via vigorous support for the Tamil insurgency. This perception quickly congealed into policy after 1983, as the Sinhalese-dominated Sri Lankan government started to request—and receive—weapons and training for its own furiously expanding armed forces from an assortment of coun-

S-21

11/24/09

4:27 PM

Page 634

634

CASE STUDIES

tries perceived to be hostile to Indian interests, including Pakistan, China, South Africa, and Israel, not to mention a shadowy firm of professional mercenaries operating out of Britain’s Channel Islands. The net consequence of the Sri Lankan government’s counterproductive policy of indiscriminate repression and the Indian role was that by the mid-1980s, northern and eastern Sri Lanka was in the grip of full-scale civil war. Until 1983, the Tamil youth insurgency had been a hit-and-run operation, consisting of sporadic, though occasionally quite deadly, strikes by very small groups of guerrillas against police stations, government offices, and military camps. By 1986, thousands of highly trained and well-equipped Tamil fighters effectively controlled large tracts of the northern and eastern provinces. In many areas, especially the thickly populated and overwhelmingly Tamil Jaffna peninsula in the far north, the government’s writ had virtually ceased to run, and its forces were effectively confined to heavily fortified encampments. The Tamil nationalist project of an independent state—Tamil Eelam—seemed to be slowly turning into a reality on the ground. The cost was high—perhaps 10,000 had been killed by 1987 and at least 200,000 Tamils had become refugees in India and elsewhere, though no precise figures are available.2 The large majority of these casualties were Tamil civilians, though sizable numbers of guerrilla combatants and government soldiers had also died. But the consensus among Tamils at this time was that the price was worth paying—in any case, war was inevitable, given the attitude of the Sinhalese government and the conduct of its armed forces. By 1986–1987, one of the several youth-led Tamil guerrilla groups active during this phase—the Liberation Tigers of Tamil Eelam (LTTE, also known as the Tamil Tigers)—had also clearly emerged as the dominant Tamil fighting force, partly because of its superior dedication, organization, leadership, and tactical skills, and partly because of its ruthless proficiency in killing members of rival guerrilla groups.

Historical Background The Sinhalese-Tamil confrontation that culminated in civil war in the 1980s has a long and checkered history. Its roots date to Sri Lanka’s colonial past and to the immediate aftermath of decolonization, when the classic Westminster-style majoritarian democracy bequeathed to Sri Lanka (then known as Ceylon) by the British was rapidly defined by the Sinhalese political elite in terms of dominance by an ethnic, linguistic, and religious majority—Sinhalese Buddhists—and the Sri Lankan state was turned into a vehicle for the assertion of the purported political and socioeconomic interests of that collective identity group. Sinhalese was declared the sole national and official language of the country in 1956, a step that apart from symbolic connotations had major repercussions for Tamil access to higher

S-21

11/24/09

4:27 PM

Page 635

THE INDO–SRI LANKA PEACE AGREEMENT

635

education and employment opportunities in the large public sector of the economy, including various government bureaucracies and services. Once it became clear to the ethnic Tamil counter-elite that an equitable share of power in the centralized, majoritarian-democratic state was no longer a feasible goal, it responded with a classic defensive strategy. This was to declare the north and east the “traditional homeland” of the Tamil people and doggedly seek political autonomy for that region of the country, where Tamils happened to constitute a majority of the population. This approach was not without its problems, given that at least a quarter of all Sri Lankan Tamils lived outside the north and east, and that the eastern province, in particular, was ethnically heterogeneous and contained a large non-Tamil population (principally Muslims, the third major community in Sri Lanka). But the demand for decentralization of the Sri Lankan state and substantive self-government for an autonomous northeastern region—as articulated by the main Tamil political party, the Federal Party (FP)—would after 1956 be the defining feature of Tamil politics. It was this autonomist claim that would, after repeated frustrations and provocations, harden into a movement for a separate state in the late 1970s and early 1980s. Autonomy agreements between the central Sinhalese elite and the Tamil federalists were indeed struck in 1958 and again in 1965, but never implemented, leaving an accumulating legacy of distrust and bitterness. In the early through middle 1970s, a new range of measures with transparent anti-Tamil intentions and effects were enacted by the Sinhalese state elite. A new state constitution enacted into law by parliament in 1972 against vociferous but ultimately ineffectual opposition by the minority of Tamil legislators amounted to a virtual charter of Sinhalese-Buddhist supremacy, formally recognizing Buddhism as the state religion, reaffirming the monopoly of the Sinhalese language in public life, and abolishing the safeguards for minority rights incorporated into the previous constitution, which had been framed by the British immediately prior to their departure in 1948. Between 1973 and 1975, a patently discriminatory scheme was put into operation to squeeze Tamil candidates out of professional higher education in medicine and engineering programs, which had become young Tamils’ principal route to professional careers and upward mobility once the public sector and government services became inhospitable terrain for Tamil speakers after 1956. This had a particularly incendiary effect on the large educated middle class among the Tamils (based primarily in Jaffna) in general, and their younger generation in particular. It provided the impetus to the formation of the first Tamil guerrilla groups. In 1977, the FP and several other Tamil groups reacted to this situation by forming a Tamil United Liberation Front (TULF) and contesting parliamentary elections that year on a single-point platform of achieving Tamil “sovereignty” and “self-determination” for their “traditional homeland” in the north and east. The TULF swept the north with an overwhelming mandate, and also did very well in Tamil-dominated areas of the eastern

S-21

11/24/09

636

4:27 PM

Page 636

CASE STUDIES

province. The TULF’s leaders were by and large constitutional politicians committed to peaceful, democratic methods, and their strategy in 1977 was intended more as a pressure tactic on the Sinhalese state elite than anything else, the primary objectives being a halt to overtly discriminatory policies in language, education, and employment and the commencement of serious negotiations on institutionalizing an autonomous Tamil-controlled political authority in the north and east. Had the Sinhalese central government responded positively and constructively to what was an unambiguous wake-up call, the slide into ethnic war would have been averted even at this late juncture. Unfortunately, not only did this not happen, but successive and progressively worsening bouts of organized anti-Tamil rioting swept the Sinhalese-majority regions of the island in 1977, 1979, and 1981, culminating in the massive violence of 1983. The 1983 riots—in ostensible reprisal for an LTTE ambush near Jaffna that had killed fourteen soldiers—were marked by the active involvement of soldiers, police officers, and activists of the ruling United National Party (UNP) in an organized campaign of murder, rape, arson, and looting directed against vulnerable Tamils living outside the relative safety of the north and east (the other major Sinhalese party, the Sri Lanka Freedom Party [SLFP], had controlled the governments responsible for the most blatant anti-Tamil policies in the 1950s and again in the 1970s). It was also the watershed event that led innumerable Tamils to conclude that even basic security, let alone equitable political representation, was not possible in a rigidly unitary, ethnically defined “democracy” like Sri Lanka’s. This view was reinforced by the fact that a largely indiscriminate and grossly disproportionate campaign of police and military repression had been under way in the north since 1979, in a futile, indeed counterproductive, government effort to stamp out the still-tiny but nonetheless troublesome Tamil guerrilla movement. This campaign made little tangible progress against the determined and elusive guerrillas, but through its brutality irrevocably radicalized large sections of Tamil society, including not just the traditionally proautonomy Tamil bourgeoisie but also increasingly the numerically preponderant nonelite, rural sectors like the fisherfolk and peasantry. This state-sponsored violence, in the form of ethnic pogroms in Sinhalese-dominated areas, and police and military attacks that greatly intensified in the north and east after 1983, would provide separatist Tamil nationalism with a powerful rationale and its fighting forces with a rich supply of recruits.3

The Indian Role Prior to July 1987 and the ISPA The Indian government, for its part, was playing with fire in its heavyhanded and self-interested meddling in the Sri Lankan crisis. The Indians, for all their moral and material support to the Tamil cause, were in the end

S-21

11/24/09

4:27 PM

Page 637

THE INDO–SRI LANKA PEACE AGREEMENT

637

simply not interested in the maximalist Tamil goal of a separate state. Indira Gandhi herself had made this much clear when she had in August 1983 unequivocally rejected calls in the Indian parliament for a full-scale Bangladesh-style Indian military intervention in Sri Lanka and affirmed India’s basic commitment to the territorial integrity of the Sri Lankan state. From an Indian perspective, Sri Lanka in the 1980s was hardly comparable in terms of strategic importance to Pakistan in 1971, and very obviously did not merit commitment of Indian resources and capabilities on any kind of similar scale. In June 1987, Rajiv Gandhi reiterated the same fundamental stance when he stated that “India wants a settlement across the table within the constitutional framework of Sri Lanka. We will not support the Eelam concept.”4 The fatal contradiction in the Indian strategy vis-à-vis the Sri Lankan conflict was, of course, that this (fundamentally sincere) declaratory position was radically at odds with the effects of India’s not-so-covert program of arming, training, and otherwise supporting the rapidly growing Tamil guerrilla movement. The contradiction was compounded by the fact that by 1985–1986, Tamil guerrilla groups more pliable to Indian control had been upstaged by the LTTE, which represented the most uncompromising, hardline tendency among the Tamil militants. Like its rival armed factions, the LTTE benefited substantially from Indian support; unlike them, it never lost its sense of independent purpose, was careful not to become too dependent on the Indians, and to that end utilized the resources and skills obtained through Indian patronage to set up its own self-sustaining fighting machine. By 1986, the Tigers were operating their own training camps in “liberated areas” (usually jungles) of the north and east, and had established an independent arms procurement and supply network drawing on the fast-expanding global diaspora of Sri Lankan Tamil expatriates and refugees. By mid-1987, when the ISPA abruptly disrupted their struggle, the Tigers had developed a formidable fighting organization of between 3,000 and 4,000 dedicated full-time cadres, aided by thousands more auxiliaries, and commanded strong popular support among a deeply radicalized Tamil population once renowned for its pacifism. Their confidence increased in direct proportion to their military strength and capabilities, and a fanatical, nonnegotiable commitment to the ideal of an independent Tamil state became the consuming obsession of this movement. This was a tiger the Indian state would find impossible to either tame or dismount. The Indians soon discovered that they had helped build up a movement that was not amenable to control or manipulation as they wished. As early as July 1985, Indian-mediated talks in a neutral location between representatives of the Sri Lankan government and the Tamils (including the increasingly marginal and irrelevant TULF leadership as well as representatives of the guerrilla groups) collapsed when the youthful Tamil delegates walked out of the proceedings after declaring in ringing rhetorical terms their nonnegotiable commitment to “national self-determination” for the Tamil peo-

S-21

11/24/09

638

4:27 PM

Page 638

CASE STUDIES

ple (to be fair, the government’s delegation was quite unconstructive as well, and an army massacre of Tamil civilians provided the catalyst to the walk-out). Indian attempts during 1986 to mediate a compromise solution, based on some degree of political autonomy for the Tamil-majority north and east, similarly came to nought. By the end of the year, Indian diplomats and politicians involved in those efforts were convinced that the LTTE was an irreconcilable extremist organization whose leadership was constitutionally incapable of any sort of compromise. The LTTE was certainly a strong and independent force by this time, but it too had its limitations. In May–June 1987, the government launched a major military offensive in the north, most of which was by this time effectively under Tiger control. While inflated government claims of decimating the Tigers once and for all were not realized, the offensive did succeed in putting the Tigers on the defensive and reasserting government control over much of the territory and population of the Jaffna peninsula. But the relative, qualified success of the military operation led to a major political and diplomatic problem for Colombo. The significant Tamil civilian casualties from the offensive, as well as the fresh flood of refugees and displaced persons it generated, enraged the population of Tamil Nadu and brought forth stern, pointed warnings from New Delhi to desist from committing “genocide.” To back up its warnings, India dispatched ships to deliver relief supplies to the beleaguered Jaffna civilians; when these were turned back by Sri Lankan gunboats, India sent military aircraft, which violated Sri Lankan airspace, to air-drop the same supplies over Jaffna. In the meantime, the Tigers began to recover from their initial disarray, and retaliated by bombing public places in Colombo, indiscriminately slaughtering noncombatant Sinhalese in the eastern province, and launching devastating suicidebombing raids against newly established military installations in the “pacified” areas of the Jaffna peninsula. Hundreds of Sinhalese military and civilian casualties were the result. By early July, the initial euphoria had evaporated in Colombo as it became increasingly evident that this was an unwinnable war, both because of the tenacity of the domestic military opposition and because of the international political and diplomatic constraints posed by India’s still hostile attitude. Just as realization was dawning on the Colombo government that there was no alternative to coming to terms with India, both because it was the dominant regional power and because it was the only country with significant leverage vis-à-vis the Tamil radicals, the Rajiv Gandhi government was coming under unprecedented pressure in India’s domestic politics. Serious allegations of corruption and cronyism implicating Rajiv and his closest associates surfaced, a damaging split took place in the ruling Congress party with the expulsion of several prominent dissidents, and the Congress suffered severe reverses in elections held in a bellwether north Indian province.

S-21

11/24/09

4:27 PM

Page 639

THE INDO–SRI LANKA PEACE AGREEMENT

639

It was in this context—mounting violence and a military stalemate in Sri Lanka, and “inconvenient domestic developments” for the ruling party and government in Indian politics5—that the “India–Sri Lanka Agreement to Establish Peace and Normalcy in Sri Lanka” was suddenly concluded in Colombo, Sri Lanka’s capital city, on July 29, 1987. The signatories to the accord were Prime Minister Rajiv Gandhi—who, in the words of one critic, “presumed to append his signature on behalf of the island’s Tamil population” 6 —and President Junius Jayewardene of Sri Lanka. No publicly known negotiation or mediation process presaged the signing of the agreement. Not only did no individuals and organizations sign up to the accord on behalf of the Tamils, one of the two belligerent parties to the conflict, but no Tamil organization was even involved in a consultative capacity in the hasty and secretive run-up to July 29. A few days before the agreement was signed in Colombo, Velupillai Prabhakaran, supreme leader of the Tigers, was flown together with several top aides from Jaffna to Delhi in a special Indian air force plane, and kept under virtual house-arrest in a luxury hotel in the Indian capital while Indian officials pressured him to accept and endorse the provisions of the ISPA as the final, definitive redressal to Tamil grievances. From his hotel suite in Delhi, Prabhakaran claimed: If we had been shown this agreement in Jaffna, we wouldn’t have come to India at all. It is only when we came here that we found that India and Sri Lanka had already come to an agreement, which I am now supposed to endorse. It was a calculated plan to persuade me to come so as to give the impression that I am a party to it. . . . This agreement was totally unacceptable to us. It was drafted without consulting the Tamil representatives. I think India has realized that without the LTTE the accord won’t work— that’s why they took the decision to even inform us, albeit at the eleventh hour.7

The conclusion of the agreement was ostentatiously trumpeted within India by the Indian government as a massive triumph of Indian statesmanship and diplomacy, an interpretation almost universally accepted in that country for several months. A few critical voices in India did, however, point out that “the Indian prime minister did have a compelling need to produce a diversion from inconvenient domestic developments, and the Sri Lankan situation was poised just at the point where it could, with Jayewardene’s cooperation, be packaged appropriately.”8 It was correctly argued, furthermore, that Jayewardene had realized that “the political dangers to him from a creeping Indian intervention, which would have been inevitable had the hostilities in the north continued, were greater than those from a one-stroke agreement that at least kept up the appearance of Indian intervention and assistance being at the request of the Sri Lankan government.”9 The motives of the Indian and Sri Lankan governments in burying the hatchet and hurriedly concluding the ISPA were thus fairly obvious. But

S-21

11/24/09

640

4:27 PM

Page 640

CASE STUDIES

there was one glaring missing link—the complete absence of Tamil representation and participation in drawing up an agreement that was supposedly motivated, above all, by the urgent imperative of protecting and advancing the security, rights, and aspirations of the Tamil people of Sri Lanka. Marginal Tamil factions might be persuaded to fall in line through various inducements or arm-twisting (indeed, this turned out to be the case, as both the TULF and the smaller Tamil guerrilla groups accepted the accord, albeit with reservations). But what of the most powerful—and intransigent— Tamil organization, the LTTE, and its concerns? Would the substantive results of the ISPA minimally satisfy the Tigers and make them a retroactive party to the accord? The early phase of implementation would prove decisive in settling this question. In late July and early August 1987, the Tigers were clearly sore at having been excluded from the mediation/negotiation process that culminated in the ISPA. They were also convinced that the agreement failed to minimally satisfy their basic security concerns and political goals. However, they also appeared resigned to implementation in the face of India’s overwhelming military power and apparent determination to end the Sri Lankan war, and Tiger leaders, however incensed, did view with considerable trepidation the prospect of taking on the vast might of the Indian armed forces, which had already begun to arrive in northern and eastern Sri Lanka to implement the accord. The Indians, for their part, badly needed the cooperation of the Tigers—however grudgingly given—in order to be able to implement any semblance of peace on the ground in the war-torn north and east. However, for this window of opportunity to be converted into a viable peace implementation process, the Tigers’ core reservations about the ISPA would have to be taken seriously by the Indians. This meant that certain key elements of the ISPA would have to be redefined—in the process of implementation—in order to at least assuage the Tigers’ core concerns. This was easier said than done, for it would, among other things, further inflame suspicions among the Sinhalese population about the purposes of the Indian intervention in northern and eastern Sri Lanka, and raise issues of Indian compliance with provisions set out in the text of the ISPA. However, given the flawed state of the ISPA from a Tamil perspective—not just in the eyes of the Tigers but in the eyes of more moderate Tamils as well—there was no alternative to such calculated flexibility if the peace agreement were to stand any chance of implementation on the ground.

The Indo-Sri Lanka Peace Agreement: Major Points and Contradictions The ISPA sought to “strengthen the unity, sovereignty and territorial integrity of Sri Lanka,” while “preserving its character as a multi-ethnic,

S-21

11/24/09

4:27 PM

Page 641

THE INDO–SRI LANKA PEACE AGREEMENT

641

multi-lingual and multi-religious plural society in which all citizens can live in equality, safety and harmony, prosper and fulfil their aspirations.” In substantive terms, the agreement temporarily merged the northern and eastern provinces into “one administrative unit,” in apparent implicit deference to the Tamil autonomist claim that those two provinces taken together constituted “areas of historical habitation of Sri Lankan Tamil-speaking people” (this was the wording used in the agreement, avoiding the politically controversial “homeland” characterization), and thus must together form a single self-governing political unit. The merger was temporary, however, pending the results of a referendum to be held in the eastern province by December 31, 1988, in which residents of the east would choose whether to remain part of an integrated northeastern autonomous entity or separate to form their own unit. The outcome of this referendum would, the peace agreement stipulated, be decided by a simple majority of polled votes. This conditional character of the merger of the north and east was unacceptable to the Tamils, including, but not limited to, the Tigers. Because ethnic Tamils composed less than 50 percent of the population of the east, it was practically certain that the merger would be undone once the referendum took place. As Prabhakaran categorically stated soon after his return to Jaffna in early August: “It is not a question of merger of the North and East. It is our homeland. There is no question of any negotiation on this.”10 This view was shared by virtually the entire Tamil political spectrum, ranging from LTTE militants to TULF moderates. Even more critically, the substantive content of the autonomous power to be enjoyed by the northeastern region left much to be desired from the Tamil point of view. The peace agreement stipulated that the body exercising autonomous authority would be a North-Eastern Provincial Council (NEPC), which would be elected preferably within three months, or at the latest by December 31, 1987 (in the hiatus, there would be a transitional administration, whose constitutive procedures and composition were left unspecified in the agreement). The NEPC would have an elected executive authority—a chief minister at the head of a board of ministers—whose activities would be supervised and overseen by a governor, to be appointed by the president of Sri Lanka. However, the quantum of devolution of power from the center to the NEPC that was proposed by the peace agreement was considered entirely unsatisfactory by Tamil opinion in general, and the Tigers in particular. Specifically, the agreement left authority in two crucial spheres—land settlement and maintenance of law and order (i.e., policing)—entirely to the prerogative of the central government, with obvious implications for both short- and longer-term security of the Tamil population of the northeastern region. Moreover, the NEPC itself had no independent fiscal powers—it could neither tax nor borrow—and was completely dependent for its finances on grants to be made at the discretion of the central government.

S-21

11/24/09

642

4:27 PM

Page 642

CASE STUDIES

The NEPC was empowered to enact legislation on a range of mostly minor matters, but even that decisionmaking authority was subject to the overriding veto of the governor, an appointee of the Sinhalese-controlled center. As the polemical but not unfounded critique of a Sri Lankan Tamil constitutional lawyer put it: The Provincial Council will be no more than a glorified local government authority, with power to enact statutes on a few innocuous questions such as probation and childcare services, market fairs, cooperatives, animal husbandry, pawnbroking—and that too subject at all times to the overriding control of the central government. And even in relation to these subjects, the [NEPC] may not pass statutes involving finance except on the recommendation of the Governor.11

Sri Lanka’s most prominent political scientist, Alfred Jeyaratnam Wilson, also wrote in the immediate aftermath of the accord that while “the scheme for devolution of powers is yet to be clarified, in its present form it is far from satisfactory, especially because the central government in Colombo would have ultimate supervision and control.” 12 Indeed, the unimplemented autonomy agreements of 1965 and especially 1958 gave much more wide-ranging and robust autonomy to the Tamils than did the ISPA, and even the word “autonomy” did not occur anywhere in the text of the agreement or in annexes to the main document. Prabhakaran plausibly spoke for the bulk of Tamil opinion when he complained: As they [the Indians] took into consideration only their own interests and hurriedly arrived at this agreement, they have not looked after the grievances of the people who have been affected for so long. . . . Having fought so much, having sacrificed so many lives and having lost 20,000 people— all this has been subordinated to India’s strategic interests. Not only that—we, the representatives of such martyrs, have not been properly respected. . . . India has not given us our due.13

As for demobilization and disarmament of combatants, the peace agreement decreed that all Tamil guerrillas would have to surrender all weapons in their possession to specially designated Sri Lankan authorities within seventy-two hours of the signing of the ISPA. This was an obviously absurd timetable, since the process would involve thousands of guerrillas spread out in a network of camps and bases all over the north and east of the island (moreover, the prospect of “surrendering” to Sri Lankan authorities with whom they had been locked in mortal combat for the preceding several years was anathema to the Tamil militants, a fact apparently unappreciated by the Indians). In return, the Sinhalese president would grant a general amnesty to all Tamil fighters and to all Tamil political prisoners incarcerated in Sri Lankan jails (in many cases under draconian antiterror-

S-21

11/24/09

4:27 PM

Page 643

THE INDO–SRI LANKA PEACE AGREEMENT

643

ism laws), and his government would “make special efforts to rehabilitate militant youth with a view to bringing them back into the mainstream of national life.” Finally, the Sri Lankan army and other state security forces would revert to their pre-May offensive positions and be confined to their barracks for an unspecified time frame. The militant youth were, predictably, not too enthused by the terms of this offer. While still in Delhi in late July, Prabhakaran stated with absolute clarity that “the crucial point on which we are rejecting this Agreement is the point about giving up our arms.” The LTTE leader argued that his movement constituted “the only deterrent force [to state violence]. If this deterrent is removed, our people become defenseless.” He pointed out that the peace agreement included absolutely no guarantees for the longer-term security of the Tamils, once the LTTE had disarmed and demobilized and any Indian peacekeeping forces sent to implement the agreement had withdrawn. This was a genuine issue, for the agreement contained no provision whatsoever for any reciprocal demobilization and disarmament. No ceilings were to be put on the strength of the Sri Lankan state-security forces (including the army and police), there were no clauses to ensure equitable representation of Tamils in those (practically all-Sinhalese) forces, and the agreement said nothing about any scaling-back or dismantling, even of a phased, gradual nature, of the enormous state security apparatus—some 200 military camps and police stations of varying sizes—located in the north and east. To the contrary, while government-sponsored paramilitary groups—deeply implicated in anti-Tamil atrocities, especially in the east— were to be disbanded, their members could be absorbed into “regular” security forces at the discretion of the Sinhalese president. In this situation, it is neither surprising nor illogical that Prabhakaran insisted that “a working arrangement has to be made on the ground that will insure the safety and security of the Tamils. Unless that working arrangement is established, the question of [LTTE] disarming doesn’t arise. It is better to fight and die than surrender the weapons in an insecure environment and die on a mass scale.”14 Toward its end, the main text of the ISPA stated that “in the event that the Government of Sri Lanka requests the Government of India to afford military assistance to implement these proposals, the Government of India will cooperate by giving to the Government of Sri Lanka such military assistance as and when requested.” The final clause of the annex to the agreement further stated that “an Indian peacekeeping contingent may be invited by the President of Sri Lanka to guarantee and enforce the cessation of hostilities, if so required.” What this diplomatic coyness sought to obscure was that in reality, a large-scale Indian military intervention to enforce the cease-fire and implement the ISPA was a foregone conclusion. There was simply no other party

S-21

11/24/09

644

4:27 PM

Page 644

CASE STUDIES

that could or would take on the role of implementer. Thus, the Rajiv Gandhi government, having precipitately signed the accord for its own selfserving reasons on behalf of the Tamils without having even given a prior hearing to the concerns of their representatives, had essentially saddled India with the sole responsibility for implementation as well. Indeed, as early as the morning of July 30, thousands of Indian army troops began to arrive by air transports in northern Sri Lanka to enforce and implement the settlement. They came to be known as the IPKF. Under the terms of the ISPA, the government of India now had exclusive responsibility for monitoring and enforcing compliance to the agreement. “If any militant groups operating in Sri Lanka do not accept this framework of proposals for a settlement,” the agreement stated, India “will take all necessary steps to insure that Indian territory is not used for activities prejudicial to the unity, integrity and security of Sri Lanka.” Concrete measures to this effect specified in the agreement and ancillary documents included collaboration between the Indian and Sri Lankan navies and coast guards to interdict guerrilla movements between Tamil Nadu and northern Sri Lanka, deportation by the Indian government of “Sri Lankan citizens [on Indian soil] found to be engaging in terrorist activities or advocating secessionism,” and even more remarkably, Indian provision of “training facilities and military supplies for Sri Lankan security forces.” There could not be a more radical transformation of the Indian role in Sri Lanka’s war. As Jayewardene gloated, “the major gain [of the government side from the ISPA] is that [Tamil] terrorism is over. . . . India is [now] willing to tackle this terrorist problem as an active partner with me. . . . Earlier, they were training the terrorists.”15 Ironically, Prabhakaran’s analysis, narrated to a 100,000-strong mass meeting organized by the Tigers in Jaffna in early August, was strikingly similar in its basic perspective and conclusions: This Agreement is primarily concerned with Indo-Sri Lankan relations. It also contains the principles and requirements for making Sri Lanka accede to India’s big-power orbit. That is why the Indian government showed such an extraordinary keenness for concluding this Agreement. However, at the same time it also happens to be an Agreement that determines the political future and fate of the people of Tamil Eelam. . . . When a great power has decided to determine our destiny in a manner that is essentially beyond our control, what are we to do? What is happening now is this. Sri Lanka and India have concluded an Agreement. The Indian Army is here and is asking for our weapons. If we don’t give them up, we will have to fight the Indian Army. We do not want that. To avert the calamitous circumstance of clashing with the Indian army, we accepted these arrangements. We love India. We love the people of India. But we have not abandoned our political objective. My beloved people. . . . The forms of struggle may change, but the goal of our struggle is not going to change. Let me make it clear to you here, beyond the shadow of a doubt, that I will continue to fight for the objective of attaining Tamil Eelam.16

S-21

11/24/09

4:27 PM

Page 645

THE INDO–SRI LANKA PEACE AGREEMENT

645

Tamil suspicions that their interests and aspirations had been sold down the river in exchange for strategic concessions by Sri Lanka to India were reinforced by a number of clauses attached to the main agreement. These essentially stipulated that Sri Lanka would expel foreign mercenaries and military instructors from countries deemed unfriendly by India, would not allow the eastern port of Trincomalee to be used in any way “prejudicial to Indian interests” but would instead develop it in cooperation with the Indian government, and would not permit intelligence agencies of countries considered hostile by India to operate from or on Sri Lankan territory.17 If the main Tamil political organization regarded the transformed Indian role as unsatisfactory bordering on treacherous, much of Sinhalese mass opinion regarded it as the culmination of a pattern of arrogant, imperialistic meddling in Sri Lankan affairs. Many Sinhalese neither shared nor understood the reasons for Jayewardene’s pragmatic rebirth vis-à-vis India, the country reviled by the government till a few weeks prior to the ISPA as the neighborhood bully and essential backer of the terrorists out to wreck Sri Lanka. Distrust of Indian intentions and motives ran deep among the Sinhalese public, subjected for years to propaganda about hegemonic Indian expansionism’s purported designs on their country. These fears were sharply aggravated by the sudden, stealthy manner in which the ISPA was concluded, and by the immediate large-scale deployment of Indian troops on the island’s territory as a result. Many Sinhalese consequently viewed the new developments as some sort of capitulation by the Colombo government to India’s hegemonic ambitions and, by extension, to the separatist agenda of India’s agents, the Tamil terrorists. The irony could not be greater, but such was the confusion generated by the remarkable reversal of Indian policy. The degree of popular Sinhalese resentment against the ISPA and the arrival of the IPKF in the north soon became clear. On July 29, the Colombo area had had to be placed under curfew to forestall demonstrations protesting the signing ceremony. On July 30, as Rajiv Gandhi was inspecting a military guard of honor at the Colombo airport prior to his return to Delhi, a Sinhalese sailor who was part of the honor guard attempted to assault the Indian leader with his rifle-butt. Jayewardene’s own prime minister, Ranasinghe Premadasa, who was to succeed him to the executive presidency in 1989, publicly expressed disquiet about the manner in which the peace agreement had been finalized and particularly the Indian military presence it entailed. A few weeks later, Jayewardene survived an assassination attempt when a grenade was lobbed into a meeting of the Sri Lankan cabinet. In the months that followed, the Janatha Vimukthi Peramuna (JVP; People’s Liberation Front), a Sinhalese youth movement that combined messianic ultranationalism with a near-totalitarian ideology of social revo-

S-21

11/24/09

646

4:27 PM

Page 646

CASE STUDIES

lution, exploited this climate in the Sinhalese regions of the country to launch a violent uprising against the UNP government (the movement had tried the same thing once earlier, in 1971; that insurrection was eventually crushed with massive loss of life among its young participants). After July 29, the JVP had a potent, if xenophobic, single-point platform for its activities: the ejection of foreign troops from Sri Lanka. By 1988, the hitherto relatively peaceful Sinhalese regions were gradually sinking into anarchy— with schools and universities (traditional hotbeds of JVP mobilization) closed down, the economic infrastructure increasingly paralyzed by strikes and terrorist violence, and members of the JVP’s armed wing and the government’s army and police forces slaughtering each other, and uninvolved innocents, by the thousands in an escalating intra-Sinhalese civil war that for sheer savagery surpassed even the conflict in the northeast.

The Breakdown of Implementation: From Peacekeeping to War-Making In contrast to the hostility in the Sinhalese south, however, the first IPKF contingents arrived in the Tamil north to a warm welcome from the warweary local population. Many Tamils had for years been eagerly anticipating an Indian intervention on their behalf—on a model similar to that of Turkey’s in Cyprus in 1974—and they thought the moment had finally come. A common sight throughout the north and east in August 1987 was of crowds of civilians cheering newly arrived Indian soldiers as they drove through local communities escorted by Tiger guerrillas, often in vehicles flying both India’s national tricolor and the LTTE’s flag, which depicts the head of a ferociously snarling tiger set against a red background. In other words, while the Indian intervention in Sri Lanka had very significant flaws and potential problems, it was not without its strengths and assets either. A major asset was the goodwill toward India in general and the IPKF in particular that existed at this point in time, in the immediate aftermath of the ISPA, among the Tamil people of the north and east. This population, for the time being at least, knew little about the specifics of the ISPA. Most ordinary Tamils were simply delighted that India had finally sent its troops en masse to deliver them from the oppression of the SinhalaBuddhist state and the violence of its army and other security forces. This reservoir of goodwill among Tamil civilians toward the Indian presence was a factor the Tigers could not discount or ignore. The Indians had other, not inconsiderable, advantages. The Tigers, while bitter for the reasons discussed above, also appeared resigned at some level to the IPKF intervention. The movement’s leader, while expressing unhappiness over the manner in which the ISPA had been concluded as well as grave reservations about key clauses relating to the

S-21

11/24/09

4:27 PM

Page 647

THE INDO–SRI LANKA PEACE AGREEMENT

647

Tamils’ basic security and political future, had also publicly called “calamitous” the idea of the Tigers engaging in military confrontation with the vastly superior forces of a “great power” like India. This reality—the overwhelmingly preponderant nature of the regional hegemon’s power—had also been instrumental in securing the acquiescence of the Sinhalese regime in Colombo to the IPKF intervention. Having called on the Indians, however reluctantly, to broker and implement a settlement, the beleaguered Jayewardene government was as resigned as the Tigers to having to accommodate the IPKF for the foreseeable future. However, in order to increase prospects that this initial leverage vis-àvis both belligerents would translate into a successful peacebuilding role, India’s implementation strategy needed to come to terms with the very real deficiencies and gaps in the ISPA. Above all, the Indians needed to recognize that the security and other concerns raised by Prabhakaran were legitimate, and find ways to meaningfully address them without mortally alienating the Sinhalese government (as I have mentioned, this may have been a tall order, but there was no alternative if the peace agreement were to stand realistic chances of implementation on the ground). Gaining the cooperation—however grudgingly given—of the Tigers was the first and most important requisite for a viable implementation process. This would mean concessions on the major points raised by the Tigers, and incentives to bring them on board as the IPKF’s de facto partner in the implementation process. First and foremost, given the genuinely iniquitous character of the ageement’s demobilization and disarmament provisions, as well as the absurdity of the seventy-two-hour timetable for the surrender of guerrilla weapons, it was both pointless and impossible for the IPKF to insist on a literal interpretation of these clauses early in the implementation process. That would have led to an immediate breakdown in IPKF-LTTE relations, given the Tigers’ adamant refusal to disarm without effective short- and long-term security guarantees for themselves and for the Tamil population of the war zones (which in turn would necessitate a renegotiation of the relevant clauses of the agreement to make them more balanced). In fact, the IPKF, having just arrived in unfamiliar terrain, was in no position to enforce LTTE compliance with the ISPA’s disarmament and demobilization provisions. The Tigers handed in approximately 15 percent of their formidable arsenal of light and medium weapons (they claimed 85 percent), including many obsolete and even completely nonfunctional pieces, in symbolic “surrenders” amid much publicity in the first half of August 1987. Once the disarmament issue had effectively been put in cold storage in this manner, the issue of the transitional administration for the north and east stipulated by the peace agreement came into the spotlight. Putting the Tigers in charge of this transitional administration might have been a useful incentive to encourage their grudging cooperation with the IPKF. At the least, it might have put the implementation process on a somewhat more

S-21

11/24/09

648

4:27 PM

Page 648

CASE STUDIES

secure footing in this delicate early phase by improving the stakes for the Tigers in the unfolding game. From a longer-term perspective, there is little doubt that successful peace implementation could only have been enabled by a renegotiation of key elements of the agreement—specifically those pertaining to disarmament and demobilization of combatants, the quantum of self-rule to be enjoyed by the Tamil autonomous authority governing the northeast, and the issue of that authority’s control over the eastern province—to better meet Tamil security concerns and political aspirations. Whether the Colombo regime would have been willing or able to acquiesce to such a renegotiation is of course an open question, given both the apparent finality of the ISPA on these issues and the growing opposition among the Sinhalese people to any Indian role that involved the stationing of tens of thousands of Indian troops on part of the island’s territory. However, in August–September 1987, these issues—making demobilization and disarmament of combatants gradual and reciprocal, improving and enlarging the scope of Tamil autonomy, and making the merger of the east with the north permanent rather than temporary, with suitable safeguards for the nonTamil population groups in the eastern province—were too large and complex to be tackled all at once. The immediate imperatives were to secure the cessation of hostilities between the Sinhalese armed forces and the guerrillas, and to establish a transitional administration that could move the implementation process forward incrementally—a step at a time (elections to constitute the NEPC, the autonomous Tamil body, were supposed to be held by December 31, 1987). Instead, control of the transitional administration proved to be the issue that generated a catastrophic decline in relations between the IPKF and the LTTE, precipitating a breakdown of the implementation process by early October. The Tigers demanded virtually monopolistic control over the transitional administration for the north and east as their price for cooperating with the IPKF. The Indians, who wanted to include some members from both the nonviolent TULF and the small anti-LTTE guerrilla groups, were reluctant to concede this demand, and negotiations on the constitution of the interim administration quickly stalled. In response, the Tigers mobilized several thousand committed civilian supporters and launched a civil disobedience campaign against the IPKF at a number of locations in the north and east in early September. At one such demonstration, nervous Indian soldiers shot and killed a civilian protester. In the second half of September, tension escalated dangerously in the Tamil areas when a twenty-two-year-old LTTE leader went on indefinite hunger strike in Jaffna demanding an end to Sinhalese “colonization” of the eastern province, permanent closure of all Sri Lankan military installations in the north and east, and establishment of the transitional administration on the LTTE’s terms. When he died from thirst in late September, the Tigers had their martyr, and public opinion in

S-21

11/24/09

4:27 PM

Page 649

THE INDO–SRI LANKA PEACE AGREEMENT

649

the Tamil areas began to turn decisively against the Indian government and the IPKF. At the beginning of October, the Indians reacted to the mounting pressure by conceding the LTTE demand for exclusive control of the transitional authority, but it was too late—events were spinning out of control. While the Indians’ desire to ensure that a plurality of Tamil groups were represented in the transitional administration was quite understandable, the imperative of moving the implementation process forward implied that they should have conceded the LTTE’s conditions on this issue much earlier. The LTTE was, after all, the only force that really mattered on the ground. Moreover, the Indians’ stonewalling on this issue for almost two months fatally aggravated the suspicions of the already resentful Tiger leadership. While negotiations over the establishment of the interim administration remained paralyzed in August and September, sporadic outbreaks of violence were proliferating on the ground, especially in the east. The main culprits were the Tigers, who were busy securing their supremacy in the eastern province by attacking Sinhalese communities as well as anti-Tiger Tamil militants based there. But the catalyst that ignited full-scale hostilities between the LTTE and the IPKF was the arrest of seventeen senior Tigers by the Sri Lankan navy off the northern coast in early October. The Tigers argued that the detention was in violation of the general amnesty granted to all militant cadres under the ISPA, and appealed urgently to the Indians to intervene and have the captives released. The Indians, annoyed by this point with what they viewed as the Tigers’ incessant tactical games, failed to respond with alacrity. The Sinhalese authorities, for their part, maintained that several of the captives—who included key Tiger field commanders—were wanted for leading massacres of Sinhalese civilians, especially in the east, as well as on charges of organizing terrorist bombings in Colombo. On October 5, as the prisoners were being loaded onto a Sri Lankan plane at an air force base on the Jaffna peninsula en route to Colombo for interrogation and incarceration, all seventeen consumed the capsules of potassium cyanide that all Tigers wear around their necks and almost always consume when faced with capture. Twelve died immediately. The following day, Prabhakaran repudiated the cease-fire, the Tigers executed eight Sinhalese prisoners of war they had in their custody, and LTTE units drastically escalated the killing spree against noncombatant Sinhalese in the eastern districts. They also ambushed and brutally killed a group of about a dozen Indian soldiers in the Jaffna area. In response, on October 10, thousands of IPKF troops, backed by tanks, artillery, and helicopter gunships, launched a massive offensive to disarm and destroy the LTTE in its stronghold, the Tamil heartland of the Jaffna peninsula. The peacekeepers had been transformed into combatants. As one Indian observer acidly commented, “what is now happening in Sri Lanka is known in the language of commerce as sub-contracting. The Sri Lankan President has subcontracted the job of restoring law and order in his country to our Prime Minister.”18

S-21

11/24/09

650

4:27 PM

Page 650

CASE STUDIES

The Indians grievously underestimated their adversaries in almost all respects—an extraordinary failure of intelligence explained partly by the equally extraordinary level of hubris of the Indian political elite and its diplomatic and military staff, partly by the abrupt, disorienting reversal of policy the ISPA represented, and partly by lack of communication and coordination between different Indian agencies, such as between the intelligence agency RAW and the regular army, whose units composed the IPKF. Far from being a motley gang of hoodlums, as they were portrayed by the Indian government, the LTTE was by this time a sophisticated politicomilitary movement whose young members had tremendous faith in their leader and in the justice of their cause and who, despite brutal propensities, were regarded as heroic resistance fighters by a large proportion of the Tamil population. Indian commanders initially estimated that the Tiger hunt would take four to six days; at the end of it, they confidently asserted, the backbone of the LTTE would have been broken permanently. What actually unfolded was four weeks of heavy fighting across the Jaffna peninsula between some 35,000 Indian troops and about 2,500 superbly organized and motivated LTTE fighters employing a dexterous mixture of conventional and guerrilla warfare tactics. The city of Jaffna (population 200,000) fell to Indian numbers and firepower at the end of October, but the cost was hundreds of Indian fatalities, most of them shot by snipers or blown up by land mines. By early November, large-scale fighting was dying down in the Jaffna peninsula, but the LTTE was far from dead. Its top leaders had almost all escaped death or capture. They, together with the most experienced fighters, simply decamped from the peninsula to jungle bases in the forested mainland of the northern province, where they successfully withstood successive Indian assaults over the next two years. The rest of the Tigers, meanwhile, simply melted into the civilian population of Jaffna to fight another day, while Tiger forces in the east opened up a new theater for a war of attrition against the IPKF. The greatest asset of the IPKF intervention in Sri Lanka in its early phase had been the goodwill of the Tamil civilian population. This asset was callously squandered during the operation to subdue the Tigers. The Indian offensive was marked by often indiscriminate mortar and artillery shelling, and strafing by helicopter gunships that caused hundreds of civilian casualties, as well as summary executions of Tamil civilians by frustrated Indian soldiers unattuned to combat with an elusive insurgent foe, and, perhaps worst of all, a large number of rapes of local Tamil women. This meant that after October 10, the IPKF would increasingly come to be looked upon as an army of occupation by most ordinary Tamils. That perception deepened into abiding hatred in the following two years as the IPKF, aided by collaborator militia raised from the remnants of the antiTiger guerrilla factions, accumulated an ignoble record of human rights abuses against the Tamil population of the north and east.19 Sustained by

S-21

11/24/09

4:27 PM

Page 651

THE INDO–SRI LANKA PEACE AGREEMENT

651

growing anti-Indian feeling, the LTTE was able “to carry on a ghostly little war—a war of midnight ambushes, mine explosions and sniper fire— fought in the classic guerrilla pattern codified by Mao Zedong; the guerrilla takes like a fish to water, becomes part of the people.”20 By the time the end to the IPKF misadventure came in 1990, the LTTE, defying all odds and expectations, were both militarily stronger and much more numerous (thousands of young people, including many women, joined during the IPKF years (August 1987–March 1990) in reaction to Indian repression), as well as more popular and legitimate in Tamil society than they had been in 1987. Indeed, the tenacity and courage of the LTTE’s resistance evoked grudging admiration even among many Sinhalese, who otherwise regarded the Tigers as devils incarnate. After the watershed represented by October 1987, the chances of implementation of the ISPA were next to nonexistent, as the energies of the IPKF—whose combat strength rose to 70,000 by early 1988 and peaked at 105,000 in early 1989—were almost entirely consumed by the Vietnamstyle guerrilla war being waged throughout the north and east by the LTTE.21 Despite some overtures in late 1987 and early 1988 for a cease-fire by the LTTE—the Tigers were, after all, themselves severely extended and in some need of a respite—the Indian government failed to change its approach to the quagmire in which it had landed itself. That would have meant deescalating offensive military operations and reopening serious talks with the Tigers, though it remains an open question whether the Tigers and their estranged mentor could have been reconciled in any lasting fashion after what had transpired in the interregnum. But while the Tigers, with their military capability still potent and their popular base still intact (indeed, expanding and deepening), remained antagonistic and at large, the realistic prospects of implementation of the ISPA were virtually nil. In the autumn of 1988, the Indians made one last, desperate attempt to resurrect the ISPA and marginalize the Tigers, by holding elections to constitute the autonomous NEPC. This proved counterproductive, and actually delivered the last nail into the coffin of the ISPA. The only Tamil participants—in addition to some Muslim and Sinhalese parties in the eastern province—were two anti-LTTE former guerrilla groups whose members had acquired a particularly unsavory reputation as the IPKF’s local torturers and executioners. Members of these groups, whose sole political principle seemed to be a pathological loathing of the Tigers, were “elected unopposed” in most instances and placed by the Indians in charge of an NEPC that, not surprisingly, proved as farcical during the next year and a half as the “election” that had brought it into being. It expired more or less simultaneously with the IPKF withdrawal from Sri Lanka in early 1990, and its leaders fled to India. The sorry episode proved once again that the IPKF’s fitful efforts at piecemeal implementation of the ISPA were a nonstarter without at least a rapprochement with the major Tamil movement, the

S-21

11/24/09

652

4:27 PM

Page 652

CASE STUDIES

Tigers, who denounced the elections as a farce and urged a popular boycott. The end-game of the ISPA/IPKF episode was played out during 1989. In February, Ranasinghe Premadasa succeeded Junius Jayewardene to the executive presidency of Sri Lanka. As president, Premadasa, who had openly harbored reservations about the IPKF intervention from the outset, immediately found himself confronted with a truly dire situation. In the north and east, the Tigers had clearly succeeded in stalemating—that is, from the point of view of the guerrillas, checkmating—close to 100,000 soldiers of the Indian army. In other words, Jayewardene’s overriding motive for agreeing to the ISPA—the hope that the Indians would if necessary do the Sinhalese elite’s dirty work for them and crush Tamil insurgency—had not been realized at all. But what made matters far worse was that the IPKF intervention had set off a vicious civil war among the Sinhalese. By early 1989, that intra-Sinhalese conflict had spiraled into desperate proportions, turning the Sinhalese regions of the island into a slaughterhouse, and the successful overthrow of the government by the JVP insurrectionists seemed an increasingly plausible possibility. In short, the ISPA had brought disaster to Sinhalese and Tamils alike, and to the fractured island of Sri Lanka as a whole. The logical course for the Premadasa government was to try and get rid of the IPKF as quickly as possible, since that alone would take the steam out of the JVP insurgency and create a breathing space for the government. In April, Premadasa initiated a strategy to respond to this imperative by offering direct, unconditional dialogue between his government and the country’s two major rebel movements, the LTTE and the JVP. The offer was directed primarily at the JVP leadership, which, in keeping with its reckless and impetuous character, refused it outright and stepped up its campaign of violence against government targets and forces. But the Tigers, after some initial hesitation, accepted, and in May 1989 talks commenced in Colombo between senior government ministers and LTTE representatives. The interests of the government and the Tigers converged on one crucial point—an urgent need to bury the ISPA and get rid of the IPKF. Hence, it should not have come as a surprise to New Delhi when on June 1, 1989, Premadasa used the occasion of a major Buddhist religious festival to declare that he desired that the last Indian soldier be out of Sri Lanka by July 29, the second anniversary of the ISPA. The vast majority of Tamils as well as the antigovernment Sinhalese had already consigned the ISPA to the dustbin of history, and now the Sinhalese establishment was doing the same. A total withdrawal from Sri Lanka by the July 29 deadline would have been a disastrous humiliation for the Rajiv Gandhi government in an election year—parliamentary elections were due in India by late 1989. So the Indian government browbeat the Sri Lankans into settling for a token withdrawal of about 600 IPKF troops by that date.

S-21

11/24/09

4:27 PM

Page 653

THE INDO–SRI LANKA PEACE AGREEMENT

653

Nonetheless, it was clear that an ignominious end was looming for Pax Indica in Sri Lanka. From the autumn of 1989 onward, fighting between the IPKF and the LTTE deescalated considerably, and on March 31, 1990, the last units of the Indian expeditionary force departed Sri Lanka’s shores. The cost in human terms was heavy—1,155 Indian soldiers had been killed in action and over 3,000 seriously injured. The final Indian fatality occurred on March 31, when an officer belonging to the last departing Indian contingents was shot dead by a Tamil sniper while boarding his ship in Trincomalee harbor. The LTTE had lost 711 of its members confronting the Indians,22 but membership in the Tiger movement more than tripled between 1987 and 1990, with almost 10,000 fighters mobilized by the time of the Indian withdrawal. These fighters moved rapidly to establish LTTE control over the north and east as the Indians pulled out. Between December 1989 and March 1990, one Tamil population center after another—Batticaloa and Trincomalee in the east and, above all, the Jaffna peninsula including the city of Jaffna in the north—were triumphantly recaptured by hardened Tiger guerrillas returning from forest and rural bases, in conjunction with local LTTE underground groups. By April 1990, the entire northeastern region, barring a few pockets, was effectively under LTTE administration. The Tigers also augmented their arsenal considerably in these final months. Since mid-1989, they had surreptitiously received weapons and supplies to fight the Indians from their old enemy, the Sri Lankan armed forces, under orders from the Premadasa government. Now they captured large quantities of arms and ammunition left by the Indians to their Tamil collaborator militias, who collapsed and disintegrated virtually without a fight as the IPKF withdrew. The worst sufferers of the ISPA/IPKF episode, however, were the Tamil and Sinhalese civilian populations. Between 5,000 and 8,000 civilian Tamils (estimates vary) died in the violence in the north and east between October 1987 and March 1990, mostly at the hands of the IPKF—which earned the sobriquet “Innocent People Killing Force” as a result—and its armed Tamil collaborators (the LTTE also killed sizable numbers of Tamils opposed to it and those viewed by the Tigers as “collaborators” for one reason or another). Between 25,000 and 60,000 Sinhalese (estimates vary) perished in the government-JVP war in the country’s south that resulted directly from the ISPA and the IPKF intervention. Between 80 and 90 percent of these were suspected JVP members or sympathizers (usually young people) massacred by the army and police forces. The rest were functionaries at various levels of the government and the ruling party, the UNP, as well as army and police personnel, murdered by the JVP. By the end of 1989, the JVP insurrection—poorly prepared, organized, and armed in comparison to that of the Tamils, lacking the same wide base of domestic popular support as well as international networks, and prone to eventually self-destructive bouts of uncontrolled, indiscriminate violence—had been largely crushed

S-21

11/24/09

654

4:27 PM

Page 654

CASE STUDIES

by the government. Almost the entire JVP leadership was physically eliminated in the process, and the ranks of its committed cadres were decimated. The India-Sri Lanka Agreement to Establish Peace and Normalcy in Sri Lanka thus ended up turning the whole country—Sinhalese and Tamil regions alike—into a killing field.

Conclusion: The Lessons and Legacy of the ISPA It is far from certain that an alternative implementation strategy, building on the strengths and assets of the IPKF intervention while acknowledging and seriously addressing its weaknesses and loopholes, could have made a success of the ISPA. But it is clear that the failure to even attempt such a strategy eventually saddled the Indians with the worst of both worlds— alienating the Tamils without mollifying the Sinhalese. Why did the Indian government persist in pursuing such a blundering course? Three factors were crucial, and they serve to underscore some of the potential pitfalls that could derail peacemaking interventions by regional powers in the internal conflicts of smaller neighbors. First, in June–July 1987 the Rajiv Gandhi government did not have a coherent, considered strategy for imposing peace in Sri Lanka. Its sudden, ambitious peacemaking intervention in the island imbroglio was, rather, motivated by a pressing domestic consideration that had very little if anything to do with the conflict in Sri Lanka. This was the need of the Congress party and the government to shore up flagging popularity, and deflect public attention from various scandals and failures, by means of a foreign policy “triumph.” Thus, India’s civilian leadership committed its military forces to police the northeast and implement a problematic settlement to the Sri Lankan civil war without any serious, prior assessment of the complexities of Sri Lanka’s own domestic political situation, the risks involved, and, most important, how to respond in the contingency of blockage or breakdown of implementation, should the agreement unravel. Second, once India was involved, its implementation strategy between late July and early October suffered from a strong hubris syndrome. An inflated sense of regional paternalism led the Indians to presume that they could impose the ISPA as it stood on the Tigers without any substantive modifications in the process of implementation. This was a grave miscalculation. The fault lay primarily not with the military leaders of the IPKF but with the politicians and diplomats who had been instrumental in sealing the peace agreement and who continued to determine the basic parameters of implementation policy. In myopia born of arrogance, these officials failed to consider the long history and strong local roots of Sri Lankan Tamil nationalism, as well as the formidable organizational infrastructure, military capability, and popular base of the LTTE, which was by that time

S-21

11/24/09

4:27 PM

Page 655

THE INDO–SRI LANKA PEACE AGREEMENT

655

unquestionably the major vehicle of that (now radicalized) nationalism. During the critical first two months of the implementation phase, Indian policymakers continued to treat the LTTE not as a serious, determined political movement whose cooperation with the IPKF was vital if the peace agreement were to be implemented on the ground, but as a “militant group,” one among several. Thus, not only was no effort made to accommodate the Tigers’ concerns and “plug” the deficiencies in the ISPA, but the problem was aggravated by the Indian decision to deny the LTTE control of the transitional authority. Conceding the LTTE’s demands on this point could have served as a most useful incentive to placate the Tigers and incorporate them into the implementation process. They were eventually conceded in early October, but by then it was too late, since IPKF-LTTE relations had been poisoned and were already on the brink of breakdown. Indian policymakers neglected, moreover, to consider that while the LTTE had certainly benefited from direct and indirect forms of Indian support during the 1983–1987 period, it was not quite India’s “client” organization. Unlike the other, smaller Tamil guerrilla groups, the Tigers not only had preserved their independent state of mind, but also had been careful not to become solely dependent on Indian sponsorship. By 1987 the Tigers had developed their own training facilities in areas under their control in the northeast, and were meticulously building their own weapons supply and procurement networks, drawing on the global arms market and the resources of the expanding global diaspora of expatriate and refugee Tamils. In an interesting interview given to an Indian newsmagazine in June 1986, Prabhakaran (then based in Madras, the capital city of the Tamil province in south India) had stated that while “India’s support is important to us, we also have a right to self-determination. We don’t think Rajiv [Gandhi] will ask us to pack our bags. But then, if we are asked to go we will have to go. Our homeland is always waiting for us with open arms.”23 Despite this growing confidence, the Tigers were still, of course, a guerrilla movement whose strength seemed puny when compared to the vast military resources the Indian state could mobilize. But their relative autonomy from Indian control meant that the Tigers certainly had enough strength to wreck the implementation of any peace deal that they considered fundamentally detrimental to their minimum interests. Third, once the IPKF became embroiled in armed conflict with the LTTE from October 1987, Indian military strategists profoundly underestimated the military capabilities and stamina of their uneasy ally turned adversary. In purely conventional terms, of course, the Tigers were far weaker than the IPKF. But this amounted to missing the point. The LTTE was not a conventional fighting force, and it did not wage a conventional form of warfare. It engaged, instead, in “protracted people’s war,” the mode of warfare codified into a systematic operational doctrine by Mao Zedong and deployed to telling effect by guerrillas against numerically and conven-

S-21

11/24/09

656

4:27 PM

Page 656

CASE STUDIES

tionally superior opponents in Algeria, Vietnam, and Afghanistan. Tiger commanders demonstrated a superb tactical sense of this mode of warfare, and the movement’s popular support grew further after October because the IPKF found it difficult to hunt the Tigers without also victimizing the Tamil civilian population on a large scale. Thus even after losing control of Jaffna and other urban centers to superior Indian numbers and firepower, the Tigers possessed a degree of resilience and staying power that was simply unappreciated in the mind-set of most Indian military planners, oriented as they were to conventional war. Indian field commanders directing operations against the Tigers gradually realized the problem and the almost insurmountable barriers it posed to any prospects of successful implementation of the ISPA,24 but even then the diplomats and political leaders remained unconvinced. Thus, a prominent Indian strategic thinker declared in November 1987 that “the bravado of the Tamil Tigers in continuing guerrilla warfare need not constitute a serious threat to the implementation of the Indo-Sri Lankan agreement,”25 while in December, Prime Minister Gandhi dismissed the LTTE as “a small outfit of 1500 to 2000 persons that represents none but itself.”26 The cumulative effect of these three factors—each crucial at different junctures—helps explain India’s rigid, blundering strategy, which degenerated by late autumn of 1987 into a counterproductive coercion-intensive approach to peace implementation and brought about the virtual collapse of the ISPA. The failure of the ISPA has contributed in two important ways to the ethnic bloodbath that continued unabated in northern and eastern Sri Lanka between 1990 and 2002, and steadily intensified in both scale and horror in a manner unimaginable even in the bloody 1980s. First, the Indian intervention between July 1987 and March 1990 helped create—in direct contrast to what it had been intended to achieve— an LTTE movement stronger both numerically and materially, and more self-confident than ever about the legitimacy and viability of its extremely radical, zero-sum interpretation of the Tamil cause. Having successfully stood down the vastly superior forces of the Indian state and consolidated its own position as the sole spokesman of the Tamil “nation” in the process, the LTTE had few qualms about breaking off its tactical truce with the Colombo government and resuming the war with renewed ferocity in June 1990. “Eelam War II” (as the 1990–1994 phase of the conflict is known in Tiger parlance) proved much more bloody than even “Eelam War I” (1983– 1987). A brief glimmer of hope for a negotiated peace appeared in the second half of 1994, when the UNP lost both presidential and parliamentary elections for the first time in seventeen years and a new Sinhalese government assumed office promising unconditional talks with the Tigers and genuine power sharing with the Tamils through radical reforms of the institutional structure of the Sri Lankan state. That hope proved to be a mirage, as the

S-21

11/24/09

4:27 PM

Page 657

THE INDO–SRI LANKA PEACE AGREEMENT

657

Tigers proved as irreconcilable as ever and resumed hostilities with a vengeance in April 1995.27 “Eelam War III” (1995–present) far surpassed the previous two phases in the scale and intensity of violence. Since 1990, 17,000 Tigers and at least 25,000 government soldiers have lost their lives, and the Tigers have assassinated, one after another, most of Sri Lanka’s top Sinhalese politicians (including President Premadasa) and many of its ranking army, navy, and air force commanders. They have also repeatedly carried out devastating suicide-bombing attacks against central areas and key government installations in Colombo, killing hundreds; and thousands of Tamil, Muslim, and Sinhalese civilians have been killed by the warring parties in the northeast. Though both sides suffered setbacks in the 1990s, and a final, total victory for either belligerent was impossible, it was only in 2002 that the parties begrudgingly accepted an offer from the Norwegian government to mediate. Until the Norwegians stepped into the breach, Sri Lanka lacked a plausible potential third-party mediator, another factor contributing to the intractability of the conflict. This is the second negative legacy of the ISPA/ IPKF episode. India would have been the natural candidate for that role, both because India is the South Asian region’s dominant power with an interest in seeing a protracted violent conflict in its backyard resolved successfully, and because of the close cultural affinities and historical links between India and Sri Lanka, particularly true of the Tamil people of Sri Lanka. That potential role, that of a genuine and effective peacemaker, was abused and squandered by the Indian state because of the arrogant manner in which a flawed agreement like the ISPA was sought to be implemented after July 1987—not through a calibrated policy of concessions, incentives, and ultimately, perhaps, selective renegotiation of key elements backed by the threat of Indian military power to deter continued noncompliance by one or both belligerents, but essentially by military power alone. It is instructive that although the Sri Lankan belligerents seem incapable of reaching any sort of compromise on their own, there has been no talk of reviving any mediatory role for India in the conflict. India’s credibility among all Sri Lankans, Sinhalese, and Tamils was destroyed by the events of 1987–1990. A Tamil civilian in Trincomalee expressed well in March 1990 this one point of Sinhalese-Tamil consensus as he watched the last IPKF units depart: “We don’t know what the future will bring, but we are glad the Indians have gone.”28 Most internationally brokered peace agreements are probably imperfect compromises between competing imperatives, and most are probably “imposed” to some degree on one or more of the parties to the conflict. However, it is also probably true that whether such settlements will hold at all or come apart at the seams during attempted implementation (which usually happens sooner rather than later) depends on the agreement crossing at least a minimum threshold of acceptability for all the significant belligerent parties. If the peace agreement as negotiated is for one reason or

S-21

11/24/09

4:27 PM

658

Page 658

CASE STUDIES

another deficient in this key requisite for prospects of successful implementation (as was the ISPA), then it is vital that the implementation strategy be fashioned in a way that compensates for the deficiency by at least attempting to address the core concerns of the dissatisfied party or parties. In demanding virtual capitulation from a well-organized, heavily armed, popularly based, and deeply committed nationalist movement like the LTTE without meaningfully redressing any of the Tigers’ core concerns about the very real gaps in the ISPA in meeting the security needs and political aspirations of the Tamils, the implementation of the ISPA clearly failed to do so. In the process, the blundering regional hegemon not only humiliated itself, but helped set the stage for intensified carnage on the divided island in the wake of its own retreat.

Notes 1. While Tamils composed only 13 percent of Sri Lanka’s total population at this time (and Sinhalese 74 percent), they made up 70 percent of the population of Sri Lanka’s northern and eastern provinces—94 percent in the north, 44 percent in the east; and 73 percent of all Sri Lankan Tamils lived somewhere in the north or east of the country at this time. 2. The total Tamil population of the north and east at this time was 1.4 million (1 million in the north, 400,000 in the east). 3. For a detailed account of the political history of Sri Lanka between the late colonial period and 1983, see Sumantra Bose, States, Nations, Sovereignty: Sri Lanka, India, and the Tamil Eelam Movement (London: Sage, 1994), chap. 2. 4. Rajiv Gandhi, interview, India Today (New Delhi), June 30, 1987, p. 28. 5. Editorial, “Beyond the Tamil Issue,” Economic and Political Weekly (Bombay), August 1, 1987, p. 1261. 6. “A.M.,” “Calcutta Diary,” Economic and Political Weekly, October 31, 1987, p. 1843. 7. Velupillai Prabhakaran, interview, India Today, August 15, 1987, p. 17. 8. “Beyond the Tamil Issue,” p. 1261. 9. Ibid. 10. Velupillai Prabhakaran, interview, Frontline (Madras, India), August 22–September 4, 1987, pp. 113–115. According to the 1981 census, the east was 44 percent Tamil, 32 percent Muslim, and 24 percent Sinhalese. Tamils were an absolute majority—72 percent—in only one of its three districts, Batticaloa, located in the center of the province. The crucial eastern district adjoining the Tamildominated north, Trincomalee, was only 35 percent Tamil. This demographic mix in itself made the status of the east problematic, but there were two further complicating factors. First, a sizable proportion of the east’s Sinhalese population had been settled there since independence under state-sponsored “colonization” schemes deliberately intended, or so Tamils claim, to undermine the Tamil concept of an integral northeastern homeland by engineering changes in the ethnic population ratios. Second, the social identity and political preferences of the East Coast Muslims, the second-largest community in the province, have historically been fluid and indeterminate. Many eastern Muslims, linguistically and culturally close to the Tamils (who are predominantly Hindu, with a large Christian minority) supported

S-21

11/24/09

4:27 PM

Page 659

THE INDO–SRI LANKA PEACE AGREEMENT

659

the Tamil federalist posture in the 1950s and 1960s; in the 1970s and 1980s, however, their support was successfully courted by the central government and the Sinhalese political parties. It was therefore fairly certain that the bulk of the Muslims would vote against making the merger permanent in the proposed referendum. Because the east is geographically and politically such hotly contested terrain, it is no coincidence that many of the worst atrocities of the Sri Lankan civil war, perpetrated by all sides, have occurred in the east. 11. Nadesan Satyendra, The Indo-Sri Lanka Accord and the Thirteenth Amendment to the Sri Lankan Constitution (London: LTTE, UK Branch, 1988), p. 12. 12. Alfred Jeyaratnam Wilson, The Breakup of Sri Lanka: The Sinhalese-Tamil Conflict (London: C. Hurst, 1988), p. 229. Wilson is also a Tamil, and is the son-inlaw of S. J. V. Chelvanayakam, leader of the Tamil autonomist cause from the founding of the Federal Party in 1951 till his death in 1977. Wilson also acted as an intermediary in abortive talks on autonomy between the TULF and the Jayewardene government in the late 1970s and early 1980s. 13. Prabhakaran, interview, Frontline, August 22–September 4, 1987, p. 115. 14. Prabhakaran, interview, India Today, August 15, 1987, p. 18. 15. Junius Jayewardene, interview, India Today, August 31, 1987, p. 44. 16. The text of this speech—Prabhakaran’s first-ever public address and the only one to date—is reproduced in Mohan Ram, Sri Lanka: The Fractured Island (New Delhi: Penguin, 1989), pp. 147–149. 17. My analysis of the substance and implementation of the ISPA in this article draws heavily from Bose, States, Nations, Sovereignty, chap. 4. 18. “A.M.,” “Calcutta Diary,” p. 1843. 19. The assassination of Rajiv Gandhi in Tamil Nadu by LTTE members in mid-1991 was almost certainly propelled by motives of revenge. 20. Ram, Sri Lanka, p. 69. 21. On IPKF numbers, see Ravi Rikhye, “How Many Troops in Sri Lanka?” Economic and Political Weekly, March 25, 1989. 22. These are the official casualty figures released in April 1990 by the IPKF and the LTTE, respectively. 23. Prabhakaran, interview, India Today, June 30, 1986, p. 74. 24. See, for example, the memoirs of the onetime IPKF commander in Jaffna, Lieutenant-General S. C. Sardeshpande, Assignment Jaffna (New Delhi: Lancer, 1992). 25. K. Subrahmanyam, writing in Frontline, November 14–27, 1987, p. 72. 26. Gandhi said this while addressing a public rally in Madras; quoted in N. Seevaratnam, ed., The Tamil National Question and the Indo-Sri Lanka Accord (New Delhi: Konark, 1989), p. 192. 27. On this period and the breakdown of hopes for peace, see Sumantra Bose, “Tamil Self-Determination in Sri Lanka: Challenges and Prospects,” Economic and Political Weekly, September 24, 1994, reprinted in Lanka Guardian (Colombo), October 15, 1994; and Sumantra Bose, “War and Peace in Sri Lanka: The Government’s Reform Proposals and Beyond,” Economic and Political Weekly, September 30, 1995. 28. Quoted in Anita Pratap’s report in Time, April 2, 1990, p. 33.

S-21

11/24/09

4:27 PM

Page 660

S-22

11/24/09

4:28 PM

Page 661

PART 4 Conclusion

S-22

11/24/09

4:28 PM

Page 662

S-22

11/24/09

4:28 PM

Page 663

22 Policy Implications STEPHEN JOHN STEDMAN

Studies of implementation suggest that the likelihood of any policy being performed and yielding expected results depends on whether: (1) objectives are well articulated, clear, consistent, and underpinned by adequate causal theory; (2) the policy is legally structured to enhance the compliance of implementing agents and target groups; (3) implementing officials are skilled and committed; (4) the policy is supported by interest groups and sovereigns; (5) the socioeconomic and political environment surrounding implementation is stable; and (6) the target of the policy is highly vulnerable.1 It is a wonder then that any peace agreement in civil war is successfully implemented. Peace agreements tend to be vague documents; the commitment of warring parties is often ambivalent at best or tactical at worst. Policy direction, especially when it involves the UN Security Council, is usually couched in rhetorical flourish or diplomatic code. Rarely, if ever, are there specific legal remedies when signatories to such agreements fail to comply. Implementing officials vary markedly in their skill and commitment. Peace is often actively opposed and undermined by factions inside and outside the warring country. Small changes in power relationships and isolated events outside the warring country can radically transform the implementation environment. And many warring factions seem beyond the reach of whatever carrots or sticks international actors dangle in front of them. And yet . . . peace happens. Sometimes. This book has sought to understand the success and failure of international attempts to implement peace agreements in civil wars. It was prompted by a desire to address a life-or-death issue for millions of people: how to end civil wars and ensure that peace is not another broken promise 663

S-22

11/24/09

664

4:28 PM

Page 664

CONCLUSION

in a litany of death and destruction. We began dissatisfied with the existing literature on peace implementation, which has failed to answer several key questions: • What is it that international actors actually do to assist the successful implementation of peace agreements in civil wars? • How do cases of civil war differ in the challenges that they pose to implementers of peace agreements? • What strategies are available to international implementers and which are the most appropriate in different circumstances? • What cases of peace implementation are likely to receive adequate attention, resources, and strategy? • Which of the various tasks of peace implementation are the most important to overall implementation success? We found that cases of peace implementation differ dramatically in terms of the difficulty of the implementation environment and of the willingness of international actors to provide resources and risk troops. These differences are predictable before a peace operation begins. These two findings mark a dramatic advance in our understanding of peace implementation in three fundamental ways. First, our results put to rest glib generalizations about peace operations based on one or few cases. To put it bluntly, the results suggest that there is no reason to assume that actions and strategies that work in a more benign conflict environment such as Guatemala or Namibia will work in a much more demanding environment such as Bosnia or Sierra Leone. Second, our results imply that implementation strategies must be designed based on the level of difficulty of the case. In certain limited situations, strategies that derive from traditional peacekeeping (with its underlying emphasis on confidence-building) can be effective. In more challenging situations, however, when predation coexists with fear, confidencebuilding will prove inadequate, and implementers will need to compel and deter to ensure compliance with a peace agreement. Third, the results raise the fundamental issue of what economists refer to as “incentive compatibility,” or what we commonly know as “political feasibility”: that strategies must be in the self-interest of critical actors in order to be implemented. Difficult implementation environments require more resources, greater international involvement, and more coercive strategies, but often such resources, involvement, and strategies are not forthcoming because no major or regional power perceives peace in a given country to be in its own vital strategic interest. The problem of incentive compatibility usually is subsumed under arguments that a lack of political will is the problem—that if only more will were found, then tougher cases would receive the needed attention and resources. But the emphasis on

S-22

11/24/09

4:28 PM

Page 665

POLICY IMPLICATIONS

665

political will misleads: it takes a relatively fixed variable—perception of vital security interest of regional and major powers—and treats it as if it were easily manipulated. Our study argues that this is a vexing analytical error that overstates the commitment of international actors to making peace in civil wars in countries of peripheral security importance. Beyond providing a differentiated understanding of implementation environments and strategies, we scrutinized claims that are often made about the importance of various implementation subgoals and their role in the overall success and failure of peace accords. Such claims grew in prominence in the 1990s, as various international nongovernmental organizations lobbied publics and governments, insisting that their single issue of concern—whether it be disarmament, elections, human rights, or refugee repatriation—was crucial to implementation success. Two major findings emerge from an examination of subgoals and the overall success of implementation. First, in terms of what can be achieved in the fulfillment of any subgoal, the desires of implementers must be commensurate with allocated resources and permissible strategies. For example, one could argue that postagreement elections should only be held under optimal conditions, or insist that every peace accord must include provisions for full accountability and prosecution for past atrocities and war crimes, or demand that all refugees be repatriated to their original homes. In the absence of commitment of adequate resources and troops, however, ambitious standards for subgoals are symbolic statements of virtue, not practical means of terminating wars. Second, in terms of the degree of investment in subgoals, priority should be given to the demobilization of soldiers and the demilitarization of politics, that is, the transformation of warring armies into political parties. Without achieving these two critical subgoals, 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. We also identified two low-cost opportunities that should be pursued during implementation: (1) civilian security through police and judicial reform, and (2) local capacity-building for human rights and reconciliation. Although the study did not discover a single case of failed implementation that resulted from a failure to pursue these opportunities, we found that the potential long-term benefits of security reform and developing local capacities for peacebuilding warrant the relatively inexpensive investments that such measures require. Several policy recommendations follow from our study. Because the United Nations remains the instrument of choice of states who advocate international implementation of peace agreements, we pitch our recommendations toward that international organization. The findings and advice, however, should resonate with member-state governments who ultimately

S-22

11/24/09

4:28 PM

Page 666

666

CONCLUSION

debate and decide which missions will be deployed by the United Nations, the level of resources that the missions will receive, and the set of strategies that are available to the implementers. Since there are occasions when individual member-states or regional coalitions eschew UN deployment and choose to take a lead in international implementation, the advice offered here can also assist them in their planning for missions. Our recommendations cluster along three dimensions: (1) the capabilities that are needed for international actors to succeed in peace implementation, (2) strategic aspects of implementation, and (3) priorities to be undertaken during peace implementation.

Policy Recommendations Capabilities Our study demonstrated that the greater the difficulty of the implementation environment, the greater the likelihood that peace implementation will fail. The two most important environmental sources of failure are the presence of spoilers—factions or leaders who oppose the peace agreement and use violence to undermine it—and neighboring states that oppose the peace agreement and assist the spoilers. Given the importance of these findings, the United Nations needs to improve its capacities for strategically assessing the implementation environment, particularly the motives, intentions, and capabilities of peace parties and spoilers. A third environmental determinant of implementation is the presence of valuable spoils. Our study found that no peace agreement has been fully successfully implemented where there are valuable, easily marketed commodities such as gems or timber. In countries where there are easily marketed valuable commodities (spoils), implementers should have a strategy, the resources, and the commitment to counter those who profit from war. Strategic Aspects of Implementation Peace implementation involves two types of strategic decisions: what cases will be implemented and, given a decision to implement, what mix of confidence-building and coercion will be used to seek compliance. Both of these decisions must reflect a prior judgment about the likelihood that given the level of difficulty of the environment, adequate resources will be available to succeed. Our research shows that cases of peace implementation differ in terms of the willingness of international actors to provide resources and risk troops. The most important variable is whether a major or regional power

S-22

11/24/09

4:28 PM

Page 667

POLICY IMPLICATIONS

667

sees assisting the war-affected country as vital to its national security interest; only when such interest is present has peace implementation succeeded in the most difficult environments. Without great or regional power interest, the United Nations can only succeed in the least difficult environments. When selecting what peace agreements the UN should implement, great power or regional power interest should be treated as a hard constraint. That major and regional power interest plays a key role in implementation success can create incompatibilities between the strategies and resources that are needed for success and the incentives of the major powers to support those strategies. As the difficulty of the implementation environment increases, there is a need for greater scope and assertiveness of the transitional authority that is supplied by international actors. Similarly, the more difficult the environment, the greater the need for coercive strategies of implementation. But, the strategies and resources available to international implementers are also a function of great and regional power interest. Unless their security interests are engaged, the resources and commitment necessary for coercive strategies to succeed will not be forthcoming. Without great or regional power interest, the United Nations should not implement the hard cases. A second incentive compatibility issue concerns mechanisms for providing strategic coherence and coordination. The more difficult the implementation environment, the greater the need for strategic coordination. When international actors lack unity, spoilers can take advantage to attack the peace process and threaten peacemakers. Similarly, the more coercive the strategy, the greater the civil-military tensions in implementation, and the greater the need for strategic coordination. But again, strategies available for international coordination, such as Friends groups, are also a function of great and regional power interest. The willingness of states to join Friends groups is indicative of a prior judgment that the specific case is in the state’s interest. Such mechanisms for coordination will not always be available: there are no “Friends of Somalia,” as there are no great or regional powers who have security interests in a peaceful Somalia. The gap between what is needed in some missions and what major or regional powers are willing to provide leads to organizational pathologies within the United Nations. In some cases, the willingness of the Security Council to authorize peace missions stems from a perception that the case will be safe and easy; where threats are perceived, the reaction of the Security Council is to cut and run. Aware of this, UN officials are reluctant to share worrisome conflict assessments and tend to “ask only for what the traffic will bear.” The need to present optimistic scenarios to the Security Council precludes realistic contingency planning, since this signals that more dire scenarios are possible. Given that most missions face unanticipated violent challenges during implementation, the UN is faced with little

S-22

11/24/09

4:28 PM

Page 668

668

CONCLUSION

more than a choice of inaction or withdrawal, both of which betray the hopes of those who want peace and humiliate those who seek to implement peace. The Security Council should demand as the basis of any authorization of a mission detailed contingency planning in the event that best-case scenarios fail to materialize. When missions are deployed the United Nations must improve the day-to-day assessment and analysis of information in missions, and incentives must be created for UN officials to communicate that assessment to the Security Council. Prioritizing Subgoals in Implementation Our project assessed the contributions of various subgoals (e.g., demobilization, disarmament, elections, human rights, refugee repatriation) to overall implementation success. We felt that “multidimensional peacekeeping,” although an accurate description of many implementation missions, does not constitute a strategy of implementation, as it simply recommends that implementers do a lot of things. Given resource limitations, implementers need to prioritize among subgoals, so that resources are best spent to end the war and provide a basis for future peacebuilding. This is especially the case when the subgoals of implementation are subject to public and NGO lobbying and possible standards of success can be unattainable. It is not just that pursuing many goals is more costly than pursuing a few, but that attempting to achieve unrealistically high standards of success for any one subgoal can also be infinitely expensive. Our project found that demobilization of soldiers is the single most important subgoal for implementation success. When allocating investment to subgoals, priority should be given to demobilization of soldiers and demilitarization of politics, that is, the transformation of warring armies into political parties. Unless demobilization is achieved, civil wars cannot be brought to an end, and other goals—be they democratization, accountability, or justice—have little chance of success. Implementers should also invest in civilian security, through police and judicial reform, and local capacity-building for human rights and reconciliation. The potential long-term benefits of security reform and local capacity-building for peacebuilding warrant the relatively inexpensive investments that such measures require.

Final Reflections The two worst humanitarian emergencies of the 1990s—Angola in 1993 and Rwanda in 1994—followed the failure of international actors to implement peace agreements. In those countries, far more people died in the aftermath of failed peace implementation than had died from the preceding

S-22

11/24/09

4:28 PM

Page 669

POLICY IMPLICATIONS

669

years of civil war. When combatants in civil wars sign a peace agreement, there is potential for progress: long-standing wars with staggering costs can be brought to an end and people gain the opportunity to rebuild shattered lives and societies. But the potential for harm is also great: fear and predation are potent threats to peace. As our project demonstrates, however, appropriate strategies and adequate resources can help to counter these threats. Our findings provide prospective implementers with useful knowledge. Employing the findings of our research, they will be better equipped to gauge the likely difficulty of the implementation environment and to make accurate assessments of whether necessary resources and correct strategies will be provided. They can also improve the allocation of the resources they have at their disposal. Our findings also help policymakers to better identify the gaps in how peace agreements are currently being implemented—in information gathering, analysis, and sharing, and in contingency planning. This said, our findings also present a daunting dilemma: to the extent that past behavior predicts future behavior, there will be more occasions where implementers face highly difficult environments with the knowledge that commensurate resources will not be forthcoming. Throughout the 1990s, the signing of a peace agreement was sufficient to trigger international involvement to implement that agreement, but nowhere near sufficient to trigger international commitment to do the job right. Our research began with the ethical concern of how best to save lives and end deadly civil wars; it concludes with a more circumspect but more daunting challenge: how best to save lives and end deadly civil wars in a world where there are limited resources and multiple civil wars—some of which pose far greater challenges to prospective implementers than others. We urge would-be implementers to remain attentive to this dilemma, to the fact that major and regional powers treat different civil wars differently, such that their commitment to making peace in war-torn countries is much greater when their motives are based on vital security interests, rather than on humanitarian concerns alone. We are not against intervention to end civil wars, but we are decidedly against failed interventions to end civil wars, the costs of which loom greater than the suffering engendered by the wars themselves. The case studies in this book of Liberia, Rwanda, and Sri Lanka suggest that the suffering and tragedy of noncombatants in war are gravely compounded by interventions on the cheap. Much is made of the lack of political will by states as an explanation for inaction in the face of humanitarian tragedy. Less is made of an equally problematic tendency: the lack of political will to eschew symbolic salves for gaping wounds on the international body politic. If, as two of our authors contend, the UN Secretariat has acted as used-car salesmen instead of Consumer Reports by indiscriminately recom-

S-22

11/24/09

670

4:28 PM

Page 670

CONCLUSION

mending cases of peace implementation, it has been able to do so because of a Security Council that acts like the only consumer in the world who wants to be ripped off. While the former has at least shown signs of recognizing the price of failure incurred by such a strategy, 2 the latter tends to act as if it will always be able to distance itself from the performance of the organization. To put it somewhat differently, if the admonition to the Secretariat to stop telling the Security Council what it wants to hear is to have any effect,3 then something must be done to change the incentives of member states to gain the benefits of symbolic action (intervention on the cheap) while deflecting the costs of failed interventions to the organization. The recommendation to avoid hard cases where there is little prospect of adequate resources or strategy worries some who believe that it will lead to a further marginalization of those areas of the globe peripheral to the major powers. Yet, it is hard to see how stopping further marginalization is promoted by half-hearted, failed interventions. As we have made abundantly clear in this book, the alternative to no implementation in these cases is not implementation adequately resourced and conceived, but implementation grudgingly funded, woefully deployed, and shackled by inadequate strategy. Moreover, the selection criteria for missions that we suggest are based on degree of difficulty, which is region-blind. Cases with less difficult environments should be chosen and then resourced to succeed. If violent challenges arise in such missions, it becomes all the more important that they are resisted and overcome. If we could imagine a history of intervention in the 1990s where outsiders had rationalized their choices of intervention cases, there never would have been a mission to Somalia, an implementation environment so challenging that 28,000 U.S. marines could not make peace. But there would have been an effective mission to Rwanda, where a much less challenging implementation environment could have been managed with much fewer troops and resources. The irony, of course, is that we would live in a world marred by inaction to save thousands of lives in Somalia, and ignorant of the successful prevention of hundreds of thousands of deaths in Rwanda.4 *

*

*

A final word concerning the vital security interests of major and regional powers: Does our analysis risk objectifying and hardening what is a socially constructed and fluid concept? We think not. What is deemed a vital national interest is highly contested, but is not vulnerable to instantaneous reassessment or the whims of fad and fashion. We can hope for a time when the Congo and its millions of noncombatant deaths is perceived as a life-ordeath issue for policymakers in Washington, New York, and Brussels, but we would be deluded to believe that day is close. Indeed, even where peace implementation intersects with the time-bound concerns of national securi-

S-22

11/24/09

4:28 PM

Page 671

POLICY IMPLICATIONS

671

ty—geography and alliances—Western policymakers are only grudgingly willing to risk blood and treasure. Implementing peace in Bosnia is more closely tied to the security interests of the United States and NATO than implementing peace in Rwanda, which explains why more than $20 billion dollars have gone to implementing the Dayton Accords, while only $35 million dollars were spent implementing the Arusha Accords. But anyone who has fought the battle to make peace in Bosnia a priority of the United States knows that its importance pales in comparison to what the Pentagon truly believes to be vital for U.S. security. It is tempting to conclude this book with a reference to the changed world of international security after September 11, 2001, and the enlarged space for more expansionist and less traditional definitions of the national security interest. The world does seem smaller and the problems of wartorn peripheral countries loom larger than ever before. But the unwillingness of the Bush administration to countenance even a minimal peace implementation force for Afghanistan, a case that scores off the charts in terms of difficulty and where consolidating peace seems so clearly in the vital security interest of the United States, gives sobering pause.

Notes 1. Joanna Spear, Carter and Arms Sales: Implementing the Carter Administration’s Arms Transfer Restraint Policy (New York: St. Martin’s Press, 1993), pp. 1–19. 2. See the Report of the Panel on United Nations Peace Operations, A/55/305, S/2000/809, (New York: United Nations, 2000). For a discussion of the so-called Brahimi Report, named after the chair of the panel, see Stephen John Stedman, “The Future of U.N. Peacekeeping Operations: The Brahimi Report and Its Discontents,” in Diana Ayton-Shenker and John Tessitore, eds., A Global Agenda: Issues Before the 56th General Assembly of the United Nations (Lanham, MD: Rowman and Littlefield, and the United Nations Association of the United States of America, 2001), pp. 1–7. 3. Ibid. 4. It is easy to find estimates of lives saved by the U.S. led intervention in Somalia running to the hundreds of thousands. There remains, however, only one scientific study of the various interventions in Somalia and their effects on Somali mortality: Steven Hansch, Excess Mortality and the Impact of Health Interventions in the Somalia Humanitarian Emergency (Washington DC and Atlanta: Refugee Policy Group and Centers for Disease Control and Prevention, 1994). This study concludes that the U.S.-led intervention of December 1992 likely prevented 10,000–25,000 deaths, an important achievement, but one with magnitudes smaller than that ascribed by various commentators and one that has to be weighed against the failure of the United States and the United Nations to implement the Arusha Accords and stop the genocide in Rwanda. The only scholarly assessment of the Somali intervention based on these figures is Thomas G. Weiss, Military-Civilian Interactions: Intervening in Humanitarian Crises (Lanham, MD: Rowman and Littlefield, 1999), pp. 69–96.

S-22

11/24/09

4:28 PM

Page 672

S-Acronyms

11/24/09

4:29 PM

Page 673

ACRONYMS

ABiH ACC ACDL AFL AFRC AHA AHLC AI AMNLAE ANC ANEP ANL ANSESAL ANSP ARENA ARSK ASC ASEAN ATC BANRURAL BBTG BiH BLDP

Army of Bosnia-Herzegovina Administrative Committee on Coordination (UN) Association for Constitutional Democracy in Liberia Armed Forces of Liberia Armed Forces Revolutionary Council (Sierra Leone) Africans for Humanitarian Action (Ethiopia) Ad Hoc Liaison Committee Amnesty International Association of Nicaraguan Women Luisa Amanda Espinosa (Nicaragua) African National Congress National Association of Private Enterprise Army for National Liberation (Kosovo) Salvadoran Security Agency National Public Security Academy/National Public Safety Academy (El Salvador) Alianza Republicana Nacionalista (National Republican Alliance, El Salvador) Army of the Republic of Serb Krajina Civil Society Assembly (Guatemala) Association of Southeast Asian Nations Rural Workers Association (Nicaragua) Rural Development Bank (Guatemala) Broad-Based Transitional Government (Rwanda) Bosnia-Herzegovina Buddhist Liberal Democratic Party (formerly KPNLF, Cambodia) 673

S-Acronyms

11/24/09

4:29 PM

674 BMATT BOS CACIF CCF CCPM CCPP CD CDF CDR CDS CEDECON CEH

CFE CGDK CHAD CIA CIAV CIDA CIMIC CIREFCA

CISAC CIVPOL CIVS CMF CMO CNE ComIFOR COPAZ COPMAGUA

CPD

Page 674

ACRONYMS

British Military Advisory and Training Team Southern Opposition Bloc (Nicaragua) Coordinating Committee of Commercial, Industrial, and Financial Associations (Guatemala) Cease-Fire Commission (Mozambique) Joint Political-Military Commission (Angola) Permanent Commissions of Guatemalan Refugees (Mexico) Democratic Convergence (El Salvador) Civil Defense Force (Sierra Leone) Coalition pour la Défense de la République (Coalition for the Defense of the Republic, Rwanda) Sandinista Defense Committees (Nicaragua) Center for the Defense of the Constitution (Guatemala) Historical Clarification Commission (Commission for the Historical Clarification of Human Rights Violations and Acts of Violence That Have Caused Suffering to the Guatemalan Population) Conventional Forces in Europe Coalition Government of Democratic Kampuchea (Cambodia) Conflict Humanitarian Affairs Department (UK) Central Intelligence Agency International Commission of Support and Verification (Nicaragua) Canadian International Development Agency civil-military cooperation Conferencia Internacional sobre Refugiados Centroamericanos (International Conference on Central American Refugees) Center for International Security and Cooperation Civilian Police (UN) International Commission for Verification and FollowUp (Nicaragua) Commonwealth Monitoring Force (Zimbabwe) chief military observer (UN) National Elections Commission (Mozambique) IFOR commander National Commission for the Consolidation of Peace (El Salvador) Coordinación de Organizaciones del Pueblo Maya de Guatemala (Coordinator of Mayan Peoples’ Organizations of Guatemala) Peace and Development Committee (Guatemala)

S-Acronyms

11/24/09

4:29 PM

Page 675

ACRONYMS

CPP CRC CRPC CRS CSC CSO CST CVDC D&D DAC DCG DFI DFiD DHA DMZ DNI DP DPA DPKO DRC EASC EBRD ECOMOG ECOWAS EGP EP5 EPRDF EPS ERD ERP ESAF ESCAP ESI EU FAA FADM FAL FALA

675

Cambodian People’s Party Central Revolutionary Council (Liberia) Commission for Real Property Claims of Displaced Persons and Refugees (Bosnia) Catholic Relief Services Supervisory and Monitoring Commission (Mozambique) civil society organization Sandinista Workers Federation (Nicaragua) Voluntary Civil Defense Committee (aka PAC, Guatemala) disarmament and demobilization Development Assistance Committee (OECD) Guatemalan Christian Democracy Defense Force of Independence (Sierra Leone) Department for International Development (UK) Department of Humanitarian Affairs (UN) Demilitarized Zone National Intelligence Directorate (El Salvador) displaced person Department of Political Affairs (UN) Department of Peacekeeping Operations (UN) Democratic Republic of Congo Election Appeals Sub-Commission (Bosnia) European Bank for Reconciliation and Development Economic Community of West African States CeaseFire Monitoring Group Economic Community of West African States Guerrilla Army of the Poor (Guatemala) Extended Permanent Five (UNTAC) Ethiopian People’s Revolutionary Democratic Front Sandinista People’s Army (Nicaragua) Emergency Response Division (UNDP) Popular Revolutionary Army (El Salvador) El Salvador’s Armed Forces Economic and Social Commission for Asia and the Pacific European Stability Initiative European Union Angolan Armed Force Forças Armadas de Defensa de Moçambique (Mozambican Defense Force) Armed Forces of Liberation (El Salvador) Forças Armadas para a Liberação de Angola (Armed Forces for the Liberation of Angola)

S-Acronyms

11/24/09

4:29 PM

676 FAO FAPLA FAR FAR FDN FDNG FDR FGT FMLN FNLA FP FPL Frelimo FRG FSLN FUNCINPEC G-7 G-8 GCE GCLW GDP GFAP HCAVM HDZ HRCC HRW HVO IAG ICCPR ICESCR ICG ICITAP ICORC ICRC ICT ICTR

Page 676

ACRONYMS

Food and Agriculture Organization Forças Armadas Populares de Liberação de Angola (People’s Armed Forces for the Liberation of Angola) Forces Armées Rwandaises Revolutionary Armed Forces (Guatemala) Nicaraguan Democratic Forces New Guatemala Democratic Front Revolutionary Democratic Front (El Salvador) Guillermo Torriello Foundation (Guatemala) Farabundo Martí National Liberation Front (El Salvador) Frente Nacional de Libertação de Angola (Angola National Liberation Front) Federal Party (Sri Lanka) Popular Liberation Forces (El Salvador) Front for the Liberation of Mozambique Guatemalan Republican Front Sandinista National Liberation Front (Nicaragua) National United Front for a Cooperative, Independent, Neutral, and Peaceful Cambodia Group of Seven (Canada, France, Germany, Greece, Italy, Japan, the United States) Group of Eight (G-7 plus Russia) Spanish Civil Guard General Confederation of Lebanese Workers gross domestic product General Framework Agreement for Peace Higher Council of the Audio-Visual Media (Lebanon) Croatian Democratic Union Human Rights Coordination Center (Bosnia) Human Rights Watch Croation Defense Force InterAfrica Group International Covenant on Civil and Political Rights International Covenant on Economic, Social, and Cultural Rights International Crisis Group International Criminal Investigations Training Assistance Program International Committee on the Reconstruction of Cambodia International Committee of the Red Cross international criminal tribunal ICT-Rwanda

S-Acronyms

11/24/09

4:29 PM

Page 677

ACRONYMS

ICTY IDB IDF IDP IEBL IFAD IFI IFOR IFP IGCC IGNU IHRO IISS ILO IMC IMF INPFL IOM IPA IPKF IPTF IRA IRIN ISPA IUDOP JNA JVP KFOR KISAN KLA KPNLF KVM LADE LBCI LCHR LDC LDF LF LNC LNM LNTG

677

ICT-Yugoslavia Inter-American Development Bank Israel Defense Forces internally displaced person Inter-Entity Boundary Line (Bosnia) International Fund for Agricultural Development international financial institution Implementation Force Inkatha Freedom Party (South Africa) Institute on Global Conflict and Cooperation (University of California) Interim Government of National Unity (Liberia) international human rights organization International Institute for Strategic Studies International Labor Organization Independent Media Commission (Bosnia) International Monetary Fund Independent National Patriotic Front of Liberia International Organization for Migration International Peace Academy Indian Peace-Keeping Force International Police Task Force Irish Republican Army Integrated Regional Information Network (UN) Indo-Sri Lanka Peace Agreement Instituto Universitario de Opinion Publica (El Salvador) Yugoslav National Army Janatha Vimukthi Peramuna (People’s Liberation Front, Sri Lanka) Kosovo Force (NATO) Kus Indian Sut Aslika Nicaragua (Union of Coastal Indians of Nicaragua) Kosovo Liberation Army Khmer People’s National Liberation Front (Cambodia) Kosovo Verification Mission Lebanese Association for the Democracy of Elections Lebanese Broadcasting Corporation International Lawyers Committee for Human Rights least developed country Lofa Defense Force (Liberia) Lebanese Forces Liberian National Conference Lebanese National Movement Liberian National Transitional Government

S-Acronyms

11/24/09

4:29 PM

678 LPC LTTE LURD MDC MDR MILCOM MINUGUA MINURSO MINUSAL MISURA MISURASATA MK MLN MNF MONUC MPLA MRND

MSU MTV NAC NATO NBN NDF NEPC NERIC NERP NGO NIF NMOG NORAD NP NPP NPFL NPRAG NRA

Page 678

ACRONYMS

Liberia Peace Council Liberation Tigers of Tamil Eelam (aka Tamil Tigers, Sri Lanka) Liberians United for Reconciliation and Democracy more developed country Mouvement Démocratique Républicain (Democratic Republican Movement, Rwanda) Military Component (UN) UN Verification Mission in Guatemala UN Mission for the Referendum in Western Sahara UN Mission in El Salvador Miskitu, Sumu, Rama (Nicaragua) Miskitu, Sumu, Rama, and Sandinistas Together (Nicaragua) Umhkonto we Sizwe (Spear of the Nation, South Africa) National Liberation Movement (Guatemala) Multinational Force (Haiti) UN Organization Mission in the Democratic Republic of the Congo Movimento Popular de Libertação de Angola (Popular Movement for the Liberation of Angola) Mouvement Révolutionnaire National pour le Développement (National Revolutionary Movement for Development, Rwanda) Multinational Specialized Unit Murr Television (Lebanon) North Atlantic Council North Atlantic Treaty Organization National Broadcasting Network (Lebanon) Namibian Defense Force North-Eastern Provincial Council (Sri Lanka) National Election Results Implementation Commission (Bosnia) National Emergency Recovery Program (Lebanon) nongovernmental organization Neutral International Force (Rwanda) Neutral Military Observer Group (OAU) Norwegian Agency for Development Cooperation National Party (South Africa) National Patriotic Party (Liberia) National Patriotic Front of Liberia National Patriotic Reconstruction Assembly Government (Liberia) National Resistance Army (Rwanda, Uganda)

S-Acronyms

11/24/09

4:29 PM

Page 679

ACRONYMS

NRP NUF OAS OAU OCHA OCI ODA OECD OED OHCHR OHR OIE ONUC ONUCA ONUMOZ ONUSAL ONUV ONUVEN ORDEN ORPA OSCE OTI PAC PAN PARMEHUTU PAT PD PDC PDC PDK PEC PGT PIC PIFWC PIN PL PLO

679

National Reconstruction Program (Liberia) National United Front (Cambodia) Organization of American States Organization of African Unity Office for the Coordination of Humanitarian Affairs (UN) Open Cities Initiative (Bosnia) Overseas Development Agency (UK) Organization for Economic Cooperation and Development Operations Evaluation Department (World Bank) Office of the High Commissioner for Human Rights (UN) Office of the High Representative (UN) State Intelligence Office (El Salvador) UN Operation in the Congo UN Observer Group in Central America UN Operation in Mozambique UN Observer Mission in El Salvador UN Verification Office UN Observer Mission for the Verification of Elections in Nicaragua National Democratic Organization (El Salvador) Organization of the People in Arms (Guatemala) Organization for Security and Cooperation in Europe Office of Transition Initiatives (USAID) Voluntary Civil Defense Committee (aka CVDC, Guatemala) National Advancement Party (Guatemala) Parti du Mouvement de l’Émancipation des Bahutu (Movement for the Emancipation of the Hutu, Rwanda) Transitory Auxiliary Police (El Salvador) Democratic Party (El Salvador) Christian Democratic Party (El Salvador) Parti Démocrate Chrétien (Democratic Christian Party, Rwanda) Party of Democratic Kampuchea Provisional Election Commission (Bosnia) Guatemalan Workers Party Peace Implementation Council person indicted for war crimes Peace Implementation Network Parti Libéral (Liberal Party, Rwanda) Palestine Liberation Organization

S-Acronyms

11/24/09

4:29 PM

680 PMA PN PNC PR PRK PRODERE PROTIERRA PRRP PRTC PSD PSP PSP PSW PTSD RAW Renamo RN ROE RPF RRTF RS RTLMC RUF SANDF SDA SDR SDS SE Accord SEPAZ SFOR SIDA SIU SLA SLFP SMC SMEs SNC SOC

Page 680

ACRONYMS

Policía Militar Ambulante (Mobile Military Police, Guatemala) National Police National Civilian Police proportional representation People’s Republic of Kampuchea (Cambodia) Development Program for Displaced Persons, Refugees, and Returnees Inter-Institutional Commission for the Development and Strengthening of Land Property Rights (Guatemala) Priority Reconstruction and Recovery Program (World Bank) Workers Revolutionary Party (El Salvador) Parti Social Démocrate (Social Democratic Party, Rwanda) problem-solving process Progressive Socialist Party (Lebanon) problem-solving workshop post-traumatic stress disorder Research and Analysis Wing (India) Mozambique National Resistance National Resistance (El Salvador) rules of engagement Rwandan Patriotic Front Reconstruction and Return Task Force (Bosnia) Republika Srpska Radio Télévision Libre des Mille Collines (Rwanda) Revolutionary United Front (Sierra Leone) South African National Defense Force Party of Democratic Action (Bosnia) special drawing right Serbian Democratic Party Agreement on Social and Economic Aspects and the Agrarian Situation (Guatemala) Secretariat for Peace (Guatemala) Stabilization Force Swedish International Development Agency Special Investigative Unit (El Salvador) Sierra Leone Army Sri Lanka Freedom Party Standing Mediation Committee (ECOWAS) small and medium enterprises Supreme National Council (Cambodia) State of Cambodia

S-Acronyms

11/24/09

4:29 PM

Page 681

ACRONYMS

SRN SRSG SWAPO SWAPOL SWATF TNA TSE TULF UEA UK ULIMO ULIMO-J ULIMO-K UN UNAMIC UNAMIR UNAMSIL UNAVEM UNDP UNESCO UNFICYP UNHCHR UNHCR UNICEF UNIDIR UNIFIL UNITA UNITAF UNMIBH UNMIH UNMIK UNMOT UNO UNOMIL UNOMIG UNOMSIL UNOMUR UNOSOM UNP UNPA UNPROFOR UNRISD UNSMA

681

National Reconstruction Secretariat (El Salvador) special representative of the Secretary-General South West African People’s Organization South West African Police South West African Territorial Force Transitional National Assembly Supreme Electoral Tribunal (El Salvador) Tamil United Liberation Front (Sri Lanka) Anti-Narcotics Executive Unit (El Salvador) United Kingdom United Liberation Movement of Liberia for Democracy ULIMO faction headed by Roosevelt Johnson ULIMO faction headed by Alhaji Kromah United Nations UN Advance Mission in Cambodia UN Assistance Mission in Rwanda UN Mission in Sierra Leone UN Angola Verification Mission UN Development Programme UN Educational, Scientific, and Cultural Organization UN Peacekeeping Force in Cyprus UN High Commission(er) for Human Rights UN High Commission(er) for Refugees UN International Children’s Emergency Fund UN Institute for Disarmament Research UN Interim Forces in Lebanon União Nacional para a Independência Total de Angola (Union for the Total Independence of Angola) Unified Task Force UN Mission in Bosnia-Herzegovina UN Mission in Haiti UN Mission in Kosovo UN Mission of Observers in Tajikistan National Opposition Union (Nicaragua) UN Observer Mission in Liberia UN Observer Mission in Guatemala UN Observer Mission in Sierra Leone UN Observer Mission Uganda-Rwanda UN Operation in Somalia United National Party (Sri Lanka) UN Protected Area UN Protection Force UN Research Institute for Social Development UN Special Mission in Afghanistan

S-Acronyms

11/24/09

682 UNTAC UNTAES UNTAET UNTAG URNG USAID USSR UXO VRS WFP WHO WOLA ZANLA ZIPRA ZOS ZWVA

4:29 PM

Page 682

ACRONYMS

UN Transitional Authority in Cambodia UN Transitional Administration for Eastern Slavonia, Baranja, and Western Sirmium UN Transitional Administration in East Timor UN Transitional Assistance Group Guatemalan National Revolutionary Unity U.S. Agency for International Development Union of Soviet Socialist Republics unexploded ordnance Republika Srpska Army World Food Program World Health Organization Washington Office on Latin America Zimbabwe African Nationalist Union, military wing Zimbabwe African People’s Union, military wing Zone of Separation (Bosnia) Zimbabwe War Veteran’s Association

S-Bibs

11/24/09

4:29 PM

Page 683

SELECTED BIBLIOGRAPHY

Adelman, Howard. “Refugees: The Right of Return.” In Group Rights. Ed. Judith Baker. Toronto: University of Toronto Press, 1994, 164–185. ———. “Why Refugee Warriors Are Threats.” Journal of Conflict Studies 18, no. 1 (spring 1998): 49–69. Adelman, Howard, and John Sorenson, eds. African Refugees: Development Aid and Repatriation. Boulder, Colo.: Westview Press, 1994. Adelman, Howard, and Astri Suhrke, with Bruce Jones. International Response to Conflict and Genocide: Lessons from the Rwanda Experience. Vol. 2, Early Warning and Conflict Management. Copenhagen: DANIDA, 1996. ———, eds. The Path to a Genocide: The Rwanda Crisis from Uganda to Zaire. New Brunswick, N.J.: Transaction, 1999. Adibe, Clement. “The Liberian Conflict and the ECOWAS-UN Partnership.” Third World Quarterly 18, no. 3 (1997): 471–488. Ajello, Aldo. “Mozambique: Implementation of the 1992 Peace Agreement.” In Herding Cats: Multiparty Mediation in a Complex World. Eds. Chester A. Crocker, Fen Osler Hampson, and Pamela Aall. Washington, D.C.: U.S. Institute of Peace, 1999, 615–642. Akhavan, Payam. “Justice and Reconciliation in the Great Lakes Region of Africa: The Contribution of the International Criminal Tribunal for Rwanda.” Duke Journal of Comparative and International Law 7, no. 2 (spring 1997): 325–348. ———. “The Yugoslav Tribunal at a Crossroads: The Dayton Peace Agreement and Beyond.” Human Rights Quarterly, 18, no. 2 (1996): 259–285. Alao, Abiodun. The Burden of Collective Goodwill: The International Involvement in the Liberian Civil War. Aldershot: Ashgate, 1998. Alvarez, Jose. “Crimes of States/Crimes of Hate: Lessons of Rwanda.” Yale Journal of International Law 24, no. 2 (1999): 365–483. Anderson, Mary B. Do No Harm: Supporting Local Capacities for Peace Through Aid. Cambridge, Mass.: Collaborative for Development Action, 1996. Anonymous. “Human Rights in Peace Negotiations.” Human Rights Quarterly 18, no. 2 (1996): 249–258.

683

S-Bibs

11/24/09

684

4:29 PM

Page 684

SELECTED BIBLIOGRAPHY

Anstee, Margaret J. Orphan of the Cold War: The Inside Story of the Collapse of the Angolan Peace Process, 1992–1993. New York: St. Martin’s Press, 1996. Arnson, Cynthia, and Carlos Basombrío, eds. Comparative Peace Processes in Latin America. Stanford: Stanford University Press, 1999. Atkinson, Philippa. The War Economy in Liberia: A Political Analysis. London: Overseas Development Institute, 1997. Austin, Reg. “New Forms of International Intervention: The United Nations Military-Civilian Intervention in Cambodia.” In Restructuring the Global Military Sector. Vol. 1, New Wars. Eds. Mary Kaldor and Basker Vashee. New York: Pinter, 1997–1998, 231–257. Ball, Nicole. “The Challenge of Rebuilding War-Torn Societies.” In Managing Global Chaos: Sources and Responses to International Conflict. Eds. Chester A. Crocker and Fen Osler Hampson, with Pamela Aall. Washington, D.C.: U.S. Institute of Peace, 1996, 607–622. ———. “Demobilizing and Reintegrating Soldiers: Lessons from Africa.” In Rebuilding Societies After Civil War: Critical Roles for International Assistance. Ed. Krishna Kumar. Boulder, Colo.: Lynne Rienner, 1997, 85–106. Ball, Nicole, and Tammy Halevy. Making Peace Work: The Role of the International Development Community. Policy Essay no. 18. Washington, D.C.: Overseas Development Council, 1996. Barnett, Michael N. “The U.N. Security Council, Indifference, and Genocide in Rwanda.” Cultural Anthropology 12, no. 4 (1997): 551–578. Bartoli, Andrea. “Mediating Peace in Mozambique: The Role of the Community of Sant’Egidio.” In Herding Cats: Multiparty Mediation in a Complex World. Eds. Chester A. Crocker, Fen Osler Hampson, and Pamela Aall. Washington, D.C.: U.S. Institute of Peace, 1999, 245–274. Bass, Warren. “The Triage of Dayton.” Foreign Affairs 77, no. 5 (September–October 1998): 98–108. Berdal, Mats, and Michael Leifer. “Cambodia.” In The New Interventionism, 1991–1994: United Nations Experience in Cambodia, Former Yugoslavia, and Somalia. Ed. James Mayall. Cambridge: Cambridge University Press, 1996, 25–58. Berman, Eric. Managing Arms in Peace Processes: Mozambique. UNIDIR Disarmament and Conflict Resolution Project. Geneva: United Nations, 1996. Black, Richard, and Khalid Koser, eds. The End of the Refugee Cycle? Refugee Repatriation and Reconstruction. New York: Berghahn Books, 1999. Bosco, David L. “Reintegrating Bosnia: A Progress Report.” The Washington Quarterly 21, no. 2 (spring 1998): 65–81. Bose, Sumantra. States, Nations, Sovereignty: Sri Lanka, India, and the Tamil Eelam Movement. London: Sage, 1994. Boyd, Charles G. “Making Bosnia Work.” Foreign Affairs 77, no. 1 (January–February 1998): 42–55. Broer, Harry, and Michael Emery. “Civilian Police in U.N. Peacekeeping Operations.” In Policing the New World Disorder: Peace Operations and Public Security. Eds. Robert B. Oakley, Michael J. Dziedzic, and Eliot M. Goldberg. Washington, D.C.: National Defense University Press, 1998, 365–398. Brown, Michael E., ed. Ethnic Conflict and International Security. Princeton: Princeton University Press, 1993. ———. The International Dimensions of Internal Conflict. Cambridge: MIT Press, 1996. Brynen, Rex. “Palestinian-Lebanese Relations: A Political Analysis.” In Peace for

S-Bibs

11/24/09

4:29 PM

Page 685

SELECTED BIBLIOGRAPHY

685

Lebanon? From War to Reconstruction. Ed. Deirdre Collings. Boulder, Colo.: Lynne Rienner, 1994, 83–96. ———. A Very Political Economy: Peacebuilding and Foreign Aid in the West Bank and Gaza. Washington, D.C.: U.S. Institute of Peace, 2000. Bugajski, Janusz, Jonathan S. Landay, John R. Lampe, Charles Lane, and Christine Wallich. “Policy: Forum: Bosnia After the Troops Leave.” Washington Quarterly 19, no. 3 (summer 1996): 69–88. Burkhalter, Holly. “The Question of Genocide: The Clinton Administration and Rwanda.” World Policy Review 11, no. 4 (winter 1994): 44–54. Byrne, Hugh. El Salvador’s Civil War: A Study of Revolution. Boulder, Colo.: Lynne Rienner, 1996. Byrne, Hugh, and William Stanley. More Haste, Less Speed: Police Reform in Guatemala. Washington, D.C.: Washington Office on Latin America, 1999. Cain, Kenneth L. “The Rape of Dinah: Civil War in Liberia and Evil Triumphant.” Human Rights Quarterly 21, no. 2 (May 1999): 265–307. Call, Charles T. “From Soldiers to Cops: ‘War Transitions’ and the Demilitarization of Policing in Latin America and the Caribbean.” Ph.D. diss., Stanford University, August 1999. ———. “Police Aid and the New World Disorder: Institutional Learning Within the U.S. International Criminal Investigative Training Assistance Program.” In Policing the New World Disorder: Peace Operations and Public Security. Ed. Robert B. Oakley, Michael J. Dziedzic, and Eliot M. Goldberg. Washington, D.C.: National Defense University Press, 1998, 315–363. Call, Charles, and Michael Barnett. “Looking for a Few Good Cops: Peacekeeping, Peacebuilding, and UN Civilian Police.” International Peacekeeping 6, no. 4 (winter 1999): 43–68. Chachiua, Martinho, and Mark Malan. “Anomalies and Acquiescence: The Mozambican Peace Process Revisited.” African Security Review 7, no. 4 (1998): 18–28. Chandler, David. “Democratization in Bosnia: the Limits of Civil Society Building Strategies.” Democratization 5, no. 4 (winter 1998): 78–102. ———. “The Limits of Peacebuilding: International Regulation and Civil Society Development in Bosnia.” International Peacekeeping 6, no. 1 (spring 1999): 109–126. Child, Jack. The Central American Peace Process, 1983–1991: Sheathing Swords, Building Confidence. Boulder, Colo.: Lynne Rienner, 1992. Chopra, Jarat. United Nations Authority in Cambodia. Watson Institute for International Studies, Occasional Paper no. 15. Providence: Brown University, 1994. ———. “The UN’s Kingdom of East Timor.” Survival 42, no. 3 (autumn 2000): 27–39. Cilliers, Jakkie, ed. Dismissed: Demobilization and Reintegration of Former Combatants in Africa. Pretoria: Institute for Defence Policy, 1995. Claesson, Paul, and Trevor Findlay. “Case Studies on Peacekeeping: UNOSOM II, UNTAC, and UNPROFOR.” In SIPRI [Stockholm International Peace Research Institute] Yearbook 1994. Oxford: Oxford University Press, 1994, 62–80. Clapham, Andrew, and Meg Henry. “Peacekeeping and Human Rights in Africa and Europe.” In Honoring Human Rights and Keeping the Peace: Lessons from El Salvador, Cambodia, and Haiti. Ed. Alice H. Henkin. Washington, D.C.: Aspen Institute, 1995, 129–160. Clapham, Andrew, and Florence Martin. “Smaller Missions, Bigger Problems.” In

S-Bibs

11/24/09

686

4:29 PM

Page 686

SELECTED BIBLIOGRAPHY

Honoring Human Rights: From Peace to Justice. Ed. Alice H. Henkin. Washington, D.C.: Aspen Institute, 1998, 133–162. Clapham, Christopher. “Rwanda: The Perils of Peacemaking.” Journal of Peace Research 35, no. 2 (1998): 193–210. Cock, Jacklyn. “The Social Integration of Demobilised Soldiers in Contemporary South Africa.” South African Defence Review no. 12 (1993): 1–17. Colletta, Nat, Markus Kostner, and Ingo Wiederhofer. The Transition from War to Peace in Sub-Saharan Africa. Washington, D.C.: World Bank, 1996. Collier, Paul. “Demobilization and Insecurity: A Study in the Economics of the Transition from War to Peace.” Journal of International Development 6, no. 3 (1994): 343–352. Collings, Deirdre, ed. Peace for Lebanon? From War to Reconstruction. Boulder, Colo.: Lynne Rienner, 1994. Costa, Gino. “The United Nations and Reform of the Police in El Salvador.” International Peacekeeping 2, no. 3 (August 1995): 365–390. Cousens, Elizabeth. “Making Peace in Bosnia Work.” Cornell International Law Journal 30, no. 3 (1997): 789–818. Cousens, Elizabeth, and Charles Cater. Toward Peace in Bosnia: Implementing the Dayton Accords. IPA Occasional Paper. Boulder, Colo.: Lynne Rienner, 2001. Cousens, Elizabeth, and Chetan Kumar, with Karin Wermester, eds. Peacebuilding as Politics: Cultivating Peace in Fragile Societies. Boulder, Colo.: Lynne Rienner, 2000. Cox, Marcus. “Strategic Approaches to International Intervention in Bosnia and Herzegovina.” Sarajevo: Center for Applied Studies in International Negotiations, October 1998. Crocker, Chester A., Fen Osler Hampson, and Pamela Aall, eds. Herding Cats: Multiparty Mediation in a Complex World. Washington, D.C.: U.S. Institute of Peace, 1999. ———. Managing Global Chaos: Sources and Responses to International Conflict. Washington, D.C.: U.S. Institute of Peace, 1996. Cuny, Frederick, Barry Stein, and Pat Reed. Repatriation During Conflict in Africa and Asia. Dallas, Tex.: Center for the Study of Societies in Crisis, 1992. Daalder, Ivo H. “Bosnia After SFOR: Options for Continued US Engagement.” Survival 39, no. 4 (winter 1997–1998): 5–28. Daadler, Ivo H., and Michael B. G. Froman. “Dayton’s Incomplete Peace.” Foreign Affairs 78, no. 6 (November–December 1999):106–113. Dallaire, Romeo. “The Changing Role of UN Peacekeeping Forces: The Relationship Between UN Peacekeepers and NGOs in Rwanda.” In The Role of Peacekeeping. Eds. Randolph Kent and Shashi Tharoor. New York: St. Martin’s Press, 1996. Daniel, Donald C. F., and Bradd C. Hayes. Beyond Traditional Peacekeeping. New York: St. Martin’s Press, 1995. David, Steven. “Internal War: Causes and Cures.” World Politics 49, no. 4 (July 1997): 552–576. de Figueiredo, Rui J. P., Jr., and Barry R. Weingast. “The Rationality of Fear: Political Opportunism and Ethnic Conflict.” In Civil War, Insecurity, and Intervention. Eds. Barbara F. Walter and Jack Snyder. New York: Columbia University Press, 1999, 261–301. de Rossanet, Bertrand. Peacemaking and Peacekeeping in Yugoslavia. Nijhoff Law Specials vol. 17. The Hague: Kluwer Law International, 1996. de Soto, Alvaro. “Ending Violent Conflict in El Salvador.” In Herding Cats: Multiparty Mediation in a Complex World. Eds. Chester A. Crocker, Fen Osler

S-Bibs

11/24/09

4:29 PM

Page 687

SELECTED BIBLIOGRAPHY

687

Hampson, and Pamela Aall. Washington, D.C.: U.S. Institute of Peace, 1999, 345–385. de Soto, Alvaro, and Graciana del Castillo. “Implementation of Comprehensive Peace Agreements: Staying the Course in El Salvador.” Global Governance 1, no. 2 (May–August 1995): 189–204. ———. “Obstacles to Peacebuilding in El Salvador.” Foreign Policy 94 (spring 1994): 69–83. Deeb, Mary-Jane, and Marius Deeb. “Internal Negotiations in a Centralist Conflict: Lebanon.” In Elusive Peace: Negotiating an End to Civil Wars. Ed. I. William Zartman. Washington, D.C.: Brookings Institution, 1995, 125–146. del Castillo, Graciana. “The Arms-for-Land Deal in El Salvador.” In Keeping the Peace: Multidimensional UN Operations in Cambodia and El Salvador. Eds. Michael W. Doyle, Ian Johnstone, and Robert C. Orr. Cambridge: Cambridge University Press, 1997, 342–365. Dercon, Stefan, and Daniel Ayalew. “Where Have All the Soldiers Gone? Demobilization and Reintegration in Ethiopia.” World Development 26, no. 9 (1998): 1661–1675. Des Forges, Alison. Leave None to Tell the Story: Genocide in Rwanda. New York: Human Rights Watch, 1999. Diller, Janelle M. Handbook on Human Rights in Situations of Conflict. Minneapolis: Minnesota Advocates for Human Rights, 1997. Dinstein, Yoram, and Mala Tabory, eds. War Crimes in International Law. The Hague: Martinus Nijhoff, 1996. Dolan, Chris. “Repatriation from South Africa to Mozambique—Undermining Durable Solutions?” In The End of the Refugee Cycle? Refugee Repatriation and Reconstruction. Eds. Richard Black and Khalid Koser. New York: Berghahn Books, 1999, 85–108. Doyle, Michael W. UN Peacekeeping in Cambodia: UNTAC’s Civil Mandate. Boulder, Colo.: Lynne Rienner, 1995. Doyle, Michael W., Ian Johnstone, and Robert C. Orr, eds. Keeping the Peace: Multidimensional UN Operations in Cambodia and El Salvador. Cambridge: Cambridge University Press, 1997. ———. “Strategies for Peace: Conclusions and Lessons.” In Keeping the Peace: Multidimensional UN Operations in Cambodia and El Salvador. Eds. Michael W. Doyle, Ian Johnstone, and Robert C. Orr. Cambridge: Cambridge University Press, 1997, 369–391. Doyle, Michael W., and Nicholas Sambanis. “International Peacebuilding: A Theoretical and Quantitative Analysis.” American Political Science Review 94, no. 4 (December 2000): 779–801. Duffy, Terence. “Cambodia Since the Election: Peace, Democracy, and Human Rights.” Contemporary Southeast Asia 15, no. 4 (1994): 407–432. ———. “Toward a Culture of Human Rights in Cambodia.” Human Rights Quarterly 16, no. 1 (February 1994): 82–104. Durch, William. The Evolution of UN Peacekeeping: Case Studies and Comparative Analysis. New York: St. Martin’s Press, 1993. ———, ed. UN Peacekeeping, American Policy, and the Uncivil Wars of the 1990s. New York: St. Martin’s Press, 1996. Dziedzic, Michael J., and Andrew Bair. “Bosnia and the International Police Task Force.” In Policing the New World Disorder: Peace Operations and Public Security. Eds. Robert B. Oakley, Michael J. Dziedzic, and Eliot M. Goldberg. Washington, D.C.: National Defense University Press, 1998, 253–314. Eaton, Chris. “The Role of Police in Institution Building.” In International

S-Bibs

11/24/09

688

4:29 PM

Page 688

SELECTED BIBLIOGRAPHY

Peacekeeping: Building on the Cambodian Experience. Ed. Hugh Smith. Canberra: Australian Defense Studies Centre, University College, 1994. Eguizábal, Cristina. “Regional Leadership and Universal Implementation in El Salvador’s Quest for Peace.” In The United Nations and Civil Wars. Ed. Thomas G. Weiss. Boulder, Colo.: Lynne Rienner, 1995, 173–190. Evans, Gareth. “The Comprehensive Political Settlement to the Cambodian Conflict: An Exercise in Cooperating for Peace.” In International Peacekeeping: Building on the Cambodian Experience. Ed. Hugh Smith. Canberra: Australian Defense Studies Centre, University College, 1994. Faris, Hani A. “The Failure of Peacemaking in Lebanon, 1975–1989.” In Peace for Lebanon? From War to Reconstruction. Ed. Deirdre Collings. Boulder, Colo.: Lynne Rienner, 1994, 17–30. Fearon, James, and David Laitin. “Explaining Interethnic Cooperation.” American Political Science Review 90, no. 4 (December 1996): 715–735. Findlay, Trevor. Cambodia: The Legacy and Lessons of UNTAC. SIPRI Research Report no. 9. New York: Oxford University Press, 1995. Fisler Damrosch, Lori, ed. Enforcing Restraint: Collective Intervention in Internal Conflicts. New York: Council on Foreign Relations, 1993. Forman, Shepard, and Stewart Patrick, eds. Good Intentions: Pledges of Aid for Postconflict Recovery. Boulder, Colo.: Lynne Rienner, 2000. Fox, Gregory H. “International Law and Civil Wars.” NYU Journal of International Law and Politics 26, no. 4 (1994): 633–653. Franck, Thomas. “A Holistic Approach to Peacebuilding.” In Peacemaking and Peacekeeping for the New Century. Eds. Olara A. Otunnu and Michael W. Doyle. Lanham, Md.: Rowman and Littlefield, 1998, 275–295. Franco, Leonardo, and Jared Kotler. “Combining Institution Building and Human Rights in Guatemala: The Challenge of Buying In Without Selling Out.” In Honoring Human Rights: From Peace to Justice. Ed. Alice H. Henkin. Washington, D.C.: Aspen Institute, 1998, 39–70. Frieson, Kate. “The Cambodian Elections of 1993: A Case of Power to the People?” In The Politics of Elections in Southeast Asia. Ed. Robert H. Taylor. Cambridge: Cambridge University Press, 1997. ———. “The Politics of Getting the Vote in Cambodia.” In Propaganda, Politics, and Violence in Cambodia: Democratic Transition Under United Nations Peace-Keeping. Eds. Steve Heder and Judy Ledgerwood. Armonk, N.Y.: M. E. Sharpe, 1996, 183–207. Goldstone, Richard. “Exposing Human Rights Abuses: A Help or Hindrance to Reconciliation?” Hastings Constitutional Law Quarterly 22, no. 3 (1995): 607–621. Goodfellow, William, and James Morrell. “From Contadora to Esquipulas to Sapoa and Beyond.” In Revolution and Counterrevolution in Nicaragua. Ed. Thomas W. Walker. Boulder, Colo.: Westview Press, 1991, 369–393. Goodwin-Gill, Guy S. Free and Fair Elections in International Law. Geneva: InterParliamentary Union, 1994. Goodwin-Gill, Guy, and Ilene Cohn. Child Soldiers: The Role of Children in Armed Conflicts. Oxford: Clarendon Press, 1994. Goose, Stephen D., and Frank Symth. “Arming Genocide in Rwanda.” Foreign Affairs 73, no. 5 (September 1994): 86–96. Gourevitch, Philip. We wish to inform you that tomorrow we will be killed with all our families: Stories from Rwanda. New York: Farrar, Straus, and Giroux, 1998. Griffin, Michèle, and Bruce Jones. “Building Peace Through Transitional Authority:

S-Bibs

11/24/09

4:29 PM

Page 689

SELECTED BIBLIOGRAPHY

689

New Directions, Major Challenges.” International Peacekeeping 7, no. 4 (winter 2000): 75–90. Griffiths, Ann L., ed. Building Peace and Democracy in Post-Conflict Societies. Halifax: Centre for Foreign Policy Studies, Dalhousie University, 1998. Gurr, Ted. Minorities at Risk: A Global View of Ethnopolitical Conflicts. Washington, D.C.: U.S. Institute of Peace, 1993. ———. Peoples Versus States: Minorities at Risk in the New Century. Washington, D.C.: U.S. Institute of Peace, 2000. Guttieri, Karen. “Symptom of the Moment: A Juridical Gap for U.S. Occupation Forces.” International Insights 13, special issue (fall 1997): 131–155. Hampson, Fen Osler. Nurturing Peace: Why Peace Settlements Succeed or Fail. Washington, D.C.: U.S. Institute of Peace, 1996. Hansen, Annika S., and Lia Brynjar. The Role of International Security Assistance in Support of Peace Agreements in War-Torn Societies. FFI Report no. 98/05291. Kjeller, Norway: Forsvarets Forskingsinstitut [Norwegian Defense Research Establishment], December 1998. Harik, Judith. “The Effects of Protracted Social Conflict on Resettlement of the Displaced in Postwar Lebanon.” In Conflict Resolution in the Arab World: Selected Essays. Ed. Paul Salem. Beirut: American University of Beirut, 1997, 384–410. Harrell-Bond, Barbara. “Repatriation: Under What Conditions Is It the Most Desirable Solution for Refugees?” African Studies Review 32, no. 1 (1989): 41–69. Harris, Peter, and Ben Reilly, eds. Democracy and Deep-Rooted Conflict: Options for Negotiators. Handbook Series no. 3. Stockholm: International Institute for Democracy and Electoral Assistance, 1998. Hayner, Priscilla B. “Fifteen Truth Commissions, 1974–1994: A Comparative Study.” Human Rights Quarterly 16, no. 4 (1994): 597–655. Heder, Steven. “The Resumption of Armed Struggle by the Party of Democratic Kampuchea: Evidence from National Army of Democratic Kampuchea ‘SelfDemobilizers.’” In Propaganda, Politics, and Violence in Cambodia: Democratic Transition Under United Nations Peace-Keeping. Eds. Steven Heder and Judy Ledgerwood. Armonk, N.Y.: M. E. Sharpe, 1996, 73–113. Heder, Steven, and Judy Ledgerwood, eds. Propaganda, Politics, and Violence in Cambodia: Democratic Transition Under United Nations Peace-Keeping. Armonk, N.Y.: M. E. Sharpe, 1996. Heininger, Janet, Peacekeeping in Transition: The United Nations in Cambodia. New York: Twentieth Century Fund Press, 1994. Helman, Gerald, and Steven Ratner. “Saving Failed States.” Foreign Policy 89 (winter 1992–1993): 3–20. Henkin, Alice H, ed. Honoring Human Rights: From Peace to Justice. Washington, D.C.: Aspen Institute, 1998. ———. Honoring Human Rights and Keeping the Peace: Lessons from El Salvador, Cambodia, and Haiti. Washington, D.C.: Aspen Institute, 1995. Herbst, Jeffrey, and Walter Clarke, eds. Learning from Somalia: The Lessons of Armed Humanitarian Intervention. San Francisco: Westview Press, 1997. ˇ cˇ anin, and Susan L. Woodward. “Bosnia and Herti´c , Zlatko, Amela Sap Herzegovina.” In Good Intentions: Pledges of Aid for Postconflict Recovery. Eds. Shepard Forman and Stewart Patrick. Boulder, Colo.: Lynne Rienner, 2000, 315–366. Hills, Alice. Policing Africa: Internal Security and the Limits of Liberalization. Boulder, Colo.: Lynne Rienner, 2000.

S-Bibs

11/24/09

690

4:29 PM

Page 690

SELECTED BIBLIOGRAPHY

Hirsch, John L., Robert Oakley, and Chester B. Crocker. Somalia and Operation Restore Hope: Reflections on Peacemaking and Peacekeeping. Washington, D.C.: U.S. Institute of Peace, 1995. Holbrooke, Richard C. To End a War. Rev. ed. New York: Random House, 1999. Holiday, David, and William Stanley. “Building the Peace: Preliminary Lessons from El Salvador.” Journal of International Affairs 46, no. 2 (winter 1993): 415–437. Hollis, Rosemary, and Nadim Shehadi, eds. Lebanon on Hold: Implications for Middle East Peace. Oxford: Royal Institute of International Affairs and the Centre for Lebanese Studies, 1996. Howe, Herbert. “Lessons of Liberia: ECOMOG and Regional Peacekeeping.” International Security 21, no. 3 (winter 1996–1997): 145–176. Hudson, Michael C. “Power-Sharing in Post-Civil War Lebanon.” International Negotiation 2, no. 1 (1997): 103–122. Hume, Cameron. Ending Mozambique’s War: The Role of Mediation and Good Offices. Washington, D.C.: U.S. Institute of Peace, 1994. Iweze, Cyril. “Nigeria in Liberia: The Military Operations of ECOMOG.” In Nigeria in International Peacekeeping, 1960–1992. Eds. Margaret Vogt and A. E. Ekoko. Lagos: Malthouse Press, 1993, 216–243. Jan, Ameen. Peacebuilding in Somalia. Policy Briefing Series. New York: International Peace Academy, July 1996. Jennar, Raoul. “UNTAC: ‘International Triumph’ in Cambodia?” Security Dialogue 25, no. 2 (1994): 145–156. Johnstone, Ian. Rights and Reconciliation in El Salvador. Boulder, Colo.: Lynne Rienner, 1995. Jones, Bruce D. “The Arusha Peace Process.” In The Path of a Genocide: The Rwanda Crisis from Uganda to Zaire. Eds. Howard Adelman and Astri Suhrke. New Brunswick, N.J.: Transaction Press, 1999, 138–139. ———. “‘Intervention Without Borders’: Humanitarian Intervention in Rwanda, 1990–1994.” Millennium 24, no. 2 (1995): 225–249. ———. Peacemaking in Rwanda: The Dynamics of Failure. Boulder, Colo.: Lynne Rienner, 2001. Joulwan, George A., and Christopher C. Shoemaker. Civilian-Military Cooperation in the Prevention of Deadly Conflict: Implementing Agreements in Bosnia and Beyond. New York: Carnegie Corporation of New York, 1998. Kaldor, Mary, and Basker Vashee. Restructuring the Global Military Sector. Vol. 1, New Wars. New York: Pinter, 1997–1998. Karl, Terry Lynn. “El Salvador’s Negotiated Revolution.” Foreign Affairs 71, no. 2 (spring 1992): 147–164. Kaufmann, Chaim. “Possible and Impossible Solutions to Ethnic Civil Wars.” International Security 20, no. 4 (spring 1996): 136–175. Keen, David, and Ken Wilson. “Engaging with Violence: A Reassessment of Relief in Wartime.” In War and Hunger: Rethinking International Responses to Complex Emergencies. Eds. Joanna Macrae and Anthony Zwi. London: Zed Books, 1994, 209–221. Kelly, Michael J. “Legitimacy and the Public Security Function.” In Policing the New World Disorder: Peace Operations and Public Security. Eds. Robert B. Oakley, Michael J. Dziedzic, and Eliot M. Goldberg. Washington, D.C.: National Defense University Press, 1998, 399–431. Kent, Randolph, and Shashi Tharoor, eds. The Role of Peacekeeping. New York: St. Martin’s Press, 1996. Khalaf, Samir. “From a Geography of Fear to a Culture of Tolerance: Reflections on

S-Bibs

11/24/09

4:29 PM

Page 691

SELECTED BIBLIOGRAPHY

691

Protracted Strife and the Restoration of Civility in Lebanon.” In Conflict Resolution in the Arab World: Selected Essays. Ed. Paul Salem. Beirut: American University of Beirut, 1997, 354–383. Khashan, Hilal. “Lebanon’s 1996 Controversial Parliamentary Elections.” Journal of South Asian and Middle Eastern Studies 20, no. 4 (summer 1997): 24–49. Kiernan, Ben. The Pol Pot Regime. New Haven, Conn.: Yale University Press, 1996. Kingma, Kees. “Demobilization of Combatants After Civil Wars in Africa and Their Reintegration into Civilian Life.” Policy Sciences 30, no. 3 (August 1997): 151–165. Kostakos, Georgios. “Division of Labor Among International Organizations: The Bosnian Experience.” Global Governance 4, no. 4 (October–December 1998): 461–484. Krayem, Hassan. “The Lebanese Civil War and the Ta’if Agreement.” In Conflict Resolution in the Arab World: Selected Essays. Ed. Paul Salem. Beirut: American University of Beirut, 1997, 411–435. Kuhlman, Tom. “Organized Versus Spontaneous Settlement of Refugees in Africa.” In African Refugees: Development Aid and Repatriation. Eds. Howard Adelman and John Sorenson. Boulder, Colo.: Westview Press, 1994, 117–142. Kumar, Krishna, ed. After the War Is Over, What Comes Next? Washington, D.C.: U.S. Agency for International Development, 1997. ———. Postconflict Elections, Democratization, and International Assistance. Boulder, Colo.: Lynne Rienner, 1998. ———. Rebuilding Societies After Civil War: Critical Roles for International Assistance. Boulder, Colo.: Lynne Rienner, 1997. Kumar, Krishna, and Marina Ottaway. From Bullets to Ballots: Electoral Assistance to Postconflict Societies. Washington, D.C.: U.S. Agency for International Development, Center for Development Information and Evaluation, 1997. Kuperman, Alan J. The Limits of Humanitarian Intervention: Genocide in Rwanda. Washington, D.C.: Brookings Institution, 2001. ———. “The Other Lesson of Rwanda: Mediators Sometimes Do More Harm Than Good.” SAIS Review 16, no. 1 (1996): 221–240. Laegreid, Turid. “U.N. Peacekeeping in Rwanda.” In The Path of a Genocide: The Rwanda Crisis from Uganda to Zaire. Eds. Howard Adelman and Astri Suhrke. New Brunswick, N.J.: Transaction, 1999, 231–251. Landsman, Stephen. “Alternative Responses to Serious Human Rights Abuses of Prosecutions and Truth Commissions.” Law and Contemporary Problems 59, no. 4 (1992): 81–92. Larkin, Mary Ann, Fred Cuny, and Barry Stein, eds. Repatriation Under Conflict in Central America. Washington, D.C.: CIPRA, Georgetown University, 1991. Lederach, John Paul. Building Peace: Sustainable Reconciliation in Divided Societies. Washington, D.C.: U.S. Institute of Peace, 1997. Ledgerwood, Judy. “Patterns of CPP Political Repression and Violence During the UNTAC Period.” In Propaganda, Politics, and Violence in Cambodia: Democratic Transition Under United Nations Peace-Keeping. Eds. Steven Heder and Judy Ledgerwood. Armonk, N.Y.: M. E. Sharpe, 1996, 114–133. Lee Kim, Cheryl M., and Mark Metrikas. “Holding a Fragile Peace: The Military and Civilian Components of UNTAC.” In Keeping the Peace: Multidimensional UN Operations in Cambodia and El Salvador. Eds. Michael W. Doyle, Ian Johnstone, and Robert C. Orr. Cambridge: Cambridge University Press, 1997, 107–133. Levine, Mark. “Peacemaking in El Salvador.” In Keeping the Peace:

S-Bibs

11/24/09

692

4:29 PM

Page 692

SELECTED BIBLIOGRAPHY

Multidimensional UN Operations in Cambodia and El Salvador. Eds. Michael W. Doyle, Ian Johnstone, and Robert C. Orr. Cambridge: Cambridge University Press, 1997, 227–254. Lodico, Yvonne C. “A Peace That Fell Apart: The United Nations and the War in Angola.” In UN Peacekeeping, American Policy, and the Uncivil Wars of the 1990s. Ed. William Durch. New York: St. Martin’s Press, 1996, 103–134. López-Pintor, Rafael. “Nicaragua’s Measured Move to Democracy.” In Postconflict Elections, Democratization, and International Assistance. Ed. Krishna Kumar. Boulder, Colo.: Lynne Rienner, 1998, 39–56. Luciuk, L. Y., and M. S. Kenzer, eds. Under Threat: Essays on the Refugee Experience. Lanham, Md.: Rowan and Littlefield, 1998. Lyons, Terrence. Voting for Peace: Post Conflict Elections in Liberia. Washington, D.C.: Brookings Institution, 1998. Macrae, Joanna, and Anthony Zwi, eds. War and Hunger: Rethinking International Responses to Complex Emergencies. London: Zed Books, 1994. Magyar, Karl, and Earl Conteh-Morgan, eds. Peacekeeping in Africa: ECOMOG in Liberia. Hampshire, London: Macmillan, 1998. Maila, Joseph. “The Ta’if Accord: An Evaluation.” In Peace for Lebanon? From War to Reconstruction. Ed. Deirdre Collings. Boulder, Colo.: Lynne Rienner, 1994, 31–44. Malan, Mark. “Peace-Building in Southern Africa: Police Reform in Mozambique and South Africa.” International Peacekeeping 6, no. 4 (winter 1999): 171–190. Malone, David. Decision Making in the UN Security Council: The Case of Haiti, 1990–1997. Oxford: Oxford University Press, 1998. Manning, Carrie. “Constructing Opposition in Mozambique: Renamo as Political Party.” Journal of Southern African Studies 24, no. 1 (March 1998): 161–190. Marenin, Otwin. Policing Change, Changing Police: International Perspectives. New York: Garland Press, 1996. Marks, Stephen P. “Elusive Justice for Victims of the Khmer Rouge.” Journal of International Affairs 52, no. 2 (spring 1999): 691–718. ———. “Forgetting ‘The Policies and Practices of the Past’: Impunity in Cambodia.” The Fletcher Forum of World Affairs 18, no. 2 (summer–fall 1994): 17–44. Martin, Ian. “Genocide: The UN Human Rights Field Operation in Rwanda.” In Honoring Human Rights: From Peace to Justice. Ed. Alice H. Henkin. Washington, D.C.: Aspen Institute, 1998, 97–132. Mayall, James, ed. The New Interventionism, 1991–1994: United Nations Experience in Cambodia, Former Yugoslavia, and Somalia. Cambridge: Cambridge University Press, 1996. McCalin, Margaret. The Reintegration of Young Ex-Combatants into Civilian Life: A Report to the International Labour Office. Geneva: International Labor Office, 1995. McCleary, Rachel. Dictating Democracy: Guatemalan Elites and the End of Violent Revolution. Gainesville: University of Florida Press, 1999. ———. “Guatemala’s Postwar Prospects.” In Journal of Democracy 8, no. 2 (April 1997): 129–143. McCoy, Jennifer L., Larry Garber, and Robert Pastor. “Pollwatching and Peacemaking.” Journal of Democracy 2, no. 3 (fall 1991): 102–114. McLean, Lyndall. “Civil Administration in Transition: Public Information and the Neutral Political Environment.” In International Peacekeeping: Building on the Cambodian Experience. Ed. Hugh Smith. Canberra: Australian Defense Studies Centre, University College, 1994.

S-Bibs

11/24/09

4:29 PM

Page 693

SELECTED BIBLIOGRAPHY

693

Melvern, Linda. “Genocide Behind the Thin Blue Line.” Security Dialogue 28, no. 3 (1997): 333–346. ———. A People Betrayed: The Role of the West in Rwanda’s Genocide. New York: Zed Books, 2000. Mendez, Juan E. “Accountability for Past Abuses.” Human Rights Quarterly 19, no. 2 (1997): 255–282. Menkhaus, Ken. “International Peacebuilding and the Dynamics of Local and National Reconciliation in Somalia.” In Learning from Somalia: The Lessons of Armed Humanitarian Intervention. Eds. Walter Clarke and Jeffrey Herbst. Boulder, Colo.: Westview Press, 1997, 42–63. Minow, Martha. Between Vengeance and Forgiveness: Facing History After Genocide and Mass Violence. Boston: Beacon Press, 1998. Montgomery, Tommie Sue. Revolution in El Salvador: From Civil Strife to Civil Peace. 2nd ed. Boulder, Colo.: Westview Press, 1995. Motumi, Tsepe, and Andrew Hudson. “Rightsizing: The Challenges of Demobilisation and Social Reintegration in South Africa.” In Dismissed: Demobilisation and Reintegration of Former Combatants in Africa. Ed. Jakkie Cilliers. Pretoria: Institute for Defence Policy, 1995, 112–129. Munck, Gerardo L. “Beyond Electoralism in El Salvador: Conflict Resolution Through Negotiated Compromise.” Third World Quarterly 14, no. 1 (1993): 75–94. Musemwa, Muchaparara. “The Ambiguities of Democracy: The Demobilisation of the Zimbabwean Ex-Combatants and the Ordeal of Rehabilitation, 1980–1993.” In Dismissed: Demobilisation and Reintegration of Former Combatants in Africa. Ed. Jakkie Cilliers. Pretoria: Institute for Defence Policy, 1995, 44–57. Narvesen, Ove, and Oona Solberg. Repatriation During Conflict. Oslo: Norwegian Refugee Council, 1992. North, Liisa L., and Alan B. Simmons, eds. “Concluding Reflections: Refugee Return, National Transformation, and Neo-Liberal Restructuring.” In Journeys of Fear: Refugee Return and National Transformation in Guatemala. Eds. Liisa L. North and Alan B. Simmons. Montreal: McGill-Queen’s University Press, 1999, 272–300. ———. “Fear and Hope: Return and Transformation in Historical Perspective.” In Journeys of Fear: Refugee Return and National Transformation in Guatemala. Eds. Liisa L. North and Alan B. Simmons. Montreal: McGill-Queen’s University Press, 1999, 3–27. ———. Journeys of Fear: Refugee Return and National Transformation in Guatemala. Montreal: McGill-Queen’s University Press, 1999. Norton, Augustus Richard. “Hizballah: From Radicalism to Pragmatism.” Middle East Policy 5, no. 4 (January 1998): 147–158. Norton, Augustus Richard, and Jillian Schwedler. “Swiss Soldiers, Ta’if Clocks, and Early Elections: Toward a Happy Ending?” In Peace for Lebanon? From War to Reconstruction. Ed. Deirdre Collings. Boulder, Colo.: Lynne Rienner, 1994, 45–65. Nowrojee, Binaifir. “Joining Forces: UN and Regional Peacekeeping, Lessons from Liberia.” Harvard Human Rights Journal 8 (spring 1995): 129–151. Oakley, Robert B., Michael J. Dziedzic, and Eliot M. Goldberg, eds. Policing the New World Disorder: Peace Operations and Public Security. Washington, D.C.: National Defense University Press, 1998. O’Flaherty, Michael, and Gregory Gisvold, eds. Post War Protection of Human Rights in Bosnia and Herzegovina. Boston: Martinus Nijhoff, 1998.

S-Bibs

11/24/09

694

4:29 PM

Page 694

SELECTED BIBLIOGRAPHY

Olonisakin, Funmi. “UN Cooperation with Regional Organizations in Peacekeeping: The Experience of ECOMOG and UNOMIL in Liberia.” International Peacekeeping 3, no. 3 (autumn 1996): 33–51. Omaar, Rakiya, and Alex de Waal. Rwanda: Death, Despair, and Defiance. London: African Rights, 1994. Orentlicher, Diane. “Settling Accounts: The Duty to Prosecute Human Rights Violators of a Prior Regime.” Yale Law Journal 100, no. 8 (June 1991): 2537–2615. Otunnu, Olara A., and Michael W. Doyle, eds. Peacemaking and Peacekeeping for the New Century. Lanham, Md.: Rowman and Littlefield, 1998. Peou, Sorpong. Conflict Neutralization in the Cambodia War: From Battlefield to Ballot-Box. New York: Oxford University Press, 1997. ———. Intervention and Change in Cambodia: Towards Democracy? New York: St. Martin’s Press, 2000. Posen, Barry. “The Security Dilemma and Ethnic Conflict.” In Ethnic Conflict and International Security. Ed. Michael E. Brown. Princeton: Princeton University Press, 1993, 103–124. Prendergast, John. Front Line Diplomacy: Humanitarian Aid and Conflict in Africa. Boulder, Colo.: Lynne Rienner, 1996. Prunier, Gerard. The Rwanda Crisis: History of a Genocide, 1959–1994. New York: Columbia University Press, 1995. Ram, Mohan. Sri Lanka: The Fractured Island. New Delhi: Penguin, 1989. Ratner, Steven B. The New UN Peacekeeping. New York: St. Martin’s Press, 1995. Reilly, Ben, and Andrew Reynolds. Electoral Systems and Conflict in Divided Societies. Papers on International Conflict Resolution no. 2. Washington, D.C.: National Academy Press, 1999. Reno, William. Warlord Politics and African States. Boulder, Colo.: Lynne Rienner, 1998. Roberts, Adam. “Communal Conflict as a Challenge to International Organization: The Case of Former Yugoslavia.” Review of International Studies 21, no. 4 (1995): 389–410. Roed-Larsen, Terje, and Rick Hooper. Command from the Saddle. Oslo: Peace Implementation Network, 1999. Rothchild, Donald. Managing Ethnic Conflict in Africa: Pressures and Incentives for Cooperation. Washington, D.C.: Brookings Institution, 1997. Rubin, Barnett R., ed. Toward Comprehensive Peace in Southeast Europe: Conflict Prevention in the South Balkans. Report of the South Balkans Working Group of the Council on Foreign Relations Center for Preventive Action. New York: Twentieth Century Fund Press, 1996. Saidi, Nasser. “The Economic Reconstruction of Lebanon: War, Peace, and Modernization.” In Peace for Lebanon? From War to Reconstruction. Ed. Deirdre Collings. Boulder, Colo.: Lynne Rienner, 1994, 195–212. Salem, Paul. “Documents: The Ta’if Agreement—Annotated Text.” The Beirut Review 1, no. 1 (spring 1991): 119–172. ———. “Two Years of Living Dangerously: General Awn and the Precarious Rise of Lebanon’s Second Republic.” The Beirut Review 1, no. 1 (spring 1991): 62–87. ———. “The Wounded Republic: Lebanon’s Struggle for Recovery.” Arab Studies Quarterly 16, no. 4 (fall 1994): 47–63. Sanderson, John M. “UNTAC: Successes and Failures.” In International Peacekeeping: Building on the Cambodian Experience. Ed. Hugh Smith. Canberra: Australian Defense Studies Centre, University College, 1994. Sbaiti, Ahmed. “Reflections on Lebanon’s Reconstruction.” In Peace for Lebanon?

S-Bibs

11/24/09

4:29 PM

Page 695

SELECTED BIBLIOGRAPHY

695

From War to Reconstruction. Ed. Deirdre Collings. Boulder, Colo.: Lynne Rienner, 1994, 163–178. Schear, James. “Bosnia’s Post-Dayton Traumas.” Foreign Policy 104 (fall 1996): 87–101. Schear, James, and Karl Farris. “Policing Cambodia: The Public Security Dimensions of U.N. Peace Operations.” In Policing the New World Disorder: Peace Operations and Public Security. Eds. Robert B. Oakley, Michael J. Dziedzic, and Eliot M. Goldberg. Washington, D.C.: National Defense University Press, 1998, 69–102. Schirmer, Jennifer. “Prospects for Compliance: The Guatemalan Military and the Peace Accords.” In Guatemala After the Peace Accords. Ed. Rachel Sieder. London: University of London, Institute of Latin American Studies, 1998, 21–32. Schmidl, Erwin A. “Police Functions in Peace Operations: An Historical Overview.” In Policing the New World Disorder: Peace Operations and Public Security. Eds. Robert Oakley, Michael J. Dziedzic, and Eliot M. Goldberg. Washington, D.C.: National Defense University Press, 1998, 19–40. Seevaratnam, N., ed. The Tamil National Question and the Indo-Sri Lanka Accord. New Delhi: Konark, 1989. Sell, Louis. “The Serb Flight from Sarajevo: Dayton’s First Failure.” Eastern European Politics and Societies 14, no. 1 (winter 2000): 179–202. Selochan, Viberto, and Carlyle A. Thayer, eds. Bringing Democracy to Cambodia. Canberra: Australian Defense Studies Centre and Australian Defense Force Academy, 1996. Shain, Yossi, and Juan J. Linz. Between States: Interim Governments and Democratic Transitions. Cambridge: Cambridge University Press, 1995. Sharp, Jane M. O. “Dayton Report Card.” International Security 22, no. 3 (winter 1997–1998): 101–137. Sieder, Rachel, ed. Guatemala After the Peace Accords. London: University of London, Institute of Latin American Studies, 1998. Sisk, Timothy D. “Elections and Conflict Management in Africa: Conclusions and Recommendations.” In Elections and Conflict Management in Africa. Eds. Timothy D. Sisk and Andrew Reynolds. Washington, D.C.: U.S. Institute of Peace, 1998, 145–171. ———. Power Sharing and International Mediation in Ethnic Conflicts. Washington, D.C.: U.S. Institute of Peace, 1996. Sisk, Timothy D., and Andrew Reynolds, eds. “Democratization, Elections, and Conflict Management in Africa: Exploring the Nexus.” Introduction to Elections and Conflict Management in Africa. Eds. Timothy D. Sisk and Andrew Reynolds. Washington, D.C.: U.S. Institute of Peace, 1998, 1–7. ———. Elections and Conflict Management in Africa. Washington, D.C.: U.S. Institute of Peace, 1998. Smith, Christopher, Peter Batchelor, and Jakkie Potgieter, eds. Small Arms Management and Peacekeeping in Southern Africa. Geneva: UNIDIR, 1996. Smith, Hugh, ed. International Peacekeeping: Building on the Cambodian Experience. Canberra: Australian Defense Studies Centre, University College, 1994. Snyder, Jack, and Robert Jervis. “Civil War and the Security Dilemma.” In Civil War, Insecurity, and Intervention. Eds. Barbara F. Walter and Jack Snyder. New York: Columbia University Press, 1999, 15–37. Spear, Joanna. “Arms Limitations, Confidence-Building Measures, and Internal Conflict.” In The International Dimensions of Internal Conflict. Ed. Michael E. Brown. Cambridge: MIT Press, 1996, 377–410.

S-Bibs

11/24/09

696

4:29 PM

Page 696

SELECTED BIBLIOGRAPHY

Spence, Jack, David R. Dye, Paula Worby, Carmen Rosa de Leon-Escribano, George Vickers, and Mike Lanchin. Promise and Reality: Implementation of the Guatemalan Peace Accords. Cambridge, Mass.: Hemisphere Initiatives, August 1998. Stanley, William. “Building New Police Forces in El Salvador and Guatemala: Learning and Counterlearning.” International Peacekeeping 6, no. 4 (winter 1999): 113–134. ———. The Protection Racket State: Elite Politics, Military Extortion, and Civil War in El Salvador. Philadelphia: Temple University Press, 1996. Stanley, William, and Charles T. Call. “Building a New Civilian Police Force in El Salvador.” In Rebuilding Societies After Civil War: Critical Roles for International Assistance. Ed. Krishna Kumar. Boulder, Colo.: Lynne Rienner, 1997, 107–134. Stanley, William, and David Holiday. “Peace Mission Strategy and Domestic Actors: United Nations Mediation, Verification, and Institution Building in El Salvador.” International Peacekeeping 4, no. 2 (summer1997): 22–49. Stanley, William, and Robert Loosle. “El Salvador: The Civilian Police Component of Peace Operations.” In Policing the New World Disorder: Peace Operations and Public Security. Eds. Robert B. Oakley, Michael J. Dziedzic, and Eliot M. Goldberg. Washington, D.C.: National Defense University Press, 1998, 103–141. Stedman, Stephen John. “Negotiation and Mediation in Internal Conflict.” In The International Dimensions of Internal Conflict. Ed. Michael E. Brown. Cambridge: MIT Press, 1996, 341–376. ———. Peacemaking in Civil War: International Mediation in Zimbabwe, 1974–1980. Boulder, Colo.: Lynne Rienner, 1991. ———. “Spoiler Problems in Peace Processes.” International Security 22, no. 2 (fall 1997): 5–53. ———. “UN Intervention in Civil Wars: Imperatives of Choice and Strategy.” In Beyond Traditional Peacekeeping. Eds. Donald C. F. Daniel and Bradd C. Hayes. New York: St. Martin’s Press, 1995, 40–63. Stedman, Stephen John, and Donald Rothchild. “Peace Operations: From ShortTerm to Long-Term Commitment.” International Peacekeeping 3, no. 2 (summer 1996): 17–35. Steil, Benn, and Susan L. Woodward. “A European New Deal for the Balkans.” Foreign Affairs 78, no. 6 (November–December 1999): 95–105. Stein, B. N., F. C. Cuny, and R. Reed, eds. Refugee Repatriation During Conflict: A New Conventional Wisdom. Dallas, Tex.: Center for the Study of Societies in Crisis, 1995. Stephen, Michael F. “Demobilisation in Mozambique.” In Dismissed: Demobilisation and Reintegration of Former Combatants in Africa. Ed. Jakkie Cilliers. Pretoria: Institute for Defence Policy, 1995, 58–69. Suhrke, Astri. “Facing Genocide: The Record of the Belgian Battalion in Rwanda.” Security Dialogue 29, no. 1 (1998): 37–48. Szayna, Thomas S., Preston Niblack, and William O’Malley. “Assessing Armed Forces’ Deficiencies for Peace Operations: A Methodology.” International Peacekeeping 3, no. 3 (autumn 1996): 77–91. Tanner, Fred. “Consensual Versus Coercive Disarmament.” In Managing Arms in Peace Processes: The Issues. Geneva: UN Institute for Disarmament Research, 1996, 169–204. Tanner, Victor. “Liberia: Railroading Peace.” Review of African Political Economy 75, no. 25 (March 1998): 133–147.

S-Bibs

11/24/09

4:29 PM

Page 697

SELECTED BIBLIOGRAPHY

697

Turner, J. Michael, Sue Nelson, and Kimberly Mahling-Clark. “Mozambique’s Vote for Democratic Governance.” In Postconflict Elections, Democratization, and International Assistance. Ed. Krishna Kumar. Boulder, Colo.: Lynne Rienner, 1998, 153–176. Uphoff Kato, Elisabeth. “Quick Impacts, Slow Rehabilitation in Cambodia.” In Keeping the Peace: Multidimensional UN Operations in Cambodia and El Salvador. Eds. Michael W. Doyle, Ian Johnstone, and Robert C. Orr. Cambridge: Cambridge University Press, 1997, 186–205. Utting, Peter, ed. Between Hope and Insecurity: The Social Consequences of the Cambodian Peace Process. Geneva: UNRISD, 1994. Van Walraven, Klaas. The Pretence of Peace-Keeping: ECOMOG, West Africa, and Liberia, 1990–1998. The Hague: Netherlands Institute of International Relations, 1999. Vinegran, Anna. “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: University of London, Institute of Latin American Studies, 1998, 207–227. Vines, Alex. Renamo: From Terrorism to Democracy in Mozambique? London: James Currey, 1996. Vogt, Margaret, ed. The Liberian Crisis and ECOMOG: A Bold Attempt at Regional Peacekeeping. Lagos: Gabumo Press, 1992. Vogt, Margaret, and A. E. Ekoko. Nigeria in International Peacekeeping, 1960–1992. Lagos: Malthouse Press, 1993. Wallensteen, Peter. “Peace-Making in Cambodia: An Analytical Framework.” In The Cambodian Conflict, 1979–1991: From Intervention to Resolution. Eds. Ramses Amer, Johan Saravanamuttu, and Peter Wallensteen. Penang, Malaysia: Sinaran Bros. Sdn. Berhad, 1996. Walter, Barbara F. “The Critical Barrier to Civil War Settlement.” International Organization 51, no. 3 (summer 1997): 335–364. ———. “Designing Transitions from Civil War.” International Security 24, no. 1 (summer 1999): 127–155. Walter, Barbara F., and Jack Snyder, eds. Civil War, Insecurity, and Intervention. New York: Columbia University Press, 1999. Weiland, Heribert, and Matthew Braham, eds. The Namibian Peace Process: Implications and Lessons for the Future. Freiburg: Arnold Bergstrasser Institut, 1994. Weiss, Thomas G. The United Nations and Civil Wars. Boulder, Colo.: Lynne Rienner, 1995. Whitfield, Teresa. “The Role of the United Nations in El Salvador and Guatemala.” In Comparative Peace Processes in Latin America. Eds. Cynthia Arnson and Carlos Basombrío. Stanford: Stanford University Press, 1999, 257–290. Williams, Michael. Civil-Military Relations and Peacekeeping. Adelphi Paper no. 321. New York: Oxford University Press, 1998. Williams, Philip J., and Knut Walter. Militarization and Demilitarization in El Salvador’s Transition to Democracy. Pittsburgh: University of Pittsburgh Press, 1997. Wilson, Alfred Jeyaratnam. The Breakup of Sri Lanka: The Sinhalese-Tamil Conflict. London: C. Hurst, 1988. Wippmann, David. “Enforcing the Peace: ECOWAS and the Liberian Civil War.” In Enforcing Restraint: Collective Intervention in Internal Conflicts. Ed. Lori Fisler Damrosch. New York: Council on Foreign Relations, 1993, 157–203. Woods, James. “Mozambique: The CIVPOL Operation.” In Policing the New World

S-Bibs

11/24/09

698

4:29 PM

Page 698

SELECTED BIBLIOGRAPHY

Disorder: Peace Operations and Public Security. Eds. Robert B. Oakley, Michael J. Dziedzic, and Eliot M. Goldberg. Washington, D.C.: National Defense University Press, 1998, 143–175. Woods, Ngaire. “Good Governance in International Organizations.” Global Governance 5, no. 1 (January–March 1999): 39–62. Woodward, Susan L. “America’s Bosnia Policy: The Work Ahead.” Brookings Institution Policy Brief no. 2. Washington D.C.: Brookings Institution, July 1996. ———. Balkan Tragedy: Chaos and Dissolution After the Cold War. Washington, D.C.: Brookings Institution, 1995. ———. Implementing Peace in Bosnia and Herzegovina: A Post-Dayton Primer and Memorandum of Warning. Brookings Discussion Papers. Washington D.C.: Brookings Institution, May 1996. Zartman, I. William, ed. Elusive Peace: Negotiating an End to Civil Wars. Washington, D.C.: Brookings Institution, 1995. ———. Ripe for Resolution. New York: Oxford University Press, 1985. Zolberg, Aristide R., Astri Suhrke, and Sergio Aguayo. Escape from Violence: Conflict and the Refugee Crisis in the Developing World. New York: Oxford University Press, 1989.

S-Contri

11/24/09

4:30 PM

Page 699

THE CONTRIBUTORS

Adekeye Adebajo is director of the Africa Program at the International Peace Academy in New York and adjunct professor at Columbia University’s School of International and Public Affairs. He is the author of Liberia’s Civil War: Nigeria, ECOMOG, and Regional Security in West Africa. He served on UN missions in Western Sahara, Iraq, and South Africa. Howard Adelman has been a professor of philosophy at York in Toronto since 1966. The founder and director of the Centre for Refugee Studies at York University and editor of Refuge until the end of 1993, he has written articles, chapters, and books on the Middle East, humanitarian intervention, membership rights, ethics, early warning, and conflict management, as well as on all aspects of the refugee issue. In 1999 he and Astri Suhrke coedited The Path to a Genocide: The Rwanda Crisis from Uganda to Zaire. Sumantra Bose teaches comparative politics in the Department of Government, London School of Economics and Political Science. His most recent books are Bosnia After Dayton: Nationalist Partition and International Intervention; and Kashmir: Sources of Conflict, Dimensions of Peace. Charles T. Call is assistant professor for research at Brown University’s Watson Institute for International Studies, and served as guest scholar at the U.S. Institute of Peace in 2001–2002. He has written on policing, human rights, and postconflict rule-of-law issues, most recently in Comparative Politics, Global Governance, and International Peacekeeping, with the 699

S-Contri

11/24/09

700

4:30 PM

Page 700

THE CONTRIBUTORS

support of a MacArthur Foundation research and writing grant. In 2001 he served as consultant on policing and justice issues to the UN Development Programme and the Department of Peacekeeping Operations. Elizabeth M. Cousens is director of the Conflict Prevention and Peace Forum in New York City. From 2000 to 2002, she served with the Office of the UN Special Coordinator for the Middle East Peace Process, heading aid coordination efforts. Previously, she was director of research at the International Peace Academy, where her research focused on postconflict reconstruction, institution-building, and strategic coordination of international assistance. George Downs is dean of social science and professor, Department of Politics, New York University. His areas of specialization are international institutions and international cooperation. Current research projects include books on the enforcement and effectiveness of multilateral environmental agreements and the role of reputation in international relations. His books include Tacit Bargaining, Arms Races, and Arms Control; Collective Security Beyond the Cold War; and Optimal Imperfection: Domestic Uncertainty and Institutions in International Relations. Michael W. Doyle is the Edwards S. Sanford Professor of Politics and International Affairs at Princeton University. His publications include Ways of War and Peace; Empires; UN Peacekeeping in Cambodia: UNTAC’s Civil Mandate; Keeping the Peace: Multidimensional UN Operations in Cambodia and El Salvador, which he edited with Ian Johnstone and Robert Orr; and Peacemaking and Peacekeeping for the New Century, edited with Olara Otunnu. He has been vice president and senior fellow of the International Peace Academy and is now a member of its board of directors. He is currently on leave from Princeton, serving as special adviser to UN Secretary-General Kofi Annan. Caroline A. Hartzell is associate professor of political science and coordinator of the Latin American Studies program at Gettysburg College in Pennsylvania. She has published articles on the negotiated settlement of civil wars in International Organization and other journals. David Holiday is the director of the Citizen Participation and Governance Project in El Salvador, implemented by Creative Associates International. A ten-year resident of Central America, he was the Central America representative of Human Rights Watch from 1991 to 1993, and has consulted for international organizations throughout Central America. He has written numerous articles and reports on peace processes, human rights, and democratization in Central America.

S-Contri

11/24/09

4:30 PM

Page 701

THE CONTRIBUTORS

701

Bruce D. Jones is currently special assistant to the UN Special Coordinator for the Middle East Peace Process. Until summer 2000, he was responsible for strategic coordination policy in the UN Office for the Coordination of Humanitarian Affairs. Prior to this, he was a Hamburg Fellow for the Prevention of Deadly Conflict at Stanford University. He is the author of Peacemaking in Rwanda: The Dynamics of Failure. Gilbert M. Khadiagala is associate professor of comparative politics and African studies and acting director of the African Studies Program at the Paul H. Nitze School of Advanced International Studies (SAIS), Washington, D.C. He is also the consulting director of the Africa Project at the Woodrow Wilson International Center for Scholars, Washington, D.C. He is the author of Allies in Adversity: The Frontline States in Southern Africa Security; Meddlers or Mediators? African Interveners in Civil Conflicts (forthcoming); and coeditor with Terrence Lyons of African Foreign Policies: Power and Process. Terrence Lyons is an assistant professor at the Institute for Conflict Analysis and Resolution, George Mason University. Previous positions have included fellow at the Brookings Institution (1990–1998) and senior researcher at the International Peace Research Institute, Oslo (1998–1999). He has participated in talks to mediate disputes in Ethiopia and served as senior program adviser to the Carter Center’s project on postconflict elections in Liberia in 1997. Among his publications are African Foreign Policies: Power and Processes (coedited with Gilbert Khadiagala); Voting for Peace: Post Conflict Elections in Liberia; Sovereignty as Responsibility: Conflict Management in Africa; and Somalia: State Collapse, Multilateral Intervention, and Strategies for Political Reconstruction. Sorpong Peou is associate professor of political science/international relations in the graduate program at Sophia University, Tokyo, Japan. He has written on UN peace operations and security issues, including Intervention and Change in Cambodia: Towards Democracy? and Conflict Neutralization in the Cambodia War: From Battlefield to Ballot-Box. Emily Plumb is an international programs specialist at the National Aeronautics and Space Administration (NASA). In 2000 she was David L. Boren National Security Education Program Graduate Fellow, studying grassroots peacebuilding initiatives in East Africa. John Prendergast is the codirector of the Africa Program of the International Crisis Group. He formerly was director for African affairs at the National Security Council and special adviser on African conflicts at the State Department. He is the author of six books on Africa, including

S-Contri

11/24/09

702

4:30 PM

Page 702

THE CONTRIBUTORS

God, Oil, and Country: Changing the Logic of War on Sudan; and Front Line Diplomacy: Humanitarian Aid and Conflict in Africa. Tonya L. Putnam is a doctoral student in political science at Stanford University. She served as a member of counsel on the international legal team of the Republic of Namibia from 1998 to 1999 in litigation before the International Court of Justice. Her primary research explores the causes and consequences of the extraterritorial reach of U.S. regulatory institutions in the human rights realm and beyond. Donald Rothchild is professor of political science at the University of California, Davis. His recent books include Managing Ethnic Conflict in Africa: Pressures and Incentives for Cooperation; Sovereignty as Responsibility: Conflict Management in Africa; The International Spread of Ethnic Conflict: Fear, Diffusion, and Escalation; and Africa in the New International Order: Rethinking State Sovereignty and Regional Security. He is currently working with Philip G. Roeder on a coedited volume titled Powersharing and Peacemaking. Joanna Spear is the director of the graduate research program, Department of War Studies, King’s College, London. In 1993–1995 she was a postdoctoral fellow at the Belfer Center, Harvard University. Her research interests include issues of conventional weapons. She is currently producing a book on the changing political economy of the defense trade for the Brookings Institution. William Stanley is associate professor of political science and interim director of the Latin American and Iberian Institute at the University of New Mexico. He has published a book titled The Protection Racket State: Elite Politics, Military Extortion, and Civil War in El Salvador, as well as numerous articles on violence, conflict resolution, and police reform in Central America. His research for this volume was funded by the U.S. Institute of Peace. Stephen John Stedman is senior fellow and interim codirector at the Center for International Security and Cooperation at Stanford University, where he directs the university’s honors program in international security studies and teaches international relations and political science. He is the author of Peacemaking in Civil War: International Mediation in Zimbabwe, 1974–1980 and The New Is Not Yet Born: Conflict Resolution in Southern Africa, and coeditor of Refugee Manipulation: War, Politics, and the Abuse of Human Suffering. His current work focuses on evaluating international efforts to manage civil wars. He has served as a consultant to the United

S-Contri

11/24/09

4:30 PM

Page 703

THE CONTRIBUTORS

703

Nations on light weapons and civil wars in Africa, peacekeeping in civil wars, and preventive diplomacy. Susan L. Woodward is professor of political science at the Graduate Center of the City University of New York. In 1994 she was head of the Analysis and Assessment Unit in the Office of the Special Representative of the Secretary-General for UNPROFOR, and in 1998 was a special adviser during the election period to the head of the OSCE mission to Bosnia and Herzegovina. She is the author of Balkan Tragedy: Chaos and Dissolution After the Cold War, selected as a Choice “Outstanding Academic Book 1995”; and Socialist Unemployment: The Political Economy of Yugoslavia, 1945–1990, winner of the 1996 Hewett Prize of the American Association for the Advancement of Slavic Studies; and numerous articles on southeastern Europe, the postcommunist transition in eastern Europe, state failure, peacekeeping, and postconflict reconstruction. Marie-Joëlle Zahar is an assistant professor of political science at the Université de Montréal. In 2000–2001 she held a postdoctoral fellowship from the Canadian Social Sciences and Humanities Research Council at the Munk Centre for International Studies at the University of Toronto.

S-Contri

11/24/09

4:30 PM

Page 704

S-Index

11/24/09

4:30 PM

Page 705

INDEX

Abacha, Sani, 604, 613 Abidjan Peace Agreement (1996), 292 ABuGiDa, 338–339 Abuja Accords, 37, 148, 225, 599. See also Liberia Accord on the Rights and Identity of Indigenous Peoples, 434 Accountability, creation of, 3 ActionAid, 105 Adebajo, Adekeye, 37, 599–626 Adelman, Howard, 28–29 Afghanistan, 106, 107; land mines in, 144; repatriation of refugees in, 279 African National Congress, 132, 147 Africans for Humanitarian Action, 339 Africa Watch, 249 Aidid, Farah, 125 Ajello, Aldo, 96, 97, 98, 99, 156, 205 Akashi, Yasushi, 17–18, 34, 81, 266n22, 505, 521 Akobo Peace Conference (1995), 343 Akosombo Agreement, 613 Aleman, Arnoldo, 374 Alianza Republicana Nacionalista, 76, 387, 392, 393, 399, 403 Amnesty International, 240, 264n1, 264n6, 322 Angola: amnesties in, 240; Angola Accords and, 155; arms embargoes in, 153, 159; Bicesse Accords and, 157, 159, 225; casualties in, 1; civil

war in, 1, 2, 284tab; confidencebuilding in, 62; demobilization in, 159; d’Hondt system in, 133; disarmament in, 142, 154, 156, 159; economic development in, 15; elections in, 160; human rights in, 46, 50, 240, 271tab; Lusaka Protocol and, 148, 149, 159; peace agreements in, 4, 14, 49, 51; Peace and National Reconciliation Fund, 161; peace implementation in, 59tab, 62, 81; peacekeeping in, 11, 51; police reform in, 308tab, 310; proportionality and, 132, 133; repatriation of refugees in, 46, 277tab; transitional governments in, 224, 225; UN Angola Verification Mission in, 82, 143, 152, 155, 159; Union for the Total Independence of Angola in, 152, 153, 157, 159, 160, 161, 181n45, 224; United Nations in, 6; violations of peace agreement, 11 Angola Accords, 155 Annan, Kofi, 64, 251 ANSESAL. See Salvadoran Security Agency Arab Fund for Economic and Social Development, 585 Arab League, 585, 593n15 Arbenz, Jacobo, 426, 435 Argentina, 357 Arias Sanchez, Oscar, 357

705

S-Index

11/24/09

706

4:30 PM

Page 706

INDEX

Armed Forces of Liberation, 385 Arnault, Jean, 96, 98, 422, 428, 432, 435, 442, 443, 459n15 Arusha Peace Agreement, 12, 279, 286, 463–493. See also Rwanda Arzú, Alvaro, 428, 430, 435 Association of Southeast Asian Nations, 501, 502 Authoritarianism, 217, 219, 384, 385, 426 Authority: administrative, 86fig; charismatic, 73–74; for decisions against the will of the parties, 83; ecologies of, 78–86; effective, 82; electoral, 27; executive, 83, 86fig; extent of, 79; formal, 82; local, 25, 73; political, 152; quasi-sovereign, 82; rational, 74; regional, 125; sources of, 72–74; state, 71, 72; successful exercise of, 74; supervisory, 83, 86fig; traditional, 73, 74, 334, 341, 343; transitional, 25, 71–86; trusteeship-like, 85; unstable factions and, 79 Autonomy: political, 126; regional, 125–128; territorial, 126 ‘Awn, Michel, 36, 570, 571, 573, 574, 575, 577, 580, 581, 594n27, 594n33 Babangida, Ibrahim, 613 Baghdad Pact, 568 Baker, James, 380n43, 388, 529n75 Barayagwiza, Jean-Bosco, 251 Belgium, 466, 467, 468, 470 Bermudez, Enrique, 355, 362 Bicesse Accords (1991), 157, 159, 225. See also Angola Birri, Nabih, 586, 587 Boley, George, 233n10, 603, 613, 625 Booh-Booh, Jacques-Roger, 485, 487, 488 Bose, Sumantra, 37, 631–658 Bosnia: ambivalence of settlement for, 532; amnesties in, 240; Arbitral Order in, 83; barriers to repatriation, 550–551; civilian implementation tasks, 540; civilian security in, 91, 547–548; civil war in, 7, 284tab, 533–539; Commission for Real Property Claims of Displaced Persons and Refugees in, 550; constraints on implementation, 541–545; Dayton Accords, 8, 35, 45–47, 82,

90, 91, 92, 225, 243, 287, 531–561; demobilization in, 161–163; democratization in, 554–558; disarmament in, 161–163; economic reconstruction in, 186; elections in, 35, 162–163, 216, 217, 534, 554–556; electoral supervision in, 229; Erdut Agreement in, 161; ethnic cleansing in, 284tab; European Union and, 190, 537, 540, 544; human rights in, 240, 247, 266n20, 271tab, 306; Human Rights Ombudsman in, 557; Implementation Force in, 6, 35, 45–47, 162, 246, 288, 540, 555; institutional reform in, 46; international actors in, 540–541, 542tab; international criminal tribunals in, 556, 558; international implementers in, 532; International Monetary Fund in, 204; international peacemaking efforts, 536–538; majority/minority repatriation, 549–550; media in, 557; military presence in, 92; minority issues in, 125; missed opportunities for peace in, 533; mixed implementation strategies in, 33–36; Mobile Support Units in, 314, 315; North Atlantic Treaty Organization in, 56, 246, 537, 538, 556; Office of the High Representative in, 91, 541; onset of war in, 534–536; Open Broadcast Network in, 557; Open Cities Initiative, 549; Organization for Security and Cooperation in Europe in, 91, 225, 226, 229, 540, 544, 554, 555; peace agreements in, 49; Peace Implementation Council in, 541; peace implementation in, 2, 59tab, 81, 531–561; peacekeeping in, 51, 536; police reform in, 308tab, 310; power sharing in, 121; preparation for implementation, 538–539; pressure for unification, 552–554; proportionality and, 132; psychosocial trauma iniatives, 335; Reconstruction and Return Task Force, 549; refugees and displaced persons, 29; repatriation of refugees in, 36, 275, 277, 277tab, 279, 287–288, 540, 548–551; reunification of, 551–554; role of international implementers in, 539–541; safe

S-Index

11/24/09

4:30 PM

Page 707

INDEX havens in, 63; security issues, 545–548; Stabilization Force in, 6; strategic coordination in, 90–92; transitional governments in, 225; trusteeship strategy in, 36, 533; UK Overseas Development Agency in, 190; unemployment in, 85, 550; UN High Commissioner for Refugees in, 91; United Nations in, 544; United States and, 90, 536; unsatisfactory compromise among warring parties, 532; UN Transitional Administration in, 162; World Bank in, 186, 204, 540; Yugoslav National Army in, 535, 536 Bosniac-Croat Federation, 35, 83 Boutros-Ghali, Boutros, 397, 403, 482, 484, 487, 490 Bowen, Hezekiah, 612, 614 Brazil, 357 Brcko, 84, 85 Burkina Faso, 604, 605, 606, 621 Burundi, 468, 469, 485, 486; media programming, 339; nongovernmental organizations in, 105; strategic coordination in, 94–95; UN special representatives of the Secretary-General in, 94 Bush, George, 363, 369, 387, 605 Buyoya, Pierre, 94 Cain, Kenneth, 255 Calderón Sol, Armando, 392, 402, 409 Calero, Adolfo, 355, 364 Call, Charles, 29, 31, 303–323, 383–414, 517 Cambodia; accommodation in, 78; alternative strategies in, 518–521; balance of power in, 34; civil administration in, 516; civilian security in, 307; civil service in, 193; civil war in, 76–77, 284tab, 501–504; coalition bias by UN Transitional Authority in Cambodia, 517–518; colonialism in, 77; compromise in, 520; confidencebuilding in, 35; culturalist approach, 525n1; culture of, 511; demilitarization of politics in, 223; demobilization in, 163–164, 508–509; disarmament in, 142, 163–164, 508–509; economic development in, 77; economic revival in, 508; elections in,

707

46, 78, 216, 217, 218, 509–510; electoral commissions in, 228; factionalism in, 510–518, 522; human rights in, 46, 247, 266n20, 271tab, 508–509; ideology in, 511–513; inflation, 522; institutionalist approach, 525n1; Khmer Rouge in, 11, 13, 77, 500, 502, 503, 508–509, 510, 512, 513, 527n35; “killing fields” in, 524; land mines in, 144, 507–508, 526n21; liberal internationalist approach, 525n1; long-term development, 504; loss of neutrality by UN Transitional Authority in Cambodia, 515–518; media programming, 339; mixed implementation strategies in, 33–36; moralist/legalist approach, 525n1; national recovery plan in, 204; National United Front for a Cooperative, Independent, Neutral Cambodia in, 502, 503, 510, 512, 513, 521, 528n74; occupation by Vietnam, 77; operational coordination in, 101; Paris Peace Accords, 11, 34, 78, 203–204, 243; peace agreements in, 4, 499–525; peace implementation in, 59tab, 62, 81, 499–525; police reform in, 308tab, 309; political instability in, 522; political violence in, 524; postholocaust syndrome in, 77; proportionality and, 132; as qualified success, 521–523; repatriation of refugees in, 46, 277, 277tab, 279, 289, 504, 507; royalist faction in, 77; security issues in, 513–515; spoilers in, 34; State of Cambodia in, 34, 77, 78, 503, 508–509, 512, 513; Supreme National Council in, 81, 503, 506; truth commissions in, 268n57; UN Civilian Police in, 517; UN Development Programme in, 508; unemployment in, 508; United Nations in, 16; United States in, 77; UN Transitional Authority in Cambodia in, 11, 17, 34, 46, 81, 184, 248, 254, 258, 266n20, 499, 500, 503, 505–518, 526n19 Canada, 362, 375; international development agencies in, 210n1; PeaceBuilding Fund, 189 Carter, Jimmy, 367, 375, 380n50

S-Index

11/24/09

708

4:30 PM

Page 708

INDEX

Catholic Bishops Conference, 339 Catholic Relief Services, 337, 339, 618 CDR. See Coalition pour la Défense de la République Cédras, Raoul, 80 Center for Conflict Resolution in South Africa, 338–339 Cerezo, Vinicio, 427, 431 Cesar, Alfredo, 362, 364 Chamorro, Violeta, 126, 131–132, 152, 353, 367, 368, 370, 371, 380n50, 380n51 Chamorro Barrios, Pedro Joaquín, 364 Chapultepec Accord (1992), 389, 416n25 Chavez, Robert, 208 China, 35, 78, 501 “Chinese Chapter Seven,” 73 Chissano, Joaquim, 129 CIAV. See International Commission of Support and Verification Civil society: composition of, 329; in Guatemala, 433; organizations in, 29, 329; peace talks and, 32; in postwar reconstruction, 331; repatriation and, 297n6; traditional authority and, 334; value of, 328 Civil war: in Angola, 1, 2; in Cambodia, 76–77; in Congo, 1; different types of, 283–290; to establish who will rule, 22; in Guatemala, 14, 30, 32; holistic approach to, 74; ideology in, 72; international assistance in ending, 19; in Lebanon, 568–569; in Liberia, 1, 2, 601–603; multiple issues in, 10; in Nicaragua, 30, 31; over people, 29, 284–285; in Rwanda, 1, 2; in El Salvador, 14, 30, 76, 383–389; undifferentiated view of, 3–20; varieties of, 5, 6–8; in Zimbabwe, 6 CIVPOL. See UN Civilian Police Claes, Willy, 487 Clark, Joe, 360 Clinton, Bill, 6, 397, 545, 555 Coalition pour la Défense de la République, 33, 93 Colombia, 392, 423 Community of Sant‘Egidio, 98, 112n9 Compaoré, Blaise, 604 Conakry Agreement (1997), 173, 292 Conciliation Resources, 104 Confidence-building, 6; in Angola, 62;

in Cambodia, 35; demobilization and, 26; disarmament and, 26, 142; effectiveness of, 11; ethnic extremism and, 62; in Guatemala, 32–33, 61, 433, 436; as implementation strategy, 33, 61; monitoring and, 33; in Mozambique, 33, 61; in Namibia, 33, 61; in Nicaragua, 32–33; in Rwanda, 33, 62; in El Salvador, 32–33, 61; transformation and, 17; transitional authority and, 25 Conflict: causes of, 2, 18, 74, 76, 79; class, 186; differentiation, 17; ecologies of, 25; ethnic, 129; internal, 31, 384; internationalization of settlement, 244; interstate, 30; intervention, 63; local, 218; management, 117, 118, 342; management training, 338–339; reduction, 47, 124; reinstitution of, 142; resolution, 2, 30, 47, 104; resource, 344; resumption of, 83; scarce-resource, 343; spread of, 47 Contadora, 357 Cooperation: incentives for, 15; in interim governments, 224 Coordination: continuity and, 97–98; effective, 89, 90; Friends groups and, 99–100; Integrated Missions and, 108–111; international financial institutions in, 98; mechanisms, 101–102; Organization of American States and, 98; in peace implementations, 89–111; Strategic Framework initiative, 105–108; UN Development Programme and, 98 COPAZ. See National Commission for the Consolidation of Peace Corruption: in Guatemala, 427, 432, 433, 449; in Liberia, 37, 601; in El Salvador, 402 Costa del Sol Declaration (1989), 363 Cotonou Agreements, 225, 610–612 Cousens, Elizabeth M., 35, 82, 90, 91, 92, 246, 287, 531–561 Crime rates, 187 Cristiani, Alfredo, 383, 384, 387, 398, 403, 410 Croatia, 533, 534, 537 Currency: devaluation, 466; erosion, 573; stabilization, 186 Cyprus: peace agreements in, 4

S-Index

11/24/09

4:30 PM

Page 709

INDEX Dallaire, General, 485, 486, 489, 490 Darby, John, 13 Dar es Salaam Conference (1991), 286 Dayton Accords (1995), 8, 45–47, 82, 91, 92, 127, 161, 225, 243, 265n11, 287, 301n49, 301n50, 531–561; Arbitral Order in, 83; coercion and, 35; contradictions in, 35 Debarge, Marcel, 474 Decisionmaking: autonomous, 125; inclusive, 121; paralysis of, 125; partitioned, 121; shared, 124 Del Castillo, Graciana, 194, 196, 197, 210n6, 411 De León Carpio, Ramiro, 427, 428, 433–435 Demilitarization of politics, 3, 27, 222–230; demobilization and, 226–227; disarmament and, 226–227; electoral commissions and, 228–230; interim governments and, 223–226; political parties and, 227–228; security and, 226– 227 Demobilization, 49–50, 141–175, 226–227; in Angola, 159; in Bosnia, 161–163; in Cambodia, 163–164, 508–509; cheating and, 156–158; child soldiers and, 145; confidencebuilding and, 26; criminality and, 305–306; defining, 145; failed, 225; financing, 193, 196; funding, 187; in Guatemala, 32, 446–448; guerrilla, 31; implementation environment in, 150–153; implementers of, 154–155; involvement of warring parties, 155–156; in Lebanon, 37, 577–579, 588; in Liberia, 148, 166–169; microinsecurity and, 146; in Mozambique, 168–169; in Namibia, 169–171; nature of conflict and, 152; in Nicaragua, 152, 171–172; in Northern Ireland, 145; peace agreements and, 148–150; as priority, 3, 26; psychological problems and, 146; reintegration and, 26, 145, 187; in El Salvador, 152, 164–166, 389; in Sierra Leone, 172–174; societal norms and, 145; in Somalia, 174–175; in South Africa, 147–148; sustained, 152; Ta’if Agreement and, 37; tasks of, 26; verification,

709

156–158; women combatants and, 145; in Zimbabwe, 152, 175–176 Democracy: bridge to, 27; creation of, 3; curtailment of, 37; electoral, 129, 216, 217; human rights and, 345–346; majoritarian, 129; power sharing and, 128; rhetorical commitment to, 20; security forces and, 304; success of peace agreements and, 14; transitional, 121 Democratic Republic of Congo, 281, 466; peace agreements in, 22; strategic coordination in, 111; UN special representatives of the SecretaryGeneral in, 113n21 Denmark, 189 Department of Peacekeeping Operations, 253 Des Forges, Alison, 268n41, 492 De Soto, Alvaro, 96, 98, 194, 196, 197, 210n6, 242, 391, 392, 393, 411, 416n25 Development: assistance, 189, 199–201; community, 30; economic, 87n17, 184, 204; long-term, 199, 320, 504; multilateral, 104; policy, 32 Development Program for Displaced Persons, Refugees, and Returnees, 337–338 Dhlakama, Afonso, 129, 235n44 Disarmament, 141–175, 226–227; in Angola, 142, 154, 159; in Bosnia, 161–163; in Cambodia, 142, 163–164, 508–509; cheating and, 156–158; coercive, 142; committment to, 5; confidence-building and, 26; cooperative, 142; in Guatemala, 446–447; implementation environment in, 150–153; implementers of, 154–155; incomplete, 306; involvement of warring parties, 155–156; in Liberia, 166–169; in Mozambique, 142, 144, 168–169; in Namibia, 169–171; nature of conflict and, 152; in Nicaragua, 152, 171–172, 369–373; peace agreements and, 148–150; process complexity, 143; role in civil war termination, 142; in El Salvador, 157, 164–166; in Sierra Leone, 172–174; societal norms and, 141; sociocultural role of weaponry and, 143–144, 151; in Somalia, 142,

S-Index

11/24/09

710

4:30 PM

Page 710

INDEX

143, 174–175; in South Africa, 144; in Sri Lanka, 142; tasks of, 26; timetables and, 148, 149–150; verification, 142, 156–158; in Zimbabwe, 175–176 Doe, Samuel, 601, 602, 604, 607 Dokie, Sam, 623 Downs, George, 24, 43–66, 247 Doyle, Michael, 25, 47, 151 Duarte, José Napoleón, 386, 387, 415n9 East Timor, 22, 80, 111; UN special representatives of the Secretary-General in, 113n21 ECOMOG. See Economic Community of West African States Cease-Fire Monitoring Group Economic: aid, 386; assistance, 184, 356; development, 15, 87n17, 184, 204; goals, 27; growth, 73, 188, 191; inequality, 186, 187, 357; interdependence, 15; neutrality, 192; organizations, 104; pluralism, 194; policy, 188; priorities in implementation, 27, 183–210; reconstruction, 27, 102, 103, 184, 191, 210n6; recovery, 197; reform, 191, 195, 198, 373; rehabilitation, 331; resources, 120; revival, 27; sustainability, 192 Economic Community of West African States, 125, 166–169, 172–174, 229, 253, 293, 599, 600, 604, 605, 606, 609, 615, 627n2 Economic Community of West African States Cease-Fire Monitoring Group, 23, 37, 56, 142, 148, 166–169, 172–174, 217, 599, 600, 602, 604, 605, 606, 607, 608, 610, 611, 615, 617 Economy: collapsed, 120; global, 14; industrial, 15; liberal, 18; war, 188 ECOWAS. See Economic Community of West African States Education, 193, 385, 424 Egypt, 568 Elections: authoritarianism and, 217; in Bosnia, 35, 216, 217, 534, 554–556; breakthrough, 216; in Cambodia, 46, 78, 216, 217, 218, 509–510; competitive, 217; for democratization, 27,

216–217, 219; demonstration, 218; disengagement and, 218; electoral commissions and, 228–230; electoral systems and, 221–222; founding, 383; funding, 188; goals of, 27; in Guatemala, 426, 427, 435–437; in Lebanon, 36, 216, 567, 579–580; legislative, 133; in Liberia, 216, 217, 599, 618–620; in Mozambique, 46, 169, 216, 217; multiparty, 388; multiple goals of, 216–219; in Namibia, 46, 87n17, 170, 216; in Nicaragua, 216, 217, 218, 353, 366–369; for organizational purposes, 27; parliamentary, 133, 579–580; postsettlement, 215–232; power sharing and, 219–222; as priority, 18; rapid, 35; return to constitutional rule and, 217; in El Salvador, 31, 46, 165, 216, 217, 383, 388, 408–409; settlement, 27; in Sierra Leone, 173–174, 216, 217; in Somalia, 175; war termination, 27, 216, 218; in Zimbabwe, 175–176, 216, 217 Erdut Agreement (1995), 83, 161, 194, 205, 308 Escobar Galindo, David, 399 Esquipulas Agreement (1987), 171–172, 273, 359, 360–366 Esquipulas Peace Process, 356–360 Ethiopia: ABuGiDa in, 338–339; Ad Hoc Committee for Peace and Development, 338–339; Ethiopian People’s Revolutionary Democratic Front in, 126; ethnic federalism in, 128; indigenous associations in, 342–343; minority issues in, 125; peace radio in, 339–340; problemsolving workshops in, 337 Ethiopian People’s Revolutionary Democratic Front, 126 Ethnic: clashes, 344; cleansing, 240, 292, 592n10; extremism, 29, 62; favoritism, 37; homogenization, 292 Ethnonationalism, 127 Eurobonds, 585 European Convention for the Protection of Human Rights and Fundamental Freedoms, 557 European Court of Human Rights, 540 European Investment Bank, 585

S-Index

11/24/09

4:30 PM

Page 711

INDEX European Parliament, 432 European Union, 104, 109, 537, 621; in Bosnia, 190, 540, 544; in Kosovo, 102, 103 Evaluation: comparative, 61; endogeneity, 24; goal levels and, 45–46; of implementation strategies, 23–26; implementation tasks, 26–30; institutional effectiveness, 52–54; in international actors, 24; issues in implementation, 43–66; mandate achievement in, 45–47; measuring success, 44–54; mission type vs. mission difficulty, 53; noncomparability of mandates, 24; operationalization of success/failure, 50–52; short-term indicators, 24; subgoal achievement, 49–50; “willingness score” and, 57–58 Evans, Gareth, 521 Executive Outcomes, 148 Farabundo Martí National Liberation Front, 76, 157, 224, 225, 383, 385, 387, 388, 391, 393, 394, 395, 396, 399, 403, 408 Farrand, William, 83, 84 Fatah, 578 Federalism, 125–128; nonterritorial, 133–134 Ferey, Azucena, 364 Financial institutions, international: conditionality policies of, 207; in implementation, 187–188; involvement in implementation, 114n25; macroeconomic restraint policy, 193; neoliberal orthodoxy of, 192; political advantages of, 198; rigidities of, 193–194; role in implementation, 27; in El Salvador, 411 First Paris Conference (1989), 503 Food and Agriculture Organization, 104, 618 France, 466, 467, 470, 475, 528n74, 573, 574, 606 Franjiyya, Suieyman, 569 Frelimo. See Front for the Liberation of Mozambique Friends of the Secretary-General, 16, 17, 25, 65, 89, 99–100 Front for the Liberation of Mozambique, 129, 229

711

Gambia, 604, 606 Gandhi, Indira, 633, 637 Gandhi, Rajiv, 38, 637, 638, 639, 645, 652 Garcia-Sayan, Diego, 256 Genocide, 62, 337, 456; in Rwanda, 33, 93, 112n7, 284tab, 293, 463, 473 Gerardi, Juan, 444 Germany, 189, 362, 464, 470 Ghana, 606, 607, 612 Gorbachev, Mikhail, 360 Goulding, Marrack, 251, 389, 392, 393 Governance: capacity for, 120; effective, 130; electoral systems and, 119; interim, 223–226; “popular,” 386; proportional distribution and, 119; proportionality principle, 130–133; proportional representation and, 119, 123, 221–222; unitary, 128 Griffiths, Martin, 97 Guatemala: Accord on the Rights and Identity of Indigenous Peoples, 434; amnesties in, 240; autogolpe in, 427, 433; civilian police force in, 422; Civil Society Assembly in, 101; civil society in, 421–422, 433; civil war in, 7, 14, 30, 32, 284tab; confidencebuilding in, 32–33, 61, 433, 436; constitutional reform in, 438–440, 450, 460n34; Coordinating Committee of Commercial, Industrial, and Financial Associations, 432; coordination mechanisms in, 89; Coordinator of Mayan Peoples’ Organizations of Guatemala, 439; corruption in, 427, 432, 433, 449; counterinsurgency in, 421, 426; demobilization in, 32, 423, 436, 446–448; democratization in, 422; disarmament in, 446–447; elections in, 426, 427, 435–437, 437; financial crises in, 428; Framework Agreement in, 433; Friends of the Secretary-General in, 99; Guatemalan Christian Democracy in, 427; Guatemalan National Revolutionary Unity in, 421, 422, 423, 426, 428, 431, 432, 433, 434, 436, 439, 440, 441, 446–447, 447, 452, 460n32; Guatemalan Republican Front in, 422, 437, 439, 440, 442; Guatemalan Workers Party

S-Index

11/24/09

712

4:30 PM

Page 712

INDEX

in, 425; Guerrilla Army of the Poor in, 425; Human Rights Council in, 248; human rights in, 240, 242, 248, 257, 271tab, 432, 434, 443–445, 456; incentives to negotiation, 429–431; indigenous rights/identity in, 454–456; Inter-American Development Bank in, 443, 444; interim security in, 317; international assistance for, 430; International Monetary Fund in, 443; international role in implementation, 441–443; judicial reform in, 424; judiciary in, 449; labor laws in, 452; land reform in, 445, 451–452; land usurpation in, 455; Mayan community in, 425–426; military budget in, 422; military context of negotiations, 425–426, 430–431; military coup in, 426, 427; multiculturalism in, 424; Multipartidaria in, 440; National Advancement Party in, 423, 427, 435, 437, 439, 440, 441, 442; National Civilian Police in, 317, 324n17, 449, 450; National Liberation Movement in, 427; New Guatemala Democratic Front in, 427, 439, 440, 444; nongovernmental organizations in, 257; Ombudsman for Human Rights in, 248; Organization of American States in, 432, 447; Organization of the People in Arms in, 425; organized labor in, 452; peace agreements in, 14, 32; peace implementation in, 30, 59tab, 62, 118, 421–458; peacekeeping in, 51; police controls in, 321, 322; police reform in, 308tab, 310; political context for negotiation, 426–428; political party system in, 438; political violence in, 444; privatization in, 152; public security reform, 448–451; repatriation of refugees in, 46, 276, 277tab, 278, 297n6, 298n11, 298n12, 421, 445–446; Revolutionary Armed Forces in, 425; Rural Development Bank, 454; SE Accord, 452; security forces in, 324n17, 448–451; Serranazo in, 427; social spending in, 442; socioeconomic issues in, 429, 451–454; Special Incorporation Commission

in, 447; strategic coordination in, 96, 102; tax policy in, 195; tax revenues in, 442, 452, 453; Timetable Agreement in, 440–441; truth commissions in, 46, 268n57, 434; UN Development Programme in, 443, 446; United Nations in, 6, 15, 441–443, 459n15; United States and, 423, 426, 432; UN-moderated negotiations in, 433–435; UN Verification Mission in Guatemala in, 424, 434, 440–441, 442, 443, 444, 446–447, 451, 453; World Bank in, 443, 444 Guatemalan National Revolutionary Unity, 421, 422, 423, 426 Guatemalan Republican Front, 422 Guinea, 606 Gutiérrez Castro, Mauricio, 403 Habyarimana, Juvenal, 92, 286, 301n47, 463, 464, 465, 466, 467, 469, 470, 472, 473, 476, 484, 485, 486, 487, 488, 489 Hackett, Ken, 339 Haiti, 22, 40n37, 80; forcible intervention in, 72; interim security in, 316–317; police reform in, 308tab; security forces in, 324n15 Hani, Chris, 341 al-Hariri, Rafiq, 581, 585, 587 Hartzell, Caroline, 30, 353–377, 387 Healthcare, 193, 385 Holiday, David, 32, 125, 421–458 Honduras, 372, 386; Contras in, 361 Houphouet-Boigny, Félix, 603 Hrawi, Ilyas, 574, 577, 586 Human rights: abuse cover-ups, 305; advocacy, 28, 256; amnesties and, 240; in Angola, 46, 271tab; in Bosnia, 240, 247, 266n20, 271tab; in Cambodia, 46, 247, 266n20, 508–509; capacity-building for, 3; in constitutions, 246–247; curtailment of, 37; education, 255–259; enforcement approach, 28, 238, 239; ethnic cleansing and, 240; grassroots promotion of, 256–257; in Guatemala, 240, 242, 257, 271tab, 432, 434, 443–445; implementation of, 30; institution-building for, 255–259, 406–407; instrumentalist approach,

S-Index

11/24/09

4:30 PM

Page 713

INDEX 28; international standards, 308; isolation of, 241; lack of resources for verification, 254; in Lebanon, 271tab, 567; in Liberia, 37, 243, 255, 271tab; limits on, 18; in Mozambique, 245, 262tab; in Namibia, 240, 245, 262tab; in Nicaragua, 242, 262tab, 374; nongovernmental organizations, 47; outside enforcement approach, 255–259; in peace implementation, 239–241; protection, 3, 187; provisions in settlements, 241–248; rhetoric of, 20, 253; in Rwanda, 249, 250, 262tab, 477, 484; in El Salvador, 31, 46, 242, 247, 257, 386, 388, 390, 399, 402–408, 403, 416n23; in Sierra Leone, 240, 243, 247, 262tab; in Somalia, 243, 262tab; in Sri Lanka, 262tab; standard-setting, 255–259; as stumbling block to agreement, 242; sustainable peace and, 237–263; turnkey justice and, 248–251; United Nations role in, 244–245; violations, 237, 238, 242, 386; violations by police, 306; in Zimbabwe, 262tab. See also Organizations, international human rights Human Rights Watch, 50, 249, 264n1, 264n6, 268n41, 484 Hun Sen, 510, 513, 524 al-Huss, Salim, 590 Hutu peoples, 33, 464, 465, 474, 485, 486 IFOR. See Implementation Force IHRO. See Organizations, international human rights Ikimi, Tom, 614 Implementation Force, 6, 35, 45–47, 73, 162, 246, 288, 540, 555 Implementation strategies: compellence, 6; confidence-building, 6 Independent National Patriotic Front of Liberia, 602 India: International Center for Peace Initiatives in, 337; peace committees in, 341; in Sri Lanka, 6, 23, 37–38, 633, 636–640, 643–654; strategic influence in Indian Ocean, 633 Indo–Sri Lanka Peace Agreement, 631–658

713

Infrastructure: funding for, 188; social, 193 Institutions: building, 117, 118, 203–205, 223, 248, 316–322, 333, 384, 406–407; domestic, 246; economic, 191; electoral, 27; evaluation of effectiveness of, 52–54; interim, 223, 226; internal security, 304; judicial, 249; legal, 46; lending, 218; liberal, 304; local, 333, 341; multiethnic, 36; political, 16, 17, 119, 215, 304, 531; reform of, 32, 46, 191; security, 187; social, 16; state, 119, 151, 568, 589–590; supranational, 126; temporary, 124; transformation of, 17; transitional sovereignty, 81; trust in, 118, 130 Insurrection, 146 InterAfrica Group, 337, 340 Interahamwe, 467 Inter-American Development Bank, 412, 443, 444 International Alert, 104, 105, 112n9 International Center for Peace Initiatives, 337 International Commission of Support and Verification, 365, 373, 376 International Committee of the Red Cross, 504, 540 International Covenant on Civil and Political Rights, 267n31, 582 International criminal tribunals: in Bosnia, 239, 250, 251, 556, 558; in Rwanda, 239, 268n50; in Yugoslavia, 239 International Crisis Group, 104, 105, 112n9 International Finance Corporation, 585 International Fund for Agricultural Development, 188 International Fund for the Reconstruction of Lebanon, 573 International Labour Organization, 104, 618 International Monetary Fund, 188, 197; in Bosnia, 204; conditionality and, 191, 198; currency boards, 204; default on loans, 194; dependence on donors, 198; economic neutrality and, 192; in Guatemala, 443; in Rwanda, 466; in El Salvador, 394;

S-Index

11/24/09

4:30 PM

Page 714

714

INDEX

sovereign lending by, 193–194; transitional authority of, 88n24 Interventions: authorization for, 72; forcible, 72; humanitarian, 72 Investment: attracting, 191; domestic, 186; foreign, 186, 191, 411, 585; foreign direct, 198; private, 195, 585 Iran-Contra scandal, 357 IRIN. See UN Integrated Regional Information Network Irish Republican Army, 13 Israel, 578; conflict with Syria, 37; Israel Defense Force, 569; Operation Grapes of Wrath, 585, 596n81; strategic coordination in, 96 Izetbegovic, Alija, 534, 535 Jayewardene, Junius, 639, 645, 647 Johnson, Prince Yeduo, 602 Johnson, Roosevelt, 125, 602, 603, 614, 615, 623, 625 Johnstone, Ian, 47 Jones, Bruce, 25, 64, 65, 89–111, 154, 194 Judiciary: collapse of, 306; creation of, 188; impartial, 238; institutions of, 249; reform of, 321, 402–408, 460n35 Jumayyil, Pierre, 568–569, 570, 592n9 Jumblatt, Walid, 579–580, 584 Kabbah, Tejan, 624 Karadzic, Radovan, 551, 556 “Katanga Rule,” 73 Kayibanda, Grégoire, 464 Khadiagala, Gilbert, 33, 463–493 Khmer Rouge, 11, 77, 502, 508–509, 510, 512, 527n35; attempts to force compliance from, 34; in Cambodia, 500; China and, 35, 78; discipline in, 77–78; egregious past of, 13; human rights and, 266n20; massacres by, 77; as spoiler, 500; withdrawal from process, 34 Klanac, 83, 84 Klein, Jacques, 83 Kosovo, 22; European Union in, 102, 103; forcible intervention in, 72; minority issues in, 125; Mobile Support Units in, 314, 315; North Atlantic Treaty Organization in, 103; Organization for Security and

Cooperation in Europe in, 102, 103; Police Support Units, 315; strategic coordination in, 90, 102, 110–111; UN High Commissioner for Refugees in, 102, 103; United Nations in, 102, 103; UN special representatives of the Secretary-General in, 97, 113n21 Kosovo Force, 102, 103, 161 Kosovo Liberation Army, 110, 162, 307 Kosovo Protection Corps, 109 Kosovo Verification Mission, 104 Kovanda, Karel, 490 Krajisnik, Momcilo, 552 Kromah, Alhaji, 125, 233n10, 602, 603, 612, 614, 615, 625 Lahoud, Emile, 590, 591 Lancaster House Agreement (1980), 6 Land: acquisition, 445; distribution, 148, 149–150; distribution inequality, 76; ownership, 148, 149–150, 391, 394, 396; redistribution, 186; reform, 297n6, 451–452; tenancy, 341; transfers, 395, 410 Land mines, 144 Law enforcement, 187 Lebanon: Arab Fund for Economic and Social Development and, 585; Chamber of Deputies in, 128; Christian Lebanese Forces, 578; civil disobedience in, 588; civil war in, 284tab, 568–569; coercive strategies in, 36–38; compliance with agreements, 574–575; confessionalism in, 128; content of agreement, 571–572; demobilization in, 37, 166–167, 577–579, 588; dissent in, 589–590; dissolution of Parliament, 574; elections in, 36, 216, 567, 579–580; ethnic cleansing in, 592n10; European Investment Bank and, 585; Fatah in, 578; foreign investment in, 585; “Greater Beirut” security plan, 577; Hizballah in, 578, 588, 589, 596n78; human rights in, 271tab, 567; implementation environment in, 572–573; Internal Security Forces, 572, 578; International Finance Corporation and, 585; International Fund for the Reconstruction of Lebanon, 573;

S-Index

11/24/09

4:30 PM

Page 715

INDEX judiciary in, 580–582; Lebanese National Movement in, 568, 569; Lebanese Popular Congress, 581; media in, 582–583, 590; military services in, 580–582; National Emergency Recovery Program, 585; National Pact in, 568, 592n2; negotiation compellence, 574–575; negotiation process, 570–571; nongovernmental organizations in, 245, 266n25, 590; Operation Clean Hands, 590; Organization of African Unity in, 600; Palestine Liberation Organization in, 578, 592n7; peace implementation in, 59tab, 572–587; political deconfessionalization in, 586–587; political reform in, 577; power sharing in, 121; private sector in, 585; Progressive Socialist Party in, 235n41; reconstruction of, 573, 584–586; refugees and displaced persons, 29; religious denominations in, 568, 569, 574, 592n2; repatriation of refugees in, 275, 276, 277tab, 288, 572, 573, 583–584, 596n66; Second Republic, 574, 587–591; state institutions in, 589–590; Syria as implementer of Ta’if Agreement in, 575–576; Syria in, 6, 36–38, 56, 567, 569, 572, 574; Ta’if Agreement, 36, 122, 567–592; unemployment in, 596n73; UN Interim Forces in Lebanon in, 569; United States and, 573; war losses, 573; World Bank in, 585 Legitimacy: political, 130 Liberalization, 191; political, 469; price, 192 Liberation Tigers of Tamil Eelam, 38, 634, 637, 641, 643 Liberia: Abuja Agreements in, 37, 225; Armed Forces of Liberia in, 600, 609; Association for Constitutional Democracy in Liberia, 607; civilian security in, 623; civil war in, 1, 2, 601–603; coercive strategies in, 36–38; collective presidency in, 124; corruption in, 37, 601; Cotonou Agreements, 225, 610–612; Council of State in, 124; demobilization in, 148, 167–168, 615, 617; disarmament in, 142, 167–168, 615,

715

616–617; Economic Community of West African States Cease-Fire Monitoring Group in, 23, 37, 56, 167–168, 217, 599, 600, 602, 604, 605, 606, 607, 608, 610, 611, 615, 617; Economic Community of West African States in, 125, 167–168, 229, 600, 604, 605, 606, 609, 615; elections in, 216, 217, 599, 618–620; electoral fraud in, 228; electoral supervision in, 229; ethnic rivalries in, 601; evaluating implementation, 620–622; humanitarian assistance to, 617–618; human rights in, 37, 243, 255, 271tab; Independent National Patriotic Front of Liberia in, 602; insolvency, 601; international aid to, 617–618; National Patriotic Front of Liberia in, 599, 600, 602, 603, 609, 610; National Patriotic Reconstruction Assembly government, 609; Operation Octupus in, 609; Organization of African Unity and, 605, 610, 611; peace agreements in, 14, 22, 51, 599–626; peace implementation in, 59tab; police reform in, 308tab, 309; power sharing in, 124; quality of peace in, 622–626; repatriation of refugees in, 276, 277tab, 278, 279, 615; repression in, 600; spoilers in, 605, 608; Standing Mediation Committee, 606; transitional governments in, 225; UN Development Programme in, 618; United Liberation Movement of Liberia for Democracy in, 602, 604, 609, 610, 615; United Nations in, 605; United States and, 37, 604; UN Observer Mission in Liberia in, 610, 611; warlords in, 603; Yamoussoukro Agreements, 225, 608 Libya, 604, 606 Lima Group, 357, 359 Lomé Peace Accord (1999), 121–122 Lopez-Pintor, Rafael, 367 LTTE. See Liberation Tigers of Tamil Eelam Lusaka Protocol (1994), 148, 149, 159, 280 Lutheran World Relief, 344 Lyons, Terrence, 27, 215–232, 345– 346

S-Index

11/24/09

716

4:30 PM

Page 716

INDEX

McNamara, Dennis, 515 Mahdi, Ali, 125 Mali, 605, 606 Managua Agreement, 370, 372 Mandates: achievement of, 45–47; ambiguity in, 46–47; comparability issues, 46; on human rights issues, 245–246; implementation, 245–246; inflation of goals and, 47; limited, 46; overlapping, 90; performance with, 46; reneging on, 47; UN Transitional Authority in Cambodia, 505–518 Mandela, Nelson, 74, 133–134, 147 Martens, Wilfried, 467 Massaquoi, Francois, 603, 614 Mediation: by international actors, 1 Mejía Victores, Humberto, 426 Méndez Montenegro, Julio César, 426 Mengistu Haile Mariam, 126 Mexico, 392, 423 Mexico Accords (1991), 419n88, 428 Military: aid, 386; alliances, 102; casual fighters, 148, 149, 151; child soldiers, 145; ghost soldiers, 148, 149; integration, 479; nonstate formations, 152; reform, 76, 152, 389–390; repression, 385 Milosevic, Slobodan, 83, 295, 534 MINUGUA. See UN Verification Mission in Guatemala MINUSAL. See UN Mission in El Salvador Mobile Support Units, 314, 315 Mobutu Seso Seko, 468, 469 Monitoring and verification of compliance, 6, 8; confidence-building and, 33; as peacekeeping strategy, 16; provision during demobilization/disarmament, 26; transitional authority and, 25 Moro Islamic Liberation Front, 126 Moro National Liberation Front, 126 Mouvement Révolutionnaire National pour le Développement, 92, 93, 464 Mozambique, 29; Cease-Fire Commission, 169; Circus of Peace in, 340; civil war in, 284tab; confidence-building in, 33, 61; coordination mechanisms in, 89; demilitarization of politics in, 223; demobilization in, 168–170; disarmament in, 142, 144, 148, 149, 156, 168–170;

elections in, 46, 169, 216, 217; electoral commissions in, 228; electoral supervision in, 229; food aid in, 194; Front for the Liberation of Mozambique in, 129, 229; human rights in, 245, 262tab; land mines in, 144; Mozambique National Resistance in, 123, 129, 205, 224, 229, 289; National Elections Commission in, 229; Norway in, 212n28; operational coordination in, 101; peace agreements in, 14, 49; peace committees in, 342; peace implementation in, 59tab, 62, 118; peacekeeping in, 51; police controls in, 321, 322; police reform in, 308tab; proportionality and, 123, 132; repatriation of refugees in, 46, 276, 277tab, 278, 279, 289; Rome Accords and, 168, 229, 245; strategic coordination in, 96, 97, 98, 102; Ta’if Agreement in, 245; technical assistance to, 213n51; transitional governments in, 224, 225; United Nations in, 6; UN Operation in Mozambique in, 148, 149, 208, 245; UN special representatives of the SecretaryGeneral in, 97, 98 Mozambique National Resistance, 123, 205, 224, 229, 289 Mu‘awwad, René, 574, 594n31 Mugabe, Robert, 122–123 Museveni, Yoweri, 467, 468, 473, 484, 485 Mwinyi, Hassan, 468, 473, 489 Namibia, 1; amnesties in, 240; civil war in, 284tab; confidence-building in, 33, 61; demobilization in, 170–171; disarmament in, 170–171; economic development in, 87n17; elections in, 46, 87n17, 170–171, 216; human rights in, 240, 262tab; LR-Hare formula in, 133; nongovernmental organizations in, 155; peace agreements in, 2, 4, 14, 49; peace implementation in, 59tab, 118; police reform in, 308tab; proportionality and, 132, 133; repatriation of refugees in, 276, 277, 277tab, 279; South West African People’s Organization in, 133, 170–171; United Nations in, 6, 19;

S-Index

11/24/09

4:30 PM

Page 717

INDEX UN Transitional Assistance Group in, 170–171, 245 National Commission for the Consolidation of Peace, 80–86, 224, 225, 391, 400 National Democratic Organization, 385 Nationalism: civic, 127; ethnic, 127 National Liberation Front, 171–172 National Peace Commission, 340 National United Front for a Cooperative, Independent, Neutral Cambodia, 502, 510, 512, 513, 521, 528n74 NATO. See North Atlantic Treaty Organization Ndadaye, Melchior, 485 Netherlands, 189, 198 NGOs. See Nongovernmental organizations Nicaragua: amnesties in, 379n29; Chamorro regime, 126, 152, 367, 368, 369, 370; civilian security in, 307; civil war in, 30, 31, 284tab, 354–356; confidence-building in, 32–33; Contadora in, 357; Contras in, 171–172, 355, 356, 361, 362, 371, 378n9, 379n38, 380n43; Costa del Sol Declaration, 363; demobilization in, 152, 171–172, 370; democratization in, 362; development zones in, 371–372, 373; disarmament in, 152, 171–172, 369–373; economic reform in, 373; elections in, 216, 217, 218, 353, 363, 366–369; evaluating the peace in, 373–375; human rights in, 242, 262tab, 374; implementation regime in, 353–377; indigenous demands in, 355; International Commission for Verification and Follow-Up, 358; International Commission of Support and Verification in, 365, 372, 373, 376; judiciary in, 374; Lima Group in, 357, 359; Managua Agreement, 370, 372; minority issues in, 125; National Guard in, 354, 379n29; National Opposition Union in, 366–369, 376; National Peace Commission in, 340; National Police Force, 368; National Reconciliation Commission in, 359, 360; Nicaraguan Resistance, 371, 372; Organization of American States in,

717

353, 380n42, 380n43, 380n50; peace agreements in, 2, 49; peace implementation in, 30, 59tab, 118; peacekeeping in, 51; peace process in, 353–377; police reform in, 308tab, 309; political repression in, 354; “Procedure for the Establishment of a Firm and Lasting Peace in Central America,” 358; proportionality in, 131–132; psychosocial trauma iniatives, 335; repatriation of refugees in, 276, 277tab, 278, 289; Rio Group in, 99, 357; Sandinista People’s Army in, 355, 356; Sandinistas in, 131–132, 171–172, 354, 361, 362, 363, 366, 367, 368, 369, 370, 380n42; Sapoa Agreement and, 362; security forces in, 323n3; Southern Opposition Bloc in, 355; structural adjustment in, 373; Technical Assistance Commission in, 361; Tela Agreement, 364, 365; Tesoro Beach Agreement, 363; Toncontin Accord, 370; transition accord in, 366–369; United States and, 355, 360, 361, 362, 377n7; United States Central Intelligence Agency in, 355, 378n13; UN Observer Group in Central America in, 171–172, 365, 370, 371, 372, 376, 381n70, 381n71; UN Observer Mission for the Verification of Elections in Nicaragua in, 367; verification/monitoring in, 358, 363 Nicaraguan Democratic Forces, 355 Niger, 605 Nigeria, 131, 600, 604, 606, 607 Nongovernmental organizations: autonomy of, 213n45; in Burundi, 105; civil society organizations and, 335–336; Community of Sant’Egidio, 98, 112n9; conflict resolution, 104, 105, 112n9; development and relief, 104; and disarmament, 154–155; diversity of operations, 213n45; humanitarian assistance, 112n9; human rights, 47, 254, 407; implementation roles, 26; international, 19, 26, 189, 331, 345–346; in Lebanon, 590; lobbying by, 47; local, 245, 329; natural disaster assistance model, 189–191; in peace implementation, 90; reconcilia-

S-Index

11/24/09

718

4:30 PM

Page 718

INDEX

tion activities by, 329; in Rwanda, 481; in the South, 270n72 Norodom Chakropong, 519 Norodom Ranariddh, 510, 513, 522 Norodom Sihanouk, 74, 77, 502, 513 North Atlantic Treaty Organization, 23, 102, 114n42; in Bosnia, 56, 246, 537, 538, 556; in Kosovo, 103; war criminals and, 556 Northern Ireland, 13, 22; civil society in, 328; demobilization and, 145; Good Friday Agreement, 310; police reform in, 308tab, 310, 311–316; power-sharing in, 121; problem-solving workshops in, 336 Norway, 189, 213n45, 423; Agency for Development Cooperation, 212n28; in Mozambique, 212n28 Nsengiyaremye, Dismas, 470, 472 OAS. See Organization of American States OAU. See Organization of African Unity Obando y Bravo, Cardinal Miguel, 360, 370, 371 OECD. See Organization for Economic Cooperation and Development Office of the High Representative, 91, 541; in Bosnia, 91 OHR. See Office of the High Representative Olurin, Adetunji, 610 ONUCA. See UN Observer Group in Central America ONUMOZ. See UN Operation in Mozambique ONUSAL. See UN Observer Mission in El Salvador ONUVEN. See UN Observer Mission for the Verification of Elections in Nicaragua Opaleye, Ekundayo, 475 Operation Support Hope, 46 ORDEN. See National Democratic Organization Organization for Economic Cooperation and Development: Development Assistance Committee, 199, 210n1 Organization for Security and Cooperation in Europe, 82, 102, 104, 108, 109, 113n13, 114n42, 225, 226,

253; in Bosnia, 91, 229, 540, 544, 554, 555, 557; electoral supervision role, 228–229; in Kosovo, 102, 103; police development and, 320 Organization of African Unity, 92, 113n13, 264n1, 468, 469, 470, 475, 476, 479, 621; in Lebanon, 600; Liberia and, 605, 610, 611, 615; in Rwanda, 93 Organization of American States, 30, 264n1, 380n42, 380n43, 380n50; coordination and, 98; in Guatemala, 432, 447; in Nicaragua, 353; El Salvador and, 31; Technical Assistance Commission, 361 Organizations, civil society, 29, 327, 340–342; commissions, 340–342; conflict management training by, 338–339; domestic, 345–346; facilitating conditions for, 329–331; indigenous mechanism for conflict management in, 342–344; initiatives in peacebuilding, 334–346; integration of local capacity by, 331–334; local knowledge and, 332; maximization of influence by, 330; network development by, 330; in Nicaragua, 374; nongovernmental organizations and, 335–336; peace committees, 340–342; peace promotion through media, 339–340; problem-solving processes, 336–338; psychosocial trauma initiatives, 334–336; roles of, 329–331; support for peacebuilding at subconstitutional level, 330 Organizations, international: in peace implementation, 90 Organizations, international human rights, 237–263; ability to take action by, 247; assignment of responsibility for rights by, 238; biases and, 249; civil society organizations and, 345–346; criticism of performance by, 255; errors of, 28; formal enumeration of provisions in settlements by, 238; in grassroots promotion of human rights, 257; insistence on prosecuting, 251; as interlocutors, 257–258; isolation of, 251; mission level issues, 259; overplaying hand by, 240; in peace implementation, 28; resources and skills of, 270n69;

S-Index

11/24/09

4:30 PM

Page 719

INDEX responsibilities of, 237; in United Nations structure, 251 Organizations, regional: coordination and, 98; Economic Community of West African States Cease-Fire Monitoring Group, 23; North Atlantic Treaty Organization, 23; operational roles, 104; in peace implementations, 73, 90; in political negotiations, 104 Orr, Robert, 47 Ortega, Daniel, 364, 374, 380n50 Ortega, Humberto, 368, 369, 380n51 OSCE. See Organization for Security and Cooperation in Europe Oslo Accords, 184 Oslo process, 428, 431 Ould-Abdallah, Ahmadou, 94 Owen, David, 537 Owen, Roberts, 84 Palestine, 568; Joint Liaison Committee and, 101; Oslo Accords and, 184; Popular Front for the Liberation of Palestine, 578; problem-solving workshops in, 336; strategic coordination in, 96 Palestine Liberation Organization, 13, 578, 592n7 Paris, Roland, 49 Paris Peace Accords (1994), 11, 34, 81, 243, 499. See also Cambodia Paris Peace Agreement (1991), 203–204 Paris Peace Conference (1989), 78, 503 Partners for Democratic Change, 341 Peace: capacity for, 75; civil, 71; commitment to, 186; compellence of, 37; consolidation of, 120; culture of, 112n9; democratic, 34; dividends, 186; enduring, 49; enforcement, 46; establishment of, 71; financing, 187–188; funding, 196–199; hegemonic, 34, 37; human rights and, 237–263; incentives, 71; negative, 289–290; positive, 289–290; real, 43; self-enforcing, 117; self-sustaining, 79, 120; social, 191, 198; social construction, 297n6; sustainable, 85, 188, 237–263; technology, 196; transitions to, 311–316 Peace agreements: in Cambodia, 499–525; coalition-based, 10; com-

719

pliance with, 2; contradictions in, 10; defining, 22–23; demobilization and, 148–150; disarmament and, 148–150; fulfillment of mandates, 88; incentives for, 10, 24, 25; international response to violations of, 11, 12; level of attention to, 5, 21; marginality of repatriation in, 276–278; mediator/implementor coordination, 8; as precondition for United Nations involvement, 56; provisions for territorial autonomy, 127; refugees as products of, 291; repatriation of refugees and, 274–276; in El Salvador, 389–391; self-implementing, 22; self-sustaining, 118; strategies for compliance, 16; vague and expedient, 8, 9 Peace agreements, failed: Angola, 4, 14, 49, 50, 51; assigning blame for, 10–11; Cyprus, 4; Liberia, 14; policy failure and, 10; Rwanda, 33, 49, 51; Somalia, 14, 51; Sri Lanka, 51 Peace agreements, successful: desire of parties to make peace and, 4; Guatemala, 14, 32, 118; Mozambique, 14, 118; Namibia, 4, 14, 49, 118; Nicaragua, 118; power sharing in, 4; reasons for, 4–5; regional democracies and, 14; El Salvador, 4, 14, 31, 118; supportive regional/international environment in, 4, 32, 33, 44; support of third parties in, 4 Peacebuilding, 2, 74, 105, 384; bottomup, 208; building local capacity and, 327–347; grassroots, 328, 341; in Haiti, 40n37; liberal, 18–20; local innovations in, 29; local knowledge in, 329; political activities of, 214n71; postconflict, 89, 90; responsibility for, 210n6; role of traditional authority in, 334; strategies for, 204; sustained, 29; transitional sovereignty institutions in, 81; triangle, 75fig Peace implementation: activities, 204; in Angola, 62, 81; barriers to, 8–11; barriers to economic support, 191–196; in Bosnia, 81, 531–561; in Burundi, 94–95; in Cambodia, 62, 81, 499–525; civil-military cooperation and, 190; coercive strategies

S-Index

11/24/09

720

4:30 PM

Page 720

INDEX

and, 3; commitment to, 21; defensive incentives, 71; demobilization and, 26, 141–175; democratization and, 32; differences in, 1, 44; difficulty of implementation environment, 1, 44, 59, 89; disarmament in, 141–175; distortions from international presence, 208–209; ecology of, 80–86; economic priorities for, 27, 183–210; embargoes and, 153; employment needs, 201–203; enhanced consent strategies, 16; in ethnic wars, 284tab; evaluating strategies, 23–26; evaluation issues in, 43–66; funding for, 187–188; gaps in, 29; in Guatemala, 30, 62, 421–458; hegemonic, 36; human rights and, 28, 237–263; institution-building and, 203–205; mandate achievement in, 45–47; measuring impact, 199–201; measuring success, 44–54; in Mozambique, 62; multidimensional strategy in, 18; need for international attention, 20, 21; in Nicaragua, 30; noncoercive cases, 30–33; offensive incentives, 71; opposition from neighboring states, 3, 44; organizational impediments to, 9; passage of time and, 49; policy implications, 663–669; in power wars, 284tab; provision of resources for, 1, 20, 21, 24, 46, 94; public sector activities, 205; refugee repatriation and, 28–29; regional, 30, 37; research on, 1–38; resource constraints in, 3, 34; reviving economies and, 185–187; role of outside actors, 20; role of women in, 192; in Rwanda, 30, 62, 463–493; in El Salvador, 30, 31, 391–393; sequencing and, 191–193; short-term, 1, 2; social transformation strategies, 15–18; in Somalia, 62; in South Africa, 11; spoilers and, 3, 11–14, 44, 88–89; strategic coordination in, 89–111; strategic deception in, 12; sustainable, 188; tradeable commodities and, 3, 44; transitional authority in, 25; variables in, 24, 43 Peace Implementation Council, Bonn Summit (1997), 83 Peacekeeping, 3, 74, 384; in Angola, 11; monitoring and verification of com-

pliance and, 16; multidimensional, 18; in Sri Lanka, 38; traditional approaches, 16, 62, 102; variability in tasks, 46; World Bank and, 212n25 Peacemaking, 74; complexity of, 10; regional initiatives, 31; role of women in, 333–334; World Bank and, 212n25 Peace operations: establishment of, 73; local authority in, 73; for maintenance of international order, 73; from negotiated consent of parties, 73; Status of Force agreements, 73 Peña Durán, Oscar, 401 Peou, Sorpong, 34, 499–525 Perez de Cuellar, Javier, 380n43, 388 Peru, 357 Philippines: minority issues in, 125; Moro Islamic Liberation Front in, 126; Moro National Liberation Front in, 126 Plavsic, Biljana, 288 PLO. See Palestine Liberation Organization Plumb, Emily, 29, 146, 327–347 Pluralism, 333, 469; economic, 194 Policy: development, 32; donor, 198; economic, 188; failure, 10; goals, 92; implications, 663–669; macroeconomic, 193; reform, 195, 204; resources needed for, 46; results from, 21; social, 192; tax, 195 Political: authority, 152; autonomy, 126; bargaining, 246; centralization, 127; certainty, 118; competition, 90; decentralization, 127; dissent, 589–590; divisions, 37; feasibility, 3; fiscal, 203; instability, 122; institutions, 16, 17, 119, 215, 304; legitimacy, 130; liberalization, 469; mobilization, 455; negotiations, 89, 90; parties, 3, 27, 31, 32, 81, 227–228, 477; power, 31; reform, 577; reintegration, 118; repression, 354; resources, 120; settlement, 25; spoilers, 35; stability, 34, 124, 132; violence, 385, 388 Politics: demilitarization of, 3, 27, 222–230; open, 32; transitional, 79fig Pol Pot, 13, 501, 513, 524, 529n75 Ponce, Rene Emilio, 397 Popular Front for the Liberation of Palestine, 578

S-Index

11/24/09

4:30 PM

Page 721

INDEX Popular Liberation Forces, 385 Popular Revolutionary Army, 385 Porcell, Gerard, 519 Porras, Gustavo, 436 Portillo, Alfonso, 428 Powell, Colin, 361 Power: balance of, 4, 22, 31, 32, 34, 72; central, 72; competition for, 27; dispersed, 151; divisibility of, 13; military, 31; monopoly, 77; political, 31; regional, 60; separation of, 222; sharing, 4, 8, 72, 94, 121, 122, 124, 128, 130, 219–222 Prabhakaran, Velupillai, 639, 641, 643, 647, 649, 658n10 Premadasa, Ranasinghe, 645 Prendergast, John, 29, 146, 327–347 Privatization, 191, 192; insider, 211n21; political consequences of, 193 Pronk, Jan, 616 Putnam, Tonya, 28, 50, 237–263, 345–346 Qadaffi, Muammar, 604 Queretaro Agreement, 428 Quesada Gomez, Agustin, 365 Quezada Toruño, Rodolfo, 432 Quiwonkpa, Thomas, 601 Ramírez-Ocampo, Augusto, 391, 416n28 Rawlings, Jerry, 612, 615 Reagan, Ronald, 355, 357, 360, 361, 362, 378n13, 387, 403 Reconciliation: curtailment of, 37 Reform: constitutional, 438–440, 450, 460n34; democratic, 30; economic, 191, 195, 198, 373; educational, 424; electoral, 408–409; institutional, 31, 32, 46, 191; intelligence, 389–390, 398–399; judicial, 3, 29, 31, 321, 402–408, 424, 460n35; land, 297n6, 451–452; military, 76, 152, 389–390, 396–398; police, 3, 29, 31, 204, 304, 307, 308, 308tab, 390, 399–402, 479; policy, 195, 204; political, 577; public security, 448–451; security, 303; socioeconomic, 410–412, 429 Refugee Convention, 290 Refugee repatriation, 28–29, 46, 50, 82, 273–296; in Bosnia, 548–551; in Cambodia, 46; in differing types of

721

civil wars, 283–290; elections and, 288; forced, 299n22; in Guatemala, 46, 297n6, 421, 445–446; internationally organized, 281, 282–283; international role in, 292–293; in Lebanon, 29; marginality in peace accords, 276–278; media and, 293; in Mozambique, 29, 46; peace agreements and, 274–276; peace implementation and, 277tab, 279–283; physical security and, 275; psychological security and, 275; refusal of, 36; remittances and, 29, 212n39; right to, 281–282; in Rwanda, 29; in El Salvador, 386; in Sierra Leone, 292; spontaneous, 281, 282–283; in third countries, 29; timing of, 281; voluntary, 281, 282, 295, 299n27; in wars over government, 284–285, 288–289; in wars over people, 284–288 Refugees and displaced persons: in Afghanistan, 279; in Angola, 46, 276, 277tab, 299n23; in Bosnia, 29, 36, 91, 275, 277tab, 279, 287–288; in Cambodia, 277, 277tab, 279, 289; defining, 296n1; in Guatemala, 276, 277tab; in Lebanon, 276, 277tab, 288; in Liberia, 276, 277tab, 279; in Mozambique, 277tab, 279; in Namibia, 277, 277tab, 299n23; in Nicaragua, 276, 277tab, 279, 289, 356; in Palestine, 298n19; as products of peace agreements, 291; regional context, 291–292; in Rwanda, 277, 277tab, 279, 281, 286–287; in El Salvador, 29, 276, 277tab, 279; in Sierra Leone, 276, 277tab, 279; in Somalia, 275, 276, 277tab; in South Africa, 279; in Sri Lanka, 276, 277tab, 288; in Western Sahara, 277, 277tab, 279, 289; in Zimbabwe, 276, 277tab Refugee warriors, 29, 273, 275, 287, 292 Relief assistance: measuring impact of, 199–201 RENAMO. See Mozambique National Resistance Republika Srpska, 35, 83 Responding to Conflict, 104, 105 Revolutionary Democratic Front, 385

S-Index

11/24/09

722

4:30 PM

Page 722

INDEX

Revolutionary United Front, 122; egregious past of, 13 Rights: citizenship, 466; cultural, 460n35; group, 119; indigenous, 454–456; individual, 25, 119, 128, 129, 130, 551; property, 188, 203; social, 460n35; in unitary states, 128. See also Human rights Rio Group, 357 Ríos Montt, Efraín, 426, 432, 439 Riyadh Conference, 569, 593n12 Riza, Iqbal, 391 Robelo, Alfonso, 364 Roed-Larsen, Terje, 96, 99, 101 Rome Accords (1992), 168, 229 Romero, Oscar ARnulfo, 385 Rosada-Granados, Héctor, 433, 460n29 Rothchild, Donald, 25, 117–135 RPF. See Rwandan Patriotic Front Rwanda: Arusha Agreement, 8, 12, 33, 92, 286, 463; Broad-Based Transitional Government in, 93, 472, 476, 480, 486, 488; casualties in, 1; cease-fires in, 467–470; civil war in, 1, 2, 284tab; Coalition pour la Défense de la République in, 33, 93; Committee for Restarting Pastoral Initiatives, 342; confidence-building in, 33, 62; conflicting strategies in, 92–94; death squads in, 486; demobilization in, 479; democratization in, 92; economic growth, 465; “Five Musketeers” in, 99; Fundamental Law in, 476; genocide in, 93, 112n7, 284tab, 293, 463, 473; historical background, 464–467; humanitarian assistance to, 112n7; human rights in, 249, 250, 262tab, 268n49, 337, 477, 484; Hutu in, 33, 463, 464, 465, 474, 485, 486; integration of military, 479; Interahamwe in, 467; international criminal tribunals in, 239, 268n50; International Monetary Fund in, 466; judiciary in, 478; Kanyarwanda in, 337; Mouvement Révolutionnaire National pour le Développement in, 92, 93, 464, 470, 472, 473, 485; national army in, 93; Neutral International Force in, 93, 479; nongovernmental organizations in, 481; Operation Support Hope in, 46; opposition to implementation,

484; Organization of African Unity and, 92, 93, 468, 469, 470, 475, 476; peace agreement mediation, 470–476; peace agreements in, 49, 51; peace implementation in, 30, 59tab, 62, 463–493; peacekeeping in, 51, 93; police reform in, 308tab, 310, 479; power sharing in, 467–470, 472, 473, 477; Protocol of Agreement on the Integration of Armed Forces, 479; Protocol of Agreement on the Rule of Law, 478; psychosocial trauma iniatives, 335; Radio Agatashya in, 339; repatriation of refugees in, 29, 92, 277, 277tab, 278, 279, 281, 286–287, 467–470, 479; Revolutionary Armed Forces in, 467; Rwandan Patriotic Front in, 92, 93, 286, 287, 463, 465, 466, 468, 469, 472, 473, 474, 482, 483; social integration in, 479; spoilers in, 93; strategic coordination in, 92–94; structural adjustment programs, 466; supremacist groups in, 33; Transitional National Assembly, 476; truth commissions in, 268n49; Tutsi in, 33, 464, 465, 486; UN Assistance Mission in Rwanda in, 34, 93, 482, 483, 484, 485, 486, 487, 488, 489; United Nations in, 6; United States and, 470, 484; UN Observer Mission Uganda-Rwanda in, 475, 482, 484; UN special representatives of the Secretary-General in, 97; violations of peace agreement, 12 Rwandan Association for Christian Workers, 337 Rwandan Patriotic Front, 92, 93, 286, 287, 463 Rwegasira, Joseph, 488 El Salvador, 1, 29, 224–225; Ad Hoc Commission in, 80, 397, 403, 417n45; Alianza Republicana Nacionalista in, 387, 392, 393, 399, 403, 409; Armed Forces of Liberation in, 385; Association of Demobilized Army Troops in, 148, 149–150; cease-fire in, 389; Chapultepec Accord, 389; Christian Democratic Party in, 415n9; civilian security in, 307, 399–402; civil war

S-Index

11/24/09

4:30 PM

Page 723

INDEX in, 14, 30, 76, 284tab, 383–389; confidence-building in, 32–33, 61; coordination mechanisms in, 89; corruption in, 402; death squads in, 87n18, 385, 400, 403; demilitarization of politics in, 223; demobilization in, 152, 164–165, 383, 389, 393–396; demobilization of guerrillas, 31; disappearances in, 421; disarmament in, 142, 156, 157, 164–165, 303, 393–396; dollarization in, 212n39; economic crisis in, 387; economic development in, 76; elections in, 31, 46, 165, 216, 217, 383, 388, 408–409; electoral fraud in, 228; electoral reform in, 408–409; electoral supervision in, 229–230; Farabundo Martí National Liberation Front in, 7, 76, 157, 224, 225, 383, 385, 387, 388, 391, 393, 394, 395, 396, 399, 403, 408; “Four plus One in,” 99, 392; Human Rights Agreement, 388; human rights in, 31, 46, 241, 242, 247, 271tab, 386, 388, 390, 399, 402–408, 416n23; institutional reform in, 46; intelligence reform in, 398–399; interim security in, 316–317; International Monetary Fund in, 394; investment in, 411; judiciary in, 390; Land Bank in, 396; land ownership in, 391, 394, 395; land redistribution in, 148, 149–150; land transfer in, 395, 410, 411; military reform in, 389–390, 396–398; military repression in, 385; National Civilian Police in, 123, 390, 398, 400, 401, 402; National Commission for the Consolidation of Peace in, 80, 224, 225, 391, 400; National Democratic Organization in, 385; National Guard in, 393, 400; National Intelligence Directorate in, 390, 399; National Police in, 390, 400; National Reconstruction Plan in, 391, 395, 410; National Resistance in, 385; nongovernmental organizations in, 257, 325n23; peace agreements in, 4, 14, 31, 49, 389–391; peace implementation in, 30, 31, 59tab, 118, 391–412; peacekeeping in, 51, 384; police controls in, 321; police reform in, 31, 308tab, 310,

723

311–316, 390, 399–402; political violence in, 385, 388, 397, 403, 419n95; Popular Liberation Forces in, 385; Popular Revolutionary Army in, 385; privatization in, 152; proportional representation in, 123; recalendarization agreements in, 384, 393, 400; repatriation of refugees in, 276, 277tab, 278, 279, 386, 410; Revolutionary Democratic Front in, 385; Salvadoran Security Agency in, 385; San José Agreement in, 248, 267n28; socioeconomic issues in, 390–391, 410–412; State Intelligence Office, 399; strategic coordination in, 96, 98, 102; Supreme Electoral Tribunal, 408; transition to peace, 383–414; truth commissions in, 46, 80, 397, 403, 407, 419n88; UN Development Programme in, 408, 412; unemployment in, 412; United Nations in, 6, 15, 16, 31, 383, 384, 388; United States in, 31, 386, 397; UN Observer Mission in El Salvador in, 17, 80, 87n18, 157, 164–165, 256, 267n28, 389, 391, 392, 393, 394, 397, 400, 401, 403, 406, 407, 409; verification/monitoring in, 388, 391–393; Workers Revolutionary Party in, 385; World Bank in, 394 Salvadoran Security Agency, 385 Sanderson, John, 34, 505 Sandline, 148 Sandoval, Mauricio, 399 Sankawulo, Wilton, 614, 615 Sankoh, Foday, 13, 207, 624 Sapoa Agreement, 362 Saudi Arabia, 571, 593n12 Save the Children, 344 Savimbi, Jonas, 11, 13, 222 Sawyer, Amos, 607 Scheverdnaze, Eduard, 388 Search for Common Ground, 105 Security, 226–227; -building, 120–128; compellence of compliance and, 6; constitution of, 82; cultural/social protections and, 133–134; demobilization and, 26; deterence of violations and, 6; dilemma, 71; elite pacts, 123–125; federalism and, 125–128; formal group-based systems, 120–128; group-based formulas,

S-Index

11/24/09

724

4:30 PM

Page 724

INDEX

130–134; guarantees, 5–8, 14, 26; immediate, 118; individually-based designs, 128–130; intergroup relations and, 119–130; international, 29; legal, 275; material, 275; physical, 275; psychological, 275; reform, 303; regional, 30; regional autonomy and, 125–128; short-term, 118, 119, 120; successful implementation and, 32; territorial decentralization and, 126 Security, civilian, 3, 29, 303–323; in Bosnia, 91; competing international interests in, 319–320; effects of war on, 305–307; international civilian police personnel and, 314–316; international military forces and, 313–314; long-term development and, 316–317; models of policing, 317–318; peace agreements and, 307–311; police reform and, 304, 307; preexisting local forces, 312–313; quick-start local civilian force, 312; in El Salvador, 303, 399–402; security force composition, 318–319; in South Africa, 303; during transitions to peace, 311–316 Security forces: accountability and, 308; competing international interests in, 319–320; composition of, 318–319; external/internal controls, 321–322; in Guatemala, 448–451; human rights violations and, 306; limited capacity of, 320; modernization of, 308; need for functioning judiciary, 306; professionalization of, 304, 308; protection of state rather than population, 305 Senegal, 605, 610 Serbia, 537 Serrano Elías, Jorge, 427, 428, 432, 445 SFOR. See Stabilization Force Sham‘un, Camille, 568, 592n9 Sharon, Ariel, 589 Sierra Leone, 22, 607; Abidjan Peace Agreement, 292; amnesties in, 240, 264n6; civil war in, 284tab; Conakry Agreement and, 173, 292; demobilization in, 173–174; disarmament in, 173–174; Economic Community of West African States Cease-Fire Monitoring Group in, 56; elections in, 174, 216, 217; human rights in,

240, 243, 247, 262tab; Lomé Peace Accord, 121–122; peace implementation in, 59tab; police reform in, 308tab, 309, 310; power sharing in, 121; repatriation of refugees in, 276, 277tab, 279, 292; Revolutionary United Front in, 13, 122; strategic coordination in, 90; UN Observer Mission in Sierra Leone in, 173– 174, 247; UN special representatives of the Secretary-General in, 113n21 Sika, Moses, 344 Slovenia, 533 Smuggling, 153 Social: capital, 185, 203; contract, 327; inequality, 357; infrastructure, 193; institutions, 16; integration, 479; interaction, 120; justice, 342; mobilization, 73; organization, 333; peace, 191, 198; policy, 192; protection, 119, 123; reconstruction, 77; revolution, 465; rights, 460n35; security, 452; spending, 442; transformation, 15 Somalia: Addis Ababa Agreement and, 243; arms embargoes in, 174; civil war in, 284tab; coercive strategy in, 63; continuation of conflict in, 125; decentralization in, 126; demobilization in, 174–175; disarmament in, 142, 143, 174–175; forcible intervention in, 72; human rights in, 243, 262tab; indigenous associations in, 343; peace agreements in, 14, 22, 51; peace implementation in, 59tab, 62; peacekeeping in, 51; peace radio in, 339; police reform in, 308tab, 310; repatriation of refugees in, 276, 277tab; security forces in, 323n3; unemployment in, 275; Unified Task Force in, 73; United Nations in, 56; United States in, 6; UN Operation in Somalia in, 324n11 Somoza Debayle, Anastasio, 354 South Africa: African National Congress in, 132; Center for Conflict Resolution in South Africa, 338–339; civilian security in, 303; civil society in, 328; demobilization in, 147–148; disarmament in, 144; federalism and, 127; Independent Complaints

S-Index

11/24/09

4:30 PM

Page 725

INDEX Directorate, 321; industrial economy in, 15; Inkatha Freedom Party in, 132; minority issues in, 127; National Party in, 132; National Peace Accord in, 341; peace agreements in, 22; peace implementation in, 11; police controls in, 321; police reform in, 308tab, 310; proportionality in, 132; repatriation of refugees in, 279; Search for Common Ground in, 340; South African National Defense Force in, 147 Southern Opposition Bloc, 355 South West African People’s Organization, 133, 169–171, 266n23 Soviet Union: ethnonationalist issues in, 127 Spain, 362, 392, 423; problem-solving workshops in, 336 Spear, Joanna, 26, 141–175, 337 Spoilers, 3, 11–14, 25, 33, 67tab, 88–89, 207; in Cambodia, 34; coercion of, 3; co-optation into peace process, 207; coordination of strategy and, 95; defining, 13; establishing peace and, 71; exploitation of opportunity by, 95; intent of, 12; isolation of, 207; management of, 12; military, 35; in peace process, 500; political, 35; in Rwanda, 93; typology of, 12; use of violence by, 12; willingness to negotiate in future, 13 Sri Lanka: breakdown of implementation, 646–654; civil war in, 284tab, 632–634; coercive strategies in, 36–38; contradictions in agreement, 640–646; decentralization in, 126; demobilization in, 642; disarmament in, 142, 642, 647; historical background, 634–636; human rights in, 262tab; India in, 6, 23, 37–38, 633, 636–640, 643–654; interventionist policy in, 633; land mines in, 650; Liberation Tigers of Tamil Eelam in, 38, 634, 637, 641, 643; major points on agreement, 640–646; peace agreements in, 51, 631–658; peace committees in, 341; peace implementation in, 37–38, 59tab; peacekeeping in, 51; pogroms in, 632; police reform in, 308tab, 309; proportionality and, 132; repatriation of refugees

725

in, 276, 277tab, 278, 288; repression in, 634; strategic context of agreement, 632–634; Tamils in, 38, 631, 632, 634, 636, 637, 638, 639, 640, 658n2, 658n10 Stability, postagreement, 117–135 Stabilization Force, 6 Stanley, William, 29, 32, 125, 303–323, 421–458, 517 State: authority, 71, 72; building, 120; capacity, 25, 27, 119, 120, 151; collapsed, 56, 73, 120; institutions, 119, 151; legitimacy, 72, 120; militarization, 76; rehabilitation, 479; revenues, 124; strengthening, 25; traditional, 76; unitary, 119, 129; weak, 120; weakness, 151 State of Cambodia, 501, 502, 503, 508–509, 512, 513 Stedman, Stephen John, 1–38, 43–66, 247, 663–669 Strategies: accomodation, 37; coercive, 36–38, 63, 64, 102; compellence, 6; competing, 91; confidence-building, 6, 33, 61; conflicting, 92–94; economic, 189–210; enhanced consent, 16; human rights advocacy, 28; implementation, 23–26, 43; inappropriate, 63; intergroup relations, 119–130; mixed, 33–36; multidimensional, 18; natural disaster model, 189–191; neoliberal, 191; peacekeeping, 16; political settlement, 25; postwar stabilization and reconstruction model, 191; social transformation, 15–18; for success, 61–65; transitional, 74–75; trusteeship, 36, 533 Sudan: minority issues in, 125; nongovernmental organizations in, 179n5; proportionality in, 131 SWAPO. See South West African People’s Organization Syria: conflict with Israel, 37; in Lebanon, 6, 36–38, 56, 567, 569, 572, 574; priorities in implementation of Ta’if Agreement, 575–576 Ta’if Agreement, 122, 245, 266n25, 567–592. See also Mozambique Tailor, Tamba, 614 Tamil Eelam, 38, 634 Tamils. See Sri Lanka

S-Index

11/24/09

726

4:30 PM

Page 726

INDEX

Tanzania, 467, 468, 469, 473, 474, 488; in Rwanda, 92 Tax(es): excise, 205; policy, 195; revenues, 422, 436, 442, 452, 453; value-added, 453 Taylor, Charles, 37, 124–125, 148, 218, 233n10, 599, 600, 602, 603, 604, 605, 608, 609, 610, 611, 612, 613, 614, 615, 619–620, 621, 623, 625, 626 Technical assistance, 202 Tela Agreement, 364, 365 Tesoro Beach Agreement (1989), 363 Togo, 605, 606 Tolbert, A. Benedict, 603 Tolbert, William, 601 Toncontin Accord, 370 Tripartite Agreement (1985), 569 Tripartite Arab Committee, 573, 593n15 Truth commissions, 80, 268n49, 268n57, 397, 403, 407, 434; in Guatemala, 46; in El Salvador, 46 Tubman, William, 601 Tudjman, Franjo, 83, 534, 551 Tutsi peoples, 33, 464, 465, 486 Twagiramungu, Faustin, 477 Uganda, 466, 467, 468, 469, 611; National Resistance Army in, 466, 467; reintegration and, 146; tax policy in, 195 UNAMIR. See UN Assistance Mission in Rwanda UN Angola Verification Mission, 82, 143, 152, 155, 159 UN Assistance Mission in Rwanda, 34, 93, 482, 483, 484, 485, 486, 487, 488, 489 UNAVEM. See UN Angola Verification Mission UN Civilian Police, 313, 314, 315, 316, 318, 320, 517 UN Development Programme, 98, 264n1, 320–321, 408, 412, 618; in Cambodia, 508; Emergency Response Division, 210n1; in Guatemala, 443, 446; Round Tables, 199 UN Educational, Scientific, and Cultural Organization, 340, 446 Unemployment: measuring needs during implementation, 201–203

UNHCR. See UN High Commissioner for Refugees UN High Commissioner for Human Rights, 251, 264n1, 270n66, 540 UN High Commissioner for Refugees, 108, 109, 253, 264n1, 274–275, 279, 287, 290, 299n25, 443, 464, 469, 479, 481, 540, 549; in Bosnia, 91; in Kosovo, 102, 103 Unified Task Force, 324n11; in Somalia, 73 UN Interim Forces in Lebanon, 569 UN International Children’s Emergency Fund, 335, 340, 342, 617 Union for the Total Independence of Angola, 152, 153, 157, 159, 160, 161, 181n45, 224 UNITA. See Union for the Total Independence of Angola UNITAF. See Unified Task Force United Kingdom: in Bosnia, 190; Department for International Development, 189, 205, 210n1; Overseas Development Agency, 190 United Liberation Movement of Liberia for Democracy, 604 United Nations: in Angola, 6; balance in core constituency, 19; in Bosnia, 544; in Cambodia, 16; Department of Peacekeeping Operations, 392; Department of Political Affairs, 105; evaluating effectiveness of, 52–54; failed peace implementations and, 33; in Guatemala, 6, 15, 441–443, 459n15; human rights role, 241, 244–245, 264n1, 270n65; liberal transformation strategies of, 15; in Liberia, 605; limited involvement by, 33; in Mozambique, 6; in Namibia, 6, 19; Office for the Coordination of Humanitarian Affairs, 106; relevance to European security, 114n42; in Rwanda, 6; in El Salvador, 6, 15, 16, 31, 383, 384, 388; Secretariat, 25, 64, 65; Security Council, 45–46, 60, 65, 97, 424, 475, 484, 487, 489, 506; in Somalia, 56; strategies for peace implementation, 15. See also specific agencies United States: Bosnia and, 90, 536; Cambodia and, 77; Federal Bureau of Investigation, 319; Guatemala and,

S-Index

11/24/09

4:30 PM

Page 727

INDEX 423, 426, 432; Lebanon and, 573; Liberia and, 37, 604; Nicaragua and, 355, 360, 361, 362, 377n7; Operation Support Hope by, 46; Rwanda and, 470, 484; El Salvador and, 31, 386, 397; Somalia and, 6 Universal Declaration of Human Rights, 504 UN Observer Group in Central America, 171–172, 365, 370, 371, 372, 376, 381n70, 381n71 UN Observer Mission for the Verification of Elections in Nicaragua, 367 UN Observer Mission in El Salvador, 17, 80, 87n18, 157, 164–166, 256, 267n28, 389, 391, 392, 393, 394, 397, 400, 401, 403, 406, 407, 409 UN Observer Mission in Liberia, 610, 611 UN Observer Mission in Sierra Leone, 172–174, 247 UN Observer Mission Uganda-Rwanda, 475, 482, 484 UNOC. See UN Operation in the Congo UNOMSIL. See UN Observer Mission in Sierra Leone UN Operation in Mozambique, 81, 148, 149, 208, 245 UN Operation in the Congo, 73 UNPROFOR. See UN Protection Force UN Protection Force, 73 UN special representatives of the Secretary-General: appointment of, 113n22; in Burundi, 94; funding issues, 205; role of, 96–97, 113n21; in Rwanda, 97; support from UN for, 96–97; timing of deployment, 97 UNTAC. See UN Transitional Authority in Cambodia UNTAG. See UN Transitional Assistance Group UN Transitional Administration, 73, 83, 162, 194 UN Transitional Assistance Group, 169–171, 245 UN Transitional Authority in Cambodia, 11, 34, 46, 81, 184, 248, 254, 258, 266n20, 499, 500, 503, 505–518, 526n19; abandonment of confidencebuilding, 35; disregard of violence, 34; resources for, 17

727

UN Verification Mission in Guatemala, 424, 434, 440–441, 442, 443, 444, 446–447, 451, 453 Uruguay, 357 U.S. Agency for International Development, 189, 210n1, 330, 332, 341, 410, 411, 444 USAID. See U.S. Agency for International Development Uwilingiyimana, Agathe, 484 Vendrell, Francesc, 432 Venezuela, 362, 392, 423 Vieira de Mello, Sergio, 97 Vietnam, 501, 502; occupation of Cambodia by, 77 Vietnam War, 77 Villalobos, Joaquín, 403 Violence: acceptance of, 305; expectation of, 305; expression of grievances and, 306; of Khmer Rouge, 34; in Lebanon, 36; political, 303, 385, 388, 524; repatriation of refugees and, 275; by spoilers, 12; termination of, 24, 50, 51 Warlordism, 343 Weber, Max, 73 Western Sahara, 39n35; civil war in, 284tab; police reform in, 308tab, 309; repatriation of refugees in, 277, 277tab, 279, 289 Woewiyu, Tom, 602, 613, 614 Woodward, Susan, 27, 183–210, 218, 332 Workers Revolutionary Party, 385 World Bank, 101, 104, 197, 394; Articles of Agreement, 191; in Bosnia, 186, 204, 540; Consultative Group, 199; Country Assistance Strategy, 203; economic neutrality and, 192; in Guatemala, 443, 444; in Lebanon, 585; National Emergency Recovery Program, 585; natural disaster model and, 190; Operations Evaluation Department, 195; PostConflict Unit, 201, 210n1, 212n26; Priority Reconstruction and Recovery Program, 204; sovereign lending by, 193–194; Transitional Support Strategy, 203 World Food Program, 618

S-Index

11/24/09

4:30 PM

Page 728

728 World Health Organization, 104 Wright, Jim, 357 Yamoussoukro Agreements, 225, 608 Yugoslavia: disintegration of, 533; ethnonationalist issues in, 127; international criminal tribunals in, 239 Zahar, Marie-Joëlle, 13, 36, 125, 567–592 Zaire, 466, 467, 468, 469; civil war in, 46 Zamora, Rubén, 409

INDEX Zimbabwe: civil war in, 6, 284tab; demobilization in, 152, 176; disarmament in, 176; elections in, 175–176, 216, 217; human rights in, 262tab; Lancaster House Agreement, 6; land distribution and, 148, 149–150; LRHare formula in, 133; nongovernmental organizations in, 155; peace agreements in, 2, 22; peace implementation in, 2, 59tab; power sharing in, 122–123; proportionality and, 132, 133; reintegration and, 145; repatriation of refugees in, 276, 277tab

S-About the Book

11/24/09

4:31 PM

Page 729

ABOUT THE BOOK

Why do some peace agreements successfully end civil wars, while others fail? What strategies are most effective in ensuring that warring parties comply with their treaty commitments? Of the various tasks involved in implementing peace agreements, which are the most important? These and related questions—life-or-death issues for millions of people today—are the subject of Ending Civil Wars. Based on a study of every intrastate war settlement between 1980 and 1997 in which international actors played a key role, Ending Civil Wars is the most comprehensive, systematic study to date of the implementation of peace agreements—of what happens after the treaties are signed. Covering both broad strategies and specific tasks and presenting a wealth of rich case material, the authors find that failure most often is related not only to the inherent difficulty of a particular case, but also to the major powers’ perception that they have no vital security interest in ending a civil war. Stephen John Stedman is senior fellow at the Center for International Security and Cooperation, Stanford University. Donald Rothchild is professor of political science at the University of California, Davis. Elizabeth Cousens is director of the Conflict Prevention and Peace Forum in New York City.

729